Available online at [Link].

com

Acta Materialia 61 (2013) 735–758

[Link]/locate/actamat

Materials challenges in nuclear energy

S.J. Zinkle a,⇑, G.S. Was b
a
Oak Ridge National Laboratory, P.O. Box 2008, Oak Ridge, TN 37831, USA
b
Nuclear Engineering and Radiological Sciences Department, University of Michigan, Ann Arbor, MI 48109, USA

Abstract

Nuclear power currently provides about 13% of electrical power worldwide, and has emerged as a reliable baseload source of elec-
tricity. A number of materials challenges must be successfully resolved for nuclear energy to continue to make further improvements
in reliability, safety and economics. The operating environment for materials in current and proposed future nuclear energy systems
is summarized, along with a description of materials used for the main operating components. Materials challenges associated with
power uprates and extensions of the operating lifetimes of reactors are described. The three major materials challenges for the current
and next generation of water-cooled fission reactors are centered on two structural materials aging degradation issues (corrosion and
stress corrosion cracking of structural materials and neutron-induced embrittlement of reactor pressure vessels), along with improved
fuel system reliability and accident tolerance issues. The major corrosion and stress corrosion cracking degradation mechanisms for
light-water reactors are reviewed. The materials degradation issues for the Zr alloy-clad UO2 fuel system currently utilized in the majority
of commercial nuclear power plants are discussed for normal and off-normal operating conditions. Looking to proposed future (Gen-
eration IV) fission and fusion energy systems, there are five key bulk radiation degradation effects (low temperature radiation hardening
and embrittlement; radiation-induced and -modified solute segregation and phase stability; irradiation creep; void swelling; and high-
temperature helium embrittlement) and a multitude of corrosion and stress corrosion cracking effects (including irradiation-assisted phe-
nomena) that can have a major impact on the performance of structural materials.
Ó 2012 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

Keywords: Nuclear materials; Radiation effects; Stress corrosion cracking; Structural alloys (steels and nickel base); Nuclear fuels

1. Introduction electricity. These provided just under 20% of the nation’s

total electric energy generation and more than 30%
Access to reliable, sustainable and affordable energy is of worldwide nuclear generating capacity. Worldwide,
viewed as crucial to worldwide economic prosperity and nuclear energy provides about 13% of the electrical
stability [1,2]. Nuclear fission energy has emerged over demand [1]. Given that nuclear power has very low carbon
the past 40 years to become a reliable baseload source of emission [2] and that energy generation currently accounts
clean and economical electrical energy. As of 2011, there for 66% of worldwide greenhouse gas emissions [4], nuclear
were 435 nuclear reactors in operation worldwide, produc- energy is considered an important resource in managing
ing 370 GWe of electricity [3]. Another 108 units or atmospheric greenhouse gases and associated climate
108 GWe are forthcoming (under construction or on change [1].
order), for a total of 543 units and 478 GWe of electrical The core of a nuclear reactor presents an exceptionally
capacity. The largest producer of power from nuclear harsh environment for materials due to the combination
energy is the USA, with 104 commercial reactors licensed of high temperature, high stresses, a chemically aggressive
to operate at 65 sites, producing a total of 103 GWe of coolant and intense radiation fluxes. Many of the features
that make reactors attractive from a physics perspective
⇑ Corresponding author. Tel.: +1 865 576 5785. (e.g. high specific power, self-sustaining reaction) exert high
E-mail address: zinklesj@[Link] (S.J. Zinkle). operational burdens on structural materials. For example,

1359-6454/$36.00 Ó 2012 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.
[Link]
736 S.J. Zinkle, G.S. Was / Acta Materialia 61 (2013) 735–758

the recoverable energy from each 235U fission reaction is ment in economics and safety, such as passive heat removal
200 MeV, which is about eight orders of magnitude per systems. There are a total of 108 Generation III and
atom higher than typical chemical reactions. As a result, Generation III+ reactors on order or under construction
typical power densities in commercial nuclear reactor cores around the world, and of those, 89 are PWRs.
are 50–75 MWth m3, which is nearly two orders of mag- Given the high representation of PWRs and BWRs in
nitude higher than the average power density in the boiler the world’s fleet, materials issues in these two types of reac-
furnace of a large-scale coal power plant. This intense pro- tors are of greatest interest. And of the many materials in a
duction of heat is accompanied by the generation of ener- reactor, those that experience the most extreme conditions
getic neutrons (which serve to sustain the fission reaction) (stress, corrosion, and radiation) are most important for
and gamma radiation, which can degrade materials by dis- maintaining plant safety and reliability. Fig. 1 shows a
placement damage and radiolysis processes, respectively. schematic of the major components in the primary and
Recent activities to extend the operating lifetime of current secondary circuits of a PWR [10]. Pressurized water
water reactors, to develop advanced fission reactor con- (15.5 MPa) in the primary circuit enters the reactor core
cepts with greater functionality and capability, and the at 275 °C, picks up heat from the reactor core with a core
coming emergence of fusion energy represent even greater exit temperature of 325 °C, and transfers the heat across
demands on materials [5–8]. the U-tubes in the steam generator to water at a lower
pressure. This water turns to steam that powers the tur-
1.1. Types of nuclear fission reactors bine, and is condensed and recirculated. Fig. 1 also lists
the alloys used throughout the primary and secondary cir-
The predominant reactor design worldwide is the pres- cuits, all of which are in contact with high-temperature
surized water reactor (PWR), accounting for two-thirds water and are subject to significant mechanical stress.
of the installed capacity, followed by boiling water reactors Alloys inside (and including) the reactor vessel are also
(BWRs) at 21% and heavy-water reactors at 14% of subject to varying levels of radiation, which produces
installed capacity, respectively (Table 1) [3]. All of these displacement damage and radiolytic decomposition of the
water-cooled reactors use ceramic fuel pellets consisting coolant water. Major pressure boundary components
of UO2 or other fissile actinide oxides to generate heat. (reactor pressure vessel, pressurizer, steam generator,
The ceramic pellets are stacked inside of long Zr alloy tubes steam lines, turbine and condenser) are made of either
(fuel cladding) that transfer the nuclear heat to flowing low carbon or low alloy steel. Austenitic stainless steels
water coolant and serve as the primary barrier containing (Types 304, 304L, 316, 316L, 321, 347) dominate the core
the volatile radioactive fission byproducts. The remaining structural materials, as well as serving for cladding (308SS
5% of installed nuclear energy comes from gas-cooled reac- and 309SS) on the inside surface of the reactor pressure
tors, graphite-moderated reactors and liquid metal cooled vessel and pressurizer. Higher strength components such
reactors (Table 1). as springs and fasteners are made of nickel-base alloys.
The vast majority of the reactors listed in Table 1 are Vessel penetrations and steam generator tubes are made
classified as Generation II reactors [9], which were designed of nickel-base alloy 690 (previously alloy 600, which was
in the 1960s and predominantly achieved initial commercial found to provide insufficient resistance to stress corrosion
operation from the 1970s through the 1990s. These reactors cracking). Condenser tubes are generally made of titanium
are distinguished from Generation I designs (1950s-60s), or stainless steel. The selection of nickel-base alloys and
which were early commercial prototype and demonstration austenitic stainless steels for core internals and the steam
reactors, and Generation III reactors, designed in the 1990s generator tubes is driven by the need for good aqueous
to incorporate significant advances in safety and economics corrosion resistance at high temperatures. These alloys
[9]. Generation III reactor construction for the past decade have low corrosion rates due to the formation of chro-
has been centered in Asia, with a few units recently built in mium-bearing spinels that form adherent, high-density
Europe. The current generation of light-water reactors protective surface layers that grow very slowly at operating
(LWRs), Generation III+, include still further advance- temperatures.

Table 1
Power reactors by type, worldwide [3].
Reactor type # Units Net MWe # Units Net MWe # Units Net MWe
(in operation) (forthcoming) (total)
Pressurized light-water reactors (PWR) 267 246555.1 89 93,014 356 339569.1
Boiling light-water reactors (BWR) 84 78320.6 6 8056 90 86376.6
Gas-cooled reactors, all models 17 8732.0 1 200 18 8932.0
Heavy-water reactors, all models 51 25610.0 8 5112 59 30722.0
Graphite-moderated reactors, all models 15 10219.0 0 0 15 10219.0
Liquid-metal-cooled reactors, all models 1 560.0 4 1016 5 2076.0
Totals 435 369996.7 108 107,896 543 477894.7
 S.J. Zinkle, G.S. Was / Acta Materialia 61 (2013) 735–758 737

Fig. 1. Schematic of the primary and secondary circuits of a pressurized water reactor and materials of construction (courtesy of R.W. Staehle) [10].

The main difference between PWRs and BWRs is that out the water circuit in both reactor types, as will be
the latter consists of a single water circuit designed for boil- discussed in Section 2.1.
ing to occur in the core with steam flowing directly to the The last reactor design that is in significant use world-
turbine, which eliminates the steam generator and pressur- wide is the pressurized heavy water reactor (PHWR), the
izer found in the PWR. The operating temperatures are most prevalent version being the CANDU (CANadian
comparable for both reactor types (300 °C), with compa- Deuterium Uranium) reactor. This reactor uses heavy
rable stress and radiation environments. As such, most of water as the moderator and primary coolant, transferring
the structural alloys are very similar between the two reac- heat to light water via a steam generator. The key charac-
tor types. The main difference is in the zirconium alloys teristic of this reactor is the use of deuterium as a modera-
used as fuel rod cladding, with BWR fuel cladding opti- tor, for which neutron absorption is low enough to permit
mized for corrosion resistance in higher oxygen potentials the use of natural (unenriched) uranium, thus bypassing
and PWR fuel cladding optimized for resistance to hydro- the need for expensive enrichment facilities. A major differ-
gen absorption in the low potential environment of the ence in materials in this system vs. LWRs is the use of Zr–
core. Typical zirconium alloy cladding materials used in Nb pressure tubes that house the Zircaloy-clad fuel and the
BWR and PWR reactors are summarized in Table 2. high pressure D2O. These tubes fit into Zircaloy-4 calan-
Differences in oxygen potential result in significant impacts dria tubes that pass through a thin walled stainless steel
on the stress corrosion degradation of materials through- calandria vessel, which also contains the low temperature

Table 2
Summary of typical commercial zirconium alloys used as cladding in PWRs and BWRs.
Reactor type Zr alloy composition Thermomechanical treatment
BWR Zircaloy-2 (1.5% Sn–0.15% Fe–0.1% Cr–0.05% Ni) Recrystallized
PWR Zircaloy-4 (1.5% Sn–0.2% Fe–0.1% Cr) Cold-worked and stress relief anneal
PWR ZIRLO (1–2% Nb–1% Sn–0.1% Fe) Quench and temper/stress relief anneal
PWR M5 (1% Nb) Recrystallized
738 S.J. Zinkle, G.S. Was / Acta Materialia 61 (2013) 735–758

D2O moderator. Thus, zirconium alloys play a larger role There are five key bulk radiation degradation effects
as pressure boundary materials in PHWRs than they do (low temperature radiation hardening and embrittlement;
in LWRs. radiation-induced and -modified solute segregation and
Most reactors in the USA and elsewhere in the world phase stability (including amorphization); irradiation
were completed in the 1970s and 1980s, and today the aver- creep; void swelling; and high-temperature helium embrit-
age age of the fleet is over 30 years. Fig. 2 shows the world- tlement) [8,12–16], and a multitude of corrosion and stress
wide distribution of nuclear power plants classified by corrosion cracking effects in water-cooled reactors [13,17–
years of commercial operation [11]. Since the original 22] and proposed advanced reactors utilizing other cool-
license period in the USA is 40 years, many reactor opera- ants [23–26] (including irradiation-assisted phenomena)
tors are seeking license renewal to allow them to operate that can have a huge impact on the performance of struc-
the plants for an additional 20 years. To date, 73 of the tural materials in nuclear energy systems. The amount of
104 operating commercial reactors in the USA have radiation damage produced in materials from exposure to
received license extensions with another 13 applications neutrons created by the nuclear energy reactions is quanti-
under review, and a key question is how long can these fied by the international standardized parameter [27,28] of
plants be safely, reliably and economically operated. The displacements per atom (dpa); a displacement damage
limiting factor is whether critical materials can continue value of 1 dpa means that, on average, each atom has been
to maintain their integrity beyond 60 years [5]. These mate- displaced from its lattice site once.
rials include reactor components, concrete, cables and bur- Neutron irradiation can produce pronounced hardening
ied piping. So the lifetime of the current reactor fleet is at low and intermediate irradiation temperatures due the
ultimately governed by the performance of materials. production of high densities of nanoscale defect clusters
(dislocation loops, helium bubbles, etc.), which serve as
1.2. Major materials degradation modes in nuclear energy obstacles to dislocation motion. This hardening is generally
systems accompanied by a reduction in tensile elongation and frac-
ture toughness. The radiation hardening and reductions in
In addition to satisfying standard materials design crite- elongation and fracture toughness typically emerge at dam-
ria based on tensile properties, thermal creep, cyclic fatigue age levels above 0.1 dpa and are generally most pro-
and creep-fatigue, structural materials for current and pro- nounced for homologous irradiation temperatures below
posed future nuclear energy systems must provide adequate 0.35TM, where TM is the absolute melting temperature
resistance to two additional overarching environmental [26,29–35]. Fig. 3 shows an example of the effect of moder-
degradation phenomena: radiation damage and chemical ate neutron displacement damage levels on the engineering
compatibility. Since the chemical compatibility issues (cor- stress–strain curve for austenitic stainless steel [36] and a 8–
rosion, stress corrosion cracking, etc.) are largely depen- 9% Cr-tempered martensitic steel [35] at 250 °C. Both
dent on the specific coolant and engineering application, materials exhibit significant radiation-induced increases in
these issues are discussed in the relevant sections on light- yield and ultimate tensile stress, large reductions in elonga-
water reactors (2.2.2) and advanced reactor concepts (3.2 tion (particularly uniform elongation) and decreased strain
and 4). hardening capacity. The reductions in elongation and
strain hardening capacity have been attributed to flow
localization (e.g. dislocation channeling) [37–44] and strain
hardening exhaustion [29–31] mechanisms. In addition to
the decreased elongation, neutron irradiation at low tem-
perature also generally produces a decrease in fracture
toughness. Fig. 4 summarizes some of the fracture tough-
ness data for Types 304 and 316 austenitic stainless steels
following irradiation at LWR-relevant conditions near
250–350 °C [32,36,45–48]. The fracture toughness decreases
rapidly with increasing irradiation dose, and approaches a
value near 50 MPa m1/2 after 5–10 dpa. The reduction in
fracture toughness can be of particular concern for body-
centered cubic materials such as ferritic/martensitic steels
if the ductile to brittle transition temperature is shifted to
temperatures above cold or warm standby temperatures.
The potential for neutron radiation-induced embrittlement
of reactor pressure vessel steels has been intensively inves-
tigated due to its importance for public safety [49].
At intermediate temperatures (homologous tempera-
Fig. 2. Age distribution of the world’s commercial nuclear power reactors tures >0.3TM), the increased mobility of the radiation
as of December 2011 [11]. defects produces a diverse range of potential microstruc-
 S.J. Zinkle, G.S. Was / Acta Materialia 61 (2013) 735–758 739

Fig. 4. Fracture toughness of Types 304 and 316 austenitic stainless steels
following irradiation at LWR-relevant conditions near 250–350 °C
[32,36,45–48].

Fig. 3. Effect of neutron irradiation to 3 dpa on the engineering stress–

strain curves for (a) solution annealed Type 316LN austenitic steel and (b)
8–9% Cr-tempered martensitic steel at 250 °C (based on Refs. [36] and
[35], respectively).

tural evolutions. The most important radiation degrada-

tion phenomena at intermediate temperatures are
radiation-induced solute segregation (and associated radia-
tion-induced or -modified precipitation), void swelling,
irradiation creep and anisotropic growth. Fig. 5 summa-
rizes the numerous radiation-induced phases that can be
induced in initially single-phase austenitic stainless steel
as a result of localized radiation-induced solute segregation Fig. 5. Precipitate phases observed in Type 316 austenitic stainless steel
processes during neutron irradiation [50]. Initial investiga- after neutron irradiation as a function of temperature and dose. Partially
tions indicated that radiation-induced precipitation was shaded data points at temperatures <400 °C denote the presence of c0
limited to temperatures above 400 °C [51,52], but recent phase and solid data points are for either G and related phases or an
unidentified phase [50].
long-term experiments have observed radiation-induced
precipitation in austenitic stainless steel for temperatures
as low as 300 °C [50]. Void swelling (due to nucleation transient regime at low doses (during the void nucleation
and growth of the supersaturation of vacancies produced and initial growth phase), followed by a steady-state swell-
by irradiation) is characterized by an initial low-swelling ing regime where the volumetric swelling increase is pro-
740 S.J. Zinkle, G.S. Was / Acta Materialia 61 (2013) 735–758

portional to the dose [12,16,53–55]. Typical post-transient ing temperature of materials in nuclear energy systems to
steady-state swelling rates in irradiated metals are 0.2– temperatures significantly lower than what would be estab-
1% dpa1, which would produce unacceptable volumetric lished by thermal creep strength considerations.
swelling in structural components exposed to high neutron
doses. Therefore, research has focused on identifying mech- 2. Materials challenges in current commercial fission
anisms that extend the low-swelling transient regime and reactors
delay the onset of the steady-state swelling regime [55,56].
Irradiation creep [12,53,57–60] and irradiation growth 2.1. Operating environment for materials in existing LWRs
[58–61] can cause substantial dimensional changes in addi-
tion to changes due to void swelling. Irradiation growth is Materials in LWRs are exposed to a variety of condi-
mainly an issue in anisotropic crystallographic systems, tions. In the following, the operating environment for nor-
such as hexagonal close-packed materials; for this phenom- mal, extended life and transient conditions are
enon, volume is conserved but pronounced anisotropic summarized. Used fuel disposition issues, while important,
expansion in one crystallographic direction (and shrinkage are not discussed in this paper.
in another direction) can occur due to preferential nucle-
ation of defect clusters, such as dislocation loops, on cer- 2.1.1. LWR materials under normal operating conditions
tain crystallographic habit planes. Materials for nuclear Core materials include both fuel materials and structural
energy systems that exhibit irradiation growth include components. The fuel consists of UO2 pellets in the shape
graphite and pure metals or alloys based on zirconium of right circular cylinders with length and diameter of
and beryllium. The amount of deformation from irradia- approximately 1 cm each, loaded into 3–4 m long zirco-
tion creep is typically proportional to the applied stress nium alloy fuel tubes (cladding), which are grouped into
and irradiation exposure, with a steady-state creep compli- fuel assemblies containing control rods or blades. In
ance coefficient of 0.5 to 1  106 MPa1 dpa1 for ferritic BWRs, the assemblies generally contain approximately
and austenitic steels, respectively [12]. Another conse- 100 fuel rods arranged in a square array. Each BWR
quence of irradiation creep is that it can induce undesirable assembly is encased in a square zirconium alloy tube or fuel
stress relaxation of bolts or springs. Fig. 6 shows the mea- channel approximately 12 cm on a side. The control rods
sured stress relaxation for neutron irradiated Inconel X750 are clustered into cruciform-shaped control blades that
springs [62]. Nearly complete relaxation of the initially pass between assemblies clustered in groups of four, as
applied stress on the springs occurred after an irradiation illustrated in Fig. 7. There are typically 700–800 fuel assem-
dose of 20 dpa at 400 °C. blies in a BWR. PWRs contain fewer (200), but larger
At high temperatures (above 0.5–0.6TM) the efficient (21 cm on a side) assemblies, containing up to 300 fuel rods
annealing of lattice defects produces recovery of most of of slightly smaller diameter than those in BWRs. The PWR
the radiation damage. One notable exception is associated
with the transmutant He produced from (n, a) reactions
within the material. The helium can diffuse to grain bound-
aries, where it can form large bubbles that weaken the grain
boundary strength and cause dramatic reductions in the
total elongation [63–66]. This phenomenon of high-temper-
ature helium embrittlement may restrict the upper operat-

1.2
Inconel X750
1 375-415°C
-8 -8
3x10 to 7x10 dpa/s
Stress Relaxation

0.8

0.6

0.4

0.2

0
0 5 10 15 20 25
Dose (dpa)

Fig. 6. Stress relaxation (normalized to the initial applied stress) for Fig. 7. Fuel assemblies and control blade used in a boiling water reactor
Inconel X750 springs irradiated in the EBR-II fast fission reactor [62]. (image credit GE).
 S.J. Zinkle, G.S. Was / Acta Materialia 61 (2013) 735–758 741

control rods are distributed throughout the square lattice ic stainless steel or nickel-base nozzles in the reactor head
and are connected to each other to form a control rod clus- (PWR) or bottom (BWR), and the RPV.
ter, as shown in Fig. 8. In both cases, control rods consist The RPV serves as both a pressure barrier and a con-
of stainless steel tubes filled with boron carbide for neutron tainment barrier for the radioactive fission products pro-
absorption. There is no channel box around a PWR assem- duced during the nuclear fission reaction, and thus plays
bly, so cross-flow of water between assemblies is possible. a key role in reactor safety. The RPV is typically con-
The PWR and BWR cores are typically operated nonstop structed from carbon and low alloy ferritic steels with 1–
for 18–24 months between refueling operations. Low and 2% Mn, 0.5–1% Ni, 0.5% Mo and 0.15–0.4% Si [67],
intermediate burn-up fuel assemblies are typically moved and has a typical wall thickness of 20 cm. Older LWR
to different positions in the core during refueling outages pressure vessels were constructed from rolled plates that
to provide optimized fuel management, with a total core were welded to form a large cylinder, whereas newer vessels
residence period of 3–4 fuel cycles (i.e. a third to a quarter are formed from ring forgings in order to eliminate welds in
of the fuel assemblies are removed each refueling cycle) the vessel “beltline” region closest to the center of the reac-
until they achieve typical cumulative burn-up levels of tor core. The top and bottom heads are usually constructed
40–60 GW days per metric ton of uranium (GWd from low alloy steel forgings, and are welded to the central
MTU1), corresponding to fissions in 4.2–6.4% of the cylindrical vessel (or bolted and gasketed in the case of the
original uranium atoms. upper head). The internal surface of the RPV is typically
Non-fuel core components consist of major structures clad with 5–10 mm of an austenitic stainless steel to pro-
such as the core shroud (BWR) or the baffle–former assem- vide corrosion compatibility with the reactor coolant. Mul-
bly (PWR), and smaller components such as bolts, springs, tiple penetrations for coolant flow and instrumentation are
support pins and clips. The core shroud in a BWR is a made through the pressure vessel. The fast neutron flux is
cylindrical barrel, open at both ends, that surrounds the three to four orders of magnitude lower at the RPV com-
fuel assemblies. Water from the condenser mixes with pared to core internal structures [67], but it is still of suffi-
water recirculated from the core between the shroud near cient intensity to cause radiation hardening, which could
the top of the vessel and is channeled down the annulus lead to fracture toughness embrittlement. It is important
formed by the shroud and the reactor pressure vessel to maintain adequate levels of fracture toughness for a
(RPV), then up through the fuel assemblies. In a PWR, wide variety of operational conditions, including normal
the baffle–former assembly plays a similar role in forcing operation, cold shutdown for refueling and other mainte-
the incoming water down the annulus formed between nance, and postulated transient accident scenarios such as
the baffle–former assembly and the RPV inner diameter pressurized thermal shock in PWRs that would introduce
and up through the fuel channels to remove heat from fis- cold water into the reactor vessel while the vessel is at oper-
sion. Moving out from the core, additional key compo- ating pressure and temperature (creating large thermal
nents are the control rod drive mechanisms and housing, stresses and potential for crack propagation).
and vessel head penetrations consisting of welded austenit- Materials utilized in LWR cores must withstand simul-
taneous application of mechanical stress, neutron irradia-
tion and corrosion due to hot water or steam (see rows 1
and 2 of Table 3). Temperatures of core components are
in the range 275–288 °C in BWRs and 290–320 °C in
PWRs. The BWR environment is characterized by an elec-
trochemical potential (ECP) in the range of 150 mV rela-
tive to the standard hydrogen electrode, or 150 mVSHE,
due to a combination of boiling of the water in the core
and radiolysis. PWRs operate at a lower potential
(<500 mVSHE) by virtue of the addition of hydrogen at
a level of 35 cc of H2 per kg of water (3 ppm) to scavenge
radiolysis products and lower the corrosion potential.
PWR primary water also contains 1000 ppm B as boric
acid (H3BO3) added for reactivity control and 2–4 ppm
Li as LiOH added for pH control. The lower ECP is better
for both corrosion and stress corrosion cracking of core
materials and is possible in a PWR due to the lack of boil-
ing. In addition to controlling pH, boron impacts the for-
mation of solid corrosion products (CRUD) and
corrosion of fuel cladding, as well as reactivity control of
Fig. 8. Fuel assembly and control rod cluster used in a pressurized water the reactor, and is briefly described in Section 2.2. Stresses
reactor (image credits: Commissariat à l0 énergie atomique and on fuel and core components come from a variety of
Westinghouse). sources, including thermal expansion, high velocity water
 742
Table 3
Reactor core environment and materials for light water reactors and advanced fission reactor concepts [7].
System Coolant Pressure Tin/ Neutron Fuel Cladding Structural materials
(MPa) Tout spectrum,
In-core Out-of-core
(°C) maximum dose
(dpa)
Pressurized Water – single 16 290/ Thermal, 80 UO2 (or Zirconium alloy Stainless steels, nickel- Stainless steels, nickel-based alloys
water reactor phase 320 MOX) based alloys

S.J. Zinkle, G.S. Was / Acta Materialia 61 (2013) 735–758

– PWR
Boiling water Water – two 7 280/ Thermal, 7 UO2 (or Zircaloy Stainless steels, nickel- Stainless steels, nickel-based alloys
reactor – phase 288 MOX) based alloys
BWR
Supercritical Supercritical 25 290/ Thermal, 30, UO2 F-M (12Cr, 9Cr, etc.) (Fe– Same as cladding options, F-M, low-alloy steels
water cooled water 600 fast, 70 35Ni–25Cr–0.3Ti), Incoloy plus low swelling stainless
reactor – 800, ODS, Inconel 690, 625, steels
SCWR and 718
Very high Helium 7 600/ Thermal, <20 UO2, UCO SiC or ZrC coating and Graphites, PyC, SiC, ZrC, Ni-based superalloys, 32Ni–25Cr–
temperature 1000 surrounding graphite vessel: F-M 20Fe–12.5W–0.05C, Ni–23Cr–18W–
reactor – 0.2C, F-M w/thermal barriers, low-
VHTR alloy steels
Gas fast reactor Helium, 7 450/ Fast, 80 MC Ceramic Refractory metals and Ni-based superalloys, 32Ni–25Cr–
– GFR supercritical 850 alloys, Ceramics, ODS, 20Fe–12.5W–0.05C, Ni–23Cr–18W–
CO2 vessel: F-M 0.2C, F-M w/therm barriers
Sodium fast Sodium 0.1 370/ Fast, 200 MOX or U– F-M or F-M ODS F-M ducts, 316SS grid Ferritics, austenitics
reactor – 550 Pu–Zr or plate
SFR MC or MN
Lead fast Lead or lead– 0.1 600/ Fast, 150 MN High-Si F-M, ODS, High-Si austenitics, ceramics, or
reactor – bismuth 800 ceramics, or refractory alloys refractory alloys
LFR
Molten salt Molten salt, for 0.1 700/ Thermal, 200 Salt Not applicable Ceramics, refractory High-Mo, Ni-based alloys (e.g.,
reactor – example: 1000 metals, Mo, Ni-alloys, INOR-8)
MSR FLiNaK (e.g., INOR-8), graphite,
Hastelloy N
Abbreviations: F-M, Ferritic–martensitic stainless steels (typically 9–12 wt.% Cr); ODS, oxide dispersion-strengthened steels (typically ferritic–martensitic); MC, mixed carbide (U,Pu)C; MN, mixed
nitride (U,Pu)N; MOX, mixed oxide (U,Pu)O2.
 S.J. Zinkle, G.S. Was / Acta Materialia 61 (2013) 735–758 743

flow, residual stresses due to welding, and stresses due to ator. Historically, the main materials degradation prob-
radiation-induced volume expansion or distortion. lems have been intergranular stress corrosion cracking
The unique environmental element of a reactor core is (IGSCC) of BWR piping and of steam generator tubes
radiation. Fission results in several different types of radia- and in vessel penetrations in PWRs. IGSCC of BWR pipes
tion that affect materials in different ways. Principal radia- and steam lines made of 304 stainless steel was due to a
tion types contributing to material degradation are the combination of weld-induced residual stresses and sensiti-
fission products, neutrons and gamma rays. Fission prod- zation of grain boundaries caused by heat treatment of
ucts consist of high-energy (100 MeV) elements of sizable high carbon steels in the temperature region in which chro-
mass (generally between 90 and 150 atomic mass units) that mium carbides formed rapidly on the boundaries, depleting
result directly from the fission process. These elements are them of chromium and making them susceptible to attack.
created as highly charged ions that deposit their energy IGSCC of Alloy 600 occurred in steam generators, on both
within 10 lm of their origin. As such, except for those that the primary and secondary sides, and was driven by a sus-
are born within this distance of the fuel pellet surface, fis- ceptible microstructure and the creation of crevices on the
sion product damage is confined to the fuel. The fission secondary side in which crevice chemistry was favorable for
process releases both neutrons and gammas. intergranular attack.
Neutrons are created with an energy of 2 MeV and
slow down via collisions with the coolant and structural 2.1.2. Life extension and power uprates
components. Neutrons are the primary source of radiation Due to the initial high capital cost for construction of
damage to core materials and fuel cladding and assembly nuclear power plants relative to the cost of the fuel and other
components. Typical displacement damage exposures to operating expenses, the levelized cost of electricity for LWRs
the fuel cladding (replaced every 5 years) are about is dominated by the amortized original cost of construction;
15 dpa. The cumulative displacement damage in core inter- the annualized costs associated with the fuel and operating
nal structures can approach 80 dpa after 40 years. The dis- and maintenance costs in a new nuclear power plant are esti-
placement damage rate decreases rapidly with increasing mated to contribute about 20% of the levelized cost of
distance from the core due to neutron moderation (lower nuclear electricity [68]. This factor, along with the high capac-
neutron energy resulting from energy loss via collisions ity factor of LWRs demonstrated during the past decade, has
with the coolant and core materials); the displacement led to significant interest in extending the operational licenses
damage levels in the RPV wall for a PWR are typically of nuclear power plants beyond their initial term (typically
0.05 dpa after 40 years of operation, and the correspond- 40 years). Extension of reactor operating licenses for an addi-
ing damage in a BWR vessel wall can be up to an order of tional 20 years means that reactor components will be
magnitude lower. This displacement damage can result in required to maintain their integrity for a period that is 50%
significant temperature- and dose-dependent changes to longer than the initial 40-year license. This increase in opera-
the microstructure (formation of dislocation loops, precip- tional life introduces a wide range of potential materials aging
itate formation/dissolution, void formation, radiation- issues that must be considered as part of the renewal license
induced segregation, etc.) [16], which affects mechanical process [5,69]. While the effect of increased irradiation expo-
properties (strength/hardness, ductility, fracture toughness sure is dependent on the component, the increase in operating
and embrittlement, creep, fatigue). When combined with life by 50% means that displacement damage at the bottom of
the environment, high temperature and stress, additional the top guide in a BWR may be >50 dpa, while the shroud
modes of degradation occur such as irradiation-assisted will acquire a damage level 100 times lower. High-fluence
stress corrosion cracking, corrosion fatigue and environ- components such as baffle bolts will reach damage levels
mentally enhanced fracture toughness degradation. exceeding 100 dpa. Fig. 9 shows a rough approximation of
The gamma radiation field is intense and extends through- the fluence (damage) levels associated with various compo-
out the core, aided by (n, c) reactions in structural compo- nent failures (top) in both BWRs and PWRs or with micro-
nents. While atom displacement by gamma rays is of minor structure/property changes (bottom) and the impact of a 20-
consequence, the main importance of gamma rays is in their year increase in operating lifetime. Life extension will increase
heating and changes to water chemistry. Gamma heating can the maximum expected damage level on the components
elevate temperatures in thicker components close to the fuel receiving the highest fluence, and it will also elevate the dam-
(such as baffle–former plates and bolts) by as much as age level on the balance of components proportionately.
60 °C above the water temperature. Gamma rays also induce Additional life extensions beyond 20 years are also being con-
radiolysis of the water and create a number of radicals that templated and the fluence (damage) level coinciding with
elevate the corrosion potential in the core. Corrosion poten- three 20-year life extensions is shown for comparison.
tial is the critical element governing stress corrosion cracking In addition to life extension activities, a second approach
of core materials at elevated temperature. to leveraging the existing capital assets of a nuclear power
Beyond the materials contained within the reactor ves- plant is to make modifications to the operating parameters
sel, the major components affected by the water chemistry (e.g. coolant flow rate) and/or changes to existing equip-
environment include piping, turbine rotors and blades, the ment (e.g. turbines) that enable high power levels to be
condenser and, in PWRs, the pressurizer and steam gener- achieved. These power uprate requests require detailed
744 S.J. Zinkle, G.S. Was / Acta Materialia 61 (2013) 735–758

plants. Many of the materials systems, such as piping and

heat exchangers, can be considered as replaceable (albeit
expensive) components and therefore are not directly influ-
enced by life extension considerations. However, improved
predictive knowledge of failure mechanisms can lead to
more economical maintenance and replacement schedules,
while also improving plant worker and public safety. For
the RPV, the primary effect of life extension and power
uprate scenarios is an increase in the cumulative neutron
fluence to the vessel. As will be discussed in Section 2.2.3,
there are significant uncertainties in the long-term embrit-
tlement behavior of irradiated RPV steels due to uncertain-
ties in potentially synergistic, late-emerging nucleation and
growth of solute–defect cluster complexes. There are also
Fig. 9. Neutron fluence (E > 1 MeV) and corresponding displacement uncertainties in the maximum useful lifetime of a variety
damage levels and the corresponding failure modes or changes in
microstructure/properties of BWR and PWR components and materials.
of other non-replaceable (or difficult to replace) materials
Dashed lines refer to the maximum expected fluence/dose in PWR core systems in nuclear power plants. For example, there are
components after the designated time period. nearly 1000 km of power, control and instrumentation
cables in a typical nuclear power plant. Considering that
safety analysis review. Since 1977, a total of 139 power electrical cables are known to be susceptible to age-related
uprates ranging from 0.4 to 18% have been approved by degradation and shorting (due to moisture, heat, etc.) in
the US Nuclear Regulatory Commission, resulting in applications ranging from electrical appliances to vehicles,
6 GWe additional generating capacity. The major impacts homes and factories, which are typically less hostile envi-
of the power uprate on the reactor materials include slightly ronments than nuclear power plants, there is interest in
higher operating temperature and higher neutron flux and developing a suite of non-destructive evaluation techniques
fluence to permanent core internal structures. to investigate the current performance and expected life-
A potential consequence of higher damage levels from time of cables [74]. Similarly, the concrete used for
power uprates and/or life extension is the appearance of safety-related structures in LWRs (e.g. primary contain-
new or unanticipated degradation modes. One such mode ment dome and base slab) is susceptible to a variety of
is void swelling. The formation and growth of voids was environmental degradation mechanisms that act on the
discovered in the development of materials for fast reactors cement matrix and steel reinforcement rods. Research top-
[70], where core temperatures range from 350 to 550 °C and ics of highest interest include long-term degradation mech-
damage rates are approximately 10 times that in thermal anisms (>50 years, including ionizing radiation effects),
reactors. Since void formation in stainless steels under fast improved noninvasive inspection techniques and improved
reactor irradiation conditions is pronounced only for irra- repair methods [75].
diation temperatures of 400–650 °C [52], void swelling was
not initially considered an issue for LWR core materials, 2.2. Materials degradation challenges in current LWRs
where maximum core component temperatures are below
this temperature range. However, as predicted by void Although there are many potential areas of concern for
swelling models [71] and recently confirmed by long-term materials in LWRs, the demonstrated highly reliable oper-
LWR irradiation experiments [72], lower damage rates ating performance of commercial reactors for the past 10–
can be more damaging than high dose rates as they induce 15 years suggests that most of these issues are tractable by
void formation at lower fluences and also extend the void appropriate materials selection and engineering design. As
swelling regime to lower temperatures. Further, as noted described in the following, three specific materials chal-
in Section 2.1.1, gamma heating causes the temperature lenges are considered to be of highest importance: (i) explo-
in thicker components to increase up to 380 °C, pulling ration of potential further improvements in fuel reliability
them into the range where void swelling readily occurs in and operational burn-up limits under normal operating
austenitic stainless steels at LWR-relevant dose rates conditions, and safety under transient accident conditions;
[52,73]. Baffle–former bolts in PWRs have developed voids (ii) corrosion and stress corrosion cracking in reactor com-
at damage levels as low as 7.5 dpa [13]. If the plates fas- ponents; and (iii) RPV integrity, particularly for life exten-
tened by the bolt also swell, they will apply a stress to sion scenarios.
the bolts, creating conditions that are ripe for irradiation
assisted stress corrosion cracking (IASCC). To date, there 2.2.1. Fuel system challenges during normal operations to
have been several instances of cracked baffle–former bolts high burn-up
caused by IASCC [20]. Reliable operation of the fuel assemblies for extended
There are a variety of life extension and power uprate time periods (several cycles, each consisting of 18–
issues for ex-core materials systems used in nuclear power 24 months of continuous operation) in an extremely harsh
 S.J. Zinkle, G.S. Was / Acta Materialia 61 (2013) 735–758 745

radiation environment is of central importance for the eco- PWRs, the capacity factor for fission reactors in the USA
nomics of nuclear power. Fuel failures (breach of the clad- has steadily increased from 60% in 1980 to 90% for
ding, allowing escape of some of the gaseous and volatile the past 10 years. Fig. 10 shows the average fuel burn-up
fission products into the primary coolant, or other damage at discharge for US BWR and PWR power plants and
to the fuel that prevents normal operation) can be managed the fleet-averaged capacity factors from 1977 to 2010 [78–
for a few fuel cladding failures in an operating reactor due 80].
to the presence of coolant clean-up systems, but it is unde- Since a wide range of phenomena can induce fuel failure,
sirable due to increased radiation exposure to personnel, a multifaceted approach involving cladding composition,
possible leakage of radioactive material, increased inspec- water chemistry and reactor operation modifications has
tion and fuel assembly replacement activities during refuel- collectively contributed to improved fuel performance
ing outages, and possible premature shutdown for refueling [81]. In the 1960s and early 1970s, the relatively poor fuel
(if too many fuel failures occur, due to limits on allowable performance was largely controlled by internal hydriding
radioactivity in the primary coolant). of the zirconium alloy cladding due to excessive moisture
The fuel concept based on monolithic UO2 fuel encased in the ceramic UO2 fuel pellets; this was solved by
in Zr alloy cladding has undergone remarkable improve- improved fuel pellet fabrication and use of moisture getters
ments in reliability since it was first developed in the in the fuel plenum in the reactor. In the 1980s and 1990s, as
1950s. The primary advantages associated with Zr for fuel fuel reliability improved, debris fretting of the cladding
cladding are very low parasitic absorption of neutrons became a significant problem; debris filters were added to
(thereby requiring less initial isotopic enrichment of the reactors and improved fuel maintenance procedures to
fuel to achieve a given burn-up level), and good fabricabil- minimize introduction of debris were successfully imple-
ity and strength; this is countered by relatively poor oxida- mented. Numerous water chemistry changes were made
tion behavior in hot water and steam, and anisotropic in an attempt to balance corrosion issues in different cool-
properties due to the hexagonal close-packed crystal struc- ant loop components. In the primary circuit of PWRs, Zn
ture. A series of alloying additions have resulted in com- was added to suppress steam generator corrosion and stress
mercial Zr alloys that have good resistance to oxidation corrosion cracking and to reduce overall system corrosion,
during normal nuclear reactor operations. Alternative clad- Li was added for pH control (to maintain constant pH as B
ding options, such as steels, have higher parasitic neutron was added for reactivity control at beginning of reactor
absorption, and may be susceptible to stress corrosion cycles) and the pH was increased from 6.9 to 7.4 in a
cracking under certain water chemistry environments in series of steps to control CRUD deposits on the surfaces
nuclear reactors. Many first-generation LWRs constructed of the fuel cladding [79]. In BWRs, noble metal additions
in the 1960s used either austenitic stainless steel (Types 304, and H were added to reduce stress corrosion cracking. As
316, 347) or zirconium alloys for the cladding. The stainless burn-ups increased, pellet–cladding interactions became
steel cladding experienced severe intergranular cracking more prominent due to increased fuel swelling. The closure
issues in the BWR coolant environment, due to the highly of the initial gap between the fuel pellets and the cladding
oxidizing high-temperature steam in combination with can produce a variety of mechanical stress effects as well as
radiation hardening that fostered stress corrosion cracking, multiple interactions with aggressive fission products. One
whereas the Zr alloy cladding was observed to function
with adequate reliability [76,77]. Stainless steel cladding
generally exhibited better performance than Zr alloy clad-
ding in first-generation PWRs (fuel pin defect rate of
0.01% vs. 0.1–0.3% for Zr alloy cladding). However, steady
improvement of the Zr alloy cladding performance during
the 1960s and early 1970s along with the superior (low)
parasitic absorption of neutrons by Zr led to nearly univer-
sal adoption of Zr alloy cladding by the early 1970s [76,77].
Cladding performance has continued to improve over the
past 30 years, with fleet-averaged fuel pin failure rates of
1  104 in 1980, 2  105 in 1990 and 3  106 in
2010 [78,79]. Considering that there are 50,000 fuel pins
in a LWR, the current fuel reliability means that most reac-
tors now routinely operate without any cladding breaches.
This achievement is even more remarkable when one recog-
nizes that the average fuel burn-up more than doubled dur-
ing the time period of 1980–2010. As a result of this
improved fuel performance, along with improvements in
other reactor operations such as utilization of online main-
tenance and improvements in steam generator reliability in Fig. 10. Summary of US LWR burn-up and capacity factors [78–80].
746 S.J. Zinkle, G.S. Was / Acta Materialia 61 (2013) 735–758

successful approach has been to utilize multiple-layered reactor [86]. This reorientation of the hydride precipitates
cladding (e.g. utilizing a soft Zr liner as a compliant barrier could lead to cladding failure in used fuel during handling
liner between the fuel and cladding) [81,82]. The fuel pin and transport, with a consequential effect on the cost for
diameter and cladding thickness were also decreased to safe management of used fuel. Current international safety
reduce heat flux and facilitate high burn-up (along with regulations for loss of coolant scenarios specify that the
operational changes that limited the reactor ramp rate dur- oxidized cladding should be less than 15–17% of the wall
ing start-up) [81], with current pellet diameters of 10 mm thickness, which corresponds to an oxide thickness of
and cladding thickness of 0.6 mm for BWRs and PWRs about 100 lm. This limits historical cladding materials
[83]. The current major fuel cladding degradation phenom- such as Zircaloy-4 to burn-up levels below 50–
ena for high burn-ups [79,84] include grid-to-rod fretting in 60 GWd MTU1, whereas more recent oxidation-resistant
PWRs and cladding corrosion/oxidation/CRUD deposi- Zr alloys, such as optimized ZIRLO and M5/AXIOM,
tion and pellet-cladding interactions in both BWRs and appear to exhibit suitable in-pile oxidation behavior for
PWRs [85]. Grid-to-rod fretting is induced by coolant- burn-ups much higher than 70 GWd MTU1 [87–89]. The
flow-induced vibration of the fuel cladding against the hor- slightly higher temperatures associated with power uprate
izontal grid assembly spacer. At high burn-ups the spring activities should pose an additional challenge, since the oxi-
force between the cladding and grid assembly is relaxed dation rate will be correspondingly increased due to the
due to irradiation creep [12]. Grid-to-rod fretting can be higher temperature.
reduced by providing additional stiffening of the fuel The fuel pellet experiences the most severe temperatures,
assembly and by design changes to minimize fuel rod or radiation damage and chemical transmutation environ-
assembly vibration, as well as to suppress cross flow and ment of any component in a nuclear reactor. The cumula-
jetting (grid-to-rod fretting is not a major issue in BWRs tive amount of radiation damage approaches 1000 dpa for
due to the presence of channel boxes that restrict flow a typical fuel pellet, with 5% or more of the original ura-
between fuel assemblies). nium atoms converted into 10 at.% or more fission prod-
Looking to the future, there is interest in continuing to ucts [90]. The microstructure of the initially monolithic
increase the fuel burn-up values while maintaining or polycrystalline UO2 fuel undergoes dramatic changes dur-
improving the reactor capacity factor and continuing to ing high burn-up irradiation due to the intense displace-
reduce the fuel failure rate toward zero (individual fuel ment damage and chemical transmutations, along with
pin failure probability of <1  107). Decreased overall high power densities and the relatively low thermal conduc-
fuel cycle costs, improved operational flexibility and tivity of the UO2 fuel that produce temperature gradients
reduced volume of radioactive waste are all benefits associ- >1000°C cm1. Of particular interest is the formation at
ated with increasing burn-up [81]. Grid-to-rod fretting burn-up levels above 50 GWd MTU1 of nanoscale
challenges may be more pronounced in reactors approved polygonized grains in the outer “rim” region of the fuel pel-
for power uprates due to increased coolant flow rates let, where the burn-up levels are highest and the tempera-
[85], and higher burn-ups would produce additional irradi- ture is lowest [90–92]. There was initially concern that
ation creep relaxation of the springs in the grid assembly, this new high burn-up structure might exhibit inferior fis-
which could exacerbate the fretting issue. Advanced ther- sion gas retention, thermal conductivity or other poor fuel
mal hydraulic modeling and other analyses would be useful properties. However, recent results indicate that the high
to determine if new designs for the spacer grid might reduce burn-up structure generally exhibits behavior comparable
the magnitude of the fretting. Distortions in the fuel bun- or favorable to the lower burn-up microstructure [90].
dles, PWR control rod guide tubes and BWR channel Good fuel behavior has been observed in test irradiations
plates at high burn-ups (including bowing of constrained for fuel irradiations up to 83 GWd MTU1 [93]. Explor-
components) due to anisotropic irradiation growth of the atory research is also being performed on a variety of alter-
zirconium alloy could impede the motion of control rods native fuel forms that are a significant departure from
or blades and therefore need to be carefully evaluated. monolithic sintered UO2, including inert matrix fuels [94],
Finally, oxidation of the cladding at very high burn-ups metallic or ceramic matrix microencapsulated article fuels
may require the development of a new generation of [95], and nanocrystalline oxide fuels [96], which may offer
ultra-high oxidation-resistant claddings. Oxidation intro- some advantages in achieving very higher burn-ups.
duces two potential degradation mechanisms: loss of sound
metal for the structural function of the cladding and the 2.2.2. Corrosion and stress corrosion cracking (SCC) in
possible introduction into the cladding of hydrogen formed structural materials
as a result of the reduction of steam during the oxidation The temperature range of the water coolant in LWRs is
process (including the formation of hydride precipitates, generally between 275 and 325 °C, and spans the saturated
which may serve as internal stress concentrators). An addi- (BWR) to sub-cooled (PWR) regimes. Although LWRs
tional concern associated with hydrogen pick-up is the have been in operation for over 50 years, corrosion remains
potential for reorientation of the hydride precipitates from a significant concern that will become even more important
circumferential to radial direction as a result of spent fuel with age. Corrosion occurs in all of the major systems
handling procedures after the fuel is withdrawn from the exposed to a water environment, including the reactor core,
 S.J. Zinkle, G.S. Was / Acta Materialia 61 (2013) 735–758 747

steam generator, turbine, condenser and piping, valves and pure water occurs only at high nickel concentrations.
fittings, and in a wide variety of alloys such as carbon and Unfortunately, alloy 600 used originally in steam genera-
low alloy steels used in piping and turbine components, tors and vessel penetrations contains approximately 78%
stainless steel used in core internals and primary flow cir- Ni and therefore is highly susceptible to SCC even in pure
cuits and the condenser, nickel-base alloys in the steam water. Neither microstructure modification (thermal treat-
generator and in reactor vessel penetrations and welds, ment) nor water chemistry control has been able to miti-
and zirconium alloy fuel cladding [17,19,20,97]. gate SCC in alloy 600. Staehle and Gorman identified
Early corrosion problems stemmed from “epidemics” seven different SCC modes defined by their potential–pH
that were generally precipitated by improper water chemis- combinations [97]. In fact, SCC in alloy 600 appears to
try control; ingress of chlorides that induced pitting in track the Ni–NiO stability line, in which cracking is at a
steam turbine discs and blades, and pitting and stress cor- maximum at the line and drops off both above and below
rosion cracking in stainless steel, or poor secondary side it (Fig. 13) [10]. Since this range of potential and pH spans
pH control that resulted in wastage and crevice corrosion those typically achievable in service, control of SCC in
in steam generator tubes, or poor microstructure or alloy alloy 600 has proved very difficult. As a result, alloy 600
chemistry control; high corrosion rates of zirconium fuel components have been replaced by alloy 690, which has a
cladding, stress corrosion cracking of stainless steel BWR nominal composition of 60Ni–30Cr–10Fe and exhibits
piping due to sensitization or weld knife-line attack [17]. much higher resistance to SCC in pure water, chlorides
More recently, corrosion degradation has emerged in the and alkaline solutions. However, historically, alloy 600
forms of stress corrosion cracking of stainless steel steam was found to be difficult to crack in pure high-temperature
lines and nickel-base steam generator tubes and reactor water in the laboratory, prompting the belief that it would
vessel penetrations, flow assisted corrosion in low alloy be resistant in service. In fact, even after years in operation,
steels, and nodular corrosion, shadow corrosion, CRUD- few incidents of cracking were reported. Over time, the
induced localized corrosion and fretting of zirconium alloy incubation period was discovered to be of the order of
fuel cladding [18]. Still more recently, irradiation has 11–12 years, and eventually, all plants began to experience
emerged to play an increasingly important role in irradia- cracking. This has led to widespread replacement by steam
tion-assisted stress corrosion cracking (IASCC) and irradi- generators constructed from alloy 690. The concern today
ation-accelerated corrosion [21,22]. As plants age, the most is that the superior resistance of alloy 690 to SCC may be
important corrosion issues will center about stress corro- due to a longer incubation period and that, in fact, it may
sion cracking, and the accelerating role of irradiation in eventually begin to crack. Current efforts are focused on
both IASCC and corrosion. A recent example of corro- trying to determine if there are microstructures, processing
sion-induced degradation occurred in 2002 at the Davis- routes or water chemistries through which alloy 690 is sus-
Besse reactor (PWR), when stress corrosion cracking of ceptible to SCC. Recently, it was found that a single cold
the weld metal between the vessel head and a control rod rolling operation to 20–30% reduction in thickness
drive housing on the vessel head caused coolant to leak increases the crack growth rate in pure water by over a fac-
onto the head. Evaporation of water on the hot surface tor of 100 [99]. Similar research activities are in progress to
resulted in a concentrated boric acid solution that resulted identify susceptibility to crack initiation.
in corrosion to such an extent as to create a hole in the
head nearly the size of a soccer ball. The 10 mm thick [Link]. Irradiation-assisted stress corrosion cracking. IAS-
stainless steel weld metal liner prevented a loss-of-coolant CC of austenitic stainless steels and some nickel-base alloys
accident. has presented a significant problem in ensuring the integ-
rity of LWR core components. IASCC is a generic chal-
[Link]. Stress corrosion cracking. Today, the major stress lenge as it apparently cuts across all LWR designs and
corrosion cracking issues are in the nickel-base alloys used materials. Table 4 shows that IASCC has been observed
in steam generators and in vessel penetrations and stainless in at least four water reactor designs, 11 different alloys
steels used in the reactor core. Since steam generator tubes and dozens of components. The specific effects of irradia-
comprise some 75% of the surface area of the primary cir- tion on IASCC are classified into two categories: water
cuit in contact with the coolant, their performance is criti- chemistry and microstructure [21,22]. Water chemistry
cal to that of the reactor. Fig. 11 shows that, while many effects include radiolysis and its effects on corrosion poten-
degradation modes of alloy 600 steam generator tubes tial, and the effects of corrosion potential on IASCC.
exist, SCC has been the dominant failure mode over the Microstructure effects include radiation-induced segrega-
past 25 years [98]. SCC of nickel-base alloys is very sensi- tion, irradiated microstructure, swelling and creep, and H
tive to composition and water chemistry. Fig. 12 shows and He generation. Leading mechanisms proposed to
the propensity for cracking of austenitic alloys as a func- explain the roles of radiation in the SCC process are: radi-
tion of nickel content in both pure water and 0.1% NaCl; olysis and crack tip strain rate, grain boundary chromium
this cracking can either be transgranular (TGSCC) or inter- depletion, irradiation hardening, localized deformation
granular (IGSCC) [10]. Note that SCC in chloride is at a and radiation-induced solute segregation of minor elements
maximum at the extremes in nickel content, but SCC in [21,22].
748 S.J. Zinkle, G.S. Was / Acta Materialia 61 (2013) 735–758

Fig. 11. Failure modes of mill-annealed Alloy 600 steam generator tubes in US PWRs over a 38-year period [98].

Fig. 12. SCC severity of austenitic alloys as a function of nickel content in pure water and 0.1% sodium chloride (courtesy of R.W. Staehle [10]).

Irradiation causes a significant change in local composi- many effects of irradiation that overlap spatially and tem-
tion near grain boundaries and other defect sinks [3,4]. The porally, more work is needed to identify their roles in the
enrichment of nickel and silicon and the depletion of chro- mechanism(s) of IASCC and develop a comprehensive pre-
mium can affect the susceptibility to IASCC, especially diction methodology. In particular, the evolution of local-
under oxidizing conditions. Irradiation also alters the ized deformation in an irradiation damage microstructure
microstructure and, under LWR conditions, faulty disloca- and the emergence of phases at high dose are the areas that
tion loops represent the primary irradiation-induced micro- require a better understanding to ensure the structural
structure defect. The loops impede the motion of integrity of core components to lives beyond 40 or 60 years.
dislocations, resulting in an increase in the yield strength The mechanism of IASCC remains a major outstanding
of stainless steels by as much as a factor of five. Radiation issue in the degradation of core components in LWRs of
hardening correlates with IASCC propensity, and also all types.
induces highly localized deformation in the form of disloca- In considering the attributes of an IASCC-resistant
tion channels, which could contribute to IASCC [100,101]. austenitic alloy, the following are likely beneficial: high
Irradiation also induces creep that can relax macroscopic Ni and Cr contents, low Si content, absence of brittle oxide
stresses, and can also enhance local dynamic deformation. and nitride inclusions, high coincident site lattice fraction
Other factors, such as swelling and formation of new of grain boundaries, low connectivity of high-angle grain
phases, may enhance IASCC at high fluence. With the boundaries and grain boundary coverage by chromium
 S.J. Zinkle, G.S. Was / Acta Materialia 61 (2013) 735–758 749

[Link]. Irradiation-accelerated corrosion. Another major

issue in LWRs operated to high doses involves the inter-
play of radiation and corrosion in irradiation-accelerated
corrosion. The mechanism of this process is unknown,
but it has been shown to have significant impacts on corro-
sion rates. For example, zirconium irradiated in-reactor in
moist carbon dioxide–air mixtures had oxygen weight gains
of more than five times that in the unirradiated state [103].
In-reactor corrosion rates of zirconium alloys were
found to be 10 times greater than those conducted out-
of-reactor, and part of the difference was attributed to
greater permeability of the oxide irradiated in-reactor
[104]. More recently, Lewis and Hunn [105] found that pro-
ton irradiation of a 316 stainless steel foil in room-temper-
ature water for 4 h produced an oxide that was 20 times
thicker than the unirradiated control. Further, old data
on in-reactor exposure of Zircaloy-2 [106] revealed ten-fold
increases in the oxide weight gain and a strong, linear
dependence on neutron flux. These data suggest that dis-
placement damage to the solid during corrosion produces
Fig. 13. Modes of SCC in Alloy 600 affected by environmental chemistry a significantly greater effect than that due to radiolysis
(courtesy of R.W. Staehle [10]). Regimes in the figure are: AcSCC:
alone.
acidic-induced SCC; AkSCC: alkaline-induced SCC; HPSCC: high
potential-induced SCC; LPSCC: low potential-induced SCC; AkIGC:
alkaline-induced intergranular corrosion; PbSCC: lead-induced SCC and 2.2.3. Reactor pressure vessel integrity issues
Sy-SCC: sulfide-induced SCC. The two major degradation issues for RPV steels are
embrittlement associated with hardening from radiation-
carbides. Finally, ferritic or ferritic–martensitic alloys induced solute–defect clusters, and corrosion and stress
should be considered due to their inherently greater resis- corrosion cracking phenomena; the corrosion and stress
tance to radiation effects and IGSCC compared to iron- corrosion cracking issues have been briefly summarized in
and nickel-base austenitic alloys [102]. Section 2.2.2.

Table 4
Summary of observed IASCC issues in LWR components [21,22].
Component Material Reactor type Possible sources of stress
Fuel cladding 304 SS BWR Fuel swelling
Fuel cladding 304 SS PWR Fuel swelling
Fuel cladding* 20%Cr25%Ni/Nb AGR Fuel swelling
Fuel cladding ferrules 20%Cr25%Ni/Nb SGHWR Fabrication
Neutron source holders 304 SS BWR Welding and be swelling
Instrument dry tubes 304 SS BWR Fabrication
Control rod absorber tubes 304/304L/316L SS BWR B4C swelling
Fuel bundle cap screws 304 SS BWR Fabrication
Control rod follower rivets 304 SS BWR Fabrication
Control blade handle 304 SS BWR Low stress
Control blade sheath 304 SS BWR Low stress
Control blades 304 SS PWR Low stress
Plate type control blade 304 SS BWR Low stress
Various bolts** A-286 PWR and BWR Service
Steam separator dryer bolts** A286 BWR Service
Shroud head bolts** 600 BWR Service
Various bolts X-750 BWR and PWR Service
Guide tube support pins X-750 PWR Service
Jet pump beams X-750 BWR Service
Various springs X-750 BWR and PWR Service
Various Springs 718 PWR Service
Baffle former bolts 316 SS cold work PWR Torque, differential swelling
Core shroud 304/316/347/L SS BWR Weld residual stress
Top guide 304 SS BWR Low stress (bending)
*
Cracking in AGR fuel occurred during storage in spent fuel pool.
**
Cracking of core internals occurred away from high neutron and gamma fluxes.
750 S.J. Zinkle, G.S. Was / Acta Materialia 61 (2013) 735–758

Radiation embrittlement is of greatest concern for PWR Pressure vessel embrittlement issues that need further
pressure vessels; the larger vessel diameter in BWRs allows research include [49,67,108]: (i) the effects of high dose,
more moderation of fast neutrons in the water between the long operation lifetimes and irradiation flux on hardening
core and vessel, resulting in 3 to 10 times lower fast neutron and embrittlement; (ii) impact of heat-to-heat variability
fluxes compared to PWR pressure vessels [67]. The funda- and the quantitative accuracy of surrogate materials such
mental mechanisms associated with radiation hardening as surveillance specimens; (iii) effect of Ni concentration
and accompanying embrittlement of RPV steels are gener- on the formation and embrittlement impact of Ni-rich “late
ally understood, including the pronounced deleterious blooming phases”; (iv) quantitative validity of the fracture
effects associated with Cu, P and Ni solute additions due toughness master curve concept (is the curve shape univer-
to the formation of nanoscale precipitates that produced sal, correlation between Charpy impact and fracture tough-
significant matrix hardening [49,67,107]. A series of ness tests, consequences of intergranular fracture, etc.); (v)
improved mechanistic models for radiation embrittlement impact of size (constraint) effects and other phenomena on
have been developed over the past 25 years that provide a the toughness correlation between pre-cracked Charpy (or
clear description of the complex and synergistic effects smaller) surveillance specimens on possible introduction of
associated with solute (Cu, Ni, P), temperature, dose and biased estimates of reference fracture toughness; (vi) cor-
irradiation flux [49,108]. Fig. 14 shows an example of the rectly accounting for the large attenuation in neutron flu-
predicted increase in the ductile-to-brittle transition tem- ence and displacement damage dose (a factor of 5) that
perature (DBTT) of RPV steel containing high levels of occurs through the wall thickness of reactor vessels; (vii)
Cu or Ni solute [49]. For the high Cu, medium Ni solute development of improved-fidelity modeling and micro-
case, a high density of Cu-rich precipitates forms rapidly structural analysis to achieve better understanding of the
at low dose and causes significant embrittlement for expo- roles and synergies of the various key experimental param-
sure levels corresponding to a few years of operation of a eters; (viii) potential for phosphorus segregation to induce
PWR pressure vessel. A relatively small amount of addi- intergranular fracture; (ix) thermal annealing and reirradi-
tional embrittlement occurs after the initial rapid embrittle- ation research to investigate feasibility and quantify opti-
ment phase due to the lack of additional Cu solute. For the mized conditions for periodic vessel annealing to mitigate
high-Ni case, predicted slow nucleation rates of the Ni-rich radiation embrittlement; and (x) investigation of long-term
precipitates result in an extended transient regime with (>50 years) thermal aging effects on the microstructure and
relatively low embrittlement, followed by pronounced properties of low alloy steels. Finally, research on neutron
embrittlement associated with the copious formation of radiation embrittlement of weldments (including the
Mn–Ni–Si precipitates. The full impact of such “late impact of various post weld heat treatments) should be
blooming phases” on possible accelerated hardening and continued.
embrittlement of reactor vessel steels near the end of their
design life is the subject of active current research, and 2.3. Materials challenges during off-normal events
could become a major impediment to additional future
extensions in the operating lifetime of PWR power plants. There are two major transient operational scenarios that
have been considered for design basis safety analyses used
to define operational safety limits for commercial reactors:
reactivity-initiated accidents (RIAs) and loss-of-cooling
accidents (LOCAs). RIA events are an unwanted increase
in fission rate (reactor power), such as might occur from
an unintended prompt ejection of a control rod [109].
The RIA scenario would produce a rapid increase in fuel
power and temperature, which could lead to potential fail-
ure of fuel cladding and release of radioactivity into the pri-
mary coolant. Due to degradation of the cladding
mechanical properties with increasing exposure, safety
authorities have established design limits for the maximum
allowable energy deposition in the fuel, which decrease
with increasing burn-up. The RIA event would be rela-
tively short in duration since the negative coefficient of
reactivity in LWRs would quickly lead to a reduction of
the excess reactivity and primary coolant flow would con-
tinue throughout the event. The LOCA scenarios are based
on the knowledge that fission reactors continue to generate
substantial amounts of decay heat for days and weeks after
Fig. 14. Predicted increase in the radiation-induced DBTT of RPV steel they are shut down, due to a variety of radioactivity decay
for two solute concentrations [49]. processes in the fuel and core materials; for example, the
 S.J. Zinkle, G.S. Was / Acta Materialia 61 (2013) 735–758 751

residual heat relative to the full-power reactor value is oxidation and would proportionally reduce the generation
about 6% at a time of 1 s after a reactor scram and of hydrogen. Possible approaches in this category include
0.5% 1 day after shutdown. Considering that commercial utilization of oxidation-resistant coatings or development
reactors typically create over 3000 MW of thermal power of new alloys with significantly improved oxidation resis-
at full power, these residual heat values represent signifi- tance (either Zr-based or new alloys). (ii) Utilize fuel clad-
cant heat removal challenges for a LOCA scenario. With- ding with improved high-temperature mechanical
out adequate heat removal, the increase in core properties and resistance to hydrogen embrittlement (high
temperature can lead to cladding rupture and release of tensile and short-term creep strength, good thermal shock
radioactivity into the primary coolant. The LOCA event resistance, high melting temperature). (iii) Utilize new fuel
could last for hours or days at elevated temperature, forms with lower operating temperature, higher margins to
depending on the severity of the loss of cooling capability. fuel melting, enhanced retention of fission products (e.g.
Whereas zirconium alloys have evolved to achieve very fuel microencapsulation or gettering techniques), and/or
impressive fuel performance under normal operating con- reduced fuel dispersal probability compared to UO2. Since
ditions in water-cooled nuclear reactors (Section 2.2.1), fuel systems are inherently a replaceable commodity, it
the behavior of UO2 monolithic fuel with Zr alloy cladding would be possible to implement new fuels with incremental
under accident scenarios is far from optimized. Hydrogen improvements in accident tolerance (using one or more of
uptake and radiation-induced hardening can lead to the approaches outlined above) as long as the geometry
reduced ductility in Zr alloy cladding that can impair the and performance under normal operations do not have a
resistance to cladding failure associated with pellet– negative impact on the reactor operations.
cladding mechanical interactions during an RIA event In addition to the potential changes to the fuel system,
(expansion of the fuel pellet due to rapid heating) [110]. there are several other core materials systems that would
If pronounced clad breaching occurs, the possibility exists be considered for improvements if robust accident toler-
for significant fragmentation and dispersal of the ceramic ant fuels were successfully developed. For example, the
UO2 fuel pellets. For LOCA scenarios, the rapid increase AgInCd control rods currently used in many PWRs have
in Zr alloy oxidation rate with increasing temperature will a relatively low melting temperature (900 °C). Similarly,
degrade the mechanical properties of the cladding and the control blades in many BWRs, which utilize B4C pellets
could lead to cladding breach and/or fracture, which in inside stainless steel tubes, begin to have a eutectic Fe–B
turn could lead to reduced flow of primary coolant within reaction at 1170 °C. Higher-temperature material options
the fuel assemblies in the core due to coolant channel for neutron control should be pursued.
blockage by cladding fragments. The high heat of oxida-
tion for Zr can also make a large contribution to the core 3. Materials challenges in future fission reactor concepts
heating; considering that a typical 1000 MWe LWR core
contains 30,000 kg of cladding, 55 MW h of heat would Construction is currently in progress worldwide on sev-
be released if it were completely oxidized (note: BWR cores eral so-called Generation III and Generation III+ LWR
also contain nearly a comparable amount of Zr in the power plants that are designed for improved efficiency, pas-
channel boxes). This could set up an autocatalytic reaction sive safety and economics. To a large extent, these reactors
where the heat from Zr oxidation drives a temperature represent an evolutionary design change utilizing materials
increase in the core, leading to more rapid oxidation and systems that are similar to current (Generation II) LWRs,
heat generation. Finally, the oxidation of Zr by steam leads and therefore the materials challenges facing the new reac-
to the production of potentially explosive hydrogen gas tors will be comparable to those faced in existing reactors.
(1200 kg if all of the cladding was oxidized by steam). Another class of light-water-cooled reactors under consid-
The shortcomings of the Zr alloy/UO2 fuel system to severe eration would utilize in-factory construction techniques
accident conditions such as occurred at the Fukushima and new designs with high emphasis on passive safety to
Dai-ichi nuclear power plant in Japan following the 2011 construct small (50–300 MWe) nuclear power plants; some
earthquake and tsunami has led to increased interest in of the materials challenges with these small modular reac-
improving the safety of nuclear reactors to rare but credible tors (SMRs) are discussed in the following section. Finally,
accident scenarios [111]. a brief discussion on the materials challenges for Genera-
Considering that the key functions of the fuel system in tion IV reactor concepts will be provided.
an accident scenario are to maintain core cooling capability
and to minimize or prevent dispersal of fuel and fission 3.1. Materials challenges for proposed light-water small
products, there are three major potential approaches to modular reactors
design LWR fuel systems with improved accident toler-
ance. (i) Utilize cladding options with reduced reaction Due to the high capital cost for constructing a large
kinetics with high-temperature steam; reduction of the (1000 MWe) nuclear power station, a variety of designs
cladding high-temperature oxidation rate by one or more for small modular reactors (<300 MWe) have been pro-
orders of magnitude compared to alloys such as Zircaloy posed [112]. In order to compete against the traditional
would nearly eliminate the contribution of heat input from economies of scale, which favor large reactor size (per
752 S.J. Zinkle, G.S. Was / Acta Materialia 61 (2013) 735–758

MWe), the proposed small reactors would utilize modular ation as part of the first phase of the collaborative Gener-
construction techniques in factories followed by shipping ation IV program in 2002 [114]. The objective was to
of major components by rail or barge for final construction identify concepts that had one or more of the following
at the reactor site. The conceptual designs include both attributes: increased efficiency, generation of process heat
light water and other coolants [112,113]. The materials to drive chemical processes such as the production of
issues facing the advanced reactor concepts (coolants other hydrogen, increased safety and reduction in waste genera-
than light water) are qualitatively similar to those for con- tion. The concepts finally selected were the supercritical-
ceptual Generation IV systems discussed in Section 3.2. water-cooled reactor (SCWR), the sodium fast reactor
Many of the specific SMR materials issues are uncertain (SFR), the lead fast reactor (LFR), the very-high-tempera-
due to the preliminary nature of their engineering designs ture reactor (VHTR), the gas fast reactor (GFR) and the
at this time. For example, it is uncertain whether there will molten salt reactor (MSR). Table 3 summarizes the basic
be any significant modification in the water chemistry for characteristics of each of these reactor types and the mate-
the SMRs compared to what has been utilized in existing rials proposed for the various major components [7]. Note
LWRs. that all designs call for higher operating temperatures and
For the light-water-cooled designs that are being pro- radiation doses, placing a higher burden on the integrity of
posed for construction in the near future (within 10 years), materials.
most of the materials issues are expected to be comparable To allow operation at much higher temperatures,
to those faced by Generation II and Generation III reac- advanced Generation IV reactor concepts utilize different
tors. One key issue will be to determine if the manufactur- coolants, including water in the supercritical state, liquid
ing processes used in the factory assembly provide any metals such as sodium and lead–bismuth, molten salts
advantage (or disadvantage) in material performance com- and high-pressure helium gas. The materials challenges
pared to standard fabrication techniques. For example, for the Generation IV reactor concepts come about
advanced manufacturing techniques such as additive man- because of the very high fuel temperatures, the intense radi-
ufacturing could potentially reduce the fabrication costs of ation flux and coolant compatibility issues. Thus, the fuel,
specialized pump components and other devices, but it the cladding, the structural materials, the reactor vessel and
would be important to compare the microstructure and the interaction of these materials with the coolants present
performance of the materials to conventional wrought the greatest challenges to new, more robust nuclear reactor
material under prototypic operating environments. One concepts for the twenty-first century. Structural materials
of the key design changes in many of the light-water small used in the cores of advanced reactors will face unprece-
modular reactors is the use of integral containment (i.e. the dented combinations of temperature, radiation dose and
steam generator for a PWR is located within the RPV). In stress. As shown in Fig. 15, a common feature of all
some concepts, the primary water flows across the outside advanced designs is a high operating temperature com-
of the tubes, resulting in a reversal in primary and second- pared to current LWRs. Another unique feature is the
ary sides of conventional steam generators used in all oper- simultaneous presence of intense knock-on displacement
ating PWRs. Integral containment has a significant safety damage by the fission neutrons. Almost all of the Genera-
advantage in that the possibility of a large-pipe LOCA is tion IV concepts call for radiation damage levels that
eliminated. However, considering that nuclear reactors still exceed those of LWR experience. One additional proposed
occasionally experience stress corrosion cracking issues
with steam generator tubing, and that the evolution of
water chemistry control is strongly coupled to the steam
generator geometry, the inversion of primary and second-
ary sides of a tubed steam generator and the limited acces-
sibility of the steam generator for routine inspection may
be a source of problems unless further improvements in
steam generator reliability are achieved.

3.2. Proposed next-generation (Generation IV) fission

reactor concepts

3.2.1. Brief overview of the six Generation IV concepts

Over the past 10 years, the United States Department of
Energy and the Generation IV International Forum have
explored six particularly appealing advanced reactor con-
Fig. 15. Temperature and dose requirements for in-core structural
cepts as potential next-generation (Generation IV) nuclear
materials for the operation of the six proposed Generation IV advanced
power systems [114,115]. These concepts were selected from reactor concepts, the traveling wave reactor and fusion reactor concepts.
hundreds of ideas submitted to the US DOE by scientists The dimensions of the colored rectangles represent the ranges of
and engineers worldwide, during a broad canvassing oper- temperature and displacement damage for each reactor concept.
 S.J. Zinkle, G.S. Was / Acta Materialia 61 (2013) 735–758 753

concept, the traveling wave reactor (TWR), would require

fuel cladding integrity to damage levels approaching
600 dpa. The need for more robust materials extends to
the pressure vessel, the primary safety structure for most
reactor designs.

3.2.2. Materials-limiting phenomena for Generation IV

concepts
The high temperatures and damage levels of all of the
designs described in Table 3 will accelerate the corrosion
and oxidation kinetics and open new pathways for materi-
als degradation. The six Generation IV concepts and the
TWR can be grouped into three general categories accord-
ing to their positions in Fig. 15, as well as the nature of the
coolant. The SFR, LFR, MSR and the TWR will all oper-
ate at elevated temperatures and very high damage levels,
and will utilize either liquid metal or molten salt as the
coolant. The GFR and VHTR will experience lower dam- Fig. 16. Comparison of the volumetric void swelling behavior of Type
age levels but even higher temperatures, and will use the 304L [116], 316 [12] and a Ti-modified (D9) [12] austenitic stainless steel
and 9–12% Cr-tempered ferritic/martensitic steels [12,117–119] following
relatively innocuous He as the working fluid. The SCWR
irradiation at 400–550 °C to high doses in a fast fission reactor spectrum.
is a unique third category in that it is the only water-cooled
reactor among the Generation IV designs; it will experience
lower damage levels relative to other concepts (comparable Many years of research and development of austenitic
damage to current LWRs), but conversely will operate at alloys have managed to extend the low-swelling transient
high temperature and very high pressure, where corrosion regime by utilizing swelling resistant microstructures such
and stress corrosion cracking issues may become as the fine TiC precipitates in Ti-modified austenitic steels,
paramount. but the delay of steady-state void growth is not sufficient to
avoid significant void swelling during operation to the high
[Link]. High-temperature, high-dose fission concepts. The doses contemplated for several Generation IV reactor con-
challenges associated with the high-temperature, high-dose cepts (cf. Figs. 9 and 10). As such, more radiation-resistant
operating environment of the first group (SFR, LFR, alloys, such as 2.5–12% Cr bainitic–ferritic–martensitic
MSR, TWR) will place increased emphasis on strength, steels are being considered for high-dose core internal
creep and creep-fatigue behavior in addition to fracture and RPV applications. Exacerbating the problem is phase
toughness at low temperature. While chemical interaction instability at high doses due to radiation-induced or -
between coolant and structural materials will present some enhanced solute segregation [121,122] and ballistic dissolu-
degradation challenges, the major concern is the very high tion of precipitates [122] by energetic displacement cas-
radiation damage levels expected in core components. At cades. Irradiation can nucleate or dissolve phases,
such high damage levels, the major degradation modes changing the solute composition of the matrix and enhanc-
are likely to be driven by void swelling and phase stability. ing void growth [123]. Further, dissolution of particles
Void swelling occurs at homologous temperatures of 0.35– added to increase the strength of the alloy results in soften-
0.55TM, which for steels (325–650 °C) overlaps the temper- ing and compromises high-temperature strength and creep.
ature range of the reactor core for these four high-temper- For example, c0 matrix precipitates that provide strength to
ature, high-dose concepts. Fig. 16 compares the void nickel-base alloys used in high-temperature applications
swelling behavior (obtained from immersion density mea- are unstable under irradiation [121,122,124–126]. Further,
surements) for Type 304L [116], 316 [12] and a Ti-modified radiation can induce the formation of brittle phases along
(D9) [12] austenitic stainless steel and 9–12% Cr-tempered grain boundaries and other defect sinks that can reduce
ferritic/martensitic steels [12,117–119] following irradiation ductility and degrade fracture toughness [124,125,127].
at 400–550 °C to high doses in a fast fission reactor spec- However, some types of second-phase particles have
trum. In all cases, the void swelling behavior consists of an demonstrated good stability to high-dose neutron irradia-
initial low-swelling transient regime followed by a high tion and can provide strength at high temperature while
swelling rate regime (approaching 1% dpa1 for the austen- ameliorating radiation damage effects. For example, con-
itic steels [12]). Although the maximum allowable volumet- trolled additions of Ti and P to austenitic Fe–Cr–Ni alloys
ric swelling for structural applications is design dependent, has been demonstrated to produce fine dispersions of TiC
void swelling levels >5% are generally unacceptable based or M2P (M = Fe, etc.) precipitates that provide dramatic
on typical engineering design considerations. Severe improvement in void swelling resistance after high-dose
embrittlement has also been observed in irradiated austen- (100 dpa) irradiation compared to standard Fe–Cr–Ni
itic steels when the volumetric swelling is >10% [120]. alloys [51,128–130]. Similarly, the nanometer-sized
754 S.J. Zinkle, G.S. Was / Acta Materialia 61 (2013) 735–758

(Y,Ti,O)-rich particles in oxide-dispersion-strengthened critical value and reduces at partial pressures below this
alloys appear to exhibit good stability under irradiation value. Similarly, chromium carbide is stable above a critical
and provide significant strength advantages over ferritic– carbon activity, and decarburization is expected to occur
martensitic alloys up to high temperatures [131–135]. An below the critical value. Oxidation, decarburization and
added benefit of nanometer-sized oxides is their role in carburization can degrade the mechanical properties of
healing radiation damage. The high particle density pre- the alloy; for example, oxidation reduces the load-bearing
sents a very large surface area for point defect trapping, cross-section of the component and internal oxide precipi-
promoting self-healing via recombination [56] and thus tates act as the preferential crack initiation sites [140] near
keeping the net accumulated radiation damage at a low the surface of the alloy, which can, potentially, decrease the
value. creep and fatigue life of the alloy. A significant reduction in
the creep-rupture ductility of alloy 800H [141], alloy 617
[Link]. Very high-temperature gas-cooled concepts. The [142] and Hastelloy X [143] has been reported in a carburiz-
structural materials challenges become magnified consider- ing environment in comparison to pure helium and air
ably when moving from medium-temperature designs to environments; Fig. 17 shows an example of the effect of
very-high-temperature designs in which materials must oxygen and methane impurities on the creep rupture
withstand temperatures approaching 1000 °C (GFR, behavior of alloy 617 (note the longer creep lifetime but
VHTR). Corrosion and oxidation of alloys are unavoid- lower tertiary creep regime for the methane impurity condi-
able at these temperatures due to the very rapid kinetics. tion). A coarse and semi-continuous film of carbides forms
In all cases, the challenge is to develop coolant–materials along the grain boundaries during carburization, and this
systems that result in the formation of protective and may act as preferential crack initiation and propagation
self-healing films to ensure the longevity of the structures paths, and could decrease the operating life of the alloys.
for the life of the reactor. Grain boundary migration and sliding has been identified
At the extreme operating temperatures envisioned for as the dominant creep deformation mechanism in the can-
gas-cooled reactor concepts, graphite and ceramic compos- didate alloys, such as alloy 617 at 1000 °C [144,145], and
ites are the leading candidates for structural materials the dissolution of carbides due to decarburization may lead
[136]. Along with the numerous engineering design issues to significant loss of the creep strength. Therefore, a
associated with utilization of low-ductility materials in a detailed knowledge of the oxidation mechanisms and rates
complex high-power energy system, the property degrada- of microstructure degradation is important to estimate the
tion associated with neutron displacement damage poses lifetime of the component and define mitigation strategies
particular challenges. The anisotropic response of graphite for improved oxidation performance of alloys. However,
to neutron displacement damage due to its hexagonal it is unlikely that unprotected alloys can maintain their
close-packed crystal structure requires the use of specially integrity at 1000 °C without protection by coatings or bar-
manufactured “nuclear” grades of graphite to achieve the rier layers. Development of protective layers without com-
desired component lifetimes [136]. For components that promising thermal conductivity is perhaps the most
are subject to relatively high displacement damage expo- important major challenge for structural materials for the
sures or engineering stresses, ceramic composites must be VHTR environment.
used instead of graphite. Nevertheless, metals are still the
most viable materials for heat transfer components such [Link]. The SCWR concept. The SCWR is unique in that it
as heat exchangers. is the only water-cooled concept among the Generation IV
In this concept, high-temperature helium gas will pass designs. Maintaining water in the supercritical state
through an intermediate heat exchanger, where it will requires very high pressures and exposes the entire circuit
transfer heat to a secondary coolant. Such temperatures
require the use of nickel-based alloys rich in chromium
(about 22 wt.%) and strengthened by additions of Mo,
Co and W (for example, Inconel 617 and Haynes 230Ò)
[137]. The helium inevitably contains parts per million
(ppm) levels of CO, CO2, H2, H2O, and CH4 as impurities,
which arise mainly from reactions between the hot graphite
core and in-leakage of O2, N2 and water vapor from seals
and welds, and degassing of reactor materials such as fuel,
thermal insulation and in-core structural materials
[138,139].
Depending on the impurity concentration, temperature
and alloy composition, the impurities react with the metal-
lic surfaces of the heat exchanger resulting in oxidation,
oxide reduction, carburization and decarburization. Chro- Fig. 17. Effect of methane and oxygen impurities in helium on the creep
mium oxide is stable at oxygen partial pressures above a rupture behavior of Alloy 617 at 843 °C [142].
 S.J. Zinkle, G.S. Was / Acta Materialia 61 (2013) 735–758 755

to water temperatures that are well above the existing and nickel-base alloys. Proton irradiations of type 316L
knowledge base. Until recently, processes such as IGSCC stainless steel and Ni-based alloy 690 showed a significant
and IASCC in supercritical water have been relatively increase in intergranular cracking relative to the unirradi-
unexplored. Recent research has shown that alloys that ated cases. The increased cracking could not be attributed
are susceptible to these degradation modes in LWR envi- to radiation-induced segregation or hardening alone, so
ronments also show susceptibility in SCW, though at combinations of factors or other defect mechanisms must
higher rates due to the higher temperatures [146]. Alloys be at play [156]. Both the density of cracks and the crack
considered for structural components fall broadly into depth increased over the unirradiated case following irradi-
two classes: austenitic stainless steels and nickel base alloys ation to 7 dpa and testing in SCW at 400 °C. One set of
are resistant to general corrosion, but susceptible to data exists on the effect of neutron irradiation on cracking
IGSCC and IASCC, whereas ferritic–martensitic alloys in SCW, where an austenitic stainless steel was irradiated
are resistant to SCC but generally exhibit higher rates of to doses of over 40 dpa and showed extreme embrittlement
oxidation. For iron-based systems, corrosion resistance is [157]. Under the same irradiation and testing conditions,
associated with the ability of iron- and chromium-based ferritic–martensitic alloys were found to be resistant to
surface oxides to function as barriers to the transport of cracking.
reactants (oxygen and metal ions). These oxides are the
same as those typically formed in other oxidizing environ- 4. Conclusions
ments, such as steam, under less extreme conditions. The
main issues related to the behavior of these oxides as pro- The continued utilization of nuclear energy systems for
tective layers in SCW are similar to those for advanced- worldwide baseload electricity offers a number of materials
steam-cycle fossil energy plants [147]. However, wall thick- research challenges. The high reliability of current light-
nesses of critical core components such as fuel cladding and water fission reactors (e.g. 90% average capacity factor
coolant tubes are an order of magnitude smaller than boiler by US reactors for the past decade) demonstrates the high
tubes in fossil plants, placing a greater burden on the devel- reliability of this energy source under normal operating
opment of thin protective oxide layers. conditions. Planned extensions in the operating lifetime
While oxide growth rates and product morphological for reactors are being supported by accompanying materi-
details are specific to the oxygen content of the fluid, the als R&D to investigate corrosion and neutron-induced
temperature and the steel composition, and possibly other materials degradation phenomena. The three major materi-
factors [148], the oxide structure on steels after exposure to als challenges for continued safe, reliable and cost-effective
SCW is similar to what is observed under steam conditions utilization of water-cooled nuclear reactors for electricity
for ferritic and ferritic–martensitic (F-M) steels [147]. production are development of improved understanding
Because it is generally accepted that chromia-containing of the synergistic fundamental mechanisms responsible
spinels are better permeation barriers to cations (metal) for corrosion and stress corrosion cracking degradation
and anions (oxygen, OH, etc.) relative to iron oxides of austenitic steels and nickel base alloys, development of
[149], it is the underlying oxide layer that can proffer the a truly predictive understanding of the multi-physics phe-
best corrosion resistance in the SCW environment. This nomena responsible for radiation hardening and degrada-
has been observed in recent work on steels under nuclear tion in ductility and fracture toughness of complex
SCW conditions [148,150,151] in terms of increasing corro- structural alloys (in particular RPV steels), and nuclear
sion resistance with increasing chromium content of the fuels innovations including investigation of further
alloy. improvements in the reliability of LWR fuel systems (can
The most daunting challenge for materials in the SCW an additional order of magnitude improvement be achieved
environment is resistance to SCC and IASCC. While on top of the three orders of magnitude improvement
nickel-base alloys and austenitic stainless steels are very accomplished over the past 40 years?), as well as explora-
resistant to corrosion in SCW, they are most susceptible tion of new LWR fuel systems with improved accident
to SCC. Intergranular stress corrosion cracking occurs tolerance without reducing the favorable performance
readily in high-purity, deaerated SCW at 400 °C and above and reliability features achieved by current fuels under nor-
in both austenitic stainless steels and nickel-base alloys. mal operating conditions. The eventual development of
Cracking severity increases exponentially with temperature advanced Generation IV fission reactor systems is directly
in both stainless steels and nickel-base alloys [152]. Over linked to successful resolution of several daunting materials
this same temperature range of 400–600 °C, ferritic–mar- challenges associated with the higher radiation damage lev-
tensitic alloys are resistant to SCC [151,153,154]. Further- els and/or operating temperatures for these concepts and
more, SCC of susceptible alloys is known to be aggressive coolants, all of which are detrimental to the reli-
exacerbated by persistent radiation damage of the metal ability of structural materials. The materials research chal-
[155]. lenges that need to be successfully resolved for fusion
Radiation effects on IGSCC are only now being investi- energy systems are even more daunting due to the high
gated for SCW conditions, yet results show that irradiation concentrations of H and He gases that will be produced
significantly increases the extent of SCC in stainless steels in the materials as a result of the high-energy fusion neu-
756 S.J. Zinkle, G.S. Was / Acta Materialia 61 (2013) 735–758

tron spectrum. A significant practical issue that is a barrier [25] Flament T, Tortorelli P, Coen V, Borgstedt HU. J Nucl Mater
to the development of structural materials for Generation 1992;191–194:132.
[26] Zinkle SJ, Ghoniem NM. Fusion Eng Des 2000;51–52:55.
IV fission and fusion energy concepts is the limited world- [27] ASTM E 521-96. Standard practice for neutron damage simulation
wide capability for high-flux materials irradiations; there by charged-particle irradiation; 2000.
are no high-intensity fusion neutron irradiation facilities, [28] Norgett MJ, Robinson MT, Torrens IM. Nucl Eng Des 1975;33:50.
and the number of fast fission test reactors is dwindling. [29] Byun TS, Farrell K. Acta Mater 2004;52:1597.
[30] Byun TS, Farrell K, Li M. Acta Mater 2008;56:1044.
[31] Byun TS, Farrell K, Li M. Acta Mater 2008;56:1056.
Acknowledgements [32] Chopra OK, Rao AS. J Nucl Mater 2011;412:195.
[33] Lucas GE. J Nucl Mater 1993;206:287.
The authors would like to acknowledge Roger Staehle [34] Rowcliffe AF, Zinkle SJ, Stubbins JF, Edwards DJ, Alexander DJ. J
for graciously providing several figures, and Jeremy Busby Nucl Mater 1998;258–263:183.
for comments on the draft manuscript. We also thank [35] Rowcliffe AF, Robertson JP, Klueh RL, Shiba K, Alexander DJ,
Grossbeck ML, et al. J Nucl Mater 1998;258–263:1275.
Robin Grimes (Imperial College) for helpful comments [36] Pawel JE, Rowcliffe AF, Alexander DJ, Grossbeck ML, Shiba K. J
on fuels and materials for nuclear reactors, and Bo Cheng Nucl Mater 1996;233–237:202.
(Electric Power Research Institute), Kurt Terrani and [37] Farrell K, Byun TS, Hashimoto N. J Nucl Mater 2004;335:471.
Lance Snead (ORNL), and Kemal Pasamehmetoglu [38] Gerold V, Karnthaler HP. Acta Metall 1989;37:2177.
(Idaho National Laboratory) for input on considerations [39] Hashimoto N, Byun TS, Farrell K. J Nucl Mater 2006;351:295.
[40] Sauzy M, Bavard K, Karlsen W. J Nucl Mater 2010;406:152.
for accident tolerant fuels. [41] Sharp JV. Radiat Eff 1972;14:71.
[42] Victoria M, Baluc N, Bailat C, Dai Y, Luppo MI, Schäublin R, et al.
References J Nucl Mater 2000;276:114.
[43] Wechsler MS. In: Reed-Hill RE, editor. The inhomogeneity of
[1] Chu S, Majumdar A. Nature 2012;488:294. plastic deformation. Metals Park, OH: American Society for Metals;
[2] Arunachalam VS, Fleischer EL. MRS Bull 2008;33:264. 1972. p. 19.
[3] World List of Nuclear Power Plants. Nucl News 2012;55:55. [44] Zinkle SJ, Singh BN. J Nucl Mater 2006;351:269.
[4] World Energy Outlook 2008. Paris: International Energy Agency; [45] Mills WJ. Int Mater Rev 1997;42:45.
2008. p. 211. [46] Odette GR, Lucas GE. J Nucl Mater 1991;179–181:572.
[5] Light Water Reactor Sustainability Program: Integrated Program [47] Rodchenkov BS, Kalinin GM, Strebkov YS, Shamardin VK,
Plan, INL/EXT-11-23452. Office of Nuclear Energy. US Depart- Prokhorov VI, Bulanova TM. J Nucl Mater 2009;386–388:277.
ment of Energy; 2012. [48] van Osch EV, Horsten MG, de Vries MI. J Nucl Mater 1998;258–
[6] Grimes RW, Konings RJM, Edwards L. Greater tolerance for 263:301.
nuclear materials. Nature Mater 2008;7:683. [49] Odette GR, Nanstad RK. PJOM 2009;61:17.
[7] Guerin Y, Was GS, Zinkle SJ. Materials challenges for advanced [50] Kenik EA, Busby JT. Mater Sci Eng B 2012;73:67.
nuclear energy systems. MRS Bull 2009;34:10. [51] Maziasz PJ. J Nucl Mater 1989;169:95.
[8] Zinkle SJ, Busby JT. Structural materials for fission and fusion [52] Zinkle SJ, Maziasz PJ, Stoller RE. J Nucl Mater 1993;206:266.
energy. Mater Today 2009;12:12. [53] Mansur LK. J Nucl Mater 1994;216:97.
[9] Marcus GH. Prog Nucl Energy 2000;37:5. [54] Singh BN, Zinkle SJ. J Nucl Mater 1994;217:161.
[10] Staehle R. Personal communication. [55] Wolfer WG. J Nucl Mater 1984;122 & 123:367.
[11] Nuclear Power Reactors in the World. Reference data series no 2. [56] Mansur LK, Lee EH. J Nucl Mater 1991;179–181:105.
Vienna: International Atomic Energy Agency; 2012. [57] Garner FA, Gelles DS. J Nucl Mater 1988;159:286.
[12] Garner FA, Toloczko MB, Sencer BH. J Nucl Mater 2000;276:123. [58] Holt RA. J Nucl Mater 1980;90:193.
[13] Was GS. Fundamentals of radiation materials science. New [59] Lemaignan C, Motta AT. In: Frost BRT, editor. Materials science
York: Springer; 2007. and technology: a comprehensive treatment, nuclear materials, vol.
[14] Zinkle SJ. Phys Plasmas 2005;12:058101. 10B. New York: VCH; 1994. p. 1.
[15] Schilling W, Ullmaier H. In: Frost BRT, editor. Materials science [60] Matthews JR, Finnis MW. J Nucl Mater 1988;159:257.
and technology: a comprehensive treatment. Nuclear materials. vol. [61] Woo CH. Radiat Eff Def Solids 1998;144:145.
10B. Weinheim: VCH; 1994. p. 179. [62] Allen TR, Busby JT. JOM 2009;61:29.
[16] Zinkle SJ. In: Konings RJM, editor. Comprehensive nuclear [63] Bloom EE. J Nucl Mater 1979;85 & 86:795.
materials, vol. 1. Amsterdam: Elsevier; 2012. p. 65. [64] Schroeder H, Stamm U. In: Packan NH, Stoller RE, Kumar AS,
[17] Danko JC. Corrosion in the nuclear power industry. Metals editors. Effects of radiation on materials: 14th international sympo-
handbook. Corrosion, vol. 13. Metals Park, OH: ASM Interna- sium. ASTM STP, vol. 1046. Philadelphia, PA: American Society for
tional; 1987. p. 927. Testing & Materials; 1990. p. 223.
[18] Feron D, Olive J-M, editors. Corrosion issues in light water reactors [65] Trinkaus H, Singh BN. J Nucl Mater 2003;323:229.
– stress corrosion cracking. Boca Raton, FL: CRC Press; 2007. [66] Ullmaier H. Nucl Fusion 1984;24:1039.
[19] Kilian E, Roth A. Mater Corros – Werkstoffe Korros 2002;53: [67] Integrity of reactor pressure vessels in nuclear power plants:
727. assessment of irradiation embrittlement effects in reactor pressure
[20] LeCalvar M, DeCurieres I. In: Feron D, editor. Nuclear corrosion vessel steels, report no NP-T-3.11. Vienna: International Atomic
science and engineering. Oxford: Woodhead Publishing Ltd.; 2012. Energy Agency; 2009.
p. 473. [68] Levelized Cost of New Generation Resources in the Annual Energy
[21] Was GS, Andresen PL. In: Feron D, editor. Nuclear corrosion Outlook 2012. Energy information administration. US Department
science and engineering. Oxford: Woodhead Publishing Ltd.; 2012. of Energy; 2012.
p. 131. [69] Generic Aging Lessons Learned (GALL) Report. NUREG-1801.
[22] Was GS, Ashida Y, Andresen PL. Corros Rev 2011;29:7. Rev 2. US Nuclear Regulatory Commission; 2010.
[23] Cabet C, Jang J, Konys J, Tortorelli PF. MRS Bull 2009;34:35. [70] Cawthorne C, Fulton EJ. Nature 1967;216:575.
[24] DeVan JH. J Nucl Mater 1979;85 and 86:249. [71] Mansur LK. J Nucl Mater 1993;206:306.
 S.J. Zinkle, G.S. Was / Acta Materialia 61 (2013) 735–758 757

[72] Garner FA. In: Tipping PG, editor. Understanding and mitigating [110] Nuclear Fuel Safety Criteria Technical Review. NEA report no
ageing in nuclear power plants. Boca Raton, FL: CRC Press; 2010. 7072, 2nd ed. Paris: Nuclear Energy Agency, OECD; 2012.
p. 308. [111] Forging a new nuclear safety construct: the ASME presidential task
[73] Porollo SI, Vorobjev AN, Konobeev YV, Dvoriashin AM, Krigan force on response to Japan nuclear power plant events. New York:
VM, Budylkin NI, et al. J Nucl Mater 1998;258–263:1613. American Society of Mechanical Engineers; 2012.
[74] Simmons KL, Hashemian HM, Ramuhalli P, Konnick R, Brenchley [112] Status and Near-term Prospects of Small and Medium Sized
DL, Ray S, et al. LWRS program non-destructive evaluation (NDE) Reactors. Vienna: IAEA; 2010.
R&D roadmap for determining remaining useful life of aging cables [113] Ingersoll DT. Nucl Technol 2012;178:125.
in nuclear power plants, PNNL-21731; 2012. [114] A Technology Roadmap for Generation IV Nuclear Energy
[75] Naus D. JOM 2009;61:35. Systems. Generation IV International, Forum; 2002.
[76] Locke DH. Nucl Eng Des 1975;33:94. [115] GIF R&D Outlook for Generation IV Nuclear Energy Systems.
[77] Simnad MT. J Nucl Mater 1981;100:93. Generation IV International, Forum; 2009.
[78] Edsinger K. Nucl News 2010;53:40. [116] Garner FA, Porter DL. Dimensional stability and mechanical
[79] Yang R, Cheng B, Deshon J, Edsinger K, Ozer O. J Nucl Sci behaviour of irradiated metals and alloys. An assessment of the
Technol 2006;43:951. swelling behavior of AISI 304 stainless steel, vol. 2. London: British
[80] Blake EM. Nucl News 2012;55:29. Nuclear Energy Society; 1983. p. 41.
[81] Very High Burn-Ups in Light Water Reactors. NEA report no 6224. [117] Dubuisson P, Gilbon D, Seran JL. J Nucl Mater 1993;205:178.
Paris: Nuclear Energy Agency, OECD; 2006. [118] Garner FA, Gelles DS. In: Packan NH, Stoller RE, Kumar AS,
[82] Edsinger K, Murty KL. JOM 2001;53:9. editors. Effects of radiation on materials: 14th international sympo-
[83] Nucl Eng Int 2004;49(September):26. sium, ASTM STP 1046, vol. II. Philadelphia: American Society for
[84] Olander DJ. Nucl Mater 2009;389:1. Testing and Materials; 1989. p. 673.
[85] Edsinger K, Stanek CR, Wirth BD. JOM 2011;63:53. [119] Toloczko MB, Garner FA, Eiholzer CR. J Nucl Mater 1994;212–
[86] Daum RS, Majumdar S, Liu Y, Billone MC. J Nucl Sci Technol 215:604.
2006;43:1054. [120] Garner FA, Toloczko MB. J Nucl Mater 1997;251:252.
[87] Bossis P, Pecheur D, Hanifi K, Thomazet J, Blat M. ASTM Int [121] Okamoto PR, Rehn LE. J Nucl Mater 1979;83:2.
2006;3:494. [122] Nelson RS, Hudson JA, Mazey DJ. J Nucl Mater 1972;44:318.
[88] Pan G, Long CJ, Garde AM, Atwood AR, Foster JP, Comstock RJ, [123] Lee EH, Maziasz PJ, Rowcliffe AF. In: Holland JR, Mansur LK,
et al. In: Proceedings of 2010 LWR fuel performance/topfuel/ Potter DI, editors. Phase stability during irradiation. New York: The
WRFPM, Orlando, FL; 2010. LaGrange Park, Illinois: American Metallurgical Society of AIME; 1981. p. 191.
Nuclear Society. Paper 74. [124] Boothby RM. J Nucl Mater 1996;230:148.
[89] Wikmark G, Hallstadius L, Yueh K. Nucl Eng Technol 2009;41:143. [125] Sklad PS, Clausing RE, Bloom EE. In: Shober FR, editor.
[90] Rondinella VV, Wiss T. Mater Today 2010;13:24. Irradiation effects on the microstructure and properties of met-
[91] Matzke H, Kinoshita M. J Nucl Mater 1997;247:108. als. Philadelphia, PA: American Society for Testing and Materials;
[92] Une K, Hirai M, Nogita K, Hosokawa T, Suzuwa Y, Shimizu S, 1976. p. 139.
et al. J Nucl Mater 2000;278:54. [126] Marwick AD. Nucl Instrum Methods 1981;182–183:827.
[93] Noirot J, Aubrun I, Desgranges L, Hanifi K, Lamontagne J, Pasquet [127] Rowcliffe AF, Mansur LK, Hoelzer DT, Nanstad RK. J Nucl Mater
B, et al. Nucl Eng Technol 2009;41:155. 2009;392:341.
[94] Matzke H, Rondinella VV, Wiss T. J Nucl Mater 1999;274:47. [128] Lee EH, Mansur LK. Philos Mag A 1990;61:733.
[95] Terrani KA, Snead LL, Gehin JC. J Nucl Mater 2012;427:209. [129] Boutard JL, Alamo A, Lindau R, Rieth M. CR Phys 2008;9:287.
[96] Spino J, Santa Cruz H, Jovani-Abril R, Birtcher RC, Ferrero C. J [130] Brager HR, Garner FA. J Nucl Mater 1978;73:9.
Nucl Mater 2012;422:27. [131] Klueh RL, Shingledecker JP, Swindeman RW, Hoelzer DT. J Nucl
[97] Staehle RW, Gorman JA. In: 10th international conference on Mater 2005;341:103.
environmental degradation of materials in nuclear power systems- [132] McClintock DA, Sokolov MA, Hoelzer DT, Nanstad RK. J Nucl
water reactors. Lake Tahoe, CA: NACE; 2001. Special bonus paper. Mater 2009;392:353.
[98] Steam Generator Progress Report. EPRI steam generator data- [133] Miller MK, Hoelzer DT. J Nucl Mater 2011;418:307.
base. Palo Alto, CA: Electric Power Research Institute; 2011. [134] Odette GR, Alinger MJ, Wirth BD. Annu Rev Mater Res
[99] Paraventi DJ, Moshier WC. In: 13th International conference on 2008;38:471.
environmental degradation of materials in nuclear power sys- [135] Ukai S, Fujiwara M. J Nucl Mater 2002;307–311:749.
tems. Whistler, British Columbia, Canada. Toronto: Canadian [136] Bonal JP, Kohyama A, van der Laan J, Snead LL. MRS Bull
Nuclear Society; 2007. 2009;34:28.
[100] Jiao Z, Was GS. J Nucl Mater 2011;408:246. [137] Burlet H, Genzbittel J-M, Cabet C, Lamagnére P, Blat M, Renaud
[101] Nishioka H, Fukuya K, Fujii K, Kitsunai Y. J Nucl Sci Technol D, et al. Evaluation of nickel-based materials for VHTR heat
2008;45:274. exchanger structural materials for innovative nuclear systems
[102] Was GS, Andresen PL. In: Shirzadi A, Wallach R, Jackson S, (SMINS). Karlsruhe: OECD Nuclear Energy Agency; 2008. p. 79.
editors. Structural alloys for power plants: operational challenges [138] Cabet C, Terlain A, Lett P, Guetaz L, Genzbittel J-M. Mater Corros
and high-temperature materials. Cambridge: Woodhead Publishing, – Werkstoffe Korros 2006;57:147.
in press [chapter 19]. [139] Rouillard F, Cabet C, Wolski K, Pijolat MJ. Oxid Metals
[103] Asher RC, Davies D, Kirstein TBA. J Nucl Mater 1973;49:189. 2007;68:133.
[104] Bradhurst DH, Shirvington PJ, Heuer PM. J Nucl Mater 1973;46:53. [140] Yun HM, Ennis PJ, Nickel H, Schuster H. J Nucl Mater
[105] Lewis MB, Hunn JD. J Nucl Mater 1999;265:325. 1984;125:258.
[106] Cox B. J Nucl Mater 1968;28:1. [141] Natesan K, Shankar PS. J Nucl Mater 2009;394:46.
[107] Phythian WJ, English CA. J Nucl Mater 1993;205:162. [142] Shankar PS, Natesan K. J Nucl Mater 2007;366:28.
[108] Nanstad RK, Odette GR, Sokolov MA. In: Proceedings of the [143] Kurata Y, Tanabe T, Mutoh I, Tsuji H, Hiraga J, Shindo M, et al. J
ASME pressure vessels and piping conference 2010, vol. 9; 2010. p. Nucl Sci Technol 1999;36:1160.
125. [144] Cook RH. Nucl Technol 1984;66:283.
[109] Nuclear Fuel Behaviour under Reactivity-initiated Accident (RIA) [145] Huchtemann B. Mater Sci Eng 1989;A121:623.
Conditions. NEA report no 6847. Paris: Nuclear Energy Agency, [146] Zhou R, West EA, Jiao Z, Was GS. J Nucl Mater 2009;395:11.
OECD; 2010.
758 S.J. Zinkle, G.S. Was / Acta Materialia 61 (2013) 735–758

[147] Wright IG, Tortorelli PF, Schütze M. Oxidation growth and [153] Ampornrat P, Gupta G, Was GS. J Nucl Mater 2009;395:30.
exfoliation on alloys exposed to steam. EPRI, report no 1013666; [154] Hirose T, Shiba K, Enoeda M, Akiba M. J Nucl Mater 2007;367–
2007. 370:1185.
[148] Was GS, Ampornrat P, Gupta G, Teysseyre S, West EA, Allen TR, [155] Was GS, Zhou R, West EA, Jiao Z. In: Proceedings of the 14th
et al. J Nucl Mater 2007;371:176. International Conference on Environmental Degradation of Mate-
[149] Kofstad P. High temperature corrosion. Amsterdam: Elsevier; 1988. rials in Nuclear Power Systems. Lagrange Park, IL: American
[150] Allen TR, Tan L, Chen Y, Ren X, Sridharan K, Was GS, et al. In: Nuclear Society; 2010. p. 1679.
Proceedings of global 2005, Tsukuba, Japan; 2005. Paper 419. [156] Teysseyre S, Jiao Z, West EA, Was GS. J Nucl Mater 2007;371:107.
[151] Hwang SS, Lee BH, Kim JG, Jang J. J Nucl Mater 2008;372:177. [157] Teysseyre S, Was GS. In: Allen TR, Busby JT, King PJ, editors. 13th
[152] Allen TR, Tan L, Chen Y, Ren X, Sridharan K, Was GS, et al. In: International conference on degradation of materials in nuclear
Konings RJM, editor. Comprehensive nuclear materials, vol. power systems – water reactors. Whistler, British Columbia,
5. Amsterdam: Elsevier; 2012. p. 279. Toronto: Canadian Nuclear Society; 2007.