Full text of DOI:10.1002/ejic.201000504

FULL PAPER
DOI: 10.1002/ejic.201000504
Thermodynamic Properties and Stability of In-Doped SrCeO
3
Proton-Conducting Ceramics
[a]
[b]
[c]
Nata I. Matskevich,* Thomas Wolf, and Tatiana I. Chupakhina
Keywords: Thermodynamics / Proton-conducting ceramics / Strontium / Cerium / Solution calorimetry / Solid-phase synthesis
The standard enthalpy of formation of SrCe0.5In0.5
been determined by solution calorimetry by combining
the solution enthalpies of SrCe0.5In0.5 2.75 and a SrCl
.5CeCl + 0.5InCl mixture in 1.0 HCl with KI as a solvent
at 298.15 K and literature data. The results show that SrCe0.5
3–δ is thermodynamically stable with respect to decom-
O
2.75 has
position to binary oxides. Calculation of the Gibbs energies
for the interaction of SrCe0.5In0.5 2.75 with aluminum oxide
and zirconium dioxide allows one to predict that SrCe0.5
2.75 does not react with Al or ZrO at ambient tem-
peratures.
O
O
2
+
-
0
3
3
M
In0.5
O
2
O
3
2
-
In0.5
O
Introduction
thermodynamic properties connected with structural pa-
rameters can be used to understand the correlation of struc-
ture and stability in solid solutions A(II)Ce1–xB(III) O
Studies of high-temperature proton conductors (HTPC)
have intensified in the last ten years,[1–12] because these ma-
terials have wide enough technological application as mate-
rials in fuel cells, sensors, electrocatalysis, hydrogen separa-
tion membranes, and so on. Conversion of hydrogen using
a solid oxide fuel cell with a proton-conducting electrolyte
occurs at an intermediate temperature (700–1100 K) with-
out expensive electrocatalysts or the necessity to recirculate
the fuel. The advancement of hydrogen separation technol-
ogy has become important in energy applications for the
production of petrochemicals and pure hydrogen, the latter
being of particular interest for use in fuel cells. Accordingly,
high-temperature proton-conducting oxides have received
increasing attention for these applications. The most well-
known HTPC materials are the perovskite oxides (ABO3)
with large basic A cations (e.g., Ba, Sr) and tetravalent B
cations (e.g., Zr, Ce). A lot of research has been devoted
to the study of ionic conductivity. However, for practical
applications, the study of other properties is required. For
example, one of the important properties is stability and,
in particular, thermodynamic stability. As has been already
found, the thermodynamic stability can have an influence
on the mechanical stability of the microstructure. However,
there has not been enough research devoted to studying of
thermodynamic stability of doped cerates. Furthermore, the
x
3–δ
[A(II) – metal of the second group, B(III) – metal of the
third group].
This work forms part of the investigations performed by
authors on barium and strontium cerates and is devoted to
a thermodynamic study of SrIn0.5Ce0.5O3–d. In an earlier
[5–6]
paper,
we reported studies on barium cerates doped with
rare-earth elements. However, as we reported, the dopant
solubility limit is less than 20% of the available B sites
(
ABO structure), and therefore, the possibility of introduc-
3
ing protonic defects into the matrix is also limited. Conse-
quently, it is interesting to study other dopants of the B
sites of ABO with the aim of investigating the possibility
3
of increasing the dopant limits and to find ionic conductors
with ionic conductivities that change over wide limits. For
this reason, we decided to investigate the compounds
SrIn0.5Ce0.5O3–d and their thermodynamic stability towards
both binary oxides and external reagents such as Al O and
2
3
ZrO2.
Results and Discussion
To determine the thermodynamic data for the SrCe0.5
-
In0.5O3–d we used the thermochemical cycle, which is similar
[
6]
to one applied in ref. for barium cerates doped with rare-
earth elements. The principal scheme is based on the dissol-
[
a] Nikolaev Institute of Inorganic Chemistry, Siberian Branch of
the Russian Academy of Sciences
Lavrentiev Prospect 3, Novosibirsk 630090, Russian Federation ution of strontium cerate doped with indium as well as mix-
Fax: +7-383-330-6449
tures of strontium, cerium, and indium chlorides (SrCl +
2
E-mail: nata@niic.nsc.ru
[
[
b] Karlsruhe Research Center GmbH, Institute of Technology
Hermann-von-Helmholtz-Platz 1, 76334 Karlsruhe, Germany
c] Institute of Solid State Chemistry, Ural Branch of the Russian
Academy of Sciences
0.5CeCl + 0.5InCl ) in hydrochloric acid [HCl (sol)] with
3 3
4+
3+
KI. The KI was added to convert Ce to Ce . In order to
realize this cycle, we needed the following samples for the
investigation: SrCe0.5In0.5O3–d, SrCl , CeCl , and InCl .
Pervomaiskaya Street 91, Ekaterinburg, Russia
2
3
3
150
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 150–153

3
In-Doped SrCeO Proton-Conducting Ceramics
Previously, we used the expensive technique from the record the temperature of the vessel, calibrate the instru-
Karlsruhe Research Center that enables one to prepare dif- ment with precise injections of electrical energy and calcu-
ferent complex oxides of high quality.[
2,6]
late calorimeter constants and enthalpies. The calorimetric
The phase purity and identity of SrCe0.5In0.5O2.75 were vessel was maintained at 298.15 K with a temperature drift
confirmed by X-ray powder diffraction. We also performed of less than 0.0003 K over 10 h. Dissolution of potassium
analysis of samples with a sequential X-ray fluorescence chloride in water was performed to check the precision of
spectrometer.
the calorimeter. The obtained dissolution heat of KCl
–1
The sample SrCe0.5In0.5
O
¯
was shown to have a cubic was 17.41Ϯ0.04 kJmol (the molality of the final solution
2.75
[13]
–1
structure (space group Pm3m). All compounds were also was
0.028 molkg ).
The
certified
values
are
characterized by chemical analysis,[
14–15]
and analytical re- 17.42Ϯ0.02 kJmol in ref.[17] and 17.47Ϯ0.07 kJmol in
–1
–1
[
18]
sults are presented in Table 1. Analyses indicated that impu- ref.
rities of Ho, Dy, Yb, La, Tm, Er, Pr, Sm, Te, Ca, Mg, Mn,
A solution of 1 m HCl with 0.1 m KI was chosen as a
–3
Pb, and Ag were present at levels of between 10 and solvent. A mixture of SrCl , CeCl , and InCl was prepared
2
3
3
–
4
10
at.-%. The oxygen contents were determined by iodo- in a ratio of 1:0.5:0.5. This solvent was also used by the
[
19]
metric titration with 0.01 m Na S O ·5H O according to the authors
to determine the enthalpies of formation of
2
2
3
2
method described in ref.[ According to the results of the BaCeO (s) and SrCeO (s). The reactions showing the ther-
16]
3
3
chemical analysis, the sample had the composition SrCe0.5
In0.5O2.75
-
mochemical cycle from which the standard enthalpy of for-
mation of SrCe0.5In0.5O2.75 was calculated are presented in
Table 2.
.
Table 1. Analytical results.
Compound Found (%)
CeCl
The molar concentration of CeCl in our experiments
3
[
19]
Calculated (%)
was the same as that in ref.
We used some data from
Ce, 56.81Ϯ0.04
In, 51.94Ϯ0.03
Sr, 55.23Ϯ0.04
Sr, 33.80Ϯ0.04
In, 22.19Ϯ0.03
Ce, 27.01Ϯ0.03
O, 16.99Ϯ0.02
Ce, 56.85
In, 51.91
Sr, 55.27
Sr, 33.82
In, 22.16
Ce, 27.04
O, 16.98
this paper for our calculation, for example, the enthalpies
3
InCl
SrCl
3
of solution of KI and I , which were reported with high
2
[19]
2
precision.
We measured only dissolution enthalpies of
SrCe0.5In0.5O
2.75
SrCe0.5In0.5O2.75 and the mixture SrCl2 + 0.5CeCl3
+
0
.5InCl . The cycle was completed with the auxiliary values
3
for the molar enthalpies of formation of H O(aq.),
2
HCl(aq.), SrCl (s), CeCl (s), InCl (s), KCl(s), and KI(s).
2
3
3
All calorimetric experiments were performed in an auto-
All manipulations with CeCl
3
, SrCl , and InCl were per-
2 3
matic calorimeter with an isothermal jacket. The calorime- formed in a dry box (pure Ar gas) to prevent interaction
ter consists of a Dewar vessel with a brass cover (V = with moisture or CO . The mass of SrCe0.5In0.5
00 mL). The platinum resistance thermometer, calibration for the experiments was about 0.1 g.
O2.75 used
2
2
heater, cooler, mixer, and the device to break the ampoules
The measured enthalpies of solution of SrCe0.5In0.5O2.75
were mounted on the lid of the Dewar vessel. Detailed de- and (SrCl2 + 0.5CeCl3 + 0.5InCl ) were determined as
3
0
Ϫ1
scriptions of the construction and the procedures for per- ΔsolH (298.15 K) = –216.5Ϯ2.8 kJmol
forming experiments and checking the calorimeter are pre- ΔsolH (298.15 K) = –158.87Ϯ0.68 kJmol (n = 6) (n =
(n = 6) and
1
0
Ϫ1
2
[
6,14–15]
sented in refs.
The computer controller for the plati- number of parallel calorimetric experiments). Errors were
num thermometer and the Matlab program were written at calculated for the 95% confidence interval by using stan-
our laboratory. The program allows one to measure and dard procedures for treatment of experimental results.[
20]
Table 2. Thermochemical cycle for the determination of the enthalpy of formation of SrCe0.5In0.5O
2.75.[a]
[a] In the table, “solution 1” refers to a 1 m solution of HCl in 0.1 m KI.
Eur. J. Inorg. Chem. 2011, 150–153
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
151

N. I. Matskevich, T. Wolf, T. I. Chupakhina
FULL PAPER
The enthalpies of dissolution measured for reactions 1
It is interesting to mention that the enthalpy of forma-
and 2 and data on reactions 3Ϫ12 (see Table 2) were used tion of complex oxides from binary oxides is usually close
for calculating the enthalpy of the reaction Sr(s) + 0.5Ce(s) to zero [for example, Δ H(298.15 K, CdSiO3)
=
ox
o
Ϫ1
+
0.5In(s) + 1.375O (g) Ǟ SrCe0.5In0.5O2.75(s) + Δ H as –6.28 kJmol ]. In our case, we have values around
2 r 13
follows: Δ H = –ΔsolH1 + ΔsolH2 + ΔsolH3 – ΔsolH4o
o
o
o
o
–
–150 kJmol , which is higher than the usual values for
complex oxides. Hence, this compound is thermodynami-
cally stable enough.
Ϫ1
f
13
ΔsolH5 + ΔsolH6o + ΔsolH7o + ΔsolH8o – ΔsolH9 + ΔsolH10
o
o
o
o
o
+
ΔsolH11 + ΔsolH12
.
o
o
Here, Δ H
= Δ H (SrCe0.5In0.5O3–δ, s, 298.15 K) =
As some researchers have pointed out, a concern for bar-
1509.4Ϯ3.4 kJmol is the standard enthalpy of forma- ium or strontium cerates is their stability in the external
f
13
f
Ϫ1
–
tion of strontium cerate doped with indium oxide. atmosphere. When pellets of SrCeO were sintered in an
3
To calculate this value, we used our experimental results alumina or zirconia container, it is possible to assume that
and literature data for the enthalpies of formation of dif- SrCeO3 can react with Al O3 or ZrO2 and decompose
2
ferent compounds and processes taken from refs.[
presented in Table 2.
17,19]
and through the following reactions: SrCeO3 + Al O3 Ǟ
2
SrAl O + CeO and SrCeO + ZrO Ǟ SrZrO + CeO .
2
4
2
3
2
3
2
Experimental data have also been used to study the
In relation to this, it was interesting to understand
thermodynamic stabilities of the investigated compounds whether SrCe0.5In0.5O2.75 reacted with Al O (s) or ZrO (s).
2
3
2
with the respect to mixtures of the same nominal composi- For this reason we considered the following reactions:
tion. Enthalpies of formation of SrO, CeO , and In O
2
2
3
taken from the literature[ were used to calculate the en-
thalpies of formation of SrCe0.5In0.5O2.75 from binary ox-
ides as follows:
17]
The possibility of interaction was estimated by using our
experimental data, formation enthalpies, and entropies of
[17]
the phases CeO , In O , SrAl O , SrZrO taken from ref.
2
2
3
2
4
3
It is necessary to note that the direction of any chemical
transformation is determined by the sign of the Gibbs en-
In order to understand whether SrCe0.5In0.5O2.75 is stable
or unstable with respect to decomposition to a SrO(s) +
o
o
o
ergy ΔG = ΔH – TΔS . Below we show the calculated
Gibbs energies for reactions 15 and 16. As was mentioned
earlier, the value of the entropy of SrCe0.5In0.5O2.75, which
is absent in the literature, was estimated by using the en-
0
.5CeO (s) + 0.25In O (s) mixture, it is necessary to know
2 2 3
the Gibbs energy (ΔG = ΔH – TΔS) of reaction 14. There
is no entropy value for the SrCe0.5In0.5O2.75 phase in the
literature. This value was estimated by using entropies of
[17,19]
tropies of In O , SrCeO , and CeO taken from refs.
2
3
3
2
The Gibbs energies for processes 15 and 16 were calculated
[17,19]
CeO , In O , and SrCeO taken from refs.
The method
o
Ϫ1
o
2
2
3
3
as Δ G (298.15 K) = +85.7Ϯ8.1 kJmol
(
and Δ G
r
r
of estimation of the entropies of complex oxides as a sum
of the binary or triple oxides is frequently used and gives
Ϫ1
298.15 K) = +69.4Ϯ6.4 kJmol , respectively.
Taking into account the Gibbs energies of reactions 15
good results. For example, the entropies of the YBa2-
and 16, it is possible to predict that SrCe0.5In0.5O2.75 does
not react with Al O and ZrO at ambient temperatures.
Cu O phase estimated as the sum of the entropies of 2Ba-
3
6.5
2
3
2
Ϫ1
CuO , CuO, and 0.5Y O is 321.8 J(Kmol) , whereas the
measured entropy is 321.7 J(Kmol)
The Gibbs energy for process 14 was estimated as Δ G
298.15 K) = –146.6Ϯ4.3 kJmol by using the formation
[6]
2
2
3
This behavior was not expected, since according to ref. ,
complex oxides BaCe0.8Nd0.2O2.9 and BaCe0.8Lu0.2O2.9 re-
act with Al O and ZrO at ambient temperatures. The pre-
Ϫ1
.
0
ox
2
3
2
Ϫ1
(
sented results indicate that the compound SrCe0.5In0.5O2.75
is a prospective material for application in fuel cells, sensors,
electrocatalysis, hydrogen separation membranes, and so
on.
enthalpy of reaction 14, the entropies of SrO, CeO , In O ,
2
2
3
as well as the estimated entropy of SrCe0.5In0.5O2.75
.
As can be seen, strontium cerate doped with indium is
thermodynamically stable with respect to its decomposition
into binary oxides at room temperature. This is not an obvi-
ous result for this class of compounds, because there is a
discussion about the thermodynamic stability of
Conclusions
[
19,21,22]
SrCeO3.
it was proved and reported in a paper that a single-phase
Here it is necessary to note the following:
[23]
In this paper we report values for the standard formation
material in Sr1+xCeO3+δ solid solutions was obtained for enthalpies of SrCe0.5In0.5O2.75 measured by solution calo-
compositions x = 0.02–0.03. Therefore, it is possible to as- rimetry in a solution of 1 m HCl in 0.1 m KI. We deter-
[
19,21,22]
sume that authors
who obtained different data on mined the thermodynamic stability of In-doped strontium
the formation enthalpies and thermodynamic stability of cerate with respect to mixtures of binary oxides, the en-
strontium cerate performed measurements for different thalpies, and Gibbs energies of interaction with alumina
compositions. It is not possible to clear the discrepancy, be- and zirconia. On the basis of these data, we established that
cause the full chemical analysis of SrCeO is not presented the above-mentioned oxide is thermodynamically stable
3
[
19,21,22]
in the literature.
with respect to decomposition into binary oxides at room
152
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 150–153

3
In-Doped SrCeO Proton-Conducting Ceramics
temperature. It is also possible to predict that SrCe0.5
-
gram of Fundamental Investigation of the Siberian Branch of the
Russian Academy of Sciences.
In0.5O2.75 does not react with Al O and ZrO at room tem-
2
3
2
perature.
[
1] N. Fernandez, M. F. Garcia, Eur. J. Inorg. Chem. 2003, 13,
2458–2462.
Experimental Section
[2] Th. Zeiske, R. Sonntag, D. Hohlwein, N. H. Andersen, Th.
Wolf, Nature 1991, 353, 542–544.
3] A. I. Ruiz, M. L. Lopez, M. L. Veiga, C. Pico, Eur. J. Inorg.
The phase of composition SrCe0.5In0.5
state synthesis from SrCO , In
formed according to the reaction: SrCO
Ǟ CO 2.75. The reagents SrCO
CERAC, TM incorporated), CeO (99.99%, Vetron GmbH), and
In (99.99%, Reacton, A Johnson Matthey Company) were used
for synthesis of the complex oxides SrCe0.5In0.5 3–d. Stoichiometric
amounts of SrCO , CeO , and In were mixed by ball milling in
an agate container with agate balls using a planetary mill
FRITSCH pulverisette) over 72 h. The ground materials were pal-
O
3–d was prepared by solid-
[
3
2
O
3
, and CeO
2
. Synthesis was per-
+ 0.25In
(99.999%,
Chem. 2000, 4, 659–664.
[4] T. Norby, M. Wideroe, R. Glockner, Y. Larring, Dalton Trans.
2004, 19, 3012–3018.
3
+ 0.5CeO
2
2
O
3
2
+ SrCe0.5In0.5
O
3
[
[
5] B. C. H. Steele, A. Heinzel, Nature 2001, 414, 345–352.
6] N. I. Matskevich, Th. Wolf, M. Yu. Matskevich, T. I. Chupakh-
ina, Eur. J. Inorg. Chem. 2009, 11, 1477–1482.
2
2
O
3
O
[
7] G. C. Mather, M. S. Islam, J. Mater. Chem. 2005, 17, 1736–
3
2
2 3
O
1744.
[
8] A. Selmi, G. Corbel, S. Kojikian, V. Voronkova, E. Khari-
(
tonova, P. Lacorre, Eur. J. Inorg. Chem. 2008, 11, 1813–1821.
letized by using a 10 mm diameter die and fired at 1300 K for 70 h,
at 1400 K for 10 h, or at 1700 K for 24 h using a CARBOLITE
furnace. Detailed procedures of the preparation of analogous com-
[9] K. D. Kreuer, S. J. Paddison, E. Spohr, M. Schusher, Chem.
Rev. 2004, 104, 4637–4678.
[10] X. Weng, P. Boldrin, I. Abrahams, S. J. Skinner, J. A. Darr,
Chem. Mater. 2007, 19, 4382–4384.
pounds were described in our earlier papers.[
2,6]
[
[
11] Z. Shao, S. M. Haile, Nature 2004, 43, 170–173.
12] G. C. Mather, F. M. Figueiredo, J. R. Paz, S. Garcia-Martin,
Inorg. Chem. 2008, 47, 921–929.
Anhydrous SrCl
2
was prepared by drying SrCl
2
(CERAC, TM in-
was also
corporated, USA, 99.9%) in argon at about 500 K. CeCl
3
purchased from CERAC (mass fraction is more then 0.999) and
purified by vacuum sublimation in order to remove the lanthanide
[13] Th. Wolf, N. I. Matskevich, C. Greaves, First Russian-German
Seminar “Thermodynamics and Materials Science” Proceedings,
2
008, 23–25 September, Novosibirsk, Russia, pp. 101–102.
[14] N. I. Matskevich, Th. Wolf, Yu. I. Pochivalov, Inorg. Chem.
008, 47, 2581–2584.
15] N. I. Matskevich, C. Greaves, Thermochim. Acta 2007, 459,
25–126.
oxychloride impurities. For this purpose, CeCl
the melting temperature (1143 K) in a vacuum better than 10 Pa.
InCl was synthesized from Cl and In. Chlorine gas was passed
over indium at a temperature of about 450 K. All manipulations
3
was sublimed above
–
5
2
3
2
[
[
[
1
with CeCl
Ar gas).
3 2 3
, BaCl , and InCl were performed in a dry box (pure
16] A. I. Nazzal, V. Y. Lee, E. M. Engler, R. D. Jacowitz, Y. Tok-
ura, J. B. Torrance, Phys. C 1988, 153–155, 1367–1368.
17] V. P. Glushko, Termicheskie Konstanty Veshchestv (Thermal
Constants of Substances), VINITI, Moscow, 1965–1982, issues
X-ray powder diffraction was performed with a STADI-P Stoe dif-
fractometer (Germany, Cu-Kα1 radiation). An ARL ADVANTЈXP
sequential X-ray fluorescence spectrometer was also used to ana-
lyze the samples.
1–10.
[
18] C. Gunter, R. Pfestorf, M. Rother, J. Seidel, R. Zimmermann,
G. Wolf, V. Schroder, J. Therm. Anal. Calorim. 1988, 33, 359–
363.
For the analysis of In and Ce, a spectrophotometric method (spec-
trophotometer SF–46) was used. Sr was determined by flame pho-
tometry (air/acetylene, Hitachi Z-8000). The content of impurities
was determined by spectroscopic methods (mass spectrometer “Ele-
ment”, Finnigan Mat, Germany).
[
[
[
[
[
19] E. H. P. Cordfunke, A. S. Booij, M. E. Huntelaar, J. Chem.
Thermodyn. 1998, 30, 437–447.
20] B. N. Oleinik, Precision Calorimetry, Publishing House of Stan-
dards, Moscow. 1973, p. 207.
21] C. W. Tanner, A. V. Virkar, J. Electrochem. Soc. 1996, 143,
1386–1389.
22] J. Goudiakas, R. G. Haire, J. Fuger, J. Chem. Thermodyn. 1990,
Acknowledgments
22, 577–587.
23] G. C. Mather, F. M. Figuiredo, J. R. Paz, S. G. Martin, Inorg.
Chem. 2008, 47, 921–929.
This work was supported by the Karlsruhe Research Center, the
Deutsche Forschungsgemeinschaft (DFG) (grant LO 250/24-1), a
NATO programme (grant CBR.NR.NRCLG 982559), and the Pro-
Received: May 5, 2010
Published Online: December 1, 2010
Eur. J. Inorg. Chem. 2011, 150–153
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
153