A Functionally Graded Composite for Service in High-Temperature Lead- and Lead-Bismuth–Cooled Nuclear Reactors—I_ Design

Nuclear Technology 

ISSN: 0029-5450 (Print) 1943-7471 (Online) Journal homepage: http://www.tandfonline.com/loi/unct20 

A Functionally Graded Composite for Service 

in High-Temperature Lead- and Lead- 

Bismuth–Cooled Nuclear Reactors—I: Design 

Michael Philip Short & Ronald George Ballinger 

To cite this article: Michael Philip Short & Ronald George Ballinger (2012) A Functionally Graded 

Composite for Service in High-Temperature Lead- and Lead-Bismuth–Cooled Nuclear Reactors—I: 

Design, Nuclear Technology, 177:3, 366-381, DOI: 10.13182/NT12-A13481 

To link to this article: http://dx.doi.org/10.13182/NT12-A13481 

Published online: 22 Mar 2017. 

Submit your article to this journal 

Article views: 4 

View related articles 

Full Terms & Conditions of access and use can be found at 

http://www.tandfonline.com/action/journalInformation?journalCode=unct20 

Download by: [Australian Catholic University] Date: 17 August 2017, At: 03:26

A FUNCTIONALLY GRADED COMPOSITE 

FOR SERVICE IN HIGH-TEMPERATURE 

LEAD- AND LEAD-BISMUTH–COOLED 

NUCLEAR REACTORS—I: DESIGN 

NUCLEAR PLANT 

OPERATIONS 

AND CONTROL 

KEYWORDS: corrosion, composite, 

lead-bismuth 

MICHAEL PHILIP SHORT* and RONALD GEORGE BALLINGER 

Massachusetts Institute of Technology, H. H. Uhlig Corrosion Laboratory 

185 Albany Street, Cambridge, Massachusetts 02139 

Downloaded by [Australian Catholic University] at 03:26 17 August 2017 

Received January 24, 2011 

Accepted for Publication May 13, 2011 

A material system that resists lead-bismuth attack 

and retains its strength at very high temperatures has 

been developed that enables increased outlet temperature 

and the promise of allowing increased coolant velocity 

and efficiency of lead- and lead-bismuth–cooled 

reactors if the behavior reported here is confirmed by 

long-term tests. The development of this system represents 

an enabling technology for lead-bismuth–cooled 

reactors. The system is a functionally graded composite 

(FGC), with separate layers engineered to perform corrosion 

resistance and structural functions. Alloy F91 was 

chosen as the structural layer of the composite because 

of its strength and radiation resistance. An Fe-12Cr-2Si 

alloy was developed based on previous work in the Fe- 

Cr-Si system, and was used as the corrosion-resistant 

I. INTRODUCTION 

A deviation from the evolutionary path of light water 

reactor design will soon be required to meet the rising 

energy demands of the world, in order to more efficiently 

extract energy from the same fuel while simultaneously 

improving safety, reducing the possibility for proliferation, 

and simplifying reactor design. These advanced systems, 

designated as Generation IV ~Gen IV! systems, 

promise higher efficiencies, improved uranium utilization, 

and increased safety margins. 1 However, each Gen 

IV system also poses unique challenges to successful 

implementation. In the case of lead-bismuth–cooled systems, 

the primary issue is the corrosion of structural 

materials. 

*E-mail: hereiam@mit.edu 

cladding layer because of its chemical similarity to F91 

and its superior corrosion resistance in lead and leadbismuth 

in both oxidizing and reducing environments. 

The availability of the FGC will have significant impacts 

on lead-bismuth reactor design. The allowable increases 

in outlet temperature and coolant velocity lead to a large 

increase in power density—either to a smaller core for 

the same power rating or to more power output for the 

same-size core. In this paper, we report on the overall 

design of the FGC. We also discuss the general implications 

for lead-bismuth–cooled reactor design. In a future 

paper, we will discuss the fabrication and the initial evaluation 

of the actual product produced using commercial 

processing methods. 

I.A. Comparison of Proposed Gen IV 

Reactor Coolants 

Six principal coolants are being investigated by countries 

pursuing Gen IV nuclear technologies as viable options 

for reactor coolants. 1–7 These coolants have been 

identified as possessing superior properties to others considered, 

and they allow for reactor designs that meet the 

goals of the Gen IV program. 1 Each coolant has specific 

advantages and disadvantages, all relating to their physical 

properties, which are summarized in Table I. 

The liquid metals and molten salts have superior heat 

conductivities and heat capacities relative to the other 

coolants. The high boiling points of the liquid metals 

ensure that the system will not need to be pressurized to 

remain in the liquid phase. Instead, an unpressurized, 

inert cover gas, such as argon, is all that is necessary to 

exclude air from the system. This means that structural 

366 NUCLEAR TECHNOLOGY VOL. 177 MAR. 2012

Short and Ballinger 

A COMPOSITE FOR HIGH-TEMPERATURE LEAD- AND LEAD-BISMUTH–COOLED REACTORS—I 

TABLE I 

Relevant Physical Properties of Selected Gen IV Reactor Coolants 

Coolant 

Melting 

Point 

~8C! 

Outlet 

Temperature 

~8C! 

Boiling 

Point 

~8C! 

rc p 

~J0cm 3 {K! a 

k 

~W0m{K! a 

Reactor 

Pressure 

~MPa! 

Viscosity 

~Pa{s! a 

Helium 8 272 950 269 0.0143 ~Ref. 9! 0.0357 7.0 4.111 10 5 

Lead-bismuth ~Ref. 10! 123.3 700 1670 1.37 17.5 0.1 1.0 10 3 

Lead 10 327.4 700 1737 1.51 19.9 0.1 1.3 10 3 

Sodium 11 97.8 550 892 1.03 ~Ref. 12! 64.9 ~Ref. 13! 0.1 2.2 10 4 ~Ref. 13! 

Molten salt 11 b 

396 700 2500 1.72 0.39 0.1 1.18 10 3 

Supercritical H 2 O ~Refs. 6 and 14! 0 500 100 0.225 0.09 22.1 3 10 5 

a 

Selected property is at the given outlet temperature and pressure. 

b 

The salt chosen for this comparison is ~3Na–2K–5Mg!Cl x . 


materials in the reactor do not have to be able to tolerate 

a higher operating pressure; they need only to withstand 

their own weight and that of the coolant. The liquid metals 

and molten salts also have higher atomic masses, and 

therefore less neutron moderation, than the other three 

proposed coolants, which make them particularly suitable 

for fast reactors. Helium shares the benefit of staying 

in one phase with the molten metals and salts; however, 

the temperatures required to raise the efficiency levels of 

helium-cooled reactors ~up to 10008C! put enormous demands 

on the materials of the reactor. Supercritical water 

possesses heat transfer properties between those of helium 

and the molten metals and salts, but it must be 

highly pressurized, again putting enormous strain on reactor 

materials. 

While some physical and thermal properties of sodium 

exceed those of lead and lead-bismuth, it should be 

noted that these latter two possess two main benefits in 

the context of fast reactors. First, the atomic number of 

lead ~and the average of lead-bismuth! is higher than that 

of sodium, leading to a lower amount of neutron moderation. 

15 This results in fewer neutrons losing energy and 

being captured by resonances in lower-nergy regions. 

Second, lead and lead-bismuth do not possess the chemical 

incompatibility with water or moist air that sodium 

does. This leads to less of a safety concern in the event of 

a coolant leak for lead and lead-bismuth. 

I.B. Material Compatibility Issues with Lead and 

Lead-Bismuth 

The most significant drawback to using lead-bismuth 

as a high-temperature coolant is corrosion, which has 

been extensively discussed in the framework of mediumpower 

lead-cooled reactors. 16 As with many working fluids, 

oxygen is soluble to some degree in lead-bismuth. 

The presence of dissolved oxygen in the working fluid 

requires materials either to be protectively coated or to 

form their own coating via passivation—the formation of 

a protective oxide film. Many materials including chromia, 

alumina, and silica formers possess the ability to do 

this, and many working fluids can carry sufficient oxygen 

to induce passivation in these materials. The main 

difference that makes lead-bismuth so incompatible with 

some of these materials, and nickel in particular, is their 

behavior in environments that are below the chemical 

potential ~a combination of oxygen concentration and 

temperature! for forming protective oxide films. 

Many metals have high solubilities in lead and leadbismuth. 

Without an oxide layer protecting the surface 

of the metal, materials are free to dissolve into the working 

fluid. Nickel has a particularly high solubility of 

4.8 wt% in lead-bismuth at 7008C ~Ref. 17!. Alloys 

with even a few percent nickel are subject to severe 

liquid-metal attack and dissolution in lead and leadbismuth 

systems. 18 This problem is further aggravated 

by the fact that the coolant is flowing in a thermal loop. 

This creates a transport cycle where because of the temperature 

dependence of solubilities, material in the hot 

leg is removed and deposited in the cold leg, simultaneously 

weakening the hot-leg structural materials and 

clogging the pipes in the cold leg. This was a problem 

with past lead-bismuth–cooled systems, experienced most 

dramatically in Soviet lead-bismuth–cooled submarines. 19 

Pure iron oxide is a poor barrier to corrosion over the 

proposed operating temperature range ~6008C to7008C! 

due to a phase transformation of the oxide from magnetite 

to wüstite at 5708C ~Ref. 20!. Wüstite is a porous, 

brittle oxide phase that does not protect the underlying 

metal. 

Both of the situations described above would lead to 

accelerated corrosion in regions of oxygen depletion, such 

as in pits or crevices, which function as powerful corrosion 

accelerators for materials in lead-bismuth. As oxygen 

inside a crevice is consumed, it is not replenished by 

oxygen-bearing fluid because of restricted transport of 

oxygen into the crevice. As a result, the oxygen potential 

can drop dramatically inside the crevice, thinning or even 

eliminating the protective passive oxide layer and leaving 

bare metal exposed to a very reducing environment. 

This could cause crevices or pits to grow, resulting in 

stress concentrators and localized wall thinning. 

NUCLEAR TECHNOLOGY VOL. 177 MAR. 2012 367




Any material expected to perform well in leadbismuth 

must be able to remain protected under crevice 

corrosion conditions, as many crevices will exist in any 

reactor design. Protection can be accomplished by means 

of either an external coating, applied to the part before 

service, or as mentioned above, the development of an 

intrinsic coating grown by the material itself. However, it 

is much more desirable from the standpoints of economics, 

maintenance, and reactor longevity to have a material 

system that can create and regenerate its own coating, 

even in very reducing environments. Such a system should 

make use of materials that are inherently resistant to dissolution 

in lead-bismuth, even in the absence of a protective 

film. The material must be able to form a fully 

dense oxide layer at extremely low potentials, and that 

layer must remain stable should the potential temporarily 

drop below the level required to reform it. In addition, 

should the protective layer become unstable, the underlying 

material should be fairly resistant to dissolution. In 

this paper, we describe the development of one such 

system. 

I.C. One Solution: A Functionally Graded Composite 

A review of work to date on materials for leadbismuth 

reveals no single material that can resist corrosion, 

withstand operating stresses at temperatures up to 

7008C, and compete economically. One solution is to 

develop a composite material, whose layers satisfy requirements 

individually, that can function effectively as 

a system in the harsh environment of lead-bismuth at 

7008C. Having two separate layers with individual functions 

greatly increases the composite system’s ability to 

perform its necessary tasks of resisting both corrosion 

and deformation. 

The general requirements for the composite system 

are as follows: 

1. A corrosion-resistant layer is required to prevent 

liquid-metal attack or oxidation in all environments, even 

below those necessary for the formation of iron oxides. 

The corrosion-resistant layer need not be structurally 

significant. 

2. The corrosion-resistant layer must be self-healing 

and resistant to flow-assisted corrosion ~FAC!. 

3. A second, structural layer is required to act as the 

backbone to provide tensile strength, creep resistance, 

and radiation resistance over the lifetime of the component 

at temperatures up to 7008C. 

4. The two layers must be microstructurally similar 

enough to form a strong interface. 

5. The composite system must be able to be inexpensively 

produced using current commercial practice in 

industrial quantities. A number of studies have been performed 

that evaluate the excellent corrosion resistances 

of many new, exotic alloys such as HCM12A ~Ref. 21!, 

Kanthal-22 ~Ref. 22!, and others 23–25 designed to resist 

lead and lead-bismuth attack. However, the majority of 

the alloys that show the most favorable results are not 

currently made in significant quantities. 

6. The composite system must be able to be produced 

in complex shapes, such as elbows and valves. 

7. The corrosion-resistant layer material must be capable 

of being easily and reliably applied. 

The composite material approach has been widely used 

in the nuclear industry in the manufacture of fuel cladding, 

where an overlay or sleeve of pure Zr is applied to 

a billet of Zircaloy and co-extruded. 26 This approach is 

also used in the fossil boiler industry, where highstrength, 

creep-resistant material is clad with an overlay 

of high chromium alloy. 27 

The overall goal of developing this composite is to 

increase the operating parameters of lead-bismuth– 

cooled reactors. Allowing a higher outlet temperature 

will provide two major benefits. First, an increased outlet 

temperature translates into an increase in thermal-toelectric 

efficiency. Second, a higher outlet temperature 

yields higher-quality process heat, which is useful for the 

purposes of driving endothermic chemical reactions, such 

as the production of hydrogen. In addition, a more protective 

oxide layer should allow for a higher coolant flow 

velocity, which translates into a higher core power density. 

I.D. General System Description 

The functionally graded composite ~FGC! developed 

here relies on a thin, corrosion-resistant layer based 

on the Fe-Cr-Si system. A schematic showing how the 

layers in the composite will be applied or formed is shown 

in Fig. 1. The presence of enough chromium and silicon 

creates a synergistic effect that promotes the formation 

of a thin, dense, and corrosion-resistant oxide scale on 

the surface of the metal. 28 This oxide layer should be able 

to resist corrosion from both liquid-metal attack ~dissolution! 

and oxidation ~both general and internal! at any 

oxygen potential that could be encountered in a leadbismuth–cooled 

reactor. The oxide will protect the FGC 

by acting as a diffusion barrier to both oxygen ingress 

and metal dissolution, eliminating the problem of corrosion 

in these systems. 29 

The corrosion-resistant layer is supported by a structural 

layer made of a ferritic0martensitic ~F0M! material 

with a similar base composition in terms of chemistry. 

Choosing layers with similar chemistries will help minimize 

diffusional dilution upon long-term exposure to high 

temperatures. A F0M material was chosen for its high 

resistance to radiation damage at high neutron dose. 30 

This structural layer must possess superior resistance to 

creep in order to withstand internal stresses at the end of 

a fuel cycle because of fission gas pressure buildup, which 





Fig. 1. Diagram showing zones expected to develop during 

operation of the composite. The structural layer of the 

composite, made of alloy F91, provides its strength, 

while a passivating layer generated by the corrosionresistant 

layer prevents internal oxidation and liquidmetal 

dissolution. The passive layer should be thin 

enough not to deplete the underlying metal of protective 

elements, so that in the event of removal of this 

layer, the composite can repassivate itself. The scale 

has been exaggerated for illustrative purposes. 

can be as high as 15 MPa in comparable fast reactors. 31 

The corrosion-resistant layer is weld-overlaid onto the 

structural layer, must provide excellent interface characteristics, 

and must ensure a smooth microstructural 

transition between the layers.Asmooth transition in composition 

and microstructure across the interface will minimize 

interfacial stresses due to lattice parameter mismatch 

and differences in thermal expansion. The F0M structural 

layer will also excel in corrosion resistance compared to 

austenitic stainless steels, which have been shown to undergo 

severe attack, due to the high solubility of nickel 

and film instability above the wüstite formation temperature 

~5708C!~Refs. 18, 32, and 33!. In the event of total 

cladding breach, the structural layer must corrode at a manageable, 

well-documented rate, as compared to the rapid, 

unpredictable nature of corrosion of austenitic stainless 

steels in lead-bismuth above 6008C ~Refs. 18 and 33!. 

II. SPECIFIC CHOICES FOR FGC LAYERS 

II.A. Structural Layer 

Because the Fe-Cr-Si system was chosen for corrosion 

resistance, the structural layer must have as similar 

a chemistry as possible to the cladding layer to ensure 

microstructural compatibility. Austenitic stainless steels 

were thus excluded from this study for the following 

reasons. Fe-Cr-Si will be entirely ferritic ~body-centered 

cubic!, while austenitic stainless steels are face-centered 

cubic. Austenitic stainless steels have been observed to 

undergo severe radiation swelling. 30 In addition, many of 

them contain nickel, which is highly soluble in leadbismuth 

~Ref. 34!. Finally, microstructural stability of 

the interface between the two layers of the composite is 

required, and using more similar layers will generally 

lead to higher microstructural stability of the interface. 

Therefore, F0M, Fe-Cr–based steels including F91, 

HT-9, NF616, EUROFER, F82H, HCM12A, Oak Ridge 

National Laboratory ~ORNL! 9Cr-2WVTa, MA957 oxidedispersion-strengthened 

~ODS! steel, and PM2000 ODS 

were considered on the bases of strength, chemistry, creep 

resistance, radiation resistance, and fabricability. The alloys 

in this list have been considered for the purposes of 

structural properties and corrosion resistance in leadbismuth–cooled 

reactors. 30 However, few of these are 

commercially available in large amounts, if at all. Applying 

the availability criteria reduced the number of 

choices to two: ~a! HT-9 ~Fe-12Cr-1MoVW! and ~b! F91 

~Fe-9Cr-1MoNbVW!. 

II.A.1. Effects of Radiation on Alloy Performance 

There is a large database on the radiation behavior of 

HT-9, due to its selection as the structural material in the 

Clinch River Breeder Reactor. Both HT-9 and F91 have 

been tested in Experimental Breeder Reactor ~EBR!-II 

and the Fast Flux Test Facility ~FFTF! to doses over 200 

dpa ~Ref. 30!. Nevertheless, enough data have been accumulated 

since then to make an informed decision between 

the two based on how their mechanical properties 

change under irradiation. Two easily measurable properties 

affected by radiation are the yield stress s y and the 

ultimate tensile strength ~UTS!, increases in which are 

indicators to the degree of radiation embrittlement suffered 

by an alloy. Figure 2a shows the change in these 

two parameters for F91 as performed in EBR-II at a dose 

of ;12 displacements per atom ~dpa!. It is clear that 

beyond 4508C, there is no notable difference in s y or 

UTS between the unirradiated and the irradiated material. 

This is likely because at higher temperatures, the 

defects produced by radiation damage are more free to 

move throughout the steel and many more of them end up 

recombining or annihilating at defect sinks, healing some 

of the radiation damage in real time. Real-time recombination 

is especially active in F0M steels, which tend to 

contain a high number of features ~grain boundaries, subgrain 

boundaries, precipitates, and dislocations! that serve 

as recombination sites for vacancy-interstitial pairs produced 

by radiation. 39 

Properties related to ductility that change with exposure 

to radiation can also be compared for the two 

alloys, such as the shift in the ductile-to-brittle transition 

temperature ~DBTT!. Figure 2b shows two such curves 

comparing irradiated HT-9 with F91 and ORNL 9Cr- 

2WVTa, a reduced activation steel developed as a nextgeneration 

iteration on F91 ~Ref. 40!. In both curves, 





Fig. 2. Shift in properties for unirradiated and irradiated HT-9 and F91. Alloy F91 undergoes less degradation in mechanical properties 

under high-irradiation doses at temperatures considered in this study. ~a! Change in s y and UTS of F91 at 

;12 dpa fast flux. 35 Note how the mechanical properties of F91 are not noticeably affected above 4508C. ~b! Shift in DBTT 

for HT-9 and F91 after 26 dpa fast flux irradiation. 36 The DBTT for unirradiated alloy F91 in small punch tests is 1538C 

~Ref. 37!, while that of alloy HT-9 in Charpy impact tests is 08C ~Ref. 38!. ~c! Comparison of creep-rupture lifetimes of 

HT-9 and alloy F91 ~Mod 9Cr-1Mo!, showing better performance by alloy F91 at higher temperatures. 30 

alloys based on F91 show better creep and radiationresistance 

properties than HT-9. The creep-rupture strength 

of F91 is also higher than that of HT-9, as Klueh and 

Nelson showed in a recent review of the literature. 30 

Note that F91 excels compared to HT-9 at higher temperatures. 

This is due to the niobium and vanadium additions 

present in F91, which promote the formation of 

fine carbides and carbonitrides that impede dislocation 

motion, helping to slow creep. 41 

A recent study has suggested that F0M steels have 

longer incubation periods for the initiation of swelling, 

eventually exhibiting similar swelling tendencies as those 

for austenitic steels. 42 This is likely because the alloys 

tested and reviewed in this study were of uniform microstructure, 

such as simple tertiary Fe-Cr-Ni alloys and 

SUS-316 stainless steel. By contrast, the fine-grained 

nature of F0M steels provides numerous sinks where defects 

can annihilate, including a high density of grain 

boundaries, dislocations, and MX precipitates, where M 

is primarily niobium or vanadium and X is carbon or 

nitrogen 30 in F91. These MX precipitates exist both on 

grain boundaries and inside the grains themselves, preventing 

the movement of dislocations through the material 

and imparting superior creep resistance compared to 

similar alloys without such intragranular particles, such 

as HT-9 ~Ref. 30!. Additionally, the F0M steels do not 





contain nickel, which is a significant source of helium 

via the ~n, a! reaction that can stabilize void nuclei. 43 

II.A.2. Microstructural Stability 

While chromium does improve corrosion resistance, 

it also can lead to deleterious microstructural effects. 

Phase diagrams for Fe-Cr, such as those in Fig. 3, show 

that a solubility limit for chromium in a-Fe exists in the 

low-Cr region at temperatures below 7008C ~Refs. 20 

and 44!. At concentrations that exceed the solubility, a 

phase separation can take place into a-Fe and s-Cr. s-Cr 

has a brittle, platelike structure and is therefore more 

susceptible to fracture. 45 Micrographs of s-Cr precipitation 

and associated cracking as measured by Babakr et al. 

are shown in Fig. 4 to illustrate this crack-susceptible 

microstructure. Note how the platelets of s-Cr are highly 

aligned, which provides fast fracture directions along 

which a crack can propagate. Therefore, the structural 

material of choice should be chosen to avoid embrittlement 

due to s-Cr precipitation. Phase separation may be 

worrisome even at 5008C, as the solubility limit of chromium 

drops to 9 wt% ~Ref. 20!. However, this should not 

be of concern for two reasons. First, the growth of any 

nucleated s-Cr is kinetically limited at lower temperatures, 

and the composite will spend most of its life at 

higher temperatures ~6008C to 7008C! where the solubility 

of Cr in a-Fe is higher. Second, the results of recent 

computational investigations and experimental literature 

reviews by Bonny et al. have suggested a change is needed 

to the currently accepted Fe-Cr phase diagram. 44 The 

conclusive figure from this effort is reproduced in Fig. 3. 

This figure suggests that Cr is actually more soluble in 

a-Fe at temperatures below 6508C than previously thought. 

In addition, the presence of manganese and nitrogen slow 

the formation of s-Cr by suppressing the sigma phase 

Fig. 3. Modified Fe-Cr phase diagram in the low-Cr region 

based on work by Bonny et al. 44 

solvus temperature to levels below annealing temperatures. 

46 This will prevent s-Cr from being formed during 

processing. Manganese can also be added as an austenite 

stabilizer to minimize s-Cr, as well as scavenge any free 

sulfur present in the metal during casting. 47 In conclusion, 

a9to10wt%Crsteel is the most logical choice to 

ensure microstructural stability. 

Alloy F91 was therefore chosen as the structural layer 

for the composite. Compared to HT-9, F91 has superior 

microstructural stability, better creep resistance, a lower 

shift in DBTT, a higher Charpy impact energy, a higher 

creep-rupture life, and an absence of change in s y under 

irradiation in the expected operating temperature range. 

In support of HT-9, it has a more extensive database of 

radiation damage properties, and it is less expensive to 

produce. F91 requires a more careful heat treatment. However, 

since F91 can handle larger stresses, parts made out 

of F91 can be thinner than their HT-9 counterparts. This 

directly reduces the material cost of making parts out of 

F91. The database for radiation performance of alloy 

F91, while more limited, is adequate to make an informed 

decision. 

II.B. Cladding Layer 

Previous studies by Lim et al. 48 and Müller et al. 49 

have shown that Fe-Cr alloys are not sufficient to protect 

against lead-bismuth dissolution at temperatures above 

the wüstite phase transformation temperature ~;5708C!. 

This is because liquid metal has been found to penetrate 

the oxide layer, allowing oxygen to diffuse through channels 

of liquid metal into the base metal to cause internal 

oxidation. 50 Therefore, the addition of one or more additional 

strong scale formers is required. 

Thermodynamics as illustrated in the Ellingham diagram 

in Fig. 5 suggest silicon and aluminum as logical 

choices, due to the stability of their oxide scales at low 

oxygen concentrations. 51 Both silicon and aluminum are 

used to “kill” steels, or deoxidize them, during casting. 

52 The Ellingham diagram shows that the oxidation 

potential for alumina is orders of magnitude lower than 

that of silica, suggesting that Al alloys would perform 

better in extremely low-oxygen environments. However, 

the addition of aluminum presents a significant 

concern for commercial fabrication. Studies have shown 

that Al-bearing alloys experience problems during welding 

due to hydrogen cracking in embrittled FeAl and 

FeAl 3 intermetallic phases formed during welding. 53 

FeCrAl alloys can become severely hydrogen embrittled, 

subsequently cracking after welding, even after a 

proper postweld heat treatment. 54 Studies by DuPont 

et al. and Regina et al. have shown that adding aluminum 

to iron in a weld metal limits the amount of chromium 

that can be added without increasing the 

susceptibility to cracking. 55,56 Figure 6 presents a weldability 

map for higher-carbon Fe-Cr-Al alloys that 

shows where hydrogen-induced cracking occurred during 





Fig. 4. Micrographs of s-Cr precipitation and its effects in alloy HK-40, an Fe-Cr-Ni alloy, after exposure to 8508C for no more 

than 4000 h ~Ref. 45!. ~a! Optical micrograph of s-Cr forming on prior austenitic grain boundaries and as highly oriented 

platelets. ~b! SEM image of a crack proceeding directly along s-Cr phase boundaries, showing severe weakening of the 

material at the phase boundary. 

have at least 10 to 12 wt% Al and 4 to 5 wt% Cr in 

order to adequately resist corrosion in gaseous NO x environments. 

This composition falls below the weldability 

line in their study. 56 In addition, adding TiC to prevent 

cracking ~by acting as a hydrogen sink! results in the 

precipitation of TiC on grain boundaries in the interdendritic 

liquid as the melt cools, 55 leading to microstructural 

inhomogeneity in the weld. 

The cladding layer for the FGC was therefore based 

on the Fe-Cr-Si system. In choosing the composition of 

this alloy, the benefits of added corrosion resistance 

were considered along with microstructural, macrostructural, 

and radiation considerations to find the best balance 

of properties. The ultimate goal of this process 

was to ensure that the cladding remains intact over decades 

of fast reactor operation. As mentioned before, 

more chromium and silicon increase the corrosion resistance 

of F0M steels. Therefore, the Fe-Cr-Si chemistry 

was chosen to maximize the concentrations of chromium 

and silicon without compromising radiation performance, 

corrosion resistance, microstructural stability, 

or ease of processing. 

II.B.1. Chromium Concentration 

Fig. 5. Ellingham diagram for metal0metal oxide systems found 

in this study. 51 Oxides are stable above each solid line, 

and metals are stable below each solid line. Oxygen 

concentrations and corresponding hydrogen–to–water 

vapor ratios are shown with dotted lines. 

welding of these alloys. 55 The weldability of similar 

Fe-Cr-Al alloys with lower carbon is worse, resulting in 

a shift of the curve to lower Cr and Al concentrations. 56 

Regina et al. have shown that an Fe-Cr-Al alloy must 

Studies have shown that corrosion resistance increases 

with added chromium content. 57 The addition 

of too much chromium can lead to s-Cr precipitation, 

as mentioned earlier. However, a chromium concentration 

of 1 to 2 wt% over the solubility limit is not expected 

to produce severe enough s-Cr precipitation to 

adversely affect a thin layer of cladding material. Therefore, 

based on the desire to increase corrosion resistance 

and on reasonable assumptions about the prevention 

of secondary phase precipitation, a chromium concentration 

of 12 wt% was chosen for the cladding layer of 

this composite. 





Fig. 6. Weldability map for Fe-Cr-Al alloys, showing which alloys crack during gas tungsten arc welding due to hydrogeninduced 

cracking. TiC is added as a hydrogen sink to mitigate this problem. 55 

II.B.2. Silicon Concentration 

Increasing the silicon level is also desirable, as it has 

been shown to encourage the formation of SiO 2 beneath 

the Cr 2 O 3 oxide layer. Lim has shown that an increasing 

Si concentration leads to the formation of a higher volume 

fraction of SiO 2 and FeSi 2 O 4 precipitates. 58 The 

total oxide layer thickness has also been shown to vary 

inversely with silicon content. These data can be interpolated 

beyond those collected by Lim to suggest that 

further increasing the silicon concentration should theoretically 

produce a fully dense, thin layer of SiO 2 . This 

effect was recently observed by Short. 29 The speed of 

formation of this layer is also expected to be proportional 

to the silicon content as well as the chromium content. 

This is because a higher chromium content would lead to 

faster formation of a thinner oxide layer, allowing little 

enough oxygen to enter so that only SiO 2 can form underneath 

the layer of Cr 2 O 3 . Once this second layer is 

complete, a very effective, stable barrier to oxygen and 

dissolution will remain regardless of the fate of the iron 

oxides facing the liquid metal. This is because oxides act 

as powerful diffusion barriers to ion transport as compared 

to metals. The diffusion coefficients for iron in 

oxides are between 3 and 12 orders of magnitude lower 

than their counterparts in relevant metals, as can be seen 

in Fig. 7 ~Ref. 29!. In addition, forming a thinner oxide 

layer results in a smaller depletion zone of Cr and Si 

directly beneath it, as forming the oxide layer consumes 

less Cr and Si from the bulk. A thinner oxide layer therefore 

results in less time required for Cr and Si concentrations 

directly beneath the oxide to return to their original 

levels by diffusion from the bulk. This results in a faster 

ability to repassivate should the oxide layer be mechanically 

removed more than once. 

Adding too much silicon does have its drawbacks. 

Experimental studies performed by Miller et al. have 

suggested that silicon can induce radiation embrittlement 

via preferential segregation at dislocations and grain 

boundaries, probably as a result of physical displacement 

by radiation. 59 The presence of too much silicon can also 

lead to the precipitation of Si-rich precipitates, mainly 

Fe-Si intermetallic compounds, that would lead to localized 

depletion of silicon in the matrix. 60 Enhanced liquidmetal 

embrittlement of silicon-enriched Fe-Cr bainitic 

steels has been confirmed by Van den Bosch et al. 61 This 

is not expected to be a concern in the context of this 

composite, as the corrosion-resistant layer is specified to 

be completely ferritic, thereby avoiding problems of lack 

of ductility that could be enhanced by lead-bismuth. 

Therefore, based on concerns relating to radiation embrittlement 

due to silicon segregation and the desire to avoid 

Si-rich phase precipitation, a maximum silicon concentration 

of 2.0 wt% was chosen for the cladding layer. At 

the same time, a lower limit of 1.25 wt% was chosen based 

on the work of Lim as being the minimum Si concentration, 

in combination with 12 wt% Cr for adequate corrosion 

resistance in lead-bismuth. The final composition of 

this composite’s cladding layer was therefore chosen to 

be Fe-12Cr-2Si. 

II.C. Comparison of the Cladding Alloy with 

Previous Work 

Fe-Cr-Si–based alloys have been developed in the 

past with the intent of meeting all the requirements for 





Fig. 7. Examples of diffusion coefficients for iron in relevant metals and oxides. 29 Data are plotted over cited temperature ranges. 

Markers for metals are solid, while those for oxides are hollow. Note how much lower diffusion coefficients are for 

chromium and silicon oxides. 

service in lead-bismuth–cooled reactors. However, single 

alloys have not shown adequate results. Studies on 

Russian Fe-Cr-Si alloys have shown unfavorable corrosion 

rates. The closest example to this work is alloy EP- 

823, whose composition is shown in Table II along with 

the alloys used in this study. EP-823 showed more favorable 

results compared to similar Fe-Cr martensitic steels 

without silicon, such as F91. Analysis demonstrated that 

alloy EP-823 produces a thinner, more Cr-rich scale than 

Optifer IV or F91 ~Ref. 62!. However, the scale is still on 

the order of tens of microns in thickness, and it contains a 

significant amount of iron. The fact that EP-823’s oxide 

thickness grows with a square root law indicates that the 

oxide layer continues to grow by diffusion through the 

oxide layer. 62 

Other methods have been developed to protect 

against lead and lead-bismuth corrosion, including applying 

a coating to the metal surface. However, many 

of these coatings suffer from problems of adhesion and 

cracking, including precipitated coatings developed by 

Weisenburger et al. 63 The Gepulste Elektronen Strahl 

Anlage ~GESA! process, however, has produced mainly 

Al-based thin surface coatings that have been immensely 

successful at resisting lead-bismuth corrosion, 

even above the magnetite-wüstite phase transformation. 

63,64 The two major drawbacks to this method are 

TABLE II 

Composition of the Russian Fe-Cr-Si Alloy EP-823 Compared to Those Used in This Study* 

Element 

Fe Cr Si Ni Mo Mn V Nb W Ti C 

EP-823 ~Ref. 62! Balance 12 1.8 0.89 0.7 0.67 0.43 0.4 1.2 — 0.14 

F91 Balance 9.43 0.35 0.28 0.96 0.51 0.19 — 0.07 — ;0.1 

MIT Fe-12Cr-2Si Balance 13.11 2.00 0.006 — 0.02 — — 0.17 — 0.01 

*In weight percent ~Ref. 62!. Compositions of F91 and Fe-12Cr-2Si were acquired by inductively coupled plasma mass spectrometry. 





its cost and its limited applicability, especially when 

considering dilution due to welding. 

III. CONFIRMATORY DATA 

A comprehensive experimental testing program has 

been undertaken to test many of the considerations put 

forth in Sec. II ~Ref. 29!. The testing program explored 

three major areas of the composite’s viability: corrosion 

resistance, diffusional stability, and commercial fabricability. 

The first two will be discussed here, while the 

commercial fabricability will be discussed in Ref. 65. 

III.A. Corrosion Resistance 

Samples of each alloy ~F91 and Fe-12Cr-2Si! 

were immersed in static lead-bismuth for up to 506 h at 

temperatures up to 7158C, at oxygen concentrations above 

~P H2 

0P H2 O ' 0.18! and below ~P H2 

0P H2 O ' 100 

1000!, those necessary to form iron oxides. Details concerning 

the specific oxygen concentrations, the test setup, 

and experiment results of this study can be found in 

another paper. 29 The corrosion resistance of the Fe-12Cr- 

2Si alloy in lead-bismuth in environments both oxidizing 

and reducing with respect to Fe-oxide formation 

was outstanding. Scanning electron microscope ~SEM! 

micrographs and associated energy dispersive X-ray 

~EDX! maps showed that a 200- to 400-nm oxide layer 

formed rapidly on the surface of alloy Fe-12Cr-2Si, 

regardless of whether the oxygen concentration in the 

lead-bismuth was above or below the stability line for 

the formation of iron oxides. An example SEM micrograph 

of the oxide layer is reproduced in Fig. 8. X-ray 

photoelectron spectroscopy confirmed that the oxide layer 

Fig. 8. Fe-12Cr-2Si after exposure to 7158C reducing leadbismuth 

for 506 h, 25 000. 

is composed principally of chromium and silicon oxides, 

which are excellent diffusion barriers to further 

oxygen ingress or metal ion escape. 29 Elemental concentration 

profiles acquired by secondary ion mass spectroscopy 

~SIMS! confirmed the thickness of the oxide 

layer. They also showed that a chromium-rich oxide 

exists at the outer surface of the alloy, followed by a 

silicon-rich oxide beneath the Cr-rich oxide layer. A 

representative SIMS profile is shown in Fig. 9, which 

shows the locations of elemental enrichment as well as 

the coincidence of oxygen with Cr and Si enrichment. 

The corrosion of alloy F91 followed predictable rates 

as functions of the temperature and the environment. The 

main mechanism of corrosion was by dissolution, as F91’s 

minimum chromium concentration of 8.50 wt% ~Ref. 66! 

is insufficient to form a protective passivating oxide 

layer. 57 SEM micrographs and EDX elemental maps in 

Fig. 10 confirm this by showing lead-bismuth penetration 

coincidental with chromium depletion. A summary 

of the corrosion data for alloy F91 acquired in this experimental 

program is shown in Fig. 11. These data show 

that diffusion of dissolved species is the rate-limiting 

factor in alloy F91 corrosion. If this process scaled as a 

normal Arrhenius-based temperature process, the difference 

between corrosion at 6008C and 7008C would have 

been far larger. 

III.B. Diffusional Stability 

The FGC has been shown to be diffusionally and 

microstructurally stable at temperatures up to 7508C. Diffusion 

couples were aged at 7008C and 7508C for times 

up to 1200 h. Data acquired at 7508C were scaled using 

the Arrhenius equation to 7008C. A graph of these aging 

times is shown in Fig. 12, along with a logarithmic regression 

equation that allows for extrapolation to longer 

times. The minimum silicon concentration of 1.25 wt% 

was chosen for the reasons stated in Sec. II.B. 

IV. IMPACT OF SUCCESS OF THE FGC 

The development of this FGC represents an enabling 

technology for lead-bismuth–cooled reactors, relaxing 

previous limits on temperature and likely flow rate, which 

made the construction and operation of lead-bismuth– 

cooled reactors prohibitively expensive. This sentiment 

has been echoed in very recent reports about the viability 

of comparable Gen IV fast reactor technologies. The most 

recent Massachusetts Institute of Technology ~MIT!0 

Electric Power Research Institute0Nuclear Energy Institute 

advanced fuel cycle report 67 states the following: 

. . . structural material corrosion has limited the actual 

peak coolant temperatures of LFRs to levels 

lower than SFRs resulting in lower thermal-toelectricity 

efficiency. Corrosion problems also limit 

the velocity of lead coolant through the reactor core, 





Fig. 9. SIMS concentration profile of alloy Fe-12Cr-2Si after exposure to oxidizing lead-bismuth for 506 h at 7008C. The data 

show a total oxide layer thickness of 200 nm, with a Cr-rich oxide at the outer surface of the sample followed by a Si-rich 

layer beneath. 

Fig. 10. EDX elemental maps of F91 exposed to 7158C reducing lead-bismuth for 506 h, 1000. A chromium depletion region 

was present throughout the region of the lead-bismuth attack, suggesting that selective dissolution of chromium was the 

dominant attack mechanism. ~a! SEI is defined as secondary electron image. 

resulting in larger, more costly, reactor cores. New 

high-temperature metal alloys that are corrosion resistant 

in lead have been developed in the laboratory 

but have not been tested under the full set of credible 

reactor conditions. If the corrosion resistant characteristics 

of these alloys are confirmed for realistic 

reactor conditions and assuming that there are no 

other unexpected challenges, LFRs could become an 

attractive alternative to SFRs. 

IV.A. Performance Gains 

Using the FGC developed in this research allows an 

increased outlet temperature. This increases the thermodynamic 

efficiency of the cycle. At increased allowable 

temperatures, the higher quality of the heat allows the 

use of lead-bismuth–cooled reactors in the production of 

hydrogen. Examples of gains in hydrogen production 

efficiency by using the FGC from this research are shown 

in Fig. 13 ~Refs. 68 and 69!. 

Amore subtle but high impact gain achieved by using 

this FGC is the potential for an increased coolant flow 

velocity. Previous designs were limited to 2 m0s because 

of concerns for FAC ~Ref. 70!. The presence of an effective 

diffusion0dissolution barrier would eliminate the concern 

for FAC. Fully dense chromia and silica resist shear 

deformation better than iron oxides formed at this temperature. 

71 They also adhere to the base metal more effectively 

because of better matches in thermal expansion, 71 

allowing the coolant to flow at a higher velocity. 





Fig. 11. Corrosion distances for alloy F91 exposed to lead-bismuth. These distances were measured from the original flat surface 

of each material to the end of the deepest observed lead-bismuth corrosion channel. The exponents show that corrosion 

of alloy F91 in lead-bismuth follows close to a diffusion-limited M t law in a range of environments and temperatures. 

Fig. 12. A log-linear curve fit to location of minimum silicon 

content during reactor operation at 7008C. 

Calculations were performed to show how moderate 

performance gains achieved by using this FGC would 

affect the performance of a typical lead-bismuth–cooled 

reactor. A 2400-MW~thermal! reactor design by Nikiforova 

et al. 70 was used as a baseline case. An outlet 

temperature increase of 508Cto6508C and flow rates up 

to6m0s were considered. Figure 14 shows the results of 

this comparison. If the reactor designer wishes to maintain 

the same power level while shrinking the core, just 

increasing the outlet temperature shrinks the pitch-todiameter 

~P0D! ratio from 1.30 to 1.17. 

Such a high coolant velocity may not necessarily be 

achievable, depending on changes made to the core geometry 

of an lead-bismuth–cooled reactor. This is due to 

the increased pumping work necessary to force more 

coolant through a core of higher power density and therefore 

a smaller P0D ratio. Therefore, the designers of the 

Fig. 13. Hydrogen production process efficiencies, showing 

the increase in efficiency by using a higher outlet 

temperature of 7008C ~green line! ~color online! as 

compared to the currently accepted limit of 5508C 

~red line! for lead-bismuth reactors. 68 Curves are shown 

for the high-temperature sulfur electrolysis ~HTSE!, 

Westinghouse sulfur process ~WSP!, and the sulfuriodide 

~S-I! processes. 69 

baseline case placed a pumping work limit of 1500 kPa 

on future designs based on current availability of liquidmetal 

pumps. 70 In addition, a power limit of 6000 

MW~thermal! was placed on future designs because of 

the limited capacity of the heat exchangers to remove 

heat from the primary coolant loop. Figure 15 summarizes 

the expanded operating region for future leadbismuth–cooled 

reactor designs as compared to the 

baseline case. 





Fig. 14. Power and P0D ratio performance gains by increasing 

the coolant flow velocity and outlet temperature. The 

baseline design point of 2400 MW~thermal! is shown 

as compared to power contours for the new design. 

The P0D ratio gives an estimate of the size of the core. 

This figure shows how increasing the outlet temperature 

by using the composite in this study can lead to 

the design of a smaller core at the same power level or 

a higher-power core of the same size. 

IV.B. Economic Gains 

A number of factors made possible by this FGC contribute 

to lowering the cost of a lead-cooled fast reactor 

~LFR! plant: 

1. Keeping the power level constant allows for shrinking 

of the core due to a higher power density, 

leading to less materials cost. 

2. Conversely, keeping the core size constant would 

allow for more electricity production out of the 

same core. 

3. Eliminating FAC reduces the risk of reactor material 

degradation. 

4. A smaller core and more passive safety systems 

~due to higher permissible temperatures and power 

densities! shrink the footprint of the plant, saving 

materials and construction costs. 

V. CONCLUSIONS 

Based on this research, the following conclusions 

may be stated: 

1. The FGC developed in this research protects 

against lead-bismuth corrosion in all expected environments, 

both oxidizing and reducing, such that corrosion 

Fig. 15. Increase in operating region based on reasonable design 

constraints for a lead-bismuth-cooled reactor. 

Pumping work, vessel size, pressure drop, heat exchangers, 

and coolant flow velocity were considered 

when restricting the recommended operation region. 

will hopefully no longer be a concern for lead-bismuth– 

cooled systems. Extrapolated corrosion rates based on 

the experiments in this study are ,1 mm0yr, which is 

negligible in terms of reactor design, even assuming a 

60-yr reactor lifetime for structural components. It should 

be noted that further corrosion studies are necessary to 

confirm both long-term corrosion behavior and corrosion 

resistance in flowing lead-bismuth. 

2. The FGC is diffusionally stable. The diffusional 

dilution zone between the two layers will not exceed 

17 mm for fuel cladding ~3-yr life! or 33 mm for coolant 

piping ~60-yr life!, both assumed to operate at 7008C. 

3. Because of the performance gains above, the FGC 

represents a potential enabling technology for leadbismuth–cooled 

reactors and systems.Asteady-state temperature 

increase of up to 1508C beyond the current 

limitation of 5508C is possible, provided that suitable 

structural materials exist. This allows reactor designers 

to increase the power density and0or increase the output 

of their reactors and to include larger safety margins in 

case of an accident. 

4. This FGC is ready for immediate deployment in 

nonirradiated or low-dose applications. The corrosion 

resistance has been demonstrated and will be verified 

pending longer-length experiments. Further work is required 

to investigate the properties of the FGC under 

flowing lead-bismuth and irradiation, especially at temperatures 

below 4508C where mechanical properties can 

be adversely affected by fast neutron irradiation. 30 





However, these results need to be confirmed with longerterm 

tests, along with irradiation performance studies, 

before deployment in high-irradiation scenarios. 

ACKNOWLEDGMENTS 

The authors wish to acknowledge the U.S. Department of 

Energy’s Nuclear Energy Research Initiative program ~award 

DE-FC07-06ID14742! for generous funding throughout this 

project. Special thanks to N. Todreas, B. Yildiz, and A. Nikiforova, 

all from MIT, for very useful discussions concerning 

this project. 

REFERENCES 

1. “A Technology Roadmap for Generation IV Nuclear Energy 

Systems,” GIF-002-00, U.S. Department of Energy Nuclear 

Energy ResearchAdvisory Committee and the Generation 

IV International Forum ~2002!. 

2. “Next Generation Nuclear Plant: A Report to Congress,” 

U.S. Department of Energy Office of Nuclear Energy; 

http:00www.ne.doe.gov0pdfFiles0NGNP_ReporttoCongress_ 

2010.pdf ~current as of Jan. 24, 2011!. 

3. M. ICHIMIYA, T. MIZUNO, and S. KOTAKE, “A Next 

Generation Sodium-Cooled Fast Reactor Concept and Its R&D 

Program,” Nucl. Eng. Technol., 39, 171 ~2007!. 

4. H.A.ABDERRAHIM, V. SOBOLEV, and E. MALAMBU, 

“Fuel Design for the Experimental ADS MYRRHA,” presented 

at International Atomic Energy Agency Technical Mtg. 

Use of LEU in ADS, Vienna, Austria, October 10–12, 2005. 

5. C. STEVENS, “Generation IV International Forum Signs 

Agreement to Collaborate on Sodium Cooled Fast Reactors,” 

Online press release, U.S. Department of Energy ~2006!; http:00 

www.energy.gov0news0archives03218.htm ~current as of Jan. 

24, 2011!. 

6. J. BUONGIORNO and P. E. MacDONALD, “Supercritical 

Water Reactor ~SCWR!: Progress Report for the FY-03 

Generation-IV R&D Activities for the Development of the 

SCWR in the U.S.,” INEEL0EXT-03-01210, Idaho National 

Engineering and Environmental Laboratory ~2003!. 

7. N. F. CYCLE and M. SECTION, “Thorium Fuel Cycle— 

Potential Benefits and Challenges,” IAEA-TECDOC-1450, International 

Atomic Energy Agency ~2005!. 

8. M.-H. KIM, H.-S. LIM, and W.-J. LEE, “A Thermal-Fluid 

Assessment of a Cooled-Vessel Concept for a VHTR,” Nucl. 

Eng. Des., 238, 3360 ~2008!. 

9. N. B. VARGAFTIK, Tables on the Thermophysical Properties 

of Liquids and Gases, Hemisphere Publishing Corporation 

~1975!. 

10. V. SOBOLEV, “Thermophysical Properties of Lead and 

Lead-Bismuth Eutectic,” J. Nucl. Mater., 362, 235 ~2007!. 

11. P. HEJZLAR et al., “Cross-Comparison of Fast Reactor 

Concepts with Various Coolants,” Nucl. Eng. Des., 239, 2672 

~2009!. 

12. P. GIERSZEWSKI, B. MIKIC, and N. TODREAS, “Property 

Correlations for Lithium, Sodium, Helium, FLiBe and 

Water in Fusion Reactor Applications,” PFC-RR-80-12, Massachusetts 

Institute of Technology Plasma Fusion Center ~1980!. 

13. G. H. HOLDEN and J. V. TOKOR, “Thermophysical Properties 

of Sodium,” ANL-7323, Argonne National Laboratory 

~1967!. 

14. INTERNATIONALASSOCIATION FOR THE PROPER- 

TIES OF WATER AND STEAM, Aqueous Systems at Elevated 

Temperatures and Pressures, R. FERNANDEZ-PRINI, A. H. 

HARVEY, D. A. PALMER, Eds., Academic Press ~2004!. 

15. J. J. DUDERSTADT and L. J. HAMILTON, Nuclear Reactor 

Analysis, John Wiley & Sons ~1976!. 

16. R. G. BALLINGER and J. LIM, “An Overview of Corrosion 

Issues for the Design and Operation of High-Temperature 

Lead- and Lead-Bismuth–Cooled Reactor Systems,” Nucl. Technol., 

147, 418 ~2004!. 

17. N. LI, “Active Control of Oxygen in Molten Lead-Bismuth 

Eutectic Systems to Prevent Steel Corrosion and Coolant Contamination,” 

J. Nucl. Mater., 300, 73~2002!. 

18. L. SOLER, F. MARTÍN, F. HERNÁNDEZ, and D. 

GÓMEZ-BRICEÑO, “Corrosion of Stainless Steels in Lead- 

Bismuth Eutectic up to 6008C,” J. Nucl. Mater., 335, 174 ~2004!. 

19. N. POLMAR and K. J. MOORE, Cold War Submarines: 

The Design and Construction of U.S. and Soviet Submarines, 

Brassey’s, Inc. ~2004!. 

20. H. WRIEDT and H. OKAMOTO, “Binary Alloy Phase 

Diagrams,” ASM Handbooks Online Vol. 3: Alloy Phase Diagrams, 

ASM International ~2004!. 

21. L. TAN, M. MACHUT, K. SRIDHARAN, and T. ALLEN, 

“Corrosion Behavior of a Ferritic0Martensitic Steel HCM12A 

Exposed to Harsh Environments,” J. Nucl. Mater., 371, 161 

~2007!. 

22. J. LIM, H. O. NAM, and I. S. HWANG, “Corrosion Test of 

Cr- and Al-Containing Alloys in Static LBE at 5508C,” Proc. 

Int. Congress Advances in Nuclear Power Plants (ICAPP’08), 

Anaheim, California, June 8–12, 2008, p. 632, American Nuclear 

Society ~2008!. 

23. S. TAKAYA et al., “Corrosion Resistance of Al Alloying 

High Cr ODS Steels in Stagnant Lead Bismuth,” J. Nucl. Mater., 

398, 132 ~2010!. 

24. D. SAPUNDJIEV, S. V. DYCK, and W. BOGAERTS, “Liquid 

Metal Corrosion of T91 and A316L Materials in Pb-Bi 

Eutectic at Temperatures 400–6008C,” Corros. Sci., 48, 577 

~2006!. 

25. F. GNECCO, E. RICCI, C. BOTTINO, and A. PASSE- 

RONE, “Corrosion Behaviour of Steels in Lead Bismuth at 

823K,” J. Nucl. Mater., 335, 185 ~2004!. 





26. P. MOINARD and Y. MILLET, “Process for the Production 

of a Tubular Zircaloy 2 Blank Internally Clad with Zirconium 

and Suitable for Ultrasound Monitoring of the Zirconium 

Thickness,” U.S. Patent 5609697 ~1997!. 

27. G. Y. LAI, High-Temperature Corrosion and Materials 

Applications, ASM International ~2007!. 

28. J. ROBERTSON and M. I. MANNING, “Healing Layer 

Formation in Fe-Cr-Si Ferritic Steels. Mater. Sci. Technol., 5, 

741 ~1989!. 

29. M. P. SHORT, “The Design of a Functionally Graded Composite 

for Service in High Temperature Lead and Lead- 

Bismuth Cooled Nuclear Reactors,” PhD Thesis, Massachusetts 

Institute of Technology ~2010!. 

30. R. KLUEH and A. NELSON, “Ferritic0Martensitic Steels 

for Next-Generation Reactors,” J. Nucl. Mater., 371,37~2007!. 

31. T. H. BAUER, G. R. FENSKE, and J. M. KRAMER, “Cladding 

Failure Margins for Metallic Fuel in the Integral Fast 

Reactor,” Structural Materials in Reactor Technology, Vol. C: 

Fuel Elements and Assemblies, p. 31, F. H. WITTMANN, Ed. 

~1987!. 

32. A. L. JOHNSON et al., “Spectroscopic and Microscopic 

Investigation of the Corrosion of 3160316L Stainless Steel by 

Lead-Bismuth Eutectic ~LBE! at Elevated Temperatures: Importance 

of Surface Preparation,” J. Nucl. Mater., 328, 88 

~2004!. 

33. T. FURUKAWA et al., “Corrosion Behavior of FBR Candidate 

Materials in Stagnant Pb-Bi at Elevated Temperature,” 

J. Nucl. Sci. Technol., 41, 6~2004!. 

34. P. N. MARTYNOV and K. D. IVANOV, “Properties of 

Lead-Bismuth Coolant and Perspectives of Non-Electric Applications 

of Lead-Bismuth Reactor,” IAEA-TECDOC-1056, 

International Atomic Energy Agency ~1997!. 

35. R. KLUEH and J. VITEK, “Elevated-Temperature Tensile 

Properties of Irradiated 9Cr-1MoVNb Steel,” J. Nucl. Mater., 

132, 27~1985!. 

36. W. L. HU and D. S. GELLES, “The Ductile-to-Brittle 

Transition Behavior of Martensitic Steels Neutron Irradiated to 

26 dpa,” Influence of Radiation on Material Properties: 13th 

International Symposium (Part II), STP956, p. 83, American 

Society for Testing and Materials ~1987!. 

37. Y. DAI, X. JIA, and K. FARRELL, “Mechanical Properties 

of Modified 9Cr-1Mo ~T91! Irradiated at 3008C in SINQ 

Target-3,” J. Nucl. Mater., 318, 192 ~2003!. 

38. F. H. HUANG, “Fracture Toughness and Tensile Properties 

of Alloy HT9 in Thin Sections Under High Neutron Fluences,” 

Effects of Radiation on Materials: 15th International 

Symposium, p. 1267, STP 1125, American Society for Testing 

and Materials ~1992!. 

39. E. A. LITTLE and D. A. STOW, “Void-Swelling in Irons 

and Ferritic Steels: II. An Experimental Survey of Materials 

Irradiated in a Fast Reactor,” J. Nucl. Mater., 87, 25~1979!. 

40. R. L. KLUEH, D. J. ALEXANDER, and E. A. KENIK, 

“Development of Low-Chromium, Chromium-Tungsten Steels 

for Fusion,” J. Nucl. Mater., 227, 11~1995!. 

41. K. SUZUKI, “Two-Phase Separation of Primary MX Carbonitride 

During Tempering in Creep Resistant 9Cr1MoVNb 

Steel,” ISIJ Int., 43, 1089 ~2003!. 

42. F. A. GARNER, M. B. TOLOCZKO, and B. H. SENCER, 

“Comparison of Swelling and Irradiation Creep Behavior of 

fcc-Austenitic and bcc-Ferritic0MartensiticAlloys at High Neutron 

Exposure,” J. Nucl. Mater., 276, 123 ~2000!. 

43. R. E. STOLLER, A. S. KUMAR, and D. S. GELLES, Eds., 

Effects of Radiation on Materials: 15th International Symposium, 

STP 1125, p. 818, American Society for Testing and 

Materials ~1992!. 

44. G. BONNY, D. TERENTYEV, and L. MALERBA, “On 

the Miscibility Gap of FeCr Alloys,” Scr. Mater., 59, 1193 

~2008!. 

45. A. M. BABAKR et al., “Sigma Phase Formation and Embrittlement 

of Cast Iron-Chromium-Nickel ~Fe-Cr-Ni! Alloys,” 

J. Miner. Mater. Charact. Eng., 7, 127 ~2008!. 

46. J. GLOWNIA, B. KALANDYK, and K. HÜBNER, “Delta 

Ferrite Predictions for Cast Duplex Steels with High Nitrogen 

Content,” Mater. Charact., 47, 149 ~2001!. 

47. G. E. TOTTEN, K. FUNATANI, and L. XIE, Handbook of 

Metallurgical Process Design, CRC Press, Boca Raton, Florida 

~2004!. 

48. J. Y. LIM, M. P. SHORT, R. B. BALLINGER, and N. 

LI, “Corrosion Studies of Fe-Cr-Si Alloys for Service in 

Lead-Bismuth Eutectic Systems up to 7008C, Part I: Initial 

Alloy Development,” J. Nucl. Mater. ~2011! ~submitted for 

publication!. 

49. G. MÜLLER et al., “Behavior of Steels in Flowing Liquid 

PbBi Eutectic Alloy at 420–6008C After 4000–7200 h,” J. 

Nucl. Mater., 335, 163 ~2004!. 

50. L. MARTINELLI and F. BALBAUD-CÉLÉRIER, “Modelling 

of the Oxide Scale Formation on Fe-Cr Steel During 

Exposure in Liquid Lead-Bismuth Eutectic in the 450–6008C 

Temperature Range,” Mater. Corros., 62, 531 ~2011!. 

51. G. MÜLLER, G. SCHUMACHER, and F. ZIMMER- 

MANN, “Investigation on Oxygen Controlled Liquid Lead Corrosion 

of Surface Treated Steels,” J. Nucl. Mater., 278, 85 

~2000!. 

52. A. R. NIKOLAEV, A. A. KORTEL, V. P. KARASEV, and 

V. I. IKONNIKOV, “Reaction of Periclase and Corundum Refractories 

with Killed Steel,” Refract. Ind. Ceram., 21, 635 

~1980!. 

53. C. T. LIU, E. H. LEE, and C. G. McKAMEY, “An Environmental 

Effect as the Major Cause for Room-Temperature 

Embrittlement in FeAl,” Scr. Metall., 23, 6, 875 ~1989!. 





54. K. D. ADAMS and J. N. DUPONT, “Influence of Ti and C 

on the Solidification Microstructure of Fe-10Al-5Cr Alloys,” 

Metall. Mater. Trans. A, 41A, 194 ~2010!. 

55. J. N. DuPONT, J. R. REGINA, and K. ADAMS, “Improving 

the Weldability of FeCrAl Weld Overlay Coatings,” Proc. 

21st Conf. Fossil Energy Materials, Knoxville, Tennessee,April 

30–May 2, 2007, p. 131 ~2007!. 

56. J. R. REGINA, J. N. DuPONT, and A. R. MARDER, “The 

Effect of Chromium on the Weldability and Microstructure of 

Fe-Cr-Al Weld Cladding,” Weld. Res., 86, 170s ~2007!. 

57. D. A. JONES, Principles and Prevention of Corrosion, 

2nd ed., Prentice Hall ~1996!. 

58. J. Y. LIM, “Effects of Chromium and Silicon on Corrosion 

of Iron Alloys in Lead-Bismuth Eutectic,” PhD Thesis, Massachusetts 

Institute of Technology ~2006!. 

59. M. K. MILLER, K. F. RUSSELL, J. KOCIK, and E. 

KEILOVA, “Embrittlement of Low Copper VVER 440 Surveillance 

Samples Neutron-Irradiated to High Fluences,” J. 

Nucl. Mater., 282, 83~2000!. 

60. B. ESMAILZADEH, A. KUMAR, and F. A. GARNER, 

“The Influence of Silicon on Void Nucleation in Irradiated 

Alloys,” J. Nucl. Mater., 133–134, 590 ~1985!. 

61. J. VAN DEN BOSCH, P. HOSEMANN, A. ALMAZOUZI, 

and S. MALOY, “Liquid Metal Embrittlement of Silicon Enriched 

Steel for Nuclear Applications,” J. Nucl. Mater., 398, 

116 ~2010!. 

62. F. BARBIER and A. RUSANOV, “Corrosion Behavior of 

Steels in Flowing Lead-Bismuth,” J. Nucl. Mater., 296, 231 

~2001!. 

63. A. WEISENBURGER, A. HEINZEL, G. MÜLLER, H. 

MUSCHER, and A. ROUSANOV, “T91 Cladding Tubes with 

and Without Modified FeCrAlY Coatings Exposed in LBE at 

Different Flow, Stress and Temperature Conditions,” J. Nucl. 

Mater., 376, 274 ~2008!. 

64. V. ENGELKO, B. YATSENKO, G. MÜLLER, and H. 

BLUHM, “Pulsed Electron Beam Facility ~GESA! for Surface 

Treatment of Materials,” Vacuum, 62, 211 ~2001!. 

65. M. P. SHORT and R. G. BALLINGER, “A Functionally 

Graded Composite for Service in High-Temperature Lead- and 

Lead-Bismuth–Cooled Nuclear Reactors—II: Processing,” Nucl. 

Technol. ~to be published!. 

66. G. GUNTZ et al., The T91 Book, Vallourec Industries, 

France ~1990!. 

67. M. KAZIMI, E. MONIZ, and C. FORSBERG, The Future 

of the Nuclear Fuel Cycle, Massachusetts Institute of Technology 

~2010!. 

68. B. YILDIZ and M. S. KAZIMI, “Efficiency of Hydrogen 

Production Systems Using Alternative Nuclear Energy Technologies,” 

Int. J. Hydrogen Energy, 31, 77~2006!. 

69. C. FORSBERG, B. BISCHOFF, L. K. MANSUR, L. 

TROWBRIDGE, and P. TORTORELLI, “ALower-Temperature 

Iodine-Westinghouse-Ispra Sulfur Process for Thermochemical 

Production of Hydrogen,” Proc. Int. Conf. Atoms for Prosperity: 

Updating Eisenhower’s Global Vision for Nuclear Energy 

(Global 2003), New Orleans, Louisiana, November 16–20, 2003, 

p. 1478, American Nuclear Society ~2003!. 

70. A. NIKIFOROVA, P. HEJZLAR, and N. E. TODREAS, 

“Lead-Cooled Flexible Conversion Ratio Fast Reactor,” Nucl. 

Eng. Des., 239, 2596 ~2009!. 

71. M. SCHÜTZE, Protective Oxide Scales and Their Breakdown, 

John Wiley & Sons ~1997!.

A Functionally Graded Composite for Service in High-Temperature Lead- and Lead-Bismuth–Cooled Nuclear Reactors—I_ Design

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?