Electrical, thermoelectric, and structural properties of La(M xFe1-x)O3 (M=Mn, Ni, Cu)

Article abstract of DOI:10.1149/1.2358840

Electrical, thermoelectric, and structural properties were studied in transition metal ion-substituted LaFe O3: La (Mnx Fe1-x) O3, La (Nix Fe1-x) O3, and La (Cux Fe1-x) O3. Structural analysis showed that a continuous series of solid solutions with no intermediate phases are forming over a wide range (0a strong temperature dependence, indicating a thermally activated carrier formation. The activation energy for carrier formation in La (Cux Fe1-x) O3 is greater than that in La (Mnx Fe1-x) O3. Thermoelectric and electrical properties evidence conduction through polaron hopping in both Mn- and Cu-substituted LaFe O3, whereas the Ni-substituted LaFe O3 shows metallic conductivity.The Electrochemical Society

Full text of DOI:10.1149/1.2358840

Journal of The Electrochemical Society, 153 ͑12͒ J133-J138 ͑2006͒
J133
0013-4651/2006/153͑12͒/J133/6/$20.00 © The Electrochemical Society
Electrical, Thermoelectric, and Structural Properties of
La„MxFe1−x…O3 „M = Mn, Ni, Cu…
,z
X.-D. Zhou,a,b, J. B. Yang,b E.-C. Thomsen,a Q. Cai,c B. J. Scarfino,b Z. Nie,a
*
G. W. Coffey,a, W. J. James, W. B. Yelon, H. U. Anderson, and
b
b,
b,
*
*
*
L. R. Pedersona
aPacific Northwest National Laboratory, Richland, Washington 99352, USA
bMaterials Research Center, University of Missouri-Rolla, Rolla, Missouri 65401, USA
cDepartment of Physics, University of Missouri-Columbia, Columbia, Missouri 65102, USA
Electrical, thermoelectric, and structural properties were studied in transition metal ion-substituted LaFeO3: La͑MnxFe1−x͒O3,
La͑NixFe1−x͒O3, and La͑CuxFe1−x͒O3. Structural analysis showed that a continuous series of solid solutions with no intermediate
phases are forming over a wide range ͑0 Ͻ x Ͻ 1͒ with substitutions of Mn and Ni, whereas the maximum Cu content is 30%
from this study. The Ni-substituted LaFeO3 specimens have substantially higher conductivity than those substituted with either Mn
or Cu, measured in air from 100 to 1000°C. The Seebeck coefficient of La͑MnxFe1−x͒O3 and La͑CuxFe1−x͒O3 has a strong
temperature dependence, indicating a thermally activated carrier formation. The activation energy for carrier formation in
La͑CuxFe1−x͒O3 is greater than that in La͑MnxFe1−x͒O3. Thermoelectric and electrical properties evidence conduction through
polaron hopping in both Mn- and Cu-substituted LaFeO3, whereas the Ni-substituted LaFeO3 shows metallic conductivity.
© 2006 The Electrochemical Society. ͓DOI: 10.1149/1.2358840͔ All rights reserved.
Manuscript submitted June 28, 2006; revised manuscript received August 8, 2006. Available electronically October 17, 2006.
Perovskite family oxides ͑ABO3͒ are of particular interest as
energy materials because, by carefully choosing the suitable cations,
their properties can be tailored to be oxide ion conductors, mixed
ionic and electronic conductors, and protonic conductors.1 The ri-
gidity of the oxygen octahedron ͑BO6͒ is the crystal chemistry ori-
gin of the diverse and unique properties in the perovskite family
oxides.1,2 The octahedron can tilt, bend, and tolerate loss of oxygen
ions. Moreover, the B site can incorporate more than one transition
metal ion, which often randomly distribute in the crystal; albeit, B
site cation ordering is reported in some perovskites.3 The rigid struc-
ture and various oxidation states of B site cations result in strongly
correlated properties in these oxides. Correlations between thermo-
electric and electric properties have been extensively studied in
p-type perovskite oxides, particularly LaCrO3- ͑chromites͒ and
LaMnO3- ͑manganites͒ based materials, in which polaron-hopping
conduction takes place.4-6
͑La,Sr͒͑Co,Fe͒O3, has been of substantial interest because of result-
ant enhancement in both electronic and ionic conductivities at el-
evated temperatures.
To further understand the nature of the electron transport and
thermoelectricity in the perovskite family oxides, it is necessary to
know the character of the exchange interaction between two B site
͑B1 and B2͒ cations in La͑B1,B2͒O3. The underlying physics of B
site cations is of interest in itself. In contrast to the well-studied
A-site-substituted perovskites, the B-site-substituted materials,
La͑B1,B2͒O3, are not well documented. Raffaelle et al.5,6 studied
solid solutions of LaMnO3 and LaCrO3. A small polaron conduction
behavior was observed in the solid solutions with a typical activa-
tion energy ϳ0.2 eV. Low-temperature electrical and magnetic
properties have been studied by Rao and his colleagues in other
B-site-substituted perovskite oxides.13-15 However, in transition
metal ion-substituted LaFeO3, there is a lack of reported data, and
consequently, electrical and thermoelectric measurements are
needed to provide further insight into the high-temperature electro-
chemical properties of these materials.
In this paper, systematic experiments to characterize electrical
conductivity, the Seebeck coefficient, and the crystal structure of
La͑MnxFe1−x͒O3, La͑NixFe1−x͒O3, and La͑CuxFe1−x͒O3 are re-
ported. Polaron hopping is found to contribute to conduction in both
La͑MnxFe1−x͒O3 and La͑CuxFe1−x͒O3, in which the majority
carriers-electron holes-are generated through a thermally activated
process. The Seebeck coefficient,S͑T͒, therefore, exponentially de-
creases with increasing temperature. In contrast, La͑NixFe1−x͒O3 is a
metallic conductor in which the absolute values of the Seebeck co-
efficient are increased with increasing temperature.
The Seebeck coefficient ͑S, V/K͒ can be experimentally deter-
mined from thermoelectric power measurements. Theoretically, S in
these materials represents the transport entropy per carrier charge
and is expressed as
k
S = ln
e
1 − c
ͩ ͪ
c
+ S
͓1͔
0
where S0 represents the vibrational entropy per particle and is esti-
mated to be on the order of or less than 10 ␮V/K; k is the Boltz-
mann constant, e is the charge of the electron, and c is the charge
carrier fraction. Equation 1 relates the Seebeck coefficient to the
fractional charge carrier concentration c, and thus to the resistivity
͑␳, ⍀ cm͒, which is inversely proportional to c at a constant tem-
perature.
Chromites are considered a typical example of small polaron
systems.7,8 Equation 1 was particularly successful in calculating the
charge-carrier concentration, which was used to compute charge-
carrier mobility from experimental conductivity data.9 In
La1−xSrxMnO3−d, Mizusaki et al. attributed the predominant elec-
tronic conduction to the electron hopping on the eg↑ level
of Mn, which was considered responsible for the thermoelectric
power.10 The substitution at both A and B sites,11,12 such as
Experimental
The glycine nitrate process was used to prepare the oxide
powders.16 Nitrate solutions were standardized using thermogravi-
metric analysis in both air and reducing environments. The A/B ratio
was set at 1.00. The raw powders, which were subsequently used to
prepare ceramic bars, were calcined at 600°C for 1 h. Bar-shaped-
substituted LaFeO3 specimens ͑4.0 cm ϫ 2mm ϫ 2 mm͒ for both
conductivity and thermoelectric power measurements were prepared
by uniaxial-pressing at 40 MPa, followed by cold isostatic pressing
at 200 MPa. Bars were sintered on substrates of the same composi-
tion at temperatures between 1000 and 1200°C, with a heating rate
of 3°C/min, a soak time of 4 h, and a cooling rate of 5°C/min. The
final density, measured by Archimedes’ method, was more than 90%
of the theoretical value.
*
Electrochemical Society Active Member.
z E-mail: xiaodong.zhou@pnl.gov
Downloaded on 2015-02-18 to IP 138.251.14.35 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

J134
Journal of The Electrochemical Society, 153 ͑12͒ J133-J138 ͑2006͒
Figure 1. XRD results of La͑MnxFe1−x͒O3 powders calcined at 1000°C.
Figure 2. ͑Color online͒ Unit cell volume as a function of Mn fraction in
La͑MnxFe1−x͒O3+␧ normalized for the different number of formula units in
the orthorhombic and rhombohedral cells.
Powder X-ray diffraction ͑XRD͒ and neutron diffraction ͑ND͒
were carried out on powders calcined at 1000°C for 1 h. XRD was
measured using a Scintag diffractometer ͑Cu K␣ radiation͒ operated
at 45 kV and 40 mA, with a 0.03° step size and 1 s dwell time. ND
measurements were carried out at the University of Missouri-
Columbia Research Reactor ͑MURR͒ using neutrons of wavelength
␭ = 1.4875 Å. The data for each sample were collected between
2␪ = 5.65 − 105.60°. Refinement of the neutron diffraction data was
carried out using the FULLPROF program,17 which permits mul-
tiple phase refinements as well as magnetic structure refinements.
Electrical conductivities ͑␴͒ were established using the four-
probe dc methods. The sintered ceramic bars were notched at four
locations along the length of the bar. Pt wires were then wrapped
around these notches. Platinum paste was used to cover both ends of
the specimen to provide a uniform current path through the ceramic
bulk. A thermocouple was placed 2 mm away from the center of the
sample and the sample was located in the constant temperature zone
in the furnace. To minimize the Joule heating effect and polarization,
a small current ͑range from 10 to 100 ␮A͒ was used to measure the
resistance.
The temperature dependence of Seebeck measurements, S͑T͒,
was measured as follows. Two Pt/Pt–Rh thermocouples were placed
at the end of the sample and spring-loaded to provide a thermal and
electrical contact between the thermocouple and both ends of the
sample. A Pt wire was used as a heater to provide variable tempera-
ture gradients ͑⌬T͒ across the specimens. The voltage gradient ͑⌬V͒
was measured through the two Pt wires of both thermocouples. A
computer-controlled Labview program was used to collect both ⌬V
and ⌬T. The measured Seebeck coefficient ͑Sm͒ was obtained by
plotting ⌬V vs ⌬T, a slope of which gives the value of Sm. The
absolute thermopower of the Pt voltage leads was subtracted from
the measured Seebeck coefficient Sm to give the sample Seebeck
coefficient.
x = 0.95 the excess oxygen is close to 7%, determined from neutron
scattering. The perovskite structure is close-packed; thus, interstitial
oxygen is not allowed. Therefore, the excess oxygen refined for the
nominal site indicates the existence of cation vacancies on both the
A and B sites.
The amount of oxygen excess ͑or cation vacancies͒ substantially
impacts structural and electrical properties in manganites. The crys-
tal structures of unsubstituted and substituted lanthanum manganites
͑LaMnO3+␦͒ are well known to be dependent on the oxygen occu-
pancy ͑3 + ␦͒. Rhombohedral symmetry was observed in
LaMnO3+␦ for ␦ Ͼ 0.08, in contrast to orthorhombic symmetry for
␦ Ͻ 0.06.19 Lanthanum ferrite ͑LaFeO3͒ has an orthorhombic struc-
ture ͑space group Pbnm͒. In La͑MnxFe1−x͒O3 solid solutions, the
Fe-rich samples ͑x Ͻ 0.6͒ have orthorhombic symmetry, and the
data were refined in space group Pbnm. Likewise, rhombohedral
symmetry ͑space group R3/c͒ is found for Mn rich specimens ͑x
ജ 0.6͒. The unit cell volume as a function of Mn fraction in
La͑MnxFe1−x͒O3+␦ is shown in Fig. 2. The orthorhombicity de-
creases with increasing Mn fraction. In perovskite oxide, ABO3,
the mismatch between equilibrium A–O and B–O bond lengths is
given by a tolerance factor1
͑A − O͒
ͱ2͑B − O͒
t =
͓2͔
The cubic perovskite ͑t = 1͒ can tolerate a change in t by a coop-
erative rotation of the BO6 octahedra to lower the symmetry to
tetragonal, rhombohedral, or orthorhombic. The decrease in
orthorthombicity shown in Fig. 2 can be attributed to the substitu-
tion of Mn and/or increase of oxygen nonstoichiometry. The ionic
radius of Fe3+ is ϳ0.645 Å for coordination number 6, which is
nearly identical with that of Mn3+ ͑0.645 Å͒.20 The oxygen stoichi-
ometry controls the unit cell volume in LaMnO3+␦. It decreases from
244.2 Å3 when ␦ = 0 to 240.5 Å3 when ␦ = 0.06.19 Hence, the
change in symmetry shown in Fig. 2 is mainly due to the change in
oxygen stoichiometry.
Results and Discussion
Solid solutions of LaFeO3 and LaMnO3.— Structure.— Mea-
surements of both XRD and ND on samples of LaMnxFe1−xO3+␦ for
0.05 ഛ x ഛ 0.95 were conducted. Detailed information on the
structural ordering in LaMnxFe1−xO3+␦ was published elsewhere.18
Shown in Fig. 1 is a plot of XRD patterns of LaMnxFe1−xO3+␦
powders calcined at 1000°C, which indicates that all samples con-
sisted of a single phase. Rietveld refinements on ND measurements
were performed with the constraint of full La occupancy and
La/͑Mn + Fe͒ equal to one. Oxygen site occupancy, therefore, was
allowed to vary. Refinement of the oxygen content shows the pres-
ence of oxygen excess, particularly when x Ͼ 0.5 ͑i.e., Mn-rich
samples͒. For x = 0.05 the oxygen excess is negligible, while for
Electrical Properties
.— Figure 3 shows a plot of ln͑␴T͒ vs 1/T of
La͑MnxFe1−x͒O3+␦ and reveals a strong temperature dependence of
␴͑T͒. For instance, ␴͑T͒ for LaMn0.5Fe0.5O3 increases nearly 2 or-
ders of magnitude from 150°C ͑ϳ0.25/⍀ cm͒ to 1000°C
͑21/⍀ cm͒. Assume ␴T follows
Downloaded on 2015-02-18 to IP 138.251.14.35 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 153 ͑12͒ J133-J138 ͑2006͒
J135
Figure 4. ͑Color online͒ The Seebeck coefficients and the carrier fraction as
a function of temperature, calculated using the small polaron model, for
La͑Fe0.40Mn0.60͒O3 measured in air.
Figure 3. ln͑␴T͒ vs 1/T for La͑MnxFe1−x͒O3 measured in air.
ing Sr ͑oxidation state of 2+͒ for La ͑oxidation state of 3+͒ in
LaFeO3 generates an electron hole, a positive charge compensating
negative substitution, SrЈ ͑Kröger–Vink notation͒.22 The holes are
affiliated with transitionLma etal ions, increasing the oxidation states
of these ions. For instance, in ͑La,Sr͒FeO3 ͑LSF͒, high oxidation
states of Fe ͑4+͒ were observed by measuring the magnetic proper-
ties of LSF.23 Much less is known about carrier formation with
substitution on the B site with transition metal ions, such as
La͑MnxFe1−x͒O3.
The carrier fraction shown in Fig. 4 was calculated from the
measured Seebeck coefficient based on Eq. 1 for LaMn0.60Fe0.40O3.
The positive dependence of the carrier fraction on temperature indi-
cates that the carrier formation is thermally activated, following
Arrhenius law
␴
− E
␴
0
␴ =
exp
͓3͔
ͩ ͪ
T
kT
where ␴ is a pre-exponential factor. A least-squares fit of a plot of
0
the data to Eq. 3 gives an activation energy of 0.30 0.01 eV for
conduction in LaMn0.5Fe0.5O3. Listed in Table I are the activation
energies for the compounds in this study.
Shown in Fig. 4 is a plot of S͑T͒ of LaFe0.40Mn0.60O3 which has
a negative temperature dependence; it is large in magnitude and
positive ͑ϳ170 ␮V/K at 150°C͒. This behavior is typical of an
oxide semiconductor with electron holes as majority charge carriers.
The observation of decreasing S͑T͒ with temperature indicates that
the generation of charge carriers is a thermally activated process,
during which the oxidation states of transition metal ions are chang-
ing. The S͑T͒ approaches to a plateau when T Ͼ 700°C, represent-
ing an equilibrium state of oxidation states.
− En
n = n exp
͓4͔
ͩ ͪ
0
kT
The activation energy for the carrier formation, however, is
rather low ͑En ϳ 0.1 eV͒ in the temperature regime from
100 to 500°C. This may be caused by the charge disproportionation
reaction, proposed in other manganite systems as 2Mn3+ → Mn4+
+ Mn2+.6 The oxidation states and spin states of Fe have been stud-
ied by Tong et al. from electron spin resonance ͑ESR͒.24 They in-
ferred the presence of an intermediate spin state of Fe3+ ͑t24geg1͒ in
LaMn1−xFexO3 ͑x ഛ 0.4͒, which is consistent with the neutron dif-
fraction results.18 Therefore, the majority of Mn and Fe cations are
at nominal oxidation states ͑3+͒ within the low-temperature regime,
and Fe2+ is balanced by Mn4+ in the elevated temperature region.
The carrier formation.— In the early 1950s, Verwey et al.21 ob-
served that when substitutions are made in the perovskite lattice, if
the valence of the substituting ion is different ͑i.e., it is aliovalent͒
from that required by the site, then the charge imbalance and overall
charge neutrality will be maintained by the formation of electrons,
holes, or charged point defects. The compensation process creates
carriers that can take part in electrical conductivity. Thus, substitut-
Table I. Activation energies (eV) for conduction and charge-
carrier formation.
Solid solutions of LaFeO3 and LaCuO3.— Cu-substituted LSF
was found to give good performance as the cathode for solid oxide
fuel cells ͑SOFCs͒.25 Therefore, Cu-substituted LaFeO3 was studied
to help understand the cathodic property in SOFCs. For synthesis
using the glycine–nitrate process, single phase samples with ortho-
rhombic symmetry were prepared to a maximum Cu content of 30%
in LaFeO3. The unit cell volume is 242.3 A3 for x = 0.01, and it
decreases to 240.2 A3 for x = 0.20, as shown in Fig. 5. Oxygen
occupancy refined from room temperature neutron diffraction data is
found to be at 3.0 and is independent of Cu content. The decrease in
unit cell volume is due to the smaller ionic radius of the substituting
Cu3+ ions ͑r = 0.54 Å͒ than that of Fe3+ ions ͑r = 0.645 Å͒.20
Shown in Fig. 6 is a plot of ln͑␴T͒ vs 1/T for La͑Fe1−xCux͒O3
measured in air. When T Ͼ 700°C, the conductivity in
Cation fraction
Ea
Temperature ͑°C͒
LaMnxFe1−xO3
x = 031
x = 0.1
x = 0.4
x = 0.5
x = 0.6
x = 0.9
x = 16
x = 0.01
x = 0.05
x = 0.05
x = 0.05
1.14
0.72
0.56
0.32
0.30
0.18
0.19
0.74
0.67
0.60
0.47
277–289
620–880
100–1000
100–1000
140–1000
30–900
30–1000
200–450
150–450
150–450
150–450
LaCuxFe1−xO3
Downloaded on 2015-02-18 to IP 138.251.14.35 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

J136
Journal of The Electrochemical Society, 153 ͑12͒ J133-J138 ͑2006͒
Figure 5. The unit cell volume of La͑CuxFe1−x͒O3 as a function of Cu fraction. ͑Inset͒ XRD patterns of powders of La͑CuxFe1−x͒O3 calcined at 1000°C.
temperature region, the activation energies for conduction in
La͑Fe1−xCux͒O3 is 0.74 eV for x = 0.01 and 0.45 eV for x = 0.2.
Figure 7 shows a plot of S͑T͒ and calculated carrier fraction for
La͑Fe0.99Cu0.01͒O3. The carrier fraction increases from 2 ϫ 10−5 at
200°C to 1 ϫ 10−2 at 600°C. The high activation energy for carrier
formation is found to be ϳ0.3 eV. A plateau is observed from
600 to 800°C, indicating a stable region for charge compensation
͑i.e., stable oxidation states for these cations͒. Hence, in
La͑Fe0.80Cu0.20͒O3 decreases with increasing temperature. From the
previous studies,26 we know that the decrease of conductivity at
very high temperatures in these compounds is caused by loss of
oxygen, confirmed by thermogravimetric analysis ͑TGA͒ measure-
ments. This decreases the total carrier density. In the low-
Figure 7. ͑Color online͒ The Seebeck coefficient and carrier fraction as a
function of temperature for La͑CuxFe1−x͒O3. The carrier fraction was calcu-
lated from the small polaron model ͑Eq. 1͒.
Figure 6. ͑Color online͒ ln͑␴T͒ vs 1/T for La͑CuxFe1−x͒O3 measured in air.
Downloaded on 2015-02-18 to IP 138.251.14.35 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 153 ͑12͒ J133-J138 ͑2006͒
J137
conductivity for LaNi0.6Fe0.4O3 around 250°C was observed with a
␴ ϳ 540 ⍀ cm in air. LaNiO3 has long been known as a metallic
oxide,27 with conductivity ϳ2000 ⍀ cm at 2 K, which gradually
decreases to ϳ500 ⍀ cm at 300 K.13,14 The metallic conduction is
caused by the strong covalent bonding between low-spin NiIII ͑t26geg1͒
and O2−, which results in ␴-bonding e orbitals being transformed
*
into itinerant ␴ states. In contrast, LaFeO3 is known as an antifer-
romagnetic insulator, which contains a high-spin, localized electron
configuration of Fe3+ ͑t23geg2͒. Low-temperature studies showed that
a critical concentration ͑x ജ 0.7͒ is required to produce a metallic
state in LaNixFe1−xO3. When x is less than 0.7, the conductivity of
La͑NixFe1−x͒O3 decreases by as much as three orders of magnitude
from 300 to 4 K,14 producing insulating behavior at low tempera-
tures. For x Ͼ 0.7, conductivity is nearly constant over this tempera-
ture regime, indicating metallic conduction. Metallic conduction is
also shown by the Seebeck coefficient of La͑NixFe1−x͒O3 ͑x
Ͼ 0.4͒, which is negative and small ͑with the absolute values
Ͻ10 ␮V/K͒. This is typical of a metallic state.28,29 The metallic
conduction in La͑NixFe1−x͒O3 contrasts with polaron-hopping in
Mn- or Cu-substituted LaFeO3. The La͑NixFe1−x͒O3 series pos-
sesses very high conductivities over a broad range of temperatures,
attributed to high carrier fraction and great mobility.
Figure 8. XRD results for Ni-substituted LaFeO3 powders calcined at
1000°C.
The conductivity maximum.— The question remains as to the
origin of the conductivity maximum. In metallic conductors, this
conductivity maximum is attributed to enhanced scattering of the
electrons because of the large phonon populations at the higher
temperatures. For instance, the transition temperature for
LaNi0.60Fe0.40O3 is ϳ200°C, as shown in Fig. 9. This mechanism is
not valid for many other p-type perovskite conductors, such as
La͑Fe1−xCux͒O3, shown in Fig. 6, because the high activation en-
ergy for carrier motion ͑Ͼ0.1 eV͒ indicates polaron conduction.
The simple electroneutrality condition in p-type conductors is given
in the Kröger–Vink notation as ͓AFЈe͔ = 2͓VؠOؠ͔ + ͓hؠ͔, where ͓AFЈe͔ is
LaFe0.99Cu0.01O3, the oxidation state of Cu at the low-temperature
region is postulated to be 3+ and is reduced to 2+ at higher tem-
peratures. The Cu2+ cations are compensated by Fe4+, acting as elec-
tron holes in the Cu-substituted LaFeO3 system. A further increase
in the carrier fraction at T Ͼ 800°C may result from a further re-
duction of Cu2+ to Cu+. Thermoelectric power measurements of
La͑MnxFe1−x͒O3 and La͑CuxFe1−x͒O3 show exponential increases
of charge-carrier fraction with increasing temperature, indicating
that charge-carrier formation is strongly dependent upon tempera-
ture.
the acceptor substitution concentration, ͓VؠOؠ͔ is the oxygen vacancy
concentration, and ͓ho͔ is the hole concentration. Oxygen vacancies
are generated at high temperatures, resulting in a higher ͓VؠOؠ͔ and a
Solid solutions of LaFeO and LaNiO
3.— LaFeO3 and LaNiO3
3
are found to form a continuous series of solid solutions without the
formation of ntermediate phases. LaNiO3 has rhombohedral symme-
try ͑space group R3c͒, and LaFeO3 is orthorhombic. Figure 8 shows
the XRD results for LaFe1−xNixO3 powder calcined at 1000°C. A
structural transition takes place around x = 0.5. At elevated tempera-
ture ͑T Ͼ 800°C͒, LaNiO3 is not stable and decomposes into
La2NiO4 and NiO. The presence of Fe in LaNiO3 increases the
lower ͓ho͔, assuming ͓AFЈe͔ is constant. Therefore, because ␴
= ␮nq, where n is the carrier concentration and q is the carrier
charge, at elevated temperature, the increased carrier mobility com-
petes with the decrease in the carrier concentration, n, resulting in
the conductivity maximum. This is valid in most A-site-substituted
ferrites, manganites, and chromites, such as La1−xSrxFeO3,
La1−xSrxMnO3, and La1−xSrxCrO3.26 Hence, it is necessary to revisit
the La͑NixFe1−x͒O3 case with respect to formation of oxygen vacan-
cies around the transition temperature. The oxygen content of
La͑NixFe1−x͒O3 remains close to 3.0 up to at least T = 1000°C.
Thus, the decreasing conductivity in La͑NixFe1−x͒O3 at elevated
temperatures cannot be attributed to the generation of oxygen va-
cancies but rather to greater scattering at higher temperatures. Simi-
lar behavior was reported in LaNi0.8Fe0.2O3, with a conductivity
maximum at ϳ110 K.14 The temperature corresponding to conduc-
tivity maximum increases with reducing the fraction of Ni ions,
which is evidenced by the maximum conductivity for
LaNi0.6Fe0.4O3 around 250°C in this work. The electrical conduc-
tivity is polaronic in higher fraction Fe compounds ͑x Ͻ 0.2͒, which
is also observed in La1−xSrxFeO3.30
high-temperature stability.
A single phase is observed for
La͑Ni0.80Fe0.20͒O3 after being calcining at 1000°C for 2 h.
Figure 9 shows the plots of ␴ vs temperature ͑°C͒ for
La͑NixFe1−x͒O3 measured in air from 200 to 1200°C. A maximum
Conclusions
In this paper, systematic measurements of electrical conductivity,
the Seebeck coefficient, and the crystal structures of
La͑MnxFe1−x͒O3, La͑NixFe1−x͒O3, and La͑CuxFe1−x͒O3 are re-
ported. Polaron hopping is found to contribute to conduction in both
La͑MnxFe1−x͒O3 and La͑CuxFe1−x͒O3, in which the majority carri-
ers, electron holes, are generated through a thermally activated pro-
cess. Thus, the Seebeck coefficient, S͑T͒, decreases exponentially
with increasing temperature. The conductivity maxima in the
Figure 9. Conductivity of La͑NixFe1−x͒O3 as a function of temperature,
measured in air.
Downloaded on 2015-02-18 to IP 138.251.14.35 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

J138
Journal of The Electrochemical Society, 153 ͑12͒ J133-J138 ͑2006͒
J. Electrochem. Soc., 143, 2722 ͑1996͒.
La͑MnxFe1−x͒O3 and La͑CuxFe1−x͒O3 are the result of increased
hopping mobility at high temperature and decreasing carrier fraction
due to the formation of oxygen vacancies. In La͑Fe1−xNix͒O3, the
maximum conductivity is caused by scattering of electron holes at
elevated temperatures.
10. J. Mizusaki, Y. Yonemura, H. Kamata, K. Ohyama, N. Mori, H. Takai, H. Tagawa,
M. Dokiya, K. Naraya, T. Sasamoto, H. Inaba, and T. Hashimoto, Solid State
Ionics, 132, 167 ͑2000͒.
11. L. W. Tai, M. M. Nasrallah, H. U. Anderson, D. M. Sparlin, and S. R. Sehlin, Solid
State Ionics, 76, 273 ͑1995͒.
12. C. Y. Park and A. J. Jacobson, J. Electrochem. Soc., 152, J65 ͑2005͒.
13. P. Ganguly, N. Y. Vasanthacharya, C. N. R. Rao, and P. P. Edwards, J. Solid State
Chem., 54, 400 ͑1984͒.
Acknowledgments
14. A. Chainani, D. D. Sarmayk, I. Dasz, and E. V. Sampathkumaranz, J. Phys.: Con-
Part of this work is supported by U.S. Department of Energy
under the High Temperature Electrochemistry Center at the Pacific
Northwest National Laboratory. Pacific Northwest National Labora-
tory is operated by Battelle for the U.S. Department of Energy under
contract DE-AC05-76RL01830.
dens. Matter, 8, L631 ͑1996͒.
15. C. N. R. Rao, O. Parkash, and P. Ganguly, J. Solid State Chem., 15, 186 ͑1975͒.
16. L. A. Chick, L. R. Pederson, G. D. Maupin, J. L. Bates, L. E. Thomas, and G. J.
Exarhos, Mater. Lett., 10, 6 ͑1990͒.
17. J. Rodriguez-Carvajal, FULLPROF program, Version 3.d ͑1998͒.
18. X.-D. Zhou, Q. Cai, J. B. Yang, B. J. Scarfino, C. Wang, M. Kim, W. J. James, W.
B. Yelon, and H. U. Anderson, J. Appl. Phys., 99, 08M918 ͑2006͒.
19. J. Topfer and J. B. Goodenough, J. Solid State Chem., 130, 117 ͑1997͒.
20. R. D. Shannon, Acta Crystallogr., 32, 751 ͑1976͒.
Pacific Northwest National Laboratory assisted in meeting the publica-
tion costs of this article.
21. E. J. W. Verwey, P. W. Haaijman, F. C. Romeijn, and G. W. Van Oosterhout, Philips
References
Res. Rep., 5, 173 ͑1950͒.
1. J. B. Goodenough, Rep. Prog. Phys., 67, 1915 ͑2004͒.
2. K. T. Lee and A. Manthiram, J. Electrochem. Soc., 153, A794 ͑2006͒.
3. J. B. Yang, X.-D. Zhou, Q. Cai, W. B. Yelon, W. J. James, and H. U. Anderson, J.
Phys.: Condens. Matter, 15, 5093 ͑2003͒; C. Perrin, N. Menguy, O. Bidault, C. Y.
Zahra, A.-M. Zahra, C. Caranoni, B. Hilczer, and A. Stepanov, J. Phys.: Condens.
Matter, 13, 10231 ͑2001͒.
4. J. H. Kuo, H. U. Anderson, and D. M. Sparlin, J. Solid State Chem., 87, 55 ͑1990͒.
5. R. Raffaelle, H. U. Anderson, D. M. Sparlin, and P. E. Parris, Phys. Rev. Lett., 65,
1383 ͑1990͒.
22. F. A. Kroger, The Chemistry of Imperfect Crystals, Wiley, New York ͑1964͒.
23. X. D. Zhou, Q. Cai, J. Yang, W. B. Yelon, W. J. James, and H. U. Anderson, Solid
State Ionics, 175, 83 ͑2004͒.
24. W. Tong, B. Zhang, S. Tan, and Y. Zhang, Phys. Rev. B, 70, 014422 ͑2004͒.
25. S. P. Simmer, M. D. Anderson, and J. W. Stevenson, J. Am. Chem. Soc., 87, 1471
͑2004͒.
26. X.-D. Zhou and H. U. Anderson, Defect Chemistry of Ternary Oxides, Woodhead
Publishing, Ltd., Cambridge, England ͑2005͒.
27. J. B. Goodenough and P. M. Raccah, J. Appl. Phys., 36, 1031 ͑1965͒.
28. K. Huang, H. Y. Lee, and J. B. Goodenough, J. Electrochem. Soc., 145, 3220
͑1998͒.
6. R. Raffaelle, H. U. Anderson, D. M. Sparlin, and P. E. Parris, Phys. Rev. B, 43,
7991 ͑1991͒.
7. J. W. Stevenson, M. M. Nasrallah, H. U. Anderson, and D. M. Sparlin, J. Solid
State Chem., 102, 175 ͑1993͒.
8. G. F. Carini, H. U. Anderson, D. M. Sparlin, and M. M. Nasrallah, Solid State
Ionics, 49, 233 ͑1991͒.
9. J. W. Stevenson, T. R. Armstrong, R. D. Carneim, L. R. Pederson, and W. J. Weber,
29. X. Granados, J. Fontcuberta, X. Obradors, L. Mañosa, and J. B. Torrance, Phys.
Rev. B, 48, 11666 ͑1993͒.
30. M. V. Patrakeev, J. A. Bahteeva, E. B. Mitberg, I. A. Leonidova, V. L. Kozhevni-
kov, and K. R. Poeppelmeier, J. Solid State Chem., 172, 219 ͑2003͒.
31. K. S. Roh, K. H. Ryu, and C. H. Yo, J. Mater. Sci., 30, 1245 ͑1995͒.
Downloaded on 2015-02-18 to IP 138.251.14.35 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).