atw - International Journal for Nuclear Power | 11/12.2020

inforum

Ever since its first issue in 1956, the atw – International Journal for Nuclear Power has been a publisher of specialist articles, background reports, interviews and news about developments and trends from all important sectors of nuclear energy, nuclear technology and the energy industry. Internationally current and competent, the professional journal atw is a valuable source of information.

www.nucmag.com

nucmag.com

2020

11/12

ISSN · 1431-5254

24.– €

An Old Promise of

Physics – Are We Moving

Closer Toward Controlled

Nuclear Fusion?

Highlights of the World Nuclear

Performance Report 2020

The EMPIrE Irradiation Test:

Lower- Enriched Fuel for High-

Performance Research Reactors


atw Vol. 65 (2020) | Issue 11/12 ı November/December

Infinite Energy

One of the great challenges for mankind has always been the availability of

sufficient energy. A glance at the sky, day or night, obviously shows which path

to take. The sun and the stars point the way, that of nuclear fusion. In addition

to nuclear fission, nuclear fusion is the second way to obtain energy directly

from the core of our matter. While in nuclear fission energy is released by

fissioning a heavy nucleus, in nuclear fusion this is done by combining two

light atomic nuclei. Under the conditions that are currently technically

possible, the two hydrogen isotopes deuterium and tritium merge most easily

to form a helium nucleus, a free neutron and released binding energy –

following E = mc 2 . This fusion reaction releases 17.58 MeV “microscopically”

at the level of the nuclei, and “macroscopically” when fused to 1 kg of helium it

releases around 120 million kWh corresponding to the energy released when

12 million kilograms of hard coal are burned. However, nature has set a hurdle

to this potential: To fuse the nuclei, a high speed is required at which they fly

towards each other, ultimately equal to around 100 million degrees Celsius, a

temperature at which atoms are present as plasma broken down into electrons

and nuclei. Since the plasma is electrically conductive, this property can be

exploited because a magnetic field is able to confine the plasma without

contact. Otherwise, the dream of a controlled sun on earth would come to an

end at this point, as no material can withstand even approximately such

temperatures.

The main objective of research and development in the field of nuclear

fusion is to develop suitable processes and systems that ultimately lead to a

controlled fusion reaction in the form of a self-sustaining reaction with usable

energy output.

However, while the physics of nuclear fusion was observed even earlier than

that of nuclear fission – which today supplies some 11 % of the electricity

generated worldwide – the technical path to its exploitation is a long and

complex one.

ITER (English for International Thermonuclear Experimental Reactor; Latin

for “the way”) is the inter national project that combines the joint efforts of the

participating countries to advance controlled nuclear fusion. The reactor is

based on the Tokamak principle and has been under construction near the

Cadarache nuclear research centre in the south of France since 2007. According

to current planning, the plant is scheduled to generate a hydrogen plasma for

the first time in December 2025. Experiments with tritium are to be carried out

from around 2035 onwards, pointing the way forward. If the results of ITER

and the accompanying research and development, including other fusion

experiments, show that the tokamak principle is promising, the successor

project DEMO will be designed as a fully operational power plant beyond 2050.

ITER is on the one hand a mammoth technical project; here are some

figures: The superconducting coils of the magnets for plasma confinement

contain niobium-tin strands with a length of 100,000 kilometres, the plasma of

ITER is to reach a peak temperature of 150 million °C, the ITER machine itself

weighs 23,000 tonnes – three times as much as the Eiffel Tower.

However, another – socio-political – aspect is even more important today:

ITER impressively demonstrates as an outstanding project worldwide how the

seven partners, China, India, Japan, the Republic of Korea, Russia, the USA and

the European Union (for its 27 member states), are jointly launching a major

project on its way with many partial successes to date.

535

EDITORIAL

Christopher Weßelmann

– Editor in Chief –

Editorial

Infinite Energy


atw Vol. 65 (2020) | Issue 11/12 ı November/December

EDITORIAL 536

Unendlich viel Energie

Eine der großen Herausforderungen für die Menschheit ist seit jeher die

Verfügbarkeit von ausreichend Energie. Ein Blick in den Himmel, bei Tag oder

Nacht, zeigt offensichtlich, welcher Weg zu beschreiten wäre. Denn die Sonne

und die Vielzahl von Sternen weisen einen Weg, den der Kernfusion. Neben

der Kernspaltung ist die Kernfusion die zweite Möglichkeit, Energie direkt aus

den „Kernbausteinen“ unserer Materie zu gewinnen. Während bei der

Kernspaltung durch Trennen eines schweren Kerns Energie frei wird, erfolgt

dies bei der Kernfusion durch Verschmelzen zweier leichter Atomkerne. Unter

den Bedingungen, die derzeit technisch möglich sind, verschmelzen am

leichtesten die beiden Wasserstoffisotope Deuterium und Tritium zu einem

Helium-Kern, einem freien Neutron sowie frei werdender Bindungsenergie –

E = mc 2 folgend. Diese Fusionsreaktion setzt „mikroskopisch“ auf Ebene der

Kerne 17,58 MeV frei, „makroskopisch“ sind es bei Fusion zu 1 kg Helium rund

120 Millionen kWh entsprechend der Energie, die beim Verbrennen von

12 Millionen Kilogramm Steinkohle frei wird. Doch die Natur setzt diesem

Potenzial eine Hürde entgegen: Zur Verschmelzung der Kerne ist eine hohe

Geschwindigkeit erforderlich, mit der diese aufeinander zufliegen, letztendlich

rund 100 Mio. Grad °C, einer Temperatur, bei der Atome als Plasma in

Elektronen und Kerne zerlegt vorliegen. Da das Plasma elektrisch leitend ist,

lässt sich diese Eigenschaft zunutze machen, denn ein Magnetfeld ist in der

Lage, das Plasma berührungslos einzuschließen. Sonst wäre der Traum von

der kontrollierten Sonne auf Erden an diesem Punkt zu Ende, da kein Material

auch nur annähernd solchen Temperaturen standhalten kann.

Das Hauptziel von Forschung und Entwicklung auf dem Gebiet der

Kernfusion sind geeignete Verfahren und Systeme, die letztendlich zu einem

kontrollierten Ablauf der Fusionsreaktion in Form einer sich selbst aufrecht

erhaltenden Reaktion mit nutzbarer Energieabgabe führen.

Doch während die Physik der Kernfusion sogar früher beobachtet wurde als

die der Kernspaltung – die heute immerhin rund 11 % des weltweit erzeugten,

Stroms liefert –, ist der technische Weg zur Nutzbarmachung ein langer und

aufwändiger.

ITER (englisch für International Thermonuclear Experimental Reactor;

lateinisch der Weg) ist dabei das internationale Projekt, in dem die gemeinsamen

Anstrengungen der beteiligten Staaten gebündelt sind, um die kontrollierte

Kernfusion voran zu bringen. Der Reaktor beruht auf dem Tokamak-Prinzip

und ist seit 2007 nahe dem südfranzösischen Kernforschungszentrum

Cadarache in Bau. Nach aktueller Planung soll in der Anlage erstmals im

Dezember 2025 ein Wasserstoffplasma erzeugt werden. Etwa ab dem Jahr

2035 sollen Experimente mit Tritium durchgeführt werden, die den weiteren

Weg weisen. Falls die Ergebnisse von ITER sowie den begleitenden Forschungen

und Entwicklungen, auch in anderen Fusionsexperimenten, zeigen, dass das

Tokamak-Prinzip erfolgversprechend ist, soll das Nachfolgeprojekt DEMO als

voll funktionsfähiges Kraftwerk jenseits des Jahres 2050 konzipiert werden.

ITER ist einerseits ein technisches Mammutprojekt; einige Zahlen dazu:

Die supraleitenden Spulen der Magneten zum Plasmaeinschluss enthalten

Niob-Zinn-Drähte mit einer Länge von 100.000 Kilometern, das Plasma des

ITER soll in der Spitze eine Temparatur von 150 Mio. °C erreichen,

die ITER-Maschine selbst wiegt 23.000 Tonnen – dreimal soviel wie der

Eiffelturm.

Aber gerade in der heutigen Zeit ist ein anderer – gesellschaftspolitischer –

Aspekt noch viel wesentlicher: ITER demonstriert eindrucksvoll als herausragendes

Projekt weltweit, wie die sieben Partner, China, Indien, Japan, die

Republik Korea, Russland, die USA und die Europäische Union (für ihre

27 Mitgliedstaaten), ein Großprojekt mit bis heute vielen Teilerfolgen

gemeinsam auf den Weg bringen.

Christopher Weßelmann

– Chefredakteur –

Editorial

Infinite energy


Kommunikation und

Training für Kerntechnik

Suchen Sie die passende Weiter bildungs maßnahme im Bereich Kerntechnik?

Wählen Sie aus folgenden Themen: Dozent/in Termin/e Ort

3 Atom-, Vertrags- und Exportrecht

WEBINAR: Dual-Use-Reform und Exportkontrolle kerntechnischer

Produkte und Dienstleistungen –

Was ist zu beachten?

WEBINAR: Atomrecht – Ihr Weg durch Genehmigungs- und

Aufsichtsverfahren

RA Kay Höft M.A. (BWL) 10.12.2020

08.06.2021

RA Dr. Christian Raetzke 19.01. - 20.01.2021

Atomrecht – Das Recht der radioaktiven Reststoffe und Abfälle RA Dr. Christian Raetzke 20.04.2021 Berlin

3 Kern- und Entsorgungstechnik

Grundlagenschulung

Einführung in die Kern- und Entsorgungstechnik

Dr. Thomas Behringer 29.04. - 30.04.2021 Berlin

3 Kommunikation und Politik

WEBINAR: Medien- und Interviewtraining Dr. Nikolai A. Behr 03.03.2021

Public Hearing Workshop –

Öffentliche Anhörungen erfolgreich meistern

Dr. Nikolai A. Behr

Termin

auf Anfrage

Berlin

3 Rückbau und Strahlenschutz

In Kooperation mit dem TÜV SÜD Energietechnik GmbH Baden-Württemberg:

3 Nuclear English

Das Strahlenschutzrecht und

seine praktische Umsetzung

Stilllegung und Rückbau in Recht und Praxis

Dr. Maria Poetsch

RA Dr. Christian Raetzke

Dr. Stefan Kirsch

RA Dr. Christian Raetzke

17.03. - 18.03.2021

22.09. - 23.09.2021

Berlin

28.09. - 29.09.2021 Berlin

English for the Nuclear Industry Angela Lloyd 16.03. - 17.03.2021 Berlin

3 Wissenstransfer und Veränderungsmanagement

Veränderungsprozesse gestalten –

Herausforderungen meistern, Beteiligte gewinnen

Erfolgreicher Wissenstransfer in der Kerntechnik –

Methoden und praktische Anwendung

Dr. Tanja-Vera Herking

Dr. Christien Zedler

Dr. Tanja-Vera Herking

Dr. Christien Zedler

03.03. - 04.03.2021 Berlin

24.03. - 25.03.2021 Berlin

Haben wir Ihr Interesse geweckt? 3 Rufen Sie uns an: +49 30 498555-30

www.kernd.de/kernd/seminare

Kontakt

INFORUM Verlags- und Verwaltungs gesellschaft mbH ı Robert-Koch-Platz 4 ı 10115 Berlin

Petra Dinter-Tumtzak ı Fon +49 30 498555-30 ı Fax +49 30 498555-18 ı Seminare@KernD.de

Die INFORUM-Seminare können je nach

Inhalt ggf. als Beitrag zur Aktualisierung

der Fachkunde geeignet sein.


atw Vol. 65 (2020) | Issue 11/12 ı November/December

Contents

538

CONTENTS

Issue 11/12

2020

November/December

Editorial

Infinite Energy E/G 535

Inside Nuclear with NucNet

Explained: Why Africa is Looking Towards Nuclear Energy

and the Obstacles it Faces 540

Did you know...? . . . . . . . . . . . . . . . . . . . . . . . . . . . 541

Feature | Major Trends in Energy Policy and Nuclear Power

An Old Promise of Physics –

Are We Moving Closer Toward Controlled Nuclear Fusion? 543

Lars Jaeger

Cover:

ASDEX Upgrade during revision

(Courtesy IPP/Bernhard Ludewig).

Serial | Major Trends in Energy Policy and Nuclear Power

Highlights of the World Nuclear Performance Report 2020 551

Jonathan Cobb

Spotlight on Nuclear Law

Export of Non-irradiated Fuel Elements – Indicator for

Fair and Functioning Foreign Trade in Germany G 555

Tobias Leidinger

Energy Policy, Economy and Law

BioKernSprit 557

Jochen K. Michels

Decommissioning and Waste Management

The Scientific Backing of the German Quiver Project 561

Wolfgang Faber, Marc Verwerft, Janne Pakarinen, Hagen Höfer and Christoph Rirschl

Fuel

The EMPIrE Irradiation Test: Lower- Enriched Fuel

for High- Performance Research Reactors 571

Bruno Baumeister, Christian Reiter and Winfried Petry

Research and Innovation

From Fission to Fusion – Transfer of Existing

Industrial Know-How to New Domains of Applications 575

Sabrina Gil Pascual and Andreas Bender

Simulation of the DEBRIS Experiments

with the Severe Accident Analysis Code ASTEC 578

Jan M. Peschel and Marco K. Koch

Analysis of the Melt Behaviour in the Lower Plenum

of the TMI-2 Reactor Using the System Code AC² – ATHLET-CD 583

Florian Krist and Marco K. Koch

Computational Heat Transfer Analysis of Tubes and

Tube Bundles with Supercritical Water as Coolant 588

Kashif Tehseen, Kamran Rasheed Qureshi, M. Abdul Basit, Rab Nawaz, Waseem Siddique and Rustam Khan

Transient Thermal-hydraulic Analysis of a Scaled Down

Test Loop for the VVER-1000 Reactor using RELAP5 Code 595

Babak Khonsha, Gholamreza Jahanfarnia, Kamran Sepanloo and Mohammadreza Nematollahi

KTG Inside . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 603

News . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 605

Nuclear Today

Japan’s Carbon-Free Policy Goal Signals New Day

in the Sun for Nuclear Energy 610

John Shepherd

G

E/G

= German

= English/German

Imprint 594

Insert: Inforum – Seminarprogramm 2021

Contents


atw Vol. 65 (2020) | Issue 11/12 ı November/December

Feature

Major Trends in Energy Policy

and Nuclear Power

539

CONTENTS

543 An Old Promise of Physics – Are We Moving Closer

Toward Controlled Nuclear Fusion?

Lars Jaeger

Serial | Major Trends in Energy Policy and Nuclear Power

551 Highlights of the World Nuclear Performance Report 2020

Jonathan Cobb

Energy Policy, Economy and Law

557 BioKernSprit

Jochen K. Michels

Decommissioning and Waste Management

561 The Scientific Backing of the German Quiver Project

Wolfgang Faber, Marc Verwerft, Janne Pakarinen, Hagen Höfer and Christoph Rirschl

Fuel

571 The EMPIrE Irradiation Test:

Lower- Enriched Fuel for High- Performance Research Reactors

Bruno Baumeister, Christian Reiter and Winfried Petry

Research and Innovation

575 From Fission to Fusion – Transfer of Existing Industrial Know-How

to New Domains of Applications

Sabrina Gil Pascual and Andreas Bender

Contents


atw Vol. 65 (2020) | Issue 11/12 ı November/December

540

INSIDE NUCLEAR WITH NUCNET

Explained: Why Africa is Looking Towards

Nuclear Energy and the Obstacles it Faces

The continent’s lack of access to electricity presents a barrier to economic and social development.

Baseload sources are needed.

Why does Africa need nuclear energy? In short, for economic

reasons. The continent’s inability to generate enough electricity

continues to hamper growth, cutting 2 to 4 % off GDP

every year, according to the Africa Progress Panel. The panel

estimates that some 600 million people on the continent of

about 1.3 billion do not have access to electricity, a figure

that will require $55bn per year in investment by 2030 to fix.

The International Atomic Energy Agency says that in

sub-Saharan Africa, only about a third of the population have

access to electricity and the number of people without access

is on the rise. This presents a significant barrier to economic

and social development and so governments across the

continent are seeking ways to improve their existing energy

infrastructure and develop new or diverse energy sources that

are reliable, affordable and sustainable.

Kenya’s reasons for considering nuclear are typical. It

needs baseload electricity to meet the demand generated by

connecting households to the grid nationwide, which is

expected to contribute significantly to the 30 % increase in

electricity demand predicted for the country by 2030.

“For a long time in our country electrification levels were

low, but the government has put in a lot of efforts towards electrifying

the entire country,” Mr Winfred Ndubai, acting director

of the Kenya Nuclear Electricity Board’s Technical Department,

told the IAEA. “Even those areas that were considered to

be remote are now vibrant. Within a period of about 10 years

we have moved from a 12 % electrification rate to 60 %.”

Nuclear In Africa Today

South Africa is the only country on the continent operating a

nuclear power station, with two units at Koeberg providing

almost 7 % of its electricity, from 1.94 GW of gross installed

capacity. This compares to 48.7 GW in China, which has a

similar population to the whole of Africa.

According to Ms Névine Schepers, a research associate

with the non-proliferation and nuclear policy programme at

the International Institute for Strategic Studies, Russia is at

various stages of negotiating nuclear cooperation agreements

with at least 16 countries in Africa. These discussions

range from projects still in their infancy ( Angola and

Ethiopia), to more fleshed-out agreements (Nigeria and

Sudan), to a fully fledged contract with Egypt for the construction

of four reactors at El Dabaa where the first plant is

scheduled to begin commercial operation in 2026.

Although many of these projects are still years or even

decades away from becoming a reality, and others will likely

remain dormant forever, some are slowly materialising. China

also has its eyes on Africa and has signed nuclear cooperation

agreements with African countries including Ghana, Algeria,

Niger, Morocco, Egypt, Tunisia, Sudan and South Africa.

Ms Névine Schepers says Russia’s incentives and reluctance

to impose additional requirements have given it a competitive

edge. By providing generous state-backed loans,

nuclear suppliers such as Russia, and to a certain extent China,

are changing the financial requirements for nuclear hopefuls.

In Egypt, Russia will be providing 85 % of the funding for

the $21bn nuclear station and will service the four reactors

for 60 years. Russian-built nuclear power reactors contracted

around the world benefit from Russian loans varying

from 49 to 90 % of the total project cost. In Turkey, Rosatom

is experimenting with a new type of contract known as

‘Build-Own-Operate’, whereby it maintains ownership of the

plant and makes a profit by selling electricity to the host

country. Western observers have criticised these Kremlinbacked

incentives, describing them as a means for Russia to

create energy dependencies (possibly mirroring its use of

gas supply as a bargaining chip in Europe) and establish a

semi-permanent foothold in South Asia, the Middle East and

Africa.

Russia complies with IAEA requirements and follows

Nuclear Suppliers Group guidelines, which regulate nuclear

trade. However, it does not seek further assurances from

partners concerning uranium enrichment and plutonium

reprocessing technologies. Many developing economies are

wary of additional conditions put on their nuclear ambitions,

which is sometimes seen as a restriction on what they con sider

to be a sovereign right to nuclear technology for peaceful

uses. Russia’s disinclination to require more guarantees puts

it in a better negotiating position than for instance the US,

which has sought to require recipients to forgo these sensitive

dual-use technologies.

Finally, Russia is one of few suppliers that takes back spent

nuclear fuel from foreign reactors. For African countries,

many of which do not possess permanent waste-storage infrastructure,

this provides yet another incentive to turn to Russia.

Obstacles To Nuclear Energy In Africa

Without proper financing, nuclear is not an option for Africa,

says the IAEA. But there are other obstacles to overcome, such

as the burden on a country’s electrical grid system. For a

country to safely introduce nuclear energy, the IAEA recommends

that its grid capacity be around 10 times the capacity

of its planned nuclear power plant. For example, a country

should have a capacity of 10,000 MW already in place to

generate 1,000 MW from nuclear power.

Few countries in Africa currently have a grid of this

capacity. “In Kenya, our installed capacity is 2,400 MW – too

small for conventional, large nuclear power plants,” says

Mr Winfred Ndubai, acting director of the Kenya Nuclear

Electricity Board’s Technical Department. “The grid would

need to increase to accommodate a large unit, or,

alternatively, other, smaller nuclear power plant options

would need to be explored.”

One option is to invest in small modular reactors (SMR),

which are among the most promising emerging technologies

in nuclear power. SMRs produce electric power up to

300 MW per unit, or around half of a traditional reactor, and

their major components can be manufactured in a factory

setting and transported to sites for ease of construction.

The IAEA says SMRs are expected to begin commercial

operation in Argentina, China and Russia in the coming

decade, but African countries are still wary.

“One of the things we are very clear about in terms of introducing

nuclear power is that we do not want to invest in a firstof-a-kind

technology,” Ms Ndubai said. “As much as SMRs

represent an opportunity for us, we would want them to be

built and tested elsewhere before introducing them here.”

Author

NucNet – The Independent Global Nuclear News Agency

Editor responsible for this story: David Dalton

Avenue des Arts 56 2/C

1000 Bruxelles

www.nucnet.org

Inside Nuclear with NucNet

Explained: Why Africa is Looking Towards Nuclear Energy and the Obstacles it Faces


atw Vol. 65 (2020) | Issue 11/12 ı November/December

Did you know...?

Nuclear, Renewables and the Levelised-Cost-Concept on the System Level

In a very recent publication, “The Failings of Levelised Cost and

the Importance of System-Level Analysis”, of October 2020 the

New Nuclear Watch Institute, an industry supported think-tank,

based in the UK and focused on the international development of

nuclear energy, analysed the limitations of the levelised cost of

electricity (LCOE) approach to comparing different types of

electricity generation and addressing system-level issues that are

important for policy decisions. According to the paper, while the

LCOE financial metric that gives the average price per unit of

generation required to balance revenues and costs of a generation

project is useful in comparing different technologies from the

perspective of an investor or operator at project level it is not

appropriate to analyse the value of technologies for the electricity

system as a whole. The failure of LCOE to reflect the complexity of

modern electricity generation systems with the interactions

between different technologies makes it an ill-advised tool for

energy policy decisions despite the fact that LCOE recently has

been more and more often used for the purpose of a general

evaluation of different technologies.

A reason for the popularity of LCOE is its simplicity by

integration of a number of cost variables into one single figure

that is easily communicated and seemingly easily understood.

This integrative function on the other hand often obscures the

inputs that went into the calculation and the categorial differences

of these inputs. One of the symptoms of this effect can be the

wide variety of LCOEs for the same technology that can be

encountered. The graph below shows the impact of different

levels of discount rates (5 % and 10 %) as input on the calculated

LCOE of different technologies. Other parameters like variable

production costs are affected by uncertainties because the data is

not freely available and the capacity factors of volatile renewable

energy (VRE) sources vary widely with geographic location. The

latter aspect of course is not a problem of the LCOE method but a

necessary input for the calculation and also generally transparent.

Nevertheless it greatly reduces the universality of LCOE figures

and enables comparison between VRE and other generation

technologies only on a regional scale.

The other major problem is that LCOE cannot capture e.g. the

important difference between intermittent and dispatchable generation

either on the system level or concerning the economic

disadvantage of VRE to not be able to produce electricity according

to demand and to the corresponding price level which decreases

their economic value. With the increasing penetration of intermittent

technologies of the same type also the correlation of the

The Impact on Levelised Cost of Increasing the Discount Rate

from 5 % to 10 % in 2010 $/MWh LCOE

Onshore Wind

61

weather induced production maxima and minima increases. This

means that the value of produced electricity is low in production

maxima and decreasing with a higher penetration of VRE in the

system and the costs of compensating for low production passed

on to the system (grid operators, other producers) and ultimately

customers is high and increasing with penetration. These system

costs estimated by different studies in a range of 10 to 80 % of the

LCOE of VRE are composed of a direct and an indirect component.

The direct costs are first balancing costs to adapt to varying VRE

production levels and prognostic uncertainties by redispatching

dispatchable power plants and maintaining an operational

reserve and second the costs of net infrastructure expansion and

adaptation to higher peak production figures and the often

unfavourable geographic distribution of VRE following the availability

of natural production resources rather than the load

centres. The indirect costs are composed by the decreasing of the

system wide capacity factor and the capacity factors of dispatchable

generation sources with increasing penetration of VRE, the

low capacity credit, i. e. low reliability necessitating back-up

capacity and the increasing likelihood of phases of overproduction

that in turn will have to be curtailed which decreases the capacity

factor of VRE again aggravated by increasing penetration.

Next to this elaboration on the limitations and blind spots of

the LCOE method particularly concerning system level policy

decisions other findings are presented. A statistical analysis of the

electricity production and its composition linked with the respective

carbon intensity of the electricity sectors for 23 European nations

from 2000 to 2018 showed that an increase in nuclear generation

leads to a 24 % greater decarbonization than an equal increase of

VRE generation. The analysis also shows that an increasing share

of gas can also decrease carbon intensity though to a considerably

lower degree than both VRE and nuclear. This effect though

diminishes considerably with a higher penetration of VRE in the

respective system when gas is more used to provide back-up for

VRE than for reducing coal generation. According to the study this

has an impact on the transition technology narrative associated

with gas power generation since VRE to the difference of nuclear

power plants cannot replace gas fired plants.

The report concludes that the LCOE method overvalues

intermittent generation technologies due to its narrow focus

and its blindness to interdependencies between generation

technologies as well as the effects on the power system as a

whole relative to energy technologies that also offer benefits at

the level of the system, particularly nuclear power.

78

p LCOE at Discount Rate 5 %

p Additional LCOE at Disocunt Rate 10 %

Source: The Failings of

Levelised Cost and the

Importance of System-

Level Analysis, The

New Nuclear Watch

Institute (NNWI),

October 2020

DID YOU EDITORIAL KNOW...?

541

Solar PV (Utility)

71

79

For further details

please contact:

Nuclear

CC Gas

Coal

0

59

65

99

110

84

100 200

160

Nicolas Wendler

KernD

Robert-Koch-Platz 4

10115 Berlin

Germany

E-mail: presse@

KernD.de

www.KernD.de

Did you know...?


atw Vol. 65 (2020) | Issue 11/12 ı November/December

CALENDAR 542

Calendar

2021

24.02. – 25.02.2021

8 th Nuclear Decommissioning & Waste

Management Summit. London, UK, Active

Communications Europe, www.wplgroup.com

Postponed to 31.05. – 04.06.2021

20 th WCNDT – World Conference on

Non-Destructive Testing. Incheon, Korea,

The Korean Society of Nondestructive Testing,

www.wcndt2020.com

This is not a full list. Dates are subject to change.

Please check the listed websites for updates.

Postponed to 30.08. – 03.09.2021

International Conference on Operational Safety

of Nuclear Power Plants. Beijing, China, IAEA,

www.iaea.org

Postponed to 08.09. – 10.09.2021

3 rd International Conference on Concrete

Sustainability. Prague, Czech Republic, fib,

www.fibiccs.org

Virtual Meeting 25.02.2021

International Power Summit 2021.

Arena International Events Group,

www.arena-international.com

03.03. – 04.03.2021

Maintenance in Power Plants 2021. Karlsruhe,

Germany, VGB PowerTech e.V., www.vgb.org

Virtual Meeting 08.03.2021

WM2021 – Waste Management Symposia.

X-CD Technologies, www.wmsym.org

16.03. – 18.03.2021

EURAD 1 st Annual Event. www.ejp-eurad.eu

Virtual Meeting 23.03. – 26.03.2021

7 th International Conference on Education and

Training in Radiation Protection.

FuseNet, www.etrap.net

30.03. – 01.04.2021

PowerGen International. Orlando, Florida, USA,

Clarion Events, www.powergen.com

13.04. – 15.04.2021

World Nuclear Fuel Cycle. Stockholm, Sweden,

WNA World Nuclear Association,

www.wnfc-event.com

26.04. – 27.04.2021

AtomExpo 2021. Sochi, Russia, Rosatom,

http://2021.atomexpo.ru/en/

03.05. – 07.05.2021

ATALANTE 202(0)1. Nimes, France, CEA + Geniors,

www.atalante2020.org

04.05.2021

2021 KTG Annual Meeting. Berlin, Germany, KTG,

www.ktg.org

Postponed to 10.05. – 15.05.2021

FEC 2020 – 28 th IAEA Fusion Energy Conference.

Nice, France, IAEA, www.iaea.org

18.05. – 20.05.2021

Power Uzbekistan 2021 – 15 th Anniversary

International Exhibition on Energy.

Tashkent, Uzbekistan, Iteca Exhibitions,

www.power-uzbekistan.uz

Postponed to 30.05. – 05.06.2021

BEPU2020 – Best Estimate Plus Uncertainty International

Conference, Giardini Naxos. Sicily, Italy,

NINE, www.nineeng.com

01.06. – 02.06.2021

Nuclear Power Plants IV. Expo & VIII. Summit

(NPPES). Istanbul, Turkey, INPPES Expo,

www.nuclearpowerplantsexpo.com

Postponed to 02.06. – 04.06.2021

HTR2020 – 10 th International Conference

on High Temperature Reactor Technology.

Yogyakarta, Indonesia, Indonesian Nuclear Society,

www.htr2020.org

04.08. – 06.08.2021

ICONE 28 28 th International Conference on

Nuclear Engineering. Nuclear Energy the Future

Zero Carbon Power. Virtual Conference, Online,

ASME, https://event.asme.org

06.06. – 10.06.2021

TopFuel 2021. Santander, Spain, ENS,

www.euronuclear.org

07.06. – 11.06.2021

International Conference on Geological

Repositories. Helsinki, Finland, EURAD,

www.ejp-eurad.eu

09.06. – 11.06.2021

NUWCEM 2021 – International Symposium on

Cement-Based Materials for Nuclear Wastes.

Avignon, France, Sfen, www.sfen-nuwcem2021.org

Postponed to June 2021

International Forum on Enhancing a Sustainable

Nuclear Supply Chain. Helsinki, Finland, Foratom,

https://events.foratom.org/mstf2020/

Postponed to June/July 2021

WNU Summer Institute. Fukui & Fukushima, Japan,

World Nuclear University,

www.world-nuclear-university.org

01.07. – 11.07.2021

4 th CORDEL Regional Workshop – Harmonization

to support the operation and new build of NPPs

including SMR. Lyon, France, World Nuclear

Association, www.events.foratom.org

25.08. – 27.08.2021

KONTEC 2021 – 15 th International Symposium

“Conditioning of Radioactive Operational &

Decommissioning Wastes”. Dresden, Germany,

atm, www.kontec-symposium.de

25.08. – 03.09.2021

The Frédéric Joliot/Otto Hahn Summer School

on Nuclear Reactors “Physics, Fuels and Systems”.

Aix-en-Provence, France, CEA & KIT, www.fjohss.eu

08.09. – 10.09.2021

World Nuclear Association Symposium 2021.

London, UK, WNA, www.wna-symposium.org

22.09. – 23.09.2021

VGB Congress 100 PLUS. Essen, Germany, VGB

PowerTech, www.vgb.org

23.09. – 27.09.2021

European Nuclear Young Generation Forum

(ENYGF). Tarragona, Spain, ENYGF, www.enygf.org

27.09. – 01.10.2021

NPC 2021 International Conference on Nuclear

Plant Chemistry. Antibes, France, SFEN Société

Française d’Energie Nucléaire,

www.sfen-npc2021.org

26.10. – 28.10.2021

VGB Conference Chemistry. Ulm, Germany, VGB

PowerTech, www.vgb.org

01.11. – 12.11.2021

COP26 – UN Climate Change Conference.

Glascow, Scotland, www.ukcop26.org

Postponed to 30.11. – 02.12.2021

Enlit (former European Utility Week and

POWERGEN Europe). Milano, Italy,

www.enlit-europe.com

30.11. – 02.12.2021

WNE2021 – World Nuclear Exhibition. Paris,

France, Gifen, www.world-nuclear-exhibition.com

Postponed to 2021

INDEX 2020: International Nuclear Digital

Experience. Paris, France, SFEN,

www.sfen-index2020.org

2022

29.03. – 30.03.2022

KERNTECHNIK 2022.

Leipzig, Germany, KernD and KTG,

www.kerntechnik.com

Calendar


atw Vol. 65 (2020) | Issue 11/12 ı November/December

An Old Promise of Physics – Are We Moving

Closer Toward Controlled Nuclear Fusion?

Lars Jaeger

“Fridays for Future” demonstrations are invigorating the masses, the European Union launches a “European

Green Deal” in which it commits to reduce net greenhouse gas emissions to zero by 2050, the new US president aims the

same for the United States, China follows by announcing a carbon free economy by 2060, Germany decides on a

multi-billion climate package, and at global climate summits government representatives and CEOs of multinational

corporations and their PR strategists are trying to make a name for themselves as well-meaning climate protectors.

It appears that the question of our future climate and energy production has finally reached the center of public attention

and debate.

And it is precisely at this time that scientists, without a

great deal of public attention, are making progress in an

area that could solve the problems of global energy supply

once and for all: the peaceful use of nuclear fusion. This is

about nothing less than fulfilling the dream of unlimited,

clean, and safe energy from the thermonuclear fusion of

atomic nuclei, the very same that supplies our sun and the

stars with seemingly endless amounts of energy.

The light of stars

Nuclear fusion research comes with a more than 80-year

history 1 . Since the 1930s, physicists have known that

under very high pressure and temperatures hydrogen

nuclei fuse into helium nuclei. This is the very mechanism

that enables the sun to generate its massive amounts of

energy. In 1938, Carl Friedrich von Weizsäcker and Hans

Bethe developed a first model for the nuclear reaction

occurring inside stars. According to their model, the fusion

of light atom nuclei, just like the splitting of very heavy

nuclei, releases a significant amount of energy. The reason

for this energy gain is that the fusion of the nuclei entails a

loss of a small amount of mass. This mass deficit manifests

itself directly in the (kinetic) energy of the particles

produced. According to Einstein’s famous formula E=mc 2

even with such low amounts of lost mass the released

energy is enormous. In fact, about 10 times more energy is

released in the fusion process of light nuclei than in the

reverse fission process of heavy nuclei.

It quickly became clear that nuclear fusion is the

fundamental process at the bottom of a.) almost every

form of energy on Earth, as well as b.) all material stuff in

the universe besides hydrogen, as within the stars the

process not only fuses hydrogen atoms to helium, but also

produces larger atomic nuclei, carbon, oxygen and finally

the heavy elements such as iron, gold, and manganese.

When a star dies, it hurls in a supernova explosion in which

the “hatched” heavy atomic nuclei spread out into the

vastness of the universe. Several billions of years ago some

of these heavy atomic nuclei eventually found their place

in the vicinity of our forming planet.

However, fusion of nuclei requires enormous pressures

(the product of temperature, i.e. kinetic particle energy,

and particle density) so that positively charged nuclei can

overcome their electrical repulsion and get close enough to

each other to fuse. In stars like our sun these pressures are

reached via the very high densities obtained by ultra-strong

gravitational forces. Such forces and thus densities are not

available on Earth. Terrestrial fusion would therefore have

to employ far higher temperatures to make up for the lower

density and thereby achieve similar pressure as in stars.

In 1934, Mark Oliphant, Paul Harteck and Ernest

Rutherford achieved the fusion of two deuterium nuclei

(an isotope of hydrogen with one extra neutron) by

shooting one deuterium atom onto a metal foil containing

other deuterium atoms. This way they measured what

physicists call the “nuclear cross section” of the fusion

reaction, a characteristic area that provides the probability

that fusion might take place, i.e. how close the nuclei must

get in order to react. This in turn allowed them to determine

the energy necessary for the deuterium-deuterium (DD)

fusion reaction to occur (under atmospheric pressure).

Their result came in at around 100,000 electron volts

(100 keV). This translates into a temperature of more than

one billion Kelvin (the factor that translates the kinetic

energy of atomic particles as measured by eV into the

macroscopic variable of temperature as measures by the

Kelvin scale is the inverse of the Boltzmann constant 1/k B ,

11,604 K/eV, i.e. one eV corresponds to 11,604).

In the late 1940s, physicists first

aimed at recreating the mechanism of

nuclear fusion on Earth, however in

an uncontrolled manner. Their goal

was to create an even more terrible

weapon than the atomic bomb (which

is based on nuclear fission). On

October 31, 1952, the US detonated

their first “hydrogen bomb” releasing

over ten megatons of TNT equivalent,

an energy equivalent to 800 times the explosive power of

the Hiroshima bomb. Less than a year later the Soviet

Union detonated its first hydrogen bomb, and another

eight years later, the Russians tested the “Tsar Bomb”, at

50 megatons and 4000 times the power of the Hiroshima

bomb the most powerful nuclear weapon ever ignited on

Earth.

Nuclear fusion – The forever future

technology?

However, as early as in the early 1940s, even before its

devastating military application, the American researcher

(and later “father” of the hydrogen bomb) Edward Teller

and the Italian Enrico Fermi (who was also the first to

| Left: Carl Friedrich

von Weizsäcker

1993, Göttingen DPI

(Source: Wikimedia);

Right: Hans Bethe

1967 (Source:

www.nobelprize.org)

543

FEATURE | MAJOR TRENDS IN ENERGY POLICY AND NUCLEAR POWER

1 Already in 1920, British astrophysicist Arthur Eddington suggested that stars draw their apparent endless energy from the fusion of hydrogen into

helium. His theory was first published in 1926.

Feature

An Old Promise of Physics – Are We Moving Closer Toward Controlled Nuclear Fusion? ı Lars Jaeger


atw Vol. 65 (2020) | Issue 11/12 ı November/December

FEATURE | MAJOR TRENDS IN ENERGY POLICY AND NUCLEAR POWER 544

perform controlled nuclear fission) developed first ideas

for power generation on the basis of controlled nuclear

fusion. Shooting nuclei at others like Rutherford and his

colleagues had done would surely not do it. Most nuclei

will not hit another one, as the cross section of the fusion

reaction is way too small. The concept Teller and Fermi

developed remains the basis for nuclear fusion researchers

today: In a kind of microwave a deuterium-tritium (DT)

mix is heated to many million degrees so that ultimately

the temperature is high enough for fusion to occur (tritium

is another isotope of hydrogen with two neutrons added;

the DT reaction has a larger cross section than the

deuterium-deuterium (DD) reaction, i.e. it requires lower

temperatures).

When heated to such high temperatures, the atoms lose

their electrons, resulting in a fluid of nuclei and electrons

called a “plasma”. At temperatures of about 100 million

degrees, around six times the temperature at the core of the

sun, terrestrial fusion can release net energy. Although the

kinetic energy of the two nuclei required to fuse is usually

higher than the equivalent temperature of 100 million

Kelvin (as we saw above, this values lies at around 100 keV,

i.e. 1 billion Kelvin), due to the distribution of energies

within the gas as given by the Maxwell- Boltzmann statistics

a gas with less temperature still contains enough particles

at high enough energies to fuse (reactions also proceed by

quantum tunneling of the electric potential energy barrier,

i.e. fusion inherently relies on quantum mechanics).

Important, however, are the conditions required for the

reaction to become self-sustaining, i.e. the energy given off

by the nuclear fusion reactions heats the surrounding fuel

rapidly enough to maintain the temperature against losses

to the environment. The ratio of the obtained fusion power

and the input power required to maintain the reaction

fusion scientists denote by the letter Q. When Q exceeds 1,

fusion produces net energy. A plasma is “ignited” when the

fusion reactions produce enough power to maintain the

temperature without any external heating. An important

variable for this to occur is the above-mentioned cross

section of the reaction. For the fusion process in most

reactors to exceed the losses of the energy to the

environment a certain function of temperature, cross

section and average particle velocity must be exceeded (in

detail: the ratio of squared temperature and the product of

cross section and average velocity of the particles, see

Lawson criterion below). This condition provides a

minimum temperature for the fusion reaction to hold up

and become net positive energy producing. For the DT

reaction this required temperature stands at around

150 million Kelvin (13.6 keV), for the DD reaction it is

around 170 million Kelvin (15 keV).

In an uncontrolled nuclear fusion, the way to get to

fusion conditions is using an atomic bomb. That is how

an H-bomb works: An exploding atomic bomb creates

the necessary pressure and temperature inside a gas

for the nuclei to fuse. That happens so fast that the

plasma does not need to be controlled in any way. In a

controlled nuclear fusion, however, the high temperature

plasma needs to be enclosed and controlled. This requires

strong forces to keep the particles within the plasma as

these are moving with those incredibly high velocities that

are necessary to overcome the electrical repulsion of their

positive charges. Thereby, the challenges are:

a. At such temperatures, the plasma possesses an

enormous amount of thermodynamic pressure and

thus, if not counteracted by another force, flies off

which quickly stops the fusion.

b. Upon contact with the “outer world” (e.g. the container

walls), the plasma immediately cools down which

interrupts the fusion almost instantly.

To address these challenges researchers and engineers

have developed enormous magnetic fields to control the

plasma. Such “magnetic confinement” of the plasma lies at

the heart of most fusion energy projects.

It is difficult not to fall into ecstatic excitement in view

of the practically unlimited possibilities of nuclear fusion.

The energy thus released is safe, carbon-free and its

required initial materials are abundantly available.

p The primary fuel – hydrogen isotopes – can be found in

normal ocean water (albeit tritium is extremely rare on

Earth and needs to be produced by irradiating lithium

in a nuclear reactor).

p One kilogram of the deuterium-tritium (DT) mix is

enough to supply an entire city with energy for a very

long period. A functioning reactor would only need five

kilograms of this hydrogen to produce the energy

equivalent of 18,750 tons of coal, 56,000 barrels of oil

or the amount of energy 755 hectares of solar collectors

produce in one year.

p The only immediate by-product is helium.

p The risk of accidents with a fusion plant is limited: If

something unexpected happens, the fusion reaction

simply stops (note that in existing nuclear power plants

based on fission the cooling of the reactor must be

assured after shut down to safely handle the decay heat

of the fuel).

Unfortunately, a 100 million degrees hot mixture of

hydrogen nuclei has proven so difficult to control that a

well-known joke among physicists is that nuclear fusion is

the most promising technology of the future – and will

remain so forever.

Between excitement and frustration –

The development of magnetic confinement

In the 1950s physicists thought magnetic confinement

would not be too difficult to achieve. Only over time did

they learn about the complexity of the thermodynamic

and magnetohydrodynamic properties of high temperature

plasmas, their inner turbulences and instabilities that

make them so extremely difficult to control. What became

clear is that in order to contain the plasma and reach

temperature and pressures sufficient to ignite the fusion

reaction one had to build up ultra-strong homogenous

magnetic fields. Top scientists all over the world have been

working on the technological challenges this entails for

decades, and so far no fusion reactor has ever been able to

achieve a Q-value larger than one.

In the first designs of magnetic confinement for plasmas

magnetic forces were designed to bring the fast-moving

particles on more and more closely aligned paths so they

can collide and fuse. Such magnetic fields can be created

through a “solenoid”, a simple coil wound into a tightly

packed helix that generates a uniform magnetic field

keeping the nuclei in line and preventing them from

drifting away. However, eventually the particles will run

out to the end of the coil and exit the magnetic field. The

obvious solution was to bend the coil into a circle, resulting

in a donut shape called a torus, in which the particles can

circle endlessly. However, this comes with a new problem:

The magnetic forces within the torus are now unevenly

distributed with their lines being tighter together at the

inside than on the outside of the torus. This leads to forces

causing the plasma particles to drift away from the center

line of the torus. A more complex arrangement of magnets

Feature

An Old Promise of Physics – Are We Moving Closer Toward Controlled Nuclear Fusion? ı Lars Jaeger


atw Vol. 65 (2020) | Issue 11/12 ı November/December

was needed in order to balance these forces and keep the

particles aligned. One design for that purpose was the

“stellarator” invented by US scientist Lyman Spitzer of

Princeton University, which twisted the entire torus at

one end of the torus compared to the other end, thus

forming a figure-eight layout. This design demonstrated

some improved confinement properties compared to a

simple torus, but also displayed a variety of effects that

caused the plasma to be lost from the reactors at too high

rates to reach fusion conditions.

| Fig. 1.

Computer graphics of plasma as well as the stellarator magnet coils and

flat magnet coils of the fusion device Wendelstein 7-X. (Source: IPP)

Another early design was the “z-pinch” concept where a

pulsed plasma is subjected to a strong electrical current

flowing in its center. Based on the principle that parallel

currents attract each other (through the Lorentz force),

the plasma would transiently compress in the process

(physicists speak of the “pinch effect”) with the goal of

reaching end stage conditions that create similar pressures

as in a star, albeit only for a few nano-seconds (a nanosecond

being one billionth of a second). This design

was envisioned to lead to a pulsed fusion concept

without magnetic coils where a regular sequence of

“ mini- implosions” would lead to pulses of net energy being

released. While this design showed some promise, it has

so far failed to reach net energy capability, as various

instabilities form during the compression process that

prevent sufficient pressure build-up.

The various configurations of magnetic and electrical

fields combined with the plasmas’ self-induced pinching

all left the plasma still too unstable. Already in 1949 David

Bohm had, based on empirical observations, conjectured a

relationship (scaling law) between the diffusion of the

plasma and, amongst other things, the strength of the

magnetic field. This relationship was supposed to by

inverse linear, rather than inverse quadratic like classical

physics would predict, so Bohm concluded. If the “Bohm

diffusion“ scaling held, there would be no hope one

could ever build a fusion reactor based on magnetic

confinement. The entire field of fusion research thus

descended into a period of intense pessimism, what

became known as “the doldrums” of nuclear fusion

research.

However, in the late 1960s a concept originally

conceptualized in the 1950s by Soviet physicists Igor Tamm

and Andrei Sakharov started showing very promising

results achieving a stable plasma equilibrium and

promising deviations from the Bohm diffusion conjecture.

In this construction, magnetic field lines wind and twist

around the torus shaped confinement chamber in a helix

like stripes on a candy cane. The asymmetry of

the magnetic fields keeps the particles from drifting

away: Each particle that finds itself at the outside edge of

the torus follows the magnetic lines around the torus and

ends up on the inside edge, where it will drift the other

way towards the outside again. The more the magnetic

field lines twist, i.e. the higher the frequency of the

particles transiting from the outside to the inside and back,

the more stable the plasma became.

In more detail, this construction consists of three arrays

of magnets:

1. External coils around the ring of the torus producing a

toroidal magnetic field, i.e. a field parallel to the inner

circle of the torus.

2. A central solenoid magnet generating with strong

energy pulses a perpendicular magnetic field and thus a

toroidal current within the plasma. The movement of

ions in the plasma then in turn creates a second poloidal

(along the inner ring of the torus) magnetic field.

3. Poloidal coils around the circumference of the torus

control the position and shape of the plasma.

| Fig. 2.

The magnetic field of Wendelstein 7-X (July 2015): The photo combines the

traces of an electron beam on its multiple revolutions along a field line through

the plasma vessel with the image points that it leaves on a fluorescent rod that

is swiveled through the image plane. (Source: IPP, Matthias Otte)

Tamm and Sakharov called their design a “tokamak”

which is a Russian acronym for “toroidal chamber with

magnetic coils”. The results they obtained were at least

10 times better than that of any other fusion machine

before. The tokamak quickly would become the new

standard in international fusion efforts in the coming

years.

The research emphasis now turned to efficient ways to

heat the plasma. Besides the conventional «Ohmic»

heating by inducing a current through the plasma three

techniques became state of the art (including combinations

thereof):

1. Magnetic compression (also called adiabatic compression),

a pinch-like technique in which a magnetic

field compresses the plasma in order to raise its

temperature.

2. Neutral beam injection, in which a particle accelerator

shoots fuel atoms into the plasma, which collide with

the particles in the plasma and thus heat it.

3. Radio-frequency heating: Like in a microwave highfrequency

electromagnetic waves with the right

frequency transfer their energy to the charged particles

in the plasma.

In 1978 by combining the first two techniques the Princeton

Large Torus (PLT) managed to reach temperatures of more

than 60 million Kelvin. The global scientific community

was more and more convinced that the road to a nuclear

FEATURE | MAJOR TRENDS IN ENERGY POLICY AND NUCLEAR POWER 545

Feature

An Old Promise of Physics – Are We Moving Closer Toward Controlled Nuclear Fusion? ı Lars Jaeger


atw Vol. 65 (2020) | Issue 11/12 ı November/December

FEATURE | MAJOR TRENDS IN ENERGY POLICY AND NUCLEAR POWER 546

fusion reactor was now wide open. And the race was on:

The Europeans created JET the „Joint European Torus“,

the Soviets continued to work on their tokamak, the Japanese

created their JT-60, and the US continued to invest

significant money and scientific effort into the Tokamak

Fusion Test Reactor (TFTR). Even private investors put

money into commercial tokamak projects, the first of them

Bob Guccione, the founder of the Penthouse Magazine.

| Fig. 3.

Princeton Large Torus. (Courtesy of Princeton Plasma Physics Laboratory)

But during the 1980s it became clear that plasmas at

such high temperatures are even more difficult to control.

New instabilities arose, and the plasma proved ever

more difficult to be confined. The promising tokamak

architecture ran into similar problems as early torus type

concepts. Todd Evans, a physicist at General Atomics in

San Diego, California, described the problem in illustrative

terms: „Think of squeezing a balloon full of water. The

harder you squeeze, the more the balloon bulges out

through your fingers.” The more the magnetic donut is

squeezed, the more likely the pressurized plasma bursts. It

became clear to physicists that much larger and more

complex (and expensive) machines were needed to solve

these problems. A second period of sobering and pessimism

arose in nuclear fusion research.

| Fig. 4.

26 May 2020: First major component installed – Nearly five years after

contractors to the European Domestic Agency poured the first concrete of

the ITER bioshield, the first major component is installed at the bottom of

the 30-metre-deep “machine well.” (Source: ITER)

ITER – Great as well as expensive hopes

Researchers have in fact been able to confine fusion

plasmas at high enough temperature for long enough to

initiate fusion reactions. However, the confinement time

reached was never long enough to allow for sufficient

fusion energy to circulate in the confined region so that the

plasma remains hot enough to maintain the appropriate

level of fusion. Tokamaks have managed confinement

times of about 30 milliseconds, but times of a second and

more are likely to be needed. Essentially, the problem of

achieving and maintaining fusion in a plasma involves

three main variables:

1. The temperature (or velocity/energy of the particles in

the plasma),

2. the density of the plasma (number of particles per

volume),

3. and the inclusion time (how long the plasma is held

together).

| Fig. 5.

ITER arial May 2020. (Source: ITER, EJF Riche)

In 1955 John Lawson published a criterion that provides

a minimum required value for the product of the plasma

density and its confinement time in order to reach ignition

and then maintain the temperature of the plasma for long

enough against all losses such that fusion energy itself

ultimately keeps the temperature up. Later an even more

useful figure of a reactor’s ability to ignite became the

triple product of density, confinement time and plasma

temperature. The minimum required value for the product

of these three variables is today referred to as a more

general form of the “Lawson criterion” 2 . According to a

rule of thumb, for DT-fusion and for temperatures over

100 million Kelvin the product of particle density and

inclusion time must be greater than 10 14 seconds per cubic

centimeter (10 16 for the deuterium-deuterium reaction).

Reaching such values should be achievable with larger

devices and stronger magnetic fields, so the hope of the

physicists. Existing tokamaks are simply not large enough

to reach burning plasma conditions, they believe. As the

costs estimates for such larger reactors kept mounting it

became clear: International cooperation and funding was

needed. This led to the creation of the project “ International

Thermonuclear Experimental Reactor” (ITER), a joint

effort by equal participation of the Soviet Union (later

Russia), the European Atomic Energy Community, the

United States, and Japan, later joined by China, South

Korea, Canada, and India. In 2005, it was decided that

ITER would be built in the European Union in Southern

2 The original Lawson criterion, however, remained the density-confinement time product, which is what the nuclear research field typically refers to

as the Lawson criterion. Many reactor kinetic equations can be normalized by the double product. However, people occasionally invoke the name

Lawson criterion with the triple product, typically by referring to a more generalized form of the original Lawson criterion.

Feature

An Old Promise of Physics – Are We Moving Closer Toward Controlled Nuclear Fusion? ı Lars Jaeger


atw Vol. 65 (2020) | Issue 11/12 ı November/December

France, in the town of Cadarache, but only recently, in July

2020, more than 30 years after the initial talks about ITER,

the assembly of the machines was launched. Fusion

experiments with DT fuel are expected to start in 2035.

ITER is projected to be the first device that can generate

and maintain a burning plasma, i.e. a plasma in which the

fusion reaction is initiated and kept running. With DT

fusion, ITER is expected to produce 500 MW of fusion

power at a Q value of 10 – fifteen times the current world

record of a Q value of 0.67 (at 16 MW) held by the JET

tokamak in the UK, attained in 1997. For this ITER will

have a central solenoid that will be the most powerful

pulsed superconducting magnet ever constructed.

However, ITER is not designed to create any electricity

output. This would only happen in a successor reactor,

already baptized DEMO (Demonstration Power Station)

and being planned by EUROfusion, the EU’s fusion

organization, with a 2 to 4 gigawatts of thermal output,

operational for electricity production at the earliest in the

late 2040s.

The total projected costs of ITER stands at over

20 billion euros to date and will, according to some experts,

go up to as high as 60 billion euros. It is already the most

expensive experiment in the history of science. Despite all

these tremendous costs and the long-time horizon their

experiments entail, the nuclear fusion researchers at ITER

do not yet know if “physics is not yet again going to bite

them in their ass”. The thermo-, fluid- and hydromagnetic

dynamics and stability properties of a plasma at such

temperature can still be subject to surprises at it has

already been quite a few times in the past 3 . On top of this,

the solution for two particular problems is not yet on the

horizon:

1. The DT fusion reaction produces neutrons of very high

energy (14.1 MeV). Since they are electrically neutral

and thus not influenced by magnetic fields, these neutrons

collide in large numbers and extremely high

speeds with the material of the reactor’s container,

causing enormous damage to it over time. The container

will therefore have to be replaced every one or two

years, which would push the operating costs of a fusion

reactor to unacceptable levels. In addition, the neutron

bombardment in the container material creates radioactive

nuclides, which generates radioactive waste and

thus makes the disposal of the material yet more costly.

2. Tritium is extremely rare on Earth. One gram

of the hydrogen isotope currently costs around

30,000 US dollars. Plus, tritium is beta-emitting

radioactive with a half-life of 12.3 years. This requires

special attention, as tritium is chemically equivalent

to ordinary hydrogen found in water.

Material scientists are working hard on container materials

that solve the first problem; however the path towards

those remains long. For the second problem the physicists

hope to be able to create enough tritium by neutron

activation of lithium-6 for which the fast neutrons of the

DT reaction themselves can be used (for this ITER will

have a “breeder blanket” of lithium located adjacent to the

vacuum vessel).

The call for an alternative to the D-T reaction, which

does not have these problems, has been made by experts

years ago. The next possible candidate for is D-He 3 . Its

neutron output is on order of 1 % of D-T. However, He 3 is

not found terrestrially, but rather abundant on the moon,

where mining would be very costly. The best candidate for

an aneutronic fusion process that does not require tritium

is perhaps the boron-proton reaction. It is “clean” as it

produces three helium nuclei, which are charged particles

that can be easily controlled by electro-magnetic fields and

cause neither lifetime limitations on reactor materials nor

any have any negative impact on the environment. Plus,

boron (and protons) is readily available on our planet. Its

problem: The reaction requires about 30 times higher

plasma temperatures to ignite.

The range of paths is widening

Remember the Lawson criterion: For this to hold it does

not matter whether the density of the plasma is low and its

inclusion time high (as in the tokamak) or vice versa, the

inclusion times being very short and the density very high.

Any combination of these two values is also feasible as long

as their product is about the same. One can thus attack the

Lawson criterion from different directions, i.e. through

different combination of the critical variables. The

tokamak, although being most prominently supported, is

thus not the only path on the road towards commercial

fusion. In fact, some alternative approaches have in recent

years generated considerable new excitement in the fusion

community, as many plasma scientists now conjecture

that in the middle between these two extremes, in the

range of medium range inclusion times and medium range

densities, could lie a very large playground, which has so

far been largely left untouched by the tokamak approach.

Is this maybe where the most promising opportunities for a

controlled nuclear fusion reaction lie?

Public versus private financing

However, public funding for alternative approaches is quite

limited, especially as ITER is taking up so much money.

Governments’ willingness to come up with more funding is

… well … confined. Not surprisingly, in 2019 the US

| Fig. 6.

Schematic view of the tokamak magnetic confinement principle. The

toroidal magnetic field coils establish a strong magnetic field (yellow lines)

within the vessel that captures charged particles on magnetic field lines.

The inner poloidal magnetic field coils are used to induce a current into

the plasma. It produces a poloidal magnetic field (blue lines) in order to

twist the magnetic field lines (green) to prevent particle outward drift.

(Source: Max Planck Institute for Plasma Physics, Christian Brandt)

FEATURE | MAJOR TRENDS IN ENERGY POLICY AND NUCLEAR POWER 547

3 Until today theoretical physicists have not found a coherent solution for the general three-dimensional equilibrium equation in magnetohydrodynamics

(MHD) for an ideal three-dimensional plasma (these equations are combination of the Navier–Stokes equations of fluid dynamics

and Maxwell’s equations for electrodynamics). The processes and turbulences in the plasma are simply too complex. In addition, MHD is not

universally applicable to many systems outside of tokamaks. For instance, the field reverse configurations (FRCs) discussed below are not

describable by MHD models.

Feature

An Old Promise of Physics – Are We Moving Closer Toward Controlled Nuclear Fusion? ı Lars Jaeger


atw Vol. 65 (2020) | Issue 11/12 ı November/December

FEATURE | MAJOR TRENDS IN ENERGY POLICY AND NUCLEAR POWER 548

National Academies of Sciences stated in its Final report of

the committee on a Strategic Plan for U. S. Burning Plasma

Research, that the landscape of fusion research has changed

substantially with the fusion research community now

being much stronger building on significant progress and

investments already made with better and better theoretical

understanding of toroidal magnetic confinement and

plasma control. All this has yielded “ remarkable new

technologies […] promise to reduce the size and cost of

future facilities”. Thus,

“a large DEMO device no longer appears to be the best

long-term goal for the U.S. program. Instead, science and

technology innovations and the growing interest and

potential for private-sector ventures to advance fusion energy

concepts and technologies suggest that smaller, more compact

facilities would better attract industrial participation and

shorten the time and lower the cost of the development path

to commercial fusion energy”.

Indeed, next to the government sponsored gigantic

tokamak project a number of private companies have

dedicated themselves to nuclear fusion research, and

instead of walking along the one and only one true (and

very expensive) path – large-scale plasma held together by

gigantic superconducting magnets – these companies

follow a variety of different ideas in order to possibly find

one path towards the jackpot of a functioning fusion

reactor. Although different in their approach all of them are

looking for paths to fusion that employ much smaller and

thus less expensive reactor technologies than ITER, aiming

at generating electricity already in the next few years and

thus also much faster than ITER with its time horizon of

several decades. They are counting on possible mistakes

and insurmountable obstacles in their ideas being found

(and fixed) much faster than in a few decades time and

before billions of dollars have been burned. The fact that

they depend on risk capital that is hungry for returns could

prove to be a decisive advantage. They simply cannot afford

to turn to large, expensive, long- lasting, and untested

projects. Rather, they must always decide step by step

which next move to take and justify every step in front of

their shareholders. In light of the nature of the described

problems around thermonuclear fusion technology such a

pragmatic approach might prove more appropriate.

| Fig. 7.

A tall electromagnet--the central solenoid--is at the heart of the ITER

Tokamak. It both initiates plasma current and drives and shapes the

plasma during operation. (Source: US ITER)

These private companies have in recent years made

some considerable progress. A real public-private race for

the best fusion technology solution has in fact developed.

How fruitful such a race can be showed the example of the

Human Genome Project some 20 years ago. The following

provides a list of private initiatives that work on tokamak

type designs:

p Commonwealth Fusion Systems (CFS) is a spin-off from

MIT’s (Massachusetts Institute of Technology) Plasma

Science and Fusion Center, one of the pioneers of the

US fusion research in the 1960s. The company is

pursuing a more or less fairly conventional tokamak

approach. However, they are trying to integrate some

recent technological advances that will not be part of

ITER, in particular new high-temperature superconducting

material for a large scale electromagnet

(barium copper oxide versus niobium-titanium in

ITER) which will allow, so the scientists hopes, for

magnetic fields in the range to 20 Tesla in overall

smaller and more efficient design. CFS is pursuing a

tokamak that would produce 50 MW to 100 MW of

fusion power, i.e. one fifth of the foreseen ITER power,

at a Q value of 3, less than one third of the foreseen

ITER value. The company is funded by MIT itself as well

as venture capital, including the Bill Gates – backed

Breakthrough Energy Ventures, and Italian oil and gas

producer Eni.

p Tokamak Energy based in the UK is employing a

tokamak with a more spherical shape. It is also privately

funded and raised about $86 million in an early 2020

funding round.

The alternatives

Other private companies have endeavored some highly

interesting alternative paths towards a fusion reactor

altogether:

p (FRC): FRC is an alternative magnetic confinement

method which still involves a toroidal plasma, however

without any magnetic coils running through the center

of the toroid like in the tokamak and also no toroidal

coils. It entails an external axial magnetic field wherein

electric currents in the plasma create a poloidal

magnetic field, which has an effective axial component

that opposes, i.e. reverses, the externally applied field.

This then self-confines the plasma torus which takes

the shape of a smoke ring or, depending on the

configuration, extends into a tubular shape. Plasma

physicists refer to this as a “compact toroid”. Its topology

represents a minimum energy state and can be made

very stable. The hope is that the less complex magnetic

field topology with high magnetic efficiency (most

of the field is produced by the plasma itself rather

than the external magnets) will allow for the

construction of dramatically simpler and less expensive

fusion reactors.

The main proponent of this method is a company

called TAE Technologies, based in Irvine, California. Its

publicly announced funding exceeds $750 million, and

known backers include venture capital firms New

Enterprise Associates and Venrock, the UK’s Wellcome

Trust, several sovereign funds, Alphabet (Google) and

other high-tech investors. Rather than relying on DT

fusion TAE seeks to ultimately fuse protons and boron.

Though this requires temperatures of more than an order

of magnitude higher than the temperatures necessary

for the DT reaction it has the advantage of being

“ aneutronic”, i.e. it does not produce the hard to control

Feature

An Old Promise of Physics – Are We Moving Closer Toward Controlled Nuclear Fusion? ı Lars Jaeger


atw Vol. 65 (2020) | Issue 11/12 ı November/December

highly energetic neutrons. Furthermore, it does not

require the rare to obtain tritium. TAE’s prototype is a

cylindrical colliding beam fusion reactor (CBFR) that

first heats hydrogen gas to form two rings of plasma

which are then merged together (see below for more

details).

p Sheared-flow stabilized Z-pinch: This is a method

extending the conventional z-pinch by trying to stabilize

the plasma with a sheared flow, i.e. plasma flowing at

different velocities at different radii. This way, the

high-temperature, high-density reactive medium is

targeted to be confined long enough for the fusion

reactions to occur, while being “orders of magnitude

cheaper” than fusion reactors requiring magnetic coils,

so the claim of its supporters. This method is employed

by the company Zap Energy, founded in 2017.

p (Laser induced) Inertial confinement fusion (ICF): While

magnetic confinement tries to solve the Lawson

criterion problem with long confinement times (several

seconds) and comparably low plasma density (10 14 ions

per cm 3 ), the ICF approach takes the inverse path:

ultra-high ion densities (10 25 ions per cm 3 , about

100 times the densest metal) and short confinement

times or even no confinement all. The high density

causes the fusion reactions to occur in around one

nanosecond which is fast enough for it to navigate

through the fusion material before this expands. In

order to achieve such a high density (and thus

temperature) ultra-strong and at the same time ultraprecise

lasers are needed. These are then focused on

the fusion fuel containing a mixture of frozen deuterium

and tritium which typically takes the form of a pellet

the size of a pinhead.

The largest ICF experiment is the National Ignition

Facility (NIF) at the Lawrence Livermore National

Laboratory (LLNL) in California with its Laser Inertial

Fusion Energy (LIFE) program. However, different to

its own predictions the NIF did not succeed in getting to

more than 1/3 of the required conditions needed for

ignition. LIFE was, therefore, stopped in 2014 and LLNL

shifted its focus toward defense applications. However,

with the power of lasers having rapidly increased in

recent year on a Moore’s-Law like path (especially with

the development of Chirped Pulse Amplification (CPA)

lasers; Physics Nobel Prize 2018) the ICF concept

has more recently attracted attention again. The

government financed company Sandia Laboratories,

based in Albuquerque, New Mexico, has dedicated itself

along this path.

p Magnetized target fusion (MTF; or magneto-inertial

fusion (MIF)): MTF attempts to work in parameter

regions between magnetic confinement and ICF aiming

for plasma densities of 10 19 ions per cm 3 and confinement

times in the order of 1 µs. Like for magnetic

confinement the fusion fuel is confined by magnetic

fields while it is heated into a plasma. However, as in

the inertial approach, the density required for fusion is

then achieved by rapidly compressing the plasma. This

approach suggests that the energy inputs to the plasma

is comparably small such that a corresponding reactor

would run more efficiently and thus be less expensive

compared to trying to achieve long confinement times

as in magnetic confinement or ultra-dense states as in

the ICF approach.

MTF is predominantly pursued by the Vancouver,

British Columbia-based company General Fusion.

General Fusion uses an array of pistons to create shock

waves in a liquid metal to compress the plasma to fusion

conditions. The company has raised $200 million in

funding or commitments. The firm is supported

amongst others by Amazon CEO Jeff Bezos, and other

venture capital sources incl. Asian sovereign wealth

funds, with the Canadian government having provided

about $40 million.

p Lockheed Martin Compact Fusion Reactor (CFR):

Lockheed Martin utilizes a different magnetic topology

and set up claiming this would produce a much more

effective magnetic field for plasma containment thus

allowing an overall smaller (and thus less expensive)

fusion reactor. However, it has yet to publicize any data

on their progress. So far, no details on temperature or

containment levels achieved have been published.

p Muon-catalyzed fusion (μCF): Muons are subatomic

particles that have similar properties as electrons

but are more than 200 times heavier. A muon can

replace an electron in a hydrogen molecule, which

due to its higher mass brings the nuclei in the

molecule much closer together which increases the

probability of nuclear fusion greatly, eventually to a

point where sufficiently many fusion events might

happen at much lower, possibly even room temperature.

One therefore speaks of “cold fusion”. However,

the creation of the (naturally unstable) muons in

sufficiently large numbers requires much more energy

than would be produced by the targeted fusion. The

company Norrønt Fusion Energy AS in Norway is

currently working on laser produced muons for

Muon-catalyzed fusion.

FEATURE | MAJOR TRENDS IN ENERGY POLICY AND NUCLEAR POWER 549

(a) (b) (c)

| Fig. 8.

Inertial confinement fusion (ICF) concepts: (a) laser indirect drive (LID); (b) laser direct drive (LDD); and (c) magnetized liner inertial fusion (MagLIF).

(Source: Laser-direct-drive program: Promise, challenge, and path forward - Scientific Figure on ResearchGate)

Feature

An Old Promise of Physics – Are We Moving Closer Toward Controlled Nuclear Fusion? ı Lars Jaeger


atw Vol. 65 (2020) | Issue 11/12 ı November/December

FEATURE | MAJOR TRENDS IN ENERGY POLICY AND NUCLEAR POWER 550

| Fig. 9.

The fusion device for field-reversed configuration. (Courtesy of TAE Technologies)

A deeper look into one alternative path

While almost all efforts focus on making the DT reaction

work and become commercially viable, the “neutron

problem” of the DT reaction demands, as we saw, a look

into alternative fuel sources. While achieving a confined

enough plasma at 100 million Kelvin is already a tough

challenge, achieving such at three billion Kelvin as required

for the best candidate, the boron-proton reaction, seems

insurmountable. Unless one finds a new, more effective

approach to reach stable confinement. Such would need to

prove more favorable at higher temperatures than at lower

temperatures. That is what TAE Technologies is trying to

achieve.

For this purpose, TAE created a reactor that appears to

be some strange combination of a particle accelerator and

an ordinary plasma. The ultra-high temperature in the

plasma is achieved by accelerating beams of fuel particles

and have them collide with plasma particles, something

particle physicists have done for decades. The typical

magnetically contained plasma donuts are thereby

replaced by a long-stretched plasma tube taking the shape

of a hollow cigar. To improve its stability, this tube would

be made to spin around itself such that the gyroscopic

effect makes it a lot more stable. This is the essence of the

advanced FRC approach pursued by TAE. In theory, this

approach can be scaled up to much higher temperatures

than those in a tokamak. TAE has found evidence that the

FRC induced stability and quiescence in the plasma

actually increases with higher temperature! It is the very

hypothesis that this beneficial scaling property will rest in

place all the way to 3 billion degrees that TAE’s approach is

based upon.

In detail, TAE’s mix of a particle accelerator and plasma

confinement works as follows:

3 It sends off short ultra-strong bursts of electric power

from two sides which generate corresponding magnetic

fields that create plasmas in each of the separate ends of

the machine.

3 A second strong electric pulse then accelerates the two

plasmas to a million km/h and makes them crash into

each other in the middle of the machine.

3 This creates a larger tube-like plasma structure that,

heated further with intense beam accelerators, shall

eventually become hot, dense, and contained for long

enough to cause the fusion reaction.

The company has just started to build its next generation

device called “Copernicus” targeting temperatures of more

than 100 million Kelvin and thus establishing deuteriumtritium

fusion conditions and the viability of achieving net

energy from DT fusion. If this proves to be viable the firm

will build a successor device to prove the commercial

viability of a fusion energy reactor designed to operate

with the proton-boron reaction, the ultimate holy grail of

fusion research.

Outlook

The science of plasma underlying nuclear fusion research

and our understanding how plasmas behave under the

required extreme circumstances have advanced a great

deal in recent years, much of that out of the public eye.

Thus, there is some optimism that the technology is well

on its way to commercial use, despite that the engineering

obstacles remain high. However, besides the immense

technological challenges, the ultimate deciding factors for

the application of fusion energy will be social and

economic. Fusion power plants will be built when investors

and public utility commissions view them as worthwhile

investments. It is worth noting that the likely time frame

of such commercial viability roughly coincides with

the period when many operating fission plants in

industrialized countries are reaching the end of their

license periods, as well as with the objective to reach

net-zero carbon emissions around 2050 or 2060. Under

such circumstances, the advantages of fusion power could

well be economically and socially compelling.

Commercially available fusion technology, if one day it

were actually available to mankind, would represent a

social, technological and economic paradigm shift. Were

we really able to produce energy like the sun does and

thus have access to the most efficient, safest and most

environmentally friendly form of energy nature provides,

we would certainly experience not only another major

technological advance, but rather a leap forward in

civilization itself, comparable only to the invention of

the steam engine that provided the energy that lifted

humanity into the modern age 250 years ago.

Author

Dr. Lars Jaeger

lars.jaeger@larsjaeger.ch

Dorfstrasse 1

6340 Baar, Switzerland

Feature

An Old Promise of Physics – Are We Moving Closer Toward Controlled Nuclear Fusion? ı Lars Jaeger


atw Vol. 65 (2020) | Issue 11/12 ı November/December

Highlights of the World Nuclear

Performance Report 2020

Jonathan Cobb

The 2020 edition of the World Nuclear Performance Report, published by World Nuclear Association, combines

the Association’s own data and analysis with data from the International Atomic Energy Agency’s PRIS database.

The report looks at long term trends in nuclear performance as well as progress in new nuclear build.

The full report, available from the Association’s website 1

also includes five case studies and interviews, looking

at individual examples of excellent performance in the

nuclear industry.

While this year’s report details the performance of

nuclear reactors globally in 2019, over much of 2020 our

focus has been on the impacts of the COVID-19 pandemic.

Throughout the pandemic, operators have worked with

great commitment to ensure that their reactors have

continued to provide electricity and support grid stability.

Staff working at reactors have had to adapt to working in

COVID-safe conditions whilst ensuring continuity.

With the dramatic drop in electricity demand seen in

some regions, reactors have had to demonstrate greater

flexibility in operation. While many renewable generators

have been cushioned from the impacts of the pandemic by

obligations to purchase their electricity, nuclear operators

have had to vary the output of their plants to support both

intermittent generation and changes in demand.

In 2020 the nuclear industry has been an essential part

of the response to the coronavirus pandemic. Prior to this,

2019 proved to be one of the most significant for nuclear

generation, with near-record levels of generation.

Nuclear reactors generated a total 2657 TWh of

electricity in 2019, up 95 TWh from 2563 TWh in 2018,

and second only to the 2661 TWh generated in 2006. This

is the seventh successive year that nuclear generation has

risen, with output 311 TWh higher than in 2012.

In 2019 reactors totalling 402.3 GWe were classed as

operable, including those that either started up or shut

down. This is fractionally higher than the 2018 figure of

402.0 GWe. The end of year capacity on 31 December 2019

was 392 GWe, down from 397 GWe in 2019.

Six reactors started up in 2019. Four large PWRs

commenced operation, one in South Korea, one in Russia

and two in China. In addition, two small reactors on the

first purpose-built floating nuclear power plant, harboured

at the town of Pevek in northeast Russia, started supplying

electricity. New construction began on five reactors, two in

China and one each in Iran, Russia and the UK.

In 2019, nuclear generation rose in Africa, Asia, South

America and East Europe & Russia. It was fractionally

down in North America, and 3 TWh lower in West &

Central Europe. Recent trends continue, with particularly

| Fig. 1.

Nuclear electricity production.

| Fig. 2.

Nuclear generation capacity operable (net).

551

SERIAL | MAJOR TRENDS IN ENERGY POLICY AND NUCLEAR POWER

Africa Asia East Europe & Russia North America South America West & Central Europe Total

BWR 21 (-5) 34 (-1) 10 (-1) 65 (-7)

FNR 2 2

GCR 14 14

LWGR 13 (-1) 13 (-1)

PHWR 24 (-1) 19 3 2 48 (-1)

PWR 2 92 (+2) 38 (+3) 64 (-1) 2 102 (-2) 300 (+2)

Total 2 137 (-4) 53 (+2) 117 (-2) 5 128 (-3) 442 (-7)

| Tab. 1.

Operable nuclear power reactors at year-end 2019.

1 https://world-nuclear.org/our-association/publications/global-trends-reports/world-nuclear-performance-report.aspx

Feature

Highlights of the World Nuclear Performance Report 2020 ı Jonathan Cobb


atw Vol. 65 (2020) | Issue 11/12 ı November/December

SERIAL | MAJOR TRENDS IN ENERGY POLICY AND NUCLEAR POWER 552

| Fig. 3.

Regional generation.

strong growth in Asia, which saw nuclear generation rise

by 17 %.

At the end of 2019 the capacity of the world’s 442

operable reactors was 392 GWe, down from 397 GWe at

the end of the previous year. Thirteen reactors shut down,

of which four were in Japan and had not generated

electricity since 2011; and three, in South Korea, Germany

and Taiwan, were shut prematurely due to political phaseout

policies.

Given the reduction in overall nuclear capacity, the

increase in generation in 2019 is all the more remarkable.

However, there is an urgent need for the pace of grid

connections and new construction starts to increase in

order to expand the essential contribution nuclear energy

makes to global clean energy provision and reach the

nuclear industry’s Harmony goal.

Operational Performance

Capacity factors are based on the performance of those

reactors that generated electricity during each calendar

year. For reactors that were grid connected or permanently

shut down during a calendar year their capacity factor

is calculated on the basis of their performance when

operable.

In 2019 the global average capacity factor was 82.5 %,

up from 79.8 % in 2018, maintaining the consistently high

capacity factors seen over the last 20 years. In general, a

high capacity factor is a good indication of excellent

operational performance. However, there is an increasing

trend in some countries for nuclear reactors to operate in a

load-following mode, which will reduce the overall

capacity factor. The ability of nuclear reactors in France to

adjust output to match varying demand and balance the

output of intermittent renewables is covered in one of the

case studies of the full World Nuclear Performance Report

2020.

Capacity factors in the different geographical regions

are also broadly consistent with those achieved over the

preceding five years. Capacity factors in North America

continue to exceed 90 %.

With 2019 seeing the first five reactors to reach 50 years

of operation and the first licences granted for 80 years of

operation, it is welcome to note that there is no significant

age-related decline in nuclear reactor performance. The

mean capacity factor for reactors over the last five years

shows little variation with age after the initial start-up of

the reactor. For reactors beyond 40 years of operation

there is a slight increase in average capacity factor. There

may be a selection effect, with those reactors performing

best more likely to be selected for long-term operation.

The spread of capacity factors in 2019 is broadly similar

to the average of the previous five years, with more than

two-thirds of reactors having a capacity factor greater than

80 %. With load-following increasing in some countries, a

greater spread of capacity factors may be seen in the

future.

There was a substantial improvement in capacity

factors achieved from the 1970s through to the 2000s.

| Fig. 4.

Global average capacity factor.

| Fig. 6.

Mean capacity factor 2015-2019 by age of reactor.

| Fig. 5.

Capacity factor by region.

| Fig. 7.

Percentage of units by capacity factor.

Feature

Highlights of the World Nuclear Performance Report 2020 ı Jonathan Cobb


atw Vol. 65 (2020) | Issue 11/12 ı November/December

Since then, this high performance has been maintained.

In the 1970s less than half of all reactors achieved a

capacity factor greater than 70 %, compared to 83 % of

reactors in 2019.

Thirteen reactors shut down in 2019. Of these, four

units at Fukushima Daini in Japan had not generated

electricity since 2011, and three reactors (Wolsong 1 in

South Korea, Philippsburg 2 in Germany and Chinshan 2,

Taiwan) were shut prematurely due to phase-out policies.

Six reactors were connected to the grid in 2019. Four of

these reactors were large-scale PWRs, two of which were in

China (Yangjiang 6 and Taishan 2), Shin Kori 4 in South

Korea and Novovoronezh II-2 in Russia.

In addition, the two reactors on the first purpose-built

floating nuclear power plant, the Akademik Lomonosov,

began supplying electricity in the town of Pevek, on the

north east coast of Russia. Those reactors also supply

district heating to the community.

The median construction time for reactors in 2019

was 117 months. As was the case in 2018, several of the

reactors starting up in 2019 featured designs that were

first-of-a-kind (FOAK).

The Akademik Lomonosov’s start date represents its

first keel-laying ceremony. The project restarted with a

second keel-laying ceremony in 2009, when construction

moved from Severodvinsk to Baltiysky Zavod.

Consequentially, the median construction time for

reactors in 2019 is significantly above the average achieved

over the last 20 years.

Not all new reactor designs have entailed such long

construction times. Yangjiang 6, which was completed in

66 months, is the second ACPR-1000 unit to be built, after

completion of its sister unit, Yanjiang 5, in 2018.

Construction times since 2015 have more typically

been between five to six years. In August 2020 we

saw the startup of Tianwan 5, after a construction period

of 56 months, less than half the 2019 average; this

is in part due to the benefits of experience gained through

series construction. Even though the reactor is only

the third of this specific design, it is a development of a

design that was used for more than 20 different reactors.

It is also partly the result of having an ongoing construction

programme that helps build and retain skills among the

workforce.

Where new reactors have been successfully deployed

there needs to be a commitment to repeating that

deployment through series build, to take advantage of the

learning gained.

Barakah 1, which started up in August 2020, will be

followed by three more reactors that will benefit from the

experience gained in starting the first unit.

Construction started on reactors, Kursk II-2 in Russia,

Bushehr 2 in Iran, Hinkley Point C2 in the United Kingdom

and two units in China, Zhangzhou 1 and Taipingling 1.

Most reactors under construction today started

construction in the last nine years. The small number that

have taken longer are either pilot plants, FOAK reactors,

or projects, where construction was suspended before

being restarted more recently (such as Mochovce 3&4 in

Slovakia)

how to restart their economies safely to generate new

employment opportunities.

| Fig. 8.

Long-term trends in capacity factors.

| Fig. 9.

Construction times of new unit’s grid connected in 2019.

| Fig. 10.

Median construction times for reactors since 1981.

SERIAL | MAJOR TRENDS IN ENERGY POLICY AND NUCLEAR POWER 553

Nuclear’s place in the post-pandemic recovery

At the time of writing, the pandemic is still affecting many

parts of the world. Its impact has not been limited to

its tragic health effects. Around the world, economies

have contracted sharply and many people have lost

their jobs. A key question for governments now is

| Fig. 11.

Operational status of reactors with construction starts since 1985.

Feature

Highlights of the World Nuclear Performance Report 2020 ı Jonathan Cobb


atw Vol. 65 (2020) | Issue 11/12 ı November/December

SERIAL | MAJOR TRENDS IN ENERGY POLICY AND NUCLEAR POWER 554

Major public infrastructure investment is the cornerstone

of many governments’ strategies for recovery. The

scale of planned stimulus packages provides a unique

opportunity to build a more sustainable world, with

societies emerging stronger, cleaner and more resilient.

Nuclear energy should play a central role in these

recovery efforts. Supporting existing nuclear generation

and promoting new nuclear build will boost economic

growth in the short-term and underpin the development of

a low-carbon, resilient and cost-effective electricity

infrastructure. Nuclear projects attract valuable inward

investment, driving sustained long-term local and national

economic growth.

Investments into nuclear projects stimulate the

economy well beyond the nuclear industry and deliver

widespread economic growth.

Studies in the USA and Europe have shown that

every euro or dollar spent on the nuclear industry

results in four times that investment in the broader

economy. And, when evaluated on an equitable basis,

nuclear energy remains a very competitive option for new

generation.

Investment in nuclear energy research, development

and deployment will stimulate job creation in new sectors

and broaden the application of nuclear technologies.

Small modular reactors and advanced reactor technologies

have the potential to decarbonize other sectors through,

for example, the provision of carbon-free heat for use in

both residential and industrial applications.

Around the world there are more than 100 planned

reactors with approval, funding or commitment in place.

With the right support, these reactors would play a crucial

role in the post-pandemic recovery, and each and every

one will create considerable societal benefits – but to

ensure these are realised governments must put

mechanisms in place to value nuclear energy’s unique

attributes.

Further opportunities exist in the form of ensuring the

long-term operation of existing nuclear reactors. Securing

continued generation from the approximately 290 reactors

which have been operating for more than 30 years, and

which have the potential to generate for decades to come,

is the cheapest way to generate low-carbon electricity.

Creating jobs and boosting local economies

Investing in nuclear energy is a very good way to create

many jobs for local and regional economies, as well as

strengthening national construction and technology

capabilities.

For example, 25,000 employment opportunities will be

created by the Hinkley Point C project in the UK, including

over 1,000 apprenticeships during the construction phase,

and 900 permanent jobs onsite during more than 60 years

of operation. About 64 % of the construction contracts are

being delivered by UK companies, and the project is

contributing some £1.5 billion to the local economy during

construction, when operating it is expected to contribute

£40 million every year.

Looking forward, governments are considering at how

to restart their economies and generate jobs, as well as

how to meet their energy and environmental goals. Each

new nuclear build project generates thousands of jobs and

boosts the local economy, as well as contributes to our

Harmony goal of a clean and reliable electricity mix.

Upgrading existing reactors also provides major

employment benefits. The Bruce nuclear power plant in

Canada is undergoing refurbishment of its six reactors.

This will sustain 22,000 jobs, as well as providing low-cost,

reliable, carbon-free electricity until 2064.

A time for action and ambition

The global nuclear industry is ready to work with

policymakers to set a greater ambition for meeting climate

goals and to create the jobs needed for sustainable

economic growth.

We cannot afford to allow a minority of countries

promoting their ill-judged anti-nuclear dogma to dictate

and restrict multilateral action on energy and the

environment. The failure to include nuclear energy in the

European Commission’s sustainable finance taxonomy

from the very start, thereby potentially hampering the

financing of new nuclear projects, runs counter to its

“do no significant harm” principle – constraining nuclear

energy will mean more pollution and higher carbon

emissions, as well as less reliable supply and higher prices

for consumers.

We need to have greater ambition to build a more

sustainable and equitable future for everyone around the

world. While the pandemic has been the focus for governments

in 2020, the essential challenge of our time remains

to ensure that no one is forced to live without reliable and

affordable energy, whilst also protecting the planet for

future generations.

We are ready to meet the Harmony goal of 1000 GWe of

new nuclear capacity before 2050, which would ensure

that at least 25 % of global electricity would be generated

by nuclear reactors, as part of a low-carbon energy mix.

World Nuclear Association therefore calls on policymakers

to consider nuclear in energy transition plans,

and enact policies to ensure that the socio-economic,

environmental and public health benefits of nuclear

technology are extended to as many people as possible.

Author

Dr Jonathan Cobb

Senior Communication Manager

World Nuclear Association

Tower House, 10 Southampton Street

London WC2E 7HA, United Kingdom

Feature

Highlights of the World Nuclear Performance Report 2020 ı Jonathan Cobb


atw Vol. 65 (2020) | Issue 11/12 ı November/December

Ausfuhr von unbestrahlten Brennelementen – Lackmustest

für fairen und funktionierenden Außenhandel in Deutschland

555

Tobias Leidinger

Die Ausfuhr von in Deutschland gefertigten, unbestrahlten Brennelementen für Kernkraftwerke im benachbarten

Ausland ist seit Jahrzehnten etablierte Praxis und Teil eines funktionierenden Außenhandels: Für den Betrieb der

Kraftwerke – und die Versorgungssicherheit im jeweiligen Empfängerland – sind pünktliche Lieferungen unverzichtbar.

Die dafür beantragten Ausfuhrgenehmigungen nach § 3 Abs. 3 AtG wurden vom Bundesamt für Wirtschaft und

Ausfuhrkontrolle (BAFA) bislang stets erteilt. Nunmehr hat eine Entscheidung des VG Frankfurt a.M. vom 16. Oktober

2020 (Az. 6 L 2470/20.F) Aufsehen erregt: Die von einer Privatperson eingelegte Klage gegen die dem Brenn element-

Hersteller erteilte Ausfuhrgenehmigung könne die Ausfuhr stoppen. Eine Gefahr für die innere Sicherheit der

Bundesrepublik sei nicht auszuschließen. Diese Entscheidung überrascht. Sie wirft grundsätzliche Fragen auf,

insbesondere, ob ein fairer und funktionierender Außenhandel in Deutschland so noch eine Perspektive hat.

I. Die Entscheidung des VG Frankfurt

vom 16.10.2020

Das BAFA hatte dem Hersteller im März 2020 die

Genehmigung zur Ausfuhr nach Belgien gemäß § 3 Abs. 3

AtG erteilt. Dagegen hatte eine Privatperson – nach

erfolgloser Erhebung von Widerspruch – im August 2020

Klage beim VG Frankfurt erhoben und beantragt, die

Ausfuhrgenehmigung aufzuheben. Zur Begründung hieß

es, dass das zu beliefernde Kernkraftwerk in Belgien

altersbedingt ein hohes Sicherheitsrisiko darstelle. Die

Ausfuhrgenehmigung verletze ihn als Kläger persönlich in

seinen Rechten auf Leben, Gesundheit und Eigentum, da

er im Grenzgebiet zu Belgien lebe. Der Hersteller verlangte

daraufhin in einem Eilantrag beim VG Frankfurt

festzu stellen, dass der Klage der Privatperson keine

auf schiebende Wirkung zukomme, da der Genehmigungstatbestand

nach § 3 Abs. 3 AtG nicht dem Schutz der

Rechte Privater diene, also durch die Erhebung der Klage

des Dritten kein Ausfuhrstopp bewirkt werden könne. Die

Klage sei bereits offensichtlich unzulässig.

Das Gericht hat den Antrag des Herstellers abgelehnt.

Die Klage des Dritten sei nicht offensichtlich unzulässig,

da nicht auszuschließen sei, dass der entscheidungserhebliche

Genehmigungstatbestand der „inneren

Sicherheit der Bundesrepublik Deutschland“ aus § 3 Abs. 3

Nr. 2 AtG der Privatperson ein subjektiv-öffentliches

Recht verleihe. Die vorgeblichen Sicherheitsmängel des

Empfängerkraftwerks überwögen jedenfalls das Interesse

an der Ausfuhr des Herstellers. Die Ausfuhrgenehmigung

sei daher nicht vollziehbar.

Der Argumentation des VG Frankfurt folgend könnten

sich gravierende Auswirkungen für den deutschen

Außenhandel ergeben. Dabei leidet die Entscheidung des

Gerichts an grundlegenden Mängeln. Sie kann keinen

Bestand haben. Das Beschwerdeverfahren beim VGH

Kassel läuft bereits.

II. Verhältnis von Außenwirtschafts- und

Atomrecht grundsätzlich verkannt

Bei der Ausfuhrgenehmigung nach § 3 Abs. 3 AtG für

Kernbrennstoffe handelt es sich um eine Zulassungsentscheidung

des Außenwirtschaftsrechts. Das Außenwirtschaftsgesetz

(AWG) bestimmt in § 14 Abs. 2 grundsätzlich,

dass Klagen gegen Ausfuhrgenehmigungen keine

aufschiebende Wirkung zukommt. Nach § 3 Abs. 4 AtG gilt

dieser Grundsatz auch für Ausfuhrgenehmigungen nach

§ 3 Abs. 3 AtG, denn die Regelungen des AWG als Rahmengesetz

bleiben nach der klaren Anordnung in § 3 Abs. 4 AtG

unberührt, d.h. sie sind zu beachten. Das AWG verdrängt

also nicht Regelungen außenwirtschaftlicher Sondervorschriften,

wie § 3 Abs. 3 AtG, sondern umgekehrt, die

Sondervorschriften – hier § 3 Abs. 4 AtG – verweisen ausdrücklich

auf die Vorgaben des darüber hinaus geltenden

Rahmengesetzes, also des AWG. Folgerichtig gilt hier die

Grundsatzanordnung aus § 14 Abs. 2 AWG. Danach dulden

Entscheidungen der Genehmigungsstellen im Außenwirtschaftsverkehr

keinen Aufschub. Denn Verzögerungen

könnten gravierende Folgen haben – hier in Bezug auf die

Versorgungssicherheit. Genau deshalb kommt Widerspruch

und Klage gegen solche Entscheidungen nach Ratio

und Inhalt des § 14 Abs. 2 AWG keine aufschiebende

Wirkung zu. Diese grund sätzliche Vorgabe im Außenwirtschaftsrecht

hat das VG Frankfurt bereits im Ausgangspunkt

verkannt.

III. Kein Drittschutz durch § 3 Abs. 3 Nr. 2 AtG

zugunsten Privater

Fehlerhaft ist die Entscheidung des Gerichts im vorläufigen

Rechtsschutzverfahren auch deshalb, weil die Frage der

Klagebefugnis des Privatklärgers prozessual unzulässig

offengelassen, d.h. nicht geklärt wurde (1.). Soweit das

Gericht in der Sache zum Drittschutz Stellung nimmt,

um ihn „jedenfalls“ nicht zu verneinen, ist das sachlich

verfehlt (2.).

1. Frage des Drittschutzes durfte nicht offen

bleiben

Durch das „Offenlassen“ der entscheidenden Rechtsfrage,

nämlich ob die Klage der Privatperson überhaupt aufschiebende

Wirkung entfalten kann, ist es zu einer

prozessual unzulässigen Verlagerung in das Hauptsacheverfahren

gekommen. Damit wurde der Rechtsgewährsanspruch

des Herstellers vereitelt: Kann jede

beliebige Privatperson – ohne dass es auf die Frage

ankäme, ob sie als außenstehender Dritter durch die in

Rede stehende (Ausfuhr-)Genehmigung überhaupt in

eigenen Rechten verletzt sein kann – Klage erheben, mit

der Folge, dass die Genehmigung nicht mehr genutzt

werden kann, liefe der Vollzugsanspruch des Herstellers

leer. Es genügte ein Rechtsmittel eines beliebigen Dritten

um das Genehmigungsrecht auszuhebeln und die Ausfuhr

erst einmal zu stoppen. Die Rechtsfrage also, ob einer

Privatperson durch § 3 Abs. 3 Nr. 2 AtG in Bezug auf eine

danach erteilte Ausfuhrgenehmigung überhaupt subjektive,

eigene Rechte eingeräumt sind, durfte das Gericht

hier nicht offenlassen. Dadurch wurden verwaltungsprozessuale

Grundsätze verletzt.

2. Vorschrift des § 3 Abs. 3 AtG vermittelt

keinen Drittschutz

Soweit das Gericht in seiner Entscheidung darauf

abstellt, dass § 3 Abs. 3 AtG „Drittschutz“ zugunsten des

SPOTLIGHT ON NUCLEAR LAW

Spotlight on Nuclear Law

Export of Non-irradiated Fuel Elements – Indicator for Fair and Functioning Foreign Trade in Germany ı Tobias Leidinger


atw Vol. 65 (2020) | Issue 11/12 ı November/December

SPOTLIGHT ON NUCLEAR LAW 556

Privatklägers jedenfalls nicht ausschließe und daher die

Vollziehung der Ausfuhrgenehmigung im Ergebnis nicht

möglich ist, fußt diese Bewertung auf einem nicht haltbaren

Rechtsverständnis: Wortlaut, Systematik und Sinn und

Zweck von § 3 Abs. 3 AtG als einer speziellen, außenwirtschaftsrechtlichen

Bestimmung ergeben ein klares Bild:

Hier geht es allein um die Wahrung der „ inneren und

äußeren Sicherheit der Bundesrepublik Deutschland“.

Damit ist allein die „Sicherheit der Bundesrepublik“ als

kollektive Rechtsträgerin erfasst, nicht aber die Sicherheit

einzelner Bürger. Der Begriff der „inneren Sicherheit der

Bundesrepublik“ wird gleichlautend in einer ganzen Reihe

anderer Gesetzen verwendet (u.a. StGB, StPO, BVerfSchG,

BNDG, BKAG, BKAG), bei denen es ebenfalls um den Schutz

des Bestandes und der Funktionstüchtigkeit des Staates

geht. Durch die Recht sprechung ist geklärt, dass dieser

Begriff enger zu verstehen ist als der Begriff der „allgemeinen

Sicherheit und Ordnung“, der in den Polizei- und

Ordnungsgesetzen Verwendung findet und auch Rechte

einzelner ein schließen kann. Systematisch wäre nicht

erklärbar, warum das Atomgesetz in seinen Zweckbestimmungen

klar zwischen dem Schutz von Leben und

Gesundheit (§ 1 Nr. 2) und – wie § 3 Abs. 3 AtG – dem

Schutz der „inneren und äußeren Sicherheit der Bundesrepublik“

unterscheidet (§ 1 Nr. 3). Nach dem Verständnis

des VG Frankfurt wäre die individualrechtsschützende

Regelung zum Schutz von Leben und Gesundheit bereits in

der Regelung zur „ inneren Sicherheit der Bundesrepublik“

enthalten, die Regelung in § 1 Nr. 2 AtG also überflüssig.

Das kann nicht richtig sein. Schließlich bezieht sich der

Schutzzweck des Ausfuhr tatbestandes in § 3 Abs. 3 AtG

ausdrücklich darauf, dass die Kernbrenn stoffe, nicht in

einer „die innere und äußere Sicherheit der Bundes republik

gefährdenden Weise verwendet werden“. Ist eine missbräuchliche

Verwendung offensichtlich aus geschlossen,

kann darüber hinaus kein weitergehender Prüfmaßstab

nach Maßgabe des deutschen Rechts angelegt werden.

Eine solche „Prüfung“ exterritorialer (technischer) Sachverhalte

am Maßstab deutschen Rechts wäre bereits mit

allgemeinen unions- und völkerrechtlichen Grund sätzen

nicht vereinbar. Das Grundgesetz gebietet es, fremde

Rechtsordnungen und – Anschauungen grundsätzlich zu

achten, selbst wenn sie im Einzelnen nicht mit den

deutschen innerstaatlichen Auffassungen übereinstimmen

(BVerfG, Beschl. Vom 15.10.2007, – 2 BvR 16880/07 –).

Das VG Frankfurt geht daher prinzipiell fehl, wenn es

meint, eigene Über legungen zur Sicherheit von aus ländischen

Kernkraftwerken anstellen zu können, um daraus

etwas für Klagerechte Privater abzuleiten. Nicht haltbar ist

schließlich die Argumentation des Gerichts, unter Verweis

auf den „ Atomausstieg“ in Deutschland eine „erweiterte“

Inter pretation des Ausfuhrtatbestands aus § 3 Abs. 3 AtG

zugunsten Dritter vornehmen zu können. Die „geordnete

Beendigung der Nutzung der Kernenergie“ als Zweck des

Atomgesetzes in § 1 Nr. 1 AtG legitimiert nicht dazu, eine

spezielle Gesetzesregelung zur Ausfuhr von Kernbrennstoffen

in § 3 Abs. 3 AtG, die enge („gebundene“)

Genehmigungsvoraussetzungen aufweist, nach „freiem

Ermessen“ plötzlich contra legem anzuwenden. Mit dem

deutschen Atomausstieg hat der Gesetzgeber weder die

Befugnis noch den Willen zum Ausdruck gebracht, den

Betrieb von Kernkraftwerken außerhalb des eigenen

Hoheitsgebiets zu verhindern oder zu erschweren. Das

kann er und das darf er nicht. Denn ein solches Vorgehen

wäre weder mit bindenden unions- noch mit geltenden

völkerrechtlichen Verpflichtungen der Bundesrepublik zu

vereinbaren. Der deutsche Atomausstieg ist und bleibt auf

Deutschland beschränkt. Im Ergebnis lässt § 3 Abs. 3 AtG

unter allen Gesichtspunkten keine Auslegung zu, um einem

Privatkläger Drittschutz gegen eine Ausfuhrgenehmigung

nach dieser Bestimmung zu vermitteln.

IV. Fazit

Die Entscheidung des VG Frankfurt vom 16. Oktober 2020

überrascht: Sie gelangt unter Verkennung grundlegender

Vorgaben des Außenwirtschafts-, Atom-, Unions- und

Völkerrechts zu einem nicht tragbaren Ergebnis. Die

Vorgaben des Rahmengesetzes aus § 14 AWG im Verhältnis

zum Atomrecht werden nicht gesehen. Die entscheidungserhebliche

Frage, ob ein Privatkläger überhaupt die

Rechtsmacht hat, Ausfuhrgenehmigungen nach § 3 Abs. 3

AtG zu „vereiteln“, wurde prozessual unzulässig „offengelassen“.

Die anschließend angestellte Abwägung der

„gegenläufigen Interessen“, die im Ergebnis zum Nachteil

des Herstellers ausfallen soll, fußt durchweg auf kaum

vertretbaren Rechtsauffassungen. Es bleibt zu hoffen, dass

diese Entscheidung baldmöglichst korrigiert wird: Zur

Rehabilitation der dadurch in Frage gestellten Rechtsgrundsätze

und im Interesse eines auch zukünftig fairen,

funktionierenden Außenhandels in Deutschland.

Author

Prof. Dr. Tobias Leidinger

Rechtsanwalt und Fachanwalt für Verwaltungsrecht

Luther Rechtsanwaltsgesellschaft

Graf-Adolf-Platz 15

40213 Düsseldorf

tobias.leidinger@luther-lawfirm.com

Spotlight on Nuclear Law

Export of Non-irradiated Fuel Elements – Indicator for Fair and Functioning Foreign Trade in Germany ı Tobias Leidinger


atw Vol. 65 (2020) | Issue 11/12 ı November/December

BioKernSprit

Jochen K. Michels

557

Introduction As commendable as all the efforts maybe to supply our society and industry with

environmental energy, they will not nearly suffice. This is most easily seen in the quantities of

imported primary energy, i.e. gas and oil, and increasingly electricity too. There is an occasional

flattening or a dent, but the increase is obvious. Considering the needs for more comfort,

communication, mobility, it quickly becomes clear that an increase rather than a decrease is to be

expected – despite all efforts to save.

This proposal addresses an important

sector: mobility. And here we focus

on road-based individual traffic rather

than rail-based mass transport or

airborne services.

BioKernSprit 1 refers to the project

(or proposal) to combine known and

proven processes in such a way that

they can bridge the coming 30 to

80 years. That is, as long as the scarce

and criticized fossil fuel supply will

not be fully replaced by other fuels.

These processes are associated with

the names of Heisenberg, Fischer,

Tropsch, Bergius, Pier, Schulten,

Kugeler and many others. They are

still familiar to many Germans from

school and university. All of them

have earned merits and fame between

1900 and 2000, but their pertinent

developments are not used in

Germany currently. Only in China

there have been approaches for

15 years to implement at least a part

of it into practice. Just recently

also the US resume research and

implemen tation projects for this

particular flavor of nuclear energy,

called “high temperature reactor

with pebble fuel”.

Basics

The basic idea is to combine these well

proven developments, methods and

inventions into one productive and

economical process to make mobility

ecological and affordable for a long

future. It should solve some of

the most urgent needs not only in

Germany and Europe but also in

other countries, even those with

limited resources in developing and

threshold regions.

The hydrogenation processes of

Fischer and Tropsch (FT) as well as

Bergius and Pier (BP) synthesized fuel

from coal and wood. Also other carbon

compounds were tested and proven as

input, e.g. wastewood. In Germany

4 billion liter of car fuel were produced

in one year (1944) by some 14 factories.

This was already 10 percent of

today’s consumption of car fuel. The

only outstanding detriment was, that

almost half of the input feedstock was

burned (oxidized) to produce the necessary

high temperature for the process.

Today this is unthinkable because

of the large CO 2 load. The reason is

well known with all chemists: if you do

not have the optimal catalyst, the

hydrogenation can only be reached by

massive high temperature heat.

For example: the sun does the

same. It converts carbon dioxide from

the atmosphere into burnable plants

(wood, eatables etc.) by low temperature

below 50 degrees Celsius. Chlorophyll

is the ideal catalyst and for plants

this is acceptable. There is time enough

for a slow hydrogenation process and

the resulting greens have a rather low

energy content per kilogram.

Today’s challenge – and answer

For car fuel we want – and need –

faster results and a much higher

density of energy. So we require much

higher temperature and/or a much

more efficient catalyst. This catalyst

has not been detected yet. Until better

results develop we need for hydrogenation

a rather high temperature –

about 900 centigrade 2 .

We propose to combine the proven

– and continuously improvable –

synthesis of FT/BP hydrogenation with

HTR-heat into an overall economical

production line. This complies with

business and market constraints, as

well as with environmental, social and

compliance regulations. Even the

German Atomgesetz (nuclear law)

does not explicitly forbid this application

of nuclear energy. Just electricity

is forbidden.

First – the necessary feedstock

Input material must be found and

provided in our current natural

environment. It seems that about 5 to

10 percent of the national fuel

consumption can be gained from

today’s bio-waste. Mostly wood, also

plastics, blast furnace gas and other

feedstocks serve as a source. Lignite

and hard coal can be used to increase

the initial quantities. Even the

Planned entry for

“Coal- Exit” can be softened by using

coal for hydrogenation. Other sources

may be developed as time progresses

and experience grows. There is no

promise to cover 100 % of our fuel

needs in the foreseeable future. We

want to make a considerable contribution

with minimal economic impact.

With the help of the forest owners’

associations, the above calculation

was made. It shows that about 5 percent

is already achievable today from

wood waste. This can be expanded by

using fallow land and special plants

without food competition.

The Viessmann company has been

showing what is possible for years.

There, 1 hectare of rolling forest

supplies around 5,000 litres of heating

oil or diesel per year.

Second – Sizeable quantities

of hydrogen

Since our proposed method does not

burn feedstock it needs additional H

for input. Currently H is mostly produced

by the Linde process from

fossil gas and other input. Obviously

for ecological reasons and import

dependency this cannot be the solution.

But with the proposed high

temperature heat and nuclear electricity,

hydrogen can also be produced

by cracking of normal water – a sustainable

method, researched and

developed in Jülich decades ago. This

electrolysis currently is rather expensive

because of the electricity needed.

But when heat and electricity are

provided by a high temperature reactor

(HTR) we can overcome this obstacle.

Third – High temperature heat

This necessary heat must be provided

without burning fresh or fossil carbonates

(wood, plants, and coal).

These input materials are too valuable

to just burn them. They should be

converted completely into highly

precious fuel for mobility.

So the necessary heat must come

from another source. High temperature

gas cooled reactors (HTGCR)

offer themselves as an almost ideal

1) BioKernSprit is an

acronym for:

Synthesized Car fuel

from Bio-Waste and

Coal by Hydrogenation

using high

temperature

nuclear heat.

2) Some people claim

to do it with lower

temperature and

optimized catalysts

– not proven yet in

industrial dimensions.

ENERGY POLICY, ECONOMY AND LAW

Energy Policy, Economy and Law

BioKernSprit ı Jochen K. Michels


atw Vol. 65 (2020) | Issue 11/12 ı November/December

ENERGY POLICY, ECONOMY AND LAW 558

3) This video

https://www.youtube.com/watch?v=

4A1uoJ1Z5iA

shows how the heat

steeply degrades

after excursion.

source of energy. While the usual

water reactors with approx. 400 centigrades

provide sufficient heat for

power turbines, the high temperature

reactor provides heat up to approx.

1,000 degrees. This can be increased

even further, if and when better

alloys for the tubes and valves are

developed. There are many concepts

and designs of HTGC Reactors since

the 1960-es. Some of them are

discussed presently in the Gen IV

projects. The HTR with pebble bed

technology seems to be the optimal

choice because its technology:

p obeys all proliferation regulations,

p can be operated decentrally close

to residential areas and industries,

p offers continuous operation essential

for any ongoing chemical production,

p has fuel elements of tennis ball

shape that allow continuous flow

through

p allows well controlled cycling of

fuel elements from top to bottom

and out

p does not need large monocoque

structures, but can work as small

modular units

p allows small and local operators instead

of large concerns

p offers secure repository because

the fuel elements themselves are

gas tight

p will in emergencies:

P cool down without human or

technical involvement

P return to normal temperature

just by the laws of physics

p has been proven in two provoked

“meltdowns” in Germany and one

in China

p allows that remaining risks can be

covered by commercial insurers

There are some more advantages of

this technology. Since there is no need

for safety gadgets these investments

and their maintenance simply do not

exist. The result: the total investment

will be lower. This leads to lower

operating expenses (opex) and lower

error susceptibility.

Why is this technology safer

than others?

The key lies in the shape and construction

of the fuel elements. Also

important is the design and construction

of the building.

The fuel itself is contained in small

particles of under 1 mm in size. These

particles are each containing the

active heavy metals uranium and/or

thorium. Each of them is coated with

three layers (TRISO) of gas tight

pyrolytic carbon and silicate carbide,

very stable layers. Some 20,000

particles are contained in the “pebble”

which is very stable too, against shock,

abrasion and heat.

Should an emergency occur, such

as missing coolant gas helium, they

would heat up to about 1.500 centigrades,

which results in more neutron

capture, so the chain reaction will

stop 3 . The residual heat then will

be cooled by heat transfer finally to –

and out of – the building structure

and into the surrounding air. This

necessitates a well-designed building

structure. Up to now steel pressure

vessels have proven optimal for sizes

up to about 100 MWe (in Juelich and

Tsinghua) and prestressed concrete

for sizes above that capacity (in

Hamm).

Another fact is, that the low density

of energy compared to most current

reactors makes the cooling faster and

easier. Compared with the absolute

safety this is neglectable.

Also the oven (instead of kiln)

principle has important safety aspects:

never is more fuel in the reactor, than

is used at any given moment. This

reduces the quantity of radioactive

material during the whole process.

That this principle also allows to

eliminate refilling stops is another

advantage, both economical, eco logical

and from a safety aspect.

Even in case where bombs or other

large objects would hit such a reactor

and would open the primary cycle,

there is only a small quantity of inert

helium that can be released to the

environment, with very little contamination.

Water is not present in

the primary cycle of the reactor. Some

hot steam out of the secondary cycle

of the heat exchanger could get

in touch with the helium and the

pebbles if worst comes to worst.

This will not cause enough oxyhydrogen

for a large explosion. Since

pebbles and particles are gastight and

quite unbreakable, the steam and

the environment would have only

minimal contamination for some

hours or days. The radioactive load

will not be much above normal

environmental levels.

Two critical points

Final storage is often demanded

without specifying its real meaning.

Doubtless it does not mean “until the

end of the Universe”. Rather one thinks

of a long time, e.g. 1 Million years.

Much better would be a definition

like: “until normal environmental

radiation is reached”. Now normal

radiation varies from about 15 mSv

(milliSievert) to more than 100 mSv

or the equivalent in Becquerel.

Current calculations of experts

in nuclear physics say that about

300 years would be a realistic time for

the spent pebbles to reach this level.

In Ahaus I saw the Castors with

pebbles standing in a hangar for

about 40 years without problems. In

Jülich the reported situation is

comparable. Also research is going

on and should be strengthened to

partition and/or transmute the

radioactive waste. So within those

300 years one might be happy to find

the waste in a secure intermediate

storage for reuse.

Another topic of endless discussion

is the residual risk and its insur ability.

None of about 440 current traditional

nuclear power plants is commercially

insured. When Fukushima had cooled

down, Prime Minister Abe told the

world, that each of the three reactors

molten down had caused about

50 billion USD in damage, so a

total 150 Billion USD was lost. The

Japanese nation, the taxpayer had to

shoulder this sum. In any comparable

accident this will be similar. This risk is

commonly understood by “residual

risk”. And it can happen with all

reactors, whose safety depends on

either human or mechanical or

electronic interaction.

Not so with the pebble bed technology.

Because it is inherently safe, this

risk simply does not exist. It needs not

be covered neither by the society nor

by an insurer. The other risks, e.g.

building damage, business continuity

will be covered by normal industrial

insurance.

How about economics?

We have to accept, that there is little

knowledge about some important

factors, which go into the business

case:

1. Nobody can tell us what is the cost

of a FT Hydrogenation plant

2. For the cost of a Pebble-Bed Reactor

we have some exaggerated figures

from Hamm, because of politically

influenced tests and delays

3. Operating cost are mainly influenced

by the costs for

a. Personnel

b. Heavy metal (Uranium, Thorium

etc.)

c. Interest rates

In the following tables some details of

the economic calculation are offered.

The complete calculation cannot be

presented here because of its size,

but is available to any interested

party by mail.

Energy Policy, Economy and Law

BioKernSprit ı Jochen K. Michels


atw Vol. 65 (2020) | Issue 11/12 ı November/December

Important components and

factors:

The goal is to produce about 1 billion

liters of ethanol or methanol per year

in an industrial complex consisting of

a modular pebble-bed reactor and a

FT hydrogenation plant:

p Pebble bed reactor, three modules

at 100 MWe (approx. 250 MWth)

each. The necessary heat exchangers,

piping, tubes, the

necessary apparatus for control

and operation.

p Hydrogenation plant with a

capacity of one billion liters of

fuel per year.

p 30 years to amortization, 7.5 percent

interest

p 180 persons at annual gross personnel

costs of 70,000 Euro

p Raw fuel costs (world market)

for yellow cake approx. 50 USD

per pound

p Hydrogen costs per year of

100 million Euro,

p Coal input per year of 70 million

Euros.

The calculation for the pebble-bed

reactor assumes construction costs of

800 million, which corresponds to a

rate of just under Euro 2,670 per kWel.

This is roughly the cost rate of the

THTR operated in Hamm, ex cluding

the externally imposed costs – essentially

for duplicating unnecessary

audits. 4

For the calculation shown

here, it is assumed that the learning

curve and any development costs are

not included in full. The calculation

does not refer to a FOAK 5

construction.

For the construction phase of

5 years, an additional third of the

construction sum is estimated, since

this also has to be financed. The

same applies to the decommissioning

costs. This results in a total rate of

3,470 per kWel. Depreciation is

estimated at a 30 th per year, because

the useful life of the building is

expected at least for 30 years. At

present, even conven tional nuclear

power plants abroad are approved

for operation for up to 60 years or

longer. The relatively high interest

rate of 7.5 percent has to be agreed

with the lenders in a specific case and

is likely to be lower at present. It is

deliberately not calculated “on the

brim”.

The operating costs are based on

experience with other power plants.

100 people are probably the upper

limit, the average gross personnel cost

rate of Euro 70,000 annually includes

all qualifications from the management

to the gatekeeper as an average

value.

Economic calculation for the Pebble Bed Oven

The market price for commercial

yellow cake uranium usually fluctuates

around USD 30 to 40 per pound. The

price of USD 50 chosen here is

the result of commercial caution.

The price of Thorium is even more

favorable. The production of coated

particles and spheres will be assumed

at twice the price of the material until

reliable values are available. The same

procedure will be followed with the

disposal costs, which are not comparable

with today’s final storage costs

due to the “cool down” storage concept.

The maintenance with an acceptable

12 percent of the construction

costs is also in line with the usual rates

in the industry, although due to the

lack of maintenance-intensive safety

devices, only little effort is to be

expected.

Based on these values, an average

internal cost price of 0.042 Euro per

kWht and total costs of around

95.5 million Euro per year is cal cu ated

for process heat. A further Euro 50 million

is added for pre heating, i.e. a

total of Euro 145.6 million per year, as

shown below in the hydrogenation

calculation.

With this heat supply the capacity

of the reactor is not fully used. The

additional energy is available for

power generation and district heating.

They generate further contribution

margins of around 93 million Euros,

so that the total costs of 238 million

Euros will be recuperated. These rates

would have to be increased if internal

profits were still to be made.

The economic calculation for the

hydrogenation plant is based on a

capacity of around 960 million liters

of fuel per year. Due to the lower

energy content of ethanol or methanol,

this quantity corresponds to

approximately 672 million liters of

petrol (Benzin).

For the construction, including

the construction phase, 2.7 billion

Euros are assumed here, resulting

in annual capital costs of just under

169 million Euros. The operating

costs include personnel costs for 80

people of 5.6 million Euros annually

and energy costs, as calculated

above, of 145.6 million Euros. Added

to this are material costs for – in

this example – 1.12 million tons of

lignite at 67.5 million Euros and

gaseous hy drogen of 5 billion liters

at 100 million Euros, maintenance

200 million Euros and miscellaneous

50 million Euros. This results in total

MWth750

assumed efficiency 0.4000 MWel300

Capital costs for the investment Calculation factors All in Euro

Construction phase approx 5 years 0.3340 267,200,000

Construction costs complete 800,000,000

Provisions for decommissioning, repository, dismantling 0.3000 240,000,000

Building costs as sum of construction and provision 1,040,000,000

Construction costs per megawatt electrical 3,466,667

Useful life in years 30

Total investment = sum of construction costs and

construction phase

1,307,200,000

Depreciation per year (total investment / 30 years) 43,573,333

Annual interest charge 7.50% 49,020,000

Annual capital costs 92,593,333

Ongoing operational costs

Personnel headcount 100

Annual gross personnel costs 70,000 7,000,000

Material costs

Market price for yellow cake (U3O8 at 50 USD/pound) Per ton 101,284

Annual consumption of yellow cake tons 75

Input energy of uranium (at 20 million KWh per Kg) kWh per Jahr 13,500,000,000

Fuel Costs for uranium/thorium p.a. 7,596,330

Production of pebbles and coated particles 200 % of fuel costs 15,192,661

Treatment and disposal of elements 200 % of fuel costs 15,192,661

Other costs 5,000,000

Maintenance as a percentage of the construction sum 12.00% 96,000,000

Preheating with lower leve heat form the “oven” 50,000,000

Annual operating costs 145,981,651

Total costs – per year 238,574,985

4) Prof. Dr. Knizia

in atw.

5) First of a Kind.

ENERGY POLICY, ECONOMY AND LAW 559

Energy Policy, Economy and Law

BioKernSprit ı Jochen K. Michels


atw Vol. 65 (2020) | Issue 11/12 ı November/December

ENERGY POLICY, ECONOMY AND LAW 560

Economic efficiency calculation hydrogenation plant

(Fischer-Tropsch / Bergius-Pier)

Cost of capital for the investment

Liter fuel 961,094,746

Which corresponds to about 0.7000 Liter petrol 672,766,322

Dimensions and factors

Euro, if not otherwise

Construction phase approx 5 years 0.3340 668,000,000

Building costs complete 2,000,000,000

Useful life years 30

Total investment = sum of construction costs and

construction phase

2,668,000,000

Depreciation per year 88,933,333

Interest per year 6.00% 80,040,000

Annual capital costs 168,973,333

Ongoing operational costs

Personnel 80

Annual gross personnel costs 70,000 5,600,000

Material costs

Available energy (high temperature heat from PBR)

highest temperature heat from the PBR kWh 2,294,244,000

Medium-temperature heat from the PBR kWh 1,204,478,100

Total energy supply from the reactor kWh 3,498,722,100

Costs for this energy supply Euro 145,530,741

possible fuel production with this

Type of liquid fuel

Ethanol

Energy content per kg kWh 8.3

Efficiency of the hydrogenation process 90.00 %

Energy content of the generated fuel kWh 6,297,699,780

Annual quantity of finished product (fuel) Kilogramm 758,759,010

Special weight of one liter kg 0.75

Annual quantity of the final product Liter 1,011,678,680

Input-Material

Type of available feedstock

Lignite coal

Energy content per kg kWh 5.6

Required annual quantity by energy content kWh 6,297,699,780

Required annual quantity by weight kg 1,124,589,246

Required annual quantity by weight to 1,124,589

Market price of input material (lignite) to 60

input material total costs annually 67,475,355

HT-Energy from the pebble bed oven see above 145,530,741

Hydrogen supply Liter 5,000,000,000

Market price of hydrogen Euro per Liter 0.020

Cost of hydrogen per year 100,000,000

Total material and energy costs 313,006,095

Maintenance 10 % 200,000,000

Total operating costs 518,606,095

Other / unforeseen 50,000,000

Total capital and operating costs 737,579,429

Output and price

Production per year Liter 1,011,678,680

thereof waste, shrinkage in percent 5.00%

of which waste, shrinkage in liters 50,583,934

Remaining usable quantity (liters of ethanol) 961,094,746

Price per liter of ethanol ex plant Euro/ Liter 0.7674

pension finan cing) become irrelevant

with such sustainable production.

The example quantified above

essentially demonstrates the simpli city

of such a calculation. Thus, this first

scenario, for

Reactor investments of

1,307 million Euros

a hydrogenation plant Investment of

2,000 million Euros

an internal price of 1 kWh of heat

0,042 Euro and

a factory outlet price for one liter of

fuel


0.76 Euro

If the parameters are changed, the

final results are of course different as

follows:

Medium scenario

Reactor investments

2,600 million Euros

Hydrogenation plant Investment

2,000 million Euros

Internal price of 1 kWh heat

0.084 Euro

Factory outlet price for one liter of fuel


1.12 Euro

High scenario

Reactor investments

4,000 million Euros

Hydrogenation plant Investment

2,000 million Euros

Internal price of 1 kWh heat

0.13 Euro

Factory outlet price for one liter of fuel


1.48 Euro

Very high scenario

Reactor investments

6,500 million Euros

Hydrogenation plant Investment

6,000 million Euros

Internal price of 1 kWh heat

0.21 Euro

Factory outlet price for one liter of fuel


2.14 Euro

The price ex-factory therefore varies

between 0.75 and 2.14 Euros per liter

of fuel. It can be assumed that the

reality will result in the lower price

range when the FOAK phase is over

after 10 years and a sufficient number

of such complexes are in operation. In

view of the reduced dependence on

energy imports and the safeguarding

of highly qualified jobs in Germany

for a technology with worldwide opportunities,

this technology appears

worthwhile considering.

annual costs of approx. 737.5 million

Euros.

With an output of 961 million liters

as mentioned above, the liter of fuel

costs 0.77 Euro ex works. Converted

to the liter of petrol, the equivalent

price would be 1.1 Euro. Although

this price is higher than the current

price of petrol ex refinery, a

number of government charges can

be elimi nated because the reasons

for them (environment, energy,

Author

Jochen K. Michels

jochen.michels@jomi1.com

Management Consultancy

Konrad-Adenauer-Ring 74

41464 Neuss, Germany

Energy Policy, Economy and Law

BioKernSprit ı Jochen K. Michels


atw Vol. 65 (2020) | Issue 11/12 ı November/December

The Scientific Backing of the

German Quiver Project

Wolfgang Faber, Marc Verwerft, Janne Pakarinen, Hagen Höfer and Christoph Rirschl

1 Brief history of the German quiver project and background for

experiments In the German history of atoms for peace the back-end perspective changed

substantially several times. Although it seemed far away when the plants started commercial

operation in the seventies, a moving target on such a long lead item can easily create problems. All

German commercial plants (table 1) were equipped with an unlimited license and there was not a

strict design lifetime.

Plant Type Utility MW el BOL EOL

Brunsbüttel KKB BWR VENE 806 1977 2007

Krümmel KKK BWR VENE 1402 1984 2009

Stade KKS PWR 15x15 PEL 672 1972 2003

Brokdorf KBR PWR 16x16 PEL 1480 1986 2021

Unterweser KKU PWR 16x16 PEL 1410 1979 2011

Emsland KKE PWR 18x18 RWE 1406 1988 2022

Grohnde KWG PWR 16x16 PEL 1430 1985 2021

Würgassen KWW BWR PEL 670 1975 1994

Mühlheim-Kärlich KMK PWR 17x17 RWE 1302 1987 1988

Biblis A KWB-A PWR 16x16 RWE 1225 1975 2011

Biblis B KWB-B PWR 16x16 RWE 1300 1977 2011

Grafenrheinfeld KKG PWR 16x16 PEL 1345 1982 2015

Philippsburg 1 KKP-1 BWR EnBW 926 1980 2011

Philippsburg 2 KKP-2 PWR 16x16 EnBW 1468 1985 2019

Neckarwestheim I GKN-I PWR 15x15 EnBW 840 1976 2011

Neckarwestheim II GKN-II PWR 18x18 EnBW 1400 1989 2022

Obrigheim KWO PWR 15x15 EnBW 357 1969 2005

Gundremmingen B KRB-B BWR RWE 1344 1984 2017

Gundremmingen C KRB-C BWR RWE 1344 1985 2021

Isar 1 KKl-1 BWR PEL 912 1979 2011

Isar 2 KKl-2 PWR 18x18 PEL 1485 1988 2022

| Tab. 1.

Start and end of operation of the commercial nuclear power plants in Germany, excluding the

Greifswald units. KMK and KWK are not relevant for the process discussed here (grey), BOL of the oldest

three plants highlighted in red.

Therefore, in the first decade of

operation emphasis was put on

enhancing enrichment to push

burnup, reducing spent fuel as a side

effect. Typically, the German plants

have a wet-storage able to accommodate

fuel assemblies over about

10 years of operation and thus fuelback-end

was not a topic with reprocessing

being required by law to make

use of the valuable resources in spent

fuel. The transport of spent fuel

to La Hague and BNFL as well as the

corresponding transportation of reprocessing

waste to the German

central storage facility at Gorleben

were the consequence of such mandatory

fuel-recycling. In 1997/1998

contamination on such transport casks

was detected, causing Angela Merkel,

back then the Federal Minister for the

Environment and Reactor Safety, to

stop transportation in May 1998. This

was in place until 2000, when Jürgen

Trittin, successor of Angela Merkel as

environmental minister, temporarily

allowed spent fuel transports in the

wake of the “Atomkonsens”, which

meant limiting the plant-lifetime to an

equivalent of about 32 years, stop of

reprocessing and erection of local dry

storage facilities to collect all spent fuel

until end-of plant life. With abandoning

reprocessing the conditioning

of failed fuel was a topic, but, considering

the time, one of rather

Planned entry for

low priority, because for Biblis A,

Neckarwestheim I, Brunsbüttel, Isar-1

and Unterweser, the oldest units next

to Stade and Obrigheim, which were

decommissioned in 2003 and 2005

making a local dry storage unnecessary,

shut-down was still more than

10 years ahead. Instead, power uprates

to increase installed capacity and the

realization of the IAEA-96 regulations

concerning transport casks on the

back-end side were the projects of

the early 2000s with highest priority.

Later, the dreams of a nuclear renaissance

took over, culminating in the

long-term operation of older plants

(beginning of life bol


atw Vol. 65 (2020) | Issue 11/12 ı November/December

DECOMMISSIONING AND WASTE MANAGEMENT 562

In July 2011 the German quiver

project as described here was started

by the utilities, asking GNS to review

the 2010 concept with respect to

regulatory robustness by increasing

margin and conservatism. Main

characteristics of this concept are the

gas-tight welded quiver as part of the

cask, not merely inventory, and the

exclusive loading of such quivers in

a CASTOR-cask without any fuelassemblies

besides. In August 2011

the concept was approved and

finalized in a joint workshop GNS/

utilities and a specification was

created, which was used in a call to

five potential vendors. Until July 2012

the vendors’ concepts were evaluated

and the utilities decided to place the

order with Höfer&Bechtel, a German

engineering company engaged in a

variety of service- and development

projects in power stations, GNS

managing the quiver-development on

behalf of the German utilities. Within

5 years the specification was turned

into a pilot-quiver that was subject to

fall-testing at -40 °C and +90 °C, a

transportable hot-cell for remotewelding

and leak-testing on top of a

transportable shielding to house the

quiver during heat-assisted vacuumdrying

in clu ding all regulatory activities

resulting in the issue of the design

approval in 2017 for PWR and in 2018

for BWR [1-3]. Within about 1 year

the industrial realization for PWR

took place which aimed at a first

campaign at Kernkraftwerk Biblis

(KWB) mid of 2018, to be followed

immediately by Kernkraftwerk Unterweser

(KKU). For reasons of practical

optimization, the places were

switched and so the first campaign

took place at KKU in October 2018,

Biblis following in January 2019 [4].

In 2011/12, taking said decision

about the concept and the realization,

utilities evaluated the risk of the

project. The time when a decommissioned

power plant does no longer

contain fuel (“fuel-free”), is a key milestone

in the decommissioning project,

delays causing financial damage

due to the necessity to maintain the

technical and regulatory status of a

facility housing high active nuclear

fuel, which constitutes about 99 % of

all radiologic activity in operating

status. While the costs of a delay are

comparatively easy to estimate, the

risk of failure in realizing a project, as

the quiver-project was in 2012, is the

more difficult to tackle. Technical

equipment can fail, unsuccessful tests,

material not suitable and licensing

might take longer than expected. The

first area must be taken care of by the

vendor accepting the offer, the second

has to be borne by his customer. The

generic questions of 2012, posing a

regulatory risk, were “by which means

and to which amount is water, which

has entered the cladding of a fuel rod,

accessible and removable from there,

and what characterizes the fuel rod

with respect to the removal of water”.

The evaluation of these days had to

rely on relatively old literature, e.g.

[5, 6], which at least allowed a positive

prognosis. Now, while generic research

is not the domain of private

utilities, they often participated in

research projects to gain and maintain

the engineering capability as a

cognizant and responsible operator.

Drying failed fuel does not exactly fit

into that reasoning, in this case it was

mitigation of generic risks in the

process by gaining timely information

that allow to react and avoid larger

losses. While the quantitative assessment

of how much should be invested

in risk-mitigation is well established

in finance and insurance, it is not

regularly applied to the nuclear industry

but can be adapted [7]. Following

an internal evaluation Eon Kernkraft,

since 2015 named PreussenElektra,

decided to initiate a research project

aiming to answer the above raised

questions. From a couple of research

institutes SCK CEN was chosen to

realize the set of experiments and

shortly after starting the project

Synatom, Belgium, joined as a

cofinancing partner, but with an own

agenda. The Synatom specific experiments

are not reported here. After

successfully finishing the first set of

experiments with one sample it was

decided to continue the project by

measuring three additional samples

from another fuel rod, financed by

GNS on behalf of all the German

nuclear utilities EnBW, VENE, RWE

and EKK (now PEL).

2 The WETFUEL

Experiments

2.1 Equipment

In 2012 phase I of the experiments

started by defining the overall goal

and devising the experimental setup.

The final goal was to demonstrate the

feasibility of drying failed fuel using

heat assisted vacuum drying and,

when Synatom had joined, hot gas

bathing (not described in this paper),

as well as characterize spent fuel rods

with respect to water removal. In the

course of the project, after successfully

demonstrating the feasibility of

drying in general, it was decided to

quantify water removal rates. A vessel

was designed that would allow testing

of a fuel rod sample of maximum

50 cm length [8]. This vessel consists

of two separate compartments, a top

and bottom reservoir, which communicate

only by the void volume of the

fuel stack in the fuel-rod sample

( Figure 1, 2).

The pressure of each compartment

is monitored and can be evacuated

individually. In the vacuum exhaustline

a cold trap and a dew-point

monitor were installed. Both reservoirs

can be filled with water or Ar or

He gas independently. The vessel was

equipped with heaters and a maximum

temperature of 130 °C could be

| Fig. 1.

Experimental setup of the WETFUEL experiments. Two separate compartments are connected by the

fuel-rod sample and can be evacuated individually. Water can be injected into both compartments [9].

Spent fuel segment rod is shown with red line in the schematic.

Decommissioning and Waste Management

The Scientific Backing of the German Quiver Project ı Wolfgang Faber, Marc Verwerft, Janne Pakarinen, Hagen Höfer and Christoph Rirschl


atw Vol. 65 (2020) | Issue 11/12 ı November/December

| Fig. 2.

WETFUEL setup installed in hot-cell [9].

| Tab. 2.

Characterization of fuel samples.

applied based on the hot-cell safety

case.

2.2 Samples

Four different samples were investigated

to study vacuum drying (WET1,

WET2, WET3, WET5b) and some

segments were tested only for their

hydraulic properties (WET4, WET5a,

WET5c). The WET1 sample was used

in phase 1 (SCK CEN/PEL/Synatom)

whereas the other segments were

used in phase 2 (SCK CEN/GNS).

The sample for WET1 was cut

from rod FT1X34-D04, irradiated in

Tihange 1 (PWR 15x15 2865 MW th )

over two 18-month cycles (cycle

20: 07.04.1998-28.08.1999 and cycle

21: 27.09.1999-04.03.2001) to an

average burnup of 49 MWd/kgHM

(see Table 2 and Figure 3).

Rod FT1X34-D04 suffered from a

cladding perforation in its second

cycle at elevation 250 cm. The openended

sample was cut at elevation

110-160 cm with a local burnup of

52 MWd/kgHM. Fuel bonding to the

cladding inner surface is sufficient

to keep the fuel from falling out in

handling these samples. Cladding

material was low tin Zircaloy 4. Figure

4 shows a cross section of rod

FT1X34- D04 at elevation 1100 mm,

displaying an average (cold) gap of

12 µm, effectively closed at 0° and

| Fig. 4.

Cross section of rod FT1X34-D04 at elevation

1100 mm. Top is 0 °, right 90 °, bottom 180 °

and left 270 °.

270° but with a gap of 30 µm and

15 µm at 90° and 180°, together

with a typical crack pattern. The

gap widths were measured using

Scanning Electron Microscopy (SEM)

images besides Figure 4.

Additional samples for the continuation

project were cut from a

different rod FT1X57-D05 from

another Tihange1 fuel assembly

irradiated in the same two cycles than

FT1X34- D04. This rod did not fail in

operation and had a slightly higher

average burnup of 51 MWd/kg. Three

samples were cut to 50 cm, 17 cm and

10 cm length (local burnup for the

50 cm and 10 cm sample 54 MWd/kg,

| Fig. 3.

Axial burnup distribution and sample WET1 (R2 segment) position of rod FT1X34-D04. The arrow

indicates the elevation for cross-section that was investigated using optical and electron microscopy.

WET1 WET2 WET3 WET5b

Rod ID FT1X34-D04 FT1X57-D05 FT1X57-D05 FT1X57-D05

Condition defect intact intact intact

Rod average burn-up (MWd/kg) 49 51 51 51

Segment burn-up (MWd/kg) 52 54 54 43

Cladding material low-Sn Zry-4 recrystallized M5 recrystallized M5 recrystallized M5

Segment length (cm) 50 50 10 17

Position from rod bottom (cm) 110 - 160 114 - 164 164 - 174 18 - 35

43 MWd/kg for the 17 cm segment)

to allow assessment of the length

dependency of water removal rates.

Another reason for using another

rod was to check the variability in

hydraulic radius of different samples

of about the same burnup.

2.3 Hydraulic resistance

from dry gas flow

The hydraulic radius r of a channel

is usually measured by the pressure

loss Dp =p 1 -p 2 over length L for an

incompressible medium with viscosity

h from the volumetric flow F=dV f /dt

(expressed in m 3 /s) as given in (1)

(1)

For compressible media this must

be corrected for the expansion along

the path with dropping pressure.

Under steady-state and isothermal

con ditions, the volumetric flow for an

ideal gas yield:

(2)

From (2) a mass flow can be calculated

using pdV g /dt=RTdn/dt.

The hydraulic radius of different

samples was measured during the

cold tests by establishing a steady

DECOMMISSIONING AND WASTE MANAGEMENT 563

Decommissioning and Waste Management

The Scientific Backing of the German Quiver Project ı Wolfgang Faber, Marc Verwerft, Janne Pakarinen, Hagen Höfer and Christoph Rirschl


atw Vol. 65 (2020) | Issue 11/12 ı November/December

DECOMMISSIONING AND WASTE MANAGEMENT 564

Rod Δp/mbar flow F

(ml/min)

Machined bar 35 1140 630

Mock-up rod 1057 339 211

| Tab. 3.

Hydraulic cold tests.

flow (F) of Ar from the bottom-to

the top- reservoir while measuring

the pressure loss. The samples were a

50 cm cylindrical bar with a machined

pinhole of about 2 mm and a mock-up

fuel rod, filled with alumina powder,

provided by Höfer&Bechtel. Details

are provided in Table 3. The hydraulic

radius for the machined bar was about

630 µm corresponding with a radius

of about 1 mm, the mock-up fuel rod

had about 200 µm (Table 3).

The same approach was intended

on the fuel rod sample WET1, but

this failed because no steady Ar-flow

could be established, although on

enhancing the bottom pressure (flow

from bottom to top) a very slight

increase of the top pressure indicated

gas communication, i.e. a leak flow.

This was finally used to determine the

hydraulic radius (Figure 5) using


r/µm

(2a)

with dp/dt the pressure evolution due

to leak flow in Volume V versus time,

p 1 and p 2 pressure at the end of

the tube of length L and Radius r, n

number of gas molecules (usually

moles), T gas temperature and R gas

constant.

As a result of the dry gas leak-rate

measurements the fuel rod sample

WET1 showed a hydraulic radius of

about 80 µm. For this evaluation

estimates from drawings for the

volumes of the top and bottom

reservoir and adjacent lines were

used, which were checked by filling

the chambers of a duplicate setup

with water as well as by expansion of a

calibrated volume.

Hydraulic radius was measured

before and after each of the tests to

check whether wetting/drying or

the vapor-flow from water pocket

experiments altered the fuel stack.

Obviously, this was not the case

(Figure 6). Even pushing liquid water

by overpressure of up to 1.5 bar (see

next section) through the fuel stack

did not alter it in a way as to change

the hydraulic radius.

Whenever theoretical vapor

removal rates are compared with

measured ones (see section 2.4) the

hydraulic radius from dry Ar gas leak

measurements were used.

Gas-connectivity through the fuel

rod samples was thus investigated

using leak-rate measurements and the

respective characteristic of the fuel

stack expressed due to (2) in terms of

an effective hydraulic radius. This can

be translated into permeability K

by using Muskat’s formulation of

Poiseuille’s law:



(2b)

(2c)

If a pressure-difference p 1 -p 2 causes

a mass-flow of q in a tube filled

with porous material of length L this

system is characterized by the same

effective hydraulic radius independently

of the cross-section area A of

the tube, whereas the permeability K

is smaller for the bigger A according

to eq. 2c.

2.4 Wetting and drying

sequences

The first set of experiments with

sample WET1 aimed at studying the

drying process itself. Each run started

with inserting water into the lower

reservoir above the lower end of the

fuel-rod sample according to Figure 7

(left hand side). Water was forced

through the fuel stack by a pressure

difference of up to 1.5 bar until water

on top of the upper end was detected

from a conductivity-meter indicating

that water had passed through the

fuel- rod. Then the Ar pressure was

lowered to 1 bar and the system kept

in that way for some hours in order to

give time to fill smaller cracks.

The time needed until the conductivity-

meter gave a signal was

30 minutes (Dp=1.1 bar for 19 min

and 1.6 bar for 11 min) and 45 minutes

(Dp=1.5 bar) in a following experiment.

From that a conservative guess

of the rate of water-ingress can be

derived by taking the void-volume of

the fuel rod sample as well as the

geometry of the upper chamber into

account. At least when all the void

volume, which is estimated to be

0.75 cm 3 for the 50 cm sample based

on thermomechanical calculations, is

filled, water must exit the upper end.

At that point detection depends on the

wetting or non-wetting characteristic

of the water at the interface to the fuel

at the upper end cross section. In case

it forms a small droplet, the conductivity

signal is expected after a total

of less than 1 g of water being percolated.

If it is wetting, it could rinse off

the side of the top cross section and fill

part of the top chamber until the

conductometer sensors were reached,

which could amount to up to 45 cm 3 .

| Fig. 5.

Hydraulic radius of fuel-rod sample measured by leak-rate.

| Fig. 6.

Hydraulic radius from leak tests on sample WET1 before and after tests.

Decommissioning and Waste Management

The Scientific Backing of the German Quiver Project ı Wolfgang Faber, Marc Verwerft, Janne Pakarinen, Hagen Höfer and Christoph Rirschl


atw Vol. 65 (2020) | Issue 11/12 ı November/December

| Fig. 7.

Wetting and drying sequence: water is forced through the fuel stack by up to 1,5 bar until water is

detected by the conductivity meter. Then the excess water is drained, the vessel heated to 120 °C and

both compartments evacuated. The pressure peaks at 6 h and 24 h are pressure rebound tests

according to ASTM [13] and IAEA [10] for dryness.

The system was not designed to

exactly determine the amount of

water that enters the rod, but only to

ensure that the fuel rod stack was wet

to demonstrate the drying procedure.

Nevertheless at least some theoretical

interpretation is intended at that

point.

Water flow in tubes is well

described with (1), and using the

effective hydraulic radius derived

from gas-flow (chapter 2.3) results in

water-transport rates that predict 9 g

for experiment 1 and 16 g for experiment

2, e.g. 18-21 g/h transport rate.

Considering the uncertainty mentioned

above this is consistent with

the experiments, the major portion

apparently being deposited outside

the rod in the upper chamber.

Nettleship [11] described a microscopic

picture of porous material by a

bundle of n circular tubes starting

from Hagen-Poiseuille (1):


(1a)

A open denotes the part of the cross

section open for gas- and water communication,

A the fuel cross section,

V void and V are the corresponding

volumes, r g the hydraulic diameter

from gas-measurements and r c the

radius of the microscopic tubes and n

their number. n and r c are determined

by the measured hydraulic radius in

the gas-flow experiments and the

void-volume from thermomechanical

calculation:

This results in about 5000 tubes

of roughly 10 µm radius, which represent

the microscopic porosity in the

fuel rod sample of high burnup. For

pressure differences of up to 1.5 bar

tubes of less than 1 µm must be considered

water tight, for which there is

obviously no indication here.

Still capillary forces have to be

considered especially in cases with

low pressure difference, see e.g. a

classical treatise by Greinacher [12].

Applying the methodology described

by Greinacher for cases with external

pressure shows that vertically oriented

tubes with 10 µm radius of length

50 cm are filled in less than a minute

from the bottom if pressure of 1.5 bar

over 50 cm is applied in addition

to the capillary force, so that the

application of Hagen-Poiseuille to

describe the wetting experiments is

appropriate.

After filling the rod with water,

the bottom reservoir was drained,

the vessel heated to 120 °C and both

compartments evacuated. Within 1 h

a pressure drop to 10 mbar indicated

that all liquid water was removed,

and a second one at 3 h that most of

the vapor was gone (Figure 7 right

hand side). The dew-point reading

down to -30 °C in the exhaust line

reflect these pressure drops. The

pressure peaks at 6 h and 24 h are

pressure rebound tests for dryness

testing akin to the ASTM standard

guide for drying of spent nuclear

fuel [13] and IAEA guidelines [10].

For this test the connection to

the vacuum pumps are closed and

the pressure increase recorded over

30 min. The volume is considered

dry when the pressure does not

increase by more than 4 mbar. This

test is especially significant in this

case because of the small volume

involved. Both tests were passed,

although the lower second peak

together with a, by about 8 K, lower

dew-point indicates that still some

moisture was removed in the 18 h

following the first test.

These wetting and drying

sequences clearly demonstrated that

water can effectively be removed

from the fuel stack in short time

with heat assisted vacuum drying

when temperatures above 100 °C are

applied.

2.5 Water pocket experiments

The next step was the quantitative

determination of water removal rates,

i.e. how many grams of water per day

are removed and which are the

relevant parameters governing the

process and how they are related with

each other. The general idea is rather

simple: a well-defined amount of

water is injected with a syringe into

the upper compartment, which is then

closed from the vacuum system,

constituting an enclosed water pocket,

and the vessel heated to a desired

temperature. Then the lower compartment

is evacuated, and the

pressure evolution of the lower

vessel is recorded, together with the

dew-point reading (see Figure 8). The

pressure drop indicates that the

flow reduces, due to the absence of

liquid water in the top vessel. The

time needed for the transport of

the injected amount of water, here

10 ml, can thus be derived. More

precisely, the time was taken when

the pressure had dropped to one half

of the initial pressure at the given

temperature.

The pressure curves of Figure 8

display a plateau of different length

depending on the chosen temperature.

The higher the temperature the

higher is the saturation pressure in

the upper compartment pushing the

vapor through the fuel cracks and

gaps. The leak rates that were

observed were between about 0.3 …

6.9 mbar l/s, which correspond

with Reynold-numbers ( )

of 99 … 1743, well below the critical

Reynold number of 2320. Accordingly,

one can expect laminar vapor-flow.

Using the hydraulic radius for the

fuel- rod sample, the length and

the saturation pressure the rate of

laminar vapor mass transport can be

calculated according to eq. 2b.

Figure 9 shows the measured data

for WETFUEL samples WET1, WET 2,

WET3 and WET 5b (Table 2) together

with the theoretical values applying

laminar flow rule, confirming the

assumption. We need to stress the

DECOMMISSIONING AND WASTE MANAGEMENT 565

Decommissioning and Waste Management

The Scientific Backing of the German Quiver Project ı Wolfgang Faber, Marc Verwerft, Janne Pakarinen, Hagen Höfer and Christoph Rirschl


atw Vol. 65 (2020) | Issue 11/12 ı November/December

DECOMMISSIONING AND WASTE MANAGEMENT 566

fact that this is no fit to the data.

The governing parameters are the

hydraulic radius, determined in a

different measurement from gas leak

rates, the sample length and the

temperature, the latter defining the

driving saturation pressure. From

these the theoretical removal rates are

calculated. The two lines of the same

color for each set of parameters mark

the uncertainty in the measured

hydraulic radius. The error-bars of

the measured rates are smaller than

the symbols.

Also added to the figure is the

consequential prognosis for a fulllength

commercial fuel rod. The thick

red line is the expected removal

rate for a “worst-case” real fuel-rod

situation with a lower bound hydraulic

radius and a maximum distance

between water pocket and cladding

breach, demonstrating industrial

feasibility considering the fact that

about 25 cm 3 of water need to be

removed in such a case, which should

take less than 3 days once 140 °C is

exceeded.

Using the laminar flow-rule

assump tion the two phases of water

removal from a water pocket can be

quantitatively described. In Phase 1

there is a constant driving pressure,

removal rate is constant and mass

transport proportional to time. In

Phase 2 the pressure in the water

pocket drops continually and thus

removal rate decreases. Based on

eq. (2b) the pressure evolution can be

described by eq. 3 (after Montgomery

et al. [14]):

(3)

with

| Fig. 8.

Determination of water removal rates. 10 ml of water were injected into the upper compartment,

which was then sealed, and the vessel heated up, here to 130 °C, 120 °C and 110 °C. Then the lower

compartment was evacuated, and the pressure recorded. The pressure drop indicates that all liquid

water was evaporated, the time for the removal of 10 ml of water were taken at ½ the initial pressure.

(4)

and

(5)

Ideal gas law gives the gas-inventory

of the water pocket at any time.

For the industrial process parameters

25 cm 3 water, r = 80 µm,

L = 4 m, T = 155 °C one would expect

the end of Phase 1 after 33 h, in which

vapor will leave the water pocket with

a rate of 20 g/d, with 70 mg vapor at

pressure 5.43 bar left in the pocket of

25 cm 3 . At 33.5 h the removal rate will

be down to 1 g/d with 16 mg vapor.

After 38 h the pressure will be reduced

to 60 mbar and 0.7 mg of vapor. Once

the driving pressure is below about

100 mbar vapor transport will effectively

stop, as will be shown in the

following section.

2.6 Connectivity

at low temperatures

In Figure 9 the temperature covers

the range from 90 °C to 160 °C,

although the measurements only

range between 110 °C to 130 °C. The

reason for the upper extrapolation is

the temperature of 150 °C chosen for

the industrial process (see 4.). The

lower limit addresses the following

interesting finding. To investigate the

vapor connectivity at progressively

lower temperatures (i.e. at lower

saturation pressures), a series of

pressure rebound tests was used. To

that effect 10 ml of water were injected

into the upper compartment, sealed

and heated to 120 °C (P sat = 2 bar).

The lower reservoir was evacuated to

about 3 mbar, then the vacuum line

was closed for 30 min and the bottom

pressure recorded. After completing

the IAEA-test the vacuum-line was

opened and the bottom evacuated

again, while the temperature was

lowered to 110 °C. After reaching

that temperature the vacuum line

was closed for another 30 min

IAEA-test. This procedure was

repeated in temperature steps of

90 °C (P sat = 0.7 bar), 70 °C

(P sat = 0.3 bar), 50 °C (P sat = 0.1 bar)

and 35 °C (P sat = 0.06 bar).

Figure 10 summarizes the pressure

rebound tests, Figure 11 shows

the corresponding dew-point reading,

which help in the interpretation of

Figure 10. During the pressure

rebound tests the vacuum line was

disconnected from the bottom reservoir

and the remaining moisture

removed within the duration of the

test (downward dips in dew point).

When reconnecting the vacuum line,

the moisture accumulated during the

test in the bottom was detected

( upward peaks in dew point). With

falling temperature these peaks get

smaller because less vapor was

transported through the rod. Below

50 °C the initial dew-point before the

test was not reached, indicating

vanishingly small communication. At

the end of the test at 35 °C, the

temperature was increased back to

120 °C and the presence of water was

confirmed by another isolation test

(not shown in Figures 10 and 11).

Leak rates for 120 °C, 110 °C and

90 °C are 4 … 0.07 mbar l/s or

17-4 g/d. Leak rate for 70 °C is

calculated to be 0.04 mbar l/s

corresponding to 0.8 g/d and at 50 °C

and 35 °C the leak rate is below

0.01 mbar l/s, removal rate smaller

than 0.6 g/d.

It can thus be concluded that at

pressures below about 100 mbar

vapor transport effectively stops. This

can be interpreted by the change in

flow-regime. Viscous flow, in this case

laminar according to the Re-numbers

as well as Figure 9, takes place

when the mean free path of the gas

particles l is very small compared to

the channel dimension w through

which it flows. If the mean free path

is comparable or even greater than

the gap, the gas particles bounce

along the wall and mass transport is

drastically reduced. The flow regimes

may thus be distinguished by the

Knudsen-number (s is the diameter of

gas molecules).

(6)

Decommissioning and Waste Management

The Scientific Backing of the German Quiver Project ı Wolfgang Faber, Marc Verwerft, Janne Pakarinen, Hagen Höfer and Christoph Rirschl


atw Vol. 65 (2020) | Issue 11/12 ı November/December

| Fig. 9.

Measured (symbols) and calculated (lines) removal rates for samples WET1, -2,-3,-5b, using laminar

flow rule and hydraulic radii from Ar-leak-rate measurements for theory. The solid red line is the

expected removal rate for a “worst-case” real fuel-rod situation with a lower bound hydraulic radius and

a maximum distance between water pocket and cladding breach, demonstrating industrial feasibility.

When the mean free path is short

compared to the dimensions of the

system, most of the molecules will interact

with other molecules and the

gas flow is in a viscous regime and can

either be laminar or turbulent. When

the mean free path is similar or larger

than the system dimensions, most

interactions are with the walls, and

one enters the molecular flow regime.

In-between these two extreme conditions,

one has a transition regime.

The Knudsen number (Kn) expresses

these regimes: with Kn < 0.01 (mean

free path less than 1/100 of the system

dimensions), one has viscous flow.

With Kn > 0.5, the system is in

molecular flow regime. If Kn > 0,01,

e.g. if the mean free path is larger

than 1/100 of the gap, flow enters the

transition regime. Figure 12 shows

the Knudsen number as function of

temperature for different gap sizes

according to eq. 6. Depending on

them flow enters the transition regime

below 80 °C with ducts of 30 µm,

whereas with 7,5 µm ducts this is the

case even with 120 °C. From the

cross-sections discussed later we can

estimate the ducts to be of the 15 µm

size, which means that a temperature

above 100 °C must be applied for

effective vapor transport [15].

2.7 Thermomechanical

calculations

Thermomechanical calculations were

performed for the fuel rods from

which the samples were taken.

Although the code does not strictly

apply to defective fuel, calculations

were performed for rod FT1X34-D04

(WET 1). Their results may tentatively

be interpreted in terms of the cold

gap (cladding inner radius – fuel

pellet radius). Figure 13 summarizes

the TRANSURANUS results for the

gap for end of cycle one and two as

well as after cooling down to isothermal

pool conditions. After the first cycle

the gap is closed only at the top, at

the end of cycle 2 almost over all the

active length. The thick red line shows

the cold gap to be between 15-20 µm.

Now the real situation with respect to

the fuel to cladding gap at end of

irradiation is displayed by the cross

section of Figure 4. There is not a

clear annular gap to be seen, but on

one side no gap, on the adjacent side

15-30 µm. An average value seems to

fit well enough with the calculations.

In addition, there are several circumferentially

and radially oriented

cracks. This pattern will of course

change axially, so that the existing

free path for vapor or gas is a crooked

and in shape greatly varying channel.

The hydraulic radius of a channel

with irregular shape is determined by

(7)

with A the cross-section area and U

the perimeter length of the wetted

cross-section. An annular gap of size

20 µm would then have the same

number as hydraulic radius, which is

much smaller than was observed.

Obviously, the crack area also adds to

the effective hydraulic radius, which

was determined to be around 80 µm.

These observations may allow a

conservative guess on the to be

expected hydraulic radius of a fuel

rod. In operation the fuel is first

densified, then swells and the cladding

creeps down under the influence

of the outer pressure until both fuel

| Fig. 10.

Pressure rebound tests at different temperatures. In all tests liquid water

was present in the upper reservoir, the saturation pressure given in

the plot. All tests except the one at 35 °C did not pass, meaning that the

pressure increased within 30 min to above 4 mbar. The test at 35 °C

passed, although liquid water was present in the top.

| Fig. 11.

Dew-point for the series of IAEA-pressure rebound tests. During the tests

the vacuum line was disconnected from the bottom reservoir and the

remaining moisture removed within the duration of the test (downward

dips). When reconnecting the vacuum line, the moisture accumulated

during the test in the bottom was detected (upward peaks).

| Fig. 12.

Knudsen number as function of vapor temperature for ducts of different

sizes. At 35 °C flow is well in the transition regime.

pellet and cladding get in contact, i.e.

the hot gap closes. Almost at the first

few power ramps, the pellets crack

due to thermal stress. Pellet swelling

acts in both directions, radially and

axially. When the pellet slices formed

by the radial cracks move outward,

bonding may occur with the cladding.

Taking all that into account, the worst

case with respect to the (theoretical)

cold gap should be reached when the

hot gap is closed, and the cladding

relaxed in this state. Then the cold

DECOMMISSIONING AND WASTE MANAGEMENT 567

Decommissioning and Waste Management

The Scientific Backing of the German Quiver Project ı Wolfgang Faber, Marc Verwerft, Janne Pakarinen, Hagen Höfer and Christoph Rirschl


atw Vol. 65 (2020) | Issue 11/12 ı November/December

DECOMMISSIONING AND WASTE MANAGEMENT 568

| Fig. 13.

Thermomechanical calculation for rod FT1X34-D04 (if it were intact) with TRANSURANUS. After the first

cycle the gap is closed only at the top, at the end of cycle 2 almost over all the active length. The thick

red line shows the cold radial gap to be between 15-20 µm.

gap opens by differential thermal

contraction of cladding and pellet:

(8)

An estimate using (8) yields with

typical numbers (DT fuel ≈ 680 K and

DT clad ≈ 350 K, a fuel ≈ 8.9 10 -6 1/K

( effective) and a clad ≈ 7.1 10 -6 1/K) an

expected cold gap from this situation

of 18 µm, which is consistent with

the thermomechanical calculations.

For this estimate a was taken from

the TRANSURANUS Handbook, version

2009 [16]. From eq.8 a general

tendency of a smaller effective

hydraulic radius for smaller cladding

radius is expected.

As pointed out earlier, the theoretical

gap, which is partly transferred

into the cracks due to bonding, cannot

account for the measured hydraulic

radius alone. Even a theoretical

transformation of the total gap area

into a crack geometry that has the

same area but minimal perimeter, that

is the separation of the pellet in two

halves that have full contact with the

cladding on the outside, results in a

hydraulic radius of less than ¾ of the

measured one. This clearly shows that

the cracks must contribute essentially

to the communication through the

fuel stack. The originating mechanism

results in a typical crack pattern,

as observed in cross sections like

Figure 4, which constitute part of the

percolation system formed early in

life. To this the cold gap is added,

which is smallest for high burnup and

low power operation end of life.

Extensive FRAPCON assessments

have been performed to investigate

axial gas flow connectivity using the

approach as outlined above, and

successfully calculated the hydraulic

properties of the WETFUEL samples

[17].

From that it may be concluded that

generally high burnup PWR fuel

should result in the smallest hydraulic

radius, which will still vary axially

with a larger hydraulic radius expected

at the extremities. Theoretically, from

two fuel rods of the same (high)

burnup, but one operated longer at

lower power, the latter will have the

smaller permeability, because the

pellet will swell almost alike the first

and the cladding has time to creep

down, but the cold gap will be smaller

from smaller differential contraction.

From generic thermomechanical

considerations a conservative guess on

the worst to be expected hydraulic

radius can be derived and should

therefore be around the number of

80 µm for the Tihange 15 × 15 rods

operated with comparable power and

coolant-temperature levels, which

was confirmed in the WETFUEL and

AGAF (footnote: AGAF was a campaign

to measure gas communication

on samples of rod FT1X57 D05, which

is not reported here) project for a

number of samples of two PWR fuel

rods with burnup of 50 MWd/kg, at

which the hot gap is well closed.

3 Discussion of

the WETFUEL results

The hydraulic radius of the WETFUEL

samples investigated here ranged

between 85 µm and 107 µm, which

correspond to permeabilities K of

2.9 10 -13 m 2 – 7.4 10 -13 m 2 .

In the Oak-Ridge sister rod program

permeabilities on 8 full length

rods from 17 × 17 FA were measured

and cover a range of 1.1 10 -14 m 2 …

8.3 10 -14 m 2 , the corresponding

hydraulic radii are 35 µm … 58 µm

[14]. Rondinella et al. measured the

permeability of a 16 × 16 fuel-rod,

yielding K = 2 × 10 -13 m 2 or r = 76 µm

[18]. These numbers clearly show a

dependency from the fuel-rod radius.

According to eq. 8, the cold gap, which

is in part responsible for the total free

cross-section, depends on the cladding

radius. Therefore, one would

expect a general trend of the permeabilities

of high burnup fuel from

15 × 15 … 17 × 17 fuel rods. This

general trend may be covered by

effects from the fuel- and cladding

temperatures especially in the last

cycle of operation. The worst situation

should occur when the fuel-rod was

operated in its last cycle with low

power, e.g. DT fuel in eq. 8 is small. The

WETFUEL rods were operated at high

power, whereas the numbers reported

in [14] point to a rather modest power

history. Moreover, the difference in

cladding radius between 15 × 15

(r clad = 5.36 mm for Tihange 1) and

16 × 16 (r clad =5.375 mm) fuel

rods is rather small, whereas the

difference from 16 × 16 to 17 × 17

(r clad = 4.75 mm) is more pronounced.

Thus, the permeability reported by

Rondinella, smaller but close to those

of the 15 × 15 WETFUEL results, and

the much smaller ones in [14] compared

to those from WETFUEL, are

well understood.

ORNL [14] compared the permeabilities

of the fuel rods with respect

to burnup, fuel-duty and claddingmaterial.

There was no dependency

on burnup between 49 MWd/kg to

60 MWd/kg, which is no surprise

because the hot gap should be closed

at this burnup. There was a correlation

observed with fuel temperature

and a fuel-duty index which is also in

line with the interpretation according

to eq. 8. They also see a certain trend

with cladding material, which is not

straight forward to explain. Creepproperties

will play a role in intermediate

burnup ranges but should not

show at high burnup. Moreover, this

trend could not be observed in the

WETFUEL experiments.

The comparison of a rod that

failed, from which sample WET1 was

taken, with an intact one from the

same cycles with almost the same

burnup, from which samples WET2,

-3 and -5b were taken, showed that

the defect in one did not cause the

hydraulic radius to be altered to a

significant extent due to interaction

of water and fuel. The intact fuel

rod was made from M5-cladding

Decommissioning and Waste Management

The Scientific Backing of the German Quiver Project ı Wolfgang Faber, Marc Verwerft, Janne Pakarinen, Hagen Höfer and Christoph Rirschl


atw Vol. 65 (2020) | Issue 11/12 ı November/December

(Framatome), the defective one from

Zry4. It must be mentioned, however,

that the trend observed in the ORNLmeasurements

does allow a different

interpretation. If the M5-cladding

would tend to smaller permeabilities,

the WETFUEL results with respect to

cladding material would mean that

the interaction of fuel with water

had an influence and lowered the

hydraulic radius.

4 The GNS/H&B campaign

October 2018 in KKU

It is of particular interest that the

information about drying time gained

in quiver campaigns is supporting the

quantitative assessment derived from

the WETFUEL experiments.

The process of drying the GNS IQ®

(Integrated Quiver System) works in

two steps according to Figure 14

[3, 4]. In phase I the water sur rounding

the defective fuel rods from loading the

quiver in the pool is removed by

vacuum-pumping at about 100 °C.

Phase II then empties water pockets

through the cladding breach by heating

to 150 °C, the saturation pressure of

5 bar pushing the vapor through the

cracks in the fuel stack.

Figure 15 shows a record from

the first GNS IQ® campaign in KKU.

The yellow and green lines are the

temperature of the quiver (inlet and

outlet of heating gas), the blue line is

the pressure in the exhaust-line. At

time A the temperature of the fuel

rods is raised slightly above 150 °C,

after the water surrounding the rods

has been evaporated. The quiver can

hold up to 32 rods, which was the

case in that campaign. Therefore, up

to 32 water- pockets with different

distances to the cladding perforation

| Fig. 14.

Phases of drying the quiver. In phase I the

water surrounding the defective fuel rods is

removed by vacuum-pumping at about

100 °C. Phase II empties water pockets

through the cladding breach by heating

slightly above 150 °C, the saturation pressure

of 5 bar pushing the vapor through the cracks

in the fuel stack.

| Fig. 15.

Figure 15: Recording from the first quiver campaign in October 2018 at Kernkraftwerk Unterweser (KKU). The yellow and green lines

are the temperature of the quiver (inlet and outlet of heating gas), the blue line is the pressure in the exhaust-line. At time A the

temperature of the fuel rods is raised to ≥150°C, after the water surrounding the rods has been evaporated. Up to 32 water-pockets

are pressurized simultaneously to 5 bar, pushing vapor through the fuel stack. The red circle marks the worst connecting fuel rod,

from which the liquid water is gone where the last drop in pressure occurs.

are pressurized simultaneously to

5 bar, pushing vapor through the fuel

stacks with different hydraulic radius,

depending on the irradiation history

of the fuel rod. The red circle marks

the worst connecting fuel rod (combination

of hydraulic radius and distance

to cladding perforation), from

which the liquid water is gone at that

point in time where the last drop in

pressure occurs.

The theoretical time to remove

water from a rod with hydraulic radius

of 80 µm and 3.5 m distance to the

cladding breach at 155 °C is 24 h,

which matches the time of phase II in

the campaign.

5 Summary

This paper describes the scientific

backing of an industrial development

process, that was performed in

parallel to the technical realization,

giving input along the way. Usually

research takes place before the

technical realization starts. In this

case the theoretical basis was

necessarily rather old, from the 80ies,

and the tremendous time pressure

from the necessity to get rid of all

fuel inventory in about 5 years did not

allow to wait for new results. Eon

Kernkraft initiated this research project

to minimize risks associated with

the development and licensing of a

technical solution in the back-end

area and was joined by the other

German nuclear utilities in a second

part of the project.

In the experiments described

above the feasibility of effectively

removing water from soaked fuel by

heat assisted vacuum drying in

practical times was demonstrated.

The governing parameters were identified:

the hydraulic radius constituted

by the difficult free path of vapor or

gas through the fuel stack, the length

of the stack along which vapor, or gas,

has to be transported as well as the

temperature applied to the fuel rod.

Moreover, quantitative measurement

of water removal rates showed that in

the temperature-regime studied here

water vapor flow can be described by

laminar flow rule, allowing a quantitative

assessment of drying times.

A simple picture how the porosity

in the fuel rod is created from cracks

and the cold gap, which allows to

draw basic conclusions with respect

to contributing effects to perme ability,

is described. Permeability of spent

fuel rods is thus dependent on rod

diameter and operating temperatures.

Lower rod diameter tends to lower

permeability, onto which temperature

effects are superimposed. Higher

temperatures will generally lead to

higher permeability by two effects,

the crack formation early in life as

well as the cold gap from differential

contraction end of life. The cold gap

will not change much once the hot

gap has been closed, which means

that permeability is essentially independent

from burnup above this

point.

DECOMMISSIONING AND WASTE MANAGEMENT 569

Decommissioning and Waste Management

The Scientific Backing of the German Quiver Project ı Wolfgang Faber, Marc Verwerft, Janne Pakarinen, Hagen Höfer and Christoph Rirschl


atw Vol. 65 (2020) | Issue 11/12 ı November/December

DECOMMISSIONING AND WASTE MANAGEMENT 570

This, together with the worst

distance between plenum and cladding

breach, which is the total active

fuel length, allows a conservative

estimate on water removal rates until

dryness. The experiments also allow a

reflection on dryness, for they clearly

show that a water-pocket cannot be

pumped directly but is rather emptied

pushing the vapor through the cracks

and fissures of the fuel stack by

its own saturation pressure. Once

the liquid water has been removed

that way, pressure drops. Once the

pressure difference is below about

100 mbar, vapor transport effectively

stops, determining the small amount

of water that cannot be removed in

practical times.

Although transport of liquid water

through the pellet stack was not an

objective in these experiments, the

process of wetting the sample before

demonstrating the general feasibility

of hot vacuum drying was interpreted

as far as the data allowed. Within the

accuracy that the experimental setup

allowed the rate of water transport

through the fuel stack by pressure

is consistent with Hagen-Poiseuille

prediction, using the hydraulic radius

from gas measurements.

The pressure evolution from the

first industrial quiver campaign in

Germany was interpreted and is in

line with the theoretical predictions

derived from the experiments.

Activity release was very small,

and fuel wash out did not happen

based on the unchanged hydraulic

radius of the samples through all the

numerous experiments.

References

[1] Bundesamt für die Sicherheit der nuklearen Entsorgung

(BASE), Transport- und Lagerbehälter CASTOR V/19,

Zulassungsschein D/4372/B(U)F-96 (Rev.5), 25.04.2017.

[2] Bundesamt für die Sicherheit der nuklearen Entsorgung

(BASE), Transport- und Lagerbehälter CASTOR V/52,

Zulassungsschein D/4373/B(U)F-96 (Rev.3), 23.04.2018.

[3] S. Bechtel, W. Faber, H. Hofer, et al., The German Quiver

Project – Quivers for Damaged and Non-Standard Fuel Rods,

ATW-Int. J. Nucl. Power 2019, 64, 151-159.

[4] C. Rirschl, H. Hofer, M. Verwerft, W. Faber, Fuel removal at

Unterweser, Nuclear Engineering International 2019, 64,

30-35.

[5] R. Kohli, D. Stahl, V. Pasupathi, A.B. Johnson Jr., E.R. Gilbert,

The Behavior of Breached Boiling Water Reactor Fuel Rods on

Long-Term Exposure to Air and Argon at 598 K, Nucl. Technol.

1985, 69, 186-197.

[6] R. Kohli, V. Pasupathi, Investigation of water-logged spent

fuel rods under dry storage conditions, United States, report

PNL-5987 (1986).

[7] M. Seidl, A. Wensauer, W. Faber, Experience with valuation

methods for the creation of real options enabling diversity of

nuclear fuel supply (submitted to the International Journal of

Forecasting).

[8] G. Cornelis, WETFUEL Conceptual design for safety study

(Rev 1), SCK CEN, Mol, Belgium, report R-5476 (2013).

[9] G. Cornelis, J. Pakarinen, W. Faber, M. Verwerft, Experimental

setup for hydraulic resistance measurements on spent nuclear

fuel, HOTLAB 2017 Annual meeting on Hot Laboratories and

Remote Handling, SCK CEN – Belgian Nuclear Research

Center, Mito (Japan), 2017, pp. 1-7.

[10] IAEA, Management of Damaged Spent Nuclear Fuel

(NF-T-3.6), INTERNATIONAL ATOMIC ENERGY AGENCY,

Vienna, report STI/PUB/1395 (2009).

[11] I. Nettleship, Applications of porous ceramics, Key Eng. Mater.

1996, 305-324.

[12] H. Greinacher, Über das Fließen in kapillaren Räumen,

Z. Phys. Chem. 1959, 19, 101-117.

[13] ASTM, C1553-16, Standard Guide for Drying Behavior of

Spent Nuclear Fuel, ASTM International, West Conshohocken,

PA, 2016.

[14] R. Montgomery, R.N. Morris, R. Ilgner, et al., Sister Rod

Destructive Test Results (FY19), United States, report ORNL/

SPR-2019/1251 Revision 1 (2019).

[15] M. Verwerft, J. Pakarinen, G. Cornelis, W. Faber, H. Höfer, C.

Rirschl, Boundary conditions to water removal from defective

spent fuel 14 th International Nuclear Fuel Cycle Conference,

GLOBAL 2019 and Light Water Reactor Fuel Performance

Conference, TOP FUEL 2019, Seattle; United States, 2020,

pp. 94-103.

[16] K. Lassmann, A. Schubert, P. Van Uffelen, C. Györi, J. Van de

Laar, TRANSURANUS Handbook, EC-JRC-ITU, report v1m1j09

(2009).

[17] K. Govers, Axial gas flow in irradiated fuel rods (AGAF) –

FRAPCON evaluation, SCK CEN, Mol, report R-6482 (2018).

[18] V. Rondinella, D. Papaioannou, R. Nasyrow, W. Goll, M. Rehm,

Measurement of gas permeability along the axis of a spent

fuel rod, TOPFUEL 2015, European Nuclear Society, Zürich,

Switzerland, 2015.

Authors

Dr. Wolfgang Faber

wolfgang.faber@

preussenelektra.de

PreussenElektra GmbH

Tresckowstraße 5

30457 Hannover, Germany

Dr. Marc Verwerft

D.Sc. (Tech.) Janne Pakarinen *)

Belgian Nuclear Research Center

SCK-CEN, Institute for Nuclear

Material Science

Boeretang 200

2400 Mol, Belgium

*)

currently Technical Research

Center Finland

VTT, Visiokatu 4

33720 Tampere, Finland

Hagen Höfer

Höfer&Bechtel GmbH

Ostring 1

63533 Mainhausen, Germany

Christoph Rirschl

GNS Gesellschaft

für Nuklearservice mbH

Fronhauser Straße 67

45127 Essen, Germany

Decommissioning and Waste Management

The Scientific Backing of the German Quiver Project ı Wolfgang Faber, Marc Verwerft, Janne Pakarinen, Hagen Höfer and Christoph Rirschl


atw Vol. 65 (2020) | Issue 11/12 ı November/December

The EMPIrE Irradiation Test:

Lower- Enriched Fuel for High-

Performance Research Reactors

Bruno Baumeister, Christian Reiter and Winfried Petry

571

FUEL

1 Introduction The Forschungs-Neutronenquelle Heinz Maier-Leibnitz (FRM II) is a Planned entry for

high-performance neutron source operated by the Technical University of Munich (TUM) with the

purpose to supply neutrons for basic science as well as industrial and medical applications like the

production of radioisotopes. These neutrons are provided by a compact core using a single

cylindrical fuel element with 113 involute-shaped fuel plates containing a total of 8.1 kg of highly

enriched uranium (HEU) with 93 % U-235 enrichment. Together with other high-performance research reactors in the

world currently operating with HEU, TUM aims to convert FRM II’s fuel to a lower-enriched type as part of worldwide

efforts to minimize the usage of HEU in the civil fuel cycle. For this conversion, a new fuel with increased uranium

density is required to compensate the decreased U-235 content in the fuel element. High-density metallic fuel with

enrichment up to 19.75% – so-called HALEU (High Assay Low-Enriched Uranium) might also be needed for some of the

currently discussed SMRs (Small Modular reactors) [1]. The currently most promising fuel candidate is a metallic

uranium-molybdenum alloy (U-Mo) in the form of either dispersion [2] or monolithic fuel [3], as depicted in Figure 1.

While dispersed U-Mo fuel consists of a mixture of spherical U-Mo powder and aluminum powder, monolithic fuel

consists of a thin U-Mo foil only. Therefore, monolithic fuel in principle allows the higher uranium density and therefore

the lowest uranium enrichment after conversion. Thus, TUM is focusing its research efforts on this fuel candidate.

| Fig. 1.

The FRM II fuel element and fuel conversion options.

| Fig. 2.

Dimensions of the EMPIrE mini-plates. The fueled zone with the dimensions 82.5 mm x 19 mm

is shown in dark grey.

Recently, both fuel types were successfully

tested in the EMPIrE (European

Mini-Plate Irradiation Experiment)

test, a joint European/US-American

irradiation experiment designed to

gather important knowledge on the

in-pile irradiation behavior of this

fuel [4]. Further, it is a first-ever

demonstra tion of newly-developed

European fabrication techniques for

dispersed and monolithic U-Mo fuel

in the mini-plate geometry, which is

illustrated in Figure 2.

While most of the fuel fabrication

steps were performed in Europe, the

irradiation of the fuel took place in the

Advanced Test Reactor ATR at the

Idaho National Laboratory (INL) in

the USA. Together with its sister irradiation

experiment SEMPER-FIDELIS

for full-size plates, EMPIrE is part of

the comprehension phase on highdensity

U-Mo fuels of HERACLES,

[5] the European consortium of

developers and fabricators of research

reactor fuel.

2 Production of irradiation

fuel plates

Metallic uranium fuels, such as U-Mo,

are known to form an interdiffusion

layer with aluminum-based cladding

materials [6] under irradiation or

during fabrication steps at elevated

temperatures. This usually amorphous

interdiffusion layer can cause gaseous

fission products to accu mulate in

macroscopic bubbles, ultimately

leading to fuel plate failures due to

severe mechanical defor mation. Thus,

it is generally necessary to prevent the

formation of such a layer, typically by

the application of a diffusion barrier.

For monolithic U-Mo fuel, a few µm

coating of a refractory metal [7]

like e.g. zirconium, molybdenum or

tungsten is possible.

For EMPIrE, bare U-Mo foils with a

nominal thickness of 0.33 mm were

obtained from BWX Technologies Inc.

(BWXT, USA) and then coated with

25 µm of zirconium on both sides at

TUM’s nuclear fuel laboratory. All

relevant processing steps, such as

wet-chemical cleaning and coating of

the foils, took place in a double

glovebox system for the safe handling

and treatment of nuclear fuel under

an inert argon atmosphere. The Zr

layer was applied by a PVD-based

(physical vapor deposition) process,

developed by TUM [8], which is

automatically controlled by a PLC

(Programmable Logic Controller)

system optimized towards safe and

stable operation for reproducible

results. Finally, the coated foils were

Fuel

The EMPIrE Irradiation Test: Lower- Enriched Fuel for High- Performance Research Reactors ı Bruno Baumeister, Christian Reiter and Winfried Petry


atw Vol. 65 (2020) | Issue 11/12 ı November/December

FUEL 572

| Fig. 3.

The European fabrication process for monolithic U-Mo fuel for research reactors.

sent to the French fuel fabricator

Framatome-CERCA, where the aluminum-

based cladding was applied

using the industrial C2TWP cladding

application process. Figure 3 illustrates

this European fabrication process

for monolithic U-Mo fuel for

research reactors.

Following two test sets with stainless

steel surrogates and three test sets

with depleted uranium, the final

fabrication of seven monolithic U-Mo

fuel plates containing low-enriched

uranium (LEU) was successfully

performed in mid-2016. Due to the

high safety requirements for nuclear

fuel, the whole production process

was monitored by a comprehensive

quality assurance program according

to NQA-1 (Nuclear Quality Assurance-

1), including materials procurement,

pre-defined working routines

and the documentation of all relevant

data. At the final quality inspection at

Framatome- CERCA, twice the required

number of plates passed the final conformance

test. By the time the fuel

plates were shipped to the ATR reactor

at the end of 2016, the monolithic

production program for the EMPIrE

project has been com pleted successfully

and on schedule, with only

18 months between the first tests and

the specification-conform fuel plates.

Figure 4 shows an electron microscopy

cross section of the interface between

U-Mo foil and Zr coating, applied by

PVD, in the unirradiated state.

To compare TUM’s PVD coating

process with the proposed US-

American process of Co-Rolling for Zr

application, seven additional monolithic

U-Mo fuel plates containing LEU

were fabricated using this process as

well. As a second part of the EMPIrE

fuel fabrication program, a high

quantity of dispersed U-Mo fuel plates

containing LEU were fabricated by

Framatome-CERCA to compare the

fabrication and irradiation behavior

of different powder production processes,

coating techniques, heat treatments

and molybdenum contents of

the U-Mo powder.

3 Irradiation in the

Advanced Test Reactor

Prior to the irradiation, a secondary

comprehensive characterization of all

fuel plates destined for irradiation

was performed at INL using ultrasonic

transmission, ultrasonic pulse-echo,

thermography, optical and scanning

electron microscopy with energy- -

dispersive x-ray spectroscopy. Based

on this characterization, which again

demonstrated the specification conformity,

two monolithic U-Mo fuel

plates from each fabrication process

route were selected for irradiation in

the ATR reactor [9].

These plates were then loaded into

irradiation capsules, designed for

insertion into the ATR reactor core, as

seen in Figure 5.

In March 2018, the irradiation of

a total of 48 fuel plates (44 dispersed

U-Mo, 2 PVD-coated monolithic U-Mo

and 2 Co-Rolled monolithic U-Mo)

began, batches of 16 plates being

irradiated for 38, 54 and 92 days,

respectively. The irradiation con ditions

for the monolithic fuel plates such as

U-235 burnup and heat flux were

chosen to conservatively approach the

operating conditions of FRM II. Hereby,

two monolithic fuel plates were irradiated

for 38 days, leading to a peak

burnup of ~38 % U-235 (low-burnup

plates) and two for 54 days, leading to

a peak burnup of ~51 % U-235 (highburnup

plates). The peak heat flux in

the fuel plates was ~380-400 W/cm 2

at the beginning of the irradiation for

all plates.

The following Table 1 indicates the

unique IDs of the four monolithic U-Mo

LEU fuel plates in EMPIrE. Hereby,

EMPI12-type IDs resemble plates

where the Zr layer was applied by PVD,

whereas EMPI11-type IDs resemble

plates with a co-rolled Zr layer.

In August 2018, the irradiation was

successfully completed without any

incident [10]. From MCNP-based [11]

(Monto-Carlo N Particles) neutronics

calculations after the irradiation, the

fission density across the fueled zone

inside the fuel plates was calculated by

INL, as illustrated in Figure 6. It is

notable that the fission density is not

homogeneous across the fueled zone,

which results from the orien tation

of the plates in the inhomo geneous

neutron flux field in the reactor.

4 Post-irradiation

examination

Post-Irradiation Examination (PIE) is

an essential part of any irradiation

experiment in order to determine the

effect of the uranium burnup on the

| Fig. 5.

Schematic view of an EMPIrE irradiation capsule with fuel plates (purple) [4].

Low burnup

High burnup

| Fig. 4.

Electron microscopy cross section of the interface between U-Mo foil and

Zr coating, applied by PVD, in the unirradiated state.

Plate IDs

(Zr layer technique)

| Tab. 1.

Monolithic U-Mo LEU fuel plates in EMPIrE.

EMPI1201 (PVD)EMPI1103

(Co-Rolling)

EMPI1207 (PVD)EMPI1102

(Co-Rolling)

Irradiation duration 38 days 54 days

Peak U-235 burnup ~ 38 % ~ 51 %

Peak heat flux ~ 380 - 400 W/cm² ~ 380 - 400 W/cm²

Fuel

The EMPIrE Irradiation Test: Lower- Enriched Fuel for High- Performance Research Reactors ı Bruno Baumeister, Christian Reiter and Winfried Petry


atw Vol. 65 (2020) | Issue 11/12 ı November/December

inside the plates is measured before

irradiation and compared with the

thickness increase of the entire fuel

plates after irradiation. Assuming no

swelling of the aluminum claddings

and Zr layers, the thickness increase of

the plates entirely results from the fuel

swelling, according to the following

equation:

FUEL 573

| Fig. 6.

Localized fission density of the monolithic U-Mo LEU fuel plates in EMPIrE. The inhomogeneity results

from the orientation of the fuel plates in the neutron flux field of the reactor. Raw data from [10].

| Fig. 7.

Visual inspection of the front and back sides of the monolithic U-Mo LEU fuel plates in EMPIrE.

Besides an expected aluminum oxide layer in the fueled zone of the plates, no obvious defects were

found. Images from [10].

| Fig. 8.

Neutron radiography of the monolithic U-Mo LEU fuel plates in EMPIrE. No fuel cracking or

dislocation could be found. Images from [10].

integrity and structure of the fuel, as

well as to ensure that no unwanted

irradiation-driven defects such as

plate warping, bursting or excessive

swelling have occurred. The data

obtained from the PIE also helps to

further optimize the fuel system as

well as the fabrication processes. Thus,

a comprehensive program of nondestructive

and destructive analyses is

carried out on the irra diated fuel

plates at INL. This paper focuses on

the results from the non-destructive

analyses carried out in 2019 and

summarized in the INL report no.

INL/LTD-19-56501 on EMPIrE Monolithic

Plates NDE Results [10]. The

results from the destructive analyses

are expected for early 2021.

By visual inspection and neutron

radiography using INL’s dedicated

NRAD TRIGA reactor [12] of the

irradiated fuel plates in hot cells, no

abnormalities or obvious defects were

found, as depicted in Figure 7 and

Figure 8. In the fueled regions of

the plates, an expected discoloration

due to aluminum oxide formation has

occurred during irradiation.

One of the most important fuel performance

data from such irra diation

tests is the swelling behavior, i.e. the

volume gain of the fuel in percent. This

swelling occurs mostly due to the accumulation

of gaseous fission products.

To quantify this irradiation- induced

swelling, the thickness of the fuel foil

Here, T plate,post is the plate thickness

after irradiation, T plate,pre is the plate

thickness before irradiation, T core,pre is

the total thickness of the fresh fuel

core (U-Mo foil plus Zr layers) before

irradiation and T Zr is the thickness of

the Zr layers. Fresh fuel thickness data

T core,pre was obtained by Framatome-

CERCA using X-Ray radiography and

destructive sampling and is illustrated

in Figure 9 for the four monolithic

U-Mo LEU fuel plates. Because the

specification for the fuel foil thickness

was 0.33 mm for both bare U-Mo

foils as well as co-rolled foils with

2 × 25 µm of Zr layers, the U-Mo fuel

thickness is typically 0.05 µm thinner

in the EMPI11-type plates with

co-rolled Zr layers.

The fuel plate thickness after irradiation

T plate,post was determined by

INL’s dedicated fuel plate measurement

bench [13] at approximately

600 points per plate by a scanning

profilometry technique using two

opposing vertical probes. Figure 10

shows the localized fuel swelling for

the four monolithic U-Mo LEU fuel

plates. It is obvious that the highburnup

plates exhibit a higher average

swelling than the low-burnup plates.

Both coating techniques further more

lead to a comparable swelling behavior.

Above-average swelling at the

edges of the fueled zone, visible

at plate EMPI1207, is believed to

originate from the plate fabrication

| Fig. 9.

Fresh fuel thickness data of the monolithic U-Mo LEU fuel plates in EMPIrE. EMPI11-type plates typically

show a lower U-Mo fuel thickness, as the specification for fuel thickness included the 2 × 25 µm Zr

layers for this plate type. Raw data from [10].

Fuel

The EMPIrE Irradiation Test: Lower- Enriched Fuel for High- Performance Research Reactors ı Bruno Baumeister, Christian Reiter and Winfried Petry


atw Vol. 65 (2020) | Issue 11/12 ı November/December

FUEL 574

| Fig. 10.

Fuel swelling data of the monolithic U-Mo LEU fuel plates in EMPIrE. It is obvious that the high-burnup

plates show a higher average swelling than the low-burnup plates. Raw data from [10].

from the Bundesministerium für

Bildung und Forschung (BMBF) and

the Baye ri sches Staatsministerium für

Wissenschaft und Kunst (StMWK).

This work was supported by the

European Commission within the

framework of HORIZON 2020

through Grant Agreement 661935

in the HERACLES-CP project and

Grant Agreement 754378 in the

LEU- FOREvER project.

We thank BWXT, Framatome-

CERCA, INL, SCK CEN and CEA for

their support with fabrication, irradiation

and post-irradiation examination

in the EMPIrE experiment.

| Fig. 11.

Fuel swelling vs. fission density of the monolithic U-Mo LEU fuel plates in EMPIrE (red and yellow).

The swelling data is in good accordance with 2012’s RERTR-12 irradiation test of monolithic U-Mo fuel

(black). Raw data from [10].

process and will be further investigated

in the destructive PIE.

Figure 11 shows the fuel swelling

in relation to the corresponding

fission density and indicates that the

swelling behavior of the fuel plates is

in good accordance with 2012’s

US-American RERTR-12 irradiation

test of monolithic U-Mo [14].

As next step, destructive PIE

will be performed on the high-burnup

plates at INL. The plates will be cut

and prepared for metallographic

examination e.g. by optical and scanning

electron microscopy, energydispersive

x-ray spectroscopy (EDS)

and electron back scattering diffraction

(EBSD). There, the U-Mo fuel, the

Zr layers, and the cladding will be

analyzed, and the possible formation

of inter mediate phases will be investigated.

Addi tionally, small lamellas of

the high- burnup plates will be cut and

shipped to the European research

institutes SCK CEN (Belgium) and

CEA (France) in early 2021 with

the goal to perform advanced destructive

PIE to gather more detailed

insight into the microstructure of the

fuel plate constituents.

5 Outlook

The EMPIrE irradiation experiment delivers

important data and know ledge

on high-density U-Mo fuels for research

reactors. The respective European fuel

fabrication processes for monolithic

U-Mo, namely TUM’s PVD coating process

and Framatome- CERCA’s C2TWP

cladding application process, successfully

demonstrated their suitability for

the next steps in the fuel qualification

and reactor conversion process. This

encompasses the further development

of these pro cesses up to an industrial

fabrication level as well as their utilization

for fabricating fuel plates for

future irradiation tests.

Therefore, TUM’s high-density fuel

development group at FRM II, in

close collaboration with Framatome-

CERCA, is currently working on the

upscaling of these processes to fuel

plate geometries suitable for operation

in research reactors like the FRM II. To

avoid the need to procure U-Mo fuel

foils from US-American fabricators in

the future, TUM has partnered with

Framatome-CERCA in 2019 to develop

and establish a pilot fabrication line

for U-Mo foils. With this fabrication

line, a complete European fabrication

capability for monolithic U-Mo fuels is

envisaged to be operational in 2022.

Acknowledgements

This work was supported by a combined

grant (FRM1318 and FRM2023)

References

[1] Euratom Supply Agency, Securing the European Supply

of 19.75 % enriched Uranium Fuel, a revised assessment,

May 2019, https://ec.europa.eu/euratom/docs/ESA_HALEU_

report_2019.pdf

[2] Van den Berghe, S.; Lemoine, P., Review of 15 Years of

High-Density Low-Enriched U-Mo Dispersion Fuel Development

for Research Reactors in Europe, Nuclear Engineering

and Technology 46 (2014) 2, pp. 125-146

[3] Meyer, M. K.; Gan, J.; Jue, J.-F.; Keiser, D. D.; Perez, E.; Robinson,

A. B.; Wachs, D. M.; Woolstenhulme, N. E.; Hofman, G. L.; Kim,

Y. S., Irradiation Performance of U-Mo Monolithic Fuel, Nuclear

Engineering and Technology 46 (2014) 2, pp. 169-182

[4] Glagolenko, I., The European MiniPlate Irradiation Experiment

(EMPIrE), RERTR 2018, Edinburgh, Scotland

[5] The HERACLES Consortium,

https://www.heracles-consortium.eu

[6] Idaho National Laboratory, RERTR-10 Irradiation Summary

Report, INL/EXT-10-18456, 2011

[7] Huang, K; Kammerer, C. C.; Keiser, D. D.; Sohn, Y. H., Diffusion

Barrier Selection from Refractory Metals (Zr, Mo and Nb) Via

Interdiffusion Investigation for U-Mo RERTR Fuel Alloy, Journal

of Phase Equilibria and Diffusion 35 (2014) 2, pp. 146-156

[8] Steyer, C., Plasma- und festkörperphysikalische Optimierung

eines Beschichtungsverfahrens für monolithische

UMo-Kernbrennstoffe, Dissertation, Technische Universität

München, 2019

[9] Idaho National Laboratory, Assessment of Type 11 and 12

Plates for the EMPIrE Experiment, CCN 241154, 2017

[10] Idaho National Laboratory, EMPIRE Monolithic Plates NDE

Results, INL/LTD-19-56501, 2019

[11] C.J. Werner, J.S. Bull, C.J. Solomon, et al. “MCNP6.2 Release

Notes”, https://laws.lanl.gov/vhosts/mcnp.lanl.gov/pdf_files/

la-ur-18-20808.pdf LA-UR-18-20808 (2018).

[12] Idaho National Laboratory, Neutron Radiography of Irradiated

Nuclear Fuel at Idaho National Laboratory,

INL/JOU-14-33699, 2015

[13] Williams, W.J.; Robinson, A. B., Non-Destructive Dimensional

Analyses at INL’s Hot Fuel Examination Facility, HOTLAB 2016,

Karlsruhe, Germany

[14] Idaho National Laboratory, RERTR–12 Post-irradiation

Examination Summary Report, INL/EXT-14-33066, 2015

Authors

Bruno Baumeister

Head of High-Density Fuel

Development

Bruno.Baumeister@frm2.tum.de

Dr. Christian Reiter

Head of Reactor Physics

Prof. Dr. Winfried Petry

Emeritus of Excellence

Forschungs-Neutronenquelle

Heinz Maier-Leibnitz (FRM II)

Technische Universität München

Lichtenbergstr. 1

85748 Garching, Germany

Fuel

The EMPIrE Irradiation Test: Lower- Enriched Fuel for High- Performance Research Reactors ı Bruno Baumeister, Christian Reiter and Winfried Petry


atw Vol. 65 (2020) | Issue 11/12 ı November/December

From Fission to Fusion – Transfer of

Existing Industrial Know-How to New

Domains of Applications

Sabrina Gil Pascual and Andreas Bender

Introduction The pursuit of new energy sources is one of mankind’s most important

challenges. This is due to the fact that energy consumption is dramatically increasing until the turn

of the century. In addition, a vast and inexpensive supply of electrical energy has always been a

guarantor of prosperity.

In Germany in particular, due to its

renewable energy policy, there is

growing awareness of the need for

CO 2 -free and at the same time base

load capable energy generation. This

will be the basis for upcoming discussions

on innovative solutions for

managing the energy transition by

2050. Basically the main question is,

how the energy supply can be secured

by that time?

A new energy source is clearly

needed, considering the potentially

devastating impacts of climate change

caused by burning fossil fuels in order

to meet this demand. In the following,

the transition from fission to fusion

and the transfer of existing industrial

nuclear power plant technology knowhow

to new domains of application

such as a future fusion power plant is

discussed.

A brief History

Kraftanlagen Heidelberg (KAH),

founded in 1923, has been working at

the forefront of innovative power

plant technologies. As it can be seen in

Figure 1, starting in 1928 with the

development of Europe’s first coal

fired power plant operated with

100 bar live steam, followed by its

dedication to nuclear power generation

in Germany, the initial involvement

in Nuclear Power Plant (NPP)

new-builds, operational services and

their decommissioning.

Nuclear fusion research has also

come a long way. Beginning with the

first research approaches in the 1920’s

up to the technical implementation

since the 1950’s in the Tokamak and

Stellarator principle, nuclear fusion

research took shape with the Joint

European Torus (JET) in the 1980’s

for the first time on a relevant scale.

In 2005, the first sod was turned in

southern France for the construction

of the International Thermonuclear

Experimental Reactor (ITER), a

nuclear fusion device that shall

demonstrate overall feasibility of

nuclear fusion for power plant application.

Since 2008, KAH has been

developing key systems of ITER. At

that time, the conceptual and detailed

design of the ITER Tokamak Complex

- Detritiation System (TC-DS) as well

as for the detailed design of the Hot

Cell Facility – Detritiation System

(HCF-DS) were executed. During the

course of four years, the design was

successfully finalized in 2012. The

main purpose of the TC-DS is to

provide tritium confinement and gas

detritiation for the buildings of the

Tokamak Complex.

The main TC-DS functions are:

1.) Maintaining sub-atmospheric

pressure

2.) Provision of a pressure cascade

3.) Ensuring detritiation function

The main tasks were:

p Process engineering of the entire

plant to the level of a detailed

design

p Safety engineering: HAZOP

stu dies, PSC, FMEA and RAMI

analysis

| Fig. 1.

Examples from left to right: Almost one hundred years of involvement in conventional

(GKM Mannheim), fission (NPP Biblis) and fusion power plant technology development –

ITER, Cadarache France. [Source: ITER Organization]

Planned entry for

p Layout design of equipment and

piping systems

p Interface control documents and

interface sheets

p Cost estimate throughout all project

phases

The TC-DS is designed to handle all

possible occurrences of tritium contamination

in the plant atmosphere.

Further, there is a dedicated glove box

detritiation system (GDS) to serve the

Tritium Plant systems which are

contained in glove boxes. The GDS

provides the glove box atmosphere

detritiation function. The standard

recirculation process has been

adapted to ITER specific needs and is

designed to cope with a tritium spill

within the glove boxes.

In 2013, Kraftanlagen Heidelberg

was awarded for the detailed design

of the ITER WDS. During the course of

three years, the design was successfully

finalized in 2016 after validation

by an international committee during

the design review meeting held at

ITER headquarters.

The main functions of the ITER-

WDS are:

1.) Provide buffer storage capacity for

tritiated water

2.) Recover tritium from tritiated

water

3.) Discharge detritiated gaseous

hydrogen

Since ITER is a nuclear facility, the

detailed design of the ITER-WDS is

driven by nuclear safety requirements

and safety requirements emerging

from hydrogen explosion risk. Consequently,

besides all the engineering

tasks, the design work included

several safety studies in order to

confirm the system.

Inside the ITER-WDS tritiated water

is converted to gaseous hydrogen and

oxygen using the Combined Electrolysis

Catalytic Exchange (CECE) technology,

which relies on the use of

electrolysers and on Liquid Phase

Catalytic Exchange (LPCE) columns

575

RESEARCH AND INNOVATION

Research and Innovation

From Fission to Fusion – Transfer of Existing Industrial Know-How to New Domains of Applications ı Sabrina Gil Pascual and Andreas Bender


atw Vol. 65 (2020) | Issue 11/12 ı November/December

RESEARCH AND INNOVATION 576

for detritiation of gaseous hydrogen

prior to its discharge to the environment.

The ITER-WDS interacts with the

ITER Isotope Separation System

(ITER- ISS) for tritium recovery.

The knowledge and lessons

learned gained from ITER shall be

used to develop a demonstration

power plant called DEMOnstration

Power Plant (DEMO), where for the

first time electricity produced by

nuclear fusion is being supplied to the

electrical grid.

KAH’s “industrial” task is to realize

a state-of-the-art and foremost

reliable Power Conversion System

(PCS) concept for the future DEMO.

The main challenge here is to cover

the design and layout of most of the

DEMO related components and

systems, and their integration into

the highly sophisticated power block

technology of the DEMO Power Conversion

System (PCS). The focus is

on the technical feasibility, reliability,

efficiency and based on the stateof-the-art

technology for a nuclear

fusion power plant. Our involvement

expanded towards the development

of the DEMO fuel cycle architecture.

Since 2016, several feasibility

studies for the DEMO Balance of

Plant (BoP) were performed in the

following fields:

p Sizing of the main DEMO BoP

components

p Optimization of the current DEMO

BoP Concept with Intermediate

Heat Transfer System (IHTS)

p Evaluation and Approximation of

a minimal Energy Storage System

(ESS) for a Direct-Coupling

Configuration

p Assessment and Optimization of

DEMO Turbine Configuration

p DEMO Thermal Storage Concept

Evaluation

p Initial Cost Estimate - DEMO BoP

p DEMO Fuel Cycle Architecture

It is our apparent understanding of

supporting the DEMO related R&D

process from the angle of “transfer of

existing industrial knowhow” to new

domains of applications. However, the

enclosed references show industry’s

ability of tackling new challenges by

properly handling extreme complex

systems.

Research for tomorrow’s

Energy Supply

As ITER being an experimental reactor,

its main goal is to demonstrate the

technological and scientific feasibility

of nuclear fusion as an energy source.

The ITER project experienced severe

Device / Reactor JET ITER DEMO

Plasma Diameter [m] 5.92 12.4 ≈13-19

Plasma Volume [m 3 ] 80 800 ≈1000-3500

Out-versus-In Power Amplification Ratio [-] ≈0.6 ≈10 ≈25

Fusion Power [MWth] ≈16 ≈500 ≈2000-4000

| Tab. 1.

Comparison between existing (JET), currently built (ITER) and future fusion demonstrator (DEMO).

[Ref.: ITER Organization]

setbacks from the more or less

science- driven project culture. One

symptom of this approach is a project

delay especially for the reactor hot

test up to the year 2035. Due to its

experimental nature (i.e. experimental

campaigns), the thermal power

of 500 MW th generated in the ITER

reactor is not used to produce electricity.

The experience gained from ITER

operation shall enable every ITER

member state to build its own

DEMO, which shall generate electricity

comparable to next generation

NPPs. The ambition for Europe’s

DEMO should be the design and

construction of an industrial prototype

that does require no further

experimental step prior the construction

of a future commercial Fusion

Power Plant (FPP).

How powerful will DEMO be?

In order to facilitate a fusion reaction

inside a hot plasma, large amounts of

external energy are required (e.g. for

a whole cryogenic plant in order to

supply superconducting magnets with

liquid helium/nitrogen as well as

heating systems for reaching plasma

temperatures of approx. 150 Mio. °C).

At JET, a power-out to powerin ratio

(Q) of approx. 0.6 was achieved. For

ITER, a Q value of 10 is anticipated,

which is still not enough to economically

generate electricity. By means of

a larger plasma volume and density

(i.e. more fuel and consequently more

fusion events), DEMO and a future

FPP as its successor will achieve a Q

value of 25 to 100.

The thermal power output of the

European DEMO is planned to be

approx. 2200 MW th . On an international

scale a DEMOnstration

reactor could reach a thermal power

output of 4000 MW th , which would

be the practical equivalent to a nextgeneration

fission reactor of the

type EPR being currently built in

Flamanville, France and Olkiluoto,

Finland. It is not in question, that only

an industry- driven DEMO project can

achieve the required design maturity

to do so.

Synergies Fission & Fusion

At first glance, nuclear fission and

fusion do not seem to be comparable.

For example, nuclear fusion uses a

completely different fuel, namely the

two hydrogen isotopes deuterium and

tritium. As a consequence, the entire

fuel cycle and the auxiliary systems

necessary for operation (e.g. plasma

heating, tritium-processing systems)

together with the reactor principle are

not comparable to nuclear fission.

Although, many similarities and

synergies from an industry’s perspective

are present:

Nuclear Island – The fusion reactor

building and its adjacent auxiliary

buildings are part of the Nuclear

Island, where the same general rules

for the design, construction, operation,

final shutdown, dismantling,

maintenance and monitoring of

nuclear installations are applicable.

Conventional Island – For the

Conventional Island, experience

gained from existing Power Conversion

Systems can be applied to its

nuclear fusion counterpart.

Nuclear Licensing – Countryspecific

licensing procedures for

nuclear installations are applicable

for both fission and fusion reactors.

Nuclear Safety – Concepts and

methodologies can be derived for

nuclear fusion. Radiation protection

rules incl. radiation zoning and occupational

safety standards apply as

well.

Codes & Standards – for design

and operation of nuclear ventilation

systems in terms of confinement, purification,

monitoring, cleaning and

conditioning.

Tritium Handling Technologies –

Experience obtained from pressurized

heavy water reactors (e.g. CANDU

reactors), especially for tritium

extraction technologies that can be

applied to nuclear fusion.

Radioactive Waste Management –

Strategies for fusion activated

materials during operation and

after decommissioning incl. recycling,

clearance, and disposal concepts.

In addition to the points mentioned

above, a wide industrial expe rience is

Research and Innovation

From Fission to Fusion – Transfer of Existing Industrial Know-How to New Domains of Applications ı Sabrina Gil Pascual and Andreas Bender


atw Vol. 65 (2020) | Issue 11/12 ı November/December

| Fig. 2.

The ITER machine’s plasma volume amounts to 800 m 3 at temperatures of approx. 150 Mio°C.

[Source: ITER Organization]

available in the management &

handling of large-scale projects as well

as the lessons learned from the

construction of next-generation

nuclear power plants, which are

extremely valuable for technical

realization of nuclear fusion. Starting

from the coding of systems and

components up to the management of

requirements and interfaces, existing

knowledge and methods have to be

applied to nuclear fusion.

Opportunities & Challenges

of Nuclear Fusion

Nuclear fusion as an energy source of

the future holds some decisive and

promising opportunities:

Nuclear Fusion as Renewable

Energy Source

The heavy hydrogen isotopes deuterium

and tritium are used as fuel.

Abundant resources are available to

provide this fuel, since deuterium is

widely available and tritium is being

bred from lithium 6 as part of the

fusion reactor design.

Nuclear fusion is excellently suited

as a base load power plant and thus

as a replacement of existing nuclear

and coal-fired power plants. A future

energy mix comprising a high amount

of highly volatile energy sources such

as renewables (i.e. wind and solar

power) requires stabilizing large-scale

FPPs.

No Need for a Final Repository

There is no need for a final disposal

of high-level radioactive waste. The

amount of activated structural

material designated for intermediate

storage is comparable with NPP

quantities.

Intrinsically Safe

A nuclear fusion reactor is intrinsically

safe. A runaway nuclear chain

reaction, under any circumstances,

is impossible. Quite the contrary,

maintaining a fusion reaction inside

the magnetically confined hot plasma

proves to be extremely difficult.

Know-How Transfer

The results of fusion research (e.g.

high temperature material science)

can be applied to other industrial

domains.

Of course, due to its innovative and

first-of-its-kind nature, nuclear fusion

has to deal with several technological

challenges:

Challenging Project Management

At the beginning of a large-scale

project such as DEMO deficient

decisions, whether they are of

technical or managerial nature,

have a significant influence on the

technical and economic success.

From an industrial point of view,

one of the challenges identified

in the pre-conceptual phase of

DEMO is the co-decision together

with industry in the preliminary

planning phase.

Maintaining the Fusion Reaction

Maintaining the hot plasma requires

that several systems interact together

at the same time (i.e. plasma heating,

magnetic confinement and an impurity

free vacuum inside the vacuum

vessel). In Figure 2, the ITER machine

is shown with an overall diameter and

height of approx. 30 meters.

Operation Mode

In contrast to the Stellarator principle,

the preferred Tokamak design has the

drawback that it can only be operated

in pulsed operation. The plasma

current is normally induced in the

plasma by a transformer coil. Therefore,

a tokamak does not operate

continuously, but pulsed like the

transformer itself.

High-temperature Materials

Especially for the plasma-facing

components, a high neutron and

consequently heat flux requires composite

materials that can withstand

these extreme conditions. Further

research is required to find optimal

materials for fusion operation conditions.

Efficiency

The power consumption of the

required auxiliary systems is very

high. For a future commercial FPP,

there are many areas that need to be

optimized in order to increase the

power plant’s efficiency.

Conclusion & Outlook

In the second half of 21 st century,

it is highly probable that human

civilization reaches a point when its

resource and energy consumption are

so immense that nuclear fusion may

well be the only way to maintain a

high level of development and prosperity.

It is inevitable to develop and

design unique technical solutions for

nuclear fusion, where research and

development is mandatory to prove

technical feasibility.

For the first fusion demonstrator

DEMO, scientists and engineers have

to cooperate intensively in order to

successfully complete this historical

task. The available industrial knowledge

and experience from the technical

exploitation of nuclear fission

must be considered and integrated in

the design of a future FPP.

However, it is clear that nuclear

fusion is the energy source of choice

for humankind’s future. As nearly

100 years ago, KAH is at the forefront

of technical innovation in order to

provide the necessary support

regarding future fusion power plant

technology for the upcoming generations.

Authors

Sabrina Gil Pascual

Business Development –

Project Manager Innovation

sabrina.gilpascual@

kraftanlagen.com

Andreas Bender

Engineering Department –

Process Engineer

Kraftanlagen Heidelberg GmbH

Im Breitspiel 7

69126 Heidelberg, Germany

RESEARCH AND INNOVATION 577

Research and Innovation

From Fission to Fusion – Transfer of Existing Industrial Know-How to New Domains of Applications ı Sabrina Gil Pascual and Andreas Bender


atw Vol. 65 (2020) | Issue 11/12 ı November/December

RESEARCH AND INNOVATION 578

Planned entry for

This work was funded

by the German

Federal Ministry of

Economic Affairs and

Energy under grant

number 1501568

on the basis of a

decision by the

German Bundestag.

The simulations are

performed with the

code version ASTEC

V2.1, developed by

IRSN.

Simulation of the DEBRIS Experiments

with the Severe Accident Analysis Code

ASTEC

Jan M. Peschel and Marco K. Koch

In the framework of the quenching

experiments the cooling of two

different debris beds was tested under

alternating starting and boundary

conditions such as temperature and

pressure profiles [Lei16; Lei17a].

For this purpose, either cylindrical

particles made of stainless steel or

particles from the PREMIX experiments

are heated up and quenched

subsequently. One of the main

features of the DEBRIS test facility

(Figure 1) is the possibility of

quenching the investigated debris

bed from both the bottom and the top.

Validation work presented in the

following shows the simulation of one

Bottom-Flooding (BF) experiment

conducted in the DEBRIS test facility

with the severe accident analysis code

ASTEC V2.1.1.6, developed by the

French Institut de Radioprotection et

Introduction During a severe accident in a light water reactor (LWR) the persistent

undersupply of cooling water might lead to the destruction of the reactor core. In this process a

debris bed can be formed by the fragmentation of the molten core when getting into contact with

the residual water in the lower reactor vessel plenum. In order to prevent damage of the reactor

pressure vessel (RPV) the cooling of these structures is of great importance. With several

experiments conducted in the DEBRIS test facility the Institute of Nuclear Technology and Energy

Systems (IKE) gained a better understanding of the thermal behaviour and the coolability of

overheated and dry debris beds.

| Fig. 1.

Overall view of the DEBRIS test facility [Lei17b].

de sûreté Nuclêaire (IRSN) [Lab19].

An objective of the simulations is the

analysis and assessment of the code

capabilities to simulate the most

relevant phenomena in particular

related to debris cooling during the

investigated experiment. For the

simulations an input deck is created

considering two different modelling

approaches for BF. The first uses a

fixed input signal for the water input

mass flow, which was also measured

during the experiments. As the input

during the experiments is realised via

a gravity driven input over a constant

water head of 950 mm another modelling

approach with an additional tank

and a gravity driven input will also be

presented.

The following work will give an

overview of the most important information

about the DEBRIS facility, test

procedure as well as the modelling

approaches. After this description,

some selected results regarding

temperature, quench front and

pressure evolution will be presented.

More detailed information can be

found in the Technical Report [Pes20].

DEBRIS Facility and

Test Procedure

The DEBRIS test section (Figure 2) is

made of stainless steel and integrated

in a pressure vessel designed for

pressures up to 4 MPa. In previous

reflooding tests materials like silica

glass, ceramic or concrete were used

for the test section that did not allow

the aspired temperature levels of up to

1,273 K. [Ras17]

The lower part of the test section is

filled with a bed of ceramic particles.

These are used to allow a uniform

input of the water flow. Furthermore,

these particles keep the actual debris

bed in position of the high frequency

inductive heating system. The debris

bed (640 mm height) is either

represented by cylindrical particles

(3x5.75 mm) or by PREMIX particles.

The latter are irregular shaped Al 2 O 3

particles mixed together with

spherical stainless steel particles in

order to improve the heat-up of the

debris bed. For the PREMIX particles a

mixture ratio of approx. 30 % Al 2 O 3

particles and 70 % stainless steel

particles was selected. [Lei17a; Ras17]

For measurement reasons a total

amount of 82 thermocouples (TC)

were installed, with 51 of them

located inside the debris bed. The TCs

are arranged on three different radial

positions: the center (C), the half

radius (HR) and close to the wall (W).

Additionally, a couple of differential

pressure sensors were installed but

as these were not active during the

Research and Innovation

Simulation of the DEBRIS Experiments with the Severe Accident Analysis Code ASTEC ı Jan M. Peschel and Marco K. Koch


atw Vol. 65 (2020) | Issue 11/12 ı November/December

III

| Fig. 2.

Schematic layout of the DEBRIS test section cf.

[Lei17a].

re-flooding experiments usage of

these signals was refrained. For

flow measurement three magnetic

inductive flow meters were installed

inside the pipes. [Lei17a]

Figure 3 gives an overview of the

test procedure for the investigated

DEBRIS experiment.

The test itself can be subdivided

into three phases with specific data

characterising:

Ia/Ib (Multi-step) heat up of the

bed by an inductive heating

coil

II Stabilization on a temperature

plateau of ~600 K

Quenching of the dry and

overheated particles

As there is no information regarding

the energy input during the heat-up

phase available, the works at hand

will concentrate on the simulation of

the quenching phase (III).

Modelling approach

with ASTEC V2.1

The simulations presented in this

work are conducted with the latest

patch version of the integral code

ASTEC V2.1.1 Patch 6, developed by

the IRSN. The code is able to perform

simulations of the most relevant

thermal hydraulic processes during an

incident with core degradation in

light water reactors (LWR). Figure 4

shows a schematic representation of

the modelled DEBRIS test section

with ASTEC V2.1 using the modules

ICARE and CESAR. Whilst ICARE

is used to simulate In-Vessel core

degradation processes, CESAR is used

for thermal hydraulics in the circuits.

The debris bed is separated into

five radial flow channels (CANAL1 –

CANAL5) with an identical modelling

of debris particles. ASTEC provides a

lot of options regarding the modelling

of debris beds. Up to three different

beds can be modelled in order to allow

a differentiation between particle

sources like FUEL, CLADDINGS and

OTHER. Each of these three debris

beds can be modelled with a selectable

number of materials and particle

size distributions. A cylindrical wall

made of stainless steel is the outer

limiting of the modelling with an

inner diameter of 154 mm.

Figure 5 shows the two different

modelling approaches developed with

ASTEC V2.1. As the wall thickness is

not further defined in the experimental

descriptions, a wall thickness of

5 mm is assumed similar to the one

of previous quenching experiments

conducted with the DEBRIS test

facility [Ras17]. The modelled test

section is axially separated into

16 nodes. Inside the area filled with

particles the nodalisation is considered

a little bit finer compared to

the area above the particles. [Bel17]

For the input of water into the

test section there are two different

modelling approaches that are presented

in this work. The main part of

the input deck, the test section, is

modelled with ICARE and is identical

in both cases. The first approach is

characterised by a fixed (F) water

input (SOURCE) on the bottom of

the ICARE module. Additionally,

there is a boundary condition on

pressure ( BCPRESS) on the outlet of

the test section.

The second approach is characterised

by an additional VOLUME and

a PIPE system modelled with CESAR

in order to simulate a gravity driven

(G) water input into the test section.

For this, connections of the type FLOW

are modelled on the inlet and outlet of

the test section. Boundary conditions

on pressure are applied on both the

VESSEL and the VOLUME.

Compared to V2.1.1 Patch 4 a

change is applied to V2.1.1 Patch 6

concerning the capillary pressure in

the momentum equation in porous

media. The capillary pressure has

been identified as responsible for a

diverging presentation of the quench

front evolution compared to the

results from the PEARL experiments,

conducted by IRSN. In the PEARL test

facility, a debris bed was surrounded

by a non-heated bypass in order to

simulate the non-degraded zone in

the reactor core [Rep13]. The results

showed a flat quench front evolution

RESEARCH AND INNOVATION 579

| Fig. 3.

Test procedure of investigated DEBRIS experiment.

| Fig. 4.

Axial and radial nodalisation of DEBRIS test

facility modelled with ASTEC V2.1.

| Fig. 5.

Differences between modelling approaches.

Research and Innovation

Simulation of the DEBRIS Experiments with the Severe Accident Analysis Code ASTEC ı Jan M. Peschel and Marco K. Koch


atw Vol. 65 (2020) | Issue 11/12 ı November/December

RESEARCH AND INNOVATION 580

| Fig. 6.

Solid-liquid contact angle cf. [Coi17].

in both the bypass and the debris bed.

Modelling this bypass with ASTEC

V2.1.1 Patch 4 caused an incorrect

reproduction of the quench front

evolution (illustrated Figure 6).

The calculation of the capillary

pressure (p c ) is described in Equation

1 [Chr06; Coi17]. It depends on

the surface tension of the liquid (σ l),

the mesh porosity (ε), the absolute

permeability (K), the Leverett function

(J (s) ) and the contact angle (θ)

between the solid particle surface and

the liquid.

(1)

In order to improve the calculation of

the capillary pressure the contact

angle was changed from 0 to .

This change indeed means that

the capillary pressure is not considered

in the further calculation of

the liquid pressure (p l ) level (see

Equation 2).

both using the approach with a fixed

input signal.

The temperature results are used

as the main reference parameter in

order to compare the simulation

results with experimental data.

Figure 7 shows the temperature

evolution for thermocouple (TC) T22,

that is located at a height of 290 mm

and radially in the center of the test

section. The respective simulation

results show the particle temperature

of the mesh that is located at the exact

same location.

Although underestimating the

quenching process, the simulation

results are overall in good accordance

with the experimental data. With a

time delay of 19 s with ASV2.1p6 the

least deviations compared to the

experimental results are calculated.

The delay is measured by comparing

the time when both calculation and

simulation fall below the saturation

temperature. ASV2.1p6/G, ASV2.1p4

and ASV2.2b show larger deviations

(up to 41 s) from the experimental

results compared to ASV2.1p6, all of

them showing a quite similar cooling

behaviour. So overall some only

minor differences can be observed

due to the newly applied contact

angle in Patch 6 but still leading

to a better simulation of the

originally observed cooling progress.

Regarding the two selected modelling

approaches ASTEC shows that the

simulations of the DEBRIS test facility

can be very well conducted with a

gravity-driven input. Although the

cooling process is slowed down, only

minor differences appear between

these two approaches.

Before the observed position is

completely cooled down, the simulations

show a noticeable gradually

reduction of the temperature. This

cooling process is linked to the

uprising steam in the nodes located

below those observed here. The

steam causes a cooling of the positions

located above. Due to this, a slowly

dropping temperature evolution can

be observed over the whole course.

When the evaporation process takes

place directly below the observed

node the cooling effect is most

obvious.

Instead of using the particle temperature

the mesh temperature

could be used in order to make a

comparison to the experimental

data. For this, the temperature

needs to be manually calculated by

Equation 3, similar to the calculation

of the averaged density [Aus19;

Pia15].

(3)

(2)

As this change caused differences in

the simulation of the DEBRIS test

facility a version comparison will be

presented in this work. Further on, a

comparison to the beta version ASTEC

V2.2b will be presented. Here, the

IRSN implemented many futures

to improve the code like changing

the thermalhydraulics calculation

( CESAR) from a 5- to a 6-equation

approach.

Simulation Results

In the following graphics results of

four simulation runs will be presented.

Two of them are simulated

with V.2.1.1.6 (ASV2.1p6) con sidering

a fixed input and a gravity driven

input (G). The other two are simulated

with V2.1.1.4 (ASV2.1p4) and

the new beta version V2.2b (ASV2.2b),

| Fig. 7.

Temperature evolution in the center region (T22).

| Fig. 8.

Temperature evolution of thermocouple, mesh and particle temperature.

Research and Innovation

Simulation of the DEBRIS Experiments with the Severe Accident Analysis Code ASTEC ı Jan M. Peschel and Marco K. Koch


atw Vol. 65 (2020) | Issue 11/12 ı November/December

| Fig. 9.

Temperature evolution in the center region (T22).

| Fig. 10.

Quench front progression after 100 s between center line and outer wall.

The mesh temperature (T Mesh ) is calculated

from the liquid temperature

(T L ), the void fraction (x_alf a) and

the gas temperature (T G ). Unlike

the particle temperature the mesh

temperature has a very fluctuating

course (see detailed view in Figure 8),

although the qualitative behaviour is

very similar. This is caused by the

ascending steam due to the cooling

of the overheated particles below.

As the thermocouples are in contact

with the particles both signals from

the simulations can be used.

Figure 9 shows the temperature

evolution of thermocouple T21 that

is located on the same height of the

half radius position. Here, the cooling

process is initiated much later though

being on the same axial position.

Although ASTEC seems to take this

behaviour into account, the experimental

results are overestimated by

all simulations. All simulations conducted

show a very similar cooling

process with a maximum deviation of

only 11 s (ASV2.1p6 and ASV2.1p4)

compared to the test results between.

Figure 10 shows the quench front

progression 100 s after the bed is

quenched for both the experiment and

the four simulations. Although these

graphics are made using interpolated

values, they still give a good overview

of the current status during both the

experiments and the simulations.

Overall all simulations are in good

accordance with the experimental

results. The graphic shows that the

| Fig. 11.

Quench front evolution.

quench front is progressing faster in

the center of the bed during the experiment.

This behaviour is only partially

reproduced by ASV2.1p6. In the half

radius and wall position the cooling

process is overestimated by the simulations.

The other two code versions

observed only show a slightly faster

progressing quench front in the center

of the test section.

Besides the thermal behaviour the

quench front evolution, illustrated in

Figure 11, can be used to compare the

conducted simulations to the experimental

data. As the water level was

not measured during the experiments

the quench front evolution is calculated

manually. For the experiments,

as soon as the temperature drops

below the saturation temperature a

height is assumed to be quenched.

This procedure can lead to a nonfunctional

plot caused by local phenomena

or measurement errors.

The graphic shows that the overall

calculation of the quench front

evolution with ASTEC is in good

accordance with the experimental

results although, especially in the end,

a little bit overestimated by the simulations.

Again, the simulation conducted

with a gravity driven water

input qualitatively show a similar

flooding behaviour compared to the

experimental results. The other

simulations conducted with a forced

water input overestimate the quench

front a little more significant. In

general, the simulations conducted

with Patch Version 4 and the new beta

version provide results closer to the

experiment.

Finally, Figure 12 shows the

released steam mass measured in the

experiment and calculated with

ASTEC V2.1 and the beta version

ASTEC V2.2b. From the beginning to

the end of the quenching progress

the released steam is overestimated by

the simulations.

RESEARCH AND INNOVATION 581

Research and Innovation

Simulation of the DEBRIS Experiments with the Severe Accident Analysis Code ASTEC ı Jan M. Peschel and Marco K. Koch


atw Vol. 65 (2020) | Issue 11/12 ı November/December

[Ras17] Rashid, M.: Coolability of volumetrically heated particle

beds. Dissertation, 2017.

RESEARCH AND INNOVATION 582

| Fig. 12.

Released steam mass.

While the three simulations

calculated with a forced water input

show more or less identical results,

the gravity driven modelling approach

leads to results that show the least

deviations from the experimental

data. With 2.48 kg (ASV2.1p4 and

ASV2.2b) and 2.45 kg (ASV2.1p6)

the experimental results are overestimated

by more than 0.5 kg. Although

the overall course of the gravity driven

setup seems closer to the experimental

data, the final results are still

overestimated by 0.46 kg

Conclusion

The results presented in this paper

show that ASTEC provides overall

good modelling options and analysing

tools in order to simulate the DEBRIS

test facility. Especially in lower positions

of the test section the simulations

are overall in good accordance

with the experimental data. This

manifests itself in the overall temperature

course that was calculated but

also in the progression of the quench

front. Major differences only appear

when analysing the resulting steam

mass released during the test and

calculated by ASTEC.

Although the available modelling

options are already developed to

such a high extent, a few barriers

still appeared while modelling the

experiment. Before the gravity driven

approach was setup with a pipe

system just a FLOW connection was

used to connect the Volume (CESAR)

to the Vessel (ICARE). As this connection

type cannot be closed a

pre-calculation phase could not be

considered. In further releases the

ability to close FLOW connections

could improve the codes capabilities

regarding the simulation of experimental

test facilities.

Furthermore, the changed contact

angle between water and the particles

only partially leads to improved

simulation results. The simulations

conducted in this work show that an

improvement might only apply for

investigated systems using a bypass.

Due to this, a possible improvement

for the models in the future could be

to allow the user to change the contact

angle manually. In [Coi17] a good

wettability is defined by a contact

angle greater than 0° and lower than

90° while greater 90° and below 180°

is a reduced wettability. Further investigations

could show that using other

contact angles would make more

sense instead of just eliminating the

capillary pressure.

In further investigations at PSS the

ability of ASTEC to simulate Top-

Flooding experiments from both the

DEBRIS and the PEARL test series,

conducted by the IRSN, will be considered.

References

[Aus19] Austregesilo, H.; Bals, C.; Langenfeld, A.; Lerchl, G.;

Schöffel, P.; Skorek, T.; Cron, D. von der; Weyermann, F.:

ATHLET – Models and Methods, GRS-P-1/Vol. 3 Rev. 5,

2019.

[Bel17] Belon, S.; Carenini, L.; Chatelard, P.; Coindreau, O.; Topin,

V.: Draft Manual for ASTEC V2.1: ICARE Module – User

manual. Technical Report, PSN-RES/SAG/2016-00421,

2017.

[Chr06] Christiansen, R.: Relative Permeability and Capillary

Pressure. In: Lake, Larry W. [Hrsg.]: Petroleum

engineering handbook. Richardson, Texas: SPE, 2006,

S. 727–765, ISBN: 9781555631130.

[Coi17] Coindreau, O.: Draft Manual for ASTEC V2.1: Physical

modelling of the ICARE module. Technical Report,

PSN-RES/SAG/2016-00422, 2017.

[Lab19] Laborde, L.; Carenini, L.; Chailan, L.; Monod, R.: ASTEC

V2.1.1.6 Release Notes, PSN-RES/SAG/2019-00373,

2019.

[Lei16]

Leininger, S.; Kulenovic, R.: Experimentelle Untersuchungen

zu Kühlbarkeit und Fluten prototypischer

Schüttbett-Konfigurationen - Phase I: Siede- und

Dryoutexperimente. Abschlussbericht zum

Forschungsvorhaben BMWi 1501507, IKE 5-267, 2016.

[Lei17a] Leininger, S.; Kulenovic, R.: Experimentelle Untersuchungen

zu Kühlbarkeit und Fluten prototypischer

Schüttbett-Konfigurationen - Phase II: Flutexperimente.

Abschlussbericht zum Forschungsvorhaben BMWi

1501507, IKE 5-269, 2017.

[Lei17b] Leininger, S.: Experimentelle Untersuchungen der

Kühlbarkeit prototypischer Schüttungskonfigurationen

unter dem Aspekt der Reaktorsicherheit. Dissertation,

IKE 8-126, 2017.

[Pes20] Peschel, J.; Koch, M. K.: Wiederflutung zerstörter Kerne

mit ASTEC. Ruhr-Universität Bochum, PSS-TR-11, 2020.

[Pia15] Piar, L.: ASTEC V2.1: CESAR physical and numerical

modelling. Technical Report, PSN-RES/SAG/2015-00332,

2015.

[Rep13] Repetto, G.; Garcin, T.; Eymery, S.; Fichot, F.:

Experimental program on debris reflooding (PEARL) –

Results on PRELUDE facility. Nuclear Engineering

and Design, Vol. 264, S. 176–186,

DOI: 10.1016/j.nucengdes.2012.11.024.

Authors

Jan M. Peschel

jan.peschel@pss.rub.de

Prof. Marco K. Koch

Ruhr-Universität Bochum

Plant Simulation and Safety Group

Universitätsstraße 150

44801 Bochum, Germany

Research and Innovation

Simulation of the DEBRIS Experiments with the Severe Accident Analysis Code ASTEC ı Jan M. Peschel and Marco K. Koch


atw Vol. 65 (2020) | Issue 11/12 ı November/December

Analysis of the Melt Behaviour in the

Lower Plenum of the TMI-2 Reactor

Using the System Code AC² – ATHLET-CD

Florian Krist and Marco K. Koch

Introduction The ongoing development and validation of severe accident analysis codes is Planned entry for

one of the key topics of current nuclear safety research. These codes applied in the numerical

reactor safety analysis contribute to the further improvement of severe accident management

(SAM) with the purpose to control the progression of postulated accidents and to mitigate its

consequences. Besides post-test calculations, which are necessarily subject to limiting factors, e.g.

experimental focus and transferability, the validation of the simulation codes’ models base is

carried out by the simulation of past nuclear accidents. The well documented Three Mile Island (TMI) accident that

occurred on March 28, 1979 in Dauphin County, near Harrisburg, is suited for both validation computations of the

entire plant and the analysis of selected code models against individual phases in the overall interaction of the accident.

The severe accident analysis code

or system code AC² 2019, is being

developed by the German Gesellschaft

für Anlagen- und Reaktorsicherheit

(GRS) gGmbH. AC² 2019 is able to

perform simulations of thermalhydraulic

processes during an incident

with core degradation as occurred in

block 2 of TMI nuclear power plant

(TMI-2). In this work simulations of

the late accident phase are conducted

using AC² focused on the melt

behaviour relocated to the lower

plenum as result of core degradation.

The processes in the lower plenum

(also referred to as lower head) are

simulated with the two different lower

head modules AIDA and LHEAD

implemented in the program part AC²

– ATHLET-CD (Analysis of Thermalhydraulics

of Leaks and Transients –

Core Degradation). After a brief overview

of the TMI-2 accident progress

the modelling of the TMI-2 plant

with AC² is presented and the major

differences between the two lower

head modules used in the simulations

are described. The analysis is focused

on the simulated temperature evolutions

in the relocated melt and the

RPV wall.

TMI-2 accident

TMI Unit 2 was a two-loop pressurized

water reactor with a design capacity

of 900 MW el (2,700 MW th ). Each heat

transport loop was equipped with

one steam generator and two reactor

coolant pumps. The accident was

initiated by an incident in the

secondary cooling system which

resulted in the loss of all main feedwater

pumps. The reduced heat transfer

in the steam generators led to an

increase of the primary pressure

which caused the reactor scram. The

overpressure could be temporarily

limited by opening of the power

operated relief valve (PORV). As the

valve did not close according to the

design a significant loss of coolant was

initiated. The coolant leakage through

the open PORV continued for approximately

140 min. until the block valve

was closed manually. 101 min. after

scram the last two main coolant

pumps were shut down in reaction to

indications of cavitation. Due to the

loss of coolant and the related fall of

the water level in the reactor the core

heated up. This process was accelerated

in addition by a reduced high pressure

injection (HPI) since an operator

bypassed the HPI actuation signal few

minutes after scram. The thermocouples

in the hotlegs give the first indication

of core un-covery after 110 min.

In the following, as a result of the core

dry out the increasing temperatures

led to a partial degradation of the

core. As result of the closure of the

coolant leakage through the PORV

and the coolant pump transient of

loop B2 beginning at 174 min. the primary

pressure rapidly increased until

the PORV block valve was opened

again at 192 min. aiming to control

the reactor system pressure. Presumably,

the resulting thermal shock

destroyed the remains of the oxidized

and porous fuel rods in the upper core

region. At 200 min. the full HPI was

automatically initiated again. The

water was injected into the cold legs

so that the core was reflooded from

bottom to top. Best estimates indicate

that the core was covered again after

207 min. [Lin98, NSA80, NRC11]

Several simultaneous events indicated

a significant relocation of

the molten core material (corium)

between 224 min. and 226 min. after

scram. According to visual analyses

after the accident the crust surrounding

the corium broke near the core

periphery and in total around 30 t of

the melt pool – that has been formed

as result of the degradation – relocated

towards the lower core support assembly

(CSA). Around 19 t corium

reached the lower plenum that was

filled with cold water by the HPI

draining through the holes in the

elliptical flow distributor (EFD) plate.

During this process that lasted

approximately 120 seconds the reactor

pressure raised by about 2 MPa

indicating strong steam production

caused by the heat transfer from

the melt to the water. The corium

decreased in temperature and

resolidified partly in the CSA. Further,

it is postulated that the initial quantity

of corium that relocated continuously

into the lower plenum rapidly formed

a thick ceramic crust that covered the

majority of the RPV wall and insulated

it from the molten material above.

[Dra04, Lin98, Wol94]

Modelling with AC²

The modelling and the simulation of

the TMI-2 accident are conducted

with the AC² in-vessel programm parts

ATHLET and ATHLET-CD. The simulations

of this work are performed

with code version 3.2. Additionally,

comparative calculations with the

former version 3.1A patch 4 (short:

p4) are conducted in order to investigate

the development progress. The

thermohydraulic processes are simulated

by ATHLET. Figure 1 shows the

modelling of the fluid channels and its

nodalisation by ATHLET’s thermofluiddynamic

module (TFD). The heat

conducting structures are modelled

according to the given masses and

This work is funded

by the German

Federal Ministry of

Economic Affairs and

Energy under grant

number 1501568

on the basis of a

decision by the

German Bundestag.

Responsibility for the

content lies with the

authors.

The results are

obtained using the

GRS program

AC² 2019 developed

by GRS (Gesellschaft

für Anlagen- und

Reaktor-sicherheit

gGmbH).

RESEARCH AND INNOVATION 583

Research and Innovation

Analysis of the Melt Behaviour in the Lower Plenum of the TMI-2 Reactor Using the System Code AC² – ATHLET-CD ı Florian Krist and Marco K. Koch


atw Vol. 65 (2020) | Issue 11/12 ı November/December

RESEARCH AND INNOVATION 584

| Fig. 1.

TFD modelling and nodalisation of the TMI-2 reactor.

heat capacities of the respective plant

components by the heat conduction

and heat transfer module (HECU).

For better clarity the heat conduction

objects (HCO) are not depicted in

Figure 1.

The primary cooling systems are

modelled by the two loops A and B

comprising the hot legs (thermo- fluiddynamic

objects (TFO): PA/B-HL),

the steam generators (PA/B-SG-IN,

PA/B-SG, PA/B-SG-OUT), and the

cold legs (PA/B-CL1 and PA/B-CL2).

The boundary conditions of the steam

generators as well as the start-up and

shutdown behaviour of the four main

coolant pumps, coupled to the cold

legs, are controlled by signals of

ATHLET’s General Control Simulation

Module (GCSM). Further, the highpressure

injection is connected by the

junctions HPIA/B1 and HPIA/B2 with

the cold legs. The HPI, the pressurizer

and the PORV are controlled by GCSM

signals containing measurement data.

The boundary conditions of the secondary

cooling circuits, modelled as

open systems, are calculated by means

of GCSM signals containing measurement

data as well.

The reactor core is modelled by

five flow paths (TFO: PV-CORE1-5)

and four HEAT (ROD1-4) objects. The

latter are modelling objects of the

ATHLET-CD module ECORE and

represent the core region. Every HEAT

object (rod) is coupled to the corresponding

TFO and represents a core

section which corresponds in total to

the number of fuel and absorber rods

that were implemented in the TMI-2

core. For a detailed description of the

nodalisation of the core region see

[Dra04].

The lower plenum is modelled by

the two TFO PV-LH and PV-LP. PV-LH

is coupled with the HCO simulating the

lower head wall, HPV-LH1 representing

the cylindrical part and HPV-

LH2 representing the bottom part.

The analysis of the simulation

results is focused on the behaviour of

the relocated melt during the late

phase of the accident. The late phase

phenomena can be simulated by

ATHLET-CD. In the current state of

ATHLET-CD the two modules AIDA

and LHEAD are applicable to simulate

the thermalhydraulic processes in

the lower plenum. AIDA, a coupled

integral simulation module, can simulate

the segregation of the corium into

a heavy oxidic phase (bottom layer)

and a light metallic phase (top layer).

The melt pool is calculated zerodimensional

using separate energy

balance equations for each layer with

empirical correlations for the calculation

of the heat flows between melt

and crust. The heat conduction in the

RPV wall is simulated by a two-dimensional

model applied on a spherical

geometry. In the AIDA simulations the

wall structure is subdivided into

30 axially nodes along its length and

5 radial layers. In ATHLET-CD 3.2 a

new extended model considering

both axial and radial heat conduction

is available and applied in the

present simulations. The heat transfer

between melt and crust is calculated

with the recommended correlation

options Kelkar and Asfia-Dhir. [Aus15,

Wie19]

In LHEAD a heat conduction model

is applied on a two-dimensional grid

with cylindrical coordinates [Aus15].

The thermohydraulics of the coolant

in the lower head is simulated by the

ATHLET module TFD – represented by

the TFO PV-LH in the modelling. The

RPV wall is modelled with ATHLET’s

module HECU. The HCO simulating

the lower head wall, HPV-LH1 and

HPV-LH2, are coupled with LHEAD.

HPV-LH1 and HPV-LH2 are each coupled

with the lower head TFO on the

one hand and to lower head module

on the other. HPV-LH2 representing

the bottom part with a diameter of

2.01 m is coupled over a length of

0.13 m to PV-LH. HPV-LH1 above

whose diameter increases from 2.01 m

to 4.34 m is coupled over 1.31 m to the

TFO without overlap with HPV-LH2.

Research and Innovation

Analysis of the Melt Behaviour in the Lower Plenum of the TMI-2 Reactor Using the System Code AC² – ATHLET-CD ı Florian Krist and Marco K. Koch


atw Vol. 65 (2020) | Issue 11/12 ı November/December

| Fig. 2.

Primary pressure evolution in TMI-2: Measured data and simulation results of different code versions.

| Fig. 3.

Void in the RPV 207 min. after scam.

In both modules AIDA and LHEAD

of AC² 2019 and in AIDA of ATHLET-

CD 3.1A p4 melt stratification is

considered. In LHEAD of code

version 3.1A melt stratification is not

yet available. In order to allow for

direct comparability, as a first step the

new model for radial relocation in

AC² 2019 (see [GRS19]) is not activated.

Further, the index for the melt

fragmentation model which is to be

input since AC² 2019 is set to the

default value.

Simulation Results

In the following some selected simulation

results are presented. Figure 2

shows the primary pressure evolution

during the first 260 min. after reactor

scram simulated by ATHLET 3.1A

patch 4 and ATHLET 3.2 compared to

the TMI-2 measurement data.

The comparison shows in general a

good qualitative accordance of the

simulated pressure behaviour with

the measurement data. The pressure

evolutions simulated by the two code

versions show only minor differences

especially during the first 60 min.

In the period between 60 min. and

110 min. the pressure is slightly

overestimated by the simulations

whereat the overestimation is marginally

reduced in the run of ATHLET 3.2.

In the subsequent period until the

begin of the pump transient the

pressure evolution is reproduced

precisely by the simulations. During

the transient, between 174 min. and

193 min. large amounts of water were

vaporized on the hot degraded core

which led to a strong increase in

pressure [And87]. The simulation

results are in good accordance with

the measurement data showing only a

slight delay of the pressure increase.

Quantitatively the pressure increase

up to 14 MPa is reproduced more

precisely with the newer code version

3.2. Also, the pressure drop as result

of the re- opening of the PORV at

192 min. is represented well however

the further decrease in pressure

during the HPI is underpredicted.

In all probability during this

period the degraded core is reflooded

completely by the cold injected

water [Lin98] which is reproduced

by the simulation as depicted in

Figure 3. The pressure decrease

starting from 200 min after scam

indicates significant condensation of

the steam generated during the

pump transient. First, the pressure

decrease agrees with the measured

data but then appears to be delayed

as the pressure shows a significant

drop only at 220 min. This deviation

has already been appeared in earlier

simulations, as shown in [Dra04,

Hof10, Hof13]. In the following in

the LHEAD simulation the pressure

rise resulting from the relocation of

corium is represented qualitatively

but is calculated slight too late as

well. In contrast the increase is not

represented in the AIDA simulations

(in both code versions). The reason

therefore is not yet clear and thus

subject of further investigation.

In those runs conducted with

AC 2019 the simulated corium mass

relocated to the lower plenum is 20.2 t

and thereby slightly higher compared

to code version 3.1A p4 which

calculates 19.2 t (AIDA) or 19.1 t

(LHEAD), respectively. The simulation

results are thus in good accordance

with estimated post-accident core

material in the lower plenum of 19.1 t

(see [Ake90]).

Figure 4 shows the melt temperature

evolution beginning at the time of

relocation simulated by AIDA and

LHEAD. In case of LHEAD the simulations

results are averaged over the four

core sections. In comparison of the

code versions a converging development

concerning the melt temperature

evolutions can be seen. While in code

version 3.1A p4 AIDA calculates

the melt temperature approximately

constant in ATHLETCD 3.2 AIDA

predicts a declining behaviour –

analogous to the decreasing course

calculated by the LHEAD runs.

Regarding the heat sink provided by

the water-filled lower plenum the

decreasing temperature evolution

simulated by ATHLET-CD 3.2 appears

more plausible. Possibly, these different

temperature evolutions are

caused by the enhanced coupling of

AIDA in ATHLET-CD 3.2. The cooling

condition in the lower plenum is

RESEARCH AND INNOVATION 585

Research and Innovation

Analysis of the Melt Behaviour in the Lower Plenum of the TMI-2 Reactor Using the System Code AC² – ATHLET-CD ı Florian Krist and Marco K. Koch


atw Vol. 65 (2020) | Issue 11/12 ı November/December

RESEARCH AND INNOVATION 586

| Fig. 4.

Averaged melt temperature evolution simulated by AIDA and LHEAD.

defined via GCSM signal containing

the void fraction of the lower head

TFO in this code version while in

ATHLETCD 3.1A p4 the cooling condition

is defined as model option

( initial configuration).

In the following paragraphs the

crust thickness and wall temperature

evolutions simulated by AIDA are

presented. Figure 5 shows a schematic

draw of the modelling with

AIDA including the nodalisation.

As mentioned above the RPV wall

is represented by a 2-dimensional

spherical geometry. Node 01 represents

the lowest part of the RPV wall,

whereas node 30 is the uppermost

node of the symmetric lower head

wall. Additionally, the corium configuration

as obtained with AIDA in

AC² 2019 is represented by the oxidic

(orange) and metallic layer (purple).

In Figure 6 the oxidic crust

thickness at the Nodes 01, 04 and 06

simulated by AIDA in ATHLET-CD

3.1A p4 and ATHLET-CD 3.2 is

depicted. While in AIDA 3.1A p4 a

distribution with values between

0.04 m and 0.09 m is calculated

AIDA 3.2 predicts values around

0.07 m and only minor differences

between the simulation results of the

individual nodes. Compared with

estimates for the thickness of the

insulating layer in TMI-2 based on the

nozzle damage profile in the lower

head of 0.15 m and 0.25 m (see

[Wol94]) the simulation results are on

a similar scale. When discussing the

crust formation, it should be noted,

that AIDA does not consider debris

bed formation as it calculates the melt

pool zero-dimensional with averaged

properties. Hence, the real melt

distribution is highly idealised.

Further, the period of relocation is

calculated significantly faster (within

10 seconds) than it is assumed

in literature (see [Wol94]). Thus, the

cool down and solidification of the

initial portion that formed presumable

the insulation layer could be

underpredicted for that reason.

Figure 7 shows the inner wall

temperature distribution simulated by

AIDA in ATHLETCD 3.2. In general,

the simulation results show a steady

progression increasing from a low

level towards approximately the

temperature level of the melt pool.

In comparison with estimated

values published in literature based

on metallurgical examinations of

peak temperatures at 1,100 °C the

simu lated temperature evolution is

moderately overestimated [Wol94].

| Fig. 5.

Sketch of the AIDA nodalisation, from CView 2.1.

Immediately after relocation AIDA

calculates a temperature level of

approx. 750 K for those nodes that

are in contact with the oxidic melt

(N01 - N06). At node 07 which is in

contact with metallic melt the

temperature is at a lower level (500 K).

The difference indicates the separate

calculation of the energy balance

of the oxidic and the metallic layer.

In the following the simulated wall

temperature in node 07 increases only

moderately which is plausible in

relation to the cooling condition

and the consideration of axial heat

conduction along the wall.

| Fig. 6.

AIDA simulations: Crust thickness simulated by ATHLET-CD 3.1A p4 and ATHLET-CD 3.2.

| Fig. 7.

AIDA simulation: Inner wall temperature evolution simulated by ATHLET-CD 3.2.

Research and Innovation

Analysis of the Melt Behaviour in the Lower Plenum of the TMI-2 Reactor Using the System Code AC² – ATHLET-CD ı Florian Krist and Marco K. Koch


atw Vol. 65 (2020) | Issue 11/12 ı November/December

| Fig. 8.

LHEAD simulation: Inner wall and fluid temperature evolution in lower head simulated by ATHLET-CD 3.2

Figure 8 shows the inner wall

temperature evolution when using

LHEAD compared to the fluid

temperature in the lower head TFO

PV-LH simulated by ATHLET-CD 3.2.

The melt pool height calculated by

LHEAD is 0.33 m and thus considerably

lower than in AIDA. Never theless,

the HCO simulating the lower head

wall, HPV-LH1 and HPV-LH2, are

geometrically at least partially both

in contact with the melt pool. The

simulation results show that the

temperature evolution corresponds

approximately to the temperature

of the coolant calculated in the

lower plenum TFO until relocation.

The decrease in temperature after

200 min. is related to the HPI which is

initiated again at this time. In contrast

to the increasing wall temperatures in

the AIDA simulation the simulated

wall temperature drops here abruptly

to 310 K (HPV-LH1) or 277 K (HPV-

LH2) respectively in the moment of

relocation. Thus, the retroactive

effect from LHEAD to HECU appears

not plausible. Further investigation

concerning the modelling should

concentrate on possible geometrical

inconsistencies which could affect

the simulation results. In addition,

variations of the geometrical dimensions

could reduce the deviations

in the calculated melt pool height.

Nevertheless, an issue concerning the

coupling between LHEAD and HECU

might also prevent the correct internal

transfer of values and could thus

cause the unexpected temperature

progression.

Conclusion

The thermohydraulic processes

during the accident progression of

TMI-2 can be reproduced in good

accordance with the available

measurement data (see also [Hof10,

Hof13]). The melt temperature

evolution simulated by AIDA and

LHEAD of AHTLETCD 3.2 show each a

decay characteristic which appears

plausible since the lower plenum is

filled with coolant. Especially compared

to the former version of AIDA

which calculates an approximately

constant melt temperature evolution

this represents a significant development.

In AIDA the simulated wall temperature

evolutions in the lower

section shows overall plausible progressions.

AIDA predicts a con tinuously

increase towards the temperature of

the melt pool. Compared with estimations

based on metallur gical examinations

of the TMI-2 lower head the

temperature increase is moderately

overestimated. When using LHEAD the

HCO simulating the lower plenum wall

do not show an increase in temperature

as it would be expectable in consequence

of the heat input by the

relocated melt. Instead, the temperature

drops abruptly in the moment

of melt relocation. An explanation

therefore would be an issue concerning

the coupling between the modules

LHEAD and HECU.

References

[Ake90] Akers, D., McCardell, R.: TMI-2 Core materials and fission

product inventory. Nuclear Engineering and Design,

Vol. 118, Issue 3, p. 451-461, 1990.

[And87] Anderson J. L.: Steam generator secondary side effects

upon primary side thermal-hydraulics during the TMI-2

accident, EGG-TMI-7482, EG&G Idaho, Inc., Idaho Falls,

1987.

[Aus15] Austregesilo, H.: Modelling of the lower head of

a BWR with the code ATHLET-CD - Technische Notiz.

TN-AUH-01/15, Gesellschaft für Anlagen- und

Reaktorsicherheit gGmbH, 2015.

[Dra04] Drath, T. et. al.: Simulation des TMI-2-Unfalls mit dem

Programmsystem ATHLET-CD (Teil 1), 7. Technischer

Fachbericht (Teil 1) zum Vorhaben BMWi 1501241,

LEE-24, Ruhr-Universität Bochum, 2004.

[GRS19] Gesellschaft für Anlagen- und Reaktorsicherheit gGmbH:

ATHLET-CD 3.2, Program Updates since Release

ATHLET-CD 3.1A, 2019.

[Hof10] Hoffmann, M. et. al.: Simulation des TMI-2-Unfalls mit

dem Programmsystem ATHLET-CD, 7. Technischer

Fachbericht zum Forschungsvorhaben BMWi 1501305,

LEE-57, 2010.

[Hof13] Hoffmann, M. et. al.: Anlagenrechnungen mit dem

Systemcode ATHLET-CD: TMI-2-Unfall und alternative

Szenarien im Rahmen des OECD/NEA Code-Benchmarks

„ATMI“, 4. Technischer Fachbericht zum Forschungsvorhaben

BMWi 1501385, LEE-85, 2013.

[Lin98]

Linnemann, T. et. al.: Review of the TMI–2 Accident and

Late Phase SFD Code Modeling with View on Material

Movement to and Behaviour in the Lower Head,

Investigation of Core Degradation (COBE), RUB E-193,

Ruhr-Universität Bochum, 1998.

[NSA80] Nuclear Safety Analysis Center: Analysis of Three Mile

Island - Unit 2 Accident, operated by the Electric Power

Research Institute, 1980.

[NRC11] United States Nuclear Regulatory Commission (USNRC):

Pressurized Water Reactor, Crosstraining Course Manual,

Three Mile Island, B&W Technology, Rev. 07/2011.

[Wie19] Wielenberg, A. et. al.: Recent improvements in the

system code package AC² 2019 for the safety analysis of

nuclear reactors. Nuclear Engineering and Design, Vol.

354, 2019.

[Wol94] Wolf, J. R. et. al.: TMI-2 Vessel Investigation Project

Integration Report, Report NUREG/CR-6197; EGG-2734,

EG&G Idaho, Inc., Idaho Falls, 1994.

Authors

Florian Krist

florian.krist@pss.rub.de

Prof. Marco K. Koch

Ruhr-Universität Bochum

Plant Simulation and Safety Group

Universitätsstraße 150

44801 Bochum, Germany

RESEARCH AND INNOVATION 587

Research and Innovation

Analysis of the Melt Behaviour in the Lower Plenum of the TMI-2 Reactor Using the System Code AC² – ATHLET-CD ı Florian Krist and Marco K. Koch


atw Vol. 65 (2020) | Issue 11/12 ı November/December

RESEARCH AND INNOVATION 588

Computational Heat Transfer Analysis

of Tubes and Tube Bundles

with Supercritical Water as Coolant

Kashif Tehseen, Kamran Rasheed Qureshi, M. Abdul Basit, Rab Nawaz, Waseem Siddique and Rustam Khan

1 Introduction The study of heat transfer in supercritical fluids has gained a significant attention and provides

a wide scope of future studies. A number of experimental as well as computational studies has been performed to

estimate the convective heat transfer coefficient (HTC) in supercritical water (SCW) and development of a suitable

correlation to predict the heat transfer at supercritical conditions. Heat transfer at supercritical conditions involve two

major phenomena i.e. heat transfer enhancement (HTE) and heat transfer deterioration (HTD). These phenomena of

heat transfer in SCW are not completely understood yet but it has been shown that the heat transfer deteriorates more

and more with increase in wall heat flux and decrease in mass flux of the inlet fluid [1].

The main purpose of development of

heat transfer correlation for different

geometries is to develop a general

correlation that would be applicable

to all practical geometries such as

tube bundles to study the heat transfer

phenomenon in the nuclear reactor

core with SCW as coolant. Detailed

experimental studies of heat transfer

in SCW flowing along vertical as well

as horizontal tubes were performed

by Yamagata et al. [1]. They presented

the profiles of HTC and suggested a

modified Dittus-Boelter correlation to

predict the heat transfer coefficient

for SCW flowing along bare circular

tubes. A number of other researchers

including Swenson et al. [2], Jackson

et al. [3], Watt et al. [4], Gorban et al.

[5], Jackson [6] and Mokry et al. [7],

developed their own correlations

to determine the heat transfer coefficient

for SCW using different geometric

configurations. But these

correlations are valid for some specific

range of operating parameters and a

single correlation predicting all

phenomena of heat transfer such

as HTE and HTD is not available.

| Fig. 1.

Variation of thermophysical properties of water with temperature.

Dyadyakin & Popov proposed the only

correlation available in literature that

is able to predict the heat transfer

in 7-element tight lattice geometry

having helical fins on the elements

using SCW as coolant [8].

To maintain the whole experimental

setup at constant operating conditions

in supercritical region is much

difficult and costly. In this scenario,

Computational Fluid Dynamics (CFD)

plays a vital role for the investigation

of heat transfer in supercritical fluids

(SCF) by providing detailed 3-D

picture of the phenomena in complex

geometries and severe operating

conditions that would have been

very difficult and costly to obtain

experimentally. One of the most

important requirement of computational

studies of heat transfer in

supercritical fluids, is to take into

account the variation of thermophysical

properties because the

accuracy of computational results

depends mainly on these properties.

Figure 1 shows the variation of these

properties with temperature at an

operating pressure of 24.52 MPa.

Different authors used different

approximations for incorporation of

the variation of these thermophysical

properties in the solver. Farah et al.

[9] discussed methods of incorporation

of IAPWS equations and NIST

real gas models in ANSYS FLUENT.

They also discussed the accuracy of

turbulence models like k-ε and k-ω by

verifying the results obtained for heat

transfer coefficient in SCW flowing

along vertical tubes. Sharabi et al.

[10] proved the validity of k-ω and k-ε

turbulence models at lower values of

mass and wall heat fluxes. Withag et

al. [11] studied the effects of mass flux

on the accuracy of results and concluded

that higher values of mass

fluxes would give more accurate

results because of reduction in effects

of gravity. The presence of recirculation

at inlet was also observed by the

authors at reduced values of mass

flux.

Among the various factors that

affect the accuracy of the results, the

most important is the generation of

grid and setting appropriate value

of y + , a dimensionless number

explaining the refinement of mesh at

heated wall. Roelofs [12] explained

the importance of y + in obtaining

accurate results during

CFD simulations of heat transfer

in SCW flowing along a vertical tube.

He tested different values of y + and

proved that highly accurate results

can be obtained with y +


atw Vol. 65 (2020) | Issue 11/12 ı November/December

[16], Palko & Anglart [17] and Yang

et al. [18] simulated the different

configurations of vertical tube and

tube bundle while Z. Shang & S. Lo

[19] have performed CFD simulations

for heat transfer in SCW for horizontal

flows using full geometry. These

authors specified that the use of k-ε

turbulence model produces a good

match between computational and

experimental results.

The present study contributes in

understanding the phenomenon of

heat transfer along different core

orientations of supercritical watercooled

reactors. Phenomenon of

heat transfer in SCW from horizontal

and vertical bare tubes and tube

bundles has been studied and effects

of flow orientation on heat transfer in

SCW have been investigated along

tubes and tube bundles. NIST real gas

model has been used to incorporate

the thermophysical properties of

SCW. The results for horizontal as

well as vertical bare circular tubes

have been validated with Yamagata et

al. [1] experimental results and

compared with values calculated by

Dittus- Boelter correlation and the

pheno mena of HTE and HTD have

been observed. Simulations of SCW

flow through tube bundles consisting

of seven similar tubes have also been

performed. In these simulations of

the tube bundles, both horizontal

and vertical orientations of tube

bundles have been studied. Moreover,

in vertical orientation of bundle,

flow of coolant both along the

gravity direction and against the

gravity direction has been considered

and results of variation of flow

direction and bundle orientation

have been compared and discussed.

The effects of change in length and

diameter of the heated tube on heat

transfer in SCW have also been investigated.

A comparison of phenomenon

of HTD has been made for vertically

upward, downward and horizontal

flows to check the severity of HTD

in these orientations. In simulation

of flow through horizontal tube

bundle, half symmetry of the geometry

has been employed and to

the knowledge of the authors,

symmetry has never been used for

simulating the flow of supercritical

water through horizontal bundle

before this study. Use of symmetry

condition results in reduced computational

time and resources and

it can prove to be of great importance

for the simulations for core of

nuclear reactors with a large number

of fuel rods.

(a) Tube with Vertically Upward Flow

| Fig. 2.

Schematics of geometries used.

Location Property Value Units

Tube Bundle

Inlet Mass flux 1260 1260 kg/m 2 -s

Inlet Turbulence intensity 5.0 5.0 %

Inlet Hydraulic diameter 7.5 4.217 Mm

Outlet Constant pressure 24.52 24.52 MPa

Wall Heat flux 233,930 600 kW/m 2 -s

Developing Length Heat flux 0 - kW/m 2 -s

| Tab. 1.

Boundary conditions.

(a) Tube for Horizontal Flow

| Fig. 3.

Mesh generated for horizontal tube and tube bundle.

2 Description of Problem

Geometry

The experiments performed by

Yamagata et al. [1] have been used as

a benchmark for the CFD simulations

using ANSYS FLUENT 14.0. Test

section used by Yamagata et al. [1]

was of 7.5 mm inner diameter and

length of 1500 mm made of AISI

type stainless steel were used for

experimentations. In their set up, a

developing length of 500 mm was

provided at the inlet to ensure

hydraulically developed flow at inlet

of test section. Figure 2 (a and b)

shows a schematic model of tube for

vertically upward flow and tube

bundle for horizontal flow used for

simulations. The same tube is used for

tube bundle, so the length of tube

bundle is 1500 mm with tube inside

diameter of 7.5 mm and pitch of

(b) Tube Bundle with Horizontal Flow

(b) Tube Bundle

9.0 mm was used. Boundary conditions

used for simulations are

summarized in Table 1. Simulations

are also performed for other flow

orientations (vertically upward,

vertically downward and horizontal)

along single tube and tube bundle

as well.

3 Mesh Independence

Study

The mesh was generated using ANSYS

Mesher. The mesh generated is

shown in Figure 3. The accuracy of

numerical results computed using

CFD methodology depends greatly

on the quality of grid as well as

the grid independency (Figure 4).

An axisymmetric model was used for

simulation of flow along vertical tube

to save computational time and the

results got mesh independent after

RESEARCH AND INNOVATION 589

Computational Heat Transfer Analysis of Tubes and Tube Bundles with Supercritical Water as Coolant

Research and Innovation

ı Kashif Tehseen, Kamran Rasheed Qureshi, M. Abdul Basit, Rab Nawaz, Waseem Siddique and Rustam Khan


atw Vol. 65 (2020) | Issue 11/12 ı November/December

RESEARCH AND INNOVATION 590

| Fig. 4.

Mesh independence study.

185,000 elements. A 3-D model was

used for simulation of flow along

horizontal tube and the results

got mesh independent after

2.6901 million elements. Half

symmetry was used for case of tube

bundle and mesh independence

was achieved after 2.922 million

elements.

4 Results and Discussion

4.1 Horizontal and vertical

tubes

4.1.1 Results validation

Experiments of Yamagata et al. [1]

were used as benchmark and Figure 4

shows that heat transfer coefficient

has been predicted very well using

ANSYS FLUENT Code 14.0 and RNG,

k-ε model with EWT. Thermal hydraulics

of vertical flows is quite simple as

compared to horizontal flows due to

influence of buoyant forces. Figure 5

shows variation of HTC with the fluidbulk

temperature. It is clear that in the

region of compressed fluid, HTC

varies almost linearly with fluid-bulk

temperature. As we enter the near

pseudocritical region, a sharp increment

in HTC can be observed. The

maximum value is attained at the

pseudocritical point (~380 °C). After

this point, heat transfer coefficient

decreases sharply which may result in

increased wall temperature which is

undesirable for the safe operation of

nuclear reactors. Due to effects of

gravity a slight variation of HTC can

be seen at bottom and top wall

which becomes more prominent by

increasing wall heat flux due to

upward movement of heated fluid. So

different values of HTC exist at top

and bottom wall due to different wall

temperatures at top and bottom

surfaces of horizontal tube.

4.1.2 Heat transfer deterioration

According to F. Wang et al. [20], if

HTC calculated using Dittus-Boelter

correlation is lesser than its predicted

value, the phenomenon of HTE occurs.

But if it is greater than the predicted

value, HTD is present. Figure 6 (a)

makes it clear that HTD cannot be predicted

with Dittus- Boelter correlation.

This phenomenon might be crucial for

the core of reactor as the wall temperature

goes on increasing sharply

due to reduction in HTC. It occurs due

to turbulence damping caused by

combined effects of buoyancy and

acceleration and can be reduced with

vertically downward flows and the

reason is the elimination of HTD due

to effects of buoyancy [21]. Figure 6

(b) provides a com parison of HTD for

case of vertically upward, downward

and horizontal flows.

For horizontal tube, HTC at top

and bottom surface is different. This

difference increases with increased

heat flux and reduced mass flow rates

as is obvious from comparison of

Figure 5 (b) and Figure 6 (b). This is

due to effects of buoyancy, upward

movement of low density liquid. This

causes a region of high temperature at

top and region of low temperature at

bottom of the horizontal tube.

| Fig. 5.

Results validation for vertical and horizontal flows.

(a) HTD for Vertical Tube

| Fig. 6.

HTD in vertical and horizontal tubes.

(b) Comparison of HTD

4.1.3 Effects of length and

diameter

Figure 7 (a) and (b) show the effects

of length and diameter on HTE and

HTD respectively for vertically upward

flows. A decrease in HTC can be

observed with increased diameter and

decreased heated length. As fluid

outlet and bulk temperature depends

on heated length, so a decreased HTC

was obtained by decreasing the heated

length. As fluid-bulk tem perature

reaches pseudocritical temperature,

HTC for both lengths becomes similar.

The value of HTC at pseudocritical

point is larger for decreased heated

length. The reason is the decrement of

HTC with tem perature after pseudocritical

point. Increase in fluid bulk

temperature is lower for reduced heated

length causing an increased value of

heat transfer coefficient for reduced

length after pseudocritical point. Same

effect was observed by increasing the

dia meter of the tube. The value of HTC

Research and Innovation

Computational Heat Transfer Analysis of Tubes and Tube Bundles with Supercritical Water as Coolant

ı Kashif Tehseen, Kamran Rasheed Qureshi, M. Abdul Basit, Rab Nawaz, Waseem Siddique and Rustam Khan


atw Vol. 65 (2020) | Issue 11/12 ı November/December

4.2 Tube bundle

(a) Wall Heat Flux = 233 kW/m 2

| Fig. 7.

Effects of length and diameter on HTC for vertically upward flows.

is lower for increased diameter and

decreased heated length, reason might

be the variation of properties of SCW

and some other thermal or hydraulic

reasons as the phenomenon of heat

transfer at supercritical con dition is

not fully understood yet.

Now wall heat flux was changed to

930 kW/m 2 -s, it was observed that by

decreasing the heated length, HTD got

more severe for the compressed liquid

region. But as we move towards the

pseudocritical point, an improvement

in heat transfer was observed. As fluid

bulk temperature reaches far away of

(b) Wall Heat Flux = 930 kW/m 2

pseudocritical point, heat transfer

deterioration got it severity again as

thermophysical properties of SCW are

constant again, shown in Figure 7 (b).

But if we increase the diameter of

the tube by keeping the heated length

constant, HTC obtained are always

lesser explaining that the phenomenon

of heat transfer deterioration gets more

and more severe by increasing the

diameter of tube. The severity of this

phenomenon is more in regions where

the properties of fluid are almost

constant like in region of compressed

liquid as shown in Figure 7 (b).

4.2.1 Wall temperature

distribution

Figure 8 (a, b and c) gives the temperature

distribution at the walls of

tube bundle. For the case of horizontal

tube bundle, the effects of buoyancy

causes the low density (high temperature)

fluid to move upwards while

the high density (low temperature)

fluid moves downwards causing an

increased temperature of top wall, so

the region of maximum temperature is

at top wall near outlet of tube bundle.

It is also clear from the Figure 8 (a)

that the temperature distri bution is

neither uniform in axial direction nor

in circumferential direction which

might result in production of thermal

stress in the tube bundle. It is also clear

that the tem perature distribution

possesses half symmetry.

Now if we look at Figure 8 (b and

c) it becomes clear that the distribution

of wall temperature is rather

uniform for the case of vertical orientations

of coolant flow. The presence

of a little lower wall temperatures

RESEARCH AND INNOVATION 591

(a) Horizontal Flow

| Fig. 8.

Wall temperature distribution.

(b) Vertically Upward Flow

(c) Vertically Downward Flow

(b) Vertically Downward Flow

(a) Horizontal Flow

| Fig. 9.

Wall temperature distribution for rods of tube bundles.

(c) Vertically Upward Flow

Computational Heat Transfer Analysis of Tubes and Tube Bundles with Supercritical Water as Coolant

Research and Innovation

ı Kashif Tehseen, Kamran Rasheed Qureshi, M. Abdul Basit, Rab Nawaz, Waseem Siddique and Rustam Khan


atw Vol. 65 (2020) | Issue 11/12 ı November/December

RESEARCH AND INNOVATION 592

(a) Horizontal Flow

| Fig. 10.

Secondary flows at a cross section 1.4 m from inlet.

(b) Vertically Downward Flow

(c) Vertically Upward Flow

(a) Horizontal Flow

| Fig. 11.

Temperature profiles at surfaces of rods.

(b) Vertical Flow

(c) Comparison for Vertical and

Horizontal Orientations

at the outlet can be observed for

vertically upward flow due to effects

of buoyancy.

Figure 9 (a, b and c) shows the

wall temperature distribution of different

rods of vertical and horizontal

tube bundles. Figure 9 (a) makes it

clear that a non-uniform wall temperature

distribution is present for

horizontal tube bundle. Maximum

temperature was observed in the

upper region of tube bundles at

rod 4 due to upward movement of

the high temperature fluid. Very

strong secondary flow is observed in

horizontal tube bundle due to this

uneven distribution of temperature.

These secondary flows tries of even

out that temperature differences present

at top and bottom of the tube bundle.

Figure 10 (a, b and c) shows the

secondary flows developed in

tube bundles. To have a good visibility,

the secondary flows for vertical tube

bundles are given a factor of 10.

Temperature profiles for different

rods of horizontal tube bundle are

shown in Figure 11 (a). The presence

of high secondary flows in horizontal

tube bundle is a reason of relatively

reduced wall temperatures.

All rods of vertical tube bundles

bears relatively uniform temperature

distributions as shown in Figure 9

(b and c). The results for horizontal

flows bears half symmetry while

1/8 symmetry is present in the

results for vertical flows. As all of

the outer rods for vertical flows bear

the same temperature distribution,

due to symmetry, only the central

and one outer rods are shown in

Figure 9. Temperature profiles of

vertical flows (upward and downward),

shown in Figure 11 (b and c),

makes it clear that the wall

temperature of outer rod is higher for

vertically downward flow as compared

to upward flow due to upward

movement of high temperature fluid

under buoyancy. This movement

causes the fluid bulk temperature

to increase to higher value as compared

to upward flow and leading

towards the phenomenon of pseudofilm

boiling which causes high wall

temperatures.

4.2.2 Temperature distribution

at a cross section 1.4 m

away from inlet

To have a better understanding of

thermal hydraulics in tube bundle

cooled with SCW, the contours of

temperature distribution are plotted

at 1.4 m downstream. Figure 12 (a, b

and c) shows that the wall temperature

of tubes present in the upper

region of the tube bundle bears high

temperature as compared to the tubes

present in the lower region and center

of the tube bundle and the reason

is again the presence of effects of

gravity which are very severe in case

of horizontal orientation of coolant

flow. The maximum temperature is

reached for the upper most tubes

and owing to the symmetry both

tubes bear the same temperature

gradients.

Figure 12 (b) explains the temperature

distribution at the same location

for the vertical upwards flow

while the contours of temperature at z

= 1.4 m for the vertically downwards

flow are given in Figure 12 (c). It is

also clear that for the case of vertical

directions of flow, all tubes present

in the tube bundle bear the same

temperature gradient. So the chances

of production of hot spots and thermal

stresses are minimum for the case of

vertical flows.

A comparison of Figure 12 (b and

c) shows that the temperature at the

central tube is larger for vertically

downward flow as compared to the

upward flow of the coolant. The

reason is the turbulence damping due

to buoyancy present in the upward

flows.

4.2.3 Density distribution

at a cross section 1.4 m

away from inlet

As for the horizontal direction of flow,

the upper region of the tube bundle

bears high temperature as compared

Research and Innovation

Computational Heat Transfer Analysis of Tubes and Tube Bundles with Supercritical Water as Coolant

ı Kashif Tehseen, Kamran Rasheed Qureshi, M. Abdul Basit, Rab Nawaz, Waseem Siddique and Rustam Khan


atw Vol. 65 (2020) | Issue 11/12 ı November/December

to the lower region, so a high density

should be present at bottom of tube

bundle as compared to top region

which is quite obvious from Figure 13

(a). It shows that due to upward

movement of low density fluid, a

region of low density is present in

upper region as compared to lower

region of the bundle.

Now if we compare Figure 13 (b

and c), it is obvious that the density

distribution is uniform for the vertical

flows and a region of relatively higher

density is seen at the center of the tube

bundle in case of vertically upward

flow as this region has a lower temperature

as explained above. Moreover,

the temperature and density distribution

is uniform for the vertical flows as

compared to the horizontal flow of the

coolant, so a uniform moderation due

to coolant is present for vertical cores

giving a simpler neutronics for the

core of the nuclear reactor making it

easier to control and operate for

normal or accidental conditions of the

nuclear reactor.

As the lighter fluid accelerates

much easier as compared to the heavier

fluid, therefore the magnitude of axial

velocity is much higher in areas where

the density is low and temperature is

high. The axial velocity will be higher

in the center of the tube bundle for

vertically downward flow rather than

the vertically upward flow. For horizontal

flows, the maximum velocity

will be present in the upper region of

the tube bundles due to presence of

low density and high temperature

liquid in the respective region.

5 Conclusion

For vertical tube, the effects of

length and diameter were studied

and found a reduction in heat transfer

coefficient with decreased heated

length and increased inside diameter

of tube. It was also observed that

severity of HTD can be reduced by

decreasing the heated length of the

tube. The effects of buoyancy were

also studied by comparing computational

results obtained for vertically

upward and downward flows

and found that HTD can be reduced

by using downward flow of the

coolant.

For the horizontal tube, it was

found that the value of HTC is

different for top and bottom walls of

same tube which becomes more

and more severe with increased heat

flux and decreased mass flux of the

coolant. The different HTC at upper

and lower wall causes temperature

of top wall to increase significantly

as compared to the bottom wall

for the case of HTD which may lead

to the production of hot spots at top

wall and conse quently, may give

rise to thermal stresses in the fuel

elements.

The comparison of vertical and

horizontal flow direction in tubes and

tube bundles makes it clear that the

heat transfer phenomenon in the

vertical directions are much simpler

than in the horizontal direction. So

the use of vertical core eliminates the

chances of pro duction of hot spots and

thermal stresses in the cores of the

SCWRs and makes the operation

of the reactor simpler and easier.

The uneven temperature distribution

and all other properties for the

horizontal cores also produce the

thermal stresses in the cores.

Acknowledgement

We would like to express our

gratitude to Dr. Zhi Shang, Faculty

of Engineering, Kingston University,

London for his kind guidance and

valuable suggestions during the

course of this research work.

RESEARCH AND INNOVATION 593

(a) Horizontal Flow

| Fig. 12.

Temperature distribution at z = 1.4 m.

(b) Vertically Downward Flow

(c) Vertically Upward Flow

(a) Horizontal Flow

| Fig. 13.

Density distribution at z = 1.4 m.

(b) Vertically Downward Flow

(c) Vertically Upward Flow

Computational Heat Transfer Analysis of Tubes and Tube Bundles with Supercritical Water as Coolant

Research and Innovation

ı Kashif Tehseen, Kamran Rasheed Qureshi, M. Abdul Basit, Rab Nawaz, Waseem Siddique and Rustam Khan


atw Vol. 65 (2020) | Issue 11/12 ı November/December

References

RESEARCH AND INNOVATION 594

Imprint

Imprint

Official Journal of Kerntechnische Gesellschaft e. V. (KTG)

Publisher

INFORUM Verlags- und Verwaltungsgesellschaft mbH

Robert-Koch-Platz 4, 10115 Berlin, Germany

Phone: +49 30 498555-33

www.nucmag.com

General Manager

Dr. Thomas Behringer

Editor in Chief

Christopher Weßelmann

+49 2324 4397723

editorial@nucmag.com

Managing Editor

Nicolas Wendler

+49 30 498555-21

nicolas.wendler@nucmag.com

Editor

@atw_Journal

@atw-international-journal-for-nuclear-power

Nicole Koch

+49 163 7772797

nicole.koch@nucmag.com

Advertising and Subscription

Petra Dinter-Tumtzak

+49 30 498555-33

petra.dinter@nucmag.com

[1] K. Yamagata, K. Nishikawa, S. Hasegawa, T. Fujii, and S. Yoshida, “Forced convective heat transfer

to supercritical water flowing in tubes,” International Journal of Heat and Mass Transfer, vol. 15,

pp. 2575-2593, 1972.

[2] H. Swenson, J. Carver, and C. d. Kakarala, “Heat transfer to supercritical water in smooth-bore

tubes,” Journal of Heat Transfer, vol. 87, pp. 477-483, 1965.

[3] J. Jackson and J. Fewster, “Forced convection data for supercritical pressure fluids,”

Simon Engineering Laboratory, University of Manchester, 1975.

[4] M. Watts and C. Chou, “Mixed convection heat transfer to supercritical pressure water,”

in Proceedings of the 7 th International Heat Transfer Conference, 1982, pp. 495-500.

[5] L. Gorban, R. Pomet’ko, and O. Khryaschev, “Modeling of water heat transfer with Freon of

supercritical pressure,” ФЭИ-1766, Institute of Physics and Power Engineering ( ФЭИ), Obninsk,

Russia, vol. 19, 1990.

[6] J. Jackson, “Consideration of the heat transfer properties of supercritical pressure water

in connection with the cooling of advanced nuclear reactors,” in The 13 th pacific basin nuclear

conference. Abstracts, 2002.

[7] S. Mokry, I. Pioro, A. Farah, K. King, S. Gupta, W. Peiman, et al., “Development of supercritical

water heat-transfer correlation for vertical bare tubes,” Nuclear Engineering and Design, vol. 241,

pp. 1126-1136, 2011.

[8] B. V. Dyadyakin and A. S. Popov, “Heat Transfer and thermal resistance of tight seven-rod bundle,

cooled with water flow at supercritical pressures,” Transactions of VTI, vol. 11, pp. 244-253, 1977.

[9] A. Farah, “Assessment of FLUENT CFD code as an analysis tool for SCW applications,” University of

Ontario Institute of Technology, 2012.

[10] M. Sharabi, W. Ambrosini, S. He, and J. D. Jackson, “Prediction of turbulent convective heat

transfer to a fluid at supercritical pressure in square and triangular channels,” Nucl. Energy,

vol. 35, no. 6, pp. 993–1005, Jun. 2008.

[11] J. A. Withag, J. L. Sallevelt, D. W. Brilman, E. A. Bramer, and G. Brem, “Heat transfer characteristics

of supercritical water in a tube: Application for 2D and an experimental validation,” The Journal of

Supercritical Fluids, vol. 70, pp. 156-170, 2012.

[12] F. Roelofs, “CFD analyses of heat transfer to supercritical water flowing vertically upward in a

tube,” NRG rapport, vol. 21353, 2004.

[13] Z. Shang, “CFD investigation of vertical rod bundles of supercritical water-cooled nuclear reactor,”

Nuclear Engineering and Design, vol. 239, pp. 2562-2572, 2009.

[14] S. Koshizuka, N. Takano, and Y. Oka, “Numerical analysis of deterioration phenomena in heat

transfer to supercritical water,” International Journal of Heat and Mass Transfer, vol. 38,

pp. 3077-3084, 1995.

[15] S. H. Lee and J. R. Howell, “Turbulent developing convective heat transfer in a tube for fluids near

the critical point,” International Journal of Heat and Mass Transfer, vol. 41, pp. 1205-1218, 1998.

[16] X. Cheng, B. Kuang, and Y. Yang, “Numerical analysis of heat transfer in supercritical water cooled

flow channels,” Nuclear Engineering and Design, vol. 237, pp. 240-252, 2007.

[17] D. Palko and H. Anglart, “Theoretical and numerical study of heat transfer deterioration in

HPLWR,” in International Conference Nuclear Energy for New Europe, 2007, pp. 10-13.

[18] J. Yang, Y. Oka, Y. Ishiwatari, J. Liu, and J. Yoo, “Numerical investigation of heat transfer in

upward flows of supercritical water in circular tubes and tight fuel rod bundles,”

Nuclear Engineering and Design, vol. 237, pp. 420-430, 2007.

[19] Z. Shang and S. Lo, “CFD in supercritical water-cooled nuclear reactor (SCWR) with horizontal tube

bundles,” Nuclear Engineering and Design, vol. 241, pp. 4427-4433, 2011.

[20] F. Wang, B. Cui, S. Zhang, and X. Qin, “Numerical Simulation of Supercritical Water Heat Transfer

in the Vertically Heated Tube,” in 2013 21st International Conference on Nuclear Engineering,

2013, pp. V003T10A035-V003T10A035.

[21] J. Y. Yoo, “The turbulent flows of supercritical fluids with heat transfer,” Annual review of fluid

mechanics, vol. 45, pp. 495-525, 2013.

Editorial Advisory Board

Frank Apel

Erik Baumann

Dr. Erwin Fischer

Carsten George

Eckehard Göring

Dr. Florian Gremme

Dr. Ralf Güldner

Carsten Haferkamp

Dr. Anton Kastenmüller

Prof. Dr. Marco K. Koch

Ulf Kutscher

Herbert Lenz

Jan-Christan Lewitz

Andreas Loeb

Dr. Thomas Mull

Dr. Joachim Ohnemus

Olaf Oldiges

Dr. Tatiana Salnikova

Dr. Andreas Schaffrath

Dr. Jens Schröder

Norbert Schröder

Prof. Dr. Jörg Starflinger

Dr. Brigitte Trolldenier

Dr. Walter Tromm

Dr. Hans-Georg Willschütz

Dr. Hannes Wimmer

Layout

zi.zero Kommunikation

Antje Zimmermann

Berlin, Germany

Printing

inpuncto:asmuth

druck + medien gmbh

Buschstraße 81, 53113 Bonn, Germany

Price List for Advertisement

Valid as of 1 July 2020

Published monthly, 9 issues per year

Germany:

Per issue/copy (incl. VAT, excl. postage) 23.55 €

Annual subscription (incl. VAT and postage) 183.50 €

All EU member states without VAT number:

Per issue/copy (incl. VAT, excl. postage) 23.55 €

Annual subscription (incl. VAT, excl. postage)183.50 €

EU member states with VAT number

and all other countries:

Per issues/copy (no VAT, excl. postage) 22.43 €

Annual subscription (no VAT, excl. postage) 174.77 €

Copyright

The journal and all papers and photos contained in it

are protected by copyright. Any use made thereof

outside the Copyright Act without the consent of the

publisher, INFORUM Verlags- und Verwaltungsgesellschaft

mbH, is prohibited. This applies to reproductions,

translations, micro filming and the input and incorporation

into electronic systems. The individual author is

held responsible for the contents of the respective

paper. Please address letters and manuscripts only

to the Editorial Staff and not to individual persons

of the association's staff. We do not assume any

responsibility for unrequested contributions.

Signed articles do not necessarily represent the views

of the editorial.

ISSN 1431-5254

Authors

Kashif Tehseena

Kamran Rasheed Qureshia

Rab Nawaz

Rustam Khan

Department of Nuclear Engineering

Pakistan Institute of Engineering and Applied Sciences

Islamabad, Pakistan

M. Abdul Basit

Waseem Siddique

Department of Mechanical Engineering

Pakistan Institute of Engineering and Applied Sciences

Islamabad, Pakistan

Corresponding Author Email:

rnawaz@pieas.edu.pk

Research and Innovation

Computational Heat Transfer Analysis of Tubes and Tube Bundles with Supercritical Water as Coolant

ı Kashif Tehseen, Kamran Rasheed Qureshi, M. Abdul Basit, Rab Nawaz, Waseem Siddique and Rustam Khan


atw Vol. 65 (2020) | Issue 11/12 ı November/December

Transient Thermal-hydraulic Analysis

of a Scaled Down Test Loop for the

VVER-1000 Reactor using RELAP5 Code

Babak Khonsha, Gholamreza Jahanfarnia, Kamran Sepanloo and Mohammadreza Nematollahi

1 Introduction

1.1 Scaling review

Scaling is an important technique in

nuclear science and it enhances the

safety of nuclear installations. Therefore,

various researches are carried

out about scaling. Here, some important

researches about the scaling are

mentioned:

A critical review of scaling criteria

was discussed by Kiang (1985) [1].

Scaling of two-phase flow transients

using reduced-pressure system and

simulant fluid was investigated

by Kocamustafaogullari and Ishii

(1987). A new scaling methodology

for simulating pressure transients is

obtained by considering the changes

of the fluid property groups which

appear within the two-phase similarity

parameters and the single-phase

to two-phase flow transition parameters

[2]. System scaling and

modeling analysis for Small Break

LOCA (SBLOCA) was presented by

Hsu et al. (1990) [3]. Scaling of

thermo hydraulic systems was presented

by Wulff (1996). It is shown

how scaling facilitates (1) the quantitative

ranking of thermohydraulic

processes in their order of priority, (2)

the rational selection of a test matrix,

and therewith (3) the efficient allocation

of resources for resolving

technical issues of reactor safety [4].

Power-to-mass scaling methodology

for reduced height and reduced

pressure conditions was developed by

Liu et al. (1997). They studied station

blackout accident conditions to

under stand the reliability of this

method [5]. Scaling analysis for the

APEX test facility was inquired by

Reyes and Hochreiter (1998). They

summarized the key aspects of a

state-of-the-art scaling analysis [6].

Scaling for the ECC bypass phenomena

during the LBLOCA reflood

phase was carried out by YUN et al.

(2004). In this study “modified linear

scaling law” is suggested for the

design of a scaled-down experimental

facility [7]. Code-based design and

stability analysis of TTL-1 test

loop was studied by Jafari et al.

(2002) [8]. A complete scheme of

scaling methods for the design of

integral system test facilities with

reduced height and reduced pressure

was studied by Liu and Lee (2004)

[9]. Scaling methodology for a

reduced-height and reduced-pressure

integral test facility (ITF) to investigate

direct vessel injection line

break SBLOCA was conducted

by Bae et al. (2008). Energy scaling

methodology (ESM) has been

developed in this study. ESM

included the effect of density

difference in the reduced pressure for

calculating the thermal power in the

test facility. Because of significant

difference in the scaling factor for

the system pressure, the enthalpy

difference according to a reduced

pressure condition should not be

ignored [10]. Design and analysis

of a thermal-hydraulic integral test

facility for Bushehr nuclear power

plant (BITF) was explored by

Khoshnevis et al. (2009). The BITF

is a volume scaled model (1:1375).

To ensure that gravitational forces

remain equal to those in the

reference reactor, the major components

and systems in the BITF

preserve 1:1 elevation equivalence

to the reference reactor [11]. The

relevance of scaling in the water

cooled nuclear reactor tech nology

reviewed by D’Auria and Galassi

(2010). The relevance of scaling in

the water cooled nuclear reactor

technology constitutes the motivation

for the present paper. The origin of

the scaling-issue, i.e. the impossibility

to get access to measured data in

case of accident in nuclear reactors,

is discussed [12]. The standard

problems of the ATLAS facility

were evaluated by Kim et al. (2014).

Three kinds of Small Break Loss of

Coolant Accidents (SBLOCAs) were

determined as target scenarios by

considering their technical importance

and incorporating interests

from participants [13].

1.2 NC review

NC is an important mechanism in

several industrial systems and the

knowledge of its behaviour is of

interest to nuclear reactor design,

operation, and safety [14, 15]. The

NC flow driven systems have been

implemented into many nuclear

reactor designs and various passive

safety system designs [16]. Single-

Phase Natural Circulation (SPNC)

regime implies no void occurrence

in the upper plenum of the system.

Therefore, coolant at the core outlet

will be subcooled up to nearly saturated.

Core flow rate is derived from

the balance between driving and

resistant forces. Driving forces are

the result of fluid density differences

occurring between {descending side

of U-tubes and vessel downcomer}

and {core and ascending side of

U-tubes}. Resistant forces are due to

irreversible friction pressure drops

along the entire loop. Resulting fluid

velocities are sufficient for removing

core power in (subcooled) nucleate

boiling or forced convection heat

transfer regimes: no film boiling

condition is experienced in the core.

SPNC may occur at any primary

system pressure, consistently with SG

pressure. However, typical primary

system pressures range between 8 and

16 MPa with secondary pressure close

to the nominal operating condition.

It is expected from the Nuclear

Power Plant (NPP) design that SPNC,

provided the availability of SG

cooling, is capable to remove the

nuclear decay heat from the core [17].

New concepts in nuclear technology

largely exploit the gravity forces for

the heat removal capability [18]. The

NC core power removal capability is

only exploited for accident situations,

basically to demonstrate the inherent

safety features of the plants [19].

Similarity analysis and scaling criteria

for LWRs under single-phase and

two-phase natural circulation was

studied by Ishii and Kataoka (1983)

[20]. Scaling laws for thermalhydraulic

system under single-phase

and two-phase natural circulations

was discussed by Ishii and Kataoka

(1984) [21]. A survey of scaling

laws for the single-phase natural

circulation loops was conducted by

RESEARCH AND INNOVATION 595

Research and Innovation

Transient Thermal-hydraulic Analysis of a Scaled Down Test Loop for the VVER-1000 Reactor using RELAP5 Code ı Babak Khonsha, Gholamreza Jahanfarnia, Kamran Sepanloo and Mohammadreza Nematollahi


atw Vol. 65 (2020) | Issue 11/12 ı November/December

RESEARCH AND INNOVATION 596

| Fig. 1.

Schematic view of SRBITF test loop.

Kvijayan and Austregesilo (1994)

[22]. Addressing the reliability evaluation

for passive thermal- hydraulic

systems was analysed by Burgazzi

et al. (1998) [23]. NC performance

of the designed thermohydraulic test

loop with the aid of the Relap5

thermohydraulic system code was

carried out by Jafari et al. (2002)

[24]. NC phenomenology, including

applications, stability studies, available

databases and evaluation of

predictive capabilities of computational

tools was conducted by

D’Auria et al. (2002b) [25]. Decay

heat removing from the reactor in the

case of station blackout with intact

primary and secondary circuits was

studied by Mousavian et al. (2002)

[26]. NC test in the PSB-VVER test

facility with RELAP5 code was presented

by Cherubini, et al. (2005).

Simulation difficulties have been

found in relation with the time of loop

seal clearance, which are attributed to

errors inherent to the mass inventory

[27]. Thermal hydraulic analysis of a

passive resi dual heat removal system

(PRHRS) for an integral pressurized

water reactor was investigated by

Gou et al. (2009) [28]. Assessment of

RELAP5/MOD3.2 for startup transients

in a NC test facility was studied

by Shanbin et al. (2018) [16].

2 Materials and methods

2.1 SRBITF test loop

description

SRBITF includes all main elements of

the VVER-1000 reactor in one circuit

including core, downcomer, upper

plenum, lower plenum, HE, pump,

ACC and PRZ (Figure 1). Core is one

of the most important element of

the test loop and it is determined as

the SRBITF test section. SRBITF core

contains eighteen heating elements

and one central guide element which

have the same diameter and pin pitch

with the same cases in VVER-1000

core. Except the diameter and pin

pitch of heating elements, all sizes

and dimensions of equipment and

pipelines including height, length,

diameter, area, volume and also

the thermal-hydraulic properties are

reduced using the ESM. ESM is the

most complete scaling method.

2.2 Scaling methodology

SRBITF is designed as a reducedheight,

reduced-pressure and re ducedtemperature

ITF and ESM is used to

scale down the coolant mass inventory

and the thermal power. The initial

values of mass inventory (M 0 ) and

average temperature (T avg ) are defined

as follows [10]:

(1)

(2)

where (T HL ) is the hot leg fluid temperature

and (T CL ) is the initial cold

leg fluid temperature.

The enthalpy difference between

the cold leg and hot leg (∆h 0 ) is:

(3)

where h HL and h CL are the local

enthalpy of the hot leg and cold leg. In

the following equations, there is flexibility

in determining a time scale, τ:

(4)

(5)

(6)

(7)

(8)

(9)

where (Q c ) is the thermal power in

the core and t is the time. For a real

time simulation, the prototype and

the test facility use the same time

scale, i.e., τ R = 1. Therefore, the

scaling criteria for conserving the

non- dimensional mass inventory

(M*) and enthalpy (h*) are as follows:

(10)

(11)

(12)

(13)

(14)

The thermal power in the core (Q c )

should be scaled by the ratio of the

total energy (M 0 ∆h 0 ) to the time scale

(τ R ). For a real time simulation, the

prototype and the test facility use the

same time scale, i.e., τ R = 1. Then,

(Q c ) is defined as follows:

(15)

(16)

where τ R = 1, then the thermal power

scaling factor in the core is defined as

follows:

(17)

The initial cold leg fluid temperature

(TCL) was determined considering

the heat transfer between the primary

system and the secondary system in

the steam generators [10]:

(14)

2.3 SRBITF scaling and

Thermal-hydraulic design

With the help of the scaling factors,

scaling and the thermal-hydraulic

design of SRBITF test loop are

accomplished. The size of the equipment

and pipelines and thermalhydraulic

parameters of SRBITF test

loop are reduced in comparison with

their sizes of the VVER-1000 reactor

and its primary circuit. ESM is used to

scale the coolant mass inventory and

Research and Innovation

Transient Thermal-hydraulic Analysis of a Scaled Down Test Loop for the VVER-1000 Reactor using RELAP5 Code ı Babak Khonsha, Gholamreza Jahanfarnia, Kamran Sepanloo and Mohammadreza Nematollahi


atw Vol. 65 (2020) | Issue 11/12 ı November/December

Parameter

Length 1:5.88

area 1:2138

density 1:0.79

Temperature 1:1.89

Pressure 1:16

Enthalpy 1:2.4

Scaling factor

Volume 1:12571

Mass inventory 1:9931

Core thermal power 1:23834

Flow rate 1:9931

Time scale 1:1

| Tab. 1.

SRBITF scaling factors.

the thermal power for the reducedpressure

and reduced-temperature

conditions. Energy Scaling Factors

(ESFs) are the elements of ESM.

Table 1 shows the results of calculations

of the scaling factors for SRBITF

test loop.

Tables 2 and 3 show the results of

SRBITF scaling for a steady state

condition of 100 % nominal core

power. The results of the Tables 2

and 3 are compared with the experimental

data of the VVER-1000 reactor

and its primary circuit. The SRBITF

results are obtained with the aid of the

scaling factors and will be applied in

the next simulations with RELAP5/

MOD3.2 code as the boundary conditions.

In order to calculate the pressure

drop of SRBITF test loop, the effective

parameters of each component and

fluid are inserted in the “pressure drop

online-calculator” software [31].

The results of SRBITF pressure

drop calculations are listed in Table 4

and are compared with the same

one in the VVER-1000 reactor and its

primary circuit.

Core parameters

| Tab. 2.

The scaling results of the SRBITF core.

SRBITF Scaling

factors

VVER-1000

[29,30]

SRBITF

(scaling)

Core height (m) 1:5.88 3.53 0.600

Volume of core (m 3 ) 1:12571 15.3 12.17e-4

Generated heat in the core (kw) 1:23834 3.0e+6 126

Average generated power

of each heating element/FE (kw)

--- 59 7

Flow rate in the core (m 3 /h) 1:9931 84,800 8.54

Pressure in the core inlet (bar) 1:16 160.81 10

Pressure in the core outlet (bar) 1:16 157 9.44

Fluid density in the inlet and outlet of the core (kg/m 3 ) 1:0.79 743-675 914 - 896

Average fluid density in the core (kg/m 3 ) 1:0.79 709 905

Average temperature in the core (°C) 1:1.89 306 161.9

Enthalpy in the inlet and outlet of the core (kJ/kg) --- 1284-1452 649 -719

Enthalpy difference between inlet and outlet

of the core (kJ/kg)

1:2.4 168 70

Temperature in the inlet and outlet of the core (°C) 1:1.89 291-321 153.96 -169.84

Temperature difference between inlet and outlet

of the core (°C)

1:1.89 30 15.88

Height of heating element/FE (m) 1:5.88 3.53 0.60

External diameter of heating element/FE (mm) 1:1 9.1 9.1

Pin pitch of heating element/FE (mm) 1:1 12.75 12.75

Diameter of Core middle guiding element/FA (mm) 1:1 10×2 10×2

Arrangement of the heating element/FE Pin pitch --- Triangular Triangular

Arrangement of the core/FA --- Hexagonal Hexagonal

Saturated temperature in the outlet of the core (°C) 1:1.89 346 183

Parameters

SRBITF Scaling

factors

VVER-1000

[29,30]

SRBITF

Scaling

Height of test loop until U-tubes of HE/SG (m) 1:5.88 15.80 2.69

Cold leg volume (m 3 ) 1:12571 4×(15.8) 50.27e-4

Hot leg volume (m 3 ) 1:12571 4×(7.4) 23.55e-4

Volume of U-tubes of HE/SG (m 3 ) 1:12571 4×(16.2) 51.54e-4

Height of U-tubes of HE/SG (m) 1: 5.88 2.670 0.454

Area of U-tubes of HE/ SG (m 2 ) 1:2138 6115 2.86

Volume of downcomer (m 3 ) 1:12571 19.2 15.27e-4

RESEARCH AND INNOVATION 597

2.4 Code input model and

boundary conditions

Thermal-hydraulic analysis of SRBITF

is carried out by using the RELAP5/

MOD3.2 code. RELAP5 was developed

at the Idaho National Engineering

Laboratory (INEL) for the US Nuclear

Regulatory Commission (NRC). Control

system and secondary system

components are included in order to

allow modeling of plant controls,

turbines, condensers, and secondary

feedwater systems. The component

models include pumps, valves, pipes,

heat structures, control system components,

etc. Heat structures are

assumed to be represented by onedimensional

heat conduction in

rectangular or cylindrical geometry.

Height of downcomer (m) 1: 5.88 7.08 1.204

Diameter of downcomer (m) 1: 5.88 0.263 0.0447

Volume of upper plenum (m 3 ) 1:12571 59.9 47.65e-4

Volume in lower plenum (m 3 ) 1:12571 13.3 10.58e-4

Volume of pump/RCP (m 3 ) 1:12571 4×(3) 9.55e-4

Flow rate of pump/RCP (m 3 /h) 1:9931 4×(21200) 8.54

Height of PRZ (m) 1: 5.88 13.235 2.25

Diameter of PRZ (m) 1: 5.88 3.330 0.57

Water level in PRZ(m) 1: 5.88 8.875 1.509

Volume of ACC (m 3 ) 1:12571 240 19.0

Water volume in the ACC (m 3 ) 1:12571 200 15.7

Inlet temperature of the HE in secondary loop (°C) 1:1.89 220 116.0

Pressure of the HE in secondary loop (bar) 1:16 61 3.81

| Tab. 3.

The scaling results of the SRBITF primary circuit.

Research and Innovation

Transient Thermal-hydraulic Analysis of a Scaled Down Test Loop for the VVER-1000 Reactor using RELAP5 Code ı Babak Khonsha, Gholamreza Jahanfarnia, Kamran Sepanloo and Mohammadreza Nematollahi


atw Vol. 65 (2020) | Issue 11/12 ı November/December

RESEARCH AND INNOVATION 598

Surface multipliers are used to convert

the unit surface of one-dimensional

calculation to the actual surface of

heat structure. Each mesh interval

Parameters VVER-1000 [29] SRBITF

Core pressure drop (bar) 3.81 0.56

HE/SG pressure drop (bar) 1.35 0.33

Pipelines pressure drop (bar) 1.08 0.47

Sum of the all components

pressure drop (bar)

6.24 1.36

Head of Pump/RCP(bar) 6.24 1.36

| Tab. 4.

The results of the Pressure drop calculation for the SRBITF elements.

contains different mesh spacing,

different material or both [32]. The

boundary conditions and initial data

are defined in Tables 2, 3 and 4.

2.5 Nodalization of SRBITF test

loop

All major components of the primary

side including core and primary

circuit are nodalized and simulated

with RELAP5 code (Figure 2). A

scaled down horizontal HE has

been modeled in the primary and

secondary side. In the primary side,

HE include four equivalent pipe

branches (Pipes #160, #170, #180

and #190). The most important

boundary conditions for modeling

have been listed in Tables 2 and 3.

2.6 Scaling validation

For scaling validation, the steady state

condition of 100 % nominal core power

is considered for VVER-1000. For

comparison between SRBITF scaling

results and VVER-1000 experimental

data, two methods are con sidered:

1. Validation with the aid of the

Reynolds (Re), Prandtl (Pr) and

Peclet (Pe) numbers;

2. Validation with the help of the new

dimensionless numbers.

2.6.1 Scaling validation with Re,

Pr and Pe numbers

Here the scaling validation is considered

for the VVER-1000 reactor

core. For comparison between SRBITF

scaling results and VVER-1000 experimental

data, Re, Pr and Pe numbers

are used. Figures 3, 4 and 5 show

comparison of variation of Re, Pr and

Pe numbers between the SRBITF and

VVER-1000 core.

Figure 3 shows that the variations

of Re number through the dimensionless

core height are very similar for

SRBITF and VVER-1000. Figure 4

shows that the variations of Pr number

through the SRBITF and the

VVER- 1000 core are very close.

Figure 5 shows that the variations

of Pe number through the SRBITF and

VVER-1000 core in the core inlet and

outlet are close.

2.6.2 Scaling validation

with new dimensionless

numbers

For comparison between SRBITF

scaling results and VVER-1000 experimental

data, four dimensionless

parameters (DP) are defined in the

steady state condition of 100 %

nominal core power according to the

following formula:

(19)

| Fig. 2.

SRBITF components nodalization.

where X can be pressure (P), enthalpy

(h), dynamic viscosity (μ) or density

(ρ).

Figures 6 shows a comparison

between dimensionless numbers of

pressure, enthalpy, dynamic viscosity

and density through the SRBITF and

VVER-1000 core.

Considering the close results

between the SRBITF and VVER-1000

in Figure 6, it can be concluded that

the scaling calculations are accurate

and valid.

Research and Innovation

Transient Thermal-hydraulic Analysis of a Scaled Down Test Loop for the VVER-1000 Reactor using RELAP5 Code ı Babak Khonsha, Gholamreza Jahanfarnia, Kamran Sepanloo and Mohammadreza Nematollahi


atw Vol. 65 (2020) | Issue 11/12 ı November/December

| Fig. 3.

Comparison of the Re number between the SRBITF and VVER-1000 core.

2.6.3 Uncertainty evaluation

Uncertainty of parameters can be

defined as follows:

(20)

where (m) is the semi-range (or halfwidth)

between the upper and lower

limits [33].

Table 5 shows the dimensionless

parameters in the inlet and outlet of

the core for the VVER-1000 and

SRBITF, the percentage of errors and

the uncertainty estimation of the

parameters.

For assessment of SRBITF scaling

calculations, the errors and uncertainties

of the important parameters of

the SRBITF core are calculated and

shown in Table 5. The errors and

uncertainties of the important parameters

are small enough and it proves

that the accuracy of the scaling calculations

are good. Therefore, it can be

concluded that the scaling of SRBITF

test loop is valid.

RESEARCH AND INNOVATION 599

| Fig. 4.

Comparison of the Pr number between the SRBITF and VVER-1000 core.

| Fig. 5.

Comparison of the Pe number between the SRBITF and VVER-1000 core.

| Fig. 6.

Comparison of the dimensionless parameter between the SRBITF and VVER-1000 core.

3 Results and discussions

Simulation results of SRBITF are

obtained with the help of the RELAP5/

MOD3.2 code for the steady state condition

of 100 % nominal core power.

Since, the scaling calculation results

and the RELAP5/MOD3.2 code results

are matched with the experimental

data of the VVER-1000 reactor,

the test loop scaling and thermalhydraulic

design are valid. The steady

state condition of 100 % nominal core

power of VVER-1000 is performed

in SRBITF test loop. The calculated

results from the scaling calculation

are inserted in RELAP5/MOD3.2 code

for SRBITF test loop.

The calculation was performed up

to 2000.0 seconds for the steady

state condition of 100 % nominal core

power [34].

3.1 Sensitivity analysis

A sensitivity study is performed to

evaluate SRBITF scaling calculations

by using of RELAP5/MOD3.2 code

for the steady state condition of 100 %

nominal core power.

3.1.1 Sensitivity analysis for the

SRBITF core

In order to analyze the sensitivity of

the scaling results, the temperature

and pressure variation for the steady

state condition of 100 % nominal core

power in the inlet, middle and outlet

of the SRBITF core are simulated

by RELAP5/MOD3.2 code (Figures 7

and 8).

Research and Innovation

Transient Thermal-hydraulic Analysis of a Scaled Down Test Loop for the VVER-1000 Reactor using RELAP5 Code ı Babak Khonsha, Gholamreza Jahanfarnia, Kamran Sepanloo and Mohammadreza Nematollahi


atw Vol. 65 (2020) | Issue 11/12 ı November/December

RESEARCH AND INNOVATION 600

Dimensionless

parameter

VVER-1000

(experiment)

SRBITF

(scaling)

Figures 7 and 8 show that the

steady state condition is happened

by 200 seconds after the simulation

beginning. Simulation of the core

parameters with RELAP5/MOD3.2

code are very close to the scaling

results (Table 2).

Error,

%

Parameter

Uncertainty

Re for the core inlet 440000 410000 6.82 8.66

Re for the core outlet 570000 530000 7.02 11.55

Pr for the core inlet 0.000902 0.001070 18.62 2.45e-5

Pr for the core outlet 0.000985 0.001010 2.53 7.22e-6

Pe for the core inlet 396.9 438.7 10.00 12.07

Pe for the core outlet 561.5 535.3 2.39 7.56

P DP for the core inlet 1.0120 1.0290 1.68 0.00982

P DP for the core outlet 0.9880 0.9710 1.72 0.00980

h DP for the core inlet 0.9383 0.9496 1.20 0.00652

h DP for the core outlet 1.0620 1.0500 1.13 0.00693

μ DP for the core inlet 1.0698 1.0530 1.57 0.00970

μ DP for the core outlet 0.9302 0.9475 1.86 0.00999

ρ DP for the core inlet 1.0479 1.0099 3.63 0.02194

ρ DP for the core outlet 0.9521 0.9900 3.98 0.02188

| Tab. 5.

Errors and uncertainties of the VVER-1000 and SRBITF core dimensionless parameters.

3.1.2 Sensitivity analysis

for HE

The simulation results for SRBITF

in the inlet and outlet of HE for

the steady state condition of 100 %

nominal core power are shown in

Figure 9.

| Fig. 7.

Temperature variation diagrams vs. time of the SRBITF core for the steady state condition

of 100 % nominal core power.

| Fig. 8.

Pressure variation diagrams vs. time of the SRBITF core for the steady state condition of 100% nominal

core power.

Diagrams of Figure 9 shows that

simulation results of HE output

temperature are in accordance with

the scaling results (Table 2). So the

scaling calculations are acceptable.

3.2 SRBITF evaluation

with RELAP5 code for the

transient condition

Thermal-hydraulic analysis is carried

out using the RELAP5/MOD3.2 code

to evaluate NC heat removal capability

of SRBITF core. Hot leg, PRZ,

HE, cold leg, downcomer, lower

plenum, core and upper plenum are

nodalized and simulated with

RELAP5/MOD3.2. The pump trip

begins by 100 seconds after operation

at the steady state condition of 100 %

nominal core power. The thermal

power decreases to 10 % of the

nominal power suddenly after the

pump trip in order to produce enough

heat to simulate the NC condition as

the prototype reactor.

Flow rate variations through the

SRBITF core for the transient condition

are shown in Figure 10. The

figure shows that the flow rate in the

primary loop drops immediately due

to the pump trip. The NC flow is well

established after the pump trip.

Figure 10 shows that by

100 seconds operation at the steady

state condition of 100 % nominal core

power, the flow rate suddenly fall with

pump trip. The NC flow rate after the

pump trip through the SRBITF core is

about 0.2 lit/s.

Figures 11 shows the simulation

diagram of the pressure in the PRZ

during the transient condition.

Figures 11 shows that the simulation

diagrams have a sharp change

after pump trip and after that, NC

fluid flow is begun. The PRZ pressure

begins to decrease rapidly with the

transient and then, it is stabilized at

about 9.4 bar.

Figures 12 shows the variation of

HE and core inlet and outlet temperature

for the transient condition for

SRBITF test loop.

Figure 12 shows that by

100 seconds after the pump trip, NC

flow is begun and by 200 seconds, the

core inlet and outlet fluid temperatures

begin to decrease gradually

with the same trend. Fluid temperature

at the core inlet suddenly fall

after pump trip. This is because of

entering the residual subcooled water

in the HE tubes to the core. Coolant

temperature in the inlet of HE is

decreased gradually after pump trip

and it is stabilized at about 410 k.

Coolant temperature in the outlet of

Research and Innovation

Transient Thermal-hydraulic Analysis of a Scaled Down Test Loop for the VVER-1000 Reactor using RELAP5 Code ı Babak Khonsha, Gholamreza Jahanfarnia, Kamran Sepanloo and Mohammadreza Nematollahi


atw Vol. 65 (2020) | Issue 11/12 ı November/December

HE is decreased after pump trip and it

is stabilized at about 390 k. The

variation of the HE temperature

diagrams are in accordance with the

variation of the core temperature

diagrams. So it proves that simulations

are valid and SRBITF test loop is

capable to remove the residual heat

with NC flow.

4 Conclusion

A typical VVER-1000 reactor and its

primary circuit has been used to scale

down of SRBITF test loop for a steady

state condition of 100 % nominal core

power with the help of ESM. SRBITF

has been designed as a reduced-height,

reduced-temperature and reducedpressure

test loop. The important

parameters of scaling results have

been compared with experimental

data. The good agreement between

SRBITF scaling results and VVER-1000

experimental data proves that scaling

calculations are valid. An uncertainty

analysis has been performed to evaluate

the results of scaling calculations.

Furthermore, a sensitivity analysis of

SRBITF has been carried out with the

help of RELAP5/MOD3.2 code for

the steady state condition of 100 %

nominal core power and it proves the

good accuracy of scaling calculations.

Also, the fallowing results is achieved

from the SRBITF simulations with

RELAP5/MOD3.2 code:

p In the steady state condition of

100 % nominal core power, the

temperature in the middle of the

SRBITF core is about 432 k and the

pressure in the middle of the

SRBITF core is about 9.8 bar.

p In the transient condition, the

flow rate through the core is about

0.2 lit/s, the core inlet temperature

is about 390 k, the core outlet

tem perature is about 410 k, the

PRZ pressure is about 9.4 bar.

The simulation results show that

SRBITF test loop can remove the heat

adequately in the transient condition

and demonstrate the inherent safety

features of the SRBITF test loop.

Therefore, SRBITF can be used to

study both the steady state and

transient conditions.

| Fig. 9.

SRBITF temperature diagrams vs. time in the inlet and outlet of the HE (Pipes#160 - #190).

| Fig. 10.

Flow rate variation diagram vs. time through the SRBITF core for the transient condition.

| Fig. 11.

Pressure variation diagram vs. time in the PRZ for the transient condition.

RESEARCH AND INNOVATION 601

Acknowledgements

I would like to convey my sincere

gratitude to Drs. G. Jahanfarnia,

K. Sepanloo and M. Nematollahi for

their times and supports throughout

my study.

| Fig. 12.

Temperature variation diagram vs. time in the inlet and outlet of the HE and core for the transient condition.

Research and Innovation

Transient Thermal-hydraulic Analysis of a Scaled Down Test Loop for the VVER-1000 Reactor using RELAP5 Code ı Babak Khonsha, Gholamreza Jahanfarnia, Kamran Sepanloo and Mohammadreza Nematollahi


atw Vol. 65 (2020) | Issue 11/12 ı November/December

RESEARCH AND INNOVATION 602

Nomenclature

Abbreviations

ACC

AEOI

APEX

APR

ATLAS

BITF

DP

ECC

ESF

ESM

FA

FE

HE

INEL

ITF

RCS

RCP

NC

NPP

NRC

LWR

Accumulator

Atomic Energy Organization of Iran

Advanced Plant Experiment

Advanced Pressurized Water Nuclear Reactor

Advanced Thermal–Hydraulic Test Loop

for Accident Simulation

Bushehr Integral Test Facility

Dimensionless Parameter

Emergency Core Cooling

Energy Scaling Factor

Energy Scaling Methodology

Fuel Assemblies

Fuel Elements

Heat Exchanger

Idaho National Engineering Laboratory

Integral Test Facility

Reactor Coolant System

Reactor Coolant Pump

Natural Circulation

Nuclear Power Plant

Nuclear Regulatory Commission

Light Water Reactor

LBLOCA Large Break Lose of Coolant Accident

LOCA

PRZ

Lose of Coolant Accident

Pressurizer

SBLOCA Small Break Loss of Coolant Accident

SG

SRBITF

VVER

Letters

a 0R

h

h fg

k

l 0R

M

P

Pr

Re

Q

t

T

Steam Generator

Science and Research Branch Integral Test Facility

Russian Pressurized Water Type Reactor

global area ratio

specific enthalpy (J/kg)

latent heat (J/kg)

coverage factor

global length ratio

primary mass coolant inventory (kg)

pressure (Pa)

Prandtl number

Reynolds number

thermal power (W)

time (s)

temperature (K)

V volume of primary system (m 3 )

ρ density (kg/m 3 )

Subscripts

0 initial parameters

avg

c

CL

comp

HL

am

p

R

s

sat

average

core

Cold Leg

components

Hot Leg

Reference

arithmetic mean

primary loop

Ratio of model to prototype

secondary loop

saturated water

[1] R.L. Kiang, Scaling criteria for nuclear reactor thermal

hydraulics. Nucl. Sci. Eng. 89, 207-216, 1985.

[2] G. Kocamustafaogullari, M. Ishii, Scaling of two-phase flow

transients using reduced pressure system and simulant fluid.

Nucl. Eng. Des. 104, 121-132, 1987.

[3] Y.Y. Hsu, Z.Y. Wang, C. Unal, M.Di. Marzo, K. Almenas,

Scaling-modelling for small break LOCA test facilities.

Nucl. Eng. Des. 122, 175-194, 1990.

[4] W. Wulf, Scaling of thermohydraulic systems. Nucl. Eng.

Des. 163, 359-395, 1996.

[5] T.J. Liu, C.H. Lee, Y.S. Way, IIST and LSTF counterpart test

on PWR station blackout transient. Nucl. Eng.

Des. 167, 357–373, 1997.

[6] J.N. Reyes, L. Hochreiter, Scaling analysis for the OSU AP600

test facility (APEX). Nucl. Eng. Des. 186, 53-109, 1998.

[7] B.J. Yun, H.K. Cho, D.J. Euh, C.H. Song, G.C. Park, Scaling for

the ECC bypass phenomena during the LBLOCA reflood phase.

Nucl. Eng. Des. 231, 315-325, 2004.

[8] J. Jafari, F. D’Auria, H. Kazeminejad, H. Davilu, Code-based

design and stability analysis of a low pressure natural

circulation test loop. Int. J. Heat Tech. 20, 2, 81-88, 2002.

[9] T.J. Liu, C.H. Lee, Adequacy of power-to-mass scaling in

simulating pwr incident transient for reduced-height,

reduced- pressure and full-height, full pressure integral system

test facilities. Nucl. Sci. Eng. 146, 274-290, 2004.

[10] B.U. Bae, K.H. Lee, Y.S. Kim, B.J. Yun, G.C. Park, Scaling

methodology for a reduced-height reduced-pressure integral

test facility to investigate direct vessel injection line break

SBLOCA. Nucl. Eng. Des. 238, 2197–2205, 2008.

[11] T. Khoshnevis, J. Jafari, M. Sohrabpour, Design and analysis of

a thermal hydraulic integral test facility for Bushehr nuclear

power plant. Prog. Nucl. Ener. 51, 443-450, 2009.

[12] F. D’Auria, G.M. Galassi, Scaling in nuclear reactor system

thermal-hydraulics. Nucl. Eng. Des. 240, 3267-3293, 2010.

[13] Y.S. Kim, K.Y. Choi, C.H. Song, W.P. Baek, Overview of the

standard problems of the ATLAS facility. Ann. Nucl. Energ.

63, 509-524, 2014.

[14] Y. Zvirin, A review of N.C. loops in PWR and other systems.

Nucl. Eng. Des. 67, 203-225, 1981.

[15] R. Grief, Natural Circulation Loops. J. Heat Trans.

110, 1243-1258, 1988.

[16] Sh. Shi, Q. Zhu, X. Sun, M. Ishii, Assessment of RELAP5/

MOD3.2 for startup transients in a natural circulation test

facility. Ann. Nucl. Energ. 112, 257-266, 2018.

[17] M. Cherubini, W. Giannotti, D. Araneo, F. D’Auria, Use of the

natural circulation flow map for natural circulation systems

evaluation. Sci. Tech. Nucl. Ins. Art. ID 479673, 2008.

[18] F. D’Auria, G.M. Galassi, M. Frogheri, Natural circulation

performance in nuclear power plants. 2 nd Conf. of the

Croation Nucl. Society, Dubrovnik (HR), 1998.

[19] IAEA, Progress in design, research and development, and

testing of safety systems for advanced water cooled reactors.

IAEA-TECDOC – 872, 2000.

[20] M. Ishii, I. Kataoka, Similarity analysis and scaling criteria for

LWRs under single-phase and two-phase natural circulation.

Argonne Nat. Lab. NUREG/CR-3267, ANL-83-32, 1983.

[21] M. Ishii, I. Kataoka, Scaling lows for thermal-hydraulic system

under single- phase and two-phase natural circulation. Nucl.

Eng. Des. 81, 411-425, 1984.

[22] P. Kvijayan, H. Austregesilo, Scaling laws for single-phase

natural circulation loops. Nucl. Eng. Des. 152, 331-347, 1994.

[23] L. Burgazzi, G.L. Fiorini, F. De Magistris, Reliability assessment

of passive safety systems. In: Proc. Sixth Int. Conf. Nucl. Eng.

ICONE 6-6340, San Diego, USA, May 10-15, 1998.

[24] J. Jafari1, F. D’Auria, H. Kazeminejad, H. Davilu, Analysis of

natural circulation performance of the designed

thermohydraulic test loop. Int. Conf. Nucl. Energ. for New

Europe 2002, Kranjska Gora, Slovenia, September 9-12,

2002.

[25] F. D’Auria, M. Frogheri, M. Misale, (Guest Eds.), Single and

two-phase natural circulation. Nucl. Eng. Des. special issue

devoted to the Eurotherm Seminar No. 3, 215, Nos.1/2,

1-170, 2002b.

[26] S. K. Mousavian, F. D’Auria, M. A. Salehi, Analysis of natural

circulation phenomena in VVER-1000. Nucl. Eng. Des. 229. 1,

25-46, 2004.

[27] M. Cherubini, F. D’Auria, J.C. Ferreri, O. Mazzantini, RELAP5

simulation of a natural circulation test in the PSB-VVER test

facility. The 11th Int. Topical Meeting on Nucl. Reactor

Thermal-Hydraulics (NURETH-11) Paper: 542, Popes’ Palace

Conf. Center, Avignon, France, October 2-6, 2005.

[28] J. Gou, S. Qiu, G. Su, D. Jia, Thermal hydraulic analysis of a

passive residual heat removal system for an integral

pressurized water reactor. Sci. Tech. Nucl. Ins. Art. ID 473795,

12 pages, 2009.

[29] AEOI, Final Safety Analysis Report (FSAR). Chapter 4, Reactor,

2003a.

[30] AEOI, Final Safety Analysis Report (FSAR). Chapter 6,

Engineering Safety Features, 2003a.

[31] Pressure Drop Online-Calculator. www.pressure-drop.com/

Online-Calculator/

[32] The RELAP5 Code Development Team, RELAP5/MOD3 Code

Manual, NUREG/CR-5535 (INEL-95/0174), Vol. I-V. Idaho

National Engineering Laboratory, Idaho Falls, Idaho, 1995.

[33] S. Bell, A beginner’s guide to uncertainty of measurement.

Measurement good practice guide. Centre for basic,

thermal and length metrology. Nat. Phys. Lab., Teddington,

Middlesex, United Kingdom, TW11 0LW, Issue 2,

ISSN 1368-6550, 1999.

[34] P. Bayless, RELAP/MOD3.2 assessment using an 11 % upper

plenum break experiment in the PSB facility. Idaho Nat. Eng.

Environ. Lab., Bechtel BWXT, Idaho. LLC, 2003.

Authors

Babak Khonsha

Gholamreza Jahanfarnia

Department of Nuclear

Engineering

Science and Research Branch

Islamic Azad University

Tehran, Iran

Kamran Sepanloo

Nuclear Science and Technology

Research Institute

Atomic Energy Organization of Iran

Tehran, Iran

Mohammadreza Nematollahi

School of Mechanical Engineering

Shiraz University

Shiraz, Iran

Research and Innovation

Transient Thermal-hydraulic Analysis of a Scaled Down Test Loop for the VVER-1000 Reactor using RELAP5 Code ı Babak Khonsha, Gholamreza Jahanfarnia, Kamran Sepanloo and Mohammadreza Nematollahi


atw Vol. 65 (2020) | Issue 11/12 ı November/December

Inside

Bericht über die KTG Mitgliederversammlung 2020

KTG Inside

603

Am 22. September 2020 fand im Crowne Plaza Hotel in Berlin

die diesjährige Mitgliederversammlung der Kerntechnischen

Gesellschaft statt. Die Versammlung wurde vor dem Hintergrund der

Corona-Pandemie als Hybridveranstaltung abgehalten, d. h. parallel

als Präsenzsitzung mit der Möglichkeit der Live-Zuschaltung über

Internet. Angemeldet waren 60 Teilnehmer, davon 25 vor Ort und

35 online.

Die Tagesordnung umfasste die im Vereinsrecht üblichen Punkte

wie folgt: Bericht des Vorsitzenden, Bericht aus dem Beirat, Bericht

des Schatzmeisters inkl. Genehmigung Rechenschaftsbericht 2019

und Genehmigung Haushaltsplanung 2021, Bericht der Rechnungsprüfer

und Entlastung des Vorstands, Bestellung der Rechnungsprüfer,

Anträge der Untergliederungen sowie Sonstiges.

Nach einem gemeinsamen Gedenken an die im Berichtsjahr

verstorbenen Mitglieder erläuterte der Vereinsvorsitzende, Frank

Apel, anhand von Charts die Aktivitäten der KTG im Berichts zeitraum,

d. h. seit der Mitgliederversammlung 2019. Zu einzelnen Punkten

wurden die jeweiligen Berichterstatter, z. B. zur Sektionsarbeit und

den Fachgruppen, online hinzugeschaltet. Der Gesamtbericht schloss

mit einem Ausblick auf die zukünftige Entwicklung. Der Geschäftsführer

der KTG, Dr. Behringer, berichtete zum politischen Umfeld in

Berlin und zur Arbeit der KTG-Geschäftsstelle. Im Einzelnen wurden

folgende Punkte hervorgehoben: Die Mitgliederzahl reduzierte sich

wie in Anbetracht des Branchenumfeldes und der Altersstruktur

der Mitglieder zu erwarten weiter auf nunmehr 1.658 (Stand

01.01.2019). Gründe hierfür sind neben einer – jedoch vergleichsweise

– geringen Anzahl von Todesfällen relativ hohe Austrittsraten

infolge sonstiger Gründe. Die Geschäftsstelle versucht dieser

Entwicklung entgegenzuwirken, u. a. mit Maßnahmen zur besseren

Informationsdarbietung für die Mitglieder. So wurde eine elektronische

Fachinformation eingeführt, die ca. 14-tägig erscheint bzw.

per E-Mail versendet wird und sich von Beginn an hoher Beliebtheit

erfreut. Die Sektionsarbeit verläuft weiterhin überwiegend in Eigenregie

und kann hinsichtlich der Aktivitäten als konstant betrachtet

werden, was ebenso für die Fachgruppen gilt. Sektionsübergreifende

Themen, über die berichtet wurden, sind die Auswirkungen von

Corona auf die Wirtschaft der Branche – insbesondere Strommarkt

und Veranstaltungen (z. B. die ausgefallene Tagung Kerntechnik

2020), das neue Konzept der Bundesregierung zur Kompetenz- und

Nachwuchsentwicklung für die nukleare Sicherheit sowie das

Helmholtz-Programm NUSAFE. Im weiteren Verlauf des Berichts

wurden zudem die Aktivitäten der Jungen Generation der KTG

von Herrn Dr. Gremme dargestellt und jene des KTG-Beirats von

Herrn Dr. Willschütz. Der Gesamtbericht schloss mit den Ausführungen

des Schatzmeisters, Herrn Dr. Fischer, über die finanzielle

Situation des Vereins, die als erfreulich stabil bzw. konsolidiert

bezeichnet wurde.

Im Anschluss an den Bericht erfolgte dann die Aussprache der

Mitgliederversammlung zu inhaltlichen und vereinsstrukturellen

Fragestellungen. Der Vorstand und die Geschäftsführung beantworteten

dabei alle Fragen der Mitgliederversammlung zur Vereinstätigkeit

und künftigen Ausrichtung.

Save the Date:

Nächste KTG Mitgliederversammlung

am 04.05.2021 in Berlin

Dr. Thomas Behringer

Verantwortlich

für den Inhalt:

Die Autoren.

Lektorat:

Kerntechnische

Gesellschaft e. V.

(KTG)

Robert-Koch-Platz 4

10115 Berlin

T: +49 30 498555-50

F: +49 30 498555-51

E-Mail:

info@ktg.org

www.ktg.org

KTG INSIDE

Nachruf

Walter Reim

27. November 1932

9. Februar 2020

….per Aspera ad Astra….

Mein Vater war mit Leib und Seele Kerntechniker und hatte viele

Freunde und Wegbegleiter, die auch atw Leser sind.

Seine wichtigsten beruflichen Meilensteine:

Ab 1959 Kerntechnische Ausbildung

P am Forschungsreaktor Garching

P bei General Electric, San Jose, California, USA

P am Vallecitos Boiling Water Reactor in Livermore,

California, USA

Ab 1960 Versuchsatomkraftwerk Kahl (VAK)

P 13. November 1960:

Schichtleiter bei der ersten Kritikalität des VAK Reaktors;

und damit die erste kontrollierte Kritikalität in Deutschland

Ab 1964 Kernkraftwerk Gundremmingen,

KRB – Block A

P Schichtleiter (14. Aug. 1966 0:12: Erste Kritikalität Block A)

P Leiter Produktion

Ab 1978 Kernkraftwerke Gundremmingen,

KGB – Block B und C

P Hauptabteilungsleiter Produktion

P Technischer Betriebsleiter ab 1. Feb. 1989

P Prokurist bis zum 27. Nov. 1997

Ein wesentlicher Meilenstein – auch für die Deutsche Kerntechnik

– jährte sich am 13. November. Vor genau 60 Jahren

war mein Vater Schichtleiter beim Versuchsatom kraftwerk

Kahl (VAK) und leitete die erste Kritikalität des VAK Reaktors.

Das war zu diesem Zeitpunkt auch die erste sich selbsterhaltende,

kontrollierte Kettenreaktion eines Kernkraftwerkes

in Deutschland.

Ein Leben im ständigen Streben nach Wissen, vom kleinsten

Atomkern bis zu den Galaxien im unendlichen Weltall.

Jetzt ist er den Sternen im Himmel ganz nah.

Uns bleiben wunderschöne Erinnerungen in Liebe.

Im Namen der Familie

Stefan Reim

KTG Inside


atw Vol. 65 (2020) | Issue 11/12 ı November/December

604

Herzlichen Glückwunsch!

Die KTG gratuliert ihren Mitgliedern sehr herzlich zum Geburtstag

und wünscht ihnen weiterhin alles Gute!

KTG INSIDE

Wenn Sie künftig eine

Erwähnung Ihres

Geburtstages in der

atw wünschen, teilen

Sie dies bitte der KTG-

Geschäftsstelle mit.

Nachträgliche

Geburtstagsnennung:

November 2020

65 Jahre | 1955

8. Hermann-Josef Marx, Kirrweiler, (Pfalz)

Dezember 2020

55 Jahre | 1965

5. Dipl.-Ing. Hartmut Grunau, Bassum

10. Michael Bächler, Gelsenkirchen

60 Jahre | 1960

6. Dr. Thomas Mull, Erlangen

65 Jahre | 1955

27. Dipl.-Ing. Udo Rieger, Hamburg

70 Jahre | 1950

7. Detlef Gründler, Euskirchen

77 Jahre | 1943

8. Dr. Dieter Herrmann, Brandis

7. Dipl.-Ing. Norbert Bauer, Limburgerhof

78 Jahre | 1942

8. Karl Georg Weber, Neckarwestheim

14. Günter Breiling, Weinheim

79 Jahre | 1942

13. Dipl.-Ing. Klaus-Dieter Hnilica,

Rodenbach/Hanau

80 Jahre | 1940

8. Dipl.-Ing. Wolfgang Heess, Laudenbach

16. Dipl.-Ing. Wolfgang Breyer, Buckenhof

19. Prof. Dr. Wernt Brewitz, Braunschweig

21. Prof. Dr. Jürgen Wehmeier, Springe

86 Jahre | 1934

28. Dipl.-Phys. Bernhard Wigger, Ettlingen

87 Jahre | 1933

10. Prof. Dr. Jürgen Vollradt,

Unna-Königsborn

95 Jahre | 1925

10. Dr. Arthur Pilgenröther,

Kleinostheim

Januar 2021

30 Jahre | 1991

9. Elvira Rapp, Dortmund

50 Jahre | 1971

31. Alexandra Sykora, Nürnberg

55 Jahre | 1966

18. Dipl.-Ing. (TH) Michael Bongartz,

Emmerthal

60 Jahre | 1961

2. Dr. Dietmar Nieder, Karlsruhe

65 Jahre | 1956

18. Dr. Hans-Jürgen Goebelbecker,

Eggenstein-Leopoldshafen

70 Jahre | 1951

19. Dr. Angelika Seibold, Fürth

75 Jahre | 1946

7. Dr. Johann Zech, München

79 Jahre | 1942

31. Dipl.-Phys. Werner Scholtyssek,

Stutensee

84 Jahre | 1937

9. Dipl.-Ing. Werner Rossbach,

Bergisch Gladbach

85 Jahre | 1936

5. Obering. Peter Vetterlein, Oberursel

23. Prof. Dr. Hartmut Schmoock,

Norderstedt

30. Dipl.-Phys. Wolfgang Borkowetz,

Rüsselsheim

30. Dipl.-Ing. Friedrich Morgenstern, Essen

86 Jahre | 1935

10. Dipl.-Ing. Walter Diefenbacher,

Karlsruhe

17. Dipl.-Ing. Helge Dyroff, Alzenau

24. Theodor Himmel, Bad Honnef

88 Jahre | 1933

9. Prof. Dr. Hellmut Wagner, Karlsruhe

89 Jahre | 1932

3. Dipl.-Ing. Fritz Kohlhaas, Kahl/Main

92 Jahre | 1929

20. Dr. Devana Lavrencic-Cannata,

Rom / IT

94 Jahre | 1927

1. Prof. Dr. Werner Oldekop,

Braunschweig

81 Jahre | 1939

6. Dipl.-Ing. Hans-Henn. Kuchenbuch,

Laboe-Brodersdorf

27. Dr. Horst Bauer, Sigless / AT

82 Jahre | 1938

1. Dr. Gert Spannagel,

Linkenheim-Hochstetten

83 Jahre | 1937

30. Dipl.-Ing. Wilhelm Weiss, Weinheim

84 Jahre | 1936

7. Dipl.-Ing. Aurel Badics, Bad Kreuznach

17. Prof. Dr.-Ing. Rolf Theenhaus, Linnich

80 Jahre | 1941

12. Dr. Hans-Gerb. Bogensberger,

Sun City / Arizona / USA

15. Dipl.-Ing. Ulf Rösser,

Heiligkreuzsteinach

82 Jahre | 1939.

11. Dipl.-Ing. Gerwin H. Rasche, Hasloch

16. Dr. Wolfgang Kersting, Blieskastel

83 Jahre | 1938

7. Dipl.-Ing. Manfred Schirra, Stutensee

12. Dipl.-Ing. Hans Dieter Adami,

Rösrath

22. Dr. Franz Müller, Erlangen

KTG Inside


atw Vol. 65 (2020) | Issue 11/12 ı November/December

nucmag.com

605

Sneak Preview

Starting 2021 atw will be published 6 times

a year, with more articles of general interest

on topics from the world of nuclear energy

and nuclear technology.

At the same time and to better serve our

German readership, a German-language

supplement will be enclosed with every

domestic issue, which will focus specifically

on German topics and the German market.

Editorial Programme

Issue

Publication

Date

1 1 Jan 2021 International Trends & New Build Worldwide

2 1 March 2021 Phase-out vs. Lifetime Extension

3 1 May 2021 Advanced reactors: Light & Heavy Water,

Gas- & Liquid-Metal-Cooled Reactors

4 1 July 2021 Waste Management and

Decommissioning & Dismantling

5 1 Sep 2021 Uranium and Nuclear Fuel Cycle

6 1 Nov 2021 Nuclear Safety and Security

NEWS

Follow Us

@atw-international-journal-for-nuclear-power

@atw_Journal

Top

World Energy Outlook 2020

shows how the response to

the Covid crisis can reshape

the future of energy

(iea) It has been a tumultuous year

for the global energy system. The

Covid- 19 crisis has caused more

disruption than any other event in

recent history, leaving scars that will

last for years to come. But whether

this upheaval ultimately helps or

hinders efforts to accelerate clean

energy transitions and reach international

energy and climate goals will

depend on how governments respond

to today’s challenges.

The World Energy Outlook 2020,

the International Energy Agency’s

flagship publication, focuses on the

pivotal period of the next 10 years,

exploring different pathways out of

the crisis. The new report provides the

latest IEA analysis of the pandemic’s

impact: global energy demand is set to

drop by 5 % in 2020, energy-related

CO 2 emissions by 7 %, and energy

investment by 18 %. The WEO’s

established approach – comparing

different scenarios that show how the

energy sector could develop – is more

valuable than ever in these uncertain

times. The four pathways presented

in this WEO are described in more

detail at the end of this press release.

In the Stated Policies Scenario,

which reflects today’s announced policy

intentions and targets, global energy

demand rebounds to its pre-crisis

level in early 2023. How ever, this does

not happen until 2025 in the event of a

prolonged pandemic and deeper

slump, as shown in the Delayed Recovery

Scenario. Slower demand

growth lowers the outlook for oil and

gas prices compared with pre-crisis

trends. But large falls in investment

increase the risk of future market volatility.

If governments and investors step

up their clean energy efforts in line

with the Sustainable Development

Scenario, the growth of both solar

and wind would be even more

spectacular – and hugely encouraging

for overcoming the world’s climate

challenge.

In the Sustainable Development

Scenario, which shows how to put the

world on track to achieving sustainable

energy objectives in full, the complete

implementation of the IEA Sustainable

Recovery Plan moves the global energy

economy onto a different post-crisis

path. As well as rapid growth of solar,

wind and energy efficiency technologies,

the next 10 years would see a

major scaling up of hydrogen and

carbon capture, utilisation and

storage, and new momentum behind

nuclear power.

In general, the IEA summarises

the role of nuclear as follows:

Nuclear power has historically been

one of the largest contributors

of carbon-free electricity globally

and it has significant potential to

contribute to power sector decarbonisation.

Countries envisaging a future role

for nuclear account for the bulk of

global energy demand and CO 2

emissions.

Nonetheless in many jurisdictions

nuclear power has trouble competing

against other, more economic alternatives,

such as natural gas or modern

World Energy

Outlook

2019

News


atw Vol. 65 (2020) | Issue 11/12 ı November/December

606

NEWS

renewables. Concerns over safety and

broader public acceptance remain

obstacles to development.

With nuclear power facing an

uncertain future in many countries,

the world risks a steep decline in its

use in advanced economies that could

result in billions of tonnes of additional

carbon emissions.

Nuclear power plants contribute

to electricity security in multiple

ways. Nuclear plants help to keep

power grids stable and can be a

good complement in decarbonisation

strategies since, to a certain extent,

they can adjust their operations to

follow demand and supply shifts. As

the share of variable renewables like

wind and solar photovoltaics (PV)

rises, the need for such services will

increase.

| www.iea.org (202621420)

Europe

EU education & training policy

must support nuclear skills

(foratom) In order to ensure that the

European nuclear industry can

continue to provide both vital medical

diagnosis and treatment, as well as

low-carbon energy, it needs people

with the right skills. According to

a new position paper issued by

FORATOM, EU education and training

policy must do more to ensure that

there is a sufficient number of people

with the right skills in the nuclear

field.

Society is facing significant

challenges in terms of climate change,

access to affordable energy, health

and employment. The European

nuclear sector stands ready to meet

these challenges. However, it is facing

a skills shortage, particularly given

that a significant part of the workforce

is reaching retirement-age and will

have to be replaced in the near

term. In addition, the implementation

of, for example, digitalisation will

require the reskilling and upskilling

of workers. The industry, as well as

policy makers at both EU and Member

State level must work together to

ensure that Europe can maintain its

highly skilled nuclear workforce, thus

ensuring long-term benefits for our

society.

“The European nuclear sector

brings significant benefits to our

day-to-day lives”, says FORATOM

Director General Yves Desbazeille.

“From helping the EU to fight climate

change, to providing access to lifesaving

treatment and supporting a

wide range of European based jobs at

Operating Results August 2020

Plant name Country Nominal

capacity

Type

gross

[MW]

net

[MW]

Operating

time

generator

[h]

Energy generated, gross

[MWh]

Month Year Since

commissioning

Time availability

[%]

Energy availability

[%] *) Energy utilisation

[%] *)

Month Year Month Year Month Year

OL1 Olkiluoto 1) BWR FI 910 880 744 670 470 4 870 299 274 335 769 100.00 93.28 99.93 90.64 97.95 90.42

OL2 Olkiluoto 1) BWR FI 910 880 744 667 454 5 073 464 264 437 550 100.00 95.42 99.94 94.85 97.51 94.19

KCB Borssele PWR NL 512 484 744 366 943 2 658 771 170 640 205 100.00 90.11 100.00 89.26 96.17 88.77

KKB 1 Beznau 7) PWR CH 380 365 744 271 051 1 892 910 132 201 730 100.00 85.64 100.00 85.43 95.69 84.96

KKB 2 Beznau 1,7) PWR CH 380 365 423 146 462 2 079 391 139 376 174 56.86 94.52 55.39 94.25 51.17 93.40

KKG Gösgen 7) PWR CH 1060 1010 744 773 594 5 702 280 327 818 515 100.00 92.87 99.95 92.34 98.09 91.88

CNT-I Trillo PWR ES 1066 1003 744 783 529 5 178 896 260 926 922 100.00 86.29 100.00 85.60 97.94 82.33

Dukovany B1 PWR CZ 500 473 744 358 695 2 889 649 118 773 833 100.00 100.00 100.00 99.95 96.42 98.71

Dukovany B2 PWR CZ 500 473 744 354 630 2 859 144 113 902 462 100.00 99.90 100.00 99.67 95.33 97.67

Dukovany B3 PWR CZ 500 473 736 331 301 1 679 060 111 930 796 98.93 60.20 92.69 58.82 89.06 57.35

Dukovany B4 PWR CZ 500 473 744 362 313 2 404 064 113 111 021 100.00 82.85 100.00 82.80 97.40 82.12

Temelin B1 PWR CZ 1080 1030 744 797 762 4 528 724 126 443 537 100.00 71.82 99.91 70.93 99.10 71.49

Temelin B2 1) PWR CZ 1080 1030 537 566 281 4 892 583 122 375 201 72.18 76.33 70.14 76.02 70.35 77.23

Doel 1 PWR BE 454 433 744 327 687 898 926 138 634 986 100.00 34.45 99.96 33.72 94.20 32.86

Doel 2 PWR BE 454 433 669 281 170 949 340 137 284 810 89.92 37.17 88.80 36.56 80.44 35.16

Doel 3 2) PWR BE 1056 1006 0 0 5 008 404 268 120 054 0 81.48 0 81.44 0 80.57

Doel 4 PWR BE 1084 1033 744 787 249 5 620 482 275 258 757 100.00 88.86 99.61 88.05 95.81 87.04

Tihange 1 2) PWR BE 1009 962 0 0 0 307 547 424 0 0 0 0 0 0

Tihange 2 PWR BE 1055 1008 744 756 061 5 852 472 263 906 990 100.00 96.52 99.98 96.19 97.03 95.60

Tihange 3 2) PWR BE 1089 1038 0 0 4 096 335 284 658 911 0 64.81 0 64.74 0 64.63

Plant name

Type

Nominal

capacity

gross

[MW]

net

[MW]

Operating

time

generator

[h]

Energy generated, gross

[MWh]

Time availability

[%]

Energy availability Energy utilisation

[%] *) [%] *)

Month Year Since Month Year Month Year Month Year

commissioning

KBR Brokdorf DWR 1480 1410 744 988 411 7 654 253 368 375 275 100.00 100.00 94.30 94.14 90.20 88.06

KKE Emsland DWR 1406 1335 744 1 025 896 7 310 311 364 910 512 100.00 90.75 100.00 90.64 98.01 88.79

KWG Grohnde DWR 1430 1360 744 1 010 366 6 482 880 394 757 726 100.00 92.03 99.93 91.78 94.33 76.93

KRB C Gundremmingen 1,2) SWR 1344 1288 743 950 355 6 243 803 347 567 356 99.80 81.51 96.40 79.63 94.23 78.73

KKI-2 Isar DWR 1485 1410 647 918 631 7 384 853 373 147 322 87.00 89.74 85.93 89.49 82.74 84.52

GKN-II Neckarwestheim DWR 1400 1310 744 1 008 400 7 063 200 347 301 444 100.00 89.01 100.00 89.00 96.98 86.27

News


atw Vol. 65 (2020) | Issue 11/12 ı November/December

different levels, nuclear helps us meet

the challenges our society faces”.

FORATOM therefore recommends

the following in order to ensure that

the EU has a sufficient number of

people – with the right skills – in order

to continue its nuclear activities:

p Promoting and making STEM

subjects attractive amongst pupils

to ensure European technological

leadership.

p Developing and implementing

policies which encourage young

people to study and work in the

nuclear field, by speaking more

positively about the benefits which

nuclear provides to society.

p Basing policies on robust scientific

facts. This means ensuring that all

technologies are treated on an

equal footing and that accurate

information is provided on

Operating Results September 2020

employment opportunities within

the EU in relation to the different

low-carbon sectors.

p EU funds allocated to nuclear

education and training should be

increased. This will help ensure that

the EU maintains its nuclear innovation

leadership, by supporting a

skilled workforce capable of both

undertaking vital R&D as well as

implementing innovative projects.

p Apply a long-term approach to EU

funded projects in the field of

Education & Training. Whilst projects

which run for only a limited

time do have some short-term

benefits, much more could be

achieved if they were to continue

for a longer period. This may also

encourage coordination between

overlapping projects and reduce

the current duplication.

p Policymakers, educational systems

and industry should work together

to ensure generation transition

and competence transfer, as well

as to help the workforce adapt to

new technologies (digitalisation,

industry 4.0).

The full position paper is available

on the webpages of the Foratom

association.

| www.foratom.org (202621431)

Company News

Framatome and General

Atomics announce

collaboration to develop

fast modular reactor

(fraatome) Framatome and General

Atomics Electromagnetic Systems

(GA-EMS) announced plans to collaborate

on the development of GA-EMS’

*)

Net-based values

(Czech and Swiss

nuclear power

plants gross-based)

1)

Refueling

2)

Inspection

3)

Repair

4)

Stretch-outoperation

5)

Stretch-inoperation

6)

Hereof traction supply

7)

Incl. steam supply

BWR: Boiling

Water Reactor

PWR: Pressurised

Water Reactor

Source: VGB

607

NEWS

Plant name Country Nominal

capacity

Type

gross

[MW]

net

[MW]

Operating

time

generator

[h]

Energy generated, gross

[MWh]

Month Year Since

commissioning

Time availability

[%]

Energy availability

[%] *) Energy utilisation

[%] *)

Month Year Month Year Month Year

OL1 Olkiluoto BWR FI 910 880 720 658 502 5 528 801 274 994 271 100.00 94.01 99.98 91.66 99.41 91.40

OL2 Olkiluoto BWR FI 910 880 720 656 351 5 729 816 265 093 901 100.00 95.92 99.93 95.41 99.09 94.72

KCB Borssele PWR NL 512 484 720 357 183 3 015 954 170 997 388 99.52 91.14 99.52 90.39 96.77 89.65

KKB 1 Beznau 7) PWR CH 380 365 720 271 347 2 164 257 132 473 077 100.00 87.21 100.00 87.02 99.14 86.51

KKB 2 Beznau 7) PWR CH 380 365 720 269 363 2 348 754 139 645 537 100.00 95.12 100.00 94.88 98.44 93.95

KKG Gösgen 7) PWR CH 1060 1010 720 753 707 6 455 987 328 572 222 100.00 93.65 99.98 93.18 98.76 92.63

CNT-I Trillo PWR ES 1066 1003 720 760 135 5 939 031 261 687 057 100.00 87.79 99.94 87.17 98.27 84.07

Dukovany B1 PWR CZ 500 473 720 351 194 3 240 843 119 125 027 100.00 100.00 99.92 99.95 97.55 98.58

Dukovany B2 PWR CZ 500 473 595 284 590 3 143 734 114 187 052 82.64 98.01 81.90 97.72 79.05 95.63

Dukovany B3 PWR CZ 500 473 720 348 385 2 027 445 112 279 181 100.00 64.56 100.00 63.33 96.77 61.67

Dukovany B4 PWR CZ 500 473 720 354 880 2 758 944 113 465 901 100.00 84.73 100.00 84.68 98.58 83.92

Temelin B1 PWR CZ 1080 1030 685 733 681 5 262 405 127 177 218 95.14 74.37 94.56 73.52 94.18 73.97

Temelin B2 PWR CZ 1080 1030 720 785 445 5 678 028 123 160 646 100.00 78.92 99.99 78.65 100.82 79.81

Doel 1 PWR BE 454 433 720 333 749 1 232 674 138 968 734 100.00 41.63 99.94 40.97 99.35 40.14

Doel 2 PWR BE 454 433 619 285 032 1 234 372 137 569 843 85.94 42.51 85.31 41.98 84.56 40.65

Doel 3 PWR BE 1056 1006 716 726 407 5 734 811 268 846 461 99.41 83.44 95.02 82.92 94.86 82.13

Doel 4 PWR BE 1084 1033 597 637 501 6 257 983 275 896 258 82.86 88.20 82.05 87.39 80.22 86.30

Tihange 1 2) PWR BE 1009 962 0 0 0 307 547 424 0 0 0 0 0 0

Tihange 2 PWR BE 1055 1008 720 740 579 6 593 050 264 647 569 100.00 96.90 99.96 96.61 98.31 95.90

Tihange 3 2) PWR BE 1089 1038 0 0 4 096 335 284 658 911 0 57.71 0 57.65 0 57.55

Plant name

Type

Nominal

capacity

gross

[MW]

net

[MW]

Operating

time

generator

[h]

Energy generated, gross

[MWh]

Time availability

[%]

Energy availability Energy utilisation

[%] *) [%] *)

Month Year Since Month Year Month Year Month Year

commissioning

KBR Brokdorf 2) DWR 1480 1410 438 595 577 8 249 829 368 970 852 60.82 95.71 56.87 90.06 55.66 84.51

KKE Emsland DWR 1406 1335 720 1 004 402 8 314 713 365 914 914 100.00 91.76 100.00 91.67 99.26 89.93

KWG Grohnde DWR 1430 1360 720 993 488 7 476 368 395 751 214 100.00 92.91 99.56 92.63 95.90 79.01

KRB C Gundremmingen SWR 1344 1288 720 960 521 7 204 324 348 527 877 100.00 83.54 100.00 81.86 98.62 80.91

KKI-2 Isar DWR 1485 1410 720 1 045 443 8 430 296 374 192 765 100.00 90.86 99.99 90.64 97.43 85.93

GKN-II Neckarwestheim DWR 1400 1310 720 989 800 8 053 000 348 291 244 100.00 90.22 99.99 90.20 98.44 87.61

News


atw Vol. 65 (2020) | Issue 11/12 ı November/December

608

NEWS

helium-cooled 50-MWe fast modular

reactor (FMR). Due to its advanced

modular design, the reactor can be

built in a factory and assembled

on-site, which helps to reduce capital

costs and enables incremental capacity

additions. Framatome’s U.S.

engineering team will be responsible

for designing several critical structures,

systems and components for the

FMR.

“This collaboration builds on

our long relationship with General

Atomics with a shared interest in

advancing nuclear energy technologies

to create a cleaner world for

generations to come,” said Bernard

Fontana, CEO of Framatome. “With

our experience and expertise in

designing reactor systems and components

for advanced and small modular

reactors, our team is helping to make

that vision a reality.”

“Designing and deploying a safe,

cost-effective, modular reactor is

critical in helping the world move

closer towards a clean energy future,”

stated Scott Forney, president of GA-

EMS. “We look forward to leveraging

our two companies’ decades of

experience in advancing nuclear

technology and demonstrating the

next generation of commercially

viable nuclear reactors.”

The FMR is designed for enhanced

safety and ease of operation with fastresponse

load following and overall

high efficiency. It offers stability for

the electricity grid and can respond

to meet demand based on fluctuations

in renewable energy sources. The

gas-cooled FMR uses a helium coolant

while eliminating the need for the

graphite common in other heliumcooled

designs. Its fuel is optimized

to support reactor operations for up

to nine years before needing to be

replaced. The power conversion

system does not use complex steam

generators and pressurizers, which

helps to drive down costs.

“We are pleased to work with

GA-EMS to advance this innovative

and promising reactor,” said Gary

Mignogna, president and CEO of

Framatome in North America. “The

synergies between our teams make

this an ideal project for demonstration

and subsequent commercialization.”

A demonstration of the FMR,

which will verify the design, manufacturing,

construction and operation

of the technology, is targeted for

completion in the early 2030s. Commercial

deployment is anticipated in

the mid-2030s.

| www.framatome.com (202621437)

Bruce Power and Westinghouse

collaborate to advance

application of Evinci battery

technology to support

Canada's net-zero initiative

(west) Bruce Power and Westinghouse

Electric Company announced an

agreement to pursue applications of

Westinghouse’s leading eVinci micro

reactor program within Canada, to

provide a reliable source of carbonfree

energy. The agreement supports

efforts by the federal and provincial

governments to study applications for

nuclear technology to reach their goal

of a Net Zero Canada by 2050.

The eVinci micro reactor is a

next-generation, small battery for

decentralized generation markets and

micro grids such as remote communities,

remote industrial mines and

critical infrastructure. It is designed to

provide competitive and resilient

power and superior reliability with

minimal maintenance and its small

size allows for standard transportation

methods and rapid, on-site

deployment.

“Small modular and micro reactors

represent an incredible opportunity to

bring GHG-emission free, affordable

energy to the farthest regions of our

province, supporting resource and

economic development across our

country,” said Greg Rickford, Minister

of Energy, Northern Development

and Mines. “Today’s announcement

further positions Bruce Power and

Ontario as a global leader in nuclear

innovation. I’m proud to see the

technology that will power tomorrow

being advanced right here in Ontario.”

“Our eVinci technology can provide

clean, reliable energy to remote areas

and industrial applications across

Canada,” said Patrick Fragman, President

and Chief Executive Officer,

Westinghouse Electric Company. “We

are proud to partner with Bruce Power

to power the future with carbon-free,

affordable electricity that will provide

Canadian communities with increased

energy flexibility and security.”

| Evinci reactor details.

Over the next year, the work

between the two companies will focus

on furthering the public policy and

regulatory framework; assessing the

economic, social and environmental

contribution of the eVinci technology

compared to alternates such as diesel

or other fossil fuels; identifying

potential industrial applications; and

accelerating the roadmap for Canada

to host a globally recognized demonstra

tion as part of the federal

small modular reactor (SMR) action

plan.

“Bruce Power and Westinghouse

Canada have a strong existing

relationship and as Canada seeks new

innovative options to build on its

existing clean, CO2-free nuclear

advantage, this is an exciting opportunity

to advance further towards

a Net Zero Canada by 2050,” said

Mike Rencheck, President and CEO

of Bruce Power. “Bruce Power

will leverage our relationships and

capacity within the Nuclear Innovation

Institute (NII) and Laurentian

University-based Mining Innovation,

Rehabilitation and Applied Research

Corporation (MIRACO) towards this

exciting opportunity for Canada.”

Bruce Power is committed to

advancing future opportunities for

nuclear energy to provide a clean,

reliable source of electricity that

provides life-saving medical isotopes

around the world, in addition to being

a source of jobs and innovation in

communities across Canada.

This agreement is the latest

partnership between Bruce Power,

Westinghouse and key Canadian

stakeholders to work towards

Canada’s Net Zero by 2050 goal.

This follows a Westinghouse presentation

on the eVinci program at a

conference hosted by the Organization

of Canadian Nuclear Industries

(OCNI) last month and attended by

200 people from leading Canadian

suppliers.

| www.westinghousenuclear.com

www.brucepower.com

News


atw Vol. 65 (2020) | Issue 11/12 ı November/December

Uranium

Prize range: Spot market [USD*/lb(US) U 3 O 8 ]

140.00

120.00

) 1

Uranium prize range: Spot market [USD*/lb(US) U 3 O 8 ]

140.00

120.00

) 1

609

100.00

100.00

80.00

60.00

40.00

20.00

0.00

1980

Jan. 2009

Yearly average prices in real USD, base: US prices (1982 to1984) *

Jan. 2010

1985

Jan. 2011

* Actual nominal USD prices, not real prices referring to a base year. Year

1990

Jan. 2012

Jan. 2013

1995

Jan. 2014

2000

Jan. 2015

Jan. 2016

2005

Jan. 2017

2010