Versatile Oxide Films Protect FeCrAl Alloys Under Normal Operation and Accident Conditions in Light Water Power Reactors
Versatile Oxide Films Protect FeCrAl Alloys Under Normal Operation and Accident Conditions in Light Water Power Reactors
The US has currently a fleet of 99 nuclear power light water reactors which generate approximately 20% of the electricity consumed in the country. Near 90% of the reactors are at least 30 years old. There are incentives to make the existing reactors safer by using accident tolerant fuels (ATF). Compared to the standard UO2-zirconium-based system, ATF need to tolerate loss of active cooling in the core for a considerably longer time while maintaining or improving the fuel performance during normal operation conditions. Ferritic iron-chromium-aluminum (FeCrAl) alloys have been identified as an alternative to replace current zirconium alloys. They contain Fe (base) + 10-22 Cr + 4-6 Al and may contain smaller amounts of other elements such as molybdenum and traces of others. FeCrAl alloys offer outstanding resistance to attack by superheated steam by developing an alumina oxide on the surface in case of a loss of coolant accident like at Fukushima. FeCrAl alloys also perform well under normal operation conditions both in boiling water reactors and pressurized water reactors because they are protected by a thin oxide rich in chromium. Under normal operation condition, the key element is Cr and under accident conditions it is Al.
RAUL B. REBAK
Worldwide, the generation of electric power has
several sources of energy that can be grouped as: (
fossil fuels (coal, petroleum and natural gas), (
nuclear and (
) renewable (wind, solar,
hydroelectric, geothermal, biomass, etc.) sources. Figure 1
shows that the world energy consumption in the
next two decades will be still dominated ( 80%) by
the burning of fossil fuels (liquid, gas and coal).
Nuclear energy represents only 6% of the energy
consumed worldwide. In the USA, 30% of the
consumed natural gas is used to generate about
20% of electrical power produced in the country. On
the other hand, 100% of the nuclear energy
produced is used to generate electricity. This nuclear
electricity also represents about 20% of all the
electrical power produced in the country. The
president of the Nuclear Energy Institute in the
US said that one of her top priorities is to ensure the
continuous ‘‘safe, reliable operation of the US
commercial nuclear reactors’’ to avoid their
premature retirement, which would be replaced by the
burning of more natural gas, increasing the
emissions of greenhouse gases.2 Currently, there are 99
operating power reactors in the US, 34 boiling water
reactors (BWR) and 65 pressurized water reactors
(PWR).3 Figure 2 shows in which year these power
reactors were connected to the grid, affirming that
89% of the current reactors are at least 30 years old
and 45% of the reactors are at least 40 years old.
Only one new reactor was connected to the grid in
the last 20 years. Currently, four new reactors are
under construction in the USA. The existing nuclear
power plants (NPP) in the USA are an aging
infrastructure, which are investing a bare minimum
in upgrades and maintenance.4 It is important to
make these NPP safer in their operation for the
remaining of their useful life.
After the Fukushima NPP black-out following the
tsunami on March 2011, the international
community has been dedicated to finding a fuel
configuration that will be more resistant to a loss of coolant
accident (LOCA) than the existing zirconium/UO2
pellets. The massive building explosions at the
Fukushima plant were caused by the ignition of
hydrogen gas formed by the fast-exothermic
reaction of zirconium with water (or steam).
Zr þ 2H2O ¼ ZrO2 þ 2H2 þ Heat
There is currently an international research effort
to delay the generation of hydrogen gas and to
expand the coping time allowed for quenching or
cooling the reactor after a severe LOCA. There are
several proposals on how to design an accident
tolerant fuel (ATF) or an advanced technology fuel
(ATF) that will permit for a safer operation of the
existing light water reactors.5,6 The ATF proposals
include the use of iron-chrome-aluminum (FeCrAl),
silicon carbide (SiC), and coated zirconium or
molybdenum (Mo) claddings. FeCrAl is the concept
that would provide the fastest implementation for
the safe operation of the remaining NPP for another
40 years.7 The proposed ATF configuration will use
the current UO2 fuel clad with a thin wall (less than
0.5 mm thick) FeCrAl tube. Eventually, other fuels
such as uranium silicide may be industrially
developed for the FeCrAl cladding.
Oxidation Behavior of FeCrAl
One of the requirements for ATF from the US
Department of Energy is that the cladding must
perform as well as or better than zirconium under
normal operation conditions and much better than
zirconium under severe accident conditions.8 That
is, under normal operation conditions, the cladding
must do well in high-purity water at 288 C for
BWRs and at 330 C for PWRs. The water may
contain dissolved hydrogen for corrosion control.
Under accident conditions, the cladding must do
well in superheated steam at 1200 C or higher
temperatures. That is, under a LOCA situation, the
cladding may not be able to oxidize through the wall
for several hours, keeping its geometry intact for
quenching operations with fresh water.
The composition of the FeCrAl alloys considered
could cover the following range: Fe-based, plus 12–
21 mass percent (%) of Cr, plus 4–6% Al plus 2–3%
Mo, with parts per million level of other elements
such as yttrium, hafnium, zirconium, etc. Some
FeCrAl alloys are made by powder metallurgy (e.g.,
APMT) and others by traditional melting, casting
and forging (e.g., C26 M and Aluchrom 418 YHf).
FeCrAl Behavior Under Normal Operation
Coupons of FeCrAl APMT (powder metallurgy
Fe + 21Cr + 5Al + 3Mo) were immersed for 1 year
in high-temperature, high-purity 18 MX water
(Table I) and the mass (weight) change was
monitored as a function of the immersion time.
The water conditions in Tables I and II are
simulated conditions to study the effect of either
oxygen or hydrogen. It is not meant to represent
actual plant conditions. Figure 3 shows the mass
change for APMT coupons tested in 288 C
highpurity water containing either excess oxygen or
excess hydrogen (Table I). Under oxidizing
conditions, the coupons gain mass due to the formation of
thicker dual layer oxide on the surface. As the
immersion time progressed, the mass gain became
smaller because of the formation of a protective
inner layer of chromia. The external oxide may
slowly dissolve in the oxygenated water. Under
hydrogen conditions, the coupons lose initially some
mass, but after a single layer chromia layer forms
on the coupon surface, the mass loss is stabilized.
Figure 4 shows comparatively the mass loss
between the BWR (288 C) and PWR (330 C)
hydrogenated environments. The mass loss was higher in
the higher PWR temperature environment.
Figure 5 and Table II shows the characteristics of
the oxides which form naturally on a FeCrAl APMT
material exposed to simulated light water reactor
environments. The chromium-rich oxides act as a
barrier for the further dissolution of the material as
shown by the low mass changes on the coupons
(Figs. 3 and 4). The results presented here suggest
that FeCrAl APMT has excellent environmental
resistance characteristics under normal operation
for both BWR and PWR coolants. There is no need to
change the water chemistry of the coolants since
FeCrAl is compatible with the current water
chemistries in light water reactors.
FeCrAl Behavior Under LOCA and Severe
Accident Conditions of NPP
The main reason FeCrAl alloys have been selected
for ATF cladding is because they have a superior
oxidation resistance in superheated steam in the
event of LOCA or a severe accident. Chromium
provides protection against oxidation in air or in
steam to all stainless steels and FeCrAl alloys. As
the temperature increases beyond 1000–1100 C,
chromium does not effectively protect these alloys
due to the evaporation of Cr2O3 in the environment.
Beyond 1100 C, the alloys must contain
approximately 4–6% Al to offer protection. The way the
FeCrAl alloys work is by the initial formation of a
Cr2O3 oxide on the surface. As the temperature
increases, a continuous thin alumina layer (Al2O3)
develops underneath the Cr2O3 film. Eventually,
PWR 330°C, H2
BWR 288°C, H2
BWR 288°C, O2
Fig. 5. Composition of the oxide film formed on the surface of APMT exposed to 330 C and 288 C water with hydrogen and 288 C water with
oxygen. Chromium oxide always develops.
the Cr2O3 oxide evaporates and the alumina layer
protects the material up to its melting point of
approximately 1500 C. The combined synergistic
action between Cr and Al in FeCrAl at very high
temperatures has been known for almost a century.
Figure 6 shows that, when an FeCrAl APMT tube
was exposed to steam at 1200 C for 4 h, it developed
an approximately 1-lm-thick oxide which was pure
alumina (it did not contain Cr, Fe or Mo). FeCrAl
alloys such as APMT generally contain small
amounts (less than 1%) of rare earth elements that
may help peg the protective oxide to the alloy
substrate. The FeCrAl producers such as Sandvik
for APMT may sell the alloy in the pre-oxidized
condition. Figure 6 shows that a pre-oxidized strip
of APMT also contains an approximately
0.5-lmthick alumina layer on the surface formed by
exposure to air at 1050 C for 8 h. The oxide formed
at 1050 C seems to contain residual amounts of Cr
on its outer layer.
It was shown in this work that, when freshly
fabricated specimens of FeCrAl alloys are exposed
to 300 C water, they develop a protective Cr2O3
oxide on the surface. Similarly, if these freshly
fabricated specimens are exposed to > 1000 C
steam or air, they form a protective alumina on
the surface. Figure 7 shows schematically the oxide
development in both scenarios (normal operation or
accident) described. A pre-oxidation treatment may
be considered for FeCrAl tubes since the presence of
an alumina layer will act as a barrier for tritium
release into the coolant.
Two questions are repeatedly asked about the
oxidation behavior of FeCrAl.
1. After the FeCrAl cladding is in high
temperature water for several months, with a Cr oxide
on its surface (Table II), would it be able to
develop a protective alumina layer on the
surface in the case of an accident?
2. If a pre-oxidation treatment is applied to the
FeCrAl tube cladding, what happens to the
alumina layer in the OD of the tube in contact
with the water? Does it dissolve? Would a Cr
oxide be able to develop on the surface (Table II)
after the alumina dissolves?
APMT tube 1200°C 4h steam
APMT Strip 1050°C 8h air SEM O
Fig. 6. continued
Question 1: Steam Oxidized APMT Tubes
Exposed to High Temperature Water
Figure 8 shows the mass change (loss) as a function
of immersion time for APMT tube specimens
exposed to both BWR (288 C) and PWR (330 C)
hydrogenated water. Each point is the average
value for two or more specimens. The mass loss for
the AR tube specimens was the same as for the AR
flat specimens (Figs. 3 and 8). The tube specimens
which were pre-oxidized (PO) in steam initially had
a higher mass change (loss) than the tube specimens
which were as-received (AR). However, once the
alumina layer dissolved in the water, the mass loss
of the PO specimens practically stopped, and the
mass loss rate became the same as for the AR
specimens. The surface area of each PO specimen
tube in Fig. 8 was approximately 3.7 cm2. When
pre-oxidized in steam at 1200 C for 2 h, the tube
specimens developed a 1-lm-thick alumina layer on
the surface (Fig. 6). The volume of the alumina
layer is 3.7 cm2 9 0.0001 cm = 0.00037 cm3
(volume of aluminum oxide). The mass of aluminum
oxide is m = V 9 density alumina = 0.00037 9
3.95 g/cm3 = 1.46 mg. Dividing the mass of
1.46 mg by the surface area of 3.7 cm2 = 0.4 mg/
cm2, which is exactly what the Fig. 8 plot shows as
the mass loss of the PO tube specimens due to the
dissolution of the alumina layer. This is a clear proof
that the initial higher mass change is just the
dissolution of the alumina, and then the dissolution
stops because a Cr2O3 should develop on the
Fig. 8. Mass change (loss) for APMT tube specimens immersed in
hydrogenated BWR and PWR water. The open symbols show
asreceived (AR) specimens and the closed symbols pre-oxidized (PO)
specimens. The round symbols are tube specimens and the square
and diamond symbols are flat specimens. The specimens that
contained an alumina layer on the surface initially had a higher mass
loss, due to the dissolution of the alumina in the hot water.
Figure 9 shows the oxide composition formed on
the surface of a APMT tube specimen after
immersion in PWR hydrogenated water at 330 C for
284 days. These tube specimens were originally
pre-oxidized in steam at 1200 C for 2 h, that is,
they originally contained a 1-lm-thick alumina
layer on the surface (Figs. 6 and 8). Figure 9 shows
that the alumina layer from the surface of the
preoxidized tube specimen dissolved in water and that
a chromium-rich oxide developed, similarly as for a
specimen which is as-prepared (not pre-oxidized)
Question 2: Water Oxidized APMT Tubes
Exposed to Superheated Steam
Figure 10 shows the oxidation resistance for 4 h
in superheated steam at 1200 C of the APMT tube
specimens, both AR and pre-exposed to
high-temperature water for 73 days. Figure 10 shows that
specimens which were immersed in
high-temperature water for 73 days had the same resistance to
superheated steam as fresh AR specimens. That is,
fuel rods clad with FeCrAl alloys in an operating
plant will resist attack by superheated steam in the
unlikely event of a LOCA. For both the AR tube
specimens and for specimens pre-exposed to
hightemperature water, the oxidation rate in
superheated steam was in the order of 0.25–0.35 mg/cm2.
Figure 11 shows schematically the evolution of
the oxides on the surface of FeCrAl alloys from
water to steam and steam to water. The alloy was
designed with the versatility to develop a protective
oxide in each situation.
Stainless steels (e.g., austenitic type 304 SS)
contain at least 13% of chromium so they can offer
passivity in most industrial applications. The
longlasting passivity is a slowing down of the corrosion
rate of the steel by the formation of a thin protective
and adherent layer of chromium oxide on the
surface. The passivity provided by chromium oxide
of the stainless steels and nickel-based alloys also
applies to FeCrAl alloys in light water reactor
environments at temperatures near 300 C.9
Nuclear power reactors have been successfully
using stainless steels and nickel alloys with
chromium in hot water for over 60 years. Chromium is
the element that provides the resistance in hot
water by the formation of a protective oxide film on
Since 2012, it has been proposed to use ferritic
FeCrAl alloys as an alternative for fuel cladding so
they can provide resistance to a severe accident
situation.11,12 Ferritic FeCrAl alloys have never
been used in light water reactors before. The FeCrAl
also contain chromium and it was expected and
shown here and elsewhere that they will also
develop a thin chromium oxide on the surface to
protect against high-temperature water attack.
FeCrAl alloys also have about 4–6% of aluminum
which will ‘‘act’’ only in the case of a severe accident.
If a plant never has an accident, the aluminum will
just sit and wait. It does not participate in the
passive film formed in high-temperature water.
Aluminum may never be needed since, for the
normal operation conditions, only the chromium is
needed. However, we have also shown here that if
an unlikely LOCA happens, the aluminum will form
the protective alumina layer required to resist the
attack by steam. Chromium can protect the
cladding tubes against steam only until near 1100 C. At
higher temperatures, aluminum offers the
protection to the tubes until their melting point. The
synergistic effect between Cr and Al in FeCrAl
makes this material ideal for both situations,
normal operation conditions and ready in the case of an
Fig. 11. Schematic representation of the oxides evolution on FeCrAl alloys under normal operation and accident conditions of a light water
1. Iron-chrome-aluminum alloys (FeCrAl) contain
both Cr and Al making it an ideal material to
replace zirconium in light water reactors’ fuel
claddings. Cr protects the alloy under normal
operation conditions and Al protects the alloy at
temperatures higher than 1100 C.
2. Under normal operation conditions, in a
hydrogenated water environment, a thin layer of Cr
oxide forms on the surface of the alloy slowing
down the degradation process. Under oxidizing
conditions, a dual layer may form. The inner
layer is the protective layer of Cr oxide.
3. If the FeCrAl tube has an alumina layer on the
surface, it will dissolve in high-temperature
( 300 C) water and a protective Cr oxide will
form in its place.
4. If a FeCrAl tube has a Cr oxide on the surface
and it is exposed to accident steam conditions,
the Cr oxide layer will evaporate and an
alumina layer will form to protect the tube.
This material is based upon work supported by
the Dept. of Energy (National Nuclear Security
Administration) under Award Number
DENE0008221. This report was prepared as an account
of work sponsored by an agency of the United States
Government. Neither the United States
Government nor any agency thereof, nor any of their
employees makes any warranty, express or implied,
or assumes any legal liability or responsibility for
the accuracy, completeness, or usefulness of any
information, apparatus, product, or process
disclosed, or represents that its use would not infringe
privately owned rights. Reference herein to any
specific commercial product, process or service by
trade name, trademark, manufacturer, or otherwise
does not necessarily constitute or imply its
endorsement, recommendation, or favoring by the
United States Government or any agency thereof.
The views and opinions of authors expressed herein
do not necessarily state or reflect those of the United
States Government or any agency thereof.
This article is distributed under the terms of the
Creative Commons Attribution 4.0 International
which permits unrestricted use, distribution, and
reproduction in any medium, provided you give
appropriate credit to the original author(s) and the
source, provide a link to the Creative Commons
license, and indicate if changes were made.
1. International Energy Outlook 2016. (US Energy Information Administration , 2016 ), https://www.eia.gov/todayine nergy/detail.php? id=26212. Retrieved 18 Aug 2017 .
2. M. Korsnick , in press conference 20 Oct 2016 ( Nuclear Energy Institute), www .nei.org/News-Media.
3. IAEA (International Atomic Energy Agency ), PRIS (Power Reactor Information System) , www.iaea.org.
4. J. Tollefson , Nature 537 , 16 ( 2016 ).
5. C. Tang , M. Stueber , H.J. Seifert , and M. Steinbrueck , Corros. Rev. 35 , 141 ( 2017 ).
6. F. Nagase , K. Sakamoto , and S. Yamashita , Corros. Rev. 35 , 129 ( 2017 ).
7. R.B. Rebak , W.P. Gassmann , and K.A. Terrani , in Paper A0042, IAEA, Top Safe 2017 conference, 12-16 February 2017 , Vienna, Austria.
8. S.M. Bragg-Sitton , M. Todosow , R. Montgomery , C.R. Stanek , R. Montgomery , and W.J. Carmack , Nucl. Technol. 195 , 111 ( 2016 ).
9. K.A. Terrani , B.A. Pint , Y.-J. Kim , K.A. Unocic , Y. Yang , C.M. Silva , H.M. Meyer III , and R.B. Rebak , J. Nucl . Mater. 479 , 36 ( 2016 ).
10. R.B. Rebak , M. Larsen , and Y. -J. Kim , Corros. Rev. 35 , 177 ( 2017 ).
11. R.B. Rebak , Metall. Trans. E 2 , 197 ( 2015 ).
12. K. Sakamoto , M. Hirai , S. Ukai , A. Kimura , A. Yamaji , K. Kusagaya , T. Kondo , and S. Yamashita, in WRFPM 2017 Conference (Jeju Island , September, 2017 ).