Note: Descriptions are shown in the official language in which they were submitted.
MIXED METAL OXIDE ELECTRODE MATERIAL FOR SOLID OXIDE AND
REVERSIBLE SOLID OXIDE FUELL CELL APPLICATIONS
BACKGROUND OF THE INVENTION
Solid oxide fuel cells (SOFCs) are electrochemical devices that can convert
chemical energy into electrical energy with very high efficiency. SOFCs also
have several
other advantages over combustion-based technologies, such as fuel flexibility
(H2,
hydrocarbon-based fuels such as CI-14, CO, etc.), low emission of pollutants
(SO x and
N0x), and serve to capture CO2 from the anode exhaust stream in high purity
form,
already separated from N2.
A typical SOFC consists of a dense electrolyte and two porous electrodes, the
anode and the cathode. As part of the efforts to develop new energy conversion
systems,
there is great interest in reversible fuel cells, particularly reversible
solid oxide fuel cells
(RSOFCs). RSOFCs are single-unit, all-solid-state, electrochemical devices
that can
operate in both the fuel cell (SOFC) and electrolysis (SOEC) mode, thus acting
as flexible
energy conversion and storage systems, particularly to store intermittent
renewable
energy, such as wind or solar. The most common degradation and cell failure
issue for
RSOFCs arises at the air electrode when the cell is operating in the
electrolysis mode
(oxygen evolution at the air electrode). This is due to delamination of the
electrocatalytic
material from the electrolyte. Although the delamination mechanism is not
fully
understood, several processes have been postulated, including high oxygen
pressure
development, morphological changes in air electrodes, and electrolyte grain
boundary
separation [1-5]. Therefore, in this work, one of the main emphases is the
development
of a mixed conducting oxide (MIEC) that can withstand electrolysis conditions
without
delamination, while also exhibiting superior oxygen evolution and reduction
activities.
To date, the most common materials used in RSOFCs are essentially the same as
those used for SOFC, namely yttria stabilized zirconia (YSZ) as the
electrolyte, a Ni-YSZ
cermet as the fuel electrode, and a Lai _x SrxMn03 (LSM)-YSZ composite as the
air
electrode. The search for higher performance electrode and electrolyte
materials for
RSOFCs has been a key focus of research in recent years, with a particular
emphasis on
the development of new air electrodes. This has included the development of
mixed ionic-
electronic conductors (MIECs), such as Fe-based perovskites e.g., SrFe03_3,
and
1
Date Recue/Date Received 2021-09-30
the use of a variety of cation dopants in both the A and B-sites [6-9]. As an
example,
LaCr03 and its doped variants are good candidates for application as
interconnect
materials and cathode materials in SOFCs [10]. Other high performance air
electrode
materials include La0.6Sr0.4Co02Fe0.803_6 (LSCF), which has exhibited a low
polarization
resistance (Rp) of 0.18 S2 cm2 at 800 C [11], La0.6Sr0 AFe0.8Cu0.203_6
(LSFCu), which
has demonstrated a very low Rp of 0.07 C2 cm2 [12], and La0.8Sr0.2Cro.5Mno.5
03 (LSCM)
(2), which has exhibited a polarization resistance of 0.3 S2 cm2 at 800 C
[13].
Recently, Chen et al. [14] have shown very good catalytic activity for both
H2/C0
oxidation and 02 reduction using the same MIEC material at both electrodes,
i.e.,
Lao.3Sr0.7Fe0.7Cro.303_6 (LSFCr), used for the first time as an SOFC
electrode. The selected
stoichiometry of the material was based on increasing the electronic and ionic
conductivity of a Fe-based perovskite by heavy A-site substitution of La by
Sr. As well,
the partial substitution of Fe at the B site by Cr was done to stabilize the
orthorhombic
perovskite and its associated high level of vacancy disorder [15].
Usually, MIECs are synthesized by solid-state reactions, where the process
involves multiple heating (> 1200 C) and regrinding steps to help overcome
the solid-
state diffusion barrier [16]. Some of the traditional methods by which MIECs
have been
prepared include the sol-gel method [6], the EDTA citrate complexing process
[12], the
auto-ignition process [7], the Pechini method [9], and most commonly, by using
combustion methods [14].
SUMMARY OF THE INVENTION
In the present study, the performance of derivatives of LSFCr containing
calcium,
i.e., La0.3Cao.7Feo.7Cro.303-6 (LCFCr), synthesized by the combustion method
is examined.
Several previous studies [17, 181 have shown that the thermal expansion
coefficient
(TEC) of perovskite materials decreases as the A-site cation size is
decreased. Herein,
the A-site of the perovskite was doped with Ca instead of Sr, to decrease the
thermal
expansion coefficient of the MIEC to more closely match that of the gadolinium
doped
ceria (GDC) electrolyte. While the thermal expansion data for perovskites
reflects both
physical and chemical expansion processes, the chemical expansion due to
oxygen loss
should dominate the thermal expansion behavior at high temperatures. The
partial
substitution of Ca for Sr may also enable the introduction of structural
inhomogeneities,
as calcium doping of LaFe03 is known to promote oxygen-vacancy ordering [19,
201.
2
Date Recue/Date Received 2021-09-30
However it was uncertain if the excellent electrochemical properties of LSFCr,
could be
maintained despite the replacement of Sr with Ca.
Consistent with these objectives, LCFCr/GDC/LCFCr symmetrical half cells,
operated at 600-800 C in stagnant air, were assessed herein and have been
found to
exhibit excellent electrochemical performance and long term durability,
similar to that of
the previously studied LSFCr material [8]. Electrochemical measurements herein
have
shown polarization resistances of only 0.07, 0.34, 0.71, 1.6 and 4 S2 cm2 at
800, 750, 700,
650 and 600 C, respectively.
This invention also provides microwave based methods for making the electrode
material of the invention.
The invention further demonstrates the resistance of the electrode materials
of this
invention to sulfur.
The invention provides electrode material, i.e., electrocatalytic material,
having
the formula:
LawMxFeyCrz03-6
where:
M is Ca or a mixture of Ca and Sr where the molar ratio of Ca to Sr ranges
from 1:1 to
100:1;
w is 0.2 to 0.4;
x is 0.6 to 0.8;
y is 0.6 to 0.8;
z is 0.2 to 0.4;
and
6 represents oxygen deficiency.
In specific embodiments, M is Ca.
In specific embodiments:
w is 0.27 to 0.33;
x is 0.67 to 0.73;
y is 0.67 to 0.73; and
z is 0.27 to 0.33.
In specific embodiments:
w is 0.29 to 0.31;
xis 0.69 to 0.71;
y is 0.69 to 0.71; and
3
Date Recue/Date Received 2021-09-30
CA 02893153 2015-05-28
z is 0.29 to 0.31.
In specific embodiments, w is 0.3; x is 0.7; y is 0.7; and z is 0.3.
In specific embodiments, the electrode material is a perovskite of the above
formula.
In a specific embodiments, M is a mixture of Ca and Sr. More specifically in
an
embodiment, the molar ratio of Ca to Sr is 1:1. Yet more specifically, the
molar ratio of
Ca to Sr is 10:1.
In a preferred embodiment, in the electrode material M is Ca. In a preferred
embodiment the electrode material is La0.3Ca0.7Fe0.7Cr0.303-6.
In a specific embodiment, the electrode material of the invention has atomic %
composition of;
La (15 0.5)
Ca (34.5 1)
Cr (15 0.5) and
Fe(35 1).
The invention further provides electrodes which comprise an electrode material
of this invention. In a specific embodiment, such electrodes are formed as a
layer on a
solid oxide electrolyte. In a specific embodiment, the electrode is a fuel
electrode,
particularly for a solid oxide fuel cell or a reversible solid oxide fuel
cell. In a specific
embodiment, the electrode is an air or oxygen electrode, particularly for an
SOFC or a
RSOFC.
The invention further provides electrodes which consist of an electrode
material
of this invention. In a specific embodiment, such electrodes are formed as a
layer on a
solid oxide electrolyte. In a specific embodiment, the electrode is a fuel
electrode,
particularly for a solid oxide fuel cell or a reversible solid oxide fuel
cell. In a specific
embodiment, the electrode is an air or oxygen electrode, particularly for an
SOFC or a
RSOFC.
The invention further provides electrodes which comprise an electrode material
of this invention as the electrocatalytic material of the electrode. Such
electrodes may
contain other supporting or non-electrocatalytic active materials. In a
specific
embodiment, such electrodes are formed having at least one layer of
electrocatalytic
material on a solid oxide electrolyte. In a specific embodiment, the electrode
is a fuel
4
CA 02893153 2015-05-28
electrode, particularly for a solid oxide fuel cell or a reversible solid
oxide fuel cell. In
a specific embodiment, the electrode is an air or oxygen electrode,
particularly for an
SOFC or a RSOFC.
The invention further provides solid oxide fuel cell having,an electrode which
comprises an electrode material of the invention.
The invention further provides a reversible solid oxide fuel cell having an
electrode which comprises an electrode material of the invention.
The invention also provides methods for generating electricity which comprises
operating a solid oxide fuel cell having at least one electrode comprising an
electrode
material of the invention
The invention also provides methods for selectively generating electricity or
employing electricity to generate a fuel which comprises selectively operating
a
reversible solid oxide fuel cell having an electrode of the invention
comprising an
electrode material of the invention to generate electricity or to generate a
fuel.
The invention also provides methods wherein the solid oxide fuel cell or
reversible solid oxide fuel cell is efficiently operated in the presence of a
fuel containing
hydrogen sulfide.
In specific embodiments, electrode materials of the invention are prepared by
microwave assisted methods. In particular, electrode materials of the
invention are
prepared by microwave-assisted combustion, microwave-assisted co-precipitation
or a
microwave-assisted sol-gel method.
BRIEF DESCRIPTION OF THE DRAWINGS
Figure 1. Rietveld refinement of LCFCr powder (synthesized by the
conventional combustion method) X-ray diffraction patterns observed (red
dotted lines),
refined (black solid lines), and their difference (bottom line). Green
vertical bars
indicate the X-ray reflection positions.
Figures 2A and 2B. In situ high temperature XRD patterns from 25-1100 C in
air (FIG. 2A). Cell parameters a, b and c and unit cell volume as a function
of
temperature (FIG. 2B).
Figure 3. HRTEM image of LCFCr crystals in a powder sample (prepared by the
combustion method) along the [101] zone axis and the corresponding digital
diffraction
pattern.
Figure 4A and B. Impedance spectra of LCFCr at 800 C, 750 C, 700 C, 650 C
and 600 C (FIG. 4A). The impedance response was obtained in stagnant air at
the OCP.
Equivalent circuit used for data fitting (FIG. 4B)
Figure 5. Arrhenius plot of total polarization resistance (Rp) resistance vs.
1/T for
LCFCr air electrode, screen-printed on GDC electrolyte and measured at the OCP
in
stagnant air over a temperature range of 600-800 C.
Figure 6. Potentiostatic response of LCFCr tested at 800 C and 0.4 V for 100
h
in stagnant air. Inset shows the OCP impedance spectra, in air, collected
before and after
the potentiostatic measurements.
Figure 7. Potentiostatic response of LCFCr tested at 800 C and -0.4 V for 100
h
in stagnant air. Inset shows the OCP impedance spectra collected before and
after the
potentiostatic measurements in stagnant air.
Figure 8. (a-c) Back-scattered electron (BSE) images of the cross-section of
the
LCFCr/GDC electrolyte interface after 100 hours of cell testing at 0.4 V
anodic and
cathodic overpotential at 800 C (Figures 8a and 8b). Figure 8c is a back-
scattered
electron (BSE) image of the cross-section of the LCFCr/GDC
electrode/electrolyte
interface in a cell before electrochemical testing.
Figure 9. Comparison of XRD patterns of LCFCr, synthesized by (a) the regular
combustion method (Method 1) and two microwave-related methods, (b) by the
microwave-combustion method (Method 2), and (c) by the microwave-assisted sol-
gel
method (Method 3) and calcined at three different temperatures, 700 C, 900 C
and 1000
C.
Figure 10. Rietveld refinement of powder X ray diffraction patterns for LCFCr
observed (red dotted lines), refined (black solid lines), and their difference
(bottom line).
Vertical bars indicate the X-ray reflection positions. The patterns are for
LCFCr powder
(a) synthesized by the microwave-combustion method (Method 2) and (b) by the
microwave-assisted sol-gel method (Method 3).
Figure 11. SEM images of LCFCr powders formed using (a) microwave-assisted
combustion synthesis (Method 2) and (b) microwave-assisted sol-gel synthesis
(Method
3).
Figure 12. HRTEM images of LCFCr crystals along the [101] zone axis and the
corresponding diffraction patterns for powders formed by (a) microwave-
assisted
combustion synthesis (Method 2) and (b) microwave-assisted sol-gel synthesis
(Method
3).
6
Date Recue/Date Received 2021-09-30
CA 02893153 2015-05-28
Figure 13. OCP impedance data for symmetrical full cell based on La o.3M
0.7Fe0.7Cr 0.303_6- (M = Sr, Ca) electrodes, at 800 C, showing the Nyquist
and Bode
(inset) plots, all in wet 30% H2/N2 gas mixtures at the fuel electrode and
with air
exposure at the 02 electrode.
Figure 14. OCP impedance data for symmetrical full cell based on
Lao.3Cao.7Feo.7Cro.303-s(LCFCr) electrodes at 800 C and showing the Nyquist
plot, all
in wet 30% H2/N2 gas mixtures at the fuel electrode and with air or 02
exposure at the
02 electrode.
Figure 15. Performance plot for symmetrical full cell based on
Lao.3Mo7Feo7Cro.303-5(M = Sr, Ca) electrodes and operated at 800 C, all in
wet 30%
H2/N2 gas mixture at the fuel electrode and with air exposure at the 02
electrode.
Figure 16. OCP impedance data for symmetrical full cell based on LCFCr
electrodes, at 800 C, showing the Nyquist plots in wet 30% H2/N2. 15% H2+ 1 5
% CO,
or 30% CO gas mixtures at the fuel electrode and with air exposure at the 02
electrode.
Figure 17. Performance plot for symmetrical full cell based on LCFCr
electrodes,
operated at 800 C in wet 30% H2/N2. 15% 112+15% CO, or 30% CO gas mixtures at
the
fuel electrode and with air exposure at the 02 electrode.
Figures 18A-D. OCP and polarized EIS response for symmetrical full cell based
on LCFCr electrodes at 800 C, with wet 30% H2/N2 with or without 9 ppm H2S fed
to
the fuel electrode and air fed to the 02 electrode, showing the Nyquist plots
acquired at
(FIG. 18A) the OCP, (FIG. 18B) -100 mV vs the cell voltage at open circuit,
(FIG. 18C)
-300 mV vs. the cell voltage at open circuit, and (FIG. 18D) the corresponding
resistances obtained from the fitted Nyquist plots using the
Rs(RHF/CPEHF)(RLF/CPELF)
equivalent circuit model and the % Rp change (inset).
Figures 19A and B. Effect of 9 ppm H2S exposure and removal on LCFCr anode
activity as a function of polarization at 800 C, showing the current versus
time (it)
plots at (FIG. 19A) -100 mV and (FIG. 19B) -300 mV vs. the full cell voltage
at open
circuit.
DETAILED DESCRIPTION OF THE INVENTION
The invention relates to certain mixed metal oxide materials which are useful
as
the active material in electrodes of solid oxide fuel cells and particularly
in reversible
solid oxide fuel cells, either anodes or cathodes, therein. In a specific
embodiment, the
7
CA 02893153 2015-05-28
electrode materials herein can be used to make symmetrical solid oxide fuel
cells where
the electrode material of the anode is the same as in the cathode.
As is known in the art fuel cells convert energy in fuels to electrical
energy.
Fuel cells can be operated in reverse (as an electrolyzer) using electrical
energy to
convert a molecule, such as water, to a fuel, such as hydrogen, Reversible
cells operate
in both modes. In a SOFC an oxidizing material, typically air or oxygen, is in
contact
with the cathode of the cell and the fuel is in contact with the anode of the
cell. During
fuel cell operation oxygen ions are transported from the cathode to the anode
to oxidize
the fuel to form water or if carbon monoxide is present to form carbon
dioxide. SOFC
cells typically operate at temperatures between about 750 to 950 C. The
electrodes are
electrically connected and operation generates a current between the
electrodes. In
reverse mode, electrical energy is used to produce oxidant and fuel. A
reversible SOFC
cell (RSOFC or Solid oxide electroyzer cell [SOECD can be operated in both
modes.
The present invention provides electrode materials that can be used as
electrodes
(anodes or cathodes or both) in SOFC cells, for example as air or oxygen
electrodes, or
as electrodes in reversible SOFC cells.
US Patent 8,354, 011 relates to reversible electrodes for solid oxide
electrolyzer
cells (SOEC). This patent provides a description of electrodes for such cells
and the
operation of such cells. FIG. 1 therein illustrates a schematic planar
configuration of
such a cell. Such a planar configuration can be employed in SOFC and SOEC of
this
invention.
Chen, M. et al. (2013) J. Power Sources 236:68-79 describes the use of certain
Sr-rich chromium ferrites for symmetrical solid oxide fuel cells. In
particular, the use of
La0n3Sr07Fe0 7C0.3 03_8 is described.
US Patent 8,617,763 provides a description of certain SOFC cells and in
particular a certain type of anode useful in such cells. Anode, cathode and
electrolyte
materials described therein can be employed in the devices of the present
invention.
As is known in the art, a cell of the invention (SOFCor RSOFC) comprises an
anode, a cathode and an ionically conductive solid oxygen electrolyte between
the
anode and the cathode. Optionally, a buffer layer is positioned between the
anode and
the electrolyte and/or between the cathode and the electrolyte. In an
embodiment, the
anode and/or the cathode is provided as a layer on one side of a layer of
electrolyte.
The other of the anode or the cathode being provided on the other side of the
layer of
electrolyte. Oxygen anions pass through the electrolyte layer from the cathode
to anode
8
CA 02893153 2015-05-28
or the reverse dependent upon the mode in which the cell is operated. The
optional
buffer can be provided as a layer between the layer of anode material and the
electrolyte
and/or between the layer of cathode material and the electrolyte.
Electrode materials of the invention are those of formula:
LawMõFeyCrz03_s
where:
M is Ca or a mixture of Ca and Sr where the molar ratio of Ca to Sr ranges
from 1:1 to
100:1;
w is 0.2 to 0.4;
x is 0.6 to 0.8;
y is 0.6 to 0.8;
z is 0.2 to 0.4;
w = x is 1;
and y =z is 1.
A preferred electrode material is La0.3Ca0.7Fe0.7Cr0.303_6. In a particularly
preferred embodiment the electrode material is a single phase material having
no
dectable second or other additional phase. In an embodiment, the electrode
material is
substantial single phase material with less than 5% by weight of a second or
other
additional phase or more preferably having less than about 2% by weight of a
second or
other additional phase. In specific embodiments, the electrode material is a
perovskite.
Various methods can be employed to prepare the mixed metal oxide compounds
of the invention. For example the following methods can be used:
A. Microwave method combined with a sol-gel methodology.
Microwave method and a sol¨gel methodology can be combined to make
electrode materials of the invention, for example
La0.3Ca0.7Fe07C0.303_8(LCFC).
Equimolar amounts of metal nitrates are dissolved in distilled water and a
saturated
polyvinyl alcohol (PVA) solution is added as the complexing agent. The amount
of
PVA added is such that the ratio of the total number of moles of cations to
that of PVA
is 1:2. Then the final solution is maintained at 80 C for 1.5 h to form a
viscous gel
solution. This gel is then irradiated with microwaves (up to 30 min) in a
porcelain
crucible placed inside another larger one filled with mullite. The microwave
source
operates at 2.45 GHz frequency and 800 W power and is uniquely able to handle
the
conditions needed. The polymeric and sponge-like-precursor is then calcined in
air at
9
CA 02893153 2015-05-28
1000 C for 6 h in order to decompose the organic remnants, rendering a black
powder
as the final product.
B. Microwave-assisted combustion
Metal nitrates are mixed in stoichiometric proportions, and then water and
glycine are added. The sample is introduced into the microwave furnace at 2.45
GHz
frequency and 800 W power for 30 minutes. When the water is evaporated,
combustion
occurred and a flame is observed inside the microwave furnace for 10 minutes.
Then the
sample is calcined in air at 900 C for 6 h in order to decompose the organic
remnants,
rendering a black powder as the final product.
C. Microwave-assisted co-precipitation
Metal nitrates are mixed in stoichiometric proportion, 25 ml of acetic acid
are
added, and then the mixture is stirred and heated at 60 C for 2 hours. When
the nitrate
vapors are evaporated, a gel formed and then it is introduced into the
microwave
furnace at 2.45 GHz frequency and 800 W power for 30 minutes, followed by
calcination at 900 C.
D. Regular combustion method
When synthesized using the regular combustion method (Method 1), the metal
nitrates are mixed in stoichiometric proportions and dissolved in deionized
water. A 2:1
mole ratio of glycine to the total cation content is used. Solutions are
slowly stirred on a
hot plate until auto-ignition and self-sustaining combustion occurred. The
sample is first
ground and then calcined in air at 1200 C for 12 hours.
The electrode materials of the invention can be employed in any SOCF or
RSOFC configurations and are particularly useful in those configurations which
employ
electrode layers.
Solid electrolytes useful in the invention include stabilized zirconia,
including
yttrium stabilized zirconia and scandia stabilized zirconia, doped ceria,
including
gadolidium-doped ceria or samarium-doped ceria, and certain mixed metal oxides
such
as LSGM (lanthanum strontium gallium magnesium oxide. One of ordinary skill in
the
art knows how to select solid oxide electrode s appropriate for use in SOFC
and RSOFC
devices.
In specific embodiments herein, the SOFC and RSOFC cells are symmetric
wherein the anode and cathode materials are the same and are electrode
materials of this
CA 02893153 2015-05-28
invention. In alternative embodiments, alternative anode having alternative
electrode
materials can be used in combination with cathodes having electrode materials
of this
invention. Alternate anode materials include among others, perovskite mixed
metal
oxide materials other than those of this invention, e.g.,
La0.3Sr0.7Fe0.7Cr0.303_6,cermets
having a metal phase, such as a nickel or nickel oxide phase, and a ceramic
phase, such
as doped ceria (samaria or gadolinium-doped); and/or stabilized zirconia.
In alternative embodiments, alternative cathode having alternative electrode
materials can be used in combination with anodes having electrode materials of
this
invention. Alternate cathode materials include among others, perovskite mixed
metal
oxide materials other than those of this invention, e.g.,
La0.3Sr0.7Fe0.7Cr0.303_6. electron
conducting phases (e.g., nickel oxide and magnesium oxide).
One of ordinary skill in the art in view of what is known in the art about
electrode materials useful in SOFC or RSOFC application can select among known
electrode materials for alternative electrode materials that are useful in
combination
with the electrode materials of this invention.
Anodes and cathodes may be formed a one or more layers on a surface of a solid
electrolyte.
Solid electrolyte can be in a planar layer configuration with one side of the
electrolyte layer containing a layer of anode material and the other a layer
of cathode
material. In symmetric cells the electrode layers are the same materials.
SOFC and RSOFC electrodes are prepared by conventional methods by
formation of at least one layer of the electrode material on an appropriate
substrate. In a
specific embodiment, an electrode is formed by application of a layer of
electrode
material on a surface of a solid oxide electrolyte material. In a preferred
method of
preparation of electrodes microwave sintering is employed. The electrode
material is
screen printed onto the solid oxide electrolyte and it is irradiated at 900 C
for 20
minutes in a Milestone MultiFAST-6 sintering microwave. It was found that the
best
performance was for the sample irradiated at 900 C and the cell performance
is
comparable to the electrodes sintered using conventional furnaces at 1200 C.
The SOFC and RSOFC of the invention can be formed into stacks as is known in
the art. Stacks of such cells are provided by this invention. US Patent
8,663,869
provide examples of such fuel cell stacks.
Molero-Sanchez, B. et al. (2015) Int'l J. Hydrogen Energy 40:1902-1910, Addo,
P. et al. (2015) ECS Transactions 66(2):219-228; and Molero-Sanchez, B. et al.
(2015)
11
CA 02893153 2015-05-28
Ceramics Int'l (article in press, available on line at web site science
direct.com) provide
details of the examples provided herein.
When a Markush group or other grouping is used herein, all individual members
of the group and all combinations and subcombinations possible of the group
are
intended to be individually included in the disclosure. Every formulation or
combination of components described or exemplified can be used to practice the
invention, unless otherwise stated. Specific names of compounds are intended
to be
exemplary, as it is known that one of ordinary skill in the art can name the
same
compounds differently.
One of ordinary skill in the art will appreciate that methods, device
elements,
starting materials, and synthetic methods other than those specifically
exemplified can
be employed in the practice of the invention without resort to undue
experimentation.
All art-known functional equivalents, of any such methods, device elements,
starting
materials, and synthetic methods are intended to be included in this
invention.
Whenever a range is given in the specification, for example, a temperature
range, a time
range, or a composition range, all intermediate ranges and subranges, as well
as all
individual values included in the ranges given are intended to be included in
the
disclosure.
Without wishing to be bound by any particular theory, there can be discussion
herein of beliefs or understandings of underlying principles relating to the
invention. It
is recognized that regardless of the ultimate correctness of any mechanistic
explanation
or hypothesis, an embodiment of the invention can nonetheless be operative and
useful.
The terms and expressions which have been employed are used as terms of
description and not of limitation, and there is no intention in the use of
such terms and
expressions of excluding any equivalents of the features shown and described
or
portions thereof, but it is recognized that various modifications are possible
within the
scope of the invention claimed. Thus, it should be understood that although
the present
invention has been specifically disclosed by preferred embodiments and
optional
features, modification and variation of the concepts herein disclosed may be
resorted to
by those skilled in the art, and that such modifications and variations are
considered to
be within the scope of this invention.
12
CA 02893153 2015-05-28
THE EXAMPLES
Example 1:
1. Material Synthesis and characterization
LaØ3Ca0.7Fe0.7Cr0.3 03_6 (LCFCr) powders were synthesized by the combustion
method. The metal nitrate precursors were mixed in stoichiometric proportions
and
dissolved in deionized water. A 2:1 mole ratio of glycine to the total cation
content was
used. Solutions were slowly stirred on a hot plate until auto-ignition and
self-sustaining
combustion occurred. Then the sample was ground and calcined in air at 1200 C
for 12
hours.
Materials were purchased from Alfa Aesar as follows: Glycine (99.5%);
La(NO3)3.6H20 (99.9%); Sr(NO3)2 (99.0%); Ca(NO3)2 (99.0%); Cr(NO3). 91120
(98.5%); and Fe(NO3)3.9H20 (98-101%).
X-ray diffraction (XRD) patterns of all of the samples synthesized in this
work
were collected using a Philips X'Pert PRO ALPHA! of Panalytical B.V.
diffractometer
with Cu Kai monochromatic radiation (X = 1.54056 A). The diffractometer was
equipped with a primary curved Gel 11 primary beam monochromator and a speed
X'Celerator fast detector, operating at 45 kV and 40 mA. XRD patterns were
collected
in the 20 range of 5 ¨ 120 at room temperature with a step size of 0.017
and 8 s
counting time in order to ensure sufficient resolution for structural
refinement.
Powder X-ray Thermodiffraction patterns were collected on an X'Pert PRO
MPD diffractometer with a high temperature reactor chamber Anton Paar HTK1200
camera, using Cu Ka radiation. The measurements were carried out at between
room
temperature and 1100 C. The standard working conditions were a 20 range of
10¨ 70
with an angle step size of 0.033 and a 25 s counting time. Sample was heated
to the
target temperatures at a ramp rate of 5 C/min and stabilized in air for 40
min prior to
the measurements. After that, the sample was cooled to RT and XRD patterns
were
acquired again in order to determine the phase stability of the LCFCr material
under
heating and cooling conditions.
Fullprof Software was employed to carry out structural refinements from
conventional XRD patterns using the Rietveld method. This method of refining
the
powder diffraction data was used to determine the crystal structure. Zero
shift, lattice
parameters, background, peak width, shape and asymmetry, atomic positions and
isotropic temperature factors were all refined. The Thompson¨Cox-Hastings
pseudo-
Voigt convoluted with axial divergence asymmetry function was used to describe
the
13
CA 02893153 2015-05-28
peak shape. Linear interpolation between set background points with refineable
heights
was used afterwards. The values were refined to improve the agreement factors.
All samples investigated by scanning electron microscopy (SEM) were first
sputter-coated with Au in an EMI _____________________________ l'ECH K550
apparatus. Field-emission SEM (FE-
SEM) was performed using a JEM 6335 F electron microscope with a field-
emission
gun operating at 10 kV. The FE-SEM was also equipped with a LINK ISIS 300
detector for the energy-dispersive analysis of the X-rays (XEDS). SEM imaging
of the
cells and attached electrode layers was carried out using a Zeiss sigma VP
field
emission SEM.
High resolution transmission electron microscopy (HRTEM) analysis of the
LCFCr powders was performed using a JEOL 3000F TEM, operating at 300 Kv,
yielding information limit of 1.1 A. Images were recorded with an objective
aperture of
70 pm centered on a sample spot within the diffraction pattern area. Fast
Fourier
Transforms (FFTs) of the HRTEM images were carried out to reveal the periodic
image
contents using the Digital Micrograph package.
2. Cell fabrication and testing
The LCFCr powders obtained from the regular combustion method were milled
(high energy planetary ball mill, Pulverisette 5, Fritsch, Germany) in an
isopropanol
medium at a rotation speed of 300 rpm for 2 h using zirconia balls. The
electrolyte-
supported symmetrical cell was constructed with a GDC electrolyte (1 mm thick)
as the
substrate. The electrolyte was fabricated by pressing the GDC powder under 200
MPa
pressure and sintering at 1400 C for 4 h. The ca. 30 pm thick LCFCr
electrodes were
then screen-printed symmetrically (over an area of 0.5 cm2) onto both sides of
the GDC
support and fired at 1000 C for 2 h. Au paste (C 5729, Heraeus Inc., Germany)
was
painted on both of the electrode layers to serve as the current collectors.
In all of this work, the electrochemical measurements to evaluate the cell
performance were performed using the 3 electrode technique in air. Impedance
spectra
were collected under open circuit conditions, between 600 C and 800 C, using
an
amplitude of 50 mV in the frequency range of 0.01 to 65 kHz using a Solatron
1287/1255 potentiostat/galvanostat/impedance analyzer. Other experiments
involved the
application of a 0.4 V anodic and -0.4 V cathodic overpotential to the LCFCr
working
electrode vs. the reference electrode and measuring the current passed through
the cell
with time. Zview software was used to fit and analyze the impedance data.
14
CA 02893153 2015-05-28
3. Results and Discussion
Structural characteristics of LCFCr electrodes formed using the combustion
method
3.1 X-ray diffraction and Rietveld refinement of LCFCr powders
XRD analysis of the La0,3Ca0.7Fe07Cr0303_6 (LCFCr) powders was performed
and structural parameters for LCFCr were obtained from the Rietveld-refined
XRD
data. The Rietveld refinement indicated that the synthesized LCFCr powders are
a pure
crystalline phase with an orthorhombic perovskite structure. FIG. 1 shows the
Rietveld
refinement fits for LCFCr, and a distorted perovskite structure with an
orthorhombic
symmetry (S.C. Pnma, #62) was confirmed. The unit cell vectors can be
represented by
.Nhap x 2ap x 42a, where ap refers to the simple cubic perovskite cell. The
cell
parameters were found to be: a = 5.4540(2) A, b = 7.7158(3) A and c =
5.4544(1) A,
while the refinement fit parameters for LCFCr were x2 = 0.96, Rp = 3.26, Rwp =
4.29,
Rexp = 4.37 and RBragg = 4. Although the orthorhombic unit cell seems to be
pseudo
tetragonal, refinements were also performed in the P4/mrnm space group, but
these
yielded higher R values (x2 = 4.03, ; = 5.53, Rwp = 8.76, Rexp = 4.37 and
RBragg =
5.59), while the lattice parameters when using this tetragonal group were: a =
b =
5.45441(1) A and c = 7.7092(1) A. Thus, the P4/mmm space group was not used
for
the LCFCr electrode material.
In order to determine the phase stability of the LCFCr material under heating
and cooling conditions, in situ high temperature XRD measurements were
performed
from room temperature to 1100 C, and then back to room temperature again, all
in air.
FIG. 2A shows that the orthorhombic structure is maintained over the full
temperature
range up to 1100 C, since peak splitting is not observed. Moreover, a shift
of all the
characteristics peaks towards lower angles is observed, which may suggest an
increase
in the cell parameters with temperature. FIG. 2B shows the cell parameters a,
b and c, as
well as the unit cell volume vs temperature, calculated from XRD data (FIG.
2A).
Table 1-1 shows the average thermal expansion coefficient calculated from the
thermal XRD data (Figure 2), using the methods described in ref [26]. The
average
TEC is 11.5 x 106 K-1 for lattice parameter a, and 12.0 x 106 K-' for lattice
parameters
b and c. These values are comparable to those reported for the well-known
cathode
material LSM (12.2 x 10 -6 K-1 ) [21-23] and noticeably lower than the TEC
values for
LSCF (16.3 x 10 -6 K') [21, 241. The measured TEC values are also considerably
lower
CA 02893153 2015-05-28
than those for the Sr-rich perovskite, La0.3Sr07Fe07Cr0.303_6 (LSFCr,),
previously
developed in our group [14]. More importantly, the measured thermal expansion
coefficient (TEC) of LCFCr (Table 1) matches very well with the TEC of ceria
(11.9 x
-6 K') [11, 21, 25-311, which is a critical requirement for minimizing
delamination
of the electrodes from the electrolyte, thus avoiding mechanical failure of
the cell.
Table 1-1. Average thermal expansion coefficient (TEC) for LCFCr material,
determined by in situ XRD analysis
Thermal expansion Average TEC
parameters (x 10 -6 K-1)
Lattice parameter (a) 11.5 (25-1100 C)
Lattice parameter (b) 12.0 (25-1100 C)
Lattice parameter (c) 12.0 (25-1100 C)
3.2 TEM analysis of LCFCr powder
Transmission Electron Microscopy (TEM) analysis was also performed on the
LCFCr powder material. The cation composition, evaluated semi-quantitatively
by X-
ray energy dispersive spectroscopy in more than ten single crystals, is in
good
agreement with the theoretical proportions of the elements in LCFCr,
indicating the
high purity of the powder. High resolution TEM micrographs recorded along the
same
zone axis [101] show nano-sized twinned domains rotated by 90 . The appearance
of
these domains can be associated with the pseudo-cubic nature of these
materials. The
presence of these domains can help to avoid the formation of tetrahedral
chains and
therefore the formation of brownmillerite-type defects [32]. Typically,
raising the
temperature leads to a phase transition of brownmillerite to perovskite at
high
temperatures, accompanied by a conductivity jump [33]. As mentioned earlier,
perovskites exhibit a higher ionic conductivity than brownmillerites and hence
they are
better candidates for air electrodes in RSOFCs.
3.3 Electrochemical performance of LCFCr as a reversible air electrode
3.3.1 Open circuit studies
The electrochemical performance of the LCFCr material, synthesized by the
regular combustion method, was then studied, with the impedance spectra of the
16
CA 02893153 2015-05-28
LCFCr/GDC/LCFCr symmetrical half cells in air at 800, 750, 700, 650 and 600 C
shown in FIG. 4A, all at the open circuit potential (OCP). From FIG. 4A two
separable
arcs are visible over the full frequency range. The best-fit equivalent
circuit is shown in
FIG. 4B, where Rs is the series ohmic resistance, the sum of R2 (high
frequency) and
R3 (low frequency) is the total polarization resistance (Rp), and the CPEs are
constant
phase elements. Rs corresponds to the intercepts of the impedance arc with the
real axis
at high frequencies and arises from the resistance to ion migration within the
electrolyte,
resistance to electron transport within the cell components, and contact
resistances [34].
Rp is the difference between the two real axis intercepts of the impedance
arcs and CPE
is a component that models the behaviour of a an imperfect capacitor [35],
with the
associated n parameter being 1 for a perfect capacitor, 0 for a pure resistor,
and 0.5 for a
Warburg element [36]. The high-frequency arc (R2) corresponds to the charge
transfer
process and the low-frequency arc (R3) has been attributed previously in the
literature
[12, 37, 381 to oxygen adsorption and desorption on the electrode surface,
combined
with the diffusion of the oxygen ions.
As can be seen in Table 1-2, the Rp values are very small, 0.07, 0.33,0.73,
1.67
and 4.24 SI cm2 at 800, 750, 700, 650 and 600 C, respectively, even lower
than what
has been reported for the well-known cathode material LSCF (0.18 II cm at 800
C)
[11, 391. However, these Rp values are comparable to what was reported for
La0.3Sr07Fe0.7Cr0.303_8 (LSFCr), previously developed in our group and studied
using a
LSGM electrolyte, giving an Rp value of 0.11 S2 cm2 at 800 C [14]. In terms
of the
capacitance values obtained from the cell examined in FIG. 4, the high-
frequency arc
(R2) has a CPE-T value of ca. 10-1 (F s)"/cm2 and an associated CPE-P value of
0.72,
while the low-frequency arc (R3) also has a CPE-T value of ca. 10-i (F s)l-
n/cm2), but an
associated CPE-P value of 0.86, very close to that of an ideal capacitor.
Table 1-2. Fitting parameters of the impedance data obtained in Figure 4
RHF RP Chi-
RLF CPE-P
Temperature (.cm2) CPE-P (12.c m2) squared
(.cm2) (LF)
(HF)
800 C 0.05 0.86 0.02 0.72 0.07 1.6x10-5
750 C 0.21 0.52 0.12 0.71 0.33 1.1x10-4
700 C 0.51 0.32 0.22 0.70 0.73 3.1x10-4
17
CA 02893153 2015-05-28
650 C 1.18 0.27 0.49 0.67 1.67 3.5x10-4
600 C 2.99 0.22 1.24 0.61 4.24 4x10-4
The Arrhenius plot of the total OCP polarization resistance for the LCFCr
material in air, obtained from the data of FIG. 4A, is presented in FIG. 5.
According to
the fitting parameters shown in Table 1-2, the resistance of the low frequency
arc is
approximately 90% of the total Rp and thus the activation energy associated
with this
arc will be dominant. As shown in FIG. 5, good linearity of the plot of the
polarization
resistance versus the inverse of temperature is obtained. The derived
activation energy
(Ea) for the ORR is 125 kJ/mol, which is lower than previously reported for
well-known
cathode materials, such as LSM (173.7 kJ/mol 1140, 411) and LSCF (178.5 kJ/mol
[421)
at the OCP in air. The lower Ea indicates that the LCFCr material is a better
catalyst for
the ORR than these two materials. Furthermore, according to the literature,
this range of
activation energies may indicate that oxygen diffusion in the gas phase is one
of the
slow steps of the reaction [36].
3.3.2 Performance of LCFCr under anodic and cathodic polarization
To further investigate the medium-term electrochemical stability of the LCFCr
air electrode for RSOFC applications, potentiostatic experiments at 800 C, at
overpotentials of 0.4 V (OER) and -0.4 V (ORR), were performed for 100 h. In
FIG. 6,
a degradation rate of 0.59 mA hi is seen over 100 h at the anodic 0.4 V
overpotential.
Impedance measurements, however, show that Rp is very similar before (0.35 Q
cm2
cm2) and after (0.30 12 cm2) the 100 h test at 0.4 V, demonstrating very good
medium-
term stability of the LCFCr air electrode performance under typical OER
operating
conditions.
These experiments were performed 24 days after commencing cell testing (FIG.
4) and some degradation of the cell performance has clearly occurred, as seen
by
comparing the results in Figs. 6 and 7 with those in FIG. 4. However, the
ohmic
resistance (Rs) is the main cause of this degradation, having changed from
0.77 to 0.99
cm2, likely due to the sintering of the current collectors. The shift of the
summit
frequency (41 Hz before testing and 2.58 Hz after testing) may be consistent
with the
densification of the Au current collector. Thus, the majority of the
degradation seen in
FIG. 6 is thus due to this increase in Rs. In support of this conclusion, our
previously
18
CA 02893153 2015-05-28
published WDX elemental map studies did not reveal an incompatibility issue
between
LCFCr and GDC [43].
The medium-term electrochemical stability of the LCFCr air electrode towards
the oxygen reduction reaction (ORR) was then investigated at a -0.4 V
overpotential,
again at 800 C for 100 h. In FIG. 7, a loss in current of 0.67 mA is seen
over this
time period, which suggests that LCFCr experiences a slightly faster
degradation as an
ORR catalyst than during the OER. Impedance measurements performed before and
after the potentiostatic experiment (FIG. 7) show that Rp increases from 0.25
C2 cm2
before cathodic polarization to 0.30 cm after the 100 h test at -0.4 V.
However, Rs
does not change, and, in fact, has the same value as that before the anodic
(+0.4 V)
polarization experiment in FIG. 6. These observations show that LCFCr performs
more poorly as an ORR catalyst than during the OER. Furthermore, the fact that
Rs in
HG. 7 has recovered to its original OCP value before anodic polarization (FIG.
6)
demonstrates that LCFCr is an excellent air electrode for the OER, and that
the loss in
performance in FIG. 6 is not permanent (the losses observed here in Rs appear
to be
reversible). Sintering of the current collectors remains the most likely
reason for the
increase of Rs with time. Furthermore, it is evident that, when the
polarization was
switched from +0.4 V (FIG. 6) to -0.4 V (FIG. 7), Rs fully recovered. Thus, it
is
plausible that dewetting of the gold current collector, which may have
occurred as a
result of sintering at +0.4 V, may have reversed upon the change of
polarization
direction. This is consistent with the known effect of electrical potential on
interfacial
tensions [44, 45].
Overall, the LCFCr material is seen to be an excellent air electrode, giving
Rp
values in the range or even lower than the best SOFC cathode materials
discussed in the
literature in this temperature range. For example, LSCF has exhibited an Rp
value of
0.18 SI cm2 at the OCP [11] and LSFCu an Rp of 0.07 Q cm2, both at 800 C [12].
3.33 Cell microstructure
The typical microstructure of the cell, examined by back-scattered SEM, is
shown after electrochemical testing in FIGs. 8A and 8B. The cell consists of a
1 mm
dense GDC electrolyte layer, with only one of the two LCFCr electrode layers
(¨ 30 pm
thick) shown. Also, a gold current collector layer is shown in the image
(white phase in
FIG. 8A, displaying very good porosity at higher magnification (FIG. 8B). For
comparison, FIG. 8C shows a somewhat higher magnification image (vs. FIG. 8A)
of
19
CA 02893153 2015-05-28
the microstructure and the interface of a cell before electrochemical
analysis, for
comparison.
The LCFCr/GDC interface after cell testing at both 0.4 V and -0.4 V, each for
100 h, is seen in Figs. 8A and 8B to have retained a continuous good contact
between
the LCFCr electrodes and the GDC electrolyte, with no delamination or cracking
detected. As mentioned earlier, the delamination of the oxygen electrode from
the
electrolyte is the most common degradation and cell failure issue for high
temperature
electrolysis cells [2]. Thus, the fact that our LCFCr-based symmetrical cell
did not show
any electrode delamination (FIG. 8B) after long times under both anodic and
cathodic
polarization is very encouraging.
3.4. Conclusions
In this example, Ca was substituted for Sr in the A site of
LaØ3Sr0.7Fe07Cr0.303_6
(LSFCr), producing a novel mixed conducting perovskite material
La0.3Ca0.7Fe07Cr0 303_5 (LCFCr). These oxides are being developed for
application as air
electrodes for use in reversible solid oxide fuel cells (RSOFCs). Due to the
smaller
ionic radius of Ca vs. Sr, we were expecting to decrease the thermal expansion
coefficient of the perovskite to more closely match that of commonly employed
RSOFC
electrolytes.
In this example, LCFCr was prepared by using the combustion method. XRD
analysis showed that the LCFCr powders are a pure crystalline phase, also
confirmed by
TEM analysis, with an orthorhombic perovskite structure. It was also shown
that the
average TEC values match closely with that of gadolinium-doped ceria (GDC),
the
electrolyte used here.
Electrochemical measurements showed very good performance, overall, with
open circuit potential (OCP) polarization resistances (Re) comparable to what
has been
reported for other well-known perovskites, used in air, at 600 - 800 C.
Further, the
activation energy of the oxygen reaction at LCFCr, at the OCP, was found to be
lower
than literature values for other well-known air electrodes. The medium-term
electrochemical stability of the LCFCr air electrode towards the OER (0.4 V)
and ORR
(-0.4 V) was also investigated at 800 C for 100 h, showing that Rp hardly
changes
during the OER, but increases by ca. 20 % during the ORR. SEM imaging of the
LCFCr/GDC interface showed no delamination or other forms of physical
degradation
of the cell after 100 hours at both 0.4 V and -0.4 V. Thus, it is clear that
LCFCr is a
very promising air electrode material for RSOFC applications.
CA 02893153 2015-05-28
References for Example 1
[1] A.V. Virkar, Mechanism of oxygen electrode delamination in solid oxide
electrolyzer cells, International Journal of Hydrogen Energy, 35 9527-9543.
[2] M.A. Laguna-Bercero, Recent advances in high temperature electrolysis
using solid
oxide fuel cells: A review, Journal of Power Sources, 203 4-16.
[3] M.A. Laguna-Bercero, J.A. Kilner, S.J. Skinner, Development of oxygen
electrodes
for reversible solid oxide fuel cells with scandia stabilized zirconia
electrolytes, Solid
State Ionics, 192 501-504.
[4] R.N. Blumenthal, R.K. Sharma, Electronic conductivity in nonstoichiometric
cerium
dioxide, Journal of Solid State Chemistry, 13 (1975) 360-364.
[5] P. Mocoteguy, A. Brisse, A review and comprehensive analysis of
degradation
mechanisms of solid oxide electrolysis cells, International Journal of
Hydrogen Energy,
38 (2013) 15887-15902.
[6] J. Lu, Y.-M. Yin, Z.-F. Ma, Preparation and characterization of new cobalt-
free
cathode Pr0.5Sr0.5Fe0.8Cu0.203-6 for IT-SOFC, International Journal of
Hydrogen
Energy, 38 (2013) 10527-10533.
[7] L. Zhao, B. He, X. Zhang, R. Peng, G. Meng, X. Liu, Electrochemical
performance
of novel cobalt-free oxide Ba0.5Sr0.5Fe0.8Cu0.203-43 for solid oxide fuel cell
cathode,
Journal of Power Sources, 195 (2010) 1859-1861.
[8] L. Zhao, B. He, Y. ling, Z. Xun, R. Peng, G. Meng, X. Liu, Cobalt-free
oxide
Ba0.5Sr0.5Fe0.8Cu0.203¨ti for proton-conducting solid oxide fuel cell cathode,
International Journal of Hydrogen Energy, 35 (2010) 3769-3774.
[9] A. Egger, E. Bucher, M. Yang, W. Sitte, Comparison of oxygen exchange
kinetics
of the IT-SOFC cathode materials La0.5Sr0.5Co03-d and La0.6Sr0.4Co03, Solid
State
Ionics, 225 (2012) 55-60.
[10] J.W. Fergus, Lanthanum chromite-based materials for solid oxide fuel cell
interconnects, Solid State Ionics, 171 (2004) 1-15.
[11] M. Chen, B.H. Moon, S.H. Kim, B.H. Kim, Q. Xu, B.G. Ahn, Characterization
of
La0.6Sr0.4Co0.2Fe0.803-6 + La2Ni04+6 Composite Cathode Materials for Solid
Oxide Fuel Cells, Fuel Cells, 12(2012) 86-96.
21
CA 02893153 2015-05-28
[12] Q. Zhou, L. Xu, Y. Quo, D. Jia, Y. Li, W.C.J. Wei, La0.6Sr0.4Fe0.8Cu0.203-
8
perovskite oxide as cathode for IT-SOFC, International Journal of Hydrogen
Energy, 37
(2012) 11963-11968.
[13] J.C. Ruiz-Morales, D. Marrero-Lopez, J. Canales-Vazquez, J.T.S. Irvine,
Symmetric and reversible solid oxide fuel cells, RSC Advances, 1 (2011) 1403-
1414.
[14] M. Chen, S. Paulson, V. Thangadurai, V. Birss, Sr-rich chromium ferrites
as
symmetrical solid oxide fuel cell electrodes, Journal of Power Sources, 236
(2013) 68-
79.
[15] I.A.L. V.L. Kozhevnikov, J.A. Bahteeva, M.V. Patrakeev, E.B. Mitergõ
C.M.e.
K.R. Poeppelmeier.
[16] J. Prado-Gonjal, R. Schmidt, J.-J. Romero, D. Avila, U. Amador, E. Moran,
Microwave-Assisted Synthesis, Microstructure, and Physical Properties of Rare-
Earth
Chromites, Inorganic Chemistry, 52 (2012) 313-320.
[17] Q. Xu, D.-p. Huang, W. Chen, F. Zhang, B.-t. Wang, Structure, electrical
conducting and thermal expansion properties of Ln0.6Sr0.4Co0.2Fe0.803
perovskite-
type complex oxides, Journal of Alloys and Compounds, 429 (2007) 34-39.
[18] F. Riza, C. Ftikos, F. Tietz, W. Fischer, Preparation and
characterization of
Ln0.8Sr0.2Fe0.8Co0.203¨x (Ln=La, Pr, Nd, Sm, Eu, Gd), Journal of the European
Ceramic Society, 21 (2001) 1769-1773.
[19] V.V. Kharton, A.V. Kovalevsky, M.V. Patrakeev, E.V. Tsipis, A.P. Viskup,
V.A.
Kolotygin, A.A. Yaremchenko, A.L. Shaula, E.A. Kiselev, J.o.C. Waerenborgh,
Oxygen Nonstoichiometry, Mixed Conductivity, and MOssbauer Spectra of
Ln0.5A0.5Fe03-43 (Ln = La¨Sm, A = Sr, Ba): Effects of Cation Size, Chemistry
of
Materials, 20 (2008) 6457-6467.
[20] M.P. J.-C. Grenier, P. Hagenrnuller. Vacancy ordering in oxygen-deficient
perovskite-related ferrites, in: Ferrites = Transitions Elements Luminescence
Structure
and Bonding 1981, pp. 1-25
[21] G. Corbel, S. Mestiri, P. Lacorre, Physicochemical compatibility of CGO
fluorite,
LSM and LSCF perovskite electrode materials with La2Mo209 fast oxide-ion
conductor, Solid State Sciences, 7 (2005) 1216-1224.
[22] A. Hammouche, E. Siebert, A. Hammou, Crystallographic, thermal and
electrochemical properties of the system La 1 ¨xSrxMn03 for high temperature
solid
electrolyte fuel cells, Materials Research Bulletin, 24 (1989) 367-380.
22
CA 02893153 2015-05-28
[23] M. Mori, Y. Hiei, H. Itoh, G.A. Tompsett, N.M. Sammes, Evaluation of Ni
and Ti-
doped Y203 stabilized ZrO2 cermet as an anode in high-temperature solid oxide
fuel
cells, Solid State Ionics, 160 (2003) 1-14.
[24] L.W. Tai, M.M. Nasrallah, H.U. Anderson, D.M. Sparlin, S.R. Sehlin,
Structure
and electrical properties of La 1 ¨ xSrxCo 1 ¨ yFey03. Part 2. The system La 1
¨
xSrxCo0.2Fe0.803, Solid State Ionics, 76 (1995) 273-283.
[25] S. Sameshima, T. Ichikawa, M. Kawaminami, Y. Hirata, Thermal and
mechanical
properties of rare earth-doped ceria ceramics, Materials Chemistry and
Physics, 61
(1999) 31-35.
[26] E.Y. Pikalova, A.N. Demina, A.K. Demin, A.A. Murashkina, V.E. Sopernikov,
N.O. Esina, Effect of doping with Co203, TiO2, Fe2O3, and Mn203 on the
properties
of Ce0.8Gd0.202-6, Inorg Mater, 43 (2007) 735-742.
[27] H. Hayashi, M. Kanoh, C.J. Quan, H. Inaba, S. Wang, M. Dokiya, H. Tagawa,
Thermal expansion of Gd-doped ceria and reduced ceria, Solid State Ionics, 132
(2000)
227-233.
[28] S. Omar, J.C. Nino, Consistency in the chemical expansion of fluorites: A
thermal
revision of the doped ceria, Acta Materialia, 61(2013) 5406-5413.
[29] T. Hisashige, Y. Yamamura, T. Tsuji, Thermal expansion and Debye
temperature
of rare earth-doped ceria, Journal of Alloys and Compounds, 408-412 (2006)
1153-
1156.
[30] V. Prashanth Kumar, Y.S. Reddy, P. Kistaiah, G. Prasad, C. Vishnuvardhan
Reddy,
Thermal and electrical properties of rare-earth co-doped ceria ceramics,
Materials
Chemistry and Physics, 112 (2008) 711-718.
[31] C. Ftikos, M. Nauer, B.C.H. Steele, Electrical conductivity and thermal
expansion
of ceria doped with Pr, Nb and Sn, Journal of the European Ceramic Society, 12
(1993)
267-270.
[32] S.V. Rompaey, W.D. , S.T. , O.Y.P. , H.T. J.V. , A.A. , J. Hadermann,
Layered oxygen vacancy ordering in Nb-doped SrCo 1-x x 0 3 d, Z. Kristallogr.,
228
(2013) 28-34.
[33] J.-C. Boivin, Structural and electrochemical features of fast oxide ion
conductors,
International Journal of Inorganic Materials, 3 (2001) 1261-1266.
[34] K.R. Cooper, M. Smith, Electrical test methods for on-line fuel cell
ohmic
resistance measurement, Journal of Power Sources, 160 (2006) 1088-1095.
23
CA 02893153 2015-05-28
[35] J.-B. Jorcin, M.E. Orazem, N. Mere, B. Tribollet, CPE analysis by local
electrochemical impedance spectroscopy, Electrochimica Acta, 51(2006) 1473-
1479.
[36] M.J. Escudero, A. Aguadero, J.A. Alonso, L. Daza, A kinetic study of
oxygen
reduction reaction on La2Ni04 cathodes by means of impedance spectroscopy,
Journal
of Electroanalytical Chemistry, 611(2007) 107-116.
[37] M.J. JAJgensen, M. Mogensen, Impedance of Solid Oxide Fuel Cell LSM/YSZ
Composite Cathodes, Journal of The Electrochemical Society, 148 (2001) A433-
A442.
[38] S. Li, Z. LA1/4, X. Huang, W. Su, Thermal, electrical, and
electrochemical
properties of Nd-doped Ba0.5Sr0.5 Co0.8Fe0.203 d" 1" as a cathode
material for SOFC, Solid State Ionics, 178 (2008) 1853-1858.
[39] V. Dusastre, J.A. Kilner, Optimisation of composite cathodes for
intermediate
temperature SOFC applications, Solid State Ionics, 126 (1999) 163-174.
[40] Y.H. Wu, Karin Vels; Norrman, Kion; Jacobsen, Torben; Mogensen, Mogens
Bjerg., Oxygen Electrode Kinetics and Surface Composition of Dense
(La0.75Sr0.25)0.95Mn03 on YSZ. /, E C S Transactions, 57 (2013) 1673-1682.
[41] Y. Takeda, R. Kanno, M. Noda, Y. Tomida, 0. Yamamoto, Cathodic
Polarization
Phenomena of Perovskite Oxide Electrodes with Stabilized Zirconia, Journal of
The
Electrochemical Society, 134 (1987) 2656-2661.
[42] B.Y. Yoon, J. Bae, Electrochemical Investigation of Composite Nano
La0.6Sr0.4Co0.2Fe0.803-6 Infiltration into LSGM Scaffold Cathode on LSGM
Electrolyte, ECS Transactions, 57 (2013) 1933-1943.
[43] B. Molero-Sanchez, P. Addo, M. Chen, S. Paulson, V. Birss,
La0.3Ca0.7Fe0.7Cr0.303-8 as a Novel Air Electrode Material for Solid Oxide
Electrolysis Cells, in: 11 th European SOFC & SOE FORUM 2014, Luzern,
Switzerland, 2014, pp. B 0804.
[44] M. Mosialek, E. Bielanska, R.P. Socha, M. Dudek, G. Mordarski, P. Nowak,
J.
Barbasz, A. Rapacz-Kmita, Changes in the morphology and the composition of the
AglYSZ and AgILSM interfaces caused by polarization, Solid State Ionics, 225
(2012)
755-759.
[45] M. Mosialek, M. Dudek, P. Nowak, R.P. Socha, G. Mordarski, E. Bielanska,
Changes in the morphology and the composition of the Ag Gd0.2Ce0.801.9
interface
caused by polarization, Electrochimica Acta, 104 (2013) 474-480.
24
CA 02893153 2015-05-28
Example 2: Microwave-assisted synthesis
1. Introduction
In this example, alternative powder processing methods are examined, with a
primary focus on microwave-based synthesis, that could both lower material
manufacturing costs and further enhance cathode performance for solid oxide
fuel cell
applications. La0.3Ca07Fe07Cr0.303_6 (LCFCr), formed using conventional solid-
state
methods, has been shown in earlier work to be a very promising catalyst for
the oxygen
reduction reaction. To further increase its performance, microwave methods
were used
to increase the surface area of LCFCr and to decrease the processing time. It
was found
that the material could be obtained in crystalline form in only 8 hours, with
the
synthesis temperature lowered by roughly 300 C as compared to conventional
methods.
Current research in the solid oxide fuel cell (SOFC) field is moving towards
the use of
mixed ionic and electronic conducting oxides (M1EC), which have been shown to
be
more durable as cathodes than conventional La1_õSrxMn03 (LSM) materials.
Usually,
MIECs are synthesized by solid-state reactions, where the process involves
multiple
heating (> 1200 C) and regrinding steps to help overcome the solid-state
diffusion
barrier (1, 2). Some of the methods by which MIECs have been traditionally
prepared
include the sol-gel method (3), the EDTA citrate complexing process (4), the
auto-
ignition process (5), the Pechini method (6), and most commonly, by using
combustion
methods (7).
In order to enhance the diffusion rate of the ceramic precursors by several
orders
of magnitude, thus shortening the reaction time and potentially lowering the
reaction
temperature, there has been an interest in the use of microwave-assisted
methods.
Furthermore, it is possible to induce interesting changes in particle
morphology and
sizes using microwave methods (8). Also, microwave-assisted techniques are
understood to be environmentally friendly, as they require less energy than
conventional
material processing methods. This makes microwave synthesis an example of
"Green
Chemistry" or "Sustainable Chemistry" (8, 9).
The main features that distinguish microwave synthesis from conventional
methods are faster energy transfer rates, i.e., more rapid heating rates, and
the selective
heating of materials. This leads to a unique temperature distribution within
the material
when it is heated in a microwave furnace. During conventional heat treatment,
energy is
transferred to a material through thermal conduction and convection, creating
thermal
gradients. However, in the case of microwave heating, energy is transferred
directly to
CA 02893153 2015-05-28
the material through an interaction of the material at the molecular level
with the
electromagnetic waves (10). The most important contribution in microwave
heating
may be that the dipoles in the material follow the alternating electromagnetic
field
associated with the microwave, with its rapidly changing electric field (ca.
2.4 x 109
times per second). The resistance to this movement generates a considerable
amount of
heat (11, 12), thus leading to more rapid heating rates.
It has been suggested that, the more complex a material is, the more difficult
it is
to prepare by using microwave-assisted synthesis. In more complex systems,
very good
diffusion is required to uniformly disperse three or more cations throughout
the sample
during the synthesis. The usual solution to this problem is to combine
microwave
irradiation with other methods, such as sol-gel or combustion synthesis, as
has been
done for the synthesis of complex perovskites, such as La0.8Sr0.2Fe05Co0.503
or
CaCu3Ti401 2 ( 1, 13).
Previous research in our group has focused on the development of a sulphur and
coke tolerant electrode-supported SOFCs, most recently based on
LaØ3Sr0.7Fe0.7Cr0.303-6
(LSFCr) as a mixed ionic-electronic conducting perovskite material. LSFCr has
shown
very good catalytic activity for both H2/C0 oxidation and 02 reduction, thus
potentially
having use in symmetrical SOFCs (7). In order to take advantage of the
excellent
performance of LSFCr, the A-site of the perovskite was doped with Ca instead
of Sr, as
Ca has a smaller ionic radius than Sr. The goal was to decrease the thermal
expansion
coefficient of this derivative of LSFCr, i.e., Lao 3Ca07Fe07Cr0 303_6 (LCFCr),
to more
closely match that of a Gd-doped ceria (GDC) electrolyte. The partial
substitution of Ca
for Sr may also enable the introduction of structural inhomogeneties, as Ca
doping of
LaFe03 is known to promote oxygen-vacancy ordering (14, 15).
In the present study, we are focused on the synthesis and characterization of
LCFCr, formed using three different methods, regular combustion (Method 1),
microwave-assisted combustion (Method 2), and microwave-assisted sol-gel
synthesis
(Method 3). We show that a single phase material can be successfully
synthesized
using microwave-assisted methods and that we can also lower the calcination
temperature by 200-300 C using this approach. Our recent work on LCFCr,
synthesized
using Method 1, has shown very good electrochemical characteristics (16). For
this
reason, parallel comparison studies (17) of the performance of LCFCr-based
cathodes,
constructed using the three methods described in this paper, are currently
being carried
out.
26
CA 02893153 2015-05-28
2. Material Synthesis
La0.3Ca0.7Cr0.3Fe0.703_8 (LCFCr) powders were synthesized using three
different
methods, the regular combustion method (Method 1), microwave-assisted
combustion
(Method 2), and microwave-assisted sol-gel synthesis (Method 3). When
synthesized
using the regular combustion method (Method 1), the metal nitrates were mixed
in
stoichiometric proportions and dissolved in deionized water. A 2:1 mole ratio
of glycine
to the total cation content was used. Solutions were slowly stirred on a hot
plate until
auto-ignition and self-sustaining combustion occurred. Then the sample was
calcined in
air at 1200 C.
LCFCr powders were also synthesized by microwave-assisted combustion
(Method 2). Here, the metal nitrates and glycine were dissolved in deionized
water
using the metal cation proportions required to generate the correct oxide
stoichiometry.
A 2:1 mole ratio of glycine to the total metal content was used. The stirred
solutions
were introduced into the microwave furnace and exposed to a 2.45 GHz frequency
and
800 W power for 30 minutes. When the water had evaporated, combustion
occurred.
Then the sample was calcined in air at 700 C, 900 C and 1000 C for 8 h in
order to
decompose the organic remnants, rendering a black powder as the final product.
In Method 3, microwave energy and a sol¨gel methodology were combined to
produce the LCFCr powders, with the metal cation proportions used based on the
desired stoichiometry. Metal nitrates were dissolved in distilled water and a
saturated
polyvinyl alcohol (PVA) solution was added to serve as the complexing agent.
The
amount of PVA added was such that the ratio of the total number of moles of
cations to
that of PVA was 1:2. Then the final solution was maintained at 80 C for 1.5 h
to form
a viscous gel. The gel was then microwave irradiated (up to 30 min) in a
porcelain
crucible placed inside another larger one filled with mullite. The microwave
source
operated at a 2.45 GHz frequency and 800 W power and was uniquely able to
handle
the conditions needed in this work. The polymeric and sponge-like-precursor
was then
calcined in air at 700 C, 900 C and 1000 C for 8 h in order to decompose
the organic
remnants, rendering a black powder as the final product.
2. Material characterization
X-ray diffraction (XRD) patterns of all samples synthesized in this work were
collected using a Philips X'Pert PRO ALPHA I of Panalytical B.V.
diffractometer with
Cu K monochromatic radiation (X = 1.54056 A). The diffractometer was equipped
27
CA 02893153 2015-05-28
with a primary curved Gelll primary beam monochromator and a speed X'Celerator
fast detector, operating at 45 kV and 40 mA. XRD patterns were collected in
the 20
range of 5 ¨ 120 at room temperature, with a step size of 0.0170 and 8 s
counting time,
in order to ensure sufficient resolution for structural refinement.
Fullprof Software was employed to carry out structural refinements from
conventional XRD patterns using the Rietveld method. This method of refining
the
powder diffraction data was used to determine the crystal structure. Zero
shift, lattice
parameters, background, peak width, shape and asymmetry, atomic positions and
isotropic temperature factors were all refined. The Thompson¨Cox-Hastings
pseudo-
Voigt convoluted with axial divergence asymmetry function was used to describe
the
peak shape. Linear interpolation between set background points with refinable
heights
was used afterwards. The values were refined to improve the agreement factors.
All samples investigated by scanning electron microscopy (SEM) were first
sputter-coated with Au in an EMITECH K550 apparatus. Field-emission scanning
electron microscopy (FE-SEM) was performed using a JEM 6335 F electron
microscope with a field-emission gun operating at 10 kV. The FE-SEM was also
equipped with a LINK ISIS 300 detector for the energy-dispersive analysis of
the X-
rays (XEDS).
High resolution transmission electron microscopy (HRTEM) analysis of the
LCFCr powders was performed using a JEOL 3000F TEM, yielding an information
limit of 1.1 A. Images were recorded with an objective aperture of 70 pm,
centered on a
sample spot within the diffraction pattern area. Fast Fourier Transforms (141-
1s) of the
HRTEM images were carried out to reveal the periodic image contents using the
Digital
Micrograph package. The experimental HRTEM images were also compared to
simulated images using MacTempas software. These computations were performed
using information from the structural parameters, obtained from the Rietveld
refinement, the microscope parameters, such as microscope operating voltage
(300 kV)
and spherical aberration coefficient (0.6 mm), and specimen parameters, such
as zone
axis and thickness. The defocus and sample thickness parameters were optimized
by
assessing the agreement between model and data.
3. Results and Discussion
3.1 Microwave-assisted synthesis of LCFCr powders: X-ray diffraction and
Rietveld refinement
28
CA 02893153 2015-05-28
FIG. 9(a) shows the XRD patterns of the LCFCr powders synthesized by the
combustion method (Method 1), as well as by microwave-assisted combustion
(Method
2), and microwave-assisted sot-gel synthesis (Method 3). The diffraction
patterns show
that a pure crystalline phase is obtained for all three synthesis methods.
Importantly, the
temperature used did not exceed 1000 'V, and without the use of microwave
methods,
the normal temperature that would have been needed to achieve the same result
is 1200
C.
FIG. 9(b) shows the XRD patterns for the material synthesized by the
microwave-combustion method (Method 2) and calcined at three different
temperatures.
It can be seen that, at 700 C, the phase is already forming and at 900 C,
the crystalline
phase for LCFC has formed FIG. 9(c) shows the XRD patterns for the material
synthesized by the microwave-assisted sol-gel (Method 3) and calcined at the
same
temperatures. It can be seen that, at 700 C and 900 C, the desired phase is
already
forming and similar to Method 2, at 1000 C, the desired product is present in
the pure
form
FIG. 10 shows the Rietveld refinement fits for the LCFCr samples produced by
microwave-combustion method (Method 2, FIG. 9a) and synthesized by the
microwave-
assisted sol-gel synthesis (Method 3, FIG. 9b). The Rieltveld refinement for
the LCFCr
powders synthesized by the regular combustion method (Method 1) has been
carried out
in our parallel comparison studies (17). A distorted perovskite structure with
an
orthorhombic symmetry (S.G. Pnma, #62) was confirmed for both samples. The
unit
cell vectors can be represented by \i2ap x 2ap x -\12a2, where ap refers to
the simple cubic
perovskite cell. The results obtained for both samples concerning the cells
parameters
and the atomic positions are summarized in Table 2-1.
3.2 Microstructural analysis of LCFCr powders synthesized using microwave-
assisted methods
3.2.1 Scanning (SEM) and transmission electron microscopy (TEM)
FIG. 11 shows the SEM images of LCFCr powders formed using microwave-
assisted combustion synthesis at 900 C (Method 2, FIG. 11(a)) and microwave-
assisted
sol-gel synthesis and (Method 3, FIG. 11(b)). As can be seen, in both cases,
the material
has a porous morphology, which makes it a good candidate as an electrode
material. A
sponge-like porous morphology can be observed for the powders formed using
Method
2 (FIG. 11(a)), which is thetypical morphology found after combustion
processes. The
29
CA 02893153 2015-05-28
sponge-like porous morphology fromMethod 2 is quite different from the
morphology
obtained using the sol-gel method(Method 3, FIG. 11(b)) whichconsists of quite
homogeneous agglomerated particles (approximate size 400nm).
Table 2-1: Structural parameters for LCFCr obtained from Rietveld refined XRD
data.
LCFCr
Conventional MW - MW- Sol-gel
combustion combustion (method 3)
(method 1) (method 2)
a (A) 5.4550(2) 5.4615 (8) 5.4476(4)
b(ii) 7.7128 (1) 7.7470 (7) 7.7194(2)
c60 5.4552 (2) 5.4619 (7) 5.4504(4)
La / Ca position 4c:
0.0145(6) 0.01959 (7) 0.0151(6)
-0.003(3) -0.003 (1) -0.0062(7)
Occ (La / Ca) 030(1)/ 030(1)/030(1) 030(1)/030(1)
030(1)
U*100 (A2) 0.40(3) 0.44(4) 0.52(2)
Fe / Cr position 4b:
Occ (Fe / Cr) 030(1)/ 0.70(1)/030(1) 030(1)/030(1)
030(1)
U*100 (A') 0.35(2) 0.32(3) 0.43(2)
0(1) position 4c:
0.502(2) 0.503(4) 0.501(3)
0.105(2) 0.106(4) 0.106(4)
Occ 1.00(1) 1.00(1) 1.00(1)
u*ion (V) 0.44(2) 0.27(3) 0.41(5)
0(2) position 8d:
0.297(4) 0.256(2) 0.257(3)
0.003(4) 0.005(3) 0.005(2)
-0.254(3) -0.31(2) -0.30(3)
occ 1.00(1) 1.00(1) 1.00(1)
u*100 (V) 0.33(2) 0.27(3) 0.41(5)
x2 1.25 1.58 1.74
4.88 / 4.37 2.42 / 1.96 2.65 /2.01
(%/%)
Rragg 7.30 3.83 5.11
S.G. Pnma: 4c (x 'Az), 4b (0 0 1/2), 8d (xyz)
CA 02893153 2015-05-28
Further characterization of the material obtained by microwave-assisted sol-
gel
synthesis (Method 3) at 1000 C and microwave-assisted combustion synthesis at
900
C (Method 2) was performed, with Table 2-2 giving the elemental analysis of
the
materials, obtained from the regions in the squares in FIG. 11. Table 2-2
shows the
atomic percentage of each component in the catalysts. The second column of
results
corresponds to the microwave-assisted combustion synthesis (Method 2) and the
third
column corresponds to microwave-assisted sol-gel synthesis (Method 3). The
atomic
percentage observed by EDX is comparable to the theoretical values, based on
the
expected stoichiometry of La0.3Ca0.7Fe0.7Cr0.303_6.
Table 2-2. EDX-determined composition (atomic %) of LCFCr powders, formed
by microwave-assisted combustion (Method 2) and microwave-assisted sol-gel
synthesis (Method 3) approaches
Atomic % composition of La0.3Caoffe0.7Cr0.303.6
MW & comb MW & sol-gel Theoretical
La 16+0.5 16+0.5 15
Ca 35+0.5 34+0.5 35
Cr 15+0.5 15+0.5 15
Fe 34+0.5 36+0.5 35
Table 2-3
Specific surface areas of l_CFCr powders. tOnned by ree.ular combustion
Method I and iniciowave-assisted combustion (Method 2, approaches.
Sample SHE r 1112
Regular combustion (Method I( OP)
Microwave-assisted combustion (Method 2i 10.1
Transmission Electron Microscopy analysis was also performed on the LCFCr
powder obtained using the different synthetic methods described above. The
cation
composition, measured semi-quantitatively by X-ray energy dispersive
spectroscopy in
31
CA 02893153 2015-05-28
more than ten single crystals is in good agreement with the theoretical
proportions in
La0.3Ca03Cro.3Fe0.703_8, indicating the high purity of the powder.
In the HTREM images of the crystals prepared by microwave-assisted
combustion synthesis (Method 2) (FIG. 12a) and assisted sol-gel synthesis
(Method 3)
(FIG. 12b), nanosized twinned domains are seen. The appearance of these
domains can
be associated with the pseudo-cubic nature of these materials. Furthermore,
their
presence avoids the formation of tetrahedral chains and therefore the
formation of
undesired brownmillerite-type defects (18). We have only detected the
formation of
defects in the sample prepared by sol-gel synthesis, Method 3 (FIG. 12b),
showing a
periodicity of 1.12 nm. This corresponds to the c axis of the A3B308 type
structure,
which results from the intergrowth of a perovskite ABO3 and a brownmillerite
phase.
HTREM images of the crystals prepared by the regular combustion method (Method
1)
are shown in our parallel electrochemical study (17). It is worth noting that
the
microwave-assisted combustion synthesis (Method 2), as it involves very fast
processes,
favors disordered phases (perovskite in our case).
4. Conclusions
The mixed ion-electron conducting perovskite, La0.3Ca0.7Fe0.7Cr0.303-8 (1-
CFCr),
was prepared here by using several microwave-assisted methods, for ultimate
use as a
cathode in solid oxide fuel cells (SOFCs). The material was successfully
prepared by
microwave-assisted combustion (Method 2) and microwave-assisted sol-gel
synthesis
(Method 3). The desired product was obtained in crystalline form in only 7 hrs
(vs. 13
hrs) and the synthesis temperature was roughly 300 C lower than what is
required for
conventional solid-state combustion synthesis. This new approach has enhanced
the rate
of formation of the LCFCr powder by several orders of magnitude, and also
increased
the specific surface area from 0.89 to 10.4 m2 These results
are very encouraging, as
they suggest that microwave synthesis can be used in the preparation of the
perovskite
materials used in this work. Whether the microwave-synthesized materials will
give
superior electro- chemical performance is currently under investigation. It is
suggested
here that the partial substitution of Ca for Sr may promote oxygen-vacancy
disordering
and thus stabilize the perovskite phase vs. the brownmillerite phase. In our
HRTEM
work, the formation of brownmillerite-type defects was detected only in the
sample
prepared by sol¨gel synthesis (Method 3). In addition, the calcination
temperature for
microwave-assisted combustion (Method2) was 900 C vs. 1000 C for microwave-
assisted sol¨gel synthesis (Method 3). Based on the lower calcination
temperature and
32
CA 02893153 2015-05-28
the absence of brownmillerite-type defects, microwave-assisted combustion
(Method 2)
would be the preferred method for the future synthesis of highly active SOFC
cathodes
composed of the La0.3Ca07Fe07Cr0.303_6material.
References for Example 2
2-1. Prado-Gonjal J, Schmidt R, Romero J-J, Avila D, Amador U, Moran E.
Microwave-Assisted Synthesis, Microstructure, and Physical Properties of Rare-
Earth
Chromites. Inorganic Chemistry. 2012 2013/01/07;52(1):313-20.
2-2. Kitchen HJ, Valiance SR, Kennedy JL, Tapia-Ruiz N, Carassiti L, Harrison
A, et
al. Modern Microwave Methods in Solid-State Inorganic Materials Chemistry:
From
Fundamentals to Manufacturing. Chemical reviews. 2013; 114(2):1170-206.
2-3. Lu J, Yin Y-M, Ma Z-F. Preparation and characterization of new cobalt-
free
cathode Pr0.5Sr0.5Fe0.8Cu0.203-6 for IT-SOFC. International Journal of
Hydrogen
Energy. 2013 8/21/;38(25):10527-33.
2-4. Thou Q, Xu L, Guo Y, Jia D, Li Y, Wei WCJ. La0.6Sr0.4Fe0.8Cu0.203-6
perovskite oxide as cathode for IT-SOFC. International Journal of Hydrogen
Energy.
2012 811;37(16):11963-8.
2-5. Zhao L, He B, Zhang X, Peng R, Meng G, Liu X. Electrochemical performance
of novel cobalt-free oxide Ba0.5Sr0.5Fe0.8Cu0.203-6 for solid oxide fuel cell
cathode.
Journal of Power Sources. 2010 4/2/;195(7):1859-61.
2-6. Egger A, Bucher E, Yang M, Sitte W. Comparison of oxygen exchange
kinetics
of the IT-SOFC cathode materials La0.5Sr0.5Co03-d and La0.6Sr0.4Co03. Solid
State
Ionics. 2012 10/4/;225(0):55-60.
2-7. Chen M, Paulson S, Thangadurai V. Birss V. Sr-rich chromium ferrites as
symmetrical solid oxide fuel cell electrodes. Journal of Power Sources. 2013
8/15/;236(0):68-79.
2-8. Prado-Gonjal J, Molero-Sanchez B, Avila-Brande D, Moran E, Perez-Flores
JC,
Kuhn A, et al. The intercalation chemistry of H2V308 nanobelts synthesised by
a
green, fast and cost-effective procedure. Journal of Power Sources. 2013
6/15/;232(0):173-80.
2-9. Prado-Gonjal JS, R.; Moran, E., Microwave-Assisted Synthesis and
Characterization of Perovskite Oxides In Perovskite: Crystallography,
Chemistry and
Catalytic Performance, Zhang, J.; Li, H., Eds. Nova Science Pub Incorporated:
2012; pp
117-140.
33
CA 02893153 2015-05-28
2-10. Gupta ML, E. W. W. Microwaves and Metals. In: Wiley, editor.2008;. p. p
228.
2-11. Zhao JY, W. Microwave-assisted Inorganic Syntheses. In Modern Inorganic
Synthetic Chemistry. In: Amsterdam E, editor.201I. p. pp 173-95.
2-12. Rao KJ, Vaidhyanathan B, Ganguli M, Ramakrishnan PA. Synthesis of
Inorganic Solids Using Microwaves. Chemistry of Materials. 1999
1999/04/01; 11(4):882-95.
2-13. Hutagalung SDI, M. I. M.; Ahmad, Z. A. . Microwave assisted sintering of
CaCu3Ti4012. Ceramics International 2008;34((4)): 939-42.
2-14. Kharton VV, Kovalevsky AV, Patrakeev MV, Tsipis EV, Viskup AP, Kolotygin
VA, et al. Oxygen Nonstoichiometry, Mixed Conductivity, and Mossbauer Spectra
of
Ln0.5A0.5Fe03-6 (Ln = La¨Sm, A = Sr, Ba): Effects of Cation Size. Chemistry of
Materials. 2008 2008/10/28;20(20):6457-67.
2-15. J.-C. Grenier MP, P. Hagenmuller. Vacancy ordering in oxygen-deficient
perovskite-related ferrites. Ferrites = Transitions Elements
Luminescence.Structure and
Bonding 471981. p. 1-25
2-16. Molero-Sanchez BA, Paul ; Chen, Min ; Paulson,Scott and Birss;Viola
editor
La0.3Ca0.7Fe0.7Cr0.303-6 as a Novel Air Electrode Material for Solid Oxide
Electrolysis Cells. 11th European SOFC & SOE FORUM 2014; 2014; Luzern,
Switzerland.
2-17. Molero-Sanchez B, Prado-Gonjal, J., Avila-Brande, D., Chen, M., Moran,
E. and
Birss ,V. High performance La0.3Ca0.7Cr0.3Fe0.703-6 air electrodes for
reversible for
solid oxide fuel cell applications, In preparation. 2014.
2-18. Rompaey SV et al. Layered oxygen vacancy ordering in Nb-doped SrCol-x x
0
3 d. Z Kristallogr. 2013;228:28-34.
34
CA 02893153 2015-05-28
Example 3: Sulfur Tolerance of Lao.3Mo.7FeoiCro.303-6- (M = Sr, Ca) Solid
Oxide
Fuel Cell Anodes
1. Introduction
Solid oxide fuel cells (SOFCs) are highly efficient electrochemical devices
that
also demonstrate excellent fuel flexibility and function well on fuels such as
H2, CO,
and methane. Traditionally, SOFCs have been based on a Ni-YSZ (yttria
stabilized
zirconia) cermet anode, a YSZ electrolyte, and a lanthanum strontium manganite
(LSM)
cathode. Although Ni-YSZ cermets are excellent SOFC anodes, largely because of
their
excellent catalytic activity towards fuel oxidation, work in our group and by
others has
shown that Ni is susceptible to poisoning at low levels (1-100 ppm) of H2S
exposure at
SOFC operating temperatures (700-1000 C) (1-5). It has been suggested that
H2S
inhibits the H2 oxidation reaction (HOR) rates because it readily dissociates
to form a
surface adsorbed Ni-S layer (Sads) on catalytic sites normally involved in 142
dissociation
and subsequent oxidation (4), thereby decreasing the performance of the SOFC.
As a result, extensive research has been carried out to develop sulfur
tolerant SOFC
anode materials based on Ni-free conducting metal oxides, such as perovskites.
For
example, Lao.75Sro.25Cra5Mno.503(LSCM) has been reported to exhibit a
comparable
electrochemical performance for hydrogen oxidation as seen at Ni-YSZ at 900 0C
(6).
However, LSCM has been shown to be less sulfur tolerant in fuels containing
10% H2S
(7). Studies by Mukundan et al (8) showed that Lao.4Sro.6TiO3(LST) is a sulfur
tolerant
SOFC anode, as it did not exhibit any form of degradation in a 5000 ppm H2S +
H2 fuel.
Studies by Haag et al (9) showed that LaSr2Fe2Cr09-0-Gdo.1Ceo.902-6composite
anodes,
exposed to 22 ppm H2S, exhibited only a slight decrease in performance
relative to the
response in H2. Also, Lao.3Sro.7Feo.7Cro.303-6, (LSFCr), operated on wet-(50%
112 CO)
containing 10 ppm H2S, showed only a small drop in cell potential, indicating
very good
stability as an anode in sulfur-containing fuels (10).
However, other perovskites have been reported to show an enhancement in the
rate of hydrogen oxidation in the presence of H2S, including
Lao.7Sro.3V03(LSV) (11),
Smo.95Ceo.o5Feo.97Nio.o303-6(SCFN) (12), and Y0.9 Sr0.1 Cr0.9Fe0.1 03-6 (YSCF)
(13). For
LSV, the observed enhancement was attributed to the formation of an active SrS
phase,
replacing an insulating phase (Sr3V208) (11), while for SCFN and YSCF, it was
suggested that the active phase that forms in the presence of H2S is probably
FeS
(12,13).
CA 02893153 2015-05-28
Previous and current studies in our group have explored the substitution of
the
A-site (M) in the mixed conducting perovskite, Lao.3Mo.7Feo.7Cro.303-s, with
Sr and Ca
to form Lao.3Sro.7Feo.7Cro.303-6(LSFCr) and Lao.3Cao.7Feo.7Cro.303-6(LCFCr)
(10, 14,
15), respectively. These Laa3Mo.7Feo.7Cro.303-6(M = Sr, Ca) perovskite oxides
have
been successfully employed as both fuel and 02 electrodes for SOFC/SOEC
applications. We have reported previously that symmetrical SOFCs, based on the
LSFCr perovskite, showed tolerance to low ppm sulfur content in the fuel
stream, also
exhibited excellent electrochemical activity towards 112 and CO oxidation, and
was also
an active oxygen reduction reaction (ORR) material (10). More recently, we
have also
demonstrated that LCFCr shows excellent catalytic properties for both the ORR
and the oxygen evolution reaction (OER) (15).
However, the performance of LCFCr as a fuel electrode has not been studied as
yet. Therefore, in this example, LCFCr is examined as a SOFC anode in H2
and/or CO
atmospheres, with or without H1S, in comparison to the more well studied
LSFCr.
Electrochemical methods employing both ac and dc techniques were used in a
symmetrical SOFC configuration in 112 and/or CO fuels, with or without the
addition of
9 ppm H2S, all at 800 'C. It is shown that LCFCr is a very good anode catalyst
in H2,
CO, and H2+CO fuels, and although an explanation is not yet available, this
mixed
conducting perovskite material demonstrates a fully reversible enhanced
catalytic
activity when ca. 10 ppm 1-1/S is added to the H2 fuel stream.
Experimental
A glycine-nitrate combustion process was employed to prepare the
La0.3M0.7Fe0.7Cro.303_6(M = Sr, Ca) perovskite powders, using methods reported
previously (10,14,15). The ash obtained from combustion was subsequently
pulverized
and pre-calcined at1200 C for 2 h in air (conditions under which single
phases are
generated). Powders were ballmilled (high energy planetary ball mill,
Pulverisette 5,
Fritsch, Germany) in an isopropanol medium at a rotation speed of 300 rpm for
2 h
using zirconia balls. The La0.3M07Fe0.7Cr0,303_8(M = Sr, Ca) powders were then
screen
printed symmetrically onto both sides of a 275 gm dense YSZ electrolyte coated
with a
porous, ca. 10 micron thick SDC buffer layer (Fuel Cell Materials), followed
by firing
at 1100 C for 2 h. Au paste (C 5729, Heraeus Inc. Germany) was painted on the
Lao.3Mo.7Feo.7Cro.303-8(M = Sr, Ca) layers on both sides of the pellet to
serve as the
current collector and Pt wires were used as the electrical leads.
36
CA 02893153 2015-05-28
The cells were fixed in a FCSH-V3 cell holder (MaterialsMate, Italy) for the
purpose of determining their electrochemical properties. A glass sealant (Type
613,
Aremco Products, USA) was used to isolate the fuel and 02 sides from each
other. The
total flow rates were 25 ml/min and 40 ml/min at the fuel and 02 electrodes,
respectively. The performance of the two electrode cells was evaluated using a
four-
probe method at 800 C. Impedance spectra were collected under both open
circuit and
polarized conditions using a 50 mV perturbation in the frequency range of 0.01
Hz to 60
kHz using a Solartron 1287/1255 potentiostat/galvanostat/impedance analyzer.
2. Results and Discussion
2.1 Performance of 1,ao.3Mo.7Feo.7Cro.303-6 (M = Sr, Ca) anodes in wet 30%
H2/Isiz at 800 "C
To carry out a first stage comparison of the performance of LCFCr and LSFCr,
impedance (EIS) and potentiodymnamic studies were carried out. It can be seen
from
the Nyquist plot in FIG. 13 that a polarization resistance of 1.06 and 1.5
SICM2was
obtained for LCFCr and LSFCr in 30% humidified H2 at 800 C, respectively. The
performance of LCFCr is slightly better than LSFCr. This is consistent with
preliminary
electronic conductivity analysis in H2 at 800 C for LCFCr which gave a value
of 0.6
S/cm, which is a little higher than the 0.2 S/cm value obtained for LSFCr
(10). Despite
this, the EIS response is quite similar for the two materials, with two time
constants
(R/CPE) seen for both LSFCr and LCFCr. From the Bode plot shown in the inset
of
FIG. 13, the dominant summit frequencies are seen to be ca. 100 Hz and ca. 1
Hz.
To determine which of the processes arises from the air (cathode) vs. fuel
(anode) electrode, the cathode in the LCFCr cell was fed with either air or
pure oxygen.
From FIG. 14A, it is clear that the high frequency (100 Hz) arc can be
attributed to the
cathode, since in pure 02, the high frequency resistance (RHO decreased from
0.37 to
0.28 11-cm2, while the low frequency resistance (RLF) remained unchanged, as
shown in
Table 3-1. The resistance values were obtained by fitting the Nyquist plot in
FIG. 14A
to the Rs (RHF/CPEHF)(RLF/CPELF ) equivalent circuit model (FIG. 14B). Rs is
the series
resistance, Rp is the polarization resistance (the sum of all of the parallel
resistances),
and (Rxr/CPEr-i() and (RLF/CPELF) are the time constants at high (100 Hz) and
low (0.5
Hz) frequencies, respectively. Therefore, the low frequency arc (RLF/CPEL() is
predominantly due to contributions from the fuel electrode (anode) while the
high
frequency arc (RHF/CPEFiF) arises from the air electrode (cathode), consistent
with what
37
CA 02893153 2015-05-28
has been reported in the literature for other mixed conducting perovskite
systems,
including LSFCr (10).
Table 3-1
Circuit element values* obtained by fitting the results in Figure 2" to the
Rs(RaF CPEhT)(12.LF.CPELF )
equivalent circuit model
Cathode gas Rs ) REF (aCMI ) RLF (11=CM: ) Rp' (acm: )
Air 1.07 0,37 0.68 1.05
02 1.06 0.28 0.63 0.91
and Ru.- obtained from the lngh (ca. 100 Hz) and low (ca. 1 Hz) frequency
arcs. respectively.
Svnunetrical fuel cell based on LCFCr electrodes, operated at SOO C in wet
30% H2:192 gas mixtures at the fuel
electrode and with air or 0, exposure at the 02 electrode.
Rp = RHF RLF
DC experiments were also carried out to evaluate the performance of
Lao.3Mo.7Feo.7Cro.303-6(M = Sr, Ca). FIG. 15 shows the performance plots of
LCFCr and
LSFCr cells operated on 30% humidified H2 at 800 C. LCFCr shows a maximum
current and power density of 270 mA/cm2and 142 mW/cm2, respectively, while the
analogous values for LSFCr are 255 mA/cm2and 134 mW/cm2, consistent with the
EIS
data in FIG. 13. The performance of our Lao.3Mo3Feo.7Cro.303-6(M = Sr, Ca)
based cells
is comparable to that of other symmetrical cells (based on perovskite
electrodes)
reported in the literature, such as La4Sr8Ti12-xFex038-ti (LSTF), which showed
power
densities of 90-100 mW/cm2at 950 C in humidified H2 (1 6).
2.2 Performance of LCFCr in other fuel mixtures at 800 C
The performance of the LCFCr electrode was also examined in CO and syngas
(H2+CO) atmospheres. FIG. 16 shows that the polarization resistance of the
cell is 0.95
fl=cm2 in H2, which is slightly smaller than both the ; values obtained in CO
(1.11
n.cm2) and CO+H2(1.00 tI=cm2) atmospheres. This indicates that the material is
only a
somewhat better catalyst for H2 oxidation than CO oxidation. This shows that
LCFCr is
a very promising SOFC anode material that can be employed in a range of fuels,
giving
a very good performance in all case. Also, from the Nyquist plot, it is seen
that it is the
low frequency arc (RLF/CPELF) that is changing with changing fuel
environments, again
confirming that the low frequency arc is associated primarily with the anode,
as
suggested earlier on. The performance plot of the cell in these three gases is
shown in
FIG. 17. The OCP in the three gases is seen to be ca. 1.06 V, which is very
close to the
theoretical value, indicating that the anode and cathode compartments are well
sealed
and that there is no gas leakage. As stated earlier, there is not much
difference in the
activity of the LCFCr material in H2, CO and H2+CO atmospheres, and this is
supported
here by the dc measurements. The maximum power density obtained is between 140
to
38
150 mW/cm2 in all of the environments, while the maximum current density is in
the
range of 250-270 mA/cm2, all at 800 C.
2-4 Effect of low ppm H2S on performance of LCFCr in 30%H2 (bat wet N2) at
800 C
FIG. 18A shows that, when 9 ppm H2S is added to 30% Hz under OCP
conditions, the polarization resistance decreased slightly, from 1.00 to 0.96
SIcnr,
translating to a ca. 4 % decrease in Rp in the presence of H2S. No
poisoning/deactivation
of the LCFCr at 800 C is seen. This was also seen for LSFCr, although only
the results
for LCFCr are shown here. In comparison, most of the sulfur-induced
performance
enhancement behavior reported for other types of perovskites has usually been
observed
at much higher concentrations of H2S (1-5%). To better understand these
results, the
effect of ac polarization on the cell was investigated.
FIGs. 18B and 18C show the polarized EIS results for the LCFCr-based cell in
Hz, with or without the addition of 9 ppm H2S at 800 C. When the cell was
polarized at
-100 mV vs. the full cell open circuit voltage, i.e., at a cell voltage of ca.
0.95 V, (Figure
18B), Rp decreased from 0.97 S2.cm2 in Hz to 0.90 Q.cm2 in the presence of
H2S, while
when the anode was polarized at -300 mV (ca. 0.75 V cell voltage), Rp
decreased from
0.80 to 0.72. S2.cm2 (Figure 18C). The plot in the inset of Figure 18D shows
the % Rp
change vs. the applied voltage, calculated based on the Rp data in FIG. 18D
(%Rp
change = I(RpHz-Rptizs)J x 100/ (Rpm)). It can be seen that Rp decreased by 4
% at the
cell OCP and by 7 % and 11 % in the presence of H2S when the cell was
polarized at -
100 and -300 mV vs. the full cell open circuit voltage, respectively. This
indicates
that the enhancement of the performance of LCFCr in Hz in the presence of 112S
improves with polarization at 800 C.
As stated earlier, some ferrite-based perovskites, such as
Sm0.95Ceo.o5Feo.97Nio.o303-6(12) and Y0.9 Sro.1Cro.9Feo.103-6(13), have been
reported to
show enhanced Hz oxidation activity or electrochemical oxidation of H2S only
in high
concentrations of H2S (1-5%), due to the formation of sulfide species (e.g.,
FeS) at 600 -
800 C. On the other hand, probably under our testing conditions, some type of
adsorbed surface sulfide species (possibly FeS) is being formed even when H2S
is
present at ppm levels at 800 C. However, detailed surface characterization
studies are required to confirm the presence of FeS in our experiments.
To further study the performance of the LCFCr electrode towards Hz oxidation
in the presence or absence of 9 ppm H2S, potentiostatic studies were also
carried out at
39
Date Recue/Date Received 2021-09-30
CA 02893153 2015-05-28
the -100 and -300 mV vs. OCP cell polarization, respectively. Figure 19A shows
the
results of polarization at -100 mV vs. the full cell voltage at open circuit
for 9 h with
and without H2S. As can be seen, upon the addition of 9 ppm H2S to the Hz
fuel, the
current density increased from about 95 mA/cmzto 98 mA/cm2, giving a 2.1 %
improvement in performance. After 4 h of removal of H2S, the current density
decreased to about 95 mA/cm2, similar to the value observed before H2S
exposure. This
shows that the enhancement is only observed in the presence of H2S and that
the cell
fully recovers in the absence of H2S, suggesting that no bulk sulfide phase
forms but
rather only a surface species is generated. This behaviour is also seen at -
300 mV vs. the
full cell voltage at open circuit (FIG. 19B), where the cell improved by 4.3 %
in the
presence of 112S and fully recovered when the H2S was removed. Although the
present
work is in a preliminary stage, a range of surface characterization methods,
such as XPS
and AES, are currently being employed to determine what surface species are
being
formed, as well as to establish the effect of temperature on these materials
in the
presence of H2S.
3. Conclusions
This application has focused on the development of mixed conducting perovskite
oxides
for use at both electrodes in reversible solid oxide fuel cells (RSOFCs). In
this
example, the performance of LCFCr, in comparison with LSFCr, has been
investigated
in a range of fuel environments, with and without ppm levels of H2S, all at
800 C. The
symmetrical full cells were constructed by screen-printing Lao.3MoiFeo.7Cro
303-6 (M =
Sr, Ca) on a Yttria-stabilized zirconia (YSZ) electrolyte covered by a thin
Samaria-
doped ceria (SDC) buffer layer, and then tested using both impedance and
potentiostatic
techniques. It was found that the LCFCr is an equally good fuel electrode
(anode) and
02 electrode (cathode) as LSFCr, exhibiting very good electrochemical
performance in
Hz, CO and syngas (Hz+CO) atmospheres, giving polarization resistance of 0.95
acm2
in wet 30% H2and 1.00 11.crnz and 1.11 II cmzin wet 15% H2 :15% CO and 30% CO
atmospheres, respectively. The maximum power density obtained using these
gases
were between 140 to 150 mW/cm2, while the maximum current density was in the
range
of 250-270 mA/cm2. The LCFCr and LSFCr anodes were also evaluated in the
presence
of 9 ppm H2S, showing a small, but reproducible and reversible, decrease in
polarization resistance (Re). Chronoamperometric studies at cell polarizations
of -100
and -300 mV vs. the full cell voltage atopen circuit showed a ca. 2-4 %
increase in
CA 02893153 2015-05-28
current density in the presence of 9 ppm H2S + 30 % H2, with the cell
recovering fully
when H2S was removed. This is very promising and may indicate that some type
of
adsorbed surface sulfide species (possibly FeS) is being formed in the
presence of low
ppm 112S at 800 "C, leading to the observe enhancement in hydrogen oxidation
activity.
References for Example 3:
3-1. M. Asif and T. Muneer, Renew. Sust. Energ. Rev., 11, 1388 (2007).
3-2. Z. Cheng, S. Zha and M. Liu, J. Power Sources, 172, 688 (2007).
3-3. L. Deleebeeck, M. Shishkin, P. Addo, S. Paulson, H. Molero, T. Ziegler
and V.
Birss, PCCP, 16, 9383 (2014).
3-4. J. B. Hansen, Electrochem. Solid-State Lett., 11, B178 (2008).
3-5. S. J. Xia and V. I. Birss, in Proceedings -Electrochem. Soc., p. 1275
(2005).
3-6. S. Tao and J. T. S. Irvine, Nat Mater, 2, 320 (2003).
3-7. S. Zha, P. Tsang, Z. Cheng and M. Liu, J Solid State Chem., 178, 1844
(2005).
3-8. R. Mukundan, E. L. Brosha and F. H. Garzon, Electrochent and Solid-State
Lett.,
7, AS (2004).
3-9. J. M. Haag, D. M. Bierschenk, S. A. Barnett and K. R. Poeppelmeier, Solid
State
Ionics, 212, 1(2012).
3-10. M. Chen, S. Paulson, V. Thangadurai and V. Birss, J.Power Sources, 236,
68
(2013).
3-11. L. Aguilar, S. Zha, S. Li, J. Winnick and M. Liu, Electrochem. and Solid-
State
Lett., 7, A324 (2004).
3-12. S. M. Bukhari, W. D. Penwell and J. B. Giorgi, ECS Trans., 57, 1507
(2013).
3-13. Y.-F. Bu, Q. Zhong, D.-D. Xu, X.-L. Zhao and W.-Y. Tan, J. Power
Sources, 250,
143(2014).
3-14. P. Addo, B. Molero-Sanchez, M. Chen, S. Paulson and V. Birss, in 11th
European
SOFC and SOE forum, p. B0314, Luzerne, Switzerland (2014).
3-15. B. Molero-Sanchez, J. Prado-Gonjal, D. Avila-Brande, M. Chen, E. Moran
and V.
Birss, mt. J. Hydrogen Energy, 40, 1902 (2015).
3-16. J. Canales-Vazquez, J. C. Ruiz-Morales, D. Marrero-Lopez, J. Perla-
Martinez, P.
Nunez and P. Gomez-Romero, J. Power Sources, 171, 552 (2007).
41
