download ENC 2007 Transactions - European Nuclear Society

© 2007 European Nuclear SocietyRue de la Loi 571040 Brussels, BelgiumPhone + 32 2 505 30 54Fax +32 2 502 39 02E-mail ens@euronuclear.orgInternet www.euronuclear.orgThese transactions contain all contributions submitted by 14 September 2007.The content of contributions published in this book reflects solely the opinionsof the authors concerned. The European Nuclear Society is not responsiblefor details published and the accuracy of data presented.

PlenaryMonday

PlenaryTuesday

PlenaryWednesday

RESEARCH – A SINE QUA NON FOR THE NUCLEARSECTORS. WEBSTER 1European Commission, Research Directorate-General, Directorate J – Energy (Euratom)B-1049, BrusselsABSTRACTNot only does 2007 mark the 50 th anniversary of the Euratom Treaty, it is also a landmarkyear for nuclear research in Europe: the 7 th Euratom Framework Programme has started andthe first European Technology Platform in the nuclear field, a major initiative fosteringenhanced cooperation between leading players in nuclear R&D, will be launched on 21 stSeptember. Furthermore, in January the Commission published its long-awaited “energypackage” framing the coming debate over the “Energy Policy for Europe” and the measuresneeded to counter the increasingly urgent problems of security of supply, competitivenessand climate change. In particular, the “Strategic Energy Technology Plan” is an extremelyimportant initiative, and as one of the technologies under investigation, the nuclear sectorhas a real opportunity to influence strategic thinking and decision making. One thing isclear – political and societal acceptance of any nuclear renaissance must go hand in handwith an integrated, effective, well-funded and long-term European research effort.1. IntroductionAs one of the original treaties of Rome, the Euratom Treaty celebrated its 50 th birthday in March thisyear. The Treaty prioritised research, in particular promoting the establishing of a Communityresearch programme in the area of nuclear science funded out of the EU budget. This led to theadoption of the first multi-annual Euratom Framework Programme in the 1980s, a model for researchfunding that was also borrowed by the more general EC Treaty. Indeed, 2007 also sees the launch ofthe EU’s 7 th research Framework Programme (FP-7), heralding a significant increase in the overall EUfunding for R&D in general. This is recognition by the European Institutions of the fundamental rolethat research must play, as one of the three pillars of the Lisbon Agenda, in the EU’s overall socioeconomicand political strategy for growth, jobs, competitiveness and the development of theknowledge-based society. Within the research field, energy has been identified as a priority in both theEC (non-nuclear) and Euratom programmes, and initiatives are being launched that could herald majorchanges to our energy supply and usage in the future. In parallel, there are important developments inthe area of energy policy and strategy at the European and global level that will also have profoundimplications. In section 2, these EU initiatives in the areas of energy policy will be outlined, followedin section 3 by a summary of the status of R&D in the nuclear field, including developments withinthe Euratom FP and the initiative to establish the “Sustainable Nuclear Energy Technology Platform”.2. Developments in EU energy policyOn 8/3/2006 the European Commission (EC) published a Green Paper entitled “A European Strategyfor Sustainable, Competitive and Secure Energy” [1] that kicked off a major debate on energy supplyand security as well as greenhouse gas (GHG) emissions and climate change. The Green Paper clearlystated that an EU energy policy should respond to three main objectives: sustainability,competitiveness and security of energy supply. In this context it considered a number of priority areas,including the diversification of the energy mix, an integrated approach to tackling climate change andthe establishing of an EU energy technology plan. Though much of the document referred to energy in1The views expressed in this paper are those of the author and do not necessarily reflect those of the EC

general, without distinction, and entire sections were devoted to energy efficiency and renewables,there were nonetheless important references to nuclear energy and innovative nuclear technology.2.1 Energy packageFollowing the Green Paper, on 10/1/2007 the EC proposed an integrated energy and climate changepackage under the banner “Energy for a Changing World” to cut GHG emissions for the 21 st Century,increase the EU’s independence and security of supply and boost competitiveness. Nuclear energyfeatures at several points and is clearly implicated in a number of the proposed measures. Anoverarching Communication entitled “An Energy Policy for Europe” (EPE) [2] addresses all thechallenges and issues. On the subject of nuclear, it recognises the important contribution that nuclearpower makes in limiting GHG emissions and in Europe’s security and independence of supply. Itreiterates that each EU Member State must decide for itself whether to resort to this form of energy,but nonetheless endorses further expansion of nuclear generation providing the highest standards ofsafety, security and non-proliferation are maintained, as required by the Euratom Treaty. Detailedinformation on the nuclear sector is presented in the “PINC” [3], or Illustrative Nuclear Programme,foreseen under Art. 40 of the Euratom Treaty and presented as part of the overall package. A furtherdocument – “Towards a European Strategic Energy Technology (SET) Plan” [4] – introduces anotherinitiative of considerable relevance to the future development of nuclear energy in Europe (see 2.3).2.2 Conclusions to the European Council summitThe initiatives put forward by the EC in the energy / climate-change package were a major topic ofdiscussion by the Member States at the spring summit in Brussels on 18-19 March 2007. This led tothe formal adoption of a number of key policies in the area of energy / climate change as well ascommitments on use of renewables, biofuels and GHG reduction targets, which must now bedeveloped further by the EU Institutions, leading to possible introduction of new EU legislation.At the summit, the EU Member States endorsed a strategy to reduce the EU’s GHG emissions by 20%relative to 1990 levels by 2020, and to increase the contribution of renewables to 20% of primaryenergy by the same date. However, the means to achieve these goals, and the respective contributionsof individual Member State, have yet to be decided. The Council Presidency, in its conclusions to thesummit [5], goes on to confirm that the Member States approve the development of an SET-Plan, and:−−−notes the Commission's assessment of the contribution of nuclear energy in meeting the growingconcerns about safety of energy supply and CO2 emissions reductions while ensuring thatnuclear safety and security are paramount in the decision-making process;confirms that it is for each and every Member State to decide whether or not to rely on nuclearenergy and stresses that this has to be done while further improving nuclear safety and themanagement of radioactive waste, and to that effect it:• supports R & D on waste management, particularly under the 7th FP;• can envisage the creation of a high-level group on nuclear safety and waste management.suggests that broad discussion takes place among all relevant stakeholders on the opportunitiesand risks of nuclear energy.”The high-level group referred to above will be coordinated by the EC’s Directorate-General forEnergy and Transport in close collaboration with the national nuclear regulatory authorities, and is anatural outcome from the two years of discussions in Council following attempts by the EC in 2003-04to introduce binding legislation in the areas of radioactive waste management and nuclear safety. Thestakeholder discussion group, or “nuclear forum”, would be established in close consultation with thenuclear sector, in particular industry, and other interested groups. Following expressions of interestfrom both the Czech Republic and Slovakia, the forum will be hosted alternately in Bratislava andPrague. However, it is too soon to know the exact composition and mandates of these two groups.

2.3 The Strategic Energy Technology PlanThe SET-Plan will be the principal vehicle for identifying where action by the EU and Member Statescan accelerate the development and market deployment of key technologies capable of responding tothe challenges of GHG emissions, sustainability, security and independence of supply. Crucially, bothnuclear fission and fusion systems are amongst the technologies under consideration. Three time-linesare considered: up to 2020, 2030 and 2050. In [4] it refers to the excellent work carried out by theexperts in the EC’s Advisory Group on Energy (AGE) over the previous 2-3 years, and their reports[6] represent an objective appraisal of the pros and cons of the various energy technologies.The SET-Plan is currently being prepared by the EC services for adoption in November and discussionby the Member States at the spring Council in 2008. In March-May 2007, the EC organised a series ofhearings with key actors in the respective technology areas, which drew heavily on expertise inexisting, and (in the case of nuclear fission – see 3.2) “embryonic”, technology platforms. The expertswere asked specifically what actions would be needed at EU and national levels to ensure that the fullpotential of the various energy technologies could be attained. These actions constitute the essence ofthe SET-Plan. In this regard, both “technology push” or “market pull” instruments, including possiblelegislation, can be used to help accelerate the various technologies to the market: a “business as usual”strategy is not an option! The challenge for the R&D community has been to provide clear, wellargued and rational messages with the correct level of ambition. The European nuclear R&D sectorhas contributed to this process, their contribution being allied closely to the ambitious vision of thenew technology platform (see 3.2) and the corresponding technology roadmaps being prepared for theplatform’s launch.3. Developments in EU researchAt the end of 2006, the EC launched the 7 th Framework Programme (FP-7, 2007-2013) and the 7 thEuratom Framework Programme (2007-2011). FP-7 will channel more than €2.3 billion to nonnuclearenergy research and, through Euratom, another €1.95 billion will be spent on research intofusion energy (of which approximately half will be for ITER). Both these funding envelopes representsignificant increases relative to FP-6 (2002-2006), though in view of the perceived R&D challenges inareas such as renewables, fuel cells, photovoltaics, etc., this level of funding is considered inadequateby many. Regarding research in nuclear fission and radiation protection, some €287 million will beavailable, though this represents no increase over inflation relative to FP-6. The reasons for thispersistent low level of funding are essentially political, stemming from the opposition of someMember States to nuclear power. Nonetheless, FP-7 Euratom [7] will continue to support research onadvanced nuclear systems for the benefit of the Community as a whole (see 3.1). Other priority areasinclude, as in FP-6, R&D on management of radioactive waste, nuclear installation safety andradiation protection, with support for infrastructures and human resources as key cross-cutting issues.3.1 Euratom support to Generation-IV researchGen-IV technology represents a revolutionary development relative to current designs of nuclearreactors. It promises vastly improved resource sustainability through the development of fast reactorsand associated fuel cycles (enabling at least 50 times more energy to be extracted from the samequantity of uranium), even higher levels of safety than current designs (via increased dependence onpassive and intrinsic safety attributes), co-generation of electricity and heat for use in a variety ofchemical or industrial processes, and full actinide recycling thereby greatly reducing quantities oflong-lived waste for disposal and minimising the risk of nuclear proliferation.Pre-conceptual design research on the six most promising innovative nuclear concepts is beingcoordinated at the global level by the Generation-IV International Forum (GIF). Euratom became amember of the GIF in May 2006 following the approval granted by Member States in the Council

Decision of December 2005. Other members include USA, Japan, Korea, Canada, France, UK andSwitzerland, with China, Russia and S. Africa all set to join during 2007. The Euratom FP-6 projectsin Table 1 are a focal point for the Euratom contribution to the GIF research effort. Further projects inthe field of Gen-IV technology are in preparation and will be supported under FP-7.Project acronym and title Key areas of R&D Coordinatingorganisation & total no.of partners*Start date &durationTotal budget / EUcontributionRAPHAEL: Reactor forProcess Heat, Hydrogen &Electricity GenerationPerformance of fuel, materialsand components of VHTRAREVA (FR)33 partners (10 countries)15/4/0548 months€19.8M / €9.0MGCFR: Gas-cooled FastReactorHPLWR Phase 2: HighPerformance LWR – Phase 2ELSY: European LeadcooledSystemALISIA: Assessment ofLiquid Salts for InnovativeApplicationsEISOFAR: Roadmap for aEuropean Innovative SodiumcooledFast ReactorConceptual design, directcoolant cycles, transmutation,safety, etc.Critical issues and technicalfeasibility of SCWRCore design, PA, main components& systems, systemintegration, safety, etc.Support action – preparationof future activities/proposalsSupport action – preparationof future activities/proposalsNNC Ltd. (UK)9 partners (7 countries)FZK (DE)10 partners (8 countries)ANSALDO ENERGIAS.p.A. Nuclear (IT) 20partners (12 countries)CEA (FR)15 partners (9 countries)CEA (FR)14 partners (9 countries)1/3/0548 months1/9/0642 months1/9/0636 monthsJan. 071 yearJan. 071 year*only partners from EU Member States and Euratom Associate Countries can receive EU fundingFor a presentation of all FP6 projects refer to http://cordis.europa.eu/fp6-euratom/projects.htmTab 1: Overview of FP6 support to Gen-IV systems€3.6M / €2.0M€4.65M / €2.5M€6.5M / €2.95M€250k / €500k€250k / €500kThe Gen-IV objectives are ambitious and will require an extensive and concerted Europeanprogramme of research, coordinated at the global level through the GIF. The EU strategy for R&D andeventual deployment of advanced reactor and fuel cycle technology is being developed in the contextof the new technology platform (3.2) and will be clearly reflected in the forthcoming SET-Plan.3.2 The Sustainable Nuclear Energy Technology PlatformA technology platform brings together all key research stakeholders – industry, research institutes,academia, even regulatory authorities – around a common “vision” for research, development anddeployment in a particular sector. The stakeholders agree collectively on a “strategic research agenda”and then cooperate using their own financial and human resources to implement this agenda. So far,some 30 technology platforms have been launched in Europe, including several in the area of nonnuclearenergy technology.The first technology platform in the nuclear field – the Sustainable Nuclear Energy TechnologyPlatform (SNE-TP) – will be formally launched on 21/9/2007 at a major event to take place inBrussels [8] under the auspices of the EC and in the presence of the EC Commissioner for Research,Janez Potočnik. Members of the European Parliament will also play an active part, as will many highranking officials from the nuclear sector. The scope of SNE-TP includes nuclear installation safety andnuclear systems (including P&T and the fuel cycle), related research infrastructures and humanresources. It is built around three “pillars”: the safety of current generations of light-water reactors; thedevelopment of next generation fast reactors with closed fuel cycles and full actinide recycling;very/high temperature reactors (V/HTR) for the cogeneration of both electricity and process heat forindustrial applications. The platform will be the key technical nuclear forum in Europe, and willensure that Europe’s world-leader status in nuclear technology can be further consolidated and

extended to include advanced nuclear technology. It is an essential complement to the SET-Planinitiative.4. ConclusionsThe EC acknowledges the role played by nuclear power and the potential it has to respond to theenergy challenges faced by Europe. The Council endorses this view, at the same time recognising thatthe choice whether or not to resort to nuclear energy must be taken at national level. It also reiteratesthe importance of safety and waste management, proposing that a high-level group of regulators be setup, and believes a forum to discuss the pros and cons of nuclear in general should also be established.The Euratom fission programme in FP-7 has not benefited from the same increase in fundingwitnessed in the fusion and non-nuclear energy sectors. However, it continues to support importantCommunity research in the area of nuclear safety and systems, including advanced nuclear technology,waste management and radiation protection, thereby stimulating and further structuring the researchefforts across Europe. This process has already started during FP-6, thanks mainly to the new fundinginstruments such as Integrated Projects and Networks of Excellence, but must now be consolidated bythe establishing of a technology platform thereby better integrating contributions from national – andindustrial – programmes. The Sustainable Nuclear Energy Technology Platform is a major initiativethat is attracting widespread support, and should enable the available resources in this field to be betterutilised. The next five years will mark a crucial period in nuclear research. In particular, the viabilityof the various Gen-IV systems will continue to be investigated, culminating in decisions onpilot/demonstration facilities to take this technology through to industrial deployment.The SET-Plan is a bold and challenging initiative. It will cover the full range of low carbon energytechnologies, and is being prepared by the EC, with input from a range of experts, for discussion bythe Member States at the spring 2008 Council. The Plan clearly demonstrates the broad portfolioapproach of the EC, eloquently summarised in a speech by the Commissioner for Research, JanezPotočnik:“The EC believes that the answers to the EU’s energy problems lie in developing a diverse mix ofoptions supported by appropriate strategies and policies. That is why we are funding, through the FPs,a comprehensive research effort looking at a broad range of energy technologies; from renewables,through clean coal, to nuclear fusion and fission. Many questions are currently being asked in allthese areas and society as a whole is not yet in a position to provide adequate responses. A wellfocussedand effective Community research programme is helping to deliver these urgently neededanswers … Ultimately, the decision whether or not to use nuclear power – just like any other energysource – is a political and societal one taken at the national level. However, this should be a decisionbased on knowledge, not one taken in ignorance. Research can and must supply this knowledge”5. References[1] COM(2006) 105 final 08/03/2006 (available on http://eur-lex.europa.eu/en/prep/index.htm)[2] COM(2007) 1 final 10/01/2007 (available on http://eur-lex.europa.eu/en/prep/index.htm)[3] COM(2006) 844 final 10/01/2007 (available on http://eur-lex.europa.eu/en/prep/index.htm)[4] COM(2006) 847 final 10/01/2007 (available on http://eur-lex.europa.eu/en/prep/index.htm)[5] 7224/07, CONCL 1, 9/3/07[6] “Future Tasks for Future European Energy R&D” Eur22395, ISSN 1018-5593, 2006 (AGE report)& “Transition to a Sustainable Energy System for Europe: The R&D Perspective”, Eur22394,ISSN 1018-5593, 2006 (AGE report)[7] 2006/970/Euratom: Council Decision of 18 December 2006 Concerning the Seventh FrameworkProgramme of the European Atomic Energy Community (Euratom) for nuclear research andtraining activities (2007 to 2011), OJ L 54 p.21 of 22/2/07[8] For more details refer to http://cordis.europa.eu/fp7/euratom/fission-events_en.html

Track 1New Reactor and energy technologies

Session 17.1.1:Advanced reactors

IRIS – AN ADVANCED, GRID-APPROPRIATE PWRFOR NEAR-TERM DEPLOYMENTM.D. CARELLI* 1 , B. PETROVIC *1 , M.E. RICOTTI 2 , C.V. LOMBARDI 2 ,S. MONTI 3 , J.M. COLLADO 4 , N. CAVLINA 5 , G. FORASASSI 6 , F. ORIOLO 6 ,F. MAGGIONI 7 , G. LOCATELLI 8 , G. CATTADORI 91Westinghouse Science and Technology, 1344 Beulah Road, Pittsburgh, PA 15235 – USA2Politecnico di Milano, v.Ponzio 34/3, 20133 Milano – ITALY3ENEA, v. Martiri di Monte Sole 4, 40129 Bologna – ITALY4ENSA, Avda. Juan Carlos I, 8, 39600 Maliaño (Cantabria) – SPAIN5FER, University of Zagreb, Unska 3, 10000 Zagreb – CROATIA6University of Pisa, Via Diotisalvi 2, 56126 Pisa – ITALY7Ansaldo Camozzi, viale Sarca 336, 20126 Milano – ITALY8Ansaldo Nucleare, Corso F.M. Perrone 25, 16161 Genova – ITALY9SIET, v.Nino Bixio 27, 29100 Piacenza – ITALYABSTRACTWith the resurgence of nuclear power there is an increasing need for a range of new reactordesigns, including smaller units of several hundred MWe. Such reactors fit not only thedeveloping and smaller countries or electric grids, but also provide commercial flexibilityto mature markets with large grids by matching the growth, reducing risk, and minimizingfinancing resources. The International Reactor Innovative and Secure (IRIS) offers anadvanced, modular 335 MWe design. IRIS features an integral primary systemconfiguration with all main components located within the reactor vessel. Thisconfiguration enables a simplified design with enhanced reliability and economics andsupports its safety-by-design approach, which results in exceptional safetycharacteristics. In addition to electricity-only production, IRIS is well suited forcogeneration, including water desalination, district heating, and process steam generation.IRIS is being developed by an international team, led by Westinghouse, incorporating 19organizations from 10 countries, about half of them European. IRIS development started in1999 and has reached the level of maturity indicating potential for being commerciallyoffered by the mid of next decade. The preliminary design has been completed and thetesting needed for design certification has started last year. The centrepiece of theexperimental program is the integral system performance testing to be performed at theSIET facility in Italy. The pre-application review process with the US NRC was initiated in2002 to address long-lead items, and enable obtaining the Final Design Approval (FDA) by2013. Economic analyses indicate that IRIS will be competitive with other nuclear andnon-nuclear energy sources, whether deployed gradually in single units in smaller grids, orin multiple twin units for larger grids. Additionally, IRIS fits well the recently announcedUS DOE initiative, GNEP (Global Nuclear Energy Partnership) aiming to supportworldwide expansion of the use of nuclear energy in a responsible and proliferationresistant manner. Within the GNEP framework, IRIS can in the near term offer anadvanced reactor design to satisfy needs for smaller, grid-appropriate reactors.1. IntroductionWith leading indicators predicting the renaissance of nuclear power, there is an ever increasing needfor a range of advanced reactor designs to satisfy diverse needs of worldwide markets. While somerequirements, such as safety, security and economics, are common to all applications, others such asreactor size (power level) are market or application dependent. While for developed, fast growing* Corresponding authors: Email: CarellMD@westinghouse.com, PetrovB@westinghouse.com

markets large reactor units may be preferred, smaller units are also needed, both for smaller/emergingmarkets (due to financial and electric grid limitations), as well as for larger stable markets with limitedgrowth rate, providing better match to needs and improved cash flow. The Westinghouse portfolio ofadvanced power plants offers designs, from the larger AP1000, to the medium and small size IRIS andPBMR, to satisfy the needs of all customers.Many emerging nations and energy markets with small grids will start introducing nuclear powerplants in the next decade. Due to their grid size, units larger than a few hundred MWe are not optimalor in many cases even not technically feasible. With its smaller size (335 MWe), simple design andoperation, exceptional safety, moderate cost, limited financing burden, and possibility to gradually addcapacity by adding more modules, IRIS offers an optimum solution that is technically andeconomically viable and technologically immediately acceptable. This same size fits well developedmarkets having large grids with a limited growth and frequently small margin in transmission lines,that need to improve energy security through redundancy, and optimize investment by “just-in-time”build.2. The IRIS ProjectIRIS[1-5] represents the latest evolution of the LWR technology which has been the overwhelmingmainstay of nuclear power development and deployment. While the integral configuration in general,and the IRIS design in particular, embodies advanced engineering solutions, no new technologydevelopment is necessary and therefore a demonstration prototype is not required to attain designcertification from the regulatory body. A first of a kind (FOAK) commercial plant is thus envisagedpast the mid of the next decade, as shown in the project schedule in Table 1.Program started 1999Assessed key technical and economic feasibility 2000Performed conceptual design, preliminary cost estimate 2001Initiated NRC pre-application licensing for Design Certification 2002Completed NSSS preliminary design 2005Initiated testing necessary for NRC Design Certification 2006Complete testing 2010Start formal Design Approval with NRC 2010Obtain Final Design Approval from NRC 2013Ready for deployment 2015-2017Table 1: IRIS project scheduleThe US Department of Energy (DOE) unveiled a major new initiative in February 2006 [6], the GlobalNuclear Energy Partnership (GNEP). Its ultimate objective is to safely expand nuclear energy withoutincreasing proliferation concerns. One of its key elements is the development of smaller-scale gridappropriatereactors: “These reactors will be safe, simple to operate, more proliferation-resistant andhighly secure. … The GNEP seeks to form international partnerships to accelerate certification ofmarketable designs, and deploy operational demonstration plants…”[6] IRIS satisfies very well theGNEP requirements and has been selected by DOE to exemplify such smaller reactors.From its very beginning, IRIS has been developed by a strong international team comprised of worldrenown organizations and led by Westinghouse.[7] The team currently includes 19 organizations from10 countries, over 4 continents, with a strong European component (about half the organizations andhalf the countries are European). These organizations represent leading nuclear manufacturers,academic institutions, national laboratories and power producers. Universities are vibrant teammembers with more than one hundred students involved to-date, a majority of them having conductedgraduate theses at the master or doctoral level.[8]

3. Innovative Approach and Solutions in the IRIS DesignWhile firmly based on the proven LWR technology, the IRIS project has introduced manyengineering and project innovations which define its unique characteristics, such as:• Design: based on simplicity to simultaneously improve safety, reliability, and economics• Primary system: integrated primary system design• Safety: a safety-by-design approach• Security: enhanced and easier to implement security, based on its design characteristics• Proliferation resistance: enhanced through extended refuelling cycle, while retaining use ofcurrent demonstrated fuel, facilitating international safeguards• Economics: simplicity, modularity, and economy of serialization in lieu of economy of scale• Operation: Simple operation, minimizing need for operators action in incident situations• Construction: Less than 3 years construction period, reduced nuclear in-country infrastructurerequired• Workers safety: significantly reduced dose to personnel in operation, maintenance, andultimately in decommissioning activities• Project management: development by an integrated international team (led by Westinghouse)of 19 organizations from 10 countries, with all team members contributing resources to IRISdevelopment• Research: effectively incorporating national laboratories and academia in the developmentefforts• Market segment: targeting markets and utilities that require a smaller-scale reactor design, dueto grid size or financial limitations• Market penetration approach: reaching more markets through the international team and widepartnership in team member countries• Licensing: based on its outstanding safety, aiming at achieving licensing with lessened and, ifpossible, eliminated off-site emergency planning requirements. Licensing supported through amultinational design evaluation program (MDEP) will be pursued.IRIS is innovative in design – employing an integrated primary system that incorporates all the mainprimary circuit components within a single vessel, i.e., the core with control rods and their drivemechanisms, eight helical coil steam generators with eight associated fully-immersed axial flowpumps, and a pressurizer (Fig. 1).The integral configuration offers intrinsic design improvements:• Pressurizer: A dedicated pressurizer is eliminated, as the vesselhead will fulfil the function. Much larger volume/power ratiogives much better control of pressure transients. Additionally,no sprays are required.• Primary coolant pumps: The axial fully immersed pumps resultin no seal leak concerns, no possibility for shaft breaks, and norequired maintenance.• Internal CRDMs: This solution eliminates head penetrations andpossibility of seal failures, as well as any future headreplacements.• Steam generators: With the primary coolant outside, tubes are incompression, thus eliminating tensile stress corrosion cracking.• Thick downcomer: The 1.7m thick downcomer reduces the fastneutron flux on the reactor vessel by 5 orders of magnitude. Thisleads to “cold” (i.e., not activated) vessel, almost no outsidedose, no vessel embrittlement, and no need for surveillance. Thevessel is essentially “eternal”, and decommissioning issimplified.Fig. 1: Integral configuration and components

• Compact layout: While leading to a larger reactor vessel, the integral layout results in a smallercontainment and overall a more compact site, with positive impact on safety, security, andeconomics (Fig. 2).• Maintenance: Intervals between maintenance outage can be extended to 48 months. A core designhas been developed enabling uninterrupted operation for up to 4 years if so desired.XXXXXXXXXFig. 2: Integral configuration and compact containment layoutIn addition to the design improvements, the integral configuration offers very significant intrinsicsafety advantages, which have led to the unique IRIS safety approach articulated over three tiers.The first tier is safety-by-design which aims at eliminating by design the possibility for an accidentto occur rather than dealing with its consequences. By eliminating some accidents, the correspondingsafety systems (passive or active) become unnecessary as well.The second tier is provided by simplified passive safety systems, which protect against the stillremaining potential accidents and mitigate their consequences.The third tier is provided by active systems, which are not required to perform safety functions (i.e.,are not safety grade) and are not considered in deterministic safety analyses, but may contribute toreducing the core damage frequency (CDF).Table 2 summarizes the IRIS design characteristics and their safety implications, together with theirimpact on accidents, with particular emphasis on condition IV events. Systematic implementation ofthe IRIS safety-by-design approach has enabled outright elimination of 3 out of 8 Design BasisEvents (DBEs) typically considered for LWRs. Severity has been reduced for another 4, while onlyone DBE (fuel handling accident) remains the same.Furthermore, by consistently applying the safety-by-design approach (guided by use of ProbabilisticRisk Assessment from the very beginning of the design process), IRIS has lowered the predicted CoreDamage Frequency (CDF) to below 10 -7 /yr and Large Early Release Frequency (LERF) to below 10 -9 /yr. While the present nuclear power plants already demonstrate remarkable safety, further safetyadvances achieved in IRIS may enable plant licensing with a reduced or even eliminated off-siteemergency planning zone[9]. This feature not only should increase public acceptance but will producea positive financial impact by reducing infrastructure cost, as well as enabling efficient co-generationfor desalination, district heating and process heat.To enable deployment in the next decade, consistent with the projected worldwide energy needsgrowth, the reference IRIS core design is based on the current, available and demonstrated LWR fueltechnology. However, the design includes features to enable future improvements in fuel managementand further enhance most of its proliferation resistance. This will be achieved by gradually increasingfuel discharge burnup and cycle length, requiring that in parallel improved fuel performance isdemonstrated.To further simplify safeguards and make them more effective (as well as to improve economy) IRISextends the fuel reloading interval. It is anchored to the IRIS optimized maintenance with outagerequired only every 48 months, therefore directly enabling refuelling interval of up to four years. The

eference IRIS design with 4.95% UO 2 fuel presently enables a 3 to 4 years cycle.[10] In the future,employing UO 2 or MOX fuel with ~10% fissile content (still comfortably below the HEU limit), aneight-year refuelling cycle with a short maintenance outage halfway through will be feasible.[11] Witha four- or eight-year refuelling cycle, and the possibility to limit spent fuel kept at site to one coreload,safeguards will be even more simple and effective, and any diversion timely identified.IRIS DesignCharacteristicSafety ImplicationAccidents AffectedCondition IV DesignBasis EventsEffect on ConditionIV Event by IRISSafety-by-DesignIntegral layout • No large primary piping • Large break LOCAs Large break LOCA Eliminated• Increased water inventory• Other LOCAs• Increased natural circulation • Decrease in heat removalvarious eventsLarge, tall vessel• Accommodates internal Control RodDrive Mechanisms• Control Rod ejection• Head penetrations failureSpectrum of ControlRod ejection accidentsEliminatedHeat removal frominside the vessel• Depressurizes primary system bycondensation and not by loss ofmass• Effective heat removal by SteamGenerator and Emergency HeatRemoval system• Other LOCAs• Other LOCAs• All events requiringeffective cooldown• Anticipated TransientWithout Scram (ATWS)Reduced size, higherdesign pressurecontainmentMultiple, integral,shaftless coolantpumpsHigh design-pressuresteam generatorsystemOnce through steamgeneratorsIntegral pressurizer• Reduced driving force throughprimary opening• No shaft• Decreased importance of singlepump failure• No Steam Generator safety valves• Primary system cannot overpressuresecondary system• Feed/Steam System Pipingdesigned for full Reactor CoolantSystem pressure reduces pipingfailure probability• Limited water inventory• Large pressurizer volume/reactorpower• Other LOCAs• Shaft seizure/break Reactor coolant pumpshaft break• Locked rotor Reactor coolant pumpseizure• Steam generator tuberupture• Steam line break• Feed line break• Feed line break• Steam line break• Overheating events,including feed line break• ATWSSteam generator tuberuptureSteam system pipingfailureFeedwater systempipe breakFuel handlingaccidentsEliminatedDowngradedDowngradedDowngradedDowngradedUnaffectedTable 2: Implementation of safety-by-design in IRISIRIS has been designed to satisfy all the current licensing requirements with the U.S. NRC. However,an additional option for IRIS licensing is being pursued through the NRC’s recent multinationaldesign evaluation program (MDEP), which would facilitate its worldwide deployment. According toRef. [12], “…NRC has formally approved moving forward with implementation of [MDEP] aimed atimproving the effectiveness and efficiency of regulatory design reviews for new reactors”. MDEP isenvisioned in three stages, with increased level and formalization of international cooperation inlicensing. One of the objectives of Stage 1 is to identify areas where national standards overlap withthe U.S. regulations and where foreign regulatory expertise could complement the expertise of theNRC’s staff. This provides an opportunity to regulatory bodies of countries potentially interested inIRIS to join the IRIS multinational licensing efforts, become familiar with relevant characteristics ofthe IRIS design while strengthening their expertise in licensing. This route has been already taken bythe Croatian regulatory agency, which in December 2006 has requested MDEP participation in theIRIS review, a request accepted by NRC.Need for potable water is even more critical in many developing nations than the need forenergy/electricity. Moreover, the co-generation market segment has specific needs as compared toelectricity-only generation. Transportation of co-generation produced water, process heat/steam,

district heating over long distances is not practical, and in many case several smaller, geographicallydistributed plants are preferable to a large single plant.With its moderate power level, simple design, and a possibility to attain licensing with a reducedEmergency Planning Zone (EPZ), and thus locating plants closer to end users, IRIS is well suited forco-generation. A preliminary design of an IRIS desalination co-generation plant[13] has beenperformed by the IRIS team member, OKBM, which has a vast experience with desalination units.Several IRIS team members (including from Brazil and Mexico) have performed economic studiesdemonstrating attractiveness of IRIS to fulfil the combined electricity and potable water needs in aridregions of their countries.[14-16] Lithuania has examined use of IRIS in district heating. Applicationof IRIS to ethanol production in the U.S. is also being considered.4. Some Current EffortsIRIS economic competitiveness in all markets is achieved through a synergistic positive effect ofseveral technical and economics factors that add up to counterbalance the negative impact of theeconomy of scale. To mention just one of the positive factors, IRIS enables a gradual increase ingenerating capacity to match growth needs. Financial risk and needed investment capital are thuslargely reduced since the staggered construction of modules deployed several years apart enablesincome to be generated from previous unit(s) while the next unit is being built. Details of theeconomics analysis are presented in a companion paper.[17]Preliminary economic analyses presented previously [18] have recently been extended to include arelatively large contingency, added up front to the estimated cost to address uncertainties and anyunforeseen factors. Additionally, the whole first core cost has been accounted as capital cost. Whilethis approach may be considered overly conservative, it does provide a very robust economic case. Inspite of this conservativism, the estimated total cost of electricity is about 4-5 ¢/kWh, competitive withother nuclear and non-nuclear sources.IRIS is currently in the pre-application review process with the U.S. Nuclear Regulatory Commission.This pre-application phase is intended to address long-lead items (such as testing) before the full-scaleformal design certification process is started, thus allowing the latter to be completed expeditiously.Additionally, in its licensing IRIS will take maximum advantage of the successfully completed DesignCertification of Westinghouse’s AP600 and AP1000 for all those design features, analyses, andsupporting tests which are similar in the three designs. However, further testing is necessary to addressthe new IRIS design features and components, including:• Integral Reactor Coolant System• Passive Safety Features specific to IRIS• Reactor Vessel and Containment InteractionThe tests have been divided into three types according to their scope and primary purpose:• Basic Engineering Development Tests.• Component Separate Effects Tests.• Integral Effects Tests.The testing program started in 2006. A detailed design of testing facilities is underway. A large part ofthe safety related tests, and in particular the integral system test will be conducted in Italy, at SIET, atthe same facilities where testing of the passive systems for AP600 was conducted in the 1990’s.Further details are provided in the companion paper [19].The safety-by-design philosophy in IRIS has lead to significant enhancement of its safetyperformance, as demonstrated in the reduction of the estimated core damage frequency (CDF) due tointernal events to below 10 -7 events per reactor-year. Currently, efforts are under way to implement thesame approach to external events, including seismic. The compact integral design facilitates this task.The reactor building is cylindrical in shape, of moderate diameter, only about 30 meters above theground, with the spherical containment fully contained within. This greatly increases security,providing robustness and resilience to external malevolent acts. Additionally, it improves the seismicresponse of the building, and if necessary it makes feasible the use of seismic isolators in regions withstrong seismic activity. More details are provided in a companion paper [20].

5. ConclusionsIRIS is an advanced integral PWR of medium power (335 MWe), developed by an international team,led by Westinghouse, and with a strong participation of European organizations. It is well suited tosatisfy the needs of both the smaller/emerging markets, as well as larger developed markets withlimited growth rate or desire to optimize cash flow. Combined with its foreseen role within the U.S.DOE GNEP program, IRIS offers potential for a worldwide deployment. A major testing program wasinitiated at SIET, Italy, in 2006, to support submittal of the FDA/DC application in 2010, thusenabling deployment mid next decade.References[1] M.D. Carelli, “IRIS: A Global Approach to Nuclear Power Renaissance,” Nuclear News, 46,No. 10, pp. 32-42 (Sep. 2003).[2] B. Petrović, M.D. Carelli, “IRIS Project Update: Status of the Design and Licensing Activities,”Proc. 5 th Intl. Conf. on Nuclear Option in Countries with Small and Medium Electricity Grids,Dubrovnik, Croatia, May 16-20, 2004, HND (2004).[3] M.D. Carelli, B. Petrović, N. Čavlina, D. Grgić “IRIS (International Reactor Innovative andSecure) – Design Overview and Deployment Prospects,” Proc. Intl. Conf. Nuclear Energy forNew Europe 2005 (NENE 2005), Bled, Slovenia, September 5-8, 2005 (2005).[4] M.D. Carelli, B. Petrović, “Here’s Looking at IRIS,” Nuclear Engineering International, Vol51. No 620, pp. 12-18 (March 2006).[5] M.D. Carelli et al., “The Design and Safety Features of the IRIS Reactor,” Nucl. Eng. Design,230, pp. 151-167 (2004).[6] Website http://www.gnep.energy.gov .[7] M.D. Carelli et al., ”The IRIS Consortium: International Cooperation in Advanced ReactorDevelopment,” Proc. 13 th International Conference on Nuclear Engineering (ICONE-13),Beijing, China, May 16-20, 2005, Paper ICONE13-50799 (2005).[8] M.D. Carelli, B. Petrović, M. Ricotti, N. Todreas, N. Čavlina, F. Oriolo, H. Ninokata, ”Role ofUniversities in the Development of IRIS,” Proc. International Youth Nuclear Congress (IYNC2006), Stockholm, Sweden, June 18-23, 2006, Paper 137 (2006).[9] M.D. Carelli, B. Petrović, P. Ferroni, ”IRIS Safety-by-Design and Its Implication to LessenEmergency Planning Requirements,” to be published in Intl. Journal of Risk Assessment andManagement (in print,2007).[10] B. Petrović, F. Franceschini, “Fuel Management Approach in IRIS Reactor”, Proceedings LAS-ANS Intl. Conf., Cancun, Mexico, July 11-16, 2004 (2004).[11] B. Petrović, F. Franceschini, “Fuel Management Options in IRIS Reactor for the ExtendedCycle”, Proceedings of the 5 th Intl. Conf. on Nuclear Option in Countries with Small andMedium Electricity Grids, Dubrovnik, Croatia, May 16-20, 2004 (2004).[12] J.W. Williams, “A Multinational Design Approval Program”, Proceedings of ICAPP ’06, Reno,NV, June 4-8, 2006 (2006).[13] V.I.Kostin, Yu.K. Panov, V.I.Polunichev, S.A.Fateev, L.V.Gureeva, “Nuclear PowerDesalinating Complex with IRIS Reactor Plant and Russian Distillation Desalinating Unit,”Proc 5 th Intl. Conf. on Nuclear Option in Countries with Small and Medium Electricity Grids,Dubrovnik, Croatia, May 16-20, 2004 (2004)[14] D. T. Ingersoll, J. L. Binder, D. Conti, M. E. Ricotti, “Nuclear Desalination Options for theInternational Reactor Innovative and Secure (IRIS) Design,” Proc 5 th Intl. Conf. on NuclearOption in Countries with Small and Medium Electricity Grids, Dubrovnik, Croatia, May 16-20,2004 (2004)[15] B.D. Baptista Filho, M. Cegalla, R.N. Raduan, A.C.O. Barroso, L. Molnary, F.R.A. Lima,C.A.B.O. Lira, R.C.F. Lima, “Social, Economic and Environmental Assessment of Energyand Water Desalination Options for the Brazilian Polygon of Drought Using the IRISReactor”, Proc 5 th Intl. Conf. on Nuclear Option in Countries with Small and MediumElectricity Grids, Dubrovnik, Croatia, May 16-20, 2004 (2004)

[16] Gustavo Alonso, Ramon Ramirez, Carmen Gomez, Jorge Viais, “IRIS Reactor: A SuitableOption to Provide Energy and Water Desalination for the Mexican Northwest Region”, Proc.LAS-ANS Intl. Conf., Cancun, Mexico, July 11-16, 2004. (2004)[17] M.D. Carelli, B. Petrovic, C. Mycoff, G. Locatelli, M. Mancini, M.E. Ricotti, P. Trucco, S.Monti, K. Miller, ”Economic Features of Smaller Size, Integral Reactors“, European NuclearConference (ENC 2007), Brussels, Belgium, September 16-19, 2007. (2007)[18] K. Miller, IRIS – Economics Review, Proc. 13 th International Conference on NuclearEngineering (ICONE-13), Beijing, China, May 16-20, 2005 (2005).[19] M.D. Carelli, B. Petrovic, M. Dzodzo, L. Oriani, L. Conway, G. Cattadori, A. Achilli, R. Ferri,F. Bianchi, S. Monti, F. Berra, M.E. Ricotti, L. Santini, D. Grgic, G.L. Yoder, “SPES-3Experimental Facility Design for IRIS Reactor Integral Testing”, European Nuclear Conference(ENC 2007), Brussels, Belgium, September 16-19, 2007. (2007)[20] S. De Grandis, G. Benamati, G. Bianchi, D. Mantegazza, F. Perotti, L. Corradi Dell’Acqua, S.Monti, “A New Approach for the Seismic Analysis and Design of the IRIS Reactor“, EuropeanNuclear Conference (ENC 2007), Brussels, Belgium, September 16-19, 2007. (2007)

ACR-1000: Product UpdateS.K.W. YU, J.M. HOPWOOD, G. LEACH, M., I.J. HASTINGS, K.BRADLEYAtomic Energy of Canada Limited2251 Speakman Drive, Mississauga, Ontario, Canada L5K 1B21. IntroductionAtomic Energy of Canada Limited (AECL) has adapted the successful features of CANDU ® *reactors to design Generation III+ Advanced CANDU Reactor ® ** (ACR ® **) technology [1-3].The ACR-1000 ® ** nuclear power plant is an evolutionary product, based on proven, traditionalCANDU reactor technology, coupled with thoroughly demonstrated innovative features to enhanceeconomics, safety, operability and maintainability. This evolutionary strategy ensures that AECL’sinnovations are based on proven experience, and focuses the development programs on a selectnumber of innovative features. The ACR-1000 basic design is complete, ready and regulatoryreview of formal licensing document is underway. Detailed pre-project design has started. TheACR program covers all activities required to achieve a first unit in-service date, and the program isbeing executed using full-scale project management principles.The ACR-1000 has been chosen for generic design assessment in the UK. Additionally, there areactive ACR-1000 new build initiatives in Canada: Ontario, Alberta and New Brunswick.2. ACR-1000 Product DescriptionThe standard ACR-1000 design is a 1200 MWe class nuclear power plant, which has evolved fromAECL’s existing successful product lines. The ACR-1000 applies the advanced CANDUtechnology developed in the ACR program. All innovative features of the ACR-1000 will be fullytested and proven before the first project. The design also makes extensive use of successfulfeatures of existing CANDU technology. By doing this, the ACR-1000 can be developed andapplied in initial projects with a high degree of confidence. Additionally, it fully exploits theconstruction techniques that contributed to the impressive schedule accomplishments at QinshanPhase III.The ACR-1000 has the following major features:• Twin-unit configuration with common control room building (can be delivered in single unitconfiguration)• Compact, horizontal pressure tube core design following traditional CANDU overallconfiguration• Enhanced inherent and passive safety with Moderator and Shield Tank heat sinks suppliedby passive water makeup from Reserve Water Tank.• Core consists of low-pressure, low-temperature calandria tank containing heavy-watermoderator, within which fuel channels are located, each containing 12 standard-length,enriched, CANFLEX-ACR fuel bundles.• Coolant is light water.• Fuel channel consists of a Zirconium alloy pressure tube, surrounded by a Zircaloy calandriatube, and attached to coolant system feeder piping by individual end fittings.• On-line core refueling is carried out via two computer-controlled fuelling machines• Reactivity control and shutdown mechanisms are located in the low-pressure calandria tankwith no possibility of accidental high-pressure ejection.• Four-quadrant layout with four-way redundancy of safety support systems,• Indirect thermal cycle (similar to PWR reactors), with the reactor coolant systemtransferring the heat from nuclear fission, through vertical shell-and-tube steam generators,to a conventional secondary turbine cycle.* CANDU ® is a registered trademark of Atomic Energy of Canada Limited (AECL).

** Advanced CANDU Reactor ® , ACR ® and ACR-1000 ® are trademarks of AECL.SMART CANDU ®*** is a registered trademark of AECL.• Customer-driven improved features for operability and ease of maintenance.While retaining proven CANDU features, innovations in the ACR-1000 design include:• Use of light-water coolant in the CANDU coolant system, in conjunction with thecontinuing use of heavy water moderator in the calandria• Design of a more compact core configuration to enable optimized reactor physicscharacteristics• Use of low enriched fuel with higher burnup than the Natural Uranium (NU) fuel used intraditional CANDU reactors.• Increased coolant system and turbine pressure to increase the overall thermal efficiency ofthe power plant.The plant layout is designed to achieve the shortest practical construction schedule. This isachieved by simplifying the design, minimizing and localizing interfaces, parallel fabrication ofmodule assemblies and civil construction, reducing construction congestion, improving access,providing flexible equipment installation sequences, and reducing material handling requirements.Security and physical protection have also been taken into consideration in the development of theplant layout. Physical protection is provided through ample separation. The reactor buildingconsists of a steel-lined, prestressed concrete containment structure and a reinforced concreteinternal structure supported on a reinforced concrete base slab. The containment structure providesan environmental boundary, biological shielding, and a pressure boundary in the event of anaccident. The building layout is arranged to provide separation by distance, elevation or barrier forsafety related structures, systems and components. These features reduce the likelihood of commonmode failure of safety systems due to malevolent acts such as an aircraft crash.The ACR-1000 reactor core has the following characteristics:• Compact size combined with on-power refuelling.• Reduced heavy water requirements due to compact core size (lattice pitch of 240 mm versus286 mm in current CANDU units) and the use of light water as the coolant.• Moderate negative coolant-void reactivity.• Simplified reactor control through negative feedback in reactor power.A high form factor of 0.94 is achieved, along with increased core stability.3. ACR-1000 Safety FeaturesThe ACR-1000 design takes advantage of inherent and engineered safety characteristics, includingdistinctive features that arise from CANDU design principles. The core is designed for smallmagnitudenegative reactivity coefficients, which provide inherent protection against transients withinadvertent increase of reactor power. Additionally, two diverse and fully capable, fast-acting,independent shutdown systems are provided. Each system can shut down the reactor for the entirespectrum of design basis and anticipated events. Also, the separate control system shuts down thereactor for Anticipated Operational Occurrences.Further defences - in-depth is derived from the inherent passive-safety design features of theCANDU fuel channel core [4]. The moderator heavy water surrounding the fuel channels water inthe calandria is itself an additional, diverse active/ passive heat sink. The calandria is filled withheavy water to a level well above the top of the calandria shell. This heavy water acts as both amoderator and reflector for the reactor, as well as an assured heat removal option. The moderatorsystem is a low-pressure and low-temperature system that is fully independent of the heat transportsystem. Moderator heat exchangers remove the heat generated in the moderator during reactor2

operation and shutdown. Passive make up to the moderator is provided, and long term coolingassured by a Reserve Water TankCore retention within the vessel includes both retention within fuel channels, and retention withinthe calandria vessel. The moderator heavy water in the ACR-1000 calandria vessel, as in any otherCANDU-type reactor, provides ample heat removal capacity in severe accidents. The ACR-1000calandria vessel design permits for passive rejection of decay heat from the moderator to the shieldwater. Also, the calandria vessel will be designed for debris retention. Core damage termination isachieved by flooding of the core components with water and keeping them flooded thereafter.Successful termination can be achieved in the fuel channels, calandria vessel or calandria vault bywater supply by the Long Term Cooling pumps and by gravity feed from the Reserve WaterSystem.The ACR-1000 containment is required to withstand external events such as earthquakes, tornados,floods and aircraft crashes. Containment integrity maintenance is achieved through control ofcontainment pressure, flammable gas control, and control/prevention of the core-concreteinteraction. The containment system includes the steel-lined, pre-stressed concrete reactor buildingcontainment structure, access airlocks, building air coolers for pressure reduction, and acontainment isolation system, consisting of valves in certain process lines and ventilation ducts thatpenetrate the containment structure.4. ACR-1000 Operability and MaintainabilityThe design basis lifetime capacity factor for ACR-1000 is 90% over the operating life of 60 years.The design basis year-to-year design capacity factor is 93%. The engineering components ofindividual systems and components uses plant-wide models such as PSA models, and operationsexpert feedback to exceed these targets. Customer feedback has resulted in many detailedoperational and maintenance improvements being incorporated into the design to meet theperformance targets. Additionally, use of CANDU operating experience facilitated by theinformation network provided by the CANDU Owners Group (COG) will further improve theperformance of the plant.On-power maintenance and testing are optimized, reducing the frequency of outages to once everythree years, with planned maintenance outages not exceeding 25 days. The unplanned outage ratetarget, is less than seven days/year through design optimization, and through the use of System-Based Maintenance strategy. The plant layout is optimized to facilitate online maintenance andinspection, to provide access for equipment exchange, and to provide effective common services forthe two-unit plant design resulting in reduced maintenance costs. For Plant Life Managementpurposes, there is provision of space and services to support a rapid, mid-life full-scale fuel channeland steam generators replacement program. Maintenance activities are enhanced to maximizecomponent life and minimize component replacement time, thereby minimizing radiation exposure,replacement costs, and the number of operating and maintenance personnel required.Application of existing CANDU computer control knowledge and experience, enhanced by state-ofthe-artinformation system technology, has produced advanced plant control and monitoringsystems that enable the plant to operate at higher capacity factors with a reduced operations staff.SMART CANDU ®*** modules provide on-line health monitoring and diagnostics for plantchemistry, predict future performance of components, determine maintenance requirements andoptimal operating conditions and ensure optimal margins and maximum power output.5. ACR-1000 Design Status5.1 GeneralThe CANDU 6, CANDU 9 and ACR-700 programs produced the foundation for the ACR-1000.AECL selected the ACR-1000 as its new reference design to meet market requirements. The ACR-1000 program focus is to plan and execute work based on risk analysis, assessment and mitigation,to ensure licensability and address customer input, and to achieve an in-service date of 2016. Theprogram plan is project based, using a comprehensive 10,000-activity schedule and is intended toensure that all required documentation is available to support the Environmental Assessment and3

Site Preparation and Construction License applications. The program is designed to have all designdocumentation completed prior to start of construction.The ACR conceptual design has been completed and the ACR-1000 is now a project under themanagement of AECL’s Commercial Operations group. A revised quality assurance manualcovering the ACR-1000 and Enhanced CANDU 6 products has been issued. The framework for theoverall project execution plan has been developed and identifies the key project execution elements.The technology issues have been successfully resolved and the licensing basis has been establishedand all elements of the basic engineering program are in progress. Project risk managementprocesses and procedures have been put in place. The 478 work packages required for thePreliminary Safety Case Package (PSCP) submission (the reference submission for use in genericpre-project reviews by regulators) and for input to the generic Preliminary Safety Analysis Report(PSAR) have been developed to support scope, cost and schedule management requirements. TheLevel-3 production schedule—covering the basic engineering program together with the remainingR&D work and licensing activities—has been issued. Approximately 400 full-time-equivalent staffwork on the project on a day-to-day basis.5.2 ACR-1000 Design EvolutionThe design of the ACR-1000’s systems, structures and components is based on the successfulCANDU 6 and Darlington nuclear steam plants (NSPs). Minimal manufacturing and supplychanges are anticipated due to the similarities of major NSP equipment and components for theACR-1000 and CANDU 6. Major equipment and components have been proven through manyyears of continuous operation of 10 CANDU 6 plants. A proven licensing and safety basis builds on40 years of CANDU licensing experience in Canada and around the world. The Balance of Plant(BOP), comprising 40% of total plant equipment, is a scale-up of the proven CANDU 6 BOP.A number of innovations were accepted by the Canadian Nuclear Safety Commission (CNSC) inthe pre-project licensability review of the CANDU 9 in 1997-98, and have been adopted in ACR-1000:• large, steel-lined containment• improved circulation in the moderator• reserve water tank for accident coolant make-up• high-pressure safety feedwater system• distributed control system/plant display system and modern control centre incorporatinghuman factors considerationsThe ACR design includes the technologies enabling compact reactor core with light water coolantand low-enriched uranium fuel developed and reviewed by regulators at earlier stages in the ACRprogram. Other ACR innovations include:• thicker pressure tubes and thicker and larger calandria tubes• mechanical zone control replacing liquid zone controllers and adjuster rods• stainless steel feeders and headers• long-term cooling system to perform long-term emergency core cooling (ECC) andmaintenance coolingAdditional design enhancements were incorporated into the ACR-1000 design to mitigatetechnology issues that represent perceived risks based on project risk management evaluations.Other changes were made to meet new Canadian regulatory guidelines and regulations including theDesign Requirements Documents (DRD) for new plants. Other design changes were made toimprove operational performance based on customer feedback:• simplified CANFLEX-ACR fuel bundle design:o 42 similarly-sized elements with 2.4% enrichmento larger centre element with burnable neutron absorber but no uraniumo uniform enrichment4

• simplified and more reliable emergency coolant injection• new system classification based on safety function categories• four-train emergency feedwater system as emergency heat removal system• additional reactor trip to meet no-dryout requirement for end-of-life conditions5.3 Design ReadinessACR-1000 design and analysis work is well underway and will meet the owner’s needs for EA, siteand construction licenses, towards a first in-service date of 2016. The next milestone—completionof the Preliminary Safety Case Package by 2008 May—will represent a standard licensingpackage appropriate for stand-alone regulatory review during any individual pre-projectphase and analysis completion for ACR-1000. The “design freeze” in 2007 March was a key stepin integrating the design and getting ready for formal safety evaluation of the ACR-1000 product.The reference plant design documents from CANDU 6, CANDU 9 and ACR-700 documents willbe updated for use in ACR-1000, as part of the project-engineering phase—with the objective ofcompleting all design documentation prior to the start of construction.6. SUMMARYThe ACR-1000 uses well-established, fundamental, CANDU design elements: core design withhorizontal pressure tubes; simple efficient fuel bundle design; on-power refuelling and a separatelow-pressure, low-temperature heavy-water moderator providing an inherent emergency heat sink.It includes adaptations for light-water coolant and low-enriched uranium fuel, and offers a compactcore configuration and higher steam pressure for greater thermodynamic efficiency. The ACR-1000links design with licensing, emphasizing operability and maintainability from the viewpoint of thecustomer—the utility operator.7. REFERENCES[1] Torgerson D.F. and Hancox W.T., “CANDU: Flexibility for Future Development”,International Conference on Engineering and Technological Sciences 2000, Beijing, 2000October 11-13.[2] Torgerson D.F. and Duffey R.B., “The Next Generation CANDU Reactor”, AmericasNuclear Energy Symposium, Miami, 2000 December 11-13.[3] Torgerson D.F., Hedges K.R. and Duffey R.B., “The Evolutionary CANDU Reactor—Past,Present and Future, Physics in Canada, Vol. 60, Number 6, 2004, p 341.[4] B. Lekakh, K. Hau and S. Ford, “ACR-1000 Passive Features”, Proc. ICONE 14, Miami,Florida, 2006 July 17-20.5

WHY THE WESTINGHOUSE ADVANCED, PASSIVEPRESSURIZED WATER REACTOR, AP1000?DAN LIPMANSenior Vice President, Nuclear Power PlantsWestinghouse Electric Company, LLCP.O. Box 355, Pittsburgh, PA 15230 USAABSTRACTWhat does the AP1000 do that is an improvement over the earlier models? For one thing, itresponds to the desire expressed by nuclear utilities for a simpler plant. It also responds toutility needs for a nuclear plant that can compete more favorably on capital investment withfossil power plants, and is even safer than current models. AP1000 meets these goals bypreserving the essentials of the proven, robust, and reliable virtues of the power generatingfeatures of earlier Westinghouse plants while incorporating simpler but highly reliablepassive cooling safety systems for the core and the containment. Combined with the use ofPRA to guide the design, the AP1000 has a Core Damage Frequency (CDF) of 5.1x10-7, ascertified by the US NRC. Compare this to currently operating plants with their active(pump-driven) safety systems that have a CDF typically at 5x10-5.To address capital cost competition, AP1000 has a highly developed construction plan tominimize the time and cost of construction. It is designed from the outset for modular and“open top” construction techniques. The whole process of construction and constructionplanning is further abetted by the lower appetite of AP1000 for construction commoditiesafforded by the passive design’s more compact dimensions and more concentrated areasrequiring less Seismic Category 1 construction. With less equipment required by thedesign, AP1000 represents a focused effort towards minimizing the traditionally high costof nuclear plant construction.Where do things stand today? Under its new licensing approach, the US NRC has reviewedand certified the AP1000 design. That makes it a licensed plant design that can bereferenced in combined Construction Operating License (COL) applications. TheAP1000’s debut has been received favorably. AP1000 has so far been identified by five USutilities for ten units as the plant design in such applications. It has also been selected forfour units to be constructed in China along with technology transfer to support additionalAP1000s to be built there under license. And on May 15 the European UtilityRequirements (EUR) organization certified that the AP1000 pressurized water reactor hassuccessfully passed all the steps of analysis for compliance with European UtilityRequirements, confirming that the AP1000 can be successfully deployed in Europe.1. IntroductionIt is worth looking back for a moment to examine the circumstances extant at the time thatdevelopment of AP1000 got underway, and how a product with such a long development timeemerges seemingly at the moment that interest in the nuclear option has taken hold world-wide. Whatwe can say is that design work began at Westinghouse on the AP600, predecessor to the AP1000, in1989. This was hardly an auspicious time to begin working on the next generation of nuclear plants.The market for nuclear power plants had for the most part retreated to Asia, where the need for newnuclear plants was still strong and supported by the electricity demands of vigorously growingeconomies. Indeed, I spent many years of my career during the 1990s with Westinghouse in theRepublic of Korea. Despite the inauspicious circumstances outside of Asia at that time, there were,nevertheless, other significant developments occurring that would give us greater confidence inproceeding with AP1000.

First, the US NRC developed, and in 1989 enacted, a new plant licensing process, Title 10 CFR (Codeof Federal Regulations) Part 52. The new process sprang from the long history of licensing all 104plants now operating in the US under 10 CFR Part 50. Part 52, would essentially re-shuffle thelicensing sequence by front-loading the approval of the site and the plant design prior to issuing alicense to build and operate. This has the effect of settling design and site issues prior to makingmajor investments. It is quite unlikely that any US utility would be considering a nuclear plant today ifthe Part 50 regulations still prevailed.In this same period, U.S. utilities, showing foresight, joined together to develop -- in cooperation withthe U.S. Department of Energy, the U.S. NRC, and nuclear power plant designers -- a set of designrequirements for the next generation of nuclear power plants. The resulting multi-volume document,the “EPRI Utility Requirements Document (URD)”, is a design specification for new nuclear powerplants, incorporating the lessons learned in construction, licensing, operation, and maintenance of theexisting fleet of operating nuclear power plants. A similar activity commenced in Europe producingthe European Utility Requirements (EUR). AP1000, and its predecessor, AP600, were developedvirtually in parallel with the URD. It was a valuable road map. Consequently, AP1000 embodies theURD design specifications for an advanced, passive plant design, and it has been certified by the EURgroup confirming its suitability to being deployed in Europe.And so we find ourselves today with a very different set of boundary conditions from 1989.2. What about the design?Over the past decades, what we have seen is steady improvement in nuclear plant performance.Capacity factors are now commonly greater than 90%. Electricity production costs of nuclear plantsare typically comparable to coal-fired plants and often are the cheapest power on the grid next tohydroelectric power. All of this has been accomplished with exemplary levels of safety.In essence, utilities have been perfecting their mastery of plant operation. Now they have becomevirtuosi. They also know what to look for in a new instrument. So it falls to plant designers to providewhat utilities need for the next generation of plants.Both the URD and the EUR have the general approach of preserving the virtues of the operating plantswhen it comes to the power producing system - the primary systems in particular - which have provedthemselves so well. But there is also a requirement for a simpler plant this time, and one that is saferand costs less to construct. Both EUR and URD anticipate and address specifically the advantages ofa passive plant for both cheaper construction and to reduce the reliance on operator action in case of anaccident. In fact, the expectation for a passive plant is to achieve and maintain safe shutdown in caseof an accident for 72 hours without operator action. This is substantially different than the 30 minuteperiod for operator action specified for an “evolutionary,” active system plant.The URD also expects that for the new generation a complete plant design will be offered to utilities,encompassing the entire plant up to its connection to the grid.3. How does AP1000 go about meeting these requirements?Retain the virtuesAP1000 is an 1117 MWe plant. The power producing primary system is a familiar one based onproven and reliable Westinghouse PWR features, but with evolutionary improvements to be expectedwith the benefit of decades of operating experience, development of improved materials and bettermanufacturing techniques. Replacing Alloy 600 steam generator tubing with Alloy 690 tubing and theuse of low cobalt-content alloys to reduce activation are some examples. This, of course, is a directoutgrowth of the steam generator replacements on the operating plants. The AP1000 reactor vessel isring-forged, eliminating longitudinal welds. And there are no circumferential welds in the high flux

core region. These features combined with improved materials allow for a 60 year vessel life. The fueldesign is closely based on the XL Robust Fuel Assembly design that has been operating in Doel,Tihange, and South Texas.One of the improvements found in the AP1000 primary system design is the use of sealless, reactorcoolant pumps. By eliminating the need for the complex shaft seal, a source of potential primarysystem leakage is eliminated. The sealless RCP requires no oil lubrication system, and is designed tobe maintenance-free. In fact, sealless motor pumps were used in the first generation of WestinghousePWRs but, as the plants became larger with the second generation designs, they out-grew the capacityof that type of pump available at the time. Since then sealless pump sizes have increased, enablingtheir application to power reactors once again.Gain the Advantages of Passive Safety SystemsHere is where we take a different path with AP1000, one that re-casts the safety-related coolingsystems in light of many decades of experience with the old systems. AP1000 features passive safetysystems for emergency core and containment cooling. It is the essential means of simplifying the PWRwhile at the same time increasing the level of safety. In designing the systems, we had the benefit ofusing highly developed Probabilistic Risk Assessment methods. This new approach of using PRA asan integral element of the design had the effect of driving the Core Damage Frequency (CDF) tounprecedented low probability levels: 5.1x10- 7 for power and shutdown conditions combined. A CDFfor the operating plants is typically 5x10- 5 . The URD and EUR goal for new plants is that they be lessthan a CDF of 1x10- 5 . The exceptionally low AP1000 CDF results from, among other things:1) Use of non-safety related active systems, such as startup feedwater, for first response totransients, backed up by the passive safety systems2) An effective design deploying not only redundant systems but systems incorporatingdiverse equipment, such as valve types, where it has the most beneficial effect3) Eliminating the extra safety related equipment needed to generate emergency ac power.The AP1000 takes a direct and simple approach for the severe accident scenario. Our design avoidsex-vessel molten core interactions altogether. The AP1000 design allows for cooling of the vesselexterior with water from the large In Containment Refueling Water Storage Tank fed by gravity to thereactor cavity. It floods the cavity and flows up along the exterior of the reactor vessel, removing heat,and then is ultimately re-circulated through a steam condensation cycle within containment. Pressurebuild up within the vessel is relieved by the automatic de-pressurization system. The cooling issufficient to maintain the vessel integrity, thereby retaining the molten core inside. The steelcontainment, meanwhile, is cooled by natural air convection channeled by the reinforced concreteshield building featuring a system of air intake and exhaust vents. The ambient air is the AP1000’sultimate heat sink in case of accidents. The air cooling can be augmented by evaporation cooling froma water storage reservoir poised on the top of the shield building.The simplicity in AP1000 derives from passive systems not needing as much equipment to carry outtheir mission. In the AP1000 the HVAC, service water and circulating water systems, among others,are re-classified as non-safety related and serve only non-safety related equipment. The elimination ofthe usual network of safety-related pumps and supporting systems results in not needing safety-relatedemergency ac power. All of this greatly reduces the volume of Category 1 seismic buildings and allowmost safety equipment to be concentrated within the containment. It also results in 40 to 50% fewercontainment penetrations for AP1000 compared to a conventional plant.

All of the forgoing achieve new levels of safety and drive down plant cost. We estimate AP1000compared to a conventional plant to have:• 50% fewer safety class valves• 80% less safety class piping length• 35% fewer pumps of all types• 70% less cable.The net effect of AP1000’s reduced requirements for equipment and the building space needed tohouse the equipment is a very compact footprint. To gauge the effect on construction and size ofbuildings, following here is a comparison of some construction quantities for AP1000 compared toSizewell B, a Westinghouse PWR commissioned in the UK in 1995:Concrete, m3 Rebar, metric tons Power, MWeSizewell B: 520,000 65,000 1188AP1000:

6. ConclusionWe are witnessing a world-wide need to re-orient our energy generating base for more secure energysupplies and for reduced greenhouse gas emissions. The resolution of these issues will ultimately relyon technology, whether for new, clean coal designs, renewable energy, or nuclear power plants. Newnuclear power plants can provide a proven, cost-competitive, non-greenhouse gas emitting option that,as indicated by the World Energy Council’s report of February 1, 2007 and the UN’sIntergovernmental Panel on Climate Change report of May 4, needs to play an important role. To usethe terminology often used these days, nuclear power is one of the “wedges” needed to make thechange. The latest Westinghouse PWR, the AP1000, is a significantly improved version of the reliablePWR, and now available to perform that role.

Session 17.1.2:Safety of advanced reactors

AP1000 PASSIVE DESIGN FOR IN-VESSEL RETENTIONC. P. KEEGAN PE PMPWestinghouse Electric CompanyNuclear Power PlantsP.O. Box 355Pittsburgh, PA 15230-0355 - USAABSTRACTAP1000 is a Westinghouse two-loop 1100 MWe advanced pressurized water reactorthat uses passive safety features to enhance plant safety and provide improvements inplant simplification, reliability, investment protection and cost. One of the passivesafety features is In-Vessel Retention which passively provides sufficient externalcooling of the reactor vessel to retain a molten core inside the vessel in the unlikelyevent of a severe accident. This concept was proven by a series of tests reviewed andaccepted by the United States Department of Energy and Nuclear RegulatoryCommission and offers numerous advantages over other severe accident coremanagement designs.The testing that proved the In-Vessel Retention Concept also identified a series of keyfeatures and functions for the Reactor Vessel Insulation System (RVIS), making itdifferent from any other reactor vessel insulation. This paper presents the keyfunctional and design requirements for the RVIS and the RVIS design configuration.1. IntroductionPassive features are defined as those that do not rely on human intervention or devices such as motorsor pumps to perform their function but instead use natural phenomena like gravity. One of the passivesafety features of the AP1000 is named “In-Vessel Retention”. In-Vessel Retention passively providessufficient external cooling of the reactor vessel to retain a core that has been relocated to the bottomhead of the reactor vessel in the unlikely event of this severe accident. The In-Vessel Retentionconcept was proven by a series of tests and analyses and offers numerous and obvious advantagesover other severe accident core management designs. The testing and analyses for AP1000 built ontesting and analyses that previously demonstrated the In-Vessel Retention concept for theWestinghouse AP600 reactor vessel.The AP1000 testing and analyses were previously presented in a number of technical papersincluding:• “In Vessel Retention of Molten Core Debris in the Westinghouse AP1000 Advanced PassivePWR”, James H. Scobel, et al, ICAPP 2003• “Westinghouse AP1000 PRA Maturity, D. McLaughlin, et al, ICAPP 2005.These papers may be reviewed for more information on the testing and analyses. This paper focuseson the mechanical design and configuration that implements the testing and analyses results.Like the reactor vessel insulation in other nuclear power plants, the AP1000 reactor vessel insulationinsulates the reactor vessel to minimize heat loss to the cooling air in the reactor cavity. As in othernuclear power plants, the AP1000 reactor vessel insulation protects temperature sensitive structuresand components from exceeding their maximum allowable temperatures.However, to provide for In-Vessel Retention the AP1000 reactor vessel insulation must be different.The AP1000 In-Vessel Retention concept uses water that has flooded the reactor cavity as the coolingmedium for the reactor vessel. The flood water must freely contact the external surface of the reactorvessel for this cooling to occur. Conventional reactor vessel insulation forms a barrier between theflood water and the reactor vessel.

Not only must the Reactor Vessel Insulation System (RVIS) allow flood water to reach the reactorvessel, the reactor vessel insulation must provide certain additional features and functions which wereidentified from the testing and analyses. These features and functions make the AP1000 RVISdifferent from the reactor vessel insulation in any other nuclear power plant, including AP600.The following sections summarize the key functional and interface requirements for the RVIS for bothnormal and severe accident conditions and show the design configuration that meets thoserequirements.2. Key Functional and Design Requirements of the RVIS2.1. Normal ConditionsFig 1. Reactor Vessel Insulation System General Arrangement. All details are not shown.Figure 1 provides a representative pictorial of the RVIS. The RVIS includes the insulation below thetop flange on the reactor vessel. As in other nuclear power plants, the RVIS is located between thereactor vessel and the reactor cavity walls. Space is maintained between the RVIS and the reactorcavity walls and floor for cooling air flow. During plant design basis conditions, the RVIS limits heatloss from the reactor vessel. The RVIS and reactor cavity cooling air limit the temperatures ofstructures and components in the reactor cavity to within allowable limits during plant designconditions. The structures and components of concern are the concrete, the neutron shielding, and theex-vessel neutron flux monitors, also called “ex-core detectors”.Reactor cavity cooling air enters the bottom of the reactor cavity (floor elevation 71’-6”), flows underand around the insulation on the bottom head of the reactor vessel, up the outside of the reactor vesselsidewall insulation, through the reactor vessel supports into the nozzle gallery (floor elevation 98’-0”).2.2 Severe Accident ConditionsTesting and analyses have shown that the reactor vessel can retain a molten core in the bottom head ofthe reactor vessel if the external surface of the vessel is sufficiently cooled. Testing and analyses havealso shown that cooling will be sufficient if:• An annulus with certain dimensions and characteristics is maintained between the reactorvessel and the reactor vessel insulation along the bottom head and up the sidewall of thereactor vessel

• Water can freely and continuously flow into the annulus at the center of the reactor vesselbottom head• Steam and water can freely vent at the top of the annulus.The scenario during a severe accident is that containment flood water flows into the reactor cavity.While the flood water rises to a level above the reactor vessel, it freely flows into the annulus betweenthe reactor vessel and the reactor vessel insulation. The water in the annulus takes heat away from thereactor vessel and forms steam bubbles. The steam bubbles rise in the annulus pushing water ahead ofthem. The rising steam-water mixture flows out of the top of the annulus, returning to thecontainment flood water.The RVIS provides the outside wall of the annulus required by the first bullet above, and musttherefore remain intact under the loadings that occur during the accident. Because of the wateroutside the insulation and the steam void fraction inside the annulus, there is a differential pressureacross the insulation. Additionally the formation and collapse of bubbles in the annulus creates apressure oscillation. Testing and analyses have quantified these numbers to be 12.95 feet (3.95meters) of differential water pressure on the outside of the RVIS with a pressure oscillation of +/-1.64feet (0.5 meters) of water. These are loads that the RVIS has been designed to withstand.The second and third bullets above are counter to the requirements during normal conditions when itis important to inhibit air flow into and out of the annulus in order to minimize heat loss from thereactor vessel, minimize the heat load on the containment cooling system, and maintain componentsand structures within their allowable temperatures. Unique RVIS features are required that inhibit airexchange during normal operations but permit the bottom and top of the annulus to passively openduring a severe accident. A water inlet assembly and steam vents were designed to provide thesediverse features. These are described in Section 3.3. RVIS Design ConfigurationThe RVIS is primarily constructed of ASTM Type 304 stainless steel metal reflective insulation (MRI)4.5 inches (11.43 cm) thick. MRI generally consists of inside and outside sheet metal enclosures andmultiple layers of metal foils inside. MRI insulates by minimizing internal conduction paths betweenthe inside and outside enclosures, minimizing internal convection currents, and minimizing internalheat transfer due to radiation. MRI is used extensively in nuclear applications.The RVIS above the nozzle gallery floor is similar in design to MRI systems used on otherpressurized water reactor (PWR) vessels. The MRI is manufactured in panels which can be handledby one or two workers. The MRI panels are attached together and allow a slight stand-off from thereactor vessel to allow for manufacturing tolerances and reactor vessel expansion due to thetemperature and pressure increases. The MRI on each nozzle is a clam-shell design. The clam-shelldesign allows the MRI to be removed, for example for in-service inspection. The clam-shell sectionsfasten together along longitudinal seams and to the MRI on the sidewall of the reactor vessel. TheMRI clam-shells are supported by the MRI panels on the reactor vessel sidewall and by the nozzle.The RVIS below the nozzle gallery floor provides the annulus for In-Vessel Retention and musttherefore be designed to withstand higher loads. Examples of these higher loads include the large andvarying differential pressure discussed above during a severe accident and differential pressure due tocontainment pressurization, such as from a hypothetical pipe break in containment. To achieve theneeded strength, the MRI panels are attached to a supporting structure which is fastened to the reactorcavity walls. In addition the MRI panels in this region have thicker inside and outside enclosures aswell as additional stiffeners inside.The structure that supports the bottom head attaches to the reactor cavity sidewall and to legs thatextend downward and attach to the reactor cavity floor. The MRI panels provide a hemispherical

shape generally conforming to the shape of the bottom head of the reactor vessel and providing theannulus prescribed by the testing.As shown in Figure 2, there is an opening in the bottom head insulation at the axial centerline of thereactor vessel. The inlet assembly is centered on and fits around this opening. The inlet assemblyextends downward from the MRI on the bottom head of the reactor vessel to the reactor cavity floor,attaching to each. The inlet assembly is constructed of MRI and includes four sides and a bottom.Each side has a vertical section at the top and an inwardly sloping side below it; similar to a hopper.A space is maintained under the bottom to assure acceptable concrete temperatures are maintained inthis area.Fig 2. Water Inlet Assembly. Far door and other details are not shown.Each of the four sloping sides of the inlet assembly contains a door. The door is hinged on the topinside edge and four open doors provide more than 6 ft 2 (0.56 m 2 ) of flow area for water to enterthe inlet assembly. This is the minimum flow area established by testing and analysis. Theweight of the door keeps it in the closed position, pressed against a continuous circumferentialstop which inhibits the free exchange of air between the inside and outside of the inlet assembly.Each door is a hollow stainless steel enclosure filled with a buoyant hydrophobic material. Thebuoyant material causes the doors to rise as the reactor cavity floods with water. There issufficient room inside the inlet assembly for the doors not to contact each other during opening.Fully open, the doors are in a vertical position with their center of gravity over the hinge. In thisposition the water flow past the doors will tend to push the doors further open. If the doors everlose their buoyancy during the severe accident, gravity and the water flow work together to causethe doors fall outward instead of inward. The inlet assembly has sufficient space for the doors tofall outward beyond the vertical position. Small vent holes prevent pressure differentials insidethe doors during reactor heat-up and cool-down.Door freedom can be periodically checked by pushing them inward and checking their swing action.Each door is mounted in a frame and the door-and-frame assembly is removable as a unit. Removingand replacing the door-and-frame assembly as a unit eliminates fit-up concerns if the door only wereto be replaced. Replacement is not anticipated over the design life of the plant.As shown in Figures 3 and 4, at the top of the annulus at the floor of the nozzle gallery are steamvents. The steam vents are comprised of a series of narrow straight doors in the nozzle gallery thatextend continuously around the reactor vessel from the neutron shield to the RVIS panels. The steamvents fit under each of the reactor vessel nozzles and their open doors provide a total flow area of atleast 12 ft 2 (1.11 m 2 ). This flow area is a design requirement from the testing and analyses.The steam vent doors are constructed of MRI panels. The doors are hinged on their outside bottomedge and slope inward toward the reactor vessel. Their top edge continuously contacts the RVISpanels in the nozzle gallery and their sides have flow barriers to inhibit air flow between the annulusand the nozzle gallery. The momentum of the steam-water mixture rising in the annulus opens thedoors during a severe accident. Once open, the doors stay open due to gravity. If necessary, these

doors could be filled with a buoyant hydrophobic material like the inlet assembly doors to provide anincreased force for opening once the flood water level reaches this point, but this is not required. Likeother MRI panels, the doors themselves are vented to prevent internal/external pressure differentialsduring heat up and cool down.Fig 3. Steam Vent and Neutron ShieldFig 4. Reactor Vessel Support Cooling DuctsThe steam vent doors are accessible during routine periodic plant shut-downs. Door freedom canbe checked by swinging them outward to check their swing action. In addition, the doors andsteam vents are removable. Removal of the steam vents or doors is not anticipated to be requiredover the design life of the plant.Also as shown in Figures 3 and 4, between the steam vents and the insulation below them is theneutron shield. The neutron shield forms part of the flow path for both the cooling air flowing upwardfrom the reactor cavity during normal conditions and the steam-water mixture flowing upward to thesteam vents during a severe accident. The neutron shielding material is enclosed in an insulatedstainless steel enclosure to maintain the shielding material to less than 400 0 F (204.44 0 ) C. Coolingair flows upward over the lower RVIS and then flows over the underside and outboard surface of theneutron shield which also protects the concrete. The neutron shield has an octagonal outside shapeand a cylindrical inside shape to interface properly with the octagonal reactor cavity and thecylindrical reactor vessel. This is the reason for the different neutron shield widths in Figures 3 and 4.As shown in Figure 4, removable ducts mounted between the neutron shield and the reactor vesselsupport channel the reactor cavity cooling air to each reactor vessel support. Each reactor vesselsupport has an opening that allows the cooling air to pass through and baffles to make cooling moreefficient. These features maintain the concrete temperature under each support below the limit.Thermal and structural analyses have confirmed that the design meets its design requirements andinterfaces for normal and severe accident conditions.

OPTIMIZING THE ACR-1000 CORE FOR SAFETY,ECONOMICS AND RELIABILITYA.BUIJS, M.OVANES, P.S.W. CHAN, M. BONECHIAtomic Energy of Canada Limited2251 Speakman Drive, Mississauga, Ontario L5K 1B2, CanadaABSTRACTAECL has adapted the successful features of CANDU ®∗ reactors to establish theGeneration III+ Advanced CANDU Reactor ∗∗ (ACR ) technology. The ACR-1000 nuclear power plant is an evolutionary product, solidly based on CANDU reactortechnology, incorporating thoroughly demonstrated innovative features to enhance safety,operability, economics and maintainability.The ACR-1000 core design is based on well-established, fundamental, CANDU designelements: fuel enclosed in horizontal pressure tubes; a simple, efficient fuel bundle design;on-power fuelling; a separate, low-pressure, low-temperature heavy-water moderatorproviding an inherent emergency heat sink.This paper summarizes design optimisation features of the ACR-1000 Reactor Coredesign. The ACR-1000 reference core is described, as well as the main core features thatdistinguish the ACR-1000 from its predecessor CANDU reactors: the reactivity controldevices and shutdown systems, including rods for maintaining the reactor in a guaranteedshutdown state; adaptations for a light-water cooled, low-enriched uranium fuel, leading tomore benign neutronic characteristics; a more compact core configuration and highersteam pressure for greater thermodynamic efficiency.IntroductionThis paper focuses on enhancements in the ACR-1000 design with respect to existing CANDUreactors[1]. Table 1 shows a comparison of relevant core parameters between a generic CANDU 6reactor as built in Qinshan, the larger CANDU reactors operating at Darlington and Bruce, and theACR-1000.The table shows that the ACR-1000 delivers a higher power in a core that is smaller than the CANDU-6 core. This is achieved by the combined effects of reduced lattice pitch, light water coolant andenriched uranium.Parameter CANDU-6 Darlington ACR-1000Heat to steam generators (MW) 2064 2657 3208Gross/net electric output (MWe) 728/666 935/881 1165/1085Number of channels 380 480 520Core diameter (m) 7.6 8.5 7.44Lattice pitch (cm) 28.6 28.6 24.0Moderator D 2 O volume (m 3 ) 265 312 235Heat Transport System D 2 O volume (m 3 ) 192 280 0Total D 2 O volume (m 3 ) 466 602 240Fuel (wt% U 235 /U) 0.71 0.71 up to 2.5Number of elements per bundle 37 37 43Total bundle weight (kg) 24.1 24.1 21.5Reference discharge burnup (MWd/Mg(U)) 7500 7800 20000Outlet header operating pressure (MPa) 9.9 9.9 11.1Outlet header operating temperature (°C) 310 310 319∗ CANDU ® is a registered trademark of Atomic Energy of Canada Limited (AECL).∗∗ Advanced CANDU Reactor TM , ACR TM and ACR-1000 TM are trademarks of AECL.

Parameter CANDU-6 Darlington ACR-1000Steam temperature (°C) 260 265 276Table 1: Comparison of relevant CANDU reactor parameters.At the same time, the similarity between the ACR-1000 and existing CANDU reactors is evident fromthe table, underlining the evolutionary nature of the ACR-1000. The following sections will describea number of key enhancements of the ACR-1000 with respect to the traditional design.2. Reactor CoreAs in existing CANDU reactors, the reactor core of the ACR-1000 consists of a calandria vesseltraversed by horizontal fuel channels.Figure 1 shows comparisons of the core layouts of CANDU-type reactors in Table 1. It clearlydemonstrates the compact design of the ACR-1000.CANDU-6 Darlington ACR-1000Figure 1: Comparison between the calandria dimensions of CANDU reactors and the ACR-1000With respect to the traditional CANDU design, the ACR-1000 shows a large reduction of the volumeof D 2 O, mostly caused by the elimination of D 2 O as a coolant, but also by the reduction of themoderator volume. This leads to a significant reduction in capital cost for the ACR-1000.3. Basic Lattice Cell3.1 Pressure and Calandria TubesFigure 2 shows the basic ACR-1000 lattice Central PinModerator (D 2O)cell, including the arrangements of the fuel Fuel Ringbundle, the H 2 O coolant, pressure tube,Gapcalandria tube, and the D 2 O moderator. Thepressure tube of the ACR-1000 has beenmade slightly thicker than that of theexisting CANDU. This is to reduce thecreep and sag of the tube over its lifetime.Creep, or diametral expansion, of thepressure tube reduces the flow through thefuel, and thus the ability to cool the fuel.The gap between the calandria tube and the24.0 cmpressure tube has been widened in order toreduce the moderator-to-fuel volume ratio,Figure 2: The basic ACR-1000 lattice cellwhich helps reduce the reactivity effect of coolant voiding.3.2 Fuel DesignCalandria TubePressure TubeCoolant (H 2O)

The fuel used in the ACR is uranium dioxide sintered in the form of cylindrical pellets and clad inzircaloy sheath using the 43-element CANFLEX-ACR fuel bundle design. The centre ring consists ofone large diameter element, whereas the outer three rings consist of 42 elements with a smallerdiameter. To reduce the coolant void reactivity during postulated accidents, dysprosium andgadolinium are blended, in a matrix of Zirconia, as burnable neutron absorbers in the centre element ofthe bundle, which does not contain any fissile material. The three outer rings of fuel elements containlow-enriched uranium (LEU) pellets. The fuel enrichment and the burnable poison concentration canbe tailored to meet desired design targets of fuel burnup and coolant-void reactivity.4. Reactivity Effects and Reactivity Control4.1 Coolant Void Reactivity and Power Coefficient of ReactivityCANDU fuel channel reactor designs have traditionally featured the advantages of small absolutemagnitudes of reactivity coefficients. The flexibility of the ACR design allows these to be optimised.An important feature of the ACR-1000 core design is the selection of fuel parameters to achieve asmall, slightly negative value of coolant void reactivity, achieved by the combined effects of a reducedlattice pitch and the addition of a neutron absorber to the central pin of the fuel element. A latticepitch of 24 cm reduces the moderator-to-fuel ratio, while still maintaining all reactor-face maintenanceactivities, including the ability to replace single fuel channels. When combined with the light-watercoolant, these features are chosen to deliver a small negative power coefficient of reactivity. Reactivitycoefficients that are small and negative allow for an easy control of the reactor by slight adjustmentsof the control rod insertions. Small magnitude coefficients avoid the need for other large-scaleemergency means of reactivity hold down such as boron injection into the coolant system as part ofemergency coolant system action, and at the same time, render the consequences of accidents morebenign.4.2 Reactivity ControlAll ACR-1000 reactivity and shut-off devices are located in the low-pressure environment of themoderator, and are hence not subject to large forces or stresses. Whereas traditional CANDU reactorsemployed light water as absorber in the zone control units, all reactivity devices (with the exception ofthe poison injection system) in the ACR-1000 are solid neutron absorbers of boron-carbide-filledstainless steel tubes sliding in zirconium alloy guide tubes. The cross section of the reactivity devicesis a flat rectangle, to permit insertion in the tight lattice while retaining sufficient reactivity worth.The ACR-1000 Reactor Regulating System (RRS) consists of zone-control units (ZCU) andmechanical control-absorber units (CAU). Each ZCU contains two independently-movable mechanicalabsorber elements, one driving in from above, one from below. The zone-control units perform thesame bulk and spatial control function as the liquid-zone controllers in CANDU 6 reactors, but withthe operational simplicity of solid absorber design. The power of each control region can be adjustedby varying the degree of insertion of the absorber in the corresponding unit. In this way, power levelsacross the core are maintained at design targets while individual fuel channels undergo refuelingoperations on line. The power in the centre of the core can be controlled by varying the degree of theoverlap of the ZCUs in the centre. This enables fine control over peak fuel channel and bundlepowers, thus increasing operating margins.The ACR CAUs perform the same function as the mechanical absorber units in CANDU 6 reactors.The CAUs are designed to provide rapid controlled reductions of reactor power when power setbackor stepback is initiated. The degree of insertion and the number of CAUs to be inserted aredetermined by the amount of power reduction required by the RRS.5. Reactor Safety Systems5.1 Shutdown SystemsACR-1000 retains the two independent fast-acting shutdown systems (SDS1 and SDS2) present inother CANDU reactors. Each system is physically, logically and functionally separate from the otherand from the RRS. Each of these shutdown systems is independently fully capable of rapidly shuttingthe reactor down in any postulated accident scenario, and to maintain the reactor in a shut down state.The ACR-1000 Shutdown System 1 (SDS1) has vertical shutoff rods (SOR), which are designed toshut the reactor down quickly under emergency conditions. Shutdown is achieved by rapidly inserting

neutron-absorbing elements into the reactor core. The insertion of the SOR is initiated by the controllogic of the SDS1. The ACR SOR’s are based on the shut-off rod technology employed in CANDU-6units, with the adaptation to plate-type absorber cross section, in deviation from the previouscylindrical cross section.The ACR-1000 Shutdown System 2 (SDS2) has horizontal liquid-injection nozzles, which cross thecore at various locations. The SDS2 is designed to inject enough liquid poison (a concentratedgadolinium nitrate solution) to blanket the core within two seconds after actuation, sufficient to shutthe reactor down rapidly during all postulated accidents. SDS2 contains enough poison to keep thereactor in the shutdown state under all foreseeable conditions. Again, the SDS2 delivery system takesadvantage of components and features from the corresponding system on the CANDU 6 reactordesign.5.2 Regional Overpower ProtectionLike the traditional CANDU reactors, the ACR-1000 is designed with a regional overpower protection(ROP) system, which triggers the shutdown systems when protective actions are warranted. The ROPsystem consists of two sets of self-powered platinum-clad inconel detectors distributed over the core.The sets are functionally independent from each other and are both further subdivided into fourindependent safety channels. The shutdown system will be activated when two out of four detectorsof a system have registered a signal above a pre-set threshold value. Traditional CANDU reactors areequipped with a three-channel ROP system. With the four-channel system design of the ACR-1000,the chance of a spurious trip during testing of one of the safety channels is eliminated, simplifyingoperator testing. This further enhances the advantages of on-power fuelling, testing and maintenancethat allow the ACR-1000 to achieve a three-year interval between maintenance outages.5.3 Guaranteed Shutdown StateThere is a strong desire from an operational point of view not to use an over-poisoned moderator toachieve a Guaranteed Shutdown State (GSS), due to significant activities and constraints requiredduring outage operations. Additional absorber rods, of a mechanical design similar to the SOR’s, areused to achieve a rod-based GSS without the need for poison addition in the moderator, except forstart-up and fuelling ahead conditions. The reactivity depth of the GSS system is sufficient to keep thereactor in the guaranteed shutdown state indefinitely.6. On-Power FuellingLike traditional CANDU reactors, the ACR-1000 has twelve fuel bundles per channel, which arereplaced on-power at a rate that compensates for the reactivity loss due to the depletion of 235 U. Theability of on-power refuelling allows the fuelling engineer to maintain an optimised channel powershape, which ensures optimum power output at maximum bundle burnup, while adhering to the safetymargins imposed on the channels powers. The on-power fuelling represents a safety feature as well,considering that no poison needs to be present in the moderator, nor does excessive core reactivityneed to be held down by reactivity devices.7. BurnupThe targeted discharge fuel burnup in the ACR-1000 is 20000 MWd/Mg(U), which is about threetimes the burnup in the current CANDU reactors using natural uranium fuel. Initially, the reactor willhave a fresh start–up core and, after going through a fuel and core transition with fuel burnup of about10000 MWd/Mg(U), will reach the equilibrium reference fuel and core configuration with fuelenrichment of up to 2.5% wt for 235 U and fuel burnup of about 20000 MWd/Mg(U).The fuel-management flexibility of the ACR design allows further improvements in fuel burnupwithout requiring any modifications to the basic reactor design.8. Alternate FuelsThe feature of on-power fuelling of individual fuel channels, combined with a flexible fuel bundledesign, allows the ACR reactor to use a variety of fuel types and management strategies. Studiesunder way indicate that the ACR-1000 is adaptable to the use of recycled plutonium in the form ofmixed plutonium and uranium oxide (MOX) fuels. In particular, the ACR-1000 design may enablethe use of 100% MOX fuel in the reactor core without the need for costly reactor design changes or

performance penalties. Furthermore, a number of very attractive options for establishing the thoriumcycle in the ACR are being considered [2,3].9. ConclusionsThe ACR-1000 achieves substantial reduction in capital cost by using H 2 O coolant, LEU fuel, in acompact D 2 O-moderated lattice with respect to the traditional CANDU design. Full-core coolant voidreactivity, as well as major reactivity feedback coefficients, are slightly negative under nominaloperating conditions. The negative power coefficient aids in a smooth control of the power shape bythe reactivity devices, and limits the speed and magnitude of power excursions following postulatedaccident scenarios. The flexibility provided by on-power fuelling and simple fuel bundle designenables the ACR to accommodate gradual increases in enrichment and hence in fuel burnup and toadapt to the use of various fuel cycles. This flexibility allows it to meet strategic energy requirementsas they evolve over time, and to respond to changes in technology and resource availability.10. References[1] The CANDU CANTEACH website (http://candu.canteach.org) provides a comprehensive reviewof all aspects of the CANDU reactors.[2] J.M. Hopwood, P. Fehrenbach, R. Duffey, S. Kuran, M. Ivanco, G.R Dyck, P.S.W. Chan, A.K.Tyagi and C. Mancuso, “CANDU Reactors with Thorium Fuel Cycles”, Proceedings of 15 thPacific Basin Nuclear Conference, 2006 October.[3] P.G. Boczar and G.R. Dyck, “CANDU Fuel Cycle Vision in China”, Proceedings of 13 thInternational Conference on Nuclear Engineering (ICONE 13), Beijing, China, 2005 May.

APPLICATION OF TECHNOLOGY NEUTRALMETHODOLOGY FOR ASSESSMENT OF DEFENCE INDEPTH APPLICATION FOR GENERATION IV REACTORSYSTEMSV.P. RANGUELOVAInstitute for Energy, European Commission, Joint Research Center,PO Box 2 1755 ZG Petten - The NetherlandsG.L. FIORINI2 Nuclear Energy Division/ Reactor Studies Department/ Innovative System SectionCEA/CADARACHE - Bt 212, 13108 Saint Paul Lez Durance cede - FranceT.J. LEAHYIdaho National Laboratory (INL), 2525 Fremont AveIdaho Falls, ID 83415 – 385, - USAABSTRACTThe Generation IV International Forum (GIF) Charter envisions the safe and reliableoperation of nuclear systems as an essential priority in the development of next-generationsystems. Safety and reliability goals broadly consider operations, improved accidentmanagement and minimization of consequences, investment protection, and reduced needfor off-site emergency response. This emphasis on enhanced safety and reliability has beenduly reflected in the Policy Group’s selection of the system designs, as well as inrecognition of the need to establish in 2004 a system cross cutting methodology workinggroup on Risk and Safety ( RSWG). The group was charged with the responsibility topromote a homogeneous approach to safety and quality in the design of Generation IVsystems and define the framework for safety design and evaluation methodology whichcould be applied to all reactor systems. It was recognized that development of advancedand enhanced safety assessment methodologies might be needed for this purpose.The paper provides an overview of the recent GIF Risk and Safety Working Groupdevelopments of technology neutral nuclear safety requirements and safety assessmentmethodologies. In particular, it addresses the applicability of “Objective Provision Tree”(OPT) methodology for verification of defence in depth application for new designs, aswell as on the need to complement this methodology with traditional safety assessmenttechniques such as deterministic accident analysis and probabilistic safety assessments. Thelater are needed to judge on the adequacy of the safety provisions and relative importanceof different provisions. The experienced gained with the pilot applications of the OPTmethodology to different reactor designs will be also discussed along with the further effortneeded to validate the applicability of this new proposed methodology for GIF reactorsystems.1. IntroductionThe Generation IV initiative concerns the identification, development and demonstration of one ormore new nuclear energy systems that offer advantages in the areas of economics, safety andreliability, and sustainability, and could be deployed commercially by 2030. Six innovative reactorsystem concepts are covered under the Generation IV International Forum (GIF) agreement, namely:Gas-Cooled Fast Reactor, Lead-Cooled Fast Reactor, Molten Salt Reactor, Sodium-Cooled FastReactor, Supercritical Water-Cooled Reactor and Very-High-Temperature Reactor System. Allconcepts potentially present a diverse set of design and safety issues. A number of these issues are

significantly different from those presented by the earlier generations of light water reactors. Theoverall success of the Generation IV program depends, among others, on the ability to develop,demonstrate, and deploy advanced system designs that exhibit excellent safety characteristics.In order to address nuclear safety concerns in a consistent manner throughout the GEN IV reactorsystems and in order to provide the designers with safety concepts and methods that can help guidingtheir R & D activities towards improved safety, a system cross cutting methodology group on Riskand Safety (RSWG) was established in 2004. The primary objective of the RSWG is to assure aharmonized approach on long-term safety, risk and regulatory issues in development of the nextgeneration systems. To this end, the RSWG focuses particularly on defining safety goals andevaluation methodology and advising and assisting the GIF Experts Group and Policy Group oninteractions with the nuclear safety regulatory community, and other relevant interested partiesincluding IAEA. The RSWG is comprised of representatives nominated from interested GIVMembers and different System Steering Committees. The Euratom is represented in this group by theInstitute for Energy of the Joint Research Centre, EC and VTT (Valtion Teknillinen Tutkimuskeskus)from Finland.2. Objective Provision TreeThe Generation IV research and development program is guided by a GIF IV Technology Roadmapdocument (Ref. [1]) which identified three specific safety goals for Generation IV systems “to be usedto stimulate the search for innovative nuclear energy systems and to motivate and guide the R&D onGeneration IV systems”:1. Generation IV nuclear energy systems operations will excel in safety and reliability.2. Generation IV nuclear energy systems will have a very low likelihood and degree of reactorcore damage.3. Generation IV nuclear energy systems will eliminate the need for offsite emergency response.While the RSWG recognizes the excellent safety record of nuclear power plants currently operating inGIF member countries, it believes that advanced technologies and a coherent safety approach hold thepromise of making Generation IV energy systems even safer and more transparent than this currentgeneration of plants. In its first two years of existence, the RSWG has focused on defining theattributes that are most likely to help meet these Generation IV safety goals, and identifyingmethodological advances that might be necessary to achieve or demonstrate achievement of thesegoals. The results of the group developments are summarized in RSWG Report on the Safety ofGeneration IV Nuclear Systems [2].One of the important issues addressed in this report is the need to apply the fundamental principle ofdefence-in-depth in a consistent manner from the very first stage of the reactor system design. Defencein depth is the key to achieve safety robustness, thereby helping to ensure that Generation IV systemsdo not exhibit any particularly dominant risk vulnerability. To meet these objectives the defence indepth has to be implemented in a way which is systematic, exhaustive, progressive, tolerant, forgivingand well-balanced.To help GIV designers to correctly implement the defence-in-depth, to assess how well the latter hasbeen applied for their reactor systems and identify areas which deserve further research, RSWG hassuggested to utilize the Objection-Provision Tree (OPT) methodology complementing it with requiredtraditional deterministic and probabilistic safety assessments.The notion of OPT is first defined in the IAEA TECDOC 1366, Considerations in the Development ofSafety Requirements for Innovative Reactors: Application to Modular High Temperature Gas CooledReactors [3]. In addition the IAEA Safety Series Report No 46: Assessment of Defence in Depth for

NPPs [4] presents a practical tool for inventorying the defence in depth capabilities of an operatingNPP, including both the design features and the operational measures. For this purpose the definitionof defence in depth and the guidance on its implementation agreed upon by international consensus(Ref. [5] & [6), have been combined into a logical graphical framework – Objective Provision Trees -that can be used for assessing the comprehensiveness and quality of defence in depth at a plant.The OPT is a graphical representation of design safety architecture which identifies for each level ofdefence-in depth, with regard to each of the safety functions, of the provisions required to realize therequired missions. All five levels of defence in depth are to be covered by the plant design.Fig 1. Simplified representation of Objective Provision TreeIn other words, for given objectives at each level of defence, a set of challenges 1 is identified, andseveral root mechanisms 2 leading to the challenges are specified. Finally, to the extent possible thecomprehensive list of safety provisions 3 , which contribute to prevent that the mechanism takes place,is provided. The broad spectrum of provisions, that encompass the inherent safety features, equipment,procedures, staff availability, staff training and safety culture aspects, are considered [4].For easier understanding, the user-friendly application of the method, including the overview of allchallenges, mechanisms and provisions for all levels of defence, is illustrated in the form of“objective provisions trees!” 4 .3. Application of OPT to GEN IV Reactor Systems1Challenges: generalized mechanisms, processes or circumstances (conditions) that may impact the intendedperformance of safety functions; a set of mechanisms have consequences which are similar in nature.2 Mechanism: specific reasons, processes or situations whose consequences might create challenges to theperformance of safety functions.3 Provisions: measures implemented in design and operation such as inherent plant characteristics, safetymargins, system design features and operational measures contributing to the performance of the safetyfunctions aimed at prevention and control of the mechanisms to occur.4 Objective provisions tree: graphical presentation, for each of the specific safety principles belonging to the fivelevels of defence in depth, of the following elements from top to bottom: (1) relevant safety functions; (2) safetyobjective of the level; (3) identified challenges; (4) constitutive mechanisms for each of the challenges; (5) listof provisions in design and operation preventing the mechanism to occur or achieving its control.

The application of the Objective Provision Tree (OPT) methodology to assess the implementation ofdefence- in depth concept for an operating LWR plant, and in particular for the WWER 440/V213reactor units at Bohunice NPP [4] was considered and analyzed by the RSWG. In addition, a pilotstudy was conducted to assess the methodology applicability to the Japan Nuclear Cycle DevelopmentInstitute Sodium Cooled Fast Reactor [7] as part of GEN IV reactor systems.In the first case it was noted, that for LWR type of designs detailed OPTs are already developed by theIAEA [4] for 68 specific safety principles/ safety functions which are linked to the three fundamentalsafety functions. The user of the methodology needs simply to assess whether the identified provisionsin the OPTs are present at his/ her plant. In some cases some alternative provisions might be identifiedby the methodology user, however the main difficulties are experienced in assessing the adequacy ofthe provisions. The need to use deterministic, probabilistic and other type of analyses for this purposeas well as additional research in some cases is clearly demonstrated.The application of OPT methodology to innovative reactor designs needs, however more efforts todevelop first OPTs which will have to evolve with the evolution of the design. For instance, at an earlydesign/conceptual stage only very general OPTs could be developed for the fundamental safetyfunctions, while at a more advanced stage OPTs have to be developed to address in detail thesubsidiary safety functions( similarly to the 68 specific safety principles identify for the LWRs). Whilethere might be some similarity with LWR, it is clear that GIV IV is deploying innovative technologiesand concepts which will require new thoughts to be given on the way safety is ensured for these newtype of reactors, possible new failure mechanisms, and why not to the safety principles themselves.Rigorous approach will have to be applied in development of the OPTs for each of the reactor systemsin order to ensure that all levels of defense-in-depth have been addressed in comprehensive manner.Difficulties could be expected in identification of all challenges and mechanisms and possibleprovisions for each of the GEN IV Reactor systems. The demonstration of the adequacy of theprovision performance will be in its own a challenge in many cases. This will have to be done throughapplying of good engineering and performance of high quality research. The role of probabilisticsafety assessment (PSA) to assess reliability of the safety provisions will also need an innovativeapproach since many of the design solutions deployed for GEN IV are not supported by anyoperational experience so far. It is however, RSWG belief that applying OPT can help designers todefine their R&D plans in the most cost effective manner by focusing on the provisions andphenomena with high contribution to safety.4. ConclusionsBoth studies, the Bohunice NPP and JSFR, have demonstrated that there is a lot of potential benefitsfor the GIF reactor designers from the application of OPT methodology. It can help to ensure that ateach stage of reactor system design adequate provisions are foreseen to ensure the application of all 5levels of the DiD concept and identify topics where more research and development activities areneeded to justify and prove this statement.In order to facilitate the designer’s use of OPT methodology it will be important for RSWG to developan application guide. This guide can be established in an electronic form to facilitate the building up ofOPTs and provide predetermined options to be selected for safety objectives, functions, challenges,mechanisms, provisions ( at least for the technology neutral ones). The reference to any availablesafety requirements for any of those items can also be incorporated and be available for designerconsultations. The experience in building detail OPTs for LWR shall be repeated in the GIF reactorscontext for the new reactor systems. The aim would be to help designers to identify all necessaryprovisions to ensure safety and define R&D activities which are needed to demonstrate the adequatecapabilities of selected provisions.

As any safety assessment methodology, the OPT has its limitations which are mainly related to theevaluation of the adequacy of the identified provisions and their prioritization or determination of theirsafety significance. It is clear that for these issues traditional deterministic (accident analyses) andprobabilistic safety assessment will be needed to complement the OPT. A number of iterations ofcombined use of all these methods will have to be done to ensure that a comprehensive and systematicassessment has been performed for each of the GIF reactor systems.5. References[1] GIF IV Technology Roadmap[2] GEN IV RSWG, On The Safety of Generation IV Nuclear Systems, June 2007, Paris, NEA[3] Considerations in the Development of Safety Requirements for Innovative Reactors:Application to Modular High Temperature Gas Cooled Reactors, IAEA TECDOC 1366,Vienna (2003)[4] Assessment of Defence in Depth for Nuclear Power Plants, Safety Reports Series No 46,IAEA, Vienna (2005)[5] Defence in Depth in Nuclear Safety, INSAG-10, A report by the International Nuclear SafetyAdvisory Group, IAEA, Vienna (1996)[6] Basic Safety Principles for Nuclear Power Plants, 75-INSAG-3 Rev.1, INSAG-12, A report bythe International Nuclear Safety Advisory Group, IAEA, Vienna (1999)[7] Findings from pilot use of the OPT methodology for JSFR, H. Niwa, S. Kubo, JAEA,Presentation given at the 4th GIF RSWG Meeting, Paris (26-28 April, 2006)

THE BALANCED SAFETY APPROACH OF EPRB. GUESDONBusiness Development, Plant Sector, AREVA NPTour AREVA, 92084 Paris La Défense Cedex – FranceABSTRACTGeneral safety objectives taken into account for EPR design considered both harmonized safetyrequirements of Safety Authorities and safety experts of France and Germany and the EuropeanUtility Requirements (EURs). This led to design an “evolutionary” PWR that has benefitedfrom the experience feedback of LWRs in operation. However innovative features were alsoconsidered and included in the design where necessary.The safety features were defined to ensure with a very high level of reliability the safetyfunctions. Reliability target is achieved through an adequate combination of redundancy,segregation and diversity. The selection and implementation either of active safety features orpassive and inherent safety provisions depend upon the advantages and drawbacks of thevarious design options.Such a balanced safety approach allows to meet most of the requirements of the Nuclear SafetyAuthorities of western countries, without the need for significant design variations, thusallowing an international generic design to be developed with the benefits of standardization.1. IntroductionAREVA's Evolutionary Power Reactor (EPR) results from a French-German cooperation set up todevelop this large 4-loop PWR of the latest generation [1]. EPR is presently under construction in Finlandfor TVO and in France for EDF. In different countries, utilities are considering EPR as a major option fortheir new builts; one of the reasons is the wide acceptability of the design due to the balanced safetyapproach that ensures a very high level of safety.This paper aims at giving a synthetic view of this approach; it is not intended to address here all aspects ofthe safety case but rather to focus on the mitigation of the faulted conditions; other important topics suchas improving radioprotection or limiting the production of radioactive waste are not here dealt with.2. The general safety objectivesFrench and German Safety Authorities and safety experts worked closely together during the EPR basicdesign phase and, in 2000, the "Technical guidelines for the design and construction of the new generationof pressurized water reactors"(TGs) [3] were endorsed. The TGs require significant safety improvement tobe incorporated in the design of the next generation plants, in comparison to NPPs presently in operation,however in harmonising the requirements of France and Germany a balanced approach was sought ratherthan simply a summation of requirements identified by each country.By this time, the European Utility Requirements (EURs) were defined by utilities from different countrieswith very diverse approaches to NPP regulation and licensing. The utilities wished to promotestandardisation of the designs of new NPPs, and therefore also sought harmonisation of design rules, inparticular with regard to nuclear safety.EPR design was developed in accordance with both the TGs and the EURs.3. The "evolutionary” designThe evolutionary design of EPR is in line with the TGs requirement as well as the wishes of manyEuropean Utilities. The reactor benefited from development, design, construction and operational

experience of the hundreds of LWRs in operation worldwide and took the best key technological featuresof the French N4 and of the German Konvoi PWRs.The designers implemented an exhaustive, progressive and robust defence to guarantee a very high levelof protection to investment, persons and the environment.- Exhaustiveness is based on a deterministic approach supplement by safety assessment, leading tocomprehensive analyses of design bases events, design extension conditions, internal and externalhazards.- Progressiveness starts with the implementation of surveillance and limiting functions which will reactshould the control system failed and avoid to actuate the protection functions and goes as far asmeasures to preserve the integrity of the containment and avoid significant radioactive releases shoulda severe accident occur.- Robustness is ensured by seeking reliability of the safety relevant features through redundancy,segregation and diversity and the use of known materials and technologies for design measureswhenever possible.3.1 Deterministic and probabilistic safety assessmentThe safety principles and criteria against which the EPR was developed are based on a strongdeterministic defence-in-depth safety concept complemented through a probabilistic safety assessment(PSA). Probabilistic consideration was incorporated from the outset into the design process in order toidentify accident sequences capable of leading to severe core damage or significant releases ofradioactivity, to evaluate their probability of occurrence, and to assist in implementing design features toreduce the contribution of such sequences to the overall risk.The early use of PSA provided also a basis for assessing the relative advantages of different designoptions, while verifying design compliance with initial project objectives.3.2 Design Basis Conditions and Design Extension Conditions (Risk Reduction Categories)Faulted conditions taken into account at an early stage of the design were extended in comparison with theprevious reactor generation. The whole plant life and all operating modes were considered, in particularthe shutdown states are explicitly addressed in the deterministic and probabilistic fault analyses, theyshould not contribute predominantly to the core melt frequency.First, Design Basis Conditions, i.e. postulated event initiated by the failure of one component or one of theI&C function, were used primarily for designing and sizing the Protection and Safeguard systems.Anticipated Operational Occurrences (Condition 2) were addressed for designing the surveillance andlimitation features which aim at avoiding that small deviation from normal operating condition couldevolve towards more adverse conditions, the benefit provided by such surveillance and limitation featureis assessed through the PSA. However it is also determistically checked that, should these limitationprovisions failed, the protection and safeguard systems ensure that criteria for core integrity are met.Infrequent accident (Condition 3) and Limiting accident (Condition 4) are analysed with a conservativeapproach to give an adequate degree of confidence in the defence efficiency.Second, three types of Design Extension Condition were considered:1) The "complex sequences" that could lead to core melt due to multiple failures. Such sequences arederived from PSA analysis, they result either from a complete loss of a safety function after occurrenceof a design basis initiating fault, e.g. Anticipated Transient Without Scram, loss of off-site power andfailure of the four emergency diesel generators (station blackout)… or from combination ofindependent events. The deterministic analysis of the complex sequences is aimed at demonstrating theeffectiveness of the safety measures implemented to reduce the risk of core melt to a very low, thusacceptable, level (e.g. additional small diesel generators to cope with the station blackout).2) Severe Accidents (SA) are analyzed to assist the design features for preventing large early releases incase of a postulated core melt; this is described in a next paragraph.

3) Specific studies address fault analysis of event that have been excluded from the design basis due toprobabilistic or deterministic reasons but are nevertheless analysed in order to check the absence of anycliff-edge effect in the plant safety demonstration, or to introduce additional safety margins in thedesign of certain systems and components, if necessary, to avoid this effect. For instance, the double–ended break on the main coolant lines (2A-LOCA) is analysed, despite the application of the breakpreclusion concept to the main primary systems, in particular it is verified that the pressure andtemperature in the containment building remain lower than their design values.3.3 Severe accidentThe emphasis given to the reduction of the core melt frequency is backed up by taking into account severeaccidents in a deterministic way, and mitigation measures are designed so that the associated maximumconceivable releases would require only very limited protective measures [4].Core melt sequences which would lead to large early releases are "practically" eliminated; provisionsagainst such a scenario have a reliability so high that this kind of event can be excluded; for instancededicated diverse valves are implemented to supplement the three pressurizer safety valves, thus avoidingany high pressure core melt sequence.To deal with low pressure core melt sequences, specific features are provided for retaining the melt withinthe containment to prevent penetration of basemat by corium-concrete interaction and to keep theconfinement integrity without the need for any containment venting.The effectiveness of the mitigation features was comprehensively verified for different possible accidentscenarios. As there is significant amount of uncertainty surrounding the main physical phenomenainvolved in the demonstration of in-vessel corium retention by outside flooding of the Reactor PressureVessel, cooling the corium on a dedicated spreading is seen as a more robust solution for large PWRs.3.4 Internal and external hazardsExternal and internal hazards that could affect the plant are identified on a generic basis, and provisionmade to ensure that the risk from the hazards is commensurate with the overall frequency and releasetargets. A deterministic analysis aims at ensuring that the safety functions needed to bring the plant in asafe shutdown state and to limit radiological releases are not unacceptably affected by hazards.Hazards are also covered by the Probabilistic Safety Assessment to verify that they do not contributepredominantly to the risk of core melt or large radioactivity releases.The layout configuration and structural technology is chosen in order to provide a high degree ofrobustness of the whole facility in relation to internal and external events, and also to accommodateunanticipated events. The plant layout contributes to the reliability of the safety functions by protectingthe relevant equipment, in particular by bunkerisation and segregation.4 Achieving reliable safety functionsSafety features were defined and designed to achieve the safety functions, with a very high level ofreliability through an adequate combination of redundancy, segregation and diversity. The challenge forthe designer was to defined an optimal mix between largely proven solutions derived from the largeexperience basis and innovative features needed to meet new requirements.Innovative features were included in the design where necessary for providing cost-competitive solutionwhile ensuring a significantly higher level of safety (especially to prevent and mitigate severe accidents,or to reduce the risk of radioactive release by mitigating any Steam Generator Tube Rupture) relative toprevious generation of PWRs.For design, selection and validation of these options, a lot of R&D work were performed, mostly inFrance, with a significant support of the Commissariat à l'Energie Atomique (CEA) and in Germany.4.1 RedundancyThe main safeguard systems (e.g. Safety Injection System, Emergency Feedwater System) are arrangedtogether with their associated control and support systems in a 4-train configuration. This arrangement

leads to a simple design concept for the fluid system: each train is connected to one of the reactor loop.This architecture makes it possible for a system to fulfil its function even if one train is unefficientbecause of the impact of the postulated initiating event, a second train is affected by a single failure whileanother train is unavailable due to preventive maintenance.However, when a four-train configuration is not necessary, a twofold configuration is adopted. Forexample the Extra Boration System or the Containment Annular Space Ventilation System are needed tomitigate postulated events which do not impact the efficiency of one train, and preventive maintenance onthese systems is not scheduled during power operation. Therefore the safety function can be fulfil with a2-train system, assuming a single failure on one train.4.2 SegregationDue consideration is given to the possibility of common cause failures that limits the benefit provided byadding identical train. The likelihood of common mode failure due to internal hazards is minimized byphysical and spatial segregation of the redundant trains of the safety systems. There are four safeguardsbuildings, every train is located in a different building with its support systems.These buildings are also protected against external hazards such as a large commercial airplane crasheither by very thick outer walls (for building #2 and #3) or by physical separation; building #1 andbuilding #4 are on both side of the reactor building, only one of these buildings could be affected by thehit, the other remaining operable.4.3 DiversityTo ensure that a diverse means can be used as a backup whenever the total failure of a safeguard systeminduces a significant risk of core melt or radioactive releases - the event sequences that fall into thiscategory being identified by probabilistic Safety Assessment - any safety grade system function can bebacked-up by another system or a group of systems. The drawbacks due to diversity such as the use ofmore complex system or additional maintenance burden are outweighed by the risk reduction provided bysuch a design.Some example of diversity implementation are:- Two small diesel generators, diverse from the four main diesel generators supply power to twosafeguard trains in case of a station blackout.- A diverse I&C channel actuates reactor and turbine trip in case of failure in the Protection System; thischannel is implemented in the Process Automation System outside the Protection System withadequate functional, equipment or software diversity between the digital I&C functions. Apart frombeing implemented between two hardware platforms, diversity is also introduce within the ProtectionSystem in order to prevent the occurrence of common mode failures, and two diverse means are usedto switch off the control rods power supply in case of a reactor trip demand: trip breakers forinterruption of power to all rods and trip contactors dedicated to every bank of four rods.4.4 Inherent safety provisions, active and passive safety featuresIn any PWR a mix of inherent characteristics, passive and active features is used. For EPR, a balanced andcomprehensive approach was sought with regard to the extent to which inherent safety provision should besized and safety functions should be achieved by passive systems. The selection depend upon theefficiency, reliability, availability and balancing cost and productivity of the various design options.For EPR, inherent characteristics provide additional design margins and extended grace periods foroperator actions thanks to components with large water inventories such as the pressurizer and the steamgenerators.Passive systems have technical assets and they may be seen as an help for communication with the public;however, passive systems do have failure modes; they may need an active triggering, they work well onlyif they are correctly aligned while dormant, they are not always easy to test and they may request specificsystem architecture and layout which make more difficult and/or costly protection against hazards.As any PWR, EPR relies on passive features like accumulators for safety injection, safety valves for overpressureprotection, gravity-driven control rod insertion…EPR designers addressed the potential inclusion

of passive features and many of them were evaluated at the beginning of the conceptual phase [5], butonly few of them were included in the design.Mitigation of Design Basis Condition relies mainly on active safeguard systems which are preferred fortheir proven design, their versatility and their ability to allow operators to keep their hands on the plantduring perturbed situation. Their main weakness is the need of electrical power, it is counteracted, in caseof a station black out, by the redundancy and the diversity of the emergency electrical sources. This allowsthe safety function to be achieved by active systems with a extremely high reliability level.Some passive features were considered valuable to mitigate the consequence of a core melt accident andto preclude significant radioactive releases. Use of passive components and means in the early phasefollowing the core melt allows to implement a simple and robust mitigation strategy which makes properallowance for the plant state in this extreme conditionsShould a core melt occurs, passive hydrogen recombiners avoid a global hydrogen detonation that couldchallenge the containment integrity. The retention, the spreading of the corium and then the flooding andcooling of the spreading area by draining water from the IRWST (In-Containment Refueling WaterStorage Tank) located inside the reactor building, are fully passive at all stages for at least twelve hours.The large heat capacity of the containment building makes not necessary an active heat removal from thecontainment during at least the first twelve hours that follow a core melt, thus providing plenty of time forrecovery.5 OutlookThe safety principles and criteria against which the EPR was developed were intended to result in aninternational standard design that should meet most of the requirements of the Nuclear Safety Authoritiesof western countries.The evolutionary approach followed by the EPR designers led to an exhaustive, graduated and robustdefence based on an optimized mix between largely proven solutions derived from a large operatingexperience and innovative features where needed to meet new requirements. Such a balanced safetyapproach achieves compliance with the harmonized safety requirements. It provides wide acceptabilitywithout the need for significant design variations, thus allowing an international generic design to bedeveloped with the benefits of standardisation. This approach protects against licensing, construction andtechnical risks and their economics impacts.This harmonized safety philosophy makes the EPR a major solution to lead the international nuclearrenaissance which is raising.References[1] EPR development – An evolutionary design processRobert C. Twilley, Jr - Nuclear News April 2004[2] Rapport préliminaire de sûreté de Flamanville 3EDFhttp://www.edf.fr/htm/epr/rps/index.pdf[3] The Technical Guidelines for the Design and Construction of the Next generation of Nuclear Power Plantswith Pressurized Water ReactorsAdopted during the GPR/German experts plenary meetings held on October 19th and 26th 2000http://www.asn.fr/sections/rubriquesprincipales/actualites/notes-d-information/prise-position-du[4] The EPR overall approach for severe accident mitigationFrançois Bouteille and all - Nuclear Engineering and Design 236 (2006)[5] Active or passive Systems ? The EPR ApproachJ.P. Py – Framatome, N. Bonhomme – NPIIAEA Advisory Group Meeting on technical Feasibility and Reliability of Passive Systems21-24 november 1994 – Jülich – Germany

Session 18.1.1:Generation IV rectors

EUROPEAN RESEARCH ON THERMAL HYDRAULICS FORHEAVY LIQUID METAL ADS APPLICATIONSF. ROELOFS 1 , A. CLASS 2 , H. JEANMART 3 , P. SCHUURMANS 4 ,A. CIAMPICHETTI 5 , G. GERBETH 6 , R. STIEGLITZ 2 , C. FAZIO 21 NRG, Westerduinweg 3,1755 LE Petten – Netherlands2 FZK, Hermann-von-Helmholtz-Platz 1,76344 Eggenstein-Leopoldshafen – Germany3 UCL, Louvain la Neuve – Belgium4 SCK•CEN, Boeretang 200, B-2400 Mol – Belgium5 ENEA,40032 Camugnano, Brasimone – Italy6 FZD, P.O.Box 510119, 01314 Dresden – GermanyABSTRACTThe objective of the European 6th framework project EUROTRANS is to demonstrate thetechnical feasibility of transmutation of high level nuclear waste using Accelerator DrivenSystems (ADS). Within this objective the design of a European experimental ADS shoulddemonstrate the technical feasibilities to transmute a sizeable amount of waste and tooperate an ADS safely. This ADS will be a subcritical reactor system having liquid leadbismutheutectic (LBE) as coolant. The liquid LBE is also intended to serve as targetmaterial for the spallation reaction which forms a crucial part to the subcritical reactor core.Since LBE is used as core coolant and spallation material, knowledge of the thermalhydraulic behaviour of LBE is essential. Within the DEMETRA domain of theEUROTRANS project, basic thermal hydraulic studies in order to support the design andsafety analysis of XT-ADS components and the development of measurement techniqueshave been started.1. IntroductionThe objective of the European 6th framework project EUROTRANS [9], sponsored by the EuropeanCommission, is to demonstrate the technical feasibility of transmutation of high level nuclear wasteusing Accelerator Driven Systems (ADS). Within this objective, the design of a Europeanexperimental ADS (XT-ADS) should demonstrate the technicalfeasibilities to transmute a sizeable amount of waste and to operatean ADS safely. Besides that, the conceptual design of a EuropeanFacility for Industrial Transmutation (EFIT) is foreseen. Bothsystems will be subcritical reactors having liquid lead-bismutheutectic (LBE) and lead as coolant, respectively. This liquid metalis also intended to serve as target material for the spallationreaction which forms a crucial part to the subcritical reactor core.Since liquid metal is used as core coolant and spallation material,knowledge of the thermal hydraulic behaviour of liquid metal isessential. Due to the functional similarity between the XT-ADS andthe so-called MYRRHA Draft 2 concept (shown in figure 1) asdeveloped by the Belgian nuclear research institute SCK•CEN (AïtAbderrahim, 2005 [1]), this design was chosen as a starting pointfor the design of the XT-ADS.Within the DEMETRA domain (Fazio et al., 2006 [7]) of theEUROTRANS project, basic thermal hydraulic studies in order tosupport the design and safety analysis of XT-ADS components andthe development of measurement techniques have been started. Inparticular, the work focuses on:• Characterisation of the free surface flow for the windowlessFig. 1: Overall configuration ofMYRRHA Draft 2 which servesas basis for the XT-ADS.

spallation target design.• The interaction of LBE with water as secondary coolant.• The development of measurement techniques for heavy liquid metal (HLM) flows.The work on the characterisation of the free surface flow for the windowless spallation target isdirectly linked to the design of the windowless spallation target for the XT-ADS within the DESIGNdomain of the EUROTRANS project. The interaction between LBE and water as secondary coolanthas an impact on the design selection and safety considerations of the heat exchanger of the ADS.These studies are also used to prepare a large scale integral experiment which is foreseen within theDEMETRA domain in the CIRCE facility at ENEA (Fazio et al., 2006 [7]). Since a large number ofEuropean lead/LBE experimental facilities are involved, this work is also closely linked to theEuropean Commission Integrated Infrastructure Initiative VELLA (Virtual European LeadLAboratory [4]). Furthermore, as the lead and LBE technologies developed within EUROTRANS arealso applicable to a lead cooled fast reactor (LFR), this work is strongly related to the EuropeanCommission Specific Targeted Research Project ELSY (European Lead-cooled System [10]).2. Characterisation of the Free Surface Flow in the Windowless Target2.1 Windowless TargetAs outlined before, due to the functional similarity between the XT-ADS and the MYRRHA Draft 2concept, the latter has served as starting point for the design of the XT-ADS. Therefore, also thedesign of the spallation target of the XT-ADS (Schuurmans et al., 2006 [14]) is based on thewindowless spallation target design of the MYRRHA Draft2 concept. The limited space available for the externalneutron source in the core of the XT-ADS and the highproton current, lead to very high proton beam densities. Atpresent, no structural material is expected to withstandsuch extreme conditions at the operational temperaturesforeseen for the XT-ADS during a reasonable lifetime ofthe spallation target of at least one year. Therefore, an LBEwindowless spallation target is chosen in which there isdirect contact between the proton beam from theaccelerator and an LBE free surface flow. This results in achallenging task for the design of the spallation target. Thedesign of the target nozzle has to be such that an LBE freesurface flow is created within the geometrical constraintsimposed by the compact sub-critical core which isadequate to remove the heat deposited by the proton beam.Furthermore, the design has to be compatible with theFig. 2: Schematic view of the verticalconfluent flow design of the MYRRHADraft 2 design of a windowless spallationtarget.vacuum requirements of the beam transport system. Theseconstraints lead to a design of the windowless spallationtarget with a vertical confluent flow as presented in figure2.2.2 Numerical Model DevelopmentAs no experiment can demonstrate the ability to transport the deposited heat in a windowlessspallation target adequately, validated numerical methods are required. For this purpose,computational fluid dynamics (CFD) simulation methods are the most appropriate to capture thespecific three-dimensional local effects of the LBE free surface including the heat deposition. Thisrequires sufficiently accurate free surface modelling, predicting a unique (sharp) interface betweenLBE and beam vacuum in combination with adequate turbulence modelling. In the European 5thframework project ASCHLIM (Arien et al., 2004 [2]), it was demonstrated that sufficiently accurateCFD modelling of such free surface targets was not possible with the state-of-the-art methodsavailable at that time. This is confirmed in other papers concerning this subject, e.g. Van Tichelen etal., 2000 [15], Van Tichelen et al., 2003 [17] and Fazio et al., 2005 [6]. Within the EUROTRANSproject, the development of CFD methods for the simulation of the removal of deposited heat in theLBE windowless target has been envisaged. Different methods are assessed by NRG, FZK, and AAA

and qualitatively compared to existing real size water flow experiments performed at UCL (VanTichelen et al., 2001 [16]), mercury experiments performed at IPUL (Van Tichelen et al., 2001 [16]),and LBE flow experiments at FZK (Schulenberg, 2005 [13]). Table 1 summarises the differentnumerical methods assessed by the different partners.Numerical Method CFD Code InstituteVolume of Fluid (VOF) STAR-CD FZKVOF + Cavitation Module STAR-CD FZKEuler-Euler CFX10 NRGMoving Mesh Algorithm (MMA) STAR-CD AAATab. 1: Evaluated numerical modelsFirst assessments have been made for the isothermal situation, i.e. without taking into account the heatdeposition of the proton beam (Batta & Class, 2007 [3], and Roelofs et al., 2007 [12]). It is concludedthat application of the VOF model in combination with the cavitation module in STAR-CD andapplication of the Euler-Euler model in CFX10 lead to promising results, although both models stillrequire improvements. Furthermore, it is concluded that the VOF model without cavitation model doesnot lead to realistic results. The MMA method is still under evaluation. First results are expected bythe end of 2007.2.3 Experimental CampaignAn experimental campaign has started for the improvement andvalidation of the developed numerical models. This campaign foreseesexperiments in a water loop of UCL, see figure 3, and in a leadbismuthloop in the KALLA laboratory of FZK. First experiments inthe water loop at UCL have already been performed using theMYRRHA Draft 2 spallation target design assessing the influence ofadding a mild swirl to the annular feeder flow on the behaviour of thefree surface. Preliminary experiments have shown that adding a swirlof about 10% leads to an unacceptable vacuum core vortex in thecentral downcomer of the spallation loop. This confirms theconclusions from numerical simulations performed by FZK and NRG.3. LBE-Water Interaction3.1 Experimental CampaignFig. 3: Water loop at UCL908070Pressure [bar]60504030201000 500 1000 1500 2000 2500 3000 3500 4000Time [ms]Fig. 4: LIFUS 5 test facility and pressure evolution in the reaction vessel (S1 - red line) and the buffertank (S5 - blue line) during campaign 1The XT-ADS and EFIT design foresee the presence of heat exchangers or steam generator modulesplaced inside the main vessel. This allows direct contact between LBE as primary coolant and water assecondary coolant in the case of a tube rupture. Since the probability of a tube rupture cannot beneglected, the consequences of such an accident have to be assessed. The experimental campaigns,which are performed in the LIFUS-5 facility of ENEA in Italy, aim at assessing the physical effectsand possible consequences related to the interaction of LBE and water in representative conditions.

For this purpose, a steam generator mock-up is placed in a reaction vessel (S1) filled with LBE. Wateris injected near a steam generator mock-up into the LBE. Fast pressure transducers and thermocouplesat various locations register the pressure and temperature evolution during the experiment. A firstexperimental campaign aimed at obtaining first of a kind data of LBE-water interaction, has beenperformed successfully injecting pressurised water at 70 bar in the reaction vessel of LIFUS-5containing LBE at 350 °C (Ciampichetti, 2007 [5]). Figure 4 shows the LIFUS-5 facility and thepressure evolution detected in the reaction tank S1 and buffer tank S5 during Test n.1. For the appliedconditions, a pressure increase of about 10 bar above the injection pressure (70 bar) was observed andno steam explosion occurred due to the fast pressurisation of the system.3.2 Numerical ProgramThe experimental data are also used for the validation of the SIMMER III code by ENEA/UNIPI andCEA. The SIMMER III code is a general fluid dynamics code coupled with a space-time and energydependentneutron transport kinetics model (Tobita et al., 2006 [18]). First simulations with a twodimensionalmodel performed by ENEA/UNIPI show a reasonable comparison between the simulationresults and the experimental values. The pressure increase above the injection pressure was predictedcorrectly. However, the exact value of the pressure peak and the time evolution give reason forimprovement of the numerical model by extending the model to three dimensions and by a moreaccurate geometrical representation of the steam generator mock-up.4. Development of Measurement TechniquesMeasurement techniques are developed for thermal-hydraulics experiments and for operationaltechniques in the XT-ADS and EFIT reactors. These techniques are tested within the laboratories ofFZD, FZK, and SCK•CEN. The focus is on local velocity meters, integral contactless flow meters, andfree surface level sensors. Two types of flow meters and two types of free surface level sensors will bedescribed hereafter.4.1 Flow metersContactless electromagnetic flow meters (EMFM) based on different principles are developed inparallel. One EMFM is based on the principle of phase shift and is developed by FZD, see Priede et al.(2006) [11]. This EMFM is validated against a commercial flow rate sensor as well as local velocitymeasurements using Ultra Doppler Velocimetry (UDV) in a GaInSn-loop at FZD. Furthermore, theEMFM measuring device is made resistant against temperatures up to 800°C. Figure 5 shows twodeveloped devices which are ready for further testing in existing liquid metal loops. The device on theleft is able to measure flow rates in channels up to 85 mm. The other device can be attached to achannel using a clamb. The latter system can be used in channels up to 34 mm and temperatures up to800°C.Another EMFM under development by FZK is based on the principle of dragging magnetic field lines.This flow meter is able to detect the flow direction. Besides that, a self calibrating method isdeveloped for this type of flow meter. First successful tests have been performed in the KALLAlaboratory at FZK.Fig. 5: Developed flow meters based on phase shift4.2 Free Surface Measuring TechniquesFree surface measuring techniques are required for experimental as well as operational purposes.Concerning the experimental purposes, the measuring technique requires accurate measurement of the

free surface shape and position. For this purpose, FZK is developing a non-invasive detection methodbased on the double layer projection technique (DLP). The proof of principle is demonstrated on astatic and a rotating mirror. Further validation is foreseen on a circular hydraulic jump experiment(Hillenbrand et al., 2007 [8]).Concerning the operational purposes, the measuring technique has to fulfil different requirements.During operation of the XT-ADS, accurate and frequent knowledge about the position of the freesurface is required for reactor and beam control. In combination with the not readily accessiblelocation of the free surface in the core of the reactor this leads to very stringent requirements for thetechnique under development: the distance between sensor and surface is about 10 m, the accuracyshould be lower than 1 mm, and the measuring frequency should be about 1 kHz. SCK•CEN hasselected a time of flight (TOF) technique for this purpose.5. SummaryThis paper summarises the ongoing work performed within the framework of the ‘advanced thermalhydraulicsand measurement techniques’ workpackage of the DEMETRA domain of the Europeanintegrated project EUROTRANS. This work focuses on the characterisation of the free surface flowfor the windowless spallation target design, the interaction of LBE with water as secondary coolant,and the development of measurement techniques for heavy liquid metal (HLM) flows. Mainachievements are:• Development of numerical methods for the simulation of the isothermal windowless target;• Determination of the influence of adding a mild swirl in a windowless target water loop;• Performance of a first of a kind LBE-water interaction experiment which shows a pressureincrease above the injection pressure and no occurence of a steam explosion;• Reasonable results for the simulation of the first campaign of the LBE-water interactionexperiments using a two-dimensional model in the SIMMER III code;• Development of EMFM devices for the contactless measuring of HLM flow rates;• Development of DLP free surface measuring technique for determination of free surface shapeand position in experiments.AcknowledgmentsThe authors would like to thank all participants (AAA, ANS, CEA, CRS4, ENEA, FZD, FZR, NRG,SCK•CEN, and UCL) involved in the workpackage ‘advanced thermal-hydraulics and measurementtechniques’ of the EUROTRANS project. The work described in this paper was carried out in andsupported by the FP6 EC Integrated Project EUROTRANS No. FI6W-CT-2004- 516520.References[1] Aït Abderrahim H. et al., 2005. MYRRHA Pre-Design File - Draft 2. SCK-CEN report R-4234,Mol, Belgium.[2] Arien et al., 2004. Assessment of Computational Fluid Dynamic codes for Heavy Liquid Metals -ASCHLIM, EC-Con. FIKW-CT-2001-80121-Final Rep.[3] Batta A., Class A., 2007. Numerical Investigations on Geometrical Designs of the Windowless XTADS Spallation Target. ICAPP07, Nice, France.[4] Benamati G. et al., 2006. Virtual European Lead Laboratory, VELLA. EC project No. FI6W-036469.[5] Ciampichetti A., 2007. LBE/water interaction in sub-critical reactors: first experimental andmodelling results. IV Workshop on Materials and for HLM-cooled Reactors and RelatedTechnologies, Rome, Italy.[6] Fazio C., Alamo A., Almazouzi A., Gomez-Briceno D., Gröschel F., Roelofs F., Turroni P.,Knebel J., 2005. Assessment of Reference Structural Materials, Heavy Liquid Metals, andThermal-hydraulics for European Waste Transmutation ADS, Global 2005, Tsukuba, Japan.[7] Fazio C., Alamo A., Henry J., Almazouzi A., Gomez-Briceno D., Soler L.,Vogt J-B., Groeschel F.,Roelofs F., Turroni P., Stieglitz R., and Knebel J., 2006. European Research on Heavy LiquidMetal Technology for Advanced Reactor Systems. 15th PBNC, Sydney, Australia.

[8] Hillenbrand M., Schmidt T., Stieglitz R., Class A., Piecha H., Neitzel G.P., 2007. Measurementand Computation of Free Surface Flows of Liquid Metals. Jahrestagung 2007, Karlsruhe,Germany.[9] Knebel J. et al., 2005. Integrated Project EUROpean Research Programme for the TRANSmutationof High Level Nuclear Waste in an Accelerator Driven System, EUROTRANS. EC project No.FI6W-CT-2004-516520.[10] Locatelli G. et al., 2006. Specific Targeted Project European Lead-cooled System, ELSY. ECproject No. FI6W-036439.[11] Priede J., Buchenau D., Gerbeth G., 2006. A contact-less electromagnetic induction flowmeterbased on phase shift measurements. EPM, Sendai, Oct. 23-27, 2006, Proceedings pp. 735-740.[12] Roelofs F., Siccama N.B., Willemsen S., 2007. Development of an Euler-Euler Two-Phase Modelfor Application in the Windowless XT-ADS Spallation Target Design. ICAPP07, Nice, France.[13] Schulenberg T, Cheng X., Stieglitz R., 2005. Thermal-hydraulics of Lead-Bismuth forAccelerator Driven Systems. NURETH11, Avignon, France.[14] Schuurmans P., Tichelen K. van, Dierckx M., Aït Abderrahim H., Guertin A., Kirchner T.,Cadiou A., Buhour J.M., Stieglitz R., Coors D., Mansani L., Roelofs F., 2006. Design andSupporting R&D of the XT-ADS Spallation Target. IEM on Actinide and Fission ProductPartitioning and Transmutation, Nîmes, France.[15] Tichelen K. van, Kupschus P., Aït Abderrahim H., Seynhaeve J.-M., Winckelmans G., JeanmartH., 2000. MYRRHA: Design of a Windowless Spallation Target for a Prototype AcceleratorDriven System, ICENES-2000, Petten, Netherlands.[16] Tichelen K. van, Kupschus P., Aït Abderrahim H., Klujkin A., Platacis E., 2001. MYRRHA:Design and Verification Experiments for the Windowless Spallation Target of the ADS PrototypeMYRRHA. AccApp01, Reno, Nevada, USA.[17] Tichelen K. van, Kupschus P., Roelofs F., Dierckx M., Aït Abderrahim H., 2003. Free SurfaceFluid Dynamics Code Adaptation by Experimental Evidence for the MYRRHA Spallaton Target.IAEA TM on Theoretical and Exp. Studies of HLM Thermal Hydraulics, Karlsruhe, Germany.[18] Tobita Y., Kondo SA., Yamano H., Morita K., Maschek W., Coste P., Cadiou T., 2006. TheDevelopment of SIMMER-III, an Advanced Computer Program for LMFR Safety Analysis, andits Application to Sodium Experiments. Nuclear Tecnology, volume 153, number 3, p.p. 245-255.

ELSY : NEUTRONIC DESIGN APPROACHC. ARTIOLI, M. SAROTTOENEA FPN-FISNUCVia Martiri di Monte Sole 4, 40129 Bologna, Italycarlo.artioli@bologna.enea.it, massimo.sarottto@bologna.enea.itE. MALAMBU , V. SOBOLEVSCK-CEN, Institute for Advanced Nuclear SystemsBE-2400 Mol, Belgium,VSOBOLEV@SCKCEN.BES. MASSARAEdF R&D1, Av. Général de Gaulle, 92141 Clamart, Francesimone.massara@edf.frM. E. RICOTTICIRTEN-Politecnico di Milanovia Ponzio 34/3, 20133 Milano, Italymarco.ricotti@polimi.itABSTRACTELSY, European Lead-cooled System, aims to fulfil European Requirements of MinorActinides burning and the GEN-IV strategic goals. It is a 600 MWe lead-cooled reactor (seea second ELSY paper in this ENC-2007). Two different core types are being preliminarilyconsidered: the core made up of hexagonal wrapped assemblies, as used in sodium-cooledreactors, and wrapperless square assemblies, as common in PWR’s. ELSY is conceived as“adiabatic reactor”, i.e. it shall feature a unitary Conversion Factor and burn its selfgeneratedMA. Consideration is also being given to the transmutation of a larger amount ofMA, to address the issue of the MA legacy. A key-point of the core design approach is thesmall delta-T between coolant mean outlet temperature (480 °C) and allowable claddingtemperature (now about 560°C), that requires a rather flat radial power distribution. Thisimplies a core sparsely fitted with Control Rods, but offering the required worth.1. IntroductionThis paper deals with the preliminary neutronic design approach of ELSY (European Lead-cooledSYstem, see a second ELSY paper in this ENC 2007 Conference): a MOX fueled 600 MWe pool-typefast reactor, aiming to fulfil the GEN-IV strategic goals. Two different schemes of core are hereconsidered: the first one made by conventional wrapped assemblies in hexagonal lattice and a secondone made of wrapperless assemblies in square lattice with pins arranged in square bundle as well.Beyond the self-evident qualitative differences rising up from the presence or not of the hexcan steeland from the quite different thermohydraulic conditions, it is mandatory to explore, evaluate andcompare how much all these differences are impacting on the most important reactor parameters.For this aim a number of parameters are assumed as a common basis, while the parameters able toexploit and show the differences in the two concepts are assumed as freedom degrees. The comparisonis carried out paying attention, and weighting, to the extension in fulfilling the goals of sustainability,economics, safety and proliferation resistance.

Both the concepts assume in common the power plant, the type of fuel, the fuel residence time, the BUpeak, the cladding max temperature and damage dose, the coolant temperatures, pointing to an unitaryConversion Factor.On the other hand each concept optimizes, according to its own design, the pin diameter, the activeheight, the core diameter, the fuel enrichment, the volumetric fractions, whether to use or not the axialblanket for reaching the required Conversion Factor, the number and position of the Absorber Rodsfor the reactivity compensation, control and safety.Minor Actinides burning is a key point. The ELSY should be an “adiabatic reactor” in the sense that itproduces its own new fuel (Conversion Factor = 1+ reprocessing losses) and burns its own selfproducedMA, without any material exchange with the environment, except loading natural ordepleted Uranium and unloading fission products. Nevertheless to cope with the MA legacy it will beable to burn in addition even MA coming from other nuclear plants.2. Specifications and aimed performancesThe main aimed specifications of the ELSY core to kept in common in the two schemes are collectedin the table 1 (see other paper of L. Cinotti et al.).ELSY PLANT AREATENTATIVE PARAMETERSPowerAbout 600 MWe (1500 MWth)Thermal efficiency About 40 %Primary coolantPure leadPrimary systemPool type, compactPrimary coolant circulation (at power) ForcedPrimary coolant circulation for DHR Natural circulation + Pony motorsCore inlet temperature ~ 400°CCore outlet temperature ~ 480°CFuelMOX with assessment also of behaviourof nitrides and dispersed minor actinidesFuel handlingSearch for innovative solutionsMain vesselAustenitic ss, hanging, short-heighte Safety VesselAnchored to the reactor pitSteam GeneratorsIntegrated in the main vesselSecondary cycleWater-supercritical steamPrimary PumpsMechanical, in the hot collectorInternalsAs much as possible removable(objective: all removable)Hot collectorSmall-volume above the coreCold collectorAnnular, outside the core, free level higher thanfree level of hot collectorDHR coolersImmersed in the cold collectorSeismic design2D isolators supporting the main vesselTab. 1 ELSY main specifications2.1 FuelAt first step, an oxide fuel without MA is proposed to be assumed: UO 2 -PuO 2 MOX with a maximumallowed enrichment of 35 wt % of reactor grade Pu in heavy metal and with 95 % of theoreticaldensity (TD). Nitride fuel will be considered later as an option.At the second step, oxide fuel containing minor actinides (MA: Am, Cm and Np) up to about 5 at.%on HM will be considered.At this stage of the ELSY conceptual design, the actinide isotopic vectors chosen in IP EUROTRANS[4] are proposed to use for minor actinides. Theses vectors were obtained in the result of mixing ofMA coming from the spent UO 2 fuel (90 %) and the spent MOX (10%) of a typical PWR unloaded atthe burnup of 45 MWd/kgHM, then cooled down and kept in storage for a period of 30 years.Plutonium is extracted from the same spent UO 2 but with the storage period of 15 years [5]. Depleted

U remaining in production of UO 2 fuel for LWRs or U obtained in reprocessing of LWR spent fuel isusually used for industrial MOX production.The isotopic vectors of the Pu is in Table 1.The relativeatomic content of neptunium, americium and curium in total MA can be expressed as follows:Np:Am:Cm = 3.88 : 91.82 : 4.30 at.%.Reactor grade PLUTONIUMIsotope Molar mass Content Contentg/mol wt.% at.%238 Pu 238,0496 2,332 2,348239 Pu 239,0522 56,873 57,015240 Pu 240,0538 26,997 26,951241 Pu 241,0569 6,104 6,069242 Pu 242,0587 7,693 7,616Pu 239,6493Tab. 2. Isotopic vector of plutonium.In order to assure a high efficiency of the reactor operation and the requirements of non-proliferation,a long fuel life-time in the core should be aimed. In ELSY, where liquid lead is used as coolant, thecorrosion of the pin cladding will play a major role. The existing laboratory studies on compatibility ofdifferent structure materials with molten lead show that some of them can resist a corrosion attack ofthe liquid metal flow with a velocity of 1.8-2.0 m/s and temperature of 560 °C during 12000 hoursunder oxygen control conditions. A small thickness of the formed corrosion layer (4-5 microns) allowsmaking a prognosis that their operation time at this temperature can be extended to 50 000 hours(more than 5 years) [1]. However, there is no similar experience under in-pile conditions. Newadvancements in the development of the corrosion resistant steels and protective layers (i.e. GESAtreatment) indicate that longer operation periods can be envisaged in a near future [2].Thus a fuel life-time in the core can be assumed to be 5 years as a realistic option and, tentatively, 10years as a futuristic option.Of course the residence time of the fuel in core is ruled even by the allowed fuel burnup and allowedcladding damage.2.2 CladdingThe choice of cladding material is of critical importance both from economic and safety viewpoints.Ferritic-martensitic steels (FMS) and austenitic steels (AUS) are within the best candidates for thecladding material. FMS show a lower swelling rate and embrittlement under irradiation at T > 350 °Cand higher resistance to dissolution in the oxygen-free Pb and Pb-Bi(e), compared to austenitic steels.However, they have a higher corrosion rate in the presence of oxygen [7].The existing experience of LMFR operation and the performed irradiation studies of the claddingmaterials for fast spectrum reactors demonstrate that optimized austenitic steel AIM1 (15-15 Ti modSi) can withstand typical LMFBR operation conditions (sodium, 400-550 °C) up to the peak dose of115 dpa [3]. At these temperatures, ferritic martensitic steels with 8-12 % chromium (such as T91,EM-10, HT9, F82H, …) are even more resistant (~200 dpa).At this stage, it is proposed to choose FMS T91 as the first candidate for cladding material, taking intoaccount its better irradiation resistance and ongoing R&D on technology of its protection againstcorrosion. The well-known austenitic steels AIM1, AISI 316 L and few Russian steels (EP823, EI852)are kept as a backup solution.

The cladding damage and the fuel burnup are correlated. In the neutronic modeling of the ADSMYRRHA, loaded with 30 % Pu MOX and cooled by liquid Pb-Bi eutectic, the mean damage of 25dpa (the peak damage of 32 dpa) was obtained for the T91 cladding in the hottest rod at the averageburnup of 28.68 MWd/kg HM in the hottest assembly (30.06 MWd/kg HM in the hottest pin) [8]. So,for an approximate estimation of the cladding damage rate, one can use a value of 0.83 dpa per 1MWD/kgHM of fuel burnup. Then the aimed average burnup of 10 at % hma will result in a damagedose of about 88 dpa (validity of this simple relationship for ELSY has still to be confirmed).In order to keep freedom in selection of cladding material, it is proposed to limit the cladding damagedose to 100 dpa, taking into account that synergy can exist between corrosion and irradiation effects.The optimistic limit can be 200 dpa, assumed to be reached in a near future with FMS and ODS steels.2.3 Other core specificationsAvailable experience with Pb-Bi(e) and Pb coolants shows that the bulk velocity has to be limited to 2m/s in order to avoid erosion problems during long-term operation [9].A supplementary parameter, i.e. a core Conversion Factor of about 1, has o be introduced in the corepre-design specifications to achieve the sustainability. Moreover this condition is important forreduction of the number of intermediate reactor shutdowns needed for the core reconfiguration. Itseems reasonable to do it not earlier than after every year of operation.Following a passive safety approach, it is proposed that RDH will be removed by natural thermalconvection; an effective height of 4 m between the heat exchanger and the core is foreseen.Table 3 summarizes the specifications, based on arguments presented above, for the preliminarydesign of the ELSY core.CharacteristicValuereal futureThermal power 1500 MWCore conversion factor ~1Minimum sub-cycle duration 1 5 yFuel residence-time (aimed) 5 10 yHottest assembly discharge burnup (aimed) 100 150 MWd/kgHMMaximum allowed Pb bulk-velocity ≤ 2.0 m/sMaximum allowed clad temperature 550 620 °CPeak clad damage 100 200 dpaResidual heat removal modenatural convection+RVACSHX-core levels difference 4 mTab. 3 Other core specifications3. Hexagonal wrapped assemblies core3.1 FuelFor the Hexagonal Wrapped scheme the following preliminary parameters have been precalculated.They need to be verified and optimized, along the core characterization, before making the selection vsthe Wrapperless Square concept.Estimated characteristicValuePellet outer diameter 8,88 mmCentral hole diameter 2,0 mmPellet height (postulated) 12,5 mmMOX density (STP) 10550 kg/m³Maximum allowed linear heating rate (EOC) 390 W/cmMaximum allowed fuel power density (EOC) 630 W/cm³Tab. 4. Estimated and pre-selected parameters of fuel pellet

Fuel rod : Variant 1Pellet diameter 8,88 mmHole diameter 2,00 mmPellet height 12,50 mmClad inner diameter 9,18 mmClad outer diameter 10,58 mmFuel column height 900 mmSupplementary breeding/burning segments 300 mmInsulation pellets 10+10 mmGas plenum length 960+240 mmCaps 50+50 mmFuel rod length 2520 mmTab. 5 Estimated and pre-selected parameters of fuel rodThe first estimate of the pitch of a rod bundle (defined as the “centre-to-centre” distance between theneighbour fuel rods) was obtained from the thermal balance. Then it was optimised in order to respecttwo other conditions: the coolant bulk velocity < 2 m/s and the pressure drop on the core < 0.12 MPa.Finally, the rod pitch of l pitch = 15.5 mm (x pitch = 1.465) has been fixed. The number of the rods in thehexagonal bundle was chosen to be 169: 168 fuel rods and 1 central bar for assembly manipulation.Then the inner plate-to-plate width of 203.0 mm was obtained for the assembly hexagonal wrapper. Awall thickness of 4.0 mm and a clearance of 5.0 mm between the neighbour assemblies were fixedtaking into account the experience of the EFIT pre-design [10]. A schematic view of the radial crosssectionof the proposed fuel assembly is presented in Figure 1 below, while Table 6 hereaftersummarises the main geometrical parameters of this variant.ELSY assembly: 168 fuel rods + 1 central tube15.54.0211.02.5Fig. 1. Assembly radial cross-section at midplane

Assembly (hexagon): Variant 1Fuel rod pitch (center-to-center) 15.50 mmRod pitch ratio 1.465Number of fuel rods in assembly168 + 1 dummy rod(rod-row number) 15Wrapper inner width 203.0 mmWraper wall thickness 4.0 mmWrapper width 211.0 mmClearence between assemblies 5.0 mmAssembly pitch (center-to-center) 216.0 mmTotal assemnly length (EFIT based) 3800 mmFuel mass in assembly 118.6 kgTab. 6 Hexagonal wrapped fuel assembly dataOutput sectionØ19521128070Hex. wrapper15.52112520Fuel bundle4.0Central rodØ122113800Joint sectionØ150Ø170900300 160Input tubeFootFig. 2 Axial schematics cross-section of the fuel assembly.

At the initial pre-design stage, it was proposed to base the axial schematics of the hexagonal fuelassembly on the mechanical design of the EFIT fuel assembly done by Ansaldo Nucleare [10], butmaking it shorter and simpler where possible. A tentative axial schematic of the proposed assembly ispresented in Figure 2.3.2 Control rodsA reactivity compensation element has about the same design as fuel assembly, except the fuel rodbundle which is replaced by the bundle of 127 rods with a diameter of 15.76 mm. Each of the rods isfilled with B 4 C pellets of Ø 14 mm over the height of 1200 mm. Other elements and axial schematic ofthe absorber rod is the same as that of the fuel rods (Fig. 2). Main measures of the absorber assemblyare given in Table 7.Absorber rod :Pellet diameter 14.00 mmHole diameternoPellet height 20.00 mmClad inner diameter 14.36 mmClad outer diameter 15.76 mmColumn height 1300 mmInsulation pellets 10+10 mmGas plenum and spring chamber lengths 560+240 mmTop and bottom plugs 50+50 mmAbsorber rod length 2520 mmAbsorber assembly (hexagon):Fuel rod pitch (center-to-center) 17.70 mmRod pitch ratio 1.123Number of absorber rods in assembly 126+1Wrapper inner width 203.0 mmWrapper wall thickness 4.0 mmWrapper width 211.0 mmClearance between assemblies 5.0 mmAssembly pitch (center-to-center) 216.0 mmTotal assembly length (EFIT based) 3800 mmTable 7 Main geometrical measures of the absorber element3.3 CoreA variant of the ELSY core configuration with the fuel and absorber elements described above ispresented in Fig. 3. The main parameters of this core, estimated on the basis of simplified thermal andhydraulic calculations, are presented in Table 8.Neutronic modeling followed by thermohydraulic and thermomechanical calculations has to beperformed to estimate the key performances of this core. The results of these calculations will be usedfor optimization of the designs of the fuel rod, assembly and core.

ELSY core variant)Control rodsInner fuel zoneOuter fuel zoneReflector zoneSafety rodsCompensation rodsCore barrelFig. 3 A variant of configuration of the hexagonal wrapped ELSY core.Normal regimeELSY-600Core:Thermal power 1500 MWCore diameter 4,54 mCore height (driver+special) 1,20 mH/D 0,26Fuel mass 38,24 tNumber of fuel assemblies (expected) 323Mean assembly power 4,65 MWMean rod power 27,7 kWRadial power form-factor (expected) 1,30Axial power form-factor (expected) 1,30Colant inlet temperature 400 °CCoolant average outlet temperature 480 °CLead mass flow-rate (maximum) 128,3 t/sLead maximum bulk-velocity 1,90 m/sPressure drop on core 0,12 MPaClad maximum temperture 540 °CTab. 8 Estimated parameters of the ELSY-600 core

4. Square Wrapperless assemblies coreThe main aim of this first approach is to verify the Control Rods (CR) worth, the Conversion Factor(CF) and the Minor Actinide burning or build-up attitude.4.1 Fuel elementAt this step no specific design of the fuel element has been done yet. Nevertheless the data have beenderived from the EFIT pre-design [10], in such a way to be confident to respect the cladding limittemperature with a Pb coolant velocity of some 1 m/s. The volumetric fractions are tuned on anaverage linear power rating of 220 W/cm. The related pressure drop through the core is expected to beof some 0.1 MPa.The postulated wrapperless fuel element is sketched in Figure 4. It contains 285 fuel pins and 5 solidsteel pins for structural needsFig. 5. Cross section of the wrapperless fuel element4.2 CoreThe arrangement of the core has been driven from the important key point of the sustainability, i.e. aunitary Conversion Factor, possibly without blanket (for proliferation resistance). The requirement onthe Conversion Factor implies a unique ratio between Pu and U in the fuel, i. e. the enrichment. Sofixed dimension and composition of the assembly, a suitable core has been laid down to reach therequired reactivity.The core is made by 259 square wrapperless fuel element (fig. 6), divided in 3 zones with differentenrichments (whose average accounts for a unitary CF) to achieve a rather flat radial powerdistribution.Taking into account two shielding rings the diameter of the barrel has to be about 5.4 m.

4.2.1 Control RodsSome attempts of placing control rods surrounding the active core, to avoid spoiling the power radialdistribution, have not shown a sufficient worth. Placing only a huge control rod in a devoted locationat the centre would account for some 1000 pcm, while surrounding completely the active core by asmany as 60 Control Rods would account for not more than 3000 pcm.So at this step an overall of 12 Control Rods are foreseen, here not yet distinguished as their functions.The next sketch in Figure 6 shows the cross section of the wrapperless square core, the threeenrichment zones and the locations of the control rods.Fig. 6. Wrapperless square ELSY core4.3 Control Rods worthThe Control Rods worth has been calculated in a cylindrical schematization. The CRs have been represented ascircular ring, whose composition is the average of the whole hexagonal ring where they are (12 CRs + 30 Fuelelements). To have an idea of some compensation margin, 3 positions have been calculated: CRs out, CRsentirely inserted, and CRs inserted 30 cm. The overall worth of the 12 CRs is rather high: about 9000 pcm.The next figure 7 shows the calculation scheme and the geometrical data of the cylindrized core.

Fig. 7. Core and Control Rods calculation schemeSince the peripheral surrounding set of 60 CRs has shown a poor effectiveness, a different surroundinghas been tried. Taking profit from the small height of the core, some absorber has been placed over thefuel element heads (placing the plenum in the lower part). Its effect has been proved to be notnegligible: using volume fractions in such a way to allow the coolant flow, a worth of some 3000 pcmhas been calculated. Figure 8

Fig. 8 Effectiveness of absorber placed over the fuel element heads4.4 Cycle and breeding.No specific cycle calculations have been done to this step, except the evaluation of the loss ofreactivity in the time. It is evaluated as about 1000 pcm/year. The Conversion Factor, simplyexpressed as ratio between odd isotopes of the Pu, results 0.99.Fig. 9 Reactivity loss during the cycle

4.5 Mass balancesThe total core inventory is 37 ton of fuel, 30 ton of depleted U and 7 ton of Pu. The sustainabilty targetis quite well reached: in fact the Pu mass does not change significantly (-3.14 kg/Twh th ), while thelarge part of the energy is coming, both directly and indirectly, from the fission of the U (-40.96kg/TWh th ).4.5.1 Minor ActinidesSince the core studied has been loaded without any MA, they have been produced during the cycle.They have an equilibrium concentration in the fuel matrix and, more or less, they point toward thisconcentration by an exponential law. From the results reported in figure 10, it is possible to deduce theexponential law, the equilibrium amount and the time constant. The MA equilibrium content is about210 Kg in the whole core, i.e. 0.57% in the heavy metal or 3% if referred to the only Plutonium. Thetime constant of the exponential is 5-6 years, that accounts for some 30 years to reach “naturally” theequilibrium.In any case it means that having some 0.5-0.6% of MA, ELSY acts as an “adiabatic reactor”, withoutany exchange with the environment except for the Uranium loaded and the Fission Product unloadedfor each refueling.At this equilibrium concentration a worsening is expected for the Doppler effect by 5%, a worseningfor the Void effect by 1-2 % and a reduction of the Delayed neutrons yield by 1% [11],Should the MA content greater, for example 2% in hm, ELSY would show a net burning of MAcoming from other nuclear plants, in addition to its self produced. For this content in MA the safetyparameters worsening would be 20% for the Doppler effect, 5% for the Void effect and a reduction of3% for the Delayed neutrons yield [11].Fig. 10. Mass balances in the cycle: Uranium, Plutonium, Minor Actinides

5. ConclusionsEven if the studies are still in the first stage and is not yet selected the most promising variant,wrapped hexagonal vs wrapperless square, it is clear how ELSY can fulfil the requirements ofsustainability and proliferation resistance (for other requirements see other ELSY paper in this ENC2007 Conference, L. Cinotti et al.).Quite interesting is the capability of reaching the unitary Conversion Factor without using neitherradial blankets nor axial (at least in the wrapperless square variant), that increases essentially theproliferation resistance.The possibility of burning its own MA without a significant worsening of the safety parameters(worsening at equilibrium: Doppler by 5%, Void by 1-2%, beta by less of 1%) makes ELSY a verypromising “Adiabatic Reactor”.Moreover ELSY can acts as a net burner receiving in addition other MA to be burnt.ACKNOWLEDGMENTSThe authors thank the Partners of the IP-EUROTRANS project for their fruitful contribution to theproject. Special thanks to the European Commission for the financial support trough the FP5 and FP6programmes.

References1. A Technology Roadmap for Generation IV Nuclear Energy Systems. GIF-002-00, USDOENuclear Energy Research Advisory Committee and the Generation IV International Forum. December2002.2. G. Müller, A. Heinzel, J. Konys, G. Schmacher, A. Weisenburger, F. Zimmermann, V. Engelgo,A. Rusanov, V. Markov, Results of steel corrosion tests in flowing liquid Pb/Bi at 420-600 °C after200 h, J. Nucl. Mater., 301 (2002) 40-46.3. Le Combustible Nucleaire des Reacteurs à Eau Sous Pression et des Reacteurs à NeutronsRapides". Ed. By H. Bailly, D. Ménissier, C. Prunier, Comissariat à l'Énergie Atomique. EYROLLES,Paris, 1996. 670 p.4. M. Köhler, H. Noel, A. Green, Overview of the European Fast Reactor (EFR). In: Proc. Int. FastReactor Safety Meeting, Snowbird, Utah, August 12-16, 1990, v.2., pp. 45-56.5. B. Verboomen, (calculation note/private communication), December 2006.6. R. L. Klueh, D. S. Gelles, S. Jitsukawa, A. Kimura, G. R. Odette, B. van der Schaaf, and M.Victoria, Ferritic/martensitic steels – overview of recent results. J. Nucl. Mater. 307-311 (2002) 455-465.7. V. Sobolev, “Compatibility of structure materials of a typical ADS with heavy liquid metalcoolants: Brief review of experimental results”. Report R-3532, SCK·CEN, Mol, May 2001, 55 p.8. E. Malambu, W Haeck, Th. Aoust, N. Messaoudi, G. Van den Eynde, Chapter 2. Sub-criticalCore Neutronics Design Calculations. In: MYRRHA Pre-Design File – Draft 2, Ed. H. AïtAbderrahim, SCK•CEN Report R-4234, Mol, 2005, 45 p.9. I.V. Gorynin, G.P. Karzov, V.G. Markov, V.S. Lavrukhin, V.A. Yakovlev, Structural materialsfor power plants with heavy liquid metals as coolants. In: Proc. Conf. on "Heavy Liquid MetalCoolants in Nuclear Technology". HLMC'98, October 5-9, 1998, Obninsk, Russian Federation.Obninsk: SSC RF - IPPE, 1999, v. 1, p. 120-133.10. C. Artioli, Specification for EFIT core and fuel element. Report Deliverable D1.6, FP6 IPEUROTRANS, Contract N°: FI6W-CT-2004-516520, EURATOM FP6, November 2006, 80 p.11. D. Warin, A. Zaetta, F. Varaine, J.P. Grouiller, S. Pillon, Neutronics, Reactor System and Fuelsfor Transmutation, PHYSOR-2006, ANS Topical Meeting on Reactor Physics

FRENCH PROGRAM TOWARDS A GENERATION IVSODIUM COOLED FAST REACTORP. MARTIN 1 , J. ROUAULT 21 CEA/Cadarache, DEN, 13108 SAINT PAUL LEZ DURANCE, France2CEA/Saclay, DEN/DDIN, 91191 GIF SUR YVETTE, FranceJ-P. SERPANTIEAREVA NP, 10 rue Juliette Récamier, 69456 LYON, FranceD. VERWAERDEEDF R&D, 1 av. du Général de Gaulle, BP 408, 92131 CLAMART, FranceABSTRACTSodium-cooled fast reactor is in France a candidate for a prototype of 4th generation system to be builtby 2020. A detailed working program has been defined to identify by 2012 the potential improvementtracks for later industrial development of these reactors. The goals for innovation are first identified:Progress of the safety with a special attention to severe accidents risk minimization and mitigation(defence in depth approach); Economic competitiveness of the system mainly by reducing the capitalcost, the investment risks by enhancing in service inspection and repair capacities, and raising theavailability; Consideration of advanced energy conversion system with particular emphasis on thereduction of risks linked to Na reactivity; Sustainability with fissile material management whileminimizing the proliferation risk; Capacity for long-lived waste transmutation. The detailed content ofthe CEA, AREVA and EDF coordinated program is then described.1. Introduction: a strategy in FranceIn March 2005, an inter-departmental committee stated that France should study sodium-cooled (SFR) and gas-cooled (GFR)fast reactors for the long term deployment of its nuclear fleet, together with Hydrogen production using high temperaturereactors. In January 2006, the French president requested for the design of a generation IV system to be operated by 2020. InJune 2006, the parliament voted a law to specify the future management of radioactive waste, and included the necessity tostudy by 2012 the options of future nuclear systems and to start operate a prototype in 2020. Finally, in December 2006, asecond inter-departmental committee agreed on a technical roadmap for SFR, GFR and fuel cycle studies leading in 2012 togather data to choose future options. This strategy is based on the necessity to save uranium and to reduce ultimate waste inthe future. In this frame, a coordinated program has been launched by the CEA, AREVA and EDF to develop innovativeSFRs. This program is presented hereafter2. Specific goals for innovationThe research goals for SFRs able to be deployed by industry in 2040-2050, are derived from the generic objectives of theGeneration IV International Forum. At first, the safety specifications are set up at a level at least equivalent to the one forgeneration III light water reactors (EPR). SFRs specificities (sodium and fast spectrum) are taken into account for behaviorunder severe accident conditions and for sodium risks. The occurrence of failing of critical components will be pushed backby extended and performing In Service Inspection (ISI).The economy of the system should be optimized to achieve a plant cost acceptable for the industry. If the investment cost istraditionally high for a nuclear plant, it is still higher for a SFR. The financial risk engaged by the utility that buy the plantmust be comparable to traditional power plants. So innovations to decrease the investment cost are to be searched fortogether with excellence in the availability of the plant.1

The interest of fast neutrons is to allow an optimized management of nuclear matters. First, the reactor should be able tobreed Pu with a breeding ratio at least equal to zero and must have the possibility to raise this ratio in the positive values inorder to allow the deployment of a fast reactor fleet by producing a sufficient mass of Pu. This implies too the deployment ofan industry that closes the fuel cycle and allows recovering Pu in a non-proliferant way.The same reactor must have the capacity to transmute long lived nuclear waste. The core must accommodate the quantity ofminor actinides (Americium, Neptunium and Curium) coming from its own closed fuel cycle and transmute most of it. Itshould be able, where needed, to transmute also wastes accumulated in the spent fuel of light water reactors. This implies themanagement of the minor actinides in the fuel cycle process.Finally, the weak points of the sodium technology must be improved. In that domain, in service inspection is concernedtogether with an easy operation, repair and dismantling. One should think to the availability of the plant on a 60 years periodwhile proving periodically that safety margins are sufficient.A coordinated research program between the CEA, AREVA and EDF was launched to answer those previous objectives. It isorganized in four main items which are reviewed in the four next paragraphs.3. An efficient Core with enhanced safety3.1 Reduction of the sodium void effectThe core performance in terms of fissile fuel management is to produce at least the same quantity of fissile Pu than the oneburnt. A basic objective is to get a zero breeding gain (IBG = production/consumption -1) for the fissile core, to ensure thatlater optimization including blankets could easily reach positive breeding gain to allow the development of a fast reactorsfleet. At that time, the blankets should be designed to avoid easy Pu separation from other actinides in a proliferationresistant way. Still for fissile material management, it is necessary to limit the mass of Pu that is needed per electrical MW.This objective to get a zero breeding gain brings too the safety advantages of a core with a low reactivity loss.In addition, future SFRs may have enhanced safety. The sodium void effect, that induces a positive reactivity for an industrialsize of the core, must be significantly reduced. The objective is to compensate it by negative effects from the Doppler andfrom other feedbacks like dilatation of the materials.Several items will be looked at. The impact of the fuel element geometry is studied while comparing pin to plate. Thegeometry (height/diameter, cylindrical, annular, modular) of the core is an open parameter as are size, total power and powerdensity, the relative volume of the components (fuel, structure and coolant) and the presence of moderator to sweeten theneutron spectrum. The sub-assembly geometry is also open to modifications: diameter of the wire spacer, impact of thesodium hydraulics, presence of a sodium plenum at the outlet of the fissile length (that voids as soon as sodium boils in thecore, inducing important neutron leaks).The nature of the fuel material is studied to compare advantage and drawbacks of oxide, carbide, metal and nitride. Oneneeds to design a specific fuel element adapted to each fuel material in order to optimize each core. The development of suchcores is conducted with a three steps program. First, neutronic calculations give reasonable materials repartition in the core.Then the fuel element and the sub-assembly design and technology are assessed. Finally, a detailed design of the core is usedfor several transient calculations that verify the performance and the safety of the given core. More details are given in [1].3.2 Compaction risk managementFast neutron cores are very sensitive to compaction. A special effort must be made to enhance their resistance to compaction,due to a seismic stress for example. The cores of the French SFRs (Phenix, Superphenix) are free to expand and compactionis limited by the contact at the pad level between sub-assemblies and especially dedicated stiff ones located at the coreperiphery. This effect could be enhanced.The performance of a ringed core will be evaluated comparatively. Finally, the dynamic behavior of the core when amechanical stress is applied will be studied using for instance the Symphony past experiments realized on a shaking table atCEA/Saclay. Modeling improvements either in static and dynamic situation will support these studies.3.3 Core instrumentationTaking advantage of the most recent technology evolution, the core instrumentation can be revised to develop new systemswith a better efficiency and a higher dynamics.The possibility of monitoring of the power distribution through in-core high dynamic fission chambers will be assessed.2

For different applications such as temperature (measurement of the temperature inside the sub-assembly head, avoiding thediscrepancy coming from the mixing jets at the core outlet), detection of boiling, presence of gas, ultra sonic detectors arebeing studied to be used under hot sodium conditions. Difficulties are to solve wetting of the transducer by sodium in orderto improve sensitiveness, and the question of a piezzo-electric material able to sustain high temperatures during long times.The clad rupture detection system can be optimized in order to reduce its response time delay. The clad rupture localizationsystem is an expensive system with tubes that cross the closure slab. Reconsidering the complete system could allow a betterperformance, simplification and enhanced security.3.4 Core performanceOne point of the core performance is the fuel burn-up. The first limitation to the lifetime of the fuel elements comes from thedose rate on the structure materials (clad, canister). A specific program, on the long term, is envisaged to improve the doserate acceptable on the clad.First, the present optimized austenitic steel AIM1 will be confirmed able to reach 120 to 130 dpa using recent irradiations inPhenix. To go beyond, a new cladding material ferritic-martensitic strengthened by oxide dispersion will be developed, whileadvanced austenitics track will be kept as a backup. The hexagonal canister should withstand the same dose than the clad andpossibly a higher temperature than presently. A ferritic steel like the T92 grade is foreseen.In order to reduce the total diameter of the core, a compact lateral neutron shield must be developed. Coming from the EFRstudies, a specific sub-assembly with neutrons absorber will be developed. Two problems are to be solved: the extraction ofthe heat produced in the material while the filling up density must be maximized, and the stability of the compounds on asufficient long time period.The lifetime of the control rods will be extended progressively.As to fuel, it is anticipated that the core of the prototype mentioned in the introduction, will be an oxide core, the only onethat allows for sufficient knowledge in the time scale of this reactor, including in accidental conditions. Innovation on theoxide fuel will be introduced as it is anticipated to be issued from so-called COEX TM process. This process allows forcoprecipitation of actinides and avoids handling of separated Pu, while allowing for simplified pelletization process.Qualification will be addressed in the timescale of the Prototype, and for that purpose an irradiation in Phenix (COPIX) iscurrently being prepared.Fig 1. Mixed (U,Pu) oxides obtained with different fabrication porocesses. From the left to the right : 11% Pu , MIMASprocess, 6%Pu COCA process and 27,5%Pi COEX processA R&D program in a more long term will be pursued on dense fuels, especially carbide. Clearly, the prototype will be usedin the frame of this program for experimental irradiations of such fuels, but application is foreseen in a longer timescale withthe view at industrial deployment.3.5 Minor actinides transmutationThe future cores of the GenerationIV SFR should be able to transmute minor actinides to reduce the quantity of ultimatewaste. Several technical options are available. Comparative studies of the global efficiency of these options are yet underwayfor various options of the future French nuclear fleet.Minor actinides can be mixed with the driver fuel, in a homogeneous way. In such a case their relative volume in the fuel isfrom 1% to 5%. They are quite easy to fabricate and handle, but all the fabrication process of the fuel needs additionaladaptation and protection against radiation.They can also be put in specific sub-assemblies with a high concentration, from 10% or more in volume. This is theheterogeneous way. In that case the number of sub-assemblies to manufacture requires a separate facility from the one for the3

driver fuel. The choice of the matrix that contents the M.A. is an essential part of the research. Most of the experimentslaunched in the past were devoted to inert matrix, either ceramic or metal. Specific irradiation experiments are yet underwayor under post irradiation examination with various materials and fuels. A new program is starting with a MA bearing UOXmatrix featuring radial blankets that could be placed around the core. Such a solution opens the way to proliferation resistantblankets.Other special concerns are on the one hand the fuel handling, according to the level of residual power even for a “fresh” fuel(especially in the case of the heterogeneous route), and on the other hand the detection of a clad failure of a sub-assemblycontaining M.A. as release of delayed neutron emitters has to be verified.3.4 Simulation toolsTools for core simulation include neutron physics, thermal-hydraulics, mechanics & fuel behavior. For neutron physics, thereference tool is ERANOS that has been validated in a wide range of situations. For thermal-hydraulics several tools areavailable, some are commercial like STAR-CD and others are specific like TRIO_U at CEA. For their use with sodium, theyneed to be validated on existing experiments from the 1980’s. Core mechanics will rely on the HARMONIE tool thatdescribes the static equilibrium of a core depending on limit conditions imposed at its boundary. The fuel behavior issimulated with the GERMINAL code. This tool was widely validated for a Phenix-like geometry. It will be extended tovarious future geometries and transferred into the PLEIADES platform, the French reference for nuclear fuel behaviorsimulation. GERMINAL will take advantage of the thermal-mechanics models yet available in PLEIADES. A special effortis intended on the long term to take into account and to validate the behavior of fuel loaded with large quantities of M.A.For all the tools, the existing experimental basis will be revisited and a set of database will be developed.On the medium term, coupling methods between the various physics will be introduced to simulate more precisely the coretransient behavior.4. Resistance to severe accidents and external hazards4.1 Safety approachGeneration IV systems require an enhanced safety. Globally, the SFR safety will be of the same level as the one of the thirdgeneration LWRs. The EPR is taken as a reference and its general objectives are already very ambitious and guarantee a veryhigh level of protection to persons and the environment. The defence-in-depth method is adopted as the basic principle tocover the risks and uncertainties inherent in this concept. Additional requirements provide both a real and demonstrablebenefit and a greater degree of assurance in the safety demonstration and therefore in its robustness. Four topics are studied:- Allowance for degraded situations and the "practical elimination" approach,- The robustness of the demonstration adapted to the system,- Consideration of the specific aspects of the sodium-cooled system,- Minimization of impacts concerning radiological protection and the environment (discharges, wastes, dismantling actions).A complete overview of this approach is given in [2].4.2 Scenarios and transients studiesThe innovative design that are envisaged may lead to specific scenarios, somewhat different from the one studied before inSPX or EFR. For instance, the fourth category transients are different whether you consider a loop-type or pool-type reactor.The presence of a power conversion system driven by gas will also induce new studies (impact of large quantities of gasunder pressure in case of suppression of the intermediate circuit for instance). A special emphasis will be put too on the longterm influence of M.A. in the sequence of events, a field of research quite new.4.3 Sub-Assembly design for core melt-down managementConcerning severe accidents and core degradation, a threefold strategy must be implemented that includes prevention againstmelting initiators (hydraulics allowing a better mitigation of a local temperature raise, design of a boiling zone capable tofavor natural convection and neutrons leakages…), enhancement of protection systems (passive 3 rd level shutdown level),and at the end, in case of core melt, provisions to minimize the consequences especially the risk of significant energetic recriticality.As a matter of fact, a major difference between LWRs and SFRs is the core reactivity. For a fast neutron spectrum,4

a compaction of the core induces a reactivity step; that is not the case in a LWR. So a specific risk linked to hypothetical coredegradation is the possibility to have a recriticality that can develop a mechanical energy release.Globally, two routes will be studied, one consisting in the dispersion of the molten core by introducing some dischargechannels either among the sub-assemblies or using neutron absorber channels, the other consisting in releasing absorbermaterial in the molten fuel to reduce its reactivity; the two solutions being potentially combined. Fig. 3 describes a qualitativescenario of core melt-down up to corium recovery with the objective to significantly reduce re-criticality risks or theirconsequences.Fig 2. Severe accident sequence of eventsFinally, to analyze the various modes of degradation, several initiators will be taken into account to study the possibledegradation of the core. This item contributes to the robustness of the demonstration. One should try to cover a wide range ofphenomenology to describe the mode of degradation of the core.4.4 Core-catcher studiesLikely they will have to cope with a reactor vessel more compact than in previous projects, while ensuring still the decayheat removal and non-criticality. The design (shape, but also location: in or ex-vessel), the materials, the modeling of thedebris, are the main R&D topics to be addressed.4.5 Strengthened systems for defense in depthIn addition to provisions mentioned for an upgraded monitoring of the core, it will be necessary:- to exclude by design scenarios such as the ingress of a large gas bubble in the core, a catastrophic failure of the coresupport structures, a compaction of the core,- to enhance the diversification of decay heat removal systems, either from the point of view of their location, the physicalprinciples used, the architecture of the plant and of its confinement,- to reinforce provisions against leakages and fires, reactions of sodium with fluid used for energy conversion,- to protect the plant against upgraded external aggressions such as earthquakes, new plane crash hypothesis.4.6 Accident modelingAccident modeling tools will be re-considered owing to their capacity to deal with the needs yielded by innovations selectedfor future SFRs. The basis for severe accidents involving very complex multi-physics aspects, will be treated with SAS4Aand SIMMER (with a refined pin model called DPIN) codes; the so-called CATHARE CEA’s code, currently being adaptedfor sodium applications, will be used for transient calculations and join SAS/SIMMER for the primary phase includingboiling but prior to the loss of geometry.So-called MC3D and PLEXUS CEA’s codes will be available respectively for corium-coolant interaction and dynamicmechanical loads of structures. Debris beads behavior is covered by LIDEB and MC3D for some aspects.5

Sodium fires will be addressed with FEUMIX and PULSAR (spray type fires).Transfer of species (including radiotoxics) in the reactor building and releases will be assessed with CONTAIN code.It is considered at this stage that the qualification of these tools can rely on the extended existing data bases, especially thenumerous experiences using simulants of fuel and coolant, and experiences in representative situations (sodium, nuclear heatand fuel) CABRI and SCARABEE [3] , as long as oxide fuel is concerned. This does not exclude that studies to come inducesome new needs, but they are not identified at now.The situation is very different for cores that could use dense fuels such as carbide; if promoted in the frame of futureindustrial commercial units, dedicated programs shall be requested timely.5. Looking for an optimized PCS to reduce sodium risk5.1 A gas power conversion systemThe main incentive for such an innovative option is to delete the risk of sodium water reaction and its potentialconsequences. Notice also that such an option opens, in the case of a loop type reactor, the possibility to suppress theintermediate sodium loop, and by the way to decrease the investment cost (see Fig 3.).NaN2Fig 3. Typical layout of a loop type SFR, without intermediate sodium loop and coupled (via an IHX Na-gas) to a nitrogenBrayton PCS (one turbine, two compressors on the same shaft, high power heat recuperator)Nevertheless at a given core outlet temperature, classical gases (such as nitrogen, or argon, eventually mixed with someamount of helium) will require a significant effort to compete with the Rankine water/steam cycle efficiency. These effortscan be made on the pressure level, on the improvement of the (indirect) Brayton cycle through optimized and enhancedcomponents, use of re-heating by sodium, and/or by raising the temperature level (at the core outlet). An alternative forrecovering an attractive efficiency (higher than 40% -Super Phenix value-) without temperature increase could be to usesupercritical CO 2 . This requires developing the necessary innovative technologies concerning components and materials.The corresponding studies aiming in a first step at establishing the feasibility are:- for Super critical - CO 2 : the cycle stability (including in load follow-up hypothesis), the components deasibility (especiallyturbine, compressors) and the sodium- CO 2 interaction through dedicated tests,- for all gases: the thermodynamical optimization and associated “hot” temperature level, the protection provisions: detectionof leakages, dedicated phases separator component, valves for insulation and decompression, possibility of a “short”intermediate loop between gas and primary sodium, the safety analysis versus the risk of massive gas ingress in core, theprospect about materials (compatibility with fluids and with required temperature level) and the preliminary studies of IHXs:heat recuperator, Na-gas IHX (including sodium plugging hazard).Good trends on this step would then allow undertaking heavier developments involving especially Na-gas IHX test at thescale of ~1MW exchanged, prior to larger ones if the option is definitely confirmed.5.2 Optimization of materials choice according to temperatures levelIndependently of the temperature level, future SFRs must be able to sustain a significant enhancement of their lifetime, up to60 years, for those components that will not be replaceable. For coping with this requirement, feedback from Phenix reactor(after its shutdown foreseen in 2009) will be used as it includes an interesting panel of steels either austenitic and ferritic,representative of relevant families and aged for a long time in representative conditions.For the non replaceable structures in the primary vessel, it is believed that hot and cold parts can be kept made out of thereference austenitic steel (Super-Phenix, EFR): Cr17-Ni12-Mo-Mn-(N). Nevertheless, in case of a temperature increase(possibly required for instance by efficiency concerns with gas PCS), by +50°C (i.e. 550 to 600°C), austenitic Cr25-20,Ni30-20 will be assessed for hot parts, with a special emphasis on creep performance and weld-ability. A more ambitious6

increment: (+100°C), if decided, will require a long term program on nickel based alloys. In both cases “corrosion” bysodium is a concern owing to Ni dissolution enhancement by temperature that will be checked by dedicated sodium tests.For other components such as heat exchangers, piping, austenitic steels could be challenged by ferritic-martensitic ones.Such a choice can be justified by mechanical properties (creep resistance), but also costs concerns as thermal properties (heatconductivity, thermal expansion coefficient), could allow for a lesser level of thermal induced stresses and a reduction ofmasses involved. The program includes the definition and optimization of a specific ferritic/martensitic grade within therange 9 to 12 chromium and to assess its attractiveness component by component.Conversely, a specific action will be devoted to evaluate the profit expectable from a limited (20°C) drop of the hottemperature of the cycle in terms of ageing of base and weld materials.At the end it is worth to mention that the outcomes of these researches will be implemented (provisions for procuring,material data, mechanical analysis methods, construction and inspection) in the code and standards dedicated to fast reactors(so called “RCC-MR”). A new release of this guide (undertaken for Super Phenix, enriched for EFR studies) is foreseen thisyear 2007.5.3 An enhanced Steam Generator energy conversion systemA first objective is to mitigate the risk of sodium-water reaction and its consequences; this yields a first set of actions:-reinforce reliability by technologies such as double walled exchange tubes, modularity etc…-assess the viability of keeping a “compact” secondary sodium loop (up to set, inside a same vessel, a SG and an IHX units,thermally coupled by a very limited amount of sodium, or by an alternative coupling fluid)-assess the possibility of replacing secondary sodium by another fluid compatible with water and sodium. With that viewdifferent metals mixtures and some salts are envisaged. They will be tested in terms of chemical stability (for salts), reactivitywith sodium, physical nature of reaction products, and corrosion of materials and provisions that could allow for its control.A second objective is to enhance the performances: with that view supercritical water cycle will be studied. It is worth tomention that such a cycle could allow for increasing performances by increasing the pressure (from 180bars for previousSFRs to 250 for instance, allow 2% efficiency earning) but a temperature increase is also to be considered (and will yield thesame type of materials concerns as for gas PCS and already presented, § IV.B). Nevertheless supercritical water rises specificcorrosion problems, that can be solved by use of nickel based alloys such as nickel based alloy 690 (to be checked also insodium environment)6. Reactor design re-examination6.1 Reactor primary systemThe previous paragraphs aimed at propose and evaluate innovations. The question is at last to see how long these innovationscan participate to coherent reactor layouts, in the frame of integration studies, and to check these layouts against the highlevel goals with dedicated tools (economy, safety..).As to the primary system, a lot has been done in Europe, and especially in France about the so-called integrated primary(pool type) system. This system provides a robust design of the primary confinement, against loss of primary sodium (and bythe way primary sodium fires), against loss of the primary hydraulic loop, ensures a high thermal inertia and guaranties agood natural circulation in the main vessel. The cold plenum contributes to the mitigation of thermal shocks and gastransport. It is also favorable to alleviate radioprotection concerns during operation and allows designing easily a hydraulicpath for cooling down the main vessel. As identified drawbacks, it is worth mentioning the difficulties to achieve a compactreactor block, to have an easy access to internal structures for monitoring and repair; it implies in vessel rotating componentsand earthquakes effects are complex because involving strong fluid-structure interaction.Loop type system has the important potential to make easier the intermediate heat transport loop suppress. It can offer someeasier maintenance and repair conditions for large components that are separated from the reactor tank, and can be integratedin a single component. There are no rotating parts in the reactor vessel and there is a potential for more compact components(main vessel). This design is likely more easy to justify vs. earthquakes because less sensitive to sloshing effects. Drawbacksconcern the risks associated to the loss of a primary loop (fire, leak, flow reversal, gas entrainment), lower thermal inertiaand risk of gas transportation. Keeping the main vessel below the creep regime is not easily achievable. Operation conditionscan be made more difficult owing to active, double walled, primary sodium transport piping. The program will consider bothoptions and will address the different topics mentioned just before through integration studies.7

Fig 4. Optimized pool type (left), and sketch of an optimization of loop type (right) primary circuitsBeyond this comparison pool vs loop, size effects will be dealt with especially in order to assess possible threshold effectsthan could incite to consider limited power output plants. When precise enough, and with the input from other systems andcomponents studies, the designs will be compared from the points of view of economics (SEMER code), safety, inspectionand repair, availability.6.2 Intermediate system optimizationFor those reactor layouts using an intermediary loop, the target for this last will be to reduce the cost (including formaintenance and manufacturing processes), looking for compactness (using ferritic-martensitic materials) and reduction orsimplification of the number of components and auxiliary circuits. Integrated components, short loops, improvement andsimplification of provisions against sodium leaks (including inert gas filled casemates) are the tracks foreseen to be followed.6.3 Components & systems optimizationAs for auxiliary systems, the sodium purity target and technologies for traps will be reconsidered. Sodium quality controlwill be adapted looking for direct measurement of impurities content (like O, H, C) in addition to conventional pluggingtemperature. Tritium management will need a particular attention to the regeneration of traps; in case of use of a gas PCS (asno hydrogen will be injected in sodium by reduction of water on the SG tube wall), a specific strategy is to be imagined.Cover gas treatment either at the input (removal of impurities) and at the ouput (gaseous FPs), is also subjected toimprovement, so is the treatment of aerosols in gas volumes above the sodium.6.4 Fast fuel handlingThis point is very important as it has a key contribution to the availability of the plant, and can be determining versus thedesign of the primary system:(1) its geometry, as it must in any case allow for access to all subassemblies and as room for in-vessel storage of used fuel isnecessary according to the option chosen,(2) its efficiency: how fast used subassemblies can be removed out of the core and replaced by fresh ones. This is particularlyimportant if in reactor vessel interim storage is not the option chosen, but it could be also a safety concern for instance if aninspection of the core support is needed or following an accident.(3) The design of the handling system can even concern the layout of the complete plant, in the case when a modulararchitecture appears to be attractive: as matter of fact the ex-vessel equipments, and especially any interim storage tank andwashing facility, can be shared by different modules8

Fig 5. Schematic sketch of a modular four units reactor with unique washing facility and interim in sodium storage tankR&D actions in the program are aimed at the following:- study three options for in-vessel fuel handling that are not indifferently applicable to the options for the primary system:1. optimization of existing system: two rotating plugs and one interim put down-take over position, with improvementsconcerning efficiency and high-power used SA discharge.2. assess one rotating plug plus pantograph solution3. assess direct handling using a dedicated hood.- study transfer and washing of the used fuel subassemblies at a power ranging in between 10 and 25 KW.6.5 Enhanced ISIRFig 6. Parts in Phenix reactor that have already been inspected with different technologies (left) and “MIR” device developedfor the volumic inspection of the welds on the primary vessel of Super-Phenix (right)Beyond these performances, for future SFRs, possibilities of in-service inspection and repair have to be clearly enhancedagain. This will be made first by considering inspection strategies at the design stage and from the point of view of thecriticality of each component or sites on this component. Strategies of access will be chosen in coherence: geometricalconsiderations and hatches, necessity or not to be able to empty the primary sodium, or part of it.Developments will be pursued on under-sodium ultrasonics technologies:For monitoring systems used during reactor operation, a key point is to define a piezzo-electric material suitable for the hightemperature of the hot pool/legs in the reactor.For periodic examination, classically made at “cold” shutdown conditions, a key point is to enhance the quality oftransmission of the US generated by the transducer to the fluid, and back from the target to the transducer.9

At the same time the modelling of US propagation and reflection will be developed in order to help the optimization ofdipped transducers technologies. Mono- and multi-elements will be developed as well, depending of the application(telemetry, far viewing, close viewing, volumic NDT against small or large defects).Development of distant US Technologies, allowing to check a structure from its external side, using it as a wave guide(already employed for examination of the Phenix distant welds on the conical shell supporting the core) will be pursued.7. ConclusionsThe program presented above, will be organized with regard to two short term milestones: 2009 and 2012. The first period isdedicated to propose and study innovations that will be integrated in very preliminary sketches. The 2009 milestone will bethe opportunity to select promising orientations. Between 2009 and 2012, selected technologies will be studied in depth, andan integration work towards one reactor layout and one backup will be performed. Each milestone with be also theopportunity to address the question of the prototype that will have the aim to feature the technologies of the power plant ofthe future as far as possible. In 2012, the main specifications of this reactor are to be fixed and will take into account itsmission regarding demonstrations on the fuel cycle. The facilities for its own cycle will also have to be defined, with regardto the first core, and to possibilities of cycle experiments at the scale of some sub-assemblies.8. References1 - L. BUIRON et al, “Innovative core design for generation IV sodium-cooled fast reactors”, Proc. of the ICAPP2007 Conference, paper 7383.2 - French Advisory Group on Safety, “First deliberations on safety aspects of 4 th generation reactors. Paths toimproving safety Approach, tools and associated R&D”, Proc. of the ICAPP 2007 Conference3 - Evaluation of material-coolant interaction and material movement and relocation in liquid metal fast reactors,IAEA Technical Committee Meeting, IWGFR-94, 09-06-1994, O-Arai, Japan4 - RCC-MR AFCEN, Tour AREVA, F 92084 Paris la Défense CEDEX10

Session 18.1.2:Status of future concepts

CURRENT STATUS OF JAPANESE SODIUM COOLED LOOPTYPE FAST REACTOR (JSFR) DEVELOPMENTK. AOTO, H.KAMIDE, H. OSHIMA, I.SATO, H.HAYAFUNE,H.NIWA, M.MORISHITAAdvanced Nuclear System Research and Development Directorate, Japan Atomic Energy Agency (JAEA)4002 Narita, Oarai-machi, Ibaraki-ken, 311-1393, JapanABSTRACTA new project, Fast Reactor Cycle Technology Development Project (FaCT Project) waslaunched in last autumn in Japan. In the project, conceptual design study on JapaneseSodium cooled loop type Fast Reactor (JSFR) which was selected as the most promisingconcept for next commercialized reactor in previous study and research and development ofinnovative technologies adopted in the concept are implemented toward an importantmilestone at 2015. In order to satisfy the high design requirements, several innovativetechnologies were identified and included in the current design. They are categorized intothree areas: for economic competitiveness, enhancement of reliability, and enhancement ofsafety. This paper describes the current status of the project, especially on the targets and thedesign requirements for JSFR as well as some related innovative technologies development,integrated IHX/Pump, thermo-hydraulic optimization in compacted reactor vessel andexperimental study on FBR core-disruption accidents aiming at establishment of advancedsafety logic, “elimination of severe re-criticality events”.1. IntroductionThe first stage of development of commercialized fast reactor cycle systems in Japan was finalized in2006. Following the results in which the sodium cooled loop type fast reactor (JSFR) with oxide fuelselected as the most promising reactor concept[1], a new project, Fast Reactor Cycle TechnologyDevelopment Project (FaCT Project) was launched focusing on development of the selected concept.In order to satisfy the high design requirements, several innovative technologies were identified (SeeFig.1) and included in the current design. They are categorized into three areas: for economiccompetitiveness, enhancement of reliability, and enhancement of safety. Shortening of piping length byadoption of high chromium steel, 2-loop system, integration of IHX with primary pump (integratedIHX/Pump), compacted reactor vessel by hot-vessel concept are in the first category. SG withdouble-wall-tubeis categorizedinto the secondarea. Passiveshutdownsystem,re-criticality freecore by specialsub-assembly,and anti-seismictechnology are inthe last.This paperdescribes thecurrent status ofJSFR in FaCTProject,Fig 1. The features of the JSFR system.

especially on the design requirements, current design as well as some related innovative technologiesdevelopment, integrated IHX/Pump, thermo-hydraulic optimization in compacted reactor vessel andexperimental study on FBR core-disruption accidents aiming at establishment of advanced safety logic,“elimination of severe re-criticality events”.2. Development targets and design requirements of JSFRIn the FaCT project, R&D activities will be carried out under the development targets as summarized inTable 1. And the design requirements of JSFR shown in Table 2 are established in order to satisfy thedevelopment targets.CategorySafety and ReliabilitySustainabilityEnvironment ProtectionWaste ManagementEfficient Utilization ofNuclear Fuel ResourcesEconomic CompetitivenessNuclear Non-ProliferationTargetsSR-1: Ensuring safety equal to or better than contemporary LWR(LWR)SR-2: Ensuring reliability equal to or better than LWREP-1: Radioactive influence through normal operation no more than LWREP-2: Emission control of environment transfer substance which can restrictin safety limitsWM-1: Reduction of the amount of radioactive waste equal to LWRWM-2: Improvement of waste manageability equal to or better that LWRWM-3: Reduction of radio-toxicity equal to or better than LWRUR-1: Breeding performance to enable transition to fast reactor, and itsflexibilityEC-1: Electricity generation cost equal to or cheaper than the competingenergy sources in the futureEC-2: Investment risks no more than LWREC-3: External costs no more than LWRNp-1: Adoption of institutional measures and application of technicalfeatures which can enhance non-proliferationNp-2: System design of physical protection and its development to preventtheft of nuclear materials and sabotageTab 1: The development targets for FaCT Project.ItemRequirementBreeding Capability Breeding ratio: ca. 1.2, System doubling time: ca. 30 yearsTRU BurningTRU burning under fast reactorMulti-recycle and long-term storage of LWR spent fuel (Transmutationof LLFP such as I-129, Tc-99 is desirable)Radioactive Release Equivalent or less than present LWR applicationPR&PPExcludes pure-Pu state throughout system flowSafetyOperability, Maintenability, Repairability and Passive safetyRe-criticality free, core damage frequency less than 10 -6 /ryElectricity Generation Cost Cost-competitiveness with other means of electricity production and avariety of market conditions, including highly competitive deregulatedor reformed marketsOperation Cycleca. 18 months, and moreConstruction Duration Large-scale: 42 months, Medium-scale modular type: 36 monthsTab 2: Major Design Requirements of JSFR System3. Current status of some main innovative technologies development3.1 Integrated IHX/primary pump[3]In order to reduce manufacturing, building, and operation cost, the primary pump and IHX are integratedand put into one vessel as shown in Fig.2 in the JSFR design. A critical issue in this component designis the fretting of the IHX heat transfer tube with a baffle plate. At present, some experiments andnumerical analyses are carried out to develop technologies for reduction of vibrations from the pumpand a vibration transfer control system.

(1) 1/4 scaled model vibration transfer experimentwith vibration oscillator and analysis: An inertialvibration oscillator was placed in a simulated pumpcasing of a test apparatus. Frequency responsecharacteristics of the 1/4 scale model with andwithout water was revealed from experimentalresults. The 1/4 scale vibration test wasnumerically analyzed by FEM-code (FINAS) witha three-dimensional shell model. This code candirectly handle the vibration behaviour of complexmulti-cylinders with considering fluid-structureinteraction effects. Fig.3 shows vibration responsesat innermost shell structure and tube of IHX. Theanalytical results were in good agreement with theexperiment, with 10% accuracy in eigenvalue. Theresults demonstrated that the numerical analysis byFINAS can be applied to the IHX designoptimization. In the IHX analysis, it is evaluatedthat there is no resonant vibration mode at pumpspeed of 100% operation capacity, 44% (initial lowpower operation), and 10% (stand by) with a safetymargin of 20%.(2) 1/4 scaled model vibration transfer experimentwith pump and analysis: Most unfavourablefeature of this design is that the oscillationeigenvalue of pump casing is lower than thevalue of revolution per second (RPS) of pump,because the pump casing is slim and long to beinstalled in IHX. This design makes avibration resonance between pump rotatingvibration and casing eigenvalue in pumpspeedup operation. The results showed thatpump speed and casing have a resonance andthe shaft vibration has the peak value ataround 1000rpm. However, the pumpstructure has enough attenuation and thevibration is lower than the limit.Fig.2 Integrated IHX/Primary pumpFig.3 Frequency response curve at dividing wall ofIHX/pump bottom3.2 Thermal-hydraulic optimization in compacted reactor vessel[4]Thermal stratification phenomena during the scram transient were investigated by using an 1/10th scaledmodel of the reactor vessel upper plenum, where water was used as working fluid. Balance betweenbuoyancy force and inertia force, i.e., Ri number in the experiment was set equal to that in the designedreactor. The temperature difference between the initial hot temperature in the plenum and cold flowfrom the core after the scram was set at 25˚C. Then the flow velocity at the core outlet became only1/10th of that in the designed reactor. Thus, Re number distortion is order of 100. The experimentalstudy was carried out to see the influences of the column type UIS (Upper Inner Structure) with a slitwhere a subassembly was transported during a fuel exchange operation and also to find mechanism ofcharacteristic phenomena. Configurations of components in the upper plenum, i.e., height of acylindrical plug in front of the UIS slit and an outer cover of the UIS were also examined as a mitigationmeasure of thermal load during the thermal stratification.(1) Experimental Setup: The 1/10th scaled model for the upper plenum of reactor vessel is shown inFig.4. The left side figure shows the reference geometry. In the reactor design of JFSR, the column typeUIS, which has the radial slit and no outer cover, is located at the centre of the upper plenum. Doubledipped plates are set to reduce the flow velocity and avoid the gas entrainment at the free surface. Other

two cases in the right figures are the geometryparameters to investigate influences on thecharacteristics of thermal stratification phenomenaand mitigation methods for the thermal stress.Temperatures in the upper plenum were measured bysome series of copper-constantan thermocouples at10Hz sampling frequency. The outlet temperatures ofseveral core and blanket fuel subassemblies were alsomeasured at the centre of each outlet hole. Themeasurement error was less than 0.1°C.(2)Results and Discussion: The temperature data inthe reference case are analyzed to see the stratificationphenomena. Vertical temperature distributions atinner position in the plenum near the UIS slit and theopposite DHX are shown in Fig. 5. The temperatureis normalized by the temperature drop (ΔT) at thecore outlet after the scram and the lower limit of coreoutlet temperature (Tc). Typical distribution in thethermal stratification can be seen in this figure. Themaximum temperature gradient at the stratificationinterface near the UIS slit was larger than near theDHX. This thin interface layer is resulted from theimpingement of the jet through the UIS slit at thestratification interface and local entrainment of thefluid at the bottom of the interface layer.Two configurations were examined to find amitigation measure. One is the FHM plug, which isinserted to the lower position in the upper plenum andthe other is a perforated outer shell of the UIS. TheFHM plug is designed to change direction of the jetthrough the UIS slit. The UIS outer shell guides thecold fluid exiting from the core to flow upward. Itwill help better mixing in the upper plenum. Thetemperatures in these cases were compared to see theinfluences of the modified structures on the thermalstratification interface. The vertical temperaturedistributions at the slit and DHX sides are shown inFig. 6 at t =1200s from the scram. In the slit side,both the FHM plug case and the UIS outer shell caseshowed that the temperature gradients at the thermalstratification interfaces were smaller than that in thereference case. The FHM plug case has advantage ofa minor impact on the in-vessel components design.Fig.4 Configuration of JSFR UISFig.5 Vertical temperature distributions nearthe UIS slit and DHX in reference case3.3 Re-criticality free core[5]Several in-pile and out-of-pile tests were conductedunder a co-operation between JAEA and NationalNuclear Centre of Republic of Kazakhstan (EAGLEprogram). One of the main objectives of these testswas demonstration of effectiveness of special FBRdesign concepts to eliminate the re-criticality issue inthe course of core disruption accidents. Figure 7Fig.6 Influences of mitigation methods onvertical temperature distributions nearthe UIS Slitshows schematic of a typical in-pile test apparatus of the EAGLE program. The geometry of this testapparatus is corresponding to a typical special design concept equipped with a “discharge duct” withineach fuel sub-assembly. The discharge duct of 2mm-thick stainless steel filled with liquid sodium wasplaced at the central part, and was surrounded by 75 UO2-fuel pins with 400mm fissile height giving

total fuel amount of ~8 kg. The test ID1 (Integral Demonstration test 1) was conducted with this testapparatus in IGR (Impulse Graphite Reactor). It was intended to produce a molten fuel-steel-mixturepool with the trapezoidal power diagram simulating the hottest part of the degraded core in a ULOFaccident. Thermocouples placed within the fuel-pin bundle region suggested molten-pool formationfrom ~27.4 second on. The duct failure took place at ~28.7 second, about one second after start of thestrong duct heating by the pool, and it was followed immediately by a rapid sodium void developmenttoward the bottom as shown in Fig. 8. Through this developed void, an effective mixture dischargelasting for about one second took place. This result showed a significant potential of core-materialrelocation even under a relatively low pressure difference (up to 0.12MPa). Although post-testexaminations to quantify final material distribution are still to be performed for this test, followingpreliminary conclusions have been drawn out through data analysis of all the in-pile and out-of-piletests.(a) Sodium-filled duct wall fails at an early phase of fuel-steel mixture pool formation corresponding tothe hottest core part.(b) Mixture discharge takes place in a short time range provided that fuel enthalpy is high enough and ameaningful pressure difference is maintained.(c) Relocated mixture can be quenched forming debris as far as sufficient amount of coolant is available.4. ConclusionThe targets and the design requirements for developmentof advanced sodium cooled loop type fast breeder reactor(JSFR) in FaCT -project are briefly discussedin this paper. And the present status of threemain R&D issues is reported. Preliminaryresult of each R&D is summarized as follows,(1) As for integrated IHX/pump, the properdesign is to be realized based on bothexperimental and numerical studies and will beapplicable to JSFR.(2) The experimental results showed that thestratification interface had steep temperaturedistribution near the UIS slit due to theimpingement of the jet through the slit. This steeptemperature gradient at the stratification interfacecan be mitigated greatly in a case where the FHMplug was located at the lower position near the coretop in the upper plenum.(3) Concerning re-criticality free core, so far mainoutcomes are encouraging the present computer-codeFig.7 Typical EAGLE in-pile testapparatusFig.8 Response of TCs and void sensors inthe discharge duct in the ID1 Testsimulations for the ULOF accident predicting that the reference FaCT design concept is free from there-criticality issue.Reference [1] M.Ichimiya, et al., Proc.GLOBAL2003, (2003) 434-446. [2] Y,Sagayama,GLOBAL2007, to be appeared (2007). [3] H.Hayafune, et al., GLOBAL2005, Paper No.207(2005). [4]H. Kamide, et al., Proc. NTHAS5, F003(2006). [5] K. Konishi, et al., NED, to be published (2007).

THE (EUROPEAN) HTR TECHNOLOGY NETWORK(HTR-TN) AND THE DEVELOPMENT OF HTRTECHNOLOGY IN EUROPED. HITTNERAREVATour AREVA, 92084 PARIS LA DEFENSE Cedex – Franceand members of the HTR-TN Executive BoardE. BOGUSCH 1) , D. BESSON 2) , D. BUCKTHORPE 3) , V. CHAUVET 4) ,M. FÜTTERER 5) , A. VAN HEEK 6) , S. LANSIART 7) , W. VON LENSA 8) ,J. PIRSON 9) , D. VERRIER 2)1) AREVA, Erlangen (Germany) 6) NRG, Petten (Netherlands)2) AREVA, Lyon (France) 7) Commissariat à l’Energie Atomique, Cadarache (France)3) AMEC NNC Ltd, Knutsford (UK) 8) Forschungszentrum Juelich (Germany)4) LGI, Paris (France) 9) Suez Tractebel, Bruxelles (Belgique)5) JRC-IE, Petten (Netherlands)ABSTRACTMastering global warming risks and securing the European energy supply cannot beobtained only by developing CO 2 free electricity generation, as electricity represents only alimited part of energy consumption, while most of the remainder is provided by fossil fuelburning. High Temperature Reactors (HTR) can contribute to reduce CO 2 emissions bysupplying heat needed by many industrial processes. Due to promising market prospectsand attractive safety features, several industrial HTR prototype projects emerged in theworld during the last decade. Europe, presently leader in nuclear energy, should maintainits rank in this race for a new frontier in nuclear energy. HTR-TN, created in 2000 forbuilding a coherent partnership for HTR development in Europe, proposes a roadmap forthe emergence of a new generation of reactors addressing heat needs of European industry.A first step should be a worldwide first-of-the-kind demonstration of the coupling of a HTRwith an industrial process heat application. HTR-TN initiated the development of advancedgeneric HTR technology in the 5 th and 6 th Framework Programmes, with already importantresults. Beyond continuation of generic R&D, the large components of the reactor, thecoupling system and the application part of the demonstrator, which are beyond state-ofthe-art,should be developed and qualified during the 7 th Framework Programme, requiringthe development of an infrastructure of large specific test facilities.1. IntroductionEurope has been a leader in High Temperature Reactors (HTR) development from the 60s’ to the 80s’(DRAGON, AVR, THTR reactors have been constructed and operated). After a break due to thenuclear phase out in the leading country, Germany, the HTR development restarted with a smallproject, INNOHTR, funded by EURATOM in the 4 th Framework Programme (FP4), followed by acluster of 10 coordinated projects in the 5 th Framework Programme (FP5) and by a large IntegratedProject, RAPHAEL, in the 6 th Framework Programme (FP6) with the additional contribution of aSpecific Targeted Research Project dedicated to the study of the potential of HTR to burn actinides,PUMA. As the development of a new type of nuclear system requires a large and long term effort farbeyond the time period of a single Framework Programme (up to two decades), the partners of theEuropean HTR development programme founded in 2000 the (European) HTR Technology Network(HTR-TN) in order to elaborate a long term R&D strategy and to organise a stable partnership forimplementing it. Significant results have already been obtained and more are expected in the comingyears. But for the time being, the HTR European programme was only dedicated to the consolidationof generic high temperature technologies and to the exploration of advanced solutions for improving

the performances of future HTR (the VHTR objective). Now the time is coming for a new stepforward towards the development of industrial high temperature systems in Europe, as it is already thecase in other parts of the world.Therefore in the next Framework Programme (FP7), on top of the continuation of programmesdedicated to the development of generic technologies, a support from the European programme to thedevelopment of a HTR prototype of industrial size and of its components should be considered, andthe large test facilities required for the qualification of these components should be built.2. The specific role of HTR/VHTR in the future fleets of nuclear reactorsIn the context of political uncertainties on the access to fossil fuel resources, of long term worldwidedepletion of these resources and of global warming risk, nuclear energy can play an important role forsecuring energy supply of Europe during the 21 st Century. But at present, nuclear fission is dedicatedalmost exclusively to electricity generation, which, however, accounts for only 16% of the energyconsumed in the world, 79% of the remaining energy consumption coming from fossil fuel burning[1]. Therefore, in order to contribute significantly to reduce the dependence on fossil fuel supply andto master CO 2 emissions, beyond electricity generation, nuclear energy should also address asignificant part of the rest of the energy market. As nuclear plants are producing large quantities ofheat that cannot be transported on long distances if they are not converted into electricity, the nonelectricitytarget for nuclear energy should be the industrial processes that consume large quantities ofheat in a limited area. Nevertheless as some of the industrial processes that can be considered aremeant at improving the extraction of fossil fuel or producing synthetic fuel (extracting oil from tarsands, lightening of heavy oil, coal to liquid transformation or hydrogen production), nuclear reactorscould also indirectly reach scattered energy uses, in particular for transport.Most of the industrial process heat applications require much higher temperatures than the operatingtemperatures of present Light Water Reactors (LWR) or of the future liquid metal fast reactors, whichare therefore doomed to be mainly stuck to electricity generation. Moreover the quantity of energyrequired is never more than a few hundred megawatts, while most of the present or future systemsbecome competitive only for a thermal production of several thousand megawatts. Because it producesheat at high or very high temperature using a smaller reactor with a very robust safety concept, themodular HTR/VHTR systems have the potential to address a wide range of industrial process heatapplications that cannot be addressed by LWR or liquid metal fast reactors, which does not excludemedium-sized electricity generation, or process heat and electricity cogeneration. Therefore it is clearthat HTR/VHTR should be included in future nuclear reactor fleets not in competition with other typesof nuclear systems, but in complement to them mainly for addressing the specific mission of providinghigh temperature heat for industrial processes.3. Possible approach for entering HTR on the industrial process heat marketThe result of a survey of present industrial processes, which require at least 100 MW of heat within asingle industrial site, is shown in figure 1 [2]. It appears that:• Potentially the HTR/VHTR heat market is far from being a niche market: such reactors couldaddress heat needs of many industrial processes. Therefore the future of HTR/VHTR is not onlydepending on the possible long term development of a hypothetical “hydrogen civilisation”, but onthe incentives (increasing costs, CO 2 tax, security of supply…) that present industries will have inthe short/medium term, to switch from heat supply by fossil fuel burning to alternative supplies.• There is a first group of processes requiring temperatures below 600°C, mainly for steamproduction, and a second group above 900°C, with practically no need between 600 and 900°C. Forthe first group, the heat could be provided by a HTR using existing industrial materials and provenTRISO HTR fuel. On the contrary, the materials and the fuel required for the VHTR, which wouldproduce the heat needed by the second type of applications, are still to be developed. Thereforeapplications at very high temperature (> 900°C) could not be considered but in the longer term.

200400600Iron Manufacturing800Electricity GenerationGasification of CoalTown GasPetroleum RefineriesDe-sulfurization of Heavy OilWood Pulp ManufactureUrea SynthesisDesalination, District HeatingReactor Temperature up to 850°C10001200Temperature (°C)1400Glass ManufacturingCement Manufacturing(Direction Reduction Methods)(Gas Turbine)1600Hydrogen (Steam Reforming)Ethylene (naphtha, ethane)Styrene (ethylbenzene)(with a Blast Furnace)ApplicationNuclear HeatFigure 1: Present heat intensive industrial processesTherefore, in order to minimise risks, HTR/VHTR penetration on the industrial process heat marketshould be undertaken step by step.First a full industrial scale demonstration of the feasibility of the coupling between an HTR and aprocess heat application should be obtained as soon as possible at a reasonable temperature level. Asnuclear energy has only been used at industrial scale for electricity generation, such a first-of-the-kinddemonstration, even at a moderate operating temperature, is already quite a major challenge. It isnecessary in order to verify that a nuclear reactor can actually be connected to an industrial processand face its requirements and hazards, which will certainly be quite different from those of utilities forelectricity generation. It is also necessary in order to demonstrate to heat intensive industries, used tofully integrate fossil fuel burning heat supply in their processes (in particular industries processingfossil fuels), that resorting to nuclear energy could not only be feasible, but also beneficial for them interms of economic competitiveness, CO 2 emissions and sparing of natural resources. Taking intoaccount the time necessary for R&D and qualification work, procurement of components, licensingand construction, the prototype demonstration could not be operated before the end of next decade.Adding to the challenging objectives of the first demonstrator the very high temperature target woulddrastically increase the risks and the length of development. But, while the first demonstrator willprovide – at a reasonable temperature level – the first experience feedback from coupling a nuclearheat source with an industrial process, which is presently missing, the R&D for higher performancematerials and fuel will be continued and will expectedly produce results allowing defining a credibledesign for VHTR. Nevertheless entering into this new phase, market needs for such a reactor shouldbe reassessed, because they could have changed from those shown in figure 1: industrial processesusually have a lifetime of one or two decades and presently the trend is to search for lower energyconsumption and lower temperature processes.4. The legacy from previous European projectsBeyond the legacy from former European HTR achievements, the large effort made in FP5 and FP6,as well as the complementary national programmes, led to significant results concerning genericaspects of the HTR technologies. These results have already been set out in different papers (see forinstance [3] and [4]) and therefore only the main achievements will be recalled here:• A steel needed for operating the vessel at higher temperature than with PWR vessel steel (limited to350°C), modified 9Cr1Mo, has been selected and the main elements of feasibility for using it for a

HTR vessel have been obtained (weldability of thick plates, no significant impact of irradiation onmechanical properties, at least 50°C increase in negligible creep limit compared to PWR steel…).• The use of nickel base alloys, which are the best candidates within existing industrial materials interms of high temperature mechanical and corrosion performances, for application to the IHX, willnot allow operating the demonstrator above 800-900°C (depending on IHX design).• Large varieties of graphite grade samples have been characterised, tested for corrosion resistance,and are presently irradiated at 750°C and 950°C beyond the turn around point. The data alreadyobtained will allow selecting the most appropriate grades for HTR graphite core application.• Conceptual designs have been obtained for the main primary components (in particular differentplate IHX concepts have been examined), but only few testing of these components could beperformed (for instance for the IHX in a helium loop (figure 3), the hot gas duct thermal barrier(figure 4), the circulator magnetic bearings (figure 5)) due to the need of large dedicated test loopswhich are not existing yet.Figure 3: HE-FUS3 heliumloop, IHX mock up testing,ENEA Brasimone (Italy)Figure 4: HETIMO test bench,thermal barrier testing, CEACadarache (France)Figure 5: FLP 500, facility for testing thedynamics of a shaft supported by magneticbearings, IPM Zittau (Germany)• The bases for fabrication of HTR fuel have been recoveredand a laboratory scale facility for manufacturing HTRTRISO fuel particles and fuel elements (compacts) has beencommissioned in France. Methods for quality control forfuel fabrication are developed.• The performance of state-of-the-art HTR fuel (recoveredfrom the best German former fabrications) at very highoperating temperature (up to 1250°C for the fuel particle)and burn-up (up to 15-20% FIMA) are being explored intwo irradiations, HFR-EU1 and HFR-EU1bis.• The main calculation tools required for HTR design andlicensing exist or are being presently developed by differentEuropean organisations. A joint qualification effort hasbeen undertaken in FP5 and FP6 through benchmarks andacquisition of new experimental data (fuel irradiation tovery high burn-up, PIE and heat-up tests for fuelperformance modelling and irradiation of fuel coating material samples for elaborating laws ofevolution of coating material properties under irradiation, isotopic analysis of very high burn-upfuel for fuel depletion calculation, recovery of operating data of the EVO 50 MW helium Braytoncycle loop for system transient calculation, NACOK loop test (figure 6) for calculation of airingress (coupling of natural convection and graphite oxidation)).• Long term leach tests of irradiated HTR fuel in geological disposal conditions have been launched.Preliminary results, to be confirmed in the continuation of these tests, already show that TRISOparticles keep their unique robustness in final disposal conditions and that their lifetime in such

conditions should be at least 10 000 years. On the other hand a preliminary survey allowedidentifying promising paths for separating fuel from graphite and for managing separately bothwaste streams in such a way that the volume of ultimate wastes to be disposed off is minimised.5. A European roadmap for the development of a HTR demonstrator for industrialheat applicationAs shown in Part 3, the development of HTR/VHTR systems should open a new frontier to nuclearenergy, allowing it not only addressing electricity needs, but also other types of energy needs, whichrepresent the largest part of the energy consumption. The European Union, faced to the challenges ofsecuring its energy supply and of mastering its CO 2 emissions, should not miss such opportunity.The development of HTR industrial prototypes already started in several leading nuclear countries,with projects of industrial prototypes: NGNP in the USA, the PBMR in South Africa, HTR-PM inChina, GT-HTR 300 in Japan and NHDD in Korea. Europe, which has been until now a worldwideleader in nuclear energy, but which limited its HTR developments to generic R&D, should not beabsent from this race for mastering nuclear high temperature technologies and should enter as soon aspossible in the development of a HTR prototype coupled to an industrial process heat application,endeavouring to embed this project in an international cooperation framework. But the progress indesign and R&D should not wait for the development of such an international partnership framework,which will take time and should be an important task for a future FP7 HTR project.Developing a partnership with industrial process heat end users is also an essential task: it is notbecause the HTR produces high temperature heat that the adequacy with industrial end users’ needs isassured. Industrial requirements for heat production must be defined in an interactive way between thenuclear reactor designer and the industrial heat end user. The reactor designer will adapt his design toindustrial requirements, and the industrial end user will also have to tune his process for integratingmore easily the nuclear heat source into the optimised scheme of its plant. An important part of theprogramme should therefore be dedicated to this evolution of the industrial process.Happily such ambitious developments will not start from scratch: they will rely on the legacy offormer European HTR projects and on the generic R&D programme started of FP5 and FP6.Nevertheless, there are still some basic R&D issues that have not been fully examined because of thelength of the necessary work, or even not been addressed at all yet, because they did not get a toppriority in the early phases development, even if they will be required for the industrial developmentof a new type of HTR/VHTR. FP7 should complete most of the remaining generic R&D tasks, even ifsome residual ones should be continued in the 8 th Framework Programme (FP8):• Even if important progress has been made in FP5 and FP6 for selecting and validating the materialsneeded by HTR/VHTR projects, the work have to be continued for completing qualification files ofthese materials, particularly when high fluence irradiation or low stress creep require long tests.• Though important code qualification elements have been obtained during FP5 and FP6, theexperimental databases required for certification of computer codes are far from beingcomprehensive. For example, in reactor physics, additional critical experiments are necessary forreducing uncertainties on neutron flux distribution; elements of validation of the thermal feedbackon the power distribution obtained through coupled neutronic / thermo-fluid dynamics calculationscould be found in some HTTR measurements; the “hot ASTRA” test just selected by ISTC willallow qualifying the calculation of the core temperature coefficient from room to the operatingtemperature, which is a key element of HTR safety demonstration. Additional fuel irradiation andsafety tests will be required for improving fuel modelling and qualifying fuel performance codes.Due to the distinctive behaviour of HTR materials (e.g. graphite) or to particular features of HTRdesign (e.g. seismic behaviour of a stack of hexahedral blocs), there are HTR specific issues inmechanical modelling that must be addressed. The thermo-fluid dynamics modelling of somecritical zones (e.g. mixing of hot and cold helium in the lower reactor vessel plenum) will have tobe qualified. Depending on the conclusions of RAPHAEL, additional qualification work mighthave to be performed on the transient system analysis codes, including perhaps some testsperformed on gas loops in configurations representative of the reactor design.

• The irradiated fuel leach tests, started in FP5, presently continued in FP6 are still to be extended inFP7 in order to reduce the uncertainty in predictions. Moreover the management of irradiatedgraphite, only touched in a preliminary way in FP5 and absent in FP6, will have to be studied inorder to develop solutions in this area which will be critical for HTR acceptability.• For having a chance to implement the technologies developed in the European programme in anactual industrial reactor prototype, some technological developments neglected until now will haveto be addressed. This is the case for the development of an instrumentation adapted to the operatingconditions of HTR/VHTR: while, until now, the in-reactor instrumentation of this type of reactorhas always been very limited, it is desirable to acquire more comprehensive operation data in afuture prototype in order to check the performance of materials and components and to beauthorised later to operate industrial reactors at a high performance level. HTR head end fuelreprocessing and recycling technologies (e.g. technologies for breaking the coating of the irradiatedparticles and recovering the kernels, as well as the development of manufacturing processes foractinide fuel and the testing of the behaviour of this fuel under irradiation) have to be developed forcompatibility with closed cycle strategies that might be needed for satisfying sustainabilityrequirements. Last but not least, after a first “state-of-the-art” study in RAPHAEL, the modellingof fission product and dust transport, for which very large uncertainties exist, will likely needimprovements in order to determine more accurately the HTR source term.Moreover in order to investigate the potential of HTR/VHTR for higher performances in the longterm, the exploration of innovative solutions for fuel and materials, started in RAPHAEL will have tobe continued, keeping carefully the balance with shorter term needs for the demonstrator.In the meantime industry started to select the main design options (pebble / block type fuel, IHXdesign concept, number of primary loops, general architecture of the system, temperature, pressureand power level, materials for the main components, etc.) in parallel to FP5 and FP6 projects. But theselection of some design options (most particularly for the IHX concept) and the validation of thesechoices require mock-up tests. As only few ones could be performed until now (see Part 4), a largeexperimental programme should be launched in FP7. It is recommended to start with separate effecttests in different facilities, each one providing one type of representative conditions: thermalloads / flow rate / coolant / chemistry. Such relatively small facilities will be more easily availablethan large integral test loops, and moreover they will allow identifying separately the sensitivity toeach parameter, facilitating design optimisation, which would not be the case with integral tests. Thenthe selection of design options should be validated in a large integral test loop providing most ofrepresentative operating parameters, like the ones planned in CEA and Forschungszentrum Karlsruhe(figure 7). For the final qualification of the largest components, a bigger loop, in the range of 10 to20 MW, will even be necessary. As the development of these large facilities will take several yearsand will require important funding, it is necessary to plan it right from the beginning of FP7.The HELITE loopThe HELOKA loop (FZK)Figure 7: large helium loops planned in EuropeDesign option selection will strongly depend on the possibility of licensing a reactor integrating suchoptions. The definition of a safety reference frame taking benefit of the specific safety features of

modular HTR is a key task that has to be undertaken during FP7 in continuity with the safety approachstudies of FP5 and FP6, to pave the way for designing the demonstrator.As already mentioned, there will be a strong interaction between the reactor design and the industrialapplication to which the reactor will provide heat. This interaction must therefore be handled rightfrom the beginning of the project. During FP7, the end users will elaborate their requirements, whichwill be an important input for the selection of the reactor design options and the reactor designer willformulate the constraints imposed by the use of a nuclear heat source, which will be taken into accountby end users for adaptation of their processes. Moreover it should be noted that, depending on thetemperature level of the heat to be provided to the industrial process and on the safety requirementsconcerning the limitation of interactions between the nuclear reactor and the chemical plant (inparticular, but not only the distance between the 2 facilities), one should pay a careful attention to thedesign of the heat transport system between them, which might be beyond the state-of-the-art.The scheduling of the different phases of the programme of development of the demonstrator isrepresented in figure 8.FP5FP6FP7FP8…2000 2005 2009 2013Base technology developmentSelection of the main design optionsValidation of technologies& design optionsDevelopment oflarge test loopsQualification of componentsREACTORReactor constraintsDesign & constructionDevelopment of a safetyapproachDefinition of a safetyreference frameConstruction of a consortium withend users & international partnersMarket approach End user requirementsDesign adaptation / developmentDesign & constructionFigure 8: Schedule for the development of the demonstratorINDUSTRIALAPPLICATION6. ConclusionIt is clear that such an ambitious project, including design of an industrial prototype, will be possibleonly with a drastic increase of its public funding, in particular the one coming from the EURATOMFramework Programme, in comparison with the present situation. But even with a large Europeaneffort to support this project, HTR-TN would recommend internationalising it in order to developsynergies and to alleviate the European burden, either by merging with an existing internationalleading project (NGNP, PBMR, etc) or by developing a European project and attracting internationalpartnership. Constructing such an international partnership should be a major task of the project, but itwill be possible only if Europe appears to have on its own a strong programme with clear objectivesand an obvious added value.The ambition is high, but the challenge is of strategic importance. Energy will play a central role forthe future of Europe: the cost and the long term security of energy supply will have a key impact in theprosperity of the EU; the increasing energy consumption is the main source of the global warmingrisk. If nuclear energy, thanks to competitive HTR/VHTR, enters the largest part of the energy market,

which is not the electricity market, it can therefore significantly contribute both to the Europeanprosperity and to the mastery of European CO 2 emissions.7. References[1] “Key world energy statistics 2005”, International Energy Agency (2006)[2] “Synthesis of information on non-electricity applications of nuclear energy: data collection onavailable non-nuclear processes and coupling with nuclear reactors”, K. Verfondern,W. von Lensa, Michelangelo Network Project of the Euratom 5 th Framework Programme,Work Package 4, MICANET-02/06-D-4.12.0, Rev.2 October 2003[3] “RAPHAEL, a European Project for the development of HTR/VHTR technology for industrialprocess heat supply and cogeneration”, D. Hittner, E. Bogusch, D. Besson, D. Buckthorpe,V. Chauvet, M. A. Fütterer, A. van Heek, W. von Lensa, M. Phélip, J. Pirson, W. Scheuermann,D. Verrier, Proceedings of the 3 rd International Topical Meeting on High Temperature ReactorTechnology, October 1-4, 2006, Johannesburg, South Africa[4] “Outlines of the French R&D programme for the development of High and Very HighTemperature Reactors”, Ph. Billot, D. Hittner, Ph. Vasseur, Proceedings of the 3 rd InternationalTopical Meeting on High Temperature Reactor Technology, October 1-4, 2006, Johannesburg,South Africa

The ELSY ProjectL. CINOTTIDel Fungo Giera EnergiaVia Durini 23, 20122 Milano – ItalyG. LOCATELLIAnsaldo NucleareCorso Perrone 25, 16161 Genova -ItalyH. AÏT ABDERRAHIMSCK-CENBoeretang 200, B2400 MOL - BelgiumS. MONTI, G. BENAMATI,ENEAVia Martiri di Montesole 4, Bologna – ItalyH. WIDERJRC/IE1755 ZG Petten - NetherlandsD. STRUWEFZKHermann-von-Helmholtzplatz 1, 76344 Eggenstein, Leopoldshafen - GermanyA. ORDENEmpresarios AgrupadosMagallares 3, E28015 Madrid - SpainABSTRACTThis paper presents the current status of the development of ELSY (theacronym for the European Lead-cooled System).The ELSY reference design is a 600 MWe pool-type reactor cooled by purelead. This concept is under development since September 2006, and issponsored by the Sixth Framework Programme of EURATOM. The ELSYproject, coordinated by Ansaldo Nucleare, is being performed by a consortiumconsisting of twenty organizations including seventeen from Europe, two fromKorea and one from the USA. The partners are from industry, researchorganisations and universities.ELSY aims to demonstrate the possibility of designing a fast critical reactorusing simple engineered technical features, whilst fully complying with theGeneration IV goals of sustainability, economics, safety, proliferation resistantand physical protection.Compactness of the reactor building is possible due to the elimination of theIntermediate Cooling System, and the adoption of innovative DHR systems.Among the critical issues, the effect of the large mass of lead has beenconsidered; this assessment allows being very confident in the feasibility of thereactor vessel and its support.1. IntroductionThe Generation IV (GEN IV) Technology Roadmap [1], prepared by GIF member countries,identified the six most promising advanced reactor systems and related fuel cycle and the R&D

necessary to develop these concepts for potential deployment. Among the promising reactortechnologies being considered by the GIF, the LFR has been identified as a technology with greatpotential to meet the needs for both remote sites and central power stations.In the GEN IV technology evaluations, the LFR system was top-ranked in sustainability because ituses a closed fuel cycle, and in proliferation resistance and physical protection because it employs along-life core. It was rated good in safety and economics. The safety was considered to be enhanced bythe choice of a relatively inert coolant. The LFR was primarily envisioned for missions in electricityand hydrogen production and actinide management. Given its R&D needs for fuel, materials, andcorrosion control, the LFR system was estimated to be deployable by 2025. The LFR system features afast-neutron spectrum and a closed fuel cycle for efficient conversion of fertile uranium. The LFR canalso be used as a burner of all actinides from spent fuel and as a burner / breeder with thorium matrices.The GIF LFR Provisional System Steering Committee has prepared a draft of the System Research Plan(SRP) for the Lead-Cooled Fast Reactor [2] with molten lead as the reference coolant and lead-bismuthas backup option. Figure 1 below illustrates the basic approach being recommended in the LFR SRP. Itportrays the dual track viability research program with convergence to a single, combineddemonstration facility (demo) leading to eventual deployment of both types of systems.SSTAR(20MWe;Preliminary design2006-2009)ELSY(600MWe;Preliminary design2006-2009)ViabilityR&DDesign of a first ofa kind LFR from(2016-2020)Demo 10-100MWeR&D engine2008-2018Prototype of acentral station LFR(2013-2025)AdvancedR&DIndustrial deployment of asmall scale LFRfrom 2020Industrial deployment of anadvanced small scale LFRfrom 2035 (H2, CO2 cycle)Industrial deployment of acentral station LFRfrom 2025Industrial deployment of acentral station LFRfrom 2035 (H2, CO2 cycle)Fig.1. LFR SRP Conceptual FrameworkThis approach consists of the design of a small transportable system of 10–100 MWe size that featuresa very long refuelling interval, and of a larger system, rated at about 600 MWe, intended for centralstation power generation. Following the successful operation of the demo around the year 2018, aprototype development effort is expected for the central station LFR leading to industrial deployment atthe horizon of 2025-2030. In the case of the small transportable (SSTAR) option the development of afirst of a kind unit in the period 2016-2020 is foreseen. The design of the industrial prototype of thecentral station LFR and that of the first of a kind SSTAR should be carried out in parallel to theconstruction of the Demo and planned in such a way as to start construction as soon as beginning of theDemo operation at full power has given the main confidence of the viability of this new technology.2. ELSY consortiumA major step in favour of the LFR occurred when EURATOM decided to fund ELSY (the acronym forthe European Lead cooled System) - a Specific Targeted Research Project of the 6th EuropeanFramework Program (FP6) – proposed to investigate the economical feasibility of a lead-cooled,critical reactor of 600 MWe power [3-4] for nuclear waste transmutation. Since September 2006, aconsortium of twenty organizations (from industry, research centres and universities) includingseventeen from Europe, two from Republic of Korea and one from United States (Tab 1) has beenpursuing the development of ELSY.The ELSY project, scheduled to last three years, aims at demonstrating the possibility to design acompetitive and safe Lead-cooled fast power reactor using simple engineered features. This prospect is

appealing also to private investors who have offered to participate in the initiative. This would createthe conditions for advancing the ELSY activity even beyond the current sponsorship under Euratom’sFP6. The use of compact, in-vessel steam generators and a simple primary circuit (Fig 2) with allinternals possibly being removable are among the reactor features needed for competitive electricenergy generation and long-term protection of investment.Participant organisationCountryAnsaldo Nucleare S.p.A ANSALDO ItalyAGH, Akademia Górniczo-Hutnicza AGH PolandCentro Elettrotecnico Sperimentale Italiano CESI ItalyInter Universities Consortium for Nuclear Technological CIRTEN ItalyResearchCentre National de la Recherche Scientifique CNRS FranceEmpresarios Agrupados Internacional S.A. EA SpainElectricité de France EDF FranceEnte Per Le Nuove Tecnologie, L'energia e L'ambiente ENEA ItalyForschungszentrum Karlsruhe GmbH FZK GermanyInstitute for Nuclear Research INR RomaniaEuropean Commission, Joint Research Centre JRC EuropeRoyal Institute of Technology-Stockholm KTH SwedenNuclear Research and Consultancy Group NRG NetherlandsUstav jaderneho vyzkumu Rez, a.s. (Nuclear ResearchCzechUJVInstitute Rez, plc.)RepublicPaul Scherrer Institut PSI SwitzerlandStudiecentrum voor Kernenergie•Centre d'Etude del'énergie NucléaireSCK•CEN BelgiumSeoul National University SNU KoreaDel Fungo Giera Energia S.p.A. DEL ItalyMassachusetts Institute of Technology MIT USAKorea Electrical Engineering and Science ResearchInstituteKESRI KoreaTab 1: Organizations involved in the ELSY projectFig 2 Preliminary scheme of the ELSY Reactor

The preliminary parameters of ELSY are specified in Tab 2.To meet the technological needs of the ELSY project, it is important to capitalize on the strong synergywith other two European initiatives, “The Integrated Infrastructure Initiative VELLA,” [5] which isdevoted to the dissemination of knowledge in the field of lead and lead-alloys technology, and the”Integrated Project EUROTRANS [6]”.Plant CharacteristicTentative Plant ParametersPower600 MWeThermal efficiency 40 %Primary coolantPure leadPrimary systemPool type, compactPrimary coolant circulation, at power ForcedPrimary coolant pressure loss, at power ~ 1,5 barPrimary coolant circulation for DHR Natural circulation + Pony motorsCore inlet temperature ~ 400°CCore outlet temperature ~ 480°CFuelMOX with consideration also of nitrides and dispersed minoractinidesFuel cladding materialT91 (aluminized)Fuel cladding temperature (max) ~ 550°CMain vessel Austenitic stainless steel, hung, short-height ~ 10 m;diameter ~ 12 mSafety vesselAnchored to the reactor pitSteam generatorsN° 8, integrated in the main vesselSecondary cycle Water-supercritical steam at 240 bar, 450°CPrimary pumpsN° 4 or 8 mechanical, in the hot collectorInternalsRemovableInner vesselCylindricalHot collectorSmall-volume, above the coreCold collectorAnnular, outside the inner vessel, free level higher than freelevel of hot collectorDHR coolers N° 4, DRC loops + a Reactor Vessel Air Cooling System .Seismic design2D isolators supporting the reactor building2. Plant power and reactor vessel sizingTab 2 Tentative parameters of the ELSY plantThe ELSY power plant is tentatively sized at 600 MWe because only plants of the order of severalhundreds MWe are expected to be economically affordable on the existing, well-interconnected grids ofEurope. Because the mass of lead of a LFR is worldwide a-priori considered a critical issue for thereactor vessel which can limit the plant power, a preliminary mechanical verification, including seismicloads, has been performed from the beginning of the design activity based on preliminary parameters.The reactor vessel has been checked to ASME III code applying response spectra of a similar nuclearplant, EUR requirements and a reactor building supported by 2D seismic insulators.The results show that the code requirements are satisfied for all service levels and allow beingconfident in the feasibility of the vessel and its support. The ongoing activity is now aimed to confirmthat the assumed relatively small vessel dimensions, are realistic thanks to innovative solutions of theprimary system layout. A LFR of a power larger than a medium power is potentially feasible accordingto these preliminary evaluations.3. Coolant and thermal cycleA large experience exists on LBE in Russia [7] and elsewhere [8-9]. Since lead is much more abundant(and less expensive) than bismuth, in case of deployment of a large number of reactors, pure lead as

coolant offers enhanced sustainability. Furthermore, the use of lead strongly reduces the production ofthe highly radioactive decay-heat generating polonium in the coolant with respect to LBE [10]. Theseare the main reasons for selecting lead as primary coolant for ELSY.Operation at a higher lower limit of the thermal cycle, required by the use of pure lead, would benecessary also in the case of LBE to improve plant efficiency and to avoid the excessive embrittlementof structural material subjected to fast neutron flux.The risk of lead freezing is reduced by the choice of a pool-type configuration.The choice of a large reactor power suggests the use of forced circulation to shorten the reactor vessel,thereby avoiding excessive coolant mass and alleviating mechanical loads on the reactor vessel.Thanks to the favorable neutronic characteristics of lead, the fuel pins of a lead-cooled reactor, similarlyto LWRs, can be spaced more apart than in the case of sodium, resulting in a lower pressure drop acrossthe core. As a consequence, in spite of the higher density of lead, the pump head can be kept low (onthe order of one to two bars) with a reduced requirement for pumping power.A possible primary-side thermal cycle of 400°C/480°C in lead, without an Intermediate CoolingSystem, offers reduced risk of steel creep and milder thermal transients, while providing the thermalefficiency above 40% with a supercritical Rankine steam cycle at 240 bar, 450°C .The reactor vessel is designed to operate at the cold temperature of 400°C, which would be a safecondition even if oxygen control in the melt is temporarily lost. All reactor internals will have tooperate at higher temperatures, at which it is necessary to rely on oxygen control, whereas fuel claddingcould be surface-treated (aluminization seems to be a promising route) for a greater safety margin. Animproved primary-side thermal cycle at higher core outlet temperature could be adopted in the longerterm, as new materials become available4. Decay heat removalAccording to the predicted low primary system pressure loss and the favorable transport properties oflead, decay heat can be removed with lead in natural circulation in the primary system.A simple system for decay heat removal is the Reactor Vessel Air Cooling System (RVACS), whichconsists basically of an annular tube bundle of U-tubes arranged in the reactor pit with atmospheric airflowing pipe-side in natural circulation. RVACS is a passive system, but its use without other systemscan only be considered for small-size reactors since the vessel outer surface is relatively large incomparison with the reactor power. In the case of ELSY, the RVACS performance is sufficient only inthe long term (about one month after shut down) and a Direct Reactor Cooling (DRC) system is neededequipped with coolers immersed in the primary system. Stringent safety and reliability requirements ofthe DRC system will be achieved by redundancy and diversification.The DRC system is made of four loops; two loops operating with water (the W-DHR loops) and theremaining loops with water and/or air (the WA-DHR loops), (Fig 3).Each W-DHR loop is made of a cooling water Storage Tank, a water-lead Dip Cooler, interconnectingpiping, and steam vent piping to discharge steam to the atmosphere. The two W-DHR loops with thecontribution of the RVACS are sufficient to remove the decay heat in order to respect the temperaturelimit of 650°C specified for the 4th Category, service level D, over a week time from reactor shut down.Each WA-DHR loop is made of an inlet air duct, an air-lead Dip Cooler and an outlet air duct. The inletair duct is equipped with an electric fan supplied by batteries. Isolation valves are installed in the inletair and outlet ducts. A connection of the WA-DHR Dip Cooler to the cooling water storage tank of aW-DHR loop is also provided to for improved cooling with a mixture of air and water. The two WA-DHR loops with the contribution of the RVACS and the use of the water of the W-DHR loops in theshort term from the reactor shut down, are sufficient to evacuate the decay heat in order to respect thetemperature limit of 650°C established for the 4th Category, service level D. In the long term operationwith air natural circulation is sufficient to respect the temperature limit.The respect of the temperature limits of the 2nd and 3rd Category is ensured in operation with three outof the four DRC loops and the RVACS.The Dip Cooler tube bundle is made of bayonet tubes. The bayonet consists of three concentric tubes,the outer two of which have the bottom end sealed. Water evaporation or air heating takes place in theannulus between inner tube and the intermediate tube. The annulus between the outer tube and

intermediate tube is filled with He gas at a pressure higher than the lead pressure at the bottom end ofthe bundle. All annuli are interconnected to form a common He gas plenum, the pressure of which iscontinuously monitored. A leak from either walls of any of the outer tubes, is promptly detectedbecause of depressurization of the common gas plenum.Fig 3. The DRC W-DHR (right-side) and WA-DHR loops,process scheme showing stored cooling water interconnection.The bayonets of the ELSY DRC Dip Coolers are different with respect to classical bayonets, whichconsist each of only a pair of concentric tubes. The two outer tubes do not constitute a double walledtube, but are mechanically and thermally decoupled. This configuration allows to localize the most partof the thermal gradient, between lead and boiling water across the gas layer, avoiding both risk of leadfreezing and excessive thermal stresses across the tube walls during DHR steady state operation andtransients.5. Primary system and reactor buildingFigure 2 shows the cylindrical inner vessel concept, a scheme evaluated as a starting point for theprimary system design of ELSY. Hot lead is pumped into the pool above the PP and driven through theSGU tube bundle into the cold pool. The free level of the hot pool inside the SGU is higher than thefree level of the cold pool outside that is higher, in turn, than the free level of the hot pool above thecore enclosed by the inner vessel.A free level difference of cold and hot collectors at normal operating condition of only 1-2 m issufficient to feed the core, eliminating the complicated, pressurized core feed system (known to sodiumfast reactor community as Liposo and Sommier, in French) typical of the pool-type, sodium-cooledreactors.Simplification of the internals will offer the possibility of removable in-vessel components, a provisionfor investment protection. In spite of the identified advantages of this scheme, design improvements arebeing developed at least to make the primary system more tolerant to Steam Generator Tube Rupture(SGTR) accidents. Compactness of the reactor building is the result of reduced footprint and height.The reduced footprint is allowed by the elimination of the Intermediate Cooling System, the reducedelevation is the result of the forced circulation, of the new DHR DRC system and of the designapproach of reduced-height components.

6. ELSY can meet the generation IV goalsThe main features identified in order to achieve the GEN IV goals are based either on the properties oflead as a coolant or are specific designs to be engineered for ELSY.SustainabilityBecause lead is a coolant with low neutron absorption and scattering, it is possible to maintain a fastneutron flux even with a large amount of coolant in the core. This allows an efficient use of neutrons, abreeding ratio of about 1 without fertile assemblies, long core life and a high fuel burn-up.The fast neutron flux significantly reduces net MA generation, Pu recycling in a closed cycle being thecondition recognized by GEN IV for waste minimization.The potential capability of the LFR system to safely burn considerable amounts of recycled minoractinides within the fuel will add to the attractiveness of the LFR. To this end, different coreconfigurations are being studied and compared (see a specific paper at this conference).Economics.A simple plant will be the basis for reduced capital and operating cost. A pool-type, low-pressureprimary system offers great potential for plant simplification. The use of in-vessel Steam GeneratorUnits (SGU’s), and hence the eliminating the intermediate circuit, is expected to provide competitivegeneration of electricity in the LFR. The configuration of the reactor internals will be as simple aspossible. The very low vapour pressure of molten lead should allow relaxation of the otherwisestringent requirements of gas-tightness of the reactor roof and possibly allow the adoption of simplefuel handling systems.Reduction in the risk to capital results from the potential of removable/replaceable in-vesselcomponents.Safety and ReliabilityMolten lead has the advantage of allowing operation of the primary system at atmospheric pressure. Alow dose to the operators can also be predicted, owing to its low vapour pressure, high capability oftrapping fission products and high shielding of gamma radiation. In the case of accidental air ingress, inparticular during refuelling, any produced lead oxide can be reduced to lead by injection of hydrogenand the reactor operation is safely resumed.The moderate ΔT between the core inlet-outlet temperature reduces the thermal stress duringtransients, and the relatively low core outlet temperature minimizes creep in steels.It is possible to design fuel assemblies with fuel pins spaced as in the case of fuel assembly of the waterreactor. This results in a moderate pressure loss through the core of about one bar, in spite of the highdensity of lead, with associated improved heat removal by natural circulation and the possibility of aninnovative reactor layout such as the installation of the primary pumps in the hot collector to improveseveral aspects affecting safety. In case of leakage of the reactor vessel, the lower free level of thecoolant will be sufficient to ensure the coolant circulation through the core and the safe decay heatremoval. Any leaked lead would solidify without significant chemical reactions affecting the operationor performance of surrounding equipment.With high-density lead as a coolant, fuel dispersion dominates over fuel compaction, making theoccurrence of complex sequences leading to re-criticality less likely. In fact lead, with its higher densitythan oxide fuel and its natural convection flow, makes it difficult to lead to fuel aggregation withsubsequent formation of a secondary critical mass in the event of postulated fuel failure.Proliferation Resistance and Physical ProtectionThe use of MOX fuel containing MA increases proliferation resistance. The use of a coolantchemically compatible with air and water and operating at ambient pressure enhances PhysicalProtection. There is reduced need for robust protection against the risk of catastrophic events, initiatedby acts of sabotage because there is a little risk of fire propagation and because of the passive safetyfunctions. There are no credible scenarios of significant containment pressurization.

7. ConclusionsA major step in favor of the LFR did occur when EURATOM decided to fund the ELSY project, inresponse to the call “Nuclear Waste Transmutation in Critical Reactors,” to investigate the economicfeasibility of using critical reactors for nuclear waste transmutation.ELSY can find relevant synergy and technical feedbacks from the ongoing FP6 activities on ADS, and theIntegrated Infrastructure Initiative (VELLA).ELSY is expected to be a simple, innovative reactor, with compact primary system and reactor building,appealing to utilities for sustainable electric energy generation with reduced capital cost andconstruction time. Based on the promising initial results, it is expected that ELSY can confirm theambitious objectives of the designers and open a phase of strong international support for LFRdevelopment and deployment.Considering that significant commonality of R&D can be found between the small, transportablesystem and the medium-or large-sized system of the two GEN IV approaches, the GIF SRP proposescoordinated research with a single demonstration facility that can serve the R&D needs of bothapproaches. Full power operation of the Demo around the year 2018 - using to the greatest extentsimple solutions, standard materials and operating at relatively low temperature, to reduce as much aspossible the technological risks - could also justify the construction, at that date, of the first of a kind orindustrial prototypes of SSTAR and ELSY and the industrial deployment at the horizon of 2025-2030as foreseen in the GEN IV Roadmap.8. References[1] GIF-002-00, “Gen IV Technology Roadmap” – December 2002.[2] System Research Plan for the Lead-cooled Fast Reactor (LFR) – Draft document of the LFRProvisional Steering Committee, May 2006.[3] L. Cinotti, H. Aït Abderrahim, G. Benamati, C. Fazio, J. Knebel, G. Locatelli, S. Monti, C. F.Smith, K. Suh “Lead-Cooled Fast Reactor”, FISA 2006, Luxembourg 13-16 March 2006.[4] L. Cinotti, C. F. Smith, J. J. Sienicki, H. Aït Abderrahim, G. Benamati, G. Locatelli, S. Monti, H.Wider, D. Struwe, A. Orden, I.S. Hwang.- The Potential of the LFR and the ELSY Project, ICAPP2007, Nice, France, May 13-18, 2007.[5] G. Benamati et al. “VELLA Project: an initiative to create a common European research area onlead technologies for nuclear applications“ presented at the same ENC 2007 Conference[6] J. Knebel et. al – “European Research Programme for the Transmutation of High Level NuclearWaste in an Accelerator Driven System: EUROTRANS” - OECD Nuclear Energy Agency 9th IEM onActinide and Fission Product P&T (9-IEMPT) - Nîmes, France, 25-29 September 2006[7] A. V. Zrodnikov, G. I. Toshinsky, O. G. Komlev, Yu. G. Dragunov, V. S. Stepanov, N. N. Klimov,I. I. Kpytov, and V. N. Krushelnitsky, "Use of Multi-Purpose Modular Fast Reactors SvBR-75/100 inMarket Conditions," Paper 6023, 2006 Congress on Advances in Nuclear Power Plants (ICAPP '06),Reno, Nevada, USA, June 4-8, 2006.[8] A.Aiello, A Azzati, G. Benamati, A. Gessi, B. Long, G. Scadozzo, “Corrosion behaviour of steels inflowing LBE at low and high oxygen concentration” Journal of Nuclear Materials, 335, 169-173, 2004.[9] C. Fazio, A. Alamo, A. Almazouzi, D. Gomez-Briceno, F. Groeschel, F. Roelofs, P. Turroni and J.U. Knebel, “Assessment of Reference structural materials, heavy liquid metal technology and thermalhydraulicsfor European waste transmutation ADS”, GLOBAL, Tsukuba, Japan, Oct 9-13, 2005.[10] E. O. Adamov and V. V. Orlov, A. Filin, Final report on the ISTC Project #1418: "Naturally SafeLead-Cooled Fast Reactor for Large Scale Nuclear Power", Moscow 2001.

Session 18.1.3:Countries’ perspectives on nuclear energypolicy

CONCEPT OF THE GLOBAL NUCLEAR ENERGY SYSTEMA.YU.GAGARINSKI, V.F. TSIBULSKIRussian Research Centre “Kurchatov Institute”Moscow, Kurchatov sq., 1ABSTRACTThe paper considers the world energy demand till the middle of the century; demonstratesthe possibilities of nuclear energy to meet this demand, and outlines innovative technologyrequirements determined by the development scope. Russia’s potential contribution inmeeting the challenges faced by the XXI century’s nuclear power is also discussed.1. IntroductionIn accordance with the IAEA data, in 2006 thirty-two countries (having about two-thirds of the worldpopulation) operated 442 nuclear power reactors with total installed capacity of 369.7 GWe (net).Construction of 29 nuclear power units of about 24 GWe total installed capacity in twelve countriescharacterizes the short-term global nuclear energy prospects. Around a dozen of other countriesofficially announced their intentions to create nuclear sectors in their national power industries.Contemporary plans of nuclear energy development by the mid-century, which the world countries areconsidering independently, have an internationally acknowledged target of about 700 GWe.The idea of consolidating the international efforts aimed at providing an open access to nuclear energyfor all countries, while preserving and maintaining the non-proliferation regime, is a global nuclearenergy system concept development incentive.2. Energy demandIn order to develop any suppositions concerning the global nuclear energy outlook (scale, structure,key requirements) in the long term, it would be necessary at first to assess the potential global nuclearenergy demand – at least, till the middle of the XXI century.The growth of the global energy consumption in the XXI century is determined by the two principalcauses: the growth of population (according to various expert assessments – 1.3-1.9-fold by the midcentury),and the rapprochement of the consumption levels in developed and developing countries.Different sources estimate the energy consumption to increase with a factor of 1.6-2.5 by the midcentury.The specific energy consumption time variance factor for the developed and the developing countriesseems to be decisive, so it was specifically analysed by the experts of the leading Russian nuclearcentre – Kurchatov Institute [1]. This analysis gave us important results (Fig. 1).

Number of peopleSpecific energy consumption levelingbetween the developed and thedeveloping countries would require atripled production of primary energyresources2520qL/qSDevelopingcountries70% of populationDeveloped countries< 30% of population15105qL, qS – average specific energy consumption ofthe two global population groups011965 1975 1985 1995 2005 2015 2025 2035Fig.1. Rapprochement by Specific Energy ConsumptionThe assessment of the rate of rapprochement between specific energy consumptions in the two groupsof countries is a key parameter determining the energy market situation. Assuming the rapprochementtrend is maintained in the perspective, it is easy to calculate the required amount of primary energy. Itcan be shown that a continuation of the current world trends would result in an energy resource deficitalready in the nearest future.Assuming the primary energy sources are growing with a rate close to IAE forecasts, the fuel mixpicture by the mid-century would contain an “unsatisfied demand” area (i.e., resources, which shouldbe used to meet the projected energy demand).Total prim ary energy supply, M toeGrowth from 2005 to 2050(maximum assessment)O il 0.9350003000025000200001500010000500001930 1940 1950 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050G as 1.1Coal 4.0H ydro 2.0Biomass& w aste 3.0Otherrenew ables 9.0Nuclear 3.0Oil Gas Coal Hydro Biomass and W aste Other Renewables Nuclear Unsatisfied demand30% deficit of primary energy resources by 2050(By 2030 – IEA forecasts (W EO-2005), after 2030 – extrapolation)Fig. 2. Primary Energy SupplyIf any new energy technology is to assure sustainable global energy development, it should releaselow emissions, be deployed on a large scale and have a long-term resource base available.

3. Nuclear energy development scenariosSupposing the “unsatisfied demand” is met by nuclear energy, installed NPP capacities should makeseveral thousand gigawatts by the mid-century.Thus, the projected XXI century’s world energy demand does not impose any upper limit onnuclear energy development, the scale of which would be determined by development opportunitiesmeeting the list of requirements, which should be analysed right now.The bottom level of nuclear energy development by the mid-century could be represented by theabove-mentioned current plans of its development in the world countries, or by the 1000 GWe level,which has been considered as a “ceiling” recently enough.It is important that the scale of nuclear energy development determines the key nuclear energyrequirements – among which we emphasize the fuel supply and the need of innovative technologies.In scenarios conditionally considered as “low” (below 1000 GW by 2050), fuel supply requirementsare met relatively easily, and there is practically no need of innovations or possibilities of extendingthe nuclear energy application sphere. In fact, such scenarios leave nuclear energy on the level of a“technological experiment” (or a result of the countries’ wish to possess nuclear technology in theirnational security interests), which has no significant impact on the energy supply of the mankind.“High” nuclear energy development scenarios, which (also conditionally) are considered starting fromthe “medium” scenarios proposed by IPCC international expert group as far as 2000 (2000 GW by themid-century, with nuclear share in primary energy not exceeding 20% by the century end) [2],determine principally new requirements to nuclear energy and, in the same time, open theopportunities to expand its application sphere.Exemplary calculations performed form the nuclear energy structure: light-water reactors in oncethroughfuel cycle (Fig. 3), systems with moderate and high breeding in the closed fuel cycle (Fig. 4)are, naturally, quite different in terms of their basic parameters. Integral (for 100 years) demand ofnatural uranium ranges from speculative 30 to realistic 10 billion tons, and the maximum annualseparation work – from 700 to 200 thousand tons of SWU (stabilized by the mid-century already in themoderate breeding scenario). Breeding systems would allow us to reduce the amount of spent nuclearfuel several times.3 5N a t u r a l U c o n s u m p t i o n i n t h e X X I c e n t u r y u n d e r d i f f e r e n ts c e n a r i o s , m i l l i o n t o n s3 02 52 01 51 05R e d B o o k , 1 5 m i l l i o n t o n s ( 2 0 0 5 )Y M c o n s t r u c t i o n f r e q u e n c y02 0 0 0 f o r e c a s t3 6 0 G W2 0 0 5 f o r e c a s t1 0 0 0 G W2 0 0 0 G WВ 2 – “ m e d iu m ” I P C C s c e n a r io ( 2 0 0 0 )Fig. 3. Once-Through Nuclear Fuel CycleWithout going into much detail, it should be nevertheless noted that the calculations have practicallyconsidered the whole “range of interest” of the global nuclear energy system development scenarios:from the very moderate approach with about 1000 GW of projected NPP capacity by the mid-century,to the so-called “aggressive” nuclear energy increasing its market attractiveness by replacing a fractionof other energy sources in the electricity generation and in other applications – such as hydrogen, heat

or potable water production, with the global nuclear energy system capacity reaching up to ∼ 10 000GWe by 2100.Electric capacity, GWFig. 4. Closed Fuel Cycle: “Moderate” Breeders (BR=1.25)Thus, assuming the uranium resource constraints based on the existing data, realization of “high”nuclear energy development scenarios leaves a two-component nuclear energy system with plutoniumbreeding for further consideration.Electric capacity, GWFig. 5. Closed Fuel Cycle: “High” Breeders (BR=1.6)Besides, these scenarios offer relatively strict conditions for the introduction rate of technologicalinnovations. The latter include the closed fuel cycle based on new reprocessing technologies, “good”and “very good” breeders (with BR ∼ 1.2÷.1.6), and “very good” LWRs with BR ∼ 0.9 and withplutonium fuel.As we see, the wish to develop nuclear energy with the given rate dictates so rapid innovations that –at least, for a considerable part of the world – its implementation would require a consolidatedinternational effort to make nuclear energy accessible for all the countries concerned.4. Russia’s contribution to nuclear energy challengesBeing one of the founders of the First Nuclear Era, Russia possesses vast experience of solving the keynuclear energy problems of the XXI century.Today 10 Russian NPPs have an installed capacity of 23.2 GWe and generate about 16% of thecountry’s electricity. In accordance with the government’s Federal Program of the Nuclear Energy

Industry Development adopted in 2006, by 2020 the total installed capacity of Russian NPPs shouldreach 41 GWe, with annual energy production of about 300 TWh. The government intends to investover 25 billion USD from the federal budget in the construction of NPPs between 2007 and 2015.Russia’s preparedness for the innovative development of nuclear energy technologies could be brieflysummarized as follows:Technologies for nuclear energy sources:• Development of designs for the next generation of VVERs (NPP-2006 and large VVER) isnearing its completion, in view of massive NPP construction in the short term.• Small NPP construction started; medium NPPs are being designed.• Fast neutron reactor BN-800 is being built, in order to demonstrate and further improve thetechnology of the uranium-plutonium fuel cycle closing.• New fast reactor designs are at various stages of development.• Technologies of energy production for non-electric applications (in particular, for hydrogenproduction, heat supply and water desalination) are at various stages of readiness.Closed fuel cycle technologies:• Water chemical reprocessing technology for thermal reactor SNF, including uranium andplutonium separation and HLW vitrification, was demonstrated on an industrial level.• Mixed uranium-plutonium oxide fuel production technologies for BN reactors were demonstratedon an experimental level.• R&D started on alternative nuclear fuel cycle technologies (dry SNF reprocessing methods, minoractinides’ transmutation, uranium-thorium cycle technology).5. ConclusionConcluding this brief review of scenarios and issues of the nuclear energy development in Russia andin the world, the following could be stated with confidence:• The world is entering the system energy crisis. Nuclear energy development could stabilize theenergy market situation till the mid-century.• Russia is interested in accelerated nuclear power development, in order to preserve and use itsresources efficiently. So Russia is capable to contribute significantly to the solution of challengesfaced by the nuclear energy in the XXI century.6. References1. E.P. Velikhov et al., “Russia in the World Energy of the XXI Century”, Moscow, IzdAt, 2006.2. International Panel on Climate Change, Special Report on Emission Scenarios, CambridgeUniversity Press, Cambridge, 2002.

Marcus NilssonManager – Nordic Reactor ServiceWestinghouse Electric Sweden ABOffice: +46 21 3437722NORDIC NUCLEAR MARKET TRENDS1 BackgroundThe Nordic market consists of the Swedish nuclear operations at the three units inForsmark, three in Oskarshamn and four at Ringhals, in total 10 units in operation. Theresults of a Swedish public vote in early 80s made the country decide against nuclear inthe long run. This decisions lead to the closing of the nuclear units, Barsebäck 1 and 2.Furthermore, The Nordic market is also represented by Finland and the two units atOlkiluoto and two units at Lovisa. In Finland AREVA is in the progress of building anew, third, N.P.P. on the Olkiluoto site.Finally, At the Oskarshamn site the Swedish plants intermediate storage facility for spentfuel is located, CLAB or Centralt Lager Använt Bränsle. At this location spent fuel isshipped annually from all Swedish plants and stored for app. 30-50 years until finalrepository.2 Electricity marketIn the 90’s associated with the deregulation of the Nordic electricity market the price forelectricity was fairly low (with a low at

etc. The trends for the 2000’s and onwards have been to do major upgrades thru changeof hardware (e.g. reactor internals such as steam dryer and steam separator,generator/turbine upgrades), better/new fuel, safety upgrade and prolong the lifetime ofthe plants from 40 to 60 years. These are massive programs, the largest ones in the region(for nuclear) going on since the plants were built reaching up to 30% increased poweroutput.4 PULSOskarshamn unit 3 is currently underway with a major upgrade to 29% increasedelectrical output, from 1200 MWe to 1450 MWe. The project name is designated PULS,Power Uprate with Licensed Safety. The upgrade includes three parts: turbine, electricalsystems and reactor systems. Furthermore the project includes lifetime extension from 40to 60 years lifetime and adherence to new SKI (Swedish regulatory commission) safetyrequirements.For the reactor systems internals a change out will be carried out of the steamseparator/shroud head, steam dryer and main steam line valves. The recirculation pumpswill be rebuilt for improved capacity.5 Reactor internals replacement programIn the Nordic region several large reactor internal change outs have been carried outduring the past 8-10 years including:• Steam dryers – Olkiluoto 1&2, 2005/2006• Steam separators – Oskarshamn 1, 1998 & Oskarshamn 2, 2003• Core shroud head - Oskarshamn 1, 1998 & Oskarshamn 2, 2003• Top guide – Forsmark 1&2, 2000• Core shroud – Forsmark 1&2, 2000• And several more smaller/medium size projectsPlanned programs for the next five years include:• Core shroud head - Forsmark 1,2&3, 2008-2010 & Oskarshamn 3, 2008• Steam separators – Forsmark 1,2&3, 2008-2010 & Oskarshamn 3, 2008• Steam dryers – Forsmark 1&2, 2008-2009 & Oskarshamn 3, 20086 Segmentation for old reactor internalsFollowing the large Reactor internals replacement programs several techniques have beendeveloped to scrap/minimize volume of the old/replaced parts. The use of band saws incombination with hydraulic pliers/cutters have show most effective. Furthermore, no

creation of airborne contamination (e.g. plasma cutting can cause airborne contamination)and debris size suitable for vacuuming post disposal are advantageous. To handle thelarge components several turning table have been used in combination with tilting/turningdevices. Segmented parts have been put in containers for final disposal.7 Field ServicesThe low electricity price on the market has also generated a lot of pressure to drive downoutage times and increase availability. This has been achieved thru systematic approachto outage durations and obstacle removal to achieve this. The durations in the Nordicregion now shows systematic outage times less then 10-15 days “braker to braker” for afull refueling outage. The record low outages at Olkiluoto show durations just above 7days.One key driver for achieving this has been close cooperation between key suppliers andthe plants with long term contracts spanning over multiple services and really driving e.g.the lessons learned feedback process, long term planning and operational mode of theplants.Finally, with the rising electricity price the trend in maintenance scope is also less pricefocus and more drivers regarding quality of performance. Outages are short enough andthe main focus in the Nordic area is to remain short rather then shorten duration.

THE EFFECT OF ELECTRICITY GENERATING PARKRENEWAL ON FOSSIL AND NUCLEAR WASTE STREAMS:THE CASE FOR THE NETHERLANDSJ. HART, A.I. VAN HEEKNRGP.O. Box 25, 1755 ZG Petten, The NetherlandsABSTRACTThe paper concentrates on options for renewing the current Dutch electricitygenerating park in the next decades. For this purpose, the existing electric generating parkof The Netherlands is modelled according to its fuel use and waste generatingcharacteristics. The present electricity park consists of four types of generating plants: thesingle nuclear plant, gas-fired plants, coal-fired plants and renewable energy (mostlybiomass and wind).In this paper the effect of a generating park transition into one with a large share ofnuclear energy on the waste streams, both fossil and nuclear, is analysed. Two demandgrowth scenarios are used, and nuclear phase-out is taken into account for comparison. Forrenewables, existing literature on planning is referenced, as well as for energy demanddevelopment. This implies a substantial growth for these sources, but their contributionremains limited in percentage. Additionally, in the high-demand scenario the demandgrowth of 1.5%/year causes a more than doubling of the electricity demand in 2060compared to 2000. In the analyzed scenarios it is assumed that fossil fuels will becomeeconomically unattractive due to high CO 2 penalties, or even partly inaccessible due tophase-out by law. Then nuclear will substitute coal and gas to a large extent, growing to acontribution of more than 50% in 2060.In a dynamic analysis, i.e. as a function of time, the electricity supply distribution bysource is being determined with the DANESS and DEEA codes, as well as the emission ofCO 2 , SO 2 , NO x and high-level radioactive waste. By 2060, the CO 2 emission of thegenerating park with nuclear plants reduces to about one-third of that without. The nuclearsector is shared by the evolutionary reactor design EPR and the smaller-scale alternativePBMR. The additional CO 2 mitigation by the PBMRs in cogeneration mode is quantified aswell: the CO 2 emission of the Dutch electricity sector could even fall below zero when theavoided emission of industrial heating is subtracted from the CO 2 emission of the fossilfiredpower plants.When replacing fossil-generated electricity by nuclear, CO 2 and other gaseous wasteis traded for radioactive waste, the CO 2 amount being in the order of a million times theamount of radioactive waste. To reduce the amount of nuclear waste further, recycling canbe applied. The options of direct spent fuel storage and reprocessing are compared for theamounts of waste until 2060, both in mass and in volume. Obviously, reprocessing of spentfuel results in a significant reduction of volume that is needed to finally dispose usedradioactive materials in geological repositories. Also, much of the volume will be occupiedby PBMR pebble fuel elements. Separation of graphite from the fuel elements, and storingthe fuel particles only, would already bring a volume reduction of over 90% for this fueltype.1

1 IntroductionMost scenarios for electricity supply development for Western Europe assume a decline fornuclear generation in the coming decades, or a small increase followed by a decline, e.g. the Europeanstudy ‘European Energy and Transport, Trends to 2030 – update 2005’[1]. Some scenarios with higheconomic growth assume an increase in nuclear generation to cover the demand growth associatedwith the economic growth, e.g. [2].This study however considers a scenario where nuclear energy is deliberately employed forcoupled economic-environmental reasons, for a real country departing from an existing electricitygenerating park.The Netherlands currently has a generating park of 21 GWe (2004), running for three quarterson natural gas, see fig. 1. Already for some decades The Netherlands is a main gas producer itself,explaining the large gas share to electricity generation and the low nuclear share, compared to theEuropean mean nuclear share of 35%. However, the main gas source at Slochteren in the north of thecountry is expected to run out in 2030, and the smaller sources below the Wadden Sea at least before2050.So if the Netherlands don’t want to rely heavily on natural gas imports in the future, someform of transition has to take place in the electricity generating sector. The government already setfairly ambitious targets for renewable generation, and forced conservation by legal bans or rationing ofelectricity is beyond the way of current thinking. Current government plans indicate obligatory CO 2sequestration for new coal plants, making the coal option economically unattractive. So the nuclearoption remains the more obvious alternative to generate base-load quantities of electricity withexisting technology.RenewableNuclearCoalGasFigure 1Installed electricity generating capacity distribution in the Netherlands. Renewableincludes biomass and wind.2 The nuclear/renewable transition scenario for The NetherlandsAs electricity generation in the first place should stimulate prosperity and economy, no capitaldestruction by forced shutdown of power stations is envisaged. We depart from the existing electricitygenerating park, and the government stimulation plans for renewables are left intact.During the past 40 years, there has been an increase in the Netherlands of the electricityconsumption by a factor of 5.8, which implies an average annual growth of 4.5% [3]. It seemsunrealistic to extrapolate this growth rate for the next 50 years, considering the decrease in growth ofthe Dutch population. Recent studies consider more moderate growth rates. The Dutch study‘Referentieramingen’ (Reference Estimates), performed for the Government to forecast the Dutchenergy consumption and the resulting environmental impact up to 2020, considered annual growthrates of 1.7% for the ‘Strong Europe’ scenario, and 2.7% for the ‘Global Economy’ scenario [4]. On2

the other hand, the CASCADE MINTS project, funded by the European Union under the support ofthe 6 th RTD Framework Programme, considered annual growth rates ranging from 0.65% in 2010 toabout 0.35% in 2030 for the “Baseline” case [5]. The recent Dutch study “Deltaplan Kernenergie”assumed a constant annual growth rate for the electricity production of 1.5% up to 2060 [6].For the present study, two different growth scenarios have been considered (see also Fig.2):1. The scenario based on the assumptions of the “Deltaplan Kernenergie”, assuming a constantannual growth rate of 1.5%;2. The scenario based on the assumptions of the CASCADE MINTS project, assuming aconstant annual growth rate declining from 0.65% in 2010 to 0.35% in 2030.In addition, the following boundary conditions have been assumed:• Phase-out of coal-fired plants: the existing coal-fired plants are serving out their plannedlifetimes and no new ones are commissioned except those that already have been planned;• The contribution of renewable energy (wind, biomass) to the total electricity production is notdetermined by the market but by government planning. It will increase by 20% in 2020, andby 30% in 2040. These assumptions are in line with the forecast of the “Referentieramingen”[4];• A gradual deployment of nuclear reactors in the next decades. Presently, only one nuclearpower plant is operated in the Netherlands, the Borssele nuclear power plant. Various reactortypes are being offered today or will be offered in the coming years. For the present analysis, afleet consisting of one type of large reactor unit and one type of smaller unit, the latter suitablefor heat and power cogeneration, has been assumed for the next decades. For the large unit theEuropean Pressurized Reactor (EPR) was selected, and for the small unit the Pebble BedModular Reactor (PBMR).• For the cases with deployment of nuclear reactors, the options of direct disposal of spent fuel(“Once Through” case), and reprocessing of spent fuel (“Reprocessing” case) have beenconsidered.Installed Capacity (TWe/yr)460410360310260210'CASCADE' Growth Scenario: 0,65-0,35 %/yr'Deltaplan' Growth Scenario: 1,5 %/yr1602000 2010 2020 2030 2040 2050 2060 2070YearFigure 2Forecast of the installed electricity generating capacity in the Netherlands for two differentgrowth scenarios.For comparison reasons, a scenario taking into account the nuclear phase out option has beenconsidered, the nuclear phase-out scenario. For that scenario, an average growth rate for the electricityconsumption of 1.5% has been assumed.An overview of the main design parameters of the fossil-fuel fired plants and the nuclearreactors and fuel cycle is given in Table 1.3

Table 1Average design parameters of the different facilitiesGas Fired Coal Fired Borssele NPP EPR PBMRPlant PlantPower per plant (MWe) 400 520 484 1600 160Plant lifetime (yr) 35 30 26 60 1 50 1Plant capacity factor (-) 0,85 0,85 0,93 0,91 0,95Efficiency factor (-) 0.45 0.39 0.35 0.37 0.41Plant construction time (yr) 2 3 (existing plant) 5 3Expected overnight cost (B€) 0,30 0,45 (existing plant) 1,3 0,19O&M Cost, (Euro/MWhe) 8,6 2 14,2 2 4,6 3,9 4,4UO 2 Enrichment (%) - - 3,1 4,2 8,1Burnup (GWd/tHM) - - 33 50 901 Including lifetime extension2 inclusive price of CO 23 Computer toolsDynamic Energy Economics Analysis (DEEA) is a system dynamics tool which is able tosimulate scenarios for the future deployment of fossil-fuel, nuclear and renewable energy systems.Driven by a future energy demand, new energy systems are introduced by means of a decision modelthat is mainly based on the profit per MWhe for each of the different electricity-generating options.DEEA is a macroscopic tool and intended to provide relatively quick results. This brings about that thecode models are relatively straightforward, taking into account the overall processes and avoiding toomuch details. Seven types of nuclear reactors as well as gas-fired and coal-fired fossil fuel plants arecharacterized by gross data. Renewables are simply modelled by power and energy demand growthrate. The economics model takes into account interest and discount rates and the price of electricity,and compares this with economic factors that are specific for each energy generating system (e.g.levelized cost, fuel cost, carbon tax).Given a future energy demand, DEEA calculates the relative contributions of nuclear, fossilfuel, and renewable energy systems to the total energy production. The development of the nuclearenergy production in time then serves as the boundary condition of a detailed analysis of the nuclearfuel cycle. This analysis is performed with the DANESS computer tool.For the assessment of the nuclear fuel cycle strategies, the DANESS code (“Dynamic Analysisof Nuclear Energy System Strategies” [7]), Version 3.2.03, was used to simulate the flows of fissilematerial, fresh fuel, spent fuel, high level waste as well as all intermediate stocks and fuel cyclefacility throughput. DANESS is an integrated dynamic nuclear process model for the analysis oftoday’s and future nuclear energy systems on a fuel batch, reactor, and country, regional or worldwidelevel. Starting from today’s nuclear reactor park and fuel cycle situation DANESS analyzes energydemanddriven nuclear energy system scenarios over time and allows the simulation of changingnuclear reactor parks and fuel cycle options. New reactors are introduced based on the energy demandand the economic and technological ability to build new reactors. The technological development ofreactors and fuel cycle facilities is modelled to simulate delays in availability of technology. Levelizedfuel cycle costs are calculated for each nuclear fuel batch for each type of reactor over time and arecombined with capital cost models to arrive at energy generation costs per reactor and, by aggregation,into a cost of energy for the whole nuclear energy system. A utility sector and government-policymodel are implemented to simulate the decision-making process for new generating assets and newfuel cycle options. The different functionalities of DANESS may be switched on or off by the useraccording the intended use. The architecture of the DANESS code is depicted schematically in fig. 3.For the calculation of the amount of nuclear waste, a fuel cycle model is used, as shown in fig.4. Properties of all fuel cycle facilities are input, including capacity and transition time. For each4

eactor, a fuel type and back-end route (direct storage or/reprocessing) is set. The amounts of waste aregiven in tonnes heavy metal (tHM), and converted to volumes in m 3 for the results in chapter 8.Figure 3Schematics of the architecture of the DANESS code.Allocation MatricesU natROW UseAllocated REPUSeparated REPUReservedU natUnat ReservationAllocated PuAllocated MASeparated PuSeparated MAReprocessingHLW InterimStorageConversionReprocessing FractionsLegacySpent FuelHLWConditioningFuel Fabrication DemandEnrichmentOperational ReactorsSpent FuelInterim StorageEnriched Uof U nat Fuel Fabrication Fresh FuelLoadedReload FuelSpent FuelAt-Reactor StorageSFConditioningDULoadedInitial FuelGeological DisposalNew Reactors StartFigure 4Fuel cycle model in DANESS code.4 Electricity supplyThe electricity supply distribution over the available sources is determined for the threeselected scenarios:• Nuclear/renewable transition with high demand rise: fig. 5,• Nuclear/renewable transition with low demand rise: fig. 6,• Nuclear phase-out with high demand rise: fig. 7.5

The initial rise in electricity production is caused by the deployment of newly-built fossil-fuel powerplants (e.g. 800MW Sloe generating plant, 800MW gas plant Eemshaven, and several others), whereasexisting plants are not yet shut down. Around 2040 a ‘bend’ in the integrated curves can be observed:after the phase-out of coal-fired plants, the demand growth is fully covered by nuclear+renewables, sono additional growth of electricity from gas-fired plants is needed.It can be seen that, with the prescribed growth rate of the renewables, in the high demandscenario nuclear energy will become the largest electricity source with 54% in 2060, whereas in thelow demand scenario the renewables take the largest share with 52% in that year. In the nuclear phaseoutscenario the electricity need that is not covered by the renewables is almost equally sharedbetween gas and coal.454035Nuclear PowerRenewable PowerFossil Power (Coal)Fossil Power (Gas)Produced Electricity (GW)3025201510502005 2012 2020 2028 2036 2044 2052 2060Time (Year)Figure 5 Produced electricity taking into account a annual growth rate of 1.5%.454035Nuclear PowerRenewable PowerFossil Power (Coal)Fossil Power (Gas)Produced Electricity (GW)3025201510502005 2012 2020 2028 2036 2044 2052 2060Time (Year)Figure 6Produced electricity taking into account the low demand scenario.5 Fossil waste generationThe amounts of the gaseous waste emissions from the fossil-fired stations for the next decades havebeen depicted in fig. 8, 9 and 10. Fig. 8 gives the carbon dioxide (CO 2 ) emissions, fig. 9 the nitrogenoxide (NO x ) emissions, and fig. 10 the sulphur dioxide (SO 2 ) emissions. The SO 2 emissions result forthe largest part from the combustion of coal. Although in the Netherlands it is required to implement6

measures to reduce the SO 2 emissions from coal-fired plants, approximately 10% of the total generatedSO 2 still is released to the atmosphere. Through better SO 2 reduction methods this percentage is likelyto decrease in the future so that the estimated values shown in fig.10 represent upper limit values.454035Nuclear EnergyRenewable PowerFossil Energy (Coal)Fossil Energy (Gas)Produced Electricity (GW)3025201510502005 2012 2020 2028 2036 2044 2052 2060Time (Year)Figure 7Produced electricity taking into account the nuclear phase-out scenario with 1.5% growthrate. The thin line for nuclear energy is the single existing plant serving out its licensedlife.The difference between the nuclear and non-nuclear scenarios is obvious: for 1.5% demandrise the accumulated CO 2 emission of the nuclear scenario in 2060 is only 12% of that of the nuclearphase-out scenario. In other words, by introducing nuclear energy on the proposed scale, 88% of theCO 2 emission of the Dutch electricity generating park can be avoided. The NO x release rate plunges atleast 90% (fig.9), whereas the SO 2 release vanishes as a result of phase-out of coal-fired plants(fig.10), which account for 95% per Gigajoule for the SO 2 release. These trends clearly indicate thatthe deployment of nuclear power for electricity generation is a serious option to reduce significantlythe emission of hazardous exhaust gases.CO2 Release Rate (Mtonnes/year)16014012010080604020Blue: "Deltaplan Kernenergie"Black: CASCADE BaselineRed: No Nuclear80007000600050004000300020001000Cumulative CO2 Release (Mtonnes)002000 2010 2020 2030 2040 2050 2060 2070YearFigure 8CO 2 release rate (thick curves) and cumulative CO 2 release from 2005 on (thin curves) forthe three considered scenarios.7

140NOx Release Rate (kilotonnes/year)12010080604020"Deltaplan Kernenergie"CASCADE BaselineNo Nuclear02000 2010 2020 2030 2040 2050 2060 2070YearFigure 9Calculated NO x release rate for the three considered scenarios.60SO2 Release Rate (kilotonnes/year)5040302010"Deltaplan Kernenergie"CASCADE BaselineNo Nuclear02000 2010 2020 2030 2040 2050 2060 2070YearFigure 10 Calculated SO 2 release rate for the three considered scenarios.6 Deployment of nuclear energyThe deployment of nuclear reactors (EPR, PBMR) as calculated by DANESS, is depicted infig. 11 and fig. 12. DANESS calculates a relatively larger deployment of the small-scale PBMRs inthe “CASCADE” case as compared to the “Deltaplan” case, because of the slower growth rate of theenergy demand in the “CASCADE” case, which makes it less attractive to deploy the larger-capacityreactor types (1600 MW EPRs). In case of slower energy demand growth rates, the deployment oflarge reactors would result in a significant over-capacity of produced electricity. The more gradualdeployment of the smaller PBMRs leads to a better match of the demand of electricity. From about2040 there is less growth in the demand curve for nuclear energy, so the deployment of the smallercapacityPBMR reactors is preferred above the large-capacity EPR reactors for the reason of demandmatching. Therefore no additional EPRs are foreseen.8

30PBMR25EPRBorssele NPPProduced Electricity (GW)201510502005 2012 2020 2028 2036 2044 2052 2060Time (Year)Figure 11 Calculated deployment of nuclear reactors for the case “Deltaplan Kernenergie” (1.5%annual growth rate)14Produced Electricity (GW)1210864PBMREPRBorssele NPP202005 2012 2020 2028 2036 2044 2052 2060Time (Year)Figure 12 Calculated deployment of nuclear reactors for the CASCADE case (moderate annualgrowth rate)7 Effect of nuclear cogeneration on CO 2 mitigationOriginally the development of the high-temperature gas-cooled reactor was started to be ableto supply not only electricity, but also process heat and cogeneration to various sectors of industry.From previous studies (e.g. [8]), it was demonstrated that high-temperature gas-cooled reactors arecapable to deliver, apart from electricity, a significant amount of heat, i.e. up to about 30% of the totalthermal power.In fig. 13 the CO 2 release is depicted for the three considered scenarios, now for the nuclearscenarios also depicting the additional avoided CO 2 emission when using the PBMR in cogenerationmode. It can be seen that the CO 2 emission of the Dutch electricity sector even falls below zero whenthe avoided emission of industrial heating is subtracted from the CO 2 emission of fossil-fired powerplants.9

CO2 Release Rate (Mtonnes/year)160140120100806040200Blue: "Deltaplan Kernenergie"Black: CASCADE BaselineRed: No Nuclear-202000 2010 2020 2030 2040 2050 2060 2070YearFigure 13 CO 2 release rate for the three considered scenarios. Thick curves: including additionalavoided CO 2 from the deployment of heat cogeneration by PBMRs. Thin curves: noadditional avoided CO 2 considered.8 Spent nuclear fuel and nuclear wasteThe amounts of nuclear wastes that have to be taken care of, have been calculated as afunction of time for two options: direct disposal (‘Once Through’) and recycling (‘reprocessing’). Forthe two options, the amount of waste in interim storage facilities is shown in fig. 14, and the amount ofwaste in final disposal in fig. 15. As can be seen in the fuel cycle model scheme of figure 4, spent fuelfirst moves from the reactor to the ‘At Reactor’ storage, consisting of the spent fuel storage ponds atthe nuclear plants. After a certain cooling down period at the reactor storage, in this study set to 5years, the spent fuel is transferred to the spent fuel interim storage facility, where it is able to cooldown further. After this, it can take two routes:• It is sent to the spent fuel conditioning facility where it is treated for final disposal in a finaldisposal facility;• It is sent to a reprocessing facility, where uranium and plutonium are recovered after which theremaining high level waste is transferred to a HLW interim storage facility for a cooling downperiod. In the high level waste conditioning facility the waste is prepared for final (geological)disposal.In fig. 14, the black curve indicates the total amount of spent fuel stored at the nuclear powerplants for the high demand scenario. The blue lines indicate the amount of spent fuel stored in interimstorage. It can be seen that after about 2045 the waste arisings decrease as a result of the decreasedgrowth in nuclear demand (cf. fig.11). The difference between the two blue lines is reflected by thethin red line: the amount of high level waste coming from the reprocessing plant.In fig. 15, the amount of high level waste coming from the reprocessing plant is still very lowin 2060. This is primarily caused by long transit and waiting periods for reprocessing. For this casestill much high level waste is in the pipeline and will arrive at the final repository after 2060. Thisillustrates the contradiction of societal demand for an operational final storage facility at the start of anuclear expansion programme on the one hand, and on the other hand the actual arriving of high levelwaste from the reprocessing plant only several decades later.10

2500Spent Fuel - at Reactor Storage2000Spent Fuel - Interim StorageHigh Level Waste - Interim StorageAmount of Waste (thm)1500100050002000 2010 2020 2030 2040 2050 2060YearFigure 14 Amount of wastes in storage arising from the nuclear reactors for the scenario “Deltaplan”.70006000Spent Fuel - Geological DisposalHigh Level Waste - Geological DisposalAmount of Waste (thm)5000400030002000100002000 2010 2020 2030 2040 2050 2060YearFigure 15 Amount of wastes in geological disposal for the scenario “Deltaplan Kernenergie” – thickcurves: once-through case; thin lines: reprocessing caseFor the low demand “CASCADE MINTS” scenario, the predicted amounts of stored waste inthe year 2060 are about 30% to 55% less, depending on the type of waste, in comparison with thehigh-demand “Deltaplan” scenario.In fig. 16, the actual container volumes are depicted that are needed to for interim storage ofspent fuel, high-level wastes and the PBMR pebbles. For vitrified high level waste, the COGEMAHLW container [9] is considered, and for spent fuel the ONDRAF-NIRAS design [10]. For the PBMRpebbles, the German design storage canister [11] is adopted. We see the volume of the high levelwaste containers completely vanishing against the large volume of containers holding non-reprocessedspent fuel. For the most part, the larger SF container volume comes from containers holding PBMRpebbles, that mainly consist of graphite and only 7% of nuclear fuel. For the CASCADE scenario, thespent fuel volume rises only to 8500 m 3 in 2060, that is 71% of the volume in that year for theDeltaplan scenario. This can be seen in the bar chart of fig. 17 as well, where the spent PBMR pebblefuel is indicated separately.From fig. 17 it is also clear that the used PBMR pebbles require by far the most storage anddisposal capacity, because for the PBMRs also the moderator material (matrix graphite) of the fuelpebble is also considered as waste. The volumes that are needed to store and dispose HLW are only aminor fraction of the total required volumes.The growth of the volume of waste containers in geological disposal over time for the wholeDutch nuclear park is similar to fig. 15, with a volume of 42000 m 3 in 2060 for the case of no11

eprocessing for the high electricity demand case, and 28000 m 3 for the low demand case. Fig. 18 iscomparing the volume of waste containers in the final storage facility for the two demand cases in theyear 2060, distinguishing between EPR and PBMR waste. For PBMR, also the amount of wasteemerging when recycling the graphite, storing only the coated particle fuel (still no reprocessing). Thismeasure already would reduce the PBMR spent pebble volume by 92%.Table 2 lists the calculated effective volumes of the waste canisters that are needed to containthe spent fuels, high-level wastes, and PBMR pebbles.1400012000Spent Fuel - Interim StorageHigh Level Waste - Interim StorageVolume of Waste Canisters (m 3 )10000800060004000200002000 2010 2020 2030 2040 2050 2060YearFigure 16 Volume of wastes containers in interim storage arising from the nuclear reactors for thescenario “Deltaplan Kernenergie” – thick curve: once-through case; thin line: reprocessingcase.Canister volume (m3)140001200010000800060004000At Reactor Storage (SF)Interim Storage (SF)Interim Storage (HLW)At Reactor Storage (Pebbles)Interim Storage (Pebbles)20000"Deltaplan" Once-Through"Deltaplan"Reprocessing"CASCADE" Once-Through"CASCADE"ReprocessingFigure 17 Comparison of the expected waste volume in storage at the nuclear plants and in theinterim storage facility in the year 2060.12

4000030000DisposalHigh Level WasteSpent FuelPBMR PebblesPBMR Pebbles, Recycled GraphiteCanister Volume (m 3 )20000100000"Deltaplan"Once-Through"Deltaplan"Reprocessing"CASCADE"Once-Through"CASCADE"ReprocessingFigure 18 Comparison of the volume of waste containers in the final storage facility for the twodemand cases in the year 2060, distinguishing between EPR and PBMR waste. The casefor recycling the PBMR graphite, storing only the coated fuel particles, is shown as well.Table 2EffectiveVolumeSpent LWR fuelHigh LevelWaste(reprocessedSpent PBMRpebble fuelComparison of canister volumes needed for the storage and disposal of nuclear wastes forthe different scenarios in the year 2060.Type ofstorageDeltaplan Deltaplan CASCADE CASCADEOnce-Through Reprocessing Once-Through ReprocessingInterim 1468 1517 489 514Final 5777 0 2038 0Interim 0 61 0 21fuel) Final 7 27 7 20Interim 10475 10729 7976 8375Final 35899 0 25675 0Final, recycledgraphite 2764 0 1977 09 Comparison with existing interim storage capacity in The NetherlandsThe facility for interim storage of spent fuel and high level nuclear waste is called HABOG(‘Hoogradioactief Afval Behandelings- en Opslag Gebouw’, Highly-radioactive Waste Treatment andStorage Building). It is located near the city of Vlissingen and the Borssele nuclear power station inthe south of the country [12]. The HABOG-building is a modular building. This means the buildingcan be extended if necessary. At this time there are three vaults for the storage of heat generatingwaste and three bunkers for the storage of non-heat generating waste. The license permits only a fullload of two of the three vaults or bunkers. It should always be possible to unload one vault or bunkerfor inspection.The capacity of each vault is 135 canisters with vitrified waste and 35 canisters with spent fuel. Thismeans a total capacity at this moment of 270 canisters with vitrified waste and 70 canisters with spentfuel. The capacity of 2 bunkers is approximately 600 drums with different types of conditioned waste.The total volume of all the waste will be 750 m 3 .13

In the high demand case, this capacity will already be used by 2024, and in the low demand case in2028. The amount of equivalent HABOG capacities for the three cases (no reprocessing, noreprocessing but with graphite recycling for the PBMR waste, and reprocessing) per demand scenarioin 2060 can be seen in fig. 19. Most storage capacity is needed in the case of high demand and directstorage of all spent fuel: 72 times the current HABOG capacity. This can be reduced to 28 byrecycling the graphite of the PBMR spent fuel, and to 17 by recycling the fuel itself (reprocessing).The figures for the low demand scenario are accordingly lower.807060504030once throughgraphite recyclingreprocessing20100deltaplancascadeFigure 19 Equivalent number of HABOG volume capacities needed in 2060 for the case withoutreprocessing, the case without reprocessing but with graphite recycling, and the case withreprocessing.10 ConclusionsTable 3 summarizes the amounts of waste generated for the high demand (‘Deltaplan’)scenario and the low demand (‘CASCADE’) scenario. In the case of the deployment of nuclearreactors, in 2060 the release of CO 2 as a result of electricity generation will be reduced to one-third ascompared to the case where nuclear electricity generation is not considered.Choosing between nuclear and non-nuclear generation parks is a trade-off of waste types. Alltypes of energy mix come with a waste mix. When replacing fossil-generated electricity by nuclear,CO 2 and other gaseous waste is traded for radioactive waste, the CO 2 amount being in the order of amillion times the amount of radioactive waste. By signing the Kyoto protocol, The Netherlandsobliged itself to reduce CO 2 emissions by 13 Mton/year in 20 years time. By implementing thenuclear/renewable transition scenario, this target could be more than achieved by the electricitygenerating sector alone (factor 1.8), leaving room for other sectors with less possibilities for CO 2reduction.The following conclusions with respect to nuclear waste reduction can be drawn from thisstudy:• Reprocessing of spent fuel results in a significant reduction of volume that is needed to finallydispose used radioactive materials in geological repositories.• Reprocessing of spent fuel impels the deployment of capacity to separate the recyclable materialfrom the HLW.• In case of not reprocessing, most of the space in the interim and final storage facilities is occupiedby spent PBMR fuel. By recycling the graphite part of the waste, a significant volume reduction canbe achieved.14

Table 3Cumulative CO 2 Release(Mton)Cumulative NO x Release(kton)Cumulative SO 2 Release(kton)Spent FuelInventory(tHM)High LevelWasteInventory(tHM)Comparison of amounts of waste of fossil and nuclear origin.No NuclearDeltaplanOnce-ThroughDeltaplanReprocessingCASCADEOnce-Through6700 2430 21305850 2160 58502075 510 2075CASCADEReprocessingAt ReactorStorage 1825 1825 770 770InterimStorage 2075 1794 923 808GeologicalDisposal 6616 0,0 2886 0,0InterimStorage 0,0 193 0,0 133GeologicalDisposal 19,9 125 19,8 60,111 References[1] European Commission, Directorate General for Energy and Transport, European Energy and transport,Trends to 2030 – update 2005, Office for Official Publications of the European Communities,Luxembourg, 2006.[2] IIASA-WEC (International Institute for Applied Systems Analysis and World Energy Council). GlobalEnergy Perspectives to 2050 and Beyond. Laxenburg, Austria: International Institute for AppliedSystems Analysis, 1995.[3] Centraal Bureau voor de Statiestiek – “Statistics Netherlands”, http://www.cbs.nl/[4] Van Dril, A.W.N., and H.E. Elzenga, “Referentieramingen energie en emissies 2005-2020”, ECN-C—05-018, Petten, May, 2005.[5] Uyterlinde, M.A., et al., “The contribution of nuclear energy to a sustainable energy system”, Vol. 3 inthe CASCADE MINTS project, ECN-C—05-085, Petten, March 2006.[6] Crok, M., A. Jaspers, E. Vermeulen, “Deltaplan Kernenergie – Knopen doorhakken over deenergievoorziening”, Natuurwetenschap en Techniek, March 2006.[7] L. Van den Durpel, A. Yacout, “Dynamic Analysis of Nuclear Energy System Strategies”, UsersManual, Argonne National Laboratory, Argonne, February 2004; alsohttp://www.daness.anl.gov/index.html.[8] A.I. van Heek, M.M. Stempniewicz, D.F. da Cruz and J.B.M. de Haas: ACACIA: a small-scale powerplant for near term deployment in new markets. Nuclear Engineering and Design 234 (2004) p.71-86.[9] De Bock, C., L. Londe, P. Blümling, T. Rothfuchs, B, Breen, F. Huertas Ilera, “Input Data andFunctional Requirements for buffer construction and monitoring equipment”, Deliverable 1 ofESDRED Module 1, Work Package 1, EC 6FWP Contract Number: FI6W-CT-2004-508851, 23December 2004.[10] Sillen, X., and J, Marivoet, “Spent Fuel Performance Assessment for a hypothetical repository in theBoom clay at the Mol site (Belgium), SCK.CEN, BLG-877, NIRAS KNT 90.95.656.02, Mol, Belgium,February 2002.[11] K. Kugeler, R. Schulten, Hochtemperaturreaktortechnik, Springer Verlag, 1989.[12] HABOG: One Building for All High-level Waste and Spent Fuel in the Netherlands. - The First Yearsof Experience, J. Kastelein, and Dr. H.D.K. Codée, 11 th International High-Level Radioactive WasteManagement Conference, April 30-May 4, 2006, Las Vegas.15

Session 18.1.4:Status of future projects

PRELIMINARY SEISMIC ANALYSIS OF A NEXTGENERATION NPPG. FORASASSI, R. LO FRANODIMNP- University of Pisa,Via Diotisalvi, n° 2, 51126 Pisa (Italy)Tel. +39-050-8386415, fax +39-050-836665E-mail: rosa.lofrano@ing.uipi.itABSTRACTThe aim of paper is to evaluate and to characterize the structural response behaviour ofreactor building internal structures under specific site seismic loading characteristics inorder to determine whether these ones satisfy present international safety regulations.Moreover, the correlations between the horizontal seismic earthquake values recorded onrock site as well as the calculated NRC ones and the in site complex nuclear buildingstructure effects are investigated and discussed, with an application example to near termnuclear power plants (NPPs) concepts (like IRIS or ELSY).To the purpose finite element method and sub-structures approach were employed forstudying the overall dynamic behaviour of the considered system also accounting for thestructure and soil interactions. The analysed results, the mentioned effects and the responseof internal components (e.g. Nuclear Building, Vessel, etc.) seem to confirm the possibilityto achieve an upgrading of geometry and performances of the proposed solutions for theconsidered NPPs.1. IntroductionEarthquake response of nuclear structures depends on both the ground motion characteristics and thedynamic properties of the structures. Integrity of structures, systems and components of a nuclearpower plant must be ensured in case of any design condition, in particular in the case of seismicaccident conditions. In fact, when a structure is subjected to dynamic loads, as seismic ones, thebehaviour of structural material may be significantly different from the one characteristic of static loadapplications. The seismic analysis of a nuclear power plant is one main regulatory requirement for thedesign and construction approval [1].The adopted analysis procedure provides minimum requirements and acceptable methods for theevaluation of safety related structures of NPPs. Moreover Soil Structure Interaction (SSI) is consideredto be important because take into account the phenomenon of coupling between a structure and itssupporting medium (soil, sand or rock) during an earthquake, due to the nonlinear behaviour of severaltype of soil etc.In this paper a preliminary application of the proposed analysis methodology to an innovative LWRreactor (International Reactor Innovative and Secure- IRIS) structure is presented. This preliminaryanalysis is intended to evaluate the dynamic loads propagation from the ground to the Internals (e.g.Reactor Pressure Vessel (RPV) or Steam Generators (SG)), considering also the mentioned SSIseffects.2. Model and structural system descriptionsBetween the new LWR concept IRIS is one of the most interesting new reactor concept under study atpresent. The IRIS integral pressure vessel (RPV) is larger than a traditional PWR one, but the size ofthe IRIS containment system (CS) volume is a fraction of that of corresponding loop reactors,resulting in a significant reduction in the overall reactor size [2-3]. This size reduction, combined withthe spherical geometry results in a CS pressure bearing capability at least three times higher than atypical loop reactor cylindrical containment, with the same metal thickness and stress levels (Fig.1).NPPs are always composed of a number of adjacent structures, so in order to ensure adequate

treatment of interaction effects the main buildings should be considered, as in the proposed examplemodel, with their real geometry and material characteristics.Fig. 1 – Scheme of the whole Nuclear IslandThe seismic response of a structure should be determined by means of setting up an adequatemathematical model and calculation of its response to the prescribed seismic input. In this applicationexample, the Nuclear Island may be subdivided into three main structures:• Auxiliary building including External Building (EB);• Inner containment structures (CS);• Containment internals including Reactor Pressure Vessel (RPV).In the considered EB the reactor type surrounds the CS. The overall structure is assumed to have arigid foundation, which is the interface between the nuclear island and the soil. The Soil was modelledas a homogeneous loose sand zone, which may influence in different ways the horizontal and verticalpropagation wave and the rocking vibration effects [4-5]. The clearly nonlinear constitutive behaviourof soil should be accounted for as an elasto-plastic Mohr-Coulomb material. The CS was one of themain structures studied, which was characterized by different mass and stiffness distribution over theheight; due mainly to the upper hemispherical steel structure and to the bottom concrete wall structure.The main internal structures such as the RPV and the suppression water pools content was consideredas lumped masses connected to the containment wall nodes. The RPV internals (e.g. Barrel, SG tubes,etc.) are considered as a set of lumped masses linked respectively to the appropriate locations.Moreover the attachments of the SG headers to the RPV internal wall were considered as rigidrestrains without mass.2.1 Method of analysisIn the numerical simulations, three dimensional models (MSC.MARC FEM code) were implementedin order to analyze the seismic behaviour of the whole nuclear island with and without soil-structureinteraction (Fig. 2). In order to ensure adequate treatment of interaction effects, firstly the mainstructures were modelled in only one 3D finite element model with some simplified assumptions, andsubsequently the seismic analysis was carried out by means of the Substructure model approach thatallows to separate the NPP seismic analysis problem into a series of simpler ones that can be solvedeach independently. Simplified structural models may be used to provide an adequate representation ofconsidered structures and to generate in-structure response spectra at the reference location orsubsystem supports [6].The performed analyses referred to the same structures coupled with the foundation depth and soileffects. The whole model was represented by a cylindrical structure resting on a shallow cylindricalfoundation that was embedded in a homogeneous soil layer. Moreover to simplify the analyses andreduce the calculation time some internal structures (e.g. RPV, SGs, etc.), in each models, wererepresented like lumped masses distributed at appropriate chosen locations. The Time Historyapproach was used in all analyses, coupled with the before mentioned Substructure method, toevaluate the effects of a Safety Shutdown Earthquake (SSE).

(a)(b)Fig. 2 (a),(b)- Complete structural model with and without SoilThe seismic excitation was simulated by means of artificial acceleration having the maximum PeakGround Acceleration (PGA) equal to 0.3 g calculated for an appropriate damping in according to theNRC Regulatory Guide 1.60 only in the horizontal translation direction and for excitation durationequal to 30s.Acceleration (g)21,510,500,1 1 10 100Frequency (Hz)0,50% 2% 5% 7% 10%Fig.3 – Input Response Spectra (PGA = 0.3g)To study the effectiveness of the damping system in mitigating the seismic response of the buildings,the maximum accelerations and displacements of the considered structures at chosen reference pointwere obtained from the results of each analysis.3. Numerical resultsThe input motion of the SSE is used to carry out results referred to the complete nuclear island modelincluding all main structure, with (Model A) and without soil (Model B) highlighting the loadingintensity decrease as the seismic input moves from the free field through the soil, EB and CS to theRPV and to the SGs tube restraints. An overview of the acceleration and displacement time histories,through the CS to RPV and through RPV are showed in Figures 4(a) and (b) and 5 (a) and (b)respectively. The response spectra (Figures 6 (a) and (b)) into the frequency domain, indicates that ifthe soil is considered (Model A) the effect of embedment on structure lead to a reduction in structureresponse due to the increased amplitudes damping effect. The SSI coupling effect results fromscattering of waves from the foundation and the transfer of energy from the structure due to structuralvibration [7]. The system damping increases considerably with increasing the foundation embedmentand the layer depth, especially for low-rise structures. Deeper is the stratum; greater is the influence ofthe embedment [8]. If the soil is not considered (Model B), the transfer function indicates that it was adecrease of seismic acceleration from the ground to the tube bundle where it is shown the transfereffect from the ground field to the RPV and from RPV to the SG tubes bundle restraints, highlighting,in this latter case, an amplification of the peak acceleration due to the in-plane internal structuresflexibility.

Acceleration (m/s 2 )Acceleration (m/s 2 )2,001,000,000 5 10 15 20 25 30-1,00-2,003,002,001,000,00-1,00-2,00-3,00Time (s)0 10 20 30Time (s)(a)Displacement (m)Displacement (m)0,050,00-0,05-0,10-0,15-0,200,05-0,10-0,15-0,200 5 10 15 20 25 30Time (s)0,00-0,050 5 10 15 20 25 30Time (s)(b)Figs. 4 (a), (b) – Acceleration and displacement to RPV- Model A and B2,000,05Acceleration (m/s 2 )Acceleration (m/s 2 )1,000,000 5 10 15 20 25 30-1,00-2,003,002,001,00-2,00-3,00Time (s)Bottom_SG Upper_SG0,00-1,00 0 5 10 15 20 25 30SG_bottomTime (s)SG_upper(a)Displacement (m)Displacement (m)0,00-0,050 5 10 15 20 25 30-0,10-0,15-0,200,05-0,10-0,15-0,20Time (s)Bottom_SG Upper_SG0,000-0,0510 20 30SG_bottomTime (s)SG_upper(b)Figs. 5 (a), (b) – Acceleration and displacement to upper and bottom SG restraints Model A and BAcceleration (g)0,80,60,40,20,00 1 10 100Frequency (Hz)Acceleration (m /s 2 )1,000,800,600,400,200,000 1 10 100Frequency (Hz)Bottom_SG Upper_SG(a)

Acceleration (g)0,90,60,30,00 1 10 100Frequency (Hz)Acceleration (g)1,51,20,90,60,30,00,1 1 10 100Frequency (Hz)Upper SG Bottom SG(b)Fig. 6(a), (b) – Response spectra to the RPV and SGs restraints- Model A and BAnalysing the response spectra it can be generally observed that there is a slight shift in thefundamental frequency of the building and a reduction in the spectral accelerations when usingcoupled models.4. ConclusionAnalysis and design of the NPP structures involve considerations not only on the available geometrybut also on the capacity of the most important structural members that transfer the seismic inertialloads from their application points.An overview of possible seismic analysis approaches has been provided, with particular emphasis onthe integral layout of the reactor coolant system accounting for the Soil Structure Interaction as well asStructure-Structure Interaction.Analysing the calculated response spectra it can be generally observed that the SSI and adjacentbuilding interaction results in rather slight shift in the fundamental frequency of the building. Theyalso depend on the dynamic properties of each structure such as strength, rigidity, and modalcharacteristics. Soil-structure interaction gives rise to kinematic and inertial effects, resulting inmodifications of the dynamic properties of the structure and the characteristics of the ground motionaround the foundation. It was shown that the effects of foundation embedment and SSI are extremelyimportant. They increase considerably the effective damping of the system relative to the damping ofthe structure alone.On the base of these very preliminary analyses, the effects of the described alternative nuclear buildingin soil embedment have been considered in order to check the possibility to achieve an upgrading ofthe NPP geometry and obtain a feed back on the critical design features (if any).5. References1. Y. Zhao et al. “Nonlinear 3-D dynamic time history analysis in the reracking modifications for a nuclearpower plant”, Nuclear Engineering and Design 165 (1996) 199 211.2. M.D. Carelli et al., “The design and safety features of the IRIS reactor”, Nuclear Engineering and Design230 (2004) 151–167.3. L. Cinotti, C. V. Lombardi et al.,”Steam generator of the International Reactor Innovative and Secure”,ASME Proceedings of ICONE 10, 2002.4. A. S. Arya, P. Nandakumaran, ”Dynamic Soil-Structure Interaction”, Int. Symposium on Soil-StructureInteraction, Jan.3-7 1977.5. Y. Kitada et al., “Models test on dynamic structure–structure interaction of nuclear power plant buildings”,Nuclear Engineering and Design 192 (1999) 205–2166. G. Forasassi, R. Lo Frano et al., “Seismic response of reactor vessel internals in the IRIS reactor”, Proc.International Congress on Nuclear Engineering, ASME, July ‘06, Miami Florida, ISBN: 0-7918-3783-1.7. ASCE Standard, “Seismic Analysis of Safety-Related Nuclear Structures and Commentary”, ASCE 4-98.8. G. Forasassi, R. Lo Frano, “Earthquake response analysis of reactor vessel internals for next generationreactors” Proc. of 19 th Int. Conference on Structural Mechanics in Reactor Technology, August ‘07,Toronto-Canada.

A NEW APPROACH FOR THE SEISMIC ANALYSISAND DESIGN OF THE IRIS REACTORS. DE GRANDIS, G. BENAMATIENEA-CR Brasimone40032 Camugnano (BO), ItalyG. BIANCHI, D. MANTEGAZZA, F. PEROTTIDepartment of Structural EngineeringPiazza Leonardo Da Vinci, 20133 Milano, ItalyL. CORRADI DELL’ACQUADepartment of Nuclear Engineeringvia Ponzio 34/3, 20133 Milano, ItalyS. MONTIENEA Bolognavia Martiri di Montesole 4, 40129 Bologna, ItalyABSTRACTThe ambitious safety goal for the IRIS reactor requires that both internal and externalevents are duly considered and treated in the PSA, with the Safety-by-Design approachadopted to reduce the overall Core Damage Frequency (in the range of 10 -8 events/year).As far as the seismic event is considered, a suitable approach has to be pursued, trying toeliminate unnecessary conservatism. Therefore, an innovative methodology for theevaluation of seismic fragility, applicable both to conventional and innovative reactorconcepts, has been developed and is here presented The two central elements of theprocedure are the use of the Response Surface Methodology (RSM) for describing theinfluence, on structural response and integrity, of all the parameters, hypotheses andmodelling criteria assumed as uncertain or random and a two-stage approach in thestructural modelling of the reactor building.1. IntroductionThe large number of seismic PRA studies performed in recent years on nuclear power plants hasshown that earthquakes are among the most important external events affecting NPP safety. In theframework of a seismic PRA, therefore, fragility evaluation of safety related components is afundamental issue for risk and reliability assessment. The seismic fragilities of individual componentsand equipments, in fact, are combined with the seismic hazard, i.e. the frequency of occurrence of agiven intensity of the earthquake motion, to evaluate the probability of different core damage states.The main objective of the seismic fragility evaluation is to estimate the capacity and the relateduncertainty of a component or a structural element relative to a given earthquake severity parameter,such as peak ground acceleration or spectral acceleration. This capacity is defined as the earthquakeseverity parameter value at which, for the considered component or structural element, the responseexceeds the available mechanical resources, leading to failure.Two sources of variability need to be incorporated in a structural fragility formulation: inherentrandomness and uncertainty. Significant randomness affects many of the parameters describing themechanical model adopted in structural analysis, such as material properties (including soil). Inaddition, the earthquake input motion is stochastic in nature, given the extremely large number ofparameters affecting the seismic source, the source-to-site transmission path and the local groundresponse. Uncertainties, on the other hand, arise from analyst’s lack of complete and accurateknowledge about models, methods for response analysis, limit-state formulation etc: uncertainties canbe reduced, in principle, with detailed studies leading to more sophisticated techniques.

In this framework, a procedure able to reduce uncertainties as much as possible, thus reducing animportant cause of unnecessary conservatism has been developed; the procedure is based on the use ofthe Response Surface Methodology for describing the structural performance, on a simulationapproach for facing the random vibration issue and on the Monte Carlo Method for computing thefailure probability.2. Definition of the problemSeismic fragility is defined as the probability of failure of a component (or structural element)conditioned to the severity (e.g. PGA) of the ground motion and can be written, for each PGA value:r rp f x dxf=∫r{ g( x) < 0}( )r is the joint probability density function of all the variables xr affecting load and responsewhere f ( x)modelling and ( x)g r is the performance function.In the proposed procedure, the Response Surface (RS) Method is used to provide an analyticalformulation to the g ( x r ) function to allow an efficient Monte Carlo evaluation of the seismic fragility.Here the performance function of the component is assumed to be expressed in the simple “capacity(C) minus demand (D)” format:r r rg ( x) = C( x) −D( x)in which the vector x r lists all input random variables affecting load and response modelling.In the case of linearity of the analysis, the performance function is expressed as follows:r A r rg( x)= C ( x) − PGA⋅y(x)from which:A rr ⎛C( x)⎞ rg%( x) = ⎜− y( x)PGA⎟⎝ ⎠r rwhere C A ( x)represents the acceleration capacity of the component under examination and y(x)isthe acceleration response of the structure, computed at the component supports, for a input timehistory having a unit PGA.The RS method is used to find an analytical expression of y(xr ) and the Monte Carlo method issubsequently employed to find the probability that the acceleration response of the studied componentD( xr r r) exceeds its maximum allowable value C A ( x), or, more in detail, that y(x)exceeds theramplification ratio C A ( x) . In this way, the behaviour of the building is characterized in terms ofPGAthe amplification ration of the peak ground acceleration.3. IRIS test caseThe methodology has been applied to the IRIS reactor as a test case.The IRIS (International Reactor Innovative and Secure) plant is a medium power (∼335 MWe)pressurized light water reactor under development by an international consortium which includes morethan 21 partners from 10 countries, led by Westinghouse Electric Company.IRIS plant development is aimed by a Safety-by-Design TM philosophy from the beginning, to reduceas much as possible both the probability of occurrence and the possible consequences of certain severeaccidents caused by internal events. The IRIS power plant, in fact, presents some peculiar features,among which a compact design and an integral layout.The IRIS safety-by-design has eliminated many initiators of internal events and consequently theinternal events CDF has decreased by at least another decade when compared to passive light waterreactors. Still, the external events initiators have not yet been addressed and thus at least for now, theCDF due to external events, such as seismic, is the preponderant factor in the total CDF for IRIS.For performing initial tests on the procedure for fragility estimation, two structural models of the IRISrector building have been set; a simplified one for performing the response computation and a refinedmodel for the validation of the previous one by comparison of the eigenproperties. The simplifiedmodel will be also used to evaluate the performance of a seismic isolation system.

3.1 Refined ModelA refined model, encompassing a degree-of-freedom number of the order of 10 6 , has been set.Modelling has been restricted to the structural system, which has been assumed as fixed at thefoundation mat base.The main structural elements of the building have been introduced according to the criteriasummarized in the following. Reference is made to the elements and modelling options available inthe ABAQUS structural analysis code.Foundation mat: 8 node solid elements have been placed in 3 layers. Approximate element size is0.65 m. Total thickness is 2 m.Structural walls: All walls directly supported by the foundation mat have been introduced, by meansof 8 node shell elements. Element size is about 1 m, while thickness is 1 m.Shielding wall: It is a cylindrical wall having a thickness of 1.5 m and surrounding the containmentsystem. Has been modeled via shell elements, with approximate size of 1 m.Slabs: A thickness of 1 m has been assumed for all structural slabs. Shell elements have been used;size is approx 1 m.Roofing system: a flat slab with stiffening girders was assumed; the slab was modelled via shellelements, 1 m thick. Girders were introduced with a 5 m spacing and a total depth of 3 m. They wererepresented by 3D beam elements; at each node, the centroidal element is connected to thecorresponding shell node with a rigid link.Containment: A 44.5 mm thick steel sphere is assumed, modelled by means of shell elements. Thelatter have an approximate size of 0.2 m. The lowest half of the sphere is supportedby a massivereinforced concrete structure, this latter modelled by means of 8 node brick elements. At the interfacewith the vessel shell, 20 node elements have been introduced. Perfect bond is assumed between steeland concrete at both sides (internal and external) of the shell.The water contained in the vessel have been treated as a rigid body, which is attached to the shell via adistributed connection (DCC), equally subdividing the inertia forces developing in the water betweenthe selected nodes on the vessel shell.Sloshing effect in the suppression pools: the suppression pools which are located within thecontainment have been modelled as rigid bodies, connected to the reinforced concrete structure bymeans of a DCC connection.The structure of the RWST pools has been modelled by means of shell elements. For modelling thewater content and taking account, though in a simplified way, of the sloshing effect it has beenassumed that the r.c. structure can be regarded as rigid in terms of interaction with the fluid.3.2 Simplified ModelThe passage from the refined to the simplified model is founded on both the simplification of somestructural components and a suitable mesh optimization, as will be described in this section.More in detail, the simplified model has been obtained by applying the criteria described in thefollowing.Walls, slabs and foundation mat: have been all modelled by means of shell finite elements.Roof: the same discretization as the one adopted for the refined model has been chosen.Containment: the lower part of the containment, encompassing the lower steel hemisphere and thesurrounding reinforced concrete supporting structure, has been modelled as a rigid body. On thecontrary, the upper part of the sphere has been represented via an equivalent two degrees of freedominverted pendulum system.Vessel: simplified modelling has been suggested by the observation of the lower vibration modes ofthe refined model, where elastic deformation is confined to a rather limited zone centered on thesupporting skirt. On this basis the lower and upper part of the vessel shell have considered rigid,whilethe central portion of the shell and the skirt have been discretized via shell elements.All the equipments located in the upper and lower parts of the vessel have been introduced as rigidmasses, lumped at the corresponding centres of gravity. The steam generators, which are located alongthe deformable zone, have been considered as rigidly attached to the upper rigid portion of the vessel.For the water, the same criterion as used in the refined model has been maintained.

Foundation ground model : it has been here assumed that the foundation mat, when stiffened by allstructural walls, can be treated as “quasi-rigid”. This means that its deformability is taken into accountin modelling it as a part of the structural system, but that it can be neglected with respect to soilstructureinteraction effects.3.3 Model comparisonIn Table 1 the natural frequencies of the most significant vibration modes are compared for the twomodels. The modes considered in the table are two cantilever modes, in each direction, a vertical and atorsional mode for the building, two rocking modes and one vertical mode for the vessel and twohorizontal translation modes for the containment.As it can be noted, natural frequencies compare satisfactory for all most significant normal modes.Mode Refined model [Hz] Simplified model [Hz] Δ [%] Mode description11 5.41 5.60 +3.5 1 “cantilever” mode in y direction12 6.42 6.66 +3.7 1 “cantilever” mode in x direction15 8.22 8.54 +3.7 Torsional mode21 12.31 12.70 +3.2 2 “cantilever” mode in y direction31 15.56 16.4 +5.4 2 “cantilever” mode in x direction24 13.55 13.52 -0.2 1 vessel rocking mode y-z plane25 13.56 13.52 -0.2 1 vessel rocking mode x-z plane92 31.38 30.79 -1.88 vessel mode translation in z direction90 30.56 30.43 -0.42 1 containment mode y direction91 31.12 31.06 -0.19 1 containment mode x direction22 13.06 13.42 +2.75 1 global mode vertical translation34 16.58 17.01 +2.6 2 global mode vertical translationTable 1. Natural frequencies of the lowest vibration modes of the FE models3.4 Estimation of the seismic fragilityOnce realized the structural model, a set of 10 time histories has been obtained from a referencespectrum. Then, three random variables have been selected to represent the main sources ofrandomness for the computation of the response of an equipment located inside the vessel:− a random variable describing the soil shear modulus G, with mean value of 200 MPa and c.o.vequal to 0.2;− a random variable for the vessel damping factor; its mean value has been chosen equal to 0.03,−and a coefficient of variation of 0.2 has been considered;a random variable to describe the viscous soil damping; more in detail, the ratio between theactual value and the mean value of each damping factor associated to foundation modes isconsidered, named δ, with a mean value of 1 and a c.o.v. of 0.2.It has to be noticed, with respect to the last two RVs, that damping has been here treated in asimplified way. This was due to the difficulty to deal, by means of the software package at hand, withcomposite damping within modal superposition analysis. In the case here shown modal damping factorwere directly stated and given in input by recognizing, with some engineering judgement, modesdominated by foundation or by vessel movements.A second degree polynomial function has been chosen to express analytically the both the mean valueand the standard deviation of the response (x)D xr .A Central Composite Design has been selected as an appropriate DoE for the RS generation.Once found an analytical representation of the response y(xr ) , the probability of exceeding the givenamplification ratio has been calculated, trough the use of the Monte Carlo method, for different valuesof the PGA amplification.y r , approximating the demand ( )

The Monte Carlo sampling technique have been used to select values of the input variablescorresponding to their probability distributions. In correspondence of each set of randomly selectedvalues of the three variables, the structural response has been calculated from the RS and then thercondition C A ( x) r< y( x)is evaluated. The obtained results are shown in figure 1.PGAProcedures are presently under investigation to refine iteratively the RS and the fragility computation.Figure 1. Probability of exceedance as a function of the structural amplification.4. ConclusionsAn accurate methodology to evaluate the seismic fragility of NPP components, trying to reduceuncertainties and conservatism of more traditional procedures, has been developed. Through the use ofthe response surface methodology, the proposed procedure offers a comprehensive and rationalframework for performing parametric studies of the sensitivity of the structural response to the variousrandomness and uncertainty sources. As already hinted, the proposed procedure represent realisticallythe uncertainty in the input ground motion. The method is affordable, being the number of simulationsto be performed rather limited; in addition it is reasonably easy to deal with, being the procedurestraightforward and the tools proposed part of the background of many engineers.References[1] De Grandis, S. and Perotti, F. “An innovative methodology for computing fragility curves ofNPP components under random seismic excitation”, submitted to SMIRT 19, Toronto,Canada, August 12-17, 2007.[2] G. Bianchi, D. Mantegazza, F. Perotti, “Dynamic Modelling for the Assessment Of SeismicFragility of NPP Components”, SMIRT 19, Toronto, Canada, August 12-17, 2007.[3] Kennedy, RP, et al., 1980, Probabilistic Seismic Safety Study of an Existing Nuclear PowerPlant, Nuclear Engineering and Design, 59, 315-338.[4] Ravindra, MK, 1997, Seismic individual plant examination of external events of US nuclearpower plants: insights and applications, Nuclear Engineering and Design, 175, 227-236.[5] Casciati F, Faravelli L, 1991, Fragility Analysis of Complex Structural Systems, Res. StudiesPress Ltd.[6] Schotanus, MIJ, Franchin P, Lupoi, A and Pinto PE, 2004, Seismic fragility analysis of 3Dstructures, Structural Safety, 26, 421-441.

Status of research reactors for future nuclear research in EuropeENC 2007 , Brussels, 16-19 September 2007Joel Guidez, Daniel Iracane, Patrick Ledermann(CEA- France )Abstract :During the 1950’s and ‘60’s, the European countries built several research reactors, partiallyto support their emerging nuclear-powered electricity programs. Now, over forty years later,the use and operation of these reactors have both widened and grown more specialized. Theirradiation reactors test materials and fuels for power reactors, produce radio-isotopes formedicine, neutrographies, doping silicon, and other materials. The neutron beam reactors arecrucial to science of matter and provide vital support to the development of nanotechnologies.Other kinds of reactors serve other specialized services such as teaching, safety tests,neutronic simulation…The modifications to the operating uses and the ageing of the nuclear facilities have led toincreasing closures year after year (ref. 1 and 2 ). Certain facilities are scheduled for closure,such as the last European fast breeder, Phenix, whose shutdown has been announced for2008/2009. For others, safety re-evaluations have had to take place, to enable extension ofreactor life. However, in the current context of streamlining and reorganization, newEuropean tools have emerged to optimally meet the changing demands for research.In 2006, in the neutron beam field, the ORPHEE reactor in Saclay returned to the “normal”number of operating days. The FRM2 in Munich has continued power escalation, extensionwork has continued on the ISIS reactor in Great Britain, and the work undertaken to bring theILL reactor up to more demanding earthquake resistance levels has reached an end.In the field of irradiation reactors, the RJH project has continued to advance. After 3 yearswith an engineering staff of 100, the definition of the reactor was reached in late 2005. TheRJH reactor is now a mature pan-European project, selected by the European Strategic Forumfor Research Infrastructure as vital to European interests. The construction phase waslaunched in 2006. The goal of commissioning is set for 2014.For the European Research Area, the RJH reactor will play a major infrastructure role in thefield of fission n, implementing international collaboration. In 2006, already 5 countriescommitted to the RJH project and contributed to construction. The European Commissionsupports the project, further strengthening the RJH Consortium.With respect to Fast Reactors, the future-oriented work developed in GEN4 has demonstratedthe strong interest in the fast reactor concept. Several countries, including Japan, Russia,India, China and South Korea, have expressed their preference for sodium-cooled reactors.This commitment has led to tangible actions well underway, including work at Monju forrestarting, construction in China of an experimental sodium reactor for divergence in 2009,construction in India of the PFBR (1200 MWth), resumption of the BN 800 constructionbudget in Russia. In the United States, the announcement of the GNEP initiative includes aprogram for the ARR sodium reactor. And in France, the satisfactory operation of the Phenixreactor continues, with an availability factor above 80 %.The Phenix reactor is scheduled to shut down 2008/early 2009, which would leave an absenceof fast reactors in Europe as of the Phenix closure date. The 28 June 2006 french law on the

subject of sustainable management of radioactive matter and waste, calls, in its article 3, forthe start-up of a prototype prior to 31 December 2020. This paper reports on the current statusof the work and organization relating to the objectives for this prototype.This paper also provides information on the status of other European projects, including thefaisability study of an experimental fast reactor in Belgium (MOL) with a technologyalternative to sodium, and the Pallas Project in Netherlands to replace HFR reactor in thefuture.To conclude, the entire group of research reactors is undergoing significant change in Europe,and moving towards a more streamlined scenario providing for optimization of resources andplant characteristics, for the entire range of users.EUROPEAN SITUATIONPRESENT SITUATIONIn Europe, nuclear electricity plays an important role and will stay for the long term a verysubstantial part of the energy mix since it contributes to the energy security of supply to thereduction of greenhouse gas production and to the competitiveness of Europe.Experimental reactors have been used to support many important fields of industry andresearch in Europe: safety, lifetime management and operation optimisation of current nuclearpower plants, development of new types of reactors with improved resources and fuel cyclemanagement, medical applications, material development for fusion reactor…European experimental reactors have been built in the 60’s and most of them have beenoperated on a national basis. With several Material Testing Reactors (BR2, Halden, HFR,LVR15, Osiris, R2, Siloe), and with demonstration reactors and prototype reactors (Rapsodie,Phenix, PFR, KNK II, and AVR, THTR) for developing the Sodium and Gas cooled reactortechnologies, Europe has gained a worldwide leadership.Some of these facilities are already stopped. The others will be more than 50 years old in thenext decade and will face increasing probability of shut-down due to their obsolescence. Sucha situation cannot be sustained on the long term.Other research infrastructures are dedicated for fundamental research application by providinghigh quality neutron beams: reactors such as ILL-RHF (1971, Laue Langevin institute,France, Germany, Great Britain), ORPHEE (1980, France), FRM2 (2004, Germany). Thesefacilities are commonly operated within European collaborations. In the field of matterscience utilizing neutron beams, a set of effective and up-to-date facilities are available withinEurope.Toward renewing some key European Experimental Reactors (EER)This survey has been discussed in depth and shared in Europe since 2002 (ref. 3).A first generation of EERs, launched in the 60’s, have provided the necessary support toindustry and research in Europe (nuclear power plants, actinides management, medicalapplications, condensed matter physics…). The question is now to define and implement aconsistent EER policy• Meeting industry & public needs, keeping a high level of scientific expertise ;

• With a limited number of EERs, specified within a rational compromise betweenspecialisation, complementarities and back-up capacities ;• To be put into effective operation in the next decade.Taking into account the needs of nuclear industry, the strategic importance of future GEN IVreactors developments, the advanced fuel cycle and the public health stakes, an Europeanpolicy must include a mid-term roadmap encompassing:• A high performance material testing reactor ;• A reactor optimised for medical applications ;• An experimental reactor for innovative fast neutron reactor technology developmentwith capabilities related to test advanced fuel cycles.RESEARCH REACTORS IN EUROPEIrradiations in support of present and future nuclear reactorsNuclear operators are bringing in operational changes and management measures to improvefuel economy and extend lifetime of nuclear power plants. While Utilities implementextended fuel burn-up, optimised fuel cycle, Safety Authorities assess this evolutionarysituation through questioning about the safety behaviour of components and systems(Generation 2 and 3).In parallel, a new generation of reactors Generation 4 will be developed to address key issuesrelated to sustainable development objectives. A variety of aspects will be addressed in thiscontext, regarding economics, safety, better use of natural resources, waste management, nonproliferation issues and new utilization of nuclear energy (process heat, hydrogen). This newgeneration will require important technological advances in material and fuel science.In the meantime closing the fuel cycle of presently used reactors remains an important topicof research to be addressed through partitioning & transmutation (P&T) and where minoractinides burning will also require reactor developments that can be commonly addressedthrough the Experimental Reactors (ERs) addressing the Generation 4 issues.European existing MTRs and ERsExisting European MTRs are ageing (seetable) this leads to a growing discrepancybetween their capabilities and the aboveindustrial and public needs.These reactors have gained a considerableinternational recognition for theiroperational flexibility and ability to set upcollaborative programmes having broadinternational participation.The R2 reactor shut down in 2005illustrates how fast the situation can evolvein Europe.Countries Reactor OperationCzech Rep. LVR 15 1957 10Norway Halden 1960 19Sweden R2 1960-2005 50Netherlands HFR 1961 45Power(MWth)Belgium BR2 1961 60/120France OSIRIS 1966 70

With the JHR-CA FP6 coordination-action (2004-2005) and through the ongoing FP6MTR+I3 (integrated infrastructure initiative), the European MTR community reinforce itsscientific capacity by sharing the development of a new generation of experimental devices.There is a need for a high performance MTR to be implemented in Europe in the comingdecade in the framework of worldwide competition. Its construction shall cope with therequirement to continuously supply irradiations. Therefore this new MTR will be networkedwith other facilities involved in material and fuel development programs (hot labs, otherreactors).As far as Experimental Reactors (ERs) in Europe (Rapsodie, Phenix, PFR, KNK II, and AVR,THTR) are concerned, all of them have been shut down except Phenix, a sodium cooled 250MWe reactor started in 1973 and planned to be shut down by 2009. Although the fuel cycleand associated advanced fuel development as well as Generation IV systems are requestingfast spectrum experimental reactors, Europe will be left without an ER after 2010.With the ongoing EUROTRANS FP6 Integrated Project (2005-2009), preparing a design of adedicated experimental minor actinide burner (XT-ADS based on the MYRRHA projectinitiated by SCK•CEN), with FP6-GCFR (Gas Cooled Fast Reactor), and with FP6-ELSY(European Lead Cooled System), the European reactor research community and nuclearindustry are integrating their efforts to put Europe in a position to decide by 2010 on therealisation of an Experimental Research Facility having a fast spectrum and able to addressthe closing of the fuel cycle. This facility can be conceived with the objective to demonstratefast reactor technology and effective burning of minor actinides. It should be conceived toserve in a later stage as a fast spectrum irradiation facility.Nuclear medicine is important for the health of European citizens with about 10 millionmedical procedures per year and 15 million in vitro analyses. This field is also important interms of market for the pharmaceutical industry in Europe. For therapeutic and diagnosisactivities, respectively 100% and 75% of the radioisotopes are produced by research reactorsin Europe more particularly in HFR, BR2 and Osiris.PERSPECTIVE FOR AN EUROPEAN RESEARCH AREA ON EXPERIMENTALREACTORS (ERAER)The Jules Horowitz Reactor (JHR), a mature project meeting nuclear industry andpublic needsThe need for a new MTR in Europe has been assessed and confirmed by the Feunmarr FP5thematic network (2002) (ref. 3) :« There is clearly a need as long as nuclear power provides a significant partof the mix of energy production sources »« Given the age of current MTRs, there is a strategic need to renew MTRs inEurope ; At least one new MTR shall be in operation in about a decade fromnow »A high performance new MTR is to be built in Europe to meet the industry and public needsrelated to safety, competitiveness and innovations for the existing generations and the futuresystems.More specifically, the JHR shall provide a secured experimental capability to support :• Plant life time management & extension for Gen 2 & 3.

• Technological evolution for Gen 3, performances improvement.• Fuel performance improvement and behaviour validation in incidental andaccidental situation.• Innovative fuel & material development for HTR and Gen 4 systems.• The expertise in the field of nuclear energy, in association with other keyinfrastructures.To meet these needs for the coming decades, JHR will be a high performance 100 MWthMTR providing high fast neutron flux in an under-moderated core (10 15 n/cm²/s perturbedflux above 0,1 MeV) and high thermal neutron flux in the moderator (5 10 14 n/cm²/s).Compared to existing MTRs, JHR will offer advanced experimental capacities such as on linefission product measurements and dedicated cells to manage safety experiments withdamaged fuel samples.JHR – RESULTS IN 20062006 was the year the development phase was launched for the Jules Horowitz reactor (JHR).This development phase corresponds to the industrialization of the project, and includesdetailed definition of the components and the preparation of the tender documents.This development phase is an important transitional step between design and construction ofthe JHR, requiring several readjustments to adapt the teams to the construction objective.Several files were sent to the Safety Authority in March 2006. These included the “DAC”(Application for Authorization to Create the Installation), the “DARPE” (Application forWater Intake and Discharge), and the “RPRS” (Preliminary Safety Report).Provision of these documents enabled the public inquiry to be held, which was thencompleted in December 2006.Processing of the Preliminary Safety Report also began, with the goal of holding a PermanentGroup in 2007.2006 was also the year of the signature of 6 bilateral agreements between the CEA and itspartners, for the construction of the JHR. These agreements are the culmination of severalyears of European cooperative efforts to define the funding of the JHR, a process which willmost likely be applied anew for other joint European research infrastructures in the field ofEURATOM-fission.The agreements were the basis for founding the JHR Consortium in 2006. The agreementproposed to the project partners in late 2006 stipulates their access rights as a function of theirfinancial participation. This agreement was signed in spring 2007.The Pallas project, securing the production of radio-nuclides for medical applicationsIn the Feunmarr 5 th FP thematic network (2002), the market for radio-nuclide production formedical applications was assessed. Securing the European production capability was stated asan important public health stake.

Nuclear medicine is important for the health of European citizens with about 10 millionmedical procedures per year and 15 million in vitro analyses. This field is also important interms of market for the pharmaceutical industry in Europe.• Nuclear imaging techniques are powerful non-invasive tools providing uniqueinformation about physiological and biochemical processes. The gamma imagingactivities represent a global annual turnover estimated at more than 1 billion €, and thedemand grows each year by about 5%. This requires typically 20 isotopes among whichthe 99Tc part represents 70%. Other techniques like the positron emission tomography(PET) or Radioimmuno-assay represent an annual turnover of some 450 million €.• For radiotherapy with radioisotopes, the overall annual turnover is roughly 250million €. If cobalt therapy is an important but declining market, new technologiesappear and are in a growing stage (gamma-knife surgery, alpha immunotherapy, brachytherapy…).For therapeutic activities (resp. diagnosis), 100% (resp. 75%) of the radioisotopes areproduced by research reactors.The Petten site, in The Netherlands, integrates on the same site the reactor HFR, hot cells andmedical-oriented production facilities. The Pallas project replacing HFR after 2015 willreinforce this medical application in Europe. A back up function from other EuropeanResearch Reactors, especially JHR, is mandatory to secure the continous supply of themedical radioisotopes.The Pallas power and main technico-economical characteristics are not yet finalised. Thisthermal power should enable both medical radionuclides production and somecomplementarity to JHR material programs.The Fast Spectrum Project, addressing the next generation energy systems andactinides recyclingFor the longer term, future nuclear energy systems should contribute, among other energysources, to secure a sustainable energy development worldwide. Generation IV fast neutronnuclear reactors with the closed fuel cycle shall play a key role to optimise the use of naturalresources and minimisation of long lived waste.Accelerator Driven Systems (ADS) which are the coupling of an accelerator with a subcriticalfast neutrons reactor, and the closed fuel cycle, are investigated as a possible alternative tocritical fast reactors, in the framework of FP6 EUROTRANS IP (Integrated Project).Future reactor research is addressing strong expectations related to energy and waste issues.Three basic needs are identified :• New power plants to be built from 2010 on will use available technologies forGeneration 3 and possibly high temperature gas cooled reactors for industrial heat(synthetic fuels…) or hydrogen production. Their development will mainly make useof generic MTRs.• Development of Gen IV reactors for deployment at the horizon of 2040-2050 requiresthe realisation of a prototype unit of middle size around 2020 for the most maturetechnology which is SFR. Nevertheless one cannot secure access to fast reactortechnology by considering a single technology only. An alternative track may be

equested at a longer term, being either GFR or LFR. A specific experimental facilitywill be needed to address technological development and demonstration of the chosenalternative track to support decision towards following step.• With the closed fuel cycle, fast reactor technology will address a specific concernabout waste management to reduce the actinide inventory to be managed in the longterm by the above mentioned fast critical reactors and/or by sub critical fast reactorsdriven by an accelerator.Based on R&D results there is an important milestone around 2010, to assess viability andperformances of GFR and LFR and to decide for an experimental facility of European Interestin the range of 50-100 MWth, either critical fast reactor or ADS.SCK•CEN volunteers to host this experimental facility on the site of Mol, Belgium.This project should involve many European countries, who will define a technical roadmapfor the selection of a second technology for fast neutron reactor and its implementation, in theframework of the European Sustainable Nuclear Energy Technology Platform (SNE-TP)which will launched in Brussels on September 21 th 2007.Prototype Sodium cooled Fast Reactor (SFR)Among the fast reactor systems, the Sodium-cooled Fast Reactor has currently the mostcomprehensive technological basis, thanks to the experience gained internationally from theoperating experimental, prototypes and commercial-size reactors (such as the Phenix plant inFrance, PFR in the UK or MONJU in Japan).The technological basis gained from these reactors includes key elements of the overallreactor design, fuel types, safety, and fuel recycling. Innovations are sought for a GenerationIV sodium cooled fast reactor in order to reduce the costs and to improve the safety. Theyinvolve design simplification, improvement of in-service inspection and repair, fuel handling,high performance materials, practical exclusion of high energy release in case of hypotheticalsevere accident.Given the maturity of sodium cooled fast reactors, the next facility to be built in Europe willbe a prototype reactor with a power conversion system of 250 to 600 MWe to demonstrateinnovations with respect to existing sodium cooled fast reactors, and to pave the way for afirst of a kind 4 th generation commercial reactor.CONCLUSIONCurrently identified large infrastructures of European interest for nuclear research are :• Jules Horowitz Material high performance Testing Reactor, identified in the ESFRIroadmap as a mature project to replace to a large extent Europe’s aging MTRs (over50 years old) when it will come in operation in 2014. The JHR, launched recently withthe support of several European countries and the European Commission, will In theshort term support studies for Gen.-II and Gen.-III Light Water Reactors on ageingand life extension, safety and fuel performances, and support material and fueldevelopments for Gen.-IV reactors.

• The prototype sodium-cooled fast reactor with a power conversion system of 250 to600 MWe, to be built through a research-industry partnership, together with a fuelfabrication pilot plant.• A Fast Spectrum Experimental System with a power range between 50 and 100 MWthto support the development and demonstration of an innovative Generation IVtechnology.• A reactor which should replace the High Flux Reactor (HFR) after 2015 as the mainEuropean provider of radio-nuclides for medical applications, and as such should besupported by the medical industry.• Besides these major infrastructures, other experimental facilities are needed to supportthe technology developments and the safety studies or to demonstrate cogenerationtechnologies, depending on the market need for hydrogen or synthetic fuel.Networking of existing facilities, and construction of new ones operated as “European userfacility”are essential for meeting the R&D needs described in the foregoing, for advancingthe European Research Area (ERA), and for attracting a new generation of scientists andengineers to contribute to new challenging programs. Modern research infrastructures areessential for enabling the scientific community to remain at the forefront of nuclear energyscience and technology, and to support the development of industrial innovations for nuclearreactors, fuels and fuel cycle.REFERENCES1. ENC 2002 Session on research reactors“Keeping tools available for future nuclear R&D”J. Guidez, S. Crutzen JRC2. ENC 2005“Keeping tools available for future nuclear research in Europe”J. Guidez, D. Iracane, P. Ledermann , L. Martin3. FEUNMARR : Future EU Needs in Material Research Reactors”,FP5 Thematic Network FIR1-CT2001-20122

Session 18.1.5:Research reactors

DEVELOPMENT OF HUMAN ENGINEERED GRAPHICDISPLAY FOR THE HANARO RESEARCH REACTORHOAN-SUNG JUNG, SANG -HOON BAE , YOUNG-KI KIM, MIN-JINKIM, HYUNG-KYU KIM, YOUNG-SAN CHOI, IN-CHOL LIMHANARO Operation Center, Korea Atomic Energy Research Institute1045 Daedeokdaero, Yuseonggu, Daejeon, 305-353, KoreaKWANG-MIN CHOSammi Information Systems Co.103-15, Galwo-dong, Yongsangu, Seoul, 140-807, KoreaABSTRACTFrom June 2006, the upgrade of the operator workstation for the HANARO research reactor hasbeen started to support modernization of the I&C system and to control the fuel test loop facility inthe main control room. The upgrade was done on both the hardware and the software. The softwarefor the monitoring and control was developed and the graphic display was added to the existingsystem. The new display was designed to meet the style guide that was developed for the designersby the human engineering specialists thru analyzing the control room environment and the hardware.The main policy for the upgrade is the consistency with the existing procedures and displays.1. IntroductionA control system for the HANARO research reactor facility is under upgrade according to the 10 yearrefurbishment plan starting from 2002. The control system consist of operator workstations (OWS),networks, controllers and panels. The plan has 5 milestones. The first milestone was the replacementof operator workstations at 2002. The second one is the upgrade of OWS in order to combine thecontrol and monitoring function of the fuel test loop (FTL) into existing system. The development forthe FTL controls has been completed and the upgrade for the HANARO controls will be finishedby November 2007. Other mile stones are the installation of the cold neutron source (CNS), thereplacement of controllers, and the digitalization of reactor protection system. This paper describes thedesign of the human machine interface (HMI) of the operator workstations.2. Control system

The upgrade of the operator workstation has been completed to control and monitor the FTL facilityby the operators in the main control room. [1] The FTL control system is independent of other systemsconceptually. But it is connected with the HANARO control system for sharing information. Allcontrollers for the FTL facility are installed in a FTL control room. But these facilities of the controlroom are used only during a start-up of the system. Normal operation is performed at the main controlroom. So that, the control and monitoring functions are integrated to the existing control system. Thesecontrollers and networks are duplicated except for the data acquisition system. Each control network isconnected to the existing HANARO data severs. The supervision network for duplicated workstationsare connected to the tag severs also. The tag server acts as a bridge between the controllers andoperator workstations. [2]The digital control system of the HANARO is duplicated from input modules to output modules. TheFTL control system is also fully redundant to ensure reliability. Two independent communicationnetwork link controllers are in each channel. The HANARO has a control local area network (LAN)and the FTL has its own control LAN. These two facilities are linked to tag servers. There are twotag servers in the main control room. The tag server is a PC for collecting information from thecontrollers and providing it to workstations and a data server. The architecture of the FTL and theHANARO is shown in figure 1. The time of all computers and controllers are synchronized with theGPS receiver. Other control systems like the CNS to be installed in 2008 will have the same structurewith the FTL control system. [3][4]Figure 1. Architecture of the control system3. Human machine interface

The old display of the operator workstation was not satisfactory in view of human engineering becauseit was designed to keep the same configuration as the old original workstation. The originalworkstation produced in 1992 has limitations in the hardware and software capability to performvarious requirements. To overcome the limitations due to the old technology, an upgrade of theoperator workstation for the FTL facility has been completed. [5] To comply with the humanengineering requirements, a style guide was made first by considering the hardware, HMI tools, andoperator requirements. Guide for developing of displays of the visual display units (VDU) consists ofthe general requirements and specific style requirements. The general requirements are representation,size, number, labeling, highlighting, and testing. The requirements for the specific displays weredeveloped by the cooperation of operators and human engineering specialists for the followingsubjects;- Configuration of the display area- Dimension of each area- Common information area- Menu area- Title area- Alarm area- Information area- IconThe 32 inch wide LCD, 1900pixels x 1200 pixels was selected as the VDU for this project. The area: Pixel1912 (Common Area)80(Alarm Area)150 (Title Area)401192(Menu Area) (Information Area)922220 1692allocation is figure 2.Figure 2. Display area allocation

Figure 3. Sample display of the OWSMenu area is at the left and key parameters are located at the top. These areas are always fixed duringnavigation for an operator to have quick access and recognition. According to the guide, graphicdisplays were developed first and were reviewed by the operators. The newly designed main display isshown in figure 3. Various new displays have been developed during this upgrade to support anoperator convenient method. The table display shows many parameters in one page and acts likeannunciators on the conventional panel. Other usual displays are trend, alarm, historical display, andX-Y plot. One of the new displays is a table display and is shown in figure 4.

Figure 4. Table display4. ConclusionThe second upgrade of operator workstations was finished to integrate the FTL control system intoHANARO control system at the end of 2006. Requirements from the human engineering aspects andoperators comments were incorporated in the style guide for designing of the displays of the visualdisplay units. The human engineered graphic displays were developed and applied to the operatorworkstations. The new displays help the operators for controlling facilities with the realistic andphysical sense easily.5. References[1] H.S. Jung, et al., The integration of the control and monitoring of the FTL and CNS into theHANARO control room, HANARO Symposium, 2006[2] H.S. Jung, et al., Upgrade of the instrumentation and control system for the HANARO researchreactor, Proceedings of European nuclear conference, 2005[3] S.H. Ahn, et al., Instrumentation and control system design for cold neutron source in HANARO,Proceedings of the international symposium on research reactor and neutron science, 2005[4] S.H. Ahn, et al., Shutdown system design for HANARO and CNS trip, KNS spring meeting, 2005[5] H.S. Jung, et al., Improvement of the HANARO control and monitoring system, Proceedings ofKNS autumn meeting, 2006

IMPROVEMENT OF INTEGRATED MANAGEMENT SYSTEMFOR THE RESEARCH REACTOR IN SOFIAA. S. STOYANOVA, K. D. ILIEVAInstitute for Nuclear Research and Nuclear Energy,Bulgarian Academy of SciencesTzarigradsko Shossee 72, 1784 Sofia, BulgariaABSTRACTThe main purpose in establishment of Integrated Management System (IMS) is to guaranteesafety operation of the nuclear facilities as well as to increase their exploitationeffectiveness. To ensure the safety operation of the nuclear facilities the Bulgarian NuclearRegulatory Agency (BNRA) has created requirements and norms to prevent potentialnuclear incidents, overdose irradiations or terrorist attacks opportunities. The IMS of theInstitute for Nuclear Research and Nuclear Energy (INRNE) has been developed in a wayto create an environment which to guarantee the ways and means for: quality managementaccording to ISO 9001:2000, environmental management in accordance with ISO14001:2004, management of safety requirements of the BNRA, security and physicalprotection, management of the safe and health working conditions for the employees. TheIMS is based on the concepts recommended by the IAEA: the entirety of work can bestructured and interpreted as a set of interacting processes that can be planned, performed,measured, assessed and improved, and, those performing assessing work, all contribute inachieving quality and ensuring safety and environment. The INRNE IMS has beendeveloped in the way to be continuously updateable and additive. The IMS will be addedwith new instructions, procedures and others formularies and documents, which willcorrespond to new activities arising during the reactor reconstruction. These instructionsand procedures should be in agreement with the quality standard requirements as well aswill be harmonized with the environment impacts aspects. The IMS developed on the baseof state-of-the-art software ARIS in is developing the way to achieve ease incommunication, visualization, possibility for assessment and continuous improvement.І. IntroductionThe Institute for Nuclear Research and Nuclear Energy (INRNE), Bulgarian Academy of Science,with its Research Reactor IRT is the biggest complex in Republic of Bulgaria for conducting researchin the field of nuclear science, nuclear technology and energy and in the field of the monitoring of theradioactive influence on the environment.The INRNE is the host and an operator of this institutes research reactor complex which is situated inSofia city, and it is responsible for reactor systems maintenance and controls the permanentlyshutdown Research Reactor. INRNE is responsible also for the reconstruction activities, whichinclude:- Planning and preparation for partial dismantling of the Research Reactor (RR)equipment;- Supply of equipment for the IRT Reconstruction;- Planning of activities and responsibility for reactor modernization;- Spent fuel control and management;- IRT radiation monitoring for all implemented activities;- Radioactive waste (RAW) management and control, for the RAW generated duringthe reconstruction process.An Integrated Management System has been elaborated on ISO 9001:2000 requirements for qualitymanagement [1], ISO 14001:2004 for environmental management [2], and safety requirements of theBulgarian Nuclear Regulatory Agency [3], governmental requirements for occupational health and

safety and security. The IMS application guarantees the safety of activities as well as reduces of theradiation influence on the environment within the governmental norms. It helps to achieve maximaleffectiveness and quality of the performed activities.ІІ. ProcessesThe processes, going along with various activities performed on the institute’s Research Reactor aredescribed in so called procedures. This procedures are developed in the way to give you anopportunity for simple process control following the Deming’s cycle trough plan, do, check, actionstages. [4] For the correct functioning of each process a process measuring indicators are formulated.In practice, indicators define the limits within which we would like a certain process to be managed.They specify the qualitative realization of activities and they are serving like comparative measuringprocess index in the time. For example: they are comparing the process execution taken alone. Themeasurement periodicity depends from indicator.Every single process is represented by a complex of activities. A responsible person delegated withpeculiar obligations and competence, is appointed for each activity. Work instructions for each activityhave been developed together with appropriate formularies where the results of the activity arerecorded. These records are applied as documentary evidence for the fulfilment of the safetyrequirements in front of the Regulatory Body, as well as in case of the internal and external audit, or togive evidence if there is civil interest.Some of the most important and specific processes and their corresponding procedures applied for ourRR management are: “The IRT nuclear and irradiation safety insurance”; “IRT Research Reactorreconstruction management”; “The radiation monitoring insurance on the Nuclear Scientific andExperimental Centre (NSEC) site”; “Radioactive waste management”; “Preparation of documentationfor licence and permissions”.There are different instructions attached to a procedure. For example, for the process “The IRT nuclearand irradiation safety insurance” there are instructions as “Instruction for distillate water full up inreactor pool and water pool spent nuclear fuel (SNF) storage”, “Instruction for the water technologycontrol in water pool SNF”, “Instruction for activity on duty for the mechanic when IRT specialsewage is used” etc. The performance of these activities is documented in the records, which provedthe IRT safety assurance.Other basic process is “The Research Reactor reconstruction management”. The realisation of thatprocess is graphically shown on Fig. 1 and it is developed in following basic procedures:- Management of “Investigations, analysis and design of the Research Reactor with 200kW power”;- The reactor equipment partial dismantling management;- The IRT reconstruction work project implementation management;The detailed schedule of the Reactor Reconstruction activities is presented in the table on Fig. 2.All processes, that are carried out on the NSEC site are accompanied with permanent radiation controland monitoring. These activities are described in the procedures “Securing the radiation monitoring onNSEC site” and “Providing safety radiation conditions for the personal, working with radioactivitysources, on the NSEC site”. Monitoring programs and instructions are implemented and being strictlydocumented by records in appropriate formularies.The „Radioactive waste management” process includes: RAW generation, collecting, sorting,minimization, and storage up to final transportation from the NSEC site. These activities control isrealized according to Bulgarian legislation requirements, which are taken in consideration in theprocedure “RAW management”. The records performed under this procedure give an account on theRAW quantity and it movement.To make easier the management of the processes as well as to evaluate them, and to give us apossibility for continuous improvement of the IMS (and in this way to satisfy the ISO 9001:2000requirements), the ARIS software has been used [5].

Governmentdecision forreconstruction ofIRT№552/06.07.2001“Management of “Thereactor 200 kWinvestigations, analysisand design”” P NSEC02Technical projectapproval“The reactor equipmentpartial dismantlingmanagement”P NSEC 07Spent nuclear fuel (SNF)shipment activitiesP NSEC 06The partialdistmantling wasdoneSpent nuclear fuelshipmentBuy stocks and materialsP 7.4 01Preparation ofdocumentation for licenceand permissionsP NSEC 04Supply ofequipment andmaterials wasacceptedThe licence andpermission wasdoneThe management of the IRTreconstruction work projectimplementationР NSEC 08The projectvalidation was donePreparation ofdocumentation for licenceand permissionsP NSEC 04Fig.1. The process chart “The IRT Research Reactor reconstruction management”

Fig.2. Detailed schedule of the Reactor Reconstruction activitiesІІІ. ResultsFrom general point of view the good practice is the approach were we plane and conduct training toachieve qualification level. To perform this we can conduct training courses not only concerningnuclear technology and energy knowledge but conducting seminars on IMS also. The necessity, toconduct education courses, is documented in the form – “Training Application”. After the end of theeducation course, the trained person has to write down report.At daily work we ran into difficulties, connected with establishment and documentation of thesuggestions for improvements in separate processes. That's why periodically we are refreshing ourcourses by the ISO standards requirements.The INRNE management policy is directed to guarantee high quality developments implementation,which are in agreement with modern world trends of continuously refreshing knowledge, of longstanding experience and cooperation with leading European and International institutions. INRNEhave a purpose to satisfy the community needs for development and maintenance of the nuclearscience, to create necessary knowledge and skills for development of applied methods and research inthe area of nuclear technology medical physics and nuclear industry.

ІV. ConclusionThe IMS is based on the concepts recommended by ISO standards and the IAEA prescriptions: theentirety of work can be structured and interpreted as a set of interacting processes that can be planned,performed, measured, assessed and improved, and, those performing assessing work, all contribute inachieving quality and ensuring safety. The IMS has been developed in the way to be continuouslyrefreshable and additive. That’s why IMS will be added with new instructions, procedures and others,which will correspond to new activities arising during the reactor reconstruction.The IMS gives strong level of certainty in the Research Reactor safety assurance, environmentalprotection, reactor physical protection and secure the normal personal working conditions.V. References1. Standard EN ISO 9001:2000, 20002. Standard EN ISO 14001:2004, 20043. Act on the safe use of nuclear energy, Promulgated in the State Gazette No. 63 of June 28, 2002.4. Kaoru Ishikava, Introduction in quality control, 19895. ARIS – software product ARIS toolset, 1997 – 2005, IDS Scheer AG

EVALUATION TESTS OF THE TELEROBOTIC SYSTEMMT200-TAOIN AREVA- NC/LA HAGUE HOT CELLSP. Garrec, F. Geffard, Y. PerrotCEA, LIST, Service Robotique Interactive18, route du Panorama, BP6, FONTENAY AUX ROSES, F- 92265 FranceG. Piolain, A.G. FreudenreichAREVA/NC/la HAGUE FranceF-50444 BEAUMONT HAGUEABSTRACTThe MT200-TAO system for hot-cells, first presented at ENC 2005, transforms aconventional wall-transmission telescopic mechanical telemanipulator (extension 4 m;capacity 20 kg), into an electrical computer-assisted telerobotic system. The workingvolume is extended to a full hemisphere (a volume approximately three times larger thanthe original telemanipulator) and operators experience a level of ergonomics and dexteritywhich sets new standards. This innovative system has been successfully evaluated in a“cold cell” of AREVA/NC/La Hague in order to prepare a complete evaluation in an activehot-cell which is currently in preparation. This paper summarizes the architecture and thecomponents of the system and details the non-active evaluation phase as well as the trainingof operators and the ergonomical revolution allowed by this system. Finally we describe theplanned active mission in the ACR workshop and the awaited benefits for the end-user.1. Introduction - Project background and objectivesAREVA/NC/La Hague is a reference plant for remote handling technology with 600 telemanipulatorsinstalled and 150 operators and is thoroughly involved in developing new tools to improve its plants aswell as its workstation ergonomics. Any advantage offered by a new system that can replace anexisting one without modifying the facility makes it possible to increase performance in the short termand to prepare changes in new plants in the medium term.In this perspective, several attempts have been made to actuate an existing disconnectabletelemanipulator from the cold side. In the past, CEA and COGEMA have cooperated on a projectnamed “MT200 Numérique” to transform a MT200 into a robot to perform repetitive tasks.The MT 200 TAO system is the result of the fruitful cooperation between the Interactive RoboticsUnit of CEA LIST and AREVA NC. It was designed to address the following specifications:guarantee similar performances than the original MT200 telemanipulatorincrease working volume allowing ceiling accessimprove workstation ergonomics of the existing MT200 telemanipulatorallow distancing of the operator from the controlled zone in certain workstationspotentially exposed to contamination or high radiation rateensure safety for difficult tasks and reduces of operator fatigue when located within thecontrol zoneallow playback of some repetitive tasks that do not require force feedback (roboticmode)2. Description of the systemThis system has been formerly described in more detail [1] but it is useful here to recall its basiccharacteristics.

MT200 - La CalhèneTelemanipulatorMT200 TAO SystemWALL TRANSMISSIONSLAVE ARMForce-Feedback MASTER ARMVirtuose 6D4040 Haption®/CEAForce-FeedbackSLAVE ARM DRIVE UNITMASTERCONTROLLER(PC Rack)SLAVECONTROLLERDatalink (optical fiber)MECHANICAL asterarmFig. 1 – Schematic principle of the MT200 TAO systemLeft picture: The original MT200 mechanical telemanipulator is disconnectable in 3 parts: the masterarm, the wall transmission and the slave arm.Right picture: The MT 200 TAO system functionally replaces the mechanical master arm. The walltransmission and the slave arm are those of the MT200 La Calhène® a design originating from theearly 80’s. The slave drive unit can be fitted to any La Calhène® wall transmission model in less thanan hour and therefore to any telescopic slave arm produced by this constructor. The compact forcefeedbackmaster arm is a Virtuose 6D/4040 constructed by ®Haption on a CEA-LIST patented designusing ball screws. It can exert a permanent effort of approximately 40 N.The TAO 2000 software also a proprietary software from CEA-LIST, allows force feedback masterslave mode between the two kinematically different arms in Cartesian coordinates. It also brings a setof powerful functions: Exact balancing and force surveillance, tool weight compensation, adjustablevelocity and effort ratio (independently), robotic modes, virtual mechanism modes, and automaticpursuit of the gripper by a telesurveillance camera.3. Performances requirements and designTeleoperation is an extension of telemanipulation which has been historically and is still today theessential way to remotely manipulate objects that can’t be handled directly because of their potentialdanger. The operator is always in view of the task (either directly through a window or indirectlythrough a television system) and master-slave telemanipulators used for this purpose havebackdrivable transmissions to ensure that efforts are transmitted whether they are applied at the masteror at the slave (property generally called bilaterality). Mechanical telemanipulators (or electricalservomanipulators) are also built with the same number of axis (rotational or translational) and thesame architecture.A teleoperator (or a telerobot) covers a wider variety of master and slave also able to perform masterslavetelemanipulation with the same above mentioned properties but through computer control. Thusthe slave arm can also playback trajectories just as a robot does and it may also be installed on atransporter (a fixed structure or a mobile platform). Powerful assistance function can be implementedsuch as “virtual mechanisms” which consists in constraints in position (or force) in certain directionsto help guide the tool (for example, to keep it normal to the surface). It is also possible to coordinatemore than one master-slave at the same time in various combinations. Master and slave can beheterogenous (mixing rotation and translation), even redundant (more than 6 axis) and therefore morefreely optimized to their tasks. The master station may consist in a force feedback joy-stick, a masterarm or an exoskeleton. The slave arm may be a dedicated design (like MT200 TAO) or an industrialrobot equipped with joint torque sensors or generalized 6 axis force sensor to compensate friction.

Furthermore, telerobots just like telemanipulators, must be “transparent” enough for the user, acombination of low friction and low inertia.In the case of the MT200 TAO, thanks to a careful design of its drive unit, the operator experiences asimilar force sensivity (including for the tong) than with the original MT200 telemanipulator,combined with a much lower inertia. This objective had to be imperatively met in order to pass theacceptation test by La Hague’s pilots, trained to daily work with conventional mechanicaltelemanipulators (principally MT200/La Calhène and A100/Wälischmiller). Moreover force feedbackis here performed without any force sensors which is a guarantee of simplicity and reliability for thesystem.4. Teleoperation: An ergonomical challenge for AREVA/NCHowever, the introduction of teleoperation represents for AREVA/NC La Hague a specific and majorchallenge in terms of ergonomics due to the cumulated experience on conventional telemanipulatorsrepresented by the whole pilot staff (about 150 persons).Moreover, teleoperation implies here to work in Cartesian coordinates. For the pilot it first means thatthe handle and the tong are no longer visually aligned. This phenomenon is usually perceived asdetrimental for everyone having tested a remote controlled model.In addition, as a consequence of energy input in the system, force and speed ratios may be selectedindependently leading to a different behaviour than a simple mechanical transmission.We can then conclude that a successful introduction of today’s teleoperation technology in the plant ishighly dependant on the perception of the new ergonomical trade-off offered by the system.To better understand evaluations, we need to recall that for the operator teleoperation is a sensorymotor activity involving visual and force feedback:• Visual feedback from the cell takes place via a shielding window (direct vision) and/or viacameras (indirect vision)• Force feedback occurs via the master arm pistol grip handle completed with a trigger.5. Validation program of the systemValidation will be fully assessed after the termination of a two phase evaluation process. The firstphase, finished in 2006, involved testing in a cold cell. Operators were trained to use the machine sothat they could assess usability and make suggestions for improvement. After incorporation of thesuggested changes, the second phase, consisting in hot cell operations, is now being scheduled (2007 –2008) under different work conditions and types of workshops such as: operating conditions in the alpha waste conditioning facility, maintenance conditions in the resin conditioning facility, repair conditions in the vitrification facilities, exceptional operation conditions (workshops to be defined).5.1 Cold-cell testing phaseTeleoperator experts first underwent a 2 day training course. This “practical” course was based onexercises involving routine works to be carried out, enabling them to get gain experience with the newtool. They also used an ergonomical test bench, a device that allows the operator to perform severalstandard teleoperation tasks, to record the operation time and measure the exerted forces, the latterbeing an important parameter to evaluate the impact on the environment.

Fig. 2 –MT 200 TAO at AREVA/NC La Hague’s training and evaluation cold-cellLeft picture: slave arm drive unit replacing the conventional mechanical master arm and itscounterweightsRight picture: slave arm in a “work at the ceiling” configurationThe training proved conclusive and the new tool received the operator’s approval at the end of thecourse. Trained operators expressed their desire to use the MT200 TAO in actual work situations. Anevaluation of the prototype was then carried out using a questionnaire with the criteria (equipment,activity, etc.) that are important to teleoperation at the end of this course. Recommendations forimproving the tool and a report were given to the AREVA NC project manager. Thanks to theflexibility of the software, most recommendations were followed by corrections and an important twofunctions have been finally added: an automatic control of the telescopic offset and a “screwdriver”function. This means that the pilot no longer needs to regularly adjust the length of the telescopicmovement using handle knobs and can thus more easily concentrate on his task.After this first phase of evaluation, positive conclusions regarding the benefits for both the plantmanagement and the operator have already been drawn in connection with each technicalcharacteristics or function of the system. They are summarized and classified in Tab 1.

Replacement of mechanical transmissions by electrical transmissionTAO software intrinsic benefitsTechnical featuresof the MT200 TAOteleoperatorReplacement of themechanical transmission bya flexible ombilicalHemispherical slaveworking volumeDisplacement homotheticalratioForce homothetical ratioAutomatic force saturationof the slave armMaster force capacityadapted to the operatorAccurate balancing of theweight of the master andslave in all positionsPartial or total balancing ofthe weight of the tool ormanipulated objectOperator visual coordinatesGeneralized offset inCartesian mode inpositions/orientations(suppression of telescopicoffset "Z electric")Virtual mechanismsPerformance/QualitybenefitsDistanciation of the operatorfrom the hot-cellIncreased working volume (3times greater) + Continuousfree displacementEffective use of slave workingvolumeIncrease operator'ssensivity/DexterityPreserve the slave arm and theenvironment (increase safety)Increased securityCoherence between vision andactionMental workload benefitsDirect vision improved as not linked to the master armHigher force fidelity=>BetterdexteritySingle push button (insteadof three for a conventionaltelemanipulator)Improved gestureaccuracy/Implementation ofcomplex proceduresDecrease attention to jointlimits managementNo necessity to anticipateparasitic efforts and theirconsequences on trajectoryexcursionsAbsence of mentalcompensation (no inversionphenomenon)Handle and tong trajectoriesoccur in the same coordinate(operator visual coordinates)Operator only controlsuseful efforts on the toolPhysical benefitsImproved operators’position (facing theshielding window)Reduction of exposition to:irradiation, contaminationElimination of electricalshock riskDecrease the force to deliverSuppression of guidingefforts on the toolRobotic modesSuppression of repetitive tasksTab 1 : Relation between teleoperation functions/performances and practical benefits for the operator/end user5.2 Hot-cell testing phaseDue to successful testing in the cold cell testing, an intervention mission has been programmed in theACR (in English, resin conditioning facility). It will predominantly consist of preventive maintenanceoperations (cleaning, checking and dismantling/reassembling in the event that equipment needsreplacing). During the intervention, the telerobot will be used at its maximum capacity. Moreover, thetask to be performed inside the cell is not aligned with the shielding window. Altogether theseconstraints represent excellent reference conditions to evaluate the benefits of the system. The ACRprocess is used to condition the resin in a cement matrix is described in the scheme below:

Fig. 3 –The ACR process and the cementing cell 9403 NPH/ACRThe Fig. 4 shows the particular arrangement of the telemanipulator layout.Operator workingzone misaligned withthe shielding windowFig. 4 – Specific telemanipulators layout in the cementing cellOperations to be carried out in the predominantly preventive maintenance:• Checking of the equipment, (frequency: 3, 6, 12 months),• Cleaning of the equipment (frequency: 6 months, 12 months),• Changing of faulty equipment, which leads to teleoperations that predominantly involveDismantling/Reassembling mechanical equipment:o valves, plugs, connectors,o large pieces of equipment such as the mixer, stirrer, hatch, chute,o Mechanical process cleaning such as tapping

6. ConclusionThe MT 200 TAO (CAT) is an ergonomically adapted tool as witnessed by the ready acceptance bytrained operators, regardless of the change in work habits that this tool will require. It is well suited tothe work to be carried out and enables the telemanipulation activity to be significantly improved interms of efficiency and reliability. The system will be brought into use on the AREVA NC site overthe next few months depending on the gains in performances obtained during hot-cell operations. TheMT2OO TAO opens a new era in teleoperation; a technical revolution such as those during the 1990’sfor control stations (migration of the wall mimic diagrams to computerized control rooms).The partners are now considering the development of an optimized telescopic arm exhibiting anextended lifetime and a higher reliability. The increased use of interactive simulation tools is alsoforecasted (hot-cells virtual mock-ups, workstation simulators…) to improve Man Machine Interfaceergonomics, training methods and task planning.7. References[1] Garrec P., Piolain Gérard, Lamy-Perbal S., Friconneau J.P., “The telerobotics system MT200-TAO replaces mechanical telemanipulators in COGEMA/AREVA-La Hague hot cells”, ENC2005, Versailles, France, December 2005.[2] Desbats P., Toubon H., Piolain G., “Status of Development of Remote Technologies Applied toCOGEMA Spent Fuel Management Facilities in France”, ANS 10 th International Topical Meetingon Robotics and Remote Systems for Hazardous Environments, Gainesville, Florida, USA,March-April 2004.[3] Garrec P. Friconneau J.P, Louveau F. " Virtuose 6D: A new force-control master arm usinginnovative ball-screw actuators» in Proceedings of ISR 2004, Paris, March 2004.[4] Goubot J.M., Garrec P., “STeP: an innovative teleoperation system for decommissioningoperations”, Workshop “Decommissioning challenges: An industrial reality ?”, French NuclearEnergy Society, November 2003, Avignon, France[5] Garrec P. "Systèmes mécaniques » in: Coiffet. P et Kheddar A., Téléopération et télérobotique,Ch 2.pages 40 to 71, Hermes, Paris, France, 2002.[6] Gicquel P., Andriot C., Coulon-Lauture F., Measson Y., Desbats P., “TAO 2000 : A genericcontrol architecture for advanced computer aided teleoperation systems”, ANS 9 th Topical meetingon Robotics and remote systems, Seattle, USA, 2001.[7] Slutski L.-I., Remote manipulation systems Quality evaluation and improvement, vol. 17 deMicroprocessor-Based and Intelligent Systems Engineering, Klüwer Academic, Publishers,Dordrecht - The Netherlands, 1998.[8] Vertut J., Coiffet P., Téléopération. Evolution des techniques, vol. 3A, Hermès, Paris, France,1984.[9] Köhler G.-W., Manipulator Type Book, Verlag Karl Thiemig, München, 1981.Patents[10] Garrec P., CEA - European patent EUR 01938347.0-2421-FR0101630: Nut-screw and cabletransmission[11] Garrec P., Piolain G., CEA/COGEMA – French patent N° 04 50358000: Telemanipulationarm in two parts[12] Streiff G., CEA– French patent N° 90 12183: Actuation unit for a manipulation arm

Session 19.1.1Plant Life Extension

MAINTENANCE ISSUES IN RELATION TO PLANT LIFEMANAGEMENT MODELSP.CONTRI, M.BIÉTHNuclear Operation Safety, European Commission,Joint Research Center, Institute of EnergyPO Box 2, 1755ZG Petten, The NetherlandsABSTRACTDue to current social and economical framework, in last years many nuclear power plantowners started a program for the Long Term Operation (LTO)/PLIM (Plant LifeManagement) of their older nuclear facilities. A PLIM framework requires both a detailedreview of the features of the main safety programs (Maintenance, ISI, Surveillance) and acomplete integration of these programs into the general management system of the plant.New external factors, such as: large use of subcontractors, need for efficient managementof spare parts, request for heavy plant refurbishment programs demand for updatedtechniques in the overall management of the plant. Therefore also new organisationalmodels have to be developed to appropriately support the PLIM framework. Last year anetwork of European Research Organisations carried out many R&D tasks aiming atcapturing the aspects of the maintenance programs where research is mostly needed and atdeveloping suitable optimised maintenance models. Using the outcome of these initiatives,this paper aims at identifying the technical attributes of the maintenance program moredirectly affecting the decision for a long-term safe operation of a nuclear facility, and theissues related to its optimal implementation.1. IntroductionDue to current social and economical framework, in last years many nuclear power plant ownersstarted a program for the Long Term Operation (LTO)/PLIM (Plant Life Management) of their oldernuclear facilities [1,2]. This process has many nuclear safety implications, other than strategic andpolitical ones. The need for tailoring the available safety assessment tools to such applications hasbecome urgent in recent years and triggered many research actions. In particular, a PLIM frameworkrequires both a detailed review of the features of the main safety programs (Maintenance, ISI,Surveillance) and a complete integration of these programs into the general management system of theplant.New external factors, such as: large use of subcontractors, need for efficient management of spareparts, request for heavy plant refurbishment programs demand for updated techniques in the overallmanagement of the plant. Therefore also new organisational models have to be developed toappropriately support the PLIM framework, integrating both safety related and non safety relatedissues.In 2003, the JRC-IE (Joint Research Center, Institute for Energy) launched a network of EuropeanOrganisations operating Nuclear Power Plants, SENUF (Safety of European Nuclear Facilities). TheSENUF Working Group on ''Safety of Nuclear Facilities in Eastern Europe dedicated to NuclearPower Plant Maintenance”, hereinafter referred to as SENUF-WG-NPPM, was founded with thefollowing objectives:1) Review and identification of the remaining open (generic/specific) maintenance relatedissues,2) Promotion of well designed and prepared maintenance plans for systems, structures andcomponents,3) Support for the implementation of advanced maintenance approaches, includingimplementation of preventive (condition based) maintenance as well as preventivemitigation measures,4) Evaluation of the advanced risk based maintenance approach and provision of assistancein its implementation.

A background report was developed by the network in 2004 on Maintenance optimisation issues in theEU, supported by a detailed questionnaire in the EU countries [3]. The report collected and evaluatedthe available and applied maintenance methods at NPPs of acceding and candidate countries to theEuropean Union (ACCs) as well as of the wider Europe (covering Russian Federation and Ukraine),and based on this evaluation, preliminary identified areas for further collaboration with them.A very successful workshop was organised in Madrid on June 19-21, 2006 on “Maintenance rules:improving maintenance effectiveness”, by the JRC-IE (SENUF network), UNESA, EPRI, Iberdrola,Soluziona and Tecnatom [4]. The workshop confirmed that improving the maintenance program is oneof the best tools to improving the overall plant performance and the cost control, even improving theoverall plant safety.A second Workshop was organised by the JRC-IE (SENUF network) and by the International AtomicEnergy Agency (IAEA), in Petten on October 2-5, 2006, on “Advanced Methods for SafetyAssessment and Optimization of NPP Maintenance” [5]. The workshop addressed the application ofadvanced probabilistic methods to the optimisation of the maintenance programmes at the EuropeanNPPs.On the basis of the outcome from the SENUF activities in the last years, the objectives of this paperare the following:• To analyze and summarize the existing strategies on nuclear power plant (NPP) maintenanceoptimization, i.e. predictive maintenance based on monitoring component condition, reliabilitycentred maintenance, and risk-informed maintenance in the NPPs of the collaborating parties• To identify the technical attributes of the maintenance programs more directly affecting thedecision for a long-term safe operation of a nuclear facility, their implementation issues andsafety review.• To identify differences and commonalities in the Western and Eastern European practice, andbased on this evaluation, to identify areas for further research and development (R&D).2. The maintenance program in the Long Term Operation perspectiveThere is a generic convincement in the nuclear community, also confirmed by the SENUFquestionnaire carried out in 2004 [3], that the maintenance program should have specific attributes inorder to support a long term operation (LTO/PLEX) program for the plant. In this sense, theInternational Standards (e.g. the IAEA) can be seen, but also the national experience of USA, Spain,Hungary, etc. More specifically, the maintenance programs based on standard preventive maintenance(time based), not oriented to the monitoring of its effectiveness, are not considered suitable to supportthe LTO/PLEX programs. Crucial attributes for maintenance programs in order to support LTO/PLEXare considered: the verification of the performance goals, the root cause analysis of failures, thefeedback from maintenance to the ISI program, and the feedback on the OLC (operational limits andconditions).All Countries implementing an LTO program applied extensive modifications to their requirements onmaintenance at first step, setting up mechanisms to monitor the effectiveness of the maintenanceactivities. In particular, the following features are believed to be indispensable for a maintenanceprogram in a PLIM framework:1) Monitor the performance of the SSCs (structures, systems and components) which mayhave impact on safety during all operational statuses of the plants;2) Assess and manage the risk that may result from the proposed maintenance activities interms of planning, prioritisation, and scheduling.In order to implement these requirements, some issues have to be addressed [6,7], namely:1) The identification of the scope of the condition based maintenance rules: typically theCountries choose the safety related SSCs, SSCs which mitigates accidents or transients,SSCs interacting with safety related SSCs, and SSCs that could cause scram or actuationof safety related systems. Therefore, many non-safety related SSCs may see theapplication of such maintenance rules, with augmented efforts in monitoring theirperformance and planning their reparation.

2) The setting of the performance goals for every component in the scope of the maintenancerules, ranking them according to their risk significance for the plant safety. This task mayend up very challenging as, when industry experience is not available, either dedicatedPSA tasks have to be developed (with special requirements on PSA quality) or specialqualification programs for the evaluation of the component reliability.3) The performance monitoring techniques for the very broad categories of structuressystems and components in the scope of the rules.4) The assessment of the safety during implementation of maintenance actions.5) The feedback from the result of the monitoring of the component reliability back into theinspection, surveillance and maintenance procedures. Root cause analysis, equipmentperformance trend analysis and corrective actions have to be developed on a case by casebasis.In this sense for example the experience of the USA and Spain (where a LTO/PLEX program is wellestablished), Hungary, and Finland (where a PLIM model is in place at the Loviisa NPP) are aconfirmation of this generic statement: all these countries modified their regulatory requirements orpractice on maintenance, in the direction mentioned above, as one of the preconditions for theLTO/PLEX of their plants.The SENUF WG carried out a detailed analysis on the experience of some of the above mentionedcountries on the interfaces between LTO/PLEX and maintenance programs, as a background for thedevelopment of a state-of-art approach to modern maintenance programs [4,5]. The most relevantconclusions are summarised in the following chapters.3. The RCM programs in the experience of the European CountriesThe objectives of the Reliability Centred Maintenance (RCM) and Maintenance Rule [8,9,10]programs as they are usually defined, are listed as in the following (with some differences according tothe country framework):1) Need to control the maintenance cost, particularly in liberalized energy markets, throughreduction of unnecessary tasks and optimized maintenance periodicity2) Improvement of plant safety through better scheduling of maintenance activities3) Optimization of the management organization, more suitable to control plant safety4) Development of pre-conditions for the plant life extension5) Support the production through minimization of outages duration and optimized workcontrol6) Minimization of the radiation doses7) Optimized integration among existing safety programs, such as: ISI, AMP, configurationmanagement, design basis reconstruction, etc.In relation to the operating cost reduction as a consequence of RCM application, the SENUF WGrecorded the following reductions [5]:• In SWE, 10 - 20% of the effort, especially for I&C calibration intervals• In SP, 20% in work, 30% in number of tasks• In HUN, expected, not quantified• In CZ, 30% on a restricted number of systems selected for a benchmark (according to theimplemented Phare project in Dukovany NPP)• In SKR, expected, not quantified.In relation to the Scoping process applied in the RCM, the WG noted that the approaches are quitedifferent in the Countries:• In SWE RCM is applied only to non-safety related SSCs. Safety SSCs are analyzed only toget a documented base for the preventive maintenance (PM) program. Analyses of safetysystem seldom result in any changes of the existing PM-program. The process to get a change

of the Technical Specification requirement are very strict and in most cases not worth theeffort.• In HUN RCM is applied to 70% of the safety related SSCs and to 30% of other systems• In SKR RCM is applied to 44 systems (100-500 components) selected on the basis of differentcriteria, including safety significance.The quality of the maintenance documentation was recognized as crucial to feed a proper feedbackmechanism. The culture of communication (including the “no blame”) may play a major role inensuring all failure mechanisms have been properly identified and all actual equipment failures havebeen recorded.It was noted that in the current dynamic industry an optimized maintenance system should be adaptive.In particular mechanisms should be put in place to deal with configuration changes, changes ofsuppliers, emerging results from the aging management programmes (AMP), etc. The need forimplementation of a living RCM program under the responsibility of the system engineer washighlighted.More in detail, the following difficulties and challenges were identified during the implementation ofoptimised maintenance systems in different EU Countries:1) The implementation of the MR poses major challenges to the organization: in some casesthe interfaces among existing departments were so many that new structures had to bedeveloped. In other cases (Spain) the organization did not change at all and only thecoordination was improved. Also in the US, the objective of the action was the redefinitionof the interfaces. It was pointed out how the interfaces are very sensitive to thechanges in plant configuration and should be promptly updated in such cases.2) The development of suitable performance criteria is a crucial task. In Spain three years ofhistorical data fed the statistical analysis, complemented by the PSA. In the USA theprocess was also reviewed by the regulator. The digital I&C cannot be monitored easily intime. Therefore the failure rate usually is provided by the supplier who can derive it on thebasis of the whole population of the installed equipment.3) There is no shared data base on maintenance among European NPPs. Only INPO andWANO provide a worldwide service to their members, though limited to some issues.There are confidentiality issues attached to it, national factors and plant dependent issuesthat still prevent such communication. Neither non-nuclear plants are involved in thisexchange of experience. Some maintenance forum (such as EPRI/NMAC) provides acertain level of experience exchange, however again restricted to members.4) The interfaces between ISI databases and MR databases are still poor, due to their history:ISI data bases are mainly related to passive components, MR to the active ones.5) There are objective difficulties in the implementation of the RCM due to the requiredchange in mentality of the personnel and amount of extra work in some cases (particularlywhen the RCM is not fully computer assisted)In general, the Ukrainian, Slovenian, Czech, Russian representatives expressed their interest to adopt aMR-like approach in their Countries, even starting on a voluntary bases, most probably closer to the”equipment reliability” model (INPO/AP-913, [11]). Many of them already created some trainingcenters which are developing procedures in this direction.The “equipment reliability” program is not mandatory in most of the Countries (including the US).However, it is gaining growing interest for its systematic approach to the management of the plantsafety. In particular, the correlation among the many existing safety related programs and theconsistent classification of items (important, critical, run-to-failure) seems to be very attractive andpractical.4. Tools for measuring maintenance performance

Recent statistics carried out in the USA (INPO) [4] show that 40% of the failures are related to humanfactors: among them, 30% are related to engineering deficiencies an 30% to work performance. Mostof the significant events in the latter category have been triggered by the supplemental workers.Therefore the contractor performance becomes a crucial issue where many utilities are investing largeeffort for their reduction. Also supplier reliability is an issue: in many cases equipment were deliveredwith wrong or different specifications.Performance Indicators for maintenance effectiveness are considered very useful. However it wasrecognized that some research work is still needed in this field. It was felt important for theInternational organization to provide assistance in this field and set up some benchmarking studies.Maintenance performance indicators are typically based upon: ownership, time from exceedance of theperformance criteria and setting of new goals, use of MR to drive performance, etc. Many Countriesuse the availability and reliability concepts defined in the MR also to monitor the performance of theageing management programs (AMP).The WG developed a special set of indicators [14] under testing at many European NPPs.A special group of indicators are now made available [4] on the “supplemental workers” and the“supplier reliability” in general, by INPO. They are recognized as very useful to monitor one of themain causes of deficiencies in the maintenance systems (they are included for example in the AP-930)The techniques for the risk monitor during maintenance are also crucial, mainly in relation to theNUMARC 93-01 [12,13] proposal. The use of panel of experts and/or PSA for the construction of therisk matrix or of the risk monitor (real time) are apparently the only two available techniques.Some data bases are available on component reliability in Europe: for example the experience ofDACNE for PSA failure probabilities and for MR performance criteria (by Tecnatom), the EPIX (byINPO) and the PKMJ (by EPRI). However, most of them remain country specific and/or restricted tothe contributing users.The WG recognized that no tools are available yet to manage the maintenance process in acomprehensive manner, even if the EPRI proposals are excellent in some fields. The user groups(EPRI/NMAC, EPRI/MRUG, etc.) are providing an invaluable contribution to this concern to theirsubscribers.5. Use of PSA for maintenance optimizationIn case the maintenance optimization is supported by the application of PSA results and models [5],the quality of the PSA becomes an important issue for the success of the maintenance optimization. Asany PSA application, the maintenance optimization has crucial requirements for the PSA quality. Thescope, completeness, modelling details and used data should be such that allow the PSA to be used foradequate support of maintenance optimization. In order to ensure an appropriate PSA quality, asminimum the following actions should be implemented:• Use appropriate guidelines during development of PSA and review of PSA• Involve both PSA experts and NPP maintenance staff in the development of PSA models• Keep in mind the intended applications at the time of scope definition and if possible take intoaccount the available standards.• Perform PSA regulatory review before maintenance optimization is implemented.Basically two guidance for qualification of PSAs for specific applications are available, namely: theASME RA-S-2002 Standard for Probabilistic Risk Assessment for Nuclear Power Plant Applicationsand the IAEA TECDOC 1511 [15]. These documents facilitate determining how suitable a given PSAis for a specific application and in particular for supporting maintenance optimizations.In particular, maintenance related special PSA needs may include the following:

• Separation of the maintenance related basic events in the component unavailability models, likeunavailability due to repair, planned maintenance, test, human errors etc.• Modeling of maintenance activities in each of the safety system trains to correctly reflect actualmaintenance activities• Use of more detailed reliability models for modeling of PSA basic events, e.g. to identify failuremodes of components affected by different type of maintenance• Additional special models to support ISI, On-line maintenance, RI configuration control, etc...In addition, it was noted that risk monitors are useful tools to support maintenance planning off-lineand on-line restoration strategies in case of equipment failures during the plant operation.6. ConclusionsThe workshop identified some areas where some R&D effort is needed to support the fullimplementation of RCM models in European Countries. These areas cover research tasks and call foran initiative at the International Organizations level.In the field of regulatory practice, support would be needed in the licensing of advanced maintenanceoptimization applications and information on the regulation in the countries with good practices in thefield. In particular, the following recommendations for future support from international organizationswere identified:• Develop detailed guidelines for regulatory review of specific maintenance optimizationapplications such as: RI TS, RI ISI, On-line maintenance, etc.• Provide training and/or training material, tutorials for regulatory review of maintenanceoptimization applications.• Promote benchmark exercises.In relation to the PSA quality issues, need for support was identified in the following tasks:• Disseminate the available PSA quality guidelines (for example the IAEA TECDOC-1511) andpromote their development towards Level 2 PSA and at least internal floods and fires in orderto facilitate the regulatory use of the PSAs• Provide support for establishment of WWER specific component reliability databaseIn terms of research tasks able to make the RCM more broadly applied, the following was identified:• Clarification of the reliability target for the different groups of components and reliabilityparameters calculation• Integrated management of the data bases available at the plants: many sources of data areavailable at the plants (ISI, maintenance, AMP, PSA, operation, etc.) but often they are notintegrated and they do not support an integrated approach to component reliability.• Development of criteria for “good” performance of SSCs (acceptance criteria)• Identification of representative maintenance effectiveness indicators• Understanding of the impact of the RCM on the workforce: in relation to different competenciesneeded and overall reduction of the workforce at the sites• Comparison of the available methodologies for RCM: the available proposals are very muchaffected by the national frameworks where they have been developed. Benchmarking onselected systems and commodity groups would be very useful to this concern• Exchange of information at the EU level, despite of the national differences and plant issues,would be very useful in the following areas:• Methodologies for RCM• Organizational aspects• Derive failure rates for commodity groups (with some assumptions on anchoring, environment,etc.)• Develop guidelines for training of personnel and use of training centres in the field of optimisedmaintenance programs oriented to PLIM.

The WG concluded that there is a potential, very important role for the IE network on safe operationof nuclear installation (in the research field) in the coordination of the efforts among the EuropeanCountries to promote a full implementation of maintenance optimization programs.In fact the implementation of RCM methods requires the availability of component data, wellestablished probabilistic techniques of appropriate quality etc. that cannot be developed at the Countrylevel only. In this framework, any future action in the EU/FP7 [1] would be most probably verywelcome and will provide concrete support to the enhancement of the safety of the European Plants.7. References[1] WORK PROGRAMME 2007, EURATOM FOR NUCLEAR RESEARCH AND TRAININGACTIVITIES, European Commission C(2007) 564 of 26.02.07[2] IAEA, Principles and Guidelines on Plant Life Management for Long Term Operation of LightWater Reactors, IAEA Technical Reports Series- 448 (2006)[3] EUR 21903 EN, P.Contri et al., “Optimization of Maintenance Programmes at NPPs -Benchmarking study on implemented organizational Schemes, Advanced Methods and Strategies forMaintenance Optimization - Summary Report”[4] EUR 22603 EN, P.Contri, Summary report on the workshop “Maintenance rules: improvingmaintenance effectiveness”, Petten, 2006[5] EUR 22604 EN, V.Ranguelova, P.Contri, I.Kouzmina, Summary report on the Workshop on“Advanced Methods for Safety Assessment and Optimization of NPP Maintenance” Petten, 2006[6] P.Contri, T.Katona, “Safety aspects of long term operation of nuclear power plants”, SmiRT 17,Prague, 18 August 2003[7] Havel, R., Contri, P., Toth, C., Inagaki, T., Guerpinar, A., Misak, J., “Long term operation –Maintaining safety margins while extending plant lifetime”, International conference on topical issuesin nuclear installation safety, 18-22 October 2004, Beijing[8] US CFR (1998): Code of Federal Regulations 10 CFR Part 50.65, “Requirements for Monitoringthe Effectiveness of Maintenance at Nuclear Power Plants,” Office of the Federal Register NationalArchives and Records Administration, U. S. Government Printing Office, Washington DC.[9] US NRC (1997): Monitoring the Effectiveness of Maintenance at Nuclear Power Plants.Regulatory Guide 1.160, Revision 2[10] US NRC (2000): Assessing and managing risk before maintenance activities at Nuclear PowerPlants. Regulatory Guide 1.182, Revision 2[11] INPO 97-013, “Guidelines for the conduct of maintenance at nuclear power stations”, Dec1997[12] NUMARC 93-01, “Industry guideline for monitoring the effectiveness of maintenance atNuclear Power Plants, Revision 2, 1996[13] NUMARC 93-01, Section 11, “Assessment of risk resulting from the performance ofmaintenance activities”2000[14] EUR 22602 EN, P.Vaisnys et al., “Monitoring maintenance effectiveness using theperformance indicators”[15] IAEA, TECDOC 1511 - Determining the Quality of Probabilistic Safety Assessment (PSA)for Applications in Nuclear Power Plants, 2006.

PLANT LIFE EXTENSION (PLEX) & REPOWERING STUDYFOR EMBALSE NUCLEAR POWER PLANTMASSIMO TEGLIA, SERGIO ORLANDI, ENRICO GARRONEAnsaldo NucleareCorso Perrone 2516161 GenovaABSTRACTEmbalse (Cordoba) PLEX pre-project work has started and is being executed by AECL andAnsaldo Nucleare, for NPP and BOP systems respectively.The job is organized in order to get within two Plant Outages the residual life evaluation ofthe as built configuration of the Plant; within the outage 2010 (end of the Plant Design Life)it is planned to produce the intervention planning to be implemented in the years 2009 and2010 to get the required life extension.Through this strategic approach it would be possible to get the life extension without beingforced to stop the Plant for updating implementation.This PLEX for Embalse BOP first phase consists, as minimum, of Condition Assessmentand Residual Life Evaluation of the BOP, and – in parallel – investigation of potentialPower Uprate of the BOP, based on evaluation of the possibility to increase the powerassociated to thermal cycle design optimization.1. IntroductionNucleoelectrica Argentina S.A. (NASA), the Argentine Utility owner and operator of the EmbalseCANDU Nuclear Power Plant, has the firm intention of refurbishing and extending the design servicelife of the Embalse Nuclear Power Plant (Cordoba) up to 2035, providing a Plant Life Extension equalto 25 years. Embalse Plant Life Extension (PLEX) pre-project work has started in 2006 and is beingexecuted by AECL and Ansaldo Nucleare (ANN), for reactor building and non nuclear systemsrespectively.ANN is performing the first study activities on the Balance of Plant (BOP) equipment and systems, aspart of an overall Plant Life Management program. Further, in the frame of PLEX program, NASAhas assigned a contract to ANN to perform a Power Uprate study on Thermal Cycle, that will includethermal cycle efficiency improvement evaluation.2. Embalse Plant Life Extension Project - OverviewFirst Phase of the Refurbishment and Life Extension project for the Embalse Nuclear Power Stationconsists of all preparatory activities that are required to define the refurbishment scope and costs, forinput into the utility business case for the Refurbishment and Life Extension project.One of these activities is the Pre-Project Plant Condition Assessment Project as part of an overall PlantLife Management program, that provides for the systematic assessment, timely detection, mitigation,recording, and reporting of significant aging effects in Systems, Structures and Components.Main elements of the project are:i) Aging AssessmentSystematic aging assessments of Systems, Structures, Components (SSCs), or groups ofcomponents with similar characteristics (Commodities), selected according to a priorityprocess. Aging assessment generally entails a review of data in order to assess the effect ofaging degradation on SCCs; it establishes their current condition and provides a prognosisfor attainment of design life and/or long term operation with associated recommendations.They include:MD00006/DNU/21

• Condition Assessment (CA): Typically applied to SSCs or Commodities. Themethodology entails a general review of design, manufacturing, installation, operations andmaintenance at a component level.• Life Assessment and Residual Life Assessment (LA & RLA): Typically applied tocritical and complex components and structures, that are designed not to be replaced as part ofnormal maintenance program, and that are subject to long term degradation mechanisms. Themethodology entails a detailed review of design, manufacturing, installation, operations andmaintenance at a sub-component level.ii)ImplementationConclusions and recommendations emerging from aging assessment studies provide inputinto the implementation stage carried out by the Plant.Aging assessments will be executed to determine which of the selected SSCs are recommended forinspection, replacement or repair during the Refurbishment Outage and which may be done duringnormal maintenance outages. They will also provide a health prognosis for continued operation of theSCCs for life attainment and life extension beyond the refurbishment outage, and may identifychanges which are necessary and sufficient in order to deal with issues related to equipmentobsolescence and aging effects.3. Embalse PLEX – BOP AssessmentANN approach to Embalse PLEX program is divided in two main phases:‣ PHASE 1: engineering activities which will govern and address the walkdown inspections inthe BOP to assure Plant Life evaluation and extension up to the Customer requirement; it ispart of this stage also the Plant Power Uprate evaluation working on the Thermal Cycleparameters and re-evaluation of existing components in the frame of Plant Life Extension;‣ PHASE 2: Detailed / Constructive Design, Hardware Procurement and Installation togetherwith in field potential supervision and assistance considering the results of the differentwalkdowns addressed between the years 2007 and 2008 and to get the goal of 25 years ofPLEX as well as of potential identified Nuclear Power Plant Repowering.According to the above approach, a Technical Study is now on going for the Phase 1 EngineeringActivities, aimed to get the following objectives:i) Evaluation of the as-built configuration of the Thermal Cycle and Essential Systems inEmbalse BOP, taking advantage of the Plant Cognitive Walkdowns;ii)iii)Through the above Cognitive Walkdown visit results, identification of all the systems whichwill need inspection and checks in order to get a residual life evaluation based on the as builtconfiguration is performed; for all these systems and for the identified and selected inspection/ checks activities, an action planning to be addressed among the 2007 and 2008 plannedOutages (respecting and not affecting the already planned duration of each outage) is going tobe issued in order to assure the extension of the life of the Thermal Cycle as well as of theBOP essential systems for 25 (twenty five) years, as required by the Customer;At the completion of the In service inspection / Non Destructive Examinations / Tests actionsamong the planned outages, it will be issued a Global Integration Report, organized into twosequential sections:- Residual Life Evaluation (RLE) of the BOP in as built configuration;MD00006/DNU/22

- Intervention Planning (integrated with technical procurement specifications, bill ofmaterials, costs-benefits analysis and budgetary economical estimate of the updating /repowering engineering, supply and installations activities) addressed to get the plantlife extension for the BOP systems up to 25 years.iv)Evaluation of a potential Repowering of the Plant considering as variables potentialmodifications of the as-built Thermal Cycle configuration and replacement of agedcomponents of the cycle itself; implementation feasibility in hardware in the Plant BOP willbe part also of the “Cognitive Walkdown” visit .At the completion of the steps described here above and following the results of the walkdown in thesite integrated in the Global Integration Report, the Phase 2 Study activities will be produced for:- Engineering, Hardware procurement and assistance in equipment replacement in the field, soas to get the Plant Life Extension goal of 25 years;- Engineering, Hardware procurement and assistance in equipment replacement for ThermalCycle in the field to get the Plant Repowering;- Planning of the Engineering and Procurement Activities in order to produce specific activitiesand implement some modification during the outages prior to the end of the design Plant Life.3.1. Plant Life Extension Study - Phase 1Independently of their implementation approach in different sequential walkdowns in the Plant, thefollowing activities are on going, as part of the whole Phase 1 job:- Investigation of the as built configuration of the Embalse BOP- Investigation of potential Power Uprate of the Embalse BOP3.1.1. Investigation of the as built configuration of the Embalse BOPBased on the available Embalse documentation in terms of Process and Instruments Diagrams /Systems Technical Descriptions for BOP BSI as well as of BOP General Arrangement / CompositeDrawings, the following documentation / information are in progress to be processed:a) Systems identification to be investigated in the current aged configuration;b) Mechanical / Electrical Components identification to be in field investigated to check /monitor their aged life;c) Piping systems lists and associated design classification to be in field investigated tocheck / monitor their aged life;d) Supports lists (and associated piping systems) to be checked during the in field visit;e) BOP Control system as built configuration analysis.For each item to be processed through the walkdown visit in order to get information on the currentaged life, dedicated check lists have been prepared, covering the following information sections:- General information (i.e. Consistency between design and as built configuration, integratingpotential modification implemented during the spent Plant Life, Identification of experimentedmonitored thermal/pressure transients of the NPP which might have affected the mechanical /electrical components / piping systems life, Identification of external accidental events whichmight have affected SSCs design life, Identification of internal accidental events which mighthave affected BOP residual life, Components identification and current availability of thesame typologies in the market, Investigation on availability of in field documentation like“Embalse Plant Operational Transients experimented during the whole Plant Life”,“Periodical Maintenance Operations registration for the most critical electrical and mechanicalcomponents”, and “In Service Inspection registration” notes, Direct contact with PlantOperators and Maintenance / In service inspection personnel to catch direct experience ofMD00006/DNU/23

problems arised during the Plant design life, Analyses of existing operational log books fordetermining the utilization degree of the various components, along with the history of theirmaintenance, preservation, and use of spares, Identification of the industry codes andstandards applied for the original Embalse Project, and proposal of an equivalent set of codesand standards that could be used in case of components refurbishment/replacement. The basicidea is to demonstrate that the application of the least stringent codes and standards is stillacceptable to the National Safety Authority, thus simplifying the market investigation on thesame equipment types)- Detailed information through investigationAs built configuration for SSCs belonging to well identified characterization (selectedElectrical Systems and Mechanical Components in 2007 outage and for example pipingsystems in 2008 Outage); Ageing as built configuration for SSCs belonging to a well selectedidentified BOP Plant Area; in this way all the disciplines might be covered simultaneously inthe Residual Life evaluation of the essential systems, as a function of SSCs available in thatselected Plant Area.Non Destructive Examinations (NDE) for mechanical components / piping systems in order tocheck residual structural thickness or evaluate weldings configurations in critical joints.Specialistic checks (insulation tests, fire resistant tests, short circuit test) will be performed onelectrical components in order to evaluated the aged configuration of each electricalcomponent;Specialistic checks on currents as-built status and aged life of instrumentation for systems andcomponents will be performed in order to judge properly on their potential systematicreplacement;Specialistic check of current available BOP Control system configuration in order to properlyevaluate the opportunity of integration of a Distributed Control System (DCS) asimplemented for Cernavoda Unit 2 BOP.Based on the Integrated Multi-disciplinary Walkdowns Visit Report, specialistic analytical evaluationsto check the residual life of the Plant in BOP Portion for specific items will be performed.For piping systems analytical evaluations (piping stress analyses) will be made in order to checkstructural behavior simulating the residual thickness as per NDE investigation, to simulate potentiallocal permanent plastic deformation or non correct structural support configuration (springs workingas rigid support or out of the spring allowed range) range; through these analyses, the followingaspects will be properly evaluated:• RLE for each simulated portion of piping system in the as built configuration andidentification of the “critical” items which dictate the residual life for the investigatedpiping portion;• Applicability of Leak Before Break (LBB) theory for high energy Lines ;• Items / Spools / Supports to be replaced with new ones in order to assure a life extensionconsistent with the Customer requirements, assuming in terms of design loads, a similarloads / design thermal / pressure transients distribution as per the spent life;• Replacement Planning for each Piping System will be prepared and submitted to a cost /benefits analysis in order to assure the life extension required by the customer;• For each item to be replaced it will be supplied a delivery time schedule and a referenceprocurement specification integrated with the design data sheets.For mechanical components analytical evaluations (components stress analyses) will be made inorder to check structural behavior simulating the residual thickness as per NDE investigation, tosimulate potential local permanent plastic deformation or incorrect structural support configuration(springs / shock absorbers working as rigid support or out of the spring allowed range) range; the jobwill have two standard approaches addressed one for static components (tanks, heat exchangers,pressure components, chillers) and the other for dynamic components (pumps, fans, compressors,valves); through these analyses, it will be properly evaluated the following:MD00006/DNU/24

• RLE for each simulated component in the as built configuration and identification of the“critical” items which dictate the residual life;• Items / Spools / Supports to be replaced with new ones in order to assure a life extensionconsistent with the Customer requirements, assuming in terms of design loads, a similarloads / design thermal / pressure transients distribution as per the spent life ;• Replacement Planning for each Components or portion of itself will be prepared andsubmitted to a cost/ benefits analysis in order to assure the life extension as required bythe Customer;• For each item (or portion of the same) to be replaced it will be supplied a delivery timeschedule and a reference procurement specification integrated with the design datasheets.For electrical components analytical evaluations (components stress analyses) will be made, in orderto check structural behavior simulating the residual thickness as per NDE investigation or otherdamage got through visual inspection; the job will have an approach similar to the two standardapproaches as defined for mechanical components; additionally the job will be addressed to checkprimarily the component electrical functionality for a potential life extension as required by theCustomer; through these analyses, the following aspects will be properly evaluated:• RLE for each simulated component in the as built configuration and identification of the“critical” items which dictate the residual life, mainly in terms of electricalfunctionality;• Items / Supports / Components (Transformers, switchgears, circuit breakers Etc.) to bereplaced with new ones in order to assure a life extension consistent with the Customerrequirements , assuming in terms of design loads, a similar loads / design current /voltage transients distribution as per the spent life ;• Replacement Planning for each Components or portion of itself will be prepared andsubmitted to a cost/ benefits analysis in order to assure the life extension as required bythe Customer;• For each item (or portion of the same) to be replaced it will be supplied a delivery timeschedule and a reference procurement specification integrated with the design datasheets.For the BOP Control System, following the results of the different walkdowns investigation, it willbe properly evaluated the possibility to implement a new distributed control system , replacing thecurrent existing one cabled control system; the job will be performed according to the latest ANNexperience gained in the construction of the Nuclear Power Plant Cernavoda Unit 2 providing for thesame BOP a Distributed Control System (DCS).At the conclusion of the Phase 1, but its preparation can be addressed in parallel, will be issued the“Safety Report” for the PLEX activity, covering all the investigations studies addressed to finalizesystem by system the residual life evaluation as well as the intervention planning to get its requiredlife extension.3.1.2 Investigation of potential Repowering of the Embalse BOPThe possibility to increase the Power of the Embalse Nuclear Power Plant will be evaluated, as anintegration of the Plant Life Extension activity. It is ANN intention to evaluate the possibility of apotential Embalse NPP Repowering into two different stages belonging nevertheless both to thePhase1 job:° Stage 1 – Repowering associated to thermal cycle design optimization, without considering anymodification to the current configuration of the turbine blades;MD00006/DNU/25

° Stage 2 – Repowering associated to updating of the turbine blades lengths or modifications to beimplemented on the Turbo-generator system.In the frame of Stage 1, the following variables of the Thermal Cycle will be investigated:1 Moisture separator efficiency in the range 95 – 97 %;2 Pre-Heaters number in the thermal cycle;3 Decreased friction losses value in the main steam interconnecting lines turbine / pre-heater;4 Decreased head losses in high and low pressure inlet valves;5 Condenser vacuum evaluation;6 Double ReHeaters (RH) integration;7 Drainage calculation for turbine;8 Discharge Pressure at the high Pressure Stage of the Turbine;9 Efficiency of Electrical GeneratorResults of potential power increasing related to each one of the selected listed variables will be shownand justified.3.2 PROJECT STATUSANN has dedicated a project team composed by engineering specialists and on site technicians inorder to perform cognitive walkdowns and data gathering during normal operation of the plant and inOutage period, inspections and survey walkdowns during Outages. In particular, for data gatheringactivity, some resident specialists have been involved for some months.3.2.1 Cognitive Walkdowns during plant operation: STATUSTwo cognitive walkdowns have been performed on site in December 2006 and February 2007, by twospecialists teams and some preliminary visual inspections have been done. A third one is now ongoing. All the systems have been identified and a first priority system list has been defined, as perTable 3.2.1.BSI System Priority36100 Main Steam Systems (inside TB) 241120 Steam system including Moisture Separator 241130 Reheater Drains and Vents 242100 Main Condenser 142120 Extraction Air 243100 FW Heating & Extraction Steam (preheaters) 143210 Condensate Systems 143230 Feedwater Systems (inside TB) 143350 Steam to Feed-water Heaters 245100 Condenser Leak Detection & Sampling Systems 245400 Chemical Addition Systems 245510 Turbine Building Inactive Drainage 451100 Transmission Line 500kv + MOT 351300 Transmission Line 132kv + SST 351400 Phase Bus 22kv + main breaker + UST 352000 Standby Emergency Generators 353000 MV Distribution (Class IV & III) 354000 LV Distribution (Class IV & III) 355000 UPS, Class II & I (EPS excluded, AECL scope) 366200 BOP Control Centre Instrumentation Racks 466300 BOP Control Centre Logic Panels 471100 Pumphouse Common Systems (screen house) 1MD00006/DNU/26

71210 Circulating Water Supply System 171310 Service Water Systems (Outside R/B) 171400 Fire Protection System 471900 Chilled Water Systems (Outside R/B) 272130 Auxiliary Steam Condensate System 272140 Auxiliary Steam Distribution 273010 Hot Water System 475100 Compressed Air System 175120 Instrument Air System (Outside RB) 1Table 3.2.1: first priority system listDuring this phase ANN has searched and involved some original manufacturers in order to havetechnical consultation related to condition assessment of Main Components.3.2.2 Walkdown during Outage 2007: STATUSSome Plant Areas and Main Components have been selected for investigation during Outage 2007, forCognitive and for Surveillance purposes, as described hereafter:- Valves inspectable during Outage- Electrical Equipment inspectable during Outage- Piping (Expansion Joints and supports inspectable during Outage)- Feed Water Pumps (FWP) and Circulating Cooling Water pumps (CCWP)- Some Main components (Condenser, MSR, exchangers, filters, ecc.)- Galvanic protection measurementsMost of the planned activities have been performed; some activities have been planned for Operationand for next Outage period.3.2.3 Repowering study on BOP: STATUSThe relevant possible scenarios of Repowering have been found and presented to NASA.Considering the impact of each modification on the Thermal Cycle As-built configuration, four mainscenarios have been studied, as described hereafter:- modification of Steam Supply conditions to the 4 th Feed Water preheaters- addition of a Feed Water High Pressure preheater (5 th preheater) with new steam extractionfrom HP turbine- addition of a second Steam Reheater- Moisture Separator / Reheater efficiency improvementAt the same time each scenarios has been evaluated with the modification of Low Pressure Turbinelast stage blades.3.3 ConclusionsIt has been developed a global view of ANN approach in RLE and related PLEX study, including thecurrent project status, the performed activities and the repowering study.First main difficulty has been the managing of such great quantity of information from archives (ANN& Plant), from Plant Staff Interviews, from Inspection Reports, etc. that forced the team to build thesoftware component/pipeline database immediately.It has been demonstrated that a good coordination between ANN team and NASA Plant team, helpedby efficient communication procedures, gives the project a very important effort to reach the maingoal of Plant Life Extension, goal that seems to be reachable in a mid-term planning.MD00006/DNU/27

CHANGE OF THE COMPONENT IMPORTANCE VALUEACCORDING TO THE INITIATING EVENT MODELLINGAND A PROBABILISTIC SAFETY ASSESSMENTQUANTIFICATION METHODMEEJEONG HWANG, WOO SIK JUNG, JOON-EON YANGIntegrated Reliability Assessment Center, Korea Atomic Energy Research Institute1045 Daeduk-daero,Yusong-Gu, 305-353, Daejon – KoreaTel: +82-42-868-2832, Fax: +82-42-868-8256mjhwang@kaeri.re.krABSTRACTThis paper presents the elements that affect importance analysis result and it also provides amethod to obtain an accurate component importance ranking by considering those sameelements in a Probabilistic Safety Assessment (PSA). According to a change of anoperational method and regulation method on Nuclear Power Plants (NPPs) by using riskinformation, it is important to select the risk-significant or safety-significant components tomake the right decision for a risk-informed regulation and application. Thus, we reviewed animportance ranking change according to the cutoff value and the Initiating Event (IE)modelling method. The results of this study revealed that a failure inducing an IE and thecutoff value affect importance analysis results. Therefore, to obtain accurate importance−13estimation results, it is recommended that the cutoff value be lower than 1× 10 and that IEFault Trees (FT) be used during a PSA quantification process.1. IntroductionIn this paper, we present the change of a component importance ranking according to the methods usedto handle the Initiating Event (IE) frequency and the cutoff value during a PSA quantification process.In the Risk-Informed Applications (RIA) and the Risk-Informed Regulations (RIR), the importancemeasures for Structures, Systems, and Components (SSCs) provide some of the most useful PSAinformation. Therefore, an importance analysis is an important tool for providing appropriate PSAinformation for an NPP. Recently, risk information has been used as a method to make a decision on theselection of the risk-significant or safety-significant in the field of the RIR and RIA such as RiskInformed-In Service Inspection (RI-ISI), Graded Quality Assurance (GQA), and maintenance rule, etc[6][7] .Section 2 briefly details the method used in this study for an identification of the elements affecting theimportance estimation results. Section 3 explains the method used in our study. Section 4 presents theresults of the comparison of importance value according to the elements selected in this study. Theconcluding remarks are given in section 5.2. Identification of the elements affecting the importance estimation resultsTo obtain more accurate importance analysis results, it is necessary to identify the elements affecting animportance ranking.We have investigated whether the IE modelling method affects the importance analysis result or not inthis study. In general, there are two approaches to estimate the frequencies of IEs. They are a Fault Tree(FT) modelling method and the Bayesian analysis method when using historical data [1][2][3] . However,the IE frequency is usually handled as a value during a quantification process even though it is obtained

through an FT analysis. Thus, we have evaluated the fact that there will be an effect on the importancevalue of components if we use an IE FT directly instead of a frequency value during a quantificationprocess. It means that we might obtain an incorrect importance estimation result by not considering theinitiator effect on the components.However, if the same component failure event is used in an IE FT and a mitigating system FTsimultaneously, it is not easy to evaluate the importance value for a component. Thus, most importancevalues for components are estimated by considering the role as the mitigating system. Therefore, weshould include not only the role of a mitigating system but also an IE to obtain a realistic componentimportance value.In addition, we have evaluated the fact that the importance estimation result might be altered accordingto the cutoff value used during a quantification process.All PSA analysts have found that the Core Damage Frequency (CDF) value is different according to thecutoff value used during a quantification process. However, most PSA analysts have overlooked theeffect of the cutoff value on the importance estimation result. Generally, we have used a value of about1x10 -10 or so as a cutoff value because of a time constraint and a quantification engine’s limitation. Thus,we reviewed an importance value change according to a cutoff value change.3. The method used in this studyIn previous PSAs, most IE frequencies have been obtained from historical data. However, if a plant hasits own specific design of a system for triggering an IE, it is not appropriate to use the generic IEfrequency value estimated from historical data. In this case, researchers usually estimate the IEfrequency value through an FT analysis by considering the specific design of a system. They generallyuse an FT analysis method to estimate an IE frequency value for the support systems such as theComponent Cooling Water System (CCWS), Service Water System (SWS), Instrument Air system(IAS), and the Electrical Power System (EPS), etc. However, once they have obtain the IE frequencyvalue of a support system through an FT analysis, the IE frequency is treated in the same way as anyother IE is treated in an Event Tree (ET). To review the change of an importance value according to theIE modelling method, we revised the IE FT for Loss of CCW (LOCCW) to use it during a quantificationprocess. The major importance measures for the components relevant to an LOCCW were estimated andthe importance results were compared with the importance estimation results obtained by using afrequency value.To review the effect of a cutoff value change on the importance result, we compared a change of thenumber of basic events classified as safety-significant by changing the cutoff value from 1x10 -8 to1x10 -15 .We performed an importance analysis for the one top model for Level 1 of Ulchin 3&4, Rev. 1 by usingthe FT quantification S/W FTREX [10, 11] . The one top model of Ulchin 3&4 is composed of 49 IEs, 2,807gates and 2,498 basic events.4. Comparison of the importance estimation results4.1 Importance estimation results according to the IE modelling methodsSeveral components in the CCW system have different importance results in the case where the IE FT isused when compared with the case where the IE frequency value is used.Table 1 shows a part of the importance estimation results of the CCW components when the IEfrequency value and the IE FT are used, respectively. As presented in Table 1, the results show that theimportance value of the components is altered according to the IE modelling method. From theviewpoint of the Risk Achievement Worth (RAW), the importance value for the CCW systemcomponents, 3461CC-V0142, V0905, 3461M-PP01A, PP02A, and the Essential Chilled Water (ECW)system components, 3633M-CH02A, 3633WO-V1014A, were changed considerably. However, theimportance value for the other components has not changed that significantly.Through the results of a component importance value for the CCW system and the ECW system, wefound that component importance value for the ECW system is altered more when compared to theCCW component importance value. We concluded that this result is because the ECW system is more

dominant against a CDF than the CCW system. That is, it is understood that the importance value changeis not considerable since the minimal cutsets induced by the LOCCW are not dominant against a CDF.Therefore, if an IE that is dominant against a CDF was modelled as an FT, the importance value for thecomponents relevant to that IE would be changed considerably.In the example of this study, the chiller unit, 3633M-CH02A, was categorized as a non-safety significantcomponent for the case of a quantification process by using the LOCCW frequency value. However, thesame chiller unit was categorized as a safety-significant component after we performed thequantification process by using the LOCCW frequency FT. If a component’s RAW is above 2, it isclassified as a safety-significant component [6], [7] . Figure 1 graphically displays the importance valechange for some components when using the two IE modelling methods.Using IE Frequency ValueUsing IE FTComponent FV RRW RAW FV RRW RAW3461CC-V0073 0.000137 1.000137 2.24 0.000135 1.000135 2.263461CC-V0074 0.000899 1.0009 2.91 0.000417 1.000417 2.493461CC-V0095 0 1 1 0.00027 1.00027 1.193461CC-V0105 0.00003 1.00003 1.21 0.00003 1.00003 1.213461CC-V0106 0.00005 1.00005 1.22 0.000035 1.000035 1.213461CC-V0141 0.000186 1.000186 2.66 0.000182 1.000182 2.683461CC-V0142 0.001774 1.001777 4.06 0.000714 1.000715 3.143461CC-V0905 0 1 1 0.0048 1.004823 1.833461CC-V0906 0 1 1 0.000056 1.000056 1.283461CC-V1001 0 1 1 0.000002 1.000002 13461CC-V1002 0.000004 1.000004 1 0 1 13461M-HX01A 0.00002 1.00002 1.83 0.00002 1.00002 1.843461M-HX01B 0.002536 1.002543 106.67 0.002426 1.002432 102.083461M-PP01A 0.00033 1.00033 403.77 0.000376 1.000376 465.363461M-PP01B 0.000328 1.000328 382.75 0.000353 1.000353 443.593461M-PP02A 0.00033 1.00033 403.77 0.000385 1.000386 465.363461M-PP02B 0.000352 1.000352 403.79 0.000384 1.000384 465.363633M-CH02A 0 1 1 0.051079 1.053829 4.133633M-CH02B 0 1 1 0.000389 1.000389 1.443633M-PP01A 0 1 1 0.000008 1.000008 13633M-PP02A 0 1 1 0.000074 1.000074 1.023633WO-V1010A 0 1 1 0.00044 1.00044 1.443633WO-V1014A 0 1 1 0.000063 1.000063 1.28Tab 1: Comparison of the Importance of the components for the CCW and ECW SystemsFig 1: Comparison of the RAW4.2 Importance estimation results according to a change of the cutoff valueFigure 2 shows that the RAW is considerably underestimated in case where the cutoff value is high.

−15We can not solve the FT in the case where the cutoff value is lower than 1× 10 . Importance analysis−15results are summarized in Table 2, when the cutoff value is 1× 10 . Figure 2 displays that the cutoff−13value should be lower than 1× 10 so as not to overlook the safety-significant SSCs or to prevent aRAW underestimation.Number of events400350300250200150100500FV > 0.005RAW > 2.08 9 10 11 12 13 14 15k (Truncation limit = 1.0E-k)Fig 2: A comparison of the number of safety-significant basic events according to the cutoff valueElementsNumberCutoff value−151×10Minimal Cutsets 27,000,850Basic Event of Minimal Cutsets / Basic Event 1,754 / 2,498Basic Event (FV>0.005) 105Basic Event (RAW>2.0) 341Basic Event (FV>0.005 or RAW>2.0) 373Basic Event (FV>0.005 and RAW>2.0) 73Tab 2: The summary of importance estimation result for the cutoff value, 1×10−155. Concluding remarksAn importance value change for a component is very important in the RIR and the RIA since theseimportance measures are used as a method to select a safety-significant or risk-significant component.Through this study, we identified that a failure inducing an IE affects an importance estimation result forcomponents. Through a review of a component’s importance estimation results for the CCW system andthe ECW system, we found that the ECW component’s importance value was altered more whencompared with the CCW component’s importance value. From the importance value comparison, weconcluded that this result is because the ECW system is more dominant against a CDF than the CCWsystem is. That is, it is understood that the importance value change is not remarkable since the MinimalCutsets (MCSs) induced by an LOCCW are not dominant against a CDF. Therefore, if an IE which isdominant against a CDF such as a Loss of Offsite Power (LOOP) was modelled with an FT, acomponent’s importance value would be change considerably. Here, we only reflected the enablerevents of the IE FT events to estimate a component’s importance value because of a technical limitation.If we could reflect the effect of an initiator, the importance value for the components would be changedconsiderably. Thus, the development of an FT for other IEs is necessary for an estimation of a moreexact importance value for components.Also, we identified that the RAW was underestimated when the cutoff value is high. Moreover, wefound that we can prevent RAW underestimation when the cutoff value is lower than 1x10 -13 . However,the Fussel-Vesely (F-V) value was not affected by the cutoff value.Therefore, to obtain an exact importance value estimation result, it is recommended that the cutoff valuebe lower than 1x10 -13 .6. References

[1] W.S. Jung, “A Method for IE FT Development for LOCCW and Quantification,”U34-1FI-MM-IE-R0-003, 2004[2] Xing L., Fleming K.N., Loh W.T., “Comparison of Markov model and fault tree approach indetermining initiating event frequency for systems with two train configuration,” RESS, Vol. 53,pp.17-29, 1996.[3] M. Hwang, J-E Yang, W.S. Jung, “Development of a FT Modelling Method for Initiating EventFrequency Estimation,” Korean Nuclear Society Meeting Proceedings Vo1. 1, 2005[4] KEPCO, “Final Probabilistic Safety Assessment Report for Ulchin Units 3&4,” 1997[5] Kilyoo Kim, Dae Il Kang, Joon-Eon Yang, ‘On the Use of the Balancing Method for CalculatingComponent RAW Involving CCFs in SSC Categorization,” Reliability Engineering & System Safety,Vol. 87, Issue 2, Feb. 2005, pp 233-242[6] 10CFR 50.69, “Risk-Informed Categorization and Treatment of SSCs for Nuclear PowerReactors,” US NRC, 2002[7] Regulatory Guide 1.174, “An Approach for using Probability Risk Assessment In Risk-InformedDecisions On Plant-Specific Changes to the Licensing Basis,” US NRC, July, 1998[8] NUREG/CR-5750, “Rates of Initiating Events at U.S. Nuclear Power Plants: 1987-1995;Technical report. 1 Jan. 87~31 Dec. 95,” US NRC, 1999[9] Fatma Yimaz, Loys Bedell, “A Review of the Initiating Fault Tree Development and Usage at theENS Sites,” ICONE14-89696, Proceedings of ICONE 14, 2006[10] Jung WS, Han SH, Ha JJ, "A Fast BDD Algorithm for Large Coherent Fault Trees Analysis,"Reliability Engineering and System Safety, Vol. 83, pp. 369–374, 2004.[11] Jung WS, Han SH, Ha JJ, "Development of an Efficient BDD Algorithm to Solve Large FaultTrees," Proceedings of the 7th International Conference on Probabilistic Safety Assessment andManagement, June, Berlin, Germany, 2004.

Session 19.1.2Maintenance and operation

THE EFFECT OF SEA SALT AEROSOLS IN THE JAPAN SEACOAST FACILITIESF. NAKAYASU,T. NOMURAM. YASHIKIFukui University of Technology3-6-1 Gakuen, 910-8505 Fukui –JapanK. MATSUKAWAM. MIZUTANIT. UMEHARANuclear Engineering Ltd1-3-7 Tosabori Nishi-ku, 550-0001 Osaka – JapanABSTRACTAll of 15 nuclear power plants in Fukui Prefecture are located in the Japan Sea Coast.Corrosion is strongly influenced by material and environmental factors. We installed the saltdamage experimental yard at Awara sea coast in March, 2006. We are doing the open air test,sheltered test without filter and with filter.The carbon steel standard specimens were exposed under three kinds of test conditionsmentioned above. The corrosivity of the open air test specimens was higher than it of MiyakoIsland, Okinawa. The corrosivity of carbon steel of winter was higher than it of summer.The measurement of chloride deposition rate was done by the dry gauze method. Thecorrelation between chloride deposition rate and the average velocity of the wind was found.The correlation between chloride deposition rate and the average rainfall was a littlecomplicate. It means the chloride deposition rate has maximum value at a certain rainfallamount. No other clear relation was found.1. IntroductionFukui Prefecture presently hosts 15 rector units, all of which are located along the Sea of Japan. Undersuch a circumstance, corrosion due to sea salt aerosols is one of the major factors causing ageingdegradation of nuclear power plants facilities. Many scientists have been engaged in research on thecorrosion of structures due to sea salt aerosols 1-8) . In pursuing sea salt aerosol-induced corrosionresearch, it is necessary to perform corrosion tests under a certain set of environmental conditions sincecorrosion of structures is highly sensitive to environmental factors. In this respect, we installed theoutdoor exposure test facilities along the seacoast in Awara-city, Fukui Prefecture. To obtain data whichcan be compared with the past research results, we performed 3 types of exposure tests; placing standardtest pieces (carbon steel) in the open air environment and placing the same test pieces in the shieldedenvironment with and without a filter in the air intake.2. Testing methodThe outdoor exposure test facility located on the seacoast. This facility was manufactured and installedaccording to JIS Z 2381 (2001). The facility has 2 air intake openings (on the front and rear sides) inparallel with the coastal line. One facility has no filter on its air intake another has a filter. The standardtest pieces are installed on the facility roof while other test pieces installed inside the test facility. Boththe standard test pieces and other test pieces installed inside of the exposure test facility were

manufactured according to JIS Z 2383 (1998). These test pieces are sized in 10cm×10cm. After goingthrough the specified test period, we weighed the test pieces after removing corrosion products todetermine the corrosivity. In the short-term testing, we changed test pieces every day and evaluated thecorrosivity in the same manner.To measure airborne chlorides, the “dry gauze method” specified in JIS Z 2382 (1998) was adopted. Weplaced a dry gauze plate inside of the outdoor test facility and replaced the plate with a new one once amonth to measure the sea salt aerosols by ion chromatography. The area capturing sea salt aerosols wasset at 100cm 2 according to JIS Z 2382. In short-term testing, we replaced the dry gauze plate everyday,which was installed outside the test facility (in front of the filter) to eliminate the effect of the filter, andmeasured sea salt aerosols by ion chromatography in the same manner.3. Test results and evaluation of the results3.1 Sea salt aerosolsFigure-1shows monthly changes in seasalt aerosols and the filter trappingefficiency R from March 31, 2006through January 10, 2007. The filtertrapping efficiency R is defined asbelow:A − B R = ×100AWhere,RFilter trapping efficiencyASea salt aerosols entering inside ofthe test facility without filtersBSea salt aerosols entering inside ofthe test facility with filtersThe solid line in Figure-1 indicates seasalt aerosols entering inside of the testfacility without filters. As can be seen inthe figure, amounts of sea salt aerosoltend to be lower in the summer seasonand higher in the winter season with adifference by a factor of over 30. Thedifference between seasons is expectedto relate to the wind speed as will bementioned later. Accordingly, in thewinter season during which the windspeed is high, greater amounts of sea saltaerosol are dispersed in the air. Thedashed line in Figure 1 shows the filtertrapping efficiency. Figure 2 plots thefilter trapping efficiency versus amountsof sea salt aerosol.As shown in Figures 1 and 2, the filtertrapping efficiency remains almostconstant (i.e., 86-88%) when the amountof sea salt aerosol is small due to a lowerwind speed. On the other hand, asNaClmg/(m2day)Filter trapping efficiency (%)Sea salt aerosols(NaClmg/m 2 day)100806040207/27~9/20 3/31~5/15 6/16~7/265/16~6/15 9/21~10/17 11/9~12/1410/18~11/8 12/15~1/10Figure 1 Seasonal changes in sea salt aerosols Sea salt aerosols(NaClmg/m 2 /day) NaClmg/(m2day)Figure 2 Sea salt aerosols and filter trapping efficiencyamounts of sea salt aerosol increase, the filter trapping efficiency also increases getting closer to 99%.This may be because the filter has clogged and the pressure loss for replacement or the capacity of windto be treated is exceeded.%Filter trapping efficiency

We measured the amount of sea salt aerosolevery day to clarify the relationshipbetween the sea salt aerosol amount andenvironmental factors. The dry gauzemethod was also used in the dailymeasurement by placing the dry gauze platein front of the test facility opening. In theshort-term testing, we call the plate placedin the ocean side as the front gauze plate andthat placed in the mountain side as the reargauze plate. Figure 3 shows the dailymeasurements of sea salt aerosols. As canbe seen in this figure, amounts of sea saltaerosol in the air largely vary depending onthe day with the largest difference of about40 times. The amount of sea salt aerosolmeasured in the front filter was about 20times larger than that measured in the rearfilter. However the ratio is not consistent.When sea wind blows, amounts of sea saltaerosol increase while land wind blows,amounts of sea salt aerosol are reduced. It isexpected that the front gauze plate is mainlyexposed to sea wind while the rear gauzeplate is to land wind.We evaluated the correlation between seasalt aerosol amounts and environmentalfactors. Figure 4 indicates the monthlyaverage wind speed versus average amountsof sea salt aerosol. In this figure, averageamounts of sea salt aerosol are representedby the star mark for each range of windspeed which is categorized every 0.5m. Thefigure suggests a clear correlation betweenthe wind speed and sea salt aerosol amount.In this figure, the amounts of sea salt aerosolwere measured inside of the outdoorexposure test facility without filters.Figures 5 show the rainfall level versus seasalt aerosol amount. The sea salt aerosolamount marks the optimal value at therainfall level of about 5mm. This suggests ahypothesis that seawater (sea salt aerosols) isdispersed in the air when rain drops hit thesea surface, as the rainfall increases, theamount of sea water aerosols dispersed in theair increases; in the beginning, an increasingamount of sea salt aerosol is dispersed in theair because only a limited amount ofgenerated sea salt aerosol returns to the seasurface accompanying rain drops. When therainfall exceeds a certain level, most ofNaClmg/(m2day)Sea salt aerosols(mg/m 2 day) Sea salt aerosols(mg/m 2 day) 12/4 12/5 12/6 12/7 12/8 12/4-10Figure 3 Daily changes in sea salt aerosol Average wind speed (m/s) Figure 4 Wind speed vs. sea salt aerosols Rainfall(mm) Figure 5 Rainfall vs. sea salt aerosolsgenerated sea salt aerosols return to the sea surface accompanying rain drops and thus amounts of seasalt aerosol in the air are reduced.3.2 Corrosivity in standard test piecesSea salt aerosolsFront gauge plateMaximal pointRear gauze plate

We measured corrosivity in the standardtest pieces, which were placed outside ofthe outdoor test facility, by the method inaccordance with JIS-Z 2383. Figure 6compares the measurements with thosemeasured in Miyako Island, Choshi andNishihara, which are described in JIS Z2383. Our measurements reveal that thecorrosivity measured during the periodfrom March 31, 2006 through December15, 2006 including the winter season was1.7 times higher than that measured fromMarch 31, 2006 through Sep. 21, 2006which does not include the winter season.This difference is suspected to be becauseamounts of sea salt aerosol in the airincrease during the winter season (SeeFigure 1).It was found that the corrosivity wemeasured using the outdoor test facilityinstalled Awara city, Fukui Prefecturewas several times higher than thatmeasured in Miyako Island. It is suspectedthat the location where the outdoor testfacility was installed is highly corrosive.Accordingly, we believe that performingcorrosion resistance tests using thespecific test facility was worthwhile. Figure 7 shows the corrosivity in thestandard test pieces placed in the open air,test pieces placed under the shieldedenvironment with a filter and test piecesplace under the shielded environmentwithout a filter. The x-axis shows thenumber of days elapsed since March 31,2006. The y-axis represents the corrosivitywhich is standardized per year. As shownin this figure, the corrosivity per year for259-day exposure including the winterseason is higher than that for 174-dayexposure. These results suggest thatregarding the standard test specimens, thecorrosivity of carbon steel increases underthe environment in which a lot of sea saltaerosols are dispersed in the air.Corrosivity under the shielded environmentare much less than those in the standard testpieces which were subject to open airexposure. Furthermore, the corrosivityunder the shielded environment with filtersCorrosivity (g/m 2 /y) Corrosivity (g/m 2 /y) Corrosivity (g/m 2 /y)(g/m 2 )24002200200026001800160014001200’06.03.31~12.15 2366g/m 2 /y’06.03.31~09.21 1405g/m 2 /y Figure 7 Corrosivity vs. exposure periodMiyako IslandChoshiNishiharaFigure 6 Corrosivity Comparison in standard test piecesw/filteTest-2 06.3.31~12.15Test-1 06.3.31~9.21w/o filter(mg/m 2 ) Sea salt aerosols (relative value)Figure 8 Amount of sea salt aerosol vs corrosivityare lower than those under the shielded environment without filters. Although the use of filter reducesthe amount of sea salt aerosol to 12% or less (See Figures 1 and 2), the reduction of corrosivity is about50%. In addition, the corrosivity in standard test pieces subject to open air exposure increased by about70% from 174-day exposure to 259 exposures. However, the increase in corrosivity was about 17% inthe test pieces placed under the shielded environment with filters and about 15% without filters.

Figure 8 shows the correlation between amounts of sea salt aerosol and corrosivity, both of which aredescribed in the cumulative amount not being standardized. As can be seen in this figure, the amount ofsea salt aerosol increases by a factor of 2.1 during certain period when filters are used and by a factor of2.0 during same period when filters are not used. The corrosivity also increases by a factor of 1.7 in bothcases with or without filters. This suggests that although corrosivity in the initial stage varies dependingon the environmental conditions.4. ConclusionWe installed the outdoor exposure test facility on the sea coast in Fukui Prefecture, which belongs tothe seashore district in the southern part of the Sea of Japan, and implemented both open air and shieldedexposure tests. As a result we confirmed that: Amounts of sea salt aerosol in the air vary among the season. There is an approximate linear correlation between the amount of sea salt aerosols and averagewind speed. Regarding the relationship between the amount of sea salt aerosols and rainfall level, a maximalvalue of sea salt aerosols exists against a certain level of rainfall. It is difficult to clearly define the relationship between the amount of sea salt aerosols andenvironmental factors including the sunshine hour and temperature. The sea salt aerosol trapping efficiency of the filter, which is about 88% when the amount of seasalt aerosols is small, increases to almost 99% when the amount of sea salt aerosols is large. The corrosivity on the seashore of Awara city was higher than that in Miyako Island. Although the use of filter reduced the amount of sea salt aerosol in the air to 12% or less, thereduction of corrosivity was about 50%. In both cases with and without a filter, the cumulative corrosivity increases by a factor of 1.7 asthe cumulative amount of sea salt aerosols in the air increases by a factor of 2.5. References1) Akira Ono, “Method to analyze components constituting an aerosol particle and its applications,”Technical Report No.1 issued by Meteorological Research Institute (1978)2) Ayaka Kishikawa, Tomoko Kojima, “Characterization of aerosol particles,” Kyushu University(2005)3) “The science of air,” The Chemical Society of Japan Editorial Centre (1990)4) Kazuhiko Miura, “Releases and transformation of marine origin aerosols,” Solas Workshop inNagoya (2002)5) Yoshinori Mori, et.al., “Results of nationwide investigation into the amount of airborne salt (1),”Public Works Research Institute, No.2203 (1985)6) Fuminori Yamada, Tokuzo Hosoyamada, “Field observation of airborne salt generating from the seasurface and development of airborne salt generation and transport numerical model,” Coastalengineering papers Vol.50, pp1176-1180 (2003)7) Kenji Horita, Masaaki Hirano, “Research of generation of sea salt particles on the coastal area(No.2),” Architectural Institute of Japan, Structural papers No.455, pp207-213 (1994)8) Nobuharu Matunaga, et.al., Generation of splash due to revetment work and transportation of splashto the land behind the revetment, Coastal engineering paper, Vol41pp1046-1050 (1994)

COLLABORATIVE MACHINING SOLUTION EXTENDS THEOPERATING LIFE OF A NUCLEAR POWER PLANTGEOFF GILMORE, PH.DClimax Portable Machine Tools, Inc.2712 East Second St., Newberg, OR 97132 USAPhone: 1.503.538.2185, Fax: 1.503.538.7600Email: ggilmore@cpmt.comJAMES VANDENBERGBabcock & Wilcox Canada, Ltd.581 Coronation Blvd.,Cambridge, ON N1R 5V3 CanadaPhone: 1.519.621.2130Email: jpvandenberg@babcock.comANDREW BECKERClimax Portable Machine Tools, Inc.2712 East Second St., Newberg, OR 97132 USAPhone: 1.503.538.2185, Fax: 1.503.538.7600Email: abecker@cpmt.comABSTRACTExamination of a CANDU 6 nuclear power plant’s steam generators during a scheduled maintenance outagerevealed that the manway ports, part of the ASME Section III, Class 1 pressure boundary, needed repair. The port’sinner cover gasket was not seating properly. Integrity was at risk. It was determined that this operation wouldrequired a specialized machine to successfully repair the man-way port.The solution included the modification of a standard portable boring machine with a custom mounting option toenlarge the counterbore in the primary head shell from a round shape to an obround shape (76mm of shell thickness,16mm radially). The shape change was needed to accommodate the new obround cover and gasket seal design.Once the new major shape was machined, the repair was finished with a Computer Numerically Controlled (CNC)machine developed by the service team to achieve the necessary gasket face location and sizing.The final result met all of the plant’s expectations and was completed well within the time allotted during themaintenance shut down. This success was due to the positive partnership and collaboration of the service team andthe machine tool manufacturer working together to successfully extend the operating life of the nuclear power plant.1.0 BackgroundExamination of a CANDU 6 nuclear power plant’s steam generators during a scheduled maintenanceoutage revealed that the manway ports, part of the ASME Section III, Class 1 pressure boundary, neededrepair. The port’s inner cover gasket, which measured 355 mm by 456 mm, was not seating properly.Integrity was at risk. The solution team needed to determine how to balance the geometry of the port andredesign the covers’ gaskets.In addition, the steam generator’s hemispherical head was developed out of forged steel and while theexact properties of the material were unknown, it was obvious that the metal would provide substantial12007 European Nuclear Conference Paper # (O 3.68)

machining challenges. It was determined that this operation would require a specialized machine tosuccessfully repair the man-way port.2.0 Machine Design SpecificationsThe service team turned to a specialized manufacturer of on-site machine tools for initial engineeringguidance and solution recommendations. A team comprisedof service team engineers and project managers workingwith the machine tool manufacture’s engineering specialistsdeveloped a cutting repair solution to quickly andaccurately machine the difficult material.First, the team needed to determine the projectrequirements. They knew the tool had to operate in a tightspace and would need to withstand cutting tough material.Second, the solution had to perform a dry cut. Third, thetool needed to be rigid, accurate, and quick to set up.Finally, with a tight 24-hour time allotment to completethe machining, the machine also needed to cut speedily.Figure 1: The boring machine with 115Vvariable electric drive and variable feedwith special fixturing for mounting to thebottom of a steam generator.The tool’s mounting requirements were anotherconsideration. The team needed one stationary bracket to mount the tool because it would be in a crowdedarea where the primary head of the generators was surrounded by process piping and system components.This made access difficult. In addition, to accommodate unforeseen machining adjustments, on-site, thetool needed to include several interchangeable cutting heads.3.0 A Standard Machine RedesignMeeting all these machining requirements limited the engineers’ tool options. After much assessment andanalysis, the team decided to modify a standard boring machine. The tool’s ability to program precisecutter movements provided the flexibility necessary to perform onsite cutting of material with unknownqualities. The machine also offered precise control of the spindle RPM, allowing feed rates and cuttermovements to be fine-tuned by the operator at the repair area.The boring machine incorporated changeable tooling and cutting technology to provide accurate cuts andinserts. An adjustable mechanical stop and an incremental adjustment process were machined into the toolhead and bit. The cutting head was automatically fed axially on a traveling bar using the standard axialfeed screw with mechanical stops. The radial feed was manually adjustable using a tapered lockingmechanism. The machine also included a 108 mm diameter x 1219 mm long chromed bar, a rotationaldrive unit and an axial feed unit. The boring machine also featured an electric drive motor, 115v 50/60Hz, with two-speed gearboxes and an 115v remote control pendant with variable speed and stop start.To meet the job’s difficult attachment concerns, a special bearing mount was developed into a slidemechanism and attached to a modified version of the hydraulic chuck supplied by the service team. Themodified chuck incorporated a slide, which held the bearing and allowed it to move from one side of thebore to the other. The chuck was mounted using stops and screws at either end of the slide. A passagebore cut into the chuck enabled the hydraulic line to connect and move to either side of the bore.22007 European Nuclear Conference Paper # (O 3.68)

The service team engineers supplied the machine tool manufacturer with a 3-meter hydraulic hose with aquick connect for attaching to a hydraulic pump for activating the chucking system. A service teammanifold attached to the top of the slide mechanisms, and the lines were routed away from the bar usinghose brackets and looms to provide stability to the machine once it was mounted on-site.Figure 2: The boring machine with specialI.D. mounting chuck and tool head foraccurately positioning the cutting tool forboring a round hole to make it obround.A special standoff bracket supported the bar’s other end andused the manway pivot block as the primary mount. Thefixture was anchored against the wall of the steam generatorand the plant’s structural steel. Adjustable legs with jackingfeet were expandable to secure the fixture in place. Theteam designed a slide mechanism into the standoff bracketfor positioning the bar on both ends.Once the new boring tool was completed, the service team’soperational staff visited the machine tool manufacture’straining center for a full education on the tool’scapabilities. Machinists tested the boring tool bysimulating the repair on a replication of the repair site.This testing and training process provided the serviceteam with a level of confidence that the tool couldcomplete the work to specification within the allottedtime. Having the opportunity to learn how to use the machine before on-site work began also helpedreduce any on-site guesswork by machinists.4.0 Approval CyclesBefore deploying the tool, the machine tool manufacturer proved the boring machine’s capabilities to avalidation committee comprised of the service team and personnel from the nuclear plant. The acceptancecriteria were based on set-up times, cutting rates and reliability of operation over what was required at thesite. This was deemed necessary to ensure that the equipment could complete the machining functionssuccessfully, considering the uncertainties of the vessel material properties.The final solution met expectations and passed a test plan before any work was completed.5.0 The RepairThe machinists removed approximately 76 mm of shell thickness, 16 mm radially to enlarge the counterbore in the primary head shell from a round shape to an obround shape. The shape change was needed toaccommodate the new obround cover and gasket seal design.Once the new major shape was machined, they completed the repair by using a service team developedComputer Numerically Controlled (CNC) machine to achieve the necessary gasket face location and size.This machine also accurately produced the desired smooth finish of the obround shape.6.0 The Result32007 European Nuclear Conference Paper # (O 3.68)

The final repair results met all of the nuclear power plant’s expectations. Together, the teams exceededthe exacting cutting requirements and beat the time requirements.By working together, the service team and the machine tool manufacturer efficiently and collaborativelydefined the requirements and customized the tool according to specifications including space and locationcriteria. Additionally, they shortened the development process and accounted for all potential problemsduring the machining process. As a result, the machine tool manufacturer completed the machine fromquote to finished job in 12 weeks — well within the targeted outage time. Together, the service team andthe specialty machine tool manufacturer formed an effective partnership that was able to quickly completethe repair and insure continued operation of the nuclear power plant for many years to come.42007 European Nuclear Conference Paper # (O 3.68)

Track 2The Nuclear Fuel Cycle /Nuclear operations

Session 17.2.1:Reprocessing

RADIOACTIVE SPENT RESINS CONDITIONING BY THEHOT SUPER-COMPACTION PROCESS AT TIHANGE NPPALAIN LEMMENSNuclear Supports Department, ElectrabelRhodestraat 125 B-1630 – Linkebeek - BelgiumBAUDOUIN CENTNERNuclear Department, Tractebel EngineeringAvenue Ariane 7 – B-1200 Brussels - BelgiumABSTRACTAt Tihange NPP, spent ion-exchange resins were conditioned by embedding in a polymermatrix with a mobile processing installation. For safety and cost reasons, Electrabeldecided to investigate by which process the former one should be replaced.After a thorough technical economical analysis of the available proven processes, TractebelEngineering selected the Resin Hot Compaction Process to be installed at Tihange NPP.The Resin Hot Compaction Process is used to make water free dense homogeneous organicblocks from a wide range of particulate waste. In this process, spent resins are firstdewatered and dried to remove the inner structural water content. After drying, the resinsare placed into special metal drums, which are automatically provided with a lid andimmediately transferred to a high force compactor. After high force compaction, the pelletsare transferred to a measuring unit, where the dose rate, height and weight areautomatically measured and recorded.1. IntroductionIon exchange is one of the most common and effective treatment methods for liquid radioactive waste.Spent ion exchanger media are considered to be problematic waste that in many cases requires specialapproaches and precautions during its immobilization to meet the acceptance criteria for disposal.With the evolution of performance-based disposal facility acceptance criteria, it is now required thatspent ion exchange materials meet specific quality requirements prior to disposal. Where final disposalfacilities exist, waste acceptance criteria define, among others, the quality of waste forms for disposal,and therefore will sometimes define appropriate treatment options; for example, disposal facilitiesnormally define acceptable levels of free liquids and requirements for waste form stability as part oftheir waste acceptance criteria.The selection of treatment options for spent ion exchange materials must consider their physical,chemical and radiological characteristics. Basically, two of the main methods for the treatment ofspent organic ion exchange materials, following pretreatment methods like dewatering, grinding,foaming or decontamination by activity stripping are [1] :• Direct immobilization, producing a stable end product by using Cement, Bitumen, Polymer or HighIntegrity Containers• The complete removal of the resin inner structural water by a thermal processIn its first part, this paper will describe the principle of the process of the Resin Hot High ForceCompaction at Philippsburg NPP, Germany.In its second part, the paper will introduce a new application of the Resin Hot High Force CompactionProcess to Tihange NPP, Belgium.2. Process Description – Philippsburg Experience

At Philippsburg NPP the spent resins are first dewatered by a centrifuge (separator and decanter)system and filled in 200 l drums. The 200 l drums will be transferred to a thermal-oil heated dryingvessel that holds the contents of two 200 l drums (Fig. 1) and the resins will be removed by a specialdevice from the drum.Fig 1: Principle of the HPA - Resin Hot High Force Compaction Process at Philippsburg NPPThis vessel can also be used for mixing bead and powder resins. In the drying and mixing unit, theresins are heated to the necessary process temperature. After meeting the drying criteria, the resins arereleased into special metal press drums, which are automatically lidded and immediately transferred toa high force compactor. After high force compaction, the pellets are transferred to a measuring unit,where the dose rate, height and weight are automatically measured and recorded.The key benefits of the process are :- a high Volume Reduction Factor (typically 4:1) – Fig. 2- an end product in the form of a homogeneous organic pellet free of water – Fig 3Fig 2: Achievable Volume Reduction Factor 4:1Fig 3: Homogeneous organic pellet3. Experience from commercial operation at Philippsburg NPP3.1. Selection of the conditioning processVolume reduction was set as the main criterion in an environment with no final disposal options and alimited interim storage capacity.After successful performance of two trial campaigns with mobile equipment in 1990 and 1991, thelessons learned had been fed back in the design of the stationary equipment that has been started up in1995.3.2 Volume of Spent Ion Exchange Resins at Philippsburg NPPIn Philippsburg there are two units, one Boiling Water Reactor (BWR) with a power of 926 MW el(Philippsburg 1) and one Pressurized Water Reactor (PWR) with a power of 1.458 MW el(Philippsburg 2).The respective systems of the BWR are operated mostly with powdered resins, but in the PWR the useof bead resins is common. The annual volume of resins at Philippsburg NPP consists of approximately23 m³ powdered resins and approximately 3 m³ bead resins.3.3 Treatment of Spent Ion Exchange Resins• Dewatering and Filling of the Spent Ion Exchange ResinsThe Spent Ion Exchange Resins are pumped in tanks for interim storage. Prior to the filling of the 200l drums, the bead and powdered resins are dewatered by a stationary centrifuge system (Decanter/Separator).

• Conditioning operationsThe equipment installed in the waste conditioning building, especially the High Force Compactor, areoperated beside their function in the Resin Hot High Force Compaction Process also for the treatmentof mixed waste, evaporator concentrates as well as sludges. In this respect, the treatment of each wastestream is performed in campaigns in one shift operations. For the operation of the waste conditioningprocesses, two operators are needed. Additionally one operator is needed for the necessary craneactivities during transport from and back to the interim drum storage area as well as the overpackfilling with the resulting pellets.The throughput of the conditioning system for the Resin Hot High Force Compaction Process ismainly determined by the drying time, which depends closely on the contents of moisture in the resins.For that reason, the resins will be dewatered upstream first.After dewatering, the residual internal structural water inventory may be as high as 50-60% of thebead resin total weight.As mentioned before, the drying of bead resins will be done in Philippsburg NPP only in combinationwith powdered resins. Consequently the volume to be dried of those mixed resins is bigger than thevolume of powdered resins. Subsequently, the typical duration of one drying cycle varies from 1,6 h(powdered resins only) to 2,3 h (mixture from bead- and powdered resins).The average height of the pellets varies from 18,2 cm (powdered resins) to 22,4 cm (mixed resins).Consequently 3 to 4 pellets can be placed in a 200 l drum (Fig. 3). The operator can select fromvarious pellets the best combination to achieve the best filling grade. The average filling gradeachieved in the commercial operation is 3,9 pellets per drum.• Data Collection and Waste Tracking SystemAll relevant data of the product are measured during the conditioning process, are combined with therespective waste packages via barcode identification and are transferred to the plants waste trackingsystem and database (BAV).3.4 Duration of a campaignThe amount of powdered resins (23m³) and bead resins (3m³) corresponds to 234 drums of powderedresins and 32 drums of bead resins to be processed per year.Based on the above throughput data (12 respectively 15 drums per week), the total duration of anannual campaign can be calculated to be approximately 19 weeks.The conditioning process generates 266 pellets (1 drum of dewatered resins = 1 pellet) per year, thatwill be placed in approximately 68 drums. The volume of a drum being 0,27 m³, the resulting volumefor interim storage is 18,36 m³.Finally the overpack drums containing the pellets are interim stored on site.3.5 Evaluation of the processThe stationary facility at Philippsburg NPP has produced more than 3.800 pellets since 1995.The pellets generated in the stationary facility at Philippsburg, NPP have been packed in 200 l drums.Since December 31 st , 2006, in total 975 x 200 l drums have been packed. Some 328 of them have beendelivered to the final repository site in Morsleben and the rest is located in the interim drum stores ofPhilippsburg NPP.The advantage of the process is that products suitable for final disposal will be generated and, at thesame time, an important volume reduction for interim storage will be achieved. Moreover, in respectof the future final disposal options, decisions on the final waste packages according to futurerequirements are very flexible, because the pellets can be very easily retrieved from the 200 l drums orthe 200 l drums can be placed in final disposal packages.4. New application of the Resin Hot High Force Compaction Process to the NuclearPower Plant Tihange, Belgium4.1 Spent Resin Conditioning Process Selection at Tihange NPP• Tihange NPP approachThe Tihange NPP (3 x 1000 MWe PWRs) spent resin production amounts to some 10-12 m 3 /year.Until 2005, these resins were immobilised in an organic matrix (styrol, epoxy based materials) by useof a mobile unit provided by external service supply companies.

The resins are immobilized in the NIRAS-ONDRAF (*) licensed 400 liter drums, on the basis of onecampaign (30-36 m 3 ) every three years. The process global VRF was found to lay in the range 0.5 0.6Total volume of dewatered resins to be processed(VRF =).Total volume of the produced 400 liter drumsDue to concerns linked to the high cost of the process, Tihange NPP requested in 2005, from TractebelEngineering, a complete survey and reassessment of the currently available and industrially provenspent resin conditioning processes with, as final aim, the recommendation of the best suited processtaking into account the internal/external constraints prevailing at the plant.Each surveyed process was assessed in accordance with the 6 following criteria :- Overall cost including, as appropriate : investment, consumables, operation/maintenance,secondary waste management, process qualification, management of the conditioned wastepackages (transportation, interim storage, final disposal), dismantling (in case of a new fixedinstallation).- Autonomy (fixed installation) versus dependence (mobile external installation).- Manpower requested qualification for the process implementation.- Qualification of the process, i.e. compliance of the end product with the NIRAS-ONDRAF WasteAcceptance Criteria (WAC).- Nuclear and industrial safety (fire risk, presence of carcinogenic and/or mutagenic components inthe conditioning process).- Industrial References.For each process, each criterion was quoted from 1 (lowest ranked process) to 6 (best ranked process).A weighing factor was then attributed to each criterion, enabling to end up with an optimised proposal.Sensitivity analysis, including variations of the weighing factor numerical values, enabled to test therobustness of the proposal against uncertainties.• ConclusionsThe implementation of this multi-criteria analysis enabled to recommend the installation of a fixeddried resin hot super-compaction unit, i.e. the by far best ranked process, under the constraintsprevailing at Tihange NPP.4.2. Process ModificationsThe most critical issue for a successful process configuration is the duroplastic behavior of the resins,in particular of the bead resins. This behavior prevents the resins from building a homogenous, solidblock and, in the worst case, bursting the metal surface of the pellet (“Spring back effect” – see Fig.4). In this respect, the presence of powdered resins occupying the gaps between the bead resin and ofthe thermoplastic AF2 material, acting as glue, optimizes the process conditions in the case ofPhilippsburg NPP application.Therefore in order to adapt the Resin Hot High Force Compaction Process to a PWR plant with beadresin arising only, various intensive laboratory and full scale test trials had to be performed to adjustthe process parameters to handle this effect.Fig 4: "Spring Back Effect"Fig 5: Homogeneous Pellets resulting from Test Trials(*)NIRAS-ONDRAF : Belgian State owned company in charge of the collection, the conditioning and the finaldisposal of the radioactive waste produced in Belgium.

The test operations mainly focused on the selection of the most suitable type of additive material andthe minimization of the volume needed for running the process. In combination with the adjustment ofthe equipment parameters of the drying and compaction unit, finally an additive (polypropylene inpowder form) has been identified that assures the required properties of the resulting waste productwith a minimum volume needed (Fig. 5).4.3. Equipment ModificationsThe main components used in the application of the Resin Hot High Force Compaction Process forTihange NPP are a 500 liter conical drying unit as well as a 2000t Supercompactor like inPhilippsburg NPP. Nevertheless, taking the specific requirements into account, the followingmodifications have been introduced (Fig. 6 and 7):Fig 6: Tihange NPP – Spent Resin Hot compaction conditioning process – Simplified flow diagramFig 7: Resin Hot High Force Compaction Process for Tihange NPP, Belgiuma. Resin feeding and dewateringThe resins at Tihange NPP are stored in tanks and will not be dewatered and filled in drums prior totreatment. Actually, the resin storage tanks are located above the conical dryer, so the resins can be fedby gravity into the dryer. Therefore a pipe work connection of the drying vessel replaces the drum liftand docking equipment used at Philippsburg NPP. The pre-treatment step of the dewatering of theresins will be performed by a special device directly in the drying vessel prior to starting the dryingoperation.b. Additive Dosing DeviceSubsequent to the results of the test trials for the process modifications, a dosing device has beenadded to the system for adding the exact volume of additive (powdered polypropylene) needed into thedrying vessel for mixing in the resin material prior to release for compaction.c. Optimized Overpack Filling SystemThe resulting pellets from the Resin Hot High Force Compaction Process are considered relativelysimilar in terms of height, weight and dose rate. The overall duration of the resin conditioningcampaigns and installation maintenance activities will be shorter than 2 months per year. Therefore, inaddition to resin conditioning, the compaction system of the installation is considered to process alsomiscellaneous other solid waste streams.In order to guarantee an optimized filling grade of the final overpacks, a selection table for bufferstoring of pellets has been added to the system.4.4. Process specific data and performances• Volume reduction factor

The dried resins are loaded into 190 liter drums which, after receiving a lid, are super-compacted. Thepellets are then piled up into the 400 liter drums licensed for final disposal. The void volume betweenthe pellets and the drum inner walls is filled with grouting mortar.The overall volume reduction, i.e. ratio :VRF = Initial volumeof wet resins to be conditioned , is predicted to lay in the range 1,5-1,8Total volumeof 400 liter drums required by the conditioningpending upon the conservatism levels attributed to some parameters such as :- the specific weight of the to be conditioned wet resin,- the specific weight of the end product.Similarly, the weight of compacted dried resins is predicted to lay in the range 193-205 kg/400 literdrum.The so obtained VRF, even assessed on the basis of conservative assumptions, largely exceed thoseobtained by cementation processes and immobilisation processes in organic matrixes, such processesexhibiting VRFs < 1.• Processing RateAt Tihange NPP, the system is designed to process the maximum yearly resin production (12m 3 )within 26 days. As indicated in § 1.3., this criterion basically impacts the sizing of the conical dryer.• Layout requirements- Ground floor surface area : 13 x 7,5 m 2- Elevation (max) : 6.2 m5. ConclusionIn more than 10 years successful commercial operations at Philippsburg NPP, more than 3.800 drumsof resins have been processed. A volume reduction factor of 4:1 has been achieved by using thisprocess. The equipment has been proven as a reliable technology with reasonable operation andmaintenance cost.The new application of the process and the equipment for a use in the PWR plant in Tihange, Belgium,demonstrates, that the process can be adapted for the use in other types of plants by using the sameprinciple equipment. In an environment of very limited space resources for interim storage on site andthe absence of an operating final repository site, the process exhibits the following key advantages:• Achieving a volume reduction factor up to 4:1 for the interim storage instead of growing thevolume, i.e. instead of processes leading to VRF < 1• Achieving a water free end product• Creating a flexible waste product for interim storage, which can be retrieved, packed in drums orcubic containers according to future requirements or reclassified in terms of activity decades later• Utilizing well proven standard technologies like drying and compaction• Enabling, in the case of Tihange NPP, the grouting of the resin pellets with a mortar alreadyqualified for the grouting of compacted solid waste pellets, i.e. avoiding, so, the complex andheavy procedure of a new recipe qualification• Flexible use of the system components also for the super-compaction of other operational solidwaste streams outside of the resin conditioning campaignsREFERENCES[1] KONTEC ’93 Conference TranscriptE. Grundke, H.-D. Harass: Conditioning Radioactive Waste at the Point of Origin

BEHAVIOUR OF ZR IN A LICL-KCL EUTECTIC MELTDURING ELECTROREFININGP. SOUČEK, L. CASSAYRE, R. MALMBECK, E. MENDES, J.-P. GLATZEuropean Commission, JRC, Institute for Transuranium ElementsP.O. 2340, 76125 Karlsruhe, GermanyABSTRACTAn electrorefining process in molten chloride salts using solid aluminium cathodes torecover actinides (An) is developed in ITU. Complete recovery of actinides inelectrorefining of An-Zr alloy fuels can not be achieved without co-dissolution of Zr. It hasbeen shown in previous experiments that the presence of Zr ions resulted in a significantdeterioration of the quality of the deposit. The present work has been focussed on a betterunderstanding of the electrochemical behaviour of Zr ions in LiCl-KCl melt. In addition,electrorefining of UPuZr alloy in a melt containing dissolved Zr at the beginning of theprocess has been carried out. At first, Zr 4+ content was chemically reduced by addition ofUPuZr alloy forming a dispersion of Zr-containing particles in the melt. Electrorefiningprocess in this modified melt yielded compact deposits consisting of UAl 3 , UAl 4 , PuAl 3and PuAl 4 alloys.1. IntroductionPartitioning and transmutation technologies are developed worldwide in order to minimize the longterm radiotoxicity of the spent nuclear fuel produced during operation of the conventional nuclearreactors. Management of partitioning processes is also a key point in a development of future reactorsystems, where application of an advanced nuclear fuel cycle is considered [1-3]. Pyrochemicaltechniques represent a promising alternative to aqueous reprocessing processes. Electrorefining andliquid – liquid metal reductive extraction are the most developed pyrochemical techniques designedfor a group selective separation of actinides (An) from a mixture with lanthanides (Ln) in a molten saltmedia [4].An electrorefining process in the molten chloride salts using solid aluminium cathodes is beingdeveloped in ITU. In this process, actinides are recovered from a metallic An-Zr alloy fuel in a form ofAn-Al alloys. In previous experiments, excellent grouped separation of actinides from lanthanides hasbeen demonstrated [5, 6]. A very high capacity of aluminium to load actinides has been alsoexperimentally proven [6, 7]. However, a complete recovery of actinides during electrorefining of An-Zr alloy fuels can not be achieved without co-dissolution of Zr [8]. It has been observed in previousexperiments that the presence of Zr ions resulted in a deterioration in quality of the deposit [7].The presented work was focussed on a better understanding of the electrochemical behaviour of Zrions in LiCl-KCl melt and a determination of Zr distribution during the electrorefining process hasbeen estimated. On basis of the obtained results, electrorefining experiments using U-Pu-Zr alloy (71-19-10 wt. %) in a melt containing Zr ions at the beginning of the process have been demonstrated.2. ExperimentalAll experiments, storage and handling of chemicals were carried out in a glovebox under purified Aratmosphere (

The reference electrode used was an Ag/LiCl-KCl-AgCl (1 wt %) prepared in a Pyrex glass tube andall potentials stated in this work are related to the corresponding reference potential. Duringelectrorefining experiments, the working electrode was replaced by an aluminium plate and a tantalumbasket containing Zr metal or UPuZr alloy material was inserted instead of the auxiliary electrode.Both anodic and cathodic potentials were monitored during the electrorefining experiments using aspecially prepared combination of two multimeters with a laboratory power supply.The initial bath was prepared by oxidation of Zr metal (ITU material) in 40.259 g of LiCl-KCl eutecticmelt (Aldrich 99.99%) by reaction with BiCl 3 . Reduced Bi metal was collected in a Bi pool at thebottom of an Al 2 O 3 crucible. The initial concentration of Zr 4+ ions in the melt was 0.55 wt.%. Thechemical composition of the salt was monitored by ICP-MS analysis [9] of regularly taken saltsamples dissolved in 1 M nitric acid.3. Results and discussion3.1 Cyclic voltammetryCyclic voltammetry (CV) measurements were performed on inert W and reactive Al workingelectrodes. A comparison of the voltammograms measured on W electrode at temperatures 450 and500°C is shown in Figure 1a. Two electrochemical systems were detected: (i) a pair of peaks II c and II a(E p,c = -1.20 V at both temperatures, E p,a = -0.85 V / -0.67 V at 450 / 500°C, respectively) and (ii)two low intensity waves I c and I a at more positive potentials. The waves I indicate a soluble-solubletransition most likely associated with Zr 4+ /Zr 2+ reduction. The peaks II have a characteristic shapewhich corresponds to the formation of an insoluble product, which is probably Zr metal. In thistransition it seems that an over-potential is required to facilitate the deposition of the metal as thedifference between the potentials of the cathodic and anodic peaks became greater with increasingtemperature. The electrochemical reduction of Zr 4+ in molten LiCl-KCl can therefore be interpreted asa two-step reaction mechanism on an inert working electrode, which is in a good agreement with theavailable published data [10].A comparison of CVs measured on both electrodes at 450°C is shown in Figure 1b. On the reactivealuminium electrode (bold curve), one intensive cathodic peak was detected at E p,c = -1.26 V, followedonly by a negative current slope. As the Zr-Al phase diagram [11] contains several different alloys,this peak was expected to be associated with ZrAl 3 alloy formation. However, the reduction peaksstarts at the same potential as the one measured on W electrode, which could indicate a metaldeposition. A sharp current increase was detected on the anodic side, which disabled the distinctionbetween re-oxidation of reduced Zr species and dissolution of Al.3.2 Drawdown of Zr dissolved in LiCl-KClIn order to avoid complications associated with the presence of Zr dissolved in the bath used for theelectrorefining process, the possibility of Zr ions removal was examined (i) by chemical reaction withaluminium and (ii) by electrorefining of zirconium metal 40using solid Al cathode.The 150 chemical drawdown 450°C of Zr was studied in stationary conditions at 450°C IIusing a an aluminium plateIIwith dimensions 14×18×0.5 mm a30500°Cimmersed into the melt, which was composed of LiCl-KCl-ZrCl 4(0.55 120 wt.% ∼ 221 mg of Zr). A formation of ZrAl 3 alloy according to the reaction scheme (1) wasexpected, as a direct reduction yielding Zr metal and AlCl20 3 is thermodynamically impossible.90II 3ZrCl 4 + 13Al → 3ZrAl 3 + 4AlCl 3 (1)a10No 60 data concerning Zr-Al alloys are available to perform relevant calculations. I a0According 30 to the ICP-MS analyses of the salt samples before and after the experiment, the complete ZrI acontent was reduced. At the end of the experiment, the concentration of Zr in the bath dropped belowthe detection limit. However, only a very small amount-10I c0of reduced Zr was found in the deposit on theAl plate (19 mg, ICP-MS analysis of the complete deposit scraped from the electrode and dissolved innitric acid). The aluminium I c-30concentration in the melt -20increased to 0.18 wt. %, which supports theoccurrence of reaction (1), however the aluminium mass balance is likely II affected by AlCl 3cevaporation. -60-30W electrodeII cAl electrodea) b)-90-40i [mA]-1,80 -1,30 -0,80 -0,30 0,20E [V] vs. Ag/AgCli [mA]-1,70 -1,40 -1,10 -0,80 -0,50 -0,20E [V] vs. Ag/AgCl

Fig. 1 a) CV of the Zr system in LiCl-KCl measured on W electrode at 450 and 500°Cb) Comparison of the Zr system measured on W and Al electrodes at 450°CThe electrodeposition of Zr on solid Al electrode was investigated by electrorefining of Zr metal in themelt from the chemical drawdown experiment, i.e. LiCl-KCl-AlCl 3 (0.18 wt.%). Two connected runswere carried out using Al plates with an active area of app. 5 cm 2 . A total charge of 4070 C waspassed during the experiment. If the charge consumed for reduction of the aluminium present in thesalt at the beginning of the experiment (0.075 g) is removed, then the remaining charge of 3265 C forZr electrodeposition corresponds to 0.772 g of reduced Zr. The deposits from both runs were scrapedfrom the electrodes, dissolved and analysed by ICP-MS. The total Zr and Al masses in both depositswere determined to be only 0.008 g and 0.011 g, respectively. However, Zr concentration in the saltremained less than 0.11 wt %, i.e. max. 0.045 g of Zr was dissolved (see Table1 below). Reduced Zrthus most likely fell down from the electrode and remained in the salt in a form of fine metallicparticles.RunE-1Total charge[C]Zr dissolvedZr non-dissolvedanalysed [wt. %] calculated [wt. %] analysed[wt. %]0 0.00 0.13 < dl 0.18252 0.00 0.28 < dl 0.28932 0.11 0.56 < dl 0.151981 0.03 1.25 1.60 0.03Al dissolvedanalysed[wt. %]2860 0.01 1.78 < dl 0.05E-2 4070 0.08 2.43 2.00 0.05Table 1 Concentration profile during the Zr electrorefining experimentsDuring the Zr electrorefining, 6 salt samples was taken and dissolved in water instead of HNO 3 .However, only in 2 samples could the existence of non dissolved particles in the melt be observed.One explanation can be a non-homogenous particle distribution in the salt phase. A calculation wasmade in order to estimate the expected mass of Zr particles in the salt m (Zr particles ) according to thefollowing equation (2):mM ( Zr)⋅Q) = m ( Zr ) + − m ( Zr ) , (2)z ⋅ F( Zrparticles0 dissolvedwhere m(Zr 0 ) is the initial expected mass of Zr particles generated during the previous experimentalsteps (estimated accordingly) and m(Zr dissolved ) is the mass of Zr analysed by ICP-MS in the salt at thecorresponding phase of electrolysis. As shown in Table 1, both analysed values fit within a reasonablerange with the expected values calculated according to equation (2).3.3 Zr distribution during electrorefining processThe melt was left equilibrated after the electrorefining runs and an additional treatment was carried outto determine the final Zr distribution in the system and to prepare a new salt melt for the UPuZr alloyelectrorefining experiment. The following steps were carried out:1. The crucible was broken and the Bi pool (containing Zr particles) was removed2. The salt was re-melted in a new crucible and a fresh Bi pool was introduced3. BiCl 3 was stepwise introduced to oxidise all Zr particles remaining in the melt4. U-Pu-Zr alloy was introduced to reduce Zr ions oxidised in step 3 by reaction with Pu and UThe mass of the metallic Zr particles present in the melt was evaluated both from the known mass ofadded BiCl 3 (0.339g Zr, step 3) and from the mass of oxidised U and Pu obtained from ICP-MSanalyses of the salt samples (0.277 g Zr, step 4) making an average of 0.308 g. It seems that a largefraction of the Zr particles ends up in the Bi pool during equilibration. Samples from the removed Bipool were analysed by ICP-MS to complete the mass balance and an average value of 0.806 g of Zrwas detected. The summary of the Zr distribution is shown in Table 2. Although the sum of input and

output masses are in a reasonable agreement, the evaluated distribution is only approximate due to ahigh heterogeneity of the sampled material and uncertainties of the mass determination during theapplied chemical processes.InputOutputDistributionmass where mass whereratio [%]221 mg initial Zr mass in the salt 32 mg Zr dissolved in the melt 2.8962 mg308 mg Zr metallic particles in the melt 26.7Zr oxidised during806 mg Zr in Bi pool 69.8electrorefining8 mg Zr deposit on the electrodes 0.71183 mg sum input 1154 mg sum output 100.0Table 2 Mass balance of the Zr distributionThe process was followed by the CV measurements and the results correspond well to the expectedcomposition of the bath. An Al system was detected after the step 2, only peaks associated to a Zrsystem were observed after the step 3 and finally, a clear spectrum with peaks corresponding only to Uand Pu systems was measured after the step 4.3.4 UPuZr electrorefiningElectrorefining of the UPuZr alloy (ITU material, 71-19-10 wt.%) was carried out after the abovedescribed treatments, i.e. the melt after step 3, LiCl-KCl-ZrCl 4 . The Zr content was calculated to beapproximately 0.75 wt.%. A piece of UPuZr alloy (m 0 = 1.746 g) was introduced in the tantalumanode basket in order to chemically reduce the Zr by a reaction with Pu and U. After the reduction wascompleted, the basket was connected as an anode and used for the electrorefining experiment. Themass and composition of the starting anodic material was calculated to be m 1 = 0.835g and U:Pu:Zr =68.5:10.5:21.0 according to the composition of the bath before start of the electrorefining, which wasanalysed as LiCl-KCl-UCl 3 (1.39 wt.%)-PuCl 3 (0.74 wt.%). An aluminium plate used as the cathodehad the dimensions of 18×14×0.5 mm.Galvanostatic electrolysis was carried out and the maximum applied current was controlled withrespect to the cathodic potential, which was limited by a value of -1.3 V in order to maintainconditions suitable for a separation of An from Ln. Currents in a range between 20 – 50 mA wereapplied and a charge of 845 C was passed during the experiment. The cathodic potential was -1.15 V,which corresponds to the actinide-aluminium alloy formation, and showed slowly decreasingtendency. Up to a charge of 600 C, the anodic potential remained constant at -1.0 V, whichcorresponds to the dissolution of U and Pu. After this it increased up to -0.8 V, which indicates thestart of Zr co-oxidation.After termination of the run, a solid, compact and metallic-shiny deposit was found on the electrodesurface. The electrode containing the adhered salt was analysed by γ-spectroscopy to determine thetotal mass of deposited actinides and the mass ratio between uranium and plutonium. In an attempt toremove adhered salt from the electrode, a vacuum distillation step of the electrode was carried our at atemperature of 450°C, but only 0.096 g (22.6 % of the calculated salt mass) was removed. The depositwas then scraped off and analysed by XRD. The results are summarized in Table 3. Thealuminium/actinides mass ratio (Al:An) was calculated using data from γ-spectroscopy. XRD analysisproved the presence of UAl 3 , UAl 4 , PuAl 3 and PuAl 4 alloys, but it was not possible to quantify themass ratio among the presented species. XRD patterns for U- and Pu-Al alloys have the samepositions. In addition, an unidentifiable species was detected, which is most likely a Zr-Al alloy,although the Zr content could not be determined by XRD.Table 3 Mass balance of the UPuZr electrorefiningPassedchargeDeposit – mass of An[g]Deposit – analysis[g]Theoretic Analysed U Pu U:PuCurrentefficiency[%]Al:Anratio

0.591 0.600 ± 0.024 0.583 0.017 34:1 101.5 ± 4.2 1:1.764. ConclusionsStudies of Zr electrochemical behaviour have been carried out in an eutectic LiCl-KCl-ZrCl 4 (0.55wt.%) melt at 450 and 500°C on inert W and reactive Al electrodes. On an inert electrode, a two stepreduction mechanism was observed. Zr ions reduce into metal at a potential of -1.1 V vs. the Ag/LiCl-KCl-AgCl (1 wt.%) reference electrode, proceeded by the reduction of Zr 4+ to Zr 2+ at more positivepotentials. On reactive Al electrode, one reduction step was observed most probably yielding Al-Zralloys. However, the reduction starts at the same potential as detected on the inert electrode.Aluminium dissolution occurs practically at the same potential as Zr metal re-oxidation detected on Welectrode. Due to the potential overlap, dissolved Zr and solid Al creates a complicated system, inwhich a spontaneous chemical reactions between Al and Zr ions is possible.It has been shown that Zr ions in the melt can be completely chemically reduced both by solid Al andby actinides from a fresh UPuZr alloy. The formed product, most likely ZrAl 3 alloy, does not stick onthe Al surface during chemical reduction. In addition, during electrorefining of Zr metal, almost no Zrcontent was detected in the deposit. Reduction of Zr ions yields a fine dispersion of Zr-containingsolid particles in the bath.The electrorefining of UPuZr alloy on solid Al electrode was successfully carried out in LiCl-KClmelt, which contained app. 0.75 wt. % of Zr 4+ at the beginning of the process. Initially, Zr ions werecompletely reduced by the fresh fuel. Electrorefining process was carried out in the formed bath andcompact metallic deposit consisting from UAl 3 , UAl 4 , PuAl 3 and PuAl 4 alloys was obtained on Alelectrode with a very high current efficiency. Although Zr was co-oxidised during the last third of theprocess, it did not significantly influence the quality of the deposit.References[1] "A Technology Roadmap for Generation IV Nuclear Energy Systems", U.S. DOE Nuclear EnergyResearch Advisory Committee at the Generation IV International Forum, December 2002, GIF-002-00, http://gif.inel.gov[2] "Accelerator Driven Systems (ADS) and Fast Reactors (FR) in Advanced Nuclear Fuel Cycles - AComparative Study", ISBN 92-64-18482-1, OECD/NEA (2002)[3] J. Magill, V. Berthou, D. Haas, J. Galy, R. Schenkel, H.-W. Wiese, G. Heusener, J. Tommasi, G.Youinou, Nuclear Energy, 42, 263 (2003)[4] O. Conocar, N. Douyere, J.-P. Glatz, J. Lacquement, R. Malmbeck, J. Serp, "PromisingPyrochemical Actinide/Lanthanide Separation Processes Using Aluminium", Nucl. Sci. Eng., 153(2006) pp 253-261[5] J. Serp, M. Allibert, A. Le Terrier, R. Malmbeck, M. Ougier, J. Rebizant, J.-P. Glatz,"Electroseparation of Actinides from Lanthanides on Solid Aluminium Electrode in LiCl-KClEutectic Melts", J. Electrochem. Soc., 152 (2005) C167[6] L. Cassayre, R. Malmbeck, P. Masset, J. Rebizant, J. Serp, P. Souček, J.-P. Glatz, "Investigationof electrorefining of metallic alloy fuel onto solid Al cathodes", J. Nucl. Mater., 360 (2006) 49[7] P. Souček, L. Cassayre, R. Malmbeck, P. Masset, J. Serp, J.-P. Glatz: "Electrorefining of metallicU-Zr and U-Pu-Zr-alloy fuel onto solid Aluminium cathodes in molten chlorides", 9th IEM onActinide and Fission Product Partitioning and Transmutation, Nimes, France, 2006[8] R. K. Ahluwalia, T. Q. Hua, H. K. Geyer, Nuclear Technology, 133, (2001), 103-118[9] S. Abousahl, P. van Belle, H. Eberle, H. Ottmar, B. Lynch, P. Vallet, K. Mayer, M. Ougier,"Development of quantitative analytical methods for the control of actinides in a pyrochemicalpartitioning process", Radiochim. Acta, 93, (2005), pp. 147 – 153

[10] V. Smolenski, A. Laplace, j. Lacquement, "A Potentiometric Study of the Interaction of Zr(IV)and O(II) Ions in the LiCl-KCl Eutectic Molten Salt", Journal of the Electrochemical Society, 151(9) E302-E305 (2004)[11] T.B. Massalski, H. Okamoto, P.R. Subramanian and L. Kacprzak, "Binary Alloy Phase Diagrams,2nd ed.", ASM International, Ohio (1990)

1Main Results of EUROPART project for actinide partitioningCharles MADIC 1 , Michael J. HUDSON 2 , Noël OUVRIER 3 , Dominique WARIN 1 ,Pascal BARON 3 , Clément HILL 3 , Françoise ARNAUD 4 , Amparo G. ESPARTERO 5 ,Jean-François DESREUX 6 , Giuseppe MODOLO 7 , Rikard MALMBECK 8 ,Stéphane BOURG 3 , Giorgio DE ANGELIS 9 and Jan UHLIR 101/ CEA, DEN Saclay, France, 2/ University of Reading, United Kingdom, 3/ CEA, DEN Marcoule, France,4/ ECPM, CNRS, Strasbourg, France, 5/ CIEMAT, Madrid, Spain, 6/ University of Liège, Belgium, 7/ ISR-FZ Jülich, Germany,8/ EC-JRC-ITU, Karlsruhe, Germany, 9/ ENEA, Casaccia, Rome, Italy, 10/ NRI, Rez, Czech Republic___________________AbstractWithin the EC Sixth Framework Program, the EUROPART Integrated Project deals with the research fordeveloping the partitioning processes aiming at the recovery of Minor Actinides from the nuclear wastes.Twenty six partners from ten European countries (and one from Japan and one from Australia) contribute to theproject over a period of 3.5 years from 2004 to mid 2007. Both hydrometallurgy and pyrometallurgy are studied.HydrometallurgyFor the first step of co-extraction of An(III) and Ln(III), numerous new organic molecules have been designedand tested, as well as worldwide known extractants, such as DMDOHEMA or TODGA. The DMDOHEMA andTODGA extractants have demonstrated satisfactory performances even with concentrated nitric acid solutions. Acounter current flowsheet was designed for the "TODGA/TBP" mixture and successfully tested on genuinePUREX raffinate, proving the scientific feasibility of the concept. Moreover, new bis-diglycolamides have beensynthesised and showed up as promising co-extractants.For the second step of separation of An(III)/Ln(III), promising ligands are BTPs and BTBPs and a new referenceN-polydentate ligand has been defined. A flowsheet is currently being designed and spiked and hot tests will beperformed before the end of EUROPART.Numerous new calixarene derivatives, podants and cosans extractants have been also prepared. Two ligandsshow promising results: the narrow rim tetra-CMPO calix[4]arene for the co-extraction of An(III) and Ln(III),and the p-Hcalix[6]arene propoxypicolinamide for the An(III)/Ln(III) separation,.Finally, three conversion methods of the actinides into oxide and carbide compounds have been studied withsome good results: i) internal gelation, ii) external gelation and iii) co-precipitation.PyrometallurgyGood progress has been made on basic properties of actinides and some fission products in molten salts (halides)and in liquid metal solvents.Two efficient processes for the separation of An from Ln have been selected as promising cores of processes anddefined: i) electrorefining process on solid aluminum cathode in molten chloride, ii) liquid-liquid reductiveextraction in liquid aluminum-molten fluoride.For the decontamination of spent chloride salts coming from electrorefining, the complementary techniques ofzeolite ion-exchange and phosphate precipitation have been selected. For the conditioning of the spent moltensalts, several conditioning matrices have been studied including: i) sodalite and ii) pollucite.System studies investigating several potential pyrochemical processes were performed including: i) double-strataconcept (ADS), ii) IFR and iii) MSTR.Most of the knowledge and achievements acquired during the FP6 EUROPART project will be further exploitedin the frame of the FP7 ACSEPT Collaborative Project submitted to the European Commission.Acknowledgements: The European Commission is acknowledged for its financial support (contractEUROPART FP6-508854).

2I. INTRODUCTIONToday, after the reprocessing of spent nuclear fuels by the PUREX process, the high activeraffinates which contain the nuclear wastes, i.e. fission products and minor actinides (MAs),are then treated in order to incorporate the nuclear wastes into a solid glass matrix. In thefuture the glass wastes will be deposited into underground repositories. The definition of theseunderground repositories is complicated owing to the fact that the most importantradiotoxicity is related to the presence of MAs in the glass waste and this radiotoxicity issignificant even after more than 10 4 years, as shown on Figure 1. So, the elimination of theseMAs from the nuclear glass wastes will simplify greatly the definition of undergroundrepositories. After partitioning, the MAs can be transmuted into stable or short-lived fissionproducts, using for example the future ADS (Accelerator Driven System) facility. Theresearch in P&T domain is an important programme in Europe and in several nuclearcountries in the world [1]. In Europe, in the previous Framework Programmes (FP) withinEURATOM, several integrated projects (IP) were related to the definition of MA Partitioningprocesses.For example, in FP-5, three separate projects for An partitioning were organised [2]:i) two in hydrometallurgy, i.e. PARTNEW [3] and CALIXPART,ii) one in pyrometallurgy, PYROREP.Figure 1. Radiotoxic inventory of an UOX spent fuel (45GWd/t).In PARTNEW, several processes were tested, using malonamide extractants (DIAMEXprocess) and N- or S- bearing extractants for the selective actinide extraction (SANEXprocesses). In CALIXPART, numerous extracting agents were tested, mostly belonging to thefamily of calixarene molecules, where several complexing arms for trivalent MAs are graftedonto the calixarene cyclic structure. The PYROREP programme was located at the origin ofthe renewal of European research in pyrometallurgy for nuclear wastes.In EUROPART, two principal targets have been chosen for actinide partitioning:i) The aqueous high active raffinate(s) (HARs) or concentrate(s) (HACs) issuing fromthe reprocessing of spent nuclear fuels (UOX or MOX) by the PUREX process. It is

3considered that soon an improved PUREX process will be able to co-extractneptunium together with uranium and plutonium. Therefore, the aim for MApartitioning is the extraction of the americium(III) and curium(III). For high burn-upfuels, also berkelium(III) and californium(III) will be partitioned,ii) The irradiated fuels or targets after transmutation for example within the ADSfacility(ies). These irradiated fuels or targets will contain several An elements,including: U, Np, Pu, Am, Cm, Bk and Cf, which are to be partitioned all together.The partners who participated within the FP-5: PARTNEW, CALIXPART and PYROREPprogrammes are also participating in EUROPART. This paper will concern EUROPART andwill deal with the following three main areas:i) The structures of the Work Packages (WPs) of both hydrometallurgy andpyrometallurgy domains,ii) Some results obtained within these WPs,iii) The partnership and the organisation.II RESEARCH PROGRAMME AND RECENT RESULTSII.1 Research Programme and Work Package organisationEach WP is divided into different tasks, which will be presented below.II.1.1 HydrometallurgyThe definition of actinide partitioning process is the subject of the WPs 1 to 4. The respectivetargets for WP1 and WP2 are the high active raffinate (HAR) and high active concentrate(HAC) issuing from the reprocessing of nuclear spent fuels (UOX or MOX) by the PUREXprocess. The research to be done in WP1 and WP2 is in continuity with the work which wasdone during the PARTNEW (WP1) and CALIXPART (WP2) projects of the previous FP-5.The targets for WP3 and WP4 are the irradiated fuels or targets of the planned ADS actinidetransmutation facilities. The work to be done in WP3 and WP4 is also in continuity with theresearch which was done in PARTNEW (WP3) and CALIXPART (WP4). In WP3 and WP4the aim is to partition all the actinides, including An= U, Np, Pu, Am, Cm and if present Bkand Cf. After An partitioning it will be necessary to prepare compounds for fabrication of newfuels for An transmutation. This is the research to be done within the WP5.WP1 and WP2For both WP1 and WP2, the work is organised with the same types of tasks, listed below:Task 1 molecular modelling of complexation and extraction,Task 2 synthesis and characterisation of ligands,Task 3 study of their extracting properties (thermodynamics, kinetics),Task 4 determination of the structures of the ligands and their metalliccomplexes at molecular and supramolecular levels,Task 5 study of the stability of the ligands vs radiolysis and hydrolysis,Task 6 scaling-up of the synthesis for the ligand(s) selected for processdevelopment,Task 7 design of process flowsheet(s),Task 8 realisation of cold test(s) of the processes,Task 9 preliminary hot test(s) of the processes.WP3 and WP4Also, for WP3 and WP4 the tasks to be done are the same:Task 1 molecular modelling of complexation and extraction for bothliquid-liquid extraction and chromatography,Task 2 synthesis and characterisation of ligands,

4Task 3Task 4Task 5Task 6Task 7Task 8Task 9study of their extracting properties (thermodynamics, kinetics),determination of the structures of the ligands, mixtures ofligands, and of their metallic complexes, at molecular andsupramolecular levels,study of the stability of the ligands and mixtures of ligands vsradiolysis and hydrolysis,scaling-up of the synthesis for the ligand(s) selected for processdevelopment,design of process flow-sheet(s),realisation of cold test(s) of the processes,preliminary hot tests of the processes.WP5The aims and the tasks for WP5 are the following. Methods for the co-conversion of separatedactinides issuing from the partitioning processes for fuel preparation. The nature of the solidsto be prepared for fuel preparation are: i) oxides, ii) nitrides and iii) carbides. The coconversionmethods to be studied are: i) crystallisation, ii) precipitation and iii) sol-gel. Foreach type of solid to be prepared and each method of preparation, the following tasks will bestudied within WP5:Task 1Task 2Task 3Task 4determination of the performances of the co-conversion method,including: the kinetics of the reactions, the yields of coconversion,as functions of the operating conditions, such as thecomposition of the aqueous solution(s), the concentration of thereagent added, the temperature, etc..,chemical, physical and structural characterisations of thecompounds formed after separation from the aqueous solution(s),study of the conversion method(s) of the solid(s) formed fromthe solution(s) into the final compounds (oxide, nitride, carbide),chemical, physical and structural characterisations of the finalcompounds (oxide, nitride, carbide) prepared, with anestimation of their suitability to prepare fuel(s).II.1.2 PyrometallurgyThe work to be done in this domain has been organised in 4 WPs: i) WP6 is related to basicresearch, ii) WP7 concerns An partitioning process development, iii) WP8 is related to theconditioning of salts wastes and iv) WP9 concerns the study of systems for futurepyrochemical partitioning processes. The aims and tasks of each WP for pyrometallurgy arepresented below.WP6The knowledge of thermodynamical data of actinides (U to Cf) and some fission products(FPs) in molten salts and in liquid metals is the key point for the development ofpyrochemical processes, such as partitioning of actinides from HARs or HACs issued fromUOX and MOX fuel reprocessing by the PUREX process and for treatment of spent fuelsfrom advanced dedicated fuel cycles, including transmutation targets from ADS concepts.The molten salts (MS) media to be studied will be: i) chlorides, ii) fluorides.The research to be done within WP6 comprises the following tasks:Task 1 completion of the determination of basic properties of An (fromU to Cf if possible) and FPs within these MS media, taking inaccount the experimental procedures,

5Task 2compilation, comparison and analysis of thermodynamical data.WP7Process development is a key issue for the development of pyropartitioning processes, such aspartitioning of actinides from HARs or HACs issued from UOX and MOX fuel reprocessingby the PUREX process and for treatment of spent fuels from advanced dedicated fuel cycles,including transmutation targets from ADS concepts.The molten salts media to be studied will be: i) chlorides, ii) fluorides.The research to be done within WP7 comprises the following tasks:Task 1 process development for the partitioning of actinides fromHARs or HACs issued from UOX and MOX fuel reprocessingby the PUREX process based on several separation concepts,e.g. electrolysis, precipitation, liquid-liquid extraction withmetallic solvents, with the aim to demonstrate the feasibility ofhigh An recovery yields (99.9 %) with sufficientdecontamination factors vs fission products,Task 2 process development for the treatment of spent fuels fromadvanced dedicated fuel cycles based on several separationconcepts, e.g. electrolysis, precipitation, liquid-liquid extractionwith metallic solvents, with the aim to demonstrate feasibility ofhigh An recovery yields (99.9 %) with sufficientdecontamination factors vs fission products,Task 3 modelling of the processes and design of experimental devices(including their modelling) for process implementation.WP8Study of the conditioning of the wastes to be generated by the implementation of thepyrochemical processes to be developed. The wastes studied will be the spent salts (chloridesand fluorides). The work within WP8 will include the following tasks:Task 1 selection of solid matrix(ces) for wastes conditioning,Task 2 determination of the chemical and physical properties of theselected matrix(ces),Task 3 determination of the resistance vs aqueous leaching of themolten salt wastes conditioned within the selected matrix(ces),Task 4 study of conversion method(s) of the chloride salt wastes intooxides for subsequent conditioning of the radioactive wasteswithin a glass matrix.WP9System studies. For some pyrochemical processes selected, system studies will be performed.These studies within WP9 include the following tasks:Task 1Task 2Task 3identification of all the steps involved for each process,calculation of the fluxes of all the media involved within theprocess(es), including the fluxes of wastes. These fluxes couldbe normalised vs a produced quantity of electricity (TWhe),design of flow-sheet diagrams for the processes.

6II.2 ResultsAs it can be imagined, owing to the large number of talented scientists involved in theresearch, after three and half years of research work the quantity of results obtained, both infundamental and applied research, is very large. So, in this paper we will only illustrate someresults obtained within the different WPs.II. 2.1 HydrometallurgyWP1Computer modeling of extractants and of their complexes with actinides(III) (An(III)) andlanthanides(III) (Ln(III)) has been carried out. For example, Figure 2 presents the calculatedstructure of a 1/2 complex formed between a Ln(III) ion an a bis-pyridine-bis-1,2,4-triazine(BTBP) ligand (Liège, Partner 19 and Reading, Partner 22). Moreover, dynamic modelingwas also carried out (Strasbourg, Partner 20). For example, Figure 3 shows the final structureof Eu(BTP) 3 3+ + 24 free BTP molecules into a mixture of 95/5 vol% (chloroform/water). TheEu(III) complex is located at the interface between the two liquids.Figure 2. Model of a bis-complex between alanthanide ion and a bis-pyridine-bis-1,2,4-triazine(BTBP) ligand.Figure 3. Dynamic simulation of Eu(BTP) 3 3+ + 24 freeBTP molecules into a mixture of 95/5vol.%(chloroform/water).The study of structure of the metal complexes with extractants and of the supramolecularorganisation of the extracted complexes into the organic phases is also an important topic. Inthe case of the extraction of An(III) and Ln(III) with malonamides, it is very important tostudy the supramolecular organisation of the organic phases. The most important methods forstudying this point are SANS and SAXS (Small Angle Neutron or X-ray Scattering). Forexample, in the case of water extraction by the malonamide DMDBTDMA in solution in thediluent TPH, it has been shown shown that 2 supramolecular structures may be formed suchas micelles and pseudo-lamellar phases, see Figure 4, (CEA, Partner 1).The points represent the maximum ofwater solubilisation in the organic phase(limit for third phase apparition). Threedomains are present.Figure 4. Correlation between extraction of water andstructure of the organic phase.CyMe 4 -BTBPFigure 5. 2,6-bis-(5,5,8,8-tetramethyl-5,6,7,8-tetrahydro-benzo[1,2,4]triazy-3-yl)-[2,2’]bipyridine.

7For trivalent MA partitioning, the most difficult problem consists in the An(III)/Ln(III)separation. It should be noted that in the HARs and HACs, the molar ratio Ln(III)/An(III) isoften larger that thirty, because the Ln represent about one third of the total FPs. The newligand CyMe 4 BTBP, prepared at Reading (Partner 22), the formula of which is shown inFigure 5, is very effective for An(III)/Ln(III) separation as shown in Figure 6 (INE-FZK,Partner 12).Figure 6. Comparison of distribution ratios of Ln(III) and An(III) for an aqueousequilibrium nitric acid concentration of 0.88 M.The malonamide DMDOHEMA acts as a phase transfer reagent for the CyMe 4 BTBPextractant and significantly enhances the extraction kinetics of An(III) and Ln(III). Sometypical extraction kinetics are shown on Figure 7 (CEA, Partner 1). The CyMe 4 BTBPextractant has been selected for the design of a SANEX process (An(III)/Ln(III) separation) tobe carried out at the ITU (Partner 13) with real hot solutions.D M(III)10SF Am/Eu10001Am Eu Selectivity Am/Eu1000.1Mixing time (min)0.01100 10 20 30 40 50 60 70Organic solution: [CyMe 4 -BTBP] ini = 0.01 mol/L in the “n-octanol/[DMDOHEMA] = 0.25 mol/L” mixture, pre-equilibrated with molarnitric acid. V org = V aq = 700 μL. Temperature = (25 ± 0.5)°C.Aqueous solution: 152 Eu(III) and 241 Am(III) trace level in a surrogate SANEX-MOX feed: [HNO 3 ] = 1 mol/L + [Ln(III)] tot = 8.8 m mol/L.Figure 7. Kinetics of extraction of Am(III) and Eu(III) by CyMe 4 -BTBP from a surrogate SANEX-MOX feed.Cold tests of the DIAMEX process (with the malonamide DMDOHEMA), using hollowfibre-module(HFM) micro-plant, have been performed at INE-FZK (Partner 12) and will bedone soon with hot HAR at the ITU (Partner 13). The flow-sheet of the process is presentedon Figure 8.

8DMDOHEMAin TPHSpent solventExtractionHFMFP scrubHFMAcid scrubHFMStrippingHFMRaffinate Feed - HARFP scrub Acid scrub Product Strip sol’nFigure 8. The Hollow Fibre Module (HFM) microplant established during the DIAMEX extraction experiment.Process using TODGA and TBP extractants done at ITU (Partner 13) on genuine HARsolution.A total of 350 mL HAR solution, produced from PUREX reprocessing of a high burn-up (~ 75GWd/tHM) UO 2 fuel, has been treated by the TODGA + TBP process in the centrifugal extractorsystem installed in the hot cell facility in ITU. The total amount of An/Ln fraction produced was about240 mL. A 28 stage centrifugal counter-current flow-sheet, see Figure 9, designed by Partner 1 (C.Sorel) was used. Presently a 16 stage centrifugal set-up is available in the hot cells, meaning that theexperiment had to be done in two consecutive days, one for extraction and scrubbing and one for backextraction.For convenience the strip section was extended from 12 to 16 centrifuges. The results ofthis process were excellent.TODGA 0.2MTBP 0.5MIn TPH60mL/hSpentSolventExtraction1 4Scrub5 12 16Strip17 32RaffinateFEEDHNO 3 4.4MOxalic Acid0.2MHEDTA 0.05M60mL/hFP ScrubHNO 3 3MOxalic Acid 0.5MHEDTA 0.08M90mL/hAcid ScrubHNO 3 0.5M60mL/hProductStrip solutionHNO 3 0.01M60mL/hFigure 9. An(III+Ln(III) extraction by TODGA+TBP extractants from a genuine HAR.WP2 and WP4The scientists involved in WP2 and WP4 consider that common studies can be done in bothWPs. That is why, some results are presented jointly. Numerous types of extracting agents,belonging to different families: i) calixarenes, ii) cosans, iii) podants, have been prepared. Forexample, Figure 10 presents the scheme for the synthesis of CMPO-calix[4]arenes, done atthe JG University at Mainz, Germany, (Partner 14).

9OHOROHOR CH 3Oi)ii)iii)t-Bu4t-But-Bu 2t-But-Bu 245R CH 3OOOOOR CH 3NH 2OR CH 3NH 2 2iv)v)NO 2NO 2 2NH 26 7O OPhC PO 2 N OPh8OOCPNHPh PhOOCPPh Ph9a R = C 3 H 7i) R-Br, K 2 CO 3 , CH 3 CN; ii) MeI, NaH, DMF; iii) HNO 3 , CH 2 Cl 2 ;iv) Raney-Ni, H 2 , toluene; v) 8, NEt 3 (cat), CHCl 3Figure 10. Synthesis scheme for the preparation of CMPO-calix[4]arenes.b R = C 5 H 11Other molecules, like cosan-CMPO have also been prepared (Rez, Partner 15), as shown onFigure 11. Moreover, calixarene-cosans were also synthesised, as the molecule shown inFigure 12.Synthesis is based on two step procedure6212`1`6` 5`1011 12 9574 83Co 3`7`8`4`11`9` 12`10`OCOSANDIOXANATEOi. benzeneii. NH 2Riii. acidiffication21`651271 3 4Co2` 3`7`4`6`5`111011`9`988`12`OONHHR10`zwitterionic intermediatei. THF/ NaHO 2NOC OiiOH 2CPCoOOOOCo4-CoOOOOOOCo1011612952 71 3 4 8Co2` 3`7`1`8`4`6` 11`5` 9` 12`10`ORO NC OH 2CPOOOO O OOHNHNO HNHNNOON ONNR= t-OctFigure 11. Synthesis scheme of cosan-CMPO.12(anionic form)Figure 12. New cosan-calix[4]rene.After their synthesis, the extractant properties were tested. For example, for the extractantsprepared at UAM (Partner 18), see Figure 13, their extraction data vs Am(III) and Eu(III)were determined at CIEMAT (Partner 4), see Table 1.OMeOOOOOMe OMeHN HNHNOOOOOON N NC 8 H C 17 8 HC 17 8H 17NHNHOOONNC 8H 17C8H17MeOOO O OMe OMeO O OHNHNHNNH NH NHO O OO OON N NC 8 H 17 C 8 H 17C 8 H 17UAM-77 UAM-79 UAM-90Figure 13. Poly-malonamide extractants from UAM.

10HNO 3Ligand10 -2 M UAM-077 1.5·10 -2 M UAM-079 10 -2 M UAM-090(M) D Am D Eu D Am D Eu D Am D Eu3 0.017 0.015 0.012 0.010 0.12 0.454 0.018 0.018 0.017 0.016 - -5 0.016 0.018 0.015 0.014 - -Table 1. Results obtained for the extraction of Am/Eu, from nitric acid solutions, withUAM077, UAM-079 and UAM-090 ligands in n-octanol.WP3The examples chosen to be presented here concern: i) the use of solid extractant (SEX) toperform An partitioning, ii) the hot experiment for U/Pu partitioning using a modifiedPUREX process. Partner 5 (CTU, Prague), prepared several SEX incorporating extractingligands, such as malonamides, TODGA, cosans and calixarenes into polyacrylonitrile (PAN)polymer beads. After preparation, these SEX can be used for An chromatographic separations.Distribution coefficients of Am(III) and Eu(III) between aqueous nitric acid solutions and aDMDOHEMA-PAN are presented on Figure 14. At FZJ (Partner 9) successfulchromatographic partitioning experiments from a synthetic HAR, using a TODGA solidextractant was achieved, as shown on Figure 15.Dg / ml.g -11E+3Dg / ml.g -11E+31E+21E+21E+11E+11E+0Eu, no carrier addedAm, no carrier added1E+0Eu, 10-5 M EuAm, 10-5 M Eu1E-10,001 0,01 0,1 1 10[HNO3] / mol.L -11E-10,001 0,01 0,1 1 10[HNO 3] / mol.L -1Figure 14. Dependence of weight distribution ratios D g of Eu and Am on DMDOHEMA–PAN(HNO 3 ) solidextractant on nitric acid concentration in absence of any carrier or in the presence of 10 –5 M Eu carrier.(0.1 M NaNO 3 + HNO 3 , V/m = 250 mL/g, 20 hrs contact time).

114Step1 2 3 4 5 6U3.53LaPrCePdRhRuSrMoZr2.5AmCmYLaPrC/C 02NdSGdNdCeSm1.5Ru, Pd, Rh, Mo,Cs, Fe, Cs; PdSmUCsFe1ZrEuEAmCmCf0.5PdSrGdYCfEu00 10 20 30 40 50 60 70 80 90Fraction numberFigure 15. Chromatographic separation of a synthetic HAR by a TODGA column.Nexia Solutions (Partner 2) performed hot tests of a modified PUREX process for Pu/Useparation. The principe of the separation of Pu/U is based on the use of a Pu(IV) complexingagent: the acetohydroxamic acid (AHA). The process flow-sheet tested with real solutions ispresented on Figure 16. The results obtained were very successful. The decontaminationfactors (DF) were : DF(U/Pu)= 1.45.10 6 , DF(Pu/U)= 360.S130% TBP/ExD804.0 ml/min12 HA 98 HA 5 4 HS 1AP1~2.6 M HNO 31.74 ml/minS230% TBP/ExD802.0 ml/min28 URX 25SP1[U] ~ 40 g/L[Pu] ~ 27 g/L0.32 M HNO 3~4 ml/minHAF[U] 150 g/L[Pu] 100 g/L[Np(VI)] 0.1 g/L[Tc] 250 mg/L3M HNO 31.6 ml/minA13M HNO 30.64 ml/min24 PuS 21 20 PuS 17 16 PuS 13SP2[U]~27g/L[Pu] 0 g/L~7.0 ml/minAP2[Pu] ~ 27 g/L[U]~0.7 g/L0.52 M HNO 3~2.4 ml/minSolventAqueousA20.5 M AHA0.2 M HNO 34.0 ml/minFigure 16. Schematic of flowsheet for the third U/Pu split trial.WP5Several methods were studied for the co-conversion of actinides into oxide compounds: i) solgeland ii) co-precipitation. Until now, most of the work has been done with An surrogates.Figure 17 presents the beads which were prepared by sol-gel with a mixture of Zr(IV) + Y(III)+ Ce(III) (CEA, Partner 1).

12At FZJ (Partner 9), co-precipitation experiments were performed. For example, hydroxides ofTh(IV) and Ce(IV,III) were co-precipitated. Figure 18 presents the kinetics of this coprecipitation.aFigure 17. Zr/Y/Ce gel beads, T bath =80 °C, dried at T amb .pH98765432ThCeTh(IV)Ce(IV)Ce(III)-2 0 2 4 6 8 10 12 14 16Time t, minFigure 18. Coprecipitation of ThO 2 -50 % CeO 2 powder. Evolution ofthe pH and element concentration during precipitation.1.00.80.60.40.20.0[element] t/ [element] initialII.2.2 PyrometallurgyWP6The determination of the basic properties of An and FPs into: i) the molten salts, i.e. chloridesand fluorides mixtures, ii) the metallic solvent(s), is required for the future definition ofpyrochemical An partitioning processes, which can be based on electrodeposition, liquidliquidextraction (with a metallic solvent), or oxide precipitation. Electrochemical methodswere used for the study of the basic properties of An and FPs. For example, Figure 19presents the cyclic voltammogram of Np chloride in LiCl-KCl eutectic on a W electrode(ITU, Partner 13 and CIEMAT, Partner 4).I / A·cm -20.350.250.150.050.06 V s-10.080.090.10.20.30.4-0.05-0.15-2.3 -2.1 -1.9 -1.7 -1.5 -1.3 -1.1E / V vs AgCl/AgFigure 19. Cyclic voltammogram for Np in LiCl-KCl – left: full scale; right: study of the Np(III)/Np(0) couple.

13Among the low melting point metals, such as Al, Bi, Zn and Cd, Al was the most promisingone for the liquid-liquid reductive extraction An partitioning process (CEA, Partner 1). Somedoubts remained on the potentialities of Ga. Therefore, the activity coefficient of Pu into Gasolvent was determined. The obtained values confirm that Al is actually the best one. Inparallel, relevant activity coefficient data of the literature were compiled in a database (EDF,Partner 7).WP7Within the WP7 three main promising core of process are today under studies: theelectrorefining on solid aluminum cathode in molten chloride, the liquid-liquid reductiveextraction in molten fluoride/liquid aluminum and an electrochemical process in moltenfluoride. Successful An partitioning process by electrorefining into LiCl-KCl eutectic usingAl cathode was performed at ITU (Partner 13). It was shown that it was possible to chargemore than 3.6g of An (U to Am) in 4.5g of Al with a convenient decontamination factor(Figure 20).Figure 20. Experimental set-up used in the electrorefining experiments and aluminum cathode.With the liquid-liquid reductive extraction, separation factors over 1000 are obtained betweenAn and Ln. A two stages process will be efficient to recover more than 99.9 % of the An(Figure 21). The electrolysis in fluoride media is less developed and the feasibility of theseparation is still to be demonstrated. Anyway, promising results were obtained on U/Lnseparation.M D M S Am/MPu 197 ±30 0.73 ±0.21Am 144 ±20 1Ce 0,142 ±0,01 1014 ±213Sm 0,062 ±0,006 2323 ±488Eu 11000La 2400Figure 21. Mass distribution coefficients (D M =X metal /X salt ) of actinides and lanthanides with Al/Cu andseparation factors with Am (D Am /D M ). Pictures: salt before and after extraction (blue colour mainly due to Puand brownish colour mainly due to remaining Sm & Eu).To develop new processes, dedicated experimental facilities are needed and processmodelling has to be increased. The Pyrel II facility was built at ENEA (Partner 8) and a builtinelectrolyser in an inert glove-box was developed at NRI (Partner 16) (Figure 22). Processmodelling is under study on liquid-liquid reductive extraction (EDF, Partner 7) andelectrorefining in chloride media (Partners 1, CEA, 8 , ENEA and 13, ITU).

14Figure 22. Built-in electrolyser for fluoride media at NRI (left) and PYREL II electrolyser facility at ENEA(right).WP8The wastes produced during future An pyrochemical processes will be conditioned into solidmatrix. For chloride salts, the solid matrix sodalite has been selected by numerous partners.Figure 23 presents, for example, the XRD pattern of a sodalite doped with numerous alkaliions.S2000CsClLin (Counts)1000SCsAlSiO 4KLiAlSiOCsClSN(K,Rb)ClRbClKLiAlSiONCsAlSiO 4CsAlSiO 4SSKLiAlSiOCsCl(K,Rb)ClSSCsCl(K,Rb)ClCsCl010 20 30 40 50Figure 23. XRD of Cs, Rb, Li, K –doped sodalite sample synthesized at 700 °C for 48 h by reaction between saltand nepheline (S = sodalite, N= nepheline, KLiAlSiO 4 = potassium lithium aluminum silicate).Also, some chemical techniques were developed to eliminate FPs from the salts: i)precipitation by phosphate, ii) sorption on zeolite.

15WP9System study for future An pyrochemical processes is very important. For example, Figure 24presents a scheme related to the pyrochemical treatment of ADS fuel.CombustibleOxyde-MgO/GaineInoxConditionnementÀdéfinirPFGazeuxDégainageConcassageTriMatériauxde gainageFusiondescoquesDéchetmétalliquePF VolatilsCs, Rb, Tc,Cd, As, Se,TeSnBroyageTamisageTraitementthermiqueHydrofluorationH 2 , ArHF(g)SoutirageSel LiF-AlF3Appoint LiFDéchetFaibleactivitédécontaminéDissolutiondans LiF-AlF 3Ni/CuDigestion desmétaux noblesdans Zn/Ni/CuGe, Nb, Ru, Rh,Pd, Ag, Sb, MoDistillationZnMétauxnobles dansNi/CuPFF 3DistillationDu selExtractionréductrice des AnLiF+AlF 3 /Al ouAl-CuDéchetMétalliqueAl/CuVitrificationDésextractionoxydante dansLiCl-KClCuCl 2O 2-VerrePrécipitation parles O 2 -An OxydesFigure 24. System for the pyrochemical treatment of ADS spent fuel.III GENERAL INFORMATIONIII.1 PartnershipA total of 26 partners participated to EUROPART programme. These partners are fromUniversities and research organisations which are principally concerned with nuclearresearch. The list of the partners with the name of the corresponding scientist is given below:1/ CEA, DEN-VRH (Marcoule), France, (Prof. Charles Madic),2/ Nexia Solutions, Sellafield, United Kingdom, (Dr. Robert G. Lewin),3/ Chalmers University, Göteborg, Sweden (Prof. Christian Ekberg),4/ CIEMAT, Madrid, Spain, (Dr. Gabriel Pina),5/ University CTU, Prague, Czech Republic, (Prof. Jan John),6/ ECPM-CNRS, Strasbourg, France, (Dr. Françoise Arnaud),7/ EDF, R&D Division, Moret-s-Loing, France (Dr. Jorgen Finne),

168/ ENEA, Casaccia, Rome, Italy, (Dr. Giorgio De Angelis),9/ ISR, FZ-Jülich, Germany, (Dr. Giuseppe Modolo),10/ ICMAB, Barcelona, Spain, (Prof. Francesc Teixidor),11/ IIC, Rez, Czech Republic, (Dr. Bohumir Gruner),12/ INE, FZ-Karlsruhe, Germany, (Dr. Andreas Geist),13/ EC-JRC-ITU, Karlsruhe, Germany, (Dr. Rikard Malmbeck),14/ Johannes Gutenberg University, Mainz, (Prof. Volker Böhmer),15/ Katchem, Rez, Czech Republic, (Dr. Bohuslav Casensky),16/ NRI, Rez, Czech Republic, (Dr. Jan Uhlir),17/ Polimi, Milano, Italy, (Dr. Mario Mariani),18/ UAM University, Madrid, Spain, (Prof. Pilar Prados),19/ Liège University, Liège, Belgium, (Prof. Jean-François Desreux),20/ Louis Pasteur University, Strasbourg, France, (Prof. Georges Wipff),21/ Parma University, Parma, Italy, (Prof. Rocco Ungaro),22/ University of Reading, Reading, United Kingdom, (Dr. Michael J. Hudson),23/ University of Twente, Twente, The Netherlands, (Prof. Willem Verboom),24/ ICHTJ University, Warsaw, Poland, (Prof. Jerzy Narbutt),25/ CRIEPI, Tokyo, Japan, (Dr. Tadashi Inoue),26/ ANSTO, Autralia, (Dr. Michael La Robina).III.2 Organisation of EUROPARTA Governing Council exists to protect the interests of the partners, with participation of theresponsible scientist of each partner. The President of the Governing Council is Dr. Michael J.Hudson, who is from the University of Reading (U.K.). The technical work withinEUROPART is organised into nine Work Packages (WPs), with five for the domain ofhydrometallurgy and four for the pyrometallurgy. WP5 provides a link with EUROTRANS.Each WP has a principal scientist. The list of the WP leaders is the following:° WP1: Dr. Clément Hill, CEA-Marcoule, France,° WP2: Dr. Françoise Arnaud, ECPM, CNRS, Strasbourg, France,° WP3: Dr. Amparo G. Espartero, CIEMAT, Madrid, Spain,° WP4: Prof. Jean-François Desreux, Liège University, Belgium,° WP5: Dr. Giuseppe Modolo, ISR, FZ-Jülich, Germany,° WP6: Dr. Rikard Malmbeck, EC-JRC-ITU, Karlsruhe, Germany,° WP7: Dr. Stéphane Bourg, CEA-Marcoule, France,° WP8: Dr. Giorgio De Angelis, Casaccia, Rome, Italy,° WP9: Dr. Jan Uhlir, NRI, Rez, Czech Republic.An Executive Board exists, which is advises the co-ordinator, is composed of: ° all the WPleaders, ° Mr. Noël Ouvrier, Scientific Secretary of EUROPART, (CEA-Marcoule, France),and ° Prof. Charles Madic, (CEA-Saclay, France), Co-ordinator of EUROPART, and Dr.Michael J. Hudson (Chairman of the Governing Council).III.3 Exploitation of knowledgeThe knowledge to be obtained within EUROPART will be important for future research ofactinide partitioning, and also the processes developed will be important for the industrialcompanies involved in the reprocessing of nuclear spent fuels, like Areva NC (France) andBNFL (U.K.).

17III.4 Future development of MAs partitioning in FP7For the future of EUROPART within FP7, a new Integrated Project has been proposed toEuropean Commission on 2 May 2007. This project is named ACSEPT. It will be organisedin several research domains: 1/ hydrometallurgy, 2/ pyrometallurgy, 3/ process development,4/ fuels and targets developments, 5/ training and mobility. The future coordinator of thisproject is Dr. Emmanuel Touron from CEA-Marcoule (France). A total of 37 partners willparticiped to the project ACSEPT.IV CONCLUSIONThe studies carried out within EUROPART have lead to significant progress for theseparation of actinides and lanthanides particularly by hydrometallurgical andpyrometallurgical means. In the two domains of hydrometallurgy and pyrometallurgy thecombined academic and industrial research has been fruitful and successful for bothfundamental research and process development. In the last year of the programme increasingefforts will be related to process developments and there will be more hot-tests. Much morework needs to be carried out and it is hoped that the studies will be excellent within ACSEPTIntegrated Project of FP-7.References1. “Actinide and Fission Product Partitioning and Transmutation””. Eight Information Exchange Meeting,Las Vegas, Nevada, USA, 9-11 November 2004. NEA-OECD Report N° 6024 (2005).2. C. Madic, M. Lecomte, J.-F. Dozol and H. Boussier, 2004, “Advanced Chemical Separation of MinorActinides from High Level Nuclear Wastes”, in the Proceedings of the Conference EURADWASTE’04, 29March -1 April 2004, Luxembourg.EUR 21027.3. C. Madic, F. Testard, M. J. Hudson, J.-O. Liljenzin, B. Christiansen, M. Ferrando, A. Facchini, A. Geist, G.Modolo, A. Gonzalez-Espartero and J. De Mendoza, 2004, “PARTNEW. Nouveaux Procédés d’Extractionpar Solvant pour les Actinides Mineurs. Rapport Final ». Rapport CEA-R-6066.4. M. R. S. Foreman et al., “Complexes formed between the quadridentate, heterocyclic molecules 6,6’-bis-(5,6-dialkyl-1,2,4-triazine-3-yl)-2,2’-bipyridine (BTBP) and lanthanides(III): implications for thepartitioning of actinides(III) and lanthanides(III). Dalton Trans. 2006, 1645-1653.5. M. Nilsson, S. Andersson, C. Eckberg, M.R.S. Foreman, M.J. Hudson and G. Skarnemark, “Inhibitingradiolysis of BTP molecules by addition of nitrobenzene”, Radiochim. Acta 94, 1-4 (2006).6. M. G.B. Drew, M. J. Hudson and T.Youngs “QSAR studies of multidentate nitrogen ligands used inlanthanide and actinide extraction processes” J. Alloys and compounds, 374 (2004) 408-415.7. M. G.B. Drew, M. R. St. J. Foreman, M. J. Hudson and K. Kennedy, “Structural Studies of LanthanideComplexes with Tetradentate Nitrogen Ligands” Inorganica Chimica Acta, 2004, 357 4102-4112.8. M. J. Hudson, J. W. Peckett and P. J.F. Harris, “Low-Temperature Sol-Gel Preparation of OrderedNanoparticles of Tungsten Carbide/Oxide” Ind. Eng Chem. Res., 44(15) 5575-5578, 2005.9. M. G.B. Drew, M. J. Hudson*, M. R.S. Foreman*, C. Hill, C. Madic. “6,6’-bis-(5-6-diethyl-[1,2,4]triazin-3-yl)-2,2’-biyridyl the first example of a new class of quadridentate heterocyclic extraction reagents for theseparation of americium(III) and europium(III)”, Inorganic Chemical Communications, 8/3 239-241, 2005.10. M. G.B. Drew, M. J. Hudson, K. Kennedy and M. R.S. Foreman, “Structural Studies with Lanthanides(III)and a Tetradentate Ligand” Inorganica Chimica Acta, 357, 2004 , 4102-4112.11. M. Weigl, M. A. Denecke, P. J. Panak, A. Geist, K. Gompper, “EXAFS and time-resolved laser fluorescencespectroscopy (TRLFS) investigations of the structure of Cm(III)/Eu(III) complexed withdi(chlorophenyl)dithiophosphinic acid and different synergistic agents”. Dalton Trans., 2005, 1281-1286.12. M.A. Denecke, A. Rossberg, P.J. Panak, M. Weigl, B. Schimmelpfennig, A. Geist,« Characterization and comparison of Cm(III) and Eu(III) complexed with 2,6-di(5,6-dipropyl-1,2,4-triazin-3-yl)pyridine using EXAFS, TRFLS, and quantum-chemical methods”. Inorg. Chem. 2005, 44 (23),8418-8425.13. A. Geist, M. Weigl, K. Gompper, “Small-scale actinide(III) partitioning processes in miniature hollow fibremodules”. Radiochim. Acta, 2005, 93, 197-202.

1814. M.R.S. Foreman, M.J. Hudson, A. Geist, C. Madic, M. Weigl, “An investigation into the extraction ofamericium(III), lanthanides and d-block metals by 6,6’-bis(5,6-dipentyl-[1,2,4]triazin-3-yl)-[2,2’]bipyridinyl (C5-BTBP)”.Solvent Extr. Ion Exch., 2005, 23 (5), 645-662.15. M.G.B. Drew, M.R.S. Foreman, A. Geist, M.J. Hudson, F. Marken, V. Norman, M. Weigl, “Synthesis,structure, and redox states of homoleptic d-block metal complexes with bis-1,2,4-triazin-3-yl-pyridine and1,2,4-triazin-3-yl-bipyridine extractants” Polyhedron (published on-line 28.11.2005).16. A. Geist, M. A. Denecke, P. J. Panak, M. Weigl, B. Schimmelpfennig, K. Gompper, « Studien zurSelektivität von Di-triazinyl-pyridinen: EXAFS, TRLFS und quantenchemische Rechnungen”. Nachrichten –Forschungszentrum Karlsruhe, 2005, 37 (4)17. A. Geist, C. Hill, G. Modolo, M.R.St.J. Foreman, M. Weigl, K. Gompper, M.J. Hudson, C. Madic, “6,6’-Bis(5,5,8,8-tetramethyl-5,6,7,8-tetrahydro-benzo-[1,2,4]triazin-3-yl)[2,2’]bipyridine, an effective extractingagent for the separation of americium(III) and curium(III) from the lanthanides”. Solvent Extr. Ion Exch.2006, 24, 463-483AcknowledgementsDr. Ved Bhatnagar, from the European Commission, who is the Scientific Officer forEUROPART is greatly acknowledged.

Session 17.2.2:Fuel

ENC 2007 Conference, Brussels, Belgium, September 17-19 2007FABRICATION OF PLUTONIUM BASED COATEDPARTICLE FUEL FOR HIGH TEMPERATURE REACTORSD. Haas, A. Fernandez, J. Somers 1Joint Research CentreInstitute for Transuranium ElementsPostfach 2340, D-76125 Karlsruhe, GermanyABSTRACTPlutonium incineration was made in both the Dragon and Peach Bottom reactors, and veryhigh burnups were achieved. At the Institute for Transuranium Elements, an experimentalprogramme on the incineration of Pu, and eventually minor actinides (MA) in HighTemperature Reactors has been initiated. Use is made of sol gel and infiltration techniquesfor the production of the fuel kernels, whereby the inert matrix fuel concept is favoured,both for ease of fabrication, and also for improved fuel irradiation performance andtransmutation efficiency.In parallel, the construction of a facility for the production of the coating layers on suchtransuranic fuel kernels is underway. The conventional chemical vapour deposition in afluidised bed route is being followed for the production of the buffer, silicon carbide andpyrocarbon layers. Eventually, zirconium carbide is also considered as a sealing layer. Theinstallation of a full set of kernel and coated particle selection and characterisationequipment is in progress.This paper outlines the scope of the project, provides details of the characterisation methodsto be deployed, and above all, the methods used to integrate this complex set of equipmentin a series of gloveboxes, so that all necessary safety and security measures are met. Thecoater facility will be commissioned in 2008 and first production of Pu based coatedparticles is planned for 2009.1. INTRODUCTIONThe first of the next generation of High Temperature Reactors (HTR) will undoubtedly be fuelled withUO 2 , but their versatility is such that they can also be used for plutonium or minor actinide (MA)incineration (1-4), i.e. the “deep burn” concept (5). For this purpose the HTR offers a uniquepossibility of using a full plutonium core, without recourse to U to maintain safety margins. Burnablepoisons (e.g. erbium) can be required, however, to preserve the inherent safety of the system (2, 6).The disposal of military Pu in a Gas Turbine-Modular Helium Reactor (GT-MHR) is being designedand developed in Russia (7, 8). Pu based coated particles have been irradiated in the past and fuelburnups of 750 GWd/tonne were demonstrated in the experimental HTR Peach Bottom (USA) (2) andDragon (UK) (9, 10) test reactors.1 Contact Author: joseph.somers@ec.europa.eu1

ENC 2007 Conference, Brussels, Belgium, September 17-19 2007There are advantages to be gained by the use of inert matrix based fuel kernel to dilute the fissilecomponent (11, 12). In particular, the fabrication process can be simplified and the extra volumegenerated in the buffer layer (for a given buffer layer thickness) can accommodate more fission gas,and higher burn-ups can be achieved. Production of the kernels is the first of three major fabricationsteps for HTR fuels. This is followed by their coating in suitable layers to prevent fission gas release,and ultimately the forming of compacts from the coated particles. In this paper, recent developmentson kernel fabrication and the on going installation of a chemical vapour deposition furnace to producethe coating layers at the JRC-ITU are presented and discussed.2. KERNEL FABRICATIONToday, the method of choice for UO 2 (or UCO) kernel production is the external gelation route (13-15). In this process, the viscosity of a solution of the uranyl nitrate salt is increased by the addition ofpolymers such as methocel or polyvinylalcohol (PVA). This broth solution is dispersed intomonodisperse droplets, which pass through ammonia gas, before being collected in an ammoniasolution. The ammonia gas causes precipitation of the uranyl hydroxide near the droplet surface,which gives it a minimal mechanical strength to sustain its impact with the aqueous ammonia solution.Once in the aqueous solution, the ammonia then diffuses into the droplet to complete the precipitationand the droplet to particle conversion. Drying, calcination and sintering steps are required to obtain thefinal product.External gelation is actually best described as a gel supported precipitation (GSP). Its success dependsintricately on a number of parameters, such as the metal and polymer concentration in the broth, itstemperature, and also on the concentration of the ammonia solution. The search for the combination ofparameters for optimal kernel production is lengthy, but once found the fabrication route has shownitself to be routinely applicable.For Pu and minor actinide (MA) based kernels, however, such long lead times are not acceptable, andfurthermore, the highly active liquid wastes produced in the process are prohibitive in its deploymenton an industrial scale. For this reason, the ITU has developed the infiltration process for thepreparation of Pu and MA containing kernels (16, 17). Within this process, the inactive cerium oxideor yttria stabilised zirconia (YSZ) kernels are produced in an inactive facility in a normal chemistrylaboratory, again using the sol gel process. The calcined kernels are then introduced into the glovebox,where they are infiltrated with a Pu or minor actinide nitrate solution. Drying, calcination and sinteringgive the final product kernel. This method greatly reduces the number of process steps in thegloveboxes. Perhaps more importantly, there are no radioactive liquid wastes. The actinide content isdetermined by the porosity of the calcined kernels, the concentration of the actinide solution, andnumber of times the infiltration is performed. A sample of kernels produced in this way is shown inFigure 1.Fig 1. (Zr,Y,Pu)O 2 kernels (φ = 700 µm) derived from (Zr,Y)O 2 produced by the external gelation and(double) infiltration routes2

ENC 2007 Conference, Brussels, Belgium, September 17-19 2007The specifications for the size and sphericity of the kernels is rather severe, so that selection steps areneeded to make sure only the kernels of highest quality are coated (see Figure 2). In a fist step, nonconforming kernels, based on size and shape, are removed by use of sieves and vibrating tables in thecold laboratory, before their introduction into the gloveboxes. This step can greatly diminish theactinide containing production scraps, which have to be recycled. Despite this precautionary step, theactinide bearing kernels need a further sieving and sphericity selection step, to remove any kernelsdamaged in the infiltration and sintering process. The kernel production is completed by mass controland the loading of the container for their coating in the CVD furnace.Control equipment related to visual inspection, sphericity and diameter, and ceromography have beenpurchased.PRODUCTIONCONTROLIMF ProductsInfiltrationCalcinationSinteringSphericityDiameterSelectionMass ControlLoadingVisualInspectionSphericityDiameterControlCeramographyFig 2. Kernel production and control. Open boxes indicate glovebox facilities already available. Shadedboxes indicate installations to be built.3. COATED PARTICLE PRODUCTIONThe typical kernel coating layer dimensions are given in Figure 3. The layers will be produced bychemical vapour deposition (CVD) of precursor gases in a fluidised bed reactor. Precursors for thebuffer, pyrocarbon (PyC) and SiC layers are acetylene, propylene and methyltrichlorosilane (MTS),respectively. For later stages, an option for TiN or ZrC coating has been included in the design phase.A new installation for the production of coated particle fuel, shown schematically in Figure 4, is beingconstructed at JRC-ITU. The order for the coater unit has been placed (18). It includes a CVD furnaceand off gas filtration to remove soot and an off gas scrubbing unit for the removal of HCl, generated inthe deposition of the SiC layer by the thermal decomposition of MTS. The furnace can house graphitecrucibles with dimensions between 35 and 50 mm, which will permit batch production of kernelquantities ranging from 20 to 100 grams. The entire CVD system will be installed in 3 or 4gloveboxes, depending on the actual dimensions of the equipment, which are now in the design phase.3

ENC 2007 Conference, Brussels, Belgium, September 17-19 2007µmd oPyC 40d SiC 35d iPy C 40d buf 90D ker 500Fig 3. TRISO coated particleThe coated particle production step represents an important radiological interface in the entire process.All remaininng steps must be performed in gloveboxes, but the contamination of the following boxeswill be much lower than those for the kernel production. For this reason several installations also usedin the kernel production are duplicated to ensure that undesired surface contamination of the coatedparticles with Pu does not occur. In particular, an independent coated particle sphericity and sizeselection installation (sieves and vibrating table) will be deployed, as will visual inspection and masscontrol facilities. An exception to this philosophy is the sphericity and diameter control, for which thesame device for the kernels will be used, with the provison that the coated particle samples used in thisequipment will not return to the boxes in the later stage of the production process.The installation foresees the use of existing cermographic facilities to determine the layer thicknessfrom samples polished to expose the interior close to the coated particle midplane. Most importantly,an optical method has been purchased for the measurement of the anisotropy of the carbon layers,revealed by the ceramographic section. The crystallographic form of the SiC is determined usingexisting conventional X-ray diffraction facilities, and an installation will be prepared for thedetermination of the layer densities using a floatation method, based on the mixture of liquidsprinciple.4

ENC 2007 Conference, Brussels, Belgium, September 17-19 2007PRODUCTIONCONTROLCoater UnitCVDCoaterOff GasFiltrationOff GasScrubbingSphericityDiameterSelectionVisualInspectionLayer DensityMassControlSphericityDiameterCarbonAnisotropyCeramographySiC StructureFigure 4: Kernel coating and control steps. Open boxes indicate gloveboxes already available. Shadedboxes indicate facilities to be built. The same device for control of diameter and sphericity will be used forcoated particles and kernels4. COMPACT FORMATIONThe preparation of fuel compacts, based on a mixture of coated particles in a graphite matrix, is notforeseen in the first phase of this project, which concentrates on the fabrication and characterisation ofthe kernels and coated particles. In addition, a further facility to prepare the coated particles forirradiation in materials testing reactors is being constructed. In a second phase a compact former basedon hot or isostatic pressing will be installed, along with the necessary diagnostic equipment.5. BIOLOGICAL PROTECTION AND SAFETYThe use of Pu necessitates that all installations are housed in α – tight gloveboxes, according to theusual radioprotection rules. In fact, only the CVD system needs extra safety precautions. Theschematic laboratory layout is shown in Figure 5. The risks posed by H 2 , either as an off gas fromacetylene or propylene precursors, or as a process gas used in conjunction with MTS, will beminimised by its dilution in nitrogen directly after the scrubbing step. An additional risk minimisationis guaranteed by the operation of the gloveboxes under a controlled atmosphere with less than 10 ppmO 2 and H 2 O. As a secondary measure the CVD coater system will be operated remotely from aposition outside the laboratory. The process lends itself to extensive automation and remote control.Visual observation is to be provided by strategically placed surveillance cameras.5

ENC 2007 Conference, Brussels, Belgium, September 17-19 2007System Monitoring and ControlGasSupplyLaboratoryMonitorsAtmospherePurifierGas ManagementPanelCVD FurnaceOff gasHandlingOffgasGloveboxesLaboratoryFigure 5: CVD furnace configuration within the Pu handling laboratory6. CONCLUSIONSThe ITU has embarked on a new project to install a coated particle fabrication facility for theproduction of Pu based HTR fuels. Novel infiltration routes are being developed to produce kernelswith a composition and size capable of reaching a high burn up. For the production of the coatinglayers, a traditional fluidised bed chemical vapour deposition furnace is now being manufactured andshould be ready for installation in 2007. Active (Pu) commissioning should begin in early 2009.7. REFERENCES1. J Bergeron, P. Mitaut, in Proceedings of International conference on evaluation of emergingnuclear fuel cycle systems – Global 95m Versailles, France Sept. 1995, p13382. A. Baxter, C. Rodriguez, Prog. Nucl. Energy, 38(2001)813. B. Bonin, D. Greneche, F. Carre, F. Damian, J.Y. Doriath, Nuclear Technology 145(2004)2664. A. Baxter, C. Rodriguez, M. Richards, J. Kuzminski, Actinide and fission product partitioing andtransmutation, 6 th Information exchange meeting Madrid, Spain, 11-13 December 2000. EUR19783en, p5935. C. Rodriguez, A. Baxter, D. McEachern, M. Fikani, F. Venneri, Nucl. Eng. Des. 222(2003)2996. X.Raepsaet, F. Damian, R. Lenain, M. Lecomte, Proc. Int. Conf. Nucl. Technol-Bridging theMillenia (GLOBAL 99), Jackson Hole, Wyoming, August 29 September 3, 1999, AmericanNuclear Society (1999) (CD-ROM)7. N. Foucher, RGN Actualites (1999)478. P.M. LeBar, A.S. Shenoy, W.A. Simon, E.M. Campbell, Nuclear News, October 2003, p289. M. Gaube, H. Bairiot, British Nuclear Energy Society, Proc. of an international conference onnuclear fuel performance, 1973, ISBN 090194897710. U. Hansen, Dragon Project Report 899, 197411. J. Baier, H. Bairiot, J. Vangeel, R. van Sinay, EURATOM report EUR 5066 e, 197412. J. Somers, A. fernandez, Progress in Nuclear Energy, 48(2006)25913. P. Naffe, E. Zimmer, Nuclear Technology, 42(1979)16314. Bao Weimin; Wan Xuejun, Atomic Energy Science and Technology, 24(1990)115. H. Huschka, M. Kadner, US Patent 4,060,497, (1977)16. K. Richter, A. Fernandez, J. Somers, J. Nucl. Mater. 249(1997)12117. A. Fernandez, D. Haas, R.J.M. Konings, J. Somers, J. Am. Ceram. Soc., 85 694 (2002)18. MPA Industrie, www.mpa.fr6

TECHNOLOGICAL ASPECTS CONCERNING THEPRODUCTION PROCEDURES OF UO 2 -GD 2 O 3 NUCLEARFUELM. DURAZZONuclear Fuel Center, Nuclear and Energy Research Institute – IPENP.O.Box 11049, Pinheiros 05499, São Paulo, BrazilH. G. RIELLAChemical Engineering Department, Santa Catarina Federal UniversityFlorianópolis, BrazilABSTRACTThe direct incorporation of Gd 2 O 3 powder into UO 2 powder by dry mechanical blending isthe most attractive process for producing UO 2 -Gd 2 O 3 nuclear fuel. However, previousexperimental results by our group indicated that pore formation due to the Kirkendall effectdelays densification and, consequently, diminishes the final density of this type of nuclearfuel. Considering this mechanism as responsible for the poor sintering behavior of UO 2 -Gd 2 O 3 fuel prepared by the mechanical blending method, it was possible to propose,discuss and, in certain cases, preliminarily test feasible adjustments in fabricationprocedures that would minimize, or even totally compensate, the negative effects of poreformation due to the Kirkendall effect. This work presents these considerations.1. IntroductionThe demand for extended fuel cycles and higher target burnups is strong incentive for the use ofGd 2 O 3 as a burnable poison in modern nuclear power reactors. This fuel has been proposed forimplantation in Brazil according to the future requirements established for Angra II Nuclear PowerPlant.The Brazilian Nuclear Industries (INB) has recently implanted a fuel pellets fabrication unit thatadopts ammonium uranyl carbonate (AUC) technology for UO 2 fuel production. Due to the goodcharacteristics of UO 2 powder derived from AUC [1], the fabrication process of UO 2 -Gd 2 O 3 fueladopts the dry mechanical blending method for preparing the mixed powders. In this process, Gd 2 O 3powder is incorporated into UO 2 powder and homogenized without additional milling, prepressing andgranulating steps, which are required when UO 2 powder is derived from other methods for theconversion of UF 6 [2].Nevertheless, the incorporation of Gd 2 O 3 powder into AUC, deriving UO 2 powder by the mostattractive commercial method of dry mechanical blending leads to difficulties in the acquisition ofsintered UO 2 -Gd 2 O 3 pellets with the minimum required density [3,4], due to the deleterious effect ofGd 2 O 3 on traditional UO 2 sintering behavior. This poor sintering behavior was confirmedexperimentally in a previous work [5], as shown in the sintering curves of Figure 1. The initialsintering phase up to 1200ºC is identical for both UO 2 and UO 2 -Gd 2 O 3 fuels. However, above 1200ºCthe shrinkage of the UO 2 -Gd 2 O 3 pellets is delayed, the sintering rate decreases and densification isshifted to higher temperatures. As a consequence, the final sintered density is significantly lower thanthe traditional density obtained when sintering pure UO 2 fuel [3].As the AUC technology is already implanted at INB, the method of producing UO 2 -Gd 2 O 3 fuel pelletswill be the dry mechanical blending method. Thus, a research program was initiated at IPEN aimed atinvestigating the possible causes of the poor sintering behavior of UO 2 -Gd 2 O 3 fuel prepared by the drymechanical blending method.In the first part of this program, a sintering blockage mechanism based on the formation of lowdiffusivity Gd rich (U,Gd)O 2 phases was studied, which could act as a diffusion barrier during thesintering process. This hypothesis was not supported by the experimental results. The investigationprogram was continued and another hypothesis was investigated, which was based on the formation of

stable pores during the formation of the solid solution simultaneously with the sintering process. Thishypothesis was confirmed experimentally.Shrinkage ∆l/l 0(%)20181614121086420600UO 2Traditional FuelUO 2- 2 wt% Gd 2O 0,253UO 2- 5 wt% Gd 2O 3UO 2-10 wt% Gd 2O 3 0,200,150,100,050,00800 1000 1200 1400 1600 0Temperature ( o C)60 120Time (min)180Shrinkage Rate (% / min)0,30600 800 1000 1200 1400 1600 0Temperature ( o C)Fig. 1 - Effect of gadolinia content on the sintering behavior of UO 2 -Gd 2 O 3 fuel pellets.60120Time (min)180The mechanism that explains the sintering behavior of UO 2 -Gd 2 O 3 fuel prepared by the drymechanical blending method is based on the occurrence of the Kirkendall effect. A significantdifference in the interdiffusion coefficients of the gadolinium into UO 2 and of the uranium into Gd 2 O 3causes a misbalance in material transport during the formation of the solid solution. As a consequence,the densification during sintering occurs simultaneously with the formation of pores in locations whereGd 2 O 3 agglomerates were originally present. The diameters of these pores are proportional to theinitial diameter of these agglomerates. The pores formed are stable, since they are formed at hightemperature in an essentially closed pore structure. Given this situation, the elimination of these poresafter their formation is not possible in the subsequent sintering process. The pores remain in thesintered pellet and are the cause of the low densities observed.The technological solution adopted industrially for this problem was the incorporation of aluminuminto the UO 2 powder in the form of Al(OH) 3 . Aluminum is incorporated in the homogenization step inconcentrations varying from 5 to 500 ppm, which depends on the gadolinium concentration in the fuel[4].Given that the mechanism that explains the insufficient densification of the UO 2 -Gd 2 O 3 fuel during thesintering process is known, it is possible to propose actions that could minimize or eliminate theproblem. These are based on changing certain fabrication procedures, as follows:a) activity optimization of the UO 2 powder used in the preparation of UO 2 -Gd 2 O 3 mixed powder,which can be achieved by adjusting the conditions for AUC reduction;b) sintering cycle optimization by adjusting the heating rate, temperature and duration of theisothermal sintering step;c) adjustment in the homogenization procedure of the UO 2 and Gd 2 O 3 powders aimed at obtaining amixture presenting a high level of homogeneity.2. Activity optimization of the UO 2 powderAccording to the results presented in Figure 1, the formation of pores due to solid solution formationbegan to occur at temperatures above 1000°C, during the second stage of sintering, and ended around1350°C. Thus, when pore formation began, the densification process of the UO 2 matrix had alreadyinitiated and when pore formation was concluded, the densification process of the UO 2 matrix was atan advanced stage and the densification rate was already in decline. Under these conditions, the poresformed are difficult to eliminate, especially when the homogeneity level in Gd 2 O 3 distribution ismacroscopic, i.e., where large sized Gd 2 O 3 agglomerates are present.One possibility for acting on this process would be to delay the sintering process. In other words, toact on the sintering kinetics in order to determine that it occurs at higher temperatures, after theformation of the solid solution and, consequently, after pore formation. This would ensure that the

201816141210pores derived from the Kirkendall effect would be formed in a more open pore structure and theycould be more easily eliminated. The delay in the sintering process could be compensated in theisothermal stage of the sintering cycle. One way to achieve this objective is to reduce the activity ofthe UO 2 powder used to prepare the mixed powder.This possibility was tested using UO 2 powder derived from AUC with different areas of specificsurface when preparing the UO 2 -Gd 2 O 3 mixed powder. Controlling the activity of UO 2 powder waspossible through controlling the parameters for AUC reduction. The results are presented in Figure 2.Shrinkage ∆l/l 0(%)86420B.E.T UO ρ(m 2 2/g) (% DT)6,0 91,525,6 91,764,5 92,042,8 90,001,6 88,95600 800 1000 1200 14001600 0 60 120 180Temperature ( o C) Time (min)Fig. 2 - Effect of specific surface area (BET) on the sintering process of the UO 2 -Gd 2 O 3 fuel.Since UO 2 powders with high specific surface begin to sinter at lower temperatures, the formation ofthe solid solution and the consequent pore formation, due to the Kirkendall effect, occur in anessentially closed pore structure, during an advanced stage of the sintering process. For this reason, theelimination of the formed pores is difficult. In contrast, when the specific surface of the UO 2 powder istoo low, although the pore formation process occurs in a more open pore structure, insufficient activityis available in the system after pore formation to promote good densification, which results in verylow densities. Thus, it would seem that an optimal specific surface for UO 2 powder exists, which is notso high that it excessively closes the pore structure prior to solid solution formation and is not so lowthat it prejudices the sintering process of the system as a whole. The optimal specific surface would bethat which provides the UO 2 powder a reserve of activity in order to guarantee that sufficientsinterability exists for densification at high temperatures, after solid solution formation. This reserveof activity for sintering at high temperatures would promote the efficient elimination of the poresformed due to the Kirkendall effect and optimize densification. In this context, densification isreinforced at temperatures above the temperature at which pore formation occurs, approximately 1350°C, by means of a reserve of activity for sintering at elevated temperatures. In this work, the best resultwas obtained using UO 2 powder with a specific surface of 4.5 m 2 /g. Figure 2 shows that densificationduring the isothermal cycle of sintering using low specific surface UO 2 powder is considerably largerthan the densification observed in the case of UO 2 powder with high specific surfaces. This findingexplains the results obtained by Agueda et al [6], who obtained sensibly superior densities aftersintering using UO 2 powder with a smaller specific surface area.The use of higher sintering temperatures, normally used to sinter the fuel UO 2 -Gd 2 O 3 (1700 to 1750°C), certainly would lead to more positive results, since the improvement observed in the sinteringbehavior of UO 2 -Gd 2 O 3 fuel in function of the control of UO 2 specific surface, by itself, did notachieve the minimum density specified for this fuel. As shown in Figure 2, the densification thatoccurred in the isothermal part of the sintering cycle is considerable, so an increase in isothermalsintering time should lead to higher densities in the sintered pellets. This observation was confirmedby previous results [7], which demonstrated that the additional elimination of approximately 2 vol% inthe residual porosity was possible when the isothermal sintering time was increased from 3 to 6 hours.3. Sintering cycle optimizationShrinkage Rate (% / min)0,300,250,200,150,100,050,00600 800 1000 1200 1400 1600 0 60 120 180Temperature ( o C) Time (min)

Other possibility of acting on the sintering kinetics by delaying the sintering process in order todetermine that it occurs after pore formation by the Kirkendall effect, would be to increase the heatingrate of the sintering cycle. If the kinetics for solid solution formation, or for pore formation due to theKirkendall effect, occurs faster than sintering kinetics, the effect of increasing the heating rate wouldbe beneficial in terms of residual porosity, since a larger fraction of pores formed could be eliminatedin the sintering process subsequent to pore formation. A delay in the sintering process could becompensated in the isothermal stage of the sintering cycle. In order to test this possibility, UO 2 -Gd 2 O 3pellets containing 10 wt% Gd 2 O 3 , prepared according to the dry mechanical blending method, weresintered under H 2 atmosphere using different heating rates. The heating rates varied from 1°C/min upto 90°C/min. The sintering curves obtained are presented in Figure 3. The shrinkage rates are alsopresented in this figure.The positive effect of increasing the heating rate can be verified in the final sintered density, asillustrated in Figure 3. Elevation of the heating rate from 5°C/min to 30°C/min led to an increase in thefinal sintered density of almost 1.5 vol%. When the heating rate was very low (1°C/min), poreformation during the formation of the solid solution was very clearly observed and densification in theisothermal period was very low. In contrast, when the heating rate was superior to 10°C/min, thedecrease in the shrinkage rate due to pore formation was smaller and densification at the onset ofisothermal treatment was pronounced, resulting in a beneficial effect in eliminating porosity, which ledto higher densities in sintered fuel pellets.Shrinkage ( % )2220181614121086420-21 o C/min5 o C/min10 o C/min30 o C/min50 o C/min90 o C/minρ (%DT) s91,4391,7692,0992,3992,6993,08600 800 1000 1200 1400 1600Fig.3 - Effect of theTemperatureheating rate(on o C)the sintering process of the UO 2 -Gd 2 O 3 fuel.4. Adjustment in the homogenization procedureTime (min)0 50 100 150Previous experimental results [5] demonstrated that homogeneity in the Gd 2 O 3 distribution within themixed powder exerted a decisive influence on the sintering process of UO 2 -Gd 2 O 3 fuel. In otherwords, the better the homogeneity in the mixed UO 2 -Gd 2 O 3 powder, the higher the final sintereddensity. This behavior is explained based on the Kirkendall effect mechanism. The development of analternative method for UO 2 and Gd 2 O 3 powder homogenization, which permits good homogeneity at amicroscopic level, is another possible action that could improve the sintering behavior of UO 2 -Gd 2 O 3fuel.The homogenization method should preserve the original morphology of the particles of the AUCpowder, which gives good flowability to the UO 2 produced and the desirable direct pressingprocedure. The use of comilling, in spite of resulting in good sintered densities (microscopic level ofhomogeneity), is not technologically interesting, because this procedure destroys the originalmorphology of the UO 2 powder, which implies the incorporation of a granulation procedure.According to previous results [5], the coprecipitation process via AUC permitted the acquisition ofgood homogeneity, at a microscopic level, in the distribution of Gd 2 O 3 in the UO 2 -Gd 2 O 3 mixedpowder, which resulted in an adequate densification level during sintering. In fact, coprecipitationdoes not occur in the AUC case. Probably, simultaneous precipitation occurs and homogenization isachieved in the liquid phase. Ravindran et al [8] was also unable to acquire solid solution by

coprecipitation via AUC. Although this method for Gd 2 O 3 incorporation during the precipitation stageresulted in a homogeneity level that was adequate for obtaining sufficiently good sintered densities,this method has the disadvantage of contaminating the precipitation reactor, which would demandexclusive equipment. To avoid duplication of the precipitation installation, an alternative would be theincorporation of the Gd 2 O 3 powder into the AUC suspension prior to filtration. Thus, homogenizationwould be achieved in a liquid media, which is much more efficient at allowing the desegregation ofGd 2 O 3 powder and would permit its dispersion in individual particles among the AUC crystals. In thiscase, after AUC precipitation in the traditional reactor, the suspension would be pumped towards ahomogenizing tank and then towards a second filter, different from that used in the process forproducing the standard UO 2 fuel. In this case, the duplication of the filtration system and theinstallation of an additional tank for homogenization would only be required.This proposed solution was tested in the present work at laboratory scale. However, the results wereunsatisfactory, since a strong tendency for segregation of the Gd 2 O 3 powder in the AUC suspensionwas observed, as well as an accentuated tendency for agglomeration of the Gd 2 O 3 powder duringfiltration. Additional work is required, aimed at determining ways that guarantee good homogeneity inthe suspension and filtration processes, possibly by using some type of dispersant.The more common methods for Gd 2 O 3 synthesis are based on the thermal decomposition of carbonatesand hydroxides. However, using such methods, the minimum particle size obtained is limited by thesensitive tendency for agglomeration during thermal decomposition. Mazdiyani and Brown [9]developed a technique based on dynamic calcination that prevented aggregation during thermaldecomposition, which made the acquisition of fine particles of Gd 2 O 3 of approximately 28 nmpossible. Moreover, as new applications for Gd 2 O 3 have been investigated recently, mainly as anadditive and dopant, new methods for its preparation have also been investigated. Ultra fine Gd 2 O 3powder was obtained by mechanochemical synthesis, resulting in particles of 0.1 µm [10,11]. The useof ultrafine Gd 2 O 3 powders presenting low tendency to agglomerate could result in a goodhomogeneity level by adopting the technique of Gd 2 O 3 incorporation into AUC, or even using the drymechanical blending method. This action would probably minimize the sintering problems, due thedecrease in the diameter of the pores formed due to the Kirkendall effect, which would lead to goodsintered densities.5. ConclusionObservation revealed that a decrease in UO 2 powder activity and an increase in the heating rate of thesintering cycle exerted a positive influence on the densification process experimentally. This behavioris in agreement with the mechanism based on the Kirkendall effect. The results presented addresspossible adjustment actions in the fuel production procedures capable of minimizing the effects of thismechanism in order to optimize the final sintered density of UO 2 -Gd 2 O 3 fuel pellets fabricatedaccording to the dry mechanical blending method, as follows:a) adopting a heating rate in the sintering cycle as high as technologically applicable and definedbased on the resulting microstructure of the sintered pellet. A heating rate of approximately30°C/minute is recommended;b) adopting a UO 2 powder with activity adjusted in such a manner that the onset of densification isdelayed without compromising densification in the final stages of the sintering process. Thespecific surface can be controlled by adjusting the conditions for reduction of the AUC powder.The ideal specific surface was not precisely determined in this work, but it should be within therange of 4 to 5 m 2 /g;c) increasing the duration of the isothermal stage of the sintering process in order to maximizedensification in this stage. An increase in the sintering temperature up to 1750 ºC should also bevery beneficial.It is likely that a combination of the proposed adjustments in the activity of UO 2 powder and in thesintering parameters will result in good density UO 2 -Gd 2 O 3 fuel pellets. However, a complementarywork with that specific objective is still required.Finally, the development of alternative techniques for homogenizing the powders, together with theuse of Gd 2 O 3 powder with special characteristics (low tendency to agglomerate), would probablypermit the acquisition of UO 2 -Gd 2 O 3 fuel pellets presenting the minimum specified density. A specificwork in that area is yet to be conducted.

AcknowledgementsThe authors wish to express their gratitude to the CTMSP (Navy Technological Center in São Paulo)for the permission to use their facilities. The authors also wish to express their sincere thanks to thestaff of the Nuclear Materials Laboratory of the CTMSP for their assistance in the course of this study.References[1] H. Assmann and M. Becker, Trans. Am. Nucl. Soc. 31 (1979), p. 147.[2] H. Assmann and W. Dörr, Materials Science Monographs n. 16 (1983), p. 707.[3] R. Manzel and W. Dörr , Am. Ceram. Soc. Bull. 59, n. 6 (1980), p. 601.[4] H. Assmann, M. Peehs and H. Roepenack, J. Nucl. Mater. 153 (1988), p. 115.[5] M. Durazzo, H. G. Riella. Key Eng. Mat.189-191 (2001), p. 60.[6] H. C. Agueda, A. D. Heredia, D. C. Amaya, M. E. Sterba and D. Russo, In: Proc. 5 GeneralCongress on Nuclear Energy, v. 2, Rio de Janeiro, 1994. p. 567.[7] H. G. Riella, M. Durazzo, M. Hirata, R. A. Nogueira, J. Nucl. Mater. 178 (1991), p. 204.[8] P. V. Revindran, K. V. Rajagopalan and P. K. Mathur, J. Nucl. Mater. 257 (1998), p. 189.[9] K. S. Mazdiyasni,and L. M. Brown, J. Am. Ceram. Soc. 54, n 10 (1971), p. 479.[10] T. Tsuzuki, W. T. A. Harrison and P. G. McCormick, Journal of Alloys and Compounds, 281(1998), p. 146.[11] T. Suzuki, E. Pirault and P. G. McCormick, Nanostructured Materials, 11, n. 1, (1999), p. 125.

THE THORIUM MOLTEN SALT REACTOR: LAUNCHINGTHE THORIUM CYCLE WHILE CLOSING THE CURRENTFUEL CYCLEE. MERLE-LUCOTTE, D. HEUER, M. ALLIBERT, V. GHETTA,C. LE BRUN, R. BRISSOT, E. LIATARD, L. MATHIEULPSC, Université Joseph Fourier, IN2P3-CNRS, INPGLPSC, 53, avenue des Martyrs, F-38026 Grenoble Cedex - FranceABSTRACTMolten salt reactors, in the configuration presented here and called Thorium Molten Salt Reactor(TMSR), are particularly well suited to fulfil the criteria defined by the Generation IV forum, andmay be operated in simplified and safe conditions in the Th/ 233 U fuel cycle with fluoride salts. Thecharacteristics of TMSRs based on a fast neutron spectrum are detailed in this paper, focusing ontheir excellent level of deterministic safety. We aimed at designing a critical TMSR able to burnthe Plutonium and the Minor Actinides produced in the currently operating reactors, andconsequently to convert this Plutonium into 233 U. This leads to closing the current fuel cyclethanks to TMSRs started with transuranic elements on a Thorium base, i.e. started in the Th/Pufuel cycle. We study the transition between the reactors of second and third generations to theThorium cycle in a European frame.1. IntroductionThe Generation-IV International Forum for the development of new nuclear energy systems hasestablished a set of goals as research directions for nuclear systems: enhanced safety and reliabilityreduced waste generation, effective use of uranium or thorium ores, resistance to proliferation,improved economic competitiveness. The Molten Salt Reactor (MSR) is one of the candidates retainedby Generation IV. MSRs are based on a liquid fuel, so that their technology is fundamentally differentfrom the solid fuel technologies currently in use. Some of the advantages specific to MSRs (in termsof safety/reliability, for example) result directly from this characteristic. Furthermore, this type ofreactor is particularly well adapted to the Thorium fuel cycle (Th- 233 U) which has the advantage ofproducing less minor actinides than the Uranium-Plutonium fuel cycle ( 238 U- 239 Pu) [1].We have reassessed the MSR concept to propose an innovative reactor called Thorium Molten SaltReactor (TMSR). Many parametric studies of the TMSR have been carried out [2,3,4,5], correlatingthe core arrangement and composition, the reprocessing performances, and the salt composition. Thesestudies have shown that the reactor design in which there is no graphite moderator inside the coreappears to be the most promising. It uses a fast neutron spectrum, as described in section II of thispaper. The characteristics (initial fissile inventory, salt composition, safety parameters, deploymentcapabilities) of such a TMSR configuration are detailed with the reactor operating either as a Thoriumbasedbreeder or as an actinide burner.Finally, the full transition between the second and third generation reactors to the Thorium cycle isoptimized in the third section by considering the deployment capacities of the TMSR concept in aEuropean frame, in association with the currently operated light water reactors.This work made use of the MCNP neutron transport code [6] coupled with an in-house materialsevolution code REM [4]. The former evaluates the neutron flux and the reaction rates in all the cellswhile the latter solves the Bateman equations for the evolution of the materials composition within thecells. These calculations take into account the input parameters (power released, criticality level,chemistry ...), by continuously adjusting the neutron flux or the materials composition of the core. Ourcalculations rest on a precise description of the geometry and consider several hundreds of nuclei withtheir interactions and radioactive decay.

2. The Non-Moderated Thorium Molten Salt Concept2.1 Neutronic Core DescriptionThe TMSR concept is a 2500 MWth (1GWe) reactor operated in the Thorium fuel cycle, using either233 U or Pu as its initial fissile fuel. As shown in Fig. 1, the core is a simple cylinder (1.25m radius and2.60m height). The nuclear reactions occur within the up-flowing molten salt. The operatingtemperature is 630°C, corresponding to a thermodynamic efficiency of 40 %. At any moment, onethird of the 20 m 3 of fuel carrying salt is outside of the core, flowing through pipes, pumps, heatexchangers and the extraction systems aimed at removing the gaseous and insoluble fission products.The core structures are protected by reflectors which ensure that 80% of the neutron flux are absorbed.To avoid thermalization of the reflected neutrons, the axial reflectors are made of ZrC. The radialreflector consists in graphite channels containing a binary fluoride salt LiF-ThF 4 with 28%- mole232 Th. This reflector is a fertile blanket, increasing the breeding ratio via a biannual extraction of 233 U.Fig.1. Schematic vertical section of the TMSR, including pumps and heat exchangers (IHX)2.2 Salt ReprocessingHere, we consider only the reprocessing aspects relevant to the reactor operation. This reprocessingfirst involves a control and adjustment of the salt composition (redox potential measurement, reactivitykept equal to one…). The reprocessing itself is done in two steps. First, an on-line gaseous extractionsystem with helium bubbling removes all gaseous fission products (neutron poisons). Our simulationsassume that helium bubbling extracts the gaseous fission products and the noble metals within 30seconds. In fact, a less efficient extraction would not affect core behaviour much. Indeed up to anextraction time of a few days, the breeding ratio stays almost unchanged.Secondly, foror the extraction of the other fission products, mainly lanthanides, a fraction of salt isperiodically set aside to be reprocessed off-line (batch). The fissile matter (uranium) is quicklyextracted by fluorination and sent back to the core. The other actinides and lanthanides are assumed tobe separated via various methods such as electrolysis, reduction into metallic solvents, solidprecipitation, or any other method studied in the frame of pyrochemistry reprocessing. Finally, theactinides are sent back to the reactor core to be burnt, while the lanthanides are evacuated as waste.The calculations below have been done assuming that reprocessing rates remain within atechnologically realistic range taken as [50kg, 200 kg] of HN reprocessed per day.2.3 Fuel Salt CompositionThe core contains a fluoride fuel salt, composed of LiF enriched in 7 Li (99.999 %) and heavy nuclei(HN) amongst which the fissile element, 233 U or Pu. We have done parametric studies of this saltcomposition by varying the proportion of heavy nuclei in the fuel salt, and the reprocessing [7]. Thishas an influence on neutron energy moderation, actinide solubility, and the necessary initial fuelinventory. For HN proportions ranging from 20 to 30 mole%, a binary salt LiF-(HN)F 4 has beenchosen whose melting point is around 570°C. For lower proportions of HN, the calculations have beendone with a salt containing 80 mole% of LiF completed with BeF 2 to lower the eutectic pointtemperature and to allow operation at 630°C. Thanks to these parametric studies, we have selected forthis presentation a typical fuel salt containing 17.5 mole % of heavy nuclei, with a reprocessing of 200

kg of heavy nuclei per day. This composition corresponds to a fast neutron spectrum. The salt densityis equal to 3.8, with a dilatation coefficient of 10 -3 /°C [8].As already stated, TMSRs can be operated in the Thorium fuel cycle, using either 233 U ( 233 U-startedTMSR) or Pu (transuranic-started TMSR or TRU-started TMSR) as initial fissile matter. The 233 U-started TMSR with 17.5 mole % of heavy nuclei requires an initial fissile load of 4.6 tons of 233 Umixed with around 37 tons of Thorium.Concerning the TRU-started TMSR, we have considered as initial load a mixture of thorium (29 tons)and, for its fissile material, the transuranic elements (Pu, Np, Am and Cm) produced in the watermoderated reactors fed at present with natural or slightly enriched uranium. Actually, to be morerealistic, these TMSRs are started with the mix of 87.5% of Pu ( 238 Pu 2.7%, 239 Pu 45.9% , 240 Pu21.5%, 241 Pu 10.7%, and 242 Pu 6.7%), 6.3% of Np, 5.3% of Am and 0.9% of Cm, corresponding to thetransuranic elements of an UOX fuel after one use in a standard Light Water Reactor followed by fiveyears of storage [9]. For the TMSR configuration considered here (17.5 mole% of HN in the salt), anamount of 7.3 tons of fissile elements is needed initially, corresponding to 4.5 mole % of Plutonium.Finally, concerning safety, the feedback coefficients of the TMSR are negative for all HN proportions,thus ensuring in both cases a very good level of deterministic safety [10]. For the TMSRconfigurations considered here, the total feedback coefficient is equal to -7 pcm/ºC at equilibrium forboth the 233 U-started and TRU-started TMSRs. The sub-coefficient called density coefficient, whichcan be viewed as a void coefficient, is also largely negative, around -3 pcm//ºC.2.4 TMSR as Actinide BurnerWe have analyzed the evolution of a typical fuel salt composition during the operation of a TMSRstarted with 233 U fuel and with a TRU initial fuel. As shown in Fig. 2, the TRU inventory of a TMSRstarted with TRU elements becomes equivalent to that of 233 U started TMSRs after about forty yearsfor our fuel salt with 17.5% of heavy nuclei: more than 85% of the initial TRU inventories are burnedso that the assets of the Thorium fuel cycle are recovered for these TRU-started TMSRs within 40years of operation.Fig.2. Heavy nuclei inventory for the 233 U-started TMSR (solid lines) and for the TRU-started TMSR(dashed lines)2.5 Deployment Capacities of TMSRsConcerning 233 U-started TMSRs, the deployment capacities are based on the amount of 233 U producedcompared to the initial fissile ( 233 U) inventory necessary to start such a reactor. The amount of 233 Uproduced and extracted all along the lifespan of the selected TMSR configuration is presented in Fig. 3(purple line): for TMSRs directly started with 233 U, the 233 U extraction follows a linear growth, of 97kg per year in the case shown. This production is directly related to the breeding ratio of theconfiguration.

TRU-started TMSRs allow the extraction of significantly larger amounts of 233 U during the first 20years of operation (Fig. 3, red upper line), thanks to the burning of TRUs which saves a part of the233 U produced in the core. This production reaches 185 kg per year in the example shown, i.e. 85%more than in the corresponding 233 U-started TMSR. After the first 20 years, the 233 U extraction rate isequivalent to that of the 233 U-started TMSR.Fig.3: 233 U extracted during the operation of TRU-started (red line) and 233 U-started (purple line)TMSRsThe operating time necessary to produce one initial fissile ( 233 U) inventory is called the reactordoubling time. On Fig. 3, the reactor doubling time of the 233 U-started TMSR is equal to 48 years andcorresponds to the crossing of the 233 U production line (red line) with the initial 233 U inventory (greendashed line). This reactor doubling time is reduced to 28 years when using transuranic elements asinitial load. This production of significantly larger amounts of 233 U in the TRU-started TMSRs duringtheir first 20 years of operation is due to the burning of TRUs which saves a part of the 233 U producedin the core. Using transuranic elements in TMSRs not only closes the current fuel cycle, but alsoimproves the deployment capacities of Thorium based reactors.3. European Deployment ScenariosThe deployment scenarios described below rest on the following nuclear power progression: startingnear 1970, nuclear power production grows to 160 GWe.y (GigaWatt electric-year) in 2000. Wepostulate that nuclear power doubles between 2000 and 2050 to later increase slowly by 0.5% per yearuntil 2100. Extrapolating up to 2100 allows us to verify that the deployment scenarios are lasting.Sodium cooled Fastneutron Reactor (SFR)Output capacity1.0 GWeFirst operating date 2040Reactor lifespan50 yrsPu amount (per load) 6 tonsLoading periodicity 5 yrsNumber of loads 2Breeding (per reactor-yr): 100 kg of PuTable 1: Characteristics of the SFRs consideredWe have simulated the deployment of several nuclear reactor fleets based on different reactortechnologies, to satisfy the anticipated energy demand stated above:- The first scenario involves light water reactors (LWRs) [9] and fast neutron breeder reactorsoperated in the U-Pu fuel cycle, specifically Sodium cooled Fast Reactors (SFRs) (characteristicsgiven in Table 1).

- The second scenario involves light water reactors and the TMSRs presented above: TRU-startedTMSRs and 233 U-started TMSR. As we aim at closing the current fuel cycle while lauching the Thfuel cycle, TRU-started TMSRs are started preferentially as long as transuranic elements producedin the light water reactors are available.A standing question was whether a fleet of TMSRs can be deployed given the absence of naturallyavailable 233 U. In fact, the question could have been asked for any other GEN-4 reactor since 239 Pu isnot naturally available either. The difference is that 239 Pu (along other TRU) is present in LWR wastes.On the other hand, the results presented in Section II establish that a TMSR is not only able to run inthe Th-U cycle but can also start its operation using the TRU produced in LWR reactors.These scenarios also led to an estimation of the production of heavy nuclei induced by the deploymentof such a nuclear reactor fleet in Europe. We aim at evaluating the complexity of the management ofthese heavy nuclei stockpiles, as well as their radio-toxicity. Ultimately when this type of electricityproduction is replaced by a novel technology (fusion for instance) all the actinides inside reactors willbecome discardable waste. The possibility of eventually shutting down the reactor fleets started hasthus to be studied, the heavy nuclei management being the key issue of such an option.Fig. 5: Stockpiles of transuranic elements and 233 U for two deployment scenarios: combination ofLWRs and SFRs (solid lines), and combination of LWRs, TRU-started TMSRs and 233 U-startedTMSRs (dashed lines)As shown in Fig 5, the scenario based on TMSRs allows a significant reduction of the transuranicelement stockpiles. For example, the Plutonium inventory of the fleet composed of LWRs and TMSRsis reduced to 200 metric tons in 2150, instead of 6000 tons for the scenario based on SFRs. TheUranium inventory in the scenario with TMSRs, replacing the Pu inventory for SFRs, lies around 4000tons. The amounts of others minor actinides are significantly reduced too thanks to TRU-startedTMSRs, which really help close the current fuel cycle.Finally, TRU-started TMSRs burn most of the TRU produced in the LWRs in the course of atransition towards a Th-U cycle operation, leading to an equilibrium salt composition whose radiotoxiccontent is significantly lower than that of the fuel in a SFR. Moreover, the uranium stockpileresulting from the fleet of TMSRs could be successfully burnt in identical TMSRs on an inert support,with a burning efficiency higher than 90% and a high deterministic safety level thanks to feedbackcoefficients ranging from -10 pcm/K to -5 pcm/K. As a result, waste management is simpler and easierto implement. Nuclear power deployment in this case is sustainable and efficient, the use of fissilematter and the production of wastes are optimized.4. ConclusionThe Thorium Molten Salt reactor (TMSR) presented here with no moderator in the core appears as avery promising, simple and suitable concept of molten salt reactor. The non-moderated TMSR

configurations considered in this paper, based on a fast neutron spectrum, present particularlyinteresting characteristics. Their deterministic safety level is excellent. They can be started with a fuelmade from the TRU wastes produced in current LWRs. Their rather large initial fissile inventory doesnot prevent fast deployment thanks to their good 233 U breeding. The technology which in principledoes not involve the transportation of radioactive materials outside the reactor site as well as thepresence of 232 U within the fuel can be considered as restricting proliferation risks.The concept itself has some appealing aspects compared to earlier versions of MSRs. The reactor coreis extremely simple. Simulation calculations do not point to major reprocessing constraints. Inparticular the fluxes considered should allow the batch mode reprocessing to be installed in thevicinity of the reactor. Initial studies of the scientific feasibility of the on-line control of the saltcomposition and of its chemical and physical properties have not unearthed a showstopper.When it comes to Generation-4, it appears that the major nuclear energy powers have given a higherpriority to the SFR concept. This mostly reflects a justified confidence in a technology which,although it has not yet reached all the performances expected for a GEN-4 reactor, has already beensuccessfully tested in numerous projects. But all the properties detailed in this paper, especially itsdeterministic safety performances and its ability to reduce the radio-toxicity of wastes currentlyproduced, put the TMSR in a very favourable position to fulfil the conditions defined by the GEN IVInternational Forum. Moreover this TMSR concept may be very appealing to countries which holdimportant thorium resources and have some remaining adjustment margins in the definition of theirnuclear energy policy. The TMSR is thus an excellent candidate to produce the large amounts ofnuclear energy that the world will need in the near future.5. AcknowledgmentsWe are very thankful to Elisabeth Huffer and Arnaud Lucotte for their help during the translation ofthis paper.6. References[1] A.Nuttin et al, “Potential of Thorium Molten Salt Reactors”, Prog. in Nucl. En., vol 46, p.77-99(2005)[2] L.Mathieu et al, “Proposition for a Very Simple Thorium Molten Salt Reactor”, GlobalConference, Tsukuba, Japan (2005).[3] L. Mathieu, D. Heuer et al, “The Thorium Molten Salt Reactor: Moving on from the MSBR”,Prog. in Nucl. En., vol 48, pp. 664-679 (2006)[4] L. Mathieu, “Cycle Thorium et Réacteurs à Sel Fondu : Exploration du champ des Paramètres etdes Contraintes définissant le Thorium Molten Salt Reactor”, PhD thesis, Institut NationalPolytechnique de Grenoble, France (2005) (in French)[5] E. Merle-Lucotte, D. Heuer C. Le Brun, L. Mathieu, et al, “Fast Thorium Molten Salt Reactorsstarted with Plutonium”, Proceedings of the International Congress on Advances in Nuclear PowerPlants (ICAPP), Reno, USA (2006)[6] J.F. Briesmeister, “MCNP4B-A General Monte Carlo N Particle Transport Code”, Los AlamosLab. report LA-12625-M (1997)[7] E. Merle-Lucotte, D. Heuer, M. Allibert, V. Ghetta, C. Le Brun, L. Mathieu, R. Brissot, E. Liatard,“Optimized Transition from the Reactors of Second and Third Generations to the Thorium Molten SaltReactor”, Proceedings of the International Congress on Advances in Nuclear Power Plants (ICAPP),Paper 7186, Nice, France (2007)[8] V. Ignatiev, E. Walle et al, “Density of Molten Salt Reactor Fuel Salts”, Nureth Conference,Avignon, France (2005)[9] C. de Saint Jean, M. Delpech, J. Tommasi, G. Youinou, P. Bourdot, “Scénarios CNE : réacteursclassiques, caractérisation à l'équilibre”, CEA report DER/SPRC/LEDC/99-448 (2000) (in French)[10] E. Merle-Lucotte, D. Heuer, L. Mathieu, C. Le Brun, “Molten Salt Reactor: Deterministic SafetyEvaluation”, Proceedings of the European Nuclear Conference, Versailles, France (2005)

Session 18.2.1:Fuel

MONITORING OF NUCLEAR FUEL ASSEMBLIES INCONDITION OF SLOVAK WET INTERIM SPENT FUELSTORAGE FACILITYV. SLUGEŇ 1 , M. MIKLOŠ 11 Department of Nuclear Physics and Technology,Slovak University of Technology Bratislava, Ilkovičova 3, 812 19 Bratislava, SlovakiaM BOŽÍK 2 , D. VAŠINA 22 JAVYS, a.s., 91931 Jaslovske Bohunice, SlovakiaABSTRACTAn advanced spent nuclear fuel management in Slovakia is presented in the paper. TheSlovak wet interim spent fuel storage facility in NPP Jaslovské Bohunice was build andstarted his operation in 1986. Since 1999, leak tests of WWER-440 fuel assemblies areprovided by special leak tightness detection system “Sipping in Pool” delivered byFramatome-anp with external heating for the precise defects determination. Although, thissystem seems to be very effective, the detection time of all fuel assemblies in one storagepool is too long. Therefore, a new “on-line” detection system, based on new sorbentNIFSIL for effective 134 Cs and 137 Cs activity was developed. Design of this detectionsystem and its application possibility in Slovak wet interim spent fuel storage facility aswell as the actual results and plans for future are presented.1. Spent fuel management in SlovakiaThe Slovak spent fuel management and safe interim storage of this fuel are among the most significantactivities provided by the new created governmental company JAVYS.The spent nuclear fuel is a product of the electric power generation in nuclear power plants and the Actof the National Council of the Slovak Republic No. 130/1998 Coll., On the peaceful use of nuclear energy,defines in its §18 the terms "spent nuclear fuel" and "spent fuel management" as follows: Spent nuclear fuel mean irradiated nuclear fuel withdrawn from a nuclear reactor. Management of spent nuclear fuel means storage, reprocessing, handling, transportation and disposalof spent nuclear fuel in a spent nuclear fuel repository. The legal person or natural person who produced the spent nuclear fuel is responsible formanagement of this fuel until its transfer to a spent nuclear fuel repository. Details of spent nuclear fuel management, especially its storage and disposal, are specified by agenerally binding legal regulation issued by the Slovak Nuclear Regulatory Authority (UJD SR).The Interim Spent Fuel Storage Facility (ISFSF) in Jaslovské Bohunice is an important component of thespent nuclear fuel management system. The facility has been used for storage purposes since 1987. ISFSF is anuclear facility providing for a safe and interim storage of the spent nuclear fuel from VVER reactors forthe time period of 50 years before the fuel is further processed in a reprocessing plant or definitively disposed of.A basic data on the ISFSF are given in the following table 1:

Table 1 – Basic technical parameters of ISFS in SlovakiaCapacitySufficient for storage of all the spent fuelproduced from expected operations of all units inJaslovské BohuniceNumber of storage pools3 operational and 1 reserveGround plan47m x 70mMax. number of the FA in a basket KZ-48 48FAT-12, T-13 30FAMass of the heavy metal stored per annum About 50 tonsDimensions of the storage pool 23,4 x 8,4 x 7,2mCooling medium in the pondsDemineralised waterMaximal thermal output of the storage fuel 1990 kWMaximal water temperature50 o C2. The Monitoring of fuel cladding leak tightness at Slovak wet interim spent fuel storagefacilityIt is necessary to keep the concentration of fission products in storage pools on the low level forassurance of acceptable activity of the coolant. This can be done with periodical monitoring of the fuelelements condition, leakages identification and closing of leaking elements in special hermetic caskets.That was the main reason for including not only “Sipping in pool” system, but also the just presentedmonitoring system based on NIFSIL sorbent to the fuel control system at the Slovak burnt fuel interimstorage in Jaslovské Bohunice [1].System “Sipping in pool” was designed by Framatome-anp for monitoring of the spent fuel fromWWER-440 reactors. It was build and implemented to the storage facility operation in 1999 andpractice shown great results in measuring. This system contain from 4 main parts: 2 electrical heatedcylindrical caskets, mechanical clamp,electrical and sampling lines and controlling and bleeding panel.The fuel assembly is inserted into the special casket, which is electrical heated by three electricalheaters (5kW). After closing of the upper end of the casket, the temperature increases. Also thetemperature of cladding and pressure inside fuel elements will increase inside. If there are anyleakages, they start to open themselves due to a higher temperature and pressure and the fissionproduct will release from the element. Then, the water and gas sample from the casket is taken andevaluated. The sample is analyzed by gamma spectroscopy (the most important nuclides are 134 Cs and137 Cs). If this test proves that there are any leakages on the fuel element cladding, the defected fuelassembly is closed into a special hermetic casket (T-13).3. acceleration of defects identification of wwer-440 fuel at the wet interim spent fuel storagefacility using “On-line” detection systemFor the acceleration of defects identification a quicker detection system was designed. This “on-line”detection system is based on group of “detectors”, which are symmetrical placed in spent fuel storagepools and used like NIFSIL holders. NIFSIL is a new type of composite sorbent based on potassiumnickel ferrocyanide incorporated in silica gel matrix and it is indissoluble in water [2]. It is made inspherical shape with diameter 0,5 - 1 mm.The sorption capacity for cesium corresponds to the total amount of ferrocyanide in the sorbent. Thedependence of K d (distribution coefficient) at the initial cesium concentration 5.10 -4 M on the amountof ferrocyanide in composite sorbent is almost linear. An increase of ferrocyanide in the silica gelmatrix causes a higher cesium uptake and the dependence is linear. This can by expressed by theΓ=0.094+0.0061.X equation, where is the sorbed amount of cesium in mmol.g -1 and X is thepercentage concentration of ferrocyanide in the silica gel matrix. From Fig.1 is clear, that cesium ions

have good access to potassium nickel hexacyanoferrates and a silica gel matrix doesn’t or doesnegligibly decrease the sorption of cesium.Fig.1 - The distribution coefficient K d of cesium vs. total amount of ferrocyanide in sorbentAv [kBq/l]Volume activity of all nuclides absorbed by10000NIFSILCO-58CO-601000NB-95I-131100CS-134CS-1371010,11 2 3 4 5MeasurementFig.2 – Results from the volume activity measurements.The experiments from 2005-2006 (Fig. 2) showed that NISFIL is great sorbent for cesium ( 134 Cs,137 Cs). Three types of samples were taken:OV57 – water sample from primary circuit of NPP V-1 in Jaslovské Bohunice,OV57R – diluted water sample from primary circuit (NPP-SE EBO) in rate 1:5,OV77 – water sample from wet interim spent fuel storage facility (pool 116/1).Five measurements were made:1 – water samples (OV57, OV57R, OV77) without NIFSIL. After these reference measurementNISFIL was filled into water samples,2 – water samples with NIFSIL, measured after 3 days,3 – water samples with NIFSIL, measured after next 7 days,4 – only NIFSIL, measured after next 12 days,5 – only water sample, measured after next 24 days.Volume activity A v [kBq/l] of all types of nuclides was spotted, but only 134 Cs and 137 Cs have had thehighest values.

The highest values are in fourth measurement. It means that NIFSIL absorbs not only cesium isotopes,but also the other nuclides.The NIFSIL holders have cylindrical shape, so it can be placed right on the head of storage containersin the pools. Both used storage containers T-12 (the old one) and KZ-48 (the new one) have the samehead, so the “cesium detector” is compatible.If the activity of nuclides (mostly 134 Cs and 137 Cs), measured on-line in the pools, exceeds the nominalrate, the pools are closed and wet sample from each pool separately is taken. Evaluation using precisegamma spectroscopy will show us, which pool has the highest activity (in that pool is also the highestprobability of leaking fuel element). Then Cs detectors are inserted to the pool and gripped on everystorage container. After few days are detectors taken out and send to the radiochemical laboratory forevaluation by gamma spectroscopy. Evaluation will show us the place (eventually the storagecontainer), where the highest activity of cesium occurred. So we can high probably find the container,which contains damage fuel element (assembly). This storage container is sending for the sippingcontrol by system “Sipping in Pool”, where the damage assembly will be found. Then is the damagefuel assembly hermetic stored in the special storage container T-13.4. ConclusionFrom the nuclear safety point of view, it is necessary to keep the fission products inside the fuelelements and to prevent their escape into environment not only during reactor operation or fueltransport, but also during the long term storage of burnt nuclear fuel. Therefore, the effective leaktightness monitoring system at all fuel interim storages is necessary. The designed system from 80-iesat the Slovak wet interim storage facility didn’t assure this task at the desired level, so the system“Sipping in Pool” was implemented in 1999. After several years of its operation, performedmeasurements showed, that this system is high effective equipment for fuel cladding defects detection.However, the time for defects finding could be too long. “On-line” detection system for determine thewater activity, which was proposed, provides us with possibility to accelerate identification of mostprobably place of damaged fuel elements. Measurements with NIFSIL showed that it is great absorberfor cesium and other important nuclides. Therefore, it was chosen as the main component for thisdetection system. Optimal use of “On-line” detection system needs to construct it and to test it inpractice.AcknowledgementFinancial contributions from the grant VEGA 1/3188/06 and EC-6FP COVERS are acknowledged.5. References[1] Limits and conditions for wet interim spent fuel storage facility operation, A-02/MSVP. (inSlovak),[2] P. Rajec, J. Orechovská, I. Novák: Nifsil a new composite sorbent for cesium, January 1999,Journal of Radio-analytical and Nuclear Chemistry , Vol.245, No.2 (2000) 317-321,[3] V. Slugen, M. Kalousek, M. Rajčok: Thermophysical verification of continuous hermetic sealingof fuel elements by coating control in the VVER-440 nuclear reactor. Journal of radioanalyticaland Nuclear Chemistry - Letters 164, 1992, p.71-80.

ADVANCED CANDU FUEL BUNDLES SOURCE TERMASSESSMENT FOR SOME ACCIDENTAL CRITICALITYSCENARIOSC.A. MARGEANU, A.C. RIZOIU, G. OLTEANUReactor Physics & Nuclear Safety Dept, Institute for Nuclear Research PitestiNo.1 Campului Str., Mioveni, 115400 - RomaniaABSTRACTThe paper aims to present the source term estimation for advanced CANDU fuel bundles insome hypothetical accidental criticality scenarios. For this purpose, advanced CANDUfresh and spent fuel bundles with 43 fuel rods have been considered. The scenarios takeinto account for a very short irradiated or spent fuel bundles immersion in the NPP spentfuel bay light water, for some configurations closed to criticality. As reference, the resultsfor CANDU standard fuel bundle with 37 fuel rods have been used. In order to estimatefuel activity, thermal power and gamma energy and to obtain the photon source profile, thecalculations have been performed by using the ORIGEN-S code, included in ORNL’sSCALE5 program package. The paper includes irradiated fuel characteristic parameterscomparison for the considered CANDU fuel bundles. For the same amount of electricenergy generated, CANDU SEU fuel cycle produces a smaller, but more radioactive weightof spent fuel than the natural UO2 one. The spent fuel characteristic parameters sustain thisaffirmation. In order to allow the spent fuel decay and reactivity decreasing, at least for 6months before sending the spent fuel bundles to final disposal or fuel reprocessingfacilities, an intermediate storage inside the NPP spent fuel bay is mandatory.1. IntroductionIn the Romanian Government policy, the nuclear energetic is considered as optimal solution for thenational energetic needs. In 2002, the Government decided to approve the Nuclear National Strategyfor the next 50 years [1]. The Fundamental Objective specifies that in 2025-2050 period RomanianNPPs must provide (20-40) % from the total electricity generated in Romania, with respecting both ofthe competitive costs conditions and the nuclear safety assurance at international agreed standards.About 4 decades ago, Romania made the option for a heavy water nuclear power plant, CANDU6type, located at Cernavoda, the decision being based on the fact that the needed nuclear fuel and heavywater could be home-bred. The Cernavoda NPP is equipped with 5 reactors PHWR CANDU 6 type,CANDU 6 standard series - 706 MW(e) each. Unit 1 is in commercial operation since December,1996, Unit 2 reactor has reached first criticality on May 7, 2007 and Unit 3 is under construction, therest of two units being under preservation stage. In about a decade of commercial operation,Cernavoda NPP Unit1 has given ~10% from the total electricity produced in Romania, this percentassuming to be around 18% after 2007 (with both Unit1 and Unit2 in operation) [2]. At the end of2006, Unit1 has registered an average capacity factor of 91.37%, being the forth world wide CANDU6 type NPP. Over the year 2006, Cernavoda NPP-Unit 1 generated an amount of electricity of 5,631GWh, out of which 5,118 GWh were supplied to National Electric System [3].In the last 15 years, both for operating reactors and future reactor projects, a general trend to increasedischarge fuel burnup (energy released by the bundle) has been world wide registered. The fuelburnup raise associated consequences are very important: spent fuel mass reduction for 1 MWhgenerated electric power; actinides mass significant reduction in the spent fuel; more rarely refuelling,leading to impressing raises in installed capacity utilization; about 15%-35% reduction in costsassociated with nuclear fuel, for 1 MWh generated electric power. For CANDU type reactors, one ofthe most attractive solutions seems to be SEU (Slight Enriched Uranium) fuels utilization. Byenriching natural uranium from 0.7% to (0.9 - 1.2) % in U235, fuel costs are lowered because lessuranium and fewer bundles are needed to fuel the reactor. This in turn reduces the quantity of used

fuel and its subsequent waste management costs. AECL (Atomic Energy of Canada Limited), alongwith the KAERI (Korea Atomic Energy Research Institute), has developed CANFLEX (CANDUFLExible fuelling), an advanced fuel bundle design, to increase fuel performance and cost efficiencythrough improved heat transfer characteristics, and to maximize advanced fuel cycle options inCANDU reactors [4]. It will deliver many benefits to current and future CANDU reactors, includinggreater operating and safety margins, extended plant life, better economics and increased power.Romanian specialists have been analyzed many advanced fuel cycles, the estimations giving the bestchance for SEU and RU fuel cycles application. In INR Pitesti there is an active preoccupation, withpromising evaluations, for the development of a fuel bundle CANDU SEU type, similar to CanadianCANFLEX fuel bundle.The paper aims to present the source term and other characteristic parameters estimation for advancedCANDU fuel bundles in some hypothetical accidental criticality scenarios. For this purpose, two typesof fuel bundles have been considered, as follows: CANDU standard fuel bundle [5], as reference, andthe Romanian CANDU SEU43 fuel bundle [6], respectively. The source term assessment and fuelcharacteristic parameters estimation were done by means of ORIGEN-S code, included in ORNL’sSCALE5 programs package [7]. A comparison between CANDU SEU study cases was obtained, theresults for CANDU standard fuel bundle being used as reference.2. Shielding analyses2.1 Shielding problem - General descriptionFrom the nuclear safety analysis point of view is very important to know the fresh and spent fuelcharacteristic parameters: radionuclides inventory – masses and activities, decay heats, neutronic andphotonic generation rates, during both the residence period in the reactor core and the cooling periodinside the NPP spent fuel bay (concrete walls, stainless-steel reinforced, as shielding material andcooling agent, light water is used).The following hypothetical accidental criticality scenarios were considered: a) fuel bundle extractionfrom the reactor and immersion in the NPP spent fuel bay, following a fuel bundle damageidentification at K eff maximum value – ″d″ (damaged) upper index for identification; b) spent fuelbundles extraction from the reactor when K eff reaches the critical value (=1.0) and immersion in theNPP spent fuel bay – ″s″ (spent) upper index for identification. For both cases, a cooling period up to10 years in the NPP damaged or spent fuel bay is considered. Fig 1 presents the K eff evolution with theirradiation period (WIMS calculation) for the considered fuel bundle projects.1.31.2NATSEU0.9SEU1.1SEU1.31.1K-eff10.90.810 50 90 130 170 210 250 290 330 370Irradiation (days)Fig 1. K-eff evolution with irradiation timeTwo CANDU fuel bundle projects have been analyzed, namely: CANDU standard fuel bundle with 37fuel rods filled with natural uranium dioxide pellets (NU), and CANDU SEU43 fuel bundle with 43fuel rods filled with SEU pellets (SEU). For SEU, the following enrichments in U235 wereconsidered: 0.9 wt% (SEUa), 1.1 wt% (SEUb) and 1.3 wt% (SEUc), respectively.The table below permits an easy identification for each study case. Along with the case notation, thecharacteristic data about the enrichment in U235, burnup and residence time inside the reactor core,respectively, are presented.Type of fuelStudyCaseEnrichment[wt% U235]Burnup[MWd/tU]Residence time[days]

Short-time irradiated fuelSpent fuelNU (d) 0.72 797.27 23SEUa (d) 0.9 648.60 15SEUb (d) 1.1 648.60 15SEUc (d) 1.3 648.60 15NU (s) 0.72 5,362 154.688SEUa (s) 0.9 8,051 186.198SEUb (s) 1.1 10,507 242.997SEUc (s) 1.3 12,756 295.0Tab 1: Study cases identificationThe geometrical arrangement of the bundles (see Fig 2) consists in 3 concentrical rings (for NU – 6elements on the inner ring, 12 elements on the intermediate ring and 18 elements on the outer ring,respectively; for SEU – 7 elements on the inner ring, 14 elements on the intermediate ring and 21elements on the outer ring, respectively) and one central element. The CANDU SEU43 fuel bundledistinctive characteristic is that the 8 inner rods (the central element and the other 7 rods from theinner ring) have a larger diameter than the other 35 outer rods (the 14 rods from the intermediate ringand the 21 rod from the outer ring). For CANDU standard fuel bundle all the 37 fuel rods have thesame diameter, no matter if they are central or on inner, intermediate or outer ring.a) b)Fig 2. Geometrical arrangement for the considered CANDU fuel bundles:a) CANDU standard; b) CANDU SEU43Both for short-time irradiated and spent fuel bundles, the considered cooling period in the NPP spentfuel bay is up to 10 years, the residence period being specified in Tab 1. Fuel characteristics andisotopic composition were considered according to [5, 6].Radionuclide inventory and irradiated fuel characteristics have been obtained by taking into accountfor all relevant isotopes generation and depletion during both the irradiation and cooling phases of thefuel history. The irradiated fuel behaviour is characterized by the values of concentration,radioactivity, thermal and γ power, both for nuclides and elements, in light materials, actinides andfission products.2.2 Short-time irradiated fuel analysisThe considered scenario takes into account for a fuel bundle damaging identification after k eff reachesthe maximum value, followed by fuel bundle extraction and storage in the NPP defected fuel bay.In Tab 2 the evolutions of the short irradiated fuel total radioactivity, total thermal power and totalgamma energy during the specified cooling period are presented.Tab 3 contains the corresponding relative differences for SEU - SEU and SEU - NU comparisons.Parameter Study CaseCooling time [years]0.5 1 3 5 10TotalNU (d) 7,924.0 2,506.0 550.7 280.6 169.9Radioactivity SEUa (d) 6,629.0 2,050.0 443.4 227.0 138.4[Ci] SEUb (d) 6,655.0 2,053.0 443.5 227.7 139.3

Total ThermalPower[W]Total GammaEnergy[W]SEUc (d) 6,672.0 2,056.0 443.7 228.3 140.0NU (d) 31.15 9.63 1.71 0.75 0.47SEUa (d) 26.05 7.86 1.36 0.60 0.38SEUb (d) 26.15 7.86 1.35 0.60 0.38SEUc (d) 26.21 7.86 1.35 0.60 0.38NU (d) 15.04 2.59 2.27 0.17 0.13SEUa (d) 12.66 2.16 1.76 0.13 0.11SEUb (d) 12.71 2.17 1.76 0.13 0.11SEUc (d) 12.75 2.17 1.76 0.13 0.11Tab 2: Short-irradiated fuel characteristic parameters evolution during cooling periodRelative differences [%]ComparisonTotalTotalTotalRadioactivity Thermal Power Gamma EnergyNU - SEU43a 18.66 19.21 19.73NU - SEU43b 18.42 19.30 19.88NU - SEU43c 18.24 19.36 19.98SEU43b – SEU43a 0.30 - 0.11 - 0.20SEU43c – SEU43a 0.51 - 0.19 - 0.33SEU43c – SEU43b 0.22 - 0.08 - 0.13Tab 3: Short-irradiated fuel characteristic parameters relative differences2.3 Spent fuel analysisThe residence periods for spent fuel cases were thus selected to justify the accidental criticalityscenario. Fig 3 illustrates the photon source profiles during cooling period for NU and SEUc cases.Fig 3. Photon source profile during cooling periodIn Fig 4 the evolution of spent fuel total radioactivity, total thermal power and total gamma energyduring the specified cooling period, for the considered CANDU fuel projects is illustrated.UNAT SEU43a SEU43b SEU43cUNAT SEU43a SEU43b SEU43cUNAT SEU43a SEU43b SEU43c1.E+051.E+031.E+02Total Radioactivity [Ci]1.E+04Total Thermal Power [W]1.E+021.E+01Total Gamma Energy [W]1.E+011.E+001.E+030 2 4 6 8 10Cooling time [years]1.E+000 2 4 6 8 10Cooling time [years]1.E-010 2 4 6 8 10Cooling time [years]Fig 4. Spent fuel characteristic parameters evolution during cooling periodThe corresponding relative differences obtained for SEU - SEU and SEU - NU comparisons arepresented in Tab 4. The considered comparisons are completed by the relative differences obtained forsome long life nuclides radioactivity (see Tab 5).

Relative differences [%]ComparisonTotalTotalTotalRadioactivity Thermal Power Gamma EnergySEU43a - NU 31.23 31.68 32.99SEU43b - NU 45.13 45.64 47.25SEU43c - NU 52.89 53.70 55.42SEU43b – SEU43a 20.29 20.53 21.62SEU43c – SEU43a 32.02 32.55 33.99SEU43c – SEU43b 14.87 15.15 16.14Tab 4: Spent fuel characteristic parameters relative differencesComparisonRelative differences [%]Kr85 Sr90 Cs137 Pu239 Pu240 Pu241SEU43a - NU 32.43 32.48 32.60 3.42 27.81 39.43SEU43b - NU 48.42 48.66 48.32 4.29 39.48 53.44SEU43c - NU 57.78 58.14 57.40 4.47 45.59 60.20SEU43b – SEU43a 23.67 23.95 23.33 0.90 16.17 23.13SEU43c – SEU43a 37.52 38.00 36.80 1.09 24.62 34.29SEU43c – SEU43b 18.14 18.47 17.57 0.19 10.09 14.51Tab 5: Spent fuel characteristic parameters relative differences3. ConclusionsBy increasing the CANDU fuel enrichment in U235 a rise in the discharge fuel burnup is registered,associated with the actinide mass reduction in the spent fuel. Meanwhile, for the same amount ofelectric energy generated, CANDU SEU fuel cycle produces a smaller, but more radioactive weight ofspent fuel than the natural UO2 one. The spent fuel characteristic parameters sustain this affirmation.For the short-time irradiation conditions, CANDU SEU fuel characteristic parameters are less than thenatural UO2 ones, the relative differences being, as follows: 18% in radioactivity, 19% in thermalpower and about 20% in gamma energy. For the fuel irradiation until k eff reaches the critical value,CANDU SEU fuel characteristic parameters are greater than the natural UO2 ones, the followingrelative differences being obtained: (31.23-52.89)% in radioactivity, (31.68-53.70)% in thermal powerand (32.99-55.42)% in gamma energy, respectively.In order to allow the spent fuel decay and reactivity decreasing, at least for 6 months before sendingthe spent fuel bundles to final disposal or fuel reprocessing facilities, an intermediate storage inside theNPP spent fuel bay is mandatory. The irradiated fuel bundles discharged from the reactor core arestored on racks, at adequate distances to avoid critical mass formation, in the NPP spent fuel bay, lightwater being used as shielding material and cooling agent.4. References[1] "Decision of the Romanian Government to approve the National Strategy for the Development ofNuclear Domain in Romania", HG No.1259, Bucharest, (2002)[2] Cernavoda NPP Direction, CNE Prod official site, www.cne.ro[3] "Annual performance indicators", Semnal „N”, nr.6, Nov.-Dec. (2006)[4] "CANFLEX advanced fuel", AECL official site, www.aecl.ca[5] Gh. Olteanu et al., "Actualization of SEU-43 fuel bundle project for CANDU type reactors", INRPitesti, RI-5984 (2001)[6] Cernavoda U1 Nuclear Generating Station–core fuel design manual, 81-37000-DM-000, Rev. 1[7] "SCALE: A Modular Code System for Performing Standardized Computer Analyses for LicensingEvaluations", ORNL/TM-2005/39,Ver.5,Vols.I–III, 2005. Avail. from RSICC at ORNL as CCC-725.

Session 18.2.2:Core design and protection

PROTOTYPE DESIGN OF AN ADVANCED COREPROTECTION CALCULATOR SYSTEMSEUNG YEOB BAEG 1 , SEUNG MIN KIM 1 , JONG SIK BAE 1 , DONG WOOKKIM 2 , SUNG HO KIM 3 , HANG BAE KIM 3 , WANG KEE IN 4 , YOUNG HOPARK 5 , AND CHANG HO CHO 11 Doosan Heavy Industries & Construction Co., Ltd., Yongin, Gyeonggi, South Korea2Korea Hydro & Nuclear Power Co. Ltd., Daejeon, South Korea3 Korea Power Engineering Company, Inc., Daejeon, South Korea4Korea Atomic Energy Research Institute, Daejeon, South Korea5 Korea Nuclear Fuel Co., Ltd., Daejeon, South KoreaABSTRACTThe core protection calculator system provides on-line calculations for Departure fromNucleate Boiling Ratio (DNBR) and Local Power Density (LPD) of a nuclear power plant. Itgenerates a reactor trip signal whenever the core conditions exceed the DNBR or LPD designlimit. The System consists of four independent channels, employing a two-out-of-four triplogic. System configuration, hardware platform and an improved algorithm of the newlydesigned core protection calculator system named Reactor COre Protection System(RCOPS)", are presented in this paper. Failure modes and effects analysis, and softwarepreliminary hazard analysis for RCOPS have been performed. One channel of RCOPSfabricated for this R&D project where final integration software test was performed. TheRCOPS consists of one Core Protection Processor (COPP) rack, one Control ElementAssembly Processor (CEAP) rack, and two Channel Communication Processor (CCP) racksper channel based on the HFC 6000 safety platform. The two CCP racks share the ReedSwitch Position Transmitter (RSPT) signal for redundancy and each CEAP of the fourchannels share penalty factor (PF). Through this scheme, RCOPS minimizes the number ofprocessors compared to conventional core protection systems, Each channel of RCOPSconsists of four sub-racks and shares the same penalty factor with the other three channels.Features of RCOPS include improvement of core thermal margin through a revised on-lineDNBR algorithm, resolution of the latching problem of control element assembly signal, andaddition of the pre-trip alarm generation.1. IntroductionThe core protection calculator system provides on-line calculations for Departure from Nucleate BoilingRatio (DNBR) and Local Power Density (LPD) of a nuclear power plant. It generates a reactor trip signalwhen the core condition exceeds the DNBR or LPD design limit. Since 2004, Doosan has worked on anR&D project to develop its own core protection calculator system, RCOPS with improved algorithm forDNBR calculation. DOOSAN has been in charge of project as the main contractor with KNFC, KAERI,KOPEC and KHNP joined as sub-contractors [1].The RCOPS includes DNBR algorithm improvement for bigger thermal margin, resolution of thelatching problem of the false CEA position signal, and addition of the pre-trip alarm generation. Thechange of on-line DNBR calculation algorithm is considered to improve the DNBR net margin by 2.5%

~ 3.3% [2]. The DNB Margin with the digital system is also higher than that for the analog system by 8~ 12.1% of rated power for commercial and advanced reactors, respectively [3].The RCOPS, based on HFC-6000 safety class PLC is a trademark of HFControls owned by Doosan.RCOPS uses an improved calculation algorithm and has a different system configuration compared tothe Shin-Kori Nuclear Power Plant Unit 1 and 2 (SKN1&2) type core protection calculator system.HFC-SBC06 is the main CPU processor board which has a Pentium100 processor and two 80386EXprocessors in a single board [4].2. Reactor COre Protection System, RCOPS2.1 System ConfigurationRCOPS is a safety related system that consists of four identical channels whose input signals should beindependent from one another. Two separate reed switches are installed in one CEA housing for thestandard PWR (Pressurized Water Reactor). Each Core Protection Processor (COPP) of a channelmonitors the CEA positions from one quadrant of the reactor core. These CEAs are called the targetCEA of that channel. COPP generates planar radial peaking factors which are used to calculate LPD andDNBR. CEA Processor (CEAP) takes all CEA positions of the reactor core through ChannelCommunication Processor (CCP) and examines the CEA deviation between subgroup positions. If thisdeviation is higher than a specified value CEAP sends the penalty factor to COPP. The single channelcabinet is shown in figure 1.Figure 1 Single channel cabinet of RCOPS2.2 Reliability and FMEA Analysis for RCOPSAccording to the reliability analysis report for RCOPS, unavailability of RCOPS is improved ascompared with the SKN1&2 type core protection calculator system [5]. This improvement is achievedthrough the simplified system structure and lower failure rate of HFC-6000 hardware components.However, from the viewpoint of redundancy, the SKN1&2 type core protection calculator system has asensor redundancy feature because, in case of channel A, the preferred source for target CEA position isRSPT1A and the alternate source for target CEA position is RSPT2A. RCOPS has only I/O redundancy

whereas only RSPT1A is measured by CCP1 and CCP2. Hence, both systems have their own advantagesand disadvantages due to the different system configurations.Reliability Block Diagram (RBD) method was used to calculate the reliability for RCOPS and SKN1&2type core protection calculator system in order to compare reliabilities and characteristics of bothsystems on the same method calculation basis. Isograph Reliability Software (Reliability WorkbenchVersion 9.0) was used as a calculation tool for reliability analysis of both systems. Both systems wereconservatively analysed. As a result, unavailability of RCOPS is 1.800x10-5 failures/demand. Thisresult is 10 times lower than the 1.328x10 -4 failures/demand of a conventional system.Figure 2 RCOPS FMEA Block DiagramFailure Mode and Effects Analysis (FMEA) is the qualitative method to analyze the effects of singlemode failures. FMEA was performed to assure safety and reliability of RCOPS at the end of the designphase. Prior to analysis, over 50 single mode failures for RCOPS were defined on the block diagram.Figure 2 is the block diagram for RCOPS marked by unique numbers for each single mode failure. It isassured that RCOPS does not lose its intended safety functions for most of the failures. In the event ofprocessor failures, RCOPS did not have enough features to satisfy the system requirement. Therefore thedry contact to generate channel trips using watch dog timer was supplemented.2.3 Improved of core thermal margin through DNBR algorithm improvementWhen the calculation of DNBR is executed in a software module of the core protection calculatorsystem, SKN1&2 type core protection calculator system uses the most simplified DNBR algorithm tosave execution time. However, RCOPS uses an improved algorithm which employs the on-lineenthalpy transport coefficient calculations. The enthalpy transport coefficient is used to relate theradially averaged enthalpies of the boundary sub-channels of lumped sub-channels to their counterparts.Its value depends on the core power distribution and the basic operating parameters.The maximum DNB-POL deviation is about 7% between two different DNBR models as shown inFigure 3. About 1 ~ 2% improvement of DNBR margin came from this DNBR model change. Another 1% improvement of DNBR margin came from the on-line calculation of enthalpy transport coefficientwhich was a pre-fixed value in conventional core protection calculator system. As a result, a 2~3% ofDNBR net margin is improved in RCOPS through the use of DNBR algorithm improvement.

The following nominal conditions were used for evaluation of the DNBR margin. Conditions; CoreOutput=100%, Core Inlet Temperature=564.5F, Pressurize Pressure=2250psia, Coolant FlowRate=105%, Peak parameter for Hot Pin Output=1.55. The DNBR Comparison between RCOPS &conventional Core Protection Calculator System is summarized in table 1.Table 1 DNBR Comparison between RCOPS & conventional Core Protection Calculator SystemCombustion rateConventional Core ProtectionCalculator SystemRCOPSBOC(Beginning of cycle) 2.230 2.292MOC(Middle of cycle) 2.073 2.112EOC(End of Cycle) 2.245 2.313DNBR Error, E DNBR(fraction)0.100.050.00-0.05-0.10-0.6 -0.4 -0.2 0.0 0.2 0.4 0.6ASIMOCTORC-SMATRACETOP-DCPCFigure 3 DNBR Error ComparisonFigure 4 DNBR ComparisonFigure 4 shows the DNBR error according to core axial power distribution at the middle of core cycle(MOC). The reason that errors of RCOPS and conventional system have negative values as a whole isthat hot channel flow factor was adjusted to get conservative results than those of TORC through out theentire operation conditions. As shown in Figure 4, when ASI is over 0.2, all of the DNBR calculationvalues are conservative.2.4 Software and Hardware Integration Test and ResultsThree cases of the operating conditions are injected to on-line program through simulator input/outputand compared each DNBR results with the expected DNBR value calculated by off-line Code (FortranCode).Table 2 Comparison between Estimated DNBR Value & Test ResultNo. of CaseEstimated Value Result Value at Deviationfor DNBR Integrated Test1 2.0031 1.9845 0.01862 1.4090 1.389 0.023 1.24 1.229 0.011The first case is to calculate DNBR at normal operation (100% power) condition. The second case is togenerate pre-trip alarm due to ASI pre-trip set point violation during 100% power operation. The thirdcase is to generate DNBR and LPD trip under trip condition. The trip condition is made by changing

ex-core neutron detector signal from the first test case. The test results re-assured that RCOPS prototypethat consists of the improved algorithm and new hardware platform calculates DNBR and LPD about1 % error against expected values during on-line operation. This error will be reduced through softwareverification and validation in the next stage. Table 2 is the result of each test case.Figure 5 is a display on the operator module for test case 1 that is normal operation.3. ConclusionFigure 5 DNBR Test Result Screen at Normal State OperationThe completion of algorithm and application software development for Reactor Core Protection System(RCOPS) concluded the project. The main results included the design of system configuration andestablishment of development facilities including single channel cabinet & I/O simulator, andconduction of integration and functional tests. For 31 months from July 2004 to February 2007, theimproved algorithm for RCOPS was developed. RCOPS has 10 times the availability and twice theresponse time than conventional Core Protection Calculator System. Also excellence of systemconfiguration as a safety system for Nuclear power plants has been analyzed through reliability analysisand FMEA .Unique system configuration reduces the number of components as compared toconventional system. The algorithm improvements employed in RCOPS would increase the plantavailability as well as the reactor core thermal margin. RCOPS has been implemented as a single channelfacility. These products will be utilized through the second stage of the project for system verificationand validation.4. References[1] Seung-Min Kim, Seung-Yeob Baeg, Jong-Sik Bae, Dong-Wook Kim, Sung-Ho Kim, Hang-BaeKim, Wang-Kee In, Young-Ho Park, and Chang-Ho Cho, "Development of an AdvancedCore Protection Calculator System Named RCOPS", 15th Pacific Basin Nuclear Conference,October 15-20, 2006[2] Wang Kee In, Dae Hyun Hwang, Young Ho Park and Tae Young Yoon, “Evaluation of DNBRcalculation methods for advanced digital core protection system,” Korean Nuclear SocietyConference, May 29-30, 2003[3] Wang Kee In, Dae Hyun Hwang, Yeon Jong Yoo, Sung Qunn Zee, "Assessment of coreprotection and monitoring systems for an advanced reactor SMART", Annals of NuclearEnergy 29, 2002.[4] HFControls Corporation, “HFC-6000 Nuclear Safety System Technical Description,” 2005.[5] Korea Power Engineering Company Inc., “Reliability Analysis Report for Reactor CoreProtection System,” 2006.

MANUFACTURE AND TEST OFPROTOTYPE REACTOR PROTECTION SYSTEMSUNG-JIN LEE, YOUNG-HO KOO, SEONG-TAE KIM, CHANG-HO CHODoosan Heavy Industries & Construction Co., Ltd.ABSTRACTThis paper describes the equipment manufacturing and testing process for the ReactorProtection System (RPS), which is being developed as a part of KNICS (Korea NuclearInstrumentation and Control System) R&D project: "Development of a Digital ReactorSafety System". The proposed RPS has a simpler structure, better maintainability and betterself-diagnoses than the conventional DPPS (Digital Plant Protection System) used in Koreanuclear power plants. We describe how the RPS control cabinet is manufactured to meetIEEE Std. 603, “IEEE Standard Criteria for Safety Systems for Nuclear Power GeneratingStations”, and how the application software for RPS is developed and validated inconformance with NUREG-0800/BTP HICB-14, USNRC R.G 1.173 and IEEE Std.1074,“IEEE Standard for Developing Software Life Cycle Processes”. We verified the functionalintegrity of bistable processor, coincidence processor, and external interface processor byobserving all input and output signals communicated between RPS and external I/Osimulator. To verify the integrity of RPS under Design Basis Accident of nuclear powerplants, we performed equipment qualification tests including environment, EMI/RFI andseismic. Through this validation program, we can conclude that the development and testprocedures of the RPS control cabinet meet the nuclear safety system regulationrequirements and that we expect the RPS cabinet to operate properly under real fieldconditions1. IntroductionA nuclear Instrumentation and Control (I&C) system is divided into two categories, one a non-safetysystem and the other a safety system. The role of the non-safety system is to maintain a Nuclear PowerPlant (NPP) in normal operation by regulating reactor control, feed-water control and BOP control andby supporting plant-monitoring systems. The safety system has the role of controlling abnormalconditions due to a Design Base Event (DBE). This is achieved by a reactor shutdown, actuatingengineered safety features to mitigate the effects of a DBE, and protecting the public from exposure toradioactive materials. The safety system is very important because it influences the determination ofNPP construction.In Korea, although system design technology was localized after Yonggwang 3&4 NPP construction,component design and equipment supply technology is still dependent on foreign technology.Nowadays, the nuclear industry has a bright prospect with the construction of new NPPs and I&Creplacement and upgrade in operating plants. The technology of the safety I&C system is moreimportant than in previous years because a digital system requires high reliability and strict developmentprocedures to avoid errors and/or malfunctions. Doosan and Korea Atomic Energy Research Institute(KAERI) have joined forces to start this safety I&C system development R&D project.2. System Design of RPSReactor Protection System (RPS) generates signals to actuate a Reactor Trip (RT) and engineered safetyfeatures function automatically whenever monitored processes reach predefined limits. RPS shall havefour channel cabinets. The RPS cabinet for each channel may be physically distributed into four separateI&C equipment rooms. Bistable Processor (BP) receives a measurement of process variables, determinesthe trip state by comparing the process variable measurement to a predefined limit, and sends the tripstate to Coincident Processor (CP). CP determines the coincident output based on the status of four trip

inputs from BP of each channel. Automatic test and Interface processor (ATIP) tests the integrity of BPand CP and provides the interface between RPS and other systems. Cabinet Operator Module made up ofIBM-compatible military computer provides the man-machine interface for the test and integritymonitoring of each controller. BP, CP and ATIP are made up of the safety grade Programmable LogicControllers (PLC) that have typically a central processor module, communication modules, digitalinput/output modules and analog input/output modules. In one channel of RPS, there are duplicate BPsand CPs for the purpose of increasing system availability. The initiation circuit made by electricaldevices only performs the 1-out-of-2 logic using the RT and ESFAS signals from both CPs individually.Main Control Room Qualified Alarm SystemPICPCENFMSSignalProcessingBPCOMCPATIPRTESFASRTSGESF-CCSFig 1. The block diagram of proposed RPS3. Software Design and ImplementationThe RPS application software was developed in accordance with NUREG-0800/BTP HICB-14,USNRC R.G 1.173 and IEEE Std.1074, “IEEE Standard for Developing Software Life CycleProcesses”. In the software planning phase, we developed fourteen software planning documentsincluding development plan, configuration management plan, software validation and verification planand safety plan. Each planning document was developed on the method proposed in NUREG/CR-6101.In the requirement phase, Software Requirement Specification (SRS) was developed on therequirements proposed in IEEE Std. 830, and SRS includes the functional requirements and interfacefeatures. In the design phase, Software Design Description (SDD) was developed on the methodproposed in IEEE Std. 1016 and includes hardware and software architecture, module decompositionand interface of modules. In the implementation phase, based on SDD, the source code was developedusing PLC engineering tools. A software coding guideline was also developed to obtain uniformity andreadability of each SDD.Fig 2. Software development life cycle of RPS4. Software Verification and Validation

The objective of the RPS software verification and validation is as follows: Software errors must beeasily detected at the initial software development phase; Developed software shall conform to systemdesign requirements; Technical and licensing standards and the RPS software verification and validationactivity shall be performed per the requirements of NUREG-0800/BTP HICB-14 and IEEEStd.1012-1998, “IEEE Standard for Software Developing Software Life Cycle Processes”. SoftwareV&V methods include license conformance review, source code inspection, formal verification,traceability matrix analysis and software test. Software test is performed by two methods. The first is asoftware unit/module test, which determines whether each application program installed in a processormodule functions properly to the design requirements. This test is performed by PLC engineering toolsconnected to PLC by serial communication. Software unit is user-made function block, composed ofseveral basic functions for the purpose of programming reuse. The second method is a softwareintegration test. This checks if the communication interface functions properly between processormodules. This test includes safety data link interface from BP to CP or vice verse, inter-processor statusmonitoring communication interface and inter-channel status communication interface.Define Functional RequirementDefine Input VariableDefine Internal VariableDefine Output VariableN.Calculation of variabletrip set pointSelectionOf InputsCalculationoftrip set pointDefine Application Logic NuSCR Specification for th_SG1_LVL_RPS_Lo_TripFormalizing by NuSCRFormal ValidationF.LFig 3. The software requirement specification process5. Hardware Design and ManufactureThe processor modules embedding each application program are installed in the RPS cabinet with otherelectronic equipment. The RPS cabinet is seismic category 1 and the equipment installed in it must bedesigned to withstand the cumulative effects of five Operational Basis Earthquakes (OBEs) and one SafeShutdown Earthquake (SSE) without loss of physical integrity. During the seismic event, no parts of theequipment shall loosen, bend or crack in a manner to impair proper operation. During and after a seismicevent, the safety-related parameters of the equipment shall be maintained as the system functionalrequirements. All equipment of the RPS cabinet must be designed and manufactured to minimize boththe generation of electromagnetic interference and the susceptibility of equipment to externallygenerated EMI. The completely assembled RPS shall be designed to operate continuously without lossof function over the entire range of environmental conditions.

The RPS cabinet is designed so that the heat generated by internal electrical components can beexhausted to the outside by an air draft fan installed in the cabinet top shelf. Each cabinet is suppliedwith an ungrounded AC power source. AC power is supplied through a main circuit breaker, a surgearrestor and distribution circuit breakers. Each power supply has proper branch circuit blocks by fuses soas not to operate improperly by a single circuit fault. For higher availability of the DC power system, twoDC power supplies are controlled by diodes. If a primary DC power supplier should fail, the DC powersource is switched to a secondary DC power supply so that the total DC power is maintained. In the RPScabinet, there are cabinet door-open switches, a cabinet internal temperature rise detector and a fan stopdetector which are classified as associated circuit. It is designed such that irrespective of the failure of anassociated circuit, safety related circuitry is not affected by a single failure of associated circuit byproper separation and isolation.Fig 4. The process of hardware design and manufacture of RPS6. System Validation TestingSystem functional requirements are verified by a system validation test. The system validation test isperformed by observing all input and output signals communicated between RPS and an external I/Osimulator. The results of the bistable processors and the coincident processors to the input conditions arecompared with the anticipated results via Man-Machine-Interface in the I/O simulator. Specific testitems include trip set point algorithm logic, operating bypass logic, trip channel bypass and all bypasslogic, coincidence logic, response time test, hardware integrity test and fail-safe testIt is required that the class-1E equipment should meet or exceed the specification requirementsthroughout its installed life. It is the primary role of qualification to ensure that the class-1E equipmentcan perform its safety functions without failure that could lead to common cause failures underpostulated service condition. RPS control cabinet is located in a mild environment where a seismic eventis only design base event of consequence. RPS control cabinet is qualified by type test, as defined inIEEE Std.323-1983, “IEEE Standard for Qualifying Class 1E Equipment for Nuclear Power GeneratingStations”. It satisfies the qualification if it accounts for significant aging mechanism, subjects theequipment to specified service condition, and demonstrates that such equipment can subsequentlyperform its intended safety function for at least the required operating time. Specific qualification test

processes include stress analysis, aging analysis/preconditioning, visual inspection, initial functionaltest, environmental test, EMI/EMC test and seismic test.SimulatedInputTripOutputFig 5. The configuration of RPS system validation test7. ConclusionAn RPS has been developed with a simpler structure, better maintainability and better self-diagnosesthan the conventional DPPS. Verification of the proposed RPS has been completed through varioussoftware tests and system validation tests. Through this validation program, we can conclude that thedevelopment and test procedures of the RPS control cabinet meet the nuclear safety system regulationrequirements and that we expect the RPS cabinet to operate properly under real field conditions.8. References1. IEEE Std. 7-4.3.2-2003, IEEE Standard Criteria for Digital Computers in Safety Systems of NuclearPower Generating Stations.2. IEEE Std. 603-1998, IEEE Standard Criteria for Safety Systems for Nuclear Power GeneratingStations.3. IEEE Std. 730.1-1989, Standard for Software Quality Assurance Plans.4. IEEE Std. 1012-1998, Standard for Software Verification and Validation.5. IEEE Std. 1074-1997, IEEE Standard for Developing Software Life Cycle Processes.6. IDiPs-RPS Design Specification of Reactor Protection System, KNICS-RPS-DS101, Rev.017. IDiPs-RPS Software Verification and Validation Plan of Reactor Protection System,KNICS-RPS-SEP110, Rev. 008. IDiPs-RPS Software Requirement Specification of Reactor Protection System,KNICS-RPS-SRS221, Rev.01

CORMORAN: CORE MONITORING &REACTOR ANALYSISD. VANTROYEN, C. SCHNEIDESCH, R . DE WOUTERSTractebel Engineering SuezAvenue Ariane, 7, B-1200 Brussels – BelgiumABSTRACTThe CORMORAN project covers two distinct applications: In-Core flux map processingand Core Monitoring.The In-Core flux map processing addresses the regulatory power distribution surveillancebased on monthly in-core flux treatments. A tighter link between cycle follow and coredesign can be ensured by integrating the measured flux map discrepancies into the coreneutronics calculations. This integration is in particular mandatory for an accuratetreatment of spatial core dissymmetry.The Core Monitoring addresses real-time core simulation based on reactor data withpredictive capability. The implementation of a 3D neutronics code in the core monitoringpackage provides the Utility with modern core monitoring capabilities, matchingsimultaneously the evolution of the support expected for core operation and the increasingheterogeneity of the neutronic properties of the core reload.IntroductionA core monitoring system and a new in-core flux map processing have been developed by the‘Core&Fuel Studies’ section of Tractebel Engineering.The development of both applications is justified by the need to- ensure core monitoring services covering core follow and core operation support for all Belgianunits;- upgrade the processing of in-core flux maps by implementing a method suitable even in case of incorepower tilts.Those two applications are included within the same project, as they share common developments, usethe same platform, and are based on the same neutronics code package. Tractebel Engineering uses thePanther code system, developed by British Energy and distributed by Serco Assurance (Winfrith,U.K.). Panther is a modular code composed of many functions which exchange data through aninternal data bank. In particular, Panther solves the diffusion equation in 2-energy groups with thenodal analytical method.It is to be mentioned that none of the new applications is yet installed and operational on the site’splatform. The project is still in progress and its validation process in its last stages. The coremonitoring system is however available in Tractebel Engineering offices for all our NPPs.

Application 1: core monitoring – current core statusThe on-line monitoring of the core status is performed by the following stages of the entire process• Data acquisition: every 2 minutes, the data server of the site acquires necessary measurements:power level, temperatures, ex-core axial imbalance (ΔI);• The application interrogates the data server and collects data on a local server;Site LANUNITACQUISITION SYSTEMEBL-TENetworkServerData Storage\ ApplicationsLinuxCore FollowPC’sTE LANLinuxCore Follow/Tests/BackupTEServerPC’sFigure 1. Schematic chart of the data acquisition• Taking account of the measured reactor data, a short transient with 3D power distribution iscalculated; all results are made available on a local machine;• The display of the current core’s status is updated with the last evaluation;• The process waits then for a new set of data.Figure 2. Core Status Display

By the help of the ‘close-to-reality’ display, one can• Easily follow the current status of the core in terms of measured data (power level, averagevessel temperature, boron concentration, ΔI, position of the ΔI in the authorized band,historical behaviour of the ΔI for the last 6 days);• Evaluate the results coming from the theoretical calculationo Axial power distribution, peak axial factor;o F Q and remaining margin towards LOCA;o Current Xenon poisoning balance and expected evolution in the close future;• Be informed of the configuration leading to an optimal situation.Application 1: core monitoring – predictive toolOnce the current status of the core is well known, a second part of the core monitoring application hasbeen developed in order to give assistance to the operator in realizing specific manoeuvres such aspower load-follow, Xenon full power transient for the excore calibration, start-up and so on.The predictive scenario is directly connected to the current core reality and may generate several trialsin order to assure the feasibility of the manoeuvre.The process is conducted by an Excel interface (Fig 3) where the operator can introduce the scenariospecification and launch the process.100%90%80%70%60%50%40%30%20%10%0%13 14 15 16 17 18 19 20 21 22 23 00 01 02 03PowerCormoran - Rx.xx - Predictive CalculationTihange 1 - Cycle 26Nom. Power Temperature Maximum Flow BoronElectr. Therm. HZP HFP Dilution Borat1008.5 2865 286.1 302.7 27 13.5 7500MWe MWT ºC ºC m3/h m3/h ppmDate Power # Comments Power Duration/Rate Temp. D Gr Boron critic Flow Delta I% ºC pas ppm search Limit Band25/06 13:58 99.8 1 Load initial state 99.8 % hr 302.2 220 1039.325/06 14:48 50.0 5 power decrease 50.0 % 10 MWe/min Réf. DDI Crit Bore Non 0.025/06 19:48 50.0 5 plateau 50.0 % 5 hr Réf. DDI Crit Bore Non 0.025/06 20:38 99.8 5 power increase 99.8 % 10 MWe/min Réf. DDI Crit Bore Non 0.026/06 01:38 99.8 5 stabilisation 99.8 % 5 hr Réf. DDI Crit Bore Non 0.0Launch calculation Clean upCompleteInsert line Delete lineFigure 3. Excel Interface for predictive calculation – user’s inputA typical prediction is made of following steps:• Load initial status: interrogates the on-line core monitoring system and gets the mostrecent data;• Introduce the chosen scenario: time steps, power level, rods positions;• Specify the transient constraintso type of criticality search (boron, rods, temperature)o flow limitation;o rules for the ΔI band: close to the reference, authorized in the ±5%, no limitation• Launch the calculation: the scenario is translated into Panther commands and copied intothe appropriate server where the calculation is performed;• Collect the calculation results: these are formatted and transferred into two specific sheetsof the interface: graph (Fig 4) and results (Tables 1a and 1b);• Modify the scenario: the results of the current scenario are analyzed; the user can easilyadapt the data and launch a new variant.

Prévision CormoranTihange 1 - Cycle 26Possible problemsBegin 25/06/2007 13:58End 26/06/2007 01:38Dilution WaterBoron30.850 m3 à 0 ppm5.857 m3 à 7500 ppmTable 1a. Predictive calculation results – Part IThe first part (Table 1a) of the Excel sheet gives the summary of the scenario and presents results andcomments such as prediction as a function of time, total volume of boration and dilution, warnings ofpotential problems (ΔI out of the band, out of the absolute trapeze, rods insertion or extraction limitsreached, subcriticality,…), global conclusion of the transient feasibility, recommendations.Date Pow D C Delta I boron React Volume Flow Volume Balance PrimGr Gr abs. discr Conc ivity dilution borat Boron/wat Temp% step step % % ppm pcm litres m3/hr litres litres Deg.C25/06/2007 13:58 99.8 220 225 2.9 . 1040 0.0 303.025/06/2007 14:08 89.9 203 225 2.4 . 1057 3.5 595 301.825/06/2007 14:17 79.9 194 225 2.1 . 1075 3.8 613 300.125/06/2007 14:28 69.9 186 225 1.8 . 1091 3.4 573 298.325/06/2007 14:38 60. 179 225 1.6 . 1105 3.0 517 296.625/06/2007 14:47 50. 173 225 1.3 . 1119 3.0 487 294.825/06/2007 15:47 50. 174 225 1.3 . 1093 5304 5.3 294.825/06/2007 16:47 50. 174 225 1.3 . 1076 3662 3.7 294.825/06/2007 17:47 50. 175 225 1.3 . 1065 2326 2.3 294.825/06/2007 18:47 50. 175 225 1.3 . 1059 1263 1.3 294.825/06/2007 19:47 50. 175 225 1.3 . 1057 428 0.4 294.825/06/2007 19:58 60. 182 225 1.6 . 1039 3828 22.5 296.625/06/2007 20:08 69.9 189 225 1.8 . 1023 3578 21.1 298.325/06/2007 20:17 79.9 197 225 2.1 . 1007 3513 22.0 300.125/06/2007 20:28 89.8 208 225 2.4 . 993 3256 19.2 301.825/06/2007 20:38 99.8 225 225 1.7 -.9 977 3693 21.7 303.725/06/2007 21:38 99.8 225 225 1.8 -.8 1010 1.1 1142 303.625/06/2007 22:38 99.8 225 225 1.9 -.8 1032 0.8 800 303.625/06/2007 23:38 99.8 225 225 1.9 -.7 1048 0.5 546 303.626/06/2007 00:38 99.8 225 225 1.9 -.7 1058 0.4 360 303.626/06/2007 01:38 99.8 225 225 1.9 -.7 1065 0.2 225 303.6Table 1b. Predictive calculation results – Part IIThe second part (Table 1b) of the Excel sheet gives the details of the results, step by step: data(power/time), rods position, absolute value of the ΔI, discrepancy between calculated ΔI and thereference (ΔΔI), boron concentration, reactivity, boron/water needs, temperature.Some results are enhanced when a limit is reached in order to draw attention on a specific difficulty.The predictive results are also presented as a time dependent graph (Fig 4). The sheet is divided intotwo parts:• Part I : description of the chosen scenario in terms of power and rods position• Part II : evolution of the calculated boron concentration and the calculated ΔΔI together withthe ΔI bounding limits

120.Cormoran - Rx.xx - Predictive CalculationTihange 1 - Cycle 26300Power (%)100.80.60.40.20.0.250200150100500D Group (steps)Pow % D Gr step C Gr step10%1175Discrepancy Delta I - ref. (%)5%11250%1075-5%1025-10%97513h 14h 15h 16h 17h 18h 19h 20h 21h 22h 23h 00h 01h 02h 03hTime (hour)Boron (ppm)Delta I discr % Di target % Di lim % boron Conc ppmFigure 4. Predicitve resultsThe predictive Calculation can be useful in the frame of transients such as:• Start-up conditions after shutdown;• Xenon oscillation: local power variation, full power excore transient;• Start-up conditions after long stay at zero power;• Load-follow manoeuvres.Application 2: in-core flux map processingThe treatment of the in-core flux maps is an off-line process: a flux map is taken monthly using a setof mobile miniature fission chambers, which provide axial distributions of relative Reaction Rates(RRs) in roughly 1/3 of the assemblies in the core.A computer program is used to process the measured data. The first function of this processor is totransform the raw measured data into a consistent set of cross-calibrated, repositioned and normalisedRRs. The second function is to produce estimates of RRs in the non-instrumented positions. The thirdfunction is to build the 3D power distribution from the completed set of RRs, using the results of offlinetheoretical simulations. The main parameters of this 3D power distribution, such as the AxialNOffset (AO), the power peaking factors F Q and F Δ H, the quadrant power tilts, are compared with theirregulatory limits to guarantee the safe continued operation of the reactor.The current in-core flux map processor relies on power reconstruction techniques developed in the70’s and suffers from two approximations:- 1/8 core symmetry is assumed in order to estimate the RRs in non-instrumented assemblies;N- The reconstruction F Q and F Δ His obtained from the neighbouring thimbles by a weightedinterpolation process, where the weights have been empirically determined through benchmarking.The consequence of the first approximation (1/8 core symmetry) is that an unpredicted core tilt wouldbe misrepresented in the 3D power reconstruction, with potential consequences on:- The safety for the current cycle, in case of underestimation of the reconstructed F Q and F Δ.NH

- The operation of the next cycle, where the tilt could be reinforced if adequate corrective measuresat reload design had not been applied.The consequence of the second approximation is a lack of method in the determination of theuncertainties affecting the reconstructed values of F Q and F Δ.There is therefore a motivation to revisit the power reconstruction stage of the current application andto replace it by a more accurate method.A survey of the open literature on the subject has shown that several recent developments in differentcountries are based on statistical adjustment of the neutron flux distribution obtained by a 3D nodalsimulator, on the measurements.It seemed quite natural to consider a method involving adjusting a Panther solution. The developmentwould however be restrained to combining existing Panther functions, as the creation of specific newfunctions was not envisaged. Consequently, the method had to be fairly simple, and could not resort toelaborate statistical methods requiring software developments.Such an “adjustment” method has been developed in the frame of Cormoran.The principle is to adjust iteratively a best-estimate set of calculated nodal reaction rates (RRs) (“trialfunction”) toward a “target” which is based on the set of measured RRs. The adjustment process isimplemented by tuning a selected cross-section recursively (e.g. the thermal absorption cross-section).The iterations are stopped when the standard deviation of the differences between the calculated andmeasured RRs has decreased to a value close to the measurement reproducibility error.The value of k-effective is then reset to the initial value by a scaling factor applied uniformly to theneutron source cross-section.Finally, the power distribution of the adjusted solution is used to determine the various safetyparameters, such as F Q and F Δ.NHThe standard geometrical model of the Panther calculations used for the Core Reload Safety Studiescontains typically 24 axial nodes, and the assembly grids are homogeneously smeared along the activelength. This model has been extensively validated on HZP and HFP core measurements on UO2 coreloading of the 7 NPPs operating in Belgium. Loadings containing MOX are not covered by thisvalidation.It has been necessary to develop a particular geometrical model for the “trial functions”, including afine axial mesh (typically 71 nodes), identical to the one of the processed measured RRs, and anexplicit model of the assembly grids.Extensive validation of the adjustment method has been done firstly by numerical benchmarkingwhere a theoretical Panther solution plays the role of the measurements. This solution is then modifiedby changing e.g. the rod insertion, or another parameter, and plays the role of the “trial function”. It isthen iteratively adjusted on the initial set of RRs. This work has shown the necessity of having the bestpossible quality of the trial function, and has also demonstrated that the adjustment process always ledto an improvement.At this stage, the choice of the adjustable cross-section and the determination of a relaxation factorensuring the best convergence behaviour were made.Secondly, 1135 flux maps previously processed with the present processor have been treated with theNadjustment method. The new values of F Q and F Δ Hare higher than the old ones by respectively 0.2%and 0.6%, in average. This is due to the better representation of the power tilts and is consistent withthe expectations.NH

F ΔNHThe uncertainties on F Q and can be now be determined in an appropriate manner. This is based onthe bounding character of the uncertainties of the “trial function” with respect to the convergedadjusted solution. An extensive validation of the model used for the “trial functions” has shown thatNthe usual uncertainties of 5% and 4% on F Q and F Δ Hrespectively can be kept with this method.However, this implies that realistic core conditions in terms of burnup, power level, coolant inlettemperature and control rod positions are modelled in the “trial function”.The various benchmarking and validation tests have shown that the adjustment method is fast, robustand reliable. The next stage of development will be to integrate it in a new flux map processing codewhich will also include the old method. This may be necessary in case e.g. of MOX loaded cores forwhich the Panther models have not been validated.Benefits for the UtilityThe technical advantages offered by the enhanced core follow system and flux map processing willbring the Utility the following benefits:- Capability to meet safety expectancies expressed by the Belgian Safety Authorities (coredissymmetry);- Improve the capacity factor by optimising power escalation or load manoeuvres;- Increase of core operational efficiency by early detection of possible operation anomaliesand by optimisation of corrective manoeuvres;- Reduction in core-follow maintenance by using standard low-cost hardware and singlecode proficiency;- Capability to introduce new functionalities such as ex-core models or more theoreticaldisplays on core status for a better on-line management of the power distribution.In addition, core operation would also benefit from a better Tractebel Engineering technical supportensured by a better knowledge of the core history and core status, as expressed in WANO SOER2004-1.Objectives for the futureFor the moment, the whole process is in its last stage of validation and qualification phase. It isforeseen that the core monitoring system will be installed at Tihange by the end of 2007. Theinstallation at Doel will be done in 2008. The new in-core flux maps treatment must be qualified bythe Belgian Safety Authorities. This process is still on its way and will be completed in 2008.Further developments will start in the next future, especially to enhance the graphical user interface:radial power, reaction rates, peak power, burnup distributions. Time dependent evolution with amoving window could also be useful, e.g. behaviour of the average ΔI together with the power levelfor the last 30 days.

Session 18.2.3New developments in fuel technology

STRATEGY OF ADVANCED PWR FUEL DEVELOPMENTIN KOREAKYU–TAE KIM, JOON-HYUNG CHOI and JUNG-MIN SUHKorea Nuclear Fuel Company493 Dukjin-dong, Yuseong-gu, Daejeon, Korea 305-353ABSTRACTIn Korea, sixteen PWRs and four PHWRs are in operation, and eight more PWRs are to bebuilt by the year 2017. Korea Nuclear Fuel Company is supplying nuclear fuel for alloperating PWRs and PHWRs in Korea since 1989. The Korean nuclear fuel technologieshave been developed by a step-by-step strategic approach. At first, the nuclear fuels wereimported since Korea did not have any fuel fabrication technology at that time. Secondly,KNFC was established in 1982 and localized successfully the nuclear fuel technologies,based on technologies transferred from foreign fuel vendors. Thirdly, to improve thetransferred fuel technologies, the “Mid- and Long-term Nuclear Fuel DevelopmentPrograms” were established in 1999. Based on these programs, PLUS7 TM fuel has beendeveloped jointly with a foreign fuel vendor and verified through LTA in-reactorirradiations, and its batch application started from June of 2006. PLUS7 TM is an advancedfuel for the OPR1000s and APR1400s. Likewise, both 16ACE7 TM and 17ACE7 TM fuels,which are advanced fuel for the 2-loop and the 3-loop Westinghouse type plants, have beendeveloped jointly with the same foreign vendor. Their in-reactor performances are nowbeing verified through four LTAs. Lastly, to be more competitive than ever and to moveinto the world fuel markets, The “X-Gen Fuel Project” started from the year 2005. Thisproject is planned to be finished by the year 2015.1. IntroductionWe are now facing new challenges as the world’s energy demand is expected to double by the year2050. Oil price is dramatically soaring as the demand is exploding and fossil fuels are being rapidlydepleted. At the same time, our world must also face such serious problems as global warming,environment destruction, and water shortage. Therefore, the world should undertake every effort tomaximize the contribution of nuclear technology to confront these problems. There are currently 429nuclear power reactors operating in 30 countries. They have about 370 gigawatts of generatingcapacity and supply about 16 percent of the world’s electricity.Considering limited natural energy sources in Korea and oil shocks, the policy of electricity generationby nuclear power was first introduced in the 1970s. In Korea, sixteen PWRs and four PHWRs are nowin operation, and eight more PWRs are to be built by the year 2017.Korea Nuclear Fuel Company (KNFC) was established in 1982 to supply nuclear fuel for all operatingnuclear power plants in Korea, based on nuclear fuel technologies transferred from foreign fuelvendors. KNFC has manufactured nine different types of PWR fuel with a 350 MTU/year productioncapacity as well as one type of PHWR fuel with a 400 MTU/year production capacity. KNFC willhave increased the PWR fuel production capacity up to 550 MWD/MTU by the end of 2008,responding to eight more PWRs to be built by the year 2017.KNFC initiated the “Mid- and Long-term Nuclear Fuel Development Programs” in 1999. It hasalready developed three kinds of improved fuel for PWRs that are PLUS7 TM fuel for the OptimizedPower Reactors (OPR 1000s) and the Advanced Power Reactors (APR 1400s), and 16ACE7 TM and17ACE7 TM fuels for the 2-loop and 3-loop Westinghouse type plants, respectively. To be morecompetitive than ever and to become one of global fuel technology leader, KNFC launched the “X-Gen Fuel Project” in September of 2005 and plans to finish it by the year 2015.2. Nuclear Power and Fuel Technology LocalizationDue to limited natural sources, Korea had to rely heavily on imported fossil fuels to generate in itsinitial stage of economic development. However, oil shocks in the 1970s became a turning point forour country to decrease our dependence on oil. Naturally the nuclear power that promised low cost and

stable energy supply was our choice. Today Korea is operating one of the most active nuclearprograms in the world. Twenty nuclear power plants are in operation in Korea and eight more PWRsare to be built by the year 2017. Sixteen operating PWRs comprise of eight Westinghouse type plantsand eight OPR1000s. Eight PWRs to be built consist of four OPR1000s and four APR1400s. The totalinstalled nuclear capacity was 17,716 megawatts as of December 2006, which accounts for about27.0% of the total electricity generation capacity. Also, nuclear power generation in 2006 reachedabout 149 billion kWh or 39.0% of the country’s electricity generation. The capacity factor of nuclearplants in Korea is maintained at more than 90% since 2000. It should be noted that in 2005 it reached95.5% which is 16.2% higher than the world average capacity factor.By the way, four additional OPR1000 units and four APR1400 units with a combined generatingcapacity of 11,600 megawatts are planned to be commissioned by 2017. If everything goes as plannedby the year 2017, a total of 28 nuclear power plants at six locations with the total installed capacity of26,637 megawatts, 31.2 percent of the total power generation capacity of 85,438 megawatts, will be inoperation, accounting for 46.4 percent of the total power generation of about 434 billion kWh.The Korean nuclear power technologies have been developed by a step-by-step strategic approach. Atfirst, Korea brought in nuclear power plants from foreign vendors on the turnkey basis in the 1970s.Secondly, Korea has developed the OPR1000 design and built eight OPR1000s, based on the proventechnologies transferred from the former Combustion Engineering in the 1980s. The transferred keytechnologies include the NSSS of Palo Verde and the T/G of Perry Unit 2 as well as the latesttechnologies and the experience in construction and operation gained from previous nuclear powerplants. The OPR1000 has key design features such as 2-loop RCS design, power level of1050MWe/2825MWt, plant design life of 40 years and plant availability of 80~87%. Thirdly, wedeveloped the APR1400 design with significantly improved and economic efficiency in the 1990s. TheAPR1400 is an evolutionary advanced light water reactor with power level of 1400MWe/4000MWt. Itis based on the experiences and technologies of the OPR1000 units operating in Korea. The APR1400incorporates almost all of the EPRI ALWR requirements and other state-of-the-art technologiesavailable today in order to enhance nuclear safety as well as construction and operational cost,operability and maintainability. The APR1400 has key design features such as 2-loop RCS system,plant design life of 60 years, fully digitalized man-machine interface system, direct vessel injectionand in-containment refueling water storage tank.Korea has also four PHWRs in operation that are CANDU 6 type reactors. The CANDU 6 is AECL700 MWe class nuclear power reactor that has key design features such as 2-loop RCS design andplant design life of 40 years.Likewise, the nuclear fuel technologies have been developed by the step-by-step strategic approach. Atfirst, nuclear fuel had been imported since Korea did not have any fuel fabrication technology at thattime. Secondly, KNFC established in 1982 to localize nuclear fuel technology, based on varioustechnologies transferred from the former Siemens/KWU, the former Combustion Engineering, andWestinghouse Electric Company. KNFC started to supply PWR and PHWR fuel from 1989 and 1997,respectively. Thirdly, KNFC began to improve the localized nuclear fuel technology through the jointdevelopment of PWR fuels with Westinghouse Electric Company from the late 1990s and to the mid2000s. Lastly, KNFC started a self-reliant advanced fuel development program from September of2005 to be more competitive than ever and to move into the world fuel markets. By the way, theKNFC’s current business areas may be summarized as follows;• Design: Reactor core design, fuel design and safety analysis• Fabrication: PWR and PHWR fuel• Engineering Service: Nuclear power operation support• Nuclear Fuel Service: Nuclear fuel examination, repair, ultrasonic-driven crud cleaning, etc.• R&Ds: Nuclear fuel components and assembly, design codes, manufacturing process, etc.KNFC has performed reactor core design, fuel design and safety analysis for more than 8 initial coresand 148 reload cores as of the end of 2006, providing engineering services for more operationalmargin and better economic efficiency of nuclear power generation such as power up-rating andrelaxed axial offset control. KNFC also provides nuclear fuel services to improve reliability of nuclearpower plants, shortening refueling outage period and reduce radiation exposure. The nuclear fuelservices cover reactor coolant activity analysis, visual inspection of fuel, repair and reconstitution ofdamaged fuel, poolside examination of irradiated fuel, and ultrasonic fuel cleaning which is used for

1999 and four LTAs fabrication was completed in March of 2002. After obtaining license for thePLUS7 TM LTA irradiation tests [2] , the four LTAs in-reactor verification test started from January of2003 in Ulchin Unit 3 and their 3 rd cycle irradiation has been completed as of February of 2007. TheLTAs in-reactor performances are as good as expected and consequently the license for the PLUS7 TMbatch applications to OPR1000s was successfully obtained [3,4,5] . The first batch application was forUlchin Unit 4 in June of 2006 and then the other OPR1000s in Korea is following, as seen in Figure 3.Figure 2. Division of Responsibilities of PLUS7 TM and ACE7 TM DevelopmentsFigure 3. Key Milestone of PLUS7 TM Fuel Development and ApplicationThe differences in design features of the Guardian and the PLUS7 TM fuels are summarized in Figure 4.The PLUS7 TM fuel has outstanding design features such as reduced fuel rod diameter, all mid-gridswith mixing vanes, and conformal grid spring and dimple. Detail information on the PLUS7 TM fueldesign and out-of-pile test results are given in the PLUS7 TM Design Review Package [6] .On the other hand, the ACE7 TM fuel development project started in August of 2001 and four LTAsfabrication was completed in July of 2004. The four LTAs of 16ACE7 TM and those of 17ACE7 TM arecurrently in the 2 nd cycle irradiation in Kori Unit 2 and Kori Unit 3, respectively. The LTAs in-reactorperformances of 16ACE7 TM and 17ACE7 TM are as good as expected. The region implementation ofthe 16ACE7 TM fuel is scheduled to start from the year 2008, while that of the 17ACE7 TM fuel from theyear 2009. The differences in design features of the current and the ACE7 TM fuels are summarized inFigures 5 and 6. The ACE7 TM fuel has outstanding design features such as reduced fuel rod diameteronly for 16 ACE7 TM , newly-introduced Intermediate Flow Mixer (IFM) or two more IFMs, I-type gridspring, and Gadolinia burnable absorber.

Both the PLUS7 TM and the ACE7 TM fuels have seven outstanding features over the current fuel, whichinclude overpower margin increase of more than 10%, high batch average burnup capability of up to55,000 MWD/MTU, neutron economy enhancement, enhanced seismic resistance capability, reducedgrid-to-rod fretting wear susceptibility, improved debris filtering efficiency and updated manufacturingprocess.Top Nozzle4 G/Ts, 1 I/TBottom NozzleFigure 4. Nuclear Fuels for the OPR1000s and the APR1400sFigure 5. Nuclear Fuels for the 2 loop Westinghouse Type PlantsFigure 6. Nuclear Fuels for the 3 loop Westinghouse Type Plants

Recently KNFC launched the “X-Gen Fuel Project,” based on the development technologiesaccumulated through the 2 nd generation fuel developments. The X-Gen Fuel is called as the 3 rdgeneration fuel. The “X-Gen Fuel Project” started from September of 2005. More than four LTAs arescheduled to fabricate later in the year 2010 and LTAs in-reactor verification tests may start from earlyin the year 2011. On the condition that LTAs in-reactor performances are as good as expected, the X-Gen Fuel batch application is planned to start from the early 2016. Development objectives of the X-Gen Fuel can be summarized as follows [7] ;• Burnup: Batch average burnup > 65GWD/MTU• Overpower Margin: Greater than 15% with respect to the Guardian fuel• Seismic Performance: Meet 0.3g seismic load requirement• Fuel Reliability: Zero defect fuel• Manufacturability: Enhance manufacturability against the Guardian fuelIn order to meet aforementioned design objectives, the proposed design features are;• Top nozzle with easy reconstitutability and minimized fuel assembly bow• Bottom nozzle with efficient debris filtering device, low pressure drop and uniform flow forfretting wear prevention• Fuel rod with advanced cladding such as HANA, large grain pellet and low volume plenumspring• Guide thimble tube with 0.3g seismic load capability and hydraulic compatibility• Mid Grid with enhanced mixing, 0.3g seismic load capability, large grid-to-rod contact areaagainst fretting wear, low pressure drop and minimized rod bow• IFM with enhanced mixing, fretting wear prevention and low pressure drop4. SummaryKorea has twenty nuclear power plants in operation and will have eight more nuclear power plants bythe year 2017. KNFC, as the sole nuclear fuel company, is designing, manufacturing and supplyingnuclear fuel for all operating nuclear power plants in Korea since 1989. KNFC initiated the “Mid- andLong-term Nuclear Fuel Development Programs” in 1999. Through these programs, KNFC hassuccessfully developed three types of advanced fuels that include PLUS7 TM fuel for the OPR1000s andthe APR1400s, the 16ACE7 TM fuel for 2-loop and 3-loop Westinghouse type plants and 17ACE7 TMfuel for the 3-loop Westinghouse type plants. To be more competitive than ever and to become one ofglobal fuel technology leaders, KNFC launched the “X-Gen Fuel Project” in the year 2005 and plansto finish it by the year 2015.References[1] “Development of Advanced Nuclear Fuel for KSNPs,” Final Report, May 2005, KNFC.[2] “Safety Evaluation Report for the PLUS7 TM LTAs to be loaded in Ulchin Unit 3,” KNF-AFD-02001, March 2002, KNFC.[3] “PLUS7 TM Fuel Design and Safety Evaluation Report for KSNPs,” KNF-TR-DMR-04001, June2004, KNFC.[4] “KCE-1 Critical Heat Flux Correlation,” KNF-TR-SGH-04001, October 2004, KNFC.[5] “Reload Transition Core Safety Analysis Report for Ulchin Units 3&4,” KNF-T-DMR-05001,January 2005, KHNP/KNFC/KOPEC.[6] “KAFD Final Design Review Package,” February 2002, KNFC/WEC.[7] “Development of Export-driving High Performance PWR Fuel,” R-2005-1-391, August 2005,KNFC.

Current Status and Development Plan on Fuel Cycle System ofFast Reactor Cycle Technology in JapanM. ITO, H. FUNASAKA, T. NAMEKAWAAdvanced Nuclear System Research and Development Directorate,Japan Atomic Energy Agency4-33 Muramatsu, Tokai-mura, Naka-gun, Ibaraki 319-1194, JapanABSTRACTThe FaCT project in Japan is implemented purposing to decide the adoption of innovativetechnologies by 2010 and to judge the prospect of the applicability of innovativetechnologies toward engineering-scale tests by 2015. Innovative technologies to bedeveloped are identified as six development issues for advanced aqueous reprocessingsystem and six ones for simplified pelletizing method fuel fabrication system. As foradvanced aqueous reprocessing technology, uranium crystallization and extractionchromatography are important to put the advanced reprocessing system into practical use.Wide range of development work from chemical fundamental study to engineering-scaleequipment operation are planned and practiced. As for simplified pelletizing method,development of the source powder preparation technology and in-cell remote handlingtechnology are important in particular, to put the low decontaminated TRU fuel fabricationinto practical use. We will develop and validate conversion and granulation process.Development of larger micro-wave heating conversion system of two kinds will be pursuedand one adequate system will be selected by 2010. We will develop modular equipments andrepairing system in a hot cell by 2015.1. IntroductionJapan Atomic Energy Agency (JAEA) launched the Fast Reactor Cycle Technology Development(FaCT) project in cooperation with the Japanese electric utilities. The FaCT project is based on theconclusion of the previous project, the Feasibility Study on Commercialized Fast Reactor CycleSystems (the FS) which was conducted for last seven years [1,2]. In the FaCT project, the combinationof the sodium-cooled fast reactor with oxide fuel, the advanced aqueous reprocessing and the simplifiedNew processConventional processMA MA recoveryrecoveryAm, CmHigh level liquidwasteU/Pu/NpConcentrationAdjusting PuPucontentcontentSpent fuelDisassembling/shearingshearingDissolutionClarificationU - - CrystallizationCo-extraction(U,Pu,NpCo-stripping(U,Pu,Np co-recovery)Solvent regenerationUpelletizing fuel fabrication was selected asthe main concept which should be developedprincipally because it was the mostpromising concept for commercialization.A conceptual design study of these conceptsand the R&D of innovative technologies tobe adopted in these concepts areimplemented purposing to decide theadoption of innovative technologies by 2010and to judge the prospect of the applicabilityof innovative technologies towardengineering-scale tests by 2015. In thispaper, current status and development planon fuel cycle system in Japan are presented.U/TRU (product)U (product)Fig.1 Process flow of advanced aqueousreprocessing

2. Development of reprocessing technologiesThe process flow of advanced aqueous reprocessing which was selected as the most promising is shownin Fig.1. Some "new" processes are adopted in advanced aqueous reprocessing such as uraniumcrystallization and extraction chromatography. Most of other process technologies already demonstratedand utilized in LWR fuel reprocessing are applied.Six development issues were identified for the advanced aqueous reprocessing. For these technologies,wide range of development works from chemical fundamental study to engineering-scale equipmentoperation are planned and/or practiced currently. For long term goal of commercializing FBR cycle,some short term objects should be settled, which are required to decide the adoption of technology by2010, to start hot engineering scale tests for adopted technologies by 2015 and to establish the concept ofcommercial scale process by the same time.Criteria for decision of technology adoption are (a) no safety problem of chemical reagents and othermaterials used in process technology, (b) promising achievement of function required for process suchdecontamination factor and recovery, (c) establish of fundamental operability, durability andmaintainability of process equipments. Based on specific feature of each process technology,engineering-scale cold and uranium tests, and fundamental-scale uranium, RI and hot tests are plannedto provide with enough technical data for evaluation in 2010.The test facility for hot engineering experiments is now in fundamental design phase. Development taskfor each process technology is to be proceeded in accordance with facility design, and also is to beplanned to address technical difficulties with scaling up to commercial size.Current status and development plan of six issues to be developed are presented below.(1) Disassembling and shearingAs for a fuel disassembling process of wrapping tube, we have been developing the fuel disassemblingsystem using with CO 2 laser or YAG laser. Cold mock-up experiments with YAG laser system in whichthe laser beam was transferred to the optical fiber result good cutting performance. However, the lasercutting method has possibility to damage to fuel pins. We will pursue developing the mechanical cuttingmethod with a grinding wheel.As for shearing, fuel pins will be sheared shorter than ordinary length of 3 cm with an existing shearingmachine in order to make the ratio of powder more than half of fragmented fuel after shearing. Controlmethod needs to be established to have sheared fuel pins of a constant length.(2) Fuel dissolutionThe highly efficient dissolution technique has been developed for preparing a highly concentrateddissolver solution which will be fed to the crystallization process. Lab-scale dissolution tests using anirradiated MOX fuel resulted that dissolution of powdered fuel was effective for a dissolver solution ofhigh concentration. We will optimize the dissolution process condition through tests which examines inpowder ratio and powder size distribution in order to efficiently obtain a dissolver solution of a specifiedconcentration. Furthermore, stable operation of rotary-drum-type continuous dissolver, e.g. control toavoid bumping of the solution, will be developed by mock-up tests.(3) Uranium Crystallization70% of uranium (U) contained in a dissolved solution is recovered as uranyl nitrate hexahydrate (UNH)crystal by lowering the temperature of the solution. Some experimental results were acquired by using Uand plutonium (Pu) mixed nitrate solutions, showing that behavior of Pu would depend on its valence.Pu of tetravalent remained in the solution and easily separated from UNH crystal. Pu of hexavalentwould be precipitated and contained in uranium crystal. This suggests that some Pu of hexavalent mighttake place of U in crystal. It can be concluded that adjustment of Pu valence in a feed solution isimportant to get a decontaminated (less Pu) UNH crystal. As for FP elements behavior, someexperiments indicated interesting results on some FP elements, especially on Cs. It was observed that Csmight form double salt of nitrate with Pu and precipitate as green crystal among yellow UNH crystal.More detailed survey is underway to identify the structure and composition of the green crystal.An operation test of semi-engineering scale crystallizer showing that a device of rotary-kiln-type isfeasible for this process. Further experiments with wider range of operation parameters are planned.

(4) Single cycle co-extraction of U, Pu and NpNp can be co-recovered with U and Pu in the simplified PUREX process because Np has extractablestates of tetra and hexavalent. Lab-scale extraction tests showed that the simplified PUREX flowsheetwith adjusting Np valence to Np(VI) was effective in co-recovery of Np with U and Pu. We will pursueoptimizing the flowsheet through process tests to obtain the extraction profile of Np and to improve anevaluation code.As for centrifugal contactor, solution flow profile and separation performance were comprehendedthrough engineering-scale U tests. We will pursue establishing a durable and well-controlled system ofcontactors.(5) MA recovery by extraction chromatographyPrior to extraction chromatography, solvent extraction process called as Solvent Extraction for Trivalentf-elements Intra-group separation in CMPO-complexant System (SETFICS) for recovery of Am and Cmwas developed utilizing CMPO for An(III)+Ln extraction and DTPA for An(III)/Ln separation.Although An(III) elements were successfully recovered by the SETFICS process, some drawbacks suchas use of an inorganic salting out reagent had to be overcome to draw the commercially available flowsheet. We will pursue selecting most suitable extractant for Am and Cm recovery and optimizing theprocess condition through lab-scale tests.The extraction chromatography technique uses a silica-based support in which various extractants canbe immobilized. This process was evaluated to be superior to the SETFICS in the respect of the amountof secondary wastes and economical load. Separation experiments using CMPO embedded supportshowed an excellent ability, and some extractants were proposed for the An(III)/Ln separation process.The leading tasks of the extraction chromatography are completion of the flowsheet, design study,manufacturing and operation of the engineering-scale apparatus with a safety for heat and gasgenerations. R&D studies for those subjects have already started. We will pursue evaluation forengineering safety and durability of the extraction column considering decomposition of extractant byradiation damage.(6) Waste treatmentSalt-free technology for solvent washing process was developed through fundamental tests using asimulated degraded solvent. We will pursue to expand the database of washing efficiencies throughfundamental hot tests and to develop equipment for decomposing treatment test of metal-free chemicals.Furthermore, we will pursue developing equipments for decomposition of extra nitrate and highlyefficient concentrator for dividing waste solution to two categories of high level and very low level.2. Fuel Fabrication technologiesU/TRU solution withadjusted Pu contentMicrowavedenitrationGranulationCalcinationReductionMOX powderPelletizingFuel elementmanufacturingSinteringFuel element inspectionO/M ratio adjustmentGrindingFuel elementPellet inspectionFuel assembling/inspectionPelletFuel assembly2.1 Feature of the advanced fabricationprocess and facilityThe MOX fabrication plant on the FaCT projecthas production ability of 200 tons of heavy metalsper year [2]. The simplified pelletizing method isadopted for this plant in order to rationalize theconventional high decontaminated MOX pelletfuel fabrication process.In this method, the Pu and minor actinides (MA)adding process is performed at the stage of thenitrate solution, and the mixture is converted tosource MOX powder with good homogeneity bymicrowave co-conversion. The source MOXpowder is granulated by a tumbling granulationmethod to improve powder flowability andcompressibility. The die cavities of the pressinghead are filled to the top with the powder, which isthen pressed into green pellets. In this process, weFig.2. Process flow of simplified pelletizingmethod

use a die lubrication method in which the lubricant is sprayed directly onto the surface of the die beforepressing. The lubricant content in a green pellet is very small, and the pre-sintering process can thus beeliminated. The green pellet is sintered to create a pellet with high density, and the outer surface of thepellet is ground to adjust its outer diameter if necessary. Then, the oxygen to metal (O/M) ratio of thepellet is reduced by heating in a hydrogen and argon gas mixture. Some pellets are placed in a claddingtube made of oxide dispersion strengthened (ODS) ferritic steel, the end plug is welded using theresistance welding method, and the produced fuel pin is obtained after various inspections. The specifiednumber of fuel pins are assembled, and the product fuel assembly is inspected.As for a feature of the plant, the main process equipments are installed in shielding cells because of highradio-activity of the low decontaminated TRU fuel.In order to put these process technologies into practical use, six development issues are identified as thefollowing.2.2 Current situation and development plan of each issue(1) Unified technology of conversion and granulation technologyAs for plutonium content adjusting technology, it is necessary to develop an advanced technology fortransferring a constant volume of Pu nitrate solution since the process involves adjusting the Pu contentto its target value at the solution mixing stage of Pu nitrate and U nitrate, with high accuracy from the Punitrate buffer vessel. A new liquid solution transfer system consisted of a volumetric feeder tank and anair-lift pump was developed and examined by cold mock-up experiment on approx. 1/2 scale of theactual equipment. Our results indicate that the adjustment accuracy was 30wt%Pu±1.8%, which wassatisfied with a target value of 30wt%Pu±2.5% within the range of normal operational conditions.As for conversion technology, the dish-shaped denitration container has been developed in thetraditional conversion process which maximum production ability of 2kg-MOX/buch by now. Adevelopment issue is expansion of the container size and micro-wave heating control for the largecontainer. A new concept of the hopper-shaped denitration container with the production ability of5kg-MOX/buch was designed in the FS. U test were carried out using with the test equipment in whichhopper-shaped container size was approximately 1/50 scale of the actual device. The physical propertiesof the obtained UO 2 powder were measured and it was confirmed that it has the same properties as thepowder produced by traditional equipment. However, it was noted that the solution boiled over from thecontainer during micro wave heating. This is important for development tasks because it is a relativelydifficult problem to solve and may be have been caused by the shape of the container. At the FaCTproject, we will pursue study and tests for both type containers in order to choose the adequate containertype by 2010.As for granulation technology, lab-scale hot tests to improve the flowability were carried out usingtumbling granulation of MOX powder. We examined the powder filling rate into a die cavity (8 mm indiameter, 20 mm in height ) of a pressing machine using the MOX powders and non-nuclear powderswith various Carr’s flowability indices[3]. We found that granulated powder with a Carr’s index of 60or more filled the die smoothly with a high filling rate. Through another lab-scale hot test, we found thatthe new method of granulationin which tumbling granulation process was applied concurrently withdenitration process had high technical feasibility.At the FaCT project, we will pursue a small scale engineering experiment for tumbling granulation usingMOX in order to develop the new method of granulation and to validate conversion and granulationprocesses.(2) Die wall lubrication pelletizing technologyThe granulated fuel source is filled up to die cavities at a die set on the pressing machine and pressed intogreen pellets. The pressing machine is a reciprocating press system with 12 punch-heads at the actualequipment and a die wall lubrication system which can spray a minimal amount of lubricant on the diewall. A die wall lubrication system was developed for a reciprocating type pressing machine with 6 dies.

Particles of the lubricant are sprayed with pressurized air from each hole on the lower part of the diecavity wall. Pelletizing tests with the die wall lubrication system using molybdenum powder werecarried out in order to evaluate the lubrication performance of the die wall lubrication system. Ourresults showed that the pellets made using the present die wall lubrication system had no defects in theirappearance such as cracks or laminations. The appearance and density of the sintered pellet made fromgranules of MOX particles made by the tumbling granulation method was excellent.At the FaCT project, we will pursue a small scale engineering experiment for die wall lubricationpelletizing using MOX in order to validate the technology.(3) Sintering and O/M ratio adjustment technologyLow decontaminated MOX fuel contains an MA of 5 wt% at maximum and small amount of FPs.Furthermore, its O/M ratio is lower than that of conventional MOX pellets. It is important to investigatephase stability and sintering behavior in fabricating such a new type of fuel. Pellets of MOX containingneptunium (Np), Am and neodymium (Nd) were prepared under various conditions, and themicrostructure and phase state of the pellets were investigated. Nd was added for the purpose ofsimulating the FP, taking into account that the FPs in low decontaminated fuel is composed primarily ofrare earth elements. In the pellets sintered in an atmosphere of high oxygen potential, all elements werefound to be uniformly distributed over the cross-section. The pellet sintered in an atmosphere of lowoxygen potential is inhomogeneous and has some spots composed of raw materials. This shows thathigher oxygen potential under sintering brings about good sintering performance. However, suchsintering in an atmosphere of high oxygen potential can not produce the specified O/M ratio of less than1.97. We adopted a two-stages heat treatment, that is a combination of sintering in an atmosphere of highoxygen potential and annealing to control the O/M ratio.The pellets were fabricated with various O/M ratios between 1.92 and 2.00 and were analyzed byEPMA. Precipitates of Pu-Am-Nd oxide were observed. The precipitates were observed only in thesample with low O/M ratio (less than 1.96). The precipitates containing concentrated Pu are expected togenerate heat locally, for example at the locations of Pu. However, it is also expected that the influenceof the precipitates on fuel behavior under irradiation is small, because the size of the precipitates is small(less than 10μm).Samples of 3% and 5% Am containing MOX (Am-MOX) were fabricated and examined in order toinvestigate the effects of a higher MA content in MOX. The composition of Am-MOX with 30wt%PuO 2 was adjusted to 3wt% and 5wt% AmO 2 content by mixing the raw oxide powders containingAmO 2 and UO 2 powder. The density of the sintered pellet and its O/M ratio were 93% TD and 1.95,respectively.(4) Studies of fuel physical propertiesAs described in section (3), it is important to investigate physical properties of such fuel which areessential to design the fuel performance and the fuel fabrication process. Various physical propertiessuch as melting point, thermal conductivity, lattice parameter, oxygen potential, oxidative rate, phasediagram etc. of Am and Np bearing MOX.We will expand these databases and develop the computer calculation methods for physical propertiesestimation through modeling of the databases.(5) In-cell remote handling technologyAt the MOX fabrication plant on the FaCT project, the main process equipments are required graterremote operability and far grater remote maintenability comparing with the equipments in glove boxeson the traditional MOX fabrication plant, because these equipments are installed in hot cells. Especially,it is essential to establish repairing system in the hot cell, because the equipments consist of precisionmachines and operators can not maintain these machines directly. We proposed the concept for repairingand maintenance system consisted of three stages; (a) replace of out ordered module in the main processcell, (b) decontamination and roughly disassembling of the module in the maintenance cell and (c)refurbishment of the module using the globe box in the maintenance room.Therefore, development of the modular equipment and the remote handling device for it has stated andcomplete by 2015.(6) TRU fuel handling technology

At the MOX fabrication plant on the FaCT project, the main process equipments are required measuresto reduce undesirable effects of heat generation of the fuel caused by MA containing. At designing theplant, the measure will be chosen and combined adequately among some methods such as todeconcentrate source fuel, to improve function of standing to cool, to prevent oxidation by surroundinginert gas, and forced cooling operation.At the fuel pin bundle assembling equipment, development of adequate method to remove the heat isessential and has been started because it is impossible to deconcentrate the heat generation of 2.6 kW perone subassembly at maximum. The fuel pins are handled horizontally and are assembled up to a bundlein this equipment as same as an equipment for a conventional MOX fuel assembly. Cooling down isperformed by air spraying at right angle to the bare pin bundle which flow channel geometry iscomplicated by lots number of pins and wrapping wire.3. Development scheduleDevelopment schedule on each issues are shown in table 1 for advanced aqueous reprocessingtechnology and table 2 for simplified pelletizing method fuel fabrication technology, respectively.JFY 2010 2015 2020 2025 2030R&D on eachprocess(CPF,EDF)Disassembling,DecladdingDissolution,ClarificationCrystallizationU-Pu-Np co-extraction,co-strippingMock-up exp. for evaluating abilityand proceduresImproving calc. code for dissolution,Evaluating undissolved residueDesign forintegratedU tests for evaluating the ability of dissolverEvaluation of scrubbing methods forcrystalEffect of several process parametersEvaluation of DF withcentrifugal contactorEndurance test,Procedures for abnormalDeveloping calc.codeMock-up tests for evaluating ability,Design for equipmentImproving calculation codeColdUHotMA recovery(Extr. Chromatography)Preparingequipment for testEffect of several processparametersMock-up exp. for evaluatingabilityR&D on totalsystem(EDF,RETF)Solt-free processfor waste reductionEngineering scaletests (110Kg/h)Mock-up tests for evaluating ability,Design for equipmentEvaluation of washing solvent anddegraion propertyEffect of several process parametersPreparing for enginering scale (1~10Kg/h)exp.UtestsEngineering scale tests(1~10Kg/h)Table 1: Development schedule on advanced aqueous reprocessing

JFY 2010 2015 2020 2025Mock-up exp. for massproduction micro-wave heatingMock-up exp. for remote maintenanceconversion equip.Unified technology ofconversion andgranulationPreparingequip. for testsS mall-s caleeng. exp. forgranulation(MOX)Validation of conversion and granulationColdUS mall-scale eng.exp. for massproducation granulation equip.Mock-up exp. for mass production equipgranulationHotDie wall lubricationpelletizing technologyPreparingequip. for testsS mall-s caleeng. exp. forpelletizingMock-up eng. exp. for remote maintenanceof die wall lubrication pelletizing equip.R&D on eachdevelopmentIssueSintering and O/M ratioadjustment technologyPreparingequip. for testsS mall-s cal eng.exp. forsintering andO/M ratioadjustmentMock-up exp. for mass productionsintering and O/M ratioadjustment equip.Validation of s intering and O/M ratioadjustmentMock-up exp. for remote maintenance andsintering and O/M ratio adjustment equip.Studies of fuel physicalpropertiesProperty measurement(melting point, thermal conductivity, state diag ram, diffusion coefficient)Development of calc. code for property estimationMock-up exp. for modularizedpelletizing equip.In-cell remote handingequipment developmentDesign studyMock-up exp.(pellet grinding , fuel element processingi )Mock-up exp. for pelletins pection andpowder property analyzerSystem design forothers analytical equip.TRU fuel handlingtechnologyMoc k-up exp. (heat removalsystem for fuel assembling)R&D on totalsystem (EDF)Engineering scale testsPreparing for eng. s cale demos traion facilityUtes tEng. scaletes tsTable 2: Development schedule on simplified pelletizing method4. References1. K. SATO, et al., “Conceptual Design Study and Evaluation of Advanced Reprocessing Plants inthe Feasibility Study on Commercialized FR Cycle Systems in Japan”, Global 2005, No.502,Tsukuba, Japan (2005).2. T. NAMEKAWA, et al., “Conceptual Design Study and Evaluation of Advanced Fuel FabricationSystems in the Feasibility Study on Commercialized FR Fuel Cycle in Japan”, Global 2005,No.424, Tsukuba, Japan (2005).3. Carr R. L., Chem. Eng., 72 (1965) 163.

PBMR HTR FUEL PRODUCTION FACILITY PROJECTSOUTH AFRICADR. GEORG BRAEHLER, KLAUS BUETTNER,WOLFRAM KRESS, KARL FROSCHAUERInternational OperationsNUKEM Technologies GmbHIndustriestrasse 13, 63755 Alzenau, GermanyABSTRACTSouth Africa’s Pebble Bed Modular Reactor (Pty) Ltd company, (PBMR), was establishedwith the intention to develop and market high-temperature reactors both locally andinternationally.Currently the Detailed Design for a pebble bed modular reactor is under preparation, whichis based on the German experience and namely on the HTR Module concept.Accordingly the PBMR fuel is based on the proven high quality German fuel designconsisting of ~10% enriched uranium triple-coated isotropic (LEU-TRISO) particlescontained in a moulded graphite sphere. A coated particle comprises of a kernel of uraniumdioxide surrounded by four ceramic layers.Following a number of studies and design activities, NUKEM Technologies GmbH wascontracted by Pebble Bed Modular Reactor (Pty) Ltd (PBMR), to prepare the detail designincluding installation supervision and cold commissioning for the process parts of the PilotFuel Plant (PFP).Currently the design documentation to support the preparation of the Safety Report hasbeen prepared, requests for equipment offers have been sent out, and bids received areunder evaluation.1. IntroductionDue to the continuous increasing world wide demand on energy, which is increasing dramatically,especially in the emerging nations and the third world, nuclear power is again being taken more andmore under serious consideration. Also in connection with the international interest and efforts inreducing the CO 2 emission, nuclear power as the cleanest energy source with regard to climaticinfluence has become of special interest again world wide and is somehow world wide experiencing arevival. Here the HTR Technology is especially considered due to its unique safety features, based onthe modular design and the relatively small reactor core. The high temperature level opens theopportunity to produce Hydrogen and to substitute fossil fuels for process heat generation underavoidance of CO 2 emission.The development of the Modular HTR Technology in Germany started after the Harrisburg accident.Besides the modular reactor design and the small dimensioned reactor core design itself, the majorsafety features of the HTR Technology are based on the fuel element as such. The development of theHTR fuel element in Germany has been systematically performed by NUKEM during this time.Nowadays this technology is again being specially considered and new activities are being undertakenin the further development of this technology in numerous countries, especially in the PBMR Projectin South Africa.

2. The PBMR Project2.1 OverviewNuclear power is not new to South Africa. Besides the operation of two Pressurized Water ReactorUnits at the Nuclear Power Station Koeberg since the mid 1980’s, the South African Atomic EnergyCorporation (today known as Necsa (South African Nuclear Energy Corporation)), was following acomprehensive nuclear program and further nuclear facilities existed at the Pelindaba site close toJohannesburg. Key personnel from these early nuclear activities later focused on the Germanexperience with a small reactor, known as the Pebble Bed Modular Reactor and its “billiard ball-like”Fuel Pebbles.The South African Pebble Bed Modular Reactor company, Pebble Bed Modular Reactor (Pty) Ltd(PBMR), was then established in 1999 with the intention of developing and marketing hightemperaturereactors, both locally and internationally.PBMR’s current investors are the South African Government, Eskom, Westinghouse and the IndustrialDevelopment Corporation (IDC). Eskom intends to phase out its shareholdings in PBMR in order tobecome a client of the technology, rather than a developer of it.Today, PBMR enjoys solid support from the South African Government, who regards it as one of themost important capital investment and development projects yet undertaken in the country.The PBMR Project itself is separated in two main projects (see Fig. 1).Koeberg:Pebble Bed Modular ReactorCommercial Fuel PlantPilot Fuel PlantFig. 1: Sites of the PBMR Project• The design and construction of the prototype Modular Reactor (165 MW electrical (400 MWt))to be built at the Koeberg site close to Cape Town and• The design and construction of the prototype Fuel Plant for the production of 270,000 fuelspheres per year to be built at the Pelindaba site about 100 km north-west of Johannesburg.

Since 2000, NUKEM has been deeply involved in the design of the Pilot Fuel Plant, and after thecompletion of a Detailed Feasibility Study and the Basic Design, is currently engaged with the DetailDesign and the support of the procurement activities for the Pilot Fuel Plant.2.2 Time ScheduleThe first criticality of the pebble bed modular reactor is planned for 2013. In order to achieve thismilestone the following time schedule for the Pilot Fuel Plant in Pelindaba has been established.The basic design was finished at the end of 2005, followed by the detail design phase, which will endat the beginning of 2008 and includes the necessary procurement and manufacturing of the hardware.PBMR has applied for the license for the construction, installation and commissioning of the PilotFuel Plant in September 2006. It is expected that the license will be received in November 2007 inorder to commence installation activities. The installation of the equipment and utilities should befinished by mid-2009, and the follow-on cold commissioning phase will end in mid-2010.After hot commissioning the production of the necessary 430 000 fuel spheres will start to provide thefirst core for the pebble bed modular reactor (see Fig. 2).PBMR ReactorConstructionCore LoadingPFP PlantManufacturingInstallationCommissioningProduction2008 2009 2010 2011 2012 2013 2014Fig. 2: Time schedule2.3 Fuel EquivalenceThe current PBMR reactor design requires a fuel design and quality which is equivalent to the Germanhigh quality fuel, as it was developed and produced in the NUKEM facilities in the 1980’s.Accordingly, all processes as well as the key equipment are identical to the processes executed in theNUKEM facility, however the handling of products and the auxiliary processes are upgraded inaccordance with the state of the art.3. The Fuel Sphere Production Facility at PelindabaThe PBMR fuel is based on a proven, high quality German fuel design consisting of ~10% enricheduranium triple-coated isotropic (LEU-TRISO) particles contained in a moulded graphite sphere. Acoated particle comprises a kernel of uranium dioxide surrounded by four ceramic layers.

A fuel sphere consists of 9 g of uranium (some 15 000 particles) and has a diameter of 60 mm; thetotal mass of a fuel sphere is 210 g (see Fig. 3).Fig. 3: Structure of a HTR Fuel SphereDuring normal operation the PBMR core contains a load of 456 000 fuel spheres. Fig. 4 gives animpression on a fuel core of a HTR (in fact it is the former German THTR).Fig. 4: First Core Loading of the THTRThe Pelindaba Fuel plant is designed to produce 270 000 fuel spheres a year. The pilot productionplant is separated into 6 areas (see Fig. 5):• Area K: Kernel production facility• Area C: Coater facility• Area M: Matrix production facility• Area F: Fuel sphere production facility• Area E: Effluent treatment facility• Area R: Uranium recovery facility

Solid WasteUraniumRecoveryU 3 O 8KernelCastingCoatingFuel SphereProductionFuel SpheresWaste WaterEffluentTreatmentMatrixMaterialGraphiteFig. 5: Fabrication processes3.1 The Processes3.1.1 Kernel ProductionIn order to achieve good in-pile behaviour and to allow the subsequent production steps, the UO2Kernels need to be of nearly theoretical density, of the required diameter and of a excellent roundness.Such Kernels are produced by casting droplets of a viscous Uranyl Nitrate solution, made up fromUranyl Nitrate solution and special chemicals (Polyvinyl Alcohol PVA, Tetrahydrofurfuryl AlcoholTHFA) into a Ammonia containing aqueous solution. The droplets form micro spheres which aregelled, washed, dried calcined to UO 3 and reduced and sintered to UO 2 Kernels. The finished Kernelsundergo special mechanical processes; sieving, sorting on vibrating tables, sampling and portioning, todeliver the right quality to the next process step. Fig. 4 shows the Kernels in different states.3.1.2 CoatingWithin the Coating facility the Kernels receive four coatings using a chemical vapour deposition(CVD) furnace to produce the Coated Particles. The first layer deposited on the kernels is porousCarbon, it allows fission products to collect as well as to accommodate any geometrical deformationof the Uranium Dioxide Kernel This is followed by a thin coating of pyrolytic Carbon (a very denseform of heat-treated Carbon), which is a first barrier for fission products, and forms the basis for thethird layer. The third layer made of Silicon Carbide serves as the main barrier for retention of fissionproducts even at temperatures far above the operational limit. Another layer of pyrolytic carboncompletes the coating is again a barrier, and as a mechanical protection it allows the followingproduction step. The coated particles undergo the same quality control like the Kernels, and inaddition the very important leak test. The fuel specification allows only one defective particle in a FuelSphere (1 particle in 15 000).

3.1.3 Matrix Material ProductionThe matrix graphite powder makes the quality of the Fuel Sphere, which has to resist numerousmechanical, thermal and radiation in pile impacts. It is generated in the matrix production facility bymixing natural and electro-Graphite powder with resins, followed by milling to a fine powder. Thisrelatively complex process is necessary to achieve the required homogeneous distribution of allcomponents in the final powder.3.1.4 Fuel Sphere ProductionIn the fuel sphere production facility the Coated Particles are overcoated with a layer of matrixgraphite powder (MGP). Again quality assurance steps are applied as with the Kernel and CoatedParticle. The overcoated particles are dosed into matrix graphite powder and pressed to the core of afuel sphere. Then an additional 5 mm layer of matrix graphite material is added to form a “non-fuel”zone. The resulting Fuel Sphere achieves its final diameter by a machining process and is thencarbonised and annealed at 2000 °C. The finished Fuel Spheres are subject to numerous QualityControl steps, amongst which are 100° % X-ray for the control of the centricity of the fuel core andthe check of the freeness of Uranium of the shell, drop tests, etc. Also very important is the check offree Uranium in the Fuel Sphere, which is a result of SiC shell breakage during fabrication (see Fig. 6)shows a couple of finished Fuel Spheres.Fig. 6: HTR Fuel Spheres3.1.5 Effluent Treatment and Uranium RecoveryThe effluents from the production processes are treated in the Effluent Treatment Facility. The mainpurpose is to recycle process liquids like IPA, THFA and Ammonium Hydroxide solution, for the reusein the Kernel facility. The resulting aqueous liquids are decontaminated prior to release. The wasteorganic liquids and solids are thermally oxidised, and Uranium residues are fed to internal recycling.The scrap material from the production process such as odd Kernels, odd Coated Particles and off-specFuel Spheres as well as other Uranium-bearing materials are recycled in the Uranium RecoveryFacility to form recycled U 3 O 8 , which is reused in the Kernel production process. As a resultradioactive and chemical emissions from the plant are extremely low.3.2 Nuclear Safety

Special attention is given to criticality safety. Different to previous plants, practically all equipment isdesigned in safe geometry, with exceptions where absolutely necessary due to the proven formerprocesses and the requirement of equivalence. The calcining furnaces, the coaters and the over-coatersare designed in accordance with the safe mass principle, but with numerous additional “defence-indepth”measures to exclude a criticality accident.To prevent the spreading of radioactivity into the room atmosphere, all processes with the risk ofrelease of Uranium-bearing dust are installed within containment enclosures, equipped with glovesand air filtration.A independent company has been engaged to check the safety of the design and the compatibility withthe South African licensing requirements.3.3 Plant DesignThe PFP has semi-continuous and batch processes.The Kernel area is designed to operate in a semi-continuous mode, with a batch size of 15 kg U and aweekly capacity of 75 kg U. The lot size of the produces Kernels is 100 kg UO 2 , which are equallydistributed into 20 portions of 5 kg UO 2 each.The coater and the overcoater are operated with the restriction of safe mass in batch sizes of 5 kg UO 2 .The presses require 40 sec per Fuel Sphere, the furnaces and the other equipment are designed for adaily throughput of 2000 Fuel Spheres.The Effluent treatment Facility is designed to operate in continuous mode, allowing the reworking ofall effluents in parallel to the operation of the main customer, the Kernel facility.The Uranium Recovery Facility is designed to rework the scrap within safe geometry equipment insidecontainment enclosures. It is batch operated with a batch size of 5 kg Uranium and a capacity of 5batches per week.The so called Uranic part of the facility, covering kernel production, coating, Uranium recovery,effluent treatment has to cope with the contamination risk from handling free Uranium. Theradiological protection downstream from that processes in the Graphitic part is easier due to the factthat the coated particles are free of contamination.To prevent spread of contamination, both parts of the facility are separated with different access andventilation systems.The whole facility has been broken down into roughly 200 work inquiry packages, typical packagesrepresent furnaces, reaction vessels, storage tanks, containment enclosures, mechanical equipment. Allinquiry packages were defined by a standardised set of documentation, with descriptions, flow sheets,lay outs, data sheets and functional specifications as core documents. They ask for offers coveringsupply of documentation, hardware, FAT, shipment, installation at site, support duringcommissioning.Those packages have been sent to potential suppliers, mainly in South Africa and in Germany, butsome of them worldwide.

4. NUKEM ContributionPBMR has started its HTR activities with the installation of lab scale facilities for investigations andpreparation of Kernels, Coated Particles, Graphite material and Fuel Spheres. Experts from formerNUKEM facilities were and are supporting these activities.Concerning engineering activities for the NUKEM has been involved in the PBMR PFP project fromthe very beginning and as one of the first steps has prepared the Detailed Feasibility Study for thewhole Fuel plant in 2000/2001.This activity was followed by the Basic Engineering, where NUKEM designed the processes forproducing the German-equivalent Fuel element to the latest up-to-date technology and safetyregulations. The Basic Engineering phase ended with the intensive Hazardous And Operability Study(HAZOP) proving the robustness of the design in respect to Safety and Operability.Since August 2005, NUKEM is carrying out the Detail Engineering, mainly to support the preparationof the Safety Report and the licensing procedure and is supporting the procurement of equipment. Theprepared by NUKEM.Currently the procurement activities, after all Enquiry Requisitions necessary for tendering have beensent out to potential suppliers, concentrate on the adjudication of incoming tenders, on clarification ofopen questions, and the preparation of recommendations for order placement.Once orders have been placed, the check of the MDD (manufacturer design documentation) willfollow.In the course of the above-mentioned design activities NUKEM has established an engineering teamof approximately 40 highly qualified and experienced engineers supported by a South Africanengineering subcontractor. This engineering team is permanently in consultation with the HTRexperts, who originally developed and produced the HTR Fuel in the 1980s, to ensure that the knowhowis transferred and that the equivalence requirements are met during the design.

Session 18.2.4Decommissioning

1. IntroductionNUCLEAR DECOMMISSIONING“THE WAY FORWARD”A CONTRACTORS VIEWPOINTGLENN ELLCOCK & TONY HANDLEYBritish Nuclear Group Project Services LtdABSTRACTThe objective of this paper is to try and take away some of the mystique behind thedecommissioning of Nuclear Facilities, to challenge the attitudes of the plant owners andidentify effective and reliable decommissioning methods. This will be done by exploringthe “Enablers” to a successful decommissioning project and by looking at how“Innovation” plays a key role in successful decommissioning projects and why we need toinnovate.The number of nuclear decommissioning sites and projects is increasing globally and there are numerous factorsthat can influence their successful outcome.This paper provides a Contractors perspective on how the outcome of a decommissioning project can be positivelyinfluenced by appropriate consideration of activities that need to be completed, how they can be performed mostefficiently and where innovative approaches to tools, techniques and applications can improve success. It providesexamples of how appropriate enablers and use of innovation has positively affected the outcome ofdecommissioning projects2. EnablersEnablers in the context of nuclear decommissioning are the actions taken by stakeholders to create the appropriateenvironment for the successful completion of the project. These include actions taken by the Customerorganisation and the Contractor3. Customer EnablersThere are a number of activities that the customer can address to ensure the success of a decommissioning project.A number of these may seem obvious but often they can lack the clarity of definition and understanding needed toensure success.Project ClarificationThe purpose of the project or task needs to be clearly understood by both Customer and Contractor, includingunambiguous definition of the desired end-state and expectations in terms of deliverables and project milestones.In addition it is important to clarify the level of detail required for supporting documents. Some projects aredriven by Regulatory or Legal requirements and it is vitally important that both the Customer and Contractor fullyunderstand what is actually required to complete the decommissioning project. Wrong or inappropriateinterpretation of the intent of these requirements can lead to overall project shortcomings and further ‘Actions’ toachieve the required end-state.

Transition PlanningTimely assembly of the ‘right’ customer team is essential. The transition from an operating plant todecommissioning site is something that takes considerable planning and the use of a range of specialisedengineering skills and operational knowledge. Some of these capabilities may not be immediately available in theexisting plant operations team and early recognition of the change in emphasis of the facility will aid the transition.The inclusion of specialist decommissioning engineers will help challenge and refocus existing operationalprocedures, safety arrangements and rules to enable effective decommissioning. Planning for the transition iscrucial and time should be allowed for training and site / work familiarisation of new team members, similarlyexisting operational staff may need training in decommissioning activities. It is important to realise that the fullyexperienced and capable Operations Manager does not necessarily equate to the ‘best’ decommissioning manager.Plan for successPlant operations often include routine, repetitive activities and decommissioning projects involve one-off shortlivedactivities. Having the right frame of mind and being prepared for change (in a managed and controlledenvironment) is essential to the success of decommissioning works.The project should focus on achieving desired outcomes. Potential obstacles should be reviewed andcontingencies built into project activities so that obstacles can be dealt with effectively. Design decommissioningprogrammes and activities to enable maximum flexibility for the Contractor to choose tooling and techniques forthe task. Provide latitude to vary approach dependent on the conditions of the work.Select appropriate tools for the taskUse or adapt low cost, reliable proprietary tooling where possible. Design tasks so that tools are provided forsingle specific tasks. It is often easier to use a number of tools for individual tasks rather than develop bespokemulti-functional tools for specific applications. Hard lessons are learned when things go wrong, tooling fails andwork is stopped: "At WAGR, (Windscale Advanced Gas cooled Reactor) during the campaign to remove theNeutron Shield there was a requirement to cut stainless steel tubes located within graphite bricks and then toremove both the cut section of tube and the corresponding brick. A complex tool was specifically developed to cutthe tube and then lift the tube and the corresponding brick into a waste basket. The complex tool was extremelyunreliable and had to be frequently repaired. The tool became radioactively contaminated and operators carryingout the repairs accumulated significant radiation dose. By replacing the special tool with two simple adaptationsof commercially available tools, one which cut the tube (an angle grinder) and the other which lifted the tube andthe brick (a ball-grab), the reliability of the operation was improved dramatically and the subsequent radiation doseto operators was reduced significantly."Initial Complex Tool DesignSimplified Ball Grab

Use an appropriate contract strategyUse a commercial strategy that aligns with the nature of the works to be carried out. Decommissioning projectsoften involve a large number of unknowns or the process for carrying out the works is still largely underdevelopment. There is no point implementing a fixed price contract if the scope is uncertain or likely to change –more effort will be expended on commercial negotiations than decommissioning implementation.Be safeThe safety requirements and hazards on an operational facility are often different to those of a decommissioningsite. A thorough review and challenge of the remaining inventory, hazards and conditions on the site should beconducted considering the often transient nature of decommissioning activities. Short-term risks may result inlong-term benefits. The Customer should challenge existing custom and use ALARP (As Low As ReasonablyPractical) and time at risk arguments when considering the benefits of the decommissioning activity in reducingthe hazard inventory.The initial team engaged on the WAGR (Windscale Advanced Gas cooled Reactor) Decommissioning Projectwere used to working under a 'fuelled ' or operational environment i.e. respecting the hazards associated with a fueland fuel debris inventory and providing appropriate Personal Protection Equipment (PPE) such as PVC/respiratorsetc. Once the fuel / debris hazard is removed or in facilities where it is known that fuel debris does not exist theplant operating and safety practices can be adjusted to reflect the change in conditions. At WAGR, operatorscontinued to prepare for work as though the plant was fuelled even when the fuel inventory was removed. Thisresulted in the use of PVC PPE suits and respirators for some decommissioning activities when it was unnecessaryto do so. This was more dangerous due to the increased conventional safety hazards introduced by the mode ofdress. It also increased the decommissioning operators time at the work face and therefore the radiation dose. Theincreased radiation dose and risk of conventional accident due to the use of unnecessary PPE was opposite to theintent. It also added hugely to the cost and time of the project.Support and help the Contractor to understand the site conditions, rules and regulations. Invest in quality trainingto ensure that Contractors fully understand site issues, permits and authorisations from the perspective that hissafety and his safety performance directly impact the Customer organisation. Ensure that appropriate supervisionand support is available to enable work to proceed. Be flexible in working approach to allow activities to becompleted.Customers should be very clear and explain what is required of Contractors personnel on site, what authorisations,site familiarity and safety inductions are required to conduct activities within the requirements determined bysafety and site licence regulations.4. Contractor / Supplier EnablersThe Contractor can also improve the chances of success by adequate preparation and understanding of thedecommissioning project. Early clarification of desired outcomes, end points and milestones will removeconfusion from critical implementation activities. Look to embed knowledgeable members of the formeroperations team within the decommissioning project as site knowledge is critical to efficient delivery. Carefulplanning and detailed method statements, including alternative approaches will help clarify understanding of tasksand reduce overall task duration and cost by removing opportunities for uncertainty.Plan for the unexpectedThe Contractor should also design activities within prescribed operating envelopes or ‘tram-lines’ rather thanprescriptive detailed descriptions of tasks. This will enable progress when obstacles occur. Designs and tasks thatconsider flexible tooling strategies will have more chance of success than approaches that consider only oneoption. A clear understanding of the facility being decommissioned, the type and quantities of materials andwastes encountered will influence the range of tooling considered and specified.Consider multiple work-fronts to enable obstacles to be resolved in one area whilst implementation progress isbeing made in others. Work processes that include hold, recycle and evaluation stages will encourage teams to regroupand deploy alternative approaches if work doesn’t progress as planned.

Use daily work planning and tool box talks to share knowledge, evaluate current activities and ensure scope andexpectations are fully understood within the team. Develop management structures and devolve appropriateaccountability to the working level to enable work to proceed within the agreed operational envelope without theneed for constant upward referral of detail decisions. Work targets should be visible and progress regularlyupdated.Deploy flexible multi-skilled resource and engage flexible working practices to ensure progress is not undulyrestrained by rigid working practices and attendance hours. Ensure alignment of work activities with Customersupervision and support availability.5. Work Delivery EnablersDon’t assumeThe Contractor needs to check details prior to implementation activities. Critical dimensions should be physicallymeasured and checked to ensure equipment and decommissioning systems brought onto site will fit properly withexisting structures. Original plant data, drawings, reports and inventories are good references but do not alwaysinclude operational history, changes to process and plant modifications. Site surveys will reveal detail changesand developments. Photographs of the current status will enable plant drawings and inventories to be verified.Photo-grammetry is an increasingly popular technique for assessing existing plant layout, particularly for areas ofrestricted man- access and the level of resolution should be determined to suit the type of information and level ofaccuracy required. This technique is an invaluable tool for the decommissioning engineer, providing data on plantcondition layout and inventory that might not be available from operational records. It has been used in a numberof decommissioning situations at Sellafield and other sites to provide detailed plant models of normallyinaccessible areas. Where this is not an option do the simple thing and “walk the site” this will often review thingsthat are not clear or have been omitted from the drawings. Recent examples of where this has proved invaluableare;“Site surveys of the ground adjacent to a fuel storage pond that is being decommissioned revealed a surface waterdrainage chamber not shown on the site drawings which could have collapsed due to applied ground pressure froma mobile crane being used to install decommissioning plant. Lifting plans were adapted to accommodate increasedreaches to enable the load to be successfully lifted.”“Safety swim rings fastened to handrails around the pond would have clashed with equipment being installed – aminor issue in itself, but it would have caused a delay to a lifting sequence that needed an appropriate weatherwindow to proceed.”Review before implementationReadiness reviews are an essential pre-requisite for any implementation activity, particularly plantdecommissioning. Checks need to address plant status, authorisations, availability of staff, detailed methodstatements and physical resources to ensure tasks can be completed.Designs should be simple and robust wherever possible. Agricultural simple equipment should always be used inpreference to complex intricate devices. Adaptation and re-use of systems and applications from other projectsand industries should be considered. Previous operating experience, reliability and availability data can be used toassess the effectiveness of an approach. Decommissioning engineers should also consider novel tools, techniquesand approaches if other simple solutions cannot be found.Adapt or re-use existing technologyOne Sellafield solid waste handling project spent considerable effort to develop complex tooling to removemagnox swarf blocking access for decommissioning, this was subsequently revised to a plant operator withappropriate PPE moving the material with a long reach tool; a simple activity. Other applications to remove solidwaste from nuclear power plant waste stores required remotely operated retrieval equipment. A proprietarybuilding plant manipulator was adapted with a bespoke shoulder joint to enable articulation into difficult accessareas; novel, re-use of related technologies.

6. InnovationThere are a wide range of descriptions of innovation and in the context of nuclear decommissioning the followingdefinitions embrace the intent:‘the introduction of something new and useful, a change that increases value to Customer or Producer,performance and growth through improvements in efficiency, productivity and quality’Within Project Services we are constantly striving to innovate by selecting tooling, solutions, process &procedures and the people to create innovative solutions for our customers.Tooling selection is critical to successful decommissioning applications. Provision of a range of alternative robusttools will improve the flexibility of decommissioning activities and reduce interruption to work activities.We have provided solutions that adapt technologies from related industries or previous applications. Sellafieldsludge handling projects have adapted waste water and sludge handling technologies for nuclear applications.Pump / retrieval selection has been heavily influenced by related industries. One example is that of harbour sand/silt clearing jet pump technology adapted to address difficult sludge retrieval projects at Sellafield.Process and procedures should be developed to suit the working environment and hazards that are present.Challenge the norm, critically review existing custom and practice, considering ALARP and time at risk argumentsto reduce programme and task complexity.Working practices, skills and capabilities are adapted to enable flexible working.3-D design modelling is often used to enable rapid assessment of potential solutions. It enables previous tools,techniques and systems to be quickly adapted to suit current requirements, learning from the experiences ofprevious work, drawing on the knowledge bank of ideas and existing models (tools in the toolbox).ConclusionsThe success of decommissioning projects relies on a number of factors that are influenced by the Customer andContractor. Project end points and drivers need to be clarified and understood, decommissioning teams need toengage skills that are appropriate for the tasks being completed. Decommissioning plans should be unambiguousand include flexibility to adapt for changes or unexpected obstacles. Safety arrangements should be appropriate tothe hazard inventory and current site conditions. Decommissioning tools should be simple and robust, avoidingcomplex special equipment.Decommissioning teams should be aware of site conditions, rules and procedures. Frequent reviews of workactivities will ensure safe operation. Implementation contracts should be designed to suit the activities andvariability of decommissioning works.Innovative use of existing and related technologies will often yield better results than complex solutions –challenge the norm.

DECOMMISSIONING OF NUCLEAR FUEL CYCLEFACILITIES IN THE IPEN-CNEN/SPP. E. O. LAINETTIChemistry and Environment Technology Centre - Institute for Nuclear and Energy Research – IPENBrazilian National Nuclear Energy Commission - CNENAv. Prof. Lineu Prestes, nº 2242, C. Universitária, ButantãSão Paulo City, SP State – Brazil, CEP 05508-090ABSTRACTIPEN has been facing the problem of the dismantling and decommissioning of theirNuclear Fuel Cycle old facilities. Those facilities already played their roles of technologicaldevelopment and personnel's training, with transfer of the technology for institutionsentrusted of the “scale up" of the units. Most of the pilot plants interrupted the activitiesmore than ten years ago, due to the lack of resources for the continuity of the researches.The appropriate facilities maintenance has been also harmed by the lack of resources, withevident signs of deterioration in structures and equipments. The existence of these facilitiesalso implicates in the need of constant surveillance, representing additional obligations,costs and problems. The decommissioning strategy for the old facilities dedicated to thetechnological domain of the Nuclear Fuel Cycle follows an approach of advancinggradually in dismantling, since the resources and technical conditions are available. Thereasons of such approach are the need of political decisions related to the destiny of thefacilities, lack of financial resources and of specialized personnel in the decommissioningissue. As some facilities had the activities suspended for about twelve to fifteen years, alsoconstitute relevant problems the equipment deterioration and personnel's loss, dueretirements and transfers for other activities and difficulties related to the availability ofoperational reports, drawings and descriptive memorials. It should be emphasized that oneof the most concerning aspects, with relationship to the future of the facilities and thepostponement of the dismantling, is the loss of the experience accumulated by thepersonnel that set up and operated the referred units. A fundamental aspect of thedismantling process of a disabled nuclear facility is the removal of the retained material ofthe process and the final disposition of the radioactive wastes generated during theoperations. The reduction of the amount/volume of radioactive waste is of vital importance.Some facilities demand special attention, requiring preliminary operations of treatment ofretained materials and/or wastes. Besides the technological development activities relatedto the subject accomplished in the IPEN, this document also presents a brief report aboutthe D & D activities performed in IPEN since 2002. The knowledge obtained thanks to theresearch project BRA-12800 supported by the IAEA was extremely useful in the decisionsand activities regarding the D & D activities.1. IntroductionThe Brazilian National Nuclear Energy Commission (CNEN) is a federal autarchy, reporting to theMinistry of Science and Technology. Accordance to the 1988 Constitution it is for the monopoly ofthe mining of radioactive products, the production and commerce of nuclear materials as well for theorientation, planning, supervision and control of Brazil’s nuclear programs. CNEN, as a superioragency of planning, orientation, supervision and inspection, is the body entitled to establish standardsand regulations on radiological protection, to issue licenses (permissions) and to survey and controlthe nuclear activities in Brazil. CNEN also develops research and development related to the use ofnuclear techniques in benefit of the society. The CNEN is divided into three directorates: Directorateof Radiation Protection and Nuclear Safety - DRS, Directorate of Logistic Support – DAL andDirectorate of Research and Development – DPD. The DPD is further subdivided into five scientificand technological institutes. The Center for Development of Nuclear Technology - CDTN, which was

created in 1952 in Belo Horizonte city of Minas Gerais State, as Brazil’s first nuclear researchinstitute. The Nuclear Engineering Institute - IEN and the Radiation Protection and Dosimetry Institute- IRD, both in the Rio de Janeiro city of Rio de Janeiro State. The Regional Center of NuclearSciences – CRCN in Recife city of Pernambuco State, the newest of all centers The Institute forEnergy and Nuclear Research – IPEN, the biggest of the CNEN’s research institutes, located in SãoPaulo city.1.1 Brazilian nuclear activitiesTo understand better the facilities dismantling & decommissioning problem at IPEN it is important todescribe the scenery of the nuclear energy in Brazil. Brazil has modest fossil energy resource and oneof the largest hydroelectric potential in the world. Nowadays, Brazil only has two nuclear power plantsin operation: Angra-I with 657 Mwe (gross electric power) and in commercial operation since January1985; and Angra-II with 1345 Mwe (gross electric power) in commercial operation since January 2001(1). Both are located in the Angra dos Reis County – Rio de Janeiro State, near the cities of Sao Paulo,Rio de Janeiro and Belo Horizonte. Still in the early steps of its construction, Angra-III depends ongovernmental decision for its conclusion (it is suspended by now). This represents only about 2% ofthe total Brazilian electric installed generation capacity of about 94.7 GWe (2003).Besides the nuclear power plants, Brazil has established a nuclear power utility / engineering company– Eletrobras Termonuclear S.A. (Eletronuclear); a heavy components manufacturer – Nuclebras HeavyEquipment (NUCLEP); a nuclear fuel manufacturing plant (FCN) and a yellow-cake production plantbelonging to Nuclear Industries of Brazil (INB). Brazil ranks in sixth in world uranium ores reserves –310,000 t U 3 O 8 recoverable at low costs. The main uranium mining activities are in the Caitité County– Bahia State and fuel elements manufacturing plant is located at Resende County – Rio de JaneiroState. Brazil has also the basic technology for uranium conversion and enrichment, as well as privateengineering companies and research and development (R&D) institutes belonging to CNEN.In comparison with some developed countries, Brazil has a relatively modest and recent nuclear powerprogram. Due to the reduced dimensions of the nuclear market in Brazil and to the lack of greatprojects of nuclear facilities shutdown in the near future, there still are not companies specialized indismantling and decommissioning.Until the 2000 year, the only decommissioning experience in Brazil was the closure of the SantoAmaro’s Mill – USAM. This facility was property of the Brazilian Nuclear Industry, INB, and duringfifty years it was dedicated to the processing and production of thorium and rare earths from monazitesands, originated of the southeast beaches of Brazil, between the Bahia and Rio de Janeiro states. Theplant was installed in a residential area and in a densely populated region of the S. Paulo, the largestcity in South America. The operations were ended in 1992. The decommissioning activities occurredbetween 1993 and 1999. The Public Ministry of São Paulo State, together with INB, had established adeadline for the plant decommissioning, with daily penalties for lack of fulfilment, and therequirement of reports about the status of the decommissioning (2).1.2 IPEN’s profile and its nuclear fuel cycle pilot plantsThe IPEN is an institution owned by the Government of Sao Paulo State, supported and operatedtechnical and administratively by the CNEN. IPEN is located at the west of Sao Paulo city, inside theCampus of the University of Sao Paulo – USP. IPEN occupies an area of nearly 500.000 m 2 (20 %buildings) and is associated to the University of Sao Paulo for teaching purposes. Through apartnership with USP, IPEN conducts a post-graduation program. The IPEN staff is currentlycomposed of about 1,100 persons of which 30 % own post-graduate degree (PhD and MSc).Since its foundation in 1956, IPEN has played a decisive role in the development of the nuclearscience and technology in Brazil. It was created with the main purpose of performing research anddevelopment of nuclear energy peaceful applications. The IPEN research centres are engaged inmultidisciplinary areas such as nuclear radiation applications, radioisotope production, nuclearreactors, nuclear fuel cycle, radiological safety, dosimetry, laser applications, biotechnology, materialsscience, chemical processes and environment. An example of a large national impact IPEN activity has

een the production and supply of radiopharmaceuticals. About 2 million diagnostic and therapeuticnuclear medicine procedures per year have been performed in 2004 with products supplied by IPEN.The main IPEN’s facilities include: the nuclear research reactor IEA-R1m that reached criticality in1957 (built with United States support under the Atoms for Peace Program) and has been upgradedrecently to operate at 5 MWth; a Zero Power Reactor IPEN/MB-01 (critical assembly); twoCyclotrons (CV-28 and Cyclone 30 MeV – for radioisotope production); two electron beamaccelerators of 1.5 MeV for irradiation applications in the industry and engineering; two Cobalt-60Irradiators (11,000 and 5,000 Ci); dispersed fuel fabrication facilities (for research reactors);laboratories for chemical and isotope characterization, micro structural and mechanical tests.The Institute recent history has shown a major participation in the technological development of allsteps of the nuclear fuel cycle. One example of the important engagement of IPEN in the technologicaldevelopment in the nuclear fuel cycle area is the isotopic enrichment of uranium by ultracentrifugation, nowadays in process of industrial implantation. This significant achievement wasperformed in cooperation with the Brazilian Navy.Nuclear fuel cycle activities at IPEN, from uranium purification to hexafluoride conversion and fuelfabrication for research reactors, besides thorium and zirconium purification, were accomplished inpilot plant scale and most facilities were built in the 70-80 years. The facilities were used to promotehuman resources, scientific research and better understanding of fuel cycle technologies. The IPEN´spilot plants were distributed in groups located in different centres:• CQMA: ADU Dissolution (Impure Yellow Cake); Uranyl Nitrate Purification; ADU Precipitation;Calcination of ADU to UO 3 ; Denitration by Fluidized Bed (NUH to UO 3 ); UF 4 Production -Aqueous route; UF 4 Production - Moving Bed Units I and II –Dry route; Thorium SulfateDissolution; Thorium Nitrate Purification; Reprocessing laboratory.• PROCON: Fluorine Production; Uranium Hexafluoride Production; UF 6 Transfer.• CCTM: UO 2 Fuel Pellets Production.2. Factors Affecting the Decision and the strategy of D & D in the IPENRadical changes of the Brazilian nuclear policy, in the beginning of 90’s, determined the interruptionof most R&D fuel cycle activities and the facilities shutdown in the IPEN. Most Nuclear Fuel CycleFacilities had the activities interrupted until 1992-1993. Since then, IPEN has faced the problem of thepilot plants dismantling and/or decommissioning. Immediately after the nuclear R&D programinterruption, the uncertainties related to an eventual retaking of the Program created some politicalhesitation about the dismantling decision. Due to total lack of resources for operation andmaintenance, the units had the production interrupted and they meet in precarious conservationsituation. As an additional problem could be mentioned the great increase in the population verified inthe neighbourhoods of IPEN and USP in the last twenty years.However, the approach has changed in the last years. Of course, the retaking of the R&D NuclearProgram is now discarded. On one hand, it has been considered the problem of the costs related tofacilities maintenance/surveillance and, on the other hand, the problem of the gradual loss ofexperience and knowledge accumulated because of retirement or dispersion in different activities ofthe personnel former involved with the different nuclear fuel cycle processes. As the activities wereinterrupted in most facilities, IPEN has promoted a professional recycling of the remaining personnelwith emphasis in environmental applications of the existent experience (chemical processes) andother Institutional different priorities, such as radioisotope production or research reactor operationand fuel production. Another problem that should be mentioned is the exhausted capacity ofradioactive waste storage at IPEN. Besides this, Brazil has yet not defined a place for a radioactivewaste national repository.The reasons to promote the dismantling of the IPEN´s Nuclear Fuel Cycle Pilot Plants as soon aspossible elapse of the follow main aspects:• The IPEN is located in the Sao Paulo City, inside the Campus of Sao Paulo University, in an areaof nearly 500,000 square meters;

• The localization is an important aspect determining the reuse of the space and buildings of thefuel cycle facilities, since this is a very valuable area and the surroundings are a very populatedarea;• Need of physical space for new activities as, for example, the Fuel Cells Program, since theactivities of R & D in the area of the nuclear fuel cycle were interrupted, not having perspectivesof retaking of the Program;• Need to take advantage, as soon as possible, of the knowledge of researchers that were indeedinvolved with the project, assembly and operation of the different facilities, since it comeshappening a gradual loss of personnel for retirements and transfers;• The long period since the interruption of the R & D activities has increased the difficulty oftracking documents and reliable information, besides to evident deterioration of equipments andstructures, increasing the concerns with relationship to the risk of liberation of radioactivecompounds or that present risks for their chemical toxicities.Besides the above mentioned aspects, in the last ten or fifteen years, IPEN has changed its “nuclearprofile” to a “comprehensive and multidisciplinary profile”. Nowadays, some Brazilian governmentaland strategic programs are: Fuel Cells, Nanotechnology, Biomaterials, Environment, Polymers,Lasers. With the end of most Fuel Cycle research and development activities, the area occupied by theformer facilities constitutes a significant and useful resource. They can be fully or partially reutilizedfor a variety of purposes and programs. Besides the full release of some facilities as ”green areas”(priority programs), some buildings can be used as interim storage facilities (for equipment andwastes) . Decision regarding the reuse of the different facilities has been made on a case-by-case basis.During this period, IPEN has been restructured in 13 Research Centers. With the end of most nuclearfuel cycle activities, the former facilities were distributed in four different centers: Environmental andChemical Technology Center - CQMA; Fuel Cell Center - CCC; Materials Science and EngineeringCenter - CCTM; Nuclear Fuel Center - CCN. Each center has adopted a different strategy and priorityto face the D&D problem and to reintegrate the areas. Resources depend on the specific programdeveloped in each area (resources from other sources, not only CNEN).3. Dismantling and decommissioning background in the IPENThe decommissioning strategy for the old facilities dedicated to the technological domain of theNuclear Fuel Cycle follows an approach of advancing gradually in dismantling, since the resourcesand technical conditions are available. The reasons of such approach are the need of political decisionsrelated to the destiny of the facilities, lack of financial resources and of specialized personnel in thedecommissioning issue, besides the fact that decommissioning is not an institutional priority in thepresent.In spite of the difficulties mentioned above, some facilities were actually dismantled at IPEN recently,even without previous experience, training support or detailed planning. Orthodox D&Dmodels/technologies could not be followed, because there is not prepared personnel for the function.Poor expertise and lack of information and experience at IPEN in the subject provoked some degree ofimprovisation. Nevertheless, the operations were accomplished with a lot of radiological andenvironmental concerns, following strict procedures (3, 4).In the first phase of the activities, in the period between 2003 and 2005, the main objectives andpriorities were a preliminary rising of the nuclear facilities status, seeking the decommissioning. Apreliminary report was prepared with the basic procedure to be adopted for the fuel cycle facilitiesdismantling at IPEN (5, 6). This rising allowed knowing each installation that should bedecommissioned better, establishing a decommissioning strategy based on the institutional needs,besides trying to fill out the main gaps in terms of lack of appropriate technical knowledge to thedecommissioning and to identify the main technical obstacles that would be faced in the facilitiesdismantling. A fundamental aspect of the dismantling process of a disabled nuclear facility is theremoval of the material retained in the process equipment and the final disposition of the radioactivewastes generated during the operations. The dismantling operations were performed in four phases:

• 2000 and 2001 years, were dismantled the Thorium Sulfate Dissolution and UF 4 Production PilotPlant - Aqueous Route in the Building 2 of CQMA;• Between 2002 and 2003 years, were dismantled the ADU Dissolution (Impure Yellow Cake) andUranyl Nitrate Purification Pilot Plants, in the Building 1 of CQMA. In the place where thefacilities were set up IPEN has built new laboratories for the Environmental Program for usewithout restrictions (as green areas). These activities have been performed in 2006 and 2007;• Between 2005 and 2006, occurred the dismantling of the Uranium Hexafluoride Conversion PilotPlant. In the place where the UF 6 conversion facility was set up it has been built part of thelaboratories for development of the Fuel Cell Program, whit releasing for use without restrictions;• In 2007, is being accomplished the decommissioning of the UO 2 Pellets Fabrication Pilot Plant.4. ConclusionSome facilities demanded special attention, or will in the future when the dismantling will beconsidered, requiring preliminary operations of treatment of retained materials and/or wastes. It is thecase of the existence of organic wastes (TBP and pyridine containing U and Pu) stored in tanks, insidethe hot cells of the CELESTE-I Lab. Besides the raising of the status of the fuel cycle facilities and theimprovement of the preliminary dismantling plan, we have accomplished some studies ofinnovative/adaptive technological solutions to solving the mentioned retained material or wasteproblems, with emphasis on the use of chemical technologies, such as molten salt oxidation, todecompose hazardous materials and wastes related to decommissioning. It is very clear that we have toimprove all steps involving decommissioning such as planning, regulatory requirements, costestimating, cost-benefit analysis, need for and extent of decontamination, selection of decontaminationtechniques, assessment of the waste amount from the dismantling, dismantling techniques, stafftraining and so on. The interchange of knowledge and personnel empowerment should not be limitedonly to the treatment processes, but to all connected areas mentioned. During the dismantling &decommissioning activities in the IPEN, we have had the support of the IAEA by means of theResearch Project IAEA BRA-12800. The knowledge obtained thanks to the project was extremelyuseful in the decisions and activities regarding the continuation of the D & D activities and releasingof the areas.5. References(1) Third National Report of Brazil for the Nuclear Safety Convention, September 2004.(2) Ferreira, P.R., Matta, L.E. Mouço, C.D.C.L., Decommissioning: learned lessons and referencefor the future, IAEA-CN-143/30, November, 2006.(3) Standard CNEN.NE 3.01 Diretrizes Básicas de Radioproteção – Resol. CNEN 12/88 (RadiationProtection Directives).(4) Serviço de Radioproteção - IPEN, IPEN Report - Serviço de Radioproteção na Desmontagem eDescontaminação de Superfícies da Instalação Nuclear de Purificação de Urânio e Processamentode Tório no IPEN/CNEN-SP, São Paulo, April, 2003 (Radioprotection Service in the Dismantlingand Surfaces Decontamination from the Uranium Purification and Thorium Processing Facilities atIPEN/CNEN-SP).(5) Lainetti, P. , Freitas, A., Ferrari, E., Ferreto, H., Ayoub, J. Seneda, J., Bergamaschi, V., IPENReport - Plano de Ação para Desmontagem das Unidades do Ciclo do Combustível, São Paulo,February, 2002 (Action Plan for Dismantling of the Fuel Cycle Units).(6) Lainetti, Paulo Ernesto de O., Freitas, Antônio A., IPEN Report - Desmontagem das Unidadesde Dissolução e Purificação de Urânio do Ciclo do Combustível – Bloco I do CQMA, São Paulo,March, 2003 (Dismantling of the uranium Dissolution and Purification Units of the Fuel Cycle –Building 1of the CQMA).

OPERATION AND DECOMMISSIONING OF THENUCLEAR RESEARCH REACTOR FRANKFURT (FRF)HOLGER STARKE(Babcock Noell GmbH, Alfred-Nobel-Strasse 20, 97080 Wuerzburg, - Germany)INGO SCHILLING(STEAG encotec GmbH, Ruettenscheider Strasse 1-3, 45128 Essen - Germany)ABSTRACTThe research reactor Frankfurt FRF 1 started operation in 1957. Due to technical reasons itwas finally shut down in 1968. After the decommissioning of major parts of the FRF 1 theupgraded FRF 2 never got critical. After dismantling of most parts of the FRF 2 someremaining parts of the FRF 1 which had been reused for the FRF 2 were put into care andmaintenance. At the beginning of this decade it was decided to reuse the area for furtherurban development of the city of Frankfurt. Following licence application with allnecessary documents including article 37 EURATOM documents the decommissioninglicence was granted at the end of 2004. At the same time the order for the decommissioningwas placed with the consortium STEAG/BNG as a prime contractor.Following the necessary preparatory phase all necessary site equipment was installed. Afterthat the essential radiological characterisation of all remaining equipment and installationstook place. Upon completion of this characterisation and confirmation by the licensingauthority the decommissioning started. Removal of all radioactivity generated by operationof the reactor began in December 2005 and took nearly 6 month. The application for freerelease of the site was filed in September 2006 and licence was granted in October 2006accordingly.1 IntroductionThe facility of the nuclear research reactor Frankfurt (“Forschungsreaktor Frankfurt”– FRF) was used as a research and training facility of the nuclear physics institute of the Johann-Wolfgang-Goethe-University, Frankfurt, Germany. The FRF 1 was a homogenous thermal reactor. Itgot critical in 1957. The nuclear fuel was an aqueous Uranium-Sulphate solution. It was enclosed inthe primary system. The maximum thermal output was 50 kW.

Facility of the FRF, about 2004, MarchOn 1968, March 19 th the reactor was shut down finally due to technical problems at the hydrogenrecombining system. In 1970 the liquid fuel including all fission products were shipped forreprocessing to “Eurochemic” at Mol, Belgium. Main parts of the reactor system weredecommissioned or got used in material testing series. In the years from 1973 to 1977 a new researchreactor was implemented in the existing building of the facility. It was called FRF 2. The FRF 2 wasan adapted TRIGA-reactor. It had a maximum thermal output of 1 MW. But finally it got nevercritical. In 1980 decommissioning was directed. In the following time parts of the facility gotdismantled. The biological shield of reactor and a few other parts were left over. Most of theremaining radioactivity was contained within the biological shield.After the decommissioning of the systems of the FRF 2 the experimental hall of the reactor wasfurther on used to pack the remaining radioactive wastes of the Johann-Wolfgang-Goethe-University.Reactor and Biological Shield of FRF, about 2004 March

2 Operation and DecommissioningThe nuclear-physics institute was located in the “Rebstock”-area in the west of Frankfurt, Germany.The “Rebstock”-area should get developed within a project called “living and working at ‘Rebstock’-park”. To get space for this project, it was necessary to dismantle the facility of the research reactorFrankfurt (FRF) including the unrestricted release of the facility from the restrictions of the atomiclaw.This background founded the application to approve the decommissioning. The responsible licencingauthority was the state Ministry for Environment of the federal state Hessen – “HessischesMinisterium für Umwelt, Ländlichen Raum und Verbraucherschutz” (HMULV). They granted theapproval in 2004, December. Also in 2004, December the order was given to the consortium ofSTEAG and BNG (ARGE STEAG / BNG). The decommissioning work on site started as planed in2005, March. The first step was to set up the site. The final one was to release the facility out ofatomic law in 2006, September.3 Concept of DecommissioningThe decommissioning of the research reactor Frankfurt (FRF) was divided in- Dismantling of the biological shield and the remaining parts of the reactor- Dismounting of components out of the controlled areasThe biological shield (LxWxH = 5.5 m x 4.5 m x 3.5 m) was made from extreme heavy concrete. Anexcavator with a hydraulic chisel knocked out a tunnel from two sides. About 60 tons of concrete werequarried out. Time pointed out, that the concrete was so hard that the excavator worked slower thanusually and also chisel was worn out pretty fast. Inside the bloc of concrete were some graphite bricks.These had together a weight of approx. 7 tons. All of them were drawn out manually with long toolsand filled in drums. The maximum dose rate was 0.6 mSv/h. Other parts, like additional shielding,went out by crane. The reactor-tank, diameter approx. 0.6 m, made from aluminium was drawn outand cut in pieces. After the dismantling of the beam tubes, the activated part of the biological shieldcould be dug out from inside. It was done combined by drilling and cutting with diamond tools. Thetorn out pieces of concrete had a weight from 4 to 8 tons. Those were size-reduced at a special placealso with the hydraulic chisel.The main steps of decommissioning components out of the controlled zone were:- Dismounting of some tanks- Removal of facing from floors or walls- Excavation of contaminated conduits out of walls, foundation or ground- Stripping down of remaining system partsThe dismantling was done with standard tools. Special arrangements were needed to preventcontamination spreading i.e. the activity was fixed in the pipes before removal.4 RadiologyPrior to decommissioning the complete radioprotection equipment had to be installed and alsoincorporated into the operating handbook. The next step was a sampling and measurement program tocheck current radiological condition of the whole facility. Key-nuclides identified were Co-60 inactivated areas and Cs-137 in contaminated areas. Remarkable was the high proportion of Sr-90. TheSr-90 proportion was found to be more than three times higher than Cs-137.In the whole facility five different nuclide-vectors could be determined. These characterised thedifferent parts of the facility. The nuclides Eu-152, Co-60 and Ba-133 were found to be characteristicfor the graphite round the core. The maximum dose rate of one graphite brick was about 250 µSv/h.The specific activity could be determined to approx. 20 kBq/g for H-3 and approx. 1,2 kBq/g for C-14.The radioactive waste is stored at the interim storage of the federal state Hessen. Measurements forfree release of residues were done by an external contractor.

Biological shield of the reactor after free releaseFinally, about 90 tons of combustible or compactable waste or waste, which needs no furtherconditioning, were packaged. The drums were documented as by the requirements of the interimstorage of the federal state Hessen for long term interim storage.The mass of all materials for free release was about 125 tons. It was packed in polyethylene-boxeswith a size of 1 cubic metres. Packaged like that, it was sent to an external contractor as a radioactivetransport.The way of external free release got proved as a good and economic way.4.1 Free releaseThe basis for the free release measurements of the remaining buildings and the site territory is theGerman radiation protection ordinance /1/. It defines for nearly all relevant radionuclides and forseveral certain scenarios discrete exemption values.The free release measurements of the remaining buildings and the site territory were in major partsdone with in-situ-gamma-spectrometry. In-situ-gamma-spectrometry is under some conditions aneffective way to perform free release measurements on building structures and land area.These conditions are:- Suitable nuclide vector, known correlation factors to non-gamma nuclides- Homogeneous contamination level- Reasonable averaging areaAs a prerequisite appropriate nuclide vectors and correlation factors between easy to measure gammaemitting radionuclides and all those not contributing to the measurement result such as Fe-55, Ni-63,Sr-90 had to be defined. These nuclide vectors were the result of the aforementioned radiologicalcharacterisation performed at the beginning of the decommissioning process.A homogeneous contamination level (exclusion of hot spots on surfaces) is not a compulsoryprerequisite but if it could be stipulated the necessary measurement time will significantly decrease.The German radiation protection ordinance stipulates to refer the measurement results to a certain areawhich is dependent on the decommissioning scenario (if the building will be further used or beingknocked down). In case the building will be further used the appropriate averaging area shall be1 square metre. Taking into account the field of view (measurement area, especially with uncollimated

detector) this leads to very conservative results and on the other hand to a significant increase inmeasurement time.For the free release of all surfaces of the controlled area of all buildings a total of approx300 measurements were performed. The necessary life time for one measurement varied between 30minutes and 22 hours. The latter representing measurements of vaults with a depth of 2 metres and across section of 15 by 15 square centimetres.An average room with a floor space of approx. 16 square metres and a height of 3.4 metres took intotal less than 4 hour measurement time (2 single measurements with diametrically opposed measuringdirections) to be well below clearance levels (depending on detector efficiency, nuclide vector).Demolition of the remaining structures after free release (April 2007)Demolition of the remaining structures after free release (May 2007)4.2 References/1/ Ordinance for the Implementation of Euratom Directives on Radiation Protection of 20 July2001, published in Bundesgesetzblatt 2001 part I p. 1714, corrected 2002, I, p. 1459

Session 18.2.5Waste and transport

VITRIFICATION EXPERIENCE AND NEW TECHNOLOGYDEVELOPMENT IN TOKAI VITRIFICATION FACILITYA.AOSHIMA, T.UENO, M.SHIOTSUKITokai Reprocessing Technology Development CenterNuclear Fuel Cycle Engineering LaboratoriesTokai Research and Development CenterJapan Atomic Energy Agency (JAEA)4-33 Muramatsu, Tokai-mura, Naka-gun, Ibaraki, Japan, 319-1194ABSTRACTTokai Vitrification Facility (TVF) started hot operation in 1996 and produced 241 canistersas of June 2007. Through TVF operation, JAEA had much experience and accumulatedmuch technical know-how which indicated that management for noble metal accumulation ina melter was key technology for smooth plant operation.JAEA should continue service operation based on a vitrification contract with the Japaneseutilities because there remains about two third of High Active Liquid Waste (HALW)produced in the reprocessing service operation of Tokai Reprocessing Plant (TRP).TVF melter is designed in condition of five years life time because of very corrosivecharacteristic of melted glass. Five years design life time is equivalent to 500 canistersproduction in TVF. Because estimated number of canisters which will be produced in thefuture is over 500 canisters, exchange of the present melter is necessary.From these situations, JAEA decided basic strategy to increase stability of the existing melteroperation and develop an advanced new melter for replacement in future which has largelyprolonged life time and high noble metal drain ability.To attain these targets, JAEA extracted necessary key technologies to assemble into a tenyears road map and started development. This development has been progressing onschedule.1. IntroductionJAEA has operated TRP successfully for about thirty years since the start of its hot test operation in1977. TRP was designed to have capacity of 0.7 tons U/day and treat LWR spent fuels burned up 28,000MWD/MTU in average. On the end of March 2006, TRP has finished reprocessing service of LWRspent fuels based on the reprocessing service contract between JAEA and the Japanese utilities. Totalamount of spent nuclear fuels reprocessed reached over eleven hundreds tones. From 2006JFY, JAEAhas changed main role of TRP operation from the reprocessing service to new reprocessing technologydevelopment by processing mainly MOX fuels of the advanced thermal reactor “Fugen”.In TRP, HALW from main process is stored and vitrified in TVF. TVF was constructed and started thehot operation in 1995 and produced 241 canisters until end of March 2007.Comparing to TRP, situation of TVF is very different. JAEA should continue service operation of TVFbased on a vitrification contract with the Japanese utilities because there remains two third of HALWproduced in the reprocessing service operation of TRP.A glass melter is main equipment in vitrification process and its design lifetime is five years in themaximum capacity operation mode because of very corrosive hot glass characteristic. The five year lifetime is equivalent to five hundreds canister production in TVF. Estimated number of canisters whichwill be produced in future exceeds five hundreds canister, exchange of the second melter is inevitable.Taking account into annually produced number of 40~50 canisters, exchange timing of the secondmelter is estimated to come after about ten years operation.Based on TVF operational experience, certain discharge of noble metal from melter should be donebefore gradual accumulation attains some amount which causes fatal damage to the melter.

Thinking of these situations, steady operation of the present melter and development of an advancedmelter with prolonged design lifetime is very important not only for more cost effective operation butalso waste volume reduction. So, JAEA listed up key technologiesand made each development plan toassembled into a ten years road map. These technologies were categorized into three main targets:(1) Stable operation of the present melter (noble metal management / melter disassembling technology)(2) Development of advanced melter and upgrading of vitrification technology (high waste loading)(3) Basic technology (simulation technology).2. Technological Accumulation through TVF Operation (1) (2)2.1 OperationHALW transferred from TRP is vitrified and poured into a 110 litre canister in TVF. The TVF melter isdesigned to have capacity equivalent to the reprocessing capacity of 0.7 tons MU/day of TRP. Type ofTVF melter is liquid fed joule-heated ceramic melter (LFCM). The LFCM type melter has been used asHALW vitrification melters in the world for its advantages of vitrifying waste liquid into solidifiedproduct directly and of designing a large scaled melter easily. (Fig.1)From the beginning of TVF operation, special attention was paid to accumulation of noble metal at thebottom of the melter. Because noble metal elements are hardly dissolved in glass matrix, noble metalelements, taking oxide or metal forms etc., are suspended in melted glass. Noble metal compoundswhich have heavier specific gravity and larger electrical conductivity than melted glass make badinfluence on melter heating capability by stray current through accumulated noble metal compounds.And this stray current seems to make fatal damage to electrodes in the worst case.So, in TVF operation, special operation method, “low temperature operation mode” which keepstemperature of melter bottom glass around 800-900 is taken. The special operation mode adjusts glasstemperature to avoid accumulation of noble metal to melter bottom by increasing viscosity of glass.Instead of these careful dealing, damage of one main electrode occurred in 2002. Soon, JAEA decided toexchange the failed melter to new one and finally JAEA successfully finished the melter exchangeproject and restarted the facility operation on October of 2004. In the design of the second melter,HLLWbottom electrode structure was improved to get moreFeed Linesmooth glass drain behaviour. (Fig.1)GlassOff-gas LineFeed LineIn the second melter operation, JAEA made resistance ofmain electrodes as a new monitoring index whichStart-upHeaterrepresents a degree of noble metal accumulation. JAEAalso set a threshold value of the resistance to stop themelter operation. The vitrification operation continued toMarch 2007 and produced 111 canisters successfully.Total number of produced canisters reached 241 from thestart of the hot operation in 1994. During the melterMain operation, operational parameters were very stable.Electrode Parallel to melter exchange activity, JAEA developed newAuxiliary melter maintenance system (JAEA calls this as “renewalElectrodeDischargeactivity”) which requires mechanical removal ofNozzleaccumulated noble metal rich glass and detail surface(a) Whole Structuremeasurement by laser measuring system. Execution timingof “renewal activity” is decided based on monitored data ofmain electrodes’ resistance. The first application of“renewal activity” is planned in September of 2007 toconfirm functions of remote devices for glass removal and(b) Bottom Electrode Structure surface measurement.Fig. 1 Present TVF Melter2.2 Melter Exchange

TVF has introduced “fully remote maintenance system” which makes equipment exchange work veryeasy and efficient comparing to direct maintenance. In the fully remote maintenance system, all mainequipments are mounted in standardized racks with dimensions of 3 m x 3 m x 6.5 m or installed on thebase framework by bolting (melter). The racks and the melter are placed in a large cell “vitrification cell”and arranged along the both sides of the cell wall. The racks and the melter are remotely removable. Formaintenance, two overhead 20t cranes and two bilateral servo-manipulator (BSM) systems are disposed.These systems are manipulated remotely from a control room based on images from radiation hardenedITV cameras mounted on the cranes and the BSMs.Manufacturing of the new melter started on April of 2003. After completion of manufacturing themelter, it was transferred to Cold Mock-up Facility (CMUF), which located adjacent to TRP, to carry outcold test for confirmation of operation condition. On May of 2003, the cold operation test of the secondmelter started and continued to the end of December. On the other hand, in advance of the second meltertransfer to TVF, the first melter has removed from original position to dismantling area of thevitrification cell. After installing the new melter on the base frame, three-dimensional measurementswas carried out remotely to determine spatial relative position of jumper piping flanges. Based on themeasured dimensional data, jumper tubes which connect wall side and melter piping were manufacturedand installed remotely. The melter installation work was finished at the end of August 2004.Improving ability of glass draining to prevent accumulation of noble metal on melter bottom, newdesign of bottom structure was introduced for the second melter. (Fig.1) Effectiveness of the designchange was checked in the cold operation test and no accumulation of noble metal was observed afterdraining all glass. In the cold operation test, operational parameters were fixed and parameters’differences from the first melter were confirmed. These new operational parameters were used toimprove the operational manuals of the TVF melter.Fig. 2 Special DevicesLeft: Glass Removal DeviceRight: Laser Scanning MeasurementDevice2.3 The Second Melter Operation and Noble MetalManagementIn the second melter operation, accumulation of noblemetal is inevitable in long operation period. So, keypoints of this technology are to grasp growth of noblemetal accumulation during operation and to stop melteroperation before damage of main electrodes occurs. Inthe second melter operation, new electrode resistancemonitoring technology is applied to detect noble metalaccumulation in advance. When resistance value of themain electrodes decreases less than the set up value, themelter operation is stopped to avoid damage of mainelectrodes. After stopping the melter, open a top flangeand insert special devices to remove glass withaccumulated noble metal. Following glass removingwork, detail surface measurement is carried out toestimate a loss of thickness by corrosion. Successive andperiodical measurement of surface roughness is veryimportant to understand degree of degradation andcorrosion trend of refractory and electrodes. For these“renewal activity”, special devices for glass removal andsurface measurement were developed. (Fig. 2) Glassremoval device is a kind of robot arm with six degree of freedom and glass removal is done by operatingthis arm grasping an air activated special tool “mechanical chisel”. The laser scanning measurementdevice is also an arm robot measuring inside surface shape by laser reflection measuring method. Thefirst renewal activity starts on October 2007.2.4 Melter Disassembling TechnologyIn the vitrification cell of TVF, there is special space for remote disassembling where the remotemaintenance devices are equipped: a ceiling crane, a power manipulator, shielding window with a pairof mechanical manipulator system and a turning table system. The TVF melter is composed of threemain materials: refractory, electrodes (INCONEL), melter outer shell (SUS). The melter disassembling

technology has been developing and special devices for dismantling were designed in collaboration withthe utilities.JAEA has designed dismantling process and selected YAG laser for cutting outer shell, electrodes andrefractory. Mechanical chisel is used for attached glass removal from refractory. Volume reduction ofglass waste and HASW is very important issue for future disposal from the viewpoint of reducingnatural burden and cost saving. So, separating technique of glass from refractory is a key technology.Decontamination technique of HASW is also important for reducing TRU waste volume. Dataacquisition on degree and distribution of contamination is useful for next melter disassembling work toperform more efficiently and will be utilized in design of the third melter.The first melter disassembling started in 2005 and is planned to finish in 2007. Detail investigation of thedamaged electrode will be done in the disassembling process and cause of the damage will be finallyfixed based on detail investigation data of taken sample from the electrode.3. Development of Advanced Melter and Upgrading of Vitrification Technology (4)3.1 Advanced MelterTVF melter is designed to exchange every fiveyears because of corrosion of electrodes andrefractory. Once the melter exchanged, theexchanging work interrupts the melter operationand disassembling work generates much radioactivewaste to be disposed. So, to decrease cost and wastevolume, an advanced melter with long lifetime isstrongly required to be developed. So, JAEA settwenty years as a design lifetime for the thirdmelter.Prolonging lifetime four times, it is necessary todecrease corrosion rates of the melter structuralmaterials one forth. To decrease corrosion ratedrastically, “skull layer formation technology” isconsidered. Fig.3 shows a very early stage conceptof a skull layer melter. Skull layer is formed onsurface of refractory by cooling from outside.Movable and remotely changeable electrodes areinserted from top of a melter. This type electrode isvery effective to avoid fatal damage of electrodes byaccumulated noble metal. In the advanced melter,Movable &ExchangeableElectrodeReuse ofOuter CasingGlassCartridgeNoble MetalRecoverysmooth drain of noble metal will be done by special temperature control of melted glass in the lowerpart.JAEA has started advanced melter development in 2005 and will promote melter design and relatedR&D activity until 2008 aiming to establish most feasible melter concept design.3.2 High Waste LoadingAt present, the solidified glass includes about 25 % waste, which is composed of 10% sodium dioxideand 15% metal oxides. If waste containment increased to 30 % (12 % sodium nitrate and 18 % metaloxides), the number of necessary canisters decreases to about 20%. This means big cost saving fortransportation, storage and disposal as well. From this point, JAEA has started the technologydevelopment of high waste loading since 2004 in collaboration with the utilities.When the glass volume is reduced, there is possibility of changing glass characteristics essential to thedeep underground disposal. So, as the first step, JAEA has been carrying out laboratory scale test of 30% waste glass by measuring basic characteristics such as appearance, density, leaching rate, electricresistance and viscosity. As a result, glass was homogeneous and satisfied the JAEA’s standardspecifications of solidified glass. Behaviour of noble metal in the melter is another important issue to beconfirmed, because there will be more possibility of noble metal accumulation in the melter when muchnoble metal is fed to the melter. So, JAEA estimated behavior of noble metal in small scale coldmock-up test.Induction HeatingWasteSkull layerMainElectrodeRefractoryForming ofSkull LayerAirCoolingJacketFig. 3 Advanced Melter Development(Example Concept)

The small scale cold mock-up test was performed in 2005. The test results were obtained satisfactorilyand successfully. The melter was stable and all operational parameters were controlled as intended. Noindication of noble metal accumulation was observed from resistance data of main electrodes. Actually,no residue of noble metal was observed after draining. Based on the test result gained in the small scalemock-up test, JAEA will start to estimate on application to the TVF melter and JAEA has started togather more detail data in the case of taking higher waste loading ratio such as FBR spent fuelreprocessing.(3)4. Other Related Technology4.1 Simulation TechnologySimulation system has been under development to analyze coupled features and phenomena in a meltersuch as temperature, heat generation, electrical potential, current density as well as noble metal particledistribution. In the simulation code, each feature is calculated step by step. Bringing the simulation codeclose to the actual melter, JAEA has been trying to upgrade the code by adding new simulation functionsuch as formation of noble metal compounds in cold cap, accumulation of noble metal particles onsurface of refractory, draining behavior of noble metal particles and so on.4.2. New Technology DevelopmentFor future vitrification technology, JAEA paid attention to sol-gel glass method and has started basic testfor confirmation of process feasibility. By utilizing sol-gel method, operation temperature was possiblyreduced remarkably(~600). It means that corrosion rate of material become too small and melterdesign with long design life time will be done more easily.Utilization of produced canisters as irradiation source has been discussed frequently. Main gammaradiation source element in a canister is caesium 137 which have longer half life (~30 years) than cobalt60(~5.3 years). Comparing to cobalt 60 as gamma ray source, energy spectrum of gamma ray of acanister is broader and its energy is weaker. Strict dose control in canister pits is also more difficult thanin cobalt 60 irradiation facility. Considering these features of gamma ray from canisters, JAEA hasstarted to study on application of canisters as radiation source.5. Conclusion(1) Through TVF operation over ten years, JAEA had much experience and accumulated muchknow-how which covers all technological area to operate vitrification facility smoothly. Among thesetechnologies, JAEA has got understanding that management of noble metal accumulation is key tooperate the melter stable and efficiently.(2) JAEA made basic strategy which is consist of stable and cost effective operation of the second melterand development the third melter with prolonged lifetime. Based on this strategy, JAEA made a decaderoad map of necessary key technologies and has started the development since 2005.(3)Besides development of main vitrification technologies, JAEA has paid attention to application ofnew technology to vitrification field and utilization of existing facility for industrial purpose.References1. A. Aoshima, T. Kozaka, K. Tanaka "Glass Melter Replacement and Melter Technology Developmentin the Tokai Vitrification Facility", 12th International Conference on Nuclear Engineering, April, 2004,Arlington, Virginia (Washington, D.C.), USA2. A. Aoshima, K. Tanaka, "Waste Treatment Experience and Future Plans in Tokai ReprocessingPlant", WM’05 Conference, March, 2005, Tucson, AZ, USA3. M. Shiotsuki, A. Aoshima, S.Nomura "Perspectives on Application and Flexibility of LWRVitrification Technology for High Level Waste Generated from Future Fuel Cycle System", WM’06Conference, February, 2006, Tucson, AZ, USA.4.A.Aoshima, T.Tanaka, "Vitrification Technology Development Plan in Tokai Reprocessing Plant",14th International Conference on Nuclear Engineering, July, 2006, Miami, Florida, USA

MINOR ACTINIDES PARTITIONING: MAIN RESULTSDURING THE FIFTEEN YEARS RESEARCH ANDPROSPECTSC. ROSTAING, P. BARON, B. LORRAINRadiochemistry & Processes Department, CEA/Valrhô Division–BP 17171, 30207 Bagnols-sur-Cèze,– FranceD. WARIN, B. BOULLISDPCD, CEA/Saclay Division91191 Gif sur Yvette Cedex – FranceABSTRACTIn the frame of the French national waste management 1991 act, the CEA had launchedresearch and development studies on the separation of the minor actinides – i.e. neptunium,americium and curium- from high active waste issuing from nuclear spent fuelreprocessing. In compliance with the 2006 deadline specified by this act, the research workinvolves two phases: demonstrating scientific feasibility (validation of the basic separationconcepts) by the end of 2001, and technical feasibility (overall validation of the processes)in 2005. Significant results obtained during these two phases are presented. In conclusion,the feasibility of the processes selected for partitioning was demonstrated. In June 2006, anew act on sustainable management of radioactive waste has been voted by the Frenchparliament. The strategy retained in the frame of this new French Act is given as aperspective.1. IntroductionUnder the terms of the French radioactive waste management act of 30 December 1991, a researchprogram has been carried out by the CEA, from the nineties with the objective of investigatingseparation processes for subsequent transmutation of long-lived radionuclides to significantly reducethe radiotoxicity of the ultimate wasteforms produced by the nuclear industry [1]. The main targetswere the minor actinides – i.e. neptunium, americium and curium- to recover from high active wasteissuing from nuclear spent fuel reprocessing. The aim was to recover them quantitatively andselectively: extraction yield at least equal to 99.9%, with a decontamination factor towards lanthanidesas large as possible. A selection of significant results related to actinides/lanthanides partitioning,issuing from this research work is given.2. StrategyThanks to the experience gained in the PUREX process operating in La Hague plant, the referencestrategy for separating the minor actinides from the spent fuel (figure 1) is based on an adaptation ofthe PUREX process for the separation of Np and the development of new liquid-liquid solventextraction processes for the others (Am, Cm). Owing to the fact that chemical properties of actinidesand lanthanides are very similar, the separation of these two series of elements is all the more difficultbecause the medium is very acidic (3 to 4 mol.L -1 ). That is why a succession of two processes calledDIAMEX and SANEX was set up.In compliance with the 2006 deadline specified by the waste management act, the research work wasdivided into two phases with the objective of demonstrating scientific feasibility (validation of thebasic separation concepts) by the end of 2001, and technical feasibility (overall validation of theprocesses) in 2005.

UPuNpAm CmSpent fuelPUREX processpotentialitiesPUREXHA raffinateF.P.DIAMEX - SANEXAm/Cm separationLnAm + CmComplementaryextraction processesFig 1. Reference strategy for minor actinides separation3. Some outstanding results related to the scientific demonstration3.1. DIAMEX processFor the first step, the objective was to find a molecule able to selectively complex actinides andlanthanides, resistant both to hydrolysis and radiolysis and containing only carbon, hydrogen, oxygenand nitrogen (called CHON principle), so that no secondary waste are produced. Diamide moleculecheck these criteria. For the first test conducted in 1993, on used fuel raffinate, the solvent was madeof DiMethylDiButylTetraDecylMAlonamide (DMDBTDMA) diluted in TetraPropyleneHydrogenated(TPH), diluent used in the PUREX process in La Hague plant [2]. Even if the actinides andlanthanides recovery yield was satisfying (more than 99%), the structure of the diamide molecule wasoptimized to get more latitude for designing the flowsheet. In this way, extracting, loading andrecycling capacities have been improved respectively by introducing an oxygen atom in the centralchain, increasing the total number of carbon atoms and at last providing a more homogeneousrepartition of the alkyl chains (so that degradation products are not too much lipophilic) [3]. Testingsome dozens of malonamides combined with structure-activity consideration lead to an improvedmalonamide: the DimethylDiOctylHexylEthoxyMAlonamide (DMDOHEMA) represented in thefigure 2. Very good performances of this molecule were confirmed by tests performed in mixerssettlers on genuine solutions in 1999: more than 99.9% of actinides plus lanthanides recovered, inagreement with the targets [4].NO ONNO OONDMDBTDMADMDOHEMAFig 2. Evolution of the reference diamide extractantIn addition, complexing scrubbings were set up, to prevent from some fission products extraction suchas zirconium, molybdenum, iron and palladium: oxalic acid and HEDTA were added (see figure 3).ExtractantEXTRACTIONSCRUBSTRIPother FPsMA + Ln+ other FPsScrubbingsolutionMA + LnStrippingsolutionFig 3. Reference flowsheet for the DIAMEX processThis flowsheet was successfully tested at laboratory scale from 1999 to 2003 in mixer-settlers andsubsequently in ECLHA centrifugal extractors 1 on active solutions issuing from the dissolution of1ECLHA: High Activity Laboratory Centrifugal Extractor

actual spent fuel samples. Actinide recovery factors above 99.9% were obtained with high purificationfactors for fission products other than the lanthanides as summarized in the table 1.CYRANO/FARITUATALANTEATALANTEITUITU199319981999200020022003Duration 16 h 4 h 45 h 38 h 3.5 hContactor M - S C - C M - S C - C C - C C - CExtractant DMDBTDMA DMDBTDMA DMDOHEMA DMDOHEMA DMDOHEMA DMDOHEMAFeed MOX UOX2 MOX MOX UOX2 MOX(concentrate)Product>99.9 % Am~ 99.9 % Cm97.6 % Mo86% Fe~ 33 % RuAn(III) >99.7%Ru ~ 9 %Pd ~ 85 %Y ~ 18 %An(III) ~ 99.9 %Zr, Mo < d.l.Ru ~ 10 %Pd ~ 60 %An(III) ~ 99.9 %Zr < d.l.Mo < d.l.Ru < d.l.Pd < d.l.Y ~ 60 %Am ~ 99.96 %Cm > 99.7 %Ln(III) ~ 99.5 %Ru ~ 3 %Pd < d.lY ~ 15 %Tc ~ 62 %Tab 1: Main results of tests performed on genuine solutionsAm > 99.7 %Cm > 99.9 %Y, Ln(III) ~99 %Zr < d.l.Mo ~ 0.6 %Ru ~ 6 %Pd < 1 %3.2. SANEX processThe second step addresses the difficult issue of separating the trivalent actinides from the trivalentlanthanides (Selective ActiNide EXtraction concept). Three approaches have been investigatedconcurrently [4]:• The basic reference strategy seeks to separate the actinides from the lanthanides following theDIAMEX process by selective extraction of the trivalent actinides without any intermediate acidityadjustment (i.e. in 0.5–1 mol·L -1 nitric acid). The extractant molecules likely to provide suchselectivity must include “soft” donors (nitrogen or sulfur atoms). The main systems investigatedcomply with the CHON principle. Thus most studies are carried out with nitrogenous ligands.• Variant 1 is based on the same criteria as the reference strategy, but includes adjustment of theacidity by adding a buffer reagent (numerous An(III)/Ln(III) separation systems are effective atpH > 2).• Variant 2 is based on an appreciably different concept: selective stripping of the trivalent actinidesin the second step of the DIAMEX process; this approach allows the trivalent actinides andlanthanides to be coextracted and separated in a single liquid-liquid extraction cycle.Some dozens of extractant molecules, including synergistic mixtures were evaluated during this periodfrom about 1994 to 2001. Finally, for the SANEX basic reference strategy, the BTP's family was wellsuited for An/Ln separation, but HA tests show insufficient stability characteristics for the two testedmolecule (nPr-BTP and then iPr-BTP). For the variant 1 (idem reference strategy with feed acidityadjustment) the main drawback, beside the need of a pH adjustment of the feed with a buffer, is thehigh sensitivity of the flowsheet to flow rates due to the limited separation factor procured by theselected extraction system. It is the reason why the variant 2 (selective An stripping in DIAMEXprocess) seemed to be the best compromise: a good extraction yield of An and good DF An/Ln hasbeen achieved in a surrogate and a HA tests and the stability of the extractant was sufficient [5].4. Some outstanding results related to the technical feasibility4.1. DIAMEX processThe main objectives of the “technical feasibility” demonstration runs at the end of 2005 with thesolution produced during the PUREX test in April 2005 were to test continuous solvent recycling(not included during the earlier tests) and to carry out essential operations in continuous contactorsrepresentative of pulsed columns that could be used at industrial scale. The demonstration was carried

out in the CBP shielded cell facility. A specificity of this facility is the large height cell, equipped withthree pulsed columns 15 mm diameter and 4 m high. About 13 kg of spent fuel were dissolved forthese tests, producing nearly fifty liters of solution requiring treatment [6].Fig 4. Pulse columns and upper shielded cells in the CBP facilityAs a preliminary a modified PUREX flowsheet was performed to provide a feed for the DIAMEX run;it was checked that a increase of nitric acid concentration allowed to extract more than 99% ofneptunium.The technological demonstration test of the DIAMEX flowsheet was performed by the end of 2005 inthe CBP facility on genuine solution issuing from the PUREX test for neptunium separation.Experimental recovery yields of americium and curium were consistent with the objectives (> 99.9%)and confirmed the results obtained in 2001.HNO 3H 2 C 2 O 4HEDTAFeedHNO 3 4MAm ~ 150 mg/LCm ~ 15 mg/LLn ~ 2,5 g/LMo, Zr, FeV = 0,96 L/hHNO 3 1.8 M0,6VHNO 3,NaOH 0.3MHEDTASolvent Treatement(ECRAN)H 2 C 2 O 4HEDTA0,2VextractionAn+Ln(CP)0.65 MDMDOHEMA/TPH0,65 M dans2VextractionAn+Ln(CP)Washing FP(CP)Back-extraction An-Ln-(MS)HNO 3 0,1 M0,75VAm, Cm, Ln :> 99.9%FPRaffinateAm ~ 0,015 %Cm < 0,002 %Fig 5. Flowsheet and main results of the run in the CBP hot cell4.2. SANEX processIn the same way, a SANEX run, based on the variant 2 flowsheet, was tested in the ATALANTEfacility in december 2005. In the same way, experimental recovery yields of americium and curiumwere consistent with the objectives (> 99.9%) and confirmed the results obtained in 2001.

DMDOHEMA+ HDEHPin TPHSolventcleanupEXTRACTION / SCRUBBINGRaffinateFEEDHNO 3An Back-extractionLn,Y Strip.Am, CmHEDTAcitric acidpH 3LnHNO 3Fig 6. Flowsheet of the SANEX process implemented in the end of 20055. Conclusion - PerspectivesSeveral effective partitioning processes are now available to meet one of the requirements of the 1991French high-level radioactive waste management act. The technological feasibility of the americiumand curium processes (DIAMEX and SANEX in two steps) was demonstrated through the treatment of13 kg of genuine spent nuclear fuel in the ATALANTE facility with actinides high yield recovery(>99.9%).In june 2006, the 28 th , a new waste management act was voted. Concerning partitioning andtransmutation, the program is now connected to 4th generation reactors, in which transmutation ofminor actinides could be operated. In this frame, the next important milestone is the evaluation of thepossible transmutation roads, which are:• minor actinides homogeneous recycling, in the whole reactors park, with a low content ofM.A (~ 3%) in the whole fuel assemblies,• minor actinides heterogeneous recycling, in about a third part of the reactors park, with ahigher content of M.A. (~ 30%) in dedicated targets put in the periphery of the reactor.In this frame, partitioning processes developed during the first waste management act (1991-2006) areconnected to heterogeneous recycling, while GANEX processes (grouped actinide separation byextraction) are connected to homogeneous recycling. So, concerning the DIAMEX and SANEX,processes already demonstrated; up to 2009, possible simplifications will be assessed. In parallel, ademonstration of GANEX process is planned before 2009; for that, all the experience gained in thecourse of the minor actinide partitioning can be valorized.The next step will involve the studies necessary prior to industrial implementation of these processes.A farther date could be the building of a small partitioning shop (micro-pilot) dedicated to theproduction of fuel assemblies containing minor actinides which would be recycled in the 4 th generationprototype reactor planned in 2020.6. References[1] B. BOULLIS, M. VIALA, C. MADIC, F. JOSSO, G. NAUD, "Separation of Long LivedRadionuclides. Main Goals and Recent Progress of the Spin program", Proceedings of the workshopon Long Lived Radionuclides Chemistry in Nuclear Waste Treatment, Villeneuve-les-Avignon(France), 18-20 June 1997, published in OECD Publications, Paris (France), 274, 39-45 (1998)[2] B. BOULLIS, M. SALVATORES and H. MOUNEY, "Separation and Transmutation of LongLived Radionuclides: Recent Advances of the French Spin Program", Proceedings of the InternationalConference on Future Nuclear Systems GLOBAL’99, Jackson Hole, Wyoming (USA), August 29 –September 3, 1999, published by American Nuclear Society, Inc (1999)[3] M.C. CHARBONNEL, C. NICOL, L. BERTHON and P. BARON, "State of Progress of DIAMEXProcess", Proceedings of the International Conference on Future Nuclear Systems GLOBAL’97,Yokohama, Japan, October 5-10, 1997, published in Atomic Energy Society of Japan, Tokyo (Japan),1588, 366-370 (1997)[4] P. BARON, X. HERES, M. LECOMTE, M. MASSON, "Separation of the Minor Actinides: theDIAMEX–SANEX Concept" Global'01, September 9/13, 2001, Paris, France (2001)

[5] P. BARON, M. LECOMTE, B. BOULLIS, D. WARIN, “Separation of the long livedradionuclides : Current status and future R&D Programm in France”, Global’03, November 16/20,2003, New Orleans, USA (2003).[6] P. BARON, B. LORRAIN, B. BOULLIS, “Progress in partitioning: activitites in ATALANTE”,9 th OECD/NEA %Information Exchange Meetings on Actinide and Fission Product Partitioning andTransmutation, September 25/29, 2006, Nîmes, France, (2006)

PUMA - PLUTONIUM AND MINOR ACTINIDESMANAGEMENT IN THERMAL HIGH-TEMPERATUREREACTORSJ.C. KUIJPER 1NRGWesterduinweg 3, P.O.Box 25, NL-1755 ZG Petten – The NetherlandsABSTRACTThe PUMA project, a Specific Targeted Research Project (STREP) of the European UnionEURATOM 6th Framework Program, is mainly aimed at providing additional keyelements for the utilisation and transmutation of plutonium and minor actinides (neptuniumand americium) in contemporary and future (high temperature) gas-cooled reactor design,which are promising tools for improving the sustainability of the nuclear fuel cycle. PUMAwould also contribute to the reduction of Pu and MA stockpiles and to the development ofsafe and sustainable reactors for CO 2 -free energy generation. The project runs fromSeptember 1, 2006 until August 31, 2009. PUMA also contributes to technological goals ofthe Generation IV International Forum. It contributes to developing and maintaining thecompetence in reactor technology in the EU and addresses European stakeholders on keyissues for the future of nuclear energy in the EU. The paper presents an overview ofplanned activities and preliminary/expected results of the PUMA project.1. IntroductionThe sustainability of the nuclear fuel cycle and the reduction of plutonium (Pu) and Minor Actinides(MA) stockpiles are key issues in the definition of the future nuclear energy mix in Europe. The HighTemperature gas-cooled Reactor (HTR) can fulfil a very useful niche for the purposes of Pu and MAincineration due to its unique and unsurpassed safety features, as well as to the attractive incentivesoffered by the nature of the coated particle fuel.No European reactor of this type is currently available, but there have been two test reactors(DRAGON, AVR) and one 300 MW el demonstrator (THTR) successfully operated creating thesubstantial know-how, which is still available. Considerable interest in the HTR is growinginternationally including projects to construct modular HTR prototype reactors in China and in SouthAfrica. In Europe, further development of HTR is undertaken within the “Reactor for Process Heat,Hydrogen and Electricity (RAPHAEL)” Integrated Project of the European Union 6th FrameworkProgram (EU FP6) [1].Apart from the inherent safety features offered by this reactor type, the nature of the CP fuel offers anumber of attractive incentives. In particular, it can withstand burn-ups far beyond that in either LWRor FR systems. Demonstrations of Pu-burning as high as 75% fissions in initial metal atoms (FIMA)have been achieved in former tests. In addition, the coated particle itself offers significantly improvedproliferation resistance, and finally with a correct choice of the kernel composition, it can be a veryeffective support for direct geological disposal of the spent fuel without significant Instant ReleaseFraction.The PUMA project is a Specific Targeted Research Project (STREP) within the EURATOM 6thFramework Programme (EU FP6) and runs from September 2006, until August 2009 [2]. The projectconsortium consists of 15 organisations from 8 European countries and 1 from the USA, as indicatedin Table 1. This paper presents an overview of planned activities and preliminary/expected results ofthe PUMA project.1 On behalf of the EU FP6 PUMA consortium (see Table 1)

Table 1: EU FP6 PUMA Consortium.PartnerNRG - Nuclear Research & consultancy Group - The NetherlandsAGH - University of Science and Technology of Cracow - PolandBN - Belgonucleaire - BelgiumCIRTEN - ItalyEDF - Electricité de France - FranceGA - General Atomics - USAUSTUTT - University of Stuttgart - GermanyJRC-ITU - Joint Research Centre - Institute for TransUranics - EUKTH - Royal Institute of Technology - SwedenLISTO bvba - BelgiumNEXIA Solutions - UKNNC - UKLGI - LaGrange Innovation - FranceDUT - Delft University of Technology - The NetherlandsFZJ - Forschungszentrum Juelich - GermanyAREVA - FranceMain contact person(s)J.C. Kuijper (coordinator)J. CetnarS. Shihab, G. TouryN. Cerullo, G. LomonacoE. GirardiF. VenneriW. BernnatJ. SomersJ. Zakova, J. WalleniusL. Van Den DurpelT. AbramD. MillingtonV. ChauvetJ.L. Kloosterman, J. JonnetH. WernerC. Trakas2. PUMA objectivesComplementary with other initiatives, the PUMA project aims at providing key results for theutilisation and transmutation of plutonium and minor actinides in HTRs. These results should furtherqualify the HTR design as a promising tool for the development of safe, sustainable and CO 2 -freeenergy generation with a high thermal efficiency.A number of important issues concerning the use of Pu and MA in gas-cooled reactors have alreadybeen studied in other projects, or are being treated in ongoing projects, e.g. as part of EU FP6. These,and other, earlier projects show favourable characteristics of HTRs with respect to Pu burning andsubsequent disposal of HTR spent fuel e.g. with regard to leach resistance and lower decay heat. It hasto be mentioned that the CP retains not only residual MA within the final repository but also the longlivedfission products, which are much more soluble and mobile than MA. However, further steps arerequired to demonstrate the potential of thermal HTRs as Pu/MA transmuters based onrealistic/feasible designs of CP Pu/MA fuel. Therefore, the overall objective of the PUMA project is toprovide additional key elements for the utilisation and transmutation of plutonium and minor actinidesin contemporary and future (high temperature) gas-cooled reactor design. High temperature reactorsare promising systems for improving the sustainability of the nuclear fuel cycle, for reducing Pu andMA stockpiles and to developing safe and sustainable reactors for CO 2 -free energy generation.For that reason, core physics of Pu/MA fuel cycles for HTRs will be investigated to optimise the CPfuel and reactor characteristics and to ensure nuclear stability of a Pu/MA HTR core.It is also envisaged to optimise the present Pu CP design and to explore the feasibility of MA fuel.New CPs will be designed that can withstand very high burn-ups and are well adapted for disposalafter irradiation. The project benefits greatly from access to past knowledge from Belgonucleaire’s PuHTR fuel irradiation tests of the 1970’s, and also secures access to materials made at that time.(Very) High Temperature Reactor V/HTR Pu/MA transmuters are envisaged to operate in a globalsystem of various reactor systems and fuel cycle facilities. Fuel cycle studies are envisaged to studythe symbiosis between LWR, GCFR and ADS, and to quantify waste streams and radiotoxicinventories. The technical, economic, environmental and socio-political impact shall be assessed aswell.

The respective activities concerning these three subjects are being carried out in the three main WorkPackages of the PUMA project, which will be described in the following sections.3. Core physics of Pu/MA loaded V/HTR systemsThe first PUMA Work Package is concerned with the core physics, including transient behaviour, ofPu/MA loaded HTRs. The main objectives of this Work Package are:• Demonstration of the full potential of contemporary and (near) future HTGR designs toutilise/transmute Pu and minor actinide fuel within the constraints of safe operation, and basedupon realistic assumptions concerning the fuel composition.• To identify necessary additional qualification of the tools employed for the assessment of Pu/MAloadedHTGR systems, and identify opportunities to obtain experimental data on which suchadditional qualification can be based. Such opportunities will e.g. be sought in cooperation withother EU FP6 projects as well as with external organisations, such as the American GT-MHRprogramme (through partner GA), as well as British (through partners NEXIA and NNC) andRussian institutes (specifically the ASTRA critical facility at the Kurchatov Institute, Moscow,through ISTC and IAEA/INPRO), and the OECD/NEA. The latter is to ensure that descriptions onwhich additional code qualification exercises are to be based, are well-defined and complete. Forthe code qualification (benchmark) exercises the intention is to adhere to OECD/NEA IRPhErequirements [3].However, the main task with this particular Work Package is concerned with the assessment of severalHTR system/fuel/fuel cycle combinations. Based on results obtained in particular in EU FP5 projectsHTR-N/-N1 [4,5], the core physics work in PUMA will establish optimised Pu/MA transmutationcharacteristics in HTRs. Transient analyses will seek to demonstrate the nuclear stability and safety ofthe optimised reactor/fuel designs. The results will also comprise measures on other performance andsafety-related parameters, such as the fast fluence in the CP coatings and the helium production inHTR coated particles, as well as an assessment of proliferation resistance. Reference systems for thesestudies are contemporary representatives of the two main HTR designs, viz. the PBMR-400 [6,7] forthe pebble-bed (continuous reload) type, and the GT-MHR [8] for the prismatic block type, loadedwith Pu-based (U-free) fuel. Studies include:• Pu/MA deep burn in pebble-bed V/HTR. This also includes investigations on the utilisation ofIMF, such as (Zr,Y,Am)O 2 /(Zr,Y,Ce)O 2 -x or (CeAm)O 2 -x;• Th/Pu fuel cycle in pebble-bed V/HTR;• Pu/MA deep burn in prismatic V/HTR;• Th/Pu fuel cycle in prismatic V/HTR;• Reactivity transients of Pu/MA fuelled V/HTR;• Integrated LWR-HTR-GCFR symbiotic fuel cycles.Some preliminary results have been obtained on the use of (U-free) Pu-containing coated particle fuelin a contemporary design of the PBMR-400 pebbled-bed HTR [9], operating in continuous multi-passre-load fuelling mode. This has also been addressed by other recent work [6], although mainlyfocusing on Pu destruction characteristics. For the application as Pu-burner it was assumed that eachfresh fuel pebble contains 2 g of first generation Pu in coated particles with a kernel diameter of240 μm (see Table 2). Noburnable poison is envisaged.The fuelling scheme employedis the continuous on-line multipassmethod similar to thedesign used in the GermanModul reactor (inMEDUL - multipass- refuelling mode). Freshfuel elements are added to thetop of the reactor while usedTable 2: Characteristics of Pu-containing PBMR fuel pebbles.DiameterDiameter of fuelled zone6.0 cm5.0 cmKernel diameter/material 0.24 mm / PuO 2Fuel loadingPu vectorPu238/239/240/241/242/2.0 g Pu tout initial per pebble2.59/53.85/23.66/13.13/6.78 wt%fuel pebbles are removed at the bottom to keep the reactor at full power. On average, each fuel pebblemakes about six passes through the reactor before being finally discharged to the spent fuel storage

tanks with a target burn-up of about 760,000 MWd/t. This target value originates from experimentalevidence (Peach Bottom) that PuO x coated particle fuel can withstand irradiation up to these levels[10]. The fuel handling system consists of a core-unloading device in each of the three de-fuellingchutes from where the fuel is moved to the burn-up assaying equipment. After the burn-up has beendetermined, the fuel is routed either to the spent fuel tanks or back to the core, depending on its burnup.The fuel spheres are re-loaded to the top of the core through three fuelling lines.To arrive at the equilibrium steady state, 1500 fuel balls, with 3000 g initial Pu, were circulated perday in the simulation. To speed up the convergence to equilibrium at least 125 fresh fuel balls wereadded on top of the bed. After about 600 days the first pebbles in the unloaded fuel at the bottom hadreached the nominal burn-up of 760 MWd/kg and were discarded. This amount was replenished at thetop by fresh balls as well. After about 8000 days the amount of discarded fuel at the bottom and addedfresh fuel at the top becomes equal, leading to a stable non-changing core inventory. At thisequilibrium state about 250 fresh pebbles per day were needed.It should be noted that this situation can be considered as equilibrium in the strictest sense, as thespatial distribution of all fuel-related nuclides has become time-independent. The shortest time framein which this situation can be established is governed by the time required for the fuel to reach theenvisaged discharge burn-up of 760 MWd/kg. At an average rating of about 0.442 MW/kg, this wouldtake about 1720 days. The actual time depends on the particular run-in scenario, and this may beconsiderably longer, because of oscillatory behaviour, as is also the case in this particular example.Table 3 shows the finallydischarged amount ofactinides, relative to freshfuel. The dischargedamount of Pu reduces to14.6 % of the originalamount, while the totalamount of actinides (U upto Cm) is reduced to23.5 %. Table 4 shows thePu-vector for fresh anddischarged fuel. The fissilePu-fraction is reducedfrom 67 % to less than20 %.Table 3: Pu-burning capabilities (masses as a fraction of initialheavy metal mass)Burn-up Pu Cm Am U Np(MWd/kg)0.0 1.0 0.0 0.0 0.0 0.07.600E+02 1.457E-01 4.859E-02 3.971E-02 6.711E-04 6.817E-06Table 4: Pu vectors for fresh and discharged fuel (masses as afraction of initial heavy metal mass)Burn-up Pu-238 Pu-239 Pu-240 Pu-241 Pu-242(MWd/kg)0.0 2.59E-02 5.385E-01 2.366E-01 1.313E-01 6.78E-027.600E+02 1.598E-02 1.887E-03 7.025E-03 5.372E-03 1.155E-014. Optimisation of Pu/MA loaded V/HTR fuel designFor transmutation purposes, ultra-high burn up plutonium (Pu) and Minor Actinides (MA)-basedcoated particle (CP) fuels will be needed and will require the development of novel fuel kernelscapable of either being reprocessed or suitable for direct geological disposal. The manufacture of HTRfuels incorporating plutonium and minor actinides probably will be more difficult than for thecorresponding pellet type fuels. There are advantages to be gained by the use of inert matrix based fuelkernel to dilute the transuranic components [11,12]. The JRC-ITU has developed a method based onthe infiltration of porous precursor kernels, which overrides some of the kernel production difficultiesand cost issues. The fabrication process can be simplified and the extra volume generated in the bufferlayer (for a given Pu mass and buffer layer thickness) can accommodate more fission gas, and higherburn-ups can be achieved.The traditional TRISO layers used in U-based fuel are taken for PUMA fuels. There is an ongoingdiscussion on the level of separation of the actinides at the reprocessing plants. Today, U and Pu areseparated individually, with the minor actinides going for disposal in the vitrified waste. GenerationIV takes the extreme position that all actinides (U, Pu, MAs) should be subjected to group separation.For proliferation resistance, this is a perfectly laudable approach, but not necessarily the most practicalsolution. For transmutation fuels today, the general consensus is to avoid U as a matrix, as it generatesfurther higher actinides on irradiation. Thus, its presence automatically leads to a decrease in thetransmutation rate. Fuels for other transmutation reactor concepts, e.g. the accelerator driven system(ADS), rely on the inert matrix concept. In addition, there is a debate on the choice of Pu and MAseparation (partitioning) process with regard to pyro- or aqueous reprocessing.

From the fabrication standpoint, it is preferred to separate the Pu, so that the bulk of the fuel can bemanufactured in a facility without excessive shielding, just as is done today in MELOX or SMP MOXfuel fabrication plants in France and the UK, respectively. In PUMA, the U/Pu separation philosophywill be followed and the fuels for PUMA should be considered as heterogeneous and will be of theform:• PuO 2 (probably with NpO 2 )• MAO 2• (Pu,MA)O 2 , whereby the Pu:MA ratio is significantly lower than in spent fuel.In principle Cm could be included in the fuel, but Cm management must be considered by alternativemeans, namely Cm storage to permit its natural decay to Pu, and then later transmutation of theresulting Pu.One real advantage of the HTR is the low power density of the core. This results in an abundance ofTable 4: Diluting inert matrices for HTR actinide (An = Pu, Am, or Np) fuel.MatrixFuel/targetYSZ Zr 0.85 Y 0.15 O 2 (Zr 0.85 Y 0.15 ) 1-z An z O 2YSZ-Ce (Zr 0.85 Y 0.15 ) 1-y Ce y O 2 ((Zr 0.85 Y 0.15 ) 1-y Ce y ) 1-z An z O 2Ceria CeO 2 Ce 1-z An z O 2PYR (pyrochlore) Zr 2 Ce 2 O 7 (Zr 2 Ce 2-z An z )O 7Zr doped Y 2 O 3 (Y,Zr) 2 O 3 (Y,Zr,An z ) 2 O 3Graphite C C + AnO 2Table 5: Novel kernel characteristics.Matrix Zr 0.85 Y 0.15 O 2 (Zr 0.85 Y 0.15 ) 1-yCe y O 2-xCeO 2 /Ce 2 O 3 Zr 2 Ce 2 O 7 Y 2 O 3 C(graphite)TD (g.cm -3 ) 5,96 7,132/6,86 6.173 5,01 2,25Mass in 500 µm 3,47E-4 * 4,15E-4/ 3,59E-4 2.92E-4 1,31E-4kernel (g) (95%TD)3,99E-4* Depends on the Ce content selectedspace, so that fuel design concepts are more flexible than for other reactor systems. The abundance ofspace makes it easier to accomodate the helium and fission gas release and the swelling of the kernels.These three aspects are very important to be able to achieve the required very high burn-up.Although the reference PuO 2 kernel has a diameter of 200 µm, this is a poor choice, and is basedlargely on historical reasons, even though it is still considered today for the GT-MHR in Russia. Amuch better option is based on actinide infiltration into a porous precursor kernel. This minimiseswastes and reduces the handling steps involving MAs. Some of the most promising materials are givenin Table 4. Carbon is still under consideration, and could be used in the infiltration process, once asuitable means to produce suitable precursors is developed.The premise in designing the diluted, coated particle with any of these inert matrices relies on taking afixed actinide mass in each kernel. This mass corresponds to the volume occupied by thecorresponding 200 µm actinide oxide kernel. In this way the power in each kernel is kept at or belowthat of a PuO 2 kernel. Sticking with tradition, justified by the excellent results obtained with NUKEMproduced UO 2 fuel, the diluted kernel size is kept at 500 µm. At this size, one can expect the in pilefuel performance, i.e. that of the sealing coating layers, to be similar to that of a UO 2 kernel, operatedat the same power/kernel. The characteristics of these diluted kernels for Pu and MA fuels are given inTable 5.Loading of fuel elements homogeneous in (Pu,MA)O 2 fuel could be handled in the same way as in aconventional UO 2 fuelled HTR (pebble bed or prismatic). Ideally the fuel should be homogeneous at

the level of the kernel, i.e. a genuine, but diluted (see Table 4), (PuMA)O 2 kernel. Given theimpracticalities of fabricating group reprocessed transuranium kernels and the pollution of a singlelarge-scale facility with minor actinides, it is preferable to maintain compositional heterogeneity at thecoated particle level, i.e. the spherical fuel elements (SFE) or cylindrical compacts should contain bothdiluted PuO 2 as well as diluted MAO 2 coated particles in the required proportions. An additionalquality control (QC) step to control the Pu and MA content would be required, as well as the need todevelop fabrication procedures for two kernel types. In this way, a quasi homogenous compact or SFEis maintained. In the case of prismatic cores, the ratio of Pu versus MA containing kernels in acompact can be selected to obtain an optimised power distribution and temperature gradient within thecore.Yttria stabilised zirconia (YSZ) and ceria have been tested in previous irradiation tests using pelletfuel. Ceria has an added chemical advantage, as it can be incorporated in the fuel in its trivalent (III)form so that oxygen released through fission of the actinides can be gettered via the Ce(III/IV) buffer.Belgonucleaire used graphite as a diluting matrix. The most readily reprocessible and proven matrixremains UO2, which depending on the strategy could be envisaged as a PUMA fuel, but is not a firstchoice.Further activities within the PUMA fuel Work Package include the further development of fuelperformance models appropriate for HTR-type coated particle fuel. These models focus on theprediction of stresses induced in coated particles during irradiation due to the pressure build-up ofvolatile/gaseous fission products, irradiation-induced dimensional changes and creep, and thermalexpansion of the layers. Contrary to uranium-based kernels, helium is produced in higher quantities inPu and MA-based kernels and this additional source of pressure has to be taken into account in thefuel performance codes of the PUMA project. The stress fields in all particles are the input for thecalculation of the fuel failure probability, hence of the release fraction of volatile species present in thekernel. This failure probability will be calculated for several design parameters of the TRISO particlesand fuel pebbles, like the fuel kernel size and the thicknesses of the various layers, the actinide densityin the fuel kernel, the concentration of TRISO particles in a pebble, pebble flow velocity in the reactorcore, etc. Such a complete calculation scheme requires the coupling of mechanical, thermal-hydraulicsand neutronics codes for a given reactor core design.The PUMA Fuel Work Package integrates previous knowledge gained in GA and Belgonucleaireprogrammes with new investigations on the optimal kernel composition. The programme makes use ofBelgonucleaire archive PuO 2 coated particles to investigate the behaviour of helium in coated particlefuel. In contrast to UO 2 fuels, this is an issue of great importance for MA bearing fuels and must beincorporated in the fuel performance models being used to determine the optimal coated particlegeometry. Finally, the design and safety study for a PUMA irradiation is being made. Independentlyof PUMA, the JRC-ITU is installing a Pu coated particle production facility, and it is planned thatPUMA fuels will be irradiated in a subsequent EU FP7 programme.5. Role for V/HTR systems with a TRU management function in Europe’s futurenuclear parkHTR reactors have a potential in a mixed nuclear park, where fast reactors would be deployed to gainfuel sustainability, whereas HTRs could provide the means to transmute the MAs. Current day lightwater reactor (LWR) UO 2 fuel is reprocessed with some of the Pu returning to the LWR in the form ofMOX. Remaining Pu not used for LWR MOX can be fabricated into HTR fuel. In addition, the Puextracted from LWR-MOX spent fuel can also be fabricated in the form of HTR fuel. Minor actinidesfrom the LWR UO 2 and MOX fuels will be separated and be processed for irradiation in the HTR.Irradiated HTR fuel can be considered for direct disposal or even reprocessing with the Pu and MAconstituents being once again prepared for HTR irradiation. Other similar scenarios based on Gen IVfast reactors rather than LWRs can also be envisaged. In either case dedicated transmutation HTRs canbe considered as a part of the nuclear park.The assessment of the role of V/HTRs delivering energy products (i.e. electricity, hydrogen andprocess heat in general), especially when fulfilling a TRU-management role, in a time-evolvingEuropean (nuclear) energy park is the prime topic addressed within the third Work Package of thePUMA-project. While the other PUMA Work Packages essentially address the scientific and technical

aspects of reactors and fuels respectively, this WP aims at providing a holistic assessment of thepotential role(s) of V/HTRs by focusing on four main facets, i.e.:• The technological impact of V/HTRs fulfilling a TRU-management function on the whole nuclearenergy park and especially on the nuclear fuel cycle. Typical questions addressed are thetechnological feasibility and expected performance of the fuel cycle operations needed toaccomplish the TRU-management mission with V/HTRs as well as the synergy between V/HTRswith other reactors (LWRs, FRs) within Europe’s nuclear reactor park;• The economics and the potential market penetration for such V/HTRs and especially the economicimpact on V/HTRs when performing a TRU-management role in future nuclear energy systems;• The environmental facet looks into a life cycle inventory analysis for (V)HTRs and especially tothe fuel cycle aspects such as (secondary) waste arising, various waste management options,separated fissile material inventories and losses in the nuclear fuel cycle, as well as into nonnuclearmaterial streams;• The socio-political facet focuses essentially on the proliferation risk of nuclear energy systemsinvolving V/HTRs with TRU-management role.The outcome of this Work Package is a characterisation and uncertainty/sensitivity analysis of thetechnical feasibility, economic viability and environmental friendliness of various V/HTR-designs, asdefined by the other PUMA Work Packages.The assessment of the potential of V/HTRs in their TRU-management function in future nuclearreactor parks will be based on internationally accepted assessment methodologies as has been recentlyproposed by the Generation-IV International Forum (GIF) activities [13] and IAEA’s INPRO’sAssessment Methodology [14] as well as practiced within the EU FP6 projects RAPHAEL and RED-IMPACT [15], among others. Specifically, a number of transmutation performance indicators will beestablished to enable comparison of different recycling scenarios and V/HTR variants as well as withother non-V/HTR scenarios based on literature review.The activities in this Work Package are essentially grouped into three major tasks with some specificsub-tasks being described in the following.Characterisation of (V)HTR and associated fuel cycle and waste management technologies withregard to the four facets described above. An assessment of the role that V/HTRs may play in aEuropean nuclear reactor park needs to cover the economic, environmental and socio-politicaldimensions in addition to the pure technical feasibility of such reactors. This first task aims atspecifically addressing these sustainability dimensions based on the technical information provided bythe other PUMA Work Packages and collected from EC and other international projects on gas-cooledreactors with the main outcome being an as complete as possible description of the reactor and fuelcycle technologies as well as the technological challenges and limitations that may be faced indeploying such V/HTR-scenarios.Simulation of (European) nuclear energy systems providing a holistic assessment framework. Basedon some (nuclear) energy demand scenarios from authoritative studies (IIASA, WEC, EC, ...), thedeployment of various nuclear energy systems incorporating V/HTRs will be analysed. As both gascooledreactors may play a specific and important role with respect to demonstrating the sustainabilityof nuclear energy, and especially also in the TRU-management for the whole nuclear energy system, adynamic analysis of the evolution from today’s European reactor park to some future reactor parkswill be analysed. Such dynamic analysis allows proper assessment of the mass-flows, waste arising,separated TRU inventories, the delay times in deployment, the economic impacts and so forth, e.g.:The timing of introduction of V/HTRs according to different fuel cycle options is a function of theevolution of the European reactor park and fuel cycle infrastructure and will therefore have to copewith different stocks of spent fuel and (separated) TRUs. The timing of the introduction of V/HTRswill therefore impact on the evolution of spent fuel, TRU-stock and other indicators as well as definethe kind of (V)HTR core management needed in order to reduce such spent fuel and TRU-stockarising.The V/HTR fuel cycle infrastructure technologies (losses, transit times, …) will define the workinginventories of TRUs in the fuel cycle and thus also the pace of V/HTR introduction in the Europeanreactor park.

Various dynamic simulation codes will be used, i.e. OSIRIS, ORION [16] and DANESS [17] andmaybe others depending on availability. A benchmark exercise will also be undertaken to verify themass-flow, economic and waste management simulations provided by these codes.An uncertainty/sensitivity analysis resulting in a mapping of the potential of V/HTR in future nuclearreactor parks and identification of the main drivers enabling such future role for V/HTRs with focuson identifying and analysing the prime important technological objectives (and associateduncertainties) to be achieved. This involves an iterative process of information exchange and scenarioanalysis based on new information gained from other Work Packages of PUMAThe uncertainty/sensitivity analysis will be covering aspects of:• Fuel and core management in V/HTRs, i.e. burn-up, core-management, fuel type, cycle length, etc.• Fuel cycle: cooling times, reprocessing and fabrication parameters, waste management options,etc.• Economic parameters and especially the impact of technological choices induced by fuel/coreoptions on expected capital, O&M and fuel cycle costs.The nuclear energy system scenarios investigating the potential future roles for (V)HTRs, i.e. thescenario families being considered are the following:• European LWR-park without (V)HTRs as reference scenario (first family of scenarios). Startingfrom the existing reactor park in Europe (i.e. taking into account the anticipated shutdown ofreactors after approximately 50 years lifetime), new Generation-III LWRs are introduced with acontinuation of once-through cycle (OTC) and mono-Pu recycling as fuel cycle option;• European LWR-park without (V)HTRs with a TRU-management objective based on MOX-EUfuel cycle is studied in the second family of scenarios. This scenario family intends to show the‘maximum’ achievable TRU-management capability of a LWR based reactor park allowing thento compare with a LWR+(V)HTR reactor park where the (V)HTRs serve a TRU-managementrole;• The third scenario family involves a mixed LWR and (V)HTR reactor park with the (V)HTRsusing uranium-fuel in OTC-mode, i.e. no TRU-management role. Various reactor park variantswith changing LWR and (V)HTR fractions shall be investigated based on energy market potential;• The fourth scenario family is comparable to the previous scenario except that (V)HTRs serves aTRU-management role. Different variants are considered, i.e.:1. (V)HTRs in Pu deep-burn mode with the Pu coming froma. first or second generation UOX/MOX from the LWRs and/or;b. from recycled CP-fuels from (V)HTRs.2. (V)HTRs in TRU deep-burn mode with the TRUs coming froma. First or second generation UOX/MOX from the LWRs and/or;b. From recycle CP-fuels from (V)HTRs.• The fifth family of scenarios involves mixed LWR+(V)HTR+(S)FR reactor parks intended toshow the potential synergies between the reactor types and highlighting the potential role for(V)HTRs as well as comparison of (V)HTRs with (S)FRs with respect to TRU-managementThe Scenario Work Package of PUMA will also aim at integrating as much as possible the results andexperience gained in previous or still ongoing EC projects such as RAPHAEL, RED-IMPACT [15],GCFR as well as from other international projects especially with the objective to integrate bestpractices developed worldwide in performing this holistic assessment of V/HTRs futures.6. ConclusionSummarizing, the PUMA project’s main goal is to provide additional key elements for the utilisationand transmutation of plutonium and minor actinides in contemporary and future (high temperature)gas-cooled reactor designs, which are promising systems for improving the sustainability of thenuclear fuel cycle, reducting Pu and MA stockpiles, and for the development of safe and sustainablereactors for CO 2 -free energy generation.The investigation on core physics aims at optimising the CP fuel and reactor characteristics, andassuring nuclear stability and safety of a Pu/MA V/HTR core. Some promising results have beenobtained already.New CP designs will be explored in order to withstand very high burn-ups and obtain optimaladaptation for disposal after irradiation. In particular, helium production in Pu- and MA-based fuel, as

well as residual thermal heat loads to a final repository, will be assessed and supported by experimentand/or calculation. Fuel irradiation performance codes, developed and used by several organisations,will permit convergence on optimised design criteria.The impact of the introduction of Pu/MA-burning (V)HTRs on the fuel cycle and future nuclearenergy mix will be assessed, with focus on the fuel cycle symbiosis with future nuclear energysystems in Europe (LWRs, Fast Reactors, ADS). This assessment involves the quantification of wastestreams and radiotoxic inventories as well as the technical, economic, environmental and sociopoliticalimpacts of introducing (V)HTRs with a TRU-management mission in a future nuclear park.PUMA also contributes to technological goals of the Generation IV International Forum. It contributesto developing and maintaining the competence in reactor technology in the EU, and addressesEuropean stakeholders on key issues for the future of nuclear energy in the EU.7. AcknowledgementThe EU FP6 Project PUMA is partly financed by the Commission of the European Communities,EURATOM 6th Framework Program, contract no. 036457, signed October 3, 2006.8. References1. M.A. Fütterer, “RAPHAEL: The European Union’s (Very) High Temperature ReactorTechnology Project”, Proc. ICAPP’06, Reno, NV, USA, June 4-8, 2006.2. “PuMA - Plutonium and Minor Actinides Management by Gas-cooled Reactors”, EuropeanUnion Sixth Framework Program contract no. 036475, signed October 3, 2006.3. “Evaluation Guide for the International Reactor Physics Experiments Evaluation Project(IRPhEP)”, Document NEA/NSC/DOC(2006)2, Revision 8.9, OECD Nuclear Energy Agency,January 20, 2006.4. “HTR-N - High-Temperature Reactor [Nuclear] Physics [, Waste] and Fuel Cycle Studies”,European Union Fifth Framework Program contracts no. FIKI-CT-2000-00020 and FIKI-CT-2001-00169.5. J.C. Kuijper et al., “HTGR Reactor Physics and Fuel Cycle Studies”, Nuclear Engineering &Design, 236, pp. 615 - 634 (2006).6. E. Mulder & E. Teuchert, “Plutonium disposition in the PBMR-400 High-Temperature Gas-Cooled Reactor”, Proc. PHYSOR 2004, Chicago, IL, USA, April 25-29, 2004.7. OECD/Nuclear Energy Agency, “Pebble-Bed Modular Reactor coupled neutronics/thermalhydraulics transient PBMR-400 design”, Document NEA 1746/01, September 2005.8. “GT-MHR, Inherently Safe Nuclear Power For The 21st Century”, General Atomics, SanDiego, CA, USA, http://gt-mhr.ga.com9. J.B.M. de Haas, J.C. Kuijper & J. Oppe, “Burn-up and Transient Analysis of a HTR-400Design Loaded With PuO 2 ”, Proc. HTR2006, 3rd International Topical Meeting on HighTemperature Reactor Technology, October 1-4, 2006, Johannesburg, South Africa.10. D. Alberstein et al., “MHTGR Plutonium Consumption Study Phase II Final Report”, TechnicalReport GA/DOE-051-94, General Atomics, PO BOX 85608, San Diego, CA 92186-9784, April1994.11. J. Baier, H. Bairiot, J. Vangeel, R. van Sinay, EURATOM report EUR 5066 e, 1974.12. J. Somers, A. Fernandez, Progress in Nuclear Energy, 48(2006)259.13. US-DOE, Generation-IV International Forum, www.nuclear.gov14. IAEA, “International Project on Innovative Reactors and Fuel Cycles”,http://www.iaea.org/OurWork/ST/NE/NENP/NPTDS/Projects/INPRO/index.html15. “RED-IMPACT”, European Union Sixth Framework Program contract no. FI6W-CT-2004-002408, http://www.red-impact.proj.kth.se/16. A Worrall, “The role of fuel cycle modelling and the capabilities within BNFL’s Research &Technology organisation”, Proceedings Global 2003, New Orleans, USA, 16-20 Nov 2003,p162917. L. Van Den Durpel, A. Yacout, “Dynamic Analysis of Nuclear Energy System Strategies”,Users Manual, Argonne National Laboratory, Argonne, February 2004,http://www.daness.anl.gov/

Session 19.2.1Waste and transport

WDC – ADVANCED SYSTEM FOR CHARACTERIZATIONOF ALPHA-BEARING WASTE IN 200L AND 400L DRUMS:PERFORMANCES AND LESSONS LEARNED FROM THEFIRST INDUSTRIAL MEASUREMENT CAMPAIGNSA.LIBENS, M.VANDORPETecnubelGravenstraat 77,B 2480 Dessel, BelgiumJ.M. CUCHETBelgonucleaireEuropalaan 20,B 2480 Dessel, BelgiumABSTRACTThe Waste Drum Characterization installation was originally developed for assay of alphabearingwaste in standard 200 l (55 gallons) drums during the dismantling operations of theSiemens mixed-oxide (MOX) facility in Hanau (Germany). That installation was validatedand qualified by the German authorities, its main performances being:- Counting efficiency for coincident neutrons: app. 1%;- Lowest Limit of Detection (LLD): 75 mg 240 Pu eq ;- Pu content per drum: up to 100 g tot. (35 g 240 Pu eq )- Measurement duration: app. 20 minutes.The success of this system, a passive neutron coincidence counter combined with a highresolution gamma spectrometer, led to the radiological characterization and qualification ofabout 1,700 drums during the period 2001 – 2004.In 2005, after completion of the dismantling operations of the Siemens MOX facility,Tecnubel took over the WDC installation which could be used in the frame of the futuredismantling of the Belgonucleaire’s MOX plant in Dessel (Belgium), which can becompared to the Siemens one.This second (and new) life for the WDC means that it must be rigorously retested andvalidated against the Belgian authorities requirements.Furthermore and additional to the future use in the Belgonucleaire’s facility, Tecnubelwere faced to new challenges, namely:- Assay of 400 l drums together with the 200 l packages;- Determination of the real LLD taking into account the background in different Belgiannuclear facilities, the determination of a value of ~5 mg 240 Pu eq being an objective;- Assay of mixed alpha/beta-gamma wastes;- Transportability of the WDC from one plant to another;- Assistance to different nuclear operators for the licensing of the WDC for their ownwaste types.This paper describes the installation itself and its performances, presents the difficultiesencountered during the new challenge and the results of the performed revalidation tests; itgives the perspectives and objectives on short time as well.1. Introduction

1.1 Scope of the installationThe WDC is currently able to determine the amounts of U and Pu present in 200 l waste drums, aswell as the isotopic composition for each element. It is also able to give the order of magnitudes of the241 Am and 237 Np masses. The principles of the measurement, the hardware and the software of theinstallation are described below.1.2 Measurement principlesPassive neutron measurement associated with a gamma measurement is an optimal choice to measurethe amounts of the mentioned isotopes. Only the coincident neutrons are taken into account during ameasurement, as the total number of neutrons is dependent on (α,n) reactions, which rate is verysensitive to the kind of element present inside the matrix of the waste. The WDC system allowstherefore a measurement of 240 Pu eq without an accurate knowledge of the physicochemical propertiesof the matrix. A passive gamma measurement can be performed during the neutron measurement,giving the possibility to compute the isotopic composition. The gamma rays emissions comes mainlyfrom the external part of the waste (a few centimetres below the surface of a medium-density matrix)therefore a total efficiency for the particular configuration of the detector with respect to a specificwaste cannot be found. A relative efficiency (see further) will be used to determine the relativeabundances of the isotopes.To achieve the assessment of the various isotopes’ masses, the measurement must be performed inseveral steps. The 240 Pu eq mass is first measured by neutron counting. During the neutron acquisition, agamma measurement allows the WDC to store a spectrum in a dedicated program called Inter Winner.The spectrum is then transferred to the MGA code, which computes the isotopic composition. Themasses of the various isotopes are then computed form the 240 Pu eq and the isotopic composition.2. Description of the installation2.1 Hardware main parts and functionsFrom an observer point of view, the 200 l drums to be measured are conveyed into an examinationcell. The measuring cell is a cell whose walls are made of polyethylene (PET) covered with a layer ofCd. Inside the walls of this cell, 37 3 He detectors are installed to capture the thermal neutrons, as ameans of estimating the 240 Pu equivalent content in g. The neutron detectors are associated in 7channels, which are grouped in 3 categories (top, bottom, walls) allowing to find the position of theepicentre of the contamination inside the drum. The electronics associated with the 3 He detectors cangive the total count rate for each channel and the coincident count rate for the whole cell. 16 positionsare preset in the Winner Neutron software managing the neutrons measurement and it chooses theclosest position to the effective one. Once the position is found the software Winner Neutron canperform the mass correction associated to this position.In order to determine the matrix present inside the drum and its filling level, an external 252 Cf source isused. Above the cell, a PET castle is installed to insure the safe storage of the 252 Cf source. Afterclosing of the measurement cell, it can be lowered along the side of the drum (high position or lowposition) to measure the neutron transmission coefficient of the matrix, as well as the filling height.For the determination of the isotopic composition of the radioactive contents, two liquid-nitrogencooled gamma detectors made with hyper pure Ge are present outside the cell. The code computing theisotopic composition, called MGA, meets the entire IAEA requirement to measure fresh or processedU and Pu samples. Furthermore, a table supporting the drums turns during the measurement cycle toobtain a homogeneous gamma measurement. All the measurement devices and the engines arecontrolled by electronic cards which are connected to a PC.

Figure 1: the WDC, rear view during theloading of a calibration drumFigure 2: plan of the measurement cell with Hetubes and Ge detectors location2.2 Efficiency of the system: gamma detectorsAs the efficiency of a gamma detector decreases with energy, correction of peak areas by a relativeefficiency is performed with a quite particular auto-calibration. MGA obtains, from the spectrum data,an efficiency curve comparing the heights of ten peaks of MOX elements to a tabulated relative heightconstructed with the branching probabilities of the gamma rays. This curve depends on the particularconfiguration of the waste and its matrix, taking into account the following absorption parameters:cadmium absorption, MOX auto-absorption and the detector efficiency. Thicknesses of these materialsare computed by MGA as well as the efficiency.Figure 3: the WDC functional scheme2.3 Calibration of the neutron detectorsThe WDC installation has already been calibrated in Germany. However, because of the installation ofa new 252 Cf source inside a new capsule, the matrix and filling height calibrations had to be performedagain. A quick check of the German calibration for epicentre finding was also performed; givingresults in statistical agreement with the ones obtained in Hanau.16 MOX standards (zyrcalloy tubes containing pellets) made at the MOX plant of Belgonucleaire wereused during all the calibrations performed in Belgium, as well as during the tests of the WDC. Twocalibration containers where used, one containing a metallic matrix (steel tubes, 580 kg / m³) and theother a plastic matrix (PVC, 290 kg / m³). The containers are equipped with 4 steel guiding tubes (atthe centre, at half radius, at ¾ radius, and on the periphery) for sources insertion. The vertical positionof the sources was controlled by the use of nacelles.

3. Performances3.1 Tests performed for the Belgian authoritiesThe NIRAS, which is the Belgian federal organisation responsible for the management of radioactivewaste, asked to perform a series of tests in order to qualify the device for operations in Belgium. Thefollowing tests where performed: linearity, reproducibility, precision with a centred source in a fulldrum, precision with source at random position with various filling heights and sensitivity to sourcesdispersions. The conclusion is that the linearity graph measured in Belgium is identical to the oneobtained in Hanau ; the error for the total Pu mass (2σ) is ±10% for the metal drum and ±20% for thePVC drum. Furthermore, the device results are reproducible at less than ±5% for repeatedmeasurements on a drum present inside the cell filled with the most penalizing matrix (PVC). Theexperience gained by Tecnubel in licensing the WDC can be useful in the assistance of potentialclients, helping them to qualify the device fulfilling the demands of their authorities.3.2 WDC transport and availabilityDuring the different phases of assembly in France, commissioning in Germany and testing in Belgium,the WDC has been disassembled, transported and re-assembled completely 4 times. The duration ofthese operations (for a transport within Europe) is more or less two to three weeks and includes ameasurement of the background, a verification of the cell efficiency for real neutrons, the efficiency ofeach channel for total neutrons and the germanium detectors calibration and efficiency curve. TheWDC is therefore a transportable installation available to perform measurement campaigns in Europefor any potential client. The client must ensure to provide the necessary authorizations as a 252 Cfsource is present inside the WDC and MOX or Pu bearing calibration standards must be available nearthe WDC site.Fût PVC353025y = 7,4723x + 0,115R 2 = 0,9988Reals (c/s)201510500 1 2 3 4 5Pu-240 eq. (g)Figure 4: linearity of the WDC for the most penalizing matrix (PVC) and measurement times of 20min.3.3 Performances of the WDC for beta-gamma waste assessmentThe two gamma detectors of the WDC are high quality detectors having a resolution below 700 eV at122 keV. Their gain can be increased to study spectrum regions above 2 MeV, but the efficiency isobviously not optimal above the MeV region. To solve that problem, higher efficiencies can beachieved installing a coaxial germanium detector.The gamma acquisition card doesn’t transfer the spectrum data directly to MGA but rather to aprogram called Inter Winner, which convert the specific format of the spectrum into an ASCII one.Inter Winner has also the capability to calibrate the spectrometer and to set some essential hardwareparameters (pole zero, coarse and fine gains, …). Moreover, Inter Winner has a table of isotopeallowing light elements identification and a procedure for efficiency calibration in order to givequantitative results in term of activities. The WDC has been calibrated several times through Inter

Winner with a wide variety of elements ( 60 Co, 137 Cs, 152 Eu, 241 Am, 88 Y …) and the results always gavesatisfaction.3.4 Detection limits of the WDCDuring the dismantling of the Hanau MOX plant, the background was quite high and the computeddetection limit was 75 mg of 240 Pu equivalent for a 20 min measurement and a 20 min backgroundacquisition. However, with natural background in Belgium (the installation was tested away fromnuclear fuel or waste) this limit is (computed according to the ISO 11929-1 standard) 7,5 mg for a 20min measurement and a 20 min background acquisition. As there are several ways to compute adetection limit, the table below summarizes the main techniques used by the different operators of theWDC and the Belgian authorities. Increasing the measurement time seems according to the data givenabove the easiest method to lower the detection limits.Place BKG Detection limit (mg 240 Pu eq.)(c/s) ISO11929-1CurrieformulaWDMformulaMontigny 0,3 6,07 6,26 3,2Hanau 0,6 8,59 8,77 75EURIDICE 0,46 7,52 7,7 8,594. WDC mechanical improvementsTable 1: LLD’s valuesTo meet the demands of various Belgian nuclear facilities, willing to characterise 400 l drums, therotating self-motorized rolls used to load the drum (see picture 4) will be changed. Smaller diameterones will be installed and the motorisation will be external. A technical design has been performed tomeet the new drums specifications as well as some test of rolls assembly and disassembly. The plansof the improved cell have been prepared and the mechanical modifications are scheduled in the secondhalf of 2007.5. Summary of the technical data sheet and performances5.1 Technical specifications• Turntable to accommodate 200-L and 400-L drums• Detectors :• 37 ³He neutron detectors in 4π configuration• 2 high resolution Ge planar detectors• Neutron measurement software allowing a mass correction by evaluation of the epicentreposition• External neutron source in order to find matrix composition and filling height• MGA software meeting IAEA demands for passive non destructive assay of fissile materials• Inter Winner software allowing to measure amounts of β-γ emitters• PC-based control computer (full Windows compatible)5.2 Performances• LLD : 8 mg 240 Pu eq. in 1.200 second count• HLD : 35 g 240 Pu eq. / 100 g Pu tot.• Counting efficiency for coincident neutrons: app. 1%• Transportability : transferable within 3 weeks• Licensing / certification: presently licensed for use in Germany and Belgium; possibleextensions are analysed.

ENC 2007 – Brussels, Belgium / 16-19 September 2007.The Packaging and Transportation of FUTURIX-FTA Fuel PinsF. L. Yapuncich, J. Raffo, D. Ohayon, L. Mariette, A. M. Ross, S. BrutBusiness Unit Logistics – AREVAD. F. Snedeker, International Nuclear ServicesBritish Nuclear FuelsRisley, Warrington, Cheshire WA3 6ASUnited KingdomS. L. HayesIdaho National LaboratoryPO Box 1625Idaho Falls, ID 83415-6188ABSTRACTThe FUTURIX-Actinide Transmutation Fuels (FTA) project is an international collaborationprogram between the United States Department of Energy (DOE) and the French Commissariat àl’Energie Atomique (CEA). The project deals with irradiation test of experimental fuels to beconducted in the Phenix Reactor located at the CEA-Marcoule site near Avignon, France. Thispaper describes the process for shipment of unirradiated plutonium based fuel pins from theIdaho National Laboratory (INL) to the Phenix Reactor utilizing a French licensed transportpackage; TN-BGC 1. This type of knowledge and experience is vital to the shipment of both fueland radioactive waste material.INTRODUCTIONThe shipment of any radioactive type material includes the three basic elements of: the transportpackage, regulatory compliance, and transport logistics. The transport package must meet thephysical requirements of the specified payload. Regulatory compliance ensures that allapplicable governmental regulations are adhered to. Regulations concerning safety, materialaccountability, and liability are incumbent on the responsible party to understand and follow.The transport logistics dictates the actual mode of conveyance including air, land, and sea andthe associated detailed planning of each transport route.The shipment of the FUTURIX-FTA plutonium based fuel pins presented specific challenges ineach of the three basic transport elements. The specified package, the TN-BGC 1, required aninternal arrangement to contain the fuel pins to prevent damage during the transport and torecord the temperature of the pins which could affect the sodium bonding of the metal fuel. TwoTN-BGC 1 transports were utilized to streamline the regulatory process. The regulatorycompliance issues emanated from the United States, France, and various international regulationsincluding the International Maritime Dangerous Goods Code (IMDG). In addition, international© Copyright TN International 2007 Page 1 on 12

ENC 2007 – Brussels, Belgium / 16-19 September 2007.liability in the form of the Price Anderson Act (PAA) and the Paris Convention Accords wereapplied appropriately throughout the transport. The logistics of the FUTURIX-FTA transportincluded ground arrangements in the United States and France, the ocean transport, and transfersat the ports of Savannah, Georgia and Cherbourg, France.PAYLOADThe FUTURIX-FTA payload consisted of four sodium bonded fast reactor fuel pins. INLfabricated the two metallic alloy fuel slugs and Los Alamos National Laboratory (LANL)fabricated two nitride fuel pellets. At INL, the four pins were clad in stainless steel tubes of0.655 cm outer diameter and 30 cm in length. The pins make use of a metallic sodium bond inthe fuel-clad gap. The metallic sodium will be solid during transportation. The projectedelemental masses of the pins are delineated in table 1.The two metallic alloy fuel slugs fabricated at INL have design dimensions of: solid cylindershaving a nominal diameter of 0.489 cm and a nominal height of 10.0 cm. The two metallic alloyshave a nominal composition of Pu-12-Am-40Zr and U-29Pu-4Am-2Np-30Zr, where alloyconstituents are given in weight-percents.The nitride fuel pellets of two compositions were fabricated at LANL and shipped to INL for usein fabricating of two more experimental fuel pins. The fuel pellets as designed have outerdiameters of 0.489 cm and the pellets are stacked in each of the two fuel pins to a nominal totalfuel column height of 10.0 cm. The two nitride fuels will have nominal compositions of(U 0.50 ,Pu 0.25 ,Am 0.15 ,Np 0.10 )N and (Pu 0.50 ,Am 0.50 )N-36wt%ZrN.Table I. Projected Elemental Masses of Metallic and Nitride Fuels in Shipment.Fuel Elemental Masses (g)Total Total Total Total Total TotalFuel Composition Alloy Np U Pu Am ZrPu-12Am-40Zr (1) 17.8 0.053 0.013 8.678 1.875 7.383U-29Pu-4Am-2Np-30Zr (2) 20.1 0.418 6.954 5.678 0.755 6.340(U0.50,Pu0.25,Am0.15,Np0.10)N (3) 18.7 1.435 8.823 4.565 2.354 0.000(Pu0.50,Am0.50)N-36wt%ZrN (4) 12.2 0.000 0.145 4.614 0.851 6.094TOTAL 68.8 1.906 15.935 23.535 5.835 19.817TRANSPORT PACKAGEThe TN-BGC 1 transport package is utilized internationally for transport of low and highenriched materials up to 95% (powder, pellets, etc). Outer dimensions of the package are 1,800mm × 600 mm × 600 mm (70.9 in × 23.6 in × 23.6 in). The maximum loaded weight is 400 kg(882 lbs). Figure 1 shows a cutaway view of this transport package.© Copyright TN International 2007 Page 2 on 12

ENC 2007 – Brussels, Belgium / 16-19 September 2007.The package is comprised of the following components:1. An inner stainless steel shell defining a useful cavity of 178 mm (7.0 inches) indiameter, and 1,475 mm (58.1 inches) in length2. A resin layer for neutron absorption for the criticality control and dose rate reduction.3. An outer stainless steel shell protecting the resin.4. A Closure Lid using a bayonet system device avoiding the use of bolts.5. An impact limiter plug which protects the Closure Lid.6. An aluminum frame surrounding the cask to facilitate the handling and tie downsystem. In addition this frame aides in the shipment criticality spacing configuration.ImpactClosure LidTest orifice= 7.0 inches(Cavity)11.6 inchesSST shellResinAluminumFrameFigure 1. TN-BGC 1 Transport Package© Copyright TN International 2007 Page 3 on 12

ENC 2007 – Brussels, Belgium / 16-19 September 2007.Primary ArrangementAs the TNBGC-1 design accommodates various radioactive contents, a set of internalarrangements called Secondary Internal Arrangement and spacers are available and licensedaccordingly for each payload. For the FUTURIX-FTA transport, the TN-90 SecondaryArrangement was utilized and a primary internal arrangement was designed specifically for thistransport. Temperature tape was required to document that the pins stayed below 80°C topreclude the need for additional sodium bond settling upon arrival at Marcoule, France.REGULATORY COMPLIANCEInternational shipment involves the understanding of all the regulatory issues for each countrythat the package will travel through. In the case of the FUTURIX-FTA shipment, the FrenchSafety Analysis Report (SAR) for the TN-BGC 1 was required to be amended by the CEA to addthe FUTURIX-FTA payload to the French competent authority certification. Once this processwas successfully completed, it was necessary to submit the revised French amendment, alongwith the payload configuration, for approval by the United States Department of Transportation(DOT) per 49CFR173 to allow for the import or export only of material meeting the FUTURIX-FTA payload profile.Furthermore, a United States export license issued by the Nuclear Regulatory Commission(NRC) was required. International liability was also a key issue to properly implement based oncontractual agreements and legal regulations.French TN-BGC 1 SAR Amendment ProcessThe TN BGC1 package design already had a French approval certificate. For the FUTURIX-FTA payload, a French extension of the approval certificate was necessary. CEA was required torevise the Safety Analysis report of the TN BGC1 package design in order to submit theFUTURIX-FTA payload. The application for the extension of the approval certificate wasperformed by CEA to the French Competent Authority (DGSNR) in February 2005. A typeB(U)F package design was applied as it was initially foreseen to transport all the pins in only onecask. The French extension F/313/B(U)F-96 (Haf) for the FUTURIX-FTA content was issued inOctober 2005.United States Revalidation ProcessThe utilization of foreign certified packages within the United States require the approval of theDOT in concert with the NRC as deemed necessary. It is important to stress that this revalidationof a foreign certificate of compliance is only valid for direct import or export activities.According to 10CFR71.15, a content is exempt from classification as fissile material if thepackage contains 15 grams or less of fissile material. The benefit of the fissile exemption is thata criticality analysis is not required which can streamline the revalidation process. However, thepackage is still a Type B package, requiring certification from DOT. By shipping two pins per© Copyright TN International 2007 Page 4 on 12

ENC 2007 – Brussels, Belgium / 16-19 September 2007.TN-BGC 1 it was possible to request the fissile exemption which reduced the supportingdocumentation to the DOT but necessitated the use of two packages.Export LicenseAn Export License is required for the export of special nuclear material such as plutonium.Because it is mandatory that the export license be approved prior to the initiation of theshipment, the effects of sea vessel timing can be heavily affected by this process. Approval of theFUTURIX-FTA Export License required 3 months due to the various governmental agenciescompleting the review both in the United States and France. Final contract negotiations with thesea vessel were not completed until this document was received because of the possibility ofcostly port costs for maintaining a vessel on standby if the Export License was delayed.International LiabilityThe U.S. Price Anderson Act (PAA) and the Paris Convention on Third Party Liability of theField of Nuclear Energy of July 29, 1960 and its protocols (Paris Convention) are the twoinstruments that afford protection regarding nuclear liability in the unlikely events of a nuclearincident during the shipment of the FUTURIX-FTA fuel pins. It is important to understandwhere each of these instruments is applicable with respect to the actual shipment route.With respect to DOE contracts and subcontracts involving transport or processing of nuclearmaterials, DOE normally includes a Nuclear Hazards Indemnity Agreement that implementsDOE’s duties under the PAA. That Indemnity Agreement is normally applicable with respect tonuclear incidents within the United States and U.S. territorial water that take place duringperformance of the contract. The DOE indemnity agreement is applicable outside of the UnitedStates only where DOE has legal title to the nuclear material that is involved in a nuclearincidentOne of the key features of the PAA is the availability of funds to compensate members of thepublic who suffer a loss as the result of a nuclear incident. In addition to the PAA’s financialprotection that results from DOE’s Nuclear Hazards Indemnity Agreements, the NuclearRegulatory Commission (NRC) rules implementing the PAA require mandatory financialprotection, with respect to nuclear incidents at power reactors and research reactors andtransportation of nuclear material within the United States to and from such facilitiesThe nuclear liability protection afforded by the PAA with respect to the FUTURIX-FTA fuelpins arises from the DOE's "Nuclear Hazards Indemnity Agreement". DOE includes thisagreement in contracts whose performance is deemed to pose a risk of a nuclear incident. ThisIndemnity Agreement is set forth in DOE's procurement rules per 48 CFR section 952.250-70.For these types of shipments it is vital that this DOE Indemnity Agreement is properly “floweddown” from the primary contractor to all subcontractors to indemnify all parties involvedthrough the various purchase orders issued for the project.The nuclear liability protection afforded by the Paris Convention and French law duringtransport of the FUTURIX-FTA fuel pins on the high seas and in France is described as followsin the authoritative Expose des Motif, published by the OECD Nuclear Energy Agency: “Wheresuch substances are being carried from a non-Contracting State [such as the U.S.] to a© Copyright TN International 2007 Page 5 on 12

ENC 2007 – Brussels, Belgium / 16-19 September 2007.Contracting Party [such as France] . . . it is vital for victims that there should always besomebody liable within the territory of the Contracting Parties: Liability in this case is imposedupon the operator for whom the substances are destined, and with whose written consent theyhave been sent, from the moment that they have been loaded on the means of transport by whichthey are to be carried from the territory of the non-Contracting State [Article 4(b)(iv)] subjectalways to the conditions described in paragraphs 27 and 28.”Article 4(c) of the Paris Convention requires the operator liable in accordance with theConvention to “provide the carrier with a certificate by or on behalf of the insurer or otherfinancial guarantor furnishing the security required pursuant to Article 10 of the ParisConvention.” The “Certificate of Financial Security” that is referenced in article 4(c) of the ParisConvention is the actual document required to be issued by the French operator (in thisshipment; CEA) to the shipping contractor. The” Certificate of Financial Security” specificallyindicates the extent to which financial protection under the Paris Convention and French lawimplementing that Convention will be applicable. The applicability of the Paris Conventionwith respect to the FUTURIX-FTA fuel pins was specified in agreements between DOE andCEA.TRANSPORT LOGISTICSA time line for the major milestones of the FUTURIX-FTA shipment is presented in Table 2.This table gives a clear indication of the time and effort involved in planning and executing aninternational shipment of radioactive material. Though less expensive, air transport was notpossible for this transport because of the presence of plutonium within the fuel pins.One of the key requirements for a successful transport is the completion of at least one completedry run of the entire process. Dry runs, utilizing qualified personnel, of the procedures with asimulated payload, the actual package, and supporting equipment such as the leak testingequipment and tie down systems are paramount to ensure the safety of the transport andcompliance to regulatory requirements. In addition a thorough dry run of all transport paperworkminimizes issues from arising during the actual transport.© Copyright TN International 2007 Page 6 on 12

ENC 2007 – Brussels, Belgium / 16-19 September 2007.Table II. FUTURIX-FTA Transportation MilestonesMilestoneFeasibilityStudyFrenchAmendmentDOTrevalidationDry Run ofloadingFuelFabricationExportLicenseUS GroundShipmentAtlanticCrossingFrenchGroundTransportFinalAcceptanceDescriptionA Packaging and Transport Scenario studywritten to describe all packaging servicesrequired to complete the FUTURIX-FTATransportReceipt of French Amendment forFUTURIX-FTA payloadReceipt of DOT approval of the Frenchamendment for FUTURIX-FTA payloadVerification of internal arrangements, leaktesting, loading proceduresFabrication of nitride pellets at LANL, metalpins at INL, fuel jacket/sodium bonding atINLActivity ActivityInitiated Completed7/2005 10/20052/2005 10/20052/2006 4/20064/16/06 4/16/0610/2005 5/2006Receipt of NRC Export License 4/2006 7/2006Load at INL for shipment via US public roadsto Port of Savannah, GATransfer of cargo from truck to ship andshipment to Cherbourg, FranceUnloading at Cherbourg, France and groundshipment to the Phenix ReactorRadiograph of fuel pins to verify sodiumbond and overall fuel pin integrity8/14/06 8/16/068/17/06 9/4/069/5/06 9/7/069/17/06 9/17/06Transport DocumentationA major aspect to any radiological shipment is the necessary paperwork to ensure safety andaccountability of the material. The major documents required for the domestic transport segmentof FUTURIX-FTA included:a. Inland Bill of Ladingb. Exclusive Use Instructionsc. Radioactive Material Emergency Response Information.The major documents required for the international segment of this transport included:a. Competent authority certification in USAb. Competent authority certification in Francec. Pro-Forma Invoice. This document reflects the value of the material and packagesd. TNBGC1 Shipping Container Shipment Checklist – This document states theRadiation and Contamination Surveys for each package.e. Dangerous Goods Declaration© Copyright TN International 2007 Page 7 on 12

ENC 2007 – Brussels, Belgium / 16-19 September 2007.f. NRC Export Licenseg. Shipper’s Export Declarationh. Ocean Bill of LadingTie Down DesignFor US road transportation of generic materials, 49CFR393 specifies the loads to be consideredfor transporting material on public roads. However, these loads apply to any carried material andare not specific to dangerous goods and class 7 transportation. To utilize a more conservativestandard, the draft American National Standards Institute(ANSI) N14.2 l, the International Codefor the Safe Carriage of Packaged Irradiated Nuclear Fuel, Plutonium and High-LevelRadioactive Wastes on Board Ships (INF) code, and the International Atomic Energy Agency(IAEA) were also reviewed. Table 3 shows the various load accelerations evaluated for the tiedown design. The final values were a compilation of the reviewed data.Table III. – Load Acceleration per Transport ModeTransport modeUS roadtransportationLongitudinal(+ forward /- rearward)+ 0.8 g- 0.5 gTransverseVertical(+ up / -down)± 0.5 g -0.2 g± 1.5 g ± 1.5 g ± 1.5 gSea transport ± 1.5 g ± 1.5 g+ 1.0 g- 2.0 gRoad Transport ±2.0 g ± 2.0 g+ 2.0 g- 3.0 gBounding Load ± 2.0 g ± 2.0 g+ 1.5 g- 4.0 gBased on the load analysis, the FUTURIX-FTA tie down system was designed of commerciallyavailable straps, lumber, and rigging bolts. The connection points within the ISO container wereverified by the supplier to meet acceptable load limits for this transport. Figure 2 shows the finaltie down configuration with the TN-BGC 1s.© Copyright TN International 2007 Page 8 on 12

ENC 2007 – Brussels, Belgium / 16-19 September 2007.Figure 2. FUTURIX-FTA tie down configurationUnited State Ground Transportation and Ocean TransportThe two TN-BGC 1 packages were shipped in an ISO container with the stated tie down systemin place from the Materials and Fuels Complex on the INL to the Port of Savannah, GA. A teamof two drivers meeting the appropriate DOT regulations for transporting radioactive materialcompleted this transport segment.The ISO container was loaded onto the Atlantic Osprey as shown in Figure 3. The FUTURIX-FTA Shipment and one other transport were the only cargo aboard ship. The Atlantic Osprey isan INF-2 vessel approved under the IMO code for the transport of irradiated nuclear fuel, highlevel waste and Plutonium with radionuclide activities of 2 x 10^6 TBq for the former two and 2x 10^5 for Plutonium. The vessel is British Flagged and entirely crewed by United Kingdomnationals. The Osprey is capable of both load/on/load off and roll/on/roll/off loading. It waspurchased from a Germany shipping company and modified in the United Kingdom. Thesemodifications included extra accommodation for armed guards and numerous shipboard securityfeatures. For this shipment none of these security features are required. The Atlantic Osprey callsinto the US roughly twice a year with Foreign Research Reactor fuel being returned to the USDOE.© Copyright TN International 2007 Page 9 on 12

ENC 2007 – Brussels, Belgium / 16-19 September 2007.French Ground Transportation and Final ReceiptUpon arrival at the Cherbourg Port, the ISO container loaded with the TN BGC1 containing theFUTURIX-FTA pins was unloaded from the vessel onto a LEMARECHEL CELESTIN vehicle.These vehicles comply with the European Agreement concerning the International carriage ofdangerous goods by road (ADR) and particularly for the transport of radioactive materials.Figure 3. Cherbourg unloading preparations (Atlantic Osprey dock side)After departing from the port, the FUTURIX-FTA transport traveled from Cherbourg to the CEAfacility of Marcoule (2 days of transport). Due to the physical category of the transport, an overnight stop in another nuclear facility of CEA (Saclay) was necessary. At the arrival at Marcoule,the TN BGC1 was transferred to CEA for unloading. Figure 4 shows the unloading of the TN-BGC 1 at the Marcoule site.© Copyright TN International 2007 Page 10 on 12

ENC 2007 – Brussels, Belgium / 16-19 September 2007.TN-BGC1Pin AssemblyTN 90Figure 4. Unloading of the FUTURIX-FTA Pin Assembly at Marcoule, FranceOnce unloaded, CEA completed radiography on the four FUTURIX-FTA pins at Marcoule. Thecondition of the fuel and bond sodium was confirmed to be consistent with the originalfabrication data.CONCLUSIONThe FUTURIX-FTA Transport successfully demonstrated the international shipment of researchplutonium based fuel pins by appropriately addressing the three basic elements of any shipmentof radioactive material: package requirements based on payload properties, regulatorycompliance of all nations involved including international liability, and transport logisticscovering all conveyance modes.© Copyright TN International 2007 Page 11 on 12

ENC 2007 – Brussels, Belgium / 16-19 September 2007.The integration of these shipment elements requires detailed scope planning prior to execution ofany actual shipment tasks. The duration of the actual “in transit” shipment was one month whilethe planning aspects and approval process consumed 13 months. The ability to properly sequenceissuance of regulatory approvals and notifications with the actual movements of the material isone of the major challenges of international shipment.© Copyright TN International 2007 Page 12 on 12

RADIOCHEMICAL CHARACTERIZATION OFACCELERATOR WASTED. SCHUMANN, J. NEUHAUSENLaboratory for Environmental and Radiochemistry, Paul Scherrer Institute Villigen5232 Villigen PSI, SwitzerlandM. WOHLMUTHERDepartment for Targets and Activetechnics, Paul Scherrer Institute Villigen5232 Villigen PSI, SwitzerlandABSTRACTNew aspects of nuclear waste management appear since the operation of large scalefacilities causes increasing amounts of activated material stemming from of accelerators aswell as from their surrounding environment. The Paul Scherrer Institute, operating the mostpowerful proton accelerator in Europe, strengthens its efforts to chemically characterizesamples from accelerator waste in order to meet the requirements from the Swissauthorities concerning the nuclear waste management. With this respect, long-livedradionuclides play an important role, their determination being difficult and timeconsuming,since chemical separation is necessary in most of the cases for samplepreparation. Classical radiochemical separation techniques have to be modified and thenadopted to the special task. Theoretical predictions using calculation codes developed atPSI together with the analytical values help to evaluate the radionuclide inventory ofproton- and neutron-activated parts foreseen for disposal.1. IntroductionRadioactive waste from particle accelerators played a minor role in the past. In few facilities dedicatedmainly to scientific purposes only small amounts of nuclear waste stemming from targets andshielding material had to be considered.This situation changed essentially with the design of new large facilities, for instance compactcyclotrons for the production of medical-used isotopes, neutron spallation sources and the build-up ofhuge radioactive ion beam setups in the recent years. This number will additionally increase in thefuture due to the efforts funded by the EC to investigate the possibilities of so-called Accelerator-Driven-Systems (ADS) for transmutation of radioactive waste.With respect to intermediate or final disposal, these determinations of the radionuclide inventory inaccelerator waste are considered as an urgent problem in Switzerland. Radionuclide analytics shouldprovide data sufficient for the needs of the safety of the repository during the operational (filling)phase and for the safety analysis for the period after closure as defined by the Swiss authorities (e.g.NAGRA). Before giving permission for conditioning the waste parts for intermediate or final disposal,a declaration of the radionuclide inventories is requested. These inventories are essentially differentfrom those already known from nuclear power plants. Important nuclides in this respect are long-livedand consequently difficult to measure. Chemical separation and time consuming sample preparation isnecessary in most cases.The Paul Scherrer Institute (PSI) operates one of the most powerful accelerators in Europe, the 590MeV ring cyclotron with a beam current of more than 1.5 mA. The activated parts from this facility,which are dismounted during maintenance or reconstruction, thus might cause radioactive wasteproblems.

2. Radionuclides and materialsRadiochemical analysis campaigns of nuclear waste are in particular intensive connected withmaintenance, dismounting and reconstruction of parts of the accelerator facility. At PSI, within the last10 years several sample taking actions were driven at a number of exposed areas like the former piontherapy station, a beam control station behind the injector cyclotron, the spallation neutron sourceSINQ, the beam dump of the former target E station and many others in phases where such ashutdown allowed the access to the section of interest. Details of the PSI accelerator facility layout canbe found in [1]. The kind of sample material varies brightly; examples are concrete, stainless steel,cast iron, aluminium, carbon, beryllium, copper and also ceramics.Radionuclides with half-lives up to 10 years emitting high-energy γ-rays are usually the maincontributors to the dose rate of a dismounted section (for instance 60 Co). They mainly determine theradioprotection safety requirements for the personnel during the dismounting process and intermediatestorage. In most of the cases, they can easily be measured without destroying the sample.For the long- and mid-term disposal strategy, the long-lived nuclear inventory of nuclear waste is ofspecial interest. Concerning accelerator waste, knowledge on the production of nuclides with longhalf-lives is rather poor. Since the extremely radio-toxic isotopes of Pu and other transuraniumelements – appearing in the nuclear fuel cycle – are missing, other long-lived, till now rarely studiedisotopes, might determine the long-term treatment of such waste casks. Depending on the surroundingmaterial and the shielding as well as on the kind and the energy of the accelerated particles a lot ofnuclides with half-lives more than 10 years have to be expected. Some of them are difficult to bemeasured due to the extremely long half-lives of ten thousand years or even more, and – in same cases-additionally because of their nuclear properties. 60 Fe (T 1/2 =1.5·10 6 a) and 59 Ni (T 1/2 =4.5·10 4 a) are twoisotopes, which are expected to be produced in considerable amounts especially from copper and ironbeam dump materials. Other radioisotopes causing similar interest are for instance 36 Cl, 26 Al, 63 Ni,10 Be, 3 H, 14 C.3. Measurement techniquesHigh purity germanium detectors are the main tool for the determination of the γ-emittingradionuclides like 60 Co, 22 Na, 65 Zn, 54 Mn and many others without sample destroying. Nevertheless,for some of the longer-lived radionuclides like 108m Ag (418 y) a chemical separation is necessary dueto the high γ-background.Low-energy-β-counting can be done by use of Liquid Scintillation Counting (LSC), requiring aprevious sample preparation including removing the radionuclide from the matrix bulk. This methodcan be applied for radionuclides like 3 H, 14 C, 44 Fe, 63 Ni.For most of the very long-lived isotopes Accelerator Mass Spectrometry (AMS) has to be used. Thesemeasurements were carried out at the tandem accelerators in Zürich (ETH, Switzerland) and Munich(TUM, Germany), respectively.4. Chemical separation techniquesDue to the large variety of matrix material, special separation techniques have to be developed forevery single task and radioisotope. Mainly, classical radiochemical separation methods likeprecipitation, ion exchange, distillation and extraction are applied. For samples foreseen for AMSmeasurement, additional purification from the corresponding isobaric element has to be carried out.5. Analytics of the beam dump of the former target "E" station as an exampleThe Target "E" station at PSI consists of a rotating wheel made from graphite which is hit by a highintensityproton beam coming from the 590-MeV-ring cyclotron. The proton beam generates intensesecondary beams of pions and muons, which are available for research in particle physics and muonspin-resonanceapplications. After the meson production targets, the proton beam is either defocusedand stopped in a beam dump, or refocused and guided to the target of the spallation neutron sourceSINQ.During the upgrade of the 590-MeV-proton accelerator facility at PSI in 1990, the old beamdump behind target E was removed and in 1997 was conditioned for final disposal. To verify the

calculated nuclide inventory of the beam dump, various samples were taken and over the last fewyears analytically analyzed for selected isotopes. The materials investigated were copper, cast iron andstainless steel. The details of the actual sample taking as well as an extended description of thechemical separation procedures are given in [2].In total, 6 independent samples were taken from the beam dump assembly, two copper samples fromfront and back side, respectively, 2 samples made from cast iron and 2 stainless steel samples, whichrepresent the construction material in the surrounding at the "hotspots" determined by dose ratemeasurements. The radionuclide inventory of these samples was analyzed completely using thedescribed chemical separation and measurement techniques. The complete data set including γ-emitting radionuclides as well as the long-lived β-emitters can also be found in [2]. The results werecompared with theoretical predictions obtained by the PSI-West-Waste-Management-BookkeepingSystem, a special tool of calculation codes developed especially for the treatment of accelerator waste[3]. While acceptable agreement of the predictions with the measured values was found for the γ-nuclides in most of the cases, the results for the long-lived isotopes show in a number of cases largediscrepancies. This refers to the problem of modeling complex nuclear reactions in thick targets andleads us to the conclusion that we cannot rely on calculations only for the characterization ofaccelerator waste in the near future. We will further need extended chemical analysis fromrepresentative sample taking positions with this number even increasing with upgrades, reconstructionand, last not least also the design of new large irradiation facilities.As the final result of our experiments it turned out, that, both by calculation and measurement only inone case the radionuclide inventory exceeded the limits given by the Swiss authority (NAGRA),namely the copper sample from the area of the beam entrance. Nevertheless, since this partrepresented only a small portion of the total waste volume, the license for decommissioning of theentire beam dump assembly could be obtained.[1] http://people.web.psi.ch/rohrer_u/weha.htm[2] D. Schumann, J. Neuhausen, R. Weinreich, F. Atchison, P. Kubik, H.-A. Synal, G. Korschinek, Th.Faestermann, G. Rugel; Determination of the radionuclide inventory in accelerator waste using calculationand radiochemical analysis, submitted to NIM B 2007[3] F. Atchison, D. Schumann, R. Weinreich, Comparison of PWWMBS calculated inventories with sampleanalysis results, PSI (TM-85-04-16), Dec. 2004

Session 19.2.2Waste and transport

ENC 2007 – Brussels, Belgium / 16-19 September 2007.Multiple water barriers: an alternative to the assessment of the fuel assembliesduring accident conditions of transportPierre MALESYSTN International1 Rue des HéronsBP 302Montigny-le-BretonneuxF - 78054 Saint-Quentin-en-YvelinesFranceTel : + 33 (0)1 39 48 74 95Fax : + 33 (0)1 39 48 74 93E-mail : pierre.malesys@areva.com1. INTRODUCTIONPackages for the transport of radioactive material have to comply with national and / or international regulations.These regulations are widely based on the requirements set forth by the International Atomic Energy Agency(IAEA) in the "Regulations for the Safe Transport of Radioactive Material".In this framework, packages to transport fuel assemblies (including spent fuel assemblies) have to meet the requirementsfor packages containing fissile material. In accident conditions of transport, the applicant for the packagedesign approval has to show that the package remains sub-critical taking due account of the status of the contentsin these conditions.In most cases, considering water ingress in the package, it is not possible to assume that the fissile material includedin the fuel assemblies is dispersed in the package with the most severe conceivable distribution regardingcriticality. In order to alleviate this difficulty, during the last years, we have provided a significant better knowledgeof the conditions of the fuel assemblies to be transported. This was part of the Fuel Integrity Project (FIP), whichprogress was regularly reported during PATRAM 2001 and PATRAM 2004 symposia.However, for packages which encounter a large g-load during accident conditions of transport and / or which containspent fuel assemblies with very high burn-up, it can be difficult to demonstrate that the fuel assemblies are notsignificantly damaged. Then, to make the criticality assessment considering water in-leakage into the flask and alarge release of fissile material within its cavity will not allow meeting the sub-criticality criteria. For that reason, forour package designs, which use a gas - and not water - as an internal coolant and which fall into that category, wehave decided to take credit of the possibilities provided by the sub-paragraph 677 (b) of the Regulations. This paragraphallows not taking into account water in the package, provided that the package exhibits “multiple high standardwater barriers”.The paper describes our experience with the implementation of this paragraph. Two different cases are considered:either a double vessel, or a double lid. It will be explained when each of these solutions is implemented, andgive examples of package designs with such features, as well as the approvals which were granted for these designsin various countries.© Copyright TN International 2007 Page 1 on 8

ENC 2007 – Brussels, Belgium / 16-19 September 2007.2. REGULATORY REQUIREMENTS FOR WATER INGRESS2.1. Individual package in isolationAs regards packages containing fissile material, requirements in the 2005 Edition of the “Regulations for the SafeTransport of Radioactive Material” set forth by the International Atomic Energy Agency (IAEA) dealing with wateringress when assessing an individual package in isolation are included in paragraph 677.It is required that “For a package in isolation, it shall be assumed that water can leak into or out of all void spacesof the packages, including those within the containment system. However, if the design incorporates special featuresto prevent such leakage of water into or out of certain void spaces, even as a result of error, absence of leakagemay be assumed in respect of those void spaces. Special features shall include the following:(a) Multiple high standard water barriers, each of which would remain watertight if the package were subject tothe tests in para. 682 (b), a high degree of quality control in the manufacture, maintenance and repair of packagingsand tests to demonstrate the closure of each package before each shipment; or(b) ... “In short, the Regulations requires to considerer water ingress in the individual package in isolation, whatever itsbehaviour during the routine, normal or accident conditions of transport, except if the design incorporates “multiplehigh standard water barriers”.2.2. Package arrayIt can be pointed out that there is no specific and / or general requirement regarding water ingress for a packagearray. Water ingress has “only” to be considered as a result of the tests for demonstrating ability to withstand normaland accident conditions of transport.3. CONFIGURATIONS FOR CALCULATIONSIn the following, the configurations which are considered for the different regulatory situations are detailed. Theyare focused on the hypothesis dealing with the water ingress and with the conditions of the fuel assemblies, whenthe package deign includes “multiple high standard water barriers” (and uses a gas as an internal coolant).3.1 Individual package in isolation - Routine conditions of transportFor this first regulatory situation, the fuel assemblies are intact.As regards water in the cavity, on a regulatory basis, and since “multiple high standard water barriers” are present,it could be considered that the cavity of the packaging is dry (or almost dry, that is to say with a quantity of waterinside the cavity equal to the residual water in the cavity after draining and drying, when the package is loaded in apond). However, our common practice, at least for packages which are loaded or unloaded with water in the cavity,is to consider the cavity flooded with water. This allows to take into account a human error during the draining anddrying operations, or a flaw in the draining and drying system.© Copyright TN International 2007 Page 2 on 8

ENC 2007 – Brussels, Belgium / 16-19 September 2007.3.2 Individual package in isolation - Normal conditions of transportFor this second regulatory situation, the fuel assemblies are almost intact: they present no damage, except thepossibility for the fuel pins to slide, until they rest on the top or bottom nozzle of the fuel assemblies.As regards water in the cavity, the hypotheses are the same as under the routine conditions of transport.- If all the operations are performed in a dry environment, it can be assumed that there is no water in thecavity since “multiple high standard water barriers” are present.- If some operations before the transport entails the presence of water, we generally consider that the cavityis flooded with water (to take into account a human error during the draining and drying operations, or aflaw in the draining and drying system), but on a regulatory basis, it could be considered that the cavity ofthe packaging is dry (or almost dry, that is to say with a quantity of water inside the cavity equal to the residualwater in the cavity after draining and drying).3.3 Individual package in isolation - Normal conditions followed by accident conditions of transportWe design a package with “multiple high standard water barriers” when the package encounters a large g-load duringaccident conditions of transport and / or when it contains spent fuel assemblies with very high burn-up. Then,the condition of the fuel assemblies after the accident conditions of transport (and particularly after the 9 m drop) isdifficult to ascertain, or at least the assessment could be considered as controversial by the competent authoritywhich has to approve the package design.Then, for this third regulatory situation, we consider two sets of hypotheses.3.3.1 In the first set of hypotheses, the fuel assemblies are considered as intact but the cavity is flooded withwater. Whilst the hypothesis about the fuel assemblies can be difficult to demonstrate for such a packagedesign with large g-load during accident conditions of transport and / or containing spent fuel withvery high burn-up, the second hypothesis (cavity flooded with water) is unduly pessimistic: the presenceof “multiple high standard water barriers” would allow to consider a restricted quantity of water asexplained in the following.3.3.2 In the second set of hypothesis, the fuel assemblies are considered as severely damaged. This can goup to considering that the fuel assemblies are completely ruined and in the most critical arrangement(as regards geometry and moderation by the restricted quantity of water in the cavity).As regards water in the cavity, credit is taken from the “multiple high standard water barriers”. Thequantity of water is restricted to:- the residual water in the cavity after draining and drying, if some operations before the transport entails thepresence of water, and- the water which can leak into the package during the tests for demonstrating ability to withstand accidentconditions of transport, that is to say, mainly,ooa 9 m drop onto a flat and unyielding surface, followed by a 800 °C / 30 minutes fire, and a 0.9 mimmersion test during eight hours, on the one hand, anda 15 m immersion test during eight hours, on the other hand.© Copyright TN International 2007 Page 3 on 8

ENC 2007 – Brussels, Belgium / 16-19 September 2007.3.4 Array of packages - Normal conditions of transportFor this fourth regulatory situation, the fuel assemblies are almost intact: they present no damage, except the possibilityfor the fuel pins to slide, until they rest on the top or bottom nozzle of the fuel assemblies.As regards water in the cavity, there is no specific and / or general regulatory requirement regarding water ingressfor a package array. Therefore,- if all the operations are performed in a dry environment, it can be assumed that there is no water in thecavity,- if some operations before the transport entails the presence of water, it can be considered that the quantityof water inside the cavity is equal to the residual water in the cavity after draining and drying.Having said that, it must be recognized that, for simplification purpose, the hypotheses we consider for an array ofpackages are very often the same as for an individual package in isolation in the normal conditions of transport.For packaging transporting spent fuel, this simplification in the choice of the hypotheses allows to decrease thenumber of calculations, while inducing no undue burden: in most cases, the neutron interaction between the packagesis reduced to an insignificant value because of the large thickness of the wall of the packages.3.5 Array of packages - Normal conditions followed by accident conditions of transportSimilarly to an individual package in isolation, because of the uncertainties regarding the condition of the fuel assembliesunder the accident conditions of transport, the fuel assemblies are considered as severely damaged.As regards, water in the cavity, and similarly to an array of packages under normal conditions of transport, there isno specific and / or general regulatory requirement regarding water package for an array of package. Therefore,the quantity of water in the cavity is restricted to this which will be “naturally” present subsequent to the tests fordemonstrating ability to withstand accident conditions of transport. Consequently, the quantity of water is restrictedto:- the residual water in the cavity after draining and drying, if some operations before the transport entails thepresence of water, and- the water which can leak into the package during the tests for demonstrating ability to withstand accidentconditions of transport, that is to say, mainly,ooa 9 m drop onto a flat and unyielding surface, followed by a 800 °C / 30 minutes fire, and a 0.9 mimmersion test during eight hours, on the one hand, anda 15 m immersion test during eight hours, on the other hand.As a consequence, when the neutron interaction between the packages is reduced to an insignificant value, thisconfiguration is the same as that yielding from the second set of hypothesis for an individual package in isolationunder the normal conditions of transport followed by the accident conditions of transport.3.6 SummaryTwo main conclusions can be pointed out.- When designing a packaging with a gas (and not water) as an internal coolant, taking credit for “multiplehigh standard water barriers” provides a significant benefit when assessing an individual package in isolationunder the normal conditions followed by the accident conditions of transport.© Copyright TN International 2007 Page 4 on 8

ENC 2007 – Brussels, Belgium / 16-19 September 2007.- Whilst the Regulations require to consider five situations (individual package / array of packages, routine /normal / accident conditions of transport), the number of situations can be drastically reduced by an adequatechoice of the hypotheses, particularly for packages with thick walls (as it is the case for packagingstransporting spent fuel). The number of situation can fall down to two, in both cases considering an individualpackage in isolation.ooFuel assemblies are intact (except the possibility for the fuel pins to slide until they rest on the topor bottom nozzle of the fuel assemblies), and• if all the operations are performed in a dry environment, it can be assumed that there is nowater in the cavity,• if some operations before the transport entails the presence of water, it can be consideredthat the quantity of water inside the cavity is equal to the residual water in the cavity afterdraining and drying, or - conservatively - it can be considered that the cavity is flooded witwater.Fuel assemblies are completely ruined, and the quantity of water is restricted to:• the residual water in the cavity after draining and drying, if some operations before thetransport entails the presence of water, and• the water which can leak into the package during the tests for demonstrating ability to withstandaccident conditions of transport, that is to say, mainly,• a 9 m drop onto a flat and unyielding surface, followed by a 800 °C / 30 minutesfire, and a 0.9 m immersion test during eight hours, on the one hand, and• a 15 m immersion test during eight hours, on the other hand.4. “MULTIPLE HIGH STANDARD WATER BARRIERS”: THICK VESSEL AND DOUBLE LIDWhen transporting spent fuel assemblies, there is a need for a significant gamma shielding to control the externalradiation level. This can be provided by several means, including multiple “thin” steel walls, a steel-lead-steel wall,a “thick” cast iron wall or a “thick” steel wall.The last type of design - “thick” steel wall - provides an intrinsic high mechanical resistance. In addition, within theconcept of “multiple high standard water barriers”, it is fair to focus on the “barrier(s)”, which is the lid(s). It is basedon the fact that the Regulations refer to multiple barriers, not multiple containments, and the sense of “barrier” includesboth a fixed obstacle and a moveable obstacle (the barrier). In this field, the fixed obstacle is the body of thepackaging, and the moveable obstacle (the barrier) is the lid.The first designs where we took credit of the “multiple high standard water barriers” were the casks of the TN 24family. The TN 24 casks are dual purpose casks which have been provided by TN International to Belgian andSwiss utilities for the transport and storage of their spent fuel assemblies.These casks include:- a thick forged steel shell and a thick forgedsteel bottom, both being jointed by a largepenetration shell / bottom weld,- a double lid, originally for storage purpose,each lid being fitted with a gasket which canbe tested before shipment.© Copyright TN International 2007 Page 5 on 8

ENC 2007 – Brussels, Belgium / 16-19 September 2007.The original designs of the TN 24 family did not take credit of“multiple high standard water barriers”. However, later on, anapproval against the latest editions of the IAEA Regulations(1996 or 2005 Editions) was needed and burnup of the fuel assembliesincreased. Therefore, whilst the use of “multiple highstandard water barriers” appeared as a worthwhile option, thechallenge was to incorporate those features on the existing design,whilst minimizing the consequences in the packaging andits operations. The idea was then to take credit of both lids, inorder to have packages with “multiple high standard water barriers”:this solution induces really no modification of the packagings.We started to discuss this issue at the end of the previous millennium with French competent authority. After gettingan agreement on the principle of the recognition that such a concept can meet regulatory requirements concerning“multiple high standard water barriers”, we applied for an approval for several package designs of the TN24 family.For each package design of interest, it was demonstrated that the leak rate of the two lids after accident conditionsof transport was quite acceptable regarding water ingress and subsequent (sub-)criticality. Also, it is specified thatthe leak rate of the lids have to be checked before each shipment, as well as during maintenance.Approvals were granted by French competent authorities, and subsequent approval / validation were obtained inBelgium and Switzerland.5. “MULTIPLE HIGH STANDARD WATER BARRIERS”: DOUBLE CONTAINMENT5.1 Transport of spent fuelAs described in paragraph 4, one option for the design of the gamma shielding of a cask transporting spent fuel isa steel-lead-steel wall.In paragraph 4, the need to focus only on the lid(s) when assessing the acceptably of a “multiple high standard waterbarriers” concept is explained. However, when discussing the implementation for casks whose body is made upof a steel-lead-steel wall with French competent authority, it appeared that this restricted focus would not be accepted,because the body do not exhibit mechanical resistance as large as a thick forged steel shell.Therefore, we decided to use a double containment for the new TN9/4 packaging. This cask is used to transport spent fuel in Switzerland,as a shuttle from the Mühleberg power plant to the Zwilagstorage facility, where the fuel assemblies are transferred in alarge payload dual purpose cask of the TN 24 family.The cask itself is a rather typical cask. The “multiple high standardwater barrier” is provided by a watertight canister fitted in the cavity;the fuel assemblies are loaded in a basket (the basket is in thecanister).Both the cask and the canister are designed to be leaktight: theleak rates measured after a series of physical drop tests lead to awater ingress quite acceptable regarding criticality purpose.© Copyright TN International 2007 Page 6 on 8

ENC 2007 – Brussels, Belgium / 16-19 September 2007.In addition, the leaktightness of the cask and of the canister arealso checked before each shipment, as well as during maintenance.Approval was applied for in France, as the country of origin of the design, and granted by French competent authority.Then, the approval was duly validated by the Swiss competent authority.5.2 Transport of fresh MOX fuelTransport of fresh MOX fuel requires a package design which meets the requirements for Type B(U) packagescontaining fissile material.The packaging designs for such material do not need a significant gamma shielding. As a consequence, and as inthe example described in paragraph 5.1 for spent fuel, due to the level of mechanical resistance which can be providedby the body of such a packaging, a double containment was needed to be in a position to obtain a packagedesign approval based on “multiple high standard water barriers”.Transport of this fuel can be made using our FS 65 packaging.This packaging is quite compact. Subsequently, a high g-load isencountered during the regulatory 9 m drop test, and it may bedifficult to convince the competent authority that the fuel assembliesare not severely damaged after this regulatory droptest. As a consequence, in order to get an approval against thelatest editions of the IAEA Regulations (1996 or 2005 Editions),the use of “multiple high standard water barriers” appeared as aworthwhile option.However, the original design of the FS 65 packaging did not incorporatesuch “multiple high standard water barriers”. Thechallenge was then to incorporate those features on the existingdesign, whilst minimizing the consequences in the packagingand its operations.For sub-criticality purpose and in order to assure the integrity ofthe fuel assembly for its use in the nuclear reactor, the fuel assemblywhich is transported in the FS 65 packaging is clampedin a basket that is inserted in the cavity of the packaging.The idea was then to modify the basket to make it watertightand to consider it as one of the “multiple high standard waterbarriers” (and more precisely as the inner one).The design of the basket was successfully modified. It wasdemonstrated that the new package design, including upgradedinstructions for operations, meets all the regulatory requirementsto be qualified as incorporating “multiple high standardwater barriers”. This terminated with a new package design approvalgranted by French competent authority.© Copyright TN International 2007 Page 7 on 8

ENC 2007 – Brussels, Belgium / 16-19 September 2007.6. CONCLUSIONFor this last decade, one of the most challenging raising technical issue in the assessment of the safety of packagedesign containing fuel assemblies is the evaluation of the fuel assemblies. This is of particular importance for thepackagings containing fresh or spent fuel and encountering high g-load, as well as for those transporting spent fuelwith high burn-up. Designing packages incorporating “multiple high standard water barriers” allows to alleviate thisissue, and then to get package design approvals with a robust technical basis and on a sound administrativeground.TN International has successfully incorporated such features on both existing designs (without modification of thepackaging or with minor modifications of the packaging, and always with limited consequences on the operations)and new designs. Approvals were applied for by TN International: they were then granted by French, Belgian andSwiss authorities.© Copyright TN International 2007 Page 8 on 8

CASTOR ® 1000/19 – A NEW CASK FOR SPENT FUELASSEMBLIES OF TEMELIN NPP, CZECH REPUBLICT. FUNKE, B. KUEHNEDepartment for Cask Development, GNS mbHHollestrasse 7A, 45127 Essen – GermanyABSTRACTAfter the commissioning of both units of the Temelin NPP the utility ČEZ established theirstrategy on management of spent fuel assemblies. Based on the positive experience of thestorage of spent fuel assemblies in CASTOR ® 440/84 casks in Dukovany NPP ČEZapplied the same approach for Temelin NPP. As a result of the bidding process in 2006GNS was appointed as the cask supplier.The concept of CASTOR ® 1000/19 follows the well-known features of ductile cast ironcasks with in-wall moderators and double lid tightness system with metallic gaskets. Thebasket is a completely new concept which complies with the loading machine interface aswell as all safety requirements. The dimensions, weight limit and handling interfacesfollows the requirements of the NPP. The new type CASTOR ® 1000/19 will be applied fora license as dual purpose cask for storage and transportation in Czech Republic.1. IntroductionThe dry storage of spent nuclear fuel from Russian type reactors was successfully realised the firsttime worldwide in CASTOR ® casks developed by the GNS Company in the nineteen eighties andnineties. These casks made of Ductile Cast Iron (DCI) are licensed for both transport and storage.They are in operation since 1983.For Temelin NPP a new type of cask was designed – the CASTOR ® 1000/19. 1000 refers to the typeof the NPP, i.e. VVER 1000, the 19 for the number of fuel assemblies to be loaded.2. Basic Design of CASTOR ® -casksThe cask body consists of a large cylindrical thick-walled casting including bottom made of ductilecast iron (DCI) with high ductility. On the outside wall, circumferential fins are machined in the bodyto improve the heat removal. For neutron moderation, axial bore holes are distributed uniformly in thecask wall. Trunnions are attached onto the cylindrical part bottom side and lid side for handling andlifting.Corrosion protection of the cask cavity and the sealing surfaces is made by nickel coating. The outersurface of the cask, including the surface of the secondary lid and the fins, is protected by a multilayerpaint system, which can be decontaminated easily.The lid system consists of two independent lids to realise the double barrier system which is requiredto fulfil the long-term storage criteria. The primary lid and the secondary lid are sealed by metal-Oringsand fastened by screws.In order to accommodate the fuel assemblies (FA), a basket consisting of tubes for positioning of eachfuel assembly is installed inside the cask cavity. The materials are steel for structural reasons andborated materials for neutron absorption. The necessary heat removal is realised by aluminium plates.

For fulfilment of the IAEA-criteria for transport of type B(U)-packaging, impact limiters are intended,which are screwed onto the lid side and the bottom side.Fig 1: CASTOR ® 1000/19 transport and storage cask3. Design Characteristics of the CASTOR ® 1000/19This cask is designed for transport and long-term storage of 19 FA’s from Temelin NPP with thefollowing specification:- Maximum U235 enrichment 5.0 wt-%- Maximum Burn up 60 GWd/MTUThe main technical data are:- Overall length 5567 mm- Outer diameter (incl. fins) 2302 mm- Cavity height 4620 mm- Cavity diameter 1472 mm- Maximum decay heat power approximately 17 kW/cask- Total mass approximately 110 t (Storage)The wall of the cask body including moderator rods and the bottom are of sufficient thickness forgamma and neutron shielding and designed according to the requirements for transport and storage. Toimprove the neutron shielding, there are moderator plates, one placed into the inter-lid space betweenprimary and secondary lid and one placed into the cask bottom, as well as axial bore holes filled withpolyethylene moderator rods. The rods distributed over two circles are inserted into the cask body side

wall from the bottom-side with the necessary clearance for thermal expansion and kept in theirposition by compression springs.On the outside wall of the cask, radial cooling fins are machined to improve the heat transfer from thecask to the environment.The cask body is designed for the reception of the following components:- Primary lid with its screwed connection (stud bolt with cap nuts and cap screws) and itssealing system- Secondary lid with its screwed connection (cap screws), its sealing system- Centring and fastening of the impact limiters during transportation.At the lid side wall surface of the cask body are four fits with thread holes to bear the trunnions. It isrequired that the casks have four top handling trunnions in order to assure redundancy during caskhandling operations. Two additional trunnions on the bottom end of the cask will be used for tiltingthe cask from vertical to horizontal position or reverse for transport.The basket is designed to accommodate the fuel assemblies inside the cask cavity. Its constructionserves to assure criticality safety under normal and accident conditions and assures sufficient heatremoval by heat conduction from fuel assemblies to cask body.Furthermore, the basket design is optimised to fulfil all requirements of maximum shielding,mechanical robustness, efficient dewatering as well as drying.It consists of 19 hexagonal fuel assembly receptacles made of borated stainless steel. The receptaclesare inserted in cylindrical stainless steel plates and cylindrical anodised aluminium plates. Thealuminium and steel plates are axially arranged in alternated order and fixed by threaded rods withstainless steel spacer tubes.4. Safety Analyses of the CASTOR ® 1000/19The analyses of shielding and thermal behavior as well as of cask strength according to IAEA TypeB(U) test-requirements (9m drop, 1m pin drop, 800 °C fire test) and of the cask behavior duringaccident scenarios at the storage site (drop, fire, gas cloud explosion, collision of casks) were carriedout by means of qualified calculation methods and programs, which are well established and acceptedby the competent authorities.The three-dimensional Monte Carlo code KENO was selected for performing the criticality analysisbecause it has been extensively used and validated by others and has all the necessary features for thisanalysis. The criticality calculations were performed with the SCALE program system.The mechanical analysis under hypothetical accident conditions for different loading cases (drop indifferent impact orientations of the cask) were performed with numerical methods, like Finite-Element-calculations (FEM) with ANSYS and LS-DYNA. The used calculation model consists oflower bound material properties, 3-dimensional simulation of the cask geometry, reasonableassumptions and clearly defined boundary conditions. As a result of the calculation the localdistribution of all stress tensors inside the cask body is known (see example in Figure 2).

Fig 2: Quasi-Statically Calculations of the Stresses during the 9-m-drop (ANSYS)For the thermal analyses, it has been assumed that the package, consisting of the cask with inventoryand impact limiters, is transported horizontally. The decay heat of the fuel assemblies is conveyed bymeans of thermal radiation and thermal conduction from the surfaces of the fuel assemblies to thebasket and then - predominantly by means of thermal conduction - to the outer surface of the basket.In the gap between the basket and the cavity wall, the heat is conveyed by means of radiation andconduction. In the cask side wall, the heat is conveyed to the surface of the cask mainly by conduction.From the surface of the cask, the heat is dissipated by means of radiation and natural convection to theenvironment.The shielding analyses were performed with MCNP, which is a Monte Carlo transport code that offersa three-dimensional combinatorial geometry modelling capability including complex surfaces. Fornormal transport conditions the cask was modelled with the impact limiters and the transport hood.The hypothetical accident conditions assume the absence of the transport hood, the impact limiters andthe neutron moderator. The shielding analysis covers the hypothetical accident conditions in therelated regulation in a conservative manner, because in reality the impact limiters remain on the caskand the total loss of the neutron moderator is not possible. Moderator regions in the shielding modelare replaced by air.With the three-dimensional modelling it is possible to model each fuel assembly. As a result, the localshielding ability of the cask can effectively been used. For each fuel assembly activity limits arederived for the main neutron and gamma generating nuclides. With these limits it is guaranteed thatduring transport the dose exposure is always lower than the limits of the IAEA regulations.5. Summary and ConclusionCASTOR ® casks have been successfully developed, manufactured and delivered for transport andinterim storage of spent fuel and high active waste (HAW). These casks fulfil both the requirementsfor type B(U) packages according to IAEA regulations and the requirements covering differentaccident situations to be assumed at storage sites.The new type CASTOR ® 1000/19 will be applied for a license as dual purpose cask for storage andtransportation in the Czech Republic. The first delivery is scheduled for 2010.

ENC 2007 – Brussels, Belgium / 16-19 September 2007.EUROFAB PROJECT :A CLEAR SUCCESS FOR INTERNATIONAL TRANSPORT OF PLUTONIUM AND MOX FUELSAuthor: Laurent BLACHET (TN International, AREVA group)Co-author: Arvid JENSEN (DCS), George MEYERS (AREVA NP Inc., AREVA group),Patrick JACOT (TN International, AREVA Group), Jean Pierre BARITEAU (AREVA NC, AREVA group), FredYAPUNCICH (Packaging Technology Inc, AREVA group)1-IntroductionAn Agreement between the United States and Russia to dispose of 68 metric tons of surplusweapons-grade plutonium provided the basis for the United States government and its agency,the Department of Energy (DOE), to enter into contracts with industry leaders to fabricate mixedoxide (MOX) fuels (a blend of uranium oxide and plutonium oxide) for use in existing domesticcommercial reactors.DukeEnergy40% 30% 30%AREVA NPAREVA NCDOE contracted with Duke, COGEMA (now AREVANC), Stone and Webster (DCS), a limited liabilitycompany comprised of Duke Energy, AREVA Inc. andStone & Webster to design a Mixed Oxide FuelFabrication Facility (MFFF) which would be built andoperated at the DOE Savannah River Site (SRS) nearAiken, South Carolina.During this same timeframe, DOE commissionedfabrication and irradiation of lead test assemblies inone of the Mission Reactors to assist in obtaining NRCapproval for batch implementation of MOX fuel priorto the operations phase of the MFFF facility.On February 2001, DOE directed DCS to initiate a predecisionalinvestigation to determine means to obtainlead assemblies including all international options formanufacturing MOX fuels.This lead to implemention of the EUROFAB project and work was initiated in earnest onEUROFAB by DCS on November 7 th , 2003.2-DescriptionEUROFAB project consisted of the following major tasks:1. Polishing and packaging 140 kg of weapons grade plutonium oxide at LANL.2. Shipment of the PuO2 from LANL to Charleston, South Carolina using security vehicles.3. Transatlantic shipment of the PuO2 powder to France via Pacific Nuclear TransportLimited (PNTL) armed convoy,4. Shipment of the PuO2 powder from Cherbourg, France through AREVA La Haguereprocessing plant to AREVA Cadarache plant.5. Fabrication of pellets and rods at Cadarache and fabrication of four MOX fuel assemblies(Lead Assemblies-LAs) at AREVA MELOX.6. Shipment of completed MOX fuel assemblies, archives and excess fuel rods from AREVA7. Transatlantic shipment via PNTL of completed fuel assemblies, archives and excess fuelrods to Charleston, South Carolina.8. Shipment of the Lead Assemblies via road to Catawba 1 and shipment of the archive andexcess fuel rods to LANL.©COPYRIGHT TN International 2007 Page 1 sur 8

ENC 2007 – Brussels, Belgium / 16-19 September 2007.Los AlamosNationalLaboratoryCatawbaNuclearStationIrradiatedLeadAssemblyPuO 2Land Transportvia secure truckLAsLand Transportvia SST/SGTsLandTransportviaSST/SGTsU.S. East CoastSea PortU.S. East CoastSea PortPuO 2Ocean Transportvia PNTL Ship SystemLAs + RodsCherbourgport, FranceCherbourgport, FrancePuO 2 PuO 2La HagueFacility, FranceLand Transportvia Secure TrucksLAs + RodsLa HagueFacility, FranceOcean TransportLand Transport viavia PNTL Ship SystemSecure TrucksArchive andScrapMaterial(Rods )PuO2 transports already completedLand Transportvia Secure TrucksLAs + RodsLand Transportvia Secure TrucksCadaracheFacility, FranceLand Transportvia Secure TrucksRodsMELOXFacility, FranceOak RidgeNationalLaboratoryLos AlamosNationalLaboratoryLead Assemblies (LAs) + MOX Fuel Rods: transports already completedIrradiated Lead Assemblies (LAs: ) Still to be realized3- TechnicalThe decision to fabricate the lead assemblies (LAs) in Europe was, in a large part, determinedbecause of the extensive fabrication experience in France, the proven shipping casks for bothPuO2 and lead assemblies and the proven maritime transport capability. The technical approachwas, therefore, to use the existing and proven infrastructures and equipment in so far as possibleand adapt them for application for providing lead assemblies in the U.S.With this approach, the technical issues were minimal while the significant issues included:interfacing different regulatory regimes, interfacing different governments and policies, differentsecurity requirements, different work laws, contracts, different engineering units andcoordination of a large number of different groups from different cultures and backgrounds invarious locations.Throughout the spectrum of EUROFAB functions, regardless of country, the respective expertstook responsibility for their functions (functions which they had performed in the past for otherlike work scopes) and performed them well. The EUROFAB management team respected theexpertise of the different groups. In addition, the team also provided program direction andmanaged the interface between performers.EUROFAB also included the translation of all necessary technical requirements between twolanguages - English and French, conversion of the U. S. fuel design and tooling requirements tometric units, complying with both English and French law as appropriate, flowing downappropriate DOE contract clause requirements to foreign suppliers, obtaining a U.S. exportlicense, obtaining both French and U.S. licenses for both shipping packages, meeting allassociated French, British, and U.S. security requirements and obtaining licensing extensionsfor U.S. weapons grade plutonium at Cadarache, Melox and La Hague facilities.4- Lead Assembly FabricationFabrication involved the interface and efforts of several entities including Duke Power, AREVA NPLynchburg, DCS, Cadarache and Melox. Deliverables required by contracts between DCS andAREVA NC and between DCS and Duke Power were completed on time and provided to theappropriate recipients.Completion of EUROFAB included the following major tasks: minor modifications to theCadarache and Melox plants to allow fabrication of the Mark-BW/MOXI (FRA-ANP AdvancedFuel Assembly design), qualification of Cadarache and Melox plant processes and personnel,supplying appropriate plant hardware and fuel assembly hardware by DCS, and qualification of©COPYRIGHT TN International 2007 Page 2 sur 8

ENC 2007 – Brussels, Belgium / 16-19 September 2007.the fabrication plants quality assurance programs to the AREVA NP (ANP) QA program which isapproved by the NRC.5- RegulatoryThere were several regulatory actions required toaccomplish the EUROFAB mission. Theseregulatory actions included:- Export License (Pu0 2 to France)- FS47 - Cask used to transport Pu0 2 toFrance- FS65 - Cask used to transport leadassemblies to U.S.The following French facilities required licenseextensions to handle U.S. weapons grade Pu:- Cadarache: pellet /rod fabrication plant- Melox: fuel bundle fabrication plant- La Hague: French transport safe haven tochange security containersFS47 Type packageto transport PuO2FS65 Type package totransport LAs and RodsThe regulatory actions by both countries were completed on time to support the EUROFABschedule.6- Institutional issuesNEPA:The DOE was responsible for the NEPA analysis and the ROD (record of decision) regardingEUROFAB. DOE was also faced with two other NEPA actions that affected the MFFF projectduring this same time period (cancellation of immobilization and incorporation of alternatefeed stock). All of these NEPA actions required coordination by DOE in order of importanceand timing which affected the issue date of the EUROFAB NEPA.On November 7, 2003, DOE issued supplemental analysis and an amended ROD forfabrication of lead assemblies in Europe (EUROFAB).The practical effect on EUROFAB was that, in general, prior to the supplemental analysis ofenvironmental Record of Decision (ROD) on November 7, 2003, only studies, plans andevaluations could be carried out. Irreversible actions such as equipment procurement andplant modifications could not be initiated.Because of the NEPA actions and constraints, DCS effectively performed all supportequipment procurements and fabrications, regulatory actions, PuO 2 shipment, lead assemblyfabrication and return shipment in an eighteen month period between November 2003 andMay 2005. The long pre-planning efforts yielded positive results in that the resulting plan waseffective and carried out on time and within budget.Interfaces and other Government Agencies:The interfaces required to carry out EUROFAB between governments, different U.S.government agencies and DCS companies and subcontractors were complex and furthercomplicated by the individual laws, the regulatory requirements and the securityrequirements of each country. DOE managed all government and inter-governmental agencyissues and actions. DCS managed all contractors' issues and performance. The DCS EUROFABmanager obtained both a DOE security clearance as well as a French security clearance tofacilitate communication and resolution of cross cutting issues.©COPYRIGHT TN International 2007 Page 3 sur 8

ENC 2007 – Brussels, Belgium / 16-19 September 2007.In general, the work scope split was such that entities from the U.S. were responsible for theoverall EUROFAB program under DOE direction, land transport within the U.S., and U.S. portand U.S. regulatory interfaces. The British were responsible for the maritime shipment andthe French were responsible for the French port, land transport within France and leadassembly fabrication with technical guidance and requirements provided by DCS.Other U.S. agencies involved in the successful completion of EUROFAB were U.S. Coast Guard,Charleston Naval Weapons Station - U.S. Department of Navy and Public Works, South CarolinaState Police and local law enforcement agencies, Federal Bureau of Investigation and the DOEOffice of Safety Transport. Other institutional bodies and Government Agencies such asNuclear Regulatory Commission (NRC), Department of Transportation, Department of PublicWork, SLED (local law enforcement agencies), Department of States, Border ProtectionServices (Agriculture, Immigrations, Customs), Department of Justice (NEPA), and the LosAlamos National Laboratory within the Department of Energy have contributed to EUROFABproject. A like list of agencies could be added to this list for the work in France and England.LANL under contract to DOE polished and packaged the plutonium with DCS QA oversight andplutonium expertise assistance. DCS provided shipping packages, support equipment, trainingand procedures. DOE entered into a memorandum of agreement with the Department of Navyfor use of the Charleston port and equipment. The DOE Savannah River Site Radiation Controlteam provided support for the shipments at the Charleston port, and a team from LANLprovided the OST/SGT truck loading/unloading functions at the Charleston port.DCS entered into a contract with ANP for the EUROFAB lead assembly design responsibility andQA oversight to assure compliance with NRC regulations. Duke Power entered into a contractwith DCS for acceptance and irradiation of the lead assemblies in their mission reactor.This brief interface discussion illustrates the vast number of entities involved in the successfulcompletion of EUROFAB. Not discussed or shown are the identical infrastructures in France andEngland between the commercial companies and their corresponding government interfaceswhich enabled successful EUROFAB performance.7- ScheduleIn April 2001, a resource loaded primavera baseline schedule was developed for EUROFAB. Itwas an optimistic schedule with a target delivery date of four lead assemblies at the missionreactor by October 2003.This baseline schedule had four parallel critical paths: 1) polishing of the Pu0 2 at LANL, 2)fabrication of the shipping packages and support equipment, 3) French and U.S. licensing ofthe shipping packages, and 4) the issuance of a mission reactor license amendment. The latterthree depended upon issuance of appropriate NEPA documentation before significant workcould be started.With the March 6, 2002 issuance of the intent to prepare supplemental EUROFAB NEPAAnalysis, DOE authorized submittal of the FS47 and FS65 shipping package request forcertificates of approval in France thereby alleviating one of the critical path bottlenecks.With the delay in issuance of the EUROFAB supplemental analysis and amended ROD untilNovember 7, 2003, it was also necessary to move the target delivery date for the four leadassemblies to April - May 2005. The resource loaded primavera baseline schedule was adjustedto this delivery schedule and the revised EUROFAB cost and schedule baseline was submittedto DOE on November 10, 2003.The lead assemblies were delivered on April 28, 2005 according to schedule.8- TransportationEUROFAB transportation was carried out by the following entities:©COPYRIGHT TN International 2007 Page 4 sur 8

ENC 2007 – Brussels, Belgium / 16-19 September 2007.1) the DOE Office of Transportation safelytransported all nuclear materials within theU.S. Commercial carriers were used to stagesupport equipment within the U.S.2) PNTL (Pacific Nuclear Transport Limited), inconjunction with the appropriate Britishsecurity forces provided maritime transportservices,3) TN International, in conjunction with Frenchsecurity forces, provided French land transportservices. Because of the classified nature ofthe shipping dates and numerous securitysupport agencies, many integration meetingswere required to confirm interfaces andprotect classified shipping dates.Specific security measures are classified information but it is recognized that the Frenchgovernment and the British government exceeded all expectation in their support and protectionof the material for the U.S.One of the main reasons for success in all transport activities was that before each transportoccurred, the following were performed: 1) support equipment was fabricated, 2) operational,handling and maintenance procedures were written, and 3) a complete dry run using theequipment and procedures was performed with the people who would be performing the actualwork to verify that the equipment performed properly, the procedures were correct and that thepeople were trained. Everything was tested and people were trained before the actual transportsoccurred.TN International provided rental use of the FS47s, procurement and fabrication of the FS65sincluding support equipment and French land transport and supporting services. PacTec providedloading interface equipment for the packages and the OST trucks as well as a simulated OSTtruck bed for training and testing purposes. Both TN International and PacTec provided training,testing and procedures for all equipment used by all entities handling the packages.A EUROFAB shipment plan outlining the equipment staging, equipment movement and the timingwas also written and followed to minimize cost and to insure transportation success. More than40 subcontracts were place by DCS and carried through to completion to accomplish EUROFAB.The following is a summary of those subcontracts:©COPYRIGHT TN International 2007 Page 5 sur 8

ENC 2007 – Brussels, Belgium / 16-19 September 2007.Consent PackagesPrior to the transport equipment fabrication, numerousconsent packages have been approved by DCS and DOE.TN International followed-up the supply and fabricationof over 20 transport equipment such as 1) SecuredCommunications, 2) Armored Closet, 3) FS47Technical Test Assistance, 4) Monilogs, 5) DummyRods, 6) AA227 Opening / Closing Machine, 7) IPCaisson & Transfer Frame & Tilting Device, 8) FS65Unloading Skid, 9) FS65 Bodies, 10) Pouring Resin,11) FS65 Tooling, 12) Forged Parts for AA431Basket, 13) Boronated Aluminum Plates, 14)Vibration Measurements for FS65, 15) FS65Clamping System, 16) Maritime Caisson PhysicalProtection, 17) FS65 Testing Facility and TechAssistance, 18) FS65 Baskets, 19) Grating Floors forFS65, 20) FS47 Tooling, 21) AA433 Rod Boxes, 22)Excess Storage Carts, 23) FS47 Repair. For PacTec:1) Transportation Services, 2) Simulated SGT, 3)Floor Assembly FS47/65 skids & air pallets & tables.Example of Equipment manufacturedNumerous subcontractors have taken part in Europe to the major procurement efforts suchas Thales, Optim, Socitec, Valorel, Sarrazin, Reel, Pôle de Plasturgie, Mécachimie, Forgesde la Loire, Eagle Picher, Sopemea, Cybernetix, Mecagest, Reel, Cimat, Sarrazin/Cybernetix,PACTEC: Reel Tri-State, PACTEC: Olympic, PacTec: Ideal.9- U.S. requirements versus foreign performanceA large majority of the EUROFAB work was performed outside of U.S. territorial waters underDCS's contract and resolution of several compatibility issues were made.Examples include:1) Flowdown of DOE orders for lead assembly fabrication in France - DCS's contract with DOErequired adherence to the requirements of a large list of DOE orders. Many of theses ordershave no applicability outside the United States. DCS put together a small team with legal helpand reviewed the DCS applicable orders for applicability in France. A large percentage of theorders were eliminated through this review. DOE Chicago, in turn, performed a review of theresultant list; adjustments were made and the fabrication contract was awarded with asignificantly reduced meaningful list for application to production operations in France.2) Law - All contracts were written such that laws in the country in which the work was beingperformed were the laws applicable to the work of that contract.3) Nuclear indemnification insurance - Price Anderson nuclear indemnification insurance wasinvoked by the DOE in the DCS contract for all activities; however, French law mandated thatParis Convention indemnification insurance be provided in France and on the high seas.10- Quality AssuranceDCS was structured such that ANP had responsibility for the fuel assembly design andcertification. Additionally, ANP's QA program had been audited and approved by the NRC. ANPis partially an AREVA-owned company who is also the parent of AREVA NC. When thecompetitive procurement for lead assembly fabrication was issued, it was determined to make©COPYRIGHT TN International 2007 Page 6 sur 8

ENC 2007 – Brussels, Belgium / 16-19 September 2007.the fabrication contractor adhere to the requirements of 10CFR 50 Appendix B so unfairadvantage would not be provided to either of the bidders. Once the contract was awarded, ANPaudited the fabricators' QA program and the fabricator was accepted as a qualified supplier toANP, an agent of DCS. Throughout the fabrication process, ANP QA also performed severalsurveillance audits to ensure compliance with the quality requirements.ANP also performed an audit of the TN International QA program and the LANL QA programsand also accepted them as qualified suppliers. TN International was responsible for theshipping casks for the lead assemblies and LANL was responsible for polishing the U.S. originmaterial.11- CommunicationBased on numerous experiences in shippingPuO 2 and MOX fuels within Europe and toJapan, French and British publiccommunication teams were involved andused a pro-active approach. The U.S.approach was to host a series of publicmeetings to address plutonium dispositionissues and to be responsive to mediainquiries during the Eurofab process. Thisdifference, in part, is caused by the fact thatthe intervener groups in France are muchmore active and militant than in the U.S.These two different methods causedadjustments during the performance ofEUROFAB.These adjustments were essentially mitigated and managed by DOE's establishment of acommunications action group and a weekly communications conference that brought togetherpublic relations representatives from all relevant agencies and the review of all press releasesand material before issuance and by DOE having input on content and timing of releases.Because of the wide publicity of EUROFAB, a special communication plan was developed at thebeginning of the project. A coordination group led by DOE Headquarters with membership fromAREVA, BNFL, Duke Power, and the DCS EUROFAB team worked together to handle allcommunication issues for EUROFAB. The Duke Power/EUROFAB representative served assecond chair in coordination the work of the group. White papers on various related topics andquestions and answers were developed, reviewed and approved by DOE and used as a commondata base by everyone as the need arose.12- Excess Material Storage SiteOn June 4, 2004, LANL was designated by DOE as the site for storage of the excessEUROFAB material. Just placing the two FS65s in storage is rather simple but whenconsidering either the extraction of a few rods for post irradiation examination or removal ofthe material to the MFFF facility, the issues become quite complex very fast. Planning,tooling, support equipment and procedures were developed for the more complex work scopesince the excess material storage site was selected late in the program.13- Equipment DispositionTo accomplish the EUROFAB project, it was necessary to buy or fabricate over 160 differentpieces of equipment (casks, tooling, lifting beams, etc). This equipment was entered into thegovernment property list maintained by DCS as it was purchased or fabricated. At the©COPYRIGHT TN International 2007 Page 7 sur 8

ENC 2007 – Brussels, Belgium / 16-19 September 2007.completion of EUROFAB, most of the equipment was no longer needed and was located inmany different places according to its last use including La Hague, Melox, Cadarache, LANL,Aiken, U.S. Port, Duke Power, etc. An equipment disposition plan was put together tomanage the equipment in accordance with DOE requirements.ConclusionPolished U.S. weapons grade Pu0 2 was loaded onto ships at Charleston,SC on September 20, 2004 bound for France. Pellet fabrication wasinitiated on October 12, 2004 and fabrication of the lead assemblies wascompleted on March 4, 2005. The lead assemblies were delivered at theU.S. mission reactor on April 28, 2005.PuO2 arrival at Cherbourg portAREVA Business Unit Logistics provided all necessary transport equipmentand expertise to realize all shipments of PuO 2 powder and MOX fuelassemblies. Based on its vast experience of nearly 200 shipments everyyear of PuO 2 and MOX fuels, AREVA Business Unit Logistics has developedreliable and proven solutions to implement with optimum safety andsecurity requirements all complex and sensitive shipments.This international effort including over 23 different subcontracts for TNInternational and its AREVA Business Unit Logistics partners wascompleted on schedule and within the initial budget for the bestsatisfaction of DCS and DOE. EUROFAB project is a major step fordisposing of surplus weapon-grade plutonium and is a proven example ofinternational transport of plutonium and MOX fuels.©COPYRIGHT TN International 2007 Page 8 sur 8

Track 3Medical applications

Session 17.3.1:Diagnostics in Nuclear Medicine

Session 17.3.2:Therapeutic Applications 2

POTENTIAL RADIONUCLIDES FORRADIOSYNOVIORTHESIS. DOSIMETRIC CRITERIA FORTHEIR SELECTIONTORRES BERDEGUEZ M. B., AYRA PARDO F. E., ALBUERNEALFONSO O.Radiation Protection Department, Isotopes CentreHavana, CubaMONTANO DELGADO M. A.Hospital General Docente Enrique CabreraHavana, CubaABSTRACTThe objectives of this work are to make a comparison of 32 P, 90 Y, 188 Re, 177 Lu, 51 Cr, 153 Sm,and 169 Er and determine the influence of the variation of the thickness of the synovialmembrane and the shape of the synovium in the S values. We modelled spheres, cylindersand ellipsoids using MCNPX code to assess the therapeutic and maximum ranges, the Svalues and the influence of the mentioned variations on it. The results achieved with thiswork indicate that an approach to the patient – specific dosimetry in RSV will improve theprescription of the dose to the patient. The selection of the radionuclide should be afunction of the inflammation of the joint and not of its size. An adequate diagnostic of thedamaged joint, could help assessing the relationship dose – effect of the treatment. Theoptimal radionuclide does not exist in RSV.1. IntroductionThe treatment of inflammatory arthropaties is aimed to diminish the inflammation, relieve the pain,improve the functional capability and induce the remission of the disease. Although every illnessrequires specific therapeutic guidelines [1], some joints may not experience clinical improvement, soother complementary therapies should be used. Most commons are intrarticular corticoids, chemicalsynovectomy, surgical synovectomy and radiosynoviorthesis (RSV). The first published reports of thislast treatment are as back as to 1924 [2]. In the clinical practice, the first results were published in1952 [3] and were followed during the 60s with the inclusion of the Y-90 [4, 5].The RSV is the injection of a radioactive substance inside any synovial cavity, in such way that theinjected radionuclide makes contact with the synovial membrane. This substance is rapidlyphagocytized by the synovial lining cells. During the radioactive decay of the injected radionuclide, atherapeutic dose will be delivered to the synovial tissue destroying it. The beta emitters used inradiosynoviorthesis range energies between 0.34 MeV (0.33 mm penetration in tissue) in the case ofEr-169 and 2.27 MeV (3.6 mm penetration in tissue) for the Y-90. The half life of these isotopes goesfrom 2.3 hrs (Dy-165) to 27.8 days (Cr-51)[6]. Nowadays, the selection of the radionuclide to be useddepends on the size of the joint to be treated. Thus, the smaller the joint, the lower should be theenergy and therefore the penetration. These facts led to the use of fixed radionuclides for specificjoints. The first works on the assessment of doses to the patients in RSV were published in the 80s andbeginning of the 90s [7, 8, 9, 10].We’ve focused our attention in radionuclides that have been used or tested in animal and humanmodels: 32 P, 90 Y, 188 Re, 177 Lu, 51 Cr, 153 Sm and 169 Er. Two were the objectives of this work. First, tocompare these radionuclides from the dosimetric point of view so we can provide selection criteriadepending on the kind and damage of the joint to be treated. Dosimetric parameters taken into accountwere the therapeutic and maximum range of the beta and electronic emission in the pannu tissue, aswell as the absorbed dose in the synovial surface for different sizes of the synovium.The second objective was to determine the influence in the S values of the proposed model, varyingthe thickness of the synovial membrane and the shape of the synovium. The method used to assess thedosimetry in radioactive synovectomy is known as the Monte Carlo method. The simulation in Monte

Carlo is the best alternative available nowadays to solve the problem of the radiation transport in thematter when we’re dealing with complex geometries.2. Materials and methods.The synovial joint is basically the articular cartilage, bone and tissue (synovial membrane). In table 1is shown the composition and density of every constituent according to Johnson and Yanch [9].ELEMENTS H C N O Na Mg P S Ca Clρ(g/cm 3 )BONE (%) 3.4 15.5 4.2 43.5 0.1 0.2 10.3 0.3 22.5 - 1.92ARTICULARCARTILAGE 9.6 9.9 2.2 74.4 0.5 - 2.2 0.9 - 0.3 1.1(%)TISSUE (%) 10.0 14.9 3.5 71.6 - - - - - - 1.0Table 1. Composition and density of components of the synovial joint.The energy of beta particles used in RSV should be enough to destroy the synovial tissue yet not ashigh as to expose unnecessarily the cartilage and bone. In table 2 are listed the energetic characteristicsof the radionuclides assessed in this work.Radionuclides Y-90 P-32 Re-188 Lu-177 Sm-153 Er-169 Cr-51T 1/2 64.1 hours 14.3 days 16.98 hours 6.71 days 46.7 hours 9.4 days 27.7 daysEmission type Beta BetaBeta, Beta, Beta, Beta,electron,gamma, gamma, gamma, gamma,gammaelectron electron electron electronTable 2. Energetic characteristic of the evaluated radionuclidesThe beta spectrum of each of these radionuclides was downloaded from www.doseinfo-radar.com andthe characteristics of the gamma and/or electronic emission are the ones in the software RADIATIONDECAY [11].The S values have been calculated for different pair of source-target organs and for a number ofradionuclides of interest in Nuclear Medicine. Here, the S factors for each proposed model wasobtained by the calculation of the mean absorbed energy by disintegration (tally f8*) in the volume ofthe cell using the MCNPX code (problem geometry)[12], thru the equation 1:*f 8 −3S = *576.7*10 (Gy/h*MBq) (1)mWhere:m: mass in grams of the cell.*f8: results of the tally (MCNPX).To determine the therapeutic range (X 90 ), a point isotropic source was simulated in the centre ofvarious concentric spheres of different radii. These radii went from 1 mm to 10 mm with 1 mmincrement for all radionuclides but for Er-169 and Cr-51. In the case of the Erbium, the radii wentfrom 0.1 mm to 1 mm with 0.1 mm increments. For Chromium, from 0.1 µm to 1 µm with 0.1 µmincrements.The mathematical model to calculate the maximum range (X max ) was a point source in the centre of asphere. Simulations were made using spheres of different radii, from 1 mm to 10 mm, with 1 mmincrements for all the radionuclides but for the Er-169 and the Cr-51. In the case of the Erbium, theradii went from 0.1mm to 1mm with 0.1mm increments. For Chromium, from 0.1µm to 1µm with0.1µm increments.The S values for the synovial membrane were calculated using as a model a cylinder with the sourceuniformly distributed in its volume. This model was simulated varying the radius of the cylinder (from0.5cm to 9cm, with 0.5cm increments) and its height (from 0.01cm to 0.04cm with 0.005cmincrements). The radii represent different sizes of the synovial surface (the area in the base of thecylinder) for small, medium and large joints. The height represent different stages of the progressionof the rheumatoid arthritis (RA)[8].To assess the influence of the variation of the shape of the synovium in the S factors, spheroidelongations with fixed mass were simulated. Keeping the mass constant at 5, 10, 30, 70 and 100

grams, the radii were changed at a rate () of 1, 3, 5 and 10 every time. The source isdistributed uniformly in the surface of the spheroid.For the variation of the thickness of the synovial membrane, the model chosen was different pairs ofconcentric spheres with masses of the smaller one of 5, 10 and 30 grams. The radius of the secondsphere will be 0.02 cm bigger than the first one. The source is uniformly distributes between the radiiof two spheres. The variation of the thickness of the membrane is simulated moving the inner spherein the X axis from 0 to 0.01 cm.3. Analysis and discussion of results.The selection of the radionuclide for the treatment will depend on the thickness of the synovium to betreated and the proximity of the non target organs of the joint (bone and articular cartilage), so it willbe based in the absorbed dose and the penetration of the radiation emitted by the radionuclide. It isparticularly important that the deepness of the target volume be in the therapeutic range (X 90 )and thatthe non target structures such as bone and cartilage be exposed the less possible to the radiation [10].The therapeutic range is defined as the deepness at which the absorbed dose equals the 10 % of themaximum dose deposited in the synovial surface. In figure 2 there is a comparison between X 90 andX max for the assessed radionuclides. The therapeutic range is the one that determines the synovialthickness that can be treated and not the maximum range.10Rango (mm)10,1Y-90 P-32 Re-188 Lu-177 Sm-153Cr-51Er-169XmaxX900,010,001Figure 1. X 90 and X máx for the 90 Y, 32 P, 188 Re, 177 Lu, 153 Sm, 51 Cr and 169 Er.As the area of the inflamed synovium increases, diminishes the value of the absorbed dose rate in thesynovial membrane (figure 3). For example, if we consider that 5 mCi of Y-90 delivers approximately100 Gy of absorbed dose in an arthritic knee with a synovium surface of 250 cm 2 , only 3 mCi of Y-90will be needed to treat an arthritic wrist of about 50 cm 2 and achieve the same therapeutic effect [10].In this case it is considered that the radionuclide decays completely in the joint, so no escape is takeninto account, important aspect to keep in mind. No significant differences in the S values (Gy/h*MBq)obtained for different synovial membrane thickness and for each assessed radionuclide were observed.The results obtained allow the estimation of the dose that will be delivered to the synovial membranefor different sizes of the joint and different stages of the disease. The values of X 90 allow theestimation of the absorbed dose to the treated synovium (pannu tissue). Nevertheless, the relationshipdose – effect of the treatment is far from been effectively assessed using this simple model due tofactors not taken into account in it and that will be analyzed bellow.Variation of the synovium shape.The variation of the shape of the synovium depends on the kind of joint and the inflammatory natureof the RA. As shown in picture 5, the mayor influence in the variation of S value is observed for Y-90,P-32 and Re-188. For the rest of radionuclides, this variation is twice as less and for the Cr-51 isbarely perceptible. For all the radionuclides this influence diminishes gradually as the synovium sizeincreases.Variation of the thickness of the synovial membrane.As shown in figure 4, the variation of the thickness of the synovial membrane for Y-90 and a verysmall synovium volume (5 g), produces a considerable increment in the variation of S value in thisvolume. As the volume of the synovium increases this influence disappears.

0.01 cmY-90 P-32 Re-188 Lu-177 Sm-1530,04 cmY-90 P-32 Re-188 Lu-177 Sm-153Cr-51Er-169Cr-51Er-169Gy/h*MBq10 50 100 150 200 250 3000,10,010,001Gy/h*MBq10 50 100 150 200 250 3000,10,010,0010,0001A (sqr cm)0,0001A(sqr cm)Figure 2. S Value for different size and thickness of synovial membraneFigure 3. S-value fraction. Influence of the variation of the synovium shape.S/So5 gY-90 P-32 Re-188 Lu-177 Sm-153 Er-169 Cr-511,61,51,51,41,41,31,31,21,21,11,11,01,00,90 0,002 0,004 0,006 0,008 0,01 0,012XdS/So1,61,51,51,41,41,31,31,21,21,11,11,01,00,90,90,830 gY-90 P-32 Re-188 Lu-177 Sm-153 Er-169 Cr-510 0,002 0,004 0,006 0,008 0,01 0,012XdFigure 4. Variation of the synovial membrane thicknees. S-value influencingPhosphorus 32 and Rhenium 188 have a similar behaviour between them. Here, the variation of thethickness of the membrane will not affect the variation of the S factors in the synovium. In the case ofEr-169, Lu-177, Sm-153, and Cr-51, the gradual increase in the variation of the thickness of thesynovial membrane (Xd), leads to an increment in the variation of the S factors for every volume ofthe synovium (5, 10 and 30 grams) been the Erbium the one that affects the most.

4. Conclusions.The results achieved with this work indicate that the use of a simple model of the joint could lead to anoverestimation or an underestimation of the dose to the damaged synovium. Nevertheless, an approachto the patient – specific dosimetry in RSV will improve the prescription of the dose to the patient,eliminating the practice of fixed radionuclides and doses for the treatment of different kind of joints,methodology applied in many clinics in Europe [14]. The selection of the radionuclide should be afunction of the inflammation of the joint and not of its size.An adequate diagnostic of the inflammatory characteristics of the damaged joint, could help assessingthe relationship dose – effect of the treatment.The optimal radionuclide does not exist in RSV. The planning of the treatment will be more effectiveas more real is the dosimetric model used. The debate is opened.5. References.1. Manual Merck, 17.ª Edición 19992. Ishido C. Über die Wirkung des radiothoriums auf die Gelenke. Strahlentherapie 1924;27:188-963. Fellinger K, Schmid J. Local therapy of rheumatics diseases. Wien Z Inn Med 1952;33:351-634. Delbarre F, Menkes C et al, Synoviorthesis with radioisotopes, Press med. 76 (1968) 1045.5. Delbarre F, Le Go A, Menhes C, Aignan M, et al, Double Blind statistical study of therapeutics effect of aradioactivity yttrium (90Y) charged colloid on rheumatoid arthritis of the knee. C R Acad Sci Hebd SeancesAcad Sci D 1974;279:1051-46. Clunie. G. A survey of radiation synovectomy in Europe, 1991-1993. European Journal of NuclearMedicine. (Sep 1995) V.22 (9).p.970-976.7. Martti Hannelin, “A dosimetric study of Dysprosium-165 macroaggreagates used in treating rheumatoidarthritis”, Commentationes physico-mathematicae 87/1988, Dissertationes No. 16, Central Hospital of Etela-Saimaa, Lappeenranta, Finland.8. Harling et al. Radiation synovectomy. Treatment of rheumatoid arthritis. Nucleat scince and engineering.Vol 110 apr 19929. L.Scott Johnson and Jacquelyn C. Yanch, “Absorbed dose profiles for radionuclides of frequent use inradiation synovectomy”, Arthritis and Rheumatism, Vol 34. No 12, December 1991.10. Johnson L.S, Yanch J.C, Shortkroff S, et.al, “Beta-particle dosimetry in radiation synovectomy”, EuropeanJournal of Nuclear Medicine, (Sep 1995) V.22 (9).p.977-988.11. Radiation decay version 2 March 1997. Charles Hacker, Engineering and Applied Science, GriffithUniversity, Australia.12. Judith F. Briesmeister, Editor. LA-12625-M, Version 4B Manual. MCNPTM. A general Monte Carlo N-particle transport code.13. Clunie. G, “A survey of radiation synovectomy in Europe”, European Journal of Nuclear Medicine, V.22(9).p.970-976, Sep 1995.

Session 18.3.1:Instrumentation

NOVEL SOLID STATE PHOTOMULTIPLIERS AND THEIRAPPLICATION TO VERY HIGH RESOLUTION PET ANDHYBRID SYSTEMSG. LLOSÁ, N. BELCARI, G. COLLAZUOL, A. DEL GUERRA,S.MARCATILI, S.MOEHRSDepartment of physics,University of Pisa and INFN - PisaLargo B. Pontecorvo 3, I-56127 Pisa – ItalyC. PIEMONTEFBK-irst, Divisione Microsistemi,Via Santa Croce 77, I-38100 Trento – ItalyT.A. CARPENTER, R.C. HAWKES, A.J. LUCAS, J.W. STEVICKWolfson Brain Imaging Centre, University of CambridgeAddenbrookes Hospital, CB2 2QQ Cambridge – UKABSTRACTPhotomultipliers (PMTs) have been employed for many years as photodetectors in medicalimaging. Position sensitive PMTs are currently used in the last generation of PET scanners,that achieve high resolution and efficiency, and they are also successfully employed incombined scanners like PET/SPECT or PET/CT. However their sensitivity to magneticfields and their size limit their use in certain applications, such as the combination of PETand MRI. Solid state photomultipliers present an alternative to vacuum photomultipliers,with important advantages. They have higher quantum efficiency, reduced size and weight,and they can operate in magnetic fields. In addition, their cost is potentially lower, sincethey can undergo a mass production process. The SiPM is a novel type of photodetectorthat operates in Geiger mode and has a gain of the order of 10 6 . The active area issegmented in small microcells (typical size 25-100 μm) in order to obtain a signalproportional to the particle energy. The use of SiPMs in PET offers the possibility to obtaindepth of interaction information and to improve the spatial resolution. In addition, theimplementation of combined PET/MRI scanners is possible. Silicon photomultipliersproduced by FBK-irst are being evaluated for their use in PET and PET/MR scanners.Results are presented.1. IntroductionFor more than six decades photodetectors have been employed in high energy physics, astrophysicsand medical imaging, among other fields. Still they are an active area of research, in the seek ofimproved photon detection efficiency (PDE), reduced size, larger area, improved spatial resolution oroptimization for different applications. PMTs continue to be the photodetector of choice for manyexperiments. Position sensitive PMTs (PSPMTs) have been employed in the construction of the lastgeneration of small animal commercial PET scanners, achieving a spatial resolution close to 1 mm and10% efficiency [1].Solid state photodetectors have gained much attention in the last years. Compared to PMTs, they offera potentially higher quantum efficiency, reduced size, and insensitivity to magnetic fields. The matureand well established silicon technology allows mass production, which can reduce significantly thecost with respect to vacuum technologies. In the case of the avalanche photodiodes (APDs), theseadvantages are at the cost of some drawbacks, such as lower gain (

A novel type of photodetector, the multicell Geiger-mode avalanche photodiode or siliconphotomultiplier (SiPM) has experienced a very fast development in the last ten years. In addition tothe advantages of silicon detectors previously mentioned, SiPMs offer high gain (~10 6 ) at low biasvoltage (

The first Silicon photomultipliers developed by FBK-irst have an active area of 1mm x 1 mm, with625 microcels of 40 μm x 40 μm size [5]. Their gain is of the order of 10 6 , and their breakdownvoltage around 30 V. Their structure has been carefully studied in order to improve the quantumefficiency in the blue region, with a very thin (4 μm) epitaxial layer. An anti-reflective coatingoptimized for 420 nm wavelength has also been implemented with the same purpose. In addition, teststructures of SiPM matrices have been fabricated and tested. The matrices consist of an array of 2x2pixel elements in the same substrate, of the same size and characteristics as the single SiPMs (Fig 1).Recently, new SiPMs with improved characteristics such as lower noise and higher GF (44-50%) havebeen developed, and are currently being tested. These include single SiPMs of various sizes andmatrices of 16 (4x4) elements.Fig 1. SiPM matrix consisting of four (2x2)pixel elements of 1 mm x 1mm in a commonsubstrate.Fig 2. Coincidence timing resolution of twoSiPMs with two LSO crystals.3.1 Electro-optical characterizationThe SiPMs from FBK-irst show a very good uniformity in their properties, which is essential for thefabrication of the matrices [5]. They have very good single photoelectron resolution, and a gainranging from 4x10 5 to 2x10 6 . The dark rate is about 1-3 MHz at the level of 1-2 photoelectrons, anddrops rapidly to kHz as the threshold is increased to 3-4 photoelectrons.A QE above 95% for blue light has been measured employing a diode structure produced in the samewafer as the SiPMs. The overall PDE is reduced due to the low triggering probability. Still, for adetector with 20% GF, the PDE is about 10% for 420 nm, and 14% at 550 nm. This parameter isenhanced for the SiPMs with 30% GF, and for the ones recently fabricated with 44% GF, for which aPDE around 20% is expected.3.2 TimingBoth the intrinsic timing and the coincidence timing with two LSO crystals coupled to the SiPMs havebeen measured.The intrinsic timing has been measured at the photoelectron (p.e.) level, employing a laser emitting 60fs pulses at 80 MHz rate (i.e, T=12.34 ns) with less than 100 ps jitter. Two wavelengths were tested,400±7 nm and 800±15 nm. The analysis of the data estimates the time difference of contiguous signalscorresponding to 1 p.e. The time difference distribution is fitted with a Gaussian function. A timingresolution around 60 ps sigma is obtained for 3-4 V overvoltage at 400 nm. Also, the timingdependence on the square root of the number of photoelectrons has been verified up to 15photoelectrons, for which the measured timing resolution is 20 ps sigma [6].For the coincidence timing measurement, two small (1mm x 1 mm x 10 mm) LSO crystals have beencoupled to two SiPMs, and the time difference is plotted. The fit of the peak with a Gaussian functionresults in a timing resolution of 600 ps sigma (Fig 2).

3.3 SiPMs and SiPM matrices as photodetectorsThe SiPM performance as readout for scintillators has also been tested. A 22 Na energy spectrum wasobtained with a 1 mm x 1 mm x 10 mm LSO crystal coupled to SiPMs with 30% GF, and operated intime coincidence with a second device. The fit of the 511 keV photopeak with a Gaussian functiongives an energy resolution of 20% FWHM.The performance of the SiPM matrices has also been evaluated. In this case, the SiPMs that composethe matrix have lower GF (20%), and therefore less PDE. In this case, the resolution obtained couplingthe crystal to one of the SiPMs in the matrix is 30%, the same as the one obtained with single SiPMsof similar characteristics. Next, the same crystal has been placed in the centre of the matrix, coveringpart of each of the SiPMs. The signals coming from the four SiPMs are summed and histogrammed.The resulting spectrum has the same energy resolution as the one obtained with a single SiPM (Fig 3).The results show that the matrix development and operation does not degrade the performance of theSiPMs.Fig 3. 22 Na energy spectrum obtained with aSiPM matrix, adding up the signals from thefour pixels.Fig 4. Single photoelectron spectrum obtainedwith the gradient fields on (red line) and off(black line). No difference is appreciated.3.4 Tests in an MR systemThe effect of MR static magnetic, pulsed gradient and radiofrequency (RF) fields on SiPMperformance has been studied. The interaction of the MR system with the electronics necessary todrive the SiPM was minimized by shielding the electronics in a box. The SiPM was left uncovered andexposed to the magnetic fields. The tests were performed both with the electronics box inside andoutside the MR system. In the latter case, the SiPM was connected to the electronics by means of a 1m long shielded coaxial cable.The performance of the SiPMs in an MR scanner has been evaluated by acquiring single photoelectonspectra and 22 Na energy spectra whilst switching magnetic fields with the SiPM positioned in the MRsystem. The resulting spectra show no degradation of the data with respect to those taken outside theMR system (Fig 4).The interference between the two systems results in a pickup in the SiPM baseline simultaneous to theswitch-on of the gradient fields or RF pulses, that must be minimized. Also, the temperature changesin the MR system result in gain variations in the SiPM that should be monitored and corrected for.4. Application to small animal imagingSilicon photomultipliers open new possibilities in medical imaging. A very high resolution PETscanner for small animals employing SiPMs has been proposed and is under construction at the

university of Pisa [7]. The tomograph consists of four 4 cm x 4 cm detector heads that rotate aroundthe object to be imaged. A detector head is made of a stack of three layers. Each layer is composed ofa 5 mm thick scintillator slab read out by a SiPM matrix structure, with 1 mm x 1mm pixels in a 1.5mm pitch. The distance between opposite heads is variable, ranging from 10 to 15 cm. The totalscintillator thickness is 15 mm, ensuring a high detection efficiency for 511 keV photons, while the 3layer structure provides discrete depth of interaction information that reduces the parallax error andimproves the spatial resolution. The use of continuous scintillators instead of pixellated blocks allowsto improve the spatial resolution, while avoiding the problems of light collection efficiency that arerelated to fine pixellation of the crystals. Additionally, the cost of the detector is reduced.In order to estimate the performance of the device, simulations have been carried out with GEANT4,that take into account the characteristics of the FBK-irst photomultipliers. For a detector head, anintrinsic spatial resolution of 0.3 mm FWHM, and 70% efficiency have been obtained. If the positionsare determined with center-of-gravity algorithms, this value degrades in the final 5 mm, near the edgesof the detectors. Skewness and barycentre based likelihood methods have been applied that allow tocorrect for the displacement error and maintain its value well below the parallax error. For theproposed tomograph, a spatial resolution of less than 1 mm 3 at the centre of the FOV, and a sensitivityabove 4% are expected. The sensitivity can easily be enhanced by adding more detector pairs.The combination of PET/CT has allowed the simultaneous acquisition of functional and structuralimages. However, the superior resolution and soft tissue contrast of MRI would provide a much betteranatomical reference for the functional data of PET. At the same time, it would avoid the radiationdose due to CT, that is significant in the case of small animal imaging where a submillimetre spatialresolution is required.The main limitations for the combination of PET and MR imaging technologies are the spaceconstraint inside the magnet, and the sensitivity of PMTs to magnetic fields. In addition, technicalproblems arise from the interference of both imaging modalities. Progress in this field has beenachieved following two main approaches. One is the use of light guides to carry the light to positionswhere the magnetic field is low enough for the PMTs to be operated [8]. The second one is the use ofAPDs that are insensitive to magnetic fields [9]. In this second approach, the shielding of theelectronics results in a decrease of the signal-to-noise ratio of the MR images.The reduced size and modularity of the SiPMs, together with their insensitivity to magnetic fieldsmakes possible their operation inside an MR system. In addition, their high signal-to-noise ratio can beprofited from to find the best configuration and shielding that would minimize the interferencebetween the two modalities. The development of a PET insert based on SiPMs to be operated inside asmall animal MR scanner will allow simultaneous PET/MR imaging.5. ConclusionsSilicon photomultipliers are an alternative photodetector that can offer significant advantages in manyfields. Being compact, light, insensitive to magnetic fields and providing high spatial resolution, theyopen new possibilities in medical imaging.FBK-irst has fabricated high performance SiPMs and matrices of SiPMs in a common substrate. Newdevices with reduced noise and larger active area have recently been produced.At the University of Pisa SiPMs are being studied for the construction of a very high resolution, highsensitivity small animal PET, and a combined PET/MR scanner.6. AcknowledgementsThis project is partially supported by the European Commission's Sixth Framework programmethrough a Marie Curie Intra-European Fellowship.7. References[1]http://www.medical.siemens.com/siemens/en_US/gg_nm_FBAs/files/multimedia/inveon/index.htm[2] C. Piemonte, “A new Silicon Photomultiplier structure for blue light detection”, Nucl. Instr. Meth.A 568, 224-232 (2006).

[3] W. G. Oldham et al, “Triggering phenomena in Avalanche diodes”, IEEE Trans. Elec. Dev., ED-19(9) (1972) 1056.[4] N. Otte “The silicon photomultiplier: A new device for High Energy Physics, AstroparticlePhysics, Industrial and Medical applications. Proc. SNIC-2006-0018.[5] C. Piemonte, et al., “Characterization of the first prototypes of Silicon Photomultiplier fabricated atITC-irst”, IEEE Trans. Nucl. Sci. 54(1), 236-244 (2007).[6] G. Collazuol et al. “Single timing resolution and detection efficiency of the ITC-irst Siliconphotomultipliers”. Presented at the XI VCI, Vienna 19-24 Feb. 2007 and to be published in Nucl.Instr. Meth. A (2007).[7] S. Moehrs, A. Del Guerra, D. Herbert, M. Mandelkern, “A detector head design for small animalPET with silicon photomultipliers”, Phys. Med. Biol. (2006), 1113-1127.[8] A.J. Lucas et al. “Development of a combined microPET/MR system”. Technol. Cancer Res.Treat. 5, 337-41 (2006).[9] B. Pichler et al. “Performance test of an LSO-APD detector in a 7T MRI scanner for simultaneousPET/MRI”. J. Nucl. Med. 47, 639-47 (2006).

HIGH SENSITIVITY PET DETECTOR DESIGN USINGMONOLITHIC SCINTILLATOR CRYSTALSP. BRUYNDONCKX, C. LEMAITRE, S. TAVERNIERDepartment of physics, Vrije Universiteit BrusselPleinlaan 2, 1050 Brussels– BelgiumD.J. VAN DER LAAN, M. MAAS, D. SCHAARTIRI, Technische Universiteit DelftMekelweg 15, 2629 JB Delft– The NetherlandsABSTRACTUsing undivided scintillator blocks yields an effective manner to significantly increase thesensitivity of high-resolution PET systems. A neural network based positioning scheme isable to localize perpendicular incident photons with an accuracy 1.7 mm FWHM for a 10mm thick LSO block and 2.0 mm FWHM for a 20 mm thick LSO block. Because thepositioning algorithm determines the incidence position as opposed to the interactionposition, no separate measurement of the interaction depth is required to obtain parallaxcorrected coordinates. Energy and time resolution were 12% FWHM and 1.6 ns FWHMrespectively.1. IntroductionPositron emission tomography (PET) is an imaging modality for measuring the spatial and temporaldistribution of an administered radio-labelled substance within the body of a patient. The quality ofPET images is usually given by three physical characteristics : spatial resolution (ability to seedetails), contrast (ability to see differences in radio-tracer levels) and noise (statistical fluctuations inthe measured radio-tracer concentrations). Despite the fact that these characteristics describe differentaspects of the image quality, they are not independent. The improvement of one parameter very oftenleads to the degradation of one of the others. Over the last decade a lot of effort has been put intoimproving the spatial resolution. Small animal PET scanners now have an image resolution wellbelow 2 mm FWHM and some get close to 1 mm FWHM. To completely utilize the resolving powerof high resolution PET scanners, clinical or small animal, the signal to noise ratio (SNR) in the imagesshould be kept at a sufficiently low level. Otherwise small details that could be discerned areswamped in the noise. For a given number of detected annihilation photon pairs in a PET scan, thesquared SNR in the image is inversely proportional to the fourth power of the image voxel size (ordetector pixel size). Hence an enhancement of the intrinsic detector resolution without an increase inthe number of detected events is detrimental for the image noise. Depending on the PET study andobject, a longer scanning time or a higher tracer dose are not always possible. The newest clinicalPET scanners now also offer time-of-flight (TOF) capability. Using the measured arrival timedifference of the two annihilation photons in the image reconstruction process allows a reduction ofthe image variance. Given a time resolution of 600 ns, TOF PET imaging is currently only useful inclinical whole body imaging [1].Another way of improving the sensitivity of a PET scanner is to increase the packing fraction (fractionof sensitive material compared to the total detector volume). Classical PET scanner designs are basedon matrices of small scintillating pixel elements that are optically separated and read out by a positionsensitive photo detector. Reducing the size of the detector pixels to enhance the spatial resolution alsoreduces the packing fraction because of the relative higher presence of material to optically separatethe crystals. The packaging of the photo detector itself also introduces some dead area unless thescintillation crystals are decoupled from the photo detector using optical fibers [2]. In this case thecrystal matrices can be mounted side-by-side at the expense of some scintillation light loss in theoptical fibers, resulting in a lower energy resolution. The spatial resolution gain obtained by using

smaller individual crystals is also partially offset by increased inter-crystal scatter and parallax errorsfor photons impinging at non-perpendicular angles.To overcome the above obstacles while trying to simultaneously improve spatial resolution andsensitivity, we studied an alternative detector design based on monolithic scintillator blocks read outby pixelated photo detectors. The philosophy behind our approach to localize the impinging photonsis to make use of the scintillation light distribution that is generated when a photon interacts in thescintillator block. This shape obviously depends on the 3D interaction position of the photon.Because we rely on the spreading of the scintillation light, the scintillator block can be larger than thesensitivity area of the photo detector en hence also cover the packaging of the photo detector.2. Materials and methods2.1 Detector components and experimental set-upIn this study two different continuous LSO scintillator blocks were used: a rectangular 20x10x10mm block and a thicker rectangular 20x10x20 mm. The LSO blocks were wrapped with Teflon on 4sides (20 mm thick blocks) or 5 sides (10 mm thick block) to maximize the light output. Thescintillation light distribution emerging at the bottom of the 10 mm thick block is sampled by an 8x4S8550 Hamamatsu avalanche photo diode (APD) array on the top side (Fig. 1 left). The pixel size ofthis device is 1.6 x 1.6 mm and the pixel pitch is 2.3 mm. In case of the 20 mm thick blocks, the topand bottom surface were read out by APDs (Fig. 1 middle).To extract the position of an impinging photon from the sampled scintillation light distribution weused a neural network (NN). The NN needs to be trained using a set of training events before it can beused. A ~1 mm FWHM electronically collimated 511 keV photon beam that can be positioned at anyknown point on the front surface of the LSO block using a motorized XZΩ stage (Fig. 1 right) wasused to generate these training events. More details about the experimental set-up can be found in [3].The beam was stepped along the central x-axis of the front surface using 250 µm steps. At each beamposition a number of events were measured, the 32/64 APD pixel values were summed along the rowsand columns to obtain the light distribution projections and stored together with the incidence photonposition (i.e. the centre of the known beam position).Photon beamPhoton beam10 mm20 mm20 mm10 mm20 mm10 mmFig. 1 : (left) A 20x10x10mm LSO block with an APD readout on the top surface, (middle) A20x10x20 mm LSO block with a double APD readout, (right) ) Schematic representation of theexperimental set-up. The photon beam is defined by coincidence measurements with a 35 mmthick BGO crystal mounted to a PMT at some distance. The 60 mm thick lead collimator has a holediameter of 5.0 mm2.2 Parallax-free positionsUsually a measurement of the depth at which a photon interacts in a scintillation crystal forms thebasis for parallax correction. This can either be achieved in a discrete way (e.g. phoswich [4] or apixel encoding scheme [5]) or in a continuous way (e.g. using the ratio of the signals measured on thetop and bottom side of the scintillator [6]). Hence the accuracy on the estimated true interactionposition in the scintillator depends on the accuracy of the estimated position of the photon in the planeof the photo detector and on the accuracy of the depth-of-interaction (DOI) measurement.Because we trained the NN to reproduce the incidence position instead of the interaction positionwithin the scintillator block, there is no need for a separate DOI measurement and DOI calibration ofthe detector module because the incidence position is independent of the incidence angle. Hence it

does not suffer from parallax problems. However the relation between the measured scintillation lightdistribution and the incidence position depends on the incidence angle. Therefore algorithms have tobe trained for different incidence angles. We set ϕ=0 o while θ was varied from 0 o to 30 o in 10 o (Fig.1).2.3 Positioning algorithmThe neural networks used to compute the incidence position consist of an input layer, one or morehidden layers and an output layer. The neurons in the input layer accept the APD pixel signals whilethe neuron in the output layer yields the corresponding photon impinging position once the networkhas been trained. The neural network computation of a coordinate x using a single hidden layernetwork can be written asx = v T σ[ W ⋅ p + d](1)= v T σ[]nwhere p is the vector of APD signals, W is the matrix of input-to-hidden layer weights, d is the vectorof hidden node biases, v is the vector of hidden-to-output layer weights. The term σ[n] indicates anelement-by-element evaluation of n using a sigmoidal function.The classical way to train a neural network is by using error back propagation [7]. For each event inthe training data set, the predicted position x i is compared with the known incidence (i.e. beam centre)position y i . There are various algorithms which tune the network parameters such that the root meansquare error is minimal. In general the neural network minimization problems are often very illconditioned.For these kind of problems the Levenberg-Marquardt algorithm is usually a good choicebecause of its higher robustness. The training of a NN with a single 10 neuron hidden layer wasimplemented using the neural network package of the Mathematica [8].3. Results3.1 Position information content in scintillation light profilesThe Cramér-Rao inequality gives a lower bound on the variance of an efficient estimator [9,10]. Weused this theory to obtain a lower limit on the accuracy with which a photon can be localized from thescintillation light distribution it generates in the scintillator blocks. Figure 2 shows the lower variancebounds on a 20x10x10 mm and 20x10x20 mm LSO block. The light distributions used in thecomputation of these lower bounds were generated by a Geant4 [11] based Monte Carlo simulation ofscintillation point sources placed on a grid in the LSO block. In most regions of the block the lowervariance bound is less than 0.5 mm. In general the position estimation accuracy degrades as thephoton interacts further away from the APD array (z=-5 in fig. 2 left). Therefore we have to use aread out on both top and bottom side of monolithic scintillator blocks that have a thickness larger than1 cm (e.g. to further optimize the stopping power and sensitivity) in order to retain good overall spatialQuickTime and aTIFF (Uncompressed) decompressorare needed to see this picture.QuickTime and aTIFF (Uncompressed) decompressorare needed to see this picture.resolution (Fig. 2 right).Fig.2: Distribution of the Cramér-Rao lower variance that can be achieved by an efficient positionestimator in a 20x10x10 mm LSO block with top one-sided read out (left) and a 20x10x20 mmLSO block with a double sided read out (right).

3.2 Detector resolution and nonlinearityFigure 3 left shows the overall intrinsic FWHM resolution measured at all beam positions along the x-axis for the 20x10x10 mm LSO block, as a function of the incidence angle and energy threshold. Forperpendicular incident photons, the measured FWHM detector resolution is 1.75 for a threshold of 100keV and 1.60 mm for a threshold of 380 keV. When the photon incidence angle increases to 30 o , thespatial resolution degrades slightly to 2.0 mm and 1.85 mm FWHM respectively. Increasing thethreshold from 100 keV to 250 keV or 380 keV reduces the detector sensitivity by respectively 11.8 %and 26.4 %.The FWHM obtained using the signals from the top and bottom APD on the 20x10x20 mm thick LSOblock is shown in figure 3 middle. The measured detector resolutions are worse compared to thethinner 20x10x10 mm LSO block. This could be due to the fact that the scintillation light is nowspread over 64 APD pixels in stead of only 32 pixels. Hence the statistical fluctuation of the numberof photons per pixel will be larger. In addition each of the 32 extra readout channels will alsocontribute electronic noise to the data.The deviation from linearity for the NN estimated coordinate is shown in figure 3 right. Thenonlinearity is negligible except for photons impinging within 1 mm from the edge of the scintillatorblock.Figure 3: (Left) FWHM resolution for a 20x10x10 mm LSO block, (Middle) FWHM resolution for a20x10x20 mm LSO block, (Right) deviation from linearity of the NN positioning algorithm for the20x10x10 mm LSO block.3.3 Energy and time resolutionFigure 4 left shows the energy spectrum for the 20x10x10 mm LSO block obtained from the eventsmeasured at all beam positions. The average energy resolution is 12% FWHM and does not varysignificantly as a function of the photon impinging position.Time resolution measurements were performed against a BaF 2 crystal coupled to an XP2020Q PMT.The PMT signal and the analogue sum of the 32 APD signals were both fed into a constant fractiondiscriminator for time pick off. The time spectrum (fig 4 right) was obtained using a time-to-analogconvertor followed by an analog-to digital convertor. The time resolution obtained had a resolution of1.6 ns FWHM. A similar measurement on the 20x10x20 mm LSO block yielded a 2.2 ns FWHM timeresolution.350030002500200015001000500100 200 300 400 500 600KeVFigure 4: (Left) Energy spectrum for a 20x10x10 mm LSO block, (Right) Time spectrum of a20x10x10 mm LYSO block measured against a BaF2 crystal coupled to an XP2020Q PMT. Thesolid line represents a Gaussian fit through the data3.4 Image resolutionA 250 µm diameter 22 Na point source (50 µCi) was placed at radial distance of 1, 6, 11 and 16 mmrespectively from the centre of a simulator set-up. The simulator consists of two detector modulesequipped with 20x10x10 mm LSO blocks. The detector modules are mounted on a rotating gantrywith a diameter of 130 mm [12]. The estimated continuous coordinates of the impinging annihilation

photons, obtained with neural networks, were first rebinned in 1x1 mm virtual pixels. The coordinatesof the centre of these pixels were then used to fill up the sinograms. Table I gives the radial andtangential resolution at the different positions of the point source in FBP or OSEM reconstructedimages. A square filter function with a cut-off at the Nyquist frequency was used in the FBPreconstruction. The OSEM algorithm used 8 subsets and 5 iterations. In both cases the image matrixconsisted of 0.5 mm 3 voxels. Close to the centre, the resolution is 1.7 mm FWHM in the FBPreconstructed image and 1.3 mm FWHM in the OSEM reconstructed image. The degradation of theradial resolution at some distance from the centre is probably due to the fact that the incidence positionof the photons was estimated using only the neural network trained for perpendicular incident photons,i.e. non-parallax corrected positions.The axial resolution was obtained from the width of the profile showing the number of coincidencesmeasured in 0.5 mm thick transaxial slices. The FWHM of this distribution was 1.8 mm.FBPOSEMPosition Radial Tangential Radial Tangential1 mm 1.7 mm 1.7 mm 1.3 mm 1.3 mm6 mm 1.8 mm 1.7 mm 1.4 mm 1.3 mm11 mm 1.9 mm 1.7 mm 1.45 mm 1.3 mm16 mm 2.0 mm 1.8 mm 1.6 mm 1.4 mmTable I : FWHM radial and tangential resolution in FBP and OSEM reconstructed images of a pointsource at various radial distances from the centre4. ConclusionA high-sensitivity, high-resolution PET detector concept using a neural network positioning algorithmwas evaluated. We achieved a 1.7 mm FWHM resolution in a 10 mm thick LSO block and 2.0 mmFWHM resolution in a 20 mm thick LSO block. These results still include the influence of the finitesize of the collimated photon beam used in the set-up. The study also indicates that a PET scannerdesign based on continuous blocks should preferably be build using a double layer of 10 mm thickLSO blocks coupled to a single APD in stead of a single 20 mm thick LSO blocks read out on bothsides. The nonlinearity in the position estimation is negligible up to 1 mm from the edge of the LSOblock. The corresponding resolution near the center of a PET image acquired on a simulator set-upwas 1.7 mm FWHM using FBP and 1.3 mm FWHM using OSEM.When training data sets are acquired for different incidence angles, we observed a slight degradationof the intrinsic detector resolution from 1.7 mm FWHM at 0 o to 2.0 mm FWHM at 30 o when a 350keV threshold is applied.The absence of multiple reflections due to the light piping effect found in the traditional crystal pixelmatrices results in a larger light output and hence a better energy resolution of 12 % that remainsrather uniform over the length of the block. Despite the summing of the 32 APD pixels signals totrigger an event, we achieved a time resolution of 1.6 ns.Given the above performances, we see this photon detection and localization principle as a viablealternative to the classical designs. Because of the high sensitivity and parallax-free photon detectionproperty, we foresee the implementation of this technology in a future BrainPET system.5. References[1] “Performance of Philips Gemini TF PET/CT Scanner with Special Consideration for Its Time-of-Flight Imaging Capabilities”, Suleman Surti, Austin Kuhn, Matthew E. Werner, Amy E. Perkins,Jeffrey Kolthammer and Joel S. Karp, The journal of nuclear medicine, Vol. 48, No. 3, pp471-480[2] “MicroPET II: an ultra-high resolution small animal PET system”,Y.C. Tai et Al.,NuclearScience Symposium Conference Record, 2002 IEEEVolume 3, Issue , 10-16 Nov. 2002 Page(s):1848 - 1852 vol.3[3] “Experimental characterization of monolithic-crystal small animal PET detectors read out by APDarrays”, M.C. Maas, D.J. van der Laan, D.R. Schaart, J. Huizenga, J.C. Brouwer, P.Bruyndonckx, S. Leonard, C. Lemaitre, C.W.E. van Eijk, IEEE Trans. Nucl. Sci., vol. 53, no. 3,pp. 1071-1077, June 2006

[4] “Digital pulse shape discrimination methods for phoswich detectors”, D. Wisniewski,M.Wisniewska, P. Bruyndonckx,M. Krieguer, S. Tavernier, O. Devroede, C. Lemaitre, J.B. Mossetand C. Morel, 2005 Nucl. Sci. Symp. Conf. record, Vol. 4, pp 2017-2021[5] “Performance evaluation of a subset of a four-layer LSO detector for a small animal DOI PETscanner; JPET-RD”, T. Tsuda, H. Murayama, K. Kitamura, N. Inadama, T. Yamaya, E.Yoshida,F. Nishikido, M. Hamamoto, H. Kawai and Y. Ono, IEEE trans. Nucl. Sci., Vol 53(2006), Issue 1, Part 1., pp 35-39[6] “Design and evaluation of the Clear-PEM scanner for positron emission mammography”, M.C.Abrue et al, IEEE Trans, Nucl. Sci., Volume 53 (2006), Issue 1, Part 1, pp 71-77[7] “Neural Networks : A comprehensive foundation”, S. Haykin, Prentice Hall, 1999[8] www.wolfram.com[9] “Performance optimization of continuous detector modules using Cramer-Rao theory combinedwith Monte Carlo simulations”, D.J. van der Laan, M.C. Maas, D.R. Schaart, P. Bruyndonckx, S.Leonard and C. van Eijk, 2004 Nucl. Sci. Symp. Conf. record, Vol. 4, pp 2417-2421[10] “Influence of sensor arrangements and scintrillator crystal properties on the 3D precision ofmonolithic scintillation detectors in PET”, P. Ojala, A. Bousselham, L. Eriksson,A. Brahma andC. Bohm, 2005 Nucl. Sci. Symp. Conf. record, Vol. 5, pp 3018-3021[11] “Geant4 – a simulation toolkit”, S. Agostinelli et al., Nucl. Instr. Mehods, A506 (2003), pp250[12] “Performance of a PET prototype demonstrator based on non-pixelated scintillators”, P.Bruyndonckx, C. Lemaitre, S. Leonard, D. Schaart, D.J. van der Laan, M. Maas, Y. Wu, M.Krieguer, S. Tavernier and O. Devroede, 2004 Nucl. Sci. Symp. Conf. record, Vol. 6, pp 3924-2927

Session 18.3.2:Radiolabelling of molecules

IN VIVO EVALUATION OF [ 18 F]-2-FLUOROMETHYL-L-PHENYLALANINE IN TUMOUR BEARING RATS WITH PETACQUISITION.KEN KERSEMANS, MATTHIAS BAUWENS, ANH THO NGUYEN,TONY LAHOUTTE, AXEL BOSSUYT, JOHN MERTENSBEFY, Radiopharm. Chemistry, Vrije Universiteit BrusselLaarbeeklaan 103, 1090 Brussels, BelgiumABSTRACTCurrently there is a vast interest for [ 18 F]-labelled amino acid-analogues for tumour specificdiagnosis with PET. We hereby present the in vivo evaluation in R1M tumour bearing ratsof [ 18 F]-2-fluoromethyl-L-phenylalanineas a new tumour specific PET tracer.The precursor, N-Boc-2-Bromomethyl-L-phenylalaine-tButylester, is radiofluorinated inconditions comparable to FDG synthesis with a good radiochemical yield. After HPLCseparation and deprotection the [ 18 F]-2-fluoromethyl-L-phenylalanine is recovered in n.c.a.conditions. Traces of free radiofluoride were removed using a custom appatite filter beforesterilisation. 3.7 – 10 MBq were injected into R1M tumour bearings rats and imaging wasperformed with a human PET camera from 5 to 45 minutes p.i.. The obtainedtumour/background and tumour/blood ratios were at least 3 coupled to an appropriatebiodistribution and cleaning pattern.N.c.a.[ 18 F]2-fluoromethyl-L-phenylalanine is currently used in a Phase I study for tumourdiagnosis with PET.1. IntroductionCurrently there is a vast interest for [ 18 F]-labelled amino acid-analogues for tumour specific diagnosiswith PET because the widespread 2-[ 18 F]fluorodeoxyglucose ([ 18 F]FDG) lacks the tumour cellspecificity required for control and follow-up after surgery, radiotherapy and chemotherapy [1,5]. Akey-point in the development of this new type of tracers is the Na+-independent LAT1 amino acidtransport system for neutral and lipophilic amino acids, which is overexpressed in tumour tissuerelative to normal tissue[6-10]. We have already proven in vitro and in vivo that 2-I-L-phenylalanine,a SPECT tracer, is taken up for the major part by LAT1 system in a R1M rhabdomyosarcoma tumourcell line and in several types of human cancer cells [11]. This radio-iodinated amino acid shows a veryhigh tumour selectivity when compared to the clinically routinely used [ 18 F]-FDG, which is taken upconsiderably in brain and inflammatory tissue [3-5]. On this basis we developed a new fluorinatedphenylalanine analogue, [ 18 F]-2-fluoromethyl-L-phenylalanine, considering that the spatial volumes ofFCH 3 and Iodine are comparable and that [ 18 F]-2-fluoromethyl-L-phenylalanine can be prepared withthe ease of [ 18 F]-FDG to allow clinical routine. The challenge was to develop a suitable precursor andradiolabeling strategy that would reduce the chemistry to a simple aliphatic nucleophilic substitutionon the methyl side chain, thus avoiding the difficult low-yield multistep radiolabeling procedures asoften encountered in previous attempts to introduce a suitable PET amino acid for routinely use. The[ 18 F - ] for bromine substitution strategy that we developed at our lab allows to prepare the [ 18 F]-2-fluoromethyl-L-phenylalanine with the ease of [ 18 F]FDG in a clinical routine and thus filling the gapbetween radiolabeled amino acids and [ 18 F]FDG. The aim of this study was to evaluate in vivo n.c.a.2-[ 18 F]fluoromethyl-L-phenylalanine in tumour-bearing rats by means of PET acquisition.2. Materials and MethodsAll the conventional products mentioned were at least analytical or clinical grade. The solvents wereof HPLC quality and were all from Sigma-Aldrich unless stated otherwise.

2.1 RadiosynthesisThe labelling has been described in detail elsewhere (submitted Nucl Med Biol). In short: 10 µmolesof N-Boc-2-bromomethyl-L-phenylalanine-tButyl ester were labelled with [18F-] in presence of K 222and K 2 CO 3 in ACN(400 µL) within 5 minutes at 120°C. The n.c.a. N-Boc-2-[18F]-fluoromethyl-Lphenylalanine-tButylester was recovered from HPLC speration..Complete deprotection was achievedin 400 µL of a dichloromethane/trifluoroacetic acid: 3/1 mixture in presence of CaCl 2 within 20minutes at 50 °C. After removal of the original solvents by evaporation using the N2-flow at roomtemperature the 2-[ 18 F]Fluoromethyl-L-phenylalanine is recovered in an isotonic aqueous solution thatis successively sent through a mini C18 column and a custom apatite column before sterilising througha 0.20 µ Cathivex filter for sterilization.2.2 Quality and shelf-life controlQuality control of 2-[ 18 F]fluoromethyl-L-phenylalanine (spiked with a small amount of KF) wasachieved by HPLC, using a Prevail C18-column (5 mm, 250 x 4 mm) (Alltech, Belgium) andethanol/water (1/99, v/v) as eluent with a flow of 1 ml/min (k’ = 5.6), while monitoring UV absorption(Shimadzu UV detector, 280 nm) and radioactivity (NaI(Tl)-detector, Harshaw Chemie, Belgium) atthe same time.2.3 In vivo experiments2.3.1 Laboratory AnimalsWater and food were ad libitum during the experimental period. For the tumour model, male Wag/Rijrats (Harlan, The Netherlands) were injected subcutaneously in the right flank (armpit region) with 15x 10 6 R1M rhabdomyosarcoma cells. Imaging experiments with 2-[ 18 F]fluoromethyl-L-phenylalaninewere performed 6 weeks after injection of the R1M cells. The average tumour size was 12 cm³. Duringall imaging experiments, the animals were anaesthetized intraperitoneal (IP) with 350 ml of a solutioncontaining 60 mg pentobarbital/ml (Nembutal, 60 mg/ml, Ceva Sante´ Animale, Belgium). 2-[ 18 F]Fluoromethyl-L-phenylalanine was injected intravenously (IV) in the penis vein. The studyprotocol was approved by the ethical committee for animal studies of our institution. Guidelines of theNational Institute of Health principles of laboratory animal care (NIH publication 86-23, revised 1985)were followed.2.3.2 PET acquisitionThe scanner is a 2D/3D human Accel PET camera (Siemens) with a resolution of 6mm at the centre ofthe field of view. R1M bearing Wag/Rij rats were injected with 3.7MBq of 2-[ 18 F]fluoromethyl-Lphenylalanine,which was eluted over a custom made appatite filter just before injection, and imagingwas performed at 5, 15 and 30 min p.i. (1 acquisition sequence lasts about 10 min) in 3D mode. Therats were positioned crosswise to allow a faster acquisition so that coronal slices are seen as sagittalrat-slices. The injected activity was calculated as the amount of radioactivity in the syringe before andafter injection (Capintec CRC-15R, Ramsey, NJ, USA). Quantification of activity uptake occurred byusing the 3D-isocontour ROI feature of the medical imaging data analysis program Amide (VA LinuxSystems), allowing accurate outlining of the tumour and tissues of interest. The ROIs for muscle weretaken in the contra-lateral background region, while the heart was taken as a measure for the bloodpool activity. Differential uptake ratios (DUR) were calculated as (mean tissue activity/voxel)/(meantotal body activity/voxel).3. Results and discussion3.1. Radiosynthesis and formulation of 2-[ 18 F]fluoromethyl-L-phenylalanine

The labeling of Boc-2-bromomethyl-L-phenylalanine-tButylester with [ 18 F - ] provided the Boc-2-[ 18 F]fluoromethyl-Lphenylalanine- tButylester with a radiochemical yield of at least 85 %. Afterdeprotection, adjusting the isotonicity and sterilising the solution, 2-[ 18 F]fluoromethyl-Lphenylalaninecould be obtained ready for use with an overall average radiochemical yield of about 57% (corrected for decay it is 83 % zelfde als artikel), with a radiochemical purity of at least 98%.Radiodefluorination during deprotection and in the aqueous radiopharmaceutical formulation weremajor problems.When applying deprotection on the ACN soluble compounds in the reaction mixture up to 50 % ofradiofluorination was noticed. The presence of CaCl 2 reduces defluorination considerably.Several shelf life tests, where 2-[ 18 F]Fluoromethyl-L-phenylalanine was incubated in milliQ water atroom temperature for several hours, showed that n.c.a. 2-[ 18 F]Fluoromethyl-L-phenylalanine was notstable in aquaous solutions and showed a defluorination rate of 33 %/hour due to hydrolysis. Asignificant stabilisation could be achieved by adding non-radioactive 2-fluoromethyl-L-phenylalaninein a concentration of 50µM to the solution containing 2-[ 18 F]Fluoromethyl-L-phenylalanine, reducingthe radiodefluorination rate to a mere 3 %/hour, as shown in Fig. 1. Applying this method should be atthe cost of the n.c.a. characteristic. The Ca 2 + ions already present from deprotection also inhibits theradiodefluorination of the n.c.a. 2-[ 18 F]Fluoromethyl-L-phenylalanine and represents an interestingalternative.3025+ 2FMethylpheN.C.A.% defluorination201510500 20 40 60 80 100 120Time (min)Fig. 1: Defluorination of [ 18 F]fluoromethyl-L-phenylalanine at room temperature in water as afunction of time, n.c.a. or in presence of 50µM 2-Fluoromethyl-L-phenylalanine (+2FMethylphe)In the animal experiments described in this work n.c.a. 2-[ 18 F]fluoromethyl-L-phenylalanine wasinjected n.c.a.3.2. In vivo results: PET acquisitionFig. 2 depicts the mean DUR values ( 3 R1M bearing rats) of the tumour and different tissues ofinterest at 5, 15 and 30 min p.i..

Fig. 2: Graph, showing the DUR values of the tumour and different tissues of interest at 5, 15 and 30minutes p.i..The PET acquisitions at three time points clearly show a high uptake in the tumour and a fastclearance of the tracer via the kidney to the bladder without renal accumulation. Thetumour/background (contralateral shoulder or flank) ratios were about 5, tumour/heart ratios about 2.5and the tumour/brain ratios about 3 at 30 min p.i. The latter shows that this new tracer allows to obtaingood contrast for brain tumour imaging.4. ConclusionThe new PET-tracer 2-[ 18 F]fluoromethyl-L-phenylalanine can be obtained in a reasonable yield with ahigh radiochemical purity using a radiochemistry similar to [ 18 F]FDG. However, in order to establish areproducible procedure which can be adapted into a routine procedure, a considerable reduction ofradiodefluorination during radiosynthesis and in the radiopharmaceutical formulation is of majorpriority. The new tracer is taken up in a high amount in the tumours in the rats and also shows afavourable biodistribution and clearance. The results coupled to the fact that this compound shows notoxicity allow that n.c.a. 2-[ 18 F]Fluoromethyl-L-phenylalanine could be used in Phase I clinicalstudies.5. AcknowledgementsThe authors thank FWO-Vlaanderen (project G.0299.04N ) and GOA-Vrije Universiteit Brussel(GOA28) for financial support.6. References[1] P.I. Jager, W.V. Vaalburg, J. Pruim, E.G.E. de Vries, K.-J. Langen, D.A. Piers, J. Nucl. Med. 42(2001) 432–445.[2] P. Laverman, O.C. Boerman, F.H.M. Cortens, W.J.G. Oyen, Eur. J. Nucl. Med. 29 (2002) 681–690.[3] D. Pauleit, A. Zimmermann, G. Stoffels, D. Bauer, J. Risse, M.O. Fluss, K. Hamacher, H.H.Coenen, K.J. Langen, J. Nucl. Med. 47 (2)(2006) 256–261.[4] C. Rau, W.A. Weber, H.J. Wester, M. Herz, I. Becker, A. Kruger, et al., Eur. J. Nucl. Med. Mol.Imag. 29 (2002) 1039–1046.[5] T. Miyagawa, T. Oku, H. Uehara, R. Desai, B. Beattie, J. Tjuvajev, et al., J. Cereb. Blood FlowMetab. 18 (1998) 500–509.

[6] O. Yanagida, Y. Kanai, A. Chairoungdua, et al., Biochim. Biophys. Acta 1514 (2001) 291–302.[7] D.B. Shennan, J. Thomson, M.C. Barber, M.T. Travers, Biochim. Biophys. Acta 1611 (2003) 81–90.[8] H. Ohkame, H. Masuda, Y. Ishii, Y. Kanai, J. Surg. Oncol. 78 (2001) 265–271.[9] T. Lahoutte, V. Caveliers, S.M.R. Camargo, R. Franca, T. Ramadan, E. Veljkovic, J.J.R. Mertens,A. Bossuyt, F. Verrey, J. Nucl. Med. 45 (9) (2004) 1591–1596. [10] F. Verrey, Pflugers Arch. 445(2003) 529–533.[10] F. Verrey, Pflugers Arch. 445 (2003) 529–533.[11] V. Kersemans, B. Cornelissen, K. Kersemans, M. Bauwens, E. Achten, R.A. Dierckx, J.J.R.Mertens, G. Slegers, J. Nucl. Med.46 (5) (2005) 32–539.

Session 18.3.3

HEALTH EFFECTS OF LOW DOSESOF IONISING RADIATIONSM.H. BOURGUIGNONCommissioner, French nuclear safety authority (ASN)6 Place du Colonel Bourgoin, 75572 Paris Cedex 12 – Francemichel.bourguignon@asn.frABSTRACTThe health effects of low doses of ionising radiations are controversial since the conceptof low doses remains ambiguous. However, it is clearly established that the risk of thecorresponding stochastic effects, i.e., cancers or hereditary disease, if it exists, is verysmall. The linear no threshold hypothesis cannot be retained to establish a collective risk.Thus progress can only result from research in radiation biology. DNA lesions have beendocumented at doses as low as 1 mGy, but the pathway between one DNA lesion and a canceris not known yet. On the basis that ionising radiations are genotoxic, the regulator must applythe current rules of radiation protection and look forward to the development of new radiationbiology paradigm.1. IntroductionLow doses of ionising radiations are usually defined as the doses below which no health effects havebeen observed, i.e., an effective dose of about 200mSv. This definition raises numerous issues:o Epidemiologic studies have shown that lower values should probably be retained, about100mSv in adults and 50mSv in children.o These values do not take into account the dose rate and the duration of exposure; indeed atotal dose of 100 mSv delivered in a few seconds and in one year are not identical in terms ofeffects because the number and type of DNA lesions and the rate at which they are producedare totally different and the corresponding repair mechanisms of DNA are not involved in thesame way in the 2 situations.o The confusion regarding the definition appears in many publications which are impossible tocategorize : the authors claim the use of low dose of ionizing radiations in experimentalanimals and it appears clearly that the total dose delivered can be greater than 200 mSvalthough delivered at low doses.With such an ambiguous definition of low doses of ionising radiations, it is not surprising that thepotential corresponding health effects are so controversial. By definition, it is not known if below thelevel of dose setting up the upper limit of the low dose domain some stochastic effects do appear.Epidemiology will remain unable to make such a demonstration because of a lack of statistical power.Therefore, since it has not been possible to demonstrate health effects of low doses of ionisingradiations, one can clearly state that the risk of low doses of ionising radiations, if it exists, is verysmall. At this level of doses, only stochastic effects, i.e., cancers or hereditary diseases might beobserved.2. The linear no treshold hypothesisA linear relationship between health effects and doses has been observed at high doses and dose ratesand has been extended as an hypothesis to the low dose domain for radiation protection purposes. Thelinear no treshold hypothesis (LNT) is by definition a hypothesis; since it states that each ray has anhealth effect, it is over-evaluating the risk of ionising radiations. Consequently, the subsequentradiation protection rules, i.e., justification, optimisation and limitation of doses, have proved useful inthe sense that they help protect the workers and the population by decreasing the dose of exposures.

However it is very clear that the LNT remains an hypothesis and a simplification: it is not establishedas a scientific fact ; consequently collective doses must not be used to evaluate the risks related to lowdoses and this has been clearly stated by ICRP in its last recommendations (still to be published).3. Progress in radiation biologyHow to make further progress regarding the potential health effects (cancer or hereditary effect) of lowdoses of ionising radiations ? We have seen above that it is not possible to clearly answer this questionwith epidemiologic studies in humans. The only way to progress is to better understand themechanisms of the effects of ionising radiations at the molecular/cellular/tissular level: progress canonly result from the research in radiation biology.Variation in individual sensitivity to ionising radiation has emerged as a new critical issue (1,2). It isclearly linked to the DNA repair mechanisms and signalling pathways. However the significance ofindividual sensitivity to low doses of ionising radiations needs to be confirmed by further research.The demonstration of the existence of non targeted effects, i.e., bystander effects and genomicinstability, has challenged the existing paradigm that DNA is the prime target of concern (3,4). Thesefindings raise more than ever the possibility that the LNT hypothesis may not be appropriate in manycicumstances.Numerous experiments have demonstrated the potential benefits of low doses of ionising radiations,this phenomenon of adaptive reponse being known as hormesis. These experiments which have beenhighlighted in the report of the French academies show real effects (5). There are no reason not toaccept these results which have been clearly published in peer review journals. However it is also clearthat these effects appear in some paradigms of exposure to ionising radiations, and this does not meanthat they appear in all type of exposures. Furthermore, many of these effects are not true beneficialeffects but appear as a decrease of the detriment caused by a large dose of ionising radiations deliveredafter chronic exposures at low “training” doses in comparison with “untrained” animals.On another hand, one can observe in the litterature the numerous reports describing cellular andmolecular effects of ionising radiations. One of these reports must be highlighted. The demonstrationby Rothkamm (6) that DNA double strand breaks can be demonstrated by anti γH2AX fluorescentantibodies and appear at doses as low as 1 mGy constitutes a major event because this level of dose isabout 100 times lower than previously reported. Indeed it is not a true surprise that ionising radiationscreate DNA lesions, but it is the first time that it is demonstrated for such low doses. It must be noticedthat 1mGy is the level of dose currently delivered by medical imaging examinations (radiology andnuclear medicine).However, there is a very large gap between one DNA lesion and a cancer. An adverse biological effectat the cellular or molecular level does not necessarily mean the appearance of adverse healthconsequences. So far it has been said that about 10 gene mutations are necessary to produce a cancer.Recent findings indicate that it is likely that many more gene mutations are necessary, up to 100.Indeed “cancer genes” are not equivalent : the activation of oncogenes is more powerful than theinactivation of tumor suppressor genes. Because gene mutations and cancer incidence increase in theelderly, mostly because of the inactivation of tumor suppressor genes, it is wise /necessary to limit theexposure to all genotoxic compounds e.g., chemicals, radiations..., which can contribute to thesemutations.Radio-induced hereditary effects have never been demonstrated in humans but have been observed inplants and animals exposed to rather high doses of ionising radiations (range of 1 Gy). The lastscientific results which have been validated by UNSCEAR indicate that the risk of hereditary effectsin humans is lower than initially expected. Consequently, ICRP has retained for its new comingrecommendation an estimation of risk for the hereditary effects 10 times lower than the risk of radioinducedcancer.

Finally, the most recent researches in genomics and proteomics tend to indicate that there might besome signature of radio-induction of cancers. Further research in these domains will certainly bringnew insights on the effects of ionising radiations at the molecular and cellular levels, and on thecellular responses to the initial injuries.4. The position of a regulatorOn the basis of the current knowledge regarding the effects of ionising radiations, what can be theposition of a regulator ?Although the health risk related to the exposure to low doses of ionising radiations is small, it is alsovery clear that ionising radiations are genotoxic. Genotoxicity does not mean health effects, i.e., canceror hereditary disease. But since a cancer results from an accumulation of gene lesions, there are noreason to remain voluntarily exposed to ionising radiations, especially when it is not so difficult todecrease the level of exposure. This is the case for both medical exposures and radon exposures whichare respectively number one in the artificial and natural exposures to ionising radiations, far greaterthan any other type of exposure such as industrial exposure.Consequently, the regulator must apply the current rules of radiation protection to protect the public,the patients and the workers and inform the population of the potential risks of exposure to ionisingradiations.In terms of policy challenge, the development of new radiation biology paradigm, including targetedand non-targeted effects and individual susceptibility to ionising radiations, may require changes to thecurrent system of radiation protection in the future (7).References1- Bourguignon MH, Gisone PA, Perez MR, Michelin S, Dubner D, Giorgio MD, Carosella ED.Genetic and epigenetic features in radiation sensitivity. Part I : cell signalling in raidation response.Eur J Nucl Med Mol Imaging 2005, 32(2): 229-2462- Bourguignon MH, Gisone PA, Perez MR, Michelin S, Dubner D, Giorgio MD, Carosella ED.Genetic and epigenetic features in radiation sensitivity. Part II : implications for clinical practice andradiation protection. Eur J Nucl Med Mol Imaging 2005, 32(3): 351-3683- Morgan WF. Non targeted and delayed effects of exposure to ionising radiations: I. Radiationinduced genomic instability and bystander effects in vitro, Radiation Research 2003, 159: 567-5804- Morgan WF. Non targeted and delayed effects of exposure to ionising radiations: II. Radiationinduced genomic instability and bystander effects in vivo, Clastogenic factors and transgenerationaleffects. Radiation Research 2003, 159: 581-5965- Tubiana M. Dose effect relationship and estimation of the carcinogenic effects of low doses ofionising radiation : the joint report of the Académie des Sciences (Paris) and of the Académienationale de médecine. Int.J. Radiation Oncology Biol.Phys. 2005,63,n°2, 317-3196- Rothkamm K, Lobrich M. Evidence for a lack of DNA double-strand break repair in human cellsexposed to very low x-ray doses. Proc Natl Acad Sci U S A. 2003 ; 100(9):5057-627- Scientific issues and emerging challenges for radiation protection. Report of the Expert Group onthe Implications of Radiation Protection Science (EGIS). OECD/NEA editor, 2007

Session 18.3.4New developments in nuclear imaging

PINHOLE SPECTCHRISTIAN VANHOVEDepartment of nuclear medicine, University Hospital BrusselsLaarbeeklaan 101, 1090 Brussels – BelgiumABSTRACTSingle-photon emission computed tomography (SPECT) is one of the image modalitieswithin a clinical environment. This technique is based on the detection of gamma photonsthat originate from injected radioactive tracers to image the body’s physiology. Nearly allclinical SPECT uses parallel-hole collimation as the image-forming aperture. Thesecollimators are a limiting component for the spatial resolution, typically >1cm. In recentyears, SPECT has become a very attractive modality for in-vivo imaging of physiologicalfunctions in small animals. Although, one might be pessimistic about the utility of SPECTfor imaging small animals due to its inadequate spatial resolution, SPECT imaging usingpinhole collimators can provide sub-millimetre spatial resolution. Pinhole collimation isbased on the camera obscura principle. Through magnification, pinhole imaging reducesthe apparent intrinsic resolution of the gamma detector. This paper demonstrates thatpinhole SPECT combined with advanced software algorithms can provide high-qualitythree-dimensional images of small animals.1. IntroductionSingle-photon emission computed tomography (SPECT) is one of the image modalities within aclinical environment. This technique is based on the detection of gamma photons that originate frominjected radioactive tracers to image the body’s physiology in a three-dimensional manner. Nearly allclinical SPECT uses parallel-hole collimation as the image-forming aperture. These collimators are alimiting component for the spatial resolution, which are typically larger than 1cm. A recent trend innuclear medicine is the growing interest for the application of SPECT in small animal studies. Thedevelopment of new experimental therapies, such as stem cell therapy and gene therapy, and of newcardiovascular drugs requires several stages of animal testing before proceeding to clinical trials.Although in-vitro imaging is a common tool in this field, it requires sacrificing the animals. This canbe a severe limitation because a same animal cannot be followed by sequential studies over time.Furthermore, the cost of in-vitro studies is dramatically higher than that of in-vivo studies, because alarger number of animals are needed, not only because sacrificed animals cannot be reused, but alsobecause a larger number of experiments must be carried out in order to eliminate variability due tosmall physiological differences between normal animals. The understanding of the advantages of noninvasivein-vivo imaging has given rise to the need for high resolution imaging devices that can imagelow concentrations of biochemical agents on very small scales. Although, one might be pessimisticabout the utility of SPECT for imaging small animals due to its inadequate spatial resolution, SPECTimaging using pinhole collimators can provide this required high spatial resolution (1, 2).Historically, the pinhole collimator was the very first collimator used on a gamma camera. Pinholecollimation is based on the camera obscura principle as shown in Fig 1.

Fig 1. Pinhole collimation is based on the camera obscura principle. Here you can see an artist’simpression using early pinhole imaging. However, in a clinical setting, pinhole imaging is not used toshrink an image, but to magnify it with the intention to improve the spatial resolution.Through magnification, pinhole imaging reduces the apparent intrinsic resolution of the gammadetector, resulting in an overall spatial resolution predominantly determined by the diameter of thepinhole opening (Fig 2). Although, sub-millimetre spatial resolution can be obtained by using pinholeinserts with a sufficiently small diameter, the use of pinhole SPECT has been mainly hampered due toa trade-off between the spatial resolution, the sensitivity, and the size of the field-of-view (FOV).Fig 2. With parallel-hole collimation (left) the image obtained of the small frog is degraded by theintrinsic camera resolution. With pinhole collimation (right), the information loss due to intrinsiccamera blurring is suppressed by the magnification.When a system is desired with high spatial resolution, very small pinhole diameters should be usedand the object should be positioned close to the pinhole opening to obtain large magnification.However, small pinhole diameters will reduce the efficiency of photon detection and the largemagnification will reduce the size of the FOV. Moreover, the high spatial resolution will make thesystem very sensitive to imperfect camera motion, which is necessary to perform a tomographicacquisition. This is equivalent to camera shake when taking pictures in conventional photographyusing large zoom lenses.Alternatively, when a system is required that efficiently detect gamma photons, large pinholediameters should be used and the object should be acquired close to the pinhole opening. This isbecause the solid angle, at which the gamma photons from each point in the object are able to passthrough the pinhole opening, increases enormously for points close to the pinhole opening and alsoincreases when the pinhole diameter increases. However, the large magnification obtained in this waywill reduce the size of the FOV and the large pinhole diameter will deteriorate the spatial resolution.Finally, when a large FOV is required, one should acquire an object far away from the pinholeopening to reduce the magnification, which will result in poor spatial resolution and sensitivity.The aim of this paper is to demonstrate that with the development of advanced software algorithmspinhole SPECT imaging can provide high-quality three-dimensional images of small animals, eventhough its resolution/sensitivity/FOV trade-off.2. Image reconstructionTomographic reconstruction algorithms are necessary to transform the acquired two-dimensionalprojection images into a three-dimensional volume. As is done in Computed Tomography (CT), thereconstruction of pinhole SPECT projection images can be performed analytically using a filteredback projection algorithm (3). Although, these algorithms are fast because they have a lowcomputational burden, they are not very adaptable to incorporate the physical properties of theimaging system and to model the transport of the gamma photons from the object to the imagingsystem. Therefore, at present, most pinhole SPECT systems use iterative reconstruction based on theMaximum-Likelihood Expectation-Maximization (MLEM) algorithm (4). Fig 3 illustrates the basic

steps involved in this iterative reconstruction approach. The slow convergence of the MLEMalgorithm results in long reconstruction times, which restricts its practical application. With theintroduction of the Ordered-Subsets Expectation-Maximization (OSEM) approach (5, 6), the algorithmwas made more efficient and could generate usable results within clinically acceptable time limits. TheOSEM algorithm was modified by our group (7) to allow iterative reconstruction of pinhole SPECTimages, by modelling the pinhole SPECT acquisition geometry.Fig 3. Iterative reconstruction. Projection images simulated from the current reconstruction estimateare arithmetically compared with the measured projections and the result is back projected to updatethe reconstruction estimate. The physical properties of the imaging system and image degradingeffects can be modelled in the forward and the back projector.3. Characterization of pinhole SPECT acquisition geometryWhen a system is desired with sub-millimetre spatial resolution, very small pinhole diameters shouldbe used and the object should be positioned close to the pinhole opening to obtain large magnification.This high spatial resolution will make the system very sensitive to imperfect camera motion, motionthat is necessary to perform a tomographic acquisition. To avoid loss of resolution, the reconstructionof the data acquired with a pinhole camera requires a detailed description of the camera geometry,which has to be incorporated into the iterative reconstruction algorithm. This can be achieved by doinga geometric calibration of the pinhole SPECT system (8, 9). The technique first acquires circular orbitSPECT projection data of a calibration object consisting of three point sources, which form a trianglewith known dimensions. The location of the three sources on each projection image is then determinedusing simple segmentation techniques. By iteratively fitting analytically calculated locations to themeasured locations, three intrinsic and six extrinsic parameters can be determined to accuratelydescribe the pinhole acquisition geometry. The three intrinsic parameters are independent of theposition of the detector during the pinhole SPECT acquisition and give information about the focallength of the pinhole collimator and the two coordinates of the orthogonal projection of the pinholeopening onto the surface of the gamma detector. For each position of the gamma detector during atomographic acquisition, six extrinsic parameters give information about the three-dimensionallocation of the pinhole opening in space together with three angles to determine the orientation of thegamma detector. Fig 4 shows the effect of accurately describing the pinhole acquisition geometryduring reconstruction.

Fig 4. Trans-axial slice of a phantom containing hot rods with diameters varying from 2mm down to1mm in steps of 0.2mm. On the left, three intrinsic parameters and only one extrinsic parameter wasused to describe the pinhole geometry, on the right side three intrinsic parameters and six extrinsicparameter were used to describe the pinhole SPECT acquisition geometry.4. Modelling the finite dimensions of the pinhole openingWhen a system is required that efficiently detect gamma photons, a large pinhole diameter should beused and objects should be scanned close to the pinhole opening, to increase the solid angle at whichthe gamma photons from each point in the object are able to pass through the pinhole opening.However, a large pinhole diameter will deteriorate the spatial resolution. This resolution-sensitivitytrade-off can be improved by modelling the finite dimension of the pinhole opening during iterativereconstruction. During the process of reconstruction it is usually assumed that the pinhole opening isinfinitesimally small. This assumption is acceptable for very small pinhole openings but may introduceartefacts when using larger pinhole openings. To incorporate the finite dimension of the pinholeopening an approach based on multi-ray projections can be used (10). The method is based on thetheory of Gaussian quadratures and prescribes a set of rays that can be incorporated in the forward andback projector of an iterative reconstruction algorithm to model the pinhole opening. Instead of usingone ray, when assuming an infinitesimally small pinhole diameter, typically seven or twenty-one raysare used in the forward and back projector of the OSEM algorithm to model the finite dimension of thepinhole opening. The effect of resolution recoveryis illustrated in Fig 5.Fig 5. Trans-axial slice of a phantom containing hot rods with diameters varying from 3mm down to1.5mm in steps of 0.3mm, reconstructed assuming that the pinhole opening was infinitesimally small(left) and reconstructed using the multi-ray technique modelling a 3mm pinhole opening (right).5. Multiple-pinhole SPECTAs mentioned above, the pinhole collimator can deliver high spatial resolution and improved detectionefficiency when imaging small subjects close to the pinhole aperture. However, this comes at theexpense of a severely reduced FOV. By setting up additional pinholes and focus them on differentregions in the FOV, it is possible to increase the size of the FOV. We have developed a three-pinholecollimator to obtain an axial FOV of 160 mm (11), allowing to reconstruct three-dimensional wholebodyimages of rats. Fig 6 illustrates the principle of image formation using this three-pinholecollimator. To characterize such a three-pinhole system, the calibration procedure described in thethird paragraph can be used. However, the three intrinsic parameters should be defined for each of theindividual pinhole openings.

Fig 6. Principle of image formation when using a three-pinhole collimator, each pinhole focussing onanother part of the object. The detector surface is tiled with three different projections of the object.6. ConclusionsThe resolution/sensitivity/FOV trade-off in pinhole SPECT can be improved using advanced softwarealgorithms. Multiple-pinhole systems, geometrically characterized with calibration software, used incombination with iterative reconstruction algorithms that allows incorporating this pinhole acquisitiongeometry, can extend the axial FOV. By modelling the finite dimensions of the pinhole aperturesduring the iterative reconstruction process, larger pinhole openings can be used to improve theefficiency gamma photons detection while remaining good spatial resolution.7. References1. Beekman F, van der Have F. The pinhole: gateway to ultra-high-resolution three-dimensionalradionuclide imaging. European Journal of Nuclear Medicine and Molecular Imaging. Feb2007;34(2):151-161.2. Madsen MT. Recent advances in SPECT imaging. J Nucl Med. Apr 2007;48(4):661-673.3. Feldkamp LA, Davis LC, Kress JW. Practical cone-beam algorithm. J Opt Soc Am.1984;A1:612-619.4. Lange K, Carson R. Em Reconstruction Algorithms for Emission and TransmissionTomography. Journal of Computer Assisted Tomography. 1984;8(2):306-316.5. Hudson HM, Larkin RS. Accelerated Image-Reconstruction Using Ordered Subsets ofProjection Data. Ieee Transactions on Medical Imaging. Dec 1994;13(4):601-609.6. Hutton BF, Hudson HM, Beekman FJ. A clinical perspective of accelerated statisticalreconstruction. European Journal of Nuclear Medicine. Jul 1997;24(7):797-808.7. Vanhove C, Defrise M, Franken PR, Everaert H, Deconinck F, Bossuyt A. Interest of theordered subsets expectation maximization (OS-EM) algorithm in pinhole single-photon emissiontomography reconstruction: a phantom study. European Journal of Nuclear Medicine. Feb2000;27(2):140-146.8. Beque D, Nuyts J, Bormans G, Suetens P, Dupont P. Characterization of pinhole SPECTacquisition geometry. Ieee Transactions on Medical Imaging. May 2003;22(5):599-612.9. Defrise M, Vanhove C, Nuyts J. Perturbative refinement of the geometric calibration inpinhole SPECT. Ieee Transactions on Medical Imaging. 2007:In Press.10. Vanhove C, Andreyev A, Defrise M, Nuyts J, Bossuyt A. Resolution recovery in pinholeSPECT based on multi-ray projections: a phantom study. European Journal of Nuclear Medicine andMolecular Imaging. Feb 2007;34(2):170-180.11. Vanhove C, Defrise M, Bossuyt A. Three-pinhole collimator to improve axial spatialresolution and sensitivity in pinhole SPECT. J Nucl Med. 2007;48:428P.

EXPERIENCE WITH A COMBINED PET/SPECT SCANNERFOR SMALL ANIMAL IMAGINGN. BELCARI, A. DEL GUERRA, A BARTOLI, S. FABBRI, D. PANETTADepartment of Physics “E. Fermi” and Center of Excellence AmbiSEN, University of Pisa and INFNLargo Bruno Pontecorvo, 3, 56127 Pisa – ItalyABSTRACTThis paper reports a brief review of the state of the art in PET, SPECT and multi-modalityscanners for small animals. At the University of Pisa a prototype version of thePET/SPECT scanner for small animal YAP-(S)PET II is available. The performance inboth PET and SPECT modalities are reported together with some examples of experimentalapplications on small animals. X-ray transmission images of a mouse obtained with a highresolution CT prototype are also shown. This CT scanner will be soon plugged on theYAP-(S)PET II.1. IntroductionThe study of biochemical processes at a molecular level is of great importance for pharmacology,genetic, and pathology investigations. This field of research is usually called “molecular imaging” [1].In-vivo imaging techniques for small animals [2] have demonstrated to be very valuable investigationmethods for molecular imaging in the pre-clinical phase. Among the various imaging techniquesPositron Emission Tomography (PET) and Single Photon Emission Computed Tomography (SPECT)have a great success that is largely due to the continuous development of high resolution, highsensitivity instrumentation for gamma ray detection. The latest advances in radiation detectiontechnology have strongly improved the performance of small animal scanners with respect to theclinical ones. This is especially true for the spatial resolution figure of merit so as to fulfil therequirements for small size animal study. On the shadow of the successful application of combinedPET-CT scanners in the clinical environment this combined technique has been recently transferred tosmall animal scanners.2. State-of-the art in PET and SPECTThe relatively small size of the objects under study in small animal imaging (small organs or brainregions of rats and mice), makes it difficult the use of imaging instruments developed for humansubjects, i.e., the spatial resolution of the available clinical PET scanners is not satisfactory for thequantitative and qualitative assessment of in vivo gene expression [1]. Molecular small animalimaging requires instruments with a finer spatial resolution than that of the available clinical scanners,that is not better than 4-6 mm FWHM. In small animals, it is usually acceptable to work with a spatialresolution better than 2 mm FWHM for rats (for example for the imaging of the brain), whilst for miceit would be ideal to use instruments with a resolution better than 1 mm FWHM.2.1 PETThe best achievable spatial resolution in PET is limited because of both the physics of the β+ decay(positron range and angular deviation from collinearity) and the available technology for the positiondetection of two gamma rays in coincidence (crystal size, crystal position readout coding and imagereconstruction algorithm). The best spatial resolution of a PET scanner can be expressed in terms ofthe FWHM of the point spread function (PSF) after a filtered backprojection (FBP) reconstructionwith the following formula [3]:

2⎛ d ⎞ 22 2 2FWHM = 1.2 ⎜ ⎟ + b + ( 0.0022D) + r + p⎝ 2 ⎠where: 1.2 = degradation factor due to tomographic reconstruction; d = crystal pitch; b = coding error;D = tomograph ring diameter; r = effective source diameter including positron range; p = parallaxerror. Considering all the contributions included in the formula it can be found that the spatialresolution in PET is intrinsically limited and cannot be better than 0.7-0.8 mm FWHM. The presentbest scanners offers a spatial resolution which is of the order of 1.4-1.6 mm FWHM at the center ofthe field-of-view (FOV). This is good enough for most cases (especially for rats). On the other hand,the sensitivity of small animal PET scanners has now reached values of the order of 8-10% at thecenter of the FOV.Siemens MicroPET® FocusTM 120 (http://www.medical.siemens.com/) and GE eXplore Vista(http://www.gehealthcare.com/usen/fun_img/pcimaging/products/vista.html) systems represent thestate-of-the-art in high resolution small animal PET scanners. These systems offer a very highresolution and a good sensitivity. For both systems a special care is devoted to keep the spatialresolution as constant as possible over the whole FOV. Two different approaches are used:microPET® FocusTM 120 uses thinner crystal elements with a larger gantry aperture, while GEeXplore Vista makes use of a more complex phoswich technology to limit the parallax error. Theperformance of these systems (expressed as the best value measured at the center of the FOV) offers aspatial resolution of 1.3 mm FWHM and 1.4 mm FWHM and a sensitivity of 6.5% and 4% for theSiemens and GE scanners, respectively. Other examples of small animal PET scanners available on themarket are the HIDAC-PET (http://www.oxpos.co.uk/), that uses a detector technology based on gasmultiwire proportional chambers coupled to solid, high Z, converters and the LabPET from AdvancedMolecular Imaging (AMI Inc.), that is based on a one-to-one coupling between APD’s and phoswichLYSO/LGSO crystal assembly.2.2 SPECTSPECT does not suffer from intrisic spatial resolution limitations. On the other hand, the main goal isto find the optimal compromise between spatial resolution, sensitivity and size of the field-of-viewthat drives the choice of the collimator type and dimensions as well as the source to collimatordistance and detector intrinsic resolution. For example, the use of pinhole collimators is a solution forultra high resolution SPECT. A pinhole collimator consists in a single hole shaped like a double cone.This type of collimators are usually made of high Z materials (such as gold) in order to reduce theradiation penetration at the edge of the hole. For this type of collimators the FWHM of the spatialresolution is given by [4]:FWHM = D e ⋅ (d + b) /bwhere d is the distance between the object and the pinhole and b is the distance between the pinholeand the scintillator. D e represents the effective diameter of the pinhole, depending also on theattenuation coefficient of the collimator material and on the hole aperture. Differently from parallelholes collimator, the sensitivity of a pinhole is proportional to 1/d 2 . Thus, the closer is the object to thehole the higher is the sensitivity. However, the sensitivity rapidly decreases as the distance increases.By using more holes (multi-pinholes) it is possible to enlarge the FOV to obtain “whole-body” highresolution images of small animals.X-SPECT® by Gamma Medica-Ideas (http://www.gammamedica.com/X-SPECT.php) andNanoSPECT system by Bioscan (http://www.bioscan.com/product.php?p=nanospect) are examples ofcommercially available SPECT imaging system for small animals. They are both based on rotatingNaI detectors. These systems can be equipped with a wide range of collimators in order to give accessto the best spatial resolution/sensitivity/FOV size compromise for the protocol under study. By usinghigh resolution pinhole collimators they can reach a spatial resolution in the submillimeter range.2.3 Multi-modality systemsMulti-modality systems handle functional imaging provided by the SPECT and/or PET modalities,fused with high-resolution anatomical imaging provided by X-ray CT. In most cases the differentmodalities are combined by assembling two or more scanners that have the same axis. In this way thedifferent systems are accessible with the same animal bed and the image fusion can be obtained by a

1-D translation. The CT part usually consists of a rotating flat-panel X-ray detector and a microfocusX-ray source.An example of such a system is the X-FLEX preclinical imaging platform from Gamma Medica-Ideas.In this case the already mentioned X-SPECT (see section 2.2) is combined with a PET ring called X-PET and a CT scanner called X-O. A similar configuration is offered by the Siemens Inveon systemthat provides integrated small animal PET, SPECT and CT with multiple configuration options. ThePET part is based on the MicroPET Focus technology upgraded in sensitivity (10 % at the center ofthe FOV) and in field-of-view (12.7 cm axially). The SPECT part is made up of two or four NaIpixellated detectors. Conversely to X-FLEX, the Inveon CT scanner is not separated from the SPECTscanner but they share the same rotation gantry and are co-planar. For this reason the CT is availablewith two SPECT head configuration only. Also the NanoSPECT is available with a CT scanner. Inthis case the CT shares the same rotational gantry with SPECT but it is displaced axially. It is made upwith a CMOS flat panel detector coupled to a microfocus X-ray source.Among the commercial systems, YAP-(S)PET II (http://www.ise-srl.com) is the only scanner thatincludes PET and SPECT on the same gantry [5]. This is possible because of the four-planar rotatingscintillator detectors, made up of a pixellated, medium Z scintillator (YAP:Ce): each matrix iscomposed of a 4 cm × 4 cm YAP:Ce matrix of 27 × 27 elements, 1.5 × 1.5 × 20 mm 3 each, and iscoupled to a PS-PMT. The SPECT imaging is obtained by simply adding a lead parallel holecollimator (0.6 mm Ø, 0.15 mm septum) in front of each crystal. Such architecture has the advantageof a reduction of number of detectors, meaning simplicity and affordability in terms of cost andmaintenance, but at the same time maintaining the resolution and sensitivity as required for preclinicalapplications. The system operates in 3-D data acquisition mode and both FBP (Filtered BackProjection) and EM (Expectation Maximization) algorithms can be used for image reconstruction. Forboth PET and SPECT modalities the scanner has an axial field of view of 4 cm and the diameter ofthe transaxial FoV is 4 cm.Fig 1. Pictures of the YAP-(S)PET II scanner prototype installed at the University of Pisa3. The YAP-(S)PET scanner prototype of the University of PisaAt the Department of Physics of the University of Pisa, a prototype version of the YAP-(S)PET IIsmall animal scanner is installed (figure 1). In this version the distance between the opposing headscan be varied between 100 mm and 250 mm. In this way it is possible to tune the sensitivity-to-spatialresolution compromise according to the experiment requirements. In fact, the sensitivity in PET modeat the center of the FOV is 17 cps/kBq at 150 mm, 24 cps/kBq at 125 mm and 35 cps/kBq at 100 mmwhile the spatial resolution slightly improves as the distance increases. The peak NEC is nearlyindependent from the head-to-head distance (about 42 kcps) but it is reached for increasing values ofactivity with increasing distance. Figure 2 reports the spatial resolution measured at the standard headto-headdistance of 125 mm in the central plane. The FWHM of the images of a 22 Na point source (1mm × 1 mm ∅ nominal size) reconstructed with FBP is plotted against the radial position of thesource. On the other hand when working in SPECT mode the minimum distance, according to theanimal size, is always preferable in order to maximize the spatial resolution. In the SPECT modalitythe spatial resolution measured on a glass capillary (1.0 mm nominal diameter) filled with a 99m Tcsolution is 2.9 mm FWHM using FBP. The measured SPECT sensitivity is 30 cps/MBq, in the 140-250 keV energy window, constant over the whole FOV.

Fig 2. Plot of the spatial resolution measured in PET mode. The FWHM of the profile of the image ofa 22 Na point-like source is plotted against the radial position in the FOVA B CFig 3. Images of the mini Derenzo phantom. The diameter of the rods are 3.0 mm, 2.5 mm, 2.0mm, 1.5mm and 1.2 mm; the center-to-center distance is twice the diameter. Left: PET images ( 18 F)reconstructed with 2D-FBP+FORE (A) and 3-D OSEM (B). Right: SPECT image ( 99m Tc)reconstructed with 2-D FBP (C).As an example of the spatial resolution figure of merit in both PET ( 18 F) and SPECT ( 99m Tc)modalities we have used a mini Derenzo phantom. Reconstructed images are shown in figure 3.We have used the YAP-(S)PET II scanner prototype in PET and SPECT for animal imaging. Thesimple procedure for switching between the two modalities allow us the user to perform both PET andSPECT experiment in the same day. As an example of PET imaging, figure 4 shows the glucosemetabolism of a normal rat obtained injecting 37 MBq of 18 F-FDG and acquiring for 45 min. after 30min. of uptake. Principal structures such as cortex, thalamus and striatum are visible.Fig 4. Coronal sections through a rat brain obtained in PET with [ 18 F]-FDG.Examples of the experiments performed in SPECT mode on rats are shown in figure 5.A CT scanner has been also developed using a CMOS flat panel detector (2048×1048 pixels, 48 μmpitch coupled to a Gd 2 O 2 S scintillator screen) and a microfocus X-ray source. The CT will be soonintegrated with the YAP-(S)PET II prototype. An image of a mouse is shown in figure 6. The imagewas obtained with a scan time of 200 seconds (single view, axial FOV 40 mm).

Fig 5. Left: rat myocardium perfusion studies performed in SPECT mode ( 99m Tc-Myoview). Right:Double tracer study on a rat heart with ischemia and subsequent re-perfusion; short axis, vertical longaxis and horizontal long axis projection are shown from left to right. Top row: perfusion study ( 99m Tc-Myoview), central row ischemia study ( 99m Tc-Annexin), bottom row: fused images.Fig 6. Left and center: transaxial and horizontal sections ( 80μm thick) of the mouse skull and thorax(single bed view). Right: side view of the volume rendering of the image. The acquisition parameterswere 40kV, 250 μA, 1mm Al filtering, continuous rotation (500 views over 360°). Total acquisitiontime 200 s. Reconstruction parameters: 512 3 image size (binning two), 80×80×80 μm 3 voxel size.4. ConclusionsOur experience with the YAP-(S)PET II indicates that its spatial resolution and sensitivity areadequate for molecular imaging investigation in both PET and SPECT. The availability of bothemission techniques on the same gantry allows multimodality study in a very easy and effective way.The future availability of an integrated CT will be a critical improvement for a better visualization ofanatomical repere, attenuation correction and morphological characterization.5. References[1] T.F. Massoud, S. S. Gambhir, Genes & Development 17 (2003) 545-580.[2] A. Del Guerra, N. Belcari, Quarterly Journal of Nuclear Medicine, 46(1) (2002) 35-47.[3] S. Derenzo, W.W. Moses, "Critical instrumentation issues for resolution

Session 18.3.5New developments in radiotherapy

TUMOUR MOTION TRACKING: EXPERIENCES WITHMICROSYSTEMS & SENSORS FOR RADIOTHERAPYM. BANDALA, M. J. JOYCEEngineering Department, Lancaster UniversityLancaster LA1 4YR – United KingdomABSTRACTThe objective of this research is to address a significant need that currently exists inmedical radiotherapy. Radiotherapy is compromised by the mobility of tumours in thechest. The motion induced while breathing often makes it difficult to target tumours,meaning that patients often have to endure extended treatment times or carry out difficultbreath-control techniques. A way to account for such motion is often desirable duringradiotherapy treatments. By using tumour tracking, physicians can irradiate tumours moreaccurately without exposing the healthy tissue around the tumour to radiation.A diminutive inertial sensor developed at Lancaster is capable of measuring position andorientation about three orthogonal axes. The sensor is intended for tracking the positionand orientation of a medical target within the human body without the need for an externalimaging system. Embodiments of this sensor may be applied to tumour tracking inradiation oncology.1. IntroductionThe major goal of radiotherapy is the delivery of a prescribed radiation dose as accurately as possibleto a tumour region while minimising the dose distribution to the surrounding healthy tissue. There areseveral factors which tend to compromise this goal such as improper placement of shielding blocks,shifting of skin marks relative to internal anatomy, incorrect beam alignment and, of course,movement of internal organs and tumours. The source of internal organ motion varies from case tocase, but generally falls into one of the following: organ filling and wriggling [1], peristalsis anddigestive processes that alter the contents of the waste-disposal organs [2], standard behaviour of theskeletal muscular, cardiac or gastrointestinal systems [3], and finally and the more significant sourceof internal motion: the respiratory system.It is well know that internal organs in the chest move due to breathing, and some tumours may moveby as much as few centimetres [4]. Radiotherapy, has traditionally addressed the issue of tumourmotion by treating what is known as the planning target volume (PTV). This includes not only theregion where the clinical target volume (CTV) is during imaging and planning, but also where it mightbe during the treatment. These tumour definitions where formally adopted in 1993 [5] by the medicalcommunity.2. BackgroundIn an ideal case, treating the CTV would be preferable, because in doing so, healthy structures are notover-irradiated. Instead, the bigger and ‘static’ PTV is generally treated since no method to ‘view’mobile tumours during treatment is normally utilized. X-rays imaging systems cannot sensibly be usedto view tumour motion throughout therapy fractions because of the unwanted increase to the radiationdose that this would cause. Instead, surrogates [6], and respiration control methods [7] are sometimesemployed.

Even though the use of surrogates (as in gated-radiotherapy) has demonstrated great effectiveness,these techniques raise another issue: the increase of the length of the treatment, which is neverdesirable in clinical practise. To account for such issues, alternative methods to track tumour motionare proposed by the scientific community.The CALYPSO system [8] is a method to perform continuous target location tracking based upon ACelectromagnetic technology. This system utilizes permanently implantable wireless transponders of 8mm x 1.85 mm diameter, without the need of additional ionizing radiation. The PeTrack [9] proposes areal-time tumour tracking system using implanted positron emission markers, containing low activitypositron emitting isotopes, such as 124 I, 74 As, or 84 Rb with half-lives comparable to the duration ofradiation therapy (from a few days to a few weeks). The size of the proposed PeTrack marker will be500-800 micrometres. Position detection requires external instrumentation for both the CALYPSO andthe PeTrack systems.3. Active navigational approachThe Powered, IN-vivo POsition Indication for Therapies (PINPOINT) system was developed atLancaster University as an inexpensive method to detect position and orientation of a moving target.The system is comprised by a six-degrees-of-freedom sensor and an embedded DSP that conform aninertial navigation system (INS).An INS uses accelerometers to measure the acceleration for object position and gyroscopes to measurethe orientation of the object. Ideally, both are deployed in orthogonal triples (for 3D position in X; Y,and Z, and 3D orientation in roll, pitch, and yaw, in order to estimate 6D pose [10]. Fig. 1 illustratesthe general operation of an INS.Fig 1. Inertial navigation performance defining the six degrees of freedom:three components of translation, and three components of rotation.The prototype consists of two symmetrically- bonded PCB boards of size of 30 mm x 20 mm x 10mm. With conditioned analog signals that are digitised, possessed, and transmitted wirelessly viaBluetooth; the power consumption is comparatively low, less than 20 mA at 3.3 V. The practicalimplementation of the sensor is shown in Fig. 3.4. Tracking algorithmInertial sensors detect and measure motion based on the laws of nature and do not rely on externalsignals. Therefore, this tracking method does not require external instrumentation. The only drawbackis the need of considering the gravity component.In order to know how much gravity affects the axes where the accelerometers are, let us consider therelation between the sensor’s reference frame and the earth’s reference frame to witch the gravity

component is linked (see fig. 2). If the vectors associated to the accelerometers are u x , u y and u z , it isnecessary to establish a equation to know their components when the systems rotates around θ 1 , θ 2 andθ 3 . ThereforeWhere x, y and z are the components of vector u x , u y and u z . Andu ' = uR φ(1)Rφ2⎛ tx + c⎜= ⎜txy− sz⎜⎝txz+ sytxy + szty2+ ctyz − sxtxz − sy⎞⎟tyz + sx⎟2tz + c⎟⎠(1)If K 0 is the accelerometer scale factor given in V/g, then the accelerometers output is affected by theaddition of the gravity offset V GV = K × cosθG 0orV Gg ⋅ u= K 0×(3)gFig 2. Representation of the earth’s reference frame (black axes)against the sensor’s reference frame (grey axes).If V 0 is the accelerometer output voltage and V Z the component when the device in immobile, then thefinal acceleration is:V0VG VZA = + −(2)K0The computation of velocity and position are made by software by implementing numericalintegration algorithms:1v [ tn] = ∑ ( tn− tn− 1) ⋅ ( a[ tn] + a[ tn−1]) + v[ tn−1](4)2i=1i=11s [ tn] = ∑( tn− tn− 1) ⋅ ( v[ tn] + v[ tn−1]) + s[ tn−1](5)2

Fig 2. 30 mm x 20 mm x 10 mm prototype of the inertial sensor.5. Experimental setupThe sensor was located inside a custom-made gimballed gyroscope (Lancaster University) for rotationtests (fig. 4a). Similarly, it was mounted on a 2-axis rig, capable of moving a small platform in twodimensions. Each individual degree of freedom was tested independently. Both the gimballedgyroscopes and the rig were marked using standard dimension measurement tools, so that the markscould be used to compare against the displacement and the angular motion of the sensor. The testingwas repeated ten times. The results of this particular experiment have been submitted to publication[11].Fig 4. (a) The sensor in a gimballed gyroscope for testing. (b) The sensor in a 2-axis rig for testing.6. Radiation effectsAs the sensor will remain implanted during the course of treatment, the total radiation dose will be afactor that may limit the operational lifetime of the sensor internal electronics. The electrical propertiesof solid-state components change upon exposure to radiation [12]. As the dose accumulates, thesechanges drive the component parameters outside of the design range for the circuits in which they areused. Ultimately, these changes cause the circuit to cease proper functioning. We are carrying out a setof experiments to determine the hardness of the DPS core. The DSP will be given similar doses as acommon treatment course. The results of this experiment will be presented in the ENC, September2007.

7. References[1] Jiang S. B. Radiotherapy of mobile tumors. Seminars in Radiation Oncology, 16:239–248, 2006.[2] Webb S. Tumour motion: many solutions to one problem. Medical Physics Web, [Online].Available: http://medicalphysicsweb.org/opinion, November 2006.[3] Keall P. J. Mageras G. S. Balter J. M. Emery R. S. Forster K. M. Jiang S.B. Kapatoes J. M. Low D.A. Murphy A. J. Murray B. R. Ramsey C. R. VanHerk M. B. Vedam S. S. Wong J. W. Yorke E.The management of respiratory motion in radiation oncology. Medical Physics, 33(10):3874–3900, 2006.[4] Murphy M. J. Tracking moving organs in real time. Seminars in Radiation Oncology, 14(1):91–100, 2004.[5] ICRU Report 50. Prescribing, recording, and reporting photon beam therapy. InternationalCommission on Radiation Units and Measurements, 1993.[6] Webb S. Motion effects in (intensity modulated) radiation therapy a review. Physics in Medicineand Biology, 51:403–425, 2006.[7] Keall P. J. Mageras G. S. Balter J. M. Emery R. S. Forster K. M. Jiang S.B. Kapatoes J. M. Low D.A. Murphy A. J. Murray B. R. Ramsey C. R. VanHerk M. B. Vedam S. S. Wong J. W. Yorke E.The management of respiratory motion in radiation oncology. Medical Physics, 33(10):3874–3900, 2006.[8] J. M. Balter J. N. Wright L. J. Newell B. Friemel S. Dimmer Y. Cheng J. Wong Vertatschitsch E.Mate T.P. \Accuracy of a wireless localization system for radiotherapy," International Journal ofRadiation Oncology, Biology, Physics, 61(3):933-7, 2005.[9] T. Xu J. T. Wong P. M. Shikhaliev J. L. Ducote M.S. Al-Ghazi S. Molloi. Real-time tumortracking using implanted positron emission markers: Concept and simulation study," MedicalPhysics, 33(7):2598{2609, 2006.[10] M. S. Grewal L. R. Weill A. P. Andrews Global Positioning Systems, Inertial Navigation, andIntegration. John Wiley and Sons, Inc., New York, p131, 2001.[11] Bandala M. Joyce M. J. Tracking of internal organ motion with MEMS: concept and simulationStudy. Meas. Sci. Technol. 2007[12] Benson C. Price R. A. Silvie J. Jaksic A. Joyce M. J. Radiation-induced statistical uncertainty inthe threshold voltage measurement of MOSFET dosimeters. Institute of Physics Publishing,Physics in Medicine and Biology, 49(2004) 3145-3159.

Session 19.3.1Human performance in safety andmaintenance

MAN, TECHNOLOGY AND ORGANISATION IN CURRENTUTILITY OPERATION AND PERSPECTIVE FOR FUTURETEA BILIĆ ZABRICEngineering, Nuclear Power Plant KrškoVrbina 12, 8270 Krško –SloveniaMILENA ČERNILOGAR RADEŽInspection, Slovenian Nuclear Safety AdministrationŽelezna cesta 16, 1001 Ljubljana –SloveniaABSTRACTMan, technology and organisation are essential attributes of a healthy nuclear safetyculture, with a goal of creating a framework for open discussion and continuing evolutionof safety culture throughout the commercial nuclear electric generating industry.A variety of operation experiences over the years have improved the safety culture atnuclear electric generating plants. The nuclear industry does not allow decreasing thesafety margins, which could cause occurrences. Many fundamental principles involvinghardware, procedures, training, and attitudes toward safety and regulation contributed tothe error prevention.Response from operation experience from both industry and regulatory organisations wasextending. Improvements were made in standards, hardware, emergency procedures,processes, training (including simulators), emergency preparedness, design andconfiguration control, testing, human performance, and attitude toward safety.But still, analysis of the available nuclear operating experience shows that humanperformance is a key factor in a large proportion of events.1. IntroductionSafety culture, what is a safety culture? “An organisation’s values and behaviours – modelled by itsleaders and internalised by its members – that serve to make nuclear safety the overriding priority”.Therefore, safety culture has to be inherent in the thoughts and actions of all the individuals at everylevel in an organization. The principles and associated attributes for a strong safety culture have abasis in plant events and operation experience. These principles and attributes influence values,assumptions, experiences, behaviours, beliefs, and norms that describe how things are done at specificutility.Utility leading personnel have to be encouraged to make comparisons between established principlesand their day-to-day policies and practices and to use any differences as a basis for improvement.Subcontractors, vendors, third-party engineering and authority support should be included in theefforts of improving safety culture. Safety culture awareness and recognition of the importance ofnuclear safety and its value to the organisation should be reflected in supplemental personnelperformance.Analysis of the available nuclear operating experience shows that man - human performance is a keyfactor in a large proportion of events. The safety gains achievable through improved humanperformance become increasingly important in deciding where to apply resources; “Arrives the time toinvest in people, not only in equipment”. Plant management establishes performance standards and1

equirements for the conduct of plant activities that are consistent with corporate policies andobjectives, nuclear safety philosophy and other requirements. Management also establishes methods tocommunicate effectively upward, downward and horizontally within the organisation.The organisation of the plant helps achieve a high level of performance and nuclear safety during plantoperation and shutdown conditions through effective implementation and control of plant activities.The organisation provides the administrative and functional structure that determines where people areassigned and defines how they are expected to accomplish their tasks. The complexity of a nucleargenerating plant requires that the organisation be clearly defined.Policies, directives, procedures, goals and objectives and performance standards provideadministrative controls and management direction to implement the organisational structure.The basic assumptions that have worked well in current organisation should be considered as a valid tonew organisation and facilities built in the future.2. Men, technology and organisationEarlier industry attempts to improve human performance focused on results and individual behaviourat the worker level, a common response to human error that exists today in many organisations.However, organisation and management influences on human behaviour are equally important, butoften overlooked or underestimated. Experience has revealed that most causes of human performanceproblems exist in the work environment, indicating weaknesses in organisation and management.To optimise successful performance, appropriate individual and leader behaviours must occur togetherwith appropriate organisational processes and values. All three must work together during all phasesof a work, from work identification through completion of documentation.2.1 Man-Human Performance and ManagementIt was learned form industry: “When a program agrees to spend less money or accelerate a schedulebeyond what the engineers and program managers think is reasonable, a small amount of overall riskis added. These little pieces of risk add up until managers are no longer aware of the total programrisk, and are, in fact, gambling”Human performance is a series of behaviours executed to accomplish specific task objectives (results).Behaviour is what people do. Results are achieved by behaviours, the mental and physical efforts toperform a task. Although results that add value are important, desired behaviour must be the target forimprovement efforts.We have to be aware that excellence in human performance becomes an important measure to reduceevents. A key way to improve human performance is to focus on instilling in people the correctbehaviour patterns in all stages of the NPP life cycle. The installation of this ‘ethos’ is of much greaterimportance than attempting to solve post event problems created by its lack. Human error is caused bya variety of conditions related to individual behaviour, management and leadership, organisationalprocesses and values. Alignment involves facilitating organisational processes and values to supportdesired behaviour.Particular organisation has to establish its own principles for excellence in human performance topromote behaviours throughout an organisation that support safe and reliable operation of the plant.Progress toward excellent human performance requires a work environment in which individuals andleaders routinely exhibit desired behaviours. Such behaviours must be clearly described,communicated and what is a most importantly reinforced. Open communication and positivereinforcement can establish a culture in which individuals, leaders and organisational processeseliminate barriers to excellent human performance, what results in reduce or even elimination ofsignificant plant events due to human error. Experience with error prevention and human performanceimprovement has revealed that the most capable defences against events are open communication andpositive reinforcement of desired behaviours.How to reach environment where we can proudly say ‘that is the way we do things here’?.Achievement of that goal includes:2

• Systematic performance assessment of the operational experience related with humanfactors/human errors, following worldwide good practices criteria and specific experience ofoperators internationally• Establishment of an analysis tool for human errors• Establishment of measures for human error prevention• Establishment of a system for near misses reporting and resolution• Safety culture improvement and safety culture audits• Establishment of basis for safety culture assessment.2.2 TechnologyChanges to plant design and improved equipment deliver high levels of safety. It based on theimplementation of operating experience programme. This contributes to safe, reliable and efficientnuclear plant operation. Events, which occur today, reveal a completely new cause or failuremechanism. Most investigations find that external operating experience was available which, if usedeffectively, could have prevented recurrence of the event. The best use of operating experience is tolook for similarities that could apply to particular plant, rather than look for differences that shouldlead to screen the experience out.Effectively using operating experience includes analysing both internal and external operatingexperience to identify fundamental weaknesses and then determining appropriate plant specificcorrective actions that will minimise the likelihood of similar events. Plants have to be encouraged toshare operating experience.The plant’s goal for operating experience is, to effectively and efficiently use lessons learned fromplant and external operating experience to improve plant safety and reliability. Learning and applyingthe lessons from operating experience is an integral part of plant safety culture and has to beencouraged by managers throughout the organisation. Plant personnel have to be encouraged to useoperating experience information at every opportunity as a helpful. Strength methods of usingoperating experience assured provision of applicable information to the right personnel in write time.When plant personnel analyse the causes of plant events, operating experience has to be routinelyreviewed to determine if and why previous lessons were not effectively learned.2.3 OrganisationAn effectively implemented organisational structure establishes the framework to accomplish theorganisation’s mission. Responsibilities and authorities for accomplishing assigned tasks have to beclearly defined and communicated in organisations. Various administrative elements have to beestablished to ensure that management's expectations are clearly communicated and understood andeffectively implemented. Administrative elements include formally established goals and objectives.Administrative elements also include policies, procedures and methods to conduct plant activities andmaintain nuclear safety. Policies and procedures have to be clearly written, technically correct andreadily available so plant personnel can easily determine and properly implement actions requiredunder varying circumstances. Management monitoring and assessment activities are an integral part ofthe administrative elements to identify strength and week performance area.3. Safety Culture and Human Performance – Perspective for the futureThe recent events like Davis-Besse Nuclear Power Station reactor vessel head and the severe damageto fuel external to the reactor at Paks Nuclear Power Plant have highlighted problems, which happenedwhen the safety environment does not have sufficient attention. A common thing in these cases is that,that they are mostly result of the plant culture. If the weaknesses are recognised and resolved on time,the events could have been prevented or their severity lessened. The series of decisions and actionsthat resulted in these events can usually be traced to the shared assumptions, values, and beliefs of theorganisation. Organisational culture is the shared basic assumptions that are developed in anorganisation as it learns and copes with problems. The basic assumptions for strong culture have to be3

taught to new employees of the organisation as the correct way to perceive, think, act, and feel.Culture is the sum total of a group’s learning.In addition to a healthy organisational culture, each nuclear power station, because of the specialcharacteristics and unique hazards of the technology – radioactive products, concentration of energy inthe reactor core, and decay heat – needs a strong safety culture.Implied in the definition of the Safety Culture is the notion that nuclear power plants are designed,built, and operated to produce power in a safe, reliable, efficient manner; that the concept of safetyculture applies to every employee in the nuclear organisation, from the board of directors to theindividual contributor; that the focus is on nuclear safety, although the same principles apply toradiological safety, industrial safety, and environmental safety; and that nuclear safety is the first valueadopted at a nuclear station and is never abandoned.Even the nuclear plants are designed to perform safely under a full range of scenarios, eventshappened. Where is the problem? Corporate cultures must embrace the reality of human fallibilitybecause even the best people make mistakes. The relieving factor is the fact that error-likely situationsare predictable, manageable, and, preventable. Organizational processes and values influenceindividual behaviour. People achieve high levels of performance based largely on the encouragementand reinforcement received from leaders and subordinates. Events can be avoided by an understandingof the reasons mistakes occur and application of the lessons learned from past events.The following statistic makes more clear answer why human performance:• 100,000 people die every year due to medical errors• 17,000 people are seriously injured or die every year due to driver errors• 21 out of 26 fuel damaging events due to human error• 75 % of all reportable nuclear power plant events are due to human error• 70 % of causes are related to weaknesses in the organizationBut it has to be clear that human error is not always a sign of poor safety culture.Could we find if Safety culture is present or it is not present? Safety culture is present, when peoplefeel good, when a well done work can be found, when the workers are satisfied, when opencommunication exists, etc. There are different indicators to “measure” safety culture and allowongoing assessment of it. Each facility has to address their own indicators, depends on their culture,people, organisation. Some of them are: Compliance with rules and licensing requirements;Challenges to operation; Radiation protection programme effectiveness; Number of human factorrelated events; Number of self-assessments.And what is perspective for future, what means improvement in human performance, what are nextsteps: emotional intelligence? What is emotional intelligence: “The ability to monitor one's own andothers' emotions, to discriminate among them, and to use the information to guide one's thinking andactions” and why emotional intelligence: “Business has become, in this last half-century, the mostpowerful institution on the planet -- the dominant institution in any society needs to take responsibilityfor the whole and emotional intelligence is emerging as a critical factor in high performance at work,school, and at home”Everyone has emotional intelligence. Research shows: can be taught or increased or can changeleadership styles. We can increase emotional intelligence by:• Practicing being more aware – Listening and Observing• Being more conscious of our choices• Deliberately blending our thinking plus feelings to generate better decisionsWhat is important emotional intelligence does not mean “being nice”, it equals “Consciously andcarefully processing and using emotional information and emotional energy”.4

If we value emotions as a source of information and energy, we will begin to get more positive resultsin our relationship with others and ourselves.4. ConclusionCommercial nuclear electric generating plants are designed, built, and operated to produce electricity.Safety, production, and cost control are necessary goals for the operation of such a plant. Theseoutcomes are quite complementary, and most plants today achieve high levels of safety, impressiveproduction records, and competitive costs, reinforced by decisions and actions made with a long-termview. This perspective keeps safety as the overriding priority for each plant and for each individualassociated with it.Nuclear safety is a collective responsibility. No one in the organisation is exempt from the obligationto ensure safety first. The following principles are important:• Everyone is personally responsible for nuclear safety.• Leaders demonstrate commitment to safety.• Trust permeates the organisation.• Decision-making reflects safety first.• Nuclear technology is recognised as special and unique.• A questioning attitude is cultivated.• Organisational learning is embraced.• Nuclear safety undergoes constant examination.The following should be sufficient:• Communication on all levels• Safety culture• Procedures and documentation• Quality assurance and motivation for it.5. References[1] IAEA TECDOC-1329, "Safety culture in nuclear installations"[2] IAEA Safety reports series No.11, "Developing safety culture in nuclear activities (practicalsuggestions)"[3 IAEA INSAG-15, "Key practical issues in strengthening Safety Culture"[4] IAEA INSAG-13, "Management of operational safety in Nuclear Power Plants"[5] IAEA Safety reports series No.1, "Examples of safety culture practices"[6] IAEA TECDOC-1479, "Human performance improvement in organizations: Potentialapplication for the nuclear industry"[7] WANO GL 2006-02, "Principles for a Strong Nuclear Safety Culture"[8] WANO GL 2001-01, "Guidelines for the Organisation and Administration of Nuclear PowerPlants"[9] WANO GL 2002-02, "Principles for Excellence in Human Performance"[10] WANO GL 2006-03, “Guidelines for Effective Nuclear Supervisor Performance”[11] INPO SER 3-05, “Weaknesses in Operator Fundamentals”[12] Kim Vicente, “The Human Factor: Revolutionizing the Way People Live with Technology”5

Human Performance to Avoid Failures and to Improve ReliabilityGünter Bäro, Thomas H. Dent, Dirk Ebert, Bernd WilkesWestinghouse Electric Germany GmbHDudenstrasse 44, 68167 MannheimReasons and GoalsHuman Performance is a program that has been recommended to the US operators of nuclear powerplants by INPO the Institute of Nuclear Power Operators. The reason is to increase the reliability andhence the availability of a plant. Having this in mind, the overall goal is to prevent errors and to avoid failures.If we reduce the failure rate, we will increase the reliability and therefore the availability of a plant 1 . Inprinciple, we know such a phenomenon from accident pyramids. These pyramids show us that thereseems to be a unique relationship for events of different significances. We are going to encounter 1 seriousevent per 10 significant events, 30 near misses and 600 minor incidents. The numbers may changeper organization, but the principle holds. If we have less minor incidents, we will reduce the probability ofincreased risks for larger problems, accidents or catastrophies, automatically.What can be done to reduce the number of failures and the failure rates. Within the Human PerformanProgram of INPO, there are several methods and tools, to accomplish this goal. In Europe, there has beensome reluctance to employ the Human Performance Program, despite the fact that WANO the World Associationof Nuclear Operators supported the utilization of the tools and methods recommended by INPO.One of the essentials of this program is a standardized process with structured procedures, structuredbriefings and debriefings as well as a structured communication to strive for an exclusion of misunderstandings.Westinghouse noted that a large portion of the problems which are encountered during the execution of aproject seems to be due to communication problems. Such problems are enhanced if you work in an internationalenvironment, in which people have to cooperate who do not have a common mother tongueand a common cultural background. Potentially country-specific attitudes and methods to accomplish acommunication have to be unified and structured such that misunderstandings are nearly excludable inour international work environments. In addition, a standardized and structured communication helps toavoid emotional problems between people who are working in such groups.This paper presents the program that Westinghouse employed to improve the performance of her employeesand how this program is being implemented and realized in Germany and in Europe.1 Human Performance Improvement – Overview Training, W. Earl Carnes, 2006 DOE Price-Anderson CoordinatorTraining April 4-6, 2006, http://www.eh.doe.gov/enforce/2006presentations/day3/Carnes(5).pdf

Human Performance – Basic ConsiderationsThe primary goal of the INPO program of Human Performance is to prevent errors which could jeopardizesafety respectively the execution of a task. During the execution of a project, each deviation or even eachirritation indicates that there is a deviation of the originally qualified parameters. A correction of such anirritation or deviation requires an additional effort and enhances the risk of a failure, because of the extensionto unknown parameters which have not been qualified, so far. This by itself may create problemswhich have not been considered and which could create unexpected disturbances at other sites respectivelyprocess steps. Such properties and such a performance are known from manufacturing materials.Changing the manufacturing parameters changes the material properties. In case of highly automatedsystems, this effect is known, too. Slight disturbances may be indicators or even causes of unexpectedand sometimes even dramatic results. Those who work with complex databases know that the solution ofa problem in one part of the program may create a new problem at another unexpected part of the program.Fig. 1: INPO Information about Human PerformanceImprovement - Overview Training by W. Earl Carnes, 2006DOE Price-Anderson Coordinator Training, April 4-6, 2006Fig. 2: INPO Information about Human PerformanceImprovement - Overview Training by W. Earl Carnes, 2006DOE Price-Anderson Coordinator Training, April 4-6, 2006INPO showed that the interaction between man, machine and organization needs to be understood, toimprove the performance of complex, profoundly jeopardized systems. Human errors account for about80% of the occurrences, and 70 % of these human errors are due to latent organizational weaknesses. Inother words, there are only 20% of occurrences or events due to equipment or material failures. We eliminatetechnically risky products before they fail unexpectedly utilizing regular, preventive and correctivemaintenance, surveillances as well as inspection and audit results. A safe and reliable, disturbance freeoperation increases the capacity factor and the availability. This is what drives us to go for methods andtools which reduce human errors and failures significantly.Event-free work and reliability are a product that is achieved by the own performance at work and theachieved results. If we are going to make mistakes or failures en route, we have to encounter additionalwork time and furthermore irritation and annoyance of the customer. This customer irritation and annoyanceby itself is creating extra time efforts and, hence, extra costs.On top of all of this, there are customer expectations to do specific tasks and provide deliveries, whichhave neither been specified nor agreed upon, explicitly. Such a delivery could be file records of a documentation,a specific warning signal of a switch or button, or the explicitly defined extended limits of servicedeliveries, which is interpreted differently by the parties involved. An early common understanding is

needed, to increase the rate of being without errors and to get such unexpressed expectations under control.The process to avoid errors and failures has the very first priority and the complete management has tofulfill the assigned role within this process. This means that the management has to ensure that all of theregulatory rules, the Westinghouse policies and procedures, as well as the product specific processes andguidelines are to be met, followed up and if necessary trained repeatedly.Avoiding Failures and Errors – Definitions and MethodsAn error is an action that unintentionally departs from an expected behavior. A violation is a deliberate, intentionalact to evade a known policy or procedure requirement for personal advantage usually adoptedfor fun, comfort, expedience, or convenience.Up to now we had the following view of a human error: A human error is the cause of accidents. To explaina failure you must seek failure. You must find people’s inaccurate assessments, wrong decisions,and bad judgments. However, human errors are often symptoms of trouble deeper inside a system.Therefore, the new view is to explain the failure, and not to try to find where people went wrong. We aretrying to find how people’s assessments and actions made sense at the time, given the circumstances thatsurrounded them. It is more important to find the reasons that caused the error than the people.Human Performance describes a behavior to achieve a work according to well described processes withoutrunning into errors or failures. If we understand and accept the situations for actions and decisions, weare enabled to develop tools that help us to avoid errors and failures. It is obvious that we have to agreeon standard behavior patterns to avoid failures and errors due to communication problems within an internationalwork group with people of different mother tongues and business / work cultures.Process Based Leadership is an indispensable process to assure that the procedural adherence has thefirst priority. Managers have to reinforce the use of methods and tools employed within this process.Tools to Optimize the Human Performance of Westinghouse• Avoid errors by applyingo Procedure use, adherence, placekeeping, i.e. use standardized processes and check listso 3-way-communicationo time out and collaborate in case of doubtso create situational awareness by monitoring tasks to identify undesirable situations• reinforce exchanges of ideas and what is done byo verbalizing and challenging - share own ideas with employees/colleagues when actingo briefings and debriefings• reveal and eliminate error causing organizational weaknesses byo self and peer checkingo persistent, repeated observation of working procedures & employeeso persistent, repeated registration of error eventso active participation in the Root Cause Analysis (RCA)• to support continuous enhancement in a learning environmento to coach and support efficient and error preventive practiceso to communicate error solutions and behaviorso to provide resources and tools

It is the task of the management to get involved with the work which is performed by their employees. Thisis essential for a manager or a lead, in order to coach and to identify themselves with the work done.Therefore, the application of these tools is to be watched, to assure the persistent implementation and usageof them. For that purpose, the management will reveal their observations in a database without violatingthe employee’s right of privacy applying a top down process. So called Learning Clocks and HumanPerformance Event Investigations shall demonstrate how Westinghouse applies and uses the same proceduresworldwide, in order to reveal and to remedy defects. In each European country an organizationalentity is established serving the site management to take care and to observe the implementation of theprocedures. For the European locations, a coordinating entity assures that in all sites the same tools andmethods are applied as in the US and that local management persistently reinforces the implementationand usage of the measures/actions. When implementing the process barriers and obstacles because ofregional cultural and linguistic phenomena are to be identified and overcome by the local groups andshould not be disregarded. A significant amount of training is given to everybody, and especially to leadsand managers. By getting involved the management and the leads will support the application, observeand coach their personnel and reinforce the use of methods and tools.Barriers in GermanyThe mentioned tools are well known since years. But we have difficulties to apply them. We face culturaland mental barriers/difficulties, which might be different in each country. It is very important that the managersincluding the directors are getting involved and clarify that we are not looking for the guilty ones butfor the latent system and organizational errors. They have to commit themselves to follow up the measures,to record observations and events and to enhance the application of the tools. Typical Germanbarriers have been:• Observations are not done to find something to be blamed for• Differences in education and culture• Differences in management systems• Linguistic incorrectness in the translations• Missing error culture (acticle in the FAZ dtd. 04-06-06)Only setting good examples and the persistent usage of the mentioned tools can eliminate these barriersand change the mindset. Managers must set a good example in order to overcome those barriers as wellas to assure the procedure.SummarizationEarly error perception and prevention has a top priority since several years. The procedure to enhance thework performance safety and reliability, e.g. in order to avoid errors, is being introduced as a top downprocedure within Westinghouse. A prerequisite for the persistent implementation is the management’s attitudethat must do their utmost to realize and support the procedure. Within Westinghouse the worldwideimplementation of identical tools and methods is assuring a harmonized procedure. The management iscommitted to follow up and to assure the application and usage of the predefined tools and methods. Theyrecord any respective observation, while events and errors and related corrective actions are published.

MONITORING THE EFFECTIVENESS OF MAINTENANCEPROGRAMMES THROUGH THE USE OF PERFORMANCEINDICATORSP. VAISNYS, M. BIETH, P CONTRIInstitute for Energy, Joint Research Center, European CommissionWesterduinweg 3, NL-1755 LE Petten, NiederlandsABSTRACTOptimization of the maintenance strategy, enhancement of the maintenance efficiency andmonitoring the performance are becoming the key attributes to ensure the survival ofnuclear utilities in the energy market. To monitor the maintenance performance in aneffective and objective way, the relevant measurable indicators were selected and amaintenance performance monitoring framework was proposed. Three attributes associatedwith the excellence of the maintenance program are proposed at the top of this structure.Using the attributes as a starting point, eight key performance indicators were proposed tocover the key aspects of maintenance. Finally each key performance indicator is supportedby a set of specific indicators representing the measurable metrics of the maintenanceprogram.1. IntroductionEconomic deregulation of electricity markets in many countries has placed nuclear power plants in anew competitive environment where capital, operating and maintenance costs must be minimized.Optimization of the maintenance strategy, enhancement of the maintenance efficiency and monitoringthe performance are becoming the key attributes to ensure the survival of nuclear utilities in the energymarket. The maintenance performance monitoring was one of the research tasks of the SENUF (Safetyof European Nuclear Facilities) network established in 2003 to facilitate the harmonization of safetycultures between the Candidate Countries and the European Union. After 4 years of successfuloperations, the SENUF network was integrated into the new Direct Action of the EuropeanCommission, SONIS (Safe Operation of Nuclear Installations), where research on maintenancemonitoring and optimisation plays a major role. The objective of this paper is to present the results ofresearch activities carried out by the institute of Energy in the area of the use of numerical indicatorsin the monitoring of maintenance performance.To monitor the maintenance performance in an effective and objective way, the relevant measurableperformance indicators should be used. However, experience has shown that focusing on any singleaspect of performance is ineffective and can be misleading. A range of specific leading and laggingindicators should be considered in order to provide a general sense of the overall performance of amaintenance programme and its trend over time. The best performance measurement systems containa mix of lagging and leading indicators. When dealing with the maintenance performance monitoringwe apply the business process approach to the maintenance function. This concept of processmanagement is based on the assumption that the process itself produces the desired results andtherefore the process has to be managed and measured. This approach ensures successful managementof the maintenance process in order to achieve optimal levels of equipment reliability, availability andcost effectiveness. For the maintenance application, the leading indicators measure the effectiveness ofthe maintenance process, while lagging indicators measure results. The necessity for tracking themaintenance performance indicators other than just equipment reliability and availability is to pinpointareas responsible for negative trends (leading indicators). In addition the performance indicatorsshould not be considered just a measure/demonstration of success but should be used as a tool tomanage successfully. The utilities should utilize performance indicators to identify opportunities forimprovement rather than measures of success or failure.

2. Maintenance performance indicators (MPI) systemAs a first step in the development of the maintenance performance monitoring framework, weconsider the definition of the maintenance monitoring concept. It is assumed that the maintenancemonitoring system is established at the power plant with the aim to achieve the maintenanceexcellence, by removing the existing or potential deficiencies. The proposed approach to monitoringof maintenance performance is presented in Fig. 1. On the top of the maintenance performancehierarchical structure we propose the Maintenance Excellence, from which we develop three attributesthat are associated with the excellence of the maintenance programme:• Preventive character of maintenance (including predictive maintenance);• Maintenance management;• Maintenance budget.Using the attributes as a starting point, a set of maintenance performance indicators is proposed. Thekey performance indicators are envisioned to provide overall evaluation of relevant aspects ofmaintenance performance. Below each attribute, key performance indicators are established. Each keyperformance indicator is supported by a set of specific indicators, some of which are already in use inthe industry. Specific or plant specific indicators represent quantifiable measures of performance.Specific indicators are chosen for their ability to identify declining performance trends or problemareas quickly, so that after proper investigation the management could take corrective actions toprevent further maintenance performance degradation.What is requiredfrom maintenanceprogramme to achieve themaintenance excellenceMAINTENANCEATTRIBUTESParameters that represent thestrategical aspects ofmaintenanceKEYPERFORMANCEINDICATORSMeasurable parametersSPECIFICINDICATORSFig 1. An approach to the monitoring of maintenance performance3. Key performance indicators3.1 Preventive maintenanceFor the preventive attributes of maintenance the following three key indicators are proposed:• System and equipment availability;• Reliability of the systems and components;• Effectiveness of preventive maintenance.The performance indicators structure for preventive maintenance is shown in Fig. 2.

PREVENTIVE MAINTENANCE(including predictive measures)System andequipmentavailabilityComponent andsystem availabilityNumber of forcedpower reductions oroutages because ofmaintenance causesTotal downtimeScheduled downtimeUnscheduleddowntimeMean time betweenmaintenance (MTBM)Reliability ofsystems andcomponentsNumber of correctivework orders issuedNumber of failures insafety related systemsMean time betweenfailures (MTBF)Mean time torepair (MTTR)PM effectivenessPreventive maintenance complianceRatio of corrective work resultedfrom PM activitiesPM work order backlogtrendPercentage of deficiencies discoveredby surveillance, testing & inspectionsRatio of PM activities to allmaintenance activitiesOverdue of PMactivitiesFig 2.Performance indicators for the preventive maintenancePreventive maintenance programs are established at the majority of nuclear facilities to maintainequipment within design operating conditions and/or to extend equipment life. In conjunction to thepredictive maintenance measures, preventive maintenance helps to correct many potential problemsbefore they occur. Preventive maintenance allows equipment to be repaired at times that do notinterfere with production schedules, thereby removing one of the largest factors from downtime cost,increasing profitability. Preventive maintenance activities can be planned in advance facilitating thecontrol of the backlog at the reasonable level.3.2 Maintenance managementA comprehensive work planning and control system applying the defense in depth principle should beimplemented so that maintenance activities can be properly authorized, scheduled and carried out byeither plant personnel or contractors, in accordance with appropriate procedures, and can be completedin a timely manner. The maintenance management system should ensure the allocation on and off thesite of the resources necessary to efficiently accomplish the maintenance activities. Effectivecoordination should be established among different maintenance groups and among the differentdepartments of the plant. To reflect the maintenance management aspects we propose the followingkey indicators:• Planning and scheduling;• Interface with operations;• Work control;• Material management.The structure of the performance indicators to reflect the maintenance management is presented inFig. 3.

MAINTENANCEMANAGEMENTPlanning &schedulingInterface withoperationsWork controlMaterial managementRatio of unplannedto planned workingordersPlanningcomplianceSchedulecomplianceRatio of correctivework orders executedto work ordersNumber ofoutstandingbacklogsWorkaroundsTemporarymodificationsRatio of downtime toallowed outage timeNumber of MCRinstruments outof serviceDuration of repairRepair time of componentssubject to the TechnicalSpecificationsWrench timeCrew efficiencyAmount of maintenancereworkSupervisor to CraftWorker RatioResponse time to callStoresservice levelNo of workspending for sparepartsStocksinventory turnsStocked MROInventory Value as aPercent ofReplacement AssetValue (RAV)planner to craftwork ratioOvertimemaintenance hours3.3 Maintenance budgetFig 3. Indicators structure for the maintenance managementThe objective of the plant management of nuclear generating utility is to maximize production ofelectricity at the lowest cost, the highest quality and within the established safety standards. Themaintenance budget is an increasingly important aspect in the new economical environment in theenergy market. Reducing the production costs, including the maintenance costs in particular is thecondition of survival in the competitive energy market. Figure 4 shows the proposed indicatorsstructure for the maintenance budgeting.MAINTENANCE BUDGETCost effective maintenanceMaintenance cost perkwh producedOvertime maintenance costUnplanned costs as percentage of totalmaintenance costsWork orders complete withinthe determined costs (20%)Ratio of replacement asset value(RAV) to craft/wage head countAnnual maintenance cost as a percent ofreplacement asset value (RAV)Fig4.Indicators for the maintenance budget

However we realize that because the main emphasis in our approach was put on the safe operation of apower plant, the economic effectiveness of the maintenance program, despite its unquestionableimportance, was not developed in such a depth that is sufficient to represent all the budgetary aspectsof a cost-effective maintenance.4. ConclusionsThe proposed system for the maintenance performance indicators is the initial step in the developmentof the framework for the monitoring of the maintenance efficiency using measurable performanceindicators. As a further step, the pilot studies should be initiated in order to validate the applicability,usefulness and viability of the approach for implementation of proposed system of maintenanceindicators at nuclear power plants. Not all indicators proposed in this paper will be found meaningfulat the specific power plant. The establishing of clear and simple definition for each selected indicatoris a key part of the programme implementation. When selecting the indicators for the validationstudies at certain power plant it is recommended to review each indicator and to modify to plantspecific definition if necessary. The elaboration of the best definition for the selected indicator is avery challenging task as it provides the evidence on how meaningful is that indicator for the powerplant. More information on the proposed maintenance monitoring framework including the definitionsof the selected performance indicators can be found in the EU Summary Report EUR 22602 [1]. TheReport also provides recommendations to nuclear power plants for the implementation of the proposedmaintenance performance monitoring system.5. References[1] P.Vaisnys, P.Contri, C.Rieg & M. Bieth, Monitoring the effectiveness of maintenance programsthrough the use of performance indicators, DG JRC-Institute of Energy Summary Report EUR 22602,2006.[2] R. Smith, Key performance Indicators, Leading or Lagging and when to use them, (2003)www.reliabilityweb.com.[3] A. Mc Neeney, Selecting the Right Key Performance Indicators, Meridium, 2005.[4] R.Smith, Key performance Indicators - Leading or Lagging and when to use them, (2003)www.reliabilityweb.com.[5] IAEA, Operational safety performance indicators for nuclear power plants, IAEA-TECDOC-1141,IAEA, Vienna, 2000.[6] WANO, 2005 Performance indicators, Coordinating Centre, London, UK, 2006.[7] IAEA, Maintenance, Surveillance and In-Service Inspection in Nuclear Power Plants, SafetyStandards Series No.NS-G-2.6, IAEA, Vienna, (2002).[8] The Society for Maintenance & Reliability Professionals (SMRP), Best Practice Metrics,www.smrp.org.[9] IAEA, Integrated approach to optimize operation and maintenance costs, IAEA-TECDOC 1509,Vienna, 2006.[10] IAEA, Economic Performance Indicators for Nuclear Power plants, Technical Report Series No.437, Vienna, 2006.[11] T. Ahren, A study of maintenance Performance indicators for the Swedish Railroad System,Licentiate thesis, Lulea university of Technology, Sweden, 2005.[12] L. Swanson, Linking maintenance strategies to performance, International Journal of ProductionEconomics, 70, 237-244, 2001.[13] E.Lehtinen, B.Wahlstrom, A.Piirto, Management of Safety through performance indicators foroperational maintenance, Proceedings of IAEA Specialist meeting on Methodology for Nuclear powerplant performance and Statistical Analysis, Vienna, 1996.

Session 19.3.2Waste

RADIO ACTIVE WASTE PLANTS – BACK TO THE FUTUREJ. E. EARP, ASSOCIATE DIRECTOR STRATEGY, D.R.THOMPSON,SENIOR PRINCIPAL ENGINEERAker Kvaerner Engineering ServicesRichardson Road, Stockton-on-Tees, TS18 3RE, United KingdomAbstractThis paper outlines the potential to provide smaller, potentially modular radwaste plants, suitable for new reactor proposals,within increasingly strict environmental control regimes, and with reduced discharge authorisations. Aker Kvaerner have beeninvolved in the design, build and commissioning of radwaste plants over many years and provided the plant for Sizewell B, oneof the last major PWR’s to be built anywhere in the world. Unit operations and design characteristics of radioactive wasteprocessing plant are discussed. It is concluded that these have changed little in the past 30 years. The traditional buildcharacteristics and metrics of large radioactive waste processing facilities are described in both the reprocessing and the powergeneration industries. New reactor and fuel characteristics are described and used to highlight areas of potential designimprovement, reducing the size, complexity, construction programme and cost of future power reactor radwaste facilities.1. IntroductionThe historical process design of large waste facilities has changed little in the past 30 years, theprincipal unit operations are given in Figure 1:SOLID LIQUID GASIncinerate Ion Exchange FiltrationCompact Precipitation / Filtration AdsorptionEncapsulateEvaporationSome Examples of these processes are given below:2. Conventional Design PracticeTable 1. Unit OperationsFigure 1. SIXEP Universal VesselAt Sellafield, the Site Ion ExchangeEffluent Plant (SIXEP) is an earlygeneration of Radwaste plant for thetreatment of site magnox sourcedliquid effluents, designed in the late1970’s, and based on the use offiltration and ion exchange. Theplant is monolithic, some 100m x50m x 30m and the processingcapacity some 4000m3/day. 70,000te of concrete were used inconstruction.Figure 1 shows a view of aUniversal Vessel designed for theSIXEP facility at Sellafield. Thiscommon vessel was designed foroperation with both media filtrationand ion exchange.

1.Figure 2 shows a view of the Enhanced Actinide Removal Plant (EARP) processing plant. Thislarge (65m x 45m x 35m) effluent treatment facility at Sellafield is still in service, and wasconstructed to meet waste requirements for medium and low active waste streams. The processbasis for this facility was advanced chemical adsorption, and ultrafiltration, engineered to meetexacting operational and performance standards. High DF’s were achieved for Actinides, Cs, Srand Co. This plant is also monolithic, and used some 60,000 te of concrete, with over 30 km ofpipework and 50 km of cabling.Sizewell B Radwaste Plant is a fully integrated facility, designed in the late 1980’s, and of similarsize to the previous examples. In contrast, this plant uses the whole range of available processtechnologies for the treatment of gaseous, liquid and solid wastes from the operation of theSizewell B PWR. This facility has nearly 400 cells, and processes almost all the active wastes onthe site.3. Process PerformanceThese plants have had a significant effect on operational performance. Figure 3 shows thereduction in historical Discharges from Sellafield in relation to SIXEP and EARP.

Annual discharge Pu 239/240 - SellafieldDischarge, TBq/yr605040302010EARPSIXEP01965 1970 1975 1980 1985 1990 1995 2000 2005 2010YearSource: EnvironmentAgencyFigure 3. Historical Discharges at Sellafield4. CommonalitiesThese examples serve to illustrate the similarities:- All use the basic unit operations to provide very effective operation- All these facilities required large buildings of bulk poured concrete, with Monolithicconstruction- All contained many interacting and inter-related systems, requiring a long and complexdesign and build.5. Impact of New TechnologiesWe now have new or intensified technologies in a number of areas such as ion exchange andfiltration. These have allowed performance improvements on a scale we could not conceive just afew years ago.6. UK Energy Supply IssuesIn the face of global energy demand, global warming and security of supply economics, thedevelopment of safe, economic advanced reactor designs is now of great importance.In the historical context above, conventional radwaste plant may not at first sight meet improvedsafety, cost and construction objectives.Accordingly, this paper describes how improved safety, cost and performance objectives can beachieved in radwaste processing for advanced reactors..7. Advanced Reactor DesignFor illustration, the AP1000 design basis has been used to assess the design of a future radwastefacility:Figure 4 shows the design metrics for the AP1000 compared with previous generation PWRtechnology: These achievements will deliver improvements in safety, operability, shortenedconstruction programme and reduced cost..

50% FewerValves35% FewerPumps80% LessPipe*45% LessSeismicBuildingVolumeFigure 4. AP1000 Metrics. Source: Westinghouse85% LessCableIn applying these metrics to a futureradwaste plant design, we also notethe following:a) Fabricated PWR FuelQuality is now higher thanever; the operational designbasis for Sizewellexpressed as %Failed Fuelcan now be exceeded by upto x100. This producesreductions in fissionproduct activity.b) Materials development and improved water chemistry have lead to much reducedcorrosion product formation.c) These in turn lead to much reduced primary and secondary coolant activities, giving:- Reduced shielding requirement in normal operation.- Reduced environmental impact, lower activity discharge.d) The use of conventional effluent treatment processes as described earlier is nowsignificantly enhanced by the availability of new materials in Ion Exchange andmembrane technology.8.0 Summary: Implications for DesignTaking these factors into account, we can now expect that compared to Sizewell B radwasteplant, it is possible to match the AP1000 metrics given in Figure 4 above.This would deliver a building and plant volume of the order of one third to one half of that forSizewell. It may be possible to make larger reductions in building volume although this isprimarily due to the removal of evaporation and relocation of gaseous waste processing.Figure 5 shows a view of one process area of a simplified future radwaste plant, designedwith these principles:Figure 5. Cutaway view of New Radwaste FacilityThe expected improvements are:- Reduced complexity and interactions significantly improve operational safety- Process intensification and close alignment of processes against reactor waste streams deliverreduced plant size.

- Accelerated construction; this becomes possible because of reduced building size, and improved,fit-for-purpose building design.- Programme: The use of skid mounted package plant, modular ‘pack’ systems for filters and ionexchangers, can shorten construction time.- Allowance for sub contractor support service, by provision of Docking Bays and subcontractorLaydown areas, allows the future use of specialist waste services.- Environmental; Reduced operational source terms gives potential for significantly reduced liquidand gaseous effluent levels.- Solid Waste; improved ion exchange processes deliver equivalent performance with reducedwaste volumes.- New Radwaste plants need not be large monolithic structures, and can make more use of modulardesign and skid mounted systems.- As a whole, these allow optimal design for the whole life cycle.- Although more intensive, the processes have still not changed from those of the past. Hence,Back to the Future.References:1. M. Howden , Progress in the reduction of radioactive discharges from Sellafield, BNFL

A 3-DIMENSIONAL HOMOGENIZED MODEL OFCOUPLED THERMO-HYDRO-MECHANICS FORNUCLEAR WASTE DISPOSAL IN GEOLOGIC MEDIAI. CAÑAMÓN VALERADepartamento de Ingeniería Civil: Infraestructura del Transporte, Universidad Politécnica de MadridE.U.I.T. Obras Públicas, C./ Alfonso XII nº 3 & 5, 28014 Madrid – SpainR. ABABOUInstitut de Mécanique de Fluides de Toulouse, Institut National Polytechnique de ToulouseAllée du Professeur Camille Soula, 31400 Toulouse – FranceF. J. ELORZA TENREIRODepartamento de Matemática Aplicada y Métodos Informáticos, Universidad Politécnica de MadridE. T. S. I. de Minas, C./ Rios Rosas. nº 21, 28003 Madrid – SpainABSTRACTThe present work is devoted to near-field coupled modelling of a hypothetical undergroundhigh level waste disposal in fractured granitic rock (“In-situ” experiment, FEBEX Project),performed at the Grimsel Test Site (GTS, Switzerland). First, a 3-dimensionalreconstruction of the fractured network surrounding the FEBEX gallery is performed byMonte Carlo simulation, taking into account geomorphological data. Then, we estimate theequivalent coefficients of the fractured network by applying a superposition method [1].Finally, a continuum equivalent model for 3D coupled Thermo-Hydro-Mechanics (THM)is developed and transient simulations of the excavation of the FEBEX gallery and theheating experiment are conducted using the Comsol Multiphysics ® software (3D finiteelements). Preliminary comparisons of simulation results with time series data collectedduring the “In-situ” experiment yield encouraging results.1. IntroductionFEBEX I and II is a demonstration and research project [2], which was carried out by an internationalconsortium led by the Spanish agency ENRESA and co-funded by the European Commission andperformed as part of the fifth EURATOM framework program, key action Nuclear Fission (1998-2002). This project aims to simulate the components of the engineering barrier system in accordancewith the ENRESA’s AGP (‘Almacenamiento Geológico Profundo’, deep geological disposal) Granitereference concept. The project includes tests on three scales: an ‘In-situ’ test at full scale in naturalconditions; a ‘Mock-up’ test at almost full scale in controlled conditions; and a series of laboratorytests to complement the information from the two large-scale experiments.The ‘In-situ’ experiment is being performed within a new drift which was excavated in the northernzone of the underground laboratory Grimsel Test Site (GTS), managed by NAGRA in Switzerland [3].An important part of the demonstration experiment lays in the modelling tasks. Being able to modelthe processes occurring in such coupled and complex conditions is a fundamental step to assess thiskind of disposals as a feasible and safe solution for the nuclear waste management.This work aims to model both the near field (fractured rock) and the coupled Thermo-Hydro-Mechanical (THM) processes occurring in the FEBEX ‘In-situ’ experiment at the GTS.

2. Monte Carlo simulation of the 3D fractured networkA reconstruction of the 3D fracture network in a 70 m×200 m×70 m block surrounding the FEBEXgallery was obtained by Monte Carlo simulation, taking into account the geo-morphological datacollected in exploratory boreholes. The following information was available [2, 3]:- Fracture orientation. Four different families of fractures were defined according to bothmorphological (exploratory boreholes measurements) and genetic criteria. Uniform distributionswithin specific angle intervals were used for the dip (maximum slope direction) and the plunge.Figure 1 shows the stereonet of the family classification;- Fracture aperture. Data on fracture aperture are only qualitative. Measurements in the maintunnel of the GTS only distinguish between filled, open and wet fractures (Figure 2). Initialvalues for aperture were assigned to each class and were later fitted to hydraulic measurements;- Fracture density. The fracture density of the 2D map of traces (Figure 3) of the cylindrical drift,taken as (Σ trace length / intersecting plane area), was adjusted. Moreover, an anisotropicdensity was obtained by considering five different density zones in the trace map;- Fracture location. A homogeneous Poisson process was used to define the coordinates of thefracture centres. However, the distribution of centres in the proximities of the drift was locallyadapted to fit the non-uniform fracture density observed in the trace map;…and the following parameters were optimized:- Fracture size. The power law distribution was used. There are three parameters in thisdistribution: R min , R max and the exponential coefficient b. The R max was set to a fixed value of100m. The other two parameters were optimized to fit the geologic data.An optimization procedure based on Simulated Annealing was used to adjust fracture size distributionso as to minimize the discrepancy between synthetic fractured medium and real fractured medium. Theoptimum values of the fracture size distribution are R min =0.1985m, R max =100m, and b=3.3048. Figure4 shows a 3D view of the optimized 3D fracture network with N = 2906474 disc fractures.-1 -0,811 0,822-0,6-0,40,60,40,20-0,20 0,2 0,4 0,6 0,8 12-0,4-0,6-0,8-1413S1+S2K4K2+LS3K1+K3S4Non clas.Fig 1: Families classification of the fracturedata coming from exploratory boreholes.Fracture systemS1S2S3K1K2K3K4LZK(1+2)# of discontinuities0 50 100Open and w et fracturesOpen fracturesFilled fracturesFig 2: Fracture aperture frequency in the GTStunnel (from [3]).Fig 3: FEBEX drift:observed tracemapalong the entire drift(above); and a zoom(left).

Fig 4: 3D view of the simulated fractured medium.3. THM simulation of the FEBEX “In-situ” experimentA continuum equivalent model for 3D coupled THM processes was developed based on [4], including:hydro-mechanical coupling via tensorial Biot equations (non-orthotropic), a Darcian flow in anequivalent porous medium (anisotropic permeability), as well as thermal stresses and heat transport bydiffusion and convection, taking into account the thermal expansivity of water.Equivalent homogenized H-M properties were determined from the simulated fractured medium basedon a linear superposition approximation, which may be applied either to the whole domain, or moregenerally, to a partition into subdomains. The superposition approach leads to a conversion of thediscrete 3D fractured medium into an equivalent continuum, by summing up all the individualcontributions due to each singular fracture [1]. Equivalent coefficients such as hydraulic conductivityand mechanical stiffness were computed. The treatment followed during all the mathematical andphysical developments considers conductivity and stiffness as tensors in the 3D space.The simulation domain is the 70x200x70m 3 block mentioned above, with the Geographic Northoriented towards –X and the origin of coordinates in the block centre. There are three connected drifts:the Main tunnel, the Laboratory tunnel and the FEBEX drift, the last one being centred in the origin ofcoordinates (Figure 5). In the test zone of the FEBEX drift there exist a heating process defined hereby a temperature gradient from 100ºC at r=0m to 35ºC at r=1.14m, being r the radial direction in theFEBEX drift. Comsol Multiphysics® was used to perform the simulations.Fig 5: Domain of the THM simulation and boundary nomenclature.

The problem was simulated for different conditions of the rock (homogeneous isotropic, homogeneousanisotropic or heterogeneous anysotropic), and for three different stages:- Hydro-lithostatic equilibrium of the rock mass: at this stage, there is no drift and a fullysaturated 365m rock mass is assumed to be lying over the upper boundary of the domain. Boththe hydrostatic and the lithostatic loads were imposed gradually for the time-dependent analysis.Relative fluid pressure was computed in all the models.- Drifts excavation simulation: the HM response of the fractured rock is analyzed during theexcavation of the drifts, which was modelled by gradually decreasing both the normal stressesand the fluid pressure in the boundaries of the excavated tunnels. Real hydraulic conditionsexisting in the GTS were used in this case.- Heating experiment simulation: at this stage, the full THM model is used. A 3-year heatingprocess is simulated around the FEBEX test zone (last 17 m of FEBEX drift). Heat load profilewas determined from FEBEX “In-situ” experiment. The test zone was filled with bentonite [2].We only present here some results of the heating experiment simulation (third stage). Figure 6 showsthe final state of the fluid pressure in the 3D domain, as well as the flow lines at z = 0m. Figure 7shows the Von-Misses stress at a cross-section through the FEBEX gallery axis, where the effect ofthermal stresses can be appreciated. Finally, Figure 8 shows a comparison of measured and simulatedtemperatures at different distances in a radial borehole drilled in the test zone.Fig 6: Final state of the fluid pressure in the heating experiment simulation.Fig 7: Final state of the Von-Misses stress (cross-section) in theheating experiment simulation.

Fig 8. Comparison of measured and simulated temperatures in theheating experiment simulation at radial borehole 70AIT-TBF14.4. ConclusionsThis work develops an integrated methodology to model coupled processes occurring in a hypotheticalradioactive waste underground repository in crystalline fractured rock (FEBEX “In-situ” experiment).First, the 3D fractured network was simulated based on geomorphological information, and the locallyheterogeneous fracture density around the FEBEX drift was notably well reproduced. Then, acontinuum thermo-hydro-mechanical model was developed and implemented in the ComsolMultiphysics® numerical package. An up-scaling methodology was used to feed the numerical modelwith homogenized coefficients computed from the simulated fractured network. The advantage of thecontinuum approach is that it can be used for modelling coupled THM processes in the presence ofmany, variously oriented fractures, while a discrete fractures approach would become rapidlyintractable as the number of fractures and their geometrical complexity increases. Fully anisotropic(non-orthotropic) spatially variable homogenized coefficients were computed to feed the model.Three different stages were simulated with this modelling approach. Here, we focused mainly on thesimulation of the FEBEX “In-Situ” heating experiment. The preliminary comparisons of simulationswith time series data collected during the experiment yield encouraging results, and provide a goodstarting point to assess the validity of such models for nuclear waste underground disposals.5. References[1] Ababou R., A. Millard, E. Treille, M. Durin, F. Plas : Continuum Modeling of Coupled Thermo-Hydro-Mechanical Processes in Fractured Rock. Comput. Methods in Water Resour. (CMWR’94Heidelberg), Kluwer Acad. Publishers, A. Peters et al. eds., Vol.1, Chap.6, pp.651-658, 1994.[2] Huertas F. et al. Full Scale Engineered Barriers Experiment for a High-level Radioactive Wastein Crystalline Host Rock (FEBEX Project). Final Report. European Commission. Report EUR19147 EN. 2000.[3] Keusen H.R., Ganguin J., Schuler P. & Buletti M. Grimsel Test Site. Geology. NAGRA, NTB87-14E. 1989.[4] Noorishad J., et al. Coupled thermo-hydro-elasticity phenomena in variably saturated fracturedporous rocks: Formulation and numerical solution. Dev. in Geot. Eng.: Coupled T-H-M Proc. ofFract. Media. (O. Stephansson, L. Jing & C-F. Tsang, eds.) 79: pp. 93-134. Elsevier. 1996.

USE OF INORGANIC HIGHLY SELECTIVE ION EXCHANGEMATERIALS FOR MINIMIZATION OF LIQUID WASTEVOLUMES AT THE LOVIISA NPP IN FINLANDESKO TUSAFortum Nuclear Services LtdPOB 100, 00048 FORTUM, FinlandABSTRACTLoviisa Nuclear Power Plant uses evaporation for treatment of different waste waters. Withhighly selective ion exchanger radioactive cesium is removed from concentrates for storageand disposal. After treatment the processed liquid can be released to the sea.Highly selective inorganic ion exchange materials, CsTreat ® for cesium, SrTreat ® forstrontium, and CoTreat for cobalt and other corrosion products, were developed. Severalindustrial applications have proven their efficiency for different waste types. Typicallydecontamination factors of several thousands can be achieved.Since 1991 only 160 liters of CsTreat ® , has been used to purify about 1100 m 3 ofevaporator concentrates at the Loviisa NPP. This volume reduction has created goodsavings in treatment, conditioning and disposal costs.Selective ion exchangers are typically used in small columns. Spent ion exchange columnscan be disposed of in concrete containers. In the Loviisa NPP 12 pieces of columns aresealed into one concrete container with inner volume of one cubic meter.1. IntroductionFortum's (formerly: Imatran Voima Oy, IVO) Loviisa Nuclear Power Plant has two VVER-typepressurised water reactor units. The commercial operation of Loviisa 1 (LO1) began on May 9, 1977,and that of Loviisa 2 (LO2) on January 5, 1981.Many improvements have been made over the years at the Loviisa NPP. The electrical power of theLoviisa NPP has been increased to a nominal output of 2x488 MW e (net), and the life time has beenextended to 50 years. The load factors of the Loviisa NPP have always been high, e.g. in 2006, theywere 93.3 % for LO1 and 88.6 % for LO2.Many improvements in radioactive waste management have been made, too. The goal has been tooptimize the whole waste management from collection of waste to its final disposal. One veryimportant area was treatment of waste waters and liquid wastes by effective volume reduction.The evaporators for floor drain waters have been modified in 1986, in order to reach a salt content of350 g/liter in concentrates. A Nuclide Removal System (NURES), based on the use of inorganic ionexchange material in columns, has been in operation since 1991 for removal of cesium fromevaporator concentrates. A solidification plant (based on cementation and using concrete containers asfinal disposal packages) has been designed and constructed, and it will be commissioned in 2007. Theunderground final repository for LILW is situated at the site of the Loviisa NPP. It was licensed in1998, and after 1 January 2007 about 7000 drums (each 200 litres) have been disposed of.Treatment of liquid wastes by highly selective ion exchangers separates radioactive elements forstorage and disposal. Even from high salt solutions radioactive target nuclides can be removed. Aftertreatment the processed liquid can be released.

Since 1991, the NURES technology has been exported also to foreign customers. A mobile NURESunithas been constructed and used in two projects, and highly selective materials have been used intens of different projects in various countries. Over 200 test cases have been realized. Best benefit forusers come from high volume reduction of waste and from high decontamination of purified liquid.2. Treatment of evaporator concentrates at the Loviisa NPPEvaporator concentrates are collected into the liquid waste storage, which is situated in a separatebuilding. Annually 50-70 m 3 of evaporator concentrates and 10-15 m 3 of spent resins are collected.Four 300 m 3 stainless steel storage tanks for evaporator concentrates and four similar tanks for spentresins are located in concrete blocks with inside stainless steel linings.The large liquid waste storage capacity made it possible to store the evaporator concentrates for manyyears prior to additional treatment, i.e. cesium removal and release to the sea, or solidification. On 1stJanuary 2007 the accumulated amounts of concentrates were 691.5 m 3 with 1297 GBq.Due to natural decay, the amount of short lived radionuclides decreases considerably during thestorage period. Most corrosion products settle onto the bottom of the tanks. During the 1980s, it wasfound that over 50% of the content of radionuclides in the evaporator concentrates consisted of Cs-137and Cs-134, and over 90% of a total solution activity was of Cs. The second dominating radionuclide,Co-60, was associated with the solid precipitates on the tank bottom. By removing cesium from thetank solution, the purified liquid could be released within licensed release limits, and Co-60 would beleft in a small waste volume on the bottom of the tank.Fortum and the Radiochemistry Laboratory of the Helsinki University developed together a cesiumselective inorganic ion exchange material CsTreat ® , which is stable, granular material (with very hightotal ion exchange capacity and extremely high selectivity for cesium). In the nuclide removal system,CsTreat ® is typically used in granular form in columns. In the column use CsTreat ® is normally used inthe grain size of 0.3-0.85 mm, and it can be used in pH area 1-13 in high salt (up to 400 g/l) liquidsand in pH 2-11.5 in very low salt solutions. Later on two other highly selective ion exchangers,SrTreat ® for strontium, and CoTreat for cobalt and other corrosion products, were developed.3. Nuclide Removal System (NURES)The principle of NURES is simple (Figure 1). The first phase is efficient particle filtering, whichremoves particle bound radioactivity as totally as possible and protects the ion exchange bed fromplugging, and the second phase is ion exchange with selective material to remove ionic radioactivity.The full size NURES, as constructed in the Loviisa NPP, is also shown in Figure 1.Table 1 shows the results for the purification of five tanks of evaporator concentrates at the LoviisaNPP since 1991 (2). After treatment of fourth tank the average processing capacity for about 900 m 3was about 12.2 m 3 /kg. After the fifth tank about 1100 m 3 has been treated, and the average processingcapacity is about 10.4 m 3 /kg. Processing capacity for the fifth tank was much lower than that for theprevious tanks. Probable reason for this was noticed to be higher amount of solids in the waste liquid.In the case of fifth tank there was about 3.4 mg/l of solid material with very small particle size. 0.1 µmparticle filter probably could not remove enough solid particles, which led to plugging of part of ionexchange capacity. Because of this more columns were used for purification of this tank than for othertanks. Efficient filtering before ion exchange is essential for nuclide removal.A final disposal container for spent ion exchange columns has the same outer dimensions (height anddiameter 1.3 m) as the container for solidified waste, having 12 disposal holes in the concrete filling ofthe container (Figure 2). Spent columns are sealed with lead plugs into holes of the concrete container.When all holes are full, a concrete cover is cast on top of the container.

Figure 1. Principle of NURES and its realization at the Loviisa NPP.Figure 2. The final disposal container for spent ion exchange columns.In treatment of about 1100 m 3 of evaporator concentrates in Loviisa NPP the average decontaminationfactors have been well over 1000 [1, 2]. Figure 3 gives a typical performance curve for one 8 litercolumn. DF is relatively low in the beginning of treatment of high salt liquids, but it rather soonreaches good decontamination levels (DF = 10 000 or above). After some time, in the case of Fig. 3after about 40 m 3 , DF starts to decrease, because of exhaustion of the ion exchange capacity. At anylevel the operator can decide, which level he accepts for final shut-down of the column.Treated inVolumetreated,m 3IX massused.litersTotal saltconc., g/lDecontamination factor,DFProcessingcapacity, l/kgVolumereductionfactorVRFTank 1 1991-92 253 24 240 >2000 16 000 2000Tank 2 1993 210 32 176 >2000 10 000 1260max 30 000Tank 3 1995 230 24 228 >1000 15 000 1840Tank 4 2000 202.6 32 220 >1000 9600 1200max >28 500Tank 5 2002 -03 200.7 48 >220 >1000max >16 1006400 800Table 1. Results from purification of five evaporator concentrate tanks at the Loviisa NPP

10000010000DF10001001012,5111525,739,752,366,679Volume Treated (m3)Fig. 3. Decontamination factor (DF) for one eight liter column during its operation time4. Volume reduction of waste with the use of selective ion exchangerHighly selective ion exchangers can be utilized in many different ways. Most efficient way is to usecompact columns with volume of 2-12 liters. Until now, totally 1100 m 3 of evaporator concentrateshave been purified at the Loviisa NPP by CsTreat ® with only 20 pieces of 8-liter columns. Between1991 and 2006, 112.7 GBq of Cs-137 and 2.6 GBq of Cs-134 (calculated on 25th April 2007) wereremoved from liquid to CsTreat ® , and the purified liquid was released to the sea.The volume reduction factor (VRF) of the waste itself, comparing the volume of the original liquid tothe volume of the CsTreat ® columns, is close to 7000.Cementation of 1100 m 3 of concentrates would result in about 2310 m 3 of solid concrete product, i.e.2310 pieces of concrete containers. Present amount of 20 spent columns will not fully occupy eventwo containers. Thus, the volume reduction of packed waste is close to 1400.5. Savings due to selective ion exchangeSelective ion exchange and release of purified liquid gives economic savings due to decreasedcementation needs, due to some reduction in storage costs, due to reduced transportation costs, anddue to reduction in final disposal volume.Remarkable savings have been achieved at the Loviisa NPP, when cementation of about 1100 m 3 ofevaporator concentrates has been avoided.Estimating that about 3-5 m 3 of sludge is left for cementation from annual 50-70 m 3 of concentratecollection, well over 90 % of total cementation, storage and transportation needs are avoided.In final disposal the savings come from reduction of disposal cavern. Low- and intermediate-levelwastes from the Loviisa NPP are disposed of in a repository constructed in the bedrock of the powerplant site. By 1st January 2007, already a total of about 7000 steel drums (each 200 litres) ofmaintenance waste have been disposed of. In the design and construction of a cavern for solidifiedwaste the reduction of cemented waste due to treatment with selective ion exchange was taken intoaccount.The cavern for solidified waste was constructed for 5040 concrete containers. When cementation of1100 m 3 was avoided, about 2310 containers were eliminated from final disposal. From annualconcentrate collection of 50-70 m 3 additionally about 100-140 containers are eliminated annually.Without the use of selective ion exchange materials the length of the cavern for solidified waste,which is now about 84 meters long, would have been at least 50 % longer.

6. Foreign applications of the NURES systemThe NURES technology has been applied successfully also in various foreign applications. A mobileNURES unit was constructed, too, and it was first used in Paldiski in Estonia to purify 760 m 3 of lowactive waste liquid. After that NURES unit was used in Murmansk in Russia to remove mainly cesiumand strontium from radioactive waste waters accumulated from nuclear-powered ice-breakers.Typical floor drain waters were purified for example in the Callaway NPP (PWR), MI, USA, and atthe Olkiluoto NPP (BWR) in Finland. Continuous purification is going on at some NPPs in USA. Poolwaters have been purified for example at the US DOE's Savannah River Site and at Sellafield'sreprocessing plant in UK.Reprocessing waste liquids were purified at Japan Atomic Energy Research Institute's (JAERI) site inTokai-mura in Japan. An interesting application of CsTreat has been at UKAEA's Dounreay site toremove cesium from Prototype Fast Reactor's sodium coolant.The construction of a new liquid waste treatment system, including utilization of CsTreat ® , has beencompleted at the Paks NPP (VVER-440) in Hungary, and the operation is scheduled to start in 2007.The system is quite similar to that used at the Loviisa NPP. Over 2000 m 3 of evaporator concentratesand other liquids will be purified (240 l/h) with this system. A boric acid recovery system has alsobeen installed. 70-90% of boric acid will be recovered from the existing liquid wastes.Additionally, plenty of smaller applications of these highly selective ion exchange materials areoperating for special needs.7. ConclusionsFortum has taken many practices into use to minimize the amount of wastes going to the on-siterepository for final disposal. The whole waste management chain has been optimized from wastecollection to final disposal. High capacities of treatment equipment and existing systems have made itpossible to store and treat liquid wastes in a different way than at many other power plants.Treatment of evaporator concentrates included originally the greatest potential for savings. Selectiveremoval of radionuclides was developed and taken into use. It has proven to be a very efficient way tominimize the volumes of evaporator concentrates and to reduce the volume of final disposal. Since1991, selective removal of radionuclides has purified 1100 m 3 of evaporator concentrates with only160 liters of selective ion exchange material, giving a volume reduction factor of about 1400 forpacked waste. Avoidance of cement solidification of the original waste volume has broughtremarkable savings in the cost of treatment, conditioning, storage and final disposal. In theconstruction of the underground repository, which is already in operation at the Loviisa NPP site,reduction in the volume of evaporator concentrate has been taken into account.Based on Fortum's own experience in the operation of the Loviisa NPP, several practices and systemsare now also used at many nuclear sites around the world.References1. E. Tusa, A. Paavola, R. Harjula and J. Lehto, Ten years' successful operation of nuclideremoval system in Loviisa NPP, Finland, Proceedings of ICEM 01 Conference, Bruges,Belgium, 30.9.-4.10.2001, Paper 40-3.2. E. Tusa, R. Harjula, and P. Yarnell, Fifteen Years of Operation with Inorganic HighlySelective ion Exchange Materials, to be published in Proceedings of Waste Management2007, Tucson, AZ, February 25 - March 1, 2007

Workshop 1:Education & Training and KnowledgeManagement / Cases

PRESERVATION AND MANAGEMENT OF NUCLEARKNOWLEDGE ON WWER REACTOR PRESSURE VESSELSV. SLUGEN, M. MIKLOŠSlovak University of Technology, 81219 Bratislava, SlovakiaL. DEBARBERIS, A. ZEMANEC JRC Institute of Energy, 1755 Petten, NetherlandsABSTRACTActivities connected to the nuclear knowledge preservation are ongoing in the EC-JRCInstitute of Energy with the intention to collect all available information about reactorpressure vessels of WWER type reactors as well as to analyze and summarize the mostimportant items and issues. This activity is in line of the European Community FP6 projectsPERFECT (Prediction of irradiation damage effects on reactor components) and mainlyCOVERS (Coordinated action on WWER safety) in which all WWER operating countriesalso take part. Actually, the electronic database was created and is accessible for young orexpired researchers in this area. The access is recommended via ODIN (Online Data andInformation Network) https://odin.jrc.nl/doma. After registration you can enter the WWERDoMa-db: “Database of references for knowledge management and Preservation onWWER reactor pressure vessel”. For the access to confidential information you have to askan indicated administrator.The nuclear knowledge management is realized not only via database creation or educationprocess during undergraduate (Bc.), graduate (MSc.) and postgraduate (PhD.) study butalso via specialised training courses in a frame of continuous education system, researchactivities and projects, workshops seminars, ect. For illustration of the actual status andpossibilities, the Slovak nuclear knowledge model is used. Unfortunately, decrease ofnumber of employees in nuclear and ”human ageing”of experts seems to be a seriousproblem not only on world but also in Slovakia.1. IntroductionIn the last decade, preservation and optimal nuclear knowledge management are becoming a risingchallenge worldwide. Many papers and experts talks at different conferences stressed attention onstagnating or decreasing expertise connecting to decreased numbers of graduates, professors orresearch workers. [1-3]. Several networks were created in the Europe in frame of the 5 th and 6 thEuratom Framework Programme accented international collaboration in training and educationphysics (EUPEN, STEPS) or in nuclear power engineering (ENEN, NEPTUNO) [4-6].From a huge amount of activities which are in favor nuclear knowledge preservation, we describe ourapproach based on the database collection of all available information about material behavior ofdifferent types of reactors, accenting the WWER ones. We are sure that this knowledge will beessential in the future for proper evaluation of plants life as well as for their lifetime extension. Baseon this knowledge, the development of new materials for Generation IV reactors will be easier.Unfortunately, experts in this field are old (with exception of you, of course) and their age is almostclose to or even within retirement and thus any preservation and use of their knowledge andexperience is becoming more and more difficult or nearly impossible in few years.2. Database creationFor the project, more than twenty precise selected specialists, mainly from WWER operatingcountries, have been asked to help with the collection of such mostly rare publications. It is clear that alarge number of publications, since the very beginning of the research studies and from the first yearsof reactors operation, were published in Russian as well as in other national languages (Czech, Slovak,Hungarian, Finish etc.). This makes the situation complicated for most of foreign experts, and today,

also for most of WWER reactor operators in individual countries. A large number of publications wasprepared also in other languages and were sent not only to international, but also to some nationaljournals as well as presented in many national conferences/workshops etc. proceedings which are nowvery complicated to find from abroad.Authors have been asked via national experts for their full lists of publications dealing with materialproperties of WWER reactor pressure vessels, and as a first step, it is the intention to concentrate onstudies and results related to the irradiation damage and testing for these type of steels as thispractically determines reactor pressure vessel lifetime and it is the reason of most vessel problems. Asa second step, full texts of the relevant publications have to been required when not be possible to findthem in standard libraries. The material and further knowledge has been discussed in the leading groupin JRC-Petten and of course, this database of the bibliographies (supported by full text of all includedpublications with abstracts in English and in electronic format) will be served and will be available toall authors immediately, thus it will also help the authors for current work and studies. Further, thesematerials will serve as a basis for elaboration of a state-of-the-art report on radiation damage inWWER reactor pressure vessels steels that will be prepared with an active participation of all activeauthors. The structure and content of the reports will be prepared in the workshop where theparticipating authors will be invited.3. Summarization of previous projectsIn the field of reactor pressure vessel (RPV) embrittlement, a generational gap is slowly appearingwith regard to in depth knowledge of materials behaviour and related neutron embrittlement issues.This is due mainly to the fact that the experts who took part in the design, construction andcommissioning of the nuclear reactors are now approaching retirement, if not already retired, and/orchanged job for different reasons. In addition, for the Russian design RPVs a significant fragmentationof the knowledge took place with the dissolution of the USSR and the dislocation of the variousWWERs into several Member States; including: Hungary, Slovakia, Check Republic, Bulgaria andUkraine. Significant knowledge on WWER-440 is also available in Finland and of course in Russia. Inthe previous decades, many of national experts were involved into following international projects.SAFELIFE - action is the JRC Action dedicated to issues of PLIM (Plant Lifetime Management) ofageing nuclear power plants [7]. Several networks [8], partnership projects and expert groups areoperated by JRC on the various PLIM disciplines; among others the major European NetworksAMES, NESC, ENIQ, NET, SENUF and AMALIA (see Figure 1). Networks is a key element for theJRC aiming at harmonisation and best practice development amongst the Member States.In particular the European Network AMES is dedicated to the study of radiation embrittlement of thereactor pressure vessels with the strongest connection to the present initiative. Within the frame of theSAFELIFE - action of JRC, and with NRI as a key partner, a new initiative in the area of knowledgemanagement has been planned and launched at the end of 2004.Fig. 1. Major European Networks operated by JRC on PLIM

The initiative of JRC –IE in co-operation with NRI is now concentrating mainly to the very first stepof knowledge management: knowledge preservation. This step is now of particular importance when itis urgent to make available as much as possible the scatter knowledge in the various WWER countries.The issues of languages is also considered, since in the Member States besides Russian in the earlytimes and English more recent, national languages were used too as they are used today. The approachfollowed in those cases is the short translation of the main conclusions and important facts arisingfrom the collected documents.The initiative is part of the JRC-IE SAFELIFE action and it is the first proto-type example of practicaldeployment of effective knowledge management in nuclear safety for the particular case of RPVembrittlement.The general scope of the project can be summarized as follows:- Collect all paper in original language from the WWER countries; in particular papers which neverreached the international circuit. The potential languages that will be encountered are those of theWWER countries: Russia, CZ, Hungary, Ukraine, Bulgaria, Slovakia. More recent papers aremainly in English.- Prepare PDF files of original.- Prepare a short English summary with key data & conclusions.- Organize a documentation data-base.- Design a method to manage such knowledge including: retrieval by keywords, tracing of authors,additional information, etc.The project and the database created in frame of this project contributed also directly to COVERS(Coordinated Action on WWER Safety).4. Development and preliminary resultsA first call with a letter (in English and Russian) of intent has been issued at the beginning of 2005 toapproximately 20 leading prospective experts identified in the various countries. The identificationwas done by direct knowledge, amongst the members of the AMES European Network, the IGRMDcommunity and the IAEA Experts.A series of Workshops was performed with the experts. Assuming that the number of identifiedexperts is a representative sample, their percentage distribution is given in Fig. 2.The response to the call was unanimously enthusiastically received and the first collection roundyielded a good number of papers and documents on the targeted subjects. Besides few papers whichcould have been anyhow found in the open literature, a large number of ‘original’ or ‘rare‘ documentsare now already available.In Fig. 3, the first round’s collected papers, the estimated available numbers and the expected finalnumbers are shown. The collection of originals will continue in the period of tree years and extrastimulation methods will be used including the workshops. From a first screening of the content wecan draw some statistic of the detailed issues appearing from the collection, see Fig. 4.5. Follow-upThe selected approach, as given in Figure 5, foresees the finalization of the first round of collectionfrom the pre-defined list of pre-identified experts. The papers are screened and systematically filed ina dedicated documentation d-base. A series of Workshops will support the analysis of the material andthe identification of further need and requirement for further collections. A few iterations should besufficient to reach a reasonable inventory of information.6. Preservation of nuclear knowledge via education in SlovakiaIn the Central-European region, there exists a very extensive and also effective internationalcollaboration in nuclear industry and education. Similarly good situation is also among universitiesand technical high schools in this area. Actually, the Slovak University of Technology in Bratislavahas established contacts with many universities abroad in the area of utilization of research andtraining reactors. One of good examples of international collaboration is ENEN – European NuclearEducation Network Association which resulted in a formation of “Eugene Wigner Training Courseson Reactor Physics Experiments” running in the last 2 years as a mutual effort of the BudapestUniversity of Technology and Economics (Budapest, Hungary), Czech Technical University (Prague,

Czech Republic), University of Technology (Vienna, Austria), and Slovak University of Technologyin Bratislava (Bratislava, Slovakia). In total 53 participants from different European countries asAustria, Belgium, Bulgaria, Czech Republic, Finland, Italy, Israel, Romania, Slovakia, Slovenia,Sweden and Switzerland took part at these international training courses so far. In the frame of thesecourses, students of nuclear engineering visited three different experimental facilities located at thecourse organisers’ institutes and carried out experimental laboratory practices.815%10%25%10%10%10%5%15%Chech RepublikHungarySlovakiaFinlandBulgariaUkraineRussiaOthers2512101540Cu-P-Ni influenceannealing of WWER-440High Ni in WWER-1000 steelsRe-embrittlementBasic on materialsothersFig. 2. Experts distribution.Fig. 4. Distribution of papers per detailed subject.collected at 1st round espected estimated- AMES & IGRDM- European Networks- IAEA Experts, etc.Identification of leading experts intarget countries1201001 st Call for list ofrare papers & list80collection6040Workshop(s)Review of collectedmaterial200Chech Hungary Slovakia Finland Bulgaria Ukraine Russia OthersRepublikFig. 3 Distribution of papers and documents.Identification of residualmaterial & further sourcesD-basedocumentationDefinition of next CallFig. 5. Methodology.The high level nuclear education is very important also due to permanent increasing of nuclear expertsage. To replace some of them are not easy. Fig.6 shows the number of graduates in the area of nuclearpower engineering and nuclear material science in the last 20 years.2003812002211320011302000102199919977721Jadrová energetika199640Materiálové inžinierstvo1995611rok199419934729199210171991161819909191989119198811171987101019861112198518150 5 10 15 20 25 30 35 40Počet absolventovFig. 6. Number of graduates in the last 20 years.

6. ConclusionsThe JRC-IE effort towards Preservation and Knowledge Management in the specific field of RPVembrittlement is a very important tasks and it is discussed in this paper. The effort is carried out in cooperationwith the Nuclear Research Institute Rez. The challenge to collect all available informationabout the behavior of WWER reactors is tackled with a systematic approach; as required in this phaseof the plants life. In fact, the information available is for a significant extent available with experts inthe specialised field which are close to or even within retirement, so the preservation and use of theirknowledge and experience is becoming more and more difficult or nearly impossible in few years. Thedocumentation d-base has been stored and serves as appropriate to suit the requirement of the keyprojects SAFELIFE, COVERS, PERFECT, etc. Created database was put on the ODIN portal which isprovide by European Commission Joint Research Centre (JRC) Petten to the European energyresearch community. It contains engineering database, document management sites and otherinformation related to European research in the area of nuclear and conventional energy.The results of the project is accessible (via https://odin.jrc.nl/doma) and useable also for youngspecialists of the new generation for PWR or WWER plants in all countries since they would be ableto be acquainted not only with current views and knowledge but also with its history and background,etc. Further, these materials can serve as a basis for elaboration of a state-of-the-art report on radiationdamage in WWER reactor pressure vessels steels that will be prepared with an active participation ofall active authors.University can contribute not only to the education but also to attract students to nuclear field, which isa base also for the safety culture at NPP as well as essential need for accepting nuclear industry by thepublic. Readers at the university (professors, assistants, etc.) can stimulate students for nuclear physicsor at least they can relieve them of distress from nuclear issues. The first contact is very important.University enables an optimal selection of students. The option for "nuclear education" is completelyfree and independent. The problem is that the amount of students taking these lectures is low. Propereducation at the university is a source of knowledge and attitudes for the whole life. Theoretical andpractical experiences, professional approach and consistency are very important also from the safetyculture point of view. University lectures and seminars are basically opened for public and thisacademic field can be made better use of in public relations. It is an investment mainly to younggeneration. During discussions with students, teachers can form their professional orientationaccording to their abilities and needs. Good teacher encourages also the growth of student and shapeshis personality. Graduated students have to learn to take responsibility for their decisions and theiracademic level of education.AcknowledgementFinancial contribution of the grant VEGA 1/3188/06 is acknowledged.7. References[1] OECD/NEA, Nuclear education and training: cause for concern? OECD 2000, ISBN 92-64-18521-6.[2] Fernandez-Ruiz, P., Forsström, H. and Van Goethem, G., The sixth European framework programe 2003-2006: a driving force for the construction of the nuclear European research area, Nuclear Engineering andDesign 235 (2005) pp. 127 - 137[3] Moons, F., Safieh, J., Giot, M., Mavko, B., Sehgal, R.R., Schäfer, A., Van Goethem, G. and D´haeseleer, W.,European Master of Science in Nuclear Engineering, Nuclear Engineering and Design, 235 (2005), pp. 165-172[4] ECTS, http://www.europa.eu.int/comm/education/programmes/socrates/ects.en.html[5] ENEN, http://www.sckcen.be/ENEN[6] NEPTUNO, http://www.sckcen.be/NEPTUNO[7] Debarberis, L., Törrönen, K., Sevnini, F., Gillemot, F., Brumovsky, M. Bieth, M. and Rieg, C., Recentadvances in R&D to support understanding of RPV embrittlement within SAFELIFE action of EC-JRC-IE.The 4th International Conference "Safety assurance of nuclear power plants with WWER", FSUE EDOGidopress, Russia, Moscow 23-25 May 2004[8] Sevini, F., Debarberis, L., Taylor, N., Gerard, R. and Brumovsky, M., Study of Ageing Mechanisms forStructural Materials within SAFELIFE Project, International Journal Strength of Materials, ISSN 0556-171X, No. 1 (367) 2004

PHYSICS AND ENGINEERING OF NUCLEAR REACTORSAT THE ECOLE NATIONALE SUPÉRIEURE DE PHYSIQUEDE GRENOBLE OF THE INSTITUT NATIONALPOLYTECHNIQUE DE GRENOBLEE. MERLE-LUCOTTE, R. BRISSOTENSPG-INPG/LPSC-CNRS53, avenue des Martyrs, F-38026 Grenoble Cedex - FranceABSTRACTIf the use of fossil fuels is to be limited to curtail greenhouse gas emissions, fission nuclearenergy is, along with new renewable energies, one of the primary energy sources able torespond significantly to the increasing worldwide demand. In this context, it is necessary todesign and evaluate new generations of nuclear reactors as defined by the Gen IVInternational Forum. The Energy and Nuclear Engineering (GEN) curriculum of the EcoleNationale Supérieure de Physique de Grenoble (ENSPG), one of the nine engineeringschools of the Grenoble Institute of Technology (INPG), includes a balanced blend of basiccourses in energy, nuclear and thermal hydraulic engineering, together with thecorresponding engineering sciences to cover the technological aspects. The objective is totrain engineers who shall master not only nuclear engineering for the production ofelectricity but, more broadly, energy and nuclear technologies and their various applicationfields.1. IntroductionNuclear reactors currently generate nearly one fourth of the electricity in the world, and nucleartechnologies have spread into many other industrial areas: instrumentation, medicine, the foodprocessingindustry, materials, etc. The worldwide demand for primary energy is increasing; solutionshave to be sought and the level to which these solutions are adapted to the stakes have to be examined.Not many options are available if use of fossil fuels is to be limited to curtail greenhouse gasemissions. Fission nuclear energy is, along with new renewable energies and, in the longer term,fusion energy, one of the primary energy sources able to respond significantly to the demand [1]. Inthis context, it is necessary:• To ensure transfer of competences between generations• To design and evaluate new types of nuclear reactors as defined by the Gen IV InternationalForum [2]Such research and development rests on a solid knowledge in physics, on technical innovations, andon efficient numerical and modeling tools allowing the management of these complex systems. TheEnergy and Nuclear Engineering (GEN) curriculum is one of the six specialty options of EcoleNationale Supérieure de Physique de Grenoble (ENSPG), one of the nine engineering schools of theInstitut National Polytechnique de Grenoble (INPG).In this paper, after a brief presentation of the Institut National Polytechnique de Grenoble or GrenobleInstitute of Technology (INPG), we will introduce the ENSPG. We will then detail the Energy andNuclear Engineering curriculum, the most complete in France for engineers in the nuclear field. Moreprecisely, three curricula are offered by ENSPG in the nuclear field: the Energy and NuclearEngineering curriculum itself, a specialized training module in safety and risk management, and aresearch Masters’ in the Physics of Energetics.

2. Grenoble Institute of Technology (INPG)The Grenoble Institute of Technology (INPG) is one of four universities in Grenoble. The EngineeringSchools of the Grenoble Institute of Technology train engineers in key industrial domains. Studentsare admitted two years after their high school graduation via a competitive entrance exam to theGrandes Ecoles, via University degrees, or an in-house Preparatory Course at Grenoble, Nancy andToulouse Institutes of Technology. They can go on to do a Research Masters’ Program and later aPhD, in one of 10 Masters’ Programs and 8 Doctoral Schools.Students graduate after three years of studies in one of the nine “Grandes Ecoles” or engineeringschools which make up INPG:• Physique fondamentale et appliquée, et génie nucléaire / Fundamental and Applied Physics andNuclear Engineering (ENSPG)• Électronique et technologies de l’information / Electronics and Information Technologies(ENSERG)• Energie et traitement de l’information / Energy and Information Processing (ENSIEG)• Fluides, mécanique et environnement / Fluids, Mechanics and Environment (ENSHMG)• Génie industriel / Industrial Engineering (ENSGI)• Industries papetières et graphiques / Papermaking and Printing Industries (EFPG)• Informatique et mathématiques appliqués / Information Technologies and Applied Mathematics(ENSIMAG)• Matériaux, électrochimie, génie des procédés / Materials, Electrochemistry, Process Engineering(ENSEEG)• Systèmes industriels embarqués / Embedded Industrial Systems (ESISAR)INPG is characterized by the following key figures:• 5,200 students of which 20 % are foreigners• 11 Engineering Degree Courses from which 1,150 engineers graduate every year• 1 Masters’ and Doctorate School College from which 180 PhDs graduate every year• 40,000 alumni working worldwide• 1 Professional Development Department• 38 laboratories among which 3 are of international standard• 1 private subsidiary for industrial valorisation, INPG Enterprise SA• 1,100 teaching and research fellows• 114 million euros budgetSeveral aspects of the organization or operation of INPG may appear exotic to a non-French reader.The engineering schools making up the INPG are actually “Grandes Ecoles”. As for all French“Grandes Ecoles”, where most of France’s top leaders are trained, each school is characterised by arather small size (under 400 students at ENSPG), and a very selective admission procedure. Studentsnormally are admitted to a school after a competitive examination (Concours d’Entrée) which takesplace after two years of intensive university-level studies. It is therefore essential to remember that thethree years of studies at our engineering schools correspond to the third, fourth and fifth year ofuniversity (post-baccalauréat) studies.We will now concentrate on the school of ‘Fundamental and Applied Physics and NuclearEngineering’ or ENSPG.3. Ecole Nationale Supérieure de Physique de Grenoble (ENSPG)ENSPG stands for Ecole Nationale Supérieure de Physique de Grenoble, in other words Grenoble’sEngineering School for Physics. ENSPG trains engineers and research physicists who master thevarious technologies originating from physics, and can make them evolve. It works at developing thestudents’ creativity, and the human qualities they will need as leaders. Over 130 students graduateeach year.

More so than other engineering schools, because of its topics and its environment, ENSPG is alsostrongly involved in graduate studies, leading to the PhD (Doctorat), normally obtained after threeyears of research work, with some courses, following completion of a research masters’ degree.Foreign students are of course welcome at ENSPG. These students can graduate from ENSPG underone of two conditions:• If these students come from one of the universities with which ENSPG has a double-degreeagreement (at the time being: Universität Karlsruhe and Politecnico di Torino), and choose, withtheir supervisor’s agreement, to take the double degree scheme. These students will then obtainboth their home University’s and INPG’s degrees.• By applying for admission as a regular student at either of the two stages where it can be done:for admission into first year (normally after a 2-year or 3-year university curriculum), or foradmission into second year (after a 4-year university curriculum). Admissions are decided on thebasis of the student’s records.3.1 Organisation of the Education at ENSPGThe school’s main objective is to train physics engineers with both a sound basic training in physicsand competence in engineering sciences, economics and social sciences, mastering several foreignlanguages. The common-core syllabus, corresponding to the three first semesters, aims at this generaleducation. More specialised training, corresponding to one of six specialty options, is given in thefollowing three semesters of the three years of studies at ENSPG.3.1.1 The common-core curriculumThe common-core syllabus covers the first three semesters or 1325 hours. They include personal workin the form of projects, and cover three main objectives:• Basic education in the School’s general specialty, the properties of matter and their theoreticalmodels, with mathematics as a tool, during 450 hours (34% of the curriculum). Physics of matter isdealt with through lectures in basic physics (quantum and statistical physics, optics, solid statephysics, semiconductors and thermodynamics, nuclear physics) and in material sciences (magneticproperties, crystallography, and physics of materials). This academic training is completed by anexperimental approach thanks to practical work and a first year group project.• Sound notions of basic engineering sciences (electronics, automatic control, signal processing,mechanical design, computer science, mathematics and numerical methods) covering 550 hours(41%), aimed at facilitating interdisciplinary exchange with specialists of other fields.• Throughout the three years, 25% of the time (around 325 hours) is devoted to languages (Englishand a second foreign language), economics and management, to an introduction to geopolitics, and tosport.3.1.2 Specialty training: the six specialty optionsThe student chooses one of the six specialty options. They cover two semesters of courses, followedby the final project. These curricula provide specialised training in areas chosen on the basis ofexisting or future industrial possibilities, and of Grenoble’s technical and scientific environment. Thesix specialty options (in decreasing number of students) offered by ENSPG are:• “Energy and nuclear engineering” (or ‘GEN’ standing for Génie Energétique et Nucléaire inFrench) covers the various aspects of energetics with a particular accent on nuclear technologies.• “Functional materials and nanophysics” deals with materials with magnetic, superconducting,semiconducting... properties, and their applications, especially at the nanoscopic level.• “Physical instrumentation” centres on the design and the technologies of instrumental devices.• “Instrumentation for biotechnologies” centres on the physical and biological processes involved inthe instrumentation used in life sciences.• “Structural materials” centres on the preparation and modelling of materials with emphasis ontheir mechanical properties.• “Physics of electronic and opto-electronic devices” covers microelectronics, optics,optoelectronics and associated technologies.

In section 4, we will focus on the first option, the “Energy and nuclear engineering” or GENcurriculum.3.1.3 The third year: Special training modules and research mastersThe third, final year at ENSPG (fifth year of university studies) is devoted to developing competencein the specialisation field chosen. It includes considerable flexibility, and can to a large extent be tunedto the student’s plans for the beginning of his or her professional career. Thus, the courses in the thirdyear include, apart from a kernel specific to each specialty option, a large choice of “specialisedcomplements” which allow in-depth study of a discipline, or an opening to other fields. Two mainoptions are offered.1. students intending to complement their training through a PhD will normally follow, inparallel with their final year, a research masters’ (graduate course) as a prerequisite fordoctoral work. The research masters’ associated to the nuclear specialty option and chosen byaround 30% of the nuclear students is the Masters’ in the Physics of Energetics described insection 4.3.2. Students can also choose one of two modules: “Project Management and Quality” or “SafetyEngineering and Risk Management” to complement their last year of studies. The module“Project management and quality” gives the students some training for jobs like productengineer, project engineer, business engineer, quality engineer. The aim of the module “Safetyengineering and risk management” is to give engineering students some basic knowledge inthe field of safety (nuclear or chemical safety). This becomes mandatory in positionsinvolving responsibilities as department or laboratory manager. This INPG teaching module,directly linked to the GEN option, will be detailed in section 4.2.An additional module concerning “Accelerator Physics and Technologies”, an international courseorganized with a strong participation from the European Particle Physics Center, CERN, is open tostudents coming either from the nuclear engineering option or from the instrumentation option. As thistraining module leads mainly to PhD theses, students are encouraged to follow it though a researchmasters’ like the Masters’ in the Physics of Energetics described in section 4.3.3.1.4 The training periodsAfter the first year, students can take a summer job in a company or a research laboratory, allowingthem to discover their future work environment. At the end of the second year, students have toperform an internship of at least two months, normally in industry in France or internationally. At theend of the third and last year of studies, they have to complete a six month final project, involving anactual engineering or research work.The second year internship and the final project lead to written reports and oral presentations.3.2 Business OpportunitiesFrom 1989 to 2001, ENSPG has awarded more than 1800 degrees. 60% of graduating studentsdecided to begin their professional career straight away. The others preferred to add to the trainingreceived in the school either a doctorate (30% of the graduates) or another specialisation (5% of theengineers chose another scientific speciality, very often abroad; 5% chose to obtain a degree ineconomics or management).The standard profile for an ENSPG alumnus a few years after graduation is a Research andDevelopment job in a high technology field corresponding to the School’s lines, either in Paris or inthe Rhône-Alpes region. But there are many variations, ranging from Theoretical Physics to Marketingand Communication. In a survey of the jobs found after ENSPG by our engineering students, wefound:• 71% in the research and development field• 13% as production, quality and security engineer• 10 % in computing• 6 % as business engineers and in marketing

The main field of opportunities for our engineers is the energetic area, and more precisely research anddevelopment in nuclear energy (34% of them), and also the area of ‘physics and materials’ and‘instrumentation’ (23% each) and finally microelectronics and optoelectronics (20%).4. Energy and Nuclear Education4.1 The ‘Energy and Nuclear Engineering’ (GEN) StreamPower production has been revolutionised for some thirty years by the increasing role played bynuclear energy, particularly in France where some 75% of the electricity is of nuclear origin. In thesame way, nuclear technologies have spread into many other industrial areas: instrumentation,medicine, the food-processing industry, materials...Nuclear energy cannot be separated from the energetic problems involved in the transportation ofenergy and in the transformation of heat into electricity. Therefore, energetics is also a hugeapplication field.Our objective is to train engineers who will master not only nuclear engineering for the production ofelectricity, but more broadly energy and nuclear technologies and their various application fields.The curriculum “Energy and nuclear engineering” of ENSPG is original, and it is the only one inFrance that trains engineers in the nuclear field.This one and a half year specialized training includes a balanced blend of basic courses in energy,nuclear and thermal hydraulic engineering, together with the corresponding technologies engineeringsciences to cover the technological aspects. It is based on the solid background in physics acquiredduring the first year and a half of the common-core syllabus of our engineering school. The details ofthe courses are indicated in table 1.SECOND YEAR COURSES: 340 hoursEnergy and Nuclear Physics: 175 hours- Advanced Nuclear Physics- Neutronics and Reactor Physics- Fluid Mechanics- Radiation-matter Interactions- Radiation Detection- Advanced quantum PhysicsComputing sciences: 20 hoursPracticals: 80 hours- Physics Lab- Practice of numerical methodsTHIRD YEAR COURSES: 350 hoursNuclear Engineering: 75 hours- Reactor Kinetics- Nuclear fuel cycle and wastes management- Nuclear Metallurgy- Nuclear Reactor SimulationsEnergy Engineering: 115 hours- Thermal hydraulics- Thermal Radiation- Simulation in thermal hydraulics- Electrochemical conversionPracticals: 65 hours- Practice of computers in process measurement Optional Lectures: 75 hoursAmong the lectures of the research master inPhysics of Energetics, materials and foreignlanguagesForeign Languages and Sports: 45 hoursForeign Languages: 20 hoursTab 1: Courses of the Energy and Nuclear Engineering StreamAmongst the graduates, roughly 30% decide to add a PhD to the education received in the school,while 60% of them decide to launch their professional career straight after the 3-years education asengineers. Most of these join the design and invention department of major companies working in thefield of nuclear power generation, like AREVA and EdF as far as France is concerned; a significantpart of our alumni are working worldwide, mainly in nuclear industries. They carry out technicalstudies, develop calculation tools and methods, prepare next generation equipments, plants andtechnologies, or improve the performance of water reactors.Another major field of employment, nuclear safety, is growing very fast: nuclear safety is ofparamount importance and this requirement underpins the organization and operation of nuclear

groups. As operating safety engineer, they ensure that safety and occupational safety criteria are met.They also work in radiation protection, radiological monitoring and risk assessment.4.2 Safety Engineering and Risk ManagementPeople are less willing to accept risks whether they are high scale risks (public safety related tonuclear, chemical or transportation activities) or more restricted risks (electrical failure, explosion...)or even natural risks. In his or her job, every engineer, as laboratory, section or firm manager, willhave to take into account the existence of risks and will thus have to acquire some basic knowledgeabout safety.The aim of the “Safety engineering and risk management” module is to give engineering students, as acomplement to their specialty field, basic knowledge about safety. Various approaches are developed:risk analysis, concepts related to operating safety (failure trees...), structure reliability, riskidentification and evaluation, regulation elements, insurance, crisis management… The courseessentially concerns two fields of application: the nuclear and industrial chemistry risks.This INPG module, though driven by ENSPG, also concerns two others engineering schools: theMaterials, Electrochemistry, Process Engineering School (ENSEEG) and the Industrial EngineeringSchool (ENSGI). The curriculum is thus organised so that students of all specialty options of the threeengineering schools concerned can follow the whole module.The 140 hours of lectures, delivered over seven weeks during the third year, deal with risk analysismethods (reliability diagrams, failure analysis, functional analysis, event trees, state diagrams,ergonomic approach), systems reliability, operation safety, technological risk, nuclear safety andradioprotection, risks of the industrial chemistry, impact of regulatory and insurance considerations,risk and crisis management…. These lectures are completed by practical studies on real cases duringthe last week of the module. Lectures are given mainly by industrial teachers from the French nuclearand chemical industries.4.3 Master of Science Degree: Physics of EnergeticsThe Master Degree is the first stage of the doctoral study scheme. Students who complete a Masters’Degree at the same time as their ENSPG engineer degree can thus directly embark on the preparationof a doctorate, which generally lasts three years. In France, the organisation of graduate studies isbased on graduate schools (Ecoles Doctorales). Université Joseph Fourier, the Grenoble University ofScience, operates the Graduate School in Physics in cooperation with INPG. Physics for Energetics isone of the specialties of this Graduate School in Physics, a specialty driven by INPG. More precisely,the specialty ‘Physics of Energetics’, a research oriented masters’ degree, is operated jointly withINSTN, the training subsidiary of the French Atomic Energy Commission, and with the ScienceUniversity of Grenoble.Each Graduate School runs a number of Masters’ Degrees, graduate courses which span a year but canbe taken in parallel with the final year of ENSPG. They consist of a semester of courses, followed byfour to six months of full-time research in an academic or industrial laboratory.The first semester is organized in three different options, all three related by a physics approach:• ‘Nuclear Energy’ option, corresponding to the research aspects of nuclear energy, enlarged by twolectures on energy to be chosen in the two other options of the master• ‘Physics of Transfers’ option, covering the fields of thermal hydraulics, heat transfers andexchanges, two phase flow• ‘Materials for Energy’ option, composed of lectures on solar energy, electrochemical conversionand energy storage (fuel cell), cryophysics, micro fluidics, physics of phase changeMore detailed information can be found in reference [3].Students from the Energy and Nuclear Engineering stream of ENSPG can prepare this Masters’through a special training scheme including:• a set of compulsory courses validated for the school and the masters’ (see list above)• a set of specialized courses, specific to the school, in the continuity of the 4th semester

It leads to positions in the nuclear industry, in public or private research or development laboratoriesinvolved in energy problems, and in engineering companies.5. AcknowledgmentsWe are very thankful to Ms. Elise Huffer for her help during the translation of this paper.6. References[1] E. Merle-Lucotte, D. Heuer, C. Le Brun, J.-M. Loiseaux, “Scenarios for a Worldwide Deploymentof Nuclear Power”, International Journal of Nuclear Governance, Economy and Ecology, Volume 1,Issue 2, pp 168-192 (2006)[2] “A Technology Roadmap for Generation IV Nuclear Energy Systems”, report GIF-002-00,http://gif.inel.gov/roadmap/pdfs/gen_iv_roadmap.pdf (2002)[3] http://master-ep.inpg.fr (in French)

NUCLEAR ENGINEERING EDUCATION: PIONEERINGACTIVITIES IN SAFETY, DECOMMISSIONING AND NEW-BUILDM.J. JOYCEEngineering Department, Lancaster UniversityLA1 4YR Lancaster – United KingdomR.J. AVESDevonport Royal Dockyard Ltd., Devonport Royal Dockyard, Plymouth PL1 4SGABSTRACTThe current professional nuclear engineering requirements of many academictraining/education initiatives are immensely demanding, with many nuclear engineeringorganisations requiring broad syllabi of the highest quality whilst requiring the minimumtime away from the workplace by their employees for the purpose of study. This is a verydifficult balance to achieve, requiring flexibility on behalf of both the education providerand the industry-based recipient of the taught provision if teaching quality and commercialinterests are both to be borne in mind. In this paper we report on an immensely innovativeand successful relationship between a research-led University and a major nuclearengineering organisation that has pioneered an approach to nuclear education to meet thisbalance of needs. This relationship has resulted in over 70 Master’s level graduands innuclear safety, including the professional chartership of numerous nuclear engineers andhas also been instrumental in the launch of the first undergraduate course in NuclearEngineering in the UK for 20 years.1. IntroductionIn the late 1990’s, in the United Kingdom, nuclear education and training was widely regarded to havereached its lowest ebb, as a result of a moribund civil nuclear industry and the paucity of public sectorfunding initiatives to substantiate new nuclear educational initiatives. This unsustainable situation wasidentified and recorded in a number of surveys [1,2,3,4,5] from a number of expert groups from 2000through to 2003. This situation was characterized at the time by the entire lack of a dedicated nuclearengineering first degree and acute difficulties in sustaining postgraduate activities.Somewhat independently of these activities a very productive relationship formed between theDepartment of Engineering at Lancaster University in the UK and the Devonport Royal Dockyard(DRDL), Plymouth UK in 1999. This collaboration across the academic-industrial divide and across500 kilometres has resulted in significant innovation in nuclear education and industry-based training,including:• Education approved by the engineering institutions, enabling routes to chartership for studentsfrom a broad selection of engineering backgrounds and experiences.• Accredited prior experiential learning, providing a variety of routes to postgraduate HigherEducation (HE) for employees without a traditional higher-education portfolio.• Industry-based training local to the sponsoring company’s location.• Postgraduate education that marries the needs of the employer with the teaching and learningrequirements of the HE provider.• A welcome blurring of the traditional roles of academia and industry, with students learningfrom each other and from industrial role models who contribute directly to the taughtprovision.

2. Course provision2.1 Safety Engineering MastersLancaster and DRDL first established their relationship with the Safety Engineering MSc in 2000.This course’s aim was to tackle an important development that occurred in the late twentieth centuryassociated with the complex mixture of engineering sub-disciplines that constitute many engineeringsystems. For these systems, the division between mechanical and electronic aspects of such a systemis often difficult to discern. In particular, it is often difficult for the non-specialist engineer to identifythe interdependencies of mechatronic engineering systems. Moreover, the inclusion of embeddedintelligence in otherwise traditional mechanical systems, such as fluid valves, introduces a subsequentinterface between software and electronic hardware. Hence, a system for which risk may have oncebeen adequately assessed by a single engineer, may now require input from several experts.Many professional engineers have an undergraduate degree as their training foundation fromwhich they continue to draw on throughout their careers, to a greater or lesser extent. The relentlessgrowth of technology has, and continues to result in the supporting syllabi of this essential preparationbeing increasingly wide. It is, in part, a reflection of this that in the UK four-year MEng degrees havebecome the standard professional requirement for engineering chartership in the UK. Indeed, there isoften greater pressure on undergraduate engineering schemes of study than, for example, Physics,since the lack of a dedicated secondary-school subject restricts the extent to which established aspectsof the syllabus can be incorporated into the secondary-level education syllabus; the capacity for flowdownto schools is limited. This is especially relevant to softer aspects of engineering training. Thusthere remains a constant requirement for both the old and new at undergraduate level.Year 1 Year 2Management in a safetycultureDesign of safety-criticalsystemsIndustrial engineeringSystemsEXAMSOperational safetyHuman factors andsafetyIndustrial safetyenvironmentEXAMSCourseworkCourseworkSafety engineering research projectFigure 1: Schematic of Safety Engineering MSc schemeA significant effect of this competition for space in the tertiary syllabus is that professionalsafety aspects are paid relatively little attention at present. This is despite the clear relevance of safetyto important softer skills, such as engineering management, economics, engineering in society,sustainability and environmental issues of engineering. Furthermore, adequate substantiation oftenrequires rigorous application of the most important engineering principles, and can draw on profoundnumerate ability in many cases. Indeed, the safety audit of a given engineering system can provideextensive exemplary material for the engineer early in their career.In response to this training shortfall, industry has established in-house training procedures thatoften constitute part of the professional engineering industrial graduate training programme.However, unless the individual concerned is likely to realise a position dedicated to safety, the depthand breadth of this training rarely extends beyond operational concerns designed to empower theindividual against individual risk. Even for the case where safety is likely to take a higher profile ofan individual career, the training rarely extends beyond the specialist immediate needs of the industryconcerned, thus neglecting the broader-based needs of the engineer’s professional development anddirect transferability of training.

2.2 Decommissioning and Environmental Clean-up MastersIn April 2005, the Nuclear Decommissioning Authority (NDA) was formed with the mission to dealwith the UK’s nuclear legacy facilities to time and cost, safely and with respect for the environment.This goal is ambitious and requires that training and education provision in nuclear topics is expandedand broadened. To appeal to this requirement, Lancaster designed and launched a postgraduatescheme in October 2004 which anticipated the postgraduate needs of engineers working in or aspiringto work in the decommissioning sector. This course built upon our experience with the SafetyEngineering MSc. and followed a similar format in order to balance the needs of the workplace andthe individual student.Figure 2: Schematic of Decommissioning and Environmental Clean-up MSc schemeThe Decommissioning MSc was designed in correspondence with the UK Government’sWhite Paper ‘Managing the Nuclear Legacy’ [6], which highlights issues regarding ProjectManagement, Safety and Environmental Restoration. This course shares modules with the SafetyEngineering MSc., since this enables students to share their experience with students with a broadersafety background. The course collaborates with the Department of Environmental Science atLancaster, which provides expertise relating to Environmental Quality Standards, also drawing onregulator expertise. The Westlakes Research Institute, in West Cumbria near the Sellafield site,provides complementary expertise in Environmental Decision Making within the essential context ofthe nearby nuclear reprocessing complex. This course has proved very popular, drawing students fromBritish Nuclear Group and Nexia Solutions, amongst others.3. Industry-based projectsA key benefit of the industry-based Masters courses described in this paper has been the opportunityfor students to pursue an extended scheme of individual study on a topic of joint interest to themselvesand their sponsoring employer. Representative examples of such studies include:• Quantitative aspects of the ‘As low as reasonably practicable’ (ALARP) concept,• A review of safety issues associated with the application of nuclear power in transport,• An investigation of cultural analysis to improve the interaction between design, safety andoperations.In some cases, these studies have stimulated further interest in the students associated with theseprojects to do further research, potentially via part-time PhD study. Very importantly, these projectshave enabled students to establish themselves within their companies as resident experts in the topicsthey have studied as part of their projects.4. Peer review and awardsAs an indication of the success of the industry-academe collaboration described in this paper the teamdelivering the course and the graduates from the course have received related peer-driven awards,including:

• British nuclear Energy Society (BNES) Young Generation Network (YGN) AchievementAward (2007)• UK Nomination for the Jan Runermark Award (2006)• Institute of Physics Award for Best Practice for Professional Development (2006),• BNES Masters’ project prize (Lancaster) (2006),• Royal Academy of Engineering Teaching Prize (2005),• The Jack Martin Prize for excellence in Radiation Protection (2004),• Lancaster University Staff Prize (2004).Although both of the Masters courses described in this paper were established without any priorfunding, the Decommissioning and Environmental Clean-up course has subsequently attracted fundingfrom the Engineering and Physical Sciences Research Council (EPSRC), as part of a CollaborativeTraining Account (CTA). This has enabled dedicated facilities to be established at Lancasterincluding what is possibly the UK’s only mock contaminated cell for laboratory-based remotecharacterization with a Brokk remote manipulation platform. As a measure of the course’s success, ithas also attracted a number of scholarships from the NDA as an incentive to Small and Medium-sizedEnterprises (SMEs).5. Future directionsRecently, modules from these degree schemes have been used as part of the Nuclear TechnologyEducation Consortium (NTEC), which brings together further provision of this type across a broaderrange of syllabi. Lancaster is currently building on its experience in nuclear education at thepostgraduate level through the launch of an undergraduate degree in Nuclear Engineering. This courseis designed as a four-year Masters of Engineering (MEng) degree and will began in October 2006.This course has already attracted sponsorship from the Institution of Nuclear Engineers (INucE), andhas already generated a great deal of interest from graduate employers in the nuclear sector andprospective students alike.6. References[1] OECD/NEA report, Nuclear education and training – causes for concern, 2000,http://www.nea.fr/html/ndd/reports/2000/nea2428-education.pdf[2] HSE/NII report, Nuclear education and training in British Universities, 17 October 2000,http://www.hse.gov.uk/nuclear/ukeduc.doc[3] HSE/NII report, Nuclear education and training in British Universities, February 2002,http://www.hse.gov.uk/nuclear/edu0202.pdf[4] Update on nuclear education and training in British Universities, December 2003,http://www.hse.gov.uk/nuclear/education03.pdf[5] COVERDALE, T., Report of the Nuclear Skills Group, Department for Trade and Industry,Nuclear and radiological skills study, 5 December 2002, http://www.dti.gov.uk/files/file23311.pdf[6] Managing the nuclear legacy – a strategy for action, HMSO Command Paper 5552, February 2002,http://www.sepa.org.uk/pdf/consultation/current/ilw_conditioning/Managing_the_Nuclear_Legacy.pdf

CHERNE – DEVELOPING A NETWORK TO ENHANCECOOPERATION FOR HIGHER EDUCATION ONRADIOLOGICAL AND NUCLEAR ENGINEERINGF. TONDEURInstitut Supérieur Industriel de Bruxelles, HE Spaak,150 rue Royale, B1000 Brussels - BelgiumJ. RÓDENASSecretary of the CHERNE NetworkDepartamento de Ingeniería Química y Nuclear, Universidad Politécnica de ValenciaCamino de Vera 14, E-46022 Valencia - SpainOn behalf of the CHERNE NetworkABSTRACTIn the last two decades, the educational capacity of many European Institutions of HigherEducation in the field of Radiological and Nuclear Engineering has decreased, because ofthe conjunction of less interest among students, academic and political authorities. Anincreasing cooperation at the international level on the educational efforts in radiologicaland nuclear science and engineering is necessary. The CHERNE network is an initiativemainly focussed on teaching and learning activities to develop a wide-scope open academicnetwork to enhance cooperation, competence as well as equipment sharing between itspartners. Typical activities organized within the network include workshops, intensivecourses, seminars and conferences. Student and professor exchanges necessary for theseactivities are organised when possible in the framework of the ERASMUS program. In thispaper, the CHERNE network and its main objectives will be presented and an account ofthe activities developed since its foundation, or foreseen in the near future, will be given.1. IntroductionThe educational capacity of many European Institutions of Higher Education in the field ofRadiological and Nuclear Engineering has decreased in the last two decades, in parallel with thedecrease of interest for this domain among students as well as among academic and politicalauthorities. Furthermore, financial restrictions have made it more difficult to maintain and developfacilities, equipment and academic staff needed for practical training of students.Each university and country presents a different situation, but many departments that were initiallyable to propose a large panel of orientations in this field had to reduce their offer and to concentrate iton a few specialities.On the other hand, a significant number of professionals at different levels of education continue to berequired for safely operating and managing the nuclear industry and all other activities involving theuse of radiations.In this situation, an increasing cooperation at the international level on the educational efforts inradiological and nuclear science and engineering is considered presently as the only viable solution.For this reason, several networks have been developed, some of them focused on specific domains,others concentrated on high level professional training, some strongly structured and others not.In particular, the CHERNE network is an initiative mainly focussed on teaching and learning activitiesto develop a wide-scope open academic network to enhance cooperation, competence as well asequipment sharing between its partners. Typical activities organized within the network includeworkshops, intensive courses, seminars and conferences on radiation protection and nuclear

measurement, radiochemistry, safety analysis, etc. Student and professor exchanges necessary forthese activities are organised when possible in the framework of the ERASMUS program.In this paper, the CHERNE network and its main objectives will be presented and an account of theactivities developed since its foundation, or foreseen in the near future, will be given.2. The CHERNE network2.1 Members of the network in 2007The network was created in 2005, involving now 12 European Institutions and one from United States.The list of members in alphabetic order of cities is the following:• UAS Aachen, University of Applied Sciences Aachen, Campus Jülich (Germany)• ETSEIB - UPC, Escola Tècnica Superior d’Enginyers Industrials de Barcelona, UniversitatPolitècnica de Catalunya (Spain)• Alma Mater Studiorum - Università degli Studi di Bologna (Italia)• ISIB, Institut Supérieur Industriel de Bruxelles (Belgique)• Dipartimento di Fisica ed Astronomia, Università di Catania (Italia)• XIOS, Hogeschool Limburg, Diepenbeek (Belgium)• KSU, Kansas State University (USA)• ITN, Instituto Tecnológico e Nuclear, Lisboa (Portugal)• Dipartimento di Fisica, Università degli Studi di Messina (Italia)• Dipartimento di Ingegneria Nucleare, Politecnico di Milano (Italia)• ČVUT, České Vysoké Učení Technické v Praze (Czech Republic)• DIQN-UPV Departamento de Ingeniería Química y Nuclear, Universidad Politécnica de Valencia(Spain)• UAS Zittau-Görlitz, University of Applied Sciences Zittau/Görlitz (Germany)It is a wide-scope open academic network mainly focussed on teaching and learning activities, whoseobjectives are to enhance cooperation, competence as well as equipment sharing between partners.A declaration, signed by all partners, contains details concerning organisation, membership andactivities. This declaration can be consulted at the web site www.upv.es/cherne/2.2 Origin of the CHERNE networkThe CHERNE network has its origin on some ERASMUS Intensive Programmes (IP) organisedduring last years [1]. The IP “PAN: Practical Approach to Nuclear techniques” was first organised in2002 in Prague, with the participation of CVUT, DIQN-UPV and ISIB. XIOS and UAS Aachen joinedthe two next editions, held in Prague (2003) and Mol-Brussels (2004). A second IP (SPERANSA,Stimulation of Practical Expertise in RAdiological and Nuclear SAfety) was first held in Prague in2005, with no European grant, by the same partners. This project was supported by the Erasmusprogramme in 2006 (Mol-Jülich), 2007 (Prague) and 2008 (Mol-Brussels).A larger partnership was considered necessary to extend the scope of this collaboration, and wasinitiated with the constitution of the CHERNE network in 2005 during a workshop organised by UPV[2].2.3 CHERNE organisation and membershipCHERNE has a minimal administrative organisation, ensured by the secretary elected at the annualmeeting. The secretary manages a Web page through which the activities of the network arecommunicated. The partners of CHERNE meet once a year to evaluate the activities of the networkand discuss any proposal to extend or modify them. For the moment no fee is foreseen for CHERNEmembership.Academic institutions, research institutions, companies or individuals are accepted as members onpresentation by two members, including at least one European academic member. Documents for thispresentation as well as the list of partners can be found at the official Web site.

3. CHERNE activities3.1 DescriptionCooperation between the institutions should enhance the mutual support by learning from each other,by exchanging experiences, and by regular mutual reflections on what we can do to counteract the 'lessinterest among students' and the 'less interest among the academic and political authorities' and also onwhat we can learn from more successful or from less successful partners.The scope of CHERNE is not limited and any activity related to higher education in radiologicaland/or nuclear engineering can be proposed.CHERNE activities will be organised mostly for students of members, mainly at Master level. Theyshould include at least a one-week/2 ECTS module. It’s necessary to include practical training inactivities for students, including when possible an access to large facilities. Teaching modules areclearly seen as a possible kind of activity, but other types of cooperation may be also developed suchas material for modules conveniently adapted in each university, e-learning, etc. The language used inCHERNE activities is English.The CHERNE activities will be organised at no cost, or very low fee, for students coming from otherpartner institutions. The organising partner will find and propose cheap accommodation for thestudents coming from abroad. When possible, the organisation of CHERNE activities will be includedin ERASMUS exchanges. Therefore, the partners are encouraged to sign bilateral ERASMUSagreements.Research collaborations are not the main goal of the network. However, they are quite naturallydeveloped as a consequence of the frequent exchanges for educational cooperation. [3, 4, 5]3.2 CHERNE activities developed or proposedActivities already realised or planned for the near future include seminars, courses, intensive courses,and research collaborations.Seminar:• Simulation of detector calibration using MCNP, by Prof. J. Ródenas (UPV) at ISIB, 2005.• Neutron Detection and Measurement, by Prof. U. Scherer (UAS Aachen) at U. Bologna, 2007.Workshops and Conferences:• Annual workshop of the CHERNE network, Valencia (2005) [2], Valencia (2006) [6], and Prague(2007) [7].• Participation in ETRAP 2005 [1].• Participation in the First EUTERP Platform Workshop [8].Courses:• Gamma spectrometry: simulation, deconvolution, applications, by Prof. J. Kluson (CVUT) atISIB, 2005.• X-Ray Photon Spectroscopy Calculations, by Prof. J. Fernández (Bologna) at ISIB, 2007.• Introduction to plasma physics, by Prof. D. Mostacci (Bologna) at ISIB, 2005 , 2007.• Protection against natural radiation, by Prof. F. Tondeur (ISIB) at U. Bologna, 2005 , 2007.• Participation of Prof. F. Tondeur (ISIB), Prof. E. Zio (Milano) and Prof. J. Fernández (Bologna) inDoctoral courses at UPV, 2004-2007.Intensive courses:• Radiation protection and nuclear measurement in non conventional sectors.2-week course organised by ISIB Brussels and XIOS Diepenbeek (Belgium), 2007.• Nuclear Chemistry.2-week course organised by UAS Aachen , 2007.• Low radiation measurements.1-week course organised by UAS Aachen, 2007.• Probabilistic Safety Analysis.2-week course organised by UPV and Politecnico di Milano, 2008.

• Radiation protection and nuclear measurement in non conventional sectors (2nd edition)ISIB & XIOS, 2008.• Nuclear Chemistry (2nd edition) , UAS Aachen, 2008.This last course is submitted as an Erasmus IP for the academic year 2008-2010. Another Erasmus IPproject will be submitted for 2009-2011, coordinated by Politecnico di Milano for a first organisationproposed to ITN Lisbon.3.3 Some Statistics.A resume of the collaborations between the CHERNE partners is schematically presented in table 1.PoliMilanoPoliMilanoISIB UPV XIOS Bologna Jülich CVUT CataniaIP EBA, IP IP IP EBA, IPISIB EBA, IP IP EBA BA, IP EBA, IP EBAUPV PE, SE PE, SE EBA, IP EBA EBA, IP EBA, IPXIOS EBA, IP EBA, IPBologna PE, SE PE EBAJülich PE, SE SE IPCVUT PE, SE PE, SE SECataniaTable 1: Summary of collaborations between CHERNE membersIPEBAPESEcooperation in one or several intensive programme(s)Erasmus bilateral agreementprofessor exchangestudent exchange.4. ConclusionsOn the basis of an existing collaboration between some institutions, the creation of the networkpermitted to enhance the educational cooperation among partners.The main target of the CHERNE network is to develop teaching activities for the benefit of students ofthe institutions belonging to the network.The network is still young and small, and does not yet propose many activities, but already representsa clear added value for the students, in particular with the intensification of Erasmus exchangesbetween the partners. Consequently, the exchange of students has been clearly increased.Furthermore, specific activities already developed and those proposed for the future are on the way toenhance the interest of students and academic authorities on Nuclear Engineering and RadiationProtection.The perspective of the network is to gradually propose more activities, while admitting new partnerswho can contribute to the network’s life with new activities and more students benefiting of them.5. References

[1] T. Cechak, F. Hoyler, H. Janssens, J. Rodenas, U. Scherer, F. Tondeur; “Sharing the access tobig nuclear facilities for safety training: experience of an ERASMUS intensiveprogramme”; 3rd International Conference on Education and Training in RadiologicalProtection ETRAP 2005, Brussels, Belgium, 23 – 25 November 2005, www.etrap.net.[2] Proceedings of the First Workshop on European Collaboration for Higher Education andResearch in Nuclear Engineering and Radiological Protection CHERNE 2005, UniversidadPolitécnica de Valencia, Valencia, Spain, 4-6 May 2005.[3] S. Gallardo, J. Ródenas, G. Verdú, J. Fernández; “Application of the Monte Carlo method to X-ray unfolding: Comparison between germanium and silicon detectors”; EXRS 2006 EuropeanConference on X-Ray Spectrometry, Paris, France, June 19-23, 2006.[4] N. Rasson, S. Gallardo, J. Ródenas, I. Gerardy, M. van Dycke, F. Tondeur, “Simulation of thedose distribution for a Brachytherapy source of Ir-192 using the Monte Carlo method”, FirstEuropean Workshop on Monte Carlo Treatment Planning “Introduction of MCTP into theclinic”, Gent, Belgium, October 22-25, 2006.[5] I. Gerardy, J. Ródenas, M. van Dycke, S. Gallardo, F. Tondeur, “Application of the MCNP5code to the modelization of vaginal and intra-uterine applicators used in intracavitarybrachytherapy: a first approach”, Third McGill Workshop on Monte Carlo Techniques inRadiotherapy Delivery and Verification, Montreal, Canada, May 29-June 1, 2007.[6] Proceedings of the Second Workshop on European Collaboration for Higher Education andResearch in Nuclear Engineering and Radiological Protection CHERNE 2006, UniversidadPolitécnica de Valencia, Valencia, Spain, 13-15 March 2006.[7] Proceedings of the Third Workshop on European Collaboration for Higher Education andResearch in Nuclear Engineering and Radiological Protection CHERNE 2007, CVUT, Prague,Czech Rep., 8-10 February 2007.[8] J. Ródenas, on behalf of the CHERNE Network; “CHERNE – Cooperation for HigherEducation on Radiological and Nuclear Engineering”; First EUTERP Platform Workshop;Vilnius, Lithuania, 22-24 May 2007.

POPULARIZING NUCLEAR SCIENCE AND TECHNOLGYTO STUDENTS OF SOME BRAZILIAN HIGH SCHOOLSW.A. SOARESBusiness and Communications Division, Nuclear Technology Development Centre (CDTN/CNEN)P.O. Box 941. CEP 30123-970 - Belo Horizonte- Minas Gerais - BRAZILF. MARETTI JR.Reactor and Irradiation Division, Nuclear Technology Development Centre (CDTN/CNEN)P.O. Box 941. CEP 30123-970 - Belo Horizonte- Minas Gerais - BRAZILABSTRACTThis paper intends to present the results of the project “Nuclear energy: itinerantexpositions” coordinated by the Nuclear Technology Development Centre (CDTN). Publichigh school students were the focus of the project. Stimulating such students for subjects,like physics, chemistry, biology, mathematics, history and awakening vocations to sciencewas the main objective. The project consisted of an exposition and a talk motivating theaudience to the nuclear theme associating at the same time to the subjects taught at suchschools. Searching information on the target public, infrastructure mounting, team training,multimedia material elaboration, interviews with students and teachers by journalists,project evaluation and divulgation were typical project activities. About 40 people ofCDTN were directly involved in the project, that reached 30 public schools and about11,100 students. The project received only high approval as the global evaluation in thereturned questionnaires.1. IntroductionNuclear energy to the public is always associated with the production of nuclear weapons or nuclearand radiological accidents. Public communication actions done by the Nuclear TechnologyDevelopment Centre (CDTN) have been contributed to popularize the social and peaceful applicationsof nuclear. The project “Nuclear energy: itinerant expositions” is one of these actions and had highschool students as target public. Stimulating such students for subjects, like physics, chemistry,biology, mathematics, history, etc, awakening vocations to science and technology and alsocontributing for having citizens able to question their uses were the purpose of the project. The projectconsisted basically of an introductory talk on fundamentals and applications of the nuclear energy andionizing radiation followed by a visit on a nuclear stand exposition. The talk had as goal motivatingthe audience to the nuclear theme and at the same time associating it to subjects taught at suchschools.2. Project methodologyThe project required the implementation of six important phases: target public survey, design, teamtraining, execution, evaluation and divulgation.2.1 Target public surveyIn this phase on target public to be reached a list of schools in the metropolitan region of BeloHorizonte-MG was prepared. Some candidate schools were selected and visited in order to know theavailable space and infrastructure needed to the project. It was important to have an idea of the schoolambient and students’ behaviour. It was observed that some schools had large spaces for the talkpresentation and other small ones. This last information helped then to define the number of talks tobe presented and the number students to be sent each time, to the stand exposition.

2.2 DesignOnce defined the selected target public, a letter was addressed to each school formalizing theinvitation. In order to keep students quiet during the talk and paying attention to the event, specialstrategies had to be used in the production of the multimedia material. Short time films and animatedimages were incorporated into the slides as a way of facilitate the understanding. Small scale modelsof a nuclear power plant, a research reactor nucleus and a gamma irradiation laboratory were prepared.Coloured gems by gamma rays were used also to compose the stand. Illustrative panels on nucleartheme were prepared to compose the exposition stand making easy the expositor explanations. About2,000 samples of a CNEN nuclear booklet on the nuclear energy applications were pre-printed and aparticipation certificate was prepared do be distributed to each school that took part in the project.2.3 Team trainingA workshop on scientific divulgation, conducted by researchers with experience in communicationmade part of the team training stage. Researchers and technicians of CDTN and members of thecommunication area from the nuclear institutions and from the science and technology area took partin this workshop. Three previous training for the team directly involved in the work at the expositionstand were also prepared using the multimedia material produced for the project.2.4 ExecutionIn general, one day before the execution, two members of project staff moved to the school with allneeded infrastructure in order to let it safely, near the space where the exposition would take place. Inthe beginning of the following day the project coordinator and members of the staff moved again toschool to install all equipment in order the talk could start. In parallel the exposition infrastructure wasmounted to receive the students after the explanation. A foreseen 45 minutes talk start then with anaudience of a order of 500 students, with the coordinator trying to relate science and technology withthe school subjects and using images as a source of concepts retention. At the end of this session, abooklet on nuclear energy applications was distributed to each participant. All students moved then totheir classrooms. Then a teacher conducted ordered groups of students to the stand, where they couldsee the exposed material and scale models, receive additional information and talk with the expositors.In general the event started at seven and finished at noon, but sometimes in function of the number ofstudents the event still run in the afternoon. Once finished the event, nuclear energy informativematerial and a participation certificate were given to the school principal.2.5 EvaluationThe project evaluation consisted of three approaches: 1) interviews with students and teachers byjournalists; 2) evaluation of the project by the schools, at the end of the execution phase; and 3)evaluation of the project by some students three months after the event. The information asked andresults related to the second approach are presented in Table 2. The methodology used for the thirdapproach was applied to the last six schools, involving only a sample of students. Items involved in theevaluation and its results are presented in Tables 3 to 4.2.6 DivulgationOnce each school was visited by the project, news on such event were disseminated internally in theCentre by a weekly mural newspaper and sometime by an electronic bulletin distributed to all workers.The CDTN printed newspaper, for the internal and external public, was also used do bring news on theproject. Releases to the media were sent before each visit to the schools and some television networksand newspaper present news on the project.3. ResultsFigures 1 and 2 illustrate exposition scenes.

Figure 1 - Talk in opened gymnasium (Photo:Santiago).Figure 2 - Exposition of a research nuclearreactor core (Photo: Santiago).Performance indicators were used in order to quantify all tasks done in the project and also its results.Table 1 shows results of main indicators used for monitoring the project development.Table 1 – Main indicators used for monitoring the project development.Indicator nameMeasuredresultNumber of people involved in the project execution 39Number of different public schools that took part in the project 30Number of visits of the project to public schools 36Number of public school classes reached by the project 294Total number of students from public schools that took part in the project 11,116Number of talks presented 64Number of hour of talks presented 48Number of students and teachers interviewed by journalists 31Number of news disseminated internally in the mural newspaper 8Percentage of the Very Good grade in global evaluation (event) 84.8Figure 3 shows the evolution of the student that received talks on nuclear science and technologygiven by CDTN. The Nuclear Energy: Itinerant Expositions project results were included and the highimpact of the project is prominent.Thirty eight people of the CDTN were directly involved by the project, working 1,535 hours,corresponding to 192 days of 8 hours or 6.4 months of work. This work duty was done by 9 doctors,11 masters, 2 graduated, 3 journalists, 8 technician from nuclear area and 2 from the administrativearea. The first 18 public schools received the project in its own installations. In 2006, a very importanttechnological and historical nuclear conference took place in Belo Horizonte, on a big conventioncentre - Minascentro. This event also celebrated the birth century of the pioneer in nuclear energy inMinas Gerais, Francisco de Assis Magalhães Gomes an it was used to reach the goal of presenting theproject at one great circulation place. Another 12 public schools took part in the itinerant project there.As free available auditorium was available at the centre, 26 private schools participated in the project,totalizing 1,524 students. In the project evaluation process at the end of each event, 33 questionnaireswere filled and returned. Grade for evaluation varied from 1 (Very Bad) do 6 (Very Good). The resultsare presented in Table 2. The project received 85% of “Very Good” and 15% of “Good”.

12000Number of students present in talks on nuclear themes10000Itinerant Exposition at Schools8000Itinerant Exposition in Minascentro - Private Schools6000National Week on Science and TechnologyCDTN Open Doors24708646400020000Talks standard CDTN program1250500855 6981093 150183412835241129 1394 1493 12262000 2001 2002 2003 2004 2005 2006Figure 3 - Evolution of number of students taking part in talks on nuclear technology in CDTNprogram.Table 2 – Evaluating the project at the event’s end.Evaluated items Evaluated themes NATEvaluation (%)Regular Good VGPreliminary contacts Preparatory contacts 32 3.1 12.5 84.4Pedagogical/didactic interest 32 0 21.9 78.1TalkTalk content 32 3. 6.3 90.6Expositors explaining capacity 31 3,3 29 67.7Global evaluation 8 0 0 100Visual impact 33 9.1 21.2 69.Pedagogical/didactic interest 33 0 21.2 78.8Exposition Exposition content 33 3 21.2 75,8Expositors explaining capacity 33 6,1 24.2 69.7Global evaluation 23 0 13 87Visual impact 33 3 24.3 72.7Distributed materialPedagogical/didactic interest 33 3 18.2 78.8Content 33 0 9.1 90.9Global evaluation 24 0 0 100Global evaluation Project global evaluation 33 0 15.2 84.8Notation – NAT: Number of answer to each theme. VG: Very Good.Three months after the last event, evaluation of the project results was done based on a sample of 162students from six public schools. 55% of these students affirm that they were stimulated to study morethe subjects presented by the expositors and 93.6% consider important the presented talk and theexposition. Table 3 presents results concerning the feelings of the students when confronted with thenuclear theme.

Table 3 - Feelings of the students hearing about radiation or nuclear area.Associated image Answers (%)Utility and applications that the nuclear area can have 42.6Atomic weapons and nuclear wars 19.8Danger to people due to the use of radiations 17.9Danger to the environment with the use of radiations 8Science and technology that the students do not yet know 6,2Not answered 4.9None of the other options 0.6The project called attention to the Education Centre for Sciences and Mathematics (CECIMIG) of theEducation Faculty of the Federal University of Minas Gerais (UFMG) that invited the coordinator toshow the project experience in a specialization course for public state high school teachers, and up tonow 500 teachers took part. This experience was also presented to 30 journalism students of theNewton Paiva Institute, in Belo Horizonte.4. ConclusionsTo be at 6 hours in the morning at the schools were challenge faced by the project staff involved in theinfrastructure mounting at each event. A negative point in the majority of the cases was a not priorpreparation of the students by the teachers in order to take advantage of the event, as for examplemaking the participation as part of a school homework. Only few schools adopted such procedure.Understandable, but negative, in many schools, was the absence of some teachers during the event.Lack of adequate infrastructure for giving talks and for locating the exposition stand was also faced.The foreseen goals to attend 30 schools and also to carry out the project in place of great circulation ofpeople were reached. The project contributed in order that CDTN could fulfil its social and strategicgoal to increase the number of public schools reached by the science and technology divulgationprogram in relation to the number of private schools. The project was a very good opportunity for theresearchers and technician of the CDTN learning more on the job training and knowing how to interactdirectly with society, speaking about the nuclear science and technology with a very questioningpublic as young students. The multimedia material developed with graphical animation resources, thesmall scale models of nuclear and radioactive installations and the irradiated foods demonstrationcreated propitious conditions in order the students could clarify its doubts and questionings with theexpositors. The project evaluation results indicate that it contributed to awake the interest of thestudents for school subjects. Designed initially only to high school students the target public wasenlarged, pointing out the importance of the project. The merit honour distinction given to the projectin the 2006 by the Minas Gerais state government in the Prize “Francisco de Assis Magalhães Gomes”recognizes the project as an important instrument of science and technology divulgation.AcknowledgmentsThanks to the Minas Gerais government Agency for Support of Research (Fapemig) that sponsored the“Nuclear Energy: Itinerant Expositions” project and also to the Minas Commerce Association(ACMinas) a very important partner in the project. Thanks to all colleagues from the Business andPublic Communication Area and from of other sectors of CDTN, who understood the importance ofdivulging the peaceful applications of the nuclear technology to the students and contributed with theirworks in this project. Thanks also to the researchers Alfonso Rodrigues de Aquino and MarthaMarques Ferreira Vieira, from the Nuclear and Energetic Research Institute (IPEN), for conducting aWorkshop on Scientific Divulgation at CDTN in 2005, as part of the team training project.

Workshop 2:Safeguards and terrorism

FIELD TRIAL OF SAFEGUARDS’ SHORT NOTICE RANDOMINSPECTIONS AT THE JUZBADO PLANTO. ZURRÓN-CIFUENTES, C. ÁLVARO-PÉREZDepartment of Safety, ENUSA Industrias Avanzadas S. A.Juzbado Nuclear Fuel Fabrication PlantCtra. Salamanca-Ledesma, km. 2637015 Juzbado, Salamanca - SpainABSTRACTDuring the last decade a number of different issues concerning either illegal traffic ordiversion of nuclear material for undeclared activities have been detected by the safeguardsagencies throughout the world. These global challenges have led the safeguards agencies torework their own procedures and objectives. In particular, both the IAEA and the EuropeanCommission (EC) have been interacting at the highest level during the last months in orderto conclude an agreement on its collaboration protocols, aiming to easy the achievement oftheir new goals and objectives and increase the effectiveness of the safeguards controls.Within this framework and in line with its transparency policy, ENUSA and the DeputyDirection for Nuclear Energy of the Spanish Ministry of Industry, Tourism and Trade havebeen working jointly with the IAEA and the EC DG-TREN to implement a six months fieldtrial of a Short Notice Random Inspections’ scenario (SNRI) at the Juzbado Nuclear FuelFabrication Plant. The implementation procedure of this field trial was agreed upon by allthe parties involved in January, 2007 and is the first of its kind within the EU. The lessonslearned are intended to be taken into account in the implementation of integrated safeguardsin the rest of the EU. This paper shows the details of such SNRI scenario for the JuzbadoFuel Fabrication Plant.1. IntroductionDuring the last decade a number of different issues concerning either illegal traffic or diversion ofnuclear material for undeclared activities have been detected by the safeguards agencies throughoutthe world. In some cases, these misled activities were committed by individuals or entities escapingthe existing States’ controls, but there is a number of examples with the governments themselves beingthe main actors of such behaviour, so challenging the global consensus on the right way to proceed inorder to avoid nuclear weapons’ proliferation and assure the development of peaceful uses of nuclearenergy worldwide. These global challenges have led the safeguards agencies to rework their ownprocedures and objectives, realizing that the control should be not only on the nuclear material itself,but also on the parallel tasks needed to be carried out to develop undeclared nuclear programmes orother malicious activities.As an outcome of such internal revisions, the IAEA issued the Additional Protocols (AP) to theSafeguards Agreements, which have been so far endorsed and implemented by a large number ofStates, including the European Union. The AP implementation is seen by the IAEA as the first step ofa much more ambitious aim called Integrated Safeguards, which is intended to be fully implementedwithin the EU by the end of 2008. The ongoing revision by the European Commission (EC) of its ownsafeguards approaches can be seen in this same regard. These so called new approaches have beendeveloped during the last two years by the EC taking into account the different sensibilities of the EUMember States and are intended to be officially issued not far long. Furthermore, both Inspectorateshave been interacting at the highest level during the last months in order to conclude an agreement on

its collaboration protocols (the so-called partnership approach), aiming to easy the achievement oftheir new goals and objectives and increase the effectiveness of the safeguards controls.Within this framework and in line with its transparency policy, ENUSA and the Deputy Direction forNuclear Energy of the Spanish Ministry of Industry, Tourism and Trade have been working jointlywith the IAEA and the EC DG-TREN to implement a six months field trial of a Short Notice RandomInspections’ scenario (SNRI) which complies with the safeguards objectives of both inspectoratestaking into account the specific constraints of the Juzbado Nuclear Fuel Fabrication Plant. Theimplementation procedure of this field trial was agreed upon by all the parties involved at thebeginning of 2007 and is the first of its kind within the EU. The lessons learned are intended to betaken into account in the implementation of integrated safeguards in the rest of the EU.The following epigraphs show the details of the above cited SNRI field trial at the Juzbado Plant.Also, an epigraph with the definitions of the main concepts discussed has been included at the end ofthe paper, with the purpose of easing its comprehension. The basic objective of such inspections is toprovide full coverage of the nuclear material involved in domestic transfers through random selectionof the timing of the inspections, so that the maximum advance notification to the Juzbado Plant is twohours. From the operator’s point of view, the benefit lies in a reduced annual number of inspections,which in the particular case of the Juzbado Plant would vary from a figure of 6-7 inspections per yearto 4-5 inspections per year.2. Notification of inspectionsThe notification that a SNRI inspection has been triggered will be received at the plant between 9:00and 9:30 of any of the agreed dates (which basically match with the Juzbado Plant working days).Should a notification is not given during the notification window no SNRI can take place during therest of that day.The Party (IAEA or EC) triggering the inspection will submit a fax to the Plant’s Control Room,backed up by telephonic confirmation. The fax’s template has also been agreed and contains veryspecific information on the details of the inspection. Later this same day, the Plant’s Control Roomwill receive another fax submitted by the second Party with similar information regarding itsparticipation in the inspection (basically, date and estimated hour of arrival of the inspectors and theirnames). All the information received at the Plant’s Control Room is promptly forwarded to the onplantorganizations responsible to host the inspection, which have to be activated upon reception of thenotification.3. Access to the plant of the inspection teamOnce notification of a SNRI has been given, the SNRI must take place. In case of force majeure eventswhich could interfere with the inspection, the operator will always grant the inspectors access to theplant and later discuss with them the situation and determine the appropriate course of action.The inspection team will arrive at the security check-point of the Juzbado Plant almost simultaneouslywith the reception of the notification and ready to produce positive identification. The inspectors willbe admitted at the facility as soon as possible, but not later than two hours from the notification time.When managing the access clearance for the inspection team, the plant’s personnel checks that all theinspectors are included in the list of designated inspectors, so that their access is granted only if theyare in the list and hold a valid identification document.Then, ENUSA produces the electronic file with the physical inventory listing (EURAPII) and hands itto the inspectors at the beginning of the SNRI. Should the IAEA being the Party triggering theinspection the file is kept under seal until the EC inspectors arrive. Besides of that, the inspection teamis provided upon arrival with a hard copy of the updated itemized list of all fuel assemblies at the finalstage of the production process. The information contained in such listing has been previously agreedand let the inspectors verify on spot the physical inventory of finished fuel assemblies at the facility.

4. SNRI verification activitiesThe maximum duration of a SNRI is three working days, its activities being performed during normalJuzbado Plant’s working hours (from 8:00 to 17:00). The second inspection day will start in amaximum of 72 hours after the arrival of the inspectors of the Party triggering the SNRI.Should the EC be the Party triggering the inspection the order of the activities described in epigraphs4.1 & 4.2 will be swapped in order to give the IAEA team the chance to carry out partial defectsverification (PDV) measurements on fuel assemblies in all interim inspections (see below). In thisparticular case, the EURAPII file will not be sealed.4.1 Scope of the verification activities on the first inspection dayUpon arrival, the inspectors are immediately led to the production areas to carry out the followingactivities:• To establish the fuel assembly inventory, by 100% item counting and comparison with theprovided itemized list and any other relevant documents, and by gross defects verification (GDV).• To randomly select two fuel assemblies for PDV. In case that no new fuel assemblies wereproduced since the last SNRI, just one the items present in the store will be selected for PDV.The activities described above are performed only on fuel assemblies and loaded shipping containersphysically present at the final stages of inspection, storing or packing and assuming there is neitherneed to move the containers with a crane nor to open them. Moreover, the containers that have beenalready enclosed within the maritime transport container or loaded onto the transport platform will notbe subjected to verification activities.Once the verification on the inventory of fuel assemblies is finished, the inspectors are conducted tothe pellet loading stations to select one pellet sample which is kept under seal for analysis (DA) duringthe next PIV.When all these activities have been conducted, it is understood that the first day of inspection is over.4.2 Scope of the verification activities on the following inspection daysThe second day of inspection is started once the inspection teams from both Parties arrive at thefacility. ENUSA personnel checks that this requirement is complied before granting the inspectors theaccess to the plant. Any event regarding the access of inspectors should be notified to the Spanishcompetent authority as soon as possible.ENUSA produces the electronic file with the Inventory Changes Report (ICR) and hands it to the ECinspectors along with the sealed EURAPII file generated on the first inspection day. If needed, asecond EURAPII file updated to the second inspection day is also produced.The activities to be carried out by the inspectors during the second part of the SNRI are the following:• To generate the sampling plan from the information included in the EURAPII file.• To select the items containing received powder which were identified in the sampling plan forPDV. Samples of these items may be taken where appropriate and kept under seal to be analyzedduring the next PIV.• To select the rods identified in the sampling plan for both GDV & PDV.• Book examination of the inventory changes which have occurred since the previous inspection.• On top of the pellet sample taken during the first day and only in the event of foreign receipt ofpellets, these will be verified by DA sampling which will be kept under seal until the next PIV.

• Finally, in case of shipment of material other than fuel assemblies, verification or sealing activitieswill be agreed on a case by case basis.All the activities described above will be performed in a maximum of two working days.5. Information to be provided to the PartiesThe inspection procedure above described is established on the basis that ENUSA provides the Partieswith timely and accurate information according to agreed formats. Thus, the Parties have theinformation on the productive activities of the Juzbado Plant needed to fulfil their objectives. Suchinformation is detailed in the following paragraphs:5.1 Juzbado calendarBy December, 20 th each year, ENUSA will submit to the Parties the preliminary version of theJuzbado calendar for the following year. This information will be updated as soon as the calendarbecomes official. In case that force majeure events which could affect or otherwise limit the inspectionactivities in a given day, ENUSA will notify the Parties as soon as the event is known.5.2 Fuel assemblies production forecastBy December, 20 th each calendar year, ENUSA will provide the Parties with the fuel assembliesproduction forecast corresponding to the following year, according to a previously specified format.5.3 Powder receipts forecastBefore January, 31 st each calendar year, ENUSA will submit to the Parties the forecast with thepowder receptions for the whole year. Again, this information is provided according to agreed formats.The forecast cited in §5.2 and §5.3 will be updated weekly if any change occurs or ratified monthly,providing there is no need to include any modification. Should the scope of the revision be importantenough, the update will be submitted as soon as the changes are known. All the information will betransmitted in electronic format by encrypted mail with copy to the Spanish authority.6. ConclusionsThis procedure is still in a 6-months trial phase which started on March 1 st , 2007. Thus, it is stillpremature to drawn conclusions on it, both from the point of view of the achievement of the IAEA andEC objectives, and from the operator’s standpoint. Nevertheless, it must be said that by the date onwhich this paper was written, two SNRI have taken place at the Juzbado Plant with satisfactory resultsfor all the parties involved (IAEA, EC, Spanish authority and ENUSA). Some minor issues regardingthe exchange of information were identified, as well as it was the need to improve the interfacebetween IAEA and EC for triggering the inspections. These issues will be discussed after finishing thetrial period along with any other which could be identified in the near future. The outcome of thesediscussions will be taken into account for the revision of the procedure, as well as the new modalitiesof co-operation between IAEA and EC being discussed at high level.7. Definitions• SNRI. Short Notice Random Inspection.• PARTY. Either the International Atomic Energy Agency (IAEA) or the European CommissionDirectorate General for Energy and Transport (EC DG-TREN).• GDV. Gross Defects Verification. It stands for assays to roughly determine whether there is orthere is not nuclear material in a given item.

• PDV. Partial Defects Verification. It stands for assays to determine in detail the safeguards relatedrelevant characteristics of a given item.• DA. Destructive Analysis. It stands for the verification of the safeguards related relevantcharacteristics of the material using destructive techniques. It is usually performed during theannual PIV. On the opposite, verification activities during interim inspection are carried out bymeans of NDA (Non-Destructive Analyses) techniques. These are also used during the PIV.• PIV. Physical Inventory Verification. This is a very specific inspection to verify the physicalinventory carried out by the Juzbado Plant once a year. It usually lasts one week and its scope iscompletely out of the SNRI scenario.• Force Majeure. Event beyond the control of the operator that prohibits or otherwise limits theinspection activities that could be carried out on a particular day.

ADVANCED SAFEGUARDS FOR THEGLOBAL NUCLEAR ENERGY PARTNERSHIP (GNEP)S.F. DEMUTH, E.T. DIXON, K.E. THOMAS, R.K. WALLACELos Alamos National LaboratoryP.O. Box 1663, Los Alamos, NM 87545 USAABSTRACTWorld energy demand and greenhouse gases are expected to significantly increase in the nearfuture. Key developing countries have identified nuclear power as a major future energy source.Consequently, the United States and other countries are currently exploring the concept of aGlobal Nuclear Energy Partnership (GNEP) to address the concerns of nuclear proliferation.Advanced safeguards will be based on new world standards for the prevention of nuclear materialsproliferation. Safeguarding nuclear facilities includes inventory accountancy, process monitoring,and containment and surveillance. An effort has been undertaken to prioritize technologydevelopment for advanced safeguards accountancy by way of using the Standard Error in theInventory Difference (SEID) as a basis for cost/benefit analyses. By performing cost/benefitstudies, technology development R&D efforts can be prioritized.1. IntroductionAs part of the Global Nuclear Energy Partnership (GNEP) the United States has begun to design areprocessing and fuel fabrication research and development (R&D) facility to support spent fueltransmutation and power production by way of a fast reactor. The reprocessing and fuel fabricationR&D facility is referred to as the Advanced Fuel Cycle Facility (AFCF) and the reactor is the AdvancedBurner Reactor (ABR). The closed fuel cycle supported by the AFCF and ABR will serve two primarypurposes (1) reduce the underground waste repository size and engineered barrier requirements, and (2)recycle more proliferation resistant fuel than the existing plutonium mixed oxide (MOX) fuel cycle.Figure 1 represents the AFCF/ABR closed fuel cycle.Existing LWRSpent FuelAFCFWaste DisposalSeparation ofWasteFabrication ofTransmutation FuelShort-LivedWasteLong-LivedWasteAdvanced BurnerReactorIncreasedProliferation-ResistantFuelFig 1: The AFCF and ABR closed fuel cycle

Principal safeguards design components used for nuclear fuel reprocessing and fuel fabrication includeinventory accountancy, process monitoring, and containment and surveillance [1]. It is the accountancycomponent of the safeguards design that this paper is concerned with. Accountancy requirements fornuclear facilities are usually related to diversion detection of a significant quantity of nuclear material.When the statistics of measurement uncertainty are considered, such as detection of a significant quantityof material at high confidence, accountancy goals for large throughput facilities become nearlyimpossible to achieve. For this reason a cost/benefit methodology for prioritizing advanced safeguardsR&D, based on modelling of the Standard Error in the Inventory Difference (SEID), Sigma-MUF (σ MUF )or Sigma-ID (σ ID ), is proposed and discussed here [2]. Examples of advanced safeguards fortransmutation fuels includes (1) destructive assay (DA) of Pu in the presence of significant actinides suchas Np, Cm and Am, (2) accountability of Np, Cm and Cm, and (3) automation of DA to reduce labourcost.2. MethodologyExtend TM simulation software was used to model UREX reprocessing for demonstration of thecost/benefit methodology [3]. An 800 MTHM/yr facility was used for the baseline simulation withmeasurement uncertainties based on nominal values for a PUREX type process, as reported by ESARDA[4]. Both interim and annual inventory difference (ID) measurement uncertainties were evaluated. Theindividual measurement uncertainty is expressed as sigma (σ), which is the square-root of the variancefor systematic error, and the square-root of the variance divided by the degrees-of-freedom (i.e. numberof measurements) for the random error. The following equation represents an individual measurementuncertainty, based on systematic and random error, for concentration and mass. Measurementuncertainty for the entire process is then represented by Sigma-ID, which is equivalent to the SEID orSigma-MUF, depending on the regulating or monitoring agency of interest [5].σ2 2( σ + )r,cσ2 2 2r,mtotal ≈ σs,c+ σs,m+nThe nominal measurement uncertainties are listed in Table 1 and represent one standard deviation. Arepresentation of the Extend TM based model is shown in Figure 2. Two separate simulations wereconducted to demonstrate the cost/benefit methodology. The first simulation relates to the interiminventory difference, which is based on plant operation rather than shutdown, and is dependent on theprocess inventories. The second simulation relates to the annual inventory difference, is based on plantshutdown and therefore does not depend on process inventory, but rather on feed, product and wasteaccountability. For international monitoring the interim inventory period is dictated by the IAEA goal ofdetecting a significant quantity of plutonium with high confidence. An increase in inventory period willproduce a cost reduction. The benefit sought for the annual inventory is not dependent on a goal, butrather any reduction in the overall ID measurement uncertainty.σ s,c (%) σ s,m (%) σ r,c (%) σ r,m (%)Feed 0.2 0.2 0.3 0.3Product 0.2 0.05 0.2 0.05Waste 0.2 0.2 0.3 0.45In-Process tanks n/a n/a 0.3 0.45In-Process solvent extraction n/a n/a 0.6 1.0In-Process conversion n/a n/a 0.6 1.0Tab 1: Systematic and random measurement uncertainties for a PUREX type process2

ev entReadOutReadOutReadOutcountkg-Cs feedkg-U f eedkgPu in Tc wsteReadOutReadOutReadOutkg-Tc feedkg-Am feedkgPu in U wsteReadOutReadOutReadOut40SpentFuelMake-Upkg-Cm feedReadOutkg-Sr f eedReadOutkgPu in productReadOutwks/y rkg-Fps f eedkg-Lns f eedkgPu in HLW800MTHM/y rReadOutkg-Np feedReadOutkg-Pu feedReadOutkgPu in CsSr5kg-Pu/can50s/batchGenerateSpentFuelReadOutCCountr #Batches PuSeparationofPuUREXProduct& WasteMake-UpProduct& WasteMeasureReadOut# measurementsReadOutkg/batch feedReadOutmeasurements/dEqnReadOutMTHM f eed1 20000Co Variance L/tankFeed & Product28Yes=0No=1 day s holdup4InitialFeedMeasureProcessInventoryMeasuremeasurements/dReadOut# measurementsReadOutHoldUp (kg)ReadOutSEIDCalcReadOutSigma ID % FeedSigma ID% of FeedReadOutSigma ID kg-PuSigma ID kg-Pu0.1NRC limit %2.4IAEA limit kg# TanksTime (mo)3. Results3.1 Interim IDFig 2: Top layer of Extend TM based UREX model for Sigma-ID calculationDue to the short interval period, plant shutdown is not practical and therefore the interim inventory ismeasured during plant operation. For this study, it is assumed approximately 90% of the processinventory exists in four tanks, which is typical for an 800 MTHM/yr plant. Additionally, it is assumedthe concentration measurement uncertainty for the four process inventory tanks is similar to the primaryaccountancy tank as defined by ESARDA, but the mass measurement uncertainty is 50% greater.Typically, primary accountancy tanks are design for more accurate mass measurement than processinventory tanks, were liquid mass is based on level and density. Advanced safeguards evaluated are (1)reducing the concentration measurement uncertainty for process inventory tanks by 1/3 and (2) reducingthe mass measurement uncertainty for process inventory tanks by 1/3. The simulation based on nominaluncertainties is shown in Figure 3, and the results of the measurement uncertainty changes are shown inTable 2.Value54.625Plotter, Discrete EventSigma-ID (kg-Pu)4.253.8753.53.1252.752.375IAEA Goal20 2.5 5 7.5 10 12.5 15 17.5 20 22.5 25 27.5 30TimeResult DivideValue ResultDays Between Inventory Difference MeasurementFig 3: SEID/Sigma-ID for Interim measurement, baseline measurement uncertainties3

3.2 Annual IDσ r,c (%) σ r,m (%) Days to IAEA GoalBaseline 0.3 0.45 8Reduced mass error 0.3 0.3 13Reduced concentration error 0.2 0.45 11Tab 2: Effect of measurement uncertainty upon interim inventory periodThe annual inventory difference is measured during plant shutdown. For these conditions, themeasurement uncertainty is based on the cumulative error for the feed accountability tank, product cansand waste forms. Sigma-ID for the nominal uncertainty case is shown in Figure 4.Value5043.75Plotter, Discrete EventSigma-ID (kg-Pu)37.531.252518.7512.56.2500 1 2 3 4 5 6 7 8 9 10 11 12TimeResult DivideValue ResultMonths Between Inventory Difference MeasurementFig 4: SEID/Sigma-ID for annual measurement, baseline measurement uncertaintiesThe advanced safeguards evaluated for annual inventory are (1) automated destructive assay (DA) in thelaboratory and (2) automated DA in the process cell. For case 1, the random error is reduced by 25% dueto automation of analysis, and the systematic is not changed. For case 2, the random error is reduced by50% due to automation of analysis and sampling, and the systematic is increased by 25% due toincreased calibration difficulty. The results of the measurement uncertainty changes are shown in Table3.σ s,c (%) σ s,m (%) σ r,c (%) σ r,m (%) σ ID (kg-Pu)BaselineFeed 0.2 0.2 0.3 0.3Product 0.2 0.05 0.2 0.0545Automate in Feed 0.2 0.2 0.225 0.3Laboratory Product 0.2 0.05 0.15 0.0544Automate in Feed 0.25 0.2 0.15 0.3Process Cell Product 0.25 0.05 0.1 0.05534. ConclusionsTab 3: Effect of measurement uncertainty upon annual inventory periodThe results of Section 3 can be used to form the basis of a cost/benefit study. Cost changes include (1)the reduced number of interim inventories required (2) decreased labour upon automation, (3) R&Drequired for advanced instrumentation and (4) implementation of advanced instrumentation. This studyis intended to demonstrate only the cost/benefit methodology; whereas, actual costs savings for specificinstrumentation is left as a follow-on activity. Quantification of the cost savings for extended interiminventory measurement is more straight-forward than reduced Sigma-ID for the annual inventory.4

However, it may be possible to rank new technologies by way of percent reduction in Sigma-ID versusadditional cost. Future efforts will need to estimate measurement uncertainties for advanced processessuch as UREX, rather than use known PUREX values.5. Nomenclaturen = number of measurements during the inventorySEID = cumulative measurement uncertainty for inventory difference at one standard deviationused by the United States Department of Energyσ ID = cumulative measurement uncertainty for inventory difference at one standard deviationused by the United States Nuclear Regulatory Agencyσ MUF = cumulative measurement uncertainty for inventory difference at one standard deviationused by the IAEAσ s,c = systematic measurement uncertainty for the concentration at one standard deviationσ s,m = systematic measurement uncertainty for the mass at one standard deviationσ r,c = random measurement uncertainty for the concentration at one standard deviation= random measurement uncertainty for the mass at one standard deviationσ r,m6. References1 S.F. DeMuth, K.E. Thomas, I. Thomas and M. Ehinger , Safeguards by Design for theAdvanced Fuel Cycle Facility (AFCF), Los Alamos National Laboratory, Los Alamos, NM,87545, USA, LA-UR-07-3488, June 2007.2 C.T. Olinger, T. Burr, P.G. Dawson, L.K. Kwei and V.L. Longmire, Realizing Benefits fromTechnology for Inventory Monitoring, Los Alamos National Laboratory, Los Alamos, NM,87545, USA, LA-UR-02-3207, 2002.3 T.S. Rudisill, M.C. Thompson, M.A. Norato, G.F. Kessinger, R.A. Pierce, and J.D. Johnson,Demonstration of the UREX Solvent Extraction Process with Dresden Reactor Fuel,Westinghouse Savannah River Company, Aiken, SC 29808, WSRC-MS-2003-00089, Rev.1, 2003. http://sti.srs.gov/fulltext/ms2003089r1/ms2003089r1.html4 European Safeguards Research and Development Association (ESARDA), Bulletin No. 31,April 2002.5 K.E. Thomas and V.L. Longmire, IAEA Safeguards Approach for Reprocessing Facilities, LosAlamos National Laboratory, Los Alamos, NM, 87545, USA, LA-UR-02-5276, August 2002.5

Workshop 3:ENS YG workshop –Challenges andopportunities for Nuclear Professionals

Workshop 4:The Politics of Sustainable Development

Poster Presentations

Poster PresentationsSession I:New reactor and energy technology

SPES-3 EXPERIMENTAL FACILITY DESIGN FOR IRISREACTOR INTEGRAL TESTINGM.D. Carelli, B. Petrovic, M. Dzodzo, L. Oriani, L. ConwayWestinghouse, Science & Technology Center1344 Beulah Road, Pittsburgh, PA 15235 – USAG. Cattadori, A. Achilli, R. FerriSIETv.Nino Bixio 27, 29100 Piacenza – ITALYF. Bianchi, S. MontiENEAv.Martiti di Monte Sole 4, 40129 Bologna – ITALYF. BerraAnsaldo-Camozziv.le Sarca 336, 20126 Milano – ITALYM.E. Ricotti, L. SantiniPolitecnico di Milano, Department of Nuclear Engineeringv.Ponzio 34/3, 20133 Milano – ITALYD. GrgicFER, University of ZagrebUnska 3, 10000 Zagreb – CROATIAG.L. YoderORNLP.O. Box 2008, Oak Ridge, TN 37831 – USAABSTRACTThe IRIS integral reactor is on the path towards licensing. The paper deals with the study,design and set up of the experimental facility for the performance of the Integral Effect Tests(IET). IRIS owns specific layout, i.e. the integral configuration, and enhanced safety features,obtained also by exploiting a coupled dynamic behaviour during the accident sequences amongthe primary vessel and its components, the steel spherical containment and the passive safetysystems. These design choices, very positive for the safety point of view, however require thatthe scaled simulator of the reactor must be able to simulate the dynamic behaviour of all thesystems, containment included. The PIRT (Phenomena Identification and Ranking Table)analysis and the H2TS (Hierarchical, Two-Tired Scaling Analysis) process were adopted,leading to a 1:1 scaling in height, 1:100 in volume and power and with the fluid in prototypicalconditions. A brief description of the facility is reported and a preliminary Text Matrix ispresented as well.1

1. IntroductionWith the resurgence of nuclear power there is an increasing need for a range of new reactor designs,including smaller units of several hundred MWe. The International Reactor Innovative and Secure (IRIS)offers an advanced, modular 335 MWe design. IRIS features an integral primary system configurationwith all main components located within the reactor vessel. This configuration enables a simplified designwith enhanced reliability and economics and supports its safety-by-design approach, which results inexceptional safety characteristics. IRIS is being developed by an international team, led by Westinghouse,incorporating 19 organizations from 10 countries, about half of them European. IRIS development startedin 1999 and has reached the level of maturity indicating potential for being commercially offered by themid of next decade. Economic analyses indicate that IRIS will be competitive with other nuclear and nonnuclearenergy sources, whether deployed gradually in single units in smaller grids, or in multiple twinunits for larger grids. Additionally, IRIS fits well the recently announced US DOE initiative, GNEP(Global Nuclear Energy Partnership) aiming to support worldwide expansion of the use of nuclear energyin a responsible and proliferation resistant manner. Within the GNEP framework, IRIS can in the nearterm offer an advanced reactor design to satisfy needs for smaller, grid-appropriate reactors.On the R&D path towards licensing, currently the preliminary design has been completed and the testingneeded for design certification has started. The centrepiece of the experimental program is the integralsystem performance testing to be performed at the SIET facility in Italy. The pre-application reviewprocess with the US NRC was initiated in 2002 to address long-lead items, and enable obtaining the FinalDesign Approval (FDA) by 2013.The SPES-3 integral test facility of the IRIS reactor project is a key point in the development of theintegral reactors concept and mainly for the IRIS licensing phase. Due to the reactor configuration, thescaling choices and the simulation constraints, SPES-3 is going to represent a unique facility in its kind.The large scaling factor and the need to simulate both the primary reactor vessel, the containment andtheir coupled dynamic behaviour, represent a significant challenge for the designers.Fig.1 reports the general scheme of the IRIS integral layout, the safety systems and the functional linksamong the reactor pressure vessel, its internals, the safety systems and the containment. Since a positivecoupling effect among the components is exploited in the safety strategy [1], all the relevant coupledphenomena have to be simulated in the facility. The scaling process and the resulting features of theSPES-3 facility are depicted in the following paragraphs.2. Scaling approach and requirements for the SPES-3 facilityScaling of the SPES-3 facility is a part of the Evaluation Model Development and Assessment Process(EMDAP) [2]. Based on the specified figures of merit, and identified and ranked phenomena andprocesses [3], which is EMDAP Element 1 (Establish Requirements for Evaluation Model Capability), thescaling analysis and identification of similarity criteria needs to be performed as a part of EMDAPElement 2 (Develop Assessment Base). The initial scaling analysis was based on Hierarchical, Two-TiredScaling Analysis (H2TS) [4, 5]. The first two Stages of H2TS analysis (IRIS System Decomposition andScale Identification) were performed [6] and followed with Sage 3 – Top-Down System Scaling Analysisand Stage 4 – Bottom-Up Process Scaling Analysis. These last two steps need to be performed iterativelyand simultaneously with the design of the test facility. In the meantime the Fractional Scaling Analysis(FSA) [7, 8, 9, 10] was applied, as well. Both, H2TS and FSA use concepts from the hierarchical theorypresented by [11], and the concept of time-scale modeling [12, 13]. However, introduction of the effectiveFractional Rate of Change (FRC) in FSA provides the proper time constant for scaling a time-dependentevolution process in an aggregate (assembled of several interacting modules) and makes it moreappropriate for scaling Integral Test Effects [8].2

Fig.1: Conceptual scheme of the IRIS pressure vessel,internals, safety systems and containment.In the case of SPES-3 test facility the overall accepted volumetric scaling factor is 1:100 and the heightscaling factor is 1:1. The fluid is water at prototypical pressure and temperature conditions.The advantages of the accepted scaling approach are summarized in the following points.Full Height of the test facility provides:• Prototypical distance between heat sources and heat sinks centers to properly simulate naturalconvection effects.• Both, single phase and two phase natural convection loops can be simulated simultaneously.• Prototype and model fluid velocities and residence times in the loops can be adjusted to be thesame.• Horizontal inter-phase areas (transfer area concentrations) are properly scaled.Prototypical Pressure and Temperature eliminates:• Distortions due to the different fluid properties are not present (scaling analysis does not generateadditional terms related to property distortions and interpretation of the results is easier).However, the overall volumetric scaling factor 1:100 and full height bring some disadvantages:• Resulting in 10 times larger Transfer Area Concentrations for heat transfer (energy exchange) andwall friction (momentum exchange) at vertical side walls.• Some components (like heat exchangers) might be represented with limited number of tubes (notadequate to address side/bundle effects).Transfer Area Concentration for Heat Transfer through the Side Walls might be adjusted passively (byapplying an insulation material), or actively (by applying electrical heaters). The design of the test facilityheat exchangers needs to address side/bundle effects (see next paragraph).3. Main features of the SPES-3 facilityThe SIET labsThe SIET company operates since 1983 as a Centre for Studies and Experiences with the primary purposeof carrying out safety tests on components and systems for nuclear power plants. Due to the “experimental3

structure” of very high technology content, SIET is able to simulate the thermal-hydraulic loops of bothexisting nuclear power stations and new generation plants, at prototypical fluid thermodynamicconditions. In the past, SIET has qualified components and systems for customers like ENEA, ENEL,Ansaldo, General Electric, Mitsubishi, Doosan, Toshiba and Westinghouse. As for the Westinghouse AP-600, for which SIET provided the experimental results for the qualification and licensing process on theSPES-2 facility, in the frame of the IRIS program, it is going to build SPES-3 simulating IRIS in all itsprimary, secondary and containment components suitable to verify the effectiveness of interaction amongthem and qualify the system.The SPES-3 IRIS facilityThe SPES-3 is an integral test facility modelling the IRIS reactor, in particular:- the primary circuit including the RPV with power channel and fuel box, lower riser and RCCA, upperriser, pressurizer, upper downcomer, steam generators, riser-to-downcomer connection check valves,lower downcomer, lower plenum, circulation pump;- the secondary circuit up to the Main Isolation Valves, including Steam Generators (SG), Feedwaterlines and Steam Lines;- the safety system including the Emergency Boration Tanks (EBT), the Emergency Heat RemovalSystem Heat Exchangers (EHRS_HE) located in the Refuelling Water Storage Tank (RWST) and theAutomatic Depressurization System (ADS);- the containment system including the Dry Well, the Quench Tank (QT), the Pressure SuppressionSystem (PSS), the Reactor cavity and DVI room, the Long Term Gravity Make-up System (LGMS),and the DVI line.Fig.2 shows the resulting general layout of the SPES-3 facility.The primary circuitThe RPV is a cylindrical tank of about 22 m height and 0.65 m diameter with the all main internalhydraulic paths reproduced.The fuel bundle consists of 235 rods powered by indirect heating plus 1 dummy rod, assembled in astandard Westinghouse 17x17 fuel assembly configuration. The maximum available power is 6.5 MWwhile the full scaled power should be 10 MW. This affects only the steady state and early phases of thetransients and is considered acceptable from the system general behaviour investigation with a correctflowrate scaling in those phases.The pressurizer has an inverted hat shape and maintains the required pressure by means of a verticalelectrical heater.The SG zone consists of three annular sections suitable to locate the helical tube rows of the SGs: a singlerow in the inner and intermediate annulus, a double row in the outer annulus. An outer pump, injectingdirectly on each SG annular section, provides the required flowrates.The secondary circuitEight IRIS SGs are simulated with 4 helical tube rows, each consisting of 14 tubes wrapped concentricallyaround the cylindrical riser. The global height of the SGs is 8.2 m with an average coil length of 32 m. Inorder to keep the same thermalhydraulic behavior for each steam generator and to ensure a prototypicalfluid dynamic behavior on the primary side, different design solutions have been identified and comparedby means of CFD analysis.As an example, Fig. 3 reports the primary flow velocity field between two helical coil tube rows,simulated as a periodic bundle in the vertical direction, for two different solutions: the two bundles arebended in a parallel way or in a crossing way. The results show a better homogeneity for the crossing way,while the parallel way offers better solution for the manufacturing and the test section instrumentation.4

Fig 2: Layout of the SPES-3 integral facility.Fig 3: Velocity field of the primary flow between two helical coil tuberows (steam generator simulators): crossing (left side) and parallel (rightside) rows layout.(Red area: space between helical coil tubes; blue area: empty space incoil bundle helix; circles: holes in the tube bundle supports)5

The safety systemsThe two EBTs are cylindrical tanks connected to the RPV and DVI lines.The three EHRS_HEs consist of 3 and 5 vertical tubes, connected to cylindrical headers, suitable tosimulate the heat exchangers belonging to a double or quadruple loop. The total height is about 3 m. Theyare located in the same RWST simulator of 12 m 3 volume and 9 m height.The three IRIS ADS trains are simulated with a single train in SPES-3 by installing proper orifices tosimulate the single or multiple intervention.The containmentThe IRIS containment compartments are simulated in SPES-3 by different tanks connected by pipe lines.Each tank shape is suitable to reproduce the same volume versus height trend of IRIS. As the pipe lines donot exists in IRIS, they are designed to limit their influence on the flow.4. Preliminary Test MatrixThe preliminary Test Matrix of the design base cases is presented in Table 1. The test types, the testinitiating events, purpose and additional comments are specified.For the Low and High Elevations SBLOCA-s several additional group of tests are under consideration,like: demonstration of the long term cooling with all, or only two ADS trains available, tests beyonddesign cases with no, or only one EHRS_HE active, and tests with split breaks (instead double-endedguillotine breaks). ADS break is to demonstrate SBLOCA response with the maximum steam space(volume in the pressurizer) involved.Test Type Test Initiating Event Purpose CommentsLow Elevation Double-ended guillotineSBLOCA DVI line breakHighElevationSBLOCAADS breakDouble-ended guillotineRHRS/CVCS line breakDouble-ended guillotineADS line breakEstablish P CV (t), ΔP CV-RV (t),P CPSS (t), LGMS injection initiation.Verify design basis case systemresponse and mixture levelAll safety systems OK except for a singlefailure on one ADS trainMaximum PZR steam space breakFeedwaterLine BreakSteam LineBreakDouble-ended guillotinefeedwater line breakinside containmentDouble-ended guillotinesteam line break insidecontainmentShow non-LOCA plant responsewith partial EHRS actuation (bydesign for all non-LOCA events).Show non-LOCA plant responsewith partial EHRS actuation.To determine if the SG Makeup tank isneeded.Safe shutdownsequenceLoss of all powerDemonstrate safe shutdownsequenceObserve primary coolant shrinkage, switchto primary coolant natural circulation,EHRS HX cool-down capabilityTable 1: Preliminary Test Matrix.6

The Feed Line and Steam Line breaks are to demonstrate non-LOCA plant response with partial EHRSactuation. Finally, the safe shutdown sequence is to demonstrate the plant response under the loss of allpower.5. ConclusionsThe Integral Effects Testing of an integral layout PWR represents a significant challenge on theexperimental side of the R&D effort. This is particularly true for the IRIS reactor, since its innovativelayout integrates the dynamic behavior not only of the components into the integral reactor pressure vesselbut also of the containment. The complexity of the item has been addressed by the IRIS R&D team andthe strategy and final layout of the facility have been identified. A significant effort will be devoted also toset up the facility instrumentation, to obtain a valuable set of data for the development and assessment ofthe evaluation models. A coupled code able to simulate both the containment and the primary andsecondary reactor systems will be developed and adopted.AcknowledgementsThe work is partially supported by USA DOE grant DE-FC07-ID14785 and by the ItalianMinistry of Economic Development (MSE), within the MSE-ENEA Plan Agreement-CERSEfunding.References[1] M.D. Carelli et al., “The Design and Safety Features of the IRIS Reactor,” Nucl. Eng. Design, 230, pp. 151-167(2004).[2] Regulatory Guide 1.203, “Transient and Accident Analysis Methods,” USNRC, December 2005.[3] WCAP-16318-P, "IRIS Small Break LOCA PIRT", WEC Internal Report.[4] Zuber N. “Appendix D: A Hierarchical, Two-Tired Scaling Analysis,” An Integrated Structure and ScalingMethodology for Severe Accident Technical Issue Resolution, U. S. Nuclear Regulatory Commission,Washington, D.C. 20555, NUREG/CR-5809, November 1991[5] Zuber N., et al., “An Integrated Structure and Scaling Methodology for Severe Accident Technical IssueResolution: Development of Methodology,” Nuclear Engineering and Design, 186, pp. 1-21, 1998[6] WCAP-16103-P, "IRIS Scaling Analysis, Part I” (Stage 1 – Decomposition & Stage 2 – Identification), WECInternal Report.[7] Zuber N., “The Effects of Complexity, of Simplicity and of Scaling in Thermal Hydraulics, NuclearEngineering and Design, 204, 1-27, 2001[8] Zuber N., Wulff W., Rohatgi U. S., Catton I., “Application of Fractional Scaling Analysis (FSA) to Loss ofCoolant Accidents (LOCA), Part 1: Methodology Development,” The 11th International Topical Meeting onNuclear Reactor Thermal-Hydraulics (NURETH-11), Paper 153, Popes’ Palace Conference Center Avignon,France, October 2-6, 2005[9] Wulff W., Zuber N., Rohatgi U. S., Catton I., “Application of Fractional Scaling Analysis (FSA) to Loss ofCoolant Accidents (LOCA), Part 2: System Level Scaling for System Depressurisation,” The 11th TopicalMeeting on Nuclear Reactor Thermal-Hydraulics (NURETH-11), Paper 111, Popes’ Palace Conference CenterAvignon, France, October 2-6, 2005[10] Catton I., Wulff W., Zuber N., Rohatgi U. S., “Application of Fractional Scaling Analysis (FSA) to Loss ofCoolant Accidents (LOCA), Part 3: Component Level Scaling for Peak Clad Temperature,” The 11th TopicalMeeting on Nuclear Reactor Thermal-Hydraulics (NURETH-11), Paper 055, Popes’ Palace Conference CenterAvignon, France, October 2-6, 2005[11] Mesarovic M. D., Macko D., and Takahara Y., “Theory of Hierarchical Multilevel Systems,” Academic Press,New York, 1970[12] Chow T. H., editor, “Time Scale Modeling of Dynamic Networks and Applications to Power Systems,” SpringerVerlag, New York, 1986[13] Kline S.T., Similitude and Approximation Theory, Springer, New York, 19867

ECONOMIC FEATURES OF SMALLER SIZE, INTEGRALREACTORSM.D. Carelli, B. Petrovic, C. MycoffWestinghouse, Science & Technology Center1344 Beulah Road, Pittsburgh, PA 15235 – USAG. Locatelli, M. Mancini, M.E. Ricotti, P. TruccoPolitecnico di Milano, Department of Nuclear Engineeringv.Ponzio 34/3, 20133 Milano – ITALYS. MontiENEAv.Martiti di Monte Sole 4, 40129 Bologna – ITALYK. MillerNexia Solutions Ltd - UKABSTRACTA misguided application of the economy of scale would label the small-medium size reactors asnot economically competitive with larger plants because of their allegedly higher capital cost(€/kWe). The historical trend of capital costs vs. plant size is estimated from literature, and areference exponent factor for the economy of scale law is obtained. Specific models are adoptedfor the estimation the various factors which, beside size, contribute in determiningdifferentiating the capital cost of smaller reactors with respect to large reactors.The results show that when all the factors are accounted the capital costs of small and largeplants installations are practically equivalent.Moreover, the historical increment in the O&M costs shows that the small size reactors are notdramatically affected.The IRIS reactor is used as the example of smaller reactors, but the analysis and conclusions areapplicable to the whole spectrum of small nuclear plants.1. IntroductionTo fulfil the growing energy needs of developing countries and emerging markets, smaller size reactorsare needed. This has been identified within the US DOE Global Nuclear Energy Partnership (GNEP)initiative as one of the key elements, “Grid-Appropriate Reactors”, needed to enable worldwide expansionof the peaceful use of nuclear power.Smaller size reactors (IAEA defines “small” those reactors with power

principle, the capital cost of a smaller size reactor is higher than for a large size reactor is simplistic andwrong.The awareness and realization of the economic potential of smaller reactors has grown significantly in thelast few years (even though some work is as old as 15 years like the seminal paper [1] by UKAEA, whichhas been the guide of the present work).In addition to individual studies, the IAEA has launched in 2006 a collaborative project to address thecompetitiveness of Small-Medium Reactors (SMRs). As part of the IRIS (International ReactorsInnovative and Secure) [2] development, Westinghouse had already initiated investigation of the economiccharacteristics of IRIS. A more comprehensive outlook at the various components which make up theeconomics of SMRs was then undertaken by Westinghouse and some of its IRIS team partners, as acontribution to the IAEA study.The general approach to smaller reactors economics and some preliminary results obtained byWestinghouse and the Polytechnic of Milan, Italy (POLIMI) are reported in this paper.2. Cost factors affecting SMRs vis-à-vis Larger nuclear plantsWhen evaluating the competitiveness of SMRs versus large reactors, the various individual factors can begrouped into two classes:• Factors which are either applicable to SMRs only or are critically affected by the difference indesign and approach brought in by the SMRs (SMR specific factors)• Factors which affect SMRs and large plants in a comparable way (common factors). Even for thecommon factors, a comparative quantitative evaluation might not be straightforward.The SMR specific and the common factors are listed in Tables 1. The list is not exhaustive and othersmight be considered. Presented here are the ones judged to have higher priority for a quantitativeevaluation; Six factors (identified by (*) in the Tables) have actually been addressed, as discussed inSection 4.SMR Specific FactorsCommon FactorsDesign Related Characteristics (*) Size (*)CompactnessModularizationCogenerationFactory FabricationMatch of Supply to Demand (*) Multiple Units at a Single Site (*)Reduction in Planning Margin Learning (*)Grid Stability Construction Time (*)Economy of ReplicationRequired Front End InvestmentBulk OrderingProgressive Construction/Operation of Multiple ModulesSerial Fabrication of ComponentsTab 1 – List of Specific and Common factors for a differential evaluation3. The life cycle cost breakdown for a nuclear power plantThe Life cycle cost for a nuclear power plant is conventionally subdivided in the following macrocategories: Capital Cost, Fuel cost, Operation and maintenance cost and Final or decommissioning costSince different studies use different drivers to allocate the cost, in the literature there are differentpercentage for breakdown cost (Table 2). There are studies (SPRU, University of Sussex) and NERA [3]that gives an indicative range representing the proportion for the accounts in the cost breakdown (Table 3).Therefore this paper is focused on two principal accounts: the Capital cost and Operation and MaintenanceCost.

Williams et Gallanti et U.S. Congress/ DOE/EIA,Miller, 2006 [4] Parozzi, 2006 [5] EIA, 1993 [6] 2005 [7]Capital Cost 48,70% 68% 62% 71,90%Fuel cost 23,25% 13% 12% 11,19%Operation and maintenance cost 27,22% 15% 26% 16,91%Final or decommissioning cost 0,84% 4% 0% 0%Tab 2 – Cost breakdown from different bibliographic sources4. Capital CostAccountRangeCapital Cost from 60% to 75%Fuel cost from 5% to 10%Operation and maintenance cost from 8% to 15%Final or decommissioning cost from 1% to 5%Tab 3 – Cost ranges according to the NERA studyThe ad hoc and common factors showed in table 1 do not by any means represent a complete list but theyare the one judged as most representative. An initial quantification of some of these factors has beenattempted. The SMR representative was the IRIS reactor, which is offered in single (335 MWe) or in twin(670 MWe) units. The large reactor used as reference was an hypothetical 1340 MWe PWR. The IRISreactor was used because of the obvious familiarity and interest of the authors, but the evaluationconducted here is fully applicable to SMRs in general.Six factors were evaluated: size; multiple units at a single site; learning; construction time; match ofsupply to demand; and, design related characteristics. The results are reported in Figure 1 and Table 4The first factor represents the economy of scale, assuming that the two plants are comparable in designand characteristics. The usual correlationn -1⎛ sizeOCC SMR = OCCSMR⎞LARGE x ⎜size ⎟ ⎝ LARGE ⎠was adopted with n = 0.62. All other things being equal, the overnight capital cost (OCC; $/KW) of theSMR would be 70% higher than the large pant. But all other things are not equal and other commonfactors will tend to reduce the SMR disadvantage.The multiple units factor was evaluated considering that there are fixed, un-repeatable costs only incurredfor the first unit and there are costs which are shared by the multiple units. The experience reported in theliterature for Korean and French units on the same site was factored in our evaluation. For the four versusone plant comparison, it was evaluated that a 14% savings exists for the multiple SMRs.The learning factor was evaluated from the various models reported in the literature [8] . It was found thatfor the four units’ case the cost reduction is between 8% and 10%. The 8% value was conservativelychosen.The next two effects, construction schedule and matching of supply to demand (or “timing”), wereevaluated together, assuming a construction schedule for the large plant and SMRs of five and three yearsrespectively, a discount rate of 5%/year and calculating the cumulative expenditures for the two cases. A6% savings was estimated for the shorter construction time SMRs capability of better following thedemand curve.The principal design related characteristics for IRIS are: elimination of the pressurizer, steam generatorsvessels, canned pump housings, all large piping, vessel head and bottom penetrations and seals;elimination of safety systems such as the high pressure emergency core cooling system due to the safetyby-designapproach which eliminates several postulated accidents; compact containment; lower amount ofcommodities. A conservative evaluation of these effects indicated a 17% cost savings.

When the various factors are combined, a single SMR of 335 MWe if deployed as part of a pack of fourhas a capital cost only 5% higher than the monolithic 1340 MWe reactor.Some sensitivity studies were also conducted, for example, to allow also the large plant to take advantageof multiple units on site and to investigate the effect of “worldwide” type learning. The reference casereported here and yielding a cumulative 1.05 factor considered four IRIS and one large plant on site, withno prior experience for either (i.e., no worldwide learning). A case of eight IRIS and two large plants onsite, still with no prior experience yielded a total factor of 1.16, reflecting the proportionally higher effectof two large units on site. On the other hand, a case of four IRIS and one large plant on site, but with aprior worldwide experience of 2680 MWe for both (which means two large plants and eight IRIS) yieldeda total factor of 1.0, reflecting the much larger learning deriving from the higher number of units.All the other sensitivity cases fell within the 1.0-1.16 range.$/KW (equivalent)2345MultipleUnitFactorsLearningCurveFactorsConstructScheduleFactorsUnitTimingFactors(1) ECONOMY OF SCALE - Assumes single unit and same design concept(large plant directly scaled down)1(2) MULTIPLE UNITS -Savings in cost for multiple small units atsame site (direct - parts and buildings shared; fixed - one timecharges; site-related costs)Economy of Scale Factor(3) LEARNING - Cost reduction due to learning (inconstruction, operation) for a series of units at a single site(4) CONSTRUCTION SCHEDULE – shorterconstruction time(5) TIMING - SMR enables gradual capacityincrease to fit energy demand growth(6) SPECIFIC DESIGN - Costreduction due to specific designconcept characteristics (e.g.,simplification)6PlantDesignFactorsPresent ValueCapital Cost“SMR Design”0 300 600 900 1,200 1,500Plant Capacity (MWe)Fig 1 Potential for Small Reactors Economic CompetitivenessFactor Individual SMR/Large 1 Cumulative SMR/Large(1) Economy of scale 1.7 1.7(2) Multiple-unit saving 0.86 1.46(3)Learning 0.92 1.34(4) (5) Construction schedule and timing 0.94 1.26(6) Design specific 0.83 1.05Tab 4 Quantification of Factors Evaluated in SMRs/Large Plant Comparison of Capital Costs (Figure 1)5. Operation and maintenance costAfter the Capital cost the most important account in the life cycle cost for a nuclear power plant is the“operation and maintenance cost”. Considering the table 3 seems that the operation and maintenance costare a small part of the total cost for a nuclear power plant. However it is important to notice that it1 SMR: One 335 MWe plant, as part of four units; Large: One single 1340 MWe plant

ecomes very important for an economic life cycle for a plant. In fact there are cases which nuclear powerplants have been closed due to a dramatic increase in the operation and maintenance cost [9] .The model used to quantify the O&M cost is based on an ORNL Technical Report [10] and identifies threemain cost categories: labour cost (on site and off site), material cost (shop supplies), and a third categoryincluding other minor cost items. The model assumes a reference cost for each cost category (estimatedfrom DOE information), that is adjusted by means of two coefficients. The first one considers theeconomy of scale effect whereas the second takes into account the number of units built in the same site.The estimation functions for both coefficients use an exponent less than one, due to the nonlinearcorrelation between the number of units, the size of reactors and O&M costs.According to the model is possible to conclude the a site with four SMR (335 MWe) has an O&M cost24% greater than a site with one 1340 MWe Large reactor, or likewise a site with three SMR has an O&Mcost 22% greater the a site with one 1005 MWe Large reactor.It is also important to notice that the model doesn’t consider the specific advantages coming from theSMR technology. A correct quantification of them will be able to reduce the gap.6. ConclusionsSmaller and larger reactors address different markets and there are many market related factors favouringone versus the other, independently from their capital cost.When however they are competing on the same market the capital cost is not a discriminator and the twotypes of nuclear plants are practically equivalent under this respect.The O&M costs increase more than the Capital Cost, but less than how is foreseen by a roughcomputation with the economy of scale canonical equation. They also represent a small part of the totalcost which is composed mainly by the capital cost.The so-called economy of scale is actually no longer an absolute advantage of larger reactors since itcould be compensated by a variety of other factors.This paper presents only the beginning of the evaluation of the competitiveness of SMRs and expanded,more detailed investigations will follow.7. References1. M. R. Hayns and J. Shepherd, “SIR-Reducing Size Can Reduce Cost,” Nucl. Energy, 1991, 30, No. 2,pp. 85- 93.2. M. D. Carelli and B. Petrovic, “Here’s Looking at IRIS,” Nuclear Engineering International, March2006, pp. 12-17.3. G. Mackerron, D. Colenutt, M. Spackman, A. Robinson, E. Linton, Paper 4: The economics ofnuclear power, report for the Sustainable Development Commission by Science & Technology PolicyResearch (SPRU, University of Sussex) and NERA Economic Consulting, March 20064. K. A. Williams, K. Miller, A User’s Manual for G4 - ECONS: A Generic EXCEL-based Model forComputation of the Projected Levelized Unit Electricity Cost (LUEC) from Generation IV ReactorSystems, Economic Modelling Working Group (EMWG) - May 20065. M. Gallanti, F. Parozzi “Valutazione dei costi di produzione dell’energia elettrica da nucleare”,Energia n. 3 , August 20066. U.S. Congress, office of technology assessment, Aging nuclear power plants: managing plant life anddecommissioning, , September 19937. Energy Information Administration (EIA), Electric Power Annual 2005, DOE/EIA-0348(2005),8. “Cost Estimating Guidelines for Generation IV Nuclear Energy Systems,” Rev. 3, November 30,2006, GIF/EMWG/2006/003, available at www.gen-4.org9. Energy Information Administration (EIA), An analysis of nuclear power plant operating costs: a1995 update, DOE/EIA, 199510. H.I. Bowers, L.C. Fuller, M.L Myers, Cost Estimating Relationships for Nuclear Power PlantOperation and Maintenance, report submitted to the DOE by ORNL, 1987

STRATEGY FOR UTILIZATION OF THE RESEARCH REACTORIRT IN SOFIA AFTER ITS REFURBISHMENTK. D. ILIEVA, T. G. APOSTOLOVInstitute for Nuclear Research and Nuclear Energy,Bulgarian Academy of SciencesTzarigradsko Shossee 72, 1784 Sofia, BulgariaABSTRACTThe reconstruction of the research reactor IRT-2000 in Sofia into a low-power reactor isbeing carried out from 2001. The future utilization of IRT aims to satisfy the society needsfor: development and preservation of nuclear science, skills, and knowledge;implementation of applied methods and research; education of students and training ofgraduated physicists and engineers in the field of nuclear science and nuclear energy;development of boron neutron capture therapy. It will support the keeping up specialistswith researcher’s approach and skills who are able to give adequate responses to thechallenges of complex modern technologies and the associated environmental problems. Itwill be used for production of isotopes needed for medical therapy and diagnostics; inelement activation analysis having a number of applications in industrial production,medicine, chemistry, criminology, etc. The reconstructed research reactor IRT will uselow-enriched uranium fuel IRT-4M, with uranium-235 enrichment below 20%, is inaccordance with the current norms on the security of transport and storage of nuclear andother radioactive materials which are vulnerable to theft by terrorists.The research reactor IRT-2000 (IRT) in Sofia to the Institute for Nuclear Research and NuclearEnergy (INRNE) was built and put into operation in 1961. It was temporarily shut down in 1989 forimprovement. The reconstruction of the IRT is being carried out under the decision of the Council ofMinisters of Republic of Bulgaria from 2001. The strategy for sustainable utilization considers the IRTas a national base and aims to satisfy the society needs for: education of students and training of graduated physicists and engineers in the field of nuclearscience and nuclear energy, implementation of applied methods and research, development and preservation of nuclear science, skills, and knowledge.The IRT Technical Design is being in process of elaboration. The IRT will be reconstructed into areactor:• of thermal power 200 kW;• with low enriched fuel, with uranium-235 enrichment below 20% in accordance with thecurrent requirements of the security of transportation and storage of nuclear and other radioactivematerials which are vulnerable to theft by terrorists;• with ten vertical and seven horizontal experimental channels which will supply maximal fastneutron flux about 3. 10 12 n/cm 2 s, and maximal thermal flux about 8.10 12 n/cm 2 s;• with channel which will supply epithermal neutron flux about 0,9.10 9 n/cm 2 s suitable formedical Boron Neutron Capture Therapy (BNCT) application.The MCNP 4C three-dimensional neutron transport code using the DLC-200 neutron cross-sectionslibrary has been applied for calculations of different variants of reactor core and BNCT filter. The coremodel used in MCNP 4C calculations is presented in Fig. 1.

MCNP Calculational Model of IRTFig.1. Core Model in MCNP 4C CalculationsThe INRNE together with the Technical University in Sofia have proposed to the Ministry ofEducation a new programme for education of students in nuclear energy. The Nuclear Energy coursewill be obligatory for obtaining the Master of Science Degree of the Technical University in Sofia.The educational classes refer: types of research reactors, main characteristics and design of thereconstructed IRT, safety assuring and licensing, reactor physics and thermo-hydraulic characteristicsdetermination, accident analyses, fresh and spent fuel management, radioactive waste managementand governmental categorization norms and rules. Acquaintance with calculational codes as theMCNP code for neutron transport and criticality calculations, WIMS-ANL code – for preparing ofneutron cross sections for diffusion calculation, REBUS code for the fuel burn depth calculation,SCALE code system for spent fuel transport and storage devices safety assessment, PLTEMP/ANLcode for calculation of thermo-hydraulic steady-state, and RELAP5 code – for transient operation, etc.is planned too. Preliminary acquaintance with the neutron activation analysis and BNCT is included inthe educational programme.The classes will be held in the INRNE and the reconstructed IRT will be used for carrying out specifictraining exercises on the reactor: reactor start, manual and automatically control, control rodcalibration, delayed neutron group measurements, sub-critical multiplication/shutdown marginmeasurements, excess reactivity and shutdown margin measurements; reactor-physics measurementsof static and kinetic reactor parameters, reactor dosimetry, measurements of the spent fuelcharacteristics in the hot laboratory, radiological characterization survey - alfa, beta and gammameasurement techniques, contamination measurement, etc.The reconstruction of the IRT includes an arragement for a BNCT facility. Preliminary neutrontransport calculations for BNCT channel regarding the geometry and material composition designhave been carried out (Fig. 2). Feasibility studies within the national network of the MedicalUniversity in Sofia, the National Centre of Radiobiology and Radiation Protection, the Institute ofExperimental Pathology and Parasitology and Institute of Electronics of the Bulgarian Academy ofSciences, and the Faculty of Physics of Sofia University are carried out. Contacts with institutes,experienced in BNCT as EC JRC, Petten, the Netherlands, VTT, Finland and NRI-Rez, the Czech

Republic, were established. Human, social and economical results due to the BNCT for patients fromBalkan region are expected.Besides the financial support of the Bulgarian government the IRT has the IAEA support through theproject BUL/4/014 “Refurbishment of the Research Reactor” and the support of the US Department ofEnergy in the frame of the RERTR program.The reconstructed IRT is a basis for keeping up specialists with researcher’s approach and skills whoare able to give adequate responses to the challenges of complex modern technologies and theassociated environmental problems. The reactor will be used for production of isotopes needed formedical therapy and diagnostics; it will be the neutron source in element activation analysis having anumber of applications in industrial production, medicine, chemistry, criminology, etc.Nuclear energy has a strategic place within the structure of the country’s energy system. A newnuclear power plant Belene with two reactors of 1000 MeV will be built. The extremely highrequirements regarding nuclear safety call for the availability of scientific and technical potential, andfor an adequate culture of safe use of nuclear energy. The acquired scientific experience andqualification in reactor operation is a basis for participation of the country in the internationalcooperation within the European structures. In that aspect, the operation and use of the IRT bringseconomic benefits for the country.Figure 2. The BNCT beamtube model:1. Vessel of Channel;2. Filter 3. Lead Shielding;4. Collimator; 5. LeadShielding of Channel; 6.Concrete.

DEVELOPMENT OF AN ITER RELEVANTINSPECTION ROBOTM. HOURY, L. GARGIULO, P. BAYETTI, V. BRUNO,J-J. CORDIER, Ch. GRISOLIA, J-C. HATCHRESSIAN,F. SAMAILLEAssociation Euratom-CEA, Département de Recherche sur la Fusion Contrôlée,CE-Cadarache F-13108, FranceJ-P. FRICONNEAU, D. KELLER, Y. PERROTCEA, LIST, Service de Robotique Interactive,18 route du Panorama, BP6, Fontenay aux Roses F-92265 FranceABSTRACTRobotic operations are one of the major maintenance challenges for ITER and future fusionreactors. In particular, in vessel inspection operations without loss of conditioning will bevery useful. In this context, an Articulated Inspection Arm (AIA) is currently beingdeveloped by CEA within the European work programme framework, which aims atdemonstrating the feasibility of a multi-purpose in-vessel Remote Handling inspectionsystem using a long reach, limited payload carrier (up to 10 kg). After qualification, the armwill constitute a promising tool for generic application. This paper deals with theintegration of the robot into Tore Supra and presents the associated processes for inspectiontasks.1 IntroductionThe aim of this R&D program is to demonstrate the feasibility of in-vessel Tokamak inspection tasksfor the future fusion reactor ITER [1]. An Articulated Inspection Arm (AIA) is being developed tosatisfy requirements in terms of maintenance and in-vessel component inspections. The meanspecifications of the project are: large operational range, ultra high vacuum and temperature ambianceat 120°C. Operations under magnetic field and nuclear ambiance are not yet considered in theimplemented technology for the AIA demonstrator.Preliminary tests of the AIA under relevant vacuum and temperature conditions are scheduled on theTORE SUPRA Tokamak at CEA/Cadarache by the end of 2007. The TORE SUPRA Tokamak isequipped with actively cooled components and operates with similar vacuum and temperatureconditions as ITER (120°C to 200°C). Integration of the AIA demonstrator, associated withdevelopment of inspection processes will demonstrate in-vessel operating capabilities (viewing, leakdetection on water loops, erosion or deposition characterization, abnormal events diagnostic…).2 The Articulated Inspection ArmThe AIA is an 8 meter long multi link carrier with 5 modules of 160 mm diameter. The length of theAIA robot is consistent with that required for ITER. The modules include pitch (±40° in the verticalplane) and yaw (±90° in the horizontal plane) joints linked with a parallelogram structure that keepyaw joints axis always vertical [2, 3, 4]. Combination of elevation and rotation motions gives to therobot 8 degrees of freedom. The total weight of the AIA is about 150 kg. The robot is moved along itssupport with a linear trolley named Deployer. The payload carrier is limited to 10 kg.

All electronic systems are embedded in each AIA module. These components are enclosed at theatmosphere pressure in tight boxes while the mechanical structure is under vacuum. Moreover,components shall sustain a temperature of 200°C during the conditioning phase and 120°C foroperations and in-vessel Tokamak deployments.A successful vacuum and temperature test campaign on a prototype module was performed in 2005 inCEA/DRFC test facilities. In particular, a baking phase at 200°C was performed during a couple ofdays and the final spectrum has shown a good component conditioning. To overcome pollution issueswith using grease, the design of free lubricant joints is based on thermal treatment with Teflon coating.An endurance testing was also performed at room conditions to qualify the 5 modules performancesunder representative loading (see Fig. 1).These encouraging first results have pre-qualified the selected technologies for the AIA project.Fig. 1: The AIA robot 5 modules assembled with a 10 kg load at its end (on the right); Assembling andtests performed in CEA/LIST laboratories.3 Integration on TORE SUPRA TokamakA scale one demonstration of the AIA under ITER requirements is planned on TORE SUPRA whichshould lead to significant improvement in R&D results for in-vessel fusion remote handlingequipments.A long storage cask was designed for conditioning of the robot (vacuum and temperature) and foraccurately guiding it during in-vessel Tokamak deployment. This stainless steel large structure (11mlong, 3m height and 5 tonnes) is carried by 2 rolling wagons. One integration objectives is to connector fold up the entire device in about 1 hour on a dedicated port of TORE SUPRA. For this purpose, thecask is equipped with a double valve that allows disconnection of the vessel without loss of thevacuum conditions. Moreover, all electro-technical equipments are embedded to realize a compact andan autonomous system (see Fig. 2).A first deployment of the AIA robot into Tore Supra vacuum-vessel in planned by the end of 2007(see Fig. 3). Robustness, reliability and flexibility will be tested and improved between successiveplasma operating campaigns [5].

Fig. 2: CAD views of the storage cask for conditioning and guiding the AIA robot.Fig. 3: Schematic view of AIA integration on Tore Supra Tokamak (upper view).4 Development of interchangeable processesThe AIA is designed to allow accurate displacements of the head in front of the Plasma FacingComponents. Several processes are in development to be implemented at the front end of the AIArobot. All these processes shall be interchangeable. This specificity will offer flexibility in theoperation tasks.The first developed process is a viewing system to make close visual inspection of Plasma FacingComponents. The video process is designed with a CCD sensor embedded in a tight box made ofstainless steel and glass (see Fig. 4). This box is linked to the head of the robot through a vertical jointactuated from inside with the same system as the yaw joint of the robot. All the electronic componentsinside this box are nitrogen-gas actively cooled by the means of a flexible umbilical. This system is

currently tested and will be integrated for the first AIA deployment inside the TORE SUPRATokamak by the end of 2007.Fig. 4: The visual inspection process.A process based on helium sniffer is being considered to improve and facilitate maintenanceoperations on water loop leak tests.Some analysis and operations on the Plasma Facing Components could be performed by laser systems[6]. Several characterizations and treatments are proposed for:1. The deposited layer depth on the Plasma Facing Components can be measured using arepetitive laser pulse in a heating regime (0.1 to 0.4 J/cm 2 , 100ns, 10 kHz, Nd-YAG laser).2. The removing of the deposited layer is possible with the same laser device in the ablationregime (0.4 J/cm 2 ). This technique can be considered for recovering the tritium trapped intoITER Plasma Facing Components.3. The composition of deposited layer can be estimated via Laser Induced BreakdownSpectroscopy (LIBS). Integration of 2 optical fibres could be added into the AIA, bothconnected respectively to laser source and spectrometer located outside of the Tokamak.5 ConclusionFuture deployment on TORE SUPRA of this multipurpose robotic device will give new perspectiveson in-vessel maintenances and operating activities for fusion reactor like ITER.Several processes are foreseen to be associated on the AIA robot carrier to inspect, diagnose or treatPlasma Facing Components.Preliminary tests with dedicated vision device under relevant vacuum and temperature conditions willbe performed in TORE SUPRA by the end of 2007. Other developments are under development for afurther integration beyond 2007.6 AcknowledgementsThis work, supported by European Communities under the contract of Association betweenEURATOM/CEA, was carried out within the framework of the European Fusion DevelopmentAgreement. This views and opinions expressed herein do not necessarily reflect those of the EuropeanCommission.

The authors would like to acknowledge the technical staff of CEA/DRFC/STEP/GARV for itsassistance.7 References[1] “ITER Final Design Report FDR 2001 and DDDs 2001” published by the IAEA under ref G A0FDR 4 01-07-21 R 0.4, also available on http://www.iter.org/reports.[2] Y. Perrot et al., Development of a long reach articulated manipulator for ITER in vessel inspectionunder vacuum and temperature, 22nd SOFT, 9-13 September 2002, Helsinki (Finland).[3] Y. Perrot et al., Scale One Field Test of a Long Reach Articulated Carrier for Inspection in SpentFuel Management Facilities, 10 th ANS, 28-31 March 2004, Florida (USA).[4] Y. Perrot et al., The articulated inspection arm for ITER, design of the demonstration in toresupra, 23rd SOFT, 20-24 September 2004, Venice (Italy).[5] L. Gargiulo et al., Towards operations on Tore Supra of an ITER relevant inspection robot andassociated processes, 24th SOFT, 11-15 September 2006, Warsaw (Poland).[6] Ch. Grisolia et al., Journal of Nuclear Materials, vol. 363-363, (2007) 1138.

CONCEPT OF A FUTUREHIGH PRESSURE - BOILING WATER REACTOR, HP-BWRF. REISCHDepartment of Nuclear Power Safety, KTH, Royal Institute of TechnologyAlba Nova University Center Roslagstullsbacken 21, SE-106 91, Stockholm – SwedenABSTRACTSome four hundred Boiling Water Reactors (BWR) and Pressurized Water Reactors (PWR)have been in operation for several decades. The presented concept, the High PressureBoiling Water Reactor (HP-BWR) make use of the operating experiences The best parts ofthe two reactor types are used and the troublesome components are left out. This meansimproved safety. The increased thermal efficiency is beneficial to the environment as lesswarm cooling water is released per produced kWh. With some modifications the presentlyused components can be used making this design cost effective and possible to realize in anot too distant future.1. IntroductionSince the 1950s several hundreds Boiling Water and Pressurized Water Reactors (BWRs and PWRs)are in use. There is a wealth of operating experiences. During the years many difficulties occurredwith a number of important components. This concept, the High Pressure – Boiling Water Reactor(HP-BWR) offers a solution to use the best parts from each type (BWR and PWR) and leave out thetroublesome components. This means an important increase of safety. As an extra benefit, also ahigher efficiency is attained beneficial for the environment as less warm cooling water is released perproduced kWh. The HP-BWR is using –with some modifications- presently manufactured partsmaking this a cost effective, realistic concept.2. The High Pressure – Boiling Water Reactor HP-BWRThe High Pressure Boiling Water Reactor (HP-BWR) is a promise for improved nuclear safety andincreased leniency to the environment. The HP-BWR is an environment friendly coast effectivealternative.

The HP-BWR is using a modified PWR reactor vessel and BWR type fuel and control rods. Howeverhere the cross formed control rods are gravity operated with ample space between the crosses and thefuel boxes. The control roads are manoeuvred electromagnetically which means that they will dropinto the core at a loss of electrical power as in the PWRs. The traditional PWR control rods are fingershaped and are surrounded by a tube with a minimum of clearance. The traditional BWR control rodsare operated from below with hydraulic pressure. Therefore in the bottom of the traditional BWRreactor vessel there are a great number of penetrations for the control rods. Directly below the reactorvessel there is an elaborate system of numerous high pressure hydraulic pipes to actuate the controlrods. Taking the best and leaving out the difficulties of both the traditional BWR and PWR systems isa substantial safety improvement.All the pipe connections to the reactor vessel are well above the reactor core. This allows the omissionof the core spray. The moisture separators and steam dryers are outside the reactor vessel, leaving freespace for the control rods.Internal circulation pumps. These allow using orifices at the inlet of the fuel boxes so that the onephase pressure drop will dominate over the two phase pressure drop. This reduces the risk forhydrodynamic oscillations. However if suitable methods are found to facilitate natural circulation eventhe circulation pumps can be left out.The use of the HP-BWR means improved Carnot cycle thermal efficiency up to about ~40% instead ofabout ~30%. The reason is that the HP-BWR steam temperature corresponds to 15MPa while thetraditional BWR’s steam temperature corresponds to 7MPa and the traditional PWR’s steamtemperature corresponds to 6MPa. The HP-BWR is lenient to the environment as less warm coolingwater is released per produced kWh to the recipient, sea or river or to the air via a cooling tower.Using direct cycle the system is simplified. Still, the usual PWR steam lines can be used through thecontainment wall to the turbine. A great advantage is that the complicated and costly steam generatorsare left out.The moisture separators and the steam dryers are outside the reactor vessel in the containment insteadof the huge troublesome steam generators.Simple dry containment is used instead of the complicated, inert, pressure suppression wetcontainment which requires a great deal of surveillance.3. The Traditional Boiling Water Reactor, BWRThe basic principles of the traditional Boiling Water reactor are well knownAs there are pipe connections to the reactor vessel below the reactor core, a pipe break can empty thevessel leaving the core uncovered, without the cooling water. Therefore a core spray is required. Thisis a common feature for the BWRs with external circulation pumps or jet pumps. However this drawback is eliminated at the later design at the Advanced Boiling Water Reactor, ABWR. All BWR

control rods are inserted to the core with hydraulic power, some with electric motors too. This makesthe lower part of the reactor both inside and outside the bottom of the reactor vessel extremelyelaborate. To make things worse, in the past, cracks, corrosion and leakage have occurred at thepenetrations at the lower part of the reactor vessel.The huge reactor vessel would require an enormous dry containment building; therefore, instead apressure suppression containment system is used. The containment is separated into two parts, theupper drywell and the lower wet well with the suppression pool. If the separation is not near perfectlyleak tight the wet well cannot fulfil its function to suppress the pressure in the drywell in case of a pipebreak. Further complication is that the traditional BWR containment operates inert, making theentrance into it more difficult.The nice thing about the BWR is that it operates in direct cycle mode without the troublesome steamgenerators4. The Traditional Pressurized Water Reactor, PWRMost of the World’s operating reactors are traditional PWRs.The control rods are operated from above. Undoubtedly some leakages were observed at thepenetrations which in a few cases led to the need to replace the reactor pressure vessel head.

.The simple electromagnetic devices which manoeuvre the rods have worked reliably. This assures ahigh degree of safety. A basically continues, uninterrupted bottom of the reactor vessel avoids causingsuspicions about its integrity.

The curse of the traditional PWRs is their steam generators. These complicated and costly huge piecesof equipment are disappointingly short-lived because of the corrosion of the internal tubes causingleakages. Many steam generators have been changed after some 15 years. This is an extremelyexpensive and troublesome and also time consuming operation.In the upper part of the steam generators there is the moisture separator and the steam dryer. The HP-BWR is “borrowing” this equipment which can be used without the troublesome steam generators.5. ReferencesAll university text books written for Nuclear Engineering students contain detailed descriptions ofboth Boiling Water Reactors and Pressurized Water Reactors. Also manufacturers in Europe, Asia andAmerica publish data about their particular designs. There is a wealth of information about BWRs andPWRs on the internet.

Poster PresentationsSession II:The nuclear Fuel cycle / Nuclear Operations

ENC 2007 Conference, September 16-20, 2007, Brussels, BelgiumA Study on the Determination of Disposal Priority for Low andIntermediate Level Radioactive Wastes (LILW) in KoreaMin Ho Ahn, Sang Chul Lee, Kun Jai LeeDepartment of Nuclear and Quantum EngineeringKorea Advanced Institute of Science and Technology (KAIST)373-1 Guseong-dong, Yuseong-gu, Daejeon 305-701, South Korea(Fax) +82-42-869-3810 (Tel) +82-42-869-3858daniel-holy@nuchen.kaist.ac.kr1. IntroductionIn Korea, the major LLW generator, called KHNP (Korea Hydro and Nuclear Power Co.)has generated about 67,000 drums (200L) of low and intermediate level radioactive waste(LILW) stored in the temporary storage facilities at each reactor sites since 1977. Especially, theamount of DAW (dry active waste) has been accumulated around 36,600 drums and this numberrepresents about 56% of total LILW generated by all NPP in Korea. The amount of evaporatedbottoms, concentrated wastes, spent ion exchange resins, and spent filters have beenaccumulated around 19,000 drums, 9,700 drums, and 1,600 drums and these numbers representabout 28%, 14%, and 2% in respectively.Figure 1. The amount of all wastes generated in Korea (As of Dec. 31, 2005)

ENC 2007 Conference, September 16-20, 2007, Brussels, BelgiumThese LILW drums have been stored in temporary storage facility of each NPP site and havebeen prepared to be disposed in the Final repository which will be operated in 2009.2. General Characteristics of LLW in Korea2.1. DAWDAW has included a variety of materials such as plastic, rubber, paper, metal and so on.The radioactivity level of DAW is turned out to be relatively low compared to other wastestreams. However, some of DAW may have possibility to contain free liquids and harmfulmaterials, including some wastes with high radioactive level. The detailed characterization ofDAW is important and required to first step prior to the disposal of the waste. The free liquids inDAW can be easily detected by NDA (non destructive analysis) method using the X-ray,neutron, microwave, etc. However, the harmful materials in DAW can not be readily checkedwhether those are in drum or not [1].2. 2. Evaporated Bottom (Concentrated Waste)In general, the evaporated bottoms and concentrated wastes are considered to behomogeneous waste that the activity of the waste can be easily estimated. Activity level ofevaporated bottoms and concentrated wastes is also relatively low compared to that of spent ionexchange resins and filters. Activity estimation of all evaporated bottoms and concentratedwastes by DTC (dose to curie) method shows that no drums exceed the value of the disposallimits, which are regulated by Korea regulation law. Some of evaporated bottoms andconcentrated wastes was solidified by paraffin wax and be relatively homogeneous. However,solidified wastes with paraffin can not be readily disposed because characterizationmethodology for those wastes has not been known yet in Korea.2.3. Spent Ion Exchange Resin and FilterFor spent ion exchange resins and filters, although the radioactivity level is relatively highcompared to other waste streams, the wastes from spent ion exchange resins and filters areshowing good homogeneity same as that of evaporated bottoms and concentrated wastes.Therefore spent ion resins and filters could be considered to have early disposal priority becauseof relatively easy and accurate determination of the contents of the waste.

ENC 2007 Conference, September 16-20, 2007, Brussels, BelgiumTable 1. The number of all waste drums generated per year (converted as the drum of 200L)DAWEvaporatedBottomSpent IonExchangeResinSpentFilterThe number of totaldrums generatedper year1977 23 0 0 0 231978 95 411 0 7 5131979 154 308 49 30 5411980 137 471 22 33 6631981 239 1104 70 125 15381982 318 353 77 28 7761983 342 417 77 29 8651984 486 788 105 44 14231985 385 628 162 9 11841986 691 965 190 32 18781987 613 1609 258 45 25251988 1165 714 599 68 25461989 1518 1397 467 89 34711990 1698 1444 453 114 37091991 1293 1467 538 81 33791992 1494 1534 521 94 36431993 1293 1192 441 105 30311994 921 1021 527 63 25321995 2154 582 551 67 33541996 1322 349 471 58 22001997 1801 258 516 60 26351998 1905 217 518 50 26901999 1786 264 525 80 26552000 1751 301 429 53 25342001 1952 251 449 57 27082002 2240 235 477 51 30032003 2356 263 418 60 30972004 2808 306 337 32 34832005 3676 191 405 24 4296The number of totaldrums generatedper stream36610 19040 9650 1588 668883. Considerations for Determination of Disposal PriorityTo determine the disposal priority of LILW in Korea, the two main parameters called theactivity and homogeneity were only considered in this study. In order words, the concept ofdisposal safety was the major consideration for the determination of disposal priority.3.1. ActivityIn this study, the DTC method was used to estimate the activity of all LILWs in Korea.The activity of γ-radionuclide derived by DTC method has some difference compared to that of

ENC 2007 Conference, September 16-20, 2007, Brussels, Belgiumγ-radionuclide derived by real detection because the DTC method has conservativecharacteristics in itself. Although DTC method has some limitation for activity evaluation ofLILWs, it is very helpful that the result of activity estimation is used for understanding theactivity tendency of LILW in Korea. Especially, this result represents that no evaporatedbottoms and concentrated wastes exceed the disposal limits. The number of drums exceedingthe disposal limits is shown orderly in DAW, Spent Filter, and Spent ion exchange resin.Although the number of DAW exceeding the disposal limits is more numerous than other wastestreams, the ratio of Spent Filter exceeding the disposal limits is more lager than that of DAW.Table 2. The number and ratio of drums exceeding the disposal limits about each waste streamThe number of totaldrumsThe number ofdrums exceeding thedisposal limitsThe ratio of drumsexceeding thedisposal limitsDAW 18,394 486 2.64 %Evaporated Bottom 16,599 - -Spent Ion ExchangeResin6,601 181 2.74 %Spent Filter 1,394 236 16.93 %Figure 2. The ratio of drums exceeding the disposal limits3.2. HomogeneityThe homogeneity of the radioactive waste may become one of main considerations for thedetermination of the priority that what kind of waste should be disposed first in order to fill in

ENC 2007 Conference, September 16-20, 2007, Brussels, Belgiumthe repository site orderly [2]. In order to allow the radioactive wastes to be safely disposed inthe repository, first of all, inner characteristics of each drum must be identified. The PCP(Process control program) is one of the methodologies to verify the homogeneity of somewastes, which are being produced in present. The NDA (non-destructive analysis) method isanother methodology for checking the homogeneity of old wastes, which were already produced.4. ConclusionSince 1977, the KHNP has generated around 67,000 drums of low and intermediate levelradioactive waste (LILW) converted as 200L drum. To dispose these drums, each drum wasonly estimated in the activity and homogeneity point of view in this study. After considering thetwo parameters to establish the disposal priority of LILW in Korea, the following order can beproposed:(1) concentrate wastes solidified by the cement and DAW whose radioactivity level is very lowand detailed characterization of the waste is identified easily because of the homogeneity, (2)spent ion exchange resin solidified by the cement, (3) spent filter, (4) DAW whose radioactivitylevel is relatively high but the characterization is well documented, (5) the wastes that requirethe additional researches including spent ion exchange resin in HIC, evaporated bottom andconcentrated waste solidified by the paraffin, and DAW contained with some possible harmfulmaterials.Table 3. The disposal priority of LILW in KoreaDISPOSAL PRIORITYWASTE STREAMS- Concentrate wastes solidified by the cement1- DAW whose radioactivity level is very low and detailedcharacterization of the waste2 - Spent Ion Exchange Resin solidified by the cement3 - Spent Filter- DAW whose radioactivity level is relatively high but the4characterization is well documented- the wastes that require the additional researchesspent ion exchange resin in HIC5evaporated bottom and concentrated waste solidified by theparaffinDAW contained with some possible harmful materialsAcknowledgement

ENC 2007 Conference, September 16-20, 2007, Brussels, BelgiumThis study was financially supported by the KHNP (Korea Hydro and Nuclear Power Co.)References[1] KHNP, “A Study on establishment of Method in Preparation for Disposal of ExistingRadioactive Waste Drums”, E05NS04, 2006[2] IAEA, “Characterization of Waste Package and Form”, Technical Reports Series, No. 383

THE PRIMARY AND SECONDARY SYSTEM WATER REGIME AT PAKS NUCLEARPOWER PLANT DURING THE EXTENDED SERVICE LIFEÁRPÁD DOMAChemistry Department, Paks Nuclear Power Plant LtdH-7030 Paks P.O.B. 71, HungaryDR. JÁNOS SCHUNKChemistry Department, Paks Nuclear Power Plant LtdH-7030 Paks P.O.B. 71, HungaryGÁBOR PATEKChemistry Department, Paks Nuclear Power Plant LtdH-7030 Paks P.O.B. 71, HungaryDR. TAMÁS PINTÉRChemistry Department, Paks Nuclear Power Plant LtdH-7030 Paks P.O.B. 71, HungaryDR. JÁNOS ŐSZBudapest University of Technology and EconomyH-1521 Budapest, P.O.B. 91 HungaryDR. TAMÁS SALAMONPannon UniversityH-8201 Veszprém, P.O.B. HungaryAbstractThe review of the water regime used for the Units of Paks Nuclear Power Plant was carried outin 2005, after 18 – 23 years of operation. It was clearly concluded after processing the huge data baseof the water regime, that there is nothing to prevent the power upgrading and service life extension ofthe Units.In 2006, a new water regime was developed that will be applied during the preparation for theservice life extension and the extended service life. In connection with this work, recommendationswere made for some modifications in the previously used water regime.Currently there is no uniform startup and shutdown water regime for WWER-440 Units.Therefore, special attention was paid to the development of a Unit startup water regime, which will beapplied for the outages as early as the year 2007.The summarised recommendations for water regime modification will be subject tointernational expert review in 2007, and the modifications judged to be implemented will be finalisedafter the review.1. The purpose and the function of the primary and secondary water regimeThe primary and secondary water regime has several purposes and functions that can besummarised in the following list:• To provide for compensation for the reactivity margin by continuous reduction of the boricacid concentration and for the control/controllability of the reactor power (with followerassemblies).• To ensure that the overall corrosion of the structural materials of the equipment is minimum.• To minimise the risk of local corrosion of the structural materials.• To minimise the deposit of corrosion products on the structural components and the fuel clad.• To keep the rate of transport of corrosion products in the coolant and their deposit on thesurfaces at low a level.

• To confine the rate of radiolytic decomposition in the primary coolant.2. Analysis of the water regime currently applied for the Paks Units2.1. The work to be doneThree fundamental issues were addressed by the complex and topical analysis:• Is there anything to prevent the service life extension of the WWER-440 Units of PaksNuclear Power Plant from water regime point of view?• To what extent the primary and secondary water regime of Paks Nuclear Power Plantcomplied with the design requirements and the knowledge level deriving from the recenttechnological developments?• How was the corrosion degradation of the major primary and secondary equipment and towhat extent was the major primary and secondary equipment degraded by the applied waterregime?2.2. Conclusions for the primary circuitIt was possible to keep the impurity concentration of the primary coolant as low as achievablefrom process system point of view.The periodic increase in the quantity of the disperse corrosion products can be attributed to theSteam Generator decontamination work performed at Unit 1, 2, and 3.Altogether two cases of fuel leakage were detected for the 79 campaigns involved. The totaliodine concentration was greater than 3.7 MBq/dm 3 , but never exceeded the Unit shutdown criteria of37 MBq/dm 3 .The average corrosion loss for the items of equipment of a material grade of 08H10N10T was0.3-0.35 μm/year, i.e. 15-17.5 μm during 50 years, which does not prevent the service life extension.2.3. Conclusions for the secondary circuitThe erosion rate of carbon steel surfaces being in contact with wet steam was high with theoriginal water regime and therefore, significant amount of deposits settled down in the SteamGenerators. This deposit had to be removed by chemical treatment method from the secondary sidesurfaces, 2 times for each Steam Generator.Due to the insufficient tightness of the initially used condensers and because of the copper alloystructural material, significant amount of corrosion activator (chloride, sulphate, copper) was carriedinto the Steam Generators. The contaminants accumulated in the structural spaces of the SteamGenerators are potential source of hazard. The shape of the gaps between the SG tube support spacersand the tubes are presented in Figure 1. The chemical cleaning work performed before were not evencompletely effective in these gaps.1629,5524Figure 1: Structural gaps in the Steam GeneratorsVarious number of heat exchange tubes had to be plugged in each Steam Generator. Thisnumber depends on several factors, e.g. on the number and extent of leakages that occurred before thereplacement of the turbine condensers of the given Steam Generator, the deviation of the materialcomposition of the Steam Generator tubes (nickel content), etc. The number of tubes plugged as aresult of the Eddy Current testing is presented in Figure 2. In this figure, the first digit of the figures on3

the horizontal axis represents the number of the Unit, while the second one is the number of the SteamGenerator. The largest leakage of primary coolant into the secondary side, during the use of the initialcondensers, was found for the Unit 2 and 3.Figure 2: Heat exchange tube plugging as a result of the leakage testsThe main condensate system, that became absolutely tight as a result of the condenserreplacement, significantly reduced the amount of corrosion activators carried into the SteamGenerators. Though, the efficiency of the blow-down of ionic contaminants from the Seam Generatorshad not changed, the decrease of the ion concentration of the feedwater resulted in the reduction ofconcentrations (chloride less than 10 µg/dm 3 ) by an order of magnitude in the Steam Generators.Due to the partial replacement of the copper containing structural elements, some coppercontent can still be detected in the feedwater and the Steam Generator.The adoption of high pH water regime resulted in a reduction, nearly by an order of magnitude,of the amount of corrosion products transported into the Steam Generators that will be further reducedby the replacement of the High Pressure Pre-heaters. The replacement has taken place for Units 3 and4 to date.The installation of the Steam Generator feedwater headers and the use of the high pH waterregime had a favourable effect on the removal of the corrosion products. The corrosion productstransported into the Steam Generators in a significantly smaller amount do no longer deposit on thesurface of the heat exchange tubes but as sludge on the bottom of the Steam Generator, thus allowingthe effective removal by blowing-down.In the following, we present a summary of the corrosion loss of the items of the secondaryequipment during the period until the end of the 50 th year of operation. s 0 represents the initial tubewall thickness, while s 50 is the wall thickness calculated for the end of the 50 th year of operation:Condenser tubes: s 0 =0.6 mm s 50 = 0.38 mmLow pressure pre-heater tubes: s 0 =1.0 mm s 50 = 0.95 mmHigh pressure pre-heater tubes: (new pre-heaters of Unit 3 and 4: s 0 =1.4 mm s 50 = 1.26 mmSuperheater, stage 1 and 2: s 0 =2.0 mm s 50 = 0.86 mmSteam Generator tubes: s 0 =1.4 mm s 50 = 1.36 mm3. Recommendation for the implementation of the service life extension3.1. Recommendations for the primary water regimeThe sampling tap (TV20) before the Water Purifier 1 does not provide representative sample forthe disperse corrosion products under startup and shutdown transitional conditions. Therefore, theinstallation of a new sampling tap is necessary. On the basis of the inspection, acceptable samples canbe taken from the sampling system of the operational boron acid measurement, which receives sample

directly from the reactor vessel.For the on-line measurement of the most important chemical parameters it is necessary toimprove the reliability and to ensure the full scope use of the primary circuit analytical monitoringsystem (FAM). The validation of FAM commenced and the results will be used to identify themeasuring systems that can be used longer and those that will need to be replaced.The development of a water regime for transitional (startup and shutdown) conditions is a taskwith highlighted importance for the service life extension. The startup and shutdown processessignificantly affect and decisively identify the water regime parameters of the Unit for the entirecampaign therefore, it is necessary to make the filtration of the primary coolant general at low (40 – 50°C) temperature in these operation modes, in addition to the more frequent checking of some majorchemical parameters. The improvement of the practice of feeding the chemicals (hydrazine, ammoniahydroxide,potassium hydroxide) to the coolant will also be initiated.Physical and/or chemical procedures will need to be developed for the removal of the corrosionproducts that accumulated in the primary circuit, which will be applied as required.The reduction of the hydrogen concentration of the primary coolant from the range of 30-60Nml/dm 3 to 25-50 Nml/dm 3 to reduce the pH fluctuation range.It is recommended that the simultaneous addition of ammonia and hydrazine, rather than the useof hydrazine water regime is considered for one or two Units.The reduction of the equivalent boron acid – potassium control range has a positive effect on themigration of corrosion products, being unavoidably present in the primary loop, and on their removalfrom the primary coolant. The difference between the presently used so called co-ordinated waterregime and the proposed overall boron acid alkalization control is presented in the Figures 3, 4, 5 and6. The reduction of the control range of the potassium equivalent (overall alkalization =potassium+lithium+sodium) and its break-down between the boron acid concentrations of 3 and 3.5g/kg and again below 1 g/kg is well marked in Figure 5. The result of these changes can be studied bya comparison of the Figures 4 and 6. It is visible in Figure 4, that the difference between the corrosionproduct concentrations (C R ) calculated for the reactor vessel and those for the Steam Generators (C SG )has a negative value for a significant portion of the campaign. Consequently, the precipitation of thecorrosion products will take place mostly in the active core, and will turn to the Steam Generators atthe end of the campaign only. This imposes unnecessary load to the active core and has an increasingeffect on the radiation doses during maintenance outage.It can be achieved by the implementation of the proposed modification as presented in Figure 6that the precipitation of the corrosion products takes place in the Steam Generators from the beginningof the campaign and it turns to the active core at the end of the campaign only. With this solution it isexpected that the active core can be maintained in rather clean condition and the portion of thecorrosion products that precipitated in the active core can be removed all years together with the fuelassemblies unloaded during refuelling outage. The reorientation of the transport before maintenanceoutage will have a favourable effect on the radiation doses received during maintenance outage.To achieve the above outlined aim, the overall alkalization concentration should be kept, fromthe beginning up to the time of achieving a concentration of 3.5 g/kg, in the upper part of the controlrange presented in Figure 5. After this point, it should be changed to the lower part of the range andbelow 0.5 g/kg boron acid concentration further reduction of alkalization is required. The change overof precipitation from the Steam Generators to the active core takes place during this period of some500 hours.

Nominal equivalent potassium-ion-boron acid co-ordinationThe orientation of the precipitation of the dissolved magnetite in the function of the boronacid concentration200,08Kekv concentration [mg/dm 3 ]18161412Kekv min10Kekv max864200 1 2 3 4 5 6 7 8 9Boron acid concentration [g/kg]dcFe [mikromol/kg]0,070,060,050,04Lmin(100%)0,03Lmax(100%)0,02Lmin(108%)0,01Lmax(108%)0-0,01 0 1 2 3 4 5 6 7 8 9-0,02-0,03Boron acid concentartion [g/kg]Figure 3: The presently used alkalization controlFigure 4: The presently used C R -C SGThe proposed equivalent potassium-ion-boron acid co-ordinationThe proposed alkalization cation-boric acid concentration: the difference of the calculatedconcentration of iron dissolved at hot leg and cold temperaturesEquivalent potassium-ion-boron acid concentration[mg/kg]20181614121086420Kekv minjavKekv maxjav0 1 2 3 4 5 6 7 8 9Boron acid concentration [g/kg]Dissolved iron concentration difference[mmol/kg]0,0350,030,0250,020,015Lmin0,01Lmax0,00500 1 2 3 4 5 6 7 8 9-0,005-0,01-0,015Boron acid concentration [g/kg]Figure 5: The proposed alkalization controlFigure 6: The proposed C R -C SG3.2. Further tasks for the secondary circuitThe small amount of deposit in the structural gaps of the Steam Generators increases the risk ofcorrosion of the austenitic steel tubes therefore, the elaboration of an effective cleaning procedure,optimised for this environment is necessary.The full replacement of the copper containing structural materials within the secondary systemshould be performed.The modification of the anion exchanger of the Water Purifier 5 to mixed bed ion exchangerwill allow a more efficient purification of the water blown down from the Steam Generators, andthanks to the cleaner water returned to the Steam Generators, the water quality of the SteamGenerators can be further improved. On line conductivity meters should be provided for testing thequality of the purified water to ensure that the exhaustion of the ion exchangers is detected in duetime.The so called additional condensate accumulating at various places in the secondary system iscollected in the condensate tank. To allow the location and elimination of the impurity sources as soonas possible, it is necessary to install an on-line conductivity measurement system for testing the waterof this tank downstream of the cation-exchange columnFor the on-line measurement of the most important chemical parameters it is necessary toimprove the reliability and to ensure the full scope use of the secondary circuit analytical monitoringsystem (FAM). The validation of FAM commenced, and the results will be used to identify themeasuring systems that can be used further and those that will need to be replaced.The blow-down efficiency of the ionic contaminants and disperse corrosion products from theSteam Generators will be reviewed by modelling and by performing dedicated measuring programs,and recommendations will be made for the improvements of the efficiency, as required.

The water regime risk factors of the Steam Generators are summarised in Table 1 below:Risk factors Expectation Actualvalues forUnits 1 to 4The condition of the heattransfer surfacesConcentration of thedisperse iron corrosionproducts transferred intothe Steam Generators[μg/dm 3 ]Free of deposits(< 50 μm)minimumdeposited< 3 – 5 < 10(3 – 8)Concentration of the stress corrosion activators in the SteamGenerator waterChloride-ionconcentration [μg/dm 3 ]Maximum chloride-ionconcentration duringhide-out re-dissolving[μg/dm 3 ]Sulphate-ionconcentration [μg/dm 3 ]Maximum sulphate-ionconcentration duringhide-out re-dissolving[μg/dm 3 ]< 10 (1 – 5) < 10(3 – 8)< 0.1 < 0.1< 20 (5 – 10) 10 – 15< 0.15 < 0.15Exclusion of oxidising materials from the water of the SteamGeneratorsConcentration of thedisperse copper corrosionproducts in the feedwater[μg/dm 3 ]Oxygen concentration ofthe main condensate[μg/dm 3 ]~ 0 < 0.5< 5 – 10 < 5Required actionsMore efficient blow-downof the Steam Generators fordisperse iron corrosionproductsMore efficient blow-downof the Steam Generators fordissolved materials.Modification of the WaterPurifier 5. Elimination ofthe periodicalcontamination effect of thecondensates. Improving theconfinement of thesecondary circuit.Complete removal of thecopper from the secondarycircuit. Removal of thecopper from the structuralgaps.Table 1: Summary of the water regime risk factors in the Steam GeneratorsLiterature[1] LG Energia Kft, 2005: Comprehensive and thematic analysis of the water regime used at PaksNuclear Power Plant during the period between 1983 and 2004,[2] LG Energia Kft: Review of the primary and secondary water regime used at A Paks NuclearPower Plant for power upgrading, 2005.[3] LG Energia Kft: Development of water regime for the primary and secondary system withconsideration of the service life extension. Development of water regime for startup andshutdown conditions, 2006.

BEST RADIOTOXICITY INDICES EVALUATION FORORIGEN 2.2 (ORNL) CODE, IN THE FRAMEWORK OFRESEARCH ACTIVITIES ON ADS SYSTEMSB.CALGARO, B.VEZZONI, N.CERULLO, G.FORASASSIDIMNP-University of Pisa,Via Diotisalvi, 2, 56122 Pisa, ItalyB.VERBOOMEN, H.A.ABDERRAHIMAdvanced Nuclear Systems Institute- SCK·CENBoeretang 200, B-2400 Mol, BelgiumABSTRACTThe European Community, in the EURATOM 6 th Framework Programme, supportsnumerous R&D projects aiming to a possible development of Minor Actinides transmuters.This study aims to provide a critical approach in the context of P&T strategy, necessary totreat radiotoxicity data from different sources. The work, performed in SCK·CEN(Belgium) and Pisa University (DIMNP), concerns the state of the art revision of internaldosimetry in ICRP Publications, 10 CFR Part.20, 96/29/EURATOM and 2001 Belgianrules on Radiological Protection. In this regard, preliminary evaluation of an IndustrialADS (400MWth, 2.5mA) burning capability, using inert matrix fuel was performed. Themain result obtained is an accurate assessment of ORIGEN2.2 last version, showing adifference of more than one order of magnitude in radiotoxicity values, if calculated usingthe most up to date coefficients from the latest ICRP Publications in comparison to theORIGEN original ones.1. IntroductionThe panorama of studies about Minor Actinides (MA) transmutation, using an Accelerator DrivenSystem, as a possible complementary approach to Spent Nuclear Fuel (SNF) geological disposal, hasuncertainties in radiotoxicity definitions and values that, nowadays, are not more acceptable.This work, joined to nuclear fuel cycle closure and in particular to Partitioning and Transmutation(P&T) strategy, aims to give a critical approach and to manage radiotoxicity definitions. In fact, datafrom different sources imply often difficulties in comparing values.Radiotoxicity coefficients evolution review across ICRP Recommendations, BEIR and UNSCEARPublications was analysed in depth. Attention was focused on their influence on spent nuclear fuelpotential radiotoxicity and on the US and European Countries rules in radiological protection[1].The background is the present Belgian situation (7 PWR, 5800 MWe installed), where, despitedecision constrained by guarantee of energy independence and by engagement to respect the Kyotoagreement, Belgian Government decided (2003) to phase-out nuclear energy production in next the 25years. Nevertheless the phase-out decision can be re-opened if certain conditions are not met(guarantee of energy independence, engagement to respect the Kyoto agreement). In this latter case,we proposed an alternative scenario: LWRs substitution, consequent installation of advanced reactors(i.e. EPR-European Pressurized Reactor) in synergy with industrial scale ADS to support energydemand and to burn consistent amount of HLW, in connection to Generation IV reactors (fastspectrum)[2].The starting point of evaluating SNF radiotoxicity evolution is the waste inventory calculation of areference reactor: a 1000 MWe PWR was chosen, with 4.3% enriched and 50GWd/tonIHM dischargeburn-up fuel, as Belgian panorama typical one (Doel 3).1 of 5

To perform these calculations ALEPH-1.1.2[3], a burn-up code under development at SCK•CEN(Belgium) inside the MYRRHA Project[4], that couples MCNPX, ORIGEN2.2 and NJOY99.112code, was used.ORIGEN2.2 was used separately from ALEPH to trace radiotoxicity curves. Apposite radiotoxicitycoefficients data-bases were created, suitable for ORIGEN2.2 and for ALEPH-1.1.2 that gives asoutput isotopic materials quantities (g/cm 3 ) only.2. Health hazard measure from Radioactive High Level WasteCorrect assessment of exposure risk for members of public shall be evaluated on the basis ofgeological disposal that will receive total or part of nuclear material from reactors, if a solution fornuclear waste management would be found.In this regard, Radiotoxicity concept is representative of the measure of health hazard from nuclearradioactive waste that is considered a biological hazard potential.In literature different radiotoxicity definitions exist, such as:• the "quantity" of water required to dilute the material to the "maximum permissible concentrationfor human consumption" (MPC), that is referred then to its own country annual law limit;• the relative measure to reference level, i.e. natural uranium radiotoxicity or an other ReferenceLevel (RL); unfortunately the definition of RL is ambiguous as it will be well shown further on;• the related to dose value, i.e. ‘dose per unit of intake’ (DPUI) in Sv/Bq, or to the number of cancerdeaths expected if a given radioisotope is swallowed by a person (CD/Bq). It is not an absolutemeasure of the biological hazard of a given amount of radioactive material, but it is very useful toconstruct a relative measure of the biological hazard potential, that do not depends on country lawannual limit of intake for members of public.We chose the last one as the best SNF radiotoxicity definition in this document; in particular, wemeasured our radiotoxicity results in Sv/GWe-y, that is a derivate unit of Sv/Bq referred to the energyproduced by a nuclear reactor for each ton discharged.Results obtained are shown in Figure 1, these curves are built with update radiotoxicity coefficientdata-bases from ICRP 72[5] applied to ORIGEN2.2. The continue curve (red) represents SNF andreaches the Reference Level, constant value (pink), in 2·10 5 years.ALEPH calculation is referred to the reference reactor described above with an irradiation period of4.5 years inside reactor and a decay period of 10 8 years after discharge (logarithmic scale).Fig 1. Spent Nuclear Fuel Radiotoxicity in Sv/GWe-y obtained by ORIGEN2.2 and updatedcoefficients from ICRP 72.2.1 Radiotoxicity concepts evolution: ICRP Recommendations.Since the first years of 20th century international organisms are involved in understanding healthhazard from buried radioactive high level waste considering at first internal dosimetry and relatedquantities. The first comprehensive publication of internal standards was ICRP Publication 2(1959)[6] where new concepts as "critical organ", "standard man" and "biological half-life" wereintroduced to describe the physiological removal of radionuclides. One of the most important2 of 5

innovation introduced by ICRP2 was the respiratory and gastrointestinal (GI) tract model. For eachradionuclide the Publication presented “Maximum Permissible Concentration values” (MPC),expressed in μCi/ml, and calculated the annual dose limits, expressed in "maximum permissible annualdoses", that included concentration in air and water (corresponding to inhalation and ingestionpathways respectively) for both a 40-h week (occupational) and a 168-h week (used also for nonoccupationalexposure).In following publications other secondary limits were introduced: "annual limit of intake", ALI and"derived air concentration", DAC. The annual limit of intake is the intake during a year of practicethat would result in a committed effective dose equivalent of 0.05 Sv or a committed effective dose toa single organ of 0.5 Sv to "Reference man" ( that updates “standard man”). These new limits and newrespiratory and GI tract models were collected inside ICRP Publication 30 (1979)[7].A fundamental change to the previous philosophy was presented in ICRP Publication 60 (1990)[8],and subsequent. In fact, during ’80s knowledge about radionuclide metabolic behaviour (minoractinides mainly) increased and more significant approach was adopted. In particular, limits of doseintake changed: 1 mSv for annual effective dose for population and 20 mSv for occupational annualeffective dose limit values (they are annual limits taken in European Countries referred to96/29/EURATOM). ICRP 30 values were, then, replaced and the new ones are expressed in terms of“Inhalation and Ingestion Dose Coefficients” or “Dose Factors” (Sv/Bq). We referred to ICRP72 lastversion coefficients. We analyzed also complementary studies about biological hazard potentialcarried out by different international organisms (i.e. BEIR and UNSCEAR) to comprehend the “cancerdose” as alternative radiotoxicity definition.2.2 Evolution in US and European Radiological Protection RulesFrom year to year, dose limit changes and drives to revise Countries rules in Radiological Protection.This work analyzed its evolution inside European and US rules.Our analysis can be resumed as follows:• European Community adopted ICRP60, and subsequent, with European Directive96/29/EURATOM, receipted by single states. In particular, this work performed in SCK•CEN isreferred to Belgium, where 96/29/EURATOM was taken in by 2001 Belgian law.• 10 CFR Part.20 includes US “Standards for Protection Against Radiations”. The code last versionis dated 1994 and it is referred to ICRP 30: it incorporates internal dose concepts and primary andsecondary dose limit coefficients adapted from the recommendations of the InternationalCommission on Radiation Protection Publication 26 (1977) and 30(1978). Although 10 CFRPart.20, 1991 revision, included the ICRP 60 recommendation to reduce the annual dose limit formembers of the public from 5 mSv to 1mSv, it didn’t adopt the new occupational dose limitrecommendation of 20 mSv/year.As just indicated above, differences in radiotoxicity coefficients affect evaluation of SNF radiotoxicityand of the time to reach Reference Level.3. Reference Level best estimateReference Level best estimation is necessary to correctly express the health hazard from SNF in termof radiotoxicity. The decay chains that lead to the achievement of secular equilibrium were simulated,taking values from present natural uranium isotopic atomic fraction, (ORIGEN2.2 calculation).In order to obtain a right value, Reference Level was calculated considering that 1 ton of U enriched at4.3% requires the use of 8 tons of natural uranium (Eq. 1). We obtained, in perfect agreement withliterature (NEA-OECD Report[9]), a reference level equals to 3.07·10 6 Sv/GWe-y that corresponds to1.61·10 5 Sv/tonIHM.4. Results obtainedtonNatU ( A − ε ) (4.3% − 0.2%)= =≅ 8(1)tonFuel ( C − ε ) (0.710971−0.2%)In order to show that data from different sources imply often difficulties in understanding values, weperformed interesting comparisons.In fact, one order of magnitude difference exists between radiotoxicity from 2001 Belgian law,referred to the latest ICRP 72 Publication, and from ORIGEN own coefficients (ORNL, 1982) taken in3 of 5

y 10 CFR Part.20 (‘Standards for Protection Against Radiations’), 1982 edition, based on ICRPPublication 2 MPC values that are old data (Figure 2). In fact, studies carried on between 70s and 90son radiobiology corrected radionuclides effects on internal organs.Fig 2. Radiotoxicity values comparison between ORIGEN2.2 original coefficients (1982, ICRP 2) andvalues obtained from ICRP 72 (1996) coefficients.Moreover, a useful comparison, Figure 3, was obtained among ORIGEN2.2 original coefficients(1982, ICRP 2), coefficients from Belgian law (2001), related to 96/29/EURATOM (“Basic SafetyStandards Directive on radiation protection”, 1996) and ICRP 72, and coefficients from 10 CFRPart.20 (1994) related to ICRP 30.10 CFR Part.20 last edition (green curves) has the most restrictive radiotoxicity values. We tried toexplain this result: 10 CFR Part.20 adopted the 1 mSv annual dose limit for members of the public butapplies dose coefficients from ICRP 30 Publication that overvalued actinides effects on tissues ifcompared with ICRP 72.Fig 3. Radiotoxicity comparison among ORIGEN2.2 original coefficients (1982, ICRP 2), valuesobtained by ICRP 72 (1996) coefficients and last 10 CFR Part.20 edition (1994).Part of this work was finalized to analyze single isotopes contribution upon the whole of radionuclidesextracted from the reactor. We found an imprecision inside ORIGEN2.2 (2002) decay library referredto Pr-143 radioactivity concentration guide for continuous ingestion of nuclide (WRCG value). In factthe value in ORIGEN is 5·10 −9 μCi/ml instead of 5·10 −5 μCi/ml; this last one is the correct valuepresent in 10CFR Part.20 (1982). This mistake was underlined during a calculation at t=0 (exit fromreactor) because t 1/2 of Pr-143 is 13.57 days. Figure 4 and Figure 5 show, respectively, the wrong and4 of 5

the correct distribution of few isotopes contribution to buried nuclear fuel radiotoxicity at time ofdischarge from reactor.Error! Not a valid link.Error! Not a valid link.Fig 4. ORIGEN2.2 (2002) RCG coefficientsoriginal one: error in Pr-143.Fig 5. ORIGEN2.2 (2002) RCG coefficients original one:correct value in Pr-143.5. ConclusionThis work enables the use of updated radiotoxicity coefficients and, also, ALEPH and ORIGENupdated versions in order to develop a more accurate approach to ADS burning capability.At first a theoretical burning capability efficiency calculation was, in fact, performed as a term ofcomparison with the industrial ADS burning capability. The theoretical burning capability acceptablevalue, that is a MAs residual concentration left in wastes, obtained is 27%wt. able to reach RL in 2-3·10 3 years. So, by the industrial scale ADS simulations performed, a transmutation efficiency equal to72% after 3 transmutation cycles was reached that is a good value because it is nearly the same onethan the theoretical transmutation analysis performed before[1],[10].References[1] B.L.Cohen. Effects of ICRP publication 30 and the 1980 BEIR report on hazard assessmentsof high-level waste. Health Physics, 42:133–143, 1981.[2] B.Calgaro, B.Vezzoni “Contribution to Long Term Minor Actinide Management throughTransmutation in Accelerator driven Systems (ADS)”, Master Thesis in Nuclear and Industrial SafetyEngineering, University of Pisa (Italy).[3] Wim Haeck and Bernard Verboomen, “An Optimum Approach to Monte Carlo Burn-Up”.Nuclear Science and Engineering, Volume 156, Number 2, Pages 180-196, American NuclearSociety, June 2007 (2007).[4] H. Aït Abderrahim and D. De Bruyn (editors) “MYRRHA, a new future for nuclear research,pre-design (draft2) report”, Technical Report (on CD-ROM), June 2005.[5] “Recommendation of the International Commission on Radiological Protection, ICRPPublication 72”, Pergamon Press, Oxford,1996.[6] “International Commission on Radiological Protection. Permissible Dose for InternalRadiation. ICRP Publication 2”, Pergamon Press, Oxford, 1959.[7] “International Commission on Radiological Protection. Limits for Intake of Radionuclides byWorkers. ICRP Publication 30 Part 1”, Pergamon Press, Oxford, 1978.[8] “International Commission on Radiological Protection. Recommendations of the InternationalCommission on Radiological Protection. ICRP Publication 60”, Pergamon Press, Oxford, 1990.[9] NEA-OECD “Physics and Safety of transmutation systems-a status report” Technical ReportN. 6090, NEA-OECD, 2006.[10] B.Calgaro, B.Vezzoni, et al. “Contribution to Long Term Minor Actinide Managementthrough Transmutation in Accelerator driven Systems (ADS)”, proceedings in IYCE2007 Conference,05/31-06/02Budapest, ISBN 978-963-420-908-9.5 of 5

PROFILING THE GAMMA-RAY DISTRIBUTION OF THECEMENT LINING OF AN ESTABLISHED NUCLEAR POND ASA FUNCTION OF DEPTHB.A. SHIPPEN AND M.J. JOYCEControl and Instrumentation Research Group, Engineering Department, Lancaster UniversityBailrigg, Lancaster LA1 4YR, UKABSTRACTAn investigation to determine the relationship between the depth of contamination within theconcrete lining of a nuclear pond and the ratio of x-ray to gamma-ray photopeak intensityobtained from sodium iodide detector is described. The environment was simulated usingestablished radiation transport codes. Caesium-137 is assumed to be the sole source ofcontamination. A detector was simulated across the energy spectrum of a source, and rasteredacross its location from ±10cm at intervals of 0.5cm. A combination of point and distributedsources were simulated within a concrete structure, as well as in free air to create acomprehensive set of data. The results were then validated against experimental data anddemonstrate a high level of agreement. This investigation indicates that the x-ray / gamma-rayratio has the potential to localise contamination on the surface of the concrete lining; furtherinvestigation is required to locate the contamination in three dimensions.1. IntroductionThe current focus of the civil nuclear industry in the UK has shifted toward the decommissioning ofobsolete facilities. This policy has been instigated by the UK government [1] via the foundation of theNuclear Decommissioning Authority (NDA) through the energy act [2]. This shift requires that manylegacy nuclear facilities are surveyed to optimise the design of processes and techniques to dismantle theassociated buildings that have often been contaminated through years of use. The early stages of thisprocess have already generated many diverse challenges that require novel measurement solutions.An interesting challenge found at many nuclear facilities is that associated with the measurement ofentrained radioactivity within the walls of nuclear fuel storage ponds. These facilities are found at manypower plants and reprocessing facilities where fuel is stored following its use in a reactor. During thelifetime of many of these facilities the pond linings absorb various nuclide species from the water in thepond, as a result of contamination from the fuel stored in it. The water, used to shield and cool the spentnuclear fuel, penetrates the concrete lining through hydroscopic action and areas of high concentration canbuild up in defect areas, i.e. cracks and holes. This is especially relevant in very old facilities, that werebuilt at a time when little was known about the long-term effects of residual activity and hydroscopicingress and, often, ponds were not lined to prevent degradation of the concrete itself.It is widely understood and expected that the penetration of radioactivity into porous media, such asconcrete, will be limited in depth. It is also expected that this depth will be dependent on the quality of theconcrete and the nature of the fuel stored in the pond, including such factors as the extent of clad corrosionetc. In many cases the distribution of the entrained radioactivity is not uniform but, often, this distributionis not known. If the concrete lining is to be safely removed during the decommissioning phase, it mustfirst be profiled in terms of nuclide penetration depth as well as resultant activity in order that doseestimates can be established. Furthermore, and very importantly, the destiny of waste arisings from thedemolition of the plant can be determined before the decommissioning process starts if the depth profile ofthe entrained activity is known, which improves planning and will potentially reduce the cost of wastedisposal significantly. A detector capable of profiling in this manner is the principal goal of theRadioactivity Depth Profile Analysis Tool (RADPAT) project that is the subject of this paper. RADPAT isa PhD bursary research project being conducted at Lancaster University, UK and is sponsored by theNDA. The research presented in this paper describes an initial method trial and concerns the localisationof caesium-137 at depth within the concrete lining of a contaminated nuclear facility.

2. BackgroundCaesium-137 is often one of the significant radioactive isotopes present in the water surroundingspent fuel stored in ponds [3] and, due to the tendency of caesium to be absorbed easily, it often becomesdistributed throughout the concrete comprising the pond structure. On this basis, caesium-137 has beenassumed to be the major nuclear species contaminating the concrete lining in this research. Caesium-137has a half-life of ~30 years, eventually decaying to a meta-stable state of barium-137m via beta decay. Theexcited barium-137m has a half-life of ~2.5 minutes and decays with via γ-ray emission to its ground state;however this decay is usually assigned to the caesium-137 nuclide. The major contribution to the resultantenergy spectrum comes from a mono-energetic γ-ray peak at 662 keV, with a smaller contribution from x-rays at lower energies; with the most significant peak situated at around 32 keV [4]. Due to the lack ofkinetic energy the emitted x-rays from a caesium source have a limited penetrative depth in comparisonwith higher-energy γ-ray emissions. This is compounded in relatively dense materials such as concretewhere, for example, the mass attenuation coefficient is 0.9601 cm 2 g -1 at 30 keV [6]. If this is compared tothe associated γ-ray, which has a mass attenuation coefficient of 0.0823 cm 2 g -1 [6], it is clear that forincreasing depth the ratio between the x-ray and γ-ray peak will vary predictably in a nonlinear fashionthat has potential uses in depth analyses.The results of an investigation into the relationship between the x-ray peak and the γ-ray peak ofthe various energy spectra obtained from rasters over a caesium-137 source are presented in this paper.Whilst scanning areas close to caesium-137 contamination, the x-ray peak at 32 keV will be relativelylarge compared to the rest of the energy spectrum. As the detector moves away from the contamination,this peak will decrease due to the attenuation of the x-rays within the medium between the detector and thesource of the radiation. The mass attenuation coefficient of the emitted 662 keV γ-ray suggests that therelative intensity of this emission will attenuate at a much slower rate than that of the x-ray peak withrespect to distance. Therefore, if the ratio is observed for each energy spectrum from a raster scan; themaxima of the x-ray/γ-ray ratio may indicate the position of contamination on a two-dimensional surface.The aim of this research was to determine whether this hypothesis is justified. Within the investigation theinfluence of two variables was examined;• The difference between an open-air caesium-137 source and a source embedded in concrete. The fallin count rate due to the attenuation of photons from a localised source embedded in concrete will beaffected as the detector moves away from it, as the photons have to travel through an increasingdistance of concrete to get to the detector. This increases the probability of attenuation in a non-linearfashion and could significantly alter the relationship between the x-ray / γ-ray ratio and the distancefrom the source. To examine this variable scans was conducted with and without a layer of concrete.• The effect of differently-shaped sources on the obtained projections. The intensity of radiationemissions from a radioactive source follow the inverse-square law with distance from the source,assuming it is a point source; otherwise the intensity distribution will exhibit an additional source ofanisotropy that is dependent on the physical shape of the source. It is convenient in many cases,especially in simulation studies, to assume that the radiation propagates from a single point in space.However, in this research the shape of the source will influence the distance the radiation has to travelbefore it hits the detector and, very importantly, anisotropic source distributions in the concrete arehighly-likely. Thus, the point source assumption can not be made. The discrepancy will be mostapparent in the contrast of a pointsource and an evenly-distributedsource across a plane normal to theaxis of symmetry of the scanningdetector. In this research both a pointsource and a distributed source havebeen examined as part of theinvestigation.Two separate approaches were usedduring the investigation to obtain results:-• Simulation – Radiation transport codeshave been used to simulate theFigure 1: The attenuation of 32keV x-rays in aluminium

environment on a standard PC prior to any experimentation taking place.• Experimentation – A set of experiments were conducted. The results from these were used to validatethe data obtained from the simulations.2.2 Simulation SetupTwo transportation codes were used in the investigation, Geant4 [7] and MCNPX [8]. The initialparameters in both codes were matched as closely as the different input methods would allow. Thistechnique was used to produce two separate sets of results that were sufficiently similar to be ascribed tothe same set, which enabled any spurious results to be identified. This approach increased the confidencein the data as a different mechanism was used to obtain data from each of the codes. For photonicsimulations Geant4 uses classical equations preceding a run to produce a set of tables that contain all therelevant simulation cross sections [9]. MCNPX uses an evaluated set of cross section libraries, such asENDF-B VI, to interpolate the relevant information during the simulation run.The material and geometric definitions of both codes were equivalent and were based upon thecollimated ORTEC detector arrangement mentioned previously. All the materials and densities weredefined from the NIST database of materials [10] and assumed to be at 270ºK. For convenience the signalprocessing chain of the detector was simulated as a single blockof Mu metal which, due to the relatively low energy of the x raysunder evaluation, would not represent a significant source oferror. The disintegration of the source was defined using aprobability distribution derived from [4]. This definition includedall possible particles and energies through the full decay chainand was assumed to be fully isotropic. The point source wasassumed to be infinitesimally small and located at the origin. Theplane source, whilst similar in every other respect, was distributedwith respect to the scanning plane from -2cm to 2cm.2.3 Experimental SetupThe approach used throughout the investigation was basedaround a collimated ORTEC 905 7.62cm sodium iodideFigure 2: The experimental setup scintillation detector [5]. The collimator used in the experimentswas manufactured from aluminium and designed such that theentire detector surface area was covered, except for the face used to create the scanning aperture. At thescanning face the collimator was designed to overhang by 0.3cm to limit the size of the aperture. The wallthickness of the collimator was specified by theoretical calculation of resultant intensity. The massattenuation coefficient for x-rays at 30 keV, taken from [6], was used to derive the relationship betweenintensity and the thickness of aluminium. This was then validated using MCNPX [8]. This relationshipcan be seen in Fig. 1, where the trend of the simulated data agrees well with that of the theory. Howeverthe intensity of the simulated data is enhanced which is due to MCNPX using 32 keV x-rays as a source;as well as geometric effects being taken into account by the simulation. On the basis of the trendexhibited in Fig. 1 it was decided that a wall thickness of ~ 0.3cm would be sufficient for the collimator,as this thickness would attenuate ~90% of the incident x-rays.The source location was defined as the origin in the experiments with the collimated detector rasteredacross from -10 cm to 10 cm at a height of 0.5 cm. Although every effort was made to keep theparameters consistent throughout the investigation, there were inevitable discrepancies between thevarious methods used that are detailed in subsequent sections.The apparatus used in the experimentation is shown in Figure 2. The detector was coupled to anORTEC Digibase module [11] which incorporates the necessary signal processing electronics. This devicewas connected to a standard portable PC through a USB cable which used ORTEC Maestro MCAemulation software to collate all the data fed back from the detector. Clamp-stands were used to hold boththe collimated detector and the source so that any contribution due to scatter from surrounding materialswas minimised. The source used in the experiment was a 39.7 kBq caesium-137 + barium-137m disk,aged at 1 year giving a relative intensity of 38.79 kBq.

3. Results0.220.07a) b)Experimental ResultPoint source in free airPlane source in free air0.20.06Plane source in concretePoint source in concrete0.180.050.16x−ray/gamma−ray ratio0.140.12x−ray/gamma−ray ratio0.040.030.10.020.080.010.060.04−15 −10 −5 0 5 10 15Distance from Origin (cm)0−15 −10 −5 0 5 10 15Distance from Origin (cm)Figure 3: MCNPX x-ray count rate for a) Free air source b) Concrete embedded sourcea) b)0.220.070.2Experimental ResultsPlane source in free airPoint source in free air0.06Plane source in concretePoint source in concrete0.180.050.16x−ray/gamma−ray ratio0.140.12x−ray/gamma−ray ratio0.040.030.10.020.080.010.060.04−15 −10 −5 0 5 10 15Distance from Origin (cm)0−15 −10 −5 0 5 10 15Distance from Origin (cm)Figure 4: Geant4 x-ray count rate for a) free air source b) source embedded in concreteDuring the simulated raster scans each energy spectrum was generated using 500,000 events as thisgave sufficiently small uncertainties. The results obtained from Geant4 were normalised during analysis.MCNPX produces an output which is already normalised. For the experimentation a spectrum at eachposition was given 5 minutes to accumulate. This corresponded to an average of 1.16 million events perscanning iteration. The FWHH values were taken for both the x-ray and γ-ray peaks to account for thestatistical response of the scintillation detector. The normalised results for the scans carried out in

MCNPX are shown in Figure 3 whilst those obtained from Geant4 are shown in Figure 4. Experimentaldata for the source are included on Figure 3a and 4a for comparison purposes.4. DiscussionThe data in both Fig. 3 and Fig. 4 demonstrate a general trend that as the detector scans across thesource (in both plane and point variants) the ratio of the 32 keV x-ray peak intensity to the 662 keV γ-raypeak increases. As expected, the distance from the detector to the source decreases the attenuation of theemitted x-rays decreases more quickly than that of the γ rays. For both the point and planar sources, thesimulated ratio data increase as the detector aperture approaches the origin, albeit at different rates withthe planar source increasing more slowly than the point source and peaking closer to the origin. As thedetector scans further toward the origin the source moves into the aperture of the detector. This removesthe effect of the collimator and increases the intensity of both the x and γ rays, though at significantlydifferent rates. The differential rate causes the local maxima at approximately ±4 cm where the collimatorinfluence is gradually removed, increasing x-ray intensity rapidly whilst the intensity of the γ-rays remainsfairly constant. However, as the position of the detector moves across the region of origin (i.e. between -5cm and + 5 cm) the ratio falls in a nonlinear but symmetric manner, in both the planar and point sourcecases and for MCNP and Geant4, with the minimum corresponding to the position at the origin. In thisregion it is reasonable to assume that the x-ray detection efficiency will not increase much more than isobserved at, say -5 cm or +5 cm. However, as more of the detector crystal moves into the path of thesource as it approaches the origin, the absolute detection efficiency for γ rays emitted by the source willincrease, as more sodium iodide material is presented within the aperture -5 cm and +5 cm. Thus the x-ray/γ-ray ratio falls in this region.The experimental results in the free-air case demonstrate a somewhat different trend. Whilst theratio increases in a consistent way with the simulated data, the fall in the ratio in the -5 cm/+5 cm region isnot replicated experimentally and this inconsistency is currently poorly understood. However, theexperimental technique employed in this research could be developed significantly to remove severalsources of error which may improve the confidence with which the data is held in future. For example,the detector position is possibly too close to the source such that its non-point disc shape influences thedata more consistent with the planar source limit.5. ConclusionThis investigation has shown that it is possible to localise contamination on a 2D surface by usingthe x to γ-ray ratio to find the local minimum exhibited by the simulated data at the origin. Furtherresearch will take into account the depth of the source; with the aim of finding a relationship between itand the x-ray to γ-ray ratio. Whilst the penetration of the x rays in this context is unlikely to exceedconcrete thicknesses of ~ 4 cm, this is largely consistent with the approximate scabbling capability ofmany decommissioning techniques currently in use to dismantle contaminated concrete surfaces.Irrespective of the distribution of caesium-137 contamination in the concrete and its specific activity i.e.whether planar or localised, the relative proportion of x ray to γ ray intensity has the potential todistinguish its depth due to the relative differences in attenuation of these photon emissions [12]. Thistechnique will aid further research to produce a 3D radioactivity profiles in legacy facilities.6. References[1] UK Government, “The Decommissioning of the UK Nuclear Industry’s Facilities”, DTI, 2004.[2] UK Government, “Energy Act 2004”, Department of Trade and Industry, 2004[3] C.A. Friskney and M. V. Speight, Journal of Nuclear Materials, Vol 62, Pages 89-94, 1976[4] Nuclear Energy Agency, JANIS 3 – JEFF 3.1. Issy-les-Moulineaux, France, 2007[5] ORTEC Ltd., 905 Series Model 4, http://www.ortec-online.com/detectors/photon/905-4.htm[6] U.S. Department of Commerce, “Table of X-ray Mass Attenuation Coefficients”, NIST, 1996[7] S. Agostinelli et al., Geant4 – A simulation tool kit, NIM A506, 250-303 (2003).[8] L.S. Waters (Ed.), MCNPX Users Guide, Doc LA-CP-05-0369, LANL, MCNPX 2.5f, 2005[9] D.H. Wright, Geant4 Physics Reference Manual, Version 4.8.3, 2007[10] U.S. Department of Commerce, “ESTAR database”, NIST[11] ORTEC Ltd., Digibase and Maestro MCA software, http://www.ortec-online.com/pdf/digibase.pdf[12] ‘Spectral detection of radiation’, M. J. Joyce, Patent, GB.07.13166.7, Application 6 July 2007.

MAESTRO: A HYDRAULIC MANIPULATORFOR MAINTENANCE AND DECOMISSIONINGAPPLICATIONO. DAVID, Y.MEASSON, C. BIDARD, C. ROTINAT-LIBERSA,F-X. RUSSOTTOInteractive Robotic Unit, CEA-LISTRoute du Panorama F92265 Cedex, Fontenay aux Roses, FranceABSTRACTCompared to electric manipulators, hydraulic manipulators can handle very high payloads with respectto their size and mass. However, due to their limitations in force-torque control, they are usuallydisqualified for precise manipulation.CEA, in collaboration with CYBERNETIX developed a complete remote handling system featuringthe advanced hydraulic robotic arm MAESTRO. The complete slave system, 10 kGy radiationhardened,is composed of the MAESTRO arm, a 2 m long, 100 kg payload 6 degrees of freedom slavehydraulic manipulator, mounted on an embedded unit made of a 210 bars hydraulic power pack and aslave controller. The master station is made of the latest generation VIRTUOSE master arm suppliedby Haption, including a dedicated master controller. The system specifications were defined accordingto the requirements of decommissioning activities in existing nuclear facilities and maintenancescenarios of the next step fusion reactor ITER. Specific attention has been paid to the decontaminationand maintainability aspect of the robot, in nuclear conditions. Validation of the system has beenachieved through test campaigns during 1000 hours of an endurance test completed with operationalevaluation on a set of tasks.Using the generic TAO (Computer Aided Teleoperation controller) control software designed byCEA, specific force-torque control loops were developed to improve the manipulator performance,thus allowing the MAESTRO arm to be used in a traditional teleoperation master/slave applicationwith force feedback. In addition to this control software, the graphical supervision software Magritteprovides the operator with an additional interface to manage and monitor the system. Repetitive taskslike tool picking can therefore be managed by the system while the operator focuses on the main task.Thanks to a collision detection algorithm, Magritte warns the operator when the tool or any part of theslave arm comes too close to a sensitive unit.This paper presents the complete MAESTRO system and gives up an overview of the research anddevelopment activities currently carried out:• Real-time failure detection: when dynamic simulation helps managing hardware failures.• Making hydraulic cleaner and safer: from oil to demineralized water hydraulic.• Collision free teleoperation: from collision detection to dynamic collision avoidance.• Decreasing working load on the operators: providing assistance functions for enhancedteleoperation tasks.

1. IntroductionThe MAESTRO hydraulic manipulator belongs to the class of servomanipulators. This class oftelerobotic systems appeared in the early 80’s with the progress on computer assisted teleoperation.Compared to traditional through the wall workstations equipped with mechanical master-slavesystems, or to power manipulators with limited control and speed performances, these new systemsprovide innovative features and improved capabilities, including:• operation from a remote control room located in an unrestricted access (cold) area,• use of different arm morphologies and technologies for the master and the slave,• work in the Cartesian coordinates,• compensation of weight of handled tool,• adjustable force and speed ratios in the force feedback loop,• handy automatic robot modes (tool picking, return to rest position...),• virtual mechanisms to assist operator in tricky tasks (sliding, axial or even more complex),• virtual reality to assist the operator in complex tasks,• real-time collision-avoidance to protect both environment and manipulators from shocks.If power of electric motors is enough to supply good force feedback capabilities to the operator inmaster arm stations, requirements in the hot zone sometimes require the capability to supply highforces that standard electric motors are unable to provide in a limited space. Starting from a hydraulicmanipulator developed for offshore applications, CEA-List, in collaboration with the French roboticscompany Cybernetix developed a complete telerobotic system for nuclear operations. The MAESTROTelerobotic System (see figure 1) is composed of:• A master station including:o a graphical 3D supervision interface,o video display monitors,o a Haption Virtuose 6D master-arm,o a master-arm controller.• A slave station with:o a 6 degrees of freedom hydraulic manipulator,o a rad-hardened embedded slave-controller,o an embedded hydraulic power pack,o a remotely controlled PTZ video camera with tool tracking capabilities.After a brief overview of the current capabilities of the MAESTRO system, this paper presents thelatest developments carried out at CEA-List in the field of terobotics.2. Design description and performances of the manipulatorMade of Titanium, the Maestro slave arm is a 6 degrees of freedom manipulator with two lengthconfiguration possible. It is typically 2 meters long with a payload capacity up to 100kg. The latestdesign work took advantage of good performances of original design from Cybernetix and Ifremerdedicated to offshore applications, improving it to satisfy easy decontamination requirements, eg: withsmooth surfaces, avoiding any contamination traps in the design, leak tight, and pressurized housingcapabilities.

Figure 1: Maestro telerobotic systemQualification of the complete system for Remote Handling (RH) in nuclear facilities application ranthrough a validation process including long term reliability testing during 1000 hours of the completearm. Tests profile was based on rehearsal of a typical working trajectory including various payloadhandling. The trajectory was defined according to position records of the real manipulator during arepresentative teleoperation task:• Tool picking.• Task completion with tool.• Tool removal.Rad resistance of the system was proved in an irradiation facility. A cumulated dose of 10.65 kGray ata dose rate of 74 Gray/hour was reached before the first stop of the system. The mock up consisted of:• a single elbow joint,• a resolver,• a servovalve,• two pressure sensors,• a low level controller.It has to be noticed that after a 2 hours power off phase it was possible to restart the mock-up due to arecovery effect of the electronics. Developments in progress for the slave arm focused on thefollowing directions:• minimize the impact of a leak in the hot zone,• improve safety of the control loop,• improve performance in force control mode,• reduce the tuning procedures time.A change of fluid from oil to water was proposed to reduce impact of leaks (see [1]). The use ofpressure controlled servovalves improved safety, force control performance, and tuning time. Drivingrequirements to adapt a joint to use water were:

• use corrosion resistant materials,• reduce clearances (direct impact on internal leaks due to water’s low viscosity),• prevent contact between water and components with poor corrosion resistance,• adapt seal materials and properties to water.In addition, attention was paid to control properties and quality of the water used during the trials.Tests were carried out with demineralized water on a single vane joint mock-up (see figure 2).Performance achieved with water is equivalent or even better than with oil.Figure 2: Water hydraulics test benchAt the time of this paper, the mock-up successfully ran during 530h before a failure of the power packoccurred, thus stopping prematurely the tests of the joint and the servovalve. Up to that level nosignificant reduction of the joint performance were noticed.Improvement of the force control loop is achieved by using pressure controlled servovalves to driveeach joint. In that scheme, the controlled parameter is directly linked to the force, and this has a directimpact on the stability of the control loop. Using these components also allows removal of all pressuresensors and therefore reduces the probability of failure of the system. Prototypes of oil pressurecontrolled servovalves with space and performances requirements needed by a Maestro manipulatorwere developed in collaboration with In-LHC. Integration of all servovalves in the slave arm provedthe feasibility of the concept. Performance was better than observed with flow servovalve and areduction of the time period of the control loop by a factor of two was possible.3. Real time failure detectionTo detect possible failure or collision with the environment, developments of model-based monitoringstrategies were tested (see [2]). The global dynamic model of the arm developed for such applicationstakes into account all inertia parameters of the arm, the centrifugal and Coriolis effects, gravitationaltorques, a contribution of the friction to each joint and offset values of the pressure sensors.Identification of all parameters is achieved with help of numerical regression methods. Thirty fivetrajectories were defined and used to identify all parameters of the model. Descriptions of all

trajectories were made according to the kinematics capabilities of the arm and real remote handlingtasks.Comparison between the estimated torque (model) and the real torque (sensors) is used to detectcollisions of the arm with the environment. Torque values need to remain within boundaries of theresidual error identified for each axis. If not, a collision (or failure) is detected.Figure 3: Collision detectionResults showed good agreement between predictions and measurements during the trials (see figure 3)although some improvements were necessary for low speed movements due to noise on speedevaluation, under-estimation of friction forces and hysteretic phenomena. Implementation of low passfiltering of the signal and definition of threshold values were necessary to overcome this trouble.4. Graphical supervisionCEA-LIST developed a 3D graphical supervisor to ease control in real environment of teleoperatedsystems such as the Maestro [3][4]. It has been assessed by operators on the prototype of the system,with indirect viewing of the environment. A complete set of maintenance tasks (cutting, welding,grinding...) were carried out with this tool.Teleoperation and especially decommissioning intervention tasks are very stressful for an operatorsince he must interact in real time with the environment dealing with complex systems. Remote tasksare not repetitive and usually undefined since work depends on observations during the interventions.Most of the time, the operator does not have direct viewing of the operating area.This supervisor (named MAGRITTE) was developed to ease operator's task during intervention. Themain requirements consist in letting the operator interact with the environment through a 3D model ofthe workplace. Through this interface, operators elaborate robot trajectories, play them and controltheir execution in an intuitive way. MAGRITTE offers graphic assistances, making easy robotsprogramming and control. It interfaces to robots and tools through an execution controller, allowingupdating model state according to the real situation. In MAGRITTE, specific processes functions(welding, ultrasounds inspection, grinding) are gathered in dedicated trade modules.

Moreover, collision test algorithm warns the operator, preventing collisions with the modeledenvironment. This feature is called passive anti-collision but can only be used to prevent collision(warns the operator or stops execution of the task) and so may not be of any help in case of use in veryconstrained environment.5. Assisted teleoperation for enhanced teleoperation taskMore recently, progress has been made in computing power and virtual reality makes it now possibleto interact with an accurate mechanical model using a haptic device. Main applications are currentlyfor training [5][6] and digital prototyping (accessibility, maintenance).Concerning the teleoperation field, the main benefit of using a physics engine with a teleoperatedsystem consists in providing the operator with an active collision avoidance feature [7][8].A prototype of a supervisor has been studied on the basis of MAGRITTE concepts and haptic softwarelibrary developed by CEA-LIST named XDE (eXtended Dynamic Engine). This function differs fromthe passive anti-collision feature as it ensures collision avoidance by generating repulsive efforts, thusmaintaining execution of the teleoperation task. At that time, the real robot is coupled with a simulatedone and the operator uses a 6 DoF mouse to generate movement on the simulation and the robot.Because the simulation not only detects the collisions but also simulates the robot behavior, theoperator can then handle the robot safely, without taking care of obstacles of the environment.Figure 4: passive collision detection

Figure 5: active collision avoidance with force feedback (future development)Future work will be done to link the system with a force reflective master arm, and then provides theoperator with force feedback on simulated collision. This functionality will allow using themaster/slave system in a real force feedback mode, preventing collision of known, modeled objects.The issue of localization of objects will also be addressed.

6. References[1] G. Dubus et al. Assessment of a water hydraulics joint for remote handling operations in thedivertor region. 8 th International Symposium on Fusion nuclear Technology. 2007 Heidelberg.Germany[2] C. Bidard et al. Dynamic identification of the hydraulic MAESTRO manipulator – Relevancefor monitoring. 23rd Symposium On Fusion Technology. 2004 Venice Italy[3] Gravez P., Leroux C., Irving M., Galbiati L., Raneda A., Siuko M., Maisonnier D. , Palmer J.“Model-based remote handling with the Maestro hydraulic manipulator”, “22nd Symp. onFusion Technology”, Helsinki, Finland, Sept. 2002.[4] Masson Y, Gravez P, Fournier R,A graphical supervision concept for telerobotics, Int. Symp. on Robotics, 1998.[5] Kühnapfel U., Çakmak H. K., Maass H., Waldhausen, S.:Models for simulating instrument-tissue interactions.9th Medicine Meets Virtual Reality 2001 (MMVR 2001), Newport Beach, CA, USA, Jan. 23-27, 2001 (2001)[6] Çakmak H. K., Kühnapfel U., Bretthauer G.;Virtual Reality Techniques for Education and Training in Minimally Invasive SurgeryProceedings of VDE World Micro Technologies Conference MICRO.tec 2000[7] Brooks, T.L.; Ince, I.;Operator vision aids for telerobotic assembly and servicing in spaceInternational Conference on Robotics and Automation, 1992. Proceedings., 1992 IEEE[8] Hernando, M.; Gambao, E.; Pinto, E.; Barrientos, A.;Collision control in teleoperation by virtual force reflection. An application to the ROBTETsystemInternational Conference on Robotics and Automation, 1999. Proceedings. 1999 IEEE

CANBERRA solutions for source location and activitydetermination for investigations of ORCADE dismantlingproject at AREVA NC La Hague siteAuthors: H. TOUBON 1 , C. CORDIER 2 , B. FERET 3 .1AREVA/CANBERRA, 1 rue des Hérons,78182 St Quentin Yvelines Cedex, Francehtoubon@canberra.com2 AREVA NC Etablissement de La HagueProjet ORCADEccordier@areva.com3AREVA/CANBERRA,50500 Beaumont Hague, Francebferet@canberra.comAbstractThe operations of dismantling of nuclear installations are often difficult owing to the lack ofknowledge about the position, identification and radiological characteristics of residualradioactivity.As part of the investigation team for the ORCADE dismantling project at AREVA NC LaHague UP2-400 reprocessing plant, CANBERRA had the opportunity to use new nuclearmeasurement systems and modeling tools to develop a methodology to locate and characteriseradioactive hold up.The methodology uses different types of dose rate meters, gamma spectrometers andmodelling tools, depending on the type of problem.In this paper, the methodology and tools are described, followed by typical case studies,including hot cell activity evaluations and drum categorization methodologies. For each case,the detector and modelling tool choices are explained and justified.This methodology helps to prepare and execute post-operational clean out and dismantlingactivities. These examples give concrete insights into their significance and the productivitygains they offer.Keywords: D&D, Gamma spectrometry, radioactivity modeling, source location, sourceidentification, MERCURAD, PASCALYS, ISOCS, CARTOGAM

1. CONTEXT OF D&D MEASUREMENT ACTIVITIESDismantling a nuclear installation is often difficult due to the lack of knowledge about theposition, the identification and the radiological characteristics of the contamination. In thatway, the contamination is particularly difficult to define in a significant global backgroundwhen the activities are relatively high. For example, identification and estimation of theactivity become more complex in hot cells, where space is limited and human intervention iscostly in terms of accumulated dose.CANBERRA, the Nuclear Measurement Business Unit of AREVA, not only designs,manufactures and sells a complete range of instruments, but provides solutions and services totake care of the global problem depending on customer needs. In that way, the followingitems have to be specified:- definition of the needed investigations,- detector choice,- dose rate modeling,- coupling dose rate measurement and model,- coupling gamma spectrometry measurement and model,- coupling neutron measurement and model.To be able to propose the best solution, CANBERRA has developed knowledge andintervention strategies based on its feedback experiences in many countries.Each part of the scene characterization is described in the following chapters. Thanks to thismethodology, CANBERRA is able to give the customer not only measurement results, butalso activities, localization and eventually to guaranty safety or process thresholdcorresponding to the customer’s needs.2. ORCADE INVESTIGATION PROJECT STAKES2.1. GeneralitiesThe ORCADE UP2-400 dismantling project at AREVA-NC La Hague began in late 2002when all new facilities in UP2-800 had started. The project concerns all the facilities of thefirst reprocessing plant: from dissolution to U, Pu and fission products (FP) extraction andstorage in tanks. To support all dismantling sub-projects, a transversal investigation projectwas created in 2005.2.2. Stakes of the Investigation ORCADE’s projectThe Investigation project support the different missions of the global ORCADE project whichare :1. Definition of dismantling scenario,2. Design of waste packaging installation,3. Good waste categorization,4. Radioactive discharges optimization,5. Safety analysis (dose rate, criticality…)

To be successful, the investigation project needs many skills and means as:- video investigation capabilities,- documentary research,- laboratory analysis,- nuclear measurements and modeling.Inside the investigation project CANBERRA is in charge of this last item.3. NUCLEAR INVESTIGATIONS IN A DISMANTLING PROJECT3.1. Necessary data retrievalThe first phase of nuclear measurement investigation is crucial. A facility prepared todismantling has usually a long history of operations with numerous events which haveoccurred its lifetime. The knowledge of these events is very important in the definition ofmeasurement strategy. These first assumptions usually come from interviews with the facilityprevious operators, allowing definition of a first model. A good expectation in an activityevaluation is already to take in account right information (detector choice for example).3.2. Scene modellingFor a high activity cell, a first model evaluation can save a lot of money for the investigationphase. From large activities panel assumptions and geometry descriptions of the scene,sources would be able to be placed in the model and dose rate evaluation would bedetermined as a range of magnitude, which will define the type of investigation:- Is the staff able to enter the cell?- Possibility to introduce nuclear measurement?- Logistics required for nuclear measurements?For gamma emitters, CANBERRA usesMERCURAD application based onMERCURE V6 (reference [1]) gammaattenuation code developed by the FrenchAtomic Commission at Saclay(CEA/SERMA). This code is used but alsosold by CANBERRA. It has a very convenientinterface which can be used by a technician formodelling a complex scene within about 30minutes. Figure 1 summarizes this first part ofthe methodology.Figure 1MERCURAD dose rate evaluation codeFor neutron emitters, only MCNP is used because no simplified code for 3D geometries existsyet.

3.3. First measurements to confirm hypothesisThe first evaluation of dose rate will allow selecting detectors, electronics and associatedshielding and/or collimators. For very complex problems (which concerns mainly highactivity cells), several iterations are needed. Simple measurements have to be done first,before complex measurement on model and first hypothesis refining.4. MEASUREMENT POSSIBILIES4.1. Main solutionsThe measurement solutions are very numerous. Usually only gamma and neutron detectorsare used. Alpha and beta emitters are detected by their gamma emission, which explains whythe ratio between gamma and alpha has to be known in the case of total gamma counting(without spectrometry). This ratio can have a deep impact in waste categorisation. Gammaspectrometry measurement will be preferred when this ratio cannot be easily identified orwhen the gamma emitters are numerous.4.2. Total counting measurement4.2.1. Gamma measurementIonization chambers are usually chosen because of their sensitivity and their ability to be usedin very hot cells. In that case, the electronics will be placed away from the detector andcurrent amplification electronics will be preferred.Geiger-Müller dose meters can also be used. Such detectors can be very small and are veryuseful when there is little space to introduce the detector in the hot cell.4.2.2. Neutron measurementBecause of their sensitivity, 3 He tubes are preferred. But they are also sensitive to gamma. Inthe case of high gamma emission, BF 3 detectors will be chosen. For very high neutronemission, fission chambers can also be used. All these detectors can only detect thermalisedneutrons. If the neutron spectrum in the cell is not a thermal spectrum, the neutrons will haveto be thermalised, usually with polyethylene blocks which increase the space needed aroundthe detector.4.3. Source location: CARTOGAM possible useThe CARTOGAM (gamma camera) will be used (see reference [2]) when the location is notknown and when it is very important to know the radioactivity’s location in order to definethe dismantling scenario. This tool can drastically decrease the number of neededmeasurements.The main advantage of the CARTOGAM system compared to others is that it superimposesvisible and gamma images using the same optics (see figure 2).

Figure 2: CARTOGAM measurement head with its PCand scheme of the image processingThe use of CARTOGAM will also simplify the need for sampling. The localization of theneeded sample is made easier as the sample contains radionuclides. Laboratory analysis onthese samples will determine all radionuclides and specific ratios.To avoid expensive laboratory analyses, gamma spectrometry measurements can also be donein the field.4.4. Source identification: use of NaI, CZT, or Ge detectorsWhile gamma spectrometers are useful for source identification, NaI detectors are a cheapersolution, with a poorer resolution (

The last and best solution is to use germanium detectorswith a resolution usually around 0.2%. This is thesolution for very complex spectra. This type of detectorcan determine the isotopic composition of U and Pu.CANBERRA developed BEGe (Broad EnergyGermanium) detectors for 15 years. Such a detectorcoupled with an ISOCS modelling code (see reference[4]) is a very versatile solution. CANBERRA alsoproposes a wide range of Ge detectors for specificproblems.Figure 5: Cart for Ge detectorand associated collimator5. SCENE MODELLING SOLUTIONS5.1. Why do we model the scene?Detectors give counts per second. The modelling of the scene with sources locationassumptions, allows determining not only counts, but also proportions, radionuclides andactivities.The use of CARTOGAM before modelling can help to make better assumptions in sourcelocation and drastically decrease the uncertainties. The following figure illustrates thismethodology.Figure 6: Methodology for activities determining

The determination of geometrical efficiency and detector efficiency allow determination ofactivities from a spectrum in counts per second via the following formula:M ( cps)A(Bq)=εgeom( γ / Bq)× ε det( cps / γ )Where: A is the activity in BqM the measurement of the peak net area in counts per second (cps)e geom the geometrical efficiency in gamma rays per Bq or g.cm -2 .s -1 / Bqe det the detector efficiency in cps per gamma ray or cps / (g.cm -2 .s -1 )Different codes can be used as gamma ray attenuation calculation. The use of the code isdiscussed below.5.2. Which model and calculation to be used?A very precise model is not often needed. The model has only to take into account the objectsand layers between the assumed sources and the detector. Often the main uncertainties comefrom not knowing the layers thickness between the source and the detector. In that case a veryprecise code is really not needed.Nevertheless, a more precise code can be useful when the scene is known more precisely,when the geometry is complex, or when there is a need to reuse the geometry model for otherapplications. This case often occurs for hot cells modeling. Where high activities areconcerned, the facility’s personnel issues are quite important and the time for modeling issignificant.5.3. ISOCS use for easy modellingISOCS gamma code attenuation is the best compromise for simple scene. The user interfaceuses templates which cover the most common geometries. With this approach, simplegeometries can be modelled by technicians in less than 15 minutes. It allows the model to bedone in the field during the measurement.5.4. MERCURAD-PASCALYS advantageWe have seen that MERCURAD is a complete 3D code for dose rate evaluation from anykind of gamma spectrum and activities. It can also be used to calculate activities from ameasurement spectrum. This way to use MERCURAD is called PASCALYS.The main advantage of PASCALYS compared to ISOCS is its modeling capabilities. Another advantage concerns the detector efficiency. As PASCALYS considers the detector as apoint, any kind of detector can be used easily with this modeling approach.The third advantage concerns the possibility of reusing the modeling scene. In a hot cellcomplete study, the geometry model can be used for other applications as health physicsstudies or dismantling scenarios.

Figure 7: Multi detector in complex geometry modellingWith the MERCURAD-PASCALYS tool, a complex geometry can be generated and stored ina database. A complex geometry can be used by PASCALYS to determine activities from onefield measurement, and then reused with MERCURAD to determine the activities dose rate.Objects can be easily removed from the scene (simulating dismantling activities) and doserates recalculated.5.5. General synthesisDuring repairing and dismantling activities, multiple types of detectors can be used. As thesource activities determination is very complex, CANBERRA proposes different approachesdepending on customer environment. CANBERRA is able to propose a large range ofsolutions.Therefore the model used in the activity determination can be reused when dose rates have tobe determined. It provides a consistent tool for engineering studies and dismantling scenarios.

6. ORCADE PROJECT EXAMPLES6.1. Calibration and quality insuranceInside the AREVA-NC investigation project, CANBERRA has deployed the previousdescribed tools and methodology to characterized different hot cells in the following facilities:- HAO : High Activity Oxide facility where fuels are previously sheared and dissolved,- HADE : High Activity facility where Decanting, fission product and actinidesExtraction are performed,- HAPF: High Activity facility where Products issued from Fission are stored.All the measurement instruments used for ORCADE project were previously calibrated at anirradiator facility (at AREVA-NC La Hague or at CANBERRA Loches irradiator). Both areCOFRAC certified. Technical notes, procedure, and storage of scene modelling insure acomplete tracability of the results6.2. Activities determination in high activity hot cellsThe following methodology is performed by:- Using laboratory sampling analysis to determine gamma & beta emitters and ratios,- Using equipment geometrical data as model and associated assumptions concerningliquid or sludge volumes,- Confirmation of sample analysis by dose rate measurements via very thin GM tubesand CZT gamma spectrometry measurement,- Model via MERCURAD code and MCNP to confirm MERCURAD results whengamma scattering effects occur,- Determination of transfer function at each measured point (Gyh -1 /Bq)- Quadratic minimization of differences between dose rate measured values and modelresults.For example, the modelling of a very complex cell in the HAO facility was performed (seefigure 8). After fitting the 44 dose rate measuring points to MERCURAD dose rate results(from few mGy/h up to 8 Gy/h), the total beta activity was estimated at around 45 TBq.Figure 8: Modelling of one complex HAO cell

A second example concerns the activity evaluation of tanks containing fission products.The only probe to be introduced in this cell was the very thin CANBERRA Geiger Müllerdetector (mounted inside an ø < 8 mm thermowell). These GM tubes where previouslycalibrated in an irradiator to establish their response up to 200 Gy/h.Figure 9: GM dose rate measurements on 2 tanks of fission productsMERCURAD modelling was performed to determine the total beta activity, establishedaround 2 300 TBq and an uncertainty about 50 %.6.3. Measurement in Uranium Middle Activity facility (MAU)In the Middle Uranium Activity facility (called MAU), the stake of ORCADE project was todetermine when the different tanks and equipments are considered as sufficiently rinsed.The aim it to make the equipments compatible with the very low activity wastes specification(called TFA). For Uranium, the TFA limit is about 100Bq/g of final waste. Consequently theORCADE project asked CANBERRA to achieve a methodology to classify the equipmentsaccording to 3 different categories:- TFA waste- Not surely TFA waste: equipment to be investigated- Surely not TFA: equipment to be rinsedThe equipments were differentiated according their geometry and weight:- Light ( < 500 kg ),- Intermediate ( from 500 kg to 1 Ton ),- Heavy ( > 1Ton ).The equipments were modelled with MERCURAD and the measure was performed insidethe facility.To types of spectrum were achieved:- almost Uranium spectrum,- almost Cs137 spectrum.

So, it was possible to categorize the waste:Figure 10: Differentiation between TFA, undetermined and non TFA wasteFor light equipment, it’s very difficult to distinguish TFA and non TFA equipment as the limitaround 80nGy/h which is nearly the background. For large tank the limit value of 200nGy/h iseasier to measure.These results which are preliminary, have to be correlated with more precise gammaspectrometry Ge detectors in situ and on final drums.7. CONCLUSION AND PERSPECTIVESCANBERRA proposes a complete set of solutions for radioactivity location anddetermination. According to the problem facing the nuclear facility, different measurementsand modeling approaches can be used. Many tools have been used for ORCADE dismantlingproject in La Hague to help the investigation team and increase the knowledge of theradioactive contents of the different facilities.Within the ORCADE project, CANBERRA contributes to the three main drivers inaccordance with sustainable development goals :- Environmental aspects : reduction of the toxicity of the waste- Finance : optimization of the global project cost by contributing to the planning ofdismantling scenarios- Social aspect : optimization of individual radiation exposure by detailed preliminarymeasurements facilitating use of ALARA methodologyNew waste categorization challenges await CANBERRA at AREVA-NC La Hague site. Sucha categorization doesn’t consist only in measuring the final waste at the end of dismantling,but also to help project teams during the realization phase. The target is also to provide a goodlevel of decontamination and to optimize the final cost of dismantling.

8. REFERENCES[1] Ali ASSAD, Maurice CHIRON, Jean Claude NIMAL, Cheikh M'backé DIOP andPhilippe RIDOUX : "General Formalism for Calculating Gamma-Ray Buildup Factorsin Multilayer Shields into MERCURE-6 Code", Nuclear Science and Technology,Supplement 1, p. 493-497, March 2000.[2] O. Gal et al., "CARTOGAM – A portable gamma camera for remote localisation ofradioactive sources in nuclear facilities", Nuclear Instruments and Methods A.[3] J.F. BREISMEISTER, “MCNP (4C) A General Monte Carlo N-Particle TransportCode”, report: LA-13709-M, Los Alamos National Laboratory, Ed. (2000).[4] F. L. BRONSON and B. M. YOUNG, “Mathematical Calibrations of Ge Detectors andthe Instruments that use them”, Proceedings of 5th Annual NDA/NDE WasteCharacterization Conference, Salt Lake City, UT, Jan 11, 1997.[5] H. TOUBON, K. BOUDERGUI, P. PIN (CANBERRA), B. NOHL (EDF), S.LEFEVRE, M. CHIRON (CEA), New methodology for source location and activitydetermination in preparation of repairing or decommissioning activities IRPA 2006 15-19 may 2006 - Paris - France

APPROACH TO RADIATION MONITIRINGAT KUDANKULAM NPPN.V. RYZHOV, E.G. CHANYSHEV, D.A. PLYSHEVSKAYAPROM Engineering29/1 Bolshaya Ordynka Str., 119017 Moscow – RussiaABSTRACTNowadays a large experience in creation and operation of automated radiation monitoringsystems (ARMS) for NPPs with WWER type reactors has been gained. ARMS developedtoday shall meet more strict requirements of monitoring efficiency in accordance withissued new radiation safety standards and shall have better indices of monitoring efficiencyand reliability. ARMS of new generation must take account of the past wide experiencepertaining both to the selection of the monitoring scope and to the arrangement of thefunctioning scheme.This paper presents the main approach to ARMS development by the example ofKudankulam NPP, which meets the most modern RF safety and reliability requirements. Itsprinciples may serve as a basis for both new power units construction and modernization ofradiation monitoring systems at nuclear power plants.1. IntroductionIn accordance with the Russian practice the NPP radiation safety is provided by the AutomatedRadiation Monitoring System (hereinafter referred to as ARMS). The aim of ARMS development is toobtain on-line comprehensive information of the radiation situation and process equipment condition.The system monitoring data shall confirm the fact that the radiation impact on the personnel andpopulation from the NPP side is within the specified limits.This paper considers the peculiarities of ARMS developed by PROM Engineering in the framework ofKudankulam NPP construction in India. ARMS has been developed on the basis of valid in RFnormative documents which regulate the requirements of the personnel and population protection fromionizing radiation impact as well as the requirements of radiation monitoring arrangement, namely:- Standards of radiation safety NRB-99;- Sanitary rules of nuclear power plants design and operation SP AS-03;- General provisions of nuclear power plants safety assurance OPB-88/97.Moreover, specific requirements of Indian national standards of radiation safety have been taken intoconsideration during the system development.2. Monitoring scopeARMS is a system of radiation safety at the NPP monitoring which allows damage to protectivebarriers to be identified at an early stage and prevents the penetration of radionuclides into theenvironment. The information exchange between ARMS and the Automatic Process Control System(hereinafter referred to as APCS) enables continuous analysis of the NPP state to be carried out andfailures of the main process equipment to be predicted.In the framework of the united ARMS system of new generation a complex solution of all tasks ofradiation monitoring at the NPP is provided. These tasks comprise the following kinds of monitoring:1. radiation monitoring of processes including:

o protective barriers (fuel elements claddings, primary circuit equipment, steamgenerators, secondary circuit equipment, containment);o radioactive media leaks in process equipment;o efficiency of gas treatment, water treatment and ventilation systems;2. monitoring of the radiation situation at the power unit and at the Site;3. monitoring of gaseous and particle releases into the atmosphere;4. monitoring of radionuclide releases into the open aquatic environment;5. monitoring of the spread of radioactive contamination at the NPP site and outside theNPP;6. monitoring of collective and individual doses received by the personnel.The crucial role of safety assurance is played by measuring channels which trace the main parameterschanges in continuous automatic mode. These very channels are capable to provide prompt assessmentof the radiation situation in case of emergency situation development.ARMS combines both continuous (remote) and periodical monitoring functions which are mutuallycomplementary. Periodical monitoring consists of measurements made using mobile and portableinstruments and it includes also manual sampling and spectrometric or radiometric analysis of samplesunder laboratory conditions. The automation of measurements data obtaining, storage and processingby means of specific programmes package is provided in ARMS for the purpose of periodicalmonitoring.One of the differences of ARMS for Kudankulam NPP from other similar systems is the integration ofAutomated Personal Dosimetry Monitoring System (APDMS) into ARMS with the uniform software.During the development the high emphasis was placed on the fact that radiation safety servicepersonnel can receive detailed information at the same time about individual doses and the radiationsituation in different NPP rooms and about the process equipment state. It will improve the radiationsafety service work and will allow to reduce radiation doses caught in the course of radiationhazardousactivities. The automation of measurements as well as the implementation of a set offunctions pertaining to calculation and planning of the individual doses for the personnel, accountingof the time of radiation-hazardous activities implementation, issue of radiation work orders and othershas been made for APDMS.ARMS allow to extend its functional possibilities and, for example, to include the following kinds ofmonitoring:- monitoring of the radioactive contamination of the environment;- radiation monitoring of processes related to radioactive wastes management;- monitoring of the meteorological situation.Thus at Kudankulam NPP the radiation situation monitoring in the observation zone is performed by aspecific automated system (ARSMS in accordance with the Russian practice) using stationarymonitoring posts. Then monitoring data are transmitted to ARMS for the purpose of their analysis andstorage as well as to obtain the prediction of the radiation situation and assessment of radiation dosesof the population in the NPP area in case of emergency situation.Hence the united automated system covers all three main objects of monitoring as follows:a) nuclear power plant (processes);b) the personnel and population living within the NPP location area;c) the environment within the NPP location area.This scale and heterogeneous scope of monitoring requires special approach to the system creation: itsdesign and the software development.3. Structure

ARMS has a distributed hierarchical structure. ARMS structural diagram for the nuclear power plants withtwo power units is provided in Figure 1. Its lower level is a set of measuring channels consisting ofindividual monitors or detection units and data processing and transmitting units which are united by datacommunication lines. Lower level devices ensure the measurement of the current values of the parametersto be monitored as well as local alarming in case the monitored parameters exceed the preventive oremergency limits. Besides, the especially important parameters preventive or emergency excess alarms arereplicated directly from lower level devices on safety panel and generalized mimic panel installed in theUnit control room.Upper level consists of concentrators made on the basis of industrial controllers, automated workplaces ofoperators (hereinafter referred to as AWP) and database servers. The upper level hardware carries out dataaccumulation and processing, recording into database, archiving and presenting the results of monitoring tothe radiation safety service. The concentrators inquire by cycle the lower level measuring channels viacommunication channels RS-485, obtain the current data from them and process this data up to the unifiedaccepted format ready to be transmitted to AWP. In addition the concentrators collecting data ofradionuclides release into atmosphere through the NPP ventilation stacks calculate total release ofradionuclides into atmosphere per day and compare the day release with the control level.AWP makes regular diagnostics of the software and hardware. The equipment right up to individualdevices, units and modules of the software are subject to diagnostics. ARMS and APCS data exchange isperformed via concentrators (interface Ethernet 100Base-FX).Power Unit 1Protected roomfor post-accident activitiesPower Unit 2ULCS (APCS)Emergency controlroom for 1 UnitEmergency controlroom for 2 UnitULCS (APCS)Unitcontrol roomUnitcontrol roomLAN - Upper Level Control System for Power Unit 1LAN - Upper Level Control System for Power Unit 2Detectionunits/devicesfor ERMConcentr.ERM-1Concentr.ERM-2DD / DUConcentr.ERM-3Concentr.ERM-4Detectionunits/devicesfor ERMDetectionunits/devicesConcent. RPMConcentr.Vent.SDD / DUConcentr.RMPConcentr.Vent.StDD / DUConcentr.Vent.StDetectionunits/devicesDetectionunits/devicesConcentr.Vent.StDD / DUConcentr.Vent.StConcentr.RMPDD / DUConcentr.Vent.SConcent. RPMDetectionunits/devicesDetectionunits/devicesfor ERMConcentr.ERM-1Concentr.ERM-2DD / DUConcentr.ERM-3Concentr.ERM-4Detectionunits/devicesfor ERMАWPRM-1АWÐIROАWPIROАWPRM-2LAN - APDMS - 4N safety classLAN - APDMS - 4N safety classLAN - 3N safety class - ARMSLAN - ARMS - 3N safety classLAN - ARMS - 3N safety classCommon buildings andthe SiteGatewayАWÐDDАWÐ in centralroom for RMARMSServerConcentr.RMP (Site)Detectionunits/devicesNotationARMS Automated Radiation Monitoring SystemAPDMS Automated Personal Dosimetry Monitoring SystemAPCS Automatic Process Control SystemDD/DU Detection devices/unitsLAN Local Area NetworkAWP Automated Workplaces:WBC Whole Body CounterPDM Personal Dosimetry MonitoringАWÐWBCRMLaboratoryАWÐPDMAPDMSServerDD Dosimetrist on dutyIRO Issue of radiation work ordersRM Radiation MonitoringInterfacesConcentrators:ERM Emergency Radiation MonitoringRS-485100 Base-TXRPM Radiation Process MonitoringVentS Ventilation SystemsVentSt Ventilation Stacks100 Base-FXRMP Radiation Monitoring of Premises (Site)Fig 1. Structural diagram of ARMS for the nuclear power plant with two power unitsTwo-level structure taken as the system basis makes it possible to improve the system reliability. Thecrucial point of the structural reliability improvement is the lower and upper levels autonomy (selfcontainedoperation). Due to this fact the monitoring is going on to be performed in case of the upperlevel equipment failure or communication lines loss, since the alarms of preventive and emergencylimits excess are transmitted to safety panel of Unit upper level control system (ULCS), and the resultsof measurement are displayed at the lower level devices displays and are saved into memory.

All AWP are unified and if one of them fails monitoring functions are transferred to any othercomputer complex. It is provided at the expense of links unification and the software proper structure.Only unified self-contained measuring channels with standard interfaces for all hardware are used inARMS. Taking into account the fact that 65% of total number of the measuring channels are used forgamma radiation dose rate monitoring a wide-range channel (covering eight decimal exponents)common for all NPP operation conditions that is for normal operation and for emergency situation – isused in the system. It further contributes to the system manufacture, adjustment and cost.In the course of design process a differential approach was used according to which the distribution ofall ARMS tasks between its sub-systems is determined by their relation to safety (classification as perthe Russian standard OPB-88/97). Sub-systems and hardware included into sub-systems are dividedinto safety-related normal operation elements (safety class 3N) and not-safety related elements (safetyclass 4N). A number of increased requirements of resistance to external impacts, electric powersupply, electromagnetic compatibility, error-free running time and etc is set for the equipment of 3Nclass. And as well a redundancy of power lines and communication lines shall be fulfilled for 3N classequipment related to the upper level. All these make ARMS design more complicated and set theadditional requirements to data exchange between sub-systems of different safety classes. Thereby thedata are transferred through interfacing gateway the role of which is played by the concentratorconnecting local area network of safety class 3N and 4N.Radiation monitoring system shall stay operable under all NPP operation conditions including designbasisaccidents, so ARMS includes independent sub-systems of normal operation and emergencycontrol. Emergency monitoring systems must continue functioning in case of any design-basisaccident, they are made with the double redundancy of the measuring channels, concentratorsconnected with them, as well as of communication and power supply lines.This flexible structure of hardware and software makes it possible to modify the system configurationand functioning conditions, if required. The same flexible structure allows ARMS to provide radiationmonitoring during the NPP decommissioning.Another problem is connected with ARMS service life. During the nuclear power plant wholeoperation life this system must stay serviceable within at least 30 years. Taking into account the factthat specific hardware service life is about 10 years, their future modernization will be required to bemade without damaging the whole system operation. The way out has been found in using thehardware manufactured only at the series enterprises. These enterprises carry out the hardwaremodernization activities without any violation of the unified interface accepted in ARMS. Theequipment manufacture is under supervision of the Federal Service for environmental, process andatomic surveillance (Rostechnadzor). It guarantees the high quality of the manufactured products.4. Software and datawareARMS software has a modular structure and consists of self-contained modules thus enabling thesystem tasks composition to be modified and expanded.The upper level software has the “client-server” architecture with the ARMS common databaselocation at the server. The software has been developed as a multi-purpose client application whichfunctions at AWP under control of the operational system Linux. Since the AWP get dataindependently of each other and work on the basis of the multi-purpose client application, they arefunctionally interchangeable thus contributing to the system viability.

Fig 2. Generalized video frame of theradiation situation at the NPP with two powerunitsFig 3. Video frame of the ventilation systemof reactor buildingFig 4. Video frame of complex monitoring ofgaseous and particle releases intothe atmosphereThe basis of the user’s interface development makes the perceptibility of the information by operators.The graphic interface is realized in the form of multiimage application using modern approaches to thedata presentation. ARMS makes it possible for the operator to monitor the general condition of all thesystems covered by the radiation monitoring, to obtain the detailed data for specific parameters andsub-systems. As an example, Figures 2 to 4 provide the patterns of video frames.5. ConclusionARMS presented in this paper is a new generation system developed for Kudankulam NPP taking intoaccount the long experience of the similar systems creation for all Russian power units with reactors ofWWER type. Due to the use of the modern measuring and computer equipment, application of theflexible two-level architecture, optimization of the monitoring scope and complex approach to thesolution of all tasks of radiation monitoring at the NPP nowadays this system is one of the advancedproducts in the world in the field of the NPP radiation safety assurance.

MONITORING SYSTEMS FOR HOT SPOT DETECTION INRADIOACTIVE WASTEGEROLD G. SIMON, MARINA SOKCIC-KOSTICNukem Technologies GmbHIndustriestrasse 13, 63755 Alzenau, GermanyABSTRACT.Radioactive waste as arised in nuclear power plants or similar installations is classified indifferent categories according to its radiological content and radiation emission rate. Thecomponents of the waste, which are small in mass but have a high activity inventory, arecalled Hot Spots. If Hot Spots can be removed more waste can be shift in a lower categoryresulting in a more economical waste management. The classification range comprehendsfrom free release to high level waste which must be specially packed and stored in safefacilities for long time. The detection of Hot Spots requires specific instrumentation andmethods which must be adapted to the radiation intensity of the Hot Spots and the speed ofmeasurement. NUKEM Technologies GmbH has developed a series of new, economicalmonitoring systems for this purpose that are presented in this paper.1. IntroductionRadioactive waste as arised in nuclear power plants or similar installations is classified in differentcategories according to its radiological content and radiation emission rate. The classification rangecomprehends from free release to high level waste which must be specially packed and stored in safefacilities for long time.For the radiological classification different limits are defined:‣ Surface dose rates (i.e. dose rates at 100mm distance from the wastesurface)‣ Beta activity content (i.e. specific beta activity in units [Bq/g])‣ Alpha activity content (i.e. specific alpha activity in units of [Bq/g])‣ Isotope concentration with life time greater than 30 yearsIn general the cost for waste handling and storage is increased with increasing activity level.Therefore, an advanced waste sorting and handling strategy should be used that considers economicalaspects.The components of the waste, which are small in mass but have a high activity inventory, are calledHot Spots.If Hot Spots can be removed more waste can be shift in a lower category resulting in a moreeconomical waste management.Typical Hot Spots are debris from nuclear fuel elements, high activated metal pieces etc.

During the last decade NUKEM Technologies GmbH was highly engaged to realise nuclear wastetreatment centres and has therefore developed and tested different methods and devices for thisapplication.In particular NUKEM Technologies GmbH has developed such kind of measurement systems for:‣ Checking of excavated soil‣ Sorting of solid waste‣ Mapping the area of nuclear sites‣ Checking of packages before leaving the nuclear plantThe measurement systems are optimised in respect to pick out small particles without the lost of therequired throughput.2. Hot Spot detection for waste designated for release and/or landfillNUKEM Technologies has developed a Measurement System for Free Release Measurement (FMA)for waste from demolition of buildings and from ground. The waste is firstly scrapped to a size below60mm and then filled in batches of 1000kg onto a conveyor belt. Above the belt four HPGe detectorsare mounted. The detector spectra are analysed for each batch to measure the total activity of thewaste. The throughput is up to 50 tons per hour. Details of the installation can be found in reference[1].The online analysis of the total gamma spectra – especially in the case of very low gamma emission –is not suitable for Hot Spot detection. Therefore, the window analysis method is implemented: foreach detector the regions of interest inside the spectrum are specified, which are readout in a shorttime sequence (typically 15 second). This procedure can be performed without interruption of thenormal data collection for a complete spectrum during the passing of the waste batch.The regions of interest are placed inside the gamma spectrum in such a way, that they cover the areafor gammas from Co-60 and Cs-137. Some regions are placed in the neighbourhood of these lineswhere no gammas are expected and therefore, they can be used for background subtraction.This method allows a time resolution of 1 second for a hot spot of 1000 Bq in a free releasemeasurment. But in practice, the read out time has to be adapted to the intensity of the expected HotSpot. The minimum time must be large enough to reliably measure a Hot Spot.If a Hot Spot is detected, the conveyor is stopped, the Hot Spot is removed, the batch is transportedback to the starting position and the measurement is started again. This procedure guarantees thesuccess of the hot spot removing.An alternative is to separate the complete batch with Hot Spot to a different place for furtherinvestigation. In this case a higher through put is possible when many hot spots are expected.Normally, this special procedure is not needed.3. Hot Spot detection for middle and high active wasteFor the detection of Hot Spots in middle or high active waste – i.e. such waste type is managed in thewaste treatment centre ICSRM in Chernobyl- NUKEM Technologies has developed a scanner systemwhich has 3 CdZnTe gamma spectrometers, each mounted with its own collimator system. Thecomplete device is moved by a crane over the sorting table. Each scanner scans a stripe of 1/3 width ofthe table.

The spectra are read out every second using the window technique as described in previous chapter.The system is able to detect Hot Spots and to associate these to specific isotopes for differentiationbetween activated or irradiated fuel material.For better local resolution the gamma camera RAYMOS was added. [2]. The gamma camera takespictures in the visual wavelength range, as well as, from gamma ray distribution affected by isotopeslike Co-60 or Cs-137. After the measurement, both pictures are superimposed and delivered easilyinterpretable information about the location of Hot Spots.The Fig. 1 shows such a picture which was taken during the tests of the camera. The source was a Cs-137 with intensity of 2x10 8 Bq, which could be detected even covered with 8cm of steel. In theexample shown in Fig. 1 the source was covered by wood and plastic material.Fig. 1 Hot Spot detection by RAYMOSThis type of RAYMOS has a relative low sensitivity: The picture presented in Fig.1 was taken from a4m distance with an exposure time of 10 minutes. In this case a low sensitivity camera was used sinceit was applied for sorting of high level waste. The camera is based on the pin-hole principle.. Forapplication in the low radiation field NUKEM Technologies has developed a multi-aperture type ofRAYMOS which has 256 pinholes instead of one single pinhole. This enhances the sensitivity by afactor 10. In addition a special background subtraction system was developed and integrated thatfurther enhances the sensitivity.4. Hot Spot detection for large areas: the Groundhog TMFor mapping large areas outside of buildings, the monitor Groundhog TM [3] was developed (Fig. 2 andFig. 3). It is a combination of a gamma measurement system and a GPS (global positioning system).The measurement data and the local position are stored in the database in real time. When themeasurement is completed, an activity map of the terrain is available. Different types of gammadetectors can be used. The most advanced system is capable of guiding itself over large terrains.

Fig. 2 Ground SurveillanceMonitoring Systems Groundhog TMFig. 3 Advanced Ground SurveillanceMonitoring System Groundhog TM5. Hot Spot detection at drumsThis Hot Spot detection system has a different purpose: The drums are normally transported bycontainers. To transport the containers the regulations about the maximum allowed surface dose ratehave to be fulfilled. If the drums have Hot Spot, i.e. the dose rate is above the allowed limit in somepositions, Since the hot spot can be placed near the surface the drum does not fulfil the requirementswhen it is stored in the container. When the location of the hot spot is known the loading procedurecan take credit of this information to place the drum in the right position. A misleading can beavoided.This method is implemented in our Drum Monitoring System DMS [4]. The drums are placed onto theturntable. From the side, as well as, from top and bottom dose rate meters equipped with rollers arepressed against the drum. In this way the distance between drum and dose rate meters is fixed. Thedose rate meters are equipped with two Geiger Mueller counter tubes of different sensitivity. The doserate meters are read out every second and the measured data are checked for Hot Spots. Once a HotSpot is detected, colour is jetted on the drum to indicate the position of the Hot Spot (Fig. 4).6. ConclusionThe detection of Hot Spots requires specific instrumentation and methods which must be adapted tothe radiation intensity of the Hot Spots and the speed of measurement. NUKEM Technologies GmbHhas developed a series of new, economical monitoring systems for this purpose that are presented inthis paper.References1 G. G. Simon, R. Leicht, M. Sokcic-Kostic, Integrated Gamma Spectrometry System forAdvanced Decommisioning Procedure, ICEM 2005, Glasgow, UK, September 20052 G. G. Simon,M. Sokcic-Kostic, Raymos, An Easy to Handle, High Sensitive Gamma Ray Imager, JahrestagungKerntechnik 2006, Aachen, Germany, May 20063 G. G. Simon, M. Sokcic-Kostic, I. Auler, L. Eickelpasch, J. Betts, Measurement Systems inthe Area of Land Remediation and Soil Segregation Activities , ICEM 2007, Brugge, Belgium,September 2007

4 M. Sokcic-Kostic, G. G. Simon, R. Schultheis, Radiological Monitoring Systems for WasteTracking and Waste Characterizations during Reactor Decommissioning Procedure, 3rd NuclearIndustry Forum, Avignon, France, October 20062431Fig. 4 Drum monitor(1) Drum (3) HPGe detector(2) Dose rate meters (4) Hot Spot

DECOMMISSIONING OF AN URANIUM HEXAFLUORIDEPILOT PLANTI. SANTOS, A. ABRÃO, M. F. S. CARVALHOCenter of Chemistry and Environment (CQMA)Institute of Nuclear and Energetic Researches (IPEN-CNEN/SP)Av. Professor Lineu Prestes 2.242 - 05508-000 - São Paulo - SP – BrazilABSTRACTThe Institute of Nuclear and Energetic Researches has completed fifty years of operation,belongs to the National Commission for Nuclear Energy, it is situated inside the city of SãoPaulo. The IPEN-CNEN/SP is a Brazilian reference in the nuclear fuel cycle, researches inthis field began in 1970, having dominancy in the cycle steps pilot scale from Yellow Caketo Uranium Hexafluoride technology. The plant of Uranium Hexafluoride produced 35metric tonnes of this gas by year, had been closed in 1992, due to domain and totaltransference of know-how for industrial scale, demand of new facilities for theimprovement of recent researches projects. The Institute initiates decommissioning in2002. Then, the Uranium Hexafluoride pilot plant, no doubt the most important unit of thefuel cycle installed at IPEN-CNEN/SP, beginning decommissioning and dismantlement(D&D) in 2005. Such D&D strategies, planning, assessment and execution are described,presented and evaluated in this paper.1. ItroductionSeveral groups of chemist and engineer at IPEN-CNEN/SP consolidated the developed researches inthe facilities of the Uranium Conversion Project (PROCON), during more than twenty-five years,being responsible for the production of 35 metric tonnes of UF 6 by year, in the end of the 1980 decade.All PROCON plants comprised section from yellow cake to UF 6 production , such as the yellow cakedissolution operations, solvent extraction purification and the production of UO 3 , its chemicalreduction to and UO 2 followed by its fluoridation to UF 4 , generation of the elementary fluorine for thefluorination of UF 4 to UF 6 .The UF 6 Production Pilot Plant began operation at IPEN-CNEN/SP in 1986 and stopped operation in1992. It was projected and assembled for a nominal capacity of production of 20 kg×h -1 of uranium,content in natural uranium hexafluoride, expressed as 27 kg×h -1 of UF 4 or 30 kg×h -1 of UF 6 . Figures 1and 2 depict some of the PROCON facilities.Fig 1. Pilot Plant BuildingFig 2. Primary Crystallizer Inside PlantDuring the operational period, the PROCON staff developed and improved components jointly withthe national industry, and soon afterwards established know-how was been transferred to the company

associated of the project. The country gained capability in the fuel cycle know-how, and trained morethan two hundreds of technicians (engineers, physicist, technologists and others). The great majorityof the those trained professionals today apply their knowledge in the national nuclear establishmentsand in the private industry in a general way. Brazil has being recognized internationally as one of thefew countries in the world the complete domain of the nuclear fuelcycle (1,7). being recognizedinternationally as one of the few countries in the world with the complete domain of the nuclear fuelcycle (1, 7). The figure 3 shows a flowchart of the Uranium Conversion Project (PROCON), that wasoperational at IPEN-CNEN/SP.DUS Receivingand StorageDissolutionPurificationPure UranylNitrateDenitrationPrecipitationDUADUS – Diuranate SulphateDUA - Diuranate AmmoniumU 0 - Metallic UraniumCalcinationUO 3Movable Bed - UF 4ChemicalReductionUF 6Wet Vie - UF 4Warehousing U 0Fig 3. Flowchart of the Nuclear Fuel Cycle steps at IPEN (1).A search in the specialized literature pointed out several papers (2, 6, 8, 9, 10, 11, 13) covering thenuclear field decommissioning and dismantlement. Nevertheless there are not available open dataconcerning on decommissioning and dismantlement including nuclear material and dangerouschemical reagents, respectively uranium, hydrogen fluoride ,as that handled in an UraniumHexafluoride production facilility. The present study comprises the development of innovativemethodology for decommissioning and dismantlement (D&D) of the Uranium Hexafluorideproduction facility. The present study comprises the development of an innovative methodology fordecommissioning and dismantlement (D&D) of the Uranium Hexafluoride Production Pilot Plant, anintegrating unit of the Uranium Conversion Project (PROCON), set up at IPEN-CNEN/SP (14).2. Strategy

The decommissioning and dismantlement (D&D) of the UF 6 pilot plant comprised the Freon gas unitand the unit of Fluorine generation. This UF 6 plant remained out of operation for a range of 12 years(1993 to 2005). During this period all plant installations not received any maintenance. Severalcomponents and accessory of the plant had not received any maintenance. Several components andaccessory of the plant were denied and relocated, impeding its operating again. Consequently, wasimpossible to discharge the residual gaseous UF 6 , and any material contained inside of the connectionsand loading lines to crystallizers tank, making the dismantlement operation difficult and laborous. Forthis reason and because part of the equipments was exposed to the plant environment conditions, adetailed study was accomplished to verify the status of the equipment and to do an inspection in theUF 6 crystallizers and their internal parts. The possibility of finding Hydrogenfluoride (HF) condensedinside components of the fluorine generation section was evaluated as well.As a consequence of the long time out of operation, all the trained operation and maintenance staffwas moved from the UF 6 Plant to another tasks execution. Then, was necessary to contract an externalcompany specialized in equipments dismantlement. However the contracted company has notinexperience in nuclear area. The alternative was to capacitate and assisted the employees of thiscompany via an intensive course involving theoretical and practical lessons and training them inRadioprotection, handling of potentially dangerous products, (HF, UF 6 ) and also in the wearing ofspecial personal protection equipments, such as total supplied air respirator and suits apparatus, asshowed in the figure 9.3. Decommissioning and Dismantlement Plan3.1 PlanningThe planning for the execution of these dismantlement tasks and decommissioning of the UF 6 Plantwas submitted to a rigorous evaluation in the feeding pipes and discharge of UF 6 , due to the possibilityto find UF 6 (crystallized and gaseous) and residual UF 4 . After the preliminary study, several blanktests were performed in the blind flanges set up with PTFE (polytetrafluorideethylen) parts. A series oftests was run for the inspection of the wash column (absorption). In the case of gaseous UF 6 lines, itwas decided begin the dismounting operation starting with one of the secondary crystallizer, since thepossibility of to find any UF 6 was remote.3.2 DismantlementThe dismantlement of the sections that comprise the UF 6 Pilot Plant (elemental fluorine and freon unit)conforms to the planning above described, except small alterations concerning handling of thehermetic flanges, with PTFE connections, mainly when inside the pipes or equipments crystallizedUF 6 was found. As the chemical reaction for formation of UO 2 F 2 is very fast, it was necessary todevelop new techniques for replacement of each part of the lines, case to case. Figures 4 to 9 presentsthe steps relatives to dismantlement, transportation, storage and safety clothe used.3.3 DecommissioningThe decommissioning of this pilot plant included specific procedures for Radioprotection (3, 4, 5),Safeguards (12), Environmental Monitoring (6) and Individual Protection (3, 4, 5, 6, 14). Thedismantlement of the sections of elemental Fluorine and Freon gases were carried out according to theplanning above described, except small alterations concerning on the handling of the hermetic flanges,with PTFE connections, mainly in the case that inside the pipes or equipments was found crystallizedUF 6 .

Fig 4. Secondary crystallizerFig 5. Transport of the CrystallizerFig6. Washing ColumnFig 7. Column TransportationFig 8. Crystallizer StorageFig 9. Technical Staff3.4 Transport and Storage of the EquipmentThe choice and selection of an appropriate place for placement of the solitary equipments of the plantsdecommissioning it was very discerning, due to the dimensions of some equipments and the containedmaterial, the proximity of the place of the plant also weighed a lot in the choice, there was the need ofmaterial remove of some deposits, to find the most appropriate place.4. ConclusionsThe UF6 Pilot Plant was successfully decommissioned and dismantled, according to initial planning,without serious adjustments. The reasons that contributed to this success were mainly planning,training, capability and technical attendance.For the subcontracted company, the group training course was successfully, including theoretical andpractical classes in Radioprotection, handling of Hydrogen fluoride, wear of special individualprotection equipments (total supplied air respirator and suits apparatus with gas filters).The decommissioning and dismantling of the UF 6 Pilot Plant was successfully accomplished withshort financial costs, due to previous planning, direct accompanying and supervising of capable incharged group and personal responsibilities (14).

The decommissioning and dismantling of the first UF 6 Pilot Plant in the Brazil and probably in theworld was a very important experience. The problems encountered in the course of decommissioningprocess allowed IPEN-CNEN/SP to find solutions, enabling their technicians to face future challengesnot only in the others nuclear plants but also in NORM(natural occurring radioactive material) andconventional industries.The works related to UF 6 Pilot Plant decommissioning and dismantling ended in 2006. All materialswere decontaminated and stored.5 References1- ABRÃO, A. The Uranium cycle in the Brazil. São Paulo, 1994, IPEN-pub-398, (in Portuguese).2- BOING, L. E. ,NECHAEV, A. F. Decommissioning of Nuclear and Radiation Facilities. SaintPetersburg Institute of Technology, Russian, 2001.3- COMISSÃO NACIONAL DE ENERGIA NUCLEAR. Requisitos de Segurança e ProteçãoRadiológica para Instalações Minero-Industriais. CNEN, CNEN-NN-4.01 Norm, Rio de Janeiro, 2005(In Portuguese).4- COMISSÃO NACIONAL DE ENERGIA NUCLEAR. Diretrizes Básicas de Proteção Radiológica.CNEN, CNEN-NN-3.01 Norm, Rio de Janeiro,2005 (In Portuguese).5- COMISSÃO NACIONAL DE ENERGIA NUCLEAR. Controle of the Nuclear Material. CNEN,CNEN-NN-2.02 Norm, Rio de Janeiro, 1999 (In Portuguese).6- EUPOPEAN COMMISSION, Environmental Impact Assessment for the Decommissioning ofNuclear Installations, Rep .EUR 20051-Rev, Feb 2002.7- FERNANDES, H. R. S. M. Subsídios ao Descomissionamento da Primeira Indústria deMineração e Beneficiamento de Urânio no Brasil - Caso do Complexo Mínero Industrial dePoços de Caldas – MG. ScD Thesis, 1997, Universidade Federal Fluminense, Rio de Janeiro (InPortuguese).8- INTERNATIONAL ATOMIC ENERGY AGENCY.Record Keeping for the Decommissioning ofNuclear Facilities: Guidelines and Experience. IAEA, Technical Reports Series Nº 411, Vienna20029- INTERNATIONAL ATOMIC ENERGY AGENCY. Safe Decommissioning for nuclearactivities proceedings of an international conference. IAEA, Berlin, 14-18 October 2002.10- INTERNATIONAL ATOMIC ENERGY AGENCY. Standard format and Content for SafetyRelated Decommissioning Documents. Vienna, 200511- INTERNATIONAL ATOMIC ENERGY AGENCY. Decommissioning of UndergroundStructures Systems and Components of Nuclear Installations. IAEA, Technical Reports Series nº439, Vienna 2006.12- MAGALHÃES, M.H;.XAVIER, A.M; GUERRERO, J.P.; MEZRAHI, A. Storage Facility inBrazil. Comissão Nacional De Energia Nuclear, CNEN, Rio de Janeiro ,Brazil.13- OECD NUCLEAR ENERGY AGENCY , The Safety of Decommissioning of Nuclear Facilities.OECD, NEA/RWM/WPDD, Paris 2004.14- Santos, I. Development of Methodology for Decommissioning and Dismantlement (D&D) of aUF 6 Pilot Plant. Sc.D Thesis, São Paulo University, São Paulo, 2007 (in Portuguese, underexecution).

DETERMINATION OF PM-147 IN SPENT FUEL SAMPLES INTHE FRAMEWORK OF THE MALIBU PROGRAML. ADRIAENSEN, M. GYSEMANS, C. HURTGEN,SCK-CENBoeretang 200, B-2400 Mol, BelgiumD. BOULANGERBelgonucléaireArianelaan 4, B-1200 Brussel, BelgiumIn this work it is shown that the concentration of Pm-147 in spent fuel samples can beaccurately determined by using liquid scintillation counting on a separated Pm fraction.Analyses were carried out on UO 2 and MOX samples. After dissolution of the fuel sample,Pm-147 has to be separated from the actinides, from the interfering fission products andfrom the other lanthanides. In this regard several seperation methods were evaluated. Thefinal separation was carried out by means of a column packed with Ln resin. To determinethe obtained separation yields, an aliquot of irradiated Nd-146 spike is added to the fuelsolution prior to separation. The progress of the separation can be monitored bymeasurement of gamma radiation of Pm-148m in several fractions. In this way separationyields of 90% could be obtained for all tested fuel samples.1. IntroductionThe objective of the MALIBU International Program was to provide improved knowledge of actinidesand fission products in fuels irradiated at high burnups. This type of information is a necessity in manynuclear areas, such as, fundamental research, safeguard issues, the calibration of source term codes andwaste management. This project proposed and managed by Belgonucleaire has been launched in June2003 with the participation of 10 organisations, representing more than 20 different companies. Aseries of UO 2 and MOX samples, originating from both BWR and PWR reactors, were chosen to beanalysed at SCK-CEN. Our group has already build up a lot of experience in the field of spent fuelanalysis and burnup determination due to participation in several spent fuel projects [1, 4, 5].After dissolution in the hotcell, the spent fuel samples were analysed by radiochemical and chemicalmethods covering an inventory of 51 different isotopes, among which Pm-147. A method wasdeveloped to determine Pm-147 in a separated Pm fraction using liquid scintillation counting. Theseparation of Pm-147 is necessary to eliminate all interferences of other β/γ emitters present in thesample during the LSC measurements.2. DiscussionThe spent fuel samples were dissolved in a hotcell following a two step procedure in which thesamples are boiled under reflux in, respectively, 8M HNO 3 and 10M HNO 3 + 0.1M HF. The obtainedsolution is diluted until its activity is low enough to be brought out of the hot cell for the subsequentseparations [2,3] and analyses.Before starting the separations, the necessary spikes and tracers were added to the fuel solution. In firstinstance U and Pu are separated from the other isotopes. For this Pu is converted to Pu(IV) byapplying a redox cycle. This solution is evaporated to dryness and redissolved in 9.0M HCl. The HCl9.0 M solution, containing the dissolved fuel, is fed onto a Dowex 1x4 column preconditioned withHCl 9.0M. The fission products, the rare earth elements, Am and Cm are eluted with 9.0M HCl andcollected for further treatment.

Consequently, the eluate fraction is evaporated to dryness and redissolved in 8.0M HNO 3 . Thissolution is fed onto a mixed PbO 2 - Dowex 1x4 column on which the tetravalent elements, e.g., Ce(IV)remain fixed. The trivalent rare earth elements and actinides, and most of the fission products areeluted with 8.0M HNO 3 . From this fraction a 0.1M HCl solution is made by multiple evaporation andredissolving. Note that in each evaporation step, some H 2 O 2 is added to help the breakdown of anyorganic material coming from the resin columns.The 0.1M HCl sample is fed onto a di(2-ethylhexyl)orthophosphoric acid-kieselguhr column (Lnresin) from which the remaining Am and Cm can be eluted with a freshly prepared mixture of 2Mlactic acid and 0.2M Na 5 DTPA (Di-ethylenetriaminepentaacetic acid).In a next step a method was developed to consecutively elute the lanthanides, and thus alsoprometium, from the Ln resin. Before applying this method to irradiated fuel samples, someexperiments were conducted on test samples whereby HNO 3 or HCl gradients were used to elute thelanthanides. Usually HNO 3 is applied in traditional spent fuel separations, but for Pm-147 it was notreally suitable. The eluate was collected in small aliquots of 2-5 ml that were analysed with ICP-MS.The Pm fractions were analysed with LSC. The results for the experiments in which a HCl gradient isused, are summarized in fig 1.40%0,18 N HCl 0,3 N HCl 0,4 N HCl 0,6 N HCl35%30%total %25%20%NdPmSmEuGdNd 146Sm 147Eu 153Gd 156Pm-14715%10%5%0%1 3 5 7 9 11 13 15 17 19 21 23 25 27aliquot numberFig 1: The elution curves of Nd-146, Sm-147, Eu-153, Gd-156 and Pm-147 using HClThese results clearly show that when HCl is used, almost 100% of the Pm-147 is eluted from thecolumn. If, on the other hand, HNO 3 is applied, the elution yield is below 50%. Moreover, it was notpossible to get a well shaped elution peak. Therefore, it was decided to use HCl for the Pm elutions ofspent fuel samples.To determine the obtained separation yields, an aliquot of irradiated Nd-146 spike is added to the fuelsolution prior to separation. This spike is irradiated in the BR2 reactor of SCK•CEN before use andcontains low concentrations of Pm-148m, which is used as a tracer to track Pm-147. The progress ofthe separation can be monitored by measurement of gamma radiation of Pm-148m in the obtainedfractions. In this way separation yields of about 90% could be obtained for all tested samples. Theresulting Pm-148m elution curves of three spent fuel samples whereby this method was used, arepresented in fig 2. The obtained separation yields do not seem to be influenced by the type of fuel(UO 2 or MOX) or the type of reactor (PWR or BWR). In all cases similar yields were produced.

40.00%35.00%30.00%Sample 1 (UO2-PWR)Sample 2 (MOX-PWR)Sample 3 (MOX-BWR)25.00%% Total20.00%15.00%10.00%5.00%0.00%5 6 7 8 9 10 11 12Aliquot numberFig 2: Elution curves for Pm-148m obtained on three different spent fuel samples measured withgamma spectrometryFor the actual analysis of Pm-147 with LSC, several fractions are selected. These are mixed and fromthis new solution a sample, suitable for LSC, is prepared. The LSC analysis is done with a PackardTri-Carb 1900CA detector which makes use of two photomultipliers to minimise the signal to noiseratio. For quantification a certified internal standard is applied. The measured concentrations are thenrecalculated to the analysis date and to the EOL (end-of-life) reference date, hereby taking intoaccount the dissolution factors and separation yields. When the concentration values at EOL arecompared, it becomes clear that for samples irradiated in the same type of reactor, the Pm-147concentration (mg/g fuel) increases with an increasing burnup.3. ConclusionTo measure Pm-147 in spent fuel samples, a suitable separation method was developed. By the use ofHCl it was possible to separate Pm from the other lanthanides with a separation yield of about 90%. Afew selected fractions were measured with LSC to obtain Pm-147 concentrations. It was found thatsamples irradiated in the same type of reactor, the Pm-147 concentration (mg/g fuel) increases with anincreasing burnup.4. References[1] Gysemans M., Dobney A., Adriaensen L., Sannen L. Proceedings Hotlab 2006, url:http//www.sckcen.be/HOTLAB/events/proceedings/2006/session2.pdf[2] P. De Regge, R. Boden, Determination of neodymium isotopes as burnup indicator of highlyirradiated (U, Pu)O 2 LMFBR fuel, J. Radioanal. Chem., 35, 1977, 173-184.[3] P. De Regge, D. Huys, R. Boden, Radiochemical analysis methods for burnup determination inirradiated fuel, presented at the Vortragstagung: Kern-, Radio- und Strahlenchemie, Grundlagen undAnwendungen, Jülich 22-26 september, 1980.[4] R. Boden, Methodology, calculation and interpretation in the destructive burnup determination ofnuclear fuel, NCS/72/D4301/RB/Ir/1398, December, 1992.[5] M. Gysemans, M. Van Bocxstoele, P. Van Bree, L. Vandevelde, E. Koonen, L. Sannen, B. Guigon.Destructive radiochemical analysis of uraniumsilicide fuel for burnup determination. Advances innuclear and radiochemistry, Jülich, 2004, p 187.

Perception of management options for contaminated milk andimplications for the decision-making processC.O. TURCANUService de Mathématiques de la Gestion, Université Libre de Bruxelles (ULB)Av. F. Roosevelt 50, C.P. 210-01, B-1050, Brussels-BelgiumB. CARLE, F. HARDEMANSociety and Policy Support, Belgian Nuclear Research Centre SCK•CENBoeretang 200, B-2400, Mol-BelgiumABSTRACTThis paper analyses public acceptance of management options for radioactivelycontaminated milk and associated consumers’ behaviour, focusing on the implications forthe decision-making process. The results derive from a public opinion survey carried out inBelgium in 2006. The instrument used was Computer Assisted Personal Interviewing,complemented with a simulated news bulletin for a fast and realistic briefing on thesituation investigated. Clean feeding of dairy cattle and disposal of contaminated milk arethe preferred options in case of contaminations above legal norms. For contaminationsbelow legal norms, normal consumption of milk seems better accepted than disposal.Nonetheless, the consumer’s behaviour reveals a precautionary tendency: the presence ofradioactivity at some step in the food chain could lead to avoiding purchasing productsfrom affected areas. Finally, public trust building appears as a key element.1. IntroductionThe management of a radiological food chain contamination requires a complex decision-makingprocess involving multiple criteria and stakeholders. The aftermath of the Chernobyl accident (IAEA,2006, pp. 83) proves that a robust and practicable restoration strategy should take into account notonly radiological and feasibility criteria, but also the “acceptability of the countermeasures, ethicaland environmental considerations, requirements for effective public communication, spatial variationand the contrasting needs of people in urban, rural and industrial environments”. The need toaccommodate social concerns in the decision-making process (Otway, 1987; Allen et al, 1996)implicitly calls for a broad stakeholder involvement including the general public.Depending on the characteristics of a radiological food chain contamination (geographical scale,type and quantities of radionuclides involved, etc), countermeasures (Nisbet et al, 2006) can be takento reduce the health risk for the population, to bring social reassurance and to facilitate the return tonormal life in the affected areas. However, little is known about the public acceptance of suchcountermeasures and one way to deal with the lack of information is by means of a survey. This paperreports on the results of a first survey on this topic. A couple of issues considered relevant for thedecision-making process are addressed: i) public acceptance of various countermeasures; ii)consumer’s behaviour and its relation with countermeasures’ acceptance. Milk was chosen as casestudy since its continuous production requires urgent decisions due to the limited storage facilities.Moreover, dairy products are an important element of the diet, especially for children, who constitutethe most radiosensitive population group. Finally, for certain radionuclides with high radiotoxicitysuch as radiocaesium and radioiodine, maximal levels of activity concentration in milk are reachedwithin few days after the ground deposition of radioactive material (Nisbet, 2002).In the next section we introduce the problem studied and its particularities. In section 3 themethodology used is described, while in section 4 some results are discussed (see Turcanu et al, 2007,for an extensive analysis). Conclusions are drawn in the last section.

2. A hypothetical radioactive contamination of the food chain: problem framingDuring recent years, the public confidence in food safety has declined due to multiple factors, e.g. thenumerous food crises or the shortcomings in communication about food safety which is too oftensolely based on scientific risk assessments (Jensen & Sandoe, 2002). Frewer et al (2004, pp.1183)observe that public risk perceptions have often been dismissed as ‘‘irrational’’, but these publicconcerns (and associated risk behaviours) “have direct consequences for human health, food safetyand security, economic expansion, and international regulation”.During a radiological emergency or in the restoration phase, authorities or food industries maydecide on countermeasures for limiting the consequences of the contamination and maintaining orrestoring public trust. Several food and agricultural countermeasures have been studied in theliterature (Howard et al, 2005). Some of these target specific radionuclides, such as administration offeed additives, for instance, which largely reduces the transfer of radiocaesium to milk and meat.Countermeasures such as feeding dairy cows with uncontaminated fodder are effective against allradionuclides. Since the presence of radioactivity in food products is not an issue of concern for theBelgian public, the question is how are these countermeasures perceived by the public and to whatdegree are they accepted? From a legal viewpoint, after the Chernobyl accident in 1986 the EuropeanUnion (CEC, 1989) laid down maximal radioactivity levels allowed for marketed food products incase of a future nuclear accident or any other radiological emergency. A radioactive contaminationinvolves however a risk that would be new, involuntary and unknown to those exposed, with delayedeffects and potentially affecting a large number of people. Due to a presumably high risk perception(cf. Slovic, 1987), the responsible authorities may decide to implement actions even forcontaminations below legal norms. There are also situations, e.g. a large scale contamination, whenconsumption of specific products with contamination levels above the current legal norms could beenvisaged. Paine (1992) reports a case related to the Chernobyl accident when the legal norms forradioactivity in reindeer meat were raised in Norway due to social and cultural reasons.Finally, if countermeasures’ acceptance is important for the political decision-maker, theknowledge about consumer’s behaviour is essential in reducing the economic loss. Therefore weexamine consumer’s behaviour for selected milk countermeasures with respect to a number ofpurchasing options, looking in particular if a high acceptance of a countermeasure is reflected by theintention of buying products originating from the area affected by contamination.3. Methodological co-ordinatesAcceptance of management options and consumer’s behaviour after an accidental radiologicalcontamination in the food chain were focal points of a public survey (Van Aeken et al, 2006) carriedout in March-April 2006 among 1063 Belgian adults. Primary data were gathered through ComputerAssisted Personal Interviewing, i.e. face-to-face interviews at the home of the respondent, the answersbeing directly stored on a portable computer's hard disk. The selection of the persons interviewed isrepresentative for Belgium for the following variables: province, region, level of urbanisation, gender,age (three categories) and professionally active status (retired or not). To better reproduce the real-lifecontext, a video clip was shown to the respondents; this simulated a news bulletin reporting on anaccidental release of radioactivity to the environment.Two hypothetical situations were analysed: i) one causing contaminations above legal normsand ii) a less severe one, with contamination remaining below legal norms. In the first situation, theauthorities are legally bound to implement countermeasures, since food products with contaminationabove legal norms may not be brought on the market. In the second situation, the main reason fortaking countermeasures is public acceptance.The milk countermeasures discussed were: keeping the dairy cows in stables and feeding themwith uncontaminated feedstuff; reducing the radioactivity concentration in milk by administration offeed additives; processing of milk to dairy products having a low radioactivity retention factor;diluting contaminated milk with clean milk; disposing of contaminated milk; and slaughtering dairycows in case of a long duration contamination. To these we added normal consumption of milk anddecreasing contamination as much as possible.

4. Results and discussion4.1 Acceptance of management optionsFor the situation when contamination of milk would exceed the legal norms (see Table 1), thecountermeasures with highest acceptance are clean feeding and disposal of contaminated milk.Slaughtering dairy cows, which is an option for minimising the volume of contaminated milkproduced in case of a long-lasting contamination, is also well accepted. These results confirm theincreased sensitivity and intolerance of Belgian consumers (Vandecasteele et al, 2005), possiblyleading to a strong pressure to dispose of any suspicious food product.Administration of feed additives and processing of contaminated milk to butter or cheese areless accepted. This could be due to a fear that these countermeasures are less controllable, or to thefact that they leave a residual activity in milk, although being effective in removing the largest part.Degree of acceptanceManagement optionstronglydisagreedisagree undecidedagree stronglyagreedon'tknowContaminations above legal normsClean Feed 7% 9% 13% 40% 31% 0%Disposal 9% 11% 11% 30% 38% 0%Slaughter 10% 11% 15% 35% 29% 1%Feed additives 15% 24% 16% 27% 17% 1%Processing to butter or cheese 19% 24% 15% 27% 15% 0%Dilution 47% 28% 9% 7% 8% 0%Normal consumption if experts ok 18% 32% 22% 18% 9% 1%Contaminations below legal normsDecrease contam. as much as possible 3% 6% 8% 38% 45% 0%Clean Feed 4% 9% 14% 38% 33% 0%Normal consumption 8% 17% 16% 38% 20% 0%Disposal 16% 29% 15% 23% 17% 0%Feed additives 11% 22% 19% 27% 20% 1%Processing to butter or cheese 16% 26% 20% 24% 13% 1%Dilution 37% 32% 12% 10% 9% 0%Tab 1: Public acceptance of milk management options (N=1063 respondents)That people prefer to take legal norms as a reference point is apparent from the low acceptanceof normal consumption of milk with contamination above norms, even if experts declared it safe forconsumption. Dilution is strongly opposed by an overwhelming majority of respondents, possibly dueto a feeling that dilution would increase the scale of the problem, rather than reduce it.For the situation when contamination of milk is expected to remain below the legal norms, cleanfeeding as a measure of preventing milk contamination enjoys again a high acceptance among therespondents. More than half or the respondents also agree with the idea that products complying withthe legal norms can be consumed as usually, but the vast majority is in favour of reducingcontamination as much as possible, even when already below the legal norms. Although one third ofthe respondents agree with disposal of all contaminated milk, the general acceptability of this radicalcountermeasure decreases compared to the previous situation: almost half of the respondents areactually against it. Processing of milk and administration of feed additives are again less accepted,while dilution is mostly considered unacceptable.For a practical approach to geographical zoning of the affected area in view of applying certaincountermeasures, it is useful to know if the countermeasures’ acceptance depends on whether theexpected contamination is above or below legal norms. Acceptance of normal consumption and ofdisposal differs significantly in the two hypothetical situations considered.This is however not the case for clean feeding and milk processing (see Table 1). For these twocountermeasures, acceptance seems more related to the countermeasure itself than to thecircumstances of its application. This conclusion is supported by a marginal homogeneity test whichindicates that there is no significant difference between the associated acceptance distributions(significance level 0.01). For administration of feed additives, the difference between thecorresponding values in the two situations considered is, for any agreement degree, less than 3%;therefore also for this countermeasure the acceptance is similar. In case of dilution, given the very

small agreement rate with this option (

From a public acceptance viewpoint, the preferred countermeasures are clean feeding anddisposal, in case of contaminations above legal norms. For contaminations below legal norms, cleanfeeding and normal consumption are both well accepted. However, the results show that if normalconsumption of milk is the chosen policy for contaminations below legal norms, a serious market losscould be expected. Public acceptance proved to be influenced by the initial contamination level mostlyfor the normal consumption and the disposal countermeasures. For clean feeding and milk processing,and to a lesser extent for feed additives and dilution, public acceptance seems to be an intrinsiccharacteristic. From a decision-making perspective this suggests that application of suchcountermeasures may be on larger areas than those where contamination of milk is expected to exceedlegal norms.Looking from the perspective of both public acceptance and consumer's behaviour, it appearsthat a majority of people favour a precautionary policy, aiming at preventing any contamination in thefood chain. This upholds options like clean feeding and disposal of contaminated foodstuff.Administration of feed additives is an alternative, but this would presumably lead to a significantdecline in the consumption of products from the affected areas. Milk processing would have to counton a limited market segment, whereas dilution of milk is strongly opposed in any circumstance.Despite the inherent uncertainties, we consider that our study is useful for emergency planningpurposes, especially for situations when there is a time constraint. It can also improve communicationon the countermeasures to be implemented and may contribute to an increased public acceptance andefficiency of the decision-making process.ReferencesAFSCA, 2004. La sécurité alimentaire: à quel prix? l’Agence fédérale pour la Sécurité de la Chaîne alimentairehttp://www.favv.be/home/pub/doc/autre/Info_FRB_Fr_Final.pdf.Allen, P., Archangelskaya, G., Belayev, S., Demin, V., Drotz-Sjöberg, B. et al, 1996. Optimisation of healthprotection of the public following a major nuclear accident: interaction between radiation protection andsocial and psychological factors. Health Phys. 71 (5): 763-765.CEC, 1989. Council Regulation (Euratom) No.2218/89 amending Regulation (Euratom) No.3954/87 layingdown maximum permitted levels of radioactive contamination of foodstuffs and feedingstuffs following anuclear accident or any other case of radiological emergency. Off. J. Eur. Comm. L-211 (1): 1-2.Frewer, L., Lassen, J., Kettlitz, B., Scholderer, J., Beekman, V., & Berdal, K.G., 2004. Societal aspects ofgenetically modified foods. Food Chem. Toxicol. 42: 1181–1193.Hanley, N., Grande, J., Alvarez-Farizo, B., Salt, C., & Wilson, M., 2001. Risk perceptions, risk-reducingbehaviour and willingness to pay: radioactive contamination in food following a nuclear accident.Working Paper. University of Glasgow, U.K., http://hdl.handle.net/1905/419.Henson, S., 1996. Consumer willingness to pay for reductions in the risk of food poisoning in the U.K.J. Agr. Econ. 47(3): 403-420.Howard, B.J., Beresford, N.A., Nisbet, A., Cox, G., Oughton, D.H. et al, 2005. The STRATEGY project:decision tools to aid sustainable restoration and long-term management of contaminated agriculturalecosystems. J. Environ. Radioact. 83: 275-295.IAEA, 2006. Environmental consequences of the Chernobyl accident and their remediation: twenty years ofexperience. Report of the Chernobyl Forum Expert Group ‘Environment’. STI/PUB/1239, InternationalAtomic Energy Agency: Vienna, Austria.Jensen, K.K., & Sandoe, P., 2002. Food safety and ethics. The interplay between science and values.J. Agr. Environ. Ethics 15: 245–253.Nisbet, A.F., 2002. A strategy for management of milk contaminated as a result of a nuclear accident. NRPB-W5. National Radiological Protection Board: Chilton, Didcot, U.K.Otway, H., 1987. Expert risk communication and democracy. Risk Analysis, 7: 125–129.Paine, R., 1992. ‘Chernobyl’ reaches Norway: the accident, science, and the threat to cultural knowledge. PublicUnderstand. Sci.1: 261-280.Slovic, P., 1987. Perception of Risk, Science 236: 280-285.Turcanu, C., Carlé, B., Hardeman, F., Bombaerts, G. & Van Aeken, K. (in press). Food safety and acceptance ofmanagement options after radiological contaminations of the food chain. Food Qual. Pref. (2007), doi:10.1016/j.foodqual.2007.05.2005.Van Aeken, K., Turcanu, C., Bombaerts, G., Carlé, B., & Hardeman, F., 2006. Risk perception of the Belgianpopulation. Results of the public opinion survey in 2006. Report SCK•CEN, Belgian Nuclear ResearchCentre, Mol: Belgium. ISBN 9789076971131Vandecasteele, C., Hardeman, F., Pauwels, O., Bernaerts, M., Carlé, B. & Sombré, L., 2005. Attitude of a groupof Belgian stakeholders towards proposed agricultural countermeasures after a radioactive contamination:synthesis of the discussions within the Belgian EC-FARMING group. J. Environ. Radioact. 83: 319-332.

ENSURING SECURITY OF SUPPLY THROUGHPARTNERSHIPJULIO S. GONZÁLEZ JIMENEZ,ENUSA Industrias Avanzadas, S.A.Santiago Rusiñol 12, 28040 Madrid, SpainMANUEL NOVO SANJURJOCC.NN. Almaraz-Trillo, AIEAvenida de Manoteras, 46 bisEdificio Delta Norte, 3 planta 5ª, 28050 – Madrid (Spain)SCOTT FERGUSONWolf Creek Nuclear Operating CorporationWolf Creek Generating StationPost Office Box 411, Burlington, KS 66839ROBERT JAMES FEAGINWestinghouse Electric Company LLCP.O. Drawer R, Columbia, South Carolina 29250ABSTRACTThe developed countries only live for years an unprecedented period of well-being. Thismeans easy access to energy sources to keep the standard of living. Nevertheless when wewatch in the news the consequences of some uncontrolable meteorological events orterrorist acts we realize that our society is not invulnerable. Additionally in the comingyears the facilities producing raw materials, components and final assemblies will likely belocated farther away one from each other and from the final customer, posing a majorinfluence in the complexity of the supply chains. All this is becoming an issue in the energymarket where the security of supply is of bigger concern especially for the countries withhigher energy demand. This background place a challenge in the national administrationsand utilities to secure energy and fuel supply from all different sources including nuclear.As part of the effort to secure nuclear fuel availability both utilities and nuclear fuelvendors are merging forces to build a sustainable supply chain that can provide a reliablenuclear fuel in case of a potential disruption.Wolf Creek Nuclear Operating Corporation (WCNOC) and Almaraz NPP (Spain) togetherwith Westinghouse and ENUSA Industrias Avanzadas (Spain) have entered into a securityof supply agreement aimed at demonstrating the feasibility of fuel supply to a US nuclearpower plant from the ENUSA Juzbado fuel factory as well as to a Spanish power plantfrom the Westinghouse Columbia fuel factory.This paper describes the project goals, challenges, planning and expected results. At theend a set of conclusions are presented despite the fact that the project is currently beingexecuted.

IntroductionEurope, North America and in general the developed countries only live for years an unprecedentedperiod of well-being. This means easy access to energy sources that allow to work to our appliancesand automobiles and to keep on moving our public transportations and industries. For the developedworld it seems inconceivable to reduce our standards of living and to give up something that we givefor fact.It is this belief the one that questions when we attend the news of some uncountrolable meteorologicalevents (hurricane Katrina) or we witness with incredulity to the collapse of the twin towers of NewYork for a terrorist act. No we are invulnerable neither to the terrorism nor the nature, neitheralthough to a lesser scale to the politics and the regulators..Additionally in the coming years the facilities producing raw materials, components and finalassemblies will likely be located farther away one from each other and from the final customer, posinga major influence in the complexity of the supply chains.Al these facts have stimulated the energy producers to revise their supply chains and to attemptprotecting the weaker links. One of the initiatives to get it it is to guarantee an alternative nuclear fuelsupply in case of a potential disruption.That is the exercise that Wolf Creek Nuclear Operating Corporation (WCNOC) and Almaraz NPP(Spain) together with Westinghouse and ENUSA Industrias Avanzadas (Spain) ENUSA, haveoutlined.These four parties have entered into a security of supply agreement aimed at demonstrating thefeasibility of fuel supply to a US nuclear power plant from the ENUSA Juzbado fuel factory as well asto a Spanish power plant from the Westinghouse Columbia fuel factory ( Fig.1)GAD rods with tubing from SANDVIKVÄSTERASWOLF CREEKW-COLUMBIAshipment MCCENUSAJUZBADOFig.1. The logistics of the program at a glanceCN ALMARAZFrom the formal point of view the project is splitted in two subprogramsOn the one hand, ENUSA will manufacture for Westinghouse four Fuel Assemblies (F/As) for one ofthe reloads that W has to deliver to WCNOC. This is known as American subprogram.

On the other hand, Westinghouse will manufacture for ENUSA four F/As for one of the reloads thatENUSA has to deliver to Almaraz NPP. This is known as Spanish subprogram.The primary goal of this project is to demonstrate the reliability of an overseas manufacturer to supplynuclear fuel for a PWR Nuclear Power Plant in the case of a potential supply disruption affecting theprimary fuel supplier. Such supply must meet not only technical but also licensing, logistics andtiming requirements which must therefore be considered in an integral manner in the project.As an aside objective, Westinghouse and ENUSA shall undertake a cross-qualification of theirrespective manufacturing facilities, by means of an auditing process, as part of which not only theproduct but also manufacturing process and inspection techniques shall be compared.Partnership is one of the key words in this project. The relationship of Westinghouse and ENUSA ofmore than three decades allows the parties to approach the exercise outlined above with trust and in anatmosphere of collaboration.Fuel DesignsIt is logical that the designs of the fuel assemblies to be manufactured for the utilities should beequivalent to the resident fuel. Also one of the assumptions in the project is for each manufacturer tominimize the number of modifications to their respective fuel product lines needed to allow insertionin the plants (Figs. 2 & 3).In the American subprogram the reference design that Wolf Creek NPP loads is 17 RFA (z+2)featuring ZIRLO TM tubing, Intermediate Flow Mixers (IFM), RFA-2 mid grids, protective grid andIntegrated Upper Nozzle (WIN). The assemblies that ENUSA will manufacture have the followingparticularities :- Bottom Nozzle fabricated in Spain by a domestic supplier (ENSA)- UO2 rods with neither annular pellets nor integrated poisons and with end plugs manufactured byENUSAIn the Spanish subprogram the reference fuel assembly design that Almaraz NPP loads it is verysimilar to the RFA design although with the specific designation of 17 MAEF. The assemblies thatWestinghouse will manufacture have the following particularities:- Westinghouse Integrated Top Nozzle (WIN)- Bottom Nozzle suplied by W (now it is manufactured by ENSA)- Gad rods manufactured in Västeras with Sandvik ZIRLO TM tubing and end plugs manufactured byWestinghouse.

MAIN DESIGN DIFFERENCESSPANISH SUBPROGRAMIntegrated Top Nozzle(WIN)End Plugs manufactured at COLAGadolinia rodsmanufactured at västeråsZIRLO tube from SANDVIK(onlyGad rods)DESIGN OF RECORDSPANISH DOMESTIC PWR PRODUCTBottom Nozzle supplied by WHFig.2. Main design differences vs. the reference design in the spanish domestic PWR productMAIN DESIGN DIFFERENCESAMERICAN SUBPROGRAMUO2 rodsWithout annular pelletsor integrated poisonEnd plugs manufacturedBy ENUSADESIGN OF RECORD 17RFA (Z+2)Bottom nozzle suppliedby ENSAFig.3. Main design differences vs. the reference design in the US domestic PWR product

The fuel designs from Westinghouse and ENUSA, which are to be supplied respectively to Almarazand WCNOC, have been analyzed by both fuel fabricators to assure compatibility with the residentfuel in the core. The analysis includes not only the core physics and mechanical design but also themanufacturing and quality processes used during fabrication of the fuel.As stated above, most of the components shall be fabricated by Westinghouse, the main componentsupplier to ENUSA. However, the Lower Nozzles for the Wolf Creek fuel assemblies shall bemanufactured by ENSA at its factory in the North of Spain. Pressure drop tests have been performedto demonstrate the feasibility of using this component at Wolf Creek.For the Almaraz fuel, a specific modification has been required by design to incorporate Gadoliniumas burnable absorber. The Gadolinium fuel rods shall be fabricated at the Westinghouse fuel facility atVästerås (Sweden) and delivered to the Columbia factory for the final fuel assembling. A specific fuelrod design study shall be carried out to allow the insertion of those Gadolinia rods.For the Wolf Creek fuel, the core was specifically designed to not require burnable absorbers in thefuel rods provided by ENUSA. Wolf Creek normally uses the Integral Fuel Burnable Absorber(IFBA), a boron coating on the fuel pellet, which is not available from ENUSA. This design changewas possible for the delivery of four fuel assemblies; however, a full reload would require the use ofgadolinia, or discrete burnable absorbers.LicensingDue to the similarity of both product lines a relatively smooth licensing effort is expected. Key pointson this process shall be the different origins and characteristics of the UO2 powder. ENUSA utilizesan IDR process with powder delivered from the Springfields facility (UK) while Westinghouse uses anADU route. Additionally, Västerås AUC process shall be used in the Gadolinia rods for the Almarazfuel.Features like the WIN nozzle are to be first delivered into the Almaraz plant.LogisticsThis Security of Supply project involves three nuclear fuel manufacturing facilities and two nuclearpower plants located at both sides of the Atlantic. Therefore, an extensive effort must be put in placeto guarantee the smooth delivery of fuel to the different facilities. The transatlantic transportation willbe carried out through shipping lines with previous experience in transport of nuclear materialsFor both deliveries the new Westinghouse Traveller container shall be used, although for the WolfCreek assemblies an additional operation at the Columbia factory must be performed to load the fuelinto MCC containers for later delivery to Wolf Creek.The Gadolinia rods for the Almaraz fuel shall be shipped following an existing route and procedure atWestinghouse for shipment of Gadolinia rods for the US market. The ZIRLO TM tubes shall bemanufactured by Sandvik Europe.The fissile materials will be accounted for as a balance of the Uranium through book transfer amongthe accounts that as much ENUSA as Westinghouse have in SFL (UK).

ConclusionsAlthough the project is currently on going some of the expected results are anticipated:- The project will show to the clients of ENUSA and Westinghouse that in the event of adisruption the alternate fuel supply from foreign manufacturing facilities is a feasible back upfrom the logistics, design and licensing standpoints.- The exercise will demonstrate that the the designs that ENUSA and Westinghouse are able tomanufacture and deliver with their current supply chains are equivalent and do not need eitherfurther qualification or additional licensing.- The program will allow exploring new routes for supply of UO2 powder and components toeventually improve the current ones and to open alternatives before the foregone increase ofthe demand in coming years.- The security of supply program intends to get the above-mentioned goals with the minimumimpact in the licensing process before the regulatory authorities. The objective is to get theauthorization of fuel loading without necessity of revising the licensed safety analyses. This isa major advantage versus qualification of an alternate vendor where new safety analyses arenormally required, nearly prohibiting any benefit in an emergency situation.Both Westinghouse and ENUSA consider that this project is a key step in their long history of mutualpartnership which shall tighten even more their historical links to better serve their customers inEurope and the US.

TECK-IN :Skills Center for Containment Techniques and Working MethodsFull Paper by Jérôme BlancherAREVA Recycling Business Unit,MELOX Research and Development DepartmentBP 93124 - 30203 Bagnols-sur-Cèze Cedex France1. Introduction: competitiveness clustersIn 2005, France launched a new industrial policy mobilizing the key factors forcompetitiveness, one of the most important of which is the capacity for innovation.A competitiveness cluster is a group of companies, research centers and training organizationsin a given area, all working together as partners and pooling their actions under a shareddevelopment strategy.The cluster is designed to harness synergies and encourage close cooperation on innovativeprojects.The French government has decided to officially approve and support 67 competitivenessclusters.The TRIMATEC competitiveness cluster (Tricastin Marcoule Technologies) covers six R&Dzones and includes a central zone covering Tricastin and the east of the Gard department.These zones bring together all the major research and development organizations.TRIMATEC received the government’s seal of approval in 2005 and aims to produce projectsthat are essentially based on technologies developed by the nuclear industry but which cannow be applied to other sectors.The cluster is located over five departments (Ardèche, Drôme, Gard, Hérault and Vaucluse).2. The project initiated by MELOX and its partnersNumerous industrial processes require the use of containments in fields as varied as thepharmaceutical industry, medicine, microelectronics or the nuclear industry.The AREVA group, world leader in the nuclear industry, has over 30 years’ experience in thisfield. The MELOX plant is at the forefront since it is here that plutonium is used to fabricatefuel assemblies.When the plant was designed, and indeed ever since it has been in operation, there have beennumerous initiatives to improve the tools, methods and techniques used when working inglove boxes. Likewise, a variety of training sessions are held regularly to maintain a highlevel of expertise in this field.It was therefore decided to set up a Skills Center for Containment Techniques and WorkingMethods, with the participation and involvement of several partner companies whoseactivities necessitate this kind of work.The project was approved by the TRIMATEC competitiveness cluster on November 23, 2006and officially launched at a kick-off meeting held at MELOX with the various partners onJanuary 19, 2007.Current partnersCompanies: MAPA Advantech, Piercan, La Calhene, Sovis (St Gobain), Plastunion, STMI,Robotic Concept, Axilya, Trihom, Sinaptec, and Novintec.Training, teaching and research organizations: École des Mines d’Alès, INSTN Valrhô, andCEA Valrhô.AREVA Recycling BU / Full paper TECK-IN ENC 07Contact MELOX COM C. Depeyre 04.66.90.64.021

3. Creation of TEKH-IN and objectives of the association• StructureThe Skills Center for Containment Techniques and Working Methods was formed as a nonprofitorganization on June 15, 2007. Its aim is to pool, develop, improve and spread thecompetencies, techniques, methods and tools used when working in containments.The project aims to create a consistent set of academic and industrial competencies to providea center specializing in containment work and focusing on three main areas:- research and development,- professional training,- consulting and expertise.Pooling skills makes it possible to launch, finance and steer actions with the joint aim ofperpetuating a high level of technicity in the target field. This is an innovative project in that itbrings into contact organizations that would normally have very few dealings with each other,such as the nuclear industry, the pharmaceutical industry, the food processing industry,depollution, asbestos removal and other activities confronted with the problems associatedwith work in glove boxes. It will encourage performance improvement as regards workertraining and system operation in containments, particularly concerning nuclear andoccupational safety and the environment in an industrial setting.• ObjectiveMore precisely, the aim is to:- develop special tools, in situ work processes and technical materials to offerworkers improved ergonomics, and greater occupational safety and precision,- devise and dispense training for those working in an industrial environment orenrolled in a specialized university course,- provide consulting services and expertise to support industrialists.A market opportunity study is currently being launched. It will provide information on themajor containment work improvement trends, and the projected requirements of companies ina number of sectors.4. Resources and partnershipsThe center will have its own resources to:- carry out the tests required before any new techniques and tools are implementedon an industrial scale,- provide training in containment work, notably the use of special tools andtechniques.And resources provided by each partner to:- design, study and produce new techniques and tools,- carry out standardized tests prior to qualification,- design appropriate training courses,AREVA Recycling BU / Full paper TECK-IN ENC 07Contact MELOX COM C. Depeyre 04.66.90.64.022

- carry out research work to acquire the basic knowledge that supports applieddevelopments.5. On-going actions• Academic trainingThe aim is to create "relevant" partnerships with schools (technical schools, engineeringschools, vocational schools, etc.) with a view to transferring knowledge and competencieswithin the framework of a diploma course. The center’s contribution should be incorporatedunder certain conditions:- participation in diplomas leading to true job opportunities.- consistency between our areas of competence and the general content of the course work.This initiative has many benefits. It will lead to Department of Education approval and thehigh-quality contribution of the center will be acknowledged.• Research and DevelopmentThe following topics have been identified so far:- introducing equipment into containments- removing equipment from containments- very long-lived HEPA filter- self-standing micro vacuums- dispensing with the need for glovesThe above R&D projects will be the subject of special agreements binding the TEKH-INmembers interested in developing the subject. New partners are currently being sought forthese topics.• Competency mappingThe current and potential know-how of the project partners will be listed to produce a map ofthe competencies required to improve containment work.The following macro-competencies have been identified to date:- Design/engineering: integration of the operating and maintenance constraints specific tocontainments- Ventilation, filtering and purification systems- Transparent materials: providing visibility and an integrated barrier- Special gloves: isolating the operator from the product without hindering the manualdexterity required for operations- Lighting: external facility generating sufficient intensity without reflecting or dazzlingAREVA Recycling BU / Full paper TECK-IN ENC 07Contact MELOX COM C. Depeyre 04.66.90.64.023

- Monitoring systems measuring the quality of the environment (special sensors to detectdust, radioactivity, etc.)- Leak detection methods- Operating and maintenance know-how: focusing on preserving containment integrity andcreating the interface between the operator and the hostile environment- workstation organization and ergonomics- operating technique with appropriate gestures- risk awareness- self-control- questioning attitude- In situ waste management: special tools and techniques for waste removal and packaging- Dismantling operations in hostile environments: coordinating the various trades andmaking allowance for the constraints associated with containments- Operating, maintenance and investigation techniques in containments:endoscopes, vacuums, airlocks, personal and collective protection devices- Specially-adapted tools: conventional mechanical tools subjected to the constraints ofcontainments and modified accordingly- Design and production of protective clothing for work in hostile environmentsAREVA Recycling BU / Full paper TECK-IN ENC 07Contact MELOX COM C. Depeyre 04.66.90.64.024

ENC 2007 – Brussels, Belgium / 16-19 September 2007.ADVANCED SOLUTIONS FOR MITIGATING HYDROGEN RISKDURING TRANSPORTATION OF RADIOACTIVE MATERIALSV. Rohr, P. Abadie, H. IssardAREVA-TN International, St Quentin en Yvelines, FranceCorresponding author: valentin.rohr@areva.comABSTRACTDuring transportation of radioactive materials, radiolysis or thermal degradation of thetransported product can lead to the generation of hydrogen, which continuously enrichesthe gaseous mixture. Among the functions to be satisfied, transportation systems shallthus allow the control of the hydrogen content below its flammability limit. This can beachieved by limiting the transportation duration so as to reopen the cask before the criticalhydrogen concentration is reached. Development of new technologies that would mitigatethe hydrogen risk is all the more motivated because it would allow an extension of thetransportation duration.AREVA-TN International has developed catalytic systems which aim at buffering thehydrogen concentration far below the flammability limit. The principle of these catalysts isto recombine the hydrogen with the oxygen contained in the gaseous mixture.Hydrogen generation is either due to radiolysis of water or organic compounds. The latterare usually transported in the form of solid waste, whereas water is either in the form ofresidual water is the case of a dry transportation or of free water in the case of a wettransportation. AREVA-TN International has thus developed two types of catalyticsystems.The first is dedicated to the recombination of hydrogen coming from radiolysis of organiccompounds and residual water. It allows a significant recombination even in the presenceof other gaseous species generated by radiolysis of organic compounds. This catalyticsystem also absorbs the water generated by the recombination reaction, preventing fromsubsequent radiolysis of this water.The second catalytic system is specifically dedicated to wet conditions of transport and isparticularly adapted to these conditions including successive immersions into boratedwater.The present paper gives an overview of both catalytic recombining systems. It describesthe laboratory tests undertaken for the evaluation of the recombining efficiencies.The catalyst dedicated to dry transportation systems was especially studied in thepresence of several gases that may be generated by radiolysis of organic compounds, i.e.CO, CO 2 , CH 4 , C 2 H 6 , C 2 H 4 , HCl, I 2 . The water absorption capacity of the catalyst has alsobeen determined.In case of wet transportation, particular attention is placed on the recombining efficiencyafter immersion of the catalysts in borated water, which would occur in a nuclear reactorpool during loading of used fuel. Laboratory investigations, carried out in an autoclavesimulating a wet transportation cask, showed that, after immersion in borated water, thecatalytic system allows the recombination of 3% hydrogen in less than 24 hours attemperatures as low as 35°C.INTRODUCTIONCatalytic systems have already been developed for mitigating hydrogen risk in case of asevere accident in a nuclear power plant . These catalysts were based on the use of amixture of platinum and palladium deposited on a support (for example alumina). Thiscatalytic system enhances the oxidation of hydrogen with respect to the following reaction:© Copyright TN International 2007 Page 1 on 13

ENC 2007 – Brussels, Belgium / 16-19 September 2007.1H2+ O2→ H2O2It also allows the recombination of about 1 µmol of hydrogen per second and per gram ofcatalyst at 90°C.As a matter of fact, the applicability of this technology to transportation of radioactivematerials requires the system to be acceptable at lower temperatures, down to about 50°C. In addition, the catalytic support for the platinum and palladium elements shall in somecases withstand the humid environment of wet transportation. In case of drytransportation, alumina support becomes an advantage since it allows the water producedby the recombination reaction to be absorbed. The latter prevents from further radiolysisand hydrogen generation.Consequently, AREVA-TN International developed two types of catalytic systems basedon the deposition of palladium and platinum on a stainless steel grid or on aluminium. Atfirst, the present paper shows the qualification results with respect to laboratory testscarried out with the stainless steel recombiner, which we dedicate to wet transportation.Secondly, details are given on the use of an alumina recombiner for dry transportation.1. QUALIFICATION OF A HYDROGEN RECOMBINER FOR WET TRANSPORTATION1.1 Experimental protocol of the qualification campaignThe qualification protocol for the catalytic grid consisted in 15 immersion/recombinationcycles simulating 15 loading/transport cycles. The experimental set-up consisted in areactor combined with gas injections and a sampling device dedicated to the chemicalanalysis of the gas . For laboratory tests, a cycle corresponds to one week of testingdefined as follows:• 2 days immersion of the catalytic stainless steel grid into borated water containing2662 mg/l of boron.• setting up a gaseous mixture of N 2 -O 2 containing 3 to 7 vol% oxygen.• homogenization of the reactor temperature at 40 °C.• injection of 3 vol % hydrogen.• measurement of the evolution of the hydrogen content as a function of time.• as soon as the hydrogen content falls below 1%, an additional injection is performedin order to readjust the hydrogen concentration to 3%.• the cycle is stopped after 5 days (1 week). The catalyst is then immersed intoborated water for 2 days (week-end).1.2 Qualification resultsAppendix 1 shows the evolution of the hydrogen and oxygen concentrations during the 15immersion/recombination cycles.The first 4 cycles were carried out with initial oxygen content of 6 to 7 vol% whereas forthe last 11 cycles, this concentration was comprised between 3 and 4 vol%. It appearsthat this initial oxygen content has a direct influence on the amount of hydrogen injectionduring the corresponding cycle. With 7 vol% O 2 initially, 4 hydrogen injections areperformed on a weekly basis. With 3 or 4 % O 2 , only 3 injections are possible per week.Indeed, with an initial concentration of 3 to 4 % O 2 , the catalytic recombination leads to adrop of the oxygen content to concentrations below 1% after about 2 days. The oxygensupply is then too small to allow the catalytic recombination with respect to the chemicalreaction presented in paragraph 3.1. Consequently, for the last 11 cycles, the hydrogencontent remains stable during the last few hours that follow the last hydrogen injection.Nevertheless, appendix 1 shows that the recombination capacity remains similar for thewhole 15 cycles.© Copyright TN International 2007 Page 2 on 13

ENC 2007 – Brussels, Belgium / 16-19 September 2007.1.3 Kinetics of the recombination reactionIt can be assumed that the H 2 recombination rate in cm 3 /h/cm 2 of catalytic grid can bewritten as follows:αV recomb= A 0[ H2]Eq. 1where:• V recomb designates the H 2 recombination rate in cm 3 /h/cm 2 of catalytic grid,• A 0 and α (the order of the oxidation reaction with respect to the hydrogen content) are thecouple of kinetic constants for each test and•[H 2 ] is the hydrogen concentration in volume %.Based on the measured evolution of hydrogen concentration in the course of time(Appendix 1), the recombination rate was calculated for each of the 15 qualification testsperformed in N 2 -4%O 2 at 40°C. Figure 1 illustrates these recombination rates V recomb forthe two tests , which show the slowest recombination reaction.With respect to Figure 1, a conservative kinetic law for hydrogen recombination can bewritten as follows:1,5V recomb= 0.014[H 2]Eq. 2Recombination rate (cm 3 H2/h/cm 2 of catalytic grid)0,160,140,120,10,080,060,040,0240°C in N2-4%O2 (1)40°C in N2-4%O2 (2)Assumed recombination rate with restect of Eq. 200 0,5 1 1,5 2 2,5 3 3,5H 2 (%vol. / incondensable gases)Figure 1 : Evolution of the H 2 recombination rate per unit surface of thecatalytic grid (in cm 3 /h/cm 2 ) as a function of the H 2 concentration in thereactor at 40 °C.For design calculations the kinetic law Eq. 2 of an order of 1.5 with respect to thehydrogen content will be considered for the oxidation of hydrogen.1.4 Applicability to the transportation of used fuelHydrogen generation due to radiolysis is usually given by a linear expression of thevolume V prod of hydrogen produced by unit time . V prod is thus given in cm 3 of H 2 /h.© Copyright TN International 2007 Page 3 on 13

ENC 2007 – Brussels, Belgium / 16-19 September 2007.Consequently, in order to recombine the total quantity of hydrogen generated byradiolysis, the minimum surface S mini of catalyst required is given by (in cm 2 ):VprodVprodSmini= =1.5V 0.014[H ]recombEq. 3As for equation Eq. 2, the hydrogen concentration in Eq. 3 is expressed in volumepercentage.An hydrogen concentration of 1% is acceptable with respect to safety since it is far belowthe flammability limit of 4% hydrogen in air. Therefore, it is conservative to consider thatthe minimum surface of catalytic grid to be put in place is given by Eq. 3 with 1 vol. %hydrogen.As an example, R62 is a wet transportation package for used fuel presented at Patram1986 . The results of measurement of hydrogen concentrations within the packageshowed a hydrogen generation rate below V prod =3.5 cm 3 /h. This means that for theparticular design of R62, a total surface of 250 cm 2 of the catalytic grid would be enoughfor maintaining the hydrogen concentration below 1%. It appears that such an amount ofrecombiner is acceptable for an application in usual transportation casks, as it should beeasy to insert this low amount of catalyst within the cavity. Even higher amounts ofrecombiner should be easy to put in place in order to allow a safety margin with regards torecombination capacity.Furthermore, with a higher amount of catalytic recombiner within the transportation cask,the hydrogen risk can be mitigated for a higher level of water radiolysis, i.e. for largercasks or for used fuels with higher radiation energies.22. QUALIFICATION OF A HYDROGEN RECOMBINER FOR DRY TRANSPORTATIONThe recombiner dedicated to dry transportation is made of alumina beads impregnatedwith palladium. Some characteristics are given in the following table:Diameterof thebeadsSpecificsurfaceDensity3 mm 300 m 2 /gr 800kg/m 3The aim of the qualification tests was to :a) Assess the hydrogen recombination capacity as well as the water absorption capacity ofthe hydrogen recombiner at different temperatures;b) Determine the recombination kinetics at different temperatures;c) Assess the influence of other radiolysis gases on the H 2 recombination capacity.2.1 Hydrogen recombination capacityThe hydrogen recombination capacity has been determined by laboratory tests at 25°C,45°C and 65°C.The catalytic beads were previously dried at 80 °C within the reactor. Tests were carriedout in a 5 L reactor, which was linked to an argon-hydrogen supply. The latter allows thehydrogen content to be maintained at about 2%. Hygrometry is measured all along thetesting duration. The composition of the gas is periodically analyzed using gaschromatography.The recombination capacity (given in moles of H 2 /kg of catalyst) is measured from thetotal quantity of oxygen consumed (2 moles of H 2 for 1 mol of O 2 ).The results are shown in the following table:25°C 45°C 65°CRecombination 18,5 23 > 40© Copyright TN International 2007 Page 4 on 13

ENC 2007 – Brussels, Belgium / 16-19 September 2007.capacity(moles H 2 /kg ofcatalyst)2.2 Water absorption capacity of the alumina beadsFor the three tests mentioned previously, the hygrometry was measured above 90%. Thewater formed by the recombination reaction is thus either absorbed by the alumina fromthe recombiner or contained in the gas phase. The ratio between the water content in thealumina and in the gas phase depends on the temperature, i.e. the higher thetemperature, the lower the amount of water absorbed by the catalytic support, thus thehigher the water content in the gas phase.The absorption capacities are given below.TestingtemperatureRatio of themass ofwateraborbed bythe catalystwith respectto the massof catalystAmount ofwaterproducedby therecombination andtrapped inthe catalystAmount ofwaterproducedby therecombination andreleased bythe gasphase25°C 32% 90% 10%45°C 31% 80% 20%65°C 17% 45% 55%2.3 Hydrogen recombination kinetics of the recombinerThe recombination kinetics has been studied at different temperatures, i.e. 28°C, 45°C,55°C et 65°C. For each test, a mass of 6,25 g of catalyst was previsouly dried at 80°C.The testing reactor had a capacity of 5 L and the starting hydrogen content was about 4 %hydrogen. The evolution of the hydrogen concentration was measured in the course oftime using gas chromatography. Experimental results are given in Figure 2.© Copyright TN International 2007 Page 5 on 13

ENC 2007 – Brussels, Belgium / 16-19 September 2007.5,004,50T = 28°C4,00T = 45°CHydrogen concentration (%)3,503,002,502,001,501,00y = 4,0988e -7,2075xR 2 = 0,9896T = 55°CT = 65°Cy = 3,508e -2,0584xR 2 = 0,99540,500,00y = 3,6806e -13,591xR 2 = 0,9958y = 4,4356e -11,178xR 2 = 0,99680,00 0,10 0,20 0,30 0,40 0,50 0,60 0,70 0,80 0,90Time (hours)Figure 2: Evolution of the hydrogen content in the reactor at differenttemperatures(caracterisation of the H 2 recombination kinetics)With respect to figure 2:a) the evolution of the hydrogen content follows a decreasing exponential law−Bt% H = Aewhere A and B are constant values (see figure 2).b) the H 2 recombination rate increases with temperature; compared to 25°C, the H 2recombination rate is multiplied by 3 at 45°C, by 4 at 55°C and by 6 at 65°C.When the temperature increases in a transportation/storage cask, gas generation from thethe radiolysis is generally enhanced. The present results thus shows that the recombinerfully answers this requirement since its recombination rate increases by increasingtemperature.2.4 Efficiency of the catalytic recombiners in presence of flammable andpotentially poisoning gasesThe efficiency of the recombiner was tested in presence of 5 other gases: CH 4 ; C 2 H 6 ;C 2 H 4 ; CO 2 ; HCl; I 2 . The purpose was to ensure that these gases have no poisoning effecton the efficiency of the catalyst. Indeed, these species were identified by AREVA as beingpotentially generated by radiolysis of organic compounds. The latter are contained incanisters dedicated to residues from the reprocessing of used fuel. Each type of gas wastested separately in combination with hydrogen. The proportions were the following:Ratio ofgases withrespect toH 2H 2 /CH 4 H 2 /C 2 H 6 H 2 /C 2 H 4 H 2 /I 2 H 2 /CO 22/1 40/1 40/1 1200/1 2/1© Copyright TN International 2007 Page 6 on 13

ENC 2007 – Brussels, Belgium / 16-19 September 2007.Concerning HCl, a capsule containing a concentrated HCl (36 %) solution was placedinside the testing reactor just above the catalysts.Figure 3 shows the evolution of the hydrogen concentration for each of the six tests.These curves can be compared directly since they refer to the same quantity ofrecombiner.It appears that the catalyst shows a significant recombination capacity since it allows thehydrogen content to fall below 1% in less than 24 hours. In addition there is no poisoningeffect for the gases CH 4 ; C 2 H 6 ; C 2 H 4 ; CO 2 ; HCl; I 2 .4,50H24,00H2 + CH4 (H2/CH4 = 2/1)H2 + C2H6 (H2/C2H6 = 40/1)Hydrogen concentration (%)3,503,002,502,001,50H2 + C2H4 (H2/C2H4 = 40/1)H2 + CO2 (H2/CO2 = 2/1)H2 + I2 (H2/I2 = 1200)H2 + HCl1,000,500,000,00 2,50 5,00 7,50 10,00 12,50 15,00 17,50 20,00 22,50time (hours)Figure 3: Evolution of the hydrogen content in presence of potential poisoninggases.Effect of CO:Moreover, the effect of carbon monoxide on the recombining efficiency of the catalyst wasalso tested.The first tests were carried out in the same experimental conditions as previously, with aratio of H 2 /CO of 2/1. The evolution of gaseous concentrations showed that on thecontrary to other gases, the carbon monoxide is consumed during the test. CO is oxidizedinto CO 2. This reaction is also catalysed by the recombiner and thus comes in competitionwith the H 2 recombining reaction. Consequently, the H 2 recombination becomes muchslower, i.e. the hydrogen oxidation into H 2 O actually starts when the quasi-total quantity ofCO is removed. (Figure 4)© Copyright TN International 2007 Page 7 on 13

ENC 2007 – Brussels, Belgium / 16-19 September 2007.3,50Evolution of the hydrogen concentration (in %) in presence of CO3,00H2CO2,50% H2 and CO2,001,501,000,500,000,00 5,00 10,00 15,00 20,00 25,00 30,00 35,00 40,00 45,00Time (hours)Figure 4: Evolution of the hydrogen and CO contents in presence of catalyticrecombiner, without any further addition of gases during the test.An additional test was carried out using the catalytic recombiner associated with carulite,which is a mixture of manganese and cupper oxides. Contrary to the above mentionedtests, this experiment was carried out in dynamic conditions. Therefore, H 2 and CO werecontinuously injected into the reactor with a ratio flow H 2 /flow CO equal to 11,5. Thedynamic conditions are thus closer to the real conditions of a transport/storage cask, werethe hydrogen and carbon monoxide are continuously generated by the radiolysis of thetransported content.Figure 5 shows the evolution of the of the hydrogen content as a function of time. It has tobe pointed out that the hydrogen concentration is rapidly stabilized at about 0.5 %. Thismeans that the hydrogen recombination takes place since the very beginning of H 2 andCO injections. Consequently, the presence of carulite allows a rapid stabilization of thehydrogen concentration far below the flammability limit of hydrogen in air.© Copyright TN International 2007 Page 8 on 13

ENC 2007 – Brussels, Belgium / 16-19 September 2007.5,004,50Evolution of H2 concentration in presence of COHydrogen concentration (%)4,003,503,002,502,001,50Hydrogen flammability limit in air1,000,500,000,00 5,00 10,00 15,00 20,00 25,00 30,00 35,00 40,00 45,00 50,00Time (hours)Figure 5: Evolution of the hydrogen concentration in presence of CO (ratioflow H 2 /flow CO = 11,5)CONCLUSIONTwo catalytic systems based on the deposition of palladium and/or platinum on either astainless steel grid or alumina beads were developed by AREVA TN International.On the one hand, the first system has shown to catalyse the oxidation of hydrogen attemperatures as low as 40 °C, even after immersion into a boric acid water solution. Thecatalysing capacity has also shown to remain stable after 1400 hours of testing.Furthermore, the recombination capacity is sufficient for stabilising the hydrogenconcentration below its flammability limit, even with a low amount of recombiner. Therequired amount of recombiner can thus easily be introduced in a free space of the caskscavity.On the other hand, concerning the palladium deposited on alumina, the recombination isensured by the noble metal whereas the alumina allows the water produced by therecombination to be absorbed and avoid further radiolysis. The efficiency of this catalyst inpresence of other radiolysis gases, i.e. CH 4 ; C 2 H 6 ; C 2 H 4 ; CO 2 ; HCl; I 2 has beendemonstrated, whereas the presence of CO requires the use of carulite in combinationwith the catalytic recombiner. This second system has already been presented in TNInternational’s patent .Both systems developed by AREVA-TN International answer the whole range ofrequirements for recombining systems either for wet transportation of used fuel or drytransportation casks such as transportation of vitrified residues.REFERENCES: Morfin F, Sabroux J-C, Renouprez A, Catalytic combustion of hydrogen formitigating hydrogen risk in case of a severe accident in a nuclear power plant: study ofcatalysts poisoning in a representative atmosphere, Applied Catalysis B: Environmental47 (2004) 47-58.: Vinson D. W., Deible R.W., Sindelar R.L., Evaluation of hydrogen generation fromradiolysis from breached used fuel, DOE document WSRC-MS-2002-000728.© Copyright TN International 2007 Page 9 on 13

ENC 2007 – Brussels, Belgium / 16-19 September 2007.: Abassin J.J., Blanchard R.J., Bochard C., Campani M., Experiences paramétriquessur la radiolyse de l’eau et application à l’emballage R62 pour le transport d’assemblagescombustibles défectueux, Proceedings of Patram 86 at Davos, Switzerland.: Patent # FR2005/050647: Procédé et dispositif d’élimination des gaz inflammablesdans une enceinte équipée d’un tel dispositif. P. Abadie, H. Issard.© Copyright TN International 2007 Page 10 on 13

ENC 2007 – Brussels, Belgium / 16-19 September 2007.APPENDIX 1 : Qualification results.© Copyright TN International 2007 Page 11 on 13

ENC 2007 – Brussels, Belgium / 16-19 September 2007.876volume%in N2Evolution of the oxygen and hydrogen contents during 15 immersion/recombinationcycles at the laboratoryO 2 contentEvolution of H 2 contentweek 1 week 2 week 3 week 4week 5 week 6 week 7 week 8week 9 week 10 week 11 week 12week 13 week 14 week 155432100 200 400 600 800 1000 1200 1400Duration (hours)© Copyright TN International 2007 Page 12 on 13

METHODS FOR FABRICATING GAMMA – URANIUM –MOLYBDENUM (γ-UMo) ALLOYS AND THEIR INFLUENCEON POWDER OBTENTION BY THE HDH TECHNIQUEF.B.V. OLIVEIRA, M. DURAZZONuclear Fuel Center, Nuclear and Energy Research Institute – IPENP.O.Box 11049, Pinheiros 05499, São Paulo, BrazilH. G. RIELLAChemical Engineering Department, Santa Catarina Federal UniversityFlorianópolis, BrazilABSTRACTGamma uranium-molybdenum (γ-UMo) alloys has been widely considered as the best lowenriched – high density fuels candidates for the substitution of the previously utilized highenriched ones, according to the RERTR requirements. For its usage as dispersions in platetype research reactor fuels, some of the techniques to transform the ingots into powder arehighly influenced by the alloys´ properties achieved in the previous steps of melting andsolidification. In this work we will briefly introduce the study of two of the maintechniques to melt (γ-UMo) alloys, the induction and arc melting, and show some of thedifferences in properties presented by the casts and its powders, obtained by the techniqueof hydration-dehydration (HDH) thermal treatments. Samples of the ingots and powdersprepared in the range of compositions of 5 to 10 weight % Mo, were characterized bymeans of scanning electron microscopy, hydrostatic density, and X-ray diffraction. It wasverified that highly homogenous alloys can be obtained by the induction technique in onlyone step, and for those produced by arc, even with smaller loadings, several microstructuralproblems arises, leading perhaps to its invalidation as a technique for the fabrication ofnuclear (γ-UMo) alloys and powders.1. IntroductionIn the nuclear area, the main techniques for fabricating the powders of γ-UMo alloys for its usage asdispersion fuels in nuclear research reactors are cryogenic milling, machining, atomization andhydration-dehydration (HDH). Atomization is the commercially most accepted, but there are worksindicating that hydration-dehydration (HDH), as studied by BALLART et al. [1], SOLONIN et al. [2],or its variation, the hydration-milling-dehydration (HMDH) technique, as studied by PASQUALINI etal [3] and PASQUALINI [4], are also suitable to produce powders which obeys more closely thespecifications requirements in terms of dimensions and granulometric distributions. This is the mainmotivation for the adoption of the HDH technique at IPEN / CNEN – Brazil, to the production of γ-UMo powders.The responsible for the success of the HDH techniques is the high hydrogen affinity presented by thealpha uranium phase. In the ranges of compositions here studied, an alloy that presents some alpha asintergranular precipitates reacts readily with hydrogen, leading to high yieldings in terms of powderproduction. The reaction:2 αU (s) + 3 H 2 (g) → 2 (αU)H 3 (s) (01)are related to the hydride formation, mainly in the grain boundaries. The hydride phase has a volumebigger than that of the alpha phase, and this volumetric difference generate tensions in the alloy thatleads to its fragmentation, even without the step of dehydration. Thus, the ease of fragmentation andsubsequent powder obtention implies necessarily in the gamma decomposition during the HDHthermal treatments. The reaction of the decomposition is:

γ matrix → α + γ poor (02)which produces precipitates with properties that are function of the HDH temperatures and, also, of themethods of the alloys preparation. The above equilibrium was extensively studied by REPAS et al. [5],VAN THYNE & McPHERSON, D.J. [6], VAN THYNE & McPHERSON, D.J.[7], SALLER, H.A., etal. [8], and more recently by HOFMAN et al. [9]. Their most important conclusions refer to the factthat the higher the molybdenum contents, the higher the gamma stability. Thus, alloys with higheralpha contents are, according to the reactions above, are more convenient to produce higher powderyieldings.But the methods of powder production are highly influenced by the methods of alloys production,since that, according to the previous works, structure stability is a function of the molybdenum contentand thus to the initial composition of the alloys. In all the alloys prepared by the arc melting technique,a subsequent thermal treatment was needed, even after remelts, to enhance their homogenization. Inthe induction ones, a high degree of homogenization could be obtained in only one step.In the present work a brief introduction about how the methods of fabricating the alloys influence thechoice of HDH condition is given. As there are no references on this issue in literature, our mainobjective is to show its importance in the nuclear technology, mainly in the fabrication of high densityγ-UMo powders.2. Experimental procedureTo the arc melted alloys, natural metallic uranium discs were used as charge. Molybdenum was usedas a charge in small cylinders with 3 mm height and 3 mm diameter. Both materials were assembled inthe copper plate of the furnace, its chamber closed and mechanical vacuum was applied to the system.After a suitable level of pressure, the vacuum valve was closed and a flow of argon was inserted insidethe chamber. The arc was opened by means of an arc-starter, and applied over the sample until a highlevel of mixing between uranium and the molybdenum charges was obtained. The typical time toreach this configuration over the charges was about 40 seconds to 1 minute, maximum values to avoiddamages in the chamber.This procedure was repeated several times until, after visual inspection, the observation that themolybdenum charge was well homogenized in the sample. The main disadvantage in applying severalremelts over the samples is the formation of an external oxide layer on the samples, deleterious to thequality of the alloy. Maximum masses of the alloys were about 30g.For the induction melts, natural metallic uranium cylinders with 7 cm height and 2 cm diameter, andthe same small cylinders of molybdenum were used as charge. Uranium cylinders were surfacecleaned with 65%vol. nitric acid, and inserted together with molybdenum, in a zircônia crucible, insidethe furnace chamber. A cycle of purge and mechanical vacuum was applied and, after 3 operations,argon was inserted and the power of the furnace was raised until the melting of the samples. Masses of700 g were utilized in each melting operations. The internal surfaces of the as cast alloys wereanalyzed by scanning electron microscopy, X-ray diffraction and its hydrostatic densities were alsomeasured. Some of the results are discussed bellow.3. Comparison between Arc and Induction Melting TechniquesTheoretical densities were obtained from some existent data in literature, for the 5 to 10 % weightmolybdenum, and were shown in figure 1, together with the experimental determinations. Theliterature data are from the paper of TRYBUS [10].

18densities (g/cm 3 )1716literaturearcinduction154 6 8 10wt % MoFigure 1 – Densities of uranium with additions of Mo, for the methods here studied.In terms of densities, alloys prepared by both methods behaved the same way up to 8% Mo, the fall inthe densities were almost at the same “rate”. After 8% Mo, the fall in the densities of the samplesprepared by arc was substantial, and must be due to some closed porosity, which doesn’t constitute aproblem in terms of powder obtention. It was observed also that this parallelism between the arc andinduction density curves was possible only after the application of at least 2 remelts in the arc samples.In the case of the induction ones, no remelts were necessary, even working with charges 30 timesheavier.However, the most important difference in terms of quality is microstructural. It was observed that thestructure of the induction alloys was mainly constituted by an homogeneous γ-UMo matrix plus someintergranular porosities, as we can see in the figure 2. In the case of the low-Mo alloys (5 to 7%wt.Mo), some α-U is also present in the grain boundaries. In the arc samples, a big number of dendriticstructures and some intragranular regions containing α-U, even after the operations of remelts, wereobserved, which indicates some incompatibility between the speeds of cooling and diffusion ofmolybdenum in uranium in the samples, the first one faster than the second.Structures containing high amounts of alpha uranium are the main responsible for the ease on thehydration-dehydration operations, but they are also related to a low degree of homogenization of bothconstituents of the samples. Thus, it is expected that, in the hydration of the arc melted samples, therates of hydrogen absorption must be higher than that the induction rates. But if we are looking forhomogeneous powders, it is necessary to work with the induction ones, and to try to find methods toenhance the hydrogen incorporation by these alloys.Factors affecting the solidification of the alloys are mainly those related to the furnace’s project, likeits geometry of melting (crucibles) and charges (how to assemble the charge into the crucible),possible impurities introduced in the charges by the crucible and the arc base materials, and mainly thecooling system.As an example, X-ray spectra and micrographies of the γ-U8Mo compositions are shown in the figures2 and 3, where we observe a high degree of homogenization presented in the induction sample.Dendrites are regions of low molybdenum concentration, and thus, the most suitable to promote highrates of hydrogen absorption, due to the high affinity presented by hydrogen and α-U. They form, bychemical reaction, uranium trihydride, which leads, after dehydration, to the formation of powders.However, the remaining alpha uranium constitutes a loss of material, because there is no possibility toreconvert it to gamma, as it is usually segregated out of the gamma matrix.

350U8Mo300250intensity (cps)2001501005000 20 40 60 80 100scattering angle (2θ)Figure 2 - MEV image and X-Ray diffraction pattern of γ-UMo alloy, induction melted.140120100UA8intensity (cps)8060402000 20 40 60 80 100Bragg angle (2θ)Figure 3 - MEV image and X-Ray diffraction pattern of γ-UMo alloy, arc melted, thermally treated,same composition as the induction sample.4. Hydrogen AbsorptionThe experiments with hydrogen absorption for both alloys were carried out exactly at the sameconditions of gas flow and sample´s masses and form. As an example, we can see below that the ratesof absorption for those produced by arc melting was higher than for those produced by induction.0,80,70,6arcinductionmass absorbed (%)0,50,40,30,20,10,0-0,10 5000 10000 15000 20000time (s)Figure 4 – Comparision of hydrogen absorption betweenarc and induction melted samples, same composition.For the figure above we conclude that arc melted samples are more capable to absorb hydrogen thaninduction samples, since the rate of hydrogen absorption by the arc sample is approximately 4,5 times

higher than that for the induction one. But, as denoted in the micrographies, inhomogenities incomposition are the main responsible for this high absorption rate. If gamma uranium is considered themore favorable phase for a high density fuel, such anomalies are undesirable, and must be treated as aprocess or method of fabrication loss.Finally, in the case of the induction-melted alloys, the structures are much more homogeneous,presenting grains of a continuous gamma matrix and, sometimes, alpha precipitates in the boundaries,mainly in those of 5 to 7% weight Mo compositions. Thus, no homogenization thermal treatment isneeded. If we take as comparison the same compositions, as the initial Mo content presented by the arcalloys, easiest is the powder obtention, but of an alloy with no homogeneous composition.As an example, in the figure 5 are presented a micrography of a powder produced after thermaltreatment under hydrogen applied on a γ-UMo induction alloy.5. ConclusionFigure 5 – Powder particles of γ-UMo alloy.The difficulty of the obtention, in one single step, a microstructural homogeneous alloy by the methodof arc melting, its abrupt reduction in the values of densities in compositions greater than 8% wt. Mo,lead us to the conclusion that, to the obtention of the same quality presented by the induction meltedalloys, some features of the arc melting process must be changed, like the furnace’s geometry, numberof remelts, geometry of loading, and mainly the cooling systems and thermal treatments conditions.Our main solution to avoid the problems of homogeneity in the arc samples was the change in thenumber of remelts. However, the remelts promotes also the formation of oxides, introducingimpurities in the alloys.Thus, the use of induction as a method of fabricating γ-UMo alloys are the choice here in IPEN-CNEN/ Brazil. However, for low molybdenum alloys, where the problem of homogeneity is not too serious,arc could be used, conditioned to the application on the samples a sufficient number of remelts.The important fact relating to the techniques of fabricating γ-UMo alloys is that the arc-melted alloyspresent, in all compositions, several dendritic structures, which are regions of low molybdenumconcentration, mainly constituted by alpha-U phase, which leads to the application of anhomogenization thermal treatment. But, as our experiences shows, they are not enough to eliminate allthe dendritic structures, which are responsible to the differences in composition inside the grains.This can be an advantage in the production of the powders, but at a cost of the loss of homogeneity.To the best preservation of the structural integrity of the alloys and their compositional

homogenization, the induction melting techniques is considered here the best choice to produce thepowders of high gamma content alloys.6. References[1] BALART, S. et al., U-Mo Alloy powder Obtained by a Hydride-Dehydride Process, Proceedingsof the RERTR Meeting, Las Vegas, Nevada, March 2000.[2] SOLONIN, M.I., et al., Development of the Method of High Density Fuel Comminution byHydride-Dehydride Processing, 2000 International Meeting on Reduced Enrichment forResearch and Test Reactors, Las Vegas, Nevada, October 1-6, 2000.[3] PASQUALINI, E.E., et al, Scaling up the Production Capacity of U-Mo Powder by HMD Process,2002 International Meeting on Reduced Enrichment for Research and Test Reactors,Bariloche, Argentina, November 3-8, 2002.[4] PASQUALINI, E.E., Set up of U-Mo powder production by HMD process, 2004 InternationalMeeting on Reduced Enrichment for Research and Test Reactors, Chicago, USA, October2004.[5] REPAS, P.E., et al., Transformation Characteristics of U-Mo and U-Mo-Ti Alloys, Transactionsof the ASM, v.57, 1964, pp.150-163.[6] VAN THYNE, R.J., McPHERSON, D.J., Transformation Kinetics of Uranium-Niobium andTernary Uranium-Molybdenum Base Alloys, Transactions of the ASM, v.49, 1957, pp.576-597.[7] VAN THYNE, R.J., McPHERSON, D.J., Transformation Kinetics of Uranium-MolybdenumAlloys, Transactions of the ASM, v.49, 1957, pp.598-621.[8] SALLER, H.A., et al., The Constitution Diagram of Uranium-Rich Uranium- MolybdenumAlloys, Battelle Memorial Institute Technical Report BMI-72, Columbus, Ohio, 1951.[9] HOFMAN, G.L., MEYER, M.K., RAY, A., Design of High Density Gamma-Phase UraniumAlloys for LEU Dispersion Fuel Applications, Proceedings of the RERTR Meeting, São Paulo,October 1998.[10] TRYBUS, C.L., et al., Design and Fabrication of High Density Uranium Dispersion Fuels,Proceedings of the RERTR Meeting, São Paulo, October 1998.

European Nuclear Conference, The Nuclear Fuel CycleBrussels, BelgiumSTUDIES ON THE AGING AND CORROSION BEHAVIOUROF SIMULATED SPENT NUCLEAR FUELC. KÜTAHYALI, J. COBOS, M. AYRANOV, D. CUI*, R. PEHRMAN,D. STAICU, T. WISS, V.V. RONDINELLA, R. KONINGSEuropean Commission, Joint Research Centre, Institute for Transuranium ElementsP.O. Box 2340, 76125 Karlsruhe, Germany* Studsvik Nuclear AB, SwedenABSTRACTTo extrapolate data obtained on a laboratory timescale to time intervals relevant forextended interim storage or final disposal of high level nuclear waste in a geologicrepository is a tough challenge. Emphasis is put on key aspects affecting waste corrosion:in particular, the effect of accumulating alpha-decay damage and helium on spent fuel isstudied by monitoring relevant properties as a function of time and through annealing tests.The mobility/precipitation behaviour of actinides and fission products exposed tocontainer/repository materials immediately after dissolution from the waste matrix is alsoconsidered. Finally, highlights from the effort to optimize suitable and complementarysolution analysis tools for leaching experiments under repository conditions areschematically described.1. IntroductionThe aim of these studies can be summarized as follows:a) Assessing the long term mechanical stability of spent fuel, which will determine the level ofsafety for transport/handling of the waste after intermediate storage times, and/or will definethe surface exposed to groundwater attack, hence the radionuclides source term for the socalledinstant release fraction in the case of final disposal scenarios.[1-3]b) Defining and understanding the mechanisms and processes governing the chemical interactionbetween aged waste and groundwater, as these will directly determine the amount and the rateof release of radionuclides from the waste form.This paper reports some highlights of these studies, including items belonging to the above mentionedlines of investigation and also an example of the effort to assess and optimize solution analysistechniques applied to the corrosion experiments on nuclear waste forms.Possibly the most challenging aspect of the studies on safe disposal of high level nuclear waste (in thiscase spent fuel) is the necessity to extrapolate to long storage times the behaviour of the waste formobserved on an experimental timescale. Spent nuclear fuel available today is not fully representative ofaged fuel after hundreds or thousands years of storage because it has not experienced thecorrespondingly long accumulation of microstructural defects and of He due to α-decay. UO 2 sinteredsamples containing short-lived alpha-emitters, the so-called alpha-doped UO 2 , with specific activitiesspanning over six orders of magnitude were used to simulate spent fuel with different ages [1]. Highactivity material was used to rapidly accumulate levels of damage corresponding to spent fuel afterthousands or tens of thousand years of storage [2].If the aged spent fuel will become exposed to groundwater, corrosion mechanisms will cause themobilization of radionuclides. It is important to assess the fate of these radionuclides, especially withregard to the capability of near field materials to act as retarding agents slowing down the transport ofthe dissolved species. This paper presents data obtained from spent fuel leaching experiments undersimulated near field repository conditions [4] (in the frame of international cooperation) indicatingimmobilization of dissolved radionuclides on the iron canister material.

European Nuclear Conference, The Nuclear Fuel CycleBrussels, BelgiumAn adequate analytic support to corrosion tests involving multiple phases under realistic repositoryconditions must be provided. In addition to high sensitivity techniques particularly effective for traceconcentration analysis, such as Inductively Coupled Plasma - Mass Spectroscopy, suitable techniquesmust be available that can be applied on a variety of solution types and capable to analyze solutionscharacterized by strong matrix effects. To this aim, a comparison of Inductively Coupled Plasma -Optical Emission Spectroscopy (ICP-OES) and Laser Uranium Fluorescence Analysis (LUFA) isschematically described.2. Microstructural evolution of spent fuel during storageAs a result of the alpha-decay process, microstructural damage associated with the defects induced inthe lattice by the energy transfer from the alpha-particle and the recoil nucleus will be produced;additionally, helium will accumulate in the fuel.The effects of alpha-damage accumulation on the microstructure of spent fuel have been studied onalpha-doped UO 2 specimens containing different amounts of either 238 Pu or 233 U. In terms ofradioactivity level, these samples represent typical UO 2 fuels from reactor discharge up to millionyears of age [1].Figure 1 shows a set of TEM micrographs of alpha-doped UO 2 specimens having cumulated alphadamageranging over 6 orders of magnitude. The early stages are characterized by formation ofdislocation loops, progressively increasing in concentration and size. Small helium bubbles (1-2 nm)can be seen on the sample after a dose of 2 displacements per atom (dpa). The precipitation of heliuminto bubbles associated to the formation of the loops will result in swelling of the fuel during storage.The internal stresses caused by the formation of such defects might result in the embrittlement of thismaterial. XRD data show an increase of the lattice parameter which is in perfect agreement with theTEM observations: a rapid increase at low dose followed by a smooth evolution towards saturation(not yet achieved) [1]. The evolution of the hardness follows the same trend [2].All together the knowledge accumulated on the behaviour of UO 2 as a function of alpha-damageaccumulation may contribute to define a model to understand the formation of the high burnupstructure in irradiated UO 2 fuels [2-3].A difficult aspect to treat in this type of studies is related to the kinetics of damage recovery. Theassessment of diffusion coefficients, for helium for instance, but also for defects would help to predictthe behaviour over long period of time of these materials. Radiation induced diffusion could contributeto the mobility of gas. Experiments are on-going to determine the solubility of helium in UO 2 at highpressure and high temperature and to study the interaction of helium with the defects, hence diffusionin a damaged lattice.Fig. 1. TEM micrographs of alpha-doped samples showing the microstructure evolution with damageaccumulation indicated as dpa values in the interval 10 -5 - 2 dpa. The main feature in this range ofaccumulated dose consists of dislocation loops whose size and concentration increase with the dose.

European Nuclear Conference, The Nuclear Fuel CycleBrussels, Belgium3. Specific heat of α-doped samplesAnother example illustrating the effects due to the accumulation of alpha-decay damage in UO 2 refersto the energy stored in the defects produced in the lattice.Figure 2 shows the results of differential scanning calorimetry (DSC) measurements on alpha-dopedUO 2 containing 10wt% of an oxide constituted mainly by 238 PuO 2 after accumulating approximately 1dpa. The apparent specific heat, C p * , of damaged samples was measured by DSC by applyingascending and descending temperature programmes in the range 400 K – 1500 K. The deviation of themeasured C p * (T) from the real heat capacity, C p (T), is related to the recovery of the latent heat of thelattice defects during thermal healing.Calorimetry of strong α-emitters is perturbed by the heat generated by radioactive decay. In fact, theapparent temperature-ascending curve of C p * is lower than the real one, whilst the descending curve ishigher. However, the average of these two curves gives exactly the value of the unbiased C p . The α-decay heat generated by the sample is known to be 0.0702 Wg -1 for 10 at% 238 Pu with 5.499 MeVenergy per α-particle and the same energy for the recoil daughter. Knowing this energy source, whoseeffects are perfectly anti-symmetric in the ascending and descending curves, the calorimetric signalproduced during damage annealing could be accurately measured and converted into energy.The real C p (T) obtained from literature data [5] for (U 0.9 ,Pu 0.1 )O 2 corresponds very well to the averageobtained between the ascending and descending curves of undamaged samples.0.35Apparent C pJ g -1 K -10.300.250.20600 800 1000 1200 1400Temperature, KAscending 2 nd seriesDescending 2 nd seriesAscending 3 rd seriesDescending 3 rd seriesFig. 2. Apparent Cp * curves for (U 0.9 , 238 Pu 0.1 )O 2 obtained by DSC during temperature ascending anddescending runs (i.e. before and after annealing). Two experiments were performed (duplicate runs).The peaks of the latent heat effects appearing during the annealing (ascending) runs at temperaturescorresponding to different healing stages can be deconvoluted and analysed in order to identify theparameters of the latent heat effects. For each stage, the quantities derived are: concentration andenergy associated to the annealing of a certain kind of defect, as well as its characteristic mobility [2].4. Reduction and incorporation of redox sensitive nuclides in corroded iron surfacesTo understand the fate of redox sensitive radionuclides released from high level radioactive waste,secondary mass ion spectroscopy (SIMS) analysis was conducted on iron coupons reacted for twoyears in spent fuel leaching experiment performed at Studsvik Nuclear AB, Sweden in the frame of aSKB project [4].

European Nuclear Conference, The Nuclear Fuel CycleBrussels, BelgiumThe concentration of radionuclides in solution from the leaching of a 47 MW/kgU spent fuel sample insynthetic groundwater dropped quickly after inserting iron metal coupons in the leaching solution.This is explained as reductive precipitation of these redox sensitive radionuclides on the iron surface.SIMS analysis (Fig. 3) confirmed that U(VI) and Pu(VI)/Pu(V) are reduced and immobilized in thecorrosion layer on the iron surface. Most U and Pu are coordinated with the inner parts of thecorrosion layer, where Fe(II)-rich corrosion products are dominating and therefore more reductivethan the Fe(III) oxides occurring in the outer periphery of the corrosion layer. The result of SIMSanalysis also showed that Si was coordinated with iron corrosion products in both inner and outer partsof the corrosion layer. From these data it can be concluded that under simulated near field conditionsthe redox sensitive radionuclides released from spent fuel can be reduced and precipitated on corrosionlayer of the proposed canister material, i.e. cast iron.99 Tc28 Si28 Si237 Np56 Fe88 Sr238 U239 PuFig. 3. SIMS analysis of the cross section of reacted surface of cast-iron coupons. Nuclide mappingshows the incorporation of U, Pu and Np in the corrosion phase of Fe. The bright yellow parts in Femapping represent the cast-iron metal matrix. Si was from the glass vessel or contained in cast iron.5-Uranium analysis by ICP-OES and laser fluorescence techniquesA comparison between ICP-OES (Horiba Jobin Yvon Ultima 2 spectrometer) and LUFA (ScintrexUA-3) has been performed on uranium solutions, to assess the complementarity and range ofeffectiveness of the two techniques. In both cases, external calibration was performed using certifieduranium standard solutions (Alfa Aesar Specpure). The samples analysed by ICP-OES had dilutionfactors between 0 and 15. A high generator power was used to minimize the matrix effects. Thewavelength was detected using the profile function, and by using a semi-quantitative analysis modeusing multiple wavelengths. Two analytical lines of uranium, 367.007 nm and 409.014 nm were usedwith best limit of detection of 1.11 ppb and 4.31 ppb, respectively. Three replicas per line and elementwere performed for each sample. The samples analysed by LUFA had dilution factors between 0 and7400. The intensity was zero for a uranium concentration

European Nuclear Conference, The Nuclear Fuel CycleBrussels, Belgiumconcentration levels (ppb to hundreds of ppm), are fast and precise, and can be used complementarilyfor the measurement of uranium. LUFA accuracy was ~5%, while in the case of ICP-OES accuracybetter than 1% is achievable.100000ICP-OESLUFAGroup VIGroup VII10000Concentration, ppb1000100Group IIGroup IIIGroup IVGroup V10Group I1Fig. 4. Measured uranium concentrations with 67 % level of confidence.A procedure to analyze uranium in solutions characterized by strong matrix effects (brines) wasdeveloped for ICP-OES. Such effects constitute a limiting factor for the use of the LUFA. ICP-OESresults for U in high concentration NaCl solutions indicated that the background shift caused bysodium can be corrected and reliable analyses of uranium can be performed in the brine matrix.6. SummaryPredicting the behaviour of spent fuel during storage and/or in a geologic repository is a complex taskrequiring simulations and extrapolations to a remote future. The related studies must cover a broadspectrum of mechanisms and processes, including the evolution of the waste under the effect ofaccumulating alpha-decay damage and He, and the corrosion behaviour in presence of groundwaterand repository agents. The first aspect is investigated by measuring property changes and recoverybehaviour of materials under the effect of accelerated dose accumulation at a microscopic andmacroscopic level; the second is studied by singling out the governing mechanisms determining therelease and mobility of radionuclides. Appropriate analytical tools must be available to characterizethe systems of interest, especially when simulating realistic, multi-phase repository configurations.References1- V.V. Rondinella, J. Cobos, T. Wiss, J.-P. Hiernaut, Proc. ICEM’03, 9 th International Conference onEnvironmental Remediation and Radioactive Waste Management Oxford, UK, Sept 21-25, 2003,paper 4593, ASME 2003.2- V.V. Rondinella, T. Wiss, J.-P. Hiernaut, Proc. ICEM’07, Bruges, Belgium, Sept. 2-6, paper 7322,ASME 2007.3- J. Jonnet, J. Rest, P. Van Uffelen, D. Staicu, T. Wiss, C. Ronchi, Proc. IAEA Technical Meeting“Fuel Behaviour Modelling under Normal, Transient and Accident Conditions, and High Burn-ups”Kendal, UK, Sept. 5 – 8, 2005.4- D. Cui, J. Low, M. Lungdsström, K. Spahiu, Mat. Res. Soc. Symp., 807 (2004) 89-94.5- C. Duriez, J.P. Alessandri, T. Gervais, Y. Philipponneau, J. Nucl. Mater., 277 (2000) 143.

THE CURRENT STATUS AND FURTHERDEVELOPMENT OF RADIATION ENGINEERING FORNUCLEAR POWER, INDUSTRY AND MEDICINEN.R.KuzeliovRussian National Technical Physics and Automation Research Institute (VNIITFA), MoscowABSTRACTIn the paper there are described the basic foundations of radiation engineering and there arereflected the questions of a current status and further development of radiation engineeringin Russia. There are given some examples of practical use the research and development inthe field of radiation engineering for the solution the problems of nuclear power engineeringand nuclear fuel cycle, metallurgy, geology, mining, gas and oil and other industries,medicine, ecology etc. There are described the main methods and means of radiation engineering.Radiation technology installations are used for food treatment, for sterilization ofmedical products as well as for purification of drinking and waste water. Radionuclidepower units are applied for power supply of the automatic equipment on the remote or unattendedobjects in Polar areas or in Space. Radiation instruments are used for quality testing,for analysis of substance composition and for control of production processes, as well as forradiation diagnostics and radiotherapy of human diseases. The information on gas dischargedetectors of radiations, which are developed and are producing in VNIITFA, is given also.1. IntroductionThe artificial radioactive nuclides and other sources of ionizing radiations are widely applied today invarious branches of industry, medicine and agriculture. It became possible only due to development ofresearches in the field of nuclear physics, progress in the nuclear power and accelerators engineering,large-scale manufacture of nuclides, success in instrumentation and computerization.The listed factors have caused appearance of a specific direction in a nuclear science and engineering– the radiation technology and engineering. The radiation technology and engineering is a complex ofnuclear and physical methods and radiation devices based on use of interaction of ionizing radiationswith substances. It is intended for treatment on materials and substance to receive new or to change theexisting properties of materials, to realize the radiation therapy of various man’s diseases, to obtain theinformation of qualitative and quantitative data of composition and properties of materials and products,to control and manage production processes, and also to transform energy, which emits as a resultof radioactive decay, in other kinds of energy.To the present time three basic directions of radiation engineering were defined: radiation technologyand equipment, radiation instrumentation and radionuclide power engineering.The aim of radiation technology is creation the methods and devices, which are based on chemical,structural or biological changes, taking place in substance under influence of the certain absorbed dozeof ionizing radiationThe radiation instrumentation covers methods and equipment for obtaining the information on parametersof a condition, quality either composition of substances or products functionally connected to aflow of ionizing radiation.The aim of radionuclide power engineering is creation various power devices, in which as a source ofprimary energy is the energy, which emits as a result of processes, accompanying radioactive decay.A complex of scientific and technical problems exists in all these directions, many of such problemswere developed in independent scientific disciplines, for example, radiation chemistry, gammaradiationtherapy, activation analysis, radiation methods of the non-destructive testing etc.Application radioactive nuclides and the radiation engineering in various areas of human activity rapidlyincrease all over the world.Many institutions and enterprises in our country and abroad are engaged in development and productionof radiation engineering equipment and equipment for use of radioactive nuclides and othersources of ionizing radiation. The international contacts in this area are intensively growing.

The current status of works in the field of radiation engineering and radionuclides application is characterizedby the following:- There is an expansion of the areas of biomedical application of radiation methods for treatment anddiagnostics of various diseases, for sterilization of medical production, for irradiation of food and agriculturalproduction, for treatment of drinking water and wastes;- There is a definite delay in an occurrence of new methods and development of essentially new typesof the radiation equipment against a background of quantitative growth of used products of radiationengineering- There are qualitative changes in developing products, which are bound up with wide application ofcomputer engineering and automatics, use of more perfect converters, detectors of radiations, optimizationof the characteristics and activity of irradiators etc., that allows essentially improve the data ofequipment on accuracy, sensitivity, productivity, enables to use the received data for automation ofmanufacture;- The data of radiation safety of used equipment are improved owing to automation of irradiationequipment control, optimum choice of sources specific activity (down to application in some cases ofsources with minimally significant activity), to application of new shielding materials and perfectionof a radiation protection design.The successes and achievement of radiation engineering in Russia in many respects are bound up withinvestigations in Russian National Technical Physics and Automation Research Institute (VNIITFA),which till 1989 referred to as All-Union Research Institute of Radiation Engineering. In institutealongside with other works in the field of technical physics, the works on radiation technology andcreation of the irradiation equipment, gamma beam therapy machines, radionuclide power units, radiationmeans for non-destructive testing, nuclear analytical technique for the elemental analysis of substance,etc. were received the large development.It is possible to show some tendencies of development of radiation engineering in Russia on examplesof the most interesting researches and development of the radiation equipment, which has been carriedout recently.2. Nuclear analytical methods and equipmentMuch attention is devoted in VNIITFA to researches and development of nuclear and physical methodsand equipment for the analysis of substance composition. The especially significant progress wasachieved in the neutron activation and neutron absorption analysis, in photonuclear methods and in radionuclideX-ray analysis.It is can be said, that VNIITFA specializes in specific questions of nuclear and physical methods ofanalysis for industry. The researches spent for this area, carry the brightly expressed applied characterdirected on creation of devices and complexes of the equipment, which finds the main application inan industry, in mining and geology.The neutron activation analysis possesses very high sensitivity of various elements determinationwhen nuclear reactor is used as a source of neutrons. A number of reactor activation analysis techniquesare developed in institute. In particular, the neutron-resonance method developed in VNIITFAis successfully used for instrumental determination of some noble metals.When small-sized generators of neutrons and powerful radionuclide sources of neutrons made of 252 Cfbecame available, the researchers of institute carried out researches and development on creation ofstationary and transportable neutron activation analytical device for an industry and geology. The specialattention in methodical development of institute was paid to activation analysis with use of shortlivednuclides. That allows to receive high efficiency of the analysis of separate samples and even toconduct the continuous analysis of technological products.The neutron absorption analyzer was developed in VNIITFA for determination of elements with highsection of neutrons absorption. In particular, now majority of atomic power stations in Russia with reactorsof VVER type and many foreign NPP (in Bulgaria, Czech Republic, Slovakia, Ukraine) areequipped with developed in VNIITFA neutron analyzers of 10 B in the coolant of the first contour ofreactor.

a) b)Fig. 1 The scheme of the neutron analyzer for boron-10 concentration determination in coolant and intechnological solutions at NPP with reactors of VVER typeThe gamma activation analysis is applied as the analytical tool for solving of various problems. Thismethod possess with sharp selectivity, sensitivity, high accuracy and rapidityIn order to achieve these advantages the scientists of institute carry out the important work on moreprecise definition the nuclear data for photonuclear reactions. The results of these researches were appliedfor development of the industrial gamma-activation equipment.The laboratories for gamma-activation analysis of gold ores were created in VNIITFA. High efficiencyand rapidity allow using the results of the ore samples analysis for directional ore mining.The linear electron accelerators are the most convenient gamma-radiation sources for realization ofgamma-activation analysis. The electron accelerators allow receiving not only intensive bremsestrahlung,but also very high neutron fluxes. So, for example, in installation "Arctica" the acceleratorLUE-15А with beryllium converter allows to have a neutron flux up to 10 13 neutrons per second. Itsignificantly expands a circle of analytical tasks can be solved by application of neutron activationanalysis in a complex with gamma-activation with use of electronic accelerators.Fig. 2 Installation for gamma-activation analysis of gold ores using the linear electron acceleratorThe radionuclide X ray analysis is the method based on excitation of atoms of analyzed elements withthe help of primary radiation from radioactive nuclide. It is one of the most widely used analyticaltechniques.Simplicity, the small dimensions and cost of the equipment in a combination with sufficient sensitivityfor solving many practical tasks made the radionuclide X ray method the most convenient analyticaltool for mass and multielement express analysis of technological or geological samples. It applies alsofor analysis of rocks and ores in-situ, automatic control of technological products in a flow etc. Allthese directions have developed in Russian National Technical Physics and Automation Research Institute.The scientists of institute have brought in the significant contribution to development of scientificbases of the radionuclide X ray analytical method and creation of devices for various purpose intendedfor operation in field, laboratory and work shop conditions.At processing results of the nuclear-physical analysis the great importance is attached to correct decodingof radiation spectra, account of interfering influence of other elements. It concerns both to activationand to radionuclide X ray analysis. Therefore in institute the special computer methods and al-

gorithms of processing of spectra are created, the great importance is attached to development of variousphysical principles for exception of interfering influence of a matrix.Fig. 3 Portable radionuclide X ray analyzer ПРАМ-1 for geology3. Non destructive testingThe wide range of problems for the production control in various branches of industry is successfullysolved with use of the equipment of the non-destructive testing and radiation gauges for determinationof various technological data.The means for non-destructive testing, applying mainly radiation methods, for an estimation of qualityof products made of various materials and possessed a complex configuration and technology ofmanufacturing are developed in VNIITFA.Due to high penetrating ability of radiation, in particular, of gamma-quanta and neutrons, it is possibleto transfer the information in form of radiation signal through walls of production equipment, pipelines,capacities etc.Development of methods and equipment for gamma and neutron radiography, self radiation radiography,radiometric flaw detection, radiation methods of the determination of such parameters, as level,density, throughput rate of materials, thickness of walls and coverings, granulometric composition ofpowders, distribution of material, the presence of interstices and cracks etc., allows to solve set ofproblems taking place in industries.Use of packaged systems and typical structures, unification and normalization of elements of the developedequipment, wide application of computers and new types of radiation sources allow to createthe equipment of the non-destructive testing and technical diagnostics at a level of the best worldmodelChoice of detectors with the maximal "signal - noise" ratio, PC-processing of the radiographic and radiometricinformation allow to increase volume of information, to receive the volumetric and level-bylevelimages of controllable products. These investigations were used as a base of industrial computertomography methods and devices, developed in institute.Fig. 4 Radionuclide computer tomography device

Fig. 5 The tomogram of the nuclear fuel elements assembly for NPP, received with help of industrialradionuclide computer tomography deviceFor realization of radiography at small focal distances the special acute-angled focal sources are createdat the basis of isotope 192 Iг. The testing of thin-walled products will be carried out with use ofsources low energy gamma radiation at the base of radionuclide 75 Se etc.Fig. 6 Gamma radiography flaw detector RID-Se/4 with a source 75 SeOn the basis of researches carried out by scientists of VNIITFA in cooperation with the experts of anindustry it was created the line of gamma radiography flaw detectors for general application and theline of radiometric flaw detectors, allowing to investigate the complex shape products, objects locatedin out-of-the way places, details and units in process of their manufacturing, welded seams of mainpipelines, power equipment etc.Fig. 7 Gamma-radiography flaw detector RID-К/100 using a 60 Co source for transmission of productswith a wall thickness up to 200 mm on steel

Development of nuclear power in our country, necessity of increase of its efficiency and the safety definethe increased requirements to quality of the NPP equipment, including nuclear fuel elements. Thehigh emphasis is placed on development at institute the means for non-destructive testing of objects innuclear power engineering.The development of the set of means for non-destructive testing in process the manufacture of nuclearfuel elements for power reactors VVER-1000, RBMK, BN will be carried out by VNIITFA in cooperationwith a number of other institutes and enterprises. The purpose of this work - to supply the enterprises,producing the nuclear fuel elements, with the automated means for testing the fuel elementsat all stages of their manufacturing, starting with nuclear fuel powders up to ready nuclear fuel assemblies.Fig.8 The automatic transfer line for fuel elements manufacture, equipped with means of non destructivetestingFor study both not irradiated, and irradiated fuel elements the methods neutron radiography can be applied.This method is used also for study of many other objects. The investigations on application ofneutrons for non-destructive testing were carried out in VNIITFA within many years. Neutron radiographyallows obtain the information on parameters of products located behind shields from heaviermaterials, to reveal inclusions of hydrogen containing materials, of substances with the large neutronabsorptioncross-sections etc. In VNIITFA there were created some installations for neutron radiographywith use of high active radionuclide 252 Cf neutron sources, and also beams of neutrons producedby nuclear reactor.A lot of industrial technological parameters can be determined with the help of radioisotope instruments.The instruments of this type developed in institute, are used for measurement of humidity of coke inmetallurgy, soils and spoils in construction works, size grading of the crushed ores and powders, inspectionof filling the bins of feeding systems of blast-furnace burden, control by anti-icier systems ofcivil aircrafts etc.

Fig. 9 Neutron moisture and density meter of coke4. Gas-discharged radiation detectorsThe gas-discharged detectors of ionizing radiations are the major element of the instruments for radiationmeasurements. The gas-discharged detectors are simple on a design, are convenient in work, providereliable measurements of various kinds of radiation in a wide range of the external influencingfactors.The fields of application of discharge detectors are wide and are various. They are used in inspectionand protection systems of nuclear power units, in systems of anti-nuclear weapons protection of industrialand civil objects, instruments of the radiation inspection of the nuclear power plants personnel,personnel of a nuclear industry enterprises and population of the country, and also in the geophysicalinstruments for minerals investigation, devices for space researches etc.There were created the essentially new types of gas-discharged detectors (such, as corona counters andintegral-pulse chambers), which are used for the solving of a number of problems in reactor engineering.Fig. 10 Gas-discharged detectors of radiationsThe development of nuclear industry and nuclear power engineering has required the creation of neutronionization chambers for registration of the neutron fluxes in control and protection systems of nuclearpower plants and research reactors.The ionization neutron chambers with solid boron radiator and ionization fission chambers containingas a radiator fissionable isotopes of uranium are widely applied in nuclear industry. They possess thehigh stability of intensive neutron fluxes registration.The nomenclature of the gas discharge detectors of nuclear radiation created and produced in institute,numbers more than 60 types. Among them there are, for example, neutron counters both with solid boronand with gaseous 3 Не radiators.

Fig. 11 Ionization chambers of a various typeBesides of gas discharge detectors of radiations VNIITFA develops and produces also the suspensionbrackets of ionization chambers - the vacuum and gas-filled communication lines of ionization chamberswith the electronic recording equipment.The cited above review shows, that the radiation engineering becomes an advanced sphere of a scienceand engineering, the areas of its application are wide - from nuclear power and metallurgy up to agricultureand medicine.5. Radiation technology equipmentThe radiation technology is based on physical, physics and chemical and biological processes takingplace in substances as result of interaction with ionizing radiations and causing the appropriate technologicaleffect.To the present time some dozens of various radiation technological processes are developed and theyare at various stages of trial and industrial realization on the following directions of radiation technology:to synthesis of new chemical compounds and initiation of chemical reactions, polymerization,vulcanization and cross-linking of polymers, clearing of waste waters and gases, sterilization of materials,irradiation of foodstuff for increase of terms of their storage, giving of new properties to solids,in particular, to semi-conductor materials etc.The significant contribution was brought by the researchers of VNIITFA to creation of scientific andtechnical bases of radiation-technological process and equipment. It was developed the complex of accountsthe basic parameters of radiation technological installations, methods of irradiator optimization,definition of the rational circuits of management, control, protection and automation of the radiationequipment. The methods of definition of dose field topography, operating ratio of radiation for variousconfigurations of irradiated objects were developed also. The large attention was given to monitoringof powerful fields of radiation.The most highly developed area of radiation technology is the radiation sterilization, which is especiallyeffective at processing of heat labile materials, for example, medical products made of polymericmaterials (syringes-ampoules of single application, catheters, catgut, devices of a taking andtransfusion of blood etc.), dressings etc. A radiation sterilization is carried out with use of both isotopesources of radiation, and electron accelerators. The plants, developed in VNIITFA with 60 Со radiationsources of activity up to 1 МCi for sterilization of blood systems, operate at the factories in St.-Petersburg and in Belgorod-Dnestrovsky. The installation with two linear electron accelerators ЛУЭ-8/5В type for sterilization of medical production operates in Kurgan. The modern enterprise for manufacturingobstetrical packages equipped with gamma-installation for their sterilization has been constructedin Kondrovo in cooperation with companies from Finland. A number of other installations forthe enterprises of medical products are created.The radiation treatment is used also for processing of food and agricultural products. It is known, thatfor the various reasons (rotting, germination, the damage by insects etc.) lose a plenty of the foodstuffsand agricultural products (raw material, seeds etc.). VNIITFA conducts also works in this very importantdirection of radiation technology. For many kinds of products the optimum modes of radiationprocessing are established, the long-term researches of their suitability and harmlessness of use arecarried out. The irradiation installation for these purposes are created, a part of them was supplied forforeign customers.

Fig. 12 Radiation installation in Republic of Peru has been constructed with the help of theVNIITFA expertsThe irradiation exerts stimulating influence on course of a number of chemical reactions. The radiation-chemicalreactions in comparison with reactions initiated by a heat or other kinds of energy, characterizedby weak dependence of velocity of initiation from temperature, opportunity of realizationwithout application of the additives and catalysts, dependence of velocity of initiation on a doze rate.Radionuclide sources of radiation ( 60 Со) as well as the powerful electron accelerators are used inplants for realization of radiation-chemical processes. The release of a lot of unique products and materials,such, as self-adhesive electric insulating tape and rubber glass cloth, thermo-shrinking polyethyleneisolation, cotton fabric with anti-microbe, haemostatic, anti-putrefactive properties, syntheticfabrics with the improved consumer properties, is under production conditions adjusted.The great importance is attached now to protection of an environment. Many problems in this importantdirection can be solved by radiation processing of gaseous and liquid industrial, household and agriculturalwastes. The installations developed by VNIITFA are used at chemical and medical instrumentsplants and can be used for prevention of pollution of an air and natural waters.Fig. 13 The circuit of installation for electronic processing of drinking and waste water1 - electron accelerator, 2 – reactive chamber, А1 - air, А2 - air in an atmosphere, Р – collection of effluents,W - treated water

Fig. 14 The building of installation for radiating processing of industrial sewages in VoronezhThe further development of works in the field of radiation technology equipment is directed on creationof the specialized industrial installations, increase of their technological and economic data, reliabilityand convenience in operation.6. Radionuclide power engineeringEnergy, released as a heat following a radioactive decay, can be then transformed into electrical energy.Radionuclide sources of thermal and electrical energy have essential advantages at operation of theequipment with the such sources, mounted in remote regions of Globe, which are difficult to access,due their high specific energy capacity (thousand W·hour/kg), long service life (10 years and more),high reliability, keeping of serviceability by short circuit.Therefore radionuclide power units (radionuclide thermoelectric generators (RTEG), the radionuclidepower devices (RPD) etc.), developed and produced in VNIITFA ("Бета-C", "Бета-M", "Эфир - МА"etc.), find rather wide application for power supply of automatic hydro meteorological stations, variousmeans of navigation - sea lights and radio beacons. The radio beacons, located on a line of NorthernSea Route, promote significant increase of navigation safety. More powerful radionuclide powerinstallations are applied to the power supply of light beacons. Such installation, created in institute,was equipped, for example, Tallinn beacon by the Baltic Sea. In Antarctic Region RTEGs are used fora power supply of magnetic variation stations.Fig. 15 RTEGs are used for power supply unattended equipment in areas of Far North.The radionuclide power engineering is on a junction of several sciences: nuclear physics, radiochemistry,thermal physics, the electrical engineering etc. Therefore, the experts of a various structure manyscientific establishments and industrial enterprises of our country in creation of RPD elements takepart.The scientific and technical bases of RPD creation, principles of designing, mathematical modeling interconnectedto nuclear physical, thermal physical and electrical processes are developed in RussianNational Institute of Technical Physics and Automation with the purposes of optimization the parametersof separate parts of RPD and device as a whole.As the basic fuel for radionuclide thermoelectric generators 90 Sr is mostly used, and 238 Рu is used forhighly power intensive RPD.In RPD, developed in VNIITFA, the thermo electric batteries are most widely used for transformationof energy, released by radioactive decay, to electrical power. The current researches are directed on rationaluse and other methods of transformation.In institute the works on creation of the radioisotope power supply for the pump of artificial heart werecarried out. Such source can be implanted directly in the body of patient together with the mechanismof artificial heart. However these works were bounded only by experimental stage, and in the near future,probably, the external systems of artificial blood circulation will be in using.

7. Radiation therapy equipmentThe research and development in the field of radiation medicine are carried out at VNIITFA from themoment of its foundation and they are directed on creation of the equipment for beam therapy (mainlymalignant neoplasm). The works were carried out in close collaboration with leading medical institutesand clinics of the country.The complex of researches, executed in VNIITFA, has allowed proving the main radiation and physicaldata of remote and contact therapy equipment with radionuclide sources of radiation, to developscientific and technical bases of its designing, to create methods and means of formation of radiationfields with the given data, to ensure their stability during beam treatment. The algorithms and computerprograms were developed for account the dose distributions from sources of gamma and neutronradiation, which are handed to medical institutions for clinical application.The modern radiological equipment for intracavity and intratissue gamma-therapy with high active radiationsources for application in oncoginaecology, proctology, urology, teratology (АГАТ-В1,АГАТ-В5, АГАТ-ВТ), developed in VNIITFA, is a great scientific and technical achievement in areaof radiation therapy.Fig. 16 The device АГАТ-ВТ for contact gamma-therapyIn institute were developed and continue to be improved remote gamma-therapy devices, static and rotationaltypes (such, as АГАТ-C and АГАТ-Р, АГАТ-РМ, АГАТ-PI). The devices with programmedcontrol systems were developed on the basis of modern microcomputer. This equipment allows passingfrom automation the separate stages of beam treatment to complex automation the process of preirradiationpreparation, to development of the individual optimum programs of beam treatment andtheir realization that considerably raises treatment efficiency of the patients.Fig. 17 The device АГАТ-Р1 for remote gamma-therapy

VNIITFA developed also the device for intracavity neutron therapy with use of a radionuclide sourceof neutrons made of Californium-252.To the present time oncological health centers, clinics, the research institutes of Russia are equippedwith several hundreds r5adiotherapy devices most of them was developed in VNIITFA.Other major direction in modern medicine is effective diagnostics of some serious diseases. In our instituteis offered radionuclide X-ray fluorescence method and is realized in equipment for diagnosticsof thyroid glands diseases and for control of its treatment. This equipment allows in tens time to reducea dose load on organism of the patient in comparison with well known radioisotope methods ofthe iodine control in thyroid gland. This method gives qualitatively new opportunities for diagnosticsand to carry out the control of process of treatment.Fig. 18 The equipment for diagnostics of thyroid glands diseases and control of their treatmentIt is already more than 40 years as VNIITFA is the leading scientific establishment on creation of newmethods and devices applying radionuclide sources of radiation and accelerators. It seems a further useof radiation engineering will develop rapidly.References1. Radiation Engineering. Moscow. Atomized, 1967-1974, issues.1-102. The Questions of the Nuclear Science and Engineering. Series. Radiation Engineering. Moscow.Energoatomizdat, 1975-1992, issues. 11-47.3. The Questions of the Nuclear Science and Engineering. Series. Technical Physics and Automation.Moscow. CNIIatominform, 1993-2000, issues.48-554. Nuclear Industry of Russia. Мoscow, Energoatopmizdat, 2000, pp. 973-990.

Poster PresentationsSession III:Medical Applications

LONG TERM RETENTION OF D- AND L- [ 123 I]-2-IODO-PHENYLALANINE IN R1M TUMOUR BEARING RATS:A POTENTIAL FOR RADIONUCLIDE THERAPYM. BAUWENS, T. LAHOUTTE, K. KERSEMANS, C. GALLEZ, A.BOSSUYT, J. MERTENSBEFY, Vrije Universiteit BrusselLaarbeeklaan 103, 1090 Brussel, BELGIUMABSTRACTThe aim of this study was to compare in vivo [ 123 I]-2-I-D-phenylalanine and[ 123 I]-2-I-D-tyrosine with their respective L-analogues [ 123 I]-2-I-Lphenylalanineand [ 123 I]-2-I-L-tyrosine, while using [ 123 I]-α-Methyl-Ltyrosineas a reference. The tumour uptake and biodistribution for all tracerswas measured in the same set of R1M rat rhabdomyosarcoma tumour bearingWag/Rij rats by dynamic planar imaging and by static planar imaging at 24and 48 hours. All tracers showed a favourable tumour uptake at 30 minutesp.i., with tumour DUR values ranging from 1.7 to 2.4. After 24 hours bothtyrosine analogues but also IMT were almost completely cleared. There wasa high retention however of both [ 123 I]-2-Iodo-D-Phe and [ 123 I]-2-Iodo-L-Pheactivity in the tumour and body, resulting in an increased tumour/backgroundcontrast over time. In conclusion, both the D and L analogues of [ 123 I]-2-Iodo-Phe have a favourable tumour uptake and biodistribution, but moreovera very long retention in the tumour allowing radionuclidetherapy.1. IntroductionThe LAT transport system is an important key in tumour imaging using radiolabeled amino acids. Thistransporter is a major nutrient transport system responsible for Na + - independent transport of largeneutral amino acids including synthetic amino acids by an obligatory exchange mechanism coupled toan anti-port system [1-4], and has several subtypes such as LAT1 and LAT2. LAT1 expression wasscarcely detected in non-tumour areas [5-6] but highly expressed (up-regulated) in proliferatingtissues, in particular malignant tumours, as it plays a critical role in cell growth and proliferation. Aremarkable characteristic of the LAT-1 amino acid transport system is its broad substrate selectivity,which enables the transporter to accept amino acid related compounds, such as D-amino acids andcancer drugs like Melphalan [3,6]. LAT2 on the other hand has a high level of expression in smallintestine, kidney, placenta, brain and in epithelia and blood-tissue barriers [4,7]. It transports all of theisomers of neutral alpha-amino acids by facilitated diffusion; however it does not transport D-aminoacids.Kersemans et al [8] have shown in tumour bearing athymic mice that D and L [ 123 I]-2-I-phenylalanine([ 123 I]-2-I-Phe) were taken up in different types of tumour cell lines. They also noticed that while thetumour clearance of [ 123 I]-2-I-D-Phe was faster than of the L analogue, at 24 hours p.i. the tumourbackgroundcontrast of the D analogue was 3 times higher than of the L analogue. Recently, Tsukadaet al. demonstrated in tumour bearing mice the advantages of D-isomers of O- 11 C-methyl-tyrosine andO- 18 F-fluoromethyl-tyrosine as Tumour-Imaging Agents as the tumour to blood ratio was higher forthe D-isomers compared to the L-isomers [9].

2-I-L-tyrosine (2-I-L-Tyr) and 2-I-L-Phe [10-13] both show a very high tumour selectivity whencompared to [ 18 F]-FDG which shows a considerable uptake in brain and inflammatory tissue. Here wecompare the pharmacokinetics of [ 123 I]-2-I-L-Phe, [ 123 I]-2-I-D-Phe, [ 123 I]-2-I-L-Tyr, [ 123 I]-2-I-D-Tyrand [ 123 I]-α-Methyl-L-Tyr in a R1M tumour bearing Wag/Rij rat model. The biodistribution, thetumour uptake and tumour retention of these tracers are investigated via dynamic and static planarimaging, both at short time p.i. (0-40 minutes) and long time (24 and 48 hours p.i.). Dissectionexperiments are used to confirm and elaborate biodistribution data of [ 123 I]-2-I-L-Phe and [ 123 I]-2-I-D-Phe.2. Material and methodsPrecursor synthesis and labellingThe synthesis of 2-I-D-Phe / 2-I-L-Phe and the nucleophilic radioiodination of 2-I-D-Phe, 2-I-L-Phe,2-I-L-Tyr and 2-I-D-Tyr were achieved by Cu 1+ assisted nucleophilic exchange as earlier described[11]. [ 123 I]-α-Methyl-L-Tyr labelling was achieved via the iodogen method [14]. The labelling yieldsfor all tracers were at least 98%. After passing through a sterile 0.22 µm Ag-membrane filter(Millipore) a radiochemical purity of > 99% was achieved.Laboratory AnimalsFood and water was ad libidum during the experimental period. For the tumour model, male Wag/Rijrats (Bioservices, The Netherlands) were injected subcutaneously in the right flank (armpit region)with 15.10 6 R1M rhabdomyosarcoma cells. All imaging experiments and all dissection experimentswere performed 6 weeks after injection of the R1M cells. All tracers were injected intravenously (IV)in the penis vein. The study protocol was approved by the ethical committee for animal studies of ourinstitution. Guidelines of the National Institute of Health principles of laboratory animal care (NIHpublication 86-23, revised 1985) were followed.Dynamic Planar Imaging (DPI)DPI was performed in a crossed (two by two) experiment with a 2-day interval using four R1Mbearing Wag/Rij rats. DPI was started immediately after i.v. injection of 18.5 MBq [ 123 I]-2-Iodo-D-Phe or [ 123 I]-2-Iodo-L-Phe and continued up to 40 min for all rats (day 0). Twenty-four hours andforty-eight hours p.i. a static image was acquired. This set-up was repeated in the same rats for [ 123 I]-2-I-L-Tyr and [ 123 I]-2-I-D-Tyr (day 7-11), after which [ 123 I]-IMT imaging was performed (day 14-16).All rats were also injected with 37 MBq [ 99m Tc]-Pyrophosphate ([ 99m Tc]-PyP) ([ 99m Tc] from agenerator, GE) for blood pool estimation (day 16). The imaging experiment with [ 123 I]-2-Iodo-D-Pheand [ 123 I]-2-Iodo-L-Phe was repeated twice, each time with 4 rats.For dissection experiments, 3.7MBq [ 125 I]-2-Iodo-L-Phe or [ 125 I]-2-Iodo-D-Phe was injected i.v.. Foreach tracer, 5 animals were euthanised 30 min p.i. and 5 animals at 24 hours p.i.3. RESULTSDynamic planar imagingThe uptake of all tracers reached its maximum between 10 and 30 minutes p.i.. The DUR value of thetumour at 20 minutes p. i. of [ 123 I]-2-Iodo-L-Phe, [ 123 I]-2-Iodo-D-Phe, [ 123 I]-2-Iodo-L-Tyr, [ 123 I]-2-Iodo-D-Tyr and [ 123 I]-IMT is respectively 2.06 ± 0.43; 2.26 ± 0.81; 2.46 ± 0.77; 1.71 ± 0.61, 2.43 ±0.56 and 2.43 ± 0.56 (data: mean ± SD (n=4)). The [ 123 I]-2-Iodo-D-Tyr uptake is lower and yet thecontrast (tumour/tissue ratio) versus heart (blood), muscle and brain is comparable with the values of[ 123 I]-2-Iodo-L-Tyr and the phenylalanine analogues. The urinary excretion of [ 123 I] activity of [ 123 I]-2-Iodo-D-Tyr is much faster than of it’s L-isomer and represents 1/8 of the total body activity after 30minutes. This phenomenon was also observed, but to a much lesser extent, for [ 123 I]-2-Iodo-D-Phe.The tumour/heart ratio of all tracers was at least 5 times higher than the value obtained for [ 99m Tc]-Pyrophosphate blood pool activity proving a specific tumour uptake.

Long term imagingTwenty-four hours p.i. the tumour is clearly visible for [ 123 I]-2-I-L-Phe and [ 123 I]-2-I-D-Phe, while for[ 123 I]-2-I-L-Tyr and [ 123 I]-2-I-D-Tyr tumour uptake is lower. In the case of [ 123 I]-IMT, the tumour canhardly be distinguished from the background. The tyrosine analogues show a clear thyroid uptake offree radioiodide which is not the case for the 2-I-Phe analogues. This indicates that the phenylalanineanalogues are less susceptible to deiodination in these rats. The DUR value of the tumour at 24 hoursp.i. of [ 123 I]-2-Iodo-L-Phe, [ 123 I]-2-Iodo-D-Phe, [ 123 I]-2-Iodo-L-Tyr and [ 123 I]-2-Iodo-D-Tyr isrespectively 2.25 ± 0.85, 2.16 ± 1.12, 1.66 ± 0.50 and 1.54 ± 0.56 (data: mean ± SD (n=4)), while onthe images of [ 123 I]-IMT the tumour is not clear enough to allow an accurate calculation of the DURvalue. At 48 hours p.i., the DUR value of [ 123 I]-2-Iodo-L-Phe and [ 123 I]-2-Iodo-D-Phe is 2.89 ± 0.83and 2.69 ± 0.87, while those of the other tracers can no longer be determined.RetentionThe retention is expressed as the ratio of counts per pixel in the organ at the appropriate time to thecounts per pixel in the organ at 30 minutes p.i. (decay corrected) and is represented for the tumour andthe total body in table 1 (n = 12 for [ 123 I]-2-I-L-Phe and [ 123 I]-2-I-D-Phe; n = 4 for [ 123 I]-2-I-L-Tyr and[ 123 I]-2-I-D-Tyr). The retention at 24 hours of the tracer in the total body is around 20% for [ 123 I]-IMT,but the majority of that activity is concentrated in the thyroid so the images of [ 123 I]-IMT hardly yieldrecognizable organs or tumours anymore. The data clearly show that the phenylalanine analogues havea much higher retention, both in tumour and total body, when compared to the tyrosine analogues.Dissection data confirmed the imaging data of 2-I-L-Phe and 2-I-D-Phe.% RetentionTumour[ 123 I]-2-Iodo-L-Phe[ 123 I]-2-Iodo-D-Phe[ 123 I]-2-Iodo-L-Tyr[ 123 I]-2-Iodo-D-Tyr24h 91 ± 10 92 ± 10 10 ± 3 6 ± 348h 81 ± 14 79 ± 11 ND NDTotal Body24h 84 ± 10 81 ± 9 15 ± 4 8 ± 248h 68 ± 11 63 ± 9 ND NDTable 1: Relative retention of radioactive tracer in the tumour, heart and total body at 24h and 48h p.i.when compared to 30 min p.i.. [ 123 I]-IMT showed no measurable retention in the tumour. (ND: Notdetectable)4. DiscussionAll tracers ([ 123 I]-2-I-L-phenylalanine, [ 123 I]-2-I-D-phenylalanine, [ 123 I]-2-I-L-tyrosine, [ 123 I]-2-I-Dtyrosineand [ 123 I]-α-Methyl-L-tyrosine) show a fast tumour uptake and a favourable biodistribution atshort times. We showed that D isomers are equivalent to L isomers with regard to tumour contrast andbiodistribution, with a significantly faster clearance to the bladder in the first 30 minutes by the D-isomers.At longer times however, there is a different pattern for the retention of radioactivity in the tumour andthe body. For the iodinated tyrosine analogues the retention in the tumour was only about 10%, while[ 123 I]-α-Methyl-Tyrosine did not yield recognizable organs anymore, as the tracer was almost fullycleared 24 hours p.i.. A considerable deiodination could be noticed for the tyrosine analoguesrepresented by the significant amount of radioactivity in the thyroid. No significant difference wasnoticed between the L and D isomers of both [ 123 I]-2-I-Phe and [ 123 I]-2-I-Tyr.Only the iodinated phenylalanine analogues showed a good tumour visualization after longer times.After 24 and 48 hours there was a similarly high retention of activity for both [ 123 I]-2-Iodo-D-Phe and[ 123 I]-2-Iodo-L-Phe in the tumour and in the body. Since the uptake in the tumour via the LATtransport system is reversible and dependent on the amount of radioactivity in the blood, the high totalbody and blood retention ensures that also the tumour retention is significantly higher for the Pheanaloguescompared to the Tyr-analogues. This high blood retention can be due to a higher re-uptake

in the kidney coupled to a high tracer stability as no thyroid uptake could be observed for theradioiodinated phenylalanine analogues. It is important to notice that the retention in the tumour ishigher compared to the retention in the total body, indicating a relative enrichment of activity in thetumour. Kersemans V. et al. already described the potential of long term imaging in a mouse modelusing [ 123 I]-2-I-L-Phenylalanine and [ 123 I]-2-I-D-Phenylalanine: at longer times the DUR values of thetumour were much higher compared to the DUR values at short times. The clearance from both thetumour and body in the athymic nude mice was much faster compared to our rat model. Samnick et al.also noticed tumour retention in human, they demonstrated that [ 123 I]-4-I-L-phenylalanine shows anintense uptake in gliomas in patients up to 24 hours p.i..The high retention of these iodinated phenylalanine analogues allows not only late imaging (due to theincreasing tumour/background ratio), but it also points to the possibility of radionuclide therapy.AcknowledgementsThe first author is currently in training as a PhD student. Financial support was given throughGOA/VUB (project number GOA28) and the FWO Vlaanderen (project number FWO G.0299.04N).5. References[1] H. Ohkame, H. Masuda, Y. Ishii, Y. Kanai, Expression of L-type amino acid transporter 1(LAT1) and 4F2 heavy chain in liver tumour lesions of rat models, Journal of Surgicaloncology, 2001, Vol. 78, p. 265-272[2] Lahoutte T., Caveliers V., Camargo SM, Franca R., Ramadan T., Veljkovic E., Mertens J.,Bossuyt A., Verrey F., SPECT and PET amino acid tracer influx via system L (h4F2hchLAT1)and its transstimulation, J Nucl Med. 2004, Vol 45 N° 9, p.1591-1596[3] Yanagida O., Kanai Y.: Human L-Type amino acid transport system 1 (LAT1) :characterisation of function and expression in tumour cell lines, Biochim Biophys Acta, Vol.1514, 2001, p. 291-302[4] Segawa H, Fukasawa Y, Miyamoto K, Takeda E, Endou H, Kanai Y.: Identification andfunctional characterisation of a Na+-independent neutral L-amino acid transporter with broadsubstrate selectivity, The Journal of Biological Chemistry, Vol. 274, 1999, p. 19745-19751[5] Kanai Y, Segawa H, Miyamoto K, Uchino H, Takeda E, Endou H. “Expression cloning andcharacterisation of a transporters for large neutral amino acids activated by heavy chain 4F2antigen (CD98)” J. Biol. Chem. (1998) 273,23629-23632[6] Kupczyk-Subotkowska L, Tamura K, Pal D, Sakaeda T, Siahaan TJ, Stella VJ, Borchardt RT.et al. “ Derivatives of melphalan designed to enhance drug accumulation in cancer cells” JDrug Target 1997; 4(6):359-70[7] Rajan DP, Kekuda R, Huang W, Devoe LD, Leibach FH, Prasad PD, Ganapathy V. , Cloningand functional characterisation of a Na(+)-independent, broad-specific neutral amino acidtransporter from mammalian intestine, Biochim BioPhys Acta 2000, Jan 15, 1463(1):6-14[8] Kersemans V, Cornelissen B, Mertens J., et al. ; In vivo evaluation and dosimetry of 123I-2-iodo-D-phenylalanine, a new potential tumour-specific tracer for SPECT, in an R1Mrhabdomyosarcoma athymic mouse model.; J Nucl Med. 2005 Dec;46(12):2104-11.[9] Hideo Tsukada, Kengo Sato, Dai Fukumoto, Shingo Nishiyama, Norihiro Harada, andTakeharu Kakiuchi: Evaluation of D-Isomers of O- 11 C-Methyl Tyrosine and O- 18 F-Fluoromethyl Tyrosine as Tumour-Imaging Agents in Tumour-Bearing Mice: Comparisonwith L- and D- 11 C-Methionine J. Nucl. Med. 2006 47: 679-688[10] T. Lahoutte, J. Mertens, V. Caveliers, PR Franken, H. Everaert, A. Bossuyt, Comparativebiodistribution of iodinated amino acids in rats: selection of the optimal analog for oncologicimaging outside the brain, Nucl Med. 2003, Vol 44 N° 9, 1489-1494[11] Mertens, J., Kersemans V., Bauwens M., et al: Synthesis, radiosynthesis and in vitrocharacterisation of [125I]-2-iodo-L-phenylalanine in a R1M rhabdomyosarcoma cell model asa new potential tumour tracer for SPECT, Nucl Med Biol, Vol 31, 2004, p. 739-746[12] Mertens J., Ytterbeke K., Lahoutte T., et al. “In Vitro and in Vivo Evaluation of L-2-radioiodo-tyrosine as a new potential tumour tracer for SPECT” J. Labelled Cmp Radiopharm2001.

[13] Kersemans V, Cornelissen B, Kersemans K, Bauwens M, Achten E, Dierckx RA, Mertens J,Slegers G.: In vivo characterization of 123/125I-2-iodo-L-phenylalanine in an R1Mrhabdomyosarcoma athymic mouse model as a potential tumour tracer for SPECT., J NuclMed. 2005 Mar;46(3):532-9.[14] Gühlke S, Biersack HJ. Simple preparation of L-3-iodo- -methyl tyrosine suitable foruse in kit preparations. Appl Radiat Isot. 1995;46:177–179

OPTIMISATION OF PRODUCTION OF MEDICALLYRADIOISOTOPES THROUGH CROSS SECTIONDETERMINATION.A. HERMANNE 1 , F TARKANYI 2 , S. TAKACS 2 , R. ADAM-REBELES 11 Cyclotron department, Vrije Universiteit Brussel, Brussels, Belgium.2 Atomki, Hungarian Academy of Science, Debrecen, Hungary.ABSTRACTDiagnostic and therapeutic applications of radioisotopes are gaining in importance. A large fractionof the used radioisotopes are produced by charged particle reactions.Improving, optimising or innovating production pathways for efficient and reliable supply is anongoing challenge for physicists around an accelerator.The standard approach for gathering information allowing prediction of production yields andcontamination rates is measurement of excitation functions with stacked foil irradiations andaccurate gamma spectrometry.We discuss three types of applications for production optimisation:- comparison of pathways: which of proton or deuteron reactions is preferable for production of103 Pd from Rh targets?- possibility to produce emerging radioisotopes: is the reaction 141 Pr(d,3n) 140 Nd an efficient choice?- alternative production approaches: can fission 99 Mo be replaced by accelerator routes like100 Mo(d,x) 99 Mo or 100 Mo(p,pn) 99 Mo?A comparison of experimental results with the reaction model based code ALICE-IPPE and thicktarget yields derived from excitation curves are discussed.1. IntroductionNeutron deficient radioisotopes are increasingly used in diagnostic and therapeutic nuclear medicine, inindustry and in research projects. Their production needs charged particle irradiations (CP) at acceleratorsites. The CP radioisotopes are mostly produced in proton induced reactions exploiting the rather highcross sections of the processes involved and low stopping power of protons. Dedicated high beam power,cost effective and simple to operate ( – H, p) cyclotrons were commercially developed. Neverthelessseveral of these isotopes can be produced by other pathways: either other charged particle reactions(deuteron or alpha induced) or reactor production starting from other target material are possible. Thechoice of pathway will influence the overall batch yield and the radionuclide purity/specific activity of thedesired end product. One of the ongoing tasks and challenges for a physicist or radio chemist workingaround an accelerator used for production of medically relevant radioisotopes is to guarantee an efficientand reliable supply to the hospital by improving, optimizing or innovating production pathways. Thestandard approach to allow prediction and optimization of batch yields and to choose routes that assurecontrolled contamination rates is accurate knowledge of the excitation curves of all (or at least most) ofthe nuclear reactions involved in the production process for given target/ beam combinations. At presentcomputer codes for calculation energy dependence of reaction cross sections, do not provide enoughaccuracy in shape and absolute values to allow production optimization. The preferred way isexperimental determination of excitation functions by irradiation of stacked targets, analysis of activationproducts, mostly by gamma spectrometry, and calculation of the cross section at each energy point byusing the classic activation formula. A comparison with results of model codes can allow betterunderstanding of the relative contribution of individual reactions, in case of use of targets with multipleisotopes (targets of natural composition), or indicate improvements to be made to the theoretical model.For optimization of production in itself, knowledge of excitation functions can have three main purposes:choose the most efficient charged particle to use on a given target material for a given end product,

investigate the possibility to use irradiation techniques to obtain new or poorly studied radioisotopes,explore the possibility to replace reactor based production by charged particle reactions.An example of these three applications will be discussed in detail, based on our experimental work in thepast years, and a discussion of the predictive value of the ALICE-IPPE code in comparison with theseexcitation curves will be given.2. Experimental methodologyNearly all results discussed here were published earlier by our group [1-5] and were obtained inirradiations using the external proton or deuteron beams of the cyclotrons of Sendai (Japan), Atomki-Debrecen (Hungary), UCL-Louvain la Neuve or VUB-Brussels (both in Belgium). For all cross sectiondeterminations the established stacked foil technique of stacked foil, with monitoring of the beam energyand current with standard reactions for which recommended excitation functions exist [6, 7], was used.2.1 Targets, Irradiation, measurementsStacked foil targets consist of an assembly of thin (10 -100mg/cm 2 ) either natural metal foils(commercially purchased) or supporting foils on which isotopically enriched material or oxides of theelement to be studies are deposited electrochemically or by other techniques. The electronic stopping ofthe mono-energetic bombarding particles in the foils results in a gradual decrease of effective particleenergy, yielding information at multiple energies for each irradiation. Insertion of monitor foils givingrise to activation products with well documented and evaluated standard cross sections allows checkingand/or correction of the energy and intensity of the incoming beam, after comparison of the remeasuredcross section of the monitor reaction with the recommended cross sections [6, 7].The irradiations took place with incident energies between 15 and 80 MeV for protons and 12 and 40MeV for deuterons, beam intensities were kept constant over the irradiation time (between 1 and 2 h) at alevel of 50 to 250 nA.The target holders acted as Faraday cups and were equipped with collimators and negatively polarisedguard rings to suppress secondary electron emitted from the target material.The irradiated stacks were allowed to cool for a period of 1 to 2 h to after EOB (End of Bombardment)allow decay of the high activities (and dose rates to experimentators) of short lived radioisotopes. Thisprocedure implies that excitation functions of isotopes with half-life shorter than 30 minutes are notdetermined as decay also occurs during the time needed to measure the 10-15 foils of a stack.For isotope identification high purity Ge detectors, coupled with acquisition/analysis software, were used.Using standard calibrated sources the detector efficiencies were accurately determined over the domain ofenergies concerned. The measurements were started shortly after EOB and were repeated for severalweeks. The activity was recorded at different detector-sample distances (5-70 cm) in order to obtain lowdead times and to eliminate other possible interferences. The contribution of background signals at thedifferent energy peaks was corrected for. No chemical separations took place.2.2 Data processingSo called elemental cross-sections are determined from the measured activities at EOB, the number ofincoming particles (derived from integrated charge) and the number of target atoms present consideringthe target material as mono-isotopic. The NUDAT data base was used for decay characteristics of theidentified radioisotopes [8]The incident beam energy, evaluated from the accelerator settings, as well as the initial current beam,measured with a Faraday cup, were adapted taking into account the excitation functions of the monitorreactions ( nat Ti(p,xn) 48 V, nat Ti(d,xn) 48 V, Al(p,x) 22,24 Na) measured over the whole energy domain.For energy degradation in the stack foils the codes like SRIM, based on polynomial approximations forthe stopping power of the elements present, was used [9].This technique allows determining the energy in every foil with high precision. The incident energy forthe different beams was estimated with an uncertainty of ±0.1 MeV.2.3 UncertaintyUncertainty on the median energy in each foil is increasing due to the cumulative effect of energy spreadand variations in foil thickness and reaches a maximum of 1.5 to 2.5 MeV for the last foil of long stacks.

The uncertainties regarding cross section values were estimated according to the prescriptions of [10] andare obtained by quadratic summation of the individual contributions.The following individual uncertainties are taken into account: absolute abundance of the used gamma rays(1-5 %), determination of the peak areas including statistical errors (1-5 %), the number of target nucleiincluding non-uniformity (5 %) and incident particle intensity (5 %). The total uncertainty is evaluated at8-15 %. The strongly non-linear effect of the possible uncertainty of the half-lives for samples measuredshortly after EOB (small T 1/2 ) was not taken into account.3. ComputercodesThe comparison of experimental results with a-priori model calculations without optimization of theparameters for individual reactions allows increasing the confidence in the predictive value of thesenumerical tools. In this report we show for some cases the cross sections of the relevant (p,x) or (d,x) onall stable isotopes of the target material calculated in the whole energy domain using the ALICE-IPPEcode. This code is a version of ALICE-91 originally proposed by Blann [11] and last modified by theObninsk group [12]. It is based on the hybrid, geometry-dependent hybrid or (HMS) pre-equilibriummodels and the Weisskopf-Ewing evaporation formalism. The level density formalism includes bothcollective and non-collective effects, and excitation energy-dependent shell effects.The individual results for the reaction products of interest in a study are weighted and summed accordingto the abundance of target isotopes. Here we present only two calculations on mono-isotopic targets infigures 2 and 5 and are discussed together with the experimental results in the appropriate sections.4. Excitation functions4.1 Production of 103 Pd from Rh targetsThis radioisotope (0.022% γ-line at 357keV and strong low energy X-lines around 20keV, 64%), used inseeds for permanent interstitial brachytherapy, is presently produced in dedicated high power acceleratorsusing 103 Rh (100%) targets and 18 MeV protons. Older data, needing an update, existed for the (p,n)reaction and we decided to also measure for the first time the cross sections for the deuteron inducedreaction. Thin, pure Rh foils (26µm for proton irradiation, 12.3 and 6.8 µm thick for deuterons) wereirradiated together with Ti monitors in a total of 9 stacks.Analysis of activity showed a discrepancy between the results obtained from γ-lines and from X-rays evenif self-attenuation of the X-rays is corrected for. The results are published in [1] and [2] and showed crosssections for the deuteron reaction that are up to 50% higher than for the proton induced reaction (see Fig.1 and 2). The thick target yields derived from the fitted experimental excitation function (based on the X-ray results) agree well with those measured earlier and are nearly double at 20 MeV incident particleenergy compared to those found for a proton irradiation (Fig. 3).Fig. 2 shows that the results of the theoretical calculations agree well with the experiments, which iscertainly not always true for deuteron induced reactions, especially when stripping is involved.4.2. Production of 140 Nd from 141 PrThe radioisotope 140 Nd decays by pure electron capture and without emission of measurable γ-rays to theβ + emitter 140 Pr. The 140 Nd/ 140 Pr pair is gaining interesting for possible combination of therapeutic andimaging properties in a single radiolabelled compound. Production by the 141 Pr(p,2n) 140 Nd route wasstudied earlier by Steyn et al. [13] In order to evaluate the possible advantage of the 141 Pr(d,3n) reactionseveral stacks of Pr 6 O 11 targets, deposited on Al or Ni backings (Julich) were irradiated in 20 and 40MeVdeuteron beams (see also [3]). The monitoring relied on the standard nat Ti(d,x) 48 V reaction. Assessment ofinduced 140 Nd activity was done through measurement of the annihilation radiation of the daughter 140 Pr.This approach asks, in the experimental condition used, for corrections for 511 keV signal from 141 Nd,48 V, 56 Co. Excitation functions for 139m Nd, 141 Nd, 140 Nd, 142 Pr, 139 Ce (high cross sections, Fig. 4) and 135 Ce,137m Ce, 140 La were determined.. Presence of 139 Pr was detected but too low counting statistics prohibitsreliable cross section calculation. No earlier data are available for these reactions.140 Nd batch production could be interesting with “dedicated” 30-40 MeV deuteron machines,radionuclidic purity has to be achieved by decay of shorter lived co-produced 138,139m,141g Nd. The thicktarget yield derived from the excitation function is 250-300MBq/µAh (comparable to the p-induced route:overview in Steyn et al. [13]: optimal range 30-15 MeV, practical TTY 275 MBq/µAh).

Cross section (mb)1000900This work X-linesThis work g-lines800Harper et al. [16]700Blaser et al. [21]600Johnson et al. [19]500Albert [20]40030020010000 5 10 15 20 25Particle energy (MeV)Cross section (mb)103 Rh(d,2n) 103 Pd1400FitAlice IPPE1200100080060040020000 5 10 15 20 25Particle energy (MeV)Fig 1. Excitation function or 103 Rh(p,n) 103 Pd(figure taken from [1]).Fig 2. Excitation function of 103 Rh(d,2n) 103 Pd.(figure taken from [2]).Thick Target Yield (MBq/µAh)30252015105Calculated thick target yieldDmitriev deuterons [4]Mukkhamedov deuterons [5]Thick target yield protons [3]Cross section (mb)1600140012001000800600400200141Nd140Nd139mNd142Pr139Ce02 6 10 14 18 22Particle energy (MeV)Fig. 3 Thick target yield for 103 Pd production:comparison of proton and deuteron induced reaction(figure taken from [2]).00 5 10 15 20 25 30 35 40 45Particle energy(MeV)Fig. 4. Excitation curves for 141,140,139m Nd, 142 Pr and139 Ce obtained in deuteron irradiation of 141 Pr4.3. Production of 99 Mo from 100 MoA vast majority of nuclear medicine investigations is presently relying on the 99 Mo- 99m Tc generator where99 Mo is a fission product recovered from research reactors. In the 90’s progressive closure of thesereactors was foreseen and possible accelerator production was studied. In [4, 5] we presented our resultsfor proton and deuteron induced reactions on 100 Mo resulting in 99 Mo (figures 5, 6).The advantages of the (d,p2n) over (p,pn) reaction can clearly be seen on figs. 5 and 6 as the crosssections for d induced are more than double compared to p induced but the maximum occurs at higherenergy. As both excitation functions remain high over a broad energy range higher energy deuteronmachines are needed to take advantage of the increased TTY (fig. 7). An upper limit is imposed bypossible contaminants production, especially 96 Tc reached by (d,6n) with Q = -35.6 MeV. In optimalconditions TTY could be 16 MBq/µAh resulting in daily batches (20 hours irradiation) at a dedicatedhigh power machine (1 mA beam current) of 3.2 10 11 Bq to be compared with the weekly needs of 8 10 12Bq in Belgium. Direct distribution of 99m Tc obtained by (p,2n) and (d,3n) on 100 Mo or by 98 Mo(d,n) couldbe other alternative. This would not require recycling of the expensive enriched target material but needsrethinking of the distribution pathways and introducing new radiopharmaceutical registration files.Fig. 5 shows that, although the shape of the excitation function is well modeled by ALICE-IPPEcalculations (data stored in MENDL2 database), the absolute values are underestimated by a factor 3.

Cross section (mb)350300250200150100100 Mo(p,x) 99 MoAlmeida (1977) [6]Levkovskii (1991) [7]Lagunas-S (1991) [8]Zhao Wenrong (1998) [11]Scholten (1999) [10]This workThis work fitMENDEL2PCross section (mb)500400300200100 Mo(d,x) 99 Mo50100Sonck, 199805 15 25 35 45Proton energy (MeV)00 10 20 30 40 50 60Particle energy (MeV)Fig. 5 Excitation function for 100 Mo(p,x) 99 Mo (figuretaken from [4]5. ConclusionsMeasurement of excitation functions allows toprove that for several reactions leading toproduction of medically relevant radionuclides,deuteron induced channels have higher thicktarget yields for the same incident energy thanproton reactions. This should be an incentive fordeveloping dedicated high power commercial 30MeV deuteron accelerators. The planned mediumenergy, high intensity deuteron acceleratordedicated to neutron production (IFMIF) couldhence be very effective in medical radioisotopeproduction.The agreement with theoretical calculationsshows often large discrepancies especially forstripping reactions. Upgrades and improvementsare being developed.Fig. 6 Excitation function for 100 Mo(d,x) 99 Mo (figuretaken from [5]Physical yield (GBq/C)706050403020100Beaver (1971) [5]Lagunas-S (1991) [8]Sholten (1999) [10]This workdeuteron induced, Sonck 1998100Mo(p or d,x) 99 Mo0 10 20 30 40 50Particle energy (MeV)References[1] A. Hermanne, M; Sonck, A. Fenivesy, L. Daraban. NIMB, 170, (2000), 281[2] A. Hermanne, M. Sonck, S. Takács, F. Tárkányi, Y. Shubin. NIMB, 187, (2002), 3[3] A. Hermanne, Abstract book Aachen, 2007[4] S. Takacs, Z. Szucs, F. Tarkanyi, A. Hermanne, M. Sonck. Radioanal Nucl Chem, 257, (2003) 1957.[5] M. Sonck, S. Takacs, F. Szelecsenyi, A. Hermanne, F. Tarkanyi. AIP Conf. Proc, editors Duggan andMorgan, 475, (1999), 987[6] F. Tarkanyi, K. Gul, A. Hermanne, M.G. Mustafa, M. Nortier, P. Oblozinsky, S.M. Qaim, B.Scholten, Yu.N. Shubin, S. Takács, Zhuang Youxiang; (2001). Charged-Particle Cross-SectionDatabase for Medical Radioisotope Production. Beam Monitor Reactions (Chapter 4). IAEA-TECDOC-1211, Int. Atomic Energy Agency, p. 49 -152, URL: http:/www-nds.iaea.org/medical.[7] S.Takacs, F. Szelecsenyi, F. Tarkanyi, M. Sonck, A. Hermanne,Yu. Shubin, A. Dityuk, M. Mustafa,Zhuang Youxiang, NIMB, 174, (2001), 235[8] National Nuclear Data Center, USA, URL http://www.nndc.bnl.gov[9] J.F. Ziegler, M.D. Ziegler, J.P. Biersack, SRIM2006, URL http://srim.org[10] Guide to expression of Uncertainty in Measurements, ISO Geneva, ISBN 92-10188-9, 1993[11] M. Blann, LLNL Report, UCRL-JC-109052, 1991[12] A.I. Dityuk, A.Yu. Konobeyev, V.P. Lunev, Yu.N. Shubin, New Version of the Advanced ComputerCode ALICE-IPPE, IAEA, Vienna, Report INDC (CCP)-410, 1998[13] G. Steyn, C. Vermeulen, F. Nortier, F. Szelecsenyi, Z. Kovacs, S. Qaim. NIMB, 252, (2006), 149

OPTIMIZING INJECTED DOSE IN EQUILIBRIUM GATEDRADIONUCLIDE VENTRICULOGRAPHY BY USING ATHREE CLUSTER MODEL AND LINEAR DISCRIMINANTANALYSISM.P. DIAZCenter forStudies on Electronics & Info Tech.,Central University of Las VillasCamajuaní Road km5 ½ ,Santa Clara 54830 – CubaO.D. RIZOInstitute for Sciences and advanced technologiest, Nuclear Physics DepartmentQuinta de los Molinos, Ave de los presidentes y Luaces, C. Havana 10400 –CubaJ.Q. GARCIANacional Center for Nuclear Safety28 Street, Miramar10500, C.Habana- CubaABSTRACTA mathematical model of three clusters is used to analyse differences in Equilibrium GatedRadionuclide Ventriculography image quality. The objective dependent variable, used tocharacterize image quality was Image Contrast. It was obtained from the Count ratiosbetween Heart and Background in some Regions of Interest. Noise and Spatial Resolutionwere also analysed as part of Image quality evaluation. The aim of the present research wasthe determination of the minimum 99m Tc-red cells activity which is able to guaranty enoughdiagnostic image quality in a traditional gamma camera. The method was applied in 30patient’s studies, injected with one of the following activities: 300, 440, 620, 800 or 1020MBq, using the ¨in vivo¨ labelled method of red cells previously stimulated withpyrophosphate. The statistical weight of each image quality determinant variable wasobtained using linear discriminant analysis. The included variables were: the radionuclidicactivity, the labelling yield, the weight of the patient, the acquisition time and the patientpathology. The cluster with optimum image quality was selected. The study showed that800 MBq is a optimised activity for the technology used and Cuban patient characteristics.1. IntroductionPatient exposure to ionizing radiations in Nuclear Medicine studies can not be indefinitely reducedwithout affecting image quality. The above problem is the cause of the necessity to establish acompromise relationship between image quality for diagnosis and radiation protection for the patient.It means that the activity should not exceed the minimum required to provide the indispensable clinicinformation. [1].IAEA recommends some reference values for each Nuclear Medicine study [2]. Nevertheless, IAEAsubjects to adjust these values to the practical reality of each country (equipment,radiopharmaceuticals, patient characteristics).Gated radionuclide ventriculography [3] presents many advantages for the diagnostic of severalcardiac diseases [4]. Therefore, there is not good consensus among doctors and medical physicistsrespect to the best activity value for this purpose [5-12].A model of two clusters combined with linear discriminant analysis of image quality has beenpreviously published as optimisation method in Nuclear Medicine [13]. The present work proposes theuse of a three cluster model looking for a best precise analysis of image quality as well as best caseclassification inside clusters, in order to justify the adequate radionuclidic dose selection according tothe radiopharmaceutical and equipment used and patient characteristics.2. Material and Methods

2.1 Experimental ProcedureAn injection of 10 mg of pyrophosphate was administered to 30 patients (16 women and 14 men, agebetween 28 and 80 years, mean 55 ± 3 years). Twenty minutes alter they received a dose of Tc 99m(divided in 6 groups, Group A: 303±6 MBq, Group B: 444±6 MBq, Group C: 617±7, Group D:803±10 and Group E: 1020±5 MBq). The labelling of patient red cells was ¨in vivo¨ [14]. The mainlabelling yield was verified by laboratory method [15]. Ten minutes after each patient is placed undergamma camera detector in LAO position, (45° detector inclination). Radioactivity is synchronised tothe R wave of electrocardiography signal obtained in CM5 derivation. Sixteen images were acquiredfor each cardiac cycle. Cycles with more than 10 % of difference with the mean cardiac frequencywere rejected. A Toshiba digital gamma camera (single head, model 500A, with RDC-44A parallelholes general purpose collimator was used. The acquisition matrix was 64 x 64 pixels and the energywindow was cantered in 140 keV ± 10 %.2.2 VariablesRegion of Interests (ROI) 5 x 5 pixels were traced over left ventricle, liver, lung, spleen and abackground zone under the left ventricle and upon the liver in diastolic image. (Fig. 1). ROIs wereprocessed by the method of the second derivative [16]. Background was automatically corrected [16].Fig. 1. Diastolic imageThe variables used to characterize image quality in terms of Image Contrast were Count ratiosbetween Useful signal in Heart respect to Organ or Background activity. They were Heart/Lung H/L,Heart/Background H/B, Heart/Spleen H/S and Heart/Liver H/L.Other (Xi) variables were also analysed as: cardiac frequency, left ejection fraction, Weight of thepatient (W), height, age, ejection time, accepted and rejected cycles, labelling efficiency (EM) and theacquisition time, looking for if they had influence over image quality. The subjective opinion aboutimage quality of an observer, blinded to the activity used, was also taken into account.2.3 Mathematical ProcedureWe follow the k means cluster model (order 3) and linear discriminant analysis [17]. If there aresignificant differences among variables belonging to the constructed clusters, then, we can affirm thateach cluster owns different image quality according to the Xi variables measured [18,19].As optimisation criterion was taken the lowest activity which permits the inclusion of cases in thecluster with the best image quality [18, 19]. Others regression and correlation analysis were alsoapplied in order to corroborate the results obtained.3. Results and DiscussionImage quality is usually analysed in terms of physical parameters as: Image Contrast, Statistical Noiseand Spatial Resolution [20]. The same gamma camera with a fixed distance patient-detector and thesame acquisition matrix 64 x 64 pixels was used for all the cases included. For this reason we considerfixed the Spatial Resolution in our experimentImage Contrast was defined as: (Counts in Useful Signal – Background Counts) / Background Countsx 100 %, using ROIs while Signal/Noise ratios were taken as (Useful Signal ROI) ½ . Then, as bothvariables are strongly correlated we took Signal/Background ratios as quantitative criterion of imagequality to conform the clusters. This variable is dependent on the administered activity and the studystopping time (taken at 300 kcounts in our experiment). As specific indexes of image contrast we tookH/B, H/S, H/L and H/P.

The expert observer graded all the images as good image quality, without distinction among activitiesused. Nevertheless, mathematically and taken into account the Xi variable measured, 3 differentclusters were constructed after 3 iterations, instead of only one, as the observer describe. After lineardiscriminant analysis we obtained that Image Contrast was only determined in clusters by: EM,Weight, Systolic counts, Activity and Acquisition time, with 96.7 % of cases correctly classified intothe clusters. The selected variables were significant taking into account their λ Wilks´s values, higherthan 0.9. EM p=0.642, W p=0.903, Syst Counts p=0.555, A p=0.155 and Acq t p=0.195.Two linear image quality discriminant functions were constructed. The first one detects changesbetween Cluster 1 and Cluster 2 and the second one between clusters 2 and 3. The canonicalcorrelation values are r=0.454 and r=0.256 respectively.H/B = 0.16 EM – 0.41 W + 0.466 Syst Counts + 0.975 A – 0.042 Acq t (1)H/B = 0.416 EM – 0.183 W + 0.810 Syst Counts – 0.255 A + 0.056 Acq t (2)Both functions were significant. λ Wilks 1 = 0.742 p = 0.709 and λ Wilks 2 = 0.934 p = 0.504.The only variable dependent on patient disease introduced in the functions was the Systolic Counts.The rest were eliminated due to multicolineality (High correlation coefficients with the selectedvariables) [17]. Fig. 2 shows the clusters in relation with the two functions.32Canonical Discriminant FunctionsH/BC/F123Ungrouped CasesGroup CentroidFunction 210123-1-2-3-2-1012Function 1Fig. 2. Cases selection by clusters and functionsThe cluster with the best image quality was cluster 3 (include the cases with the best H/B ratios).Table 1 shows the correlation coefficients of the selected variables in relation to H/B. The importanceof these values is that each coefficient establishes the statistical weight of each variable to distinguishclusters with differentiated image quality [17-19].VariableCorrelationCoefficient withfunction 1CorrelationCoefficient withfunction 2A 0.698 0.628Acq t 0.658 0.550EM 0.330 0.300Syst counts 0.062 0.803W 0.145 0.186Tab. 1 Correlation coefficients with image quality discriminant functionsThe above table expresses clearly that in this type of study the activity is the most important variablefor image quality, following by the Acquisition Time and the Labelling Efficiency. These results showsimilarity with others discriminant studies of image quality in Nuclear Medicine [21-23].Nevertheless, although the weight of the patient reveals some influence over image quality, this

variable was not very important for image quality discrimination, which has different behaviourrespect to others Nuclear Medicine Studies [24].Fig. 3. provides information about the variables behaviour to determine H/B ratios. A high dispersionis observed around the linear behaviour for almost all the variables, except for the weight, whose rangeincluded in this experiment was very similar and typical of Cuban population. Systolic counts neitherhad significant variations for the sample analysed. This variable was less determinant for imagequality than the rest of the analysed variables.1200100080060040020000 2 4 6y = 118,03x + 359,36 C/FR 2 = 0,120560050040030020010000 2 4 6y = -45,136x + 436,19 C/FR 2 = 0,0925T adquis (seg)97969594939291908988870 2 4 6y = 0,7354x + 90,534C/FR 2 = 0,05611008060402000 2 4 6y = 1,303x + 64,731C/FR 2 = 0,0195Peso (Kg)4000030000200001000000 2 4 6y = -344,7x + 20976C/FR 2 = 0,0039Cont SistFig. 3. Variables behaviour in relation to Image qualityTable 2 shows the case distribution by clusters and the correspondence with the activity administered.The activity with the best results from the point of view of H/B, H/L, H/P and H/S was 803 MBq(most of the cases are included in cluster 3). This result subjects that activities higher than 800 MBqsaturates the detector for the technology used, diminishing its sensitivity and consequentlySignal/Background ratios.Cluster 1 Cluster 2 Cluster 3Cases 3, 4 and 6 (303 MBq),Cases 8,9,1 and 12 (444 MBq),Cases 15 and 17 (617 MBq),Cases 21 and 24 (803 MBq) andCases 26 and 29 (1020 MBq)Cases 1, 2 and 5 (303cMBq), Case 7 (444MBq)and Cases 13, 14, 16 and18 (617 MBq)Tab 2. Case distribution by clustersCases 19, 20, 22 and 23(803 MBq), Case 10(444MBq) and case 27(1020 MBq)Fig. 4 shows the Heart/Background ratio relative to Heart/Background maximum vs. the normalisedactivity respect to the activity that permits the highest H/B ratio. This result corroborates the results oftable 2, after the three cluster model application. The value of 800 MBq is obtained as the optimisedactivity for the acquisition technology and radiopharmaceutical used, and the patients characteristicssampled.Fig. 4. Curvas de respuesta relativa para 30 estudios de ventriculografía nuclear gatillada en reposo.

The activity obtained as optimum was coincident with previous results for this type of NuclearMedicine study in Cuba [13]. In the referenced study was used a model with lower clusterisationdimension. The method of two cluster combined with discriminant analysis to optimise radionuclidicactivity has been successfully also proved in other studies of Nuclear Medicine [24] and it has beenvalidated by comparison with the well-known ROC analysis [25]. The present work shows how theincrement in the cluster number increases the classification precision but did not change the optimisedactivity value selected. Special attention is required respect to the optimisation of the clusterisationdimension for futures works.4. ConclusionsThe use of a combination of three clusters and linear discriminant for mathematical image qualitydetermination is a useful tool to optimise activity in Gated Radionuclide Ventriculography. Themethod is sensible to small variations in Image Contrast, not detectable for expert observers. Themethod is more precise than the model of two clusters. The value of 800 MBq was reported as theminimum radionuclidic activity tested for obtaining good image quality with the technology andradiopharmaceutical used.5. References[1] International Commission on Radiological Protection. 1990 ¨Recommendations of the InternationalCommission on Radiological Protection¨. ICRP Publication 60 Ann. ICRP 21 (1-3), Oxford:Pergamon Press; 1991[2] Colección de Seguridad 115. ¨Norma Internacional Básica de Protección contra la RadiaciónIonizante y de Seguridad de las Fuentes de Radiación¨, OIEA, Viena, 1996.[3] R. Callahan R., J. Froelish ., J. Leppo., W. Strauss, ¨Modified method for “in vivo” labelling ofred blood cells with Tc-99m´. J. Nucl. Med. 23, 1982, pp 315-318.[4] K. Edmund, P. Thomas P. ¨Nuclear diagnostic Imaging. Practical Clinical Applications¨ en. McMillam Publishing Company, New York.: 1987, pp 340-359[5] ARSAC: ¨Notes for Guidelines on the Administration of Radioactive Substances to Persons forPurposes of Diagnosis, Treatment and Research¨ en ARSAC, London, 1993.[6] The Medical Exposure Directive (97/43). ´List of administered activities used in nuclear medicinepractice in the Member States¨. EURATOM, 1994[7] S. Mattsson , L. Jacobsson., E. Vestergren. ¨The Basic Principles in Assessment and Selection ofReference Doses: Considerations in Nuclear Medicine¨. Rad. Prot. Dosim. Vol 80,: pp 23-27; 1998[8] B. Millari, ¨ First Pass and Gated Equilibrium Methods¨ J. Nuc. Med. Vol 32: pp 1849-1853 ,1991.[9] K. Riden. ¨Intravenous lines vs. Vinipucture in Nuclear Medicine¨. J. Nucl. Med. Vol 11, pp 830-833, 1986.[10] C. Wayne, D. Manuel, ¨Four Radionuclide methods for left ventricular volume determination:comparison of a manual and an automated technique¨. J. Nucl. Med. Vol 33, pp: 763-770, 1992.[11] V. Green., I. Stephen., S. Jeffrey ¨A theoretical comparison of First Pass and Gated EquilibriumMethods in the Measurements of Systolic Left Ventricular Function´, J. Nucl. Med. Vol 32, pp1801-1807, 1991.[12] J.M. Castro Beiras. ´Cardiología Nuclear y otras técnicas no invasivas de imagen en Cardiología¨en MT Editores. Madrid, 2005[13] M. P. Díaz , O. D. Rizo, J. Q. García., F.P. Vicente, ¨Administered Activity Optimization inpatients studied by Gated Radionuclide Ventriculography using Tc-99m labeling red cells´, NucMed Commun vol 23 (4), pp 137-142, 2002.[14] D. Pavel, A. Zimmer, V. Patterson, ¨In vivo labelling of red blood cell with Tc-99m. A newapproach to blood pool visualization¨, J.Nuc. Med.vol 18, pp 305-308, 1977.[15] P. Alderson, R. Hamilton R, ´A comparative evaluation of technique for rapid and efficient “ïnvivo” labelling of red blood cells with Tc-99m pertechnetato¨, J. Nuc. Med .vol 18, pp 1008-1011, 1978[16] T. Gerbrands, C. Hook, J. Reiber , S. Lie, M. Simoons, ¨Automated left ventricular boundaryextraction from Tc-99m gated blood pool scintigraphy with fixed or moving regions of interest¨.

Proceedigs of The 2 nd International Conference on Visual Psichophysics and Medical imaging.IEEE vol 81, pp 155-159, 1976.[17] W. N. Venables, B. D. Ripley, ¨Statistics and Computing. Modern Applied Statistics with S-Plus¨. En Springer-Verlag New York, pp 311-318, 1994[18] M. P. Diaz, O. D. Rizo. ¨Fundamentos físicos de calidad de imagen en Medicina Nuclear.Métodos para su valoración´, ALASBIMN Journal vol 9(32), Nro AJ32-3, 2007.[19] M. P. Diaz, O.D. Rizo, N.Ferrer. ¨Métodos de optimización de la actividad a administrar alpaciente en estudios de Medicina Nuclear (Revisión del Tema)¨. Rev Fis. Med .vol , pp: 22-25;2006.[20] A. Ll. Evans. ¨Medical Physics Handbook¨. The evaluation of Medical Images. In ImageEvaluation by Signal Detection Theory. Adams Hilger Ltd, Techno House, pp: 80-113, UK,1981.[21] M. P. Díaz, O.D Rizo, J.Q. García J., R.D. Hernández, E. E. Aparicio, A.V.A.V. Marin.¨Administered Activity Optimization in Skeletal Scanning using MDP labeled Tc99m¨. AlasbimnJournal vol 4(16), 2002.[22] M. P. Díaz, O. D Rizo, R. D. Hernández., E.E. Aparicio. ¨Administered Activity Optimization inRenal Scintigraphy with DMSA - 99mTc¨ ALASBIMN Journal, vol 5 (20), 2003.[23] M. P. Díaz E. E. Aparicio, O. D. Rizo, R. R. Diaz, C. Hernández. ¨Administered Activityoptimization in 99m Tc-MAG3 Renography for Adults¨. J Nucl Med Technol vol 31(4), pp 206-209,2003.[24] E.Vestergren ¨PhD. Thesis¨. Administered radipharmaceutical activity and radiation dosimetryin paediatric nuclear medicine. Department of Radiation Physics. Gotemburgo University.Sweeden, 1998.[25] M. P. Díaz, O. D. Rizo, A. L.Diaz, E. E. Aparicio, R. R.Diaz ¨Activity Optimization Method inSPECT: A comparison with ROC analysis¨. Applied Physics & Enging. (JZUS-B), vol 7(12), pp947-956, 2006.

THERMAL FLUX ANALYSIS OF THE PFMA-1 DEVICES. MANNUCCI, D. MOSTACCI, F. ROCCHI, M. SUMININuclear Engineering Laboratory of Montecuccolino, University of Bolognavia dei Colli 16, 40136 Bologna (BO) – ItalyE. ANGELI, A. TARTARIDepartment of Physics, University of Ferraravia Saragat 1, 44100 Ferrara (FE) – Italy1. IntroductionABSTRACTThis work presents experimental results pertaining to the heating by plasma-wall interaction ofthe upper flange of the vacuum chamber of the PFMA-1 Plasma Focus at 18 kV and at 1 Torrof 4 He. The 2D temperature distributions over the flange have been obtained using an infraredcamera. From these distributions average temperature profiles are obtained and used toestimate the heat imparted to the flange by plasma-wall interaction. It is found that at least 15%of the input energy is converted into thermal energy of the flange. The heat flux has beenestimated too; it is found that its average value is comparable to that for ITER in the cases ofplasma disruptions. It is therefore proposed that a powerful PF can be used similarly to plasmaguns and e.m. railguns to test materials for fusion reactors. Other studies are under way tocorrelate the model presented to the experimental results obtained with an aluminium diskwhich resulted ablated due to the interaction with the plasma.One of the various issues that has to be overcome along the pathway to the reliable, industrial applicabilityof the Plasma Focus (PF) technology in the repetitive regime is the management of the thermal loads tothe components of a given device. Order of magnitude estimates predict that about 98% of the inputenergy imparted to a device in a shot is, at equilibrium, directly converted into thermal content of its parts.The largest fraction of the thermal loads is most probably generated and deposited in the vacuum chamberregion, where the plasma is formed and pinched. Experimental and/or theoretical data about themechanisms by which energy is deposited into the materials of PFs are very scanty, as well data about thecooling of these devices [1]. Small scale PFs, devices with input energy below about 1 kJ, have beenproduced with shot repetition frequencies up to 10 Hz and operated for some minutes without the need ofcooling apparatuses [2]. Many nuclear reactions that are triggered in the pinched plasma of PFs are knownto scale with a given power law of the input energy, most often the exponent being ≈ 2. This means that acareful optimization design has to be carried out in choosing between a low input energy device with highrepetition rates, or a high input energy device with rather low repetition frequencies. Since the nuclearyields scale with a more-than-linear power law and since it is quite difficult to raise the repetitionfrequency of powerful devices due to problems in the switching technology, it is often convenient toincrease, for output efficiency purposes, the input energy rather than the repetition rate. An increase by afactor four in the output can be obtained either by enhancing the input energy by a factor two, or byenhancing the repetition frequency by a factor four; this second option however implies an increase by afactor four in the heat loads, whereas in the first instance the heat loads are increased by a factor two only.The most important problem related to the thermal analysis of repetitive PFs is the distribution of the heatloads; a previous experimental study [1] indicated that about 10% of the input energy is imparted to theinner electrode in a positive polarity device; in this study the energy released to the upper flange of thevacuum chamber of PFMA-1 is evaluated and reported from experimental evidence.

2. PFMA-1 and experimental setupPFMA-1 [3,4] is a 150 kJ Mather-type PF designed for a repetition frequency up to 1 Hz and dedicated tothe neutron-free endogenous production [5] of 18 F, a short-lived radioisotope used in the preparation ofFDG drugs used in PET medical examinations. A photograph of the device is given in Fig. 1. Theoreticalestimates predict a yield of about 1 Ci of 18 F in two consecutive hours of repetitive operation at 1 Hz. Thedevice has a total equivalent inductance of about 40 nH, included the contribution of the cylindricalelectrodes; its input energy is stored in a 350 μF capacitor bank charged at 30 kV. The vacuum chamber isfilled with 16 O and 3 He at a total pressure of about 10 Torr.Peak total current is predicted by 2-D Snowplow calculations to be about 1.5 MA. Active cooling of allthe system is achieved with closed-circuit SF6 flow, open-circuit dry air, closed-circuit deionized waterand partially closed-circuit demineralized water refrigerated by a 150 kW evaporative cooling tower.The temperature distributions over the external surface of the upper flange, Fig. 2, of the vacuum chamberhave been obtained with a Flir Thermacam PM 675 infrared camera. The flange is entirely made of AISI304L stainless steel and has some ports that connect the vacuum chamber to various diagnostics as well asto the vacuum system. A special 3M black tape was used to cover entirely the area to be investigated, sothat no reflections in the visible spectrum due to the very flat surface could affect the measurement; aspecific cross-calibration for obtaining the emissivity coefficient of the black tape was done with the useof a contact thermocouple; error in the temperature measurements is estimated to be +/- 0.3 °C. Theinfrared images were taken with about 25-30 s delay after each shot, which is about 4-6 times thetemperature relaxation time in the thickness of the flange (estimated at first approximation as2 2s ρ cpπ k , s = 25 mm being the thickness, ρ the mass density, k the thermal conductivity and c p thespecific heat); this is the time the temperature gradient along the thickness of the flange takes to becomeso small that the measured external surface temperature T ~ () r can be practically considered coincidentwith the average temperature:s~ 1T () r ≡ ∫T( r,z)dz(1)s0T(r,z) being the real temperature distribution after a shot. The cooling of the flange can be neglectedbecause the experimental evidence showed that 25-30 s are not enough to modify substantially the radialprofiles.Fig 1. Photograph of PFMA-1.Fig 2. Drawing of the upper flange of thevacuum chamber.The radius of the flange is 17.5 cm. Its mass is estimated as being that of a full and perfectly circular plateflange without holes for ports, to compensate for the extra mass of the ports themselves; it results 18.9 kg.

The distance between the pinch point and the flange is about 14 cm. The experiments were carried out at18 kV charging voltage, which corresponds to an input energy of 54 kJ, and 1 Torr pressure of 4 He. Thetime to plasma-pinch is about 2 μs, and the whole half-period of the discharge lasts about 10 μs. Shotswere performed every 4-5 minutes and no external cooling apparatuses were connected.3. Experimental results and discussionFig. 3 shows a typical postprocessed image taken with the infrared camera; the point of maximum heatingis at the center of the flange. This is due to two mechanisms: plasma-wall interaction and ion impingingfrom the post-pinch MHD particle emission in the forward direction (this being the direction toward thegeometrical center of the flange). The contribution of the first mechanism is probably the most significantfor the energy deposition, as will be shown below. Fig. 4 is the plot of the isothermal lines from Fig. 3; itshows rather uniform gradients in the radial direction. Both figures are plotted against the matrix indicesby which spatial information is stored in the infrared camera.Fig 3. Infrared camera picture. Fig 4. Isothermal lines from Fig 3.~Fig. 5 shows in the upper row the temperature profiles () rT nalong the dashed line drawn in Fig. 3, afterFig 5. Upper: temperature profile sequence.Lower: average incremental temperatureprofile for a single shot.Fig 6. Internal side of flange, with circularaluminium plate before and after (upper leftcorner) 40 shots.

each shot of a series of 11 shots; it is clearly seen that a difference of about 5-6 °C exists between the~ ~center of the flange and its borders. In the lower row is plotted Δ T () r = Tn+ 1() r − Tn() r , the averagenprofile of temperature increase between two consecutive shots, as can be determined from the profiles ofthe upper row; ± 1σ error bars are also shown. It was also determined that a series of 20 shots is capable ofgenerating a temperature increase of about 18 °C above room temperature in the center of the flange. Theamount of energy Q deposited in the flange per shot can therefore be estimated asR= mcpQ ∫ Δ T () r 2πrdr(2)2πR0R being the radius of the flange, m its mass and c p the specific heat of stainless steel. It results Q ≈ 8 kJ.This corresponds roughly to 15% of the whole input energy stored in the capacitor bank; such a largeamount of energy can only be transmitted by the direct contact with the plasma. After about 200 shots ithad also been found that a large circular area of the inner surface of the flange resulted eroded or ablated.This area is rather well defined; Fig. 6 shows the ablated area delimited by a white circle. It was thereforedecided to investigate more deeply this phenomenon. An aluminium circular plate or disk (diameter 15.1cm, thickness 1.6 mm) was installed in the peripheral part of the flange and exposed to 40 shots. Afterexposure it was found to be heavily eroded or ablated in an area that roughly corresponded to that of thestainless steel (inset in the upper left corner of Fig. 6). Gaussian shapes in r and t are assumed for the heatflux:( r,t)2⎛ r ⎞−⎜⎟⎝ σ ⎠2⎛ t ⎞−⎜⎟⎝ α ⎠F = Ne e(3)where N is a normalization constant, σ is taken to be R 3 and α is taken to be 0.1 μs (0.6 μs is assumed tobe the duration of the flange heating phase by plasma interaction). The normalization constant can bedetermined by:Q =∞ R∫∫−∞ 0F( r t), 2π rdrdt(4)12It results N ≈ 4× 10 W/m 2 . The average heat flux is found to be:F∞ R1F1026απR= ∫∫−∞ 011( r,t) 2πrdrdt≈ 1.4×W/m 2 . This value is of the same order of magnitude as thatexpected in ITER in the case of plasma disruption [6]. Therefore it can be suggested that a powerful PFcan be used, likewise plasma-guns and electromagnetic railguns [7], to test fusion reactor components andmaterials. In Fig. 7 it is plotted the ablation threshold of aluminium as well as the profileFig 7. Q’’(r) over the flange and Q Th ’’ for aluminium.

∞∫−∞'() r F( r t)Q ' = , dt along the radial dimension of the flange. It is seen that the local heat flux resultshigher than the threshold in an area that corresponds roughly to that seen in the experiments (see Fig. 6).The ablation threshold has been calculated as Q'' = Ω ρkc τ ≈ 22.2 J/cm 2 , Ω being thevaporization specific enthalpy, ρ the mass density, k the thermal conductivity, c p the specific heat and τ thetime-duration of the heat flux (taken as 6α) [8].This formula is valid for the nanosecond or longer energy deposition-times, this regime being typical ofPFs. Moreover Vorobyev and coworkers [9] have shown that energy deposition to aluminium withambient pressures higher than 0.1 Torr (the case of PFMA-1) is independent from the specific ambientgas, and that the energy coupling is maximal, with residual energy coefficient between 0.8 and 1, so thatthe estimated value of Q is not only the residual heat in the flange, but is also very close to the whole heatimparted to the flange before ablation itself.Further studies are currently on the way to correlate the ablation rate of aluminium with the observedchange in mass of the disk and with the energy deposited per shot.4. References[1] E. Angeli, M. Frignani, S. Mannucci, F. Rocchi, M. Sumini, A. Tartari, The Heating of Plasma FocusElectrodes, Plasma Sources Science and Technology, vol. 15, n. 1, 2006, 91-8.[2] F. Malik, H. Schmidt, S. M. Hassan, R. S. Rawat, T. Zhang, S. Mahmood, J. J. Lin, T. L. Tan, P. Lee,S. V. Springham, Effet of Anode Shapes on Neutron Emission from a Repetitive Plasma Focus Device,submitted to IEEE Pulsed Power & Plasma Science Conference, Albuquerque, NM, June 2007.[3] M. Sumini, D. Mostacci, F. Rocchi, M. Frignani, A. Tartari, E. Angeli, D. Galaverni, U. Coli, B.Ascione, G. Cucchi, Preliminary Design of a 150kJ Repetitive Plasma Focus for the Production of 18-F,Nuclear Instruments & Methods A, vol. 562, n. 2, 2006, 1068-71.[4] F. Rocchi, S. Mannucci, D. Mostacci, M. Sumini, E. Angeli, A. Tartari, R. E. Beverly III, PFMA-1: A1-Hz, 150kJ Pulsed Power System for Plasma Focus Generation, submitted to IEEE Pulsed Power &Plasma Science Conference, Albuquerque, NM, June 2007.[5] D. Mostacci, E. Angeli, A. Tartari, M. Frignani; F. Rocchi, M. Sumini, Preliminary Results on theProduction of Short-lived Radioisotopes with a Plasma Focus Device, Applied Radiation and Isotopes,vol. 63, n. 5-6, 2005, 545-51.[6] J. G. Gilligan, M. A. Bourham, E. C. Tucker, Effect of Disruptions on Plasma-Facing Components,16th IEEE/NPSS Symposium Fusion Engineering, SOFE '95, vol.1, 1995, 424-429.[7] K. B. Nornoo, T. L. King, Ablation Measurement Technique for Fusion Reactor Components, 17thIEEE/NPSS Symposium Fusion Engineering, vol.2, 1998, 877-880.[8] B. N. Chichkov, C. Momma, S. Nolte, F. Von Alvensleben, A. Tuennermann, Femtosecond,Picosecond and Nanosecond Laser Ablation of Solids, Applied Physics A, vol. 63, 1996, 109-115.[9] A. Y. Vorobyev, V. M. Kuzmichev, C. Guo, N. G. Kokody, P. Kohns, Thermal Energy Coupling to Alin Ablation with ms-, ns-, and fs-laser Pulses, Proceedings of LFNM 2004, 6th International Conferenceon Laser and Fiber-Optical Networks Modeling, 2004, 281-283.Thp

NEW PERSPECTIVES FOR MEDICAL APPLICATIONSUSING THE TRIGA MARK I IPR-1 RESEARCH REACTORA.S LEAL, R.S. GOUVÊA, M. Â. B. C. MENEZES, J. SOPRANI,P. R. O. SILVA, J. L. MATTOSLaboratory of Nuclear Measurements, Nuclear Technology Development Centre (CDTN)/ National NuclearEnergy Commission (CNEN),Rua Mario Werneck s/n C.P. 941, CEP 30123-970, Belo Horizonte, Minas Gerais, BrazilM. A. SOARES, E. PEREIRA-MAIA, W. GUERRA, F. C. S. DE PAULADepartment of Chemistry, Federal University of Minas Gerais,CAMPUS UFMG, CEP 30123-970, Belo Horizonte, Minas Gerais, BrazilABSTRACTThis work presents the more recent enhancements of the TRIGA MARK I IPR-R1 researchreactor located at Nuclear Technology Development Centre/Brazilian Commission forNuclear Energy (CDTN/CNEN) focusing medical applications. It is presented thepreliminary results of the preparation of some specific molecules of platinum compoundsto be used as radiotracer in pharmacokinetical studies as well for therapeutic purposes asantitumoral or antibiotic.1. IntroductionResearch reactors have contributed to the development of the nuclear science and technology for thepast 50 years, but the new discoveries and innovations need newer tools and more powerful reactorwith special attributes. Even though many research reactors worldwide are underutilized and manyolder ones will be shut down next years, the need for research reactors in not waning. According to arecent diagnostic of a survey performed by IAEA, many research reactors worldwide, as is the case ofTRIGA, are underutilized due to several factors, mainly due to the inefficient communication ownerclientof the reactor [1,2]. The starting up of the TRIGA MARK I IPR-RI research reactor in 1960marked the beginning of the activities of the Nuclear Technology Development Center (CDTN) one ofthe five institutes of the National Commission of Nuclear Energy, (CNEN) ] . Since then, the reactorhas been used mainly for neutron activation analysis and training of the reactor operators. The TRIGAIPR-R1 CDTN´s research reactor is of the type MARK I, where the core is below the floor level, as itis shown in Figure 1 [3,4].Figure 1 - Views of the well (left) and the core of the TRIGA IPR - MARK 1 reactor [5].1

More recently, new research projects were initiated to enhance the reactor utilization. Since 2001,special labelled molecules, to be used in biodistribution experiments for the investigation of new drugsdesign, have been obtained [6]. Brazilian gemstones, specially topaz, are being irradiated to improvethe color [7,8] and neutron radiography images a limited to the samples of small size have been taken[9]. The reactor is operating at 100kW but the actual configuration of the core allows the increasing ofthe power up to 250kW.Specially during the last five years, results of researches involving new drugs or new strategies ofadministration of drugs already in the market opened a very interesting and important field ofapplication for the TRIGA reactor: the labelling of special molecules to be used in biodistribution andfor pharmacokinetical studies in vitro and in vivo. The first compound irradiated in the TRIGA withthis purpose was the CDDP, cis-dichlorodiammineplatinum (II), Pt(NH 3 ) 2 Cl 2 . CDDP is an effectivechemotherapeutic largely used to treat systemic tumours in several organs: testicles, ovary, head, neck,bladder. However, its side effects are serious principally the nephrotoxicity [10-14]. Investigation ofnew formulations containing CDDP to minimise or eliminate these effects is extremely relevant. Adetailed description of irradiation of CDDP by CDTN and its applications are described elsewhere [6].Besides CDDP, in the last few years new compounds of platinum and other metals with antitumoraland/or antibiotic activity has being investigated as, tetracycline-platinum II (Tc), Pt(C 22 H 24 N 2 O 8 )Cl 2complex . Tetracycline is one of most important antimicrobial agents with broad spectrum but its lowtoxicity, low cost and oral administration led to and indiscriminate use and appearance of bacterialresistance. Previous studies performed with Tc, showed a more pronounced antibacterial effectcompared to tetracycline [15]. More recent results obtained by our group showed also andenhancement of the antitumoral activity of the radiolabelled Tc. Further in vivo studies must beperformed to confirm this important potential use of Tc. Once a chemotherapic agent is radiolabelledit is interesting to investigate the possible synergic effect of its chemotherapy and radiotherapyactions. The preliminary results obtained using the radiolabelled platinum compounds of in vitrostudies showed very interesting results [16].2. Material and MethodsThe development of new medicines or a new formulation of those already existent for clinical userequires many stages of assessment such as biodistribution, pharmacokinetics and toxicity studies.The evaluation of in vivo biodistribution using radiolabelled compounds has many advantagescompared with others available analytical methods, such as, spectrophotometry of atomic absorptionwhich requires laborious sample treatment, and consequently, longer time for the analysis. In the caseof platinum compounds it may allows, in the future, to investigate its potential clinical use inpostoperative treatment using gamma imaging systems already. The radionuclides produced byirradiation of Pt compounds CDDP and Tc are showed in Table 1 [6].Table 1. Main radionuclides produced by platinum compounds after irradiation and nuclear data.StableNuclideNaturalabundance(%)NuclideProducedHalf-life γ energy in keV , (abundance %)190 Pt 0.01191 Pt 2.96d 359.93 (6), 409.48 (8), 538.9 (13.7)192 Pt 0.79193m Pt 4.33d -----194 Pt 32.9195m Pt 4.02d 30.8 (2.3), 98.8(11.4), 129.7 (2.8)196 Pt 25.3197 Pt 18.3h 279.11(2.3), 191.36( 3.7)198 Pt 7.2199 Pt 30.8m 317.06 (4.87), 493.74 (5.7), 542.9 (14.8)37 Cl 24,238 Cl 37.24m 1642.69(31.0), 2167.68(42.0)2

The direct irradiation of Pt compounds is undoubtedly the most simple procedure in spite of itslimitations related to maximum specific activity, chemical and radiochemical purity to be reached. Itoccurs due to the breaking of Pt -NH 3 and Pt -CI bonds as it showed in Figure 1, by the recoil of theexcited nuclei of Pt and Cl, - the Szillard-Chalmers effect - and by the collision with high energyneutrons [17,19]. Thus, the preparation of Pt compounds showing an high specific activity, goodchemical and radiochemical purity by the direct irradiation process is important in order to optimisethe in vivo biodistribution studies, avoiding eventual corrections of decay time in the activity for eachsample and also allowing the assessment of its potential clinical use with gamma imaging systems.Figure 1. Molecular structures of CDDP (left) and tetracycline-platinum (right) [15]2.1 Irradiation and Gamma SpectrometryCDDP was initially irradiated under the flux of average thermal neutrons of 6.4 x 10 11 n.cm -2 .s -1 at100 kW for 2, 3, 4, 6 and 8hours. After 4 hours of irradiation the sample turned from yellow, itsoriginal colour, to grey, the colour of elemental platinum, confirming the molecule degradation, asalready expected. As some platinum isotopes have high integrals of resonance related to the thermalcross sections, the CDDP samples covered by a Cd capsule of 1mm thickness were irradiated in orderto obtain an higher specific activity with acceptable chemical and radiochemical purity. Tc samplewas initially irradiated by 8h covered by Cd capsule but the best results of antitumoral activityperformed using cells of leukaemia was obtained with bare sample of Tc irradiated by 2h. Anyevidence of disruption of the Tc molecule was verified for both conditions of irradiation. Thedetermination of radiochemical purity for CDDP and Tc are described elsewhere [6, 20] Gammaspectrometry was applied in a HPGe detector with 15% nominal efficiency and 1.85 keV FWHM for1332 keV peak of 60 Co GENIE 2000, CANBERRA software was used for data preparation andspectrum analysis.3. Results and Discussion3.1 Irradiation of Pt compoundsIn the Figure 2, it is showed the gamma spectra of bare samples and Cd-covered samples irradiated for2 hours, after the decay time of 30 min.. It can be observed that the spectra are similar up to 300keV,but above this limit, photopeaks of 191 Pt of 359.6 keV, 409.4 keV and 538.9 keV were detected in thebare sample but not in the spectrum of the sample irradiated with cadmium. This result can beexpected according to the values of (σ th ) and RI of 191 Pt [6].Energy(keV)Figure 2. Pt-compounds spectra of 2 h of irradiation: upper: bare sample; lower, Cd-covered sample.3

3.2 Radiation counting in miceMice were used to evaluate the biodistribution of CDDP after its administration . The results ofdetection of radiation of CDDP* in liver are shown in the Figure 3 The liver was collected andsubmitted to the gamma spectrometry using the well-type NaI detector, of 20% efficiency. Themeasurement time was of 2 min. The irradiation time of 8 hours allowed to a great level of detection.Figure 3. Gamma counting of CDDP in the liver of mice samples [21].The specific activity was of 11.5 kBq.mg -1 for the bare sample irradiated during 2hours andapproximately 57 kBq.mg -1 for the Cd-covered sample during 8hours.3.3 In vitro-effects of radiolabelled CDDPFigure 4 shows the comparative results of the antitumoral effect against GH3 tumoral cells betweenirradiated and non irradiated CDDP. This result is very important and open new interestingperspectives for the TRIGA IPR1 research reactor.Figure 4. Effect of the irradiated and irradiated CDDP over the metabolism of tumoral GH3 cells.Preliminary investigations of citotoxicity of the radiolabelled Tc against cells of human leukaemiaK5562 showed an enhancement of 22 times if compared with similar results obtained with nonirradiated Tc. With the same purpose, new antitumoral metallic complex of Cu and Sn has beeninvestigated and will be published briefly.4. ConclusionsResults obtained through irradiation in the TRIGA reactor operating at 100kW open new andpromising prospects to enhance the use of the reactor for research and development of new drugs ornew formulation of those existent ones. Optimistic prospects for radiolabeled drugs production opennew possibilities for its use in biological research at CDTN/CNEN and partners confirming the4

importance of nuclear area as an indispensable research tool .AcknowledgementsThe authors would like to thank the FAPEMIG, CNPq for their financial support and also colleaguesof the Reactor and the Irradiation Service of CDTN: F.M. Júnior, A. Z. Mesquita, P.F.de Oliveira,L.O.I.S. Câmara and A. Amaral, for their valuable technical co-operation.5. References[1] IAEA, TECDOC 1234, 2001.[2] Research Reactors Purpose and Future, IAEA, 2004[3] P. C. Tófani, M. Paiano: NUCLEBRÁS, CDTN, CNEN/CDTN-611, Belo Horizonte, 1989.[4] Menezes, M. Â.B.C.; Sabino, V.S.C.; Kastner, G. F.; Proceedings, pp.303-306, EncontroNacional de Química Analítica, Belo Horizonte, 1995.[5] Private Communication, Mesquita, A., Z, CDTN/CNEN, 2002[6] Leal, A.S., Carvalho Júnior, A.D., Abrantes, F.M., Menezes, M.A.B.C., Ferraz, V., Cruz, T.S.,Cardoso, V.N., De Oliveira, M.C., Appl. Rad .Isotop. 2006, 64, 178.[7] A.S. Leal, K. Krambrock, L.G.M. Ribeiro, M.Â.B.C. Menezes, P. Vermaercke and L. Sneyers, ,Nuclear Instruments and Methods in Physics Research A (2007), doi:10.1016/j.nima.2007.05.069[8] K. Krambrock, L. G. M. Ribeiro, M. V. B. Pinheiro, A. S. Leal, M. Â. de B. C. Menezes, J. M.Spaeth, Phys Chem Minerals (2007) doi: 10.1007/s00269-007-0160-z[9] A.S. Leal, R. S. Damaso, R.R. Rodrigues, Proceedings of 3rd World TRIGA Users Conference,2006.[10] Calvert, H.; Judson, I., Van der Vijgh, W. J. Cancer Sur. 1993, 17, 189.[11] Daugaard, g.; Abildgaard, U. Cancer Chemother. Pharmacol. 1989, 25, 1.[12] Reynolds, J. E. F.; Martindale The Extra Pharmacopeia. 1999, 513.[13] Stewart, C. F.; Hampton, E. M. Am. J. Hosp. Pharm. 1989, 46, 1400.[14] Trissel, L. A.; Martinez, J. F. Am. J. Health-Syst Pharm. 1996, 53, 1041.[15] E., Chartone-Souza, T.L.Loyla, M. Buciarelli –Rodriguez, M.A.B.C. Menzezes, N. A. Rey, E.C.Pereira-Maia Journal of Inorganic Biochemistry 2005, 99, 1001.[16] Private Communication, Gouvea, R., S, Maia, E.C. Pereira, CDTN/CNEN, 2007[17] Hoeschelle, J. D.; Butle, T. A.; Roberts, J. A.; Guyer, C. E. Radiochimica Acta. 1982, 31, 27.[18] Kawai, K.; Maki, H.; Ehrlich, W.; Akaboshi, M. J. Radional. Nucl. Chem. Lett. 1989, 136, 67.[19] Sykes, T. R.; Stephens-Newsham, L. G.; Noujaim, A. A. Appl. Radiat. Isot. 1986, 37, 231.[20] Friedlander, G. et al. Nuclear and Radiochemistry, 3 rd Ed. John Wiley & Sons New ork, 1981.[21] Leal, A.S., Carvalho Júnior, A.D., Abrantes, F.M., Menezes, M.A.B.C., Ferraz, V., Cruz, T.S.,Cardoso, V.N., De Oliveira, M.C., Brazilian Archives of Biology and Technology Rad .Isotop. 2005,48, 85.5

PROBABILITY OF CAUSATION BETWEEN CANCERINCIDENCE AND GO BEFORE OCCUPATIONAL EXPOSUREAMONG NUCLEAR WORKERS IN BULGARIAN. CHOBANOVA, At. JAGOVA, J. DJUNOUVARadiation Epidemiology Sector, National Center of Radiobiology and Radiation Protection132 “Kl. Ohridsky” blvd., 1756 Sofía, BulgariaABSTRACTPersons, working in ionising radiation (IR) environment are exposed to continuous radiationwith low doses and low dose rate. The assessment method of the Probability of Causation (PC)provides possibility to determine the causative connection between the occupational exposureand the consequent radiation-dependant cancer. An analysis has been carried out to 27 people.The probability for such a connection depends on the duration of the employment period in anIR environment, on the cumulative dose for that period, age and sex and the latent period forthe corresponding localization. Data have been processed with a software product “Survrad”.For 24 of the cases there has been established no causative connecting between cancer diseaseand a foregoing occupational radiation. In 3 of them a 1% (0.0-0.02) PC is found. The presentstudy does not find out a causative connection between cancer and a foregoing occupationalradiation by normal work. The results show that the doses and the dose rates obtained by theworkers of the Nuclear Power Plant are low.KEY WORDS: Nuclear Power Plant, cancer, employees, probability of causation.INTRODUCTIONStudies of workers in the nuclear industry have the potential to provide information on radiationrisk after low doses delivered over many years at low rate. There have been carried out a number ofcomprehensive studies in USA (1) United Kingdom (2) Canada (3, 4) and Japan (5). Studies of Chernobylcleanup workers have also been conducted (6). The results of the studies occupied in an IR environmentare controversial. This is due to the relatively low doses as well as to the limited similar studies. Withview to this in the last years the efforts have been aimed at the unifying of the studied cohorts (7, 8).Epidemiological studies among workers in the nuclear industry are being preferred because ofthat they offer a possibility for a direct assessment of the effect of the chronically expose with low dosesby a human being.The purpose of this study is to determine the probability of the occupational radiation to be areason for the diagnosed newly found malignant disease with the group of workers from NPP.MATERIALS AND METHODSThe concept of Probability of Causation (PC) determines what is the probability for the foregoingradiation to be the cause for the cancer disease with a given individual. The PC was defined by the UnitedStates National Institutes of Health Ad Hoc Committee as the fraction of the risk at the age of occurancefor the given cancer that is attributable to the exposure, i.e.Δr(D,t,e,s)PC =where r 0 (a, s) is the cancer rate for age a and sex s for ther0(a,s)+ Δr(D,t,e,s)particular cancer type under consideration and ∆r(d ,t ,e, s) is the excess cancer rate due to a dose ofradiation D at age e and time since exposure t (= a – e). The rate for a given cancer is the probability perunit time for a person of sex s and age a to develop the cancer (9).The probability of a causative connection is accepted to be expressed by excess relative risk (ERR) andthe dependence looks like the following:PC =ERR(D,e,s)1+ERR(D,e,s)All calculations have been conducted with the software product "Survrad", with which it can bedetermined the cancer risk and the probability reason for 17 radiogenic malignant neoplasm.For a period of 10 years 67 cases with cancer have been diagnosed altogether. Data foroccupational radiation of 27 workers with malignant illnesses have been submitted by Division “Safety”

in the NPP. 40 workers are not under control – they are not included in the system for individualdosimetric control due to lack of necessity. As a risk factor for the cancer emergency only theoccupational exposure has been analyzed, as data for the rest radiation and non-radiation harms isincomplete.Necessary data to conduct the analyses is year of birth, sex, year of first employment in IRenvironment and inclusion in the system for individual dosimetric control, occupational experience in anIR environment with individual dosimetric control, annual dose (for each year of employment, includingnull values of the dose), cumulative dose for the whole employment period in the IR environment,diagnosis, year of diagnosis.RESULTSThe distribution of all cases (number of cases and their percentage proportion) according tumorlocalization by organs and systems is presented on Tab. 1.Tab. 1: Number and percentage of cancer cases diagnosed among worker studied.Number of casesPercentageSexMale 43 64Female 24 36Type of cancerLung cancer 12 18Breast cancer 8 12Cancer of the Digestive tract 12 18Cancer of the Urinary tract 6 9Cancer of the Genital Tract 10 15Other 19 28Total 67 100On table 2 are shown the main characteristics of diagnosed cancer cases, for which full dosimetricdata is available and PC is calculated. Analysis of the data (Table 2) shows that the dose loading of theexamined employees of NPP is low, below 200 mSv. Only one case (No. 18) is an exception when thecumulated dose for the period prior to the diagnosed disease is approximately 300 mSv.With 24 out of 27 workers the probability their occupational radiation to be the reason for thecancer is zero. There is a probability that the additional exposure to radiation could be the reason forcarcinogenic illnesses in 3 employees (Сase No. 6, 14, 18).DISCUSSIONThe studies of the health status of persons, professionally occupied in an IR environment can giveus a direct assessment of the correlation between incidence and chronically radiation with relatively lowdoses, as interpretation of the results should be careful (it is very often when no other carcinogens fromthe professional and living environment are not taken into consideration et ct).Basically the NPP workers are men because of which in the present study prevalent are cancercases by men. The distribution of the malignances by localizations in the studied cohort is in accordancewith their distribution characteristic for the whole country (10).Cancer cases are late health radiation effects which mean that between the radiation and theappearance of the disease there passes a certain period of time. The duration of that latent period isdifferent for the different cancer localizations.

Tab. 2: Assessment of probability of causation between occupational exposure and cancer diseases among workers in NPP.№ Sex DiagnosisAge at cancer Employment period Cumulative dose in mSv for the Probability of causation anddiagnosed before diagnosed cancer whole period before diagnosis 95% confidence interval1 М Skin cancer 49 8 4.02 0.002 M Lung cancer 61 1 0.1 0.003 M Lung cancer 47 9 133.49 0.004 M Lung cancer 41 4 3.88 0.005 M Lung cancer 45 24 14.3 0.006 M Lung cancer 50 12 153.17 1% (0.0-0.02)7 F Breast cancer 49 5 0.91 0.008 F Breast cancer 56 6 0.49 0.009 F Breast cancer 50 16 4.38 0.0010 F Mastitis carcinomatosis 57 12 0.26 0.0011 M Laryngeal cancer 51 25 16.3 0.0012 М Esophageal cancer 34 13 34.06 0.0013 F Stomach cancer 54 11 3.15 0.0014 M Pancreatic cancer 43 14 181.15 1% (0.0-0.02)15 M Rectal cancer 50 4 1.47 0.0016 F Rectal cancer 56 11 5.03 0.0017 M Sigmoid cancer 53 28 118.49 0.0018 M Sigmoid cancer 48 29 281.87 1% (0.0-0.02)19 M Sigmoid cancer 54 22 0.7 0.0020 M Kidney cancer 42 7 2.93 0.0021 F Kidney cancer 50 8 1.25 0.0022 M Bladder cancer 51 14 0.61 0.0023 F Ovarian cancer 51 15 4.76 0.0024 M Cancer cerebri 48 13 200.88 0.0025 M Meningeoma 57 9 211.5 0.0026 M Oligoastrocitoma 44 8 48.28 0.0027 M Meta hepatic 50 24 22.06 0.00NPP – Nuclear Power Plant; M - male; F - female;

The comparison of latent period continuity including different cancer cases in this presentresearch involving the relevant localizations as per specialized data shows that the first opion is shorter fora great deal of the cases. This comparison made on the ground of occupational exposure evaluation aboutexamined cancer cases is rather unlikely.Research aiming the analysis of the healthy effects of low dose and capacity exposure amongoccupational exposed persons has been restrictedly performed in Bulgaria. The observation of the healtystatus of NPP workers did not show any registered cases of occupational exposure so far. This is due tolow dose of occupational exposure (average annual effective dose for the NPP crew in the last few years isfrom 0.5 to 3.0 mSv per capita) as well as to emergency exposure of the crew (11, 12).The analysis confirms that the dose load of the examined workers is low. 90% of all examined PCis practically 0 as expected, having in mind the low values of registered occupational exposure. These aremalignant tumors that had spontaneously apperead and which would have been developed in a life timeperiod of tha said person regardless his working place.There is only 1% probability in every 3 cases as a connection between a previous occupationalexposure and a disease diagnostically confirmed. Tha main etiologic factor for lung cancer is smoking(13). The lack of detailed information about harmful habits (smoking) (case No. 6) of smokers in thisresearch limits the particular participation of the said factors. The most affected age for pancreatic canceris above 65 years old compared to the present case (case No. 14, the diagnosis has been confirmed in anearlier age). The most affected age groups of colon cancer are between 60 - 80 years (case No. 18) Theconfirmation of the genetic factors has been proved as a determined importance of the nutrive regime(food rich in animal fat increases the risk of cancer) (13).Meantime, the evaluations of inductive-radiation cancer corresponding to the probability ofreasonable connection are rather unlikely. They are bounded to inevitable extrapolations, dosimetry aswell as of the participation of other cancerous professional and domestic conditions.PC calculation offers the best method of systematically quantifying the probability that aparticular cancer may have been induced by radiation in an individual (9).CONCLUSIONThe present research is one among the fewest making an attempt to clarify the healthy effects ofoccupational exposure by IR върху among workers in the от nuclear industry. The lack of radiogenicsolid tumors and leukaemia cases in appliance of modern research methods manifests the good level ofexisting radiation preventive measures in NPP – “Kozloduy”.REFERENCES1. Gilbert E.S., D. Cragle, L. Wiggs. Updated analyses of combined mortality data for workers at theHanford site, Oak Ridge National Laboratory and Rocky Flats Weapons Plant. Rad. Res. 1993; 136: 408 -421.2. Muirhead C.R., A.A. Coodill, RG.Hylock, J. Vokes, M.P. Little, D.A Jackson et al. Occupationalradiation exposure and mortality; Second analysis of the National Registry for Radiation Workers. J.Radiol. Prot. 1999; 19: 3-26.3. Sont W. N., J. M. Zielinski, J. P. Ashmore, H. Jiang, D. Krewski, M.E. Fair, et al. First analysis ofcancer incidence and occupational radiation exposure based on the National Dose Registry of Canada. Am.J. Epidemiol. 2001; 153: 309-318.4. Gilbert E .S. Studies of workers to low doses of radiation. Am. J. Epidemiol. 2001; 153: 319-324.5. Iwasaki T., M. Murata, S. Ohshima, T. Miyake, Sh. Kudo, Y. Inoue, et al. Second analysis of mortalityof nuclear industry workers in Japan, 1986-1997. Rad. Res. 2003; 159: 228-238.6. Hatch M., E. Ron, A. Bouville, E. Zablotska, G. Howe The Chernobyl disaster: cancer following theaccident at the Chernobyl nuclear power plant. Epidemiol. Rev. 2005; 27: 56-66.7. Cardis E., E. S. Gilbert, L. Carpenter, G. Howe, I. Kato, BK Armstrong et al. Effects of low doses andlow dose rates for external ionizing radiation: cancer mortality among nuclear industry workers in threecountries. Rad. Res. 1995; 142: 117-132.8. Cardis E., M. Vrijheid, M. Blettner, E. Gilbert, M. Hakama, C. Hill et al. Risk of cancer after low dosesof ionizing radiation: retrospective cohort study in 15 countries. Br. Med. J. 2005; 331: 77-83.

9. IAEA, Methods for estimating the probability of cancer from occupational radiation exposure. IAEA,Vienna, 1996, IAEA-TECDOC-870.10. National Cancer Register. Cancer Incidence in Bulgaria, vol. X-XIV Oncologia, Suppl. (Sofia)Medizina and Fizkultura; 1999 - 2003.11. NCRRP, “40 years of NCRRP 1962-2003”, Scientific Proceeding of Conference, Sofia, 2003.12. Vasilev, G. Ionizing radiation as a source of both occupational and public exposure. Is there anydifference between them? Conclusion for radiation protection practice. In: The VIII National Conferenceon Biomedical Physics and Engineering. Proceedings of Conference, Sofia, 12-14 Oct. 2000.13. Oncologia 2001, Ciela Soft and Publishing; 2000.

Poster PresentationsWS I:Education, Training and KnowledgeManagement

VELLA PROJECT: AN INITIATIVE TO CREATE ACOMMON EUROPEAN RESEARCH AREA ON LEADTECHNOLOGIES FOR NUCLEAR APPLICATIONSG. BENAMATI, S. DE GRANDIS, C. FOLETTIENEA CR Brasimonelocalità Brasimone, 40032 Camugnano (BO), ItalyH. AIT ABDERRAHIM, A. AL MAZOUZIThe Belgian Nuclear Research CentreBoeretang 200 - LHMABE-2400 MolD. BRICENOCIEMATAv. Complutense, 22, E 28040 MadridC. FAZIO, J. KONYS, R. STIEGLITZForschungszentrum Karlsruhe GmbHHermann-von-Helmholtz-Platz 1, D 76344 Eggenstein-LeopoldshafenD. GORSECNRS / CECM15, rue Georges Urbain, 94407 VITRY SUR SEINE Cedex, FranceF. GROESCHELPaul Scherrer InstituteCH 5232 VilligenC. LATGELe Commissariat à l’Energie Atomique, CEA-Cadarache13108 Saint Paul lez Durance, FranceABSTRACTVELLA (Virtual European Lead Laboratory) is a Euratom FP6 project which emergedfrom the idea to create a common research area among the European Union and itsassociate countries in the field of lead technologies.A wide use of lead and lead alloys is foreseen in several nuclear-related fields, i.e. it isstudied as coolant for critical and sub-critical nuclear reactors, as spallation target forneutron generation and for tritium generation in fusion systems. Given this possible futureextensive use of lead in nuclear systems, which require a deep understanding of its physicalproperties and engineering applications, large efforts are dedicated to lead technologies. Inparticular, the EU has launched large R&D programmes, strongly interconnected. Amongthose programs VELLA, has the ambitious aim to homogenize the European research areain the field of lead technologies by exploiting the tools of Networking Activities,Transnational Accesses activities and Joint Research Activities..1. IntroductionA wide use of pure lead, as well as its alloys (such as lead-bismuth, lead-lithium), thanks to itsfavourable properties, is foreseen in several nuclear-related fields, i.e. it is studied as coolant for

critical and sub-critical nuclear reactors, as spallation target for neutron generation and for tritiumgeneration in fusion systems.Given this possible future extensive use of lead in nuclear systems, a deep understanding of itsphysical properties and engineering applications is mandatory. As a consequence, given the quitelimited nowadays experience, large efforts both at national level as well as within the EuropeanCommission are dedicated to the development and understanding of heavy liquid metal technologies.In particular, the European Commission has launched several large R&D programmes, stronglyinterconnected, such as TECLA (TEChnologies for Lead Alloys), MEGAPIE-TEST (MEGAwatt PIlotExperiment), EUROTRANS-DEMETRA (EUROpean research programme for the TRANSmutation ofhigh level nuclear waste in Accelerator Driven Systems-DEvelopment and assessment of structuralmaterials and heavy liquid MEtal technologies for TRAnsmutation systems), ELSY (European Lead-Cooled SYstem), and VELLA (Virtual European Lead LAboratory).VELLA is an FP6 project which has the ambitious intent to create a virtual laboratory for leadtechnologies. More in detail, the driving idea of VELLA is to homogenize the European research areain the field of lead technologies for nuclear applications in order to produce a common platform ofwork which will continue also beyond the VELLA initiative.Above all, VELLA has the ambitious intent to both create a network of all the principal laboratoriesand to strongly connect the different groups of experts, in order to have a common definition of thegood operational practices and to promote the exchange of the scientific results by means ofappropriate and innovative tools and procedures. It also has the important objectives to promote theaccess to the main existing facilities in the EU to different specialist groups. This will allowsupporting the technological development and the qualification activities as well as the formation of aEuropean “scientific community”, organized to meet all the technological challenges and the necessaryresearch requirements.In this framework, detailing the abovementioned goals, VELLA is articulated in Networking Activities(NA), Transnational Accesses activities (TA) and Joint Research Activities (JRA).2. Networking ActivitiesThe Networking Activities have as main objectives to create a virtual community of researchers, todefine common standards and protocols for the use of the facilities and to interact with theprogrammes and the institutions operating in this field.2.1 Networking Activity 1: “Management of the consortium activities”VELLA is a Consortium gathering thirteen European Research Institutions and Universities, from nineCountries, established in 2006, with the support of the European Commission, under a three-yearscontract.The initiative is coordinated by ENEA (Italy). The other partners, at the moment, are: Le Commissariatà l’Energie Atomique (CEA), CIEMAT, Consiglio Nazionale delle Ricerche, sezione di Genova (CNR-IENI), CNRS, Forschungszentrum Karlsruhe GmbH (FZK), Forschungszentrum Dresden -Rossendorf e. V.(FZD), Institut Quimic de Sarria -Universitat Ramon Llull (IQS), Kungliga Tekniska Högskolan (KTH),Nuclear Research Institute Rez (NRI), Paul Scherrer Institut (PSI), The Belgian Nuclear Research Centre(SCK-CEN), and Institute of Physics, University of Latvia (IPUL).The management structure consists of: an Executive Board (EB), formed of representatives from eachparticipant institution, which is the Consortium decision-making and arbitration body. A TechnicalAdvisor Committee (TAC) has the responsibility for scientific advices and recommendations coveringthe topics of the Project. A Scientific Access Panel (SAP) has the duty of selecting the accesses to theinfrastructures during the whole duration of the Project, selecting the Human Mobility (HM)applications, monitoring the accepted actions. A Scientific and Technical Panel (STP) has theresponsibility to manage the Joint Research Activities within the project execution. Finally, the Coordinatoris the single point of contact between the EC and the Consortium and it is responsible for theproject management and the Project Office (PO) is the body which manage the administrative, legal,financial and other non-technical aspects of the Project and assist the abovementioned bodies and theCoordinator in their activities.

2.2 Networking Activity 2: “Virtual community creation and results dissemination”The activity has the main objective to create a real “virtual” community of researchers in the field ofheavy liquid metal (HLM) technologies for nuclear applications through: the creation of an interactive website where it is possible to access all the publications relatedto the project, to have direct contact with researchers for further information and to visit the“virtual laboratory”; the realization of a “dedicated chat line” (and/or programmed teleconferences) for discussionamong researchers; the edition of a VELLA electronic newsletter reporting the most important news; the organization of dedicated workshops on specific issues related to the HLM technology.2.3 Networking Activity 3: “Harmonization of knowledge and establishment of goodpractices”The activity has a twofold objective: the improvement and the harmonization of the scientific knowledge of the present generationof scientists and engineers coming from different areas of the nuclear R&D projects related toHLM; the preparation of the new generation of scientists, by giving them a common platform ofknowledge and developing their skills in all aspects of HLM technologies and related areas.For attaining these objectives, two types of actions are considered: thematic workshops on HLM technologies to be organized by the VELLA partners during thethree-years duration of the initiative,; ‘Good practices’ workshops to be organized for young researchers involved in the field ofnuclear science coming from Universities, Agencies or Companies in the different researchteams or laboratories of the Organizations grouped in VELLA.2.4 Networking Activity 4: “Guidelines and procedures”The activity is aimed at developing guidelines for relevant issues for future HLM systems, collectingthe available information, analysing the experimental procedures used in different laboratories,evaluating the quality of the available data, identifying items not covered and realizing, if needed,some experiments to fill the existing gaps.The expected impact of this NA, therefore, is to provide the research community withrecommendations, procedures and data to be used for the design and construction of new HLMsystems. It is intended to represent a forum for discussion and understanding of the most relevanttopics. The recommendations and procedures issued, as results of this activity, are intended to be usedby the labs involved in HLM investigations. If possible, standard proposal will be written to besubmitted to ISO.2.5 Networking Activity 5: “Collaborations with other programmes and future R&D”The main goal of the NA5 is to promote a real integration among the EU activities on HLM alreadygoing on within Europe in several fields and to establish regular and coherent links among nationaland international programmes in the field of HLM technologies in order to increase the possiblecooperation.This objective will be pursued setting up a group of experts, including the major EU specialists in thefield of HLM, establishing links with other teams working on programmes related to the HLMtechnologies in order to have a complete overview of the different research actions going on andpromoting collaboration and exchange of knowledge.3. Transnational Access Activities

The TA are activities aimed to promote the access of researchers, universities and companies to theexisting infrastructures and knowledge, in order to increase the competitiveness of the Europeanindustry, to train the researchers in using the EU infrastructures during the three years duration of theproject and to help the human mobility between and towards the laboratories. These objectives will berealized using two different tools: access to the infrastructures, granted to the users interested in technological developmentand/or basic research; Human Mobility, to support researchers inside and outside the VELLA Consortium.For both the access to the infrastructures (Transnational Access) and the Human Mobility, a call forproposals will be published on the VELLA website (www.3i-vella.eu ) every year, solicitingapplications from interested candidates.3.1 Access to the InfrastructuresThe infrastructures made available within VELLA are the two virtual structures MATLAB andCHEMLAB, andthe large devices considered reference laboratories themselves. These infrastructurescan give access to researchers and users creating the unique opportunity of the accessibility to a largeEuropean common laboratory.MATLAB (MATerial LABoratory) is a virtual laboratory where it is possible to perform compatibilitytests among different structural materials and heavy liquid metals. The possible tests include tensile,fatigue, fracture toughness, creep and corrosion investigations on irradiated and unirradiated materialsThe corrosion science laboratory comprises both facilities for flowing liquid metal and facilities forstagnant tests.Parts of MATLAB, in fact, are LECOR & CHOEPE III (ENEA, Italy), COLONRI I&II (NRI, CzechRepublic), LINCE (CIEMAT, Spain), CORRIDA (FZK, Germany) and CICLAD (CEA, France) forcorrosion tests of unirradiated materials in flowing Lead and Lead-Bismuth. Also LIMETS I&II(SCK.CEN, Belgium), COSTA 1-6 (FZK, Germany) and COLIMESTRA (CEA, France) belong toMATLAB. They are stagnant devices for irradiated materials (the first one) and for unirradiatedmaterials (the last two devices).CHEMLAB (CHEMistry LABoratory) is a virtual laboratory on physic and chemistry science which isdevoted to the chemistry control related studies of Lead-Bismuth eutectic systems, such as the oxygencontrol and monitoring, as well as the other impurities control in both the liquid and the gas phases. Itcomprises ELEFANT (FZD, Germany) STELLA (CEA, France) OCEAN & THESYS (FZK,Germany) and CHEOPE II (ENEA, Italy).Finally, also three large facilities offer access in the framework of the VELLA TA: “CIRCE“, pooltype facility for the study of thermal – hydraulics issues, “TALL”, originally designed to investigatethermal-hydraulic phenomena for the ADS normal and transient conditions, and “THEADES”, a looptype facility having relevant dimension for thermal-hydraulic studies and large components testing.4. Joint Research ActivitiesThe JRA are technical activities with the objective to: develop and the technologies needed for theoperation of large facilities for future Gen IV reactors and ADS cooled by HLM; develop the neededcomponents and instrumentation; study the liquid metal thermal-hydraulics and analyse the effects ofirradiation in presence of LBE. Finally, to homogenise and complete the results obtained in the otherresearch programmes related to the HLM technologies for nuclear applications.4.1 Joint Research Activity 1: “Lead technology”The objective of this JRA is the development and validation of the needed technologies and therequired know-how for the future Gen-IV reactors and Accelerator Driven Systems, providing, aboveall, data on:

the oxygen control in large lead pool-type configurations, including the development ofoxygen sensors for these applications ; the corrosion behaviour of potential materials in stagnant lead; the influence of lead on the mechanical properties (tensile, creep rupture).4.2 Joint Research Activity 2: “HLM components, instrumentation development andsystem operation”The JRA2 has the objectives of: defining and developing optimised systems and dedicated components for future experimentscharacterizing the features and the potentialities of the existing dedicated instrumentation; developing new instrumentation (e.g. pumps, flowmeters, coolant handling systems, etc.) foroptimised experimental facilities; defining and validating experimentally, if necessary, the basic operational procedures; developing a standardized In Service Inspection and Repair methodology.4.3 Joint Research Activity 3: “Liquid metal thermal-hydraulics”Three experiments are proposed in this JRA, which will be conducted by at least by two associationswith a good experience in liquid metals and are followed by the whole HLM community. single phase turbulent heat transfer from highly heat loaded surfaces; single phase buoyant and mixed convection in loop systems; free surface shaping in heavy liquid metal flows in dependence of the geometric and fluiddynamic boundary conditions.4.4 Joint Research Activity 4: “Irradiation in presence of LBE”The aim of this JRA is to cover a major lack in the capabilities of the existing facilities to performexperiments by combining HLM and neutron irradiation. Hence a high temperature lead alloy loopwill be developed of, to be installed in the MTR reactor BR2. The in-pile HLM loop should bedesigned in such a way that it can be used for the purposes that are common to the development ofGEN-IV LFR and Pb(-Bi) ADS, whereby some of these points are as well covered in LiSoR.5. AcknowledgmentsThis work is supported by the European Commission under the Contract VELLA number 036469.6. References[1] G. Benamati, S. De Grandis, VELLA Project Presentation, December 2006.[2] Vella official website: www.3i-vella.eu

Poster PresentationsWK II:Safeguards and terrorism

DIGITAL ADVANCES IN PULSED FAST NEUTRONANALYSIS FOR THE USE IN COUNTER TERRORISMM.D. ASPINALL, B. D’MELLOW, M.J. JOYCE, R.O. MACKINControl and Instrumentation Group, Engineering Department, Lancaster UniversityLA1 4YR Lancaster – United KingdomA.J. BOSTON, P.J. NOLANDepartment of Physics, University of LiverpoolL69 3BX Liverpool – United KingdomN. HAWKES, D.J. THOMASNational Physical LaboratoryHampton Road, TW11 0LW Teddington – United KingdomA.J. PEYTONSchool of Electrical and Electronic Engineering, University of ManchesterSackville Street, M26 1QD Manchester – United KingdomP. MONKSNIS LtdAckhurst Road, PR7 1NH Chorley – United KingdomABSTRACTPulsed Fast Neutron Analysis (PFNA) has long been suggested as a potential means for thenon-destructive detection of hidden explosives and contraband. The premise for thetechnique is to use fast neutrons to stimulate γ-ray emissions that are characteristic of thetarget isotope. However, it has often proved difficult to 1) separate the response of thematerial under detection from materials that have a similar isotopic composition, 2)distinguish the radiation response from scattered neutrons and 3) to process the datasufficiently quickly for real-time processing. In this paper we report on a novel approach toPFNA that exploits the digital discrimination of neutrons and γ rays to provide informationon the shape and composition of suspicious objects in mass transport systems.1. IntroductionIonising radiation has long been used for the non-destructive screening of opaque objects ever since x-ray use was pioneered by Roentgen. The recent global escalation in the threat level associated withterrorist attacks, and horrific events such as the Madrid bombings and the July 7 th 2005 attacks inLondon, have renewed research interest in a variety of potential screening capabilities, and especiallythe use of neutrons. Neutrons are very useful because of they have great penetrating ability, as a resultof their having no electrostatic charge, and because they are easily detected by relatively inexpensivedetector systems that are now widely available. Furthermore, inelastic interactions between theincident neutron and the atomic nuclei making up the object under scrutiny often result in the emissionof prompt γ rays. The energies of these photons are characteristic of the isotope that emitted them and,since most explosives comprise of carbon, nitrogen, oxygen and/or hydrogen, it is thus possible toidentify the substance emitting them. A variety of possibilities based on this concept have been studiedover the past twenty years and a few are now deployed commercially.However, the measurement of neutron energy is notoriously difficult, predominantly becauseall indication of the energy of the neutron is often lost via the indirect basis on which neutrons areusually detected, in order for them to be converted to electronic signals. Limited spectroscopicinformation can be gleaned by unfolding the neutron response from several detectors with

complementary spectral sensitivities, but this is sophisticated and often results in too crude a responseto extract detailed analytical information. Alternatively, the time taken for neutrons to traverse a givendistance, the time-of-flight (TOF), can be used to extract the neutron energy and/or to determine theorigin of the neutron but then it is necessary to also know the time at which the primary neutron wasemitted. This requires the use of neutron sources that are pulsed very rapidly indeed. Interestingly, formost inelastic reactions relevant to freight and baggage screening, neutron energies in excess 10 MeVare required. Whilst 14 MeV neutron generators are available, based on the deuterium-tritium (D-T)reaction, the traverse-time of a standard candidate container by a neutron with such energies will be ofthe order of tens of nanoseconds.Fast neutrons are often detected with scintillation detectors. Instruments using proprietaryorganic liquid scintillation compounds respond with typical pulse-lengths of the order of a fewnanoseconds. Whilst high-end commercial electronic systems have been available for some time thatare able to digitise γ-ray events arising from semiconductor detectors (see for example DESCOVICHet al. (2005)), it has only been possible to sample at a rate compatible with digitisers and until recentlydigitisers were not fast enough for scintillators. Thus, neutron detection and measurement has beenlimited by the capabilities of analogue data acquisition systems, many of which were developed thirtyyears ago. The discrimination and rejection of contaminant γ-ray events (to which organic scintillatorsalso respond) has long been performed in the analogue domain. Amongst the principal restrictions ofsuch systems are the inability to post-process the pulse data and the requirement of very careful skilledoptimisation. These result in difficulties in the automation of such processing and the inherentincompatibility with computer-based control. Most importantly, all record of the continuous signalthat describes each event is lost and only the salient features, such as the pulse amplitude or rise time,can be recorded.In this paper we report on the digital processing of neutron events for use in counter-terrorismapplications. The potential flexibility of this advance is described in the context of improved neutron/γdiscrimination and the digital measurement of TOF. Although much of the data presented here hasbeen acquired with a standalone digital data acquisition system more consistent with a laboratoryexperiment, the performance requirements are compatible with the burgeoning capability of fieldprogrammableembedded systems.2. Pulse-shape discrimination2.1 Experimental detailsThe experimental set-up for this aspect of the research consisted of a 4.5 ml cylindrical cellscintillation detector filled with EJ-301 organic liquid (John Caunt Scientific Ltd., UK), opticallycoupledto a Hamamatsu R5611 photomultiplier tube (PMT). The PMT was operated with a negativehigh-tension (HT) supply voltage of –840 V dc. An 8 m length of 50 Ω coaxial cable was used totransport the signal from the PMT to an Infiniium digital oscilloscope (Agilent Tech.). Data wererecorded digitally with a sampling frequency of 4 GHz and 12 bit amplitude resolution. Dataacquisition was automated by driving the oscilloscope remotely via a Transmission ControlProtocol/Internet Protocol (TCP/IP) connection to a personal computer running a bespoke Matlabprogram. The personal computer used for this system was a Dell Latitude D400 laptop.Events were acquired from an americium-beryllium (Am-Be) neutron source at the PhysicsDepartment, University of Liverpool, UK. This is suspended in a moderating water tank and externallyshielded on three sides with 30 mm thick of plate cadmium, encased in sheet plywood. This source canbe moved forwards and backwards to vary the degree of neutron moderation. The source waspositioned 180 mm from the outside edge of the front of the tank; the detector was positioned in linewith the source (360 mm above ground-level) at a distance of 800 mm from the front edge of the tank,as shown in Figure 1. A dataset comprising 5000 events was collected.The Am-Be source has a low neutron/γ ratio and the high proportion of γ rays frombackground and scattering results in a flux consisting mainly of γ-ray events. By including a γ-rayabsorber, a more even spread of neutron and γ-ray events can be obtained. A 50 mm thick lead wall,placed immediately in front of the detector was enough to significantly attenuate the γ-ray flux to aneutron/γ ratio of ~50:50. Presented in Figure 2 are twenty superimposed pulses acquired using theset-up described above.

Figure 1: A schematic diagram of the apparatusused in this research on digital discriminationstudies (distances in mm).Figure 2: A plot of amplitude versus time for 20superimposed ~50:50 neutron/γ-ray pulsesacquired from an Am-Be source using an EJ-301liquid scintillator shielded with a 50 mm thicklead brick wall, at 4 GHz sampling frequency.2.2 Results and discussionPulse gradient analysis (PGA) (D’MELLOW et al. (2007)) is a simple and inherently fast method forthe classification between γ-ray and neutron particle detection in organic liquid scintillators. Incommon with many analogue discrimination methods, it exploits the principle that the scintillationlight function due to a neutron interaction decays more slowly than that for a γ-ray interaction.However, in the digital domain the elegance of the technique arises from the comparison of sampleamplitudes early in a given pulse’s evolution.Evidence of digital PGA is presented in Figure 3, as a scatter plot of peak amplitude versusdiscrimination sample amplitude, where the discrimination sample amplitude is a measure of thedecrease of the pulse amplitude in a given time after the peak. Clearly, two characteristic plumes areobserved. The data were sampled at 250 MHz, 8-bit amplitude resolution and the amplitude of thesample 10 ns after the peak was measured. Figure 4 suggests that the optimum sample to use fordiscrimination lies between 15 ns and 25 ns after the peak of the pulse for the EJ-301 liquidscintillator. In this time window the separation between pulse shapes is greatest. Whilst there is anoptimum discrimination sample for synthetic pulse shapes (D’MELLOW (2006)), it is favourable todiscriminate as early as reasonably possible to minimise the possibility of pile-up events corrupting thesingle pulse amplitudes.Figure 3: A scatter plot of peak amplitudesagainst 10 ns discrimination sample amplitudes.PGA applied to 2436 pulses from ~50:50 n/γ data.Figure 4: A plot of the normalised pulseamplitude against time for an average neutron andγ-ray pulse, average of 1182 n and 1254 γ-rays.

3. Digital time-of-flight3.1 Experimental detailsThe Van de Graaff accelerator at the National Physical Laboratory (NPL) Neutron Irradiation Facilitywas used to produce a 2.924 MeV proton beam which was incident on a thin 60 μg cm –2 lithiumfluoride target. Through the 7 Li(p,n) 7 Be reaction neutrons at 1.225 MeV and 0.745 MeV energies wereproduced corresponding to transitions to the ground and first excited states respectively of the productnucleus. The accelerator was operated in pulse mode, and a capacitive-type detector was positioned inthe beam line at about 1.5 m from the lithium fluoride target in order to detect the arrival of the protonbeam pulse immediately before they collided with the target. This provides the beam pickup signalwhich is used to provide timing information for the time-of-flight measurement and to provideinformation on the proton pulse duration and frequency. The transit time from the pickup to the targetis constant because the proton beam is mono-energetic and the target itself is very thin. Therefore,there is a constant delay between the beam pickup signal and the emission of neutrons from the7 Li(p,n) 7 Be reaction. Thus the beam pickup is used to identify the start, pulse duration and period ofthe subsequent neutron pulses. The proton pulse in this experiment had a period of 400 ns and pulsewidth of 5 ns.The detector set-up consisted of the same scintillation detector described in Section 2. In thiscase, the HT was connected to the PMT via a standard RG58 cable. The output signal from thescintillator was connected to channel 1 of an Infiniium® digital oscilloscope, via a Huber+Suhner SX07262 BD cable. This cable preserved pulse shape information whereas the standard RG58 did not.The beam pickup signal was coupled to the control room from the main bay via a 30 m length ofRG58 cable. The beam pickup signal was then passed through a discriminator and connected tochannel 3 of the digital oscilloscope via a further 4 m length of 50 Ω coaxial cable. The scintillatorwas taped onto the NPL plastic scintillator and was placed 54 mm vertically off-axis and 1613 mmhorizontally from the face of the lithium fluoride target.Scintillator pulses were used to trigger acquisition. Acquisition of the scintillator pulse andcorresponding beam pickup pulse data were automated by driving the oscilloscope remotely via aTCP/IP connection to a personal computer running a bespoke Matlab® program.3.2 A comparison of analogue time pick-off and digital leading-edge triggeringThe plastic scintillation detector in the experimental set-up at NPL has a constant-fractiondiscriminator built into the base of the detector’s photo-multiplier. The signal from this detector isattenuated by a long 50 Ω cable from the experimental area to the control room; therefore, an Ortec436 100 MHz discriminator is used to boost the signal. The output from this discriminator provides atime-to-amplitude converter (TAC) unit with its start pulse. The beam pickup pulse acts as the TACstop pulse. The TAC data are presented in Figure 5(a).The leading-edge triggering method is the simplest time pick-off method. It is based on theidentification of the time that the detector signal crosses a fixed trigger-level. Leading-edge triggeringis a commonly used timing method and can be quite effective provided the dynamic range of the inputpulses is small. In the present context it is the simplest method to replicate digitally in order forcomparison to be made with the established analogue method. Amplitude walk can be a source ofuncertainties with leading-edge timing methods and arises as a result of variations in amplitude andpulse shape. Timing uncertainty resulting from amplitude walk is unacceptable in situations whereaccurate timing is required over a wide amplitude range. Detectors that have variable charge collectiontimes, such as germanium counters, will be subject to amplitude walk irrespective of input pulses ofconstant amplitude.Amplitude walk uncertainties are reduced by setting the trigger-level as low as possible.However, the trigger-level should be in a region of fast rise time to minimise uncertainties due to jitter.An optimum trigger-level is usually between 10% – 20% of the average pulse amplitude. For the TOFspectra presented in Figure 5(b) the trigger-level was set to 20% of the average pulse height.For the TOF data collected, the purpose was to use the TOF information to provide n/γdiscrimination rather than to achieve the best timing. The TOF spectra presented in this work do nothave the best timing resolution possible, but rather demonstrate the importance of digital TOF data.

Some points which should be clarified are, (1) the NPL TOF system is under development workingtowards optimisation of the timing; (2) the Lancaster scintillator may have better timing characteristicsbecause it is smaller than the NPL scintillator and; (3) background scatter has not been subtracted fromeither spectrum. Nevertheless, the significance of Figure 5(b) with respect to Figure 5(a) is a sharperspectral line for the 0.745 MeV neutrons and a lower baseline is achieved.γ 1.225 MeV n 0.745 MeV n γ 1.225 MeV n 0.745 MeV n(a) TOF spectra using an analogue constantfractiontime pick-off method.(b) TOF spectra using a digital leading edge timepick-off method.Figure 5: Time-of-flight (TOF) spectra for the lithium-7 reaction over a flight path of 1.614 m usingthe Neutron Irradiation Facility at the National Physical Laboratory.4. ConclusionsIn this paper we have described two digital methods for the processing of neutron events that arefundamental to advanced PFNA methods of screening in counter-terrorism. The digital capture ofneutron data enables post-processing and optimisation of entire data sets because each complete pulseis stored. In particular, mixed radiation fields can be sorted in terms of their neutron and γ-raycomponents and this discrimination process is carried out in software; therefore the efficiency of themethod is only reliant upon the algorithm used. By replicating analogue timing methods in the digitaldomain it is possible to time-stamp neutron events in real-time. Once again, the opportunity exists topost-process the data to optimise the confidence with which the timing information is held.5. ReferencesD’MELLOW, B. et al., Digital n-γ discrimination in liquid scintillators using pulse gradient analysis,Nuclear Instruments and Methods A (2007), doi:10.1016/j.nima.2007.04.174D'MELLOW, B., Digital Processing in Neutron Spectrometry, PhD Thesis, Lancaster University,2006.DESCOVICH, M. et al., The position response of a large-volume segmented germanium detector,Nuclear Instruments and Methods A 553 (2005) 512–521.6. AcknowledgementsThe authors would like to acknowledge the help of Dr. Tim Sheldon at the Home Office ScientificDevelopment Branch. This research was funding by the Engineering and Physical Sciences ResearchCouncil (EPSRC) via the DISTINGUISH project (http://www.distinguish.org.uk).

MEGAPORTS: DETECTION OF NUCLEAR SMUGGLING INTHE PORT OF ANTWERPP. FIAS, N. BERGANS, S. SCHEURSNucleair Technologisch Centrum (NuTeC), XIOS Hogeschool LimburgAgoralaan, 3590 Diepenbeek – BelgiumT. PEETERSMegaports – MPI, Douane en AccijnzenFrans Tijsmanstunnel, 2040 Antwerpen - BelgiumABSTRACTSince the beginning of 2007 Belgian customs performs nuclear inspections with the widelyused radiation portals. The detection of nuclear smuggling with these plastic scintillatordetectors is not as simple as it seems. In this paper we will present the results of a 6 monthmeasurement period of container traffic in the port of Antwerp. The main conclusion is thatalmost all alarms on radiation portals are due to natural occurring radioactive materials(NORM). The combination of a scanning device and radiation portals in the inspection isan important aid to distinct between NORM alarms and the few alarms that are due toradiological contaminations or artificial radioactive sources.1. IntroductionThe Megaports Initiative is a worldwide effort to prevent nuclear smuggling in sea containers and wasstarted after 9/11. In the port of Antwerp the installation of a nuclear detection system is co-funded bythe American and Belgian governments. Starting from 2007 Belgian customs inspects container trafficon nuclear smuggling in the port of Antwerp, the third biggest port in the world. The container trafficis permanently inspected with radiation portals. The portals are equipped with 4 helium tubes forneutron detection and 4 plastic scintillators for gamma detection.Belgian customs officers make a decision on releasing or holding a container after it alarmed on aradiation portal. This is not an easy or straightforward task and therefore Belgian customs askedNuTeC to be the scientific partner in the inspections during the first year.In 2006 NuTeC performed a survey of the container traffic with a radiation portal. Thesemeasurements were needed to estimate the effect of the inspections on the workings of the port and tohave a first glimpse of the kind of radioactive materials that are shipped through the port of Antwerp.The survey was performed at the scan site of Antwerp customs. At this site containers are scanned forsmuggling. The container traffic at the scan site is selected from all over the port and therefore can beestimated as a site with an average container traffic. In figure 1 the set-up for the survey is presented.It consisted of two radiation portals and one x-ray scan facility or scan tunnel (2 linear accelerators).

Fig 1. Set-up for the screening of container trafficScan tunnelLinearRadiation portal2. Results and discussionMeasurements were carried out during a 6 month period in 2006-2007. In general the results are thefollowing:Number of scanned containers 9145Number of alarming containers 624 (6,8%)Number of containers with radiation levels above legalinspection limit27 (0,3%)Number of measurements with handheld equipment 37 (0,4%)Number of containers opened for physical inspection 3 (>0,1%)Number of contaminated materials found 1Number of unlicensed radioactive sources found 0Tab 1: General resultsPortals give an alarm when a radioactive load passes through them. During the survey almost allalarms were caused by natural occurring radioactive materials (NORM). All measured activities werebelow exemption levels for radioactive materials as stated in the Belgian and European legislation.Therefore no containers were blocked during the survey period.The materials causing alarms can be quite divers, the reason for the radioactive nature of thesematerials is always the same: NORM. Only one case of a Co-60 contamination was detected in aperiod of 6 months.Type of material% of alarmsCeramics 38Stone 16Biological materials 7CRT (televisions) 6Ores 5Kitty Litter 2TENORM 3Glass fibre and glass 4Chemical products 5Others 13Unknown 2Tab2. Type materials causing alarms during the survey period

Alarms on radiation portals do not give information about the nature of the alarm. To distinct betweenNORM, contaminated materials, radioactive sources or nuclear smuggling measurements withhandheld equipment are a possible technique.During the survey very few inspections with handheld equipment were needed due to the presence ofthe x-ray scanner. The scanner allowed a comparison between the radiation alarm profile and thecontents of the container. In almost all cases alarms could be sufficiently explained by this technique.For example if a container causes an alarm on a radiation portal and this is due to the presence ofceramics (NORM), the scan image clearly shows this.Information about the contents of the container is also needed to verify if the radiation has a naturalsource. For example if steel rods cause an alarm this is not due to natural radiation, but a result of thepresence of radioactive sources during the recycling of scrap metal.3. ConclusionIn conclusion we can state that techniques and working protocols for nuclear inspections should beselected so that a distinction between NORM, contaminated materials, radioactive sources and nuclearsmuggling is easy and straightforward. For example the combination of a radiation portal and a X-rayscanner is a strong and efficient technique. Furthermore information about the contents of containersshould be easily accessible and reliable.

European Nuclear SocietyRue de la Loi 571040 Brussels, BelgiumTelephone +32 2 505 3 50Fax + 32 2 502 39 02ens@euronuclear.orgwww.euronuclear.orgLayout and Design: Marion Brünglinghaus, ENS