Journal of Power Sources 293 (2015) 806e814

Contents lists available at ScienceDirect

Journal of Power Sources

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

Synthesis and characterization of calcium and iron co-doped

lanthanum silicate oxyapatites by solegel process for solid oxide fuel
cells
Xiao Guo Cao a, *, San Ping Jiang b, *, Yun Yong Li a
a
b

School of Materials and Energy, Guangdong University of Technology, Guangzhou 510006, Guangdong, PR China
Fuels and Energy Technology Institute & Department of Chemical Engineering, Curtin University, Perth, WA 6102, Australia

h i g h l i g h t s

Oxyapatite co-doped with Ca and Fe is synthesized by a solegel process (SGP).

The ion conductivity of LCSFO(SGP) is signicantly higher than that of LCSFO(SSP).

Ca and Fe Co-doping for oxyapatite could improve the sinterability and ion conductivity.

a r t i c l e i n f o

a b s t r a c t

Article history:
Received 1 April 2015
Received in revised form
26 May 2015
Accepted 2 June 2015
Available online 11 June 2015

Lanthanum silicate oxyapatites with and without calcium (Ca) and iron (Fe) doping, La10Si5FeO26.5 (LSFO)
and La9.5Ca0.5Si5.5Fe0.5O26.5 (LCSFO), are synthesized by solegel process (SGP) and solid state reaction
process (SSP). The phase formation, microstructure and conductivities of LSFO and LCSFO oxyapatites are
characterized by X-ray diffraction (XRD), scanning electron spectroscopy (SEM) and complex impedance
analysis. The morphologies of LCSFO oxyapatite nanoparticles synthesized by SGP were characterized by
transmission electron microscope (TEM). The thermal and decomposition properties of the LCSFO gel
were analyzed by simultaneous differential scanning calorimetry and thermal gravimetric analysis (DSC
eTGA). The results show that the phase formation of LCSFO synthesized by SGP occurs at temperatures as
low as 750  C, signicantly lower than ~1500  C required for LCSFO synthesized by SSP. Co-doping of Ca
and Fe signicantly improves the densication, sinterability and oxide-ion conductivity of lanthanum
silicate oxyapatites. The best results were obtained on LCSFO synthesized by SGP, achieving oxide-ion
conductivity of 2.08  102 S cm1 at 800  C, which is higher than 5.68  103 S cm1 and
1.04  102 S cm1 for LSFO and LCSFO synthesized by SSP, respectively, under the identical test
conditions.
2015 Elsevier B.V. All rights reserved.

Keywords:
Solid oxide fuel cells
Electrolyte
Lanthanum silicate oxyapatite
Sol-gel process
Oxide-ion conductivity

1. Introduction
Oxide-ion conductors are important in their application to
sensors, solid oxide fuel cells (SOFCs) and oxygen pumps. A number
of oxide-ion conductors, such as the doped lanthanum gallate,
doped ceria, yttria-stabilized zirconia (YSZ) have being developed
for applications as the electrolytes in SOFCs [1e4]. Apatite-type
oxides with a general formula RE9.33xSi6O263x/2 (RE rareearth elements) are attracting considerable attention as a new class

* Corresponding authors.
E-mail addresses: xgcao@gdut.edu.cn
(S.P. Jiang).

(X.G.

http://dx.doi.org/10.1016/j.jpowsour.2015.06.008
0378-7753/ 2015 Elsevier B.V. All rights reserved.

Cao),

s.jiang@curtin.edu.au

of electrolyte materials for SOFCs [5e16]. The oxyapatite lattice

consists of covalent SiO4 tetrahedra and ionic-like RE/O channels
[12,17]. The RE-site cations occupy cavities created by SiO4 units
with four distinct oxygen positions, additional oxygen sites form
channels through the lattice. The open structure of oxyapatite
suggests that this material should be appropriate for the electrolyte
applications for intermediate temperature SOFCs [18,19]. Mineshige
et al. [16] studied the ionic and electronic conductivity of
lanthanum silicate (La9.333xSi6O261.5x with and without Al doping
and showed that the electronic conductivities are in the range of
105e104 S cm1 at temperatures of 800e1000  C, orders of
magnitudes lower than ~101 S cm1 for ionic conductivities. The
ionic transfer number is very close to one and independent of

X.G. Cao et al. / Journal of Power Sources 293 (2015) 806e814

oxygen partial pressure. The lanthanum silicate electrolyte has

basically ionic conducting characteristics regardless of Al doping.
This shows that lanthanum silicate apatite materials are pure oxygen ion conductors with negligible electronic conductivity,
showing the promising potential as alternative electrolytes for next
generation of SOFCs.
Direct evidences for the importance of the silicate units were
obtained from the solid state 29Si NMR studies, which shows a close
correlation between the silicon environment and the observed
oxide-ion conductivity [20]. The computer modeling studies also
indicated the importance of the silicate substructure in aiding the
motion of the oxide ions down the channels in the oxyapatites.
Detailed studies of the Fe-doped lanthanum silicate oxyapatite
systems, La10Si6xFexO27x/2, by Kharton et al. [21e23] showed that
the ionic conductivity increases with the Fe doping level, reaching a
maximum value for x 1.0. Further Fe incorporation resulted in a
decrease in the conductivity up to the fully stoichiometric
composition, La10Si4Fe2O26. La-site vacancies, particularly in positions enveloping oxygen channels, obviously affect the ionic
transport in the apatite lattice. The La-site deciency inuences the
SiO4 tetrahedra relaxation and may cause displacement of the anions from channels into new interstitial sites, thus creating vacancies at xed oxygen content [12,24e26]. Studies of doping
alkaline earth elements on La site show that the doped oxyapatites
containing oxygen excess, e.g. La9MSi6O26.5 (M Ca, Sr, Ba), exhibit
high oxide-ion conductivities [25e27]. The results of doping lower
valent ions on the La or Si sites as shown above indicate clearly that
the formation of either cation vacancies or oxygen excess can
improve the oxide-ion conductivity of the oxyapatites.
The ionic conductivity and microstructure of lanthanum silica
oxyapatites strongly depends on the synthesis methods and conditions. The traditional method to prepare the lanthanum silicate
oxyapatite is the high-temperature solid state reaction process
(SSP) [28]. However, lanthanum silica oxyapatites with a high purity level are very difcult to obtain, as the samples are often
contaminated with either La2SiO5 or La2Si2O7 secondary phases
[29e31]. In addition, oxyapatite powders synthesized by SSP have
to be calcined at high temperatures for a long period, which would
result in the signicant increase in the manufacturing cost of the
cells. To reduce the sintering and phase formation temperatures,
other synthesis routes have been developed to synthesize oxyapatite powders, such as freeze-drying [32], gel-casting technique
[33,34] and solegel process (SGP) [35,36]. Among them, SGP is a
low-cost, low temperature, versatile with few steps in the production process for the synthesis of lanthanum silicate oxyapatites
[37e39]. The SGP allows the synthesis of compounds in solution,
achieving homogeneity on the molecular scale in the nal product.
After calcining at 800e1000  C, oxyapatite nanoparticles with pure
apatite phase can be obtained.
In this paper, we employed SGP to synthesize the lanthanum
silicate oxyapatite co-doped with Ca and Fe, La9.5Ca0.5Si5.5Fe0.5O26.5.
As shown early [28], doping on the La site of aluminum lanthanum
oxyapatites improves the densication, sinterability and oxide-ion
conductivity. The effects of SGP and co-doping on the microstructure and oxide-ion conductivity of lanthanum silicate oxyapatites
were discussed.
2. Experimental
2.1. Sample preparation
Tetraethyl orthosilicate (TEOS), La(NO3)3$6H2O, Fe(NO3)3$9H2O,
Ca(NO3)3$4H2O and acetic acid were obtained from SigmaeAldrich
and used as received. Lanthanum silicate oxyapatite co-doped with
Ca and Fe, La9.5Ca0.5Si5.5Fe0.5O26.5 (LCSFO) was synthesized from

807

TEOS, La(NO3)3$6H2O, Fe(NO3)3$9H2O and Ca(NO3)3$4H2O, which

was weighed in appropriate ratio and dissolved in ethanol and
deionized water to form a mixture solution. The pH of the solution
was adjusted to 2 by adding slowly acetic acid. TEOS was then
slowly added to the solution at room temperature under stirred
condition. The stabilized sol was heated to 70  C, becoming
increasingly viscous due to polymerization reaction. Thereafter, the
viscous resin was dried at 100  C for 8 h, followed by calcination at
550  C for 4 h. The powders were then calcined at different temperatures (700, 750, 800, 900 and 1000  C) for 10 h in air. The ascalcined powders were dispersed in isopropanol and pulverized
in a ball-mill using an Y2O3-stabilized zirconia (YSZ) ball medium.
The Ca and Fe co-doped lanthanum silicate oxyapatites synthesized
by SGP were denoted as LCSFO(SGP).
La10Si5.5Fe0.5O26.5 (LSFO) and La9.5Ca0.5Si5.5Fe0.5O26.5 (LCSFO)
were also prepared from high purity La2O3, SiO2, Fe(NO3)3$9H2O
and CaCO3 (all from SigmaeAldrich) by solid state reaction process
(SSP). The raw materials were weighed in appropriate ratio and
mixed in plastic vessels for 24 h. The details of the procedures for
the synthesis of oxyapatite powders by SSP can be found in our
previous publications [28]. The obtained products were denoted as
LSFO(SSP) and LCSFO(SSP), respectively.
As-synthesized oxyapatite powders by SGP and SSP were
pressed uniaxially into pellets under a pressure of 150 MPa and
sintered at different temperature from 1500 to 1600  C for 4 h in air.
The sintered pellets were ~9 mm in diameter and 1.5 mm thick.
2.2. Characterizations
Phase formation of the oxyapatite powders synthesized by SGP
and SSP were determined by X-ray diffraction (XRD, Philips MPD
1880 e Cu Ka1). Simultaneous differential scanning calorimetry and
thermal gravimetric analysis (DSCeTGA) of the LCSFO(SGP) gel
were carried out using TA STD-2960 V3.0F at a scan rate of 10  C/
min in air. The morphologies of LCSFO oxyapatite nanoparticles
synthesized by SGP were characterized by TEM (JEM-2100, JEOL).
The microstructure of LSFO(SSP), LCSFO(SSP) and LCSFO(SGP) pellets were examined by scanning electron microscope (SEM, JSM5600/LV). The bulk densities of the sintered oxyapatite pelletss
were obtained from the mass and geometric dimensions of the
pellet samples. Silver paste (Ferro Corporation USA) was painted
onto both sides of the sintered pellets as the electrodes. Ionic
conductivity of the oxyapatites was obtained from the impedance
measurement, using a Solartron 1260 frequency response analyzer
in conjunction with a 1287 electrochemical interface. The impedance curves were obtained in the frequency range from 0.1 Hz to
10 M Hz in 50  C interval in air between 300 and 800  C. The
impedance data were analyzed by Zview software.
3. Results and discussion
3.1. Phase and microstructure of the oxyapatites
The phase formation of oxyapatite powders synthesized by SSP
and SGP was investigated by XRD. Fig. 1a and b shows the XRD
patterns of LSFO(SSP) and LCSFO(SSP) powders calcined at the
temperatures of 1000e1400  C, respectively. The main phase of the
powders is primarily apatite, however, a secondary phase, La2SiO5,
was also found. According to the La2O3/SiO2 phase diagram [40],
La2SiO5 and La2Si2O7 phases are more stable than apatite phase
below 1600  C. As shown previously [28,33,35], La2SiO5 is difcult
to remove once it is formed during the synthesis process. The
reason could be due to the difculties to obtain a homogeneous
mixture of the oxide precursors by the SSP route. With the increase
of the calcined temperature, the intensity of the La2SiO5 second

808

X.G. Cao et al. / Journal of Power Sources 293 (2015) 806e814

the La2SiO5 secondary phase. However, as shown previously, phase

pure lanthanum silicate apatite-type powders synthesized by SSP
can be formed at temperature above 1500  C [29,41]. Fig. 1c is the
XRD patterns of LCSFO powders synthesized by SGP at the temperatures of 700e1000  C. The oxyapatite phase starts to form at
700  C and the complete formation of apatite phase for LCSFO(SGP)
powders occurs at 750  C. This is substantially lower than complete
phase formation temperature of ~1500 observed for the lanthanum
silicates such as La10Si6O27 and La9Ba1Si6O26.5 synthesized by SSP
route [29,41]. La2SiO5 was not detected. These results indicate that
the co-doped oxyapatite powders can be successfully synthesized
by SGP with a substantially reduced phase formation temperature,
similar to the previous results on La10Si6O27 [35] and La9.33Si6O26
[36] powders. The low phase formation temperature is most likely
due to the better homogeneity of the precursor formed during the
gel formation step. The intensity of peaks of apatite phase increases
with the increase of calcined temperature, indicating the increased
crystallinity of LCSFO(SGP) oxyapatite phase.
The thermal decomposition and the phase formation behavior
of the LCSFO gel were analyzed by a DSC-TGA and the results are
shown in Fig. 2a. The TGA curve exhibits four distinct weight loss
steps. The rst 20.4% weight loss step from 25 to 150  C is due to the
loss of residual water and ethanol in the dried gel. As expected, the
intense exothermic peak was observed at 88  C in the DSC curve.
The second step with 31% weight loss in the temperature range of
150e430  C and the intense exothermic peak at 390  C correspond
to the evaporation and decomposition of organic components. The

Fig. 1. XRD patterns of (a) LSFO(SSP) and (b) LCSFO(SSP) powders calcined at 1000,
1100, 1200, 1300 and 1400  C for 10 h in air. XRD patterns of (c) LCSFO(SGP) powders
calcined at 700, 750, 800, 900 and 1000  C for 10 h in air.

phase decreased, indicating the reduction in the formation of the

secondary phase. The intensity of the La2SiO5 second phase of
LCSFO(SSP) is lower than that of LSFO(SSP) synthesized under
identical conditions and temperatures, indicating that doping Ca on
the La site for LSFO oxyapatite can further retard the formation of

Fig. 2. (a) DSC-TGA curves of the LCSFO(SGP) xerogel. (b) TEM image of LCSFO(SGP)
apatite nano-particles calcined at 900  C for 10 h in air.

X.G. Cao et al. / Journal of Power Sources 293 (2015) 806e814

third weight loss step from 450 to 550  C with the corresponding
intense exothermic peak at 499  C can be attributed to the
decomposition of the residue organics in ashes. A small weight loss
(~2.3%) observed from 550 to 750  C is most likely associated with
the formation of apatite phase. Fig. 2b shows the TEM image of
LCSFO(SGP) oxyapatite nanoparticles calcined at 900  C for 10 h in
air. The LCSFO(SGP) oxyapatite nanoparticles exhibite a welldispersed irregular shape with an average size of ~20 nm.
Fig. 3 shows the fractured cross-section SEM micrographs of
LSFO(SSP), LCSFO(SSP) and LCSFO(SGP) oxyapatite pellets sintered
at 1500  C for 4 h in air. The doped lanthanum silicate oxyapatite
samples are porous. However, LCSFO(SSP) oxyapatite pellets show a
much denser and compact structure. The porosity of LCSFO(SSP) is
signicantly smaller than that of LSFO(SSP), indicating that codoping of Ca and Fe signicantly benets the sintering and densication process. The LCSFO(SGP) shows a high porosity, as
compared to LCSFO(SSP) (Fig. 3b and c). In the case of LCSFO synthesized by SGP, LCSFO(SGP), the apatite phase occurs at low
temperature, 750  C (Fig. 1c) and also forms nanoparticles with size
of ~20 nm (Fig. 2b). We studied the effect of YSZ particle size on the
sintering and densication of YSZ thin lm electrolyte on anodesupported SOFCs and found that the nano-sized YSZ is difcult to
form a dense thin electrolyte lm on the Ni/YSZ anode substrate, as
compared to micron-sized YSZ powders [42]. The reason is due to
the low density of the green tapes of the nano-sized YSZ because of
the large surface area, which can lead to the signicant differences
of the sintering behavior of the YSZ lm and anode substrates. The
reason for the high porosity of LCSFO(SGP) pellet samples as
compared to that of LCSFO(SSP) is most likely related to the low
density of the green pellets. Tao and Irvine also reported similar
phenomena of the low density of the La10(SiO4)6O3 oxyapatite
prepared synthesized by SGP [35].

809

Fig. 4 shows the SEM micrographs of fractured cross-sections

and surface of LCSFO(SSP) oxyapatite pellets sintered at different
temperatures for 4 h in air. The results show that the porosity of
LCSFO(SSP) oxyapatite samples decreases with the increase of
sintering temperature. However, the grain size of the samples increases with the increase of sintering temperature. When the sintering temperature increased to 1600  C, the microstructure of
LCSFO(SSP) oxyapatite pellet shows dense structure with no pores.
EDS mapping of an LCSFO(SSP) oxyapatite pellet shows the uniform
distribution of the constitute elements with no detection of the
second phase formation (Fig. 5).
The differences in the microstructure of LSFO(SSP), LCSFO(SSP)
and LCSFO(SGP) oxyapatite ceramics are also indicated by the bulk
densities of the oxyapatite ceramics, as shown in Table 1. The bulk
densities of LSFO(SSP) and LCSFO(SGP) oxyapatite ceramics sintered at 1500  C are 4.59 and 4.62 g cm3, respectively, which are
signicantly lower than 5.01 g cm3 of LCSFO(SSP), consistent with
the SEM observation. As the sintering temperature increased from
1500 to 1600  C, the bulk density of LCSFO(SSP) is increased from
5.01 to 5.51 g cm3.
3.2. Conductivity
The oxide-ion conductivities of LSFO(SSP), LCSFO(SSP) and
LCSFO(SGP) oxyapatite ceramics sintered 1500  C were investigated by impedance analysis in temperature range of 300e800  C.
Fig. 6 presents the complex impedance plots obtained for
LSFO(SSP), LCSFO(SSP) and LCSFO(SGP), measured at 500  C. The
impedance responses of the oxyapatite materials consist of two
distorted semicircles at high and low frequencies. The impedance
data were analyzed by equivalent circuit with two RQ (R: resistance; Q: constant phase element) elements in series [34,43,44].

Fig. 3. SEM micrographs of fractured cross-sections of (a) LSFO(SSP), (b) LCSFO(SSP) and (c) LCSFO(SGP) apatite pellets sintered at 1500  C for 4 h in air.

810

X.G. Cao et al. / Journal of Power Sources 293 (2015) 806e814

Fig. 4. SEM micrographs of (a, b, c) fractured cross-sections and (d, e, f) surface of LCSFO(SSP) apatite pellets sintered at1500, 1550 and 1600  C for 4 h in air. The oxyapatite powders
were calcined at 1300  C for 10 h.

The symbols are the experimental data and lines are the tted results based on equivalent circuit. The high and low frequency arcs
can be easily separated by the equivalent circuit analysis software,
and can be identied with the grain bulk (Rg) and grain boundary
(Rgb) resistances, respectively, similar to the previous studies
[45e48]. The size of the impedance arc of LCSFO(SSP) is signicantly smaller than that of LSFO(SSP) (Fig. 6a and c), indicating the
higher ionic conductivity of LCSFO(SSP). This is most likely due to
the higher densication and less La2SiO5 impurity of LCSFO(SSP)
specimens. This indicates that co-doping of Ca and Fe obviously
improves the oxide-ion conductivity of lanthanum silicate oxyapatite. The most interesting result is that the overall impedance of
LCSFO(SGP) is signicantly lower than that of LCSFO(SSP) (Fig. 6b
and c), despite the fact that LCSFO(SGP) oxyapatite ceramics is more
porous than LCSFO(SSP) oxyapatite ceramics (Fig. 3). This is most
likely due to the pure apatite phase for LCSFO(SGP). The results
indicate that phase purity has signicant effect on the oxygen ion
conductivity of lanthanum silicate oxyapatites. However, the
impedance of LCSFO(SSP) decreases signicantly with the increase

of sintering temperature (Fig. 6c, b and c), which is likely due to the
higher densication of the oxyapatite ceramics sintered at higher
temperatures. For example, as the sintering temperature increased
from 1500 to 1600  C, the oxide-ion conductivity of LBSFO(SSP)
increased from 2.36  104 to 5.23  104 S cm1 at 500  C. This
indicates that sintering temperature is a crucial parameter in the
achieving high oxide-ion conductivity of lanthanum silicate oxyapatite electrolyte.
Fig. 7 shows the distribution of grain bulk and grain boundary
resistances of LCSFO(SSP) oxyapatite electrolytes as a function of
sintering temperature and LSFO, LCSFO(SSP) and LCSFO(SGP) oxyapatite electrolytes sintered at 1500  C, measured at 500  C. The
grain bulk and grain boundary resistances of the LCSFO(SSP) oxyapatite ceramics decrease with the increase of the sintering temperature (Fig. 7a), which is most likely due to the denser structures
and larger grain size of LCSFO(SSP) samples sintered at high temperatures (Fig. 4). As the sintering temperature increases from 1500
to 1600  C, the contributions of grain bulk resistance to the overall
resistance, Rg/(Rgb Rg), decrease from 63.3 % to 52.1 %, indicating

X.G. Cao et al. / Journal of Power Sources 293 (2015) 806e814

811

Fig. 5. EDS mapping of an LCSFO(SSP) oxyapatite pellet sintered at 1500  C.

of the LCSFO(SGP) are characterized by a large depressed curve,

indicating the dominance of the electrode process. The oxide-ion
conductivity of LCSFO(SGP) calculated from the high frequency
intercept and the geometry of the electrolyte substrate is
2.08  102 S cm1 at 800  C, which is higher than
2.71  103 S cm1 reported on La9CaSi6O26.5 [34] and also higher
than 5.68  103 S cm1 and 1.04  102 S cm1 measured on
LSFO(SSP) and LCSFO(SSP), respectively (Table 1).
The activation energy plots of LSFO(SSP), LCSFO(SSP) and
LCSFO(SGP) oxyapatite electrolytes sintered at 1550  C are shown
in Fig. 8b. The oxide-ion conductivity of LCSFO(SSP) oxyapatite is

that the grain bulk resistance contribution to the overall resistance

of LCSFO(SSP) oxyapatites decreases with the increase of the sintering temperature. The grain boundary resistance contribution to
the overall resistance of LCSFO(SSP) oxyapatite ceramic is 36.7%,
smaller than 53.5% of LSFO(SSP) and 53.6% of LCSFO(SGP) sintered
at 1500  C (Fig. 7b). This is mostly due to the porosity of LCSFO(SSP)
oxyapatite ceramic is signicantly smaller than that of LSFO(SSP)
and LCSFO(SGP) oxyapatite ceramics (Fig. 3).
Fig. 8a shows the complex impedance plots of LCSFO(SGP)
oxyapatites sintered at 1500  C, and measured at 800  C. In the case
of impedance curves measured at 800  C, the impedance responses

Table 1
Oxide-ion conductivity, activation energy and bulk density of LSFO(SSP), LCSFO(SSP) and LCSFO(SGP) oxyapatite electrolytes sintered at different temperature.
Sintering temperature ( C)

Conductivity (S cm1)

LSFO
LCSFO(SSP)

1.84
1.25
4.09
4.50
1.78

300

LCSFO(SGP)

1500
1600
1550
1500
1500

400






106
105
106
106
105

1.98
9.03
5.71
1.10
1.10

500






105
105
105
105
104

1.25
5.23
3.87
2.36
6.03

600






104
104
104
104
104

4.57
2.10
1.58
1.03
2.43

700






104
103
103
103
103

1.69
8.16
6.28
4.05
7.51

Ea (kJ mol1)

Bulk density (gcm3)

86.47
81.36
90.71
88.20
79.85

4.59
5.01
5.43
5.51
4.62

800






103
103
103
103
103

5.68
1.62
1.39
1.04
2.08







103
102
102
102
102

812

X.G. Cao et al. / Journal of Power Sources 293 (2015) 806e814

Fig. 6. Complex impedance plots of (a) LSFO(SSP) and (b) LCSFO(SGP) oxyapatite electrolytes sintered at 1500  C for 4 h. Complex impedance plots of LCSFO(SSP) oxyapatite
electrolytes sintered at (c)1500, (d)1550 and (e)1600  C. Impedance was measured at 500  C and numbers are frequencies in Hz. The symbols are the measured data and the lines
are the tting results.

higher than that of LSFO(SSP) oxyapatite, but lower than that of

LCSFO(SGP) oxyapatite under identical test conditions. For
example, the oxide-ion conductivity of LCSFO(SSP) oxyapatite is
2.36  104 S cm1 at 500  C, which is higher than
1.25  104 S cm1 of LSFO(SSP) oxyapatite and lower than
6.03  104 S cm1 of LCSFO(SGP) oxyapatite. The best oxide-ion
conductivity was obtained on LCSFO oxyapatite synthesized by
solegel method.
The activation energies of LCSFO oxyapatites are in the range of
80e90 kJmol1 in this study, generally lower than ~100 kJmol1
[49]. Even low activation energies of ~50e60 kJ/mol for oxygen
excess lanthanum silicate electrolyte were reported by Mineshige
et al. [16] and for Ba doped lanthanum silicate electrolyte [10]. Low
activation energies were also reported in lanthanum silicate-based
electrolytes by others [46]. The signicantly lower activation

energies of oxyapatites are advantageous for the SOFC operating

particularly at intermediate and low temperatures, as in the
reduced temperatures, the ionic conductivity of lanthanum silicatebased electrolytes is higher than that of zirconia-based electrolytes.
The low activation energy of lanthanum silicate based materials is
essentially related to the fact that in the case of lanthanum silicate,
the oxygen conduction is through the interstitial oxygen ions via a
non-linear (sinusoidal-like) pathway along the c-axis oxide channel
[12,20], while in the case of YSZ, ionic conductivity is via the oxygen
vacancy mechanism. The ceria based electrolyte shows high ionic
conductivity at reduced temperature, however, ceria-based materials are mixed ionic and electronic conductivity and the ionic
transfer number is signicantly lower than unit [50,51]. Thus in
SOFC application, an electronic barrier such as YSZ is required to
prevent the electronic short circuit and to increase the open circuit

X.G. Cao et al. / Journal of Power Sources 293 (2015) 806e814

Fig. 7. Grain bulk and grain boundary resistances of (a) LCSFO(SSP) oxyapatite electrolytes as a function of sintering temperature and (b) LSFO(SSP), LCSFO(SSP) and
LCSFO(SGP) oxyapatite electrolytes sintered at 1500  C, measured at 500  C.

potential of the cells. The dual electrolyte approaches increase the

cost and complicity of the cell manufacturing.
4. Conclusions
Ca and Fe co-doped lanthanum silicate oxyapatite, LCSFO, with
pure apatite phase was synthesized by SGP. Both thermal and XRD
analyses indicate that the phase formation temperature of LCSFO
powders with single apatite phase synthesized by SGP occurs at
750  C. This is substantially lower than complete phase formation
temperature of ~1500  C reported for lanthanum silicates synthesized by SSP. The SGP produces LCSFO nanoparticles with a welldispersed irregular shape and an average size of ~20 nm. The
impedance analysis indicates that co-doping of Ca and Fe substantially increases the oxide-ion conductivity of lanthanum silicate oxyapatite. The oxide-ion conductivity is 2.08  102 S cm1 at
800  C for LCSFO(SGP), higher than 1.04  102 S cm1 and
5.68  103 S cm1 for LCSFO(SSP) and LSFO(SSP) oxyapatite,
respectively, despite the fact that LCSFO(SGP) oxyapatite ceramics
is more porous than LCSFO(SSP) oxyapatite ceramics. The results in
the present study indicate that phase purity play an important role

813

Fig. 8. (a) Complex impedance plots of LCSFO(SGP) oxyapatites sintered at 1500  C.

Impedance was measured at 800  C. (b) Activation energy plots of LSFO(SSP),
LCSFO(SSP) and LCSFO(SGP) oxyapatite specimens sintered at 1500  C.

in the ionic conductivity of the lanthanum silicate-based oxyapatites and demonstrate that the SGP is an effective method to
synthesize the lanthanum silicate oxyapatite nanoparticles with
pure phase at low temperature.
Acknowledgment
The project is supported by the National Natural Science Foundation of China (Grant No. 51302042 and 21176045), the Pearl River
New Star Plan of Science and Technology of Guangzhou City of
China (Grant No. 2013J2200038), and the Foundation of the
Training Plan of Outstanding Young Teachers of Universities in
Guangdong Province.
References
[1] C. Brahim, A. Ringuede, E. Gourba, M. Cassir, A. Billard, P. Briois, J. Power
Sources 156 (2006) 45e49.
[2] D.G. Goodwin, H. Zhu, A.M. Colclasure, R.J. Kee, J. Electrochem. Soc. 156 (2009)
B1004eB1021.
[3] T. Ishihara, J. Tabuchi, S. Ishikawa, J. Yan, M. Enoki, H. Matsumoto, Solid State

814

X.G. Cao et al. / Journal of Power Sources 293 (2015) 806e814

Ionics 177 (2006) 1949e1953.

[4] H. Huang, J.H. Shim, C.C. Chao, R. Pornprasertsuk, M. Sugawara, T.M. Guer,
F.B. Prinz, J. Electrochem. Soc. 156 (2009) B392eB396.
[5] W. Liu, S. Yamaguchi, T. Tsuchiya, S. Miyoshi, K. Kobayashi, W. Pan, J. Power
Sources 235 (2013) 62e66.
[6] A. Mineshige, T. Nakao, M. Kobune, T. Yazawa, H. Yoshioka, Solid State Ionics
179 (2008) 1009e1012.
[7] A. Mineshige, T. Nakao, Y. Ohnishi, M. Kobune, T. Yazawa, H. Yoshioka, Electrochemistry 77 (2009) 146e148.
[8] A.L. Shaula, V.V. Kharton, J.C. Waerenborgh, D.P. Rojas, E. Tsipis, N.P. Vyshatko,
M. Patrakeev, F.M.B. Marques, Mater. Res. Bull. 39 (2004) 763e773.
[9] D.-Y. Kim, S.-G. Lee, Mater. Res. Bull. 47 (2012) 2856e2858.
[10] Y. Nojiri, S. Tanase, M. Iwasa, H. Yoshioka, Y. Matsumura, T. Sakai, J. Power
Sources 195 (2010) 4059e4064.
[11] A. Laghzizil, N. Elherch, A. Bouhaouss, G. Lorente, T. Coradin, J. Livage, Mater.
Res. Bull. 36 (2001) 953e962.
[12] J.R. Tolchard, M.S. Islam, P.R. Slater, J. Mater. Chem. 13 (2003) 1956e1961.
[13] H. Yoshioka, J. Alloys Compd. 408 (2006) 649e652.
[14] X.G. Cao, S.P. Jiang, Int. J. Hydrog. Energy 38 (2013) 2421e2431.
[15] T. Nakao, A. Mineshige, M. Kobune, T. Yazawa, H. Yoshioka, Solid State Ionics
179 (2008) 1567e1569.
[16] A. Mineshige, T. Nakao, Y. Ohnishi, R. Sakamoto, Y. Daiko, M. Kobune,
T. Yazawa, H. Yoshioka, T. Fukutsuka, Y. Uchimoto, J. Electrochem. Soc. 157
(2010) B1465eB1470.
[17] J.E.H. Sansom, D. Richings, P.R. Slater, Solid State Ionics 139 (2001) 205e210.
[18] J.E.H. Sansom, P.R. Slater, Solid State Ionics 167 (2004) 23e27.
[19] X.G. Cao, S.P. Jiang, Int. J. Hydrogen Energy 39 (2014) 19093e19101.
[20] J.E.H. Sansom, J.R. Tolchard, M.S. Islam, D. Apperley, P.R. Slater, J. Mater. Chem.
16 (2006) 1410e1413.
[21] V.V. Kharton, A.L. Shaula, M.V. Patrakeev, J.C. Waerenborgh, D.P. Rojas,
N.P. Vyshatko, E.V. Tsipis, A.A. Yaremchenko, F.M.B. Marques, J. Electrochem.
Soc. 151 (2004) A1236eA1246.
[22] A.L. Shaula, V.V. Kharton, J.C. Waerenborgh, D.P. Rojas, F.M.B. Marques, J. Eur.
Ceram. Soc. 25 (2005) 2583e2586.
[23] J. McFarlane, S. Barth, M. Swaffer, J.E.H. Sansom, P.R. Slater, Ionics 8 (2002)
149e154.
[24] E.J. Abram, D.C. Sinclair, A.R. West, J. Mater. Chem. 11 (2001) 1978e1979.
[25] P.R. Slater, J.E.H. Sansam, The synthesis and characterisation of new apatitetype oxide ion conductors, in: Solid State Chemistry V, Trans Tech Publications Ltd, Zurich-Uetikon, 2003, pp. 195e200.
[26] J.E.H. Sansom, J.R. Tolchard, P.R. Slater, M.S. Islam, Solid State Ionics 167
(2004) 17e22.

[27] A. Vincent, S.B. Savignat, F. Gervais, J. Eur. Ceram. Soc. 27 (2007) 1187e1192.
[28] X.G. Cao, S.P. Jiang, J. Alloys Compd. 523 (2012) 127e133.
[29] S. Beaudet-Savignat, A. Vincent, S. Lambert, F. Gervais, J. Mater. Chem. 17
(2007) 2078e2087.
[30] Y. Masubuchi, M. Higuchi, T. Takeda, S. Kikkawa, Solid State Ionics 177 (2006)
263e268.
[31] L. Leon-Reina, M.C. Martin-Sedeno, E.R. Losilla, A. Cabeza, M. Martinez-Lara,
S. Bruque, F.M.B. Marques, D.V. Sheptyakov, M.A.G. Aranda, Chem. Mater. 15
(2003) 2099e2108.
[32] A. Chesnaud, G. Dezanneau, C. Estournes, C. Bogicevic, F. Karolak, S. Geiger,
G. Geneste, Solid State Ionics 179 (2008) 1929e1939.
[33] S.P. Jiang, L. Zhang, H.Q. He, R.K. Yap, Y. Xiang, J. Power Sources 189 (2009)
972e981.
[34] L. Zhang, H.Q. He, H. Wu, C.-Z. Li, S.P. Jiang, Int. J. Hydrogen Energy 36 (2011)
6862e6874.
[35] S.W. Tao, J.T.S. Irvine, Mater. Res. Bull. 36 (2001) 1245e1258.
[36] S. Celerier, C. Laberty-Robert, J.W. Long, K.A. Pettigrew, R.M. Stroud,
D.R. Rolison, F. Ansart, P. Stevens, Adv. Mater. 18 (2006) 615e618.
[37] S. Celerier, C. Laberty-Robert, F. Ansart, C. Calmet, P. Stevens, J. Eur. Ceram.
Soc. 25 (2005) 2665e2668.
[38] S. Celerier, C. Laberty, F. Ansart, P. Lenormand, P. Stevens, Ceram. Int. 32
(2006) 271e276.
[39] C. Tian, J. Liu, J. Cai, Y. Zeng, J. Alloys Compd. 458 (2008) 378e382.
[40] J. Felsche, J. Solid State Chem. 5 (1972) 266e275.
[41] S. Nakayama, T. Kageyama, H. Aono, Y. Sadaoka, J. Mater. Chem. 5 (1995)
1801e1805.
[42] Y.J. Leng, S.H. Chan, K.A. Khor, S.P. Jiang, P. Cheang, J. Power Sources 117
(2003) 26e34.
[43] A. Babaei, L. Zhang, S.L. Tan, S.P. Jiang, Solid State Ionics 181 (2010)
1221e1228.
[44] S.P. Jiang, J.G. Love, Y. Ramprakash, J. Power Sources 110 (2002) 201e208.
[45] S. Che, A. Madani, H. Boussetta, C. Roux, A. Hammou, J. Power Sources 177
(2008) 464e469.
[46] E. Bechade, I. Julien, T. Iwata, O. Masson, P. Thomas, E. Champion, K. Fukuda,
J. Eur. Ceram. Soc. 28 (2008) 2717e2724.
[47] S.H. Jo, P. Muralidharan, D.K. Kim, Electrochimica Acta 54 (2009) 7495e7501.
[48] X.G. Cao, S.P. Jiang, J. Mater. Chem. A 2 (2014) 20739e20747.
[49] S.P.S. Badwal, Solid State Ionics 52 (1992) 23e32.
[50] Y.P. Xiong, H. Kishimoto, K. Yamaji, M. Yoshinaga, T. Horita, M.E. Brito,
H. Yokokawa, Solid State Ionics 192 (2011) 476e479.
[51] M. Mogensen, N.M. Sammes, G.A. Tompsett, Solid State Ionics 129 (2000)
63e94.