i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 0 ( 2 0 1 5 ) 5 3 0 5 e5 3 1 2

Available online at www.sciencedirect.com

ScienceDirect

journal homepage: www.elsevier.com/locate/he

Dual atmosphere study of the K41X stainless steel

for interconnect application in high temperature
water vapour electrolysis

M.R. Ardigo a, I. Popa a,*, L. Combemale a, S. Chevalier a, F. Herbst a,

P. Girardon b
a
ICB, UMR 6303 CNRS-Universite de Bourgogne, 21078 Dijon Cedex, France
b
APERAM, Centre de Recherche, 62330 Isbergues, France

article info abstract

Article history: High temperature water vapour electrolysis (HTE) is one of the most efficient technologies
Received 11 July 2014 for mass hydrogen production. A major technical difficulty related to high temperature
Received in revised form water vapour electrolysis is the development of interconnects working efficiently for a long


1 December 2014 period. Working temperature of 800 C enables the use of metallic materials as in-
Accepted 20 January 2015 terconnects. High temperature corrosion behaviour and electrical conductivity of a com-
Available online 21 February 2015 mercial stainless steel, K41X (AISI 441), were tested in HTE dual atmosphere (95%O2-5%H20/
10%H2-90%H2O) at 800
C. The alloy exhibits a very good oxidation resistance compared to
Keywords: single atmosphere tests. However, a supplied electrical current significantly changes the
High temperature electrolysis nature of the oxides that form during the test. A very good Area Specific Resistance (ASR)
Interconnect parameter was measured in dual atmosphere, much lower than the values obtained in
Dual atmosphere single atmosphere tests.
Corrosion behaviour Copyright © 2015, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights
Electrical current influence reserved.

temperature and a good electrical conductivity with an area

Introduction
High temperature water vapour electrolysis (HTE) is one of the
most efficient technologies for mass hydrogen production. It
works as an inverse fuel cell, using air at the anode side and
water vapour at the cathode side. To increase hydrogen pro-
duction, cells are overlapped, to create a stack. Stack con-
struction requires the presence of interconnects, which
physically connect the anode of one individual cell to the
cathode of the neighbouring one.
The most important criteria required for the interconnect
are a good oxidation resistance in air and in H2/H2O at high
specific resistance (ASR) below 0.1 U cm2 for 25000 h (expected
service lifetime) [1,2]. The targeted operating temperature of
800
C enables the use of metallic interconnects and ferritic
stainless steels seem to be the best option [3,23].
A few works have only been done in HTE conditions. As
HTE is a reverse process compared to SOFC, literature data
dealing with SOFC interconnects can be considered. Several
works have been performed in O2/H2O gas environment, but
only a few studies have been done under H2/H2O atmo-
spheres. However, tests in dual atmosphere, by exposing the
interconnect at the anode atmosphere at one side and at the

* Corresponding author. Tel.: þ33 (0)380396159; fax: þ33 (0)380396132.

E-mail address: ioana.popa@u-bourgogne.fr (I. Popa).
http://dx.doi.org/10.1016/j.ijhydene.2015.01.116
0360-3199/Copyright © 2015, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
5306 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 0 ( 2 0 1 5 ) 5 3 0 5 e5 3 1 2
cathode atmosphere at the other side, are required in order to
do a more realistic evaluation of the interconnect material.
The SOFC literature reports very few studies done in dual at-
mosphere and they all suggest that the nature of the oxide
scale that grows at the O2 side is influenced by the H2 gas
present at the other side, while any evolution was not
observed at the H2 side [4e8].
In a previous study performed in HTE single atmospheres
[9], the high temperature corrosion resistance and the elec-
trical conductivity of a commercial ferritic stainless steel with
18% Cr-content, the K41X (AISI 441), were assessed in both
anode (95%O2e5%H2O) and cathode atmosphere (10%H2e90%
H2O) at 800
C on as-rolled samples. In both atmospheres, the
formation of a duplex oxide scale was revealed. In O2-H2O, a
Cr-Mn spinel-type oxide grew at the top of an inner chromia
layer, of about 0.2e1.4 mm after 100 h of exposition. In H2-H2O,
an inner layer consisting of a Cr-Fe mixed oxide and an outer
layer formed by a magnetite-type iron oxide were observed,
with a total thickness of about 2e3 mm. The result obtained in
single H2-H2O atmosphere was very surprising and not in
agreement with the literature data concerning chromia-
forming alloys in general [10] and K41X alloy in particular
[11]. This specific behaviour could be related to the mechani-
cal surface state [12].
In parallel, electrical resistivity measurements were per-
formed in both HTE atmospheres during 100 h at 800
C. A
rather high, non-suitable, ASR parameter value (0.45 U cm2)
was obtained in the cathode atmosphere, while a better
behaviour, but still too high (0.1 U cm2) was observed in the
anode atmosphere [13].
In this paper, we performed very new and innovative dual
atmosphere tests in HTE conditions on the K41X material, the
interconnect being exposed at one side to H2-H2O and at the
other side to O2-H2O. In a first type of tests, the interconnect
reactivity at 800
C was evaluated in dual atmosphere and
compared to single atmosphere results. In parallel, to better
approach functioning situation, a second type of tests was
done by applying an electrical current through the intercon-
nect during the dual atmosphere exposition. The possible in-
fluence of the electrical current on the oxidation phenomena
was checked by comparing microstructural analysis results
obtained during the two different tests.
Material and methods
The material selected for this study, K41X, is a commercial
stainless steel with 18% Cr content (Fee- 18%Cr e 0.58%Si e
0.52%Nb e 0.25%Mn e 0.14%Ti), provided by Aperam Isber-
gues, France. Prior to dual atmosphere tests, the as-rolled
samples were cut into discs of 24 mm diameter and 1 mm
thick. Specimens were ultrasonically cleaned in ethanol and
dried at room temperature.
Dual atmosphere tests were performed at 800
C by using a
commercial set-up (ProboStat™), distributed by NorECs (Nor-
wegian Electro Ceramics AS). The sample is placed between
two Pt grid electrodes, that are assembled together by using a
resort system. This assembly is placed on top of an alumina
tube and then in the middle of a vertical furnace. The system's
sealing is insured by a silver paste put between the sample
and the alumina tube. In this way, one surface of the sample is
exposed at the anode atmosphere (95%O2e5%H2O) while the
other surface is simultaneously exposed to the cathode at-
mosphere (10%H2e90%H2O). An electrical current can be
applied and the resulting voltage can be measured. The Ohm's
law allows to calculate the resistance of the growing oxide
layer (R ¼ V/I). The ASR parameter is obtained by multiplying
this resistance with the interconnect surface. These conduc-
tivity measurements can be performed in-situ during the dual
atmosphere exposure of the interconnect. The value of the
supplied electrical current was identical to single atmosphere
tests, 150 mA, corresponding to a current density of
130 mA cm2.
For safety reasons, on the cathode side of the set-up, H2
was diluted in Ar and the real gas composition used was Ar-1%
H2e9%H2O (that allows maintaining the H2-H2O ratio con-
stant). Gas mixture (Ar/1%H2 or O2) was passed through
boiling distilled water. The H2O fraction (9% or 5% respec-
tively) was controlled and adjusted by bubbling gas through a
vertical tube cooled at a well-defined temperature.
The oxide surface and cross-section morphologies, as well
as the chemical composition of the corrosion products, were
analysed using a JEOL JSM-6400F scanning electron micro-
scope. Phase composition of the oxide scales was determined
by X-ray diffraction (XRD) with an Inel CPS 120 diffractometer
using Cu Ka (l ¼ 0.154 nm) radiation with a fixed incidence
angle of 3
.
Results and discussion
Dual atmosphere influence on the oxidation behaviour (test
without electrical current)
Fig. 1 shows the surface and cross-section morphologies of the
K41X alloy after ageing during 100 h at 800
C in dual
atmosphere.
Dual atmosphere influence at the O2-H2O side (HTE anode side)
At the anode side (Fig. 1a), the SEM images show a compact
oxide layer, without any cracks or gap. The corrosion scale
consists in some pyramidal grains randomly distributed on a
continuous and compact layer made of smaller grains. The
XRD pattern showed the presence of two oxides: (Cr,Mn)3O4
and Cr2O3. Cross-section observations (Fig. 1b) together with
EDX analysis confirm the presence of a duplex oxide scale: an
inner scale formed by protective chromia and an outer scale
mainly formed by the Cr-Mn spinel type oxide grains. The
oxide scale is adherent to the substrate and its thickness is
found to vary between 0.4 mm and 1.5 mm.
At the O2-H2O side of the dual atmosphere exposition, the
oxide scale composition, morphology and thickness are very
similar to those obtained in single atmosphere. This result is
very interesting, as the few studies that were performed in
dual SOFC conditions at 800
C report an anomalous oxidation
at the air side (dry air or moist air). The grown scales are
significantly different of those grown on samples exposed to
air on both sides [4,5]. This anomalous oxidation was related
to the transport of hydrogen through the steel (that was
proved to be over 60 mm in 1 h at 800
C) [14]. In consequence,
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 0 ( 2 0 1 5 ) 5 3 0 5 e5 3 1 2
Fig. 1 e SEM observations after dual atmosphere exposition of the K41X alloy during 100 h at 800
C: (a) surface at the anode
side (O2-H2O); (b) cross-section at the anode side; (c) surface at the cathode side (H2-H2O); (d) cross-section at the cathode
side.
an important increase of the Fe concentration or the forma-
tion of Fe-rich oxide nodules within the air side scale can be
observed [15]. This effect strongly depends on the alloy
composition: it becomes less pronounced when the Cr content
increases [5,6]. Nevertheless, even for high Cr content (~27%),
where any unusual phases are not observed, the air side scales
are less dense than in single atmosphere [5,6].
However, not only the Cr content is an important factor,
but also the minor alloying elements. For instance, the K41X
alloy exhibits substantially accelerated and anomalous air-
side corrosion related to its Nb content, as Nb is known to
have a high solubility for hydrogen. By segregating at the grain
boundaries, it can serve as a short-circuit diffusion pathway
for hydrogen within the steel [7,15].
In our study, even if the tests were done with the same
duration and temperature conditions (100 h at 800
C) as most
of the above cited studies, the anomalous oxidation behaviour
was not observed at the O2-H2O side of the dual HTE exposi-
tion, as can be observed in Fig. 2. EDX mapping was done on
the oxide scale that grew on this side of the K41X sample. The
presence of Cr and Mn can be noticed inside the oxide scale;
however Fe signal is absent, proving that no Fe diffusion took
place during the experiment. Several hypotheses were taken
into account to explain this very good behaviour.
In the SOFC literature, the Fe transport through the scale
formed at the air side is strongly related to the hydrogen
diffusion from the opposite side through the entire metal
thickness. We can suppose that in our case, the low H2 partial
5307
pressure present at the H2-H2O side (1% against 97% in normal
SOFC dual experiments) can be discussed. However, Zhao
et al. [16] and Yang et al. [5] obtained similar results at the air
side of the K41X alloy after dual experiments that were per-
formed in Ar-5%H2e3%H2O/dry air and respectively in 97%
H2e3%H2O/dry air. In both cases, the anomalous oxidation
occurred, leading to Fe-rich nodules formation.
Another parameter taken into account in the SOFC litera-
ture is the H2 flow rate. Zhao et al. [16] showed that the
number of Fe-rich nodules becomes higher with increasing
gas velocity between 1.5 cm s1 and 11 cm s1, but a certain
amount of nodules was already present at the lower tested
value. In our study, the H2 velocity was kept constant at
0.75 cm s1, rather similar to the minimum value tested by
Zhao et al. The H2 flow rate is probably not responsible for our
observations.
Water vapour level at the O2 side is also discussed in the
SOFC literature, as a higher growth occurs with a higher hu-
midity level [16,17]. However, it cannot be considered in our
case as the values are very similar between our study (5%) and
the SOFC studies done in wet air (3%).
The last parameter that can be discussed is the water
vapour level at the H2 side, as the oxide scales formed at this
side are thicker than those formed under dry H2 and, conse-
quently, will decrease hydrogen transport through the sample
and oxidation at the air side [8]. A variation from 0% to 3% of
the H2O fraction leads to a significant decrease of the oxide
thickness and of the Fe content at the air side [8]. In our case, it
5308 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 0 ( 2 0 1 5 ) 5 3 0 5 e5 3 1 2
Fig. 2 e EDX mapping of the oxide scale grown on K41X at the anode side (O2-H2O) after the dual atmosphere experiment of
100 h at 800
C.
is possible to imagine that a 9% H2O level at the H2 side is high
enough to inhibit the H2 diffusion to the opposite side and
might explain our results. Moreover, the H2O:H2 ratio is
probably also a key factor, as the SOFC atmosphere is H2 rich,
while the HTE atmosphere is H2O rich (ratio of 3:97 in the first
case and 9:1 in the latter case).
Dual atmosphere influence at the H2-H2O side (HTE cathode
side)
At the cathode side of the sample (Fig. 1c), the surface
morphology of the oxide layer is slightly different with respect
to the anode case. Although it is still compact, without any
cracks, the oxide layer is now entirely formed of small grains
(smaller than those present at the anode side). However, the
XRD pattern showed the presence of the same two oxides:
(Cr,Mn)3O4 and Cr2O3. The cross-section observations (Fig. 1d)
show a thin oxide scale, with a thickness that varies between
0.6 and 1.2 mm. Some porosity can be observed at the interface
with the metallic substrate. EDX analyses showed that in this
case the oxide scale is formed by a continuous Cr-Mn spinel
oxide outer layer grown on a chromia inner layer.
At the H2-H2O side, the oxide scale obtained in dual at-
mosphere, composed of (Cr,Mn)3O4 and Cr2O3 phases, is
completely different than the one obtained in single atmo-
sphere and formed of Fe3O4 and (Cr,Fe)2O3 phases. This
change of composition is also correlated with a lower thick-
ness in dual conditions.
This result is surprising for two reasons. First of all, any
change in the oxide scale composition and morphology was
not observed during dual atmosphere SOFC tests compared to
single atmosphere experiments [9]. Secondly, as said in the
introduction part, the mechanical state of the sample surface
determines the growth of iron-based oxides in H2-H2O atmo-
spheres. As the tests in single atmosphere and dual atmo-
sphere were done on similar as-rolled samples, a complete
change of the oxide scale composition was not expected.
At this moment of our study, we have no explanation for
the evolution between the non-protective behaviour observed
in single H2-H2O atmosphere to the protective one observed at
the H2-H2O side of the dual atmosphere exposition. The K41X
drastic evolution in HTE dual conditions is certainly related to
the chemical potential difference compared to the H2-H2O
single atmosphere.
Electrical current influence on the oxidation behaviour in
dual atmosphere conditions (test with applied electrical
current)
After exposing the K41X alloy during 100 h at 800
C in dual
atmosphere conditions under an applied electrical current of
150 mA, two different regions could be identified on the
sample: a first one outside the electrodes and a second one
that was located under the electrodes during the test.
Fig. 3 shows the surface and cross-section morphologies of
the regions situated outside the electrodes. Any important
difference with respect to the previous experiment done
without electrical current could not be detected. At both sides
of the sample, a very similar surface morphology is observed.
The oxide layers are composed of (Cr,Mn)3O4 and Cr2O3. The
cross-section analysis shows that the thickness of the oxide
scale is very low in both cases, of about 400 nm.
In the electrodes region, the grown oxide scales are
completely different than those evidenced outside the elec-
trodes. On both sides of the K41X sample, a rather similar
microstructure was observed, as shown in Fig. 4a,b in cross-
section SEM images. The oxide scales are formed of separate
large nodules connected by very thin Cr2O3/(Cr,Mn)3O4 layers.
The nodules have similar thicknesses at both sides, but they
are more elongated and more numerous at the cathode side.
A zoom on such a nodule is shown in Fig. 5. Three different
layers can be identified inside the nodule. The superior layer
seems to have grown in top of the initial metal surface. EDX
analysis showed that it contains only Fe and O, which should
correspond to a Fe3O4 oxide, according to EDX quantification.
The other two layers grew inside the metal. The middle layer
is a Cr-Fe mixed oxide, while the inferior layer contains only
Cr and O and probably corresponds to chromia. If the inner
Cr2O3 layer is very thin and compact, it is not the case of the
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 0 ( 2 0 1 5 ) 5 3 0 5 e5 3 1 2 5309

Fig. 3 e SEM observations done outside the electrode region after dual atmosphere exposition of the K41X alloy with
electrical current supplied: (a) surface at the anode side (O2-H2O); (b) cross-section at the anode side; (c)surface at the cathode
side (H2-H2O); (d) cross-section at the cathode side.

two exterior layers, which are very porous (in particular the
Cr-Fe oxide) and rather thick. The total thickness of the nod-
ules is generally situated between 5 mm and 15 mm.
The nodules structure was already observed at the air side
of dual SOFC experiments [5,18] and also in single Ar-4%
H2e20%H2O atmosphere at lower temperatures [19]. However,
it was never observed before on K41X steel after tests done at
800
C in HTE single or dual atmosphere. As it is only observed
in the regions where the electrodes were placed on the sample
during the test, it is possible to relate this microstructure to
the electrical current effect on the HTE dual atmosphere
behaviour of the K41X alloy.
There is a very small number of works in the literature
concerning the electrical current effect on the stainless steels
high temperature oxidation [18,20,21]. They all show a positive
influence at the cathode side and a detrimental one at the
anode side. By taking into account chromia growth on Fe-25Cr
steel at 800
C, Kawamura et al. showed that, when an elec-
trical current is supplied, the flux of Cr3þ is changed, being
promoted at the anode side and restrained in the cathode side.

Fig. 4 e Cross-section observations done under the electrode after dual atmosphere exposition of the K41X alloy with
electrical current supplied: (a) anode side (O2-H2O); (b) cathode side (H2-H2O).
5310 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 0 ( 2 0 1 5 ) 5 3 0 5 e5 3 1 2
Fig. 5 e Cross-section observations and EDX mapping of a nodule grown in the electrode area during the dual atmosphere
exposition of the K41X alloy for 100 h at 800
C with electrical current supplied.
Compared to the situation without electrical current, it results
a thicker anode scale and a thinner cathode scale.
In our case, the thickness of the oxide scale is not very
different between the O2-H2O side and the H2-H2O side. This
result might be related to the 4 times lower electrical current
density, 130 mA cm2, compared to that used in the SOFC
literature of 500 mA cm2. In parallel, the total time of our
experiment, 100 h, is probably too short in order to lead to a
visible difference of thickness between the two sides.
The evolution of the oxide scale microstructure when an
electrical current is supplied was equally surprising as no
such effect was observed in single atmosphere. Indeed, in this
latter case, the same morphology and composition was
observed in the oxide scales obtained after conductivity
measurements or after simple ageing treatments. It seems
that the nodules morphology results from the combined effect
of the dual atmosphere and the electrical current.
Electrical behaviour in dual atmosphere
The area specific resistance (ASR) parameter measured in dual
atmosphere conditions has a very good value, of 0.04 U cm2
after 100 h at 800
C, lower that both the values obtained in
single O2-H2O and H2-H2O atmospheres and lower than the
limit value of 0.1 U cm2.
The ASR parameter reflects both the thickness and the
composition of the oxide layer. Even if the magnetite present
in the nodules has a very good electrical conductivity [22], the
chromia layer shows a very poor electrical behaviour. More-
over, the very dense chromia layer seems to be continuous
inside all the observed nodules. Together with a very impor-
tant thickness (5e15 mm), it is obvious that the very good ASR
parameter cannot be related to the nodules presence.
The chromia thin layers that connect the nodules might be
the explanation for the very interesting conductivity as their
thickness is extremely low (200e500 nm), as shown in Fig. 6.
Moreover, Cr-Mn spinel oxide grains can be observed in top of
the chromia layer and they are expected to enhance the
electrical conductivity of the scale [3,23e25]. The hypothesis
that the good conductivity is directly related to thin Cr2O3/
(Cr,Mn)3O4 layer is supported by the fact that the ASR
parameter has almost no variation during the experiment, in
connection with the well-known very slow chromia growth.
However, the literature discusses the influence of the
hydrogen (from H2 or H2O) dissolution in the chromia scale on
the conductivity [26]. By injecting protons in the oxide layer,
the hydrogen might be responsible of the default concentra-
tion change and, consequently, of its conductivity decrease
[27]. Nevertheless, all these observations were done during
single atmosphere experiments and it is obvious that they do
not agree with our results. In dual atmosphere, the difference
of chemical potential between the two sides can be again the
reason of different phenomena than those observed in singles
atmospheres.
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 0 ( 2 0 1 5 ) 5 3 0 5 e5 3 1 2
Fig. 6 e Cross-section observations and EDX mapping of the thin oxide layer that connects the nodules grown in the
electrode area during the dual atmosphere exposition of the K41X alloy for 100 h at 800
C with electrical current supplied.
Conclusion
In order to better approach interconnect HTE functioning
conditions, very new and innovative tests were performed on
K41X in dual atmosphere at 800
C during 100 h. If the oxide
layer grown at the anode (O2-H2O) side is the same as obtained
previously in single atmosphere (Cr2O3 and (Cr,Mn)3O4), a
totally different situation was observed at the cathode (H2-
H2O) side: the corrosion scale is now composed by inner
chromia and outer (Cr,Mn)3O4 protective oxides (instead of
Cr2O3 and Fe3O4).
Unlike SOFC dual tests, the Fe poisoning is avoided at the
O2-H2O side of the HTE tests, most likely because of the higher
H2O content at the opposite side. The very interesting evolu-
tion of the oxide layer at the H2-H2O side compared to single
atmosphere tests is not fully understood yet and it is probably
related to the different chemical potential.
The influence of the electrical current on the corrosion
behaviour was tested by performing a second series of ex-
periments. Outside the electrode area, the sample behaves
exactly like in the tests without current. Under the electrode, a
very interesting microstructure was observed, composed of
large and thick nodules of Fe3O4/Fe,Cr mixed oxide/Cr2O3
connected by very thin chromia layers. We believe that this
complex microstructure is the result of electrical current and
dual atmosphere combined effects, as it was never observed
when only one of these factors was present.
5311
A very interesting ASR parameter value was measured
(0.04 U cm2) in dual atmosphere, lower than both the values
obtained in single atmospheres. The very good electrical
behaviour was related to the very low thickness of the
chromia/Cr-Mn spinel oxide layer that connects the nodules.
Acknowledgement
This work has been supported by French Research National
Agency (ANR) through the project ICARE n
ANR-08-PANH-009.
references
[1] Zhu WZ, Deevi SC. Opportunity of metallic interconnects for
solid oxide fuel cells: a status on contact resistance. Material
Res Bull 2003;38:957e72.
[2] Zhu WZ, Deevi SC. Development of interconnect materials
for solid oxide fuel cells. Material Sci Eng 2003;A348:227e43.
[3] Fergus JW. Metallic interconnects for solid oxide fuel cells.
Mater Sci Eng 2005;A397:271e83.
[4] Holcomb GR, Ziomek-Moroz M, Cramer SD, Covino Jr BS,
Bullard SJ. Dual-environment effects on the oxidation of
metallic inteconnects. J Mat Eng Perform 2006;15:404e9.
[5] Yang Z, Walker MS, Singh P, Stevenson JW, Norby T.
Oxidation behavior of ferritic stainless steels under SOFC
5312 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 0 ( 2 0 1 5 ) 5 3 0 5 e5 3 1 2
interconnect exposure conditions. J Electrochem Soc
2004;151:B669e78.
[6] Yang Z, Xia G, Walker MS, Wang C, Stevenson JW, Singh P.
High temperature oxidation/corrosion behavior of metals
and alloys under a hydrogen gradient. Int J Hydrogen Energy
2007;32:3770e7.
[7] Rufner J, Gannon P, White P, Deibert M, Teintze S, Smith R,
et al. Oxidation behavior of stainless steel 430 and 441 at
800
C in single (air/air) and dual atmosphere (air/hydrogen)
exposures. Int J Hydrogen Energy 2008;33:1392e8.
[8] Bredvei Skilbred AW, Haugsrud R. The effect of dual
atmosphere conditions on the corrosion of Sandvik Sanergy
HT. Int J Hydrogen Energy 2012;37:8095e101.
[9] Ardigo MR, Parry V, Popa I, Chevalier S, Chandra-
Ambhorn W, Phakpetinaan P, et al. Optimisation of metallic
interconnects for hydrogen production by high temperature
water vapour electrolysis. Defect Diffusion Forum
2012;323e325:239e44.
[10] Quadakkers WJ, Zurek J. Oxidation in steam and steam/
hydrogen environments. Shreir's Corrosion 2010:407. Elsevier
Oxford.
[11] Guillou S, Desgranges C, Chevalier S. Influence of a coating
on oxidation resistance and resistivity of a cromia former
alloy for high temperature vapor electrolysis application.
Oxid Metals 2013;80:341e61.
[12] Ardigo MR. [PhD thesis]. University of Burgundy, 2011.
[13] Chevalier S, Favergeon J. French activity on high temperature
corrosion in water vapour. Trans Tech Publications; 2014.
[14] Yen SK, Tsai YC. Critical hydrogen concentration for the
brittle fracture of AISI 430 stainless steel. J Electrochem Soc
1996;143:2736e41.
[15] Gannon P, Amendola R. High-temperature, dual-atmosphere
corrosion of solid-oxide fuel cell interconnects. JOM
2012;64(12):1470e6.
[16] Zhao Y, Fergus JW. Oxidation of alloys 430 and 441 in SOFC
dual amospheres: effects of flow rate and humidity. J
Electrochem Soc 2012;159:C109e13.
[17] Yang Z, Xia G, Singh P, Stevenson JW. Effects of water vapor
on oxidation behavior of ferritic stainless steels under solid
oxide fuel cell interconnect exposure conditions. Solid State
Ionics 2005;176:1495e503.
[18] Horita T, Kshimoto H, Yamaji K, Sakai N, Xiong Yuping,
Brito ME, et al. Anomalous oxidation of ferritic interconnects
in solid oxide fuel cells. Int J Hydrogen Energy
2008;33:3962e9.
[19] Young DJ, Zurek J, Singheiser L, Quadakkers WJ.
Temperature dependence of oxide scale formation on
high-Cr ferritic steels in Ar-H2-H2O. Corros Sci
2011;53:2131e41.
[20] Kawamura K, Nitobe T, Kurokawa H, Ueda M, Maruyama T.
Effect of electric current on growth of oxide scale on Fe-25Cr
alloy for SOFC interconnect at 1073 K. J Electrochem Soc
2012;159(3):B259e64.
[21] Kodjamanova P, Fu Q, Gautier L. Electric current effects on
the corrosion behaviour of high chromium ferritic steels.
Oxid Met 2013;79:53e64.
[22] Dieckmann R, Witt CA, Manson TO. Defects and cation
diffusion in magnetite (V): electrical conduction, cation
distribution and point defects in Fe3-zO4. Berichte der Busen-
Gesellschaft fur Physikalische Chemie 1983;87:495e503.
[23] Quadakkers WJ, Piron-Abellan J, Shemet V, Singheiser L.
Metallic interconnects for solid oxide fuel cells - a review.
Mater High Temp 2003;20:115e27.
[24] Qu W, Li J, Ivey DG. Sol-gel coatings to reduce oxide growth in
interconnects used for solid oxide fuel cells. J Power Sources
2004;138:162e73.
[25] Stanislowski M, Wessel E, Hilpert K, Markus T, Singheiser L.
Chromium vaporization from high-temperature alloys I.
Chromia-forming steels and the influence of outer oxide
layers. J Electrochem Soc 2007;154:A295e306.
[26] Holt A, Kofstad P. Electrical conductivity and defect structure
of Mg-doped Cr2O3. Solid State Ionics 1997;100:201e9.
[27] Guillou S. [PhD thesis]. University of Burgundy, 2011.