Double Perovskite Sr2B B O6 Oxides For H

INVITED REVIEW
Downloaded from https://www.cambridge.org/core. IP address: 107.173.20.55, on 02 Nov 2018 at 20:32:30, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1557/jmr.2018.376
Double perovskite (Sr2B9B0O6) oxides for high-temperature

thermoelectric power generation—A review
Tanmoy Maiti,a) Mandvi Saxena, and Pinku Roy
Plasmonics and Perovskites Laboratory, Department of Materials Science and Engineering, Indian Institute of
Technology Kanpur, Uttar Pradesh 208016, India
(Received 1 June 2018; accepted 20 September 2018)
Recently, double perovskite-based oxide materials have been proposed for thermoelectric (TE)
applications due to their environment-friendly nature, high-temperature stability, better oxidation
resistance, and lower processing cost compared to conventional chalcogenides and intermetallics. In
this review article, we have comprehensively summarized our recent research studies on Sr2B9B0O6-
based double perovskites for high-temperature TE power generation. We have shown that decoupling
of phonon-glass and electron-crystal behavior is possible in oxides by reducing thermal conductivity
due to induced dipolar glassy state as a result of relaxor ferroelectricity. We have also introduced
metal-like electrical conductivity (;105 S/m) in these ceramics that are inherently insulator in nature.
Moreover, we have observed interesting behavior of temperature-driven p–n type conduction switch-
ing assisted colossal change in thermopower in some of these oxides, hitherto, obtained only in
chalcogenides. The charge transport mechanism in these complex oxides has been analyzed by small
polaron hopping conduction model in conjugation with defect chemistry.
Dr. Tanmoy Maiti is an Associate Professor in the Department of Material Science and Engineering at the Indian
Institute of Technology Kanpur, India. Dr. Maiti received his Ph.D. in Materials Science and Engineering from The
Pennsylvania State University, University Park, PA, USA, in December 2007. He earned his Masters Degree in
Ceramic Engineering from Institute of Technology, BHU in 2004 and Bachelors in Ceramic Technology from College
of Ceramic Technology, University of Calcutta in 2002. Prior to joining IIT Kanpur in 2010, he did his post-doctoral
research in Lawrence Berkeley National Laboratory and Pennsylvania State University. He specializes in electronic
materials and nanophotonics. Prof. Maiti has formed his research group in newly set-up Plasmonics and Perovskites
Laboratory at IIT Kanpur. His research interests span the areas of Thermoelectrics, Plasmonics, Photovoltaics and
Oxide electronic materials and devices. However, a common thread in his research is to address (i) the global energy
Tanmoy Maiti problem by developing novel materials for clean energy generation (e.g., thermoelectric power generator, perovskite
solar cell) and (ii) the design of next generation chip-scale technology based on nanophotonics and nano-electronics
(e.g., optical memory based on plasmonic lenses and RRAM).
I. INTRODUCTION transforms thermal energy (waste heat) into electrical

Energy is one of the most important core requirements energy using the Seebeck effect. Moreover, it is a solid-
for the progress of modern civilization. As of today, most state device without any moving parts, offering great
of our energy demand is fulfilled by the supply of fossil flexibility in operational integration.3,4 The performance
fuels. Burning the fossil fuels for energy generation has of a TE device is generally evaluated by a figure-of-merit
been shown to be one of the reasons behind global called ZT, which corresponds to the intrinsic properties of
warming and climate change. Interestingly, only one- the material as expressed by the following equation:
third of the primary energy input is utilized efficiently, S2 r S2 r
and two-third of it is wasted in the form of heat.1,2 Hence, ZT ¼ T¼ T ; ð1Þ
j je þ jl
it is desirable to develop an energy conversion device that
can produce electricity using the huge amount of waste where S is the Seebeck coefficient, r is the electric
heat generated from automotive exhausts, industrial pro- conductivity, j is the thermal conductivity, and T is the
cesses, and residential heating. Thermoelectric (TE) temperature. Thermal conductivity has a contribution
power generation offers a fascinating prospect as it from two parameters namely electrical thermal conduc-
tivity (je) and lattice thermal conductivity (jl). A good
a)
Address all correspondence to this author. TE material with a high ZT value requires a high Seebeck
e-mail: [email protected] coefficient (S), high electrical conductivity (r), and low
DOI: 10.1557/jmr.2018.376 lattice thermal conductivity (j). The primary obstacle to
J. Mater. Res., 2018 Ó Materials Research Society 2018. This is an Open Access article, distributed under the terms of the 1
Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0/), which permits
unrestricted re-use, distribution, and reproduction in any medium, provided the original work is properly cited.
T. Maiti et al.: Double perovskite (Sr2B9B0O6) oxides for high-temperature thermoelectric power generation—A review
increase ZT is that these fundamental properties of direction. We demonstrated that using glassy state of
materials are interconnected. As a result, enhancing one relaxor ferroelectrics above the Burns temperature can be
property deteriorates the others. instrumental in achieving very low thermal conductivity
Conventional TE materials are based on chalcogenides in these double perovskites to enhance the ZT values.
such as Bi2Te3,5 Sb2Te3,6 Zn4Sb3,7 Bi2Se3,8 and SnTe.9 Moreover, we discovered a new class of oxide materials
However, most of the waste heat generated from various exhibiting temperature driven p–n type resistance switch-
sources is .900 K and these materials decompose and ing behavior coupled with a large change in thermo-
oxidize at such high temperatures. Oxides are a better power, which can pave way for the fabrication of novel
alternative for high-temperature TE applications due to high temperature multifunctional devices.
their low costs, and higher oxidation resistance with
better thermal stability. II. DESIGN OF DOUBLE PEROVSKITES
Recently oxide TE materials10,11 such as NaxCoO2,
Ca3Co4O9, etc. showed good ZT values with p-type Double perovskites have stoichiometric formula
conductivity. However, their ZT values are not compara- A2B9B0O6, where the B-sites, i.e., B9, B0 can be occupied
ble with the conventional chalcogenides. Besides, there by the different combination of transition metals.42,43 The
are not many options available among oxide-based crystal structure and space group of double perovskites
materials, to choose as n-type element in the TE module. mainly depend on the arrangement of B9O6 and B0O6
It is beneficial to have both n-type and p-type oxides with octahedra, especially their ordering and tilting.44 There
good ZT values to reduce the strain mismatch in the TE are three types of B-site cationic arrangements in the 3-
module in which n-type and p-type materials are con- dimensional crystal structure, i.e., random, rock salt, and
nected thermally in parallel and electrically in series.12 In layered arrangement.43 The space-group of double per-
recent years, Nb and La-doped SrTiO3 perovskite-based ovskites with randomly ordered octahedra is generally
n-type materials showed very promising high- Pm3m, as shown schematically in Fig. 1(a). In the unit
temperature TE properties, but their ZT values need to cell of a randomly ordered (Pm3m) double perovskite, the
be improved to be considered for commercialization. corner positions are occupied by A-site (0, 0, 0) divalent
In the present work, we hypothesize that SrTiO3-based cations. However, B-site tetravalent cations (B9 and B0)
double perovskites (Sr2TiB9O6) are environmentally are placed in the body centered position (1/2, 1/2, 1/2)
friendly high-temperature TE materials. Double perov- and oxygen anions sit at the face-centered positions (1/2,
skite materials were extensively investigated over the 1/2, 0). On the contrary, in double perovskite structure
years for their wide range of exciting properties,36–40 e.g., with rock salt ordering, B9O6 and B0O6 octahedra are
colossal magneto-resistance, ferroelectricity, multiferroic- alternatively arranged in three-dimension space resulting
ity, high Tc superconductivity, multi-band Mott insula- different space groups, such as Fm3m, I4/m, and P4/nmc.
tors, and half-metallic behavior. A2B2O6 double The least common crystal structure of the double perov-
perovskites are very amenable to compositional modifi- skite is layered-type in which, the layers of B9O6 and
cation, whereby using different types of B site cations, B0O6 octahedra are alternately arranged only in one
their electrical, magnetic, and thermal properties can be direction.
tailored. Furthermore, these materials have complex Over the years, various types of perovskite materials
crystal structure capable of accommodating multiple were designed based on the Goldschmidt tolerance
cations, which can act as phonon scattering centers factor.45 Tolerance factor “t” is nearly 1 for cubic crystal
required for low lattice thermal conductivity. Although structure. If the A-site cations have a smaller radius
some papers reported13–15,17,18 on TE properties of resulting in t , 1, cation size mismatch is generally
double perovskites as presented in Table I, most of these compensated by tilting the BO6 octahedra. However, if
materials were primarily designed for investigating mag- t . 1, due to the larger radius of A-site ion, the mismatch
netoresistance, mutiferroics properties. is not offset by octahedral tilting, and a hexagonal phase
Recently, we published a series of papers on SrTiO3- may form. The equation for tolerance factor43 was further
based double perovskites for high-temperature TE appli- modified for our double perovskite structure,
cation. We showed that by selecting appropriate elements Ax9A2xB9B0O6, where A9-cation was used as the dopant
for A-site and B-site cations in A2B2O6 double perovskite in the A-site. The modified equation for tolerance factor
and manipulating the dopant concentrations along with is given by the following equation:
optimized processing parameters,24,25 one can achieve ð xÞrA0 þð2xÞrA
very large thermopower,19,22,41 metal-like electrical con- þ rO
t ¼ pffiffiffirB20 þrB00 ; ð2Þ
ductivity,24–26 and glass-like thermal conductivity.22 In 2 2 þ rO
this article, we present a comprehensive review on the
recent progress of Sr2B9B0O6-based double perovskites where, rA ; rA0 ; rB0 ; rB00 ; and rO are the ionic radii of
for high-temperature TE applications and the future ions.
2 J. Mater. Res., 2018

TABLE I. TE properties such as electrical conductivity, Seebeck coefficient, power factor, and ZT of different oxide TE materials.
Crystal structure/ Electrical conductivity Seebeck coefficient Power factor (S2rmax)

Composition space group (rmax)(S/m) (Smax)(lV/K) (lW/mK2) ZTmax Ref.
Double perovskites
Sr2YRuO6 Monoclinic (P21/n) 4.5 (1150 K) 475 (300 K) ;0.27 (1150 K) ... 13
Sr2ErRuO6 Monoclinic (P21/n) 9.09 (1150 K) 400 (300 K) ;0.56 (1150 K) 0.025 13
(1250 K)
La2CoRuO6 Monoclinic (P21/n) 10 (300 K) 290 (400 K) 2.5 (1073 K) ... 14
La2CrRhO6 ... 1.7 (300 K) 643.2 (300 K) 0.70 (300 K) ... 15
La2CuRhO6 ... 1.7 103 (300 K) 149.5 (300 K) 39.4 (300 K) ... 15
Sr2CoTiO6 Tetragonal (I4/m) 1.3 103 (720 K) 120 (300 K) 4 (1230 K) ... 16
La0.2Sr1.8CoTiO6 Tetragonal (I4/m) 2.6 103 (720 K) 70 (300 K) 3 (1230 K) ... 16
Ca2FeMoO6 Monoclinic (P21/n) ;105 (300 K) ;110 (1200 K) 350 (1200 K) 0.15 (1250 17
K)
Sr2MnMoO6 Monoclinic (P21/n) 183 (1220 K) 368 (1240 K) 120 (1250 K) 0.13 (1100 18
K)
Ba0.15Sr1.85FeTiO6
Cubic with (Pm3m) 273 (900 K) ;680 (1123 K) 26 (1123 K) ... 19
Ba0.15Sr1.85Ti0.8Fe0.8Nb0.4O6 Cubic (Pm3m) ;150 (930 K) 478 (935 K) 13 (930 K) ... 20
La0.1Sr0.9FeTiO6 Cubic (Pm3m) 510 (900 K) 1282 (1188 K) 89.5 (1139) ... 21
Ba0.2Sr1.8TiCoO6 Cubic (Pm3m) 790 (1220 K) 328 (1220 K) 85 (1220 K) 0.3 (1220 22
K)
Bi0.1Sr1.9TiCoO6 Cubic (Pm3m) 3.8 103 (714 K) 106.5 (333 K) 5.7 (761 K) ... 23
Sr2TiMoO6
Cubic (Pm3m) 1.3 105 (300 K) 7.73 (1220 K) 7.8 (1220 K) ... 24
La0.25Sr1.75TiMoO6 x 5 Cubic (Pm3m) 1.6 105 (500 K) 4.44 (1220 K) 4.49 (700 K) ... 25
0.25
Ba2TiMoO6 Cubic (Pm3m) 1.17 105 (700 K) 15 (1220 K) 20.79 (1220 K) ... 26
Ba0.1Sr1.9TiFe0.5Mo0.5O6 Cubic (Pm3m) 6.4 103 (1220 K) 73 (1220 K) 34.2 (1220 K) ... 27
Sr2TiCo0.5Mo0.5O6 Cubic (Pm3m) 3.7 103 (1220 K) 40.6 (1220 K) 6.1 (1220 K) ... 28
Sr2CoMoO6 (after Tetragonal (I4/m) 2.3 103 (;473 K) 210 (1173 K) ;45 (1173 K) ... 29
annealing)
Perovskite
SrTi0.85Nb0.15O3 Cubic ; 900 (1220 K) ; 195 (1220 K) 35 (1220 K) ... 30
La doped SrTiO3 (single ... 105 (300 K) ; 294 (1050 K) ;7 102 (1050 K) 0.27 (1073 31
crystal) K)
Nb doped SrTiO3 (single ... 3.5 104 (300 K) ; 370 (1050 K) ;4 102 (1050 K)- 0.17 (1073 31
crystal) K)
Ca0.8Lu0.2MnO3 Orthorhombic 1.6 104 (1100 K) 122 (973 K) 2.6 102 (973 K) 0.16 (973 32
K)
La0.6K0.4TiO3 Cubic (Pm3m) 3.2 104 (520 K) 150 (850 K) 6 102 (720 K) 0.18 (800 33
K)
Nonperovskite oxides
Ca2.8Lu0.2Co4O91ӧ Layered cobaltite 1 104 (1200 K) 195 (1045 K) 4 102 (1070 K) 0.35 (1070 34
K)
BaAgCa2.8Co4O9 Layered cobaltite 1.3 104 (1000 K) 171 (1000 K) 3.8 102 (1000 K) 0.3 (1000 35
K)
To design the materials’ composition and dopant SrTiO3-based double perovskites are stable with a cubic
percentage, the tolerance factor [Eq. (2)] was calculated crystal structure.
for Sr2TiMO6 double perovskites, where M 5 Fe, Co,
and Mo. Tolerance factors with respect to the percentages
of ‘A’-site doping for these double perovskite systems are III. PROCESSING OF DOUBLE PEROVSKITES
plotted in Fig. 1(b). It is evident from the figure that pure In general, double perovskites are electrical insulators.
Sr2TiCoO6 (STC) shows higher tolerance factor than However, in the TE applications, very good electrical
Sr2TiMoO6 (STM) and Sr2TiFeO6 (STF) ceramics. Tol- conductivity of the material is desired. To engineer
erance factor ‘t’ was found to increase with the sub- double perovskite oxides with reasonable electrical con-
stitution of Ba for the Sr-site in all the double perovskites ductivity, it is essential to attain a single phase solid
such as STC, STF, and STM, whereas ‘t’ decreased with solution. The presence of impurities or secondary phase
the La doping. Nevertheless, the tolerance factor values deteriorates the electrical conductivity. We used the
for all the compositions developed in the present work solid-state reaction method in the present work to prepare
found to be within the range of 1 6 0.03. It predicts that solid solution of double perovskites. This method
J. Mater. Res., 2018 3

FIG. 1. (a) Schematic of A2B9B0O6 double perovskite structure and (b) tolerance factor of Ba and La-doped Sr2TiCoO6 (STC), Sr2TiFeO6 (STF),
and Sr2TiMoO6 (STM).
includes calcination of a homogeneous mixture of differ- temperature. The XRD pattern of calcined and sintered
ent powders such as oxides, carbonates, etc. This solid- powder showed that solid state reactions were not
state reaction process contains repeated grinding, milling, completed in air. We found a secondary phase, SrMoO4,
and sintering at high temperature. During sintering and as confirmed by the XRD profile in Fig. 3(a). SrMoO4
calcination, various parameters such as temperature, phase segregation suggests that Mo is more stable with
heating/cooling rate, gas flow, etc. were optimized to 16 oxidation state as depicted by Eq. (3). The SrMoO4
acquire the required resultant phase of ceramic materials. phase exhibits insulating behavior and significantly
There was another important step that we included in the reduces the overall conductivity of the system. How-
current work, i.e., nanomilling of calcined powder. After ever, Mo has a formal valence of (14) in the double
nanomilling, the particle size of the starting powder for perovskite structure. To troubleshoot this problem,
sintering was reduced to ,100 nm. Smaller particle size repeated process of grinding followed by calcination
is important to achieve the small grain size in the was carried out in reducing atmosphere as described by
microstructure of the sample, which can enhance the Eq. (4). Single-phase solid solution was achieved when
phonon scattering at grain boundaries resulting in low calcination and sintering were carried out in reducing
lattice thermal conductivity (jl) of the ceramic materials. atmosphere, as shown in Fig. 3(a). To the best of our
The calcined powder was milled by planetary micromill knowledge, there was no report on STM-based double
(Fritsch, PULVERISETTE 7 premium line, Rhineland perovskites demonstrating single-phase solid solution.
Palatinate, Germany) at 600 rpm using zirconia grinding Therefore, the sintering temperature, as well as the
balls and ethanol as milling media. The schematic of the atmosphere, played significant role in the fabrication of
processing steps involved in the fabrication of double these double perovskites.
perovskite ceramics by the solid-state reaction method is
in air
presented in Fig. 2. 2SrO þ MoO3 þ TiO2 ! SrTiO3 þ SrMoO4 ; ð3Þ
A. Effect of processing atmosphere in 10% H2 þAr 1

2SrO þ MoO3 þ TiO2 ! Sr2 TiMoO6 þ O2 [ :
To obtain good electrical conductivity, we used tran- 2
sition metals with multiple oxidation states in the B-site ð4Þ
of the double perovskite, which has the formal valence of
14. Hence, unwanted secondary phases might form in
the system. In calcination and sintering, the processing To determine the crystal structure of double perovskite
atmosphere, i.e., inert (Ar, N2), oxidizing (Air) and oxides, Rietveld refinement of room temperature powder
reducing (H2) play a significant role in the formation of XRD data was carried out using Full Prof Program46 for
a single phase. Initially, we calcined and sintered La- and all the double perovskite samples. We obtained a single-
Ba-doped Sr2TiMoO6 in air. To analyze the phase purity phase cubic symmetry for all the SrTiMO6 double
and crystal structure of calcined and sintered sample, X- perovskites, where M 5 Co, Mo, and Fe. The observed
ray diffraction (XRD) was carried out by using a PAN- (open circle), calculated (continuous line), and the
analytical Empyrean ACMS 101 (Malvern Panalytical difference profiles obtained after refinement of the XRD
Ltd., Malvern, United Kingdom) diffractometer at room pattern of LSTM x 5 0.1 are shown in Fig. 3(b). The

FIG. 2. Schematic of processing double perovskite ceramics by the solid-state reaction method and its TE measurement.
XRD data show a well-fitted graph with respect to Pm3m SrMoO4 phase of the air sintered sample. The Sr2Co-
space group with very low-value of v2 (1.41), which MoO6 and SrMoO4 were fitted with I4/m and I4/a space
signifies the goodness of fit in the Rietveld refinement.25 group, respectively. However, the amount of this
The microstructure of all the ceramic samples (;95– SrMoO4 secondary phase was found to be 8.9 wt%. To
97% density) was analyzed by scanning electron micros- reduce the amount of the SrMoO4 phase, heat treatment
copy (SEM) (Carl Zeiss GmbH, EV050, Oberkochen, was carried out in reducing atmosphere at 1273 K for 2 h.
Germany). All the double perovskite ceramic samples After annealing, there was an absence of SrMoO4 phase,
showed well-sintered, dense microstructure, as shown in as shown in Fig. 4(b). However, the extra peaks corre-
Fig. 3(c) for LSTM.25 These double perovskites demon- sponding to the SrMoO3 perovskite phase was also
strated a wide range of grain size distribution scaling recorded. The annealing process not only changed the
from nm to lm range. Due to nanoscale high energy ball phase of the system but also showed the effect on the
milling of calcined powder, we obtained a significant microstructure of SCMO ceramic. Before annealing, all
amount of grain size ,500 nm reduction. the grains were distinguished with clearly defined grain
boundaries, as shown in Fig. 4(c). However, the micro-
structure of SCMO after annealing demonstrated molten
B. Effect of annealing phase with micron-scale grain.29
We studied the effect of annealing on the TE properties Annealing of SCMO ceramic in reducing atmosphere
of Sr2CoMoO6-based double perovskite oxide.29 Figure 4 affected its TE properties significantly. Air-sintered
shows the structural and TE properties of as-sintered SCMO ceramic exhibited insulator like behavior up to
Sr2CoMoO6 (SCMO) oxide in comparison with SCMO 913 K (;0.2 S/m at room temperature) with p-type
sample annealed in reducing atmosphere. XRD of air- Seebeck coefficient, as shown in Fig. 4(e). Around
sintered SCMO implied the signature of a small concen- 913 K, a sharp decrease in the Seebeck coefficient (S)
tration of nonperovskite phase SrMoO4, as shown in was found and SCMO showed p–n type transition depicted
Fig. 4(a). Furthermore, we carried out the Rietveld in Fig. 4(e). After annealing, SCMO demonstrated n-type
refinement using Fullprof program to reveal the crystal conduction behavior in the entire temperature range, as
structure and weight percentage of Sr2CoMoO6 and shown in Fig. 4(f). Furthermore, electrical conductivity was

FIG. 3. (a) XRD pattern of STM powder calcined and sintered sample in the air and reducing atmospheres, (b) Rietveld analysis, and (c) SEM image of
the fracture surface of the LSTM x 5 0.15 ceramic sample. Reproduced from Ref. 25 with permission from The Royal Society of Chemistry.
enhanced by ten times after annealing. With increasing as expressed by defect reactions (5) and (6). These
temperature, electrical conductivity was decreased indicat- oxygen vacancies are the sources of the positive charge
ing the metal-like behavior. The maximum power factor carriers in the SCMO sample. However, the presence of
(S2r) value, ;45 lW/mK2, was achieved at 1173 K, which Mo61 in SCMO donates negative charge carrier, electron
is significantly higher as compared to air-sintered SCMO as depicted by Eq. (7). Hence the presence of both low
ceramics (;13 lW/mK2 at 1173 K).29 Hence, annealing mobility oxygen vacancies and high mobility electrons
was found to be an interesting way to enhance the electrical are expected to exist in the SCMO system. The p-type
conductivity of any n-type TE materials and to improve the conduction observed till 1023 K in the air-sintered
overall performance. The conduction mechanism of air- sample suggests the presence of positively charged
sintered and annealed sample was further explained by the oxygen vacancies as majority charge carriers. The air-
following defect reactions: sintered sample showed the presence of SrMoO4 as the
secondary phase, which was reported as insulating in
2BO2
Co2 O3 ! 2Co0 B þ 3O nature.47 It validates the low electrical conductivity
O þ VO ; ð5Þ
observed in the as-sintered sample. Moreover, at low
BO2
temperature, electrons produced by defect reaction (7) are
CoO ! Co00 B þ O
O þ VO ; ð6Þ expected to be localized at defect sites in the ceramic
sample. To overcome the energy barrier, the charge
BO2 1 carriers (e) required the activation energy for delocal-
MoO3 ! Mo 0
B þ 2OO þ 2e þ O2 [ : ð7Þ ization. Small polaron hopping phenomena were found29
2
to be the conductivity mechanism in SCMO. With
The B-site cation in double perovskite oxides pos- increasing temperature, these electrons were expected to
sesses formal valance of 41. Hence, the B-site occu- become delocalized contributing to the conductivity. The
pancy by Co31 and Co21 produces the oxygen vacancies n-type conductivity observed in SCMO above 1023 K

FIG. 4. (a) and (b) XRD, (c) and (d) SEM, (e) and (f) TE properties for SCMO ceramics before and after annealing, respectively. Reproduced from
Ref. 29 with the permission of AIP Publishing.
suggests the high mobility electrons as majority charge the dominant charge carriers. The Seebeck coefficient of
carriers, which further corroborates the higher conduc- STF was found to be higher than STC composition.
tivity values obtained beyond 1023 K. On the other hand, Furthermore, we studied the effect of barium doping at
the SrMoO3 phase found in the annealed sample was A-site on TE properties of STC and STF. Doping of
reported to be metallic in nature.48 As a result the barium in the place of strontium increased the Seebeck
annealed sample showed higher conductivity values as coefficient of both STC and STF. Very large S value
well as metal-like (dr/dT , 0) behavior in the ;800 lV/K at 1100 K was obtained in the Ba0.25Sr1.75-
temperature-dependent conductivity plot, as shown in TiFeO6 double perovskite. In the temperature-dependent
Fig. 4(f) unlike the air-sintered SCMO sample. conductivity curve, both STC and STF showed semi-
conductor (dr/dT . 0) to metal-like (dr/dT , 0)
transition as evident in Figs. 5(a) and 5(d). Furthermore,
IV. LARGE THERMOPOWER IN DOUBLE the doping of Ba21 at the Sr-site decreased the value of
PEROVSKITES electrical conductivity in both STC and STF ceramics.
For the potential TE materials, it is essential to have Maximum conductivity value for STC was found to be
a high thermopower (S . 100 lV/K) in combination with higher than that of STF. The highest value of power
a high value of electrical conductivity resulting in high factor (S2r) was found to be ;100 lW/mK2 at 1200 K
TE power factor (S2r). In the present work, we engi- for BaxSr2xTiCoO6 with x 5 0.15 composition. In the
neered the SrTiO3-based double perovskite TE materials, pure STC sample, two semiconductors to metal transition
Sr2TiB9O6, by using ferromagnetic elements (Co and Fe) peaks were found at 700 K and 1170 K, respectively,
at the B9-site, which resulted in double perovskites such unlike the STF sample as shown in Fig. 5(a). The first
as Sr2CoTiO6 (STC)22 and Sr2FeTiO6 (STF).19 Figure 5 semiconductor to metal transition temperature (TS–M) was
shows the Seebeck coefficient and the electrical conduc- found to be shifted to the higher temperature with the
tivity of these materials measured in the temperature incorporation of Ba in the Sr-site. There could be a phase
range from 300 to 1200 K (ZEM-3 M10 apparatus, transition responsible for semiconductor to metal transi-
ADVANCE RIKO Inc., Yokohama, Japan). Both STF tion observed in STC. However, our laboratory scale
and STC compositions demonstrated positive Seebeck high-temperature XRD did not show any peak splitting
coefficient confirming the p-type behavior with holes as suggesting cubic phase in the whole temperature range of

FIG. 5. TE properties such as (a) and (b) electrical conductivity, (c) and (d) Seebeck coefficient and (e) and (f) power factor of STC and STF-based
TE materials. (a-c) reproduced from Ref. 22 with the permission of AIP Publishing and (d-f) reproduced from Ref. 19 with permission from The
Royal Society of Chemistry.
measurement. High-resolution XRD or neutron diffrac- However, the doping in A-site and B-site of double
tion is required to unravel the possible phase transition in perovskite with metals having much higher oxidation
the STC sample. It is to be noted that these materials state is expected to introduce e-s in these oxides. As a
exhibited relaxor ferroelectricity as discussed in the result, these double perovskites give rise to some
subsequent section of this manuscript. Hence it is difficult new exciting behavior as discussed in a subsequent
to solve the crystal structure of BSTC since the metal– section.
semiconductor transition temperatures belonged to the
paraelectric region, which is known to possess the cubic V. DECOUPLING OF ‘PHONON-GLASS’ AND
crystal structure. Hence one may need to resolve the local ‘ELECTRON-CRYSTAL’ BEHAVIOR IN OXIDES
structural ordering in this system. USING RELAXOR FERROELECTRIC INDUCED
Majority charge carriers (holes, h ) in STC and STF-
d
DIPOLAR SPIN GLASS
based double perovskites can be explained by the charge
compensation mechanism due to the substitution of B-site To achieve good ZT values in TE materials, it is
having formal charge 14 with multivalent cations such as essential to have high electrical conductivity and low
Fe and Co. Defect reactions can be expressed by using thermal conductivity. It means that we need materials
Krӧger–Vink notation. where electron transport in the medium should be as
smooth as possible, but on the contrary, heat transport in
1 2TiO2 the material should be as hard as possible. Slack
Co2 O3 þ O2 ! 2Co0 Ti þ 2h þ 4O
O ; ð8Þ summarized well this phenomenon by proposing “pho-
2
non glass electron crystal” model.49 To determine the
1 TiO2 heat transport in the BSTC double perovskite, thermal
CoO þ O2 ! Co00 Ti þ 2h þ 2O
O ; ð9Þ diffusivity (D) and heat capacity (Cp) were measured in
2
the temperature range from 300 to 1273 K. The formula
1 2TiO2 j 5 qCpD was used to calculate the thermal conductivity
Fe2 O3 þ O2 ! 2Fe0 Ti þ 2h þ 4O
O : ð10Þ of these materials, where q is the density of samples.
2
Thermal conductivity (j) of all BSTC samples varies
Due to the substitution of isovalent Ba21 in the place from 0.01 to 0.55 W/mK22, as shown in Fig. 6(a). Lattice
of Sr21 ions, no changes in charge carrier dynamics thermal conductivity (jl) of the system was calculated by
was expected, which explains p-type behavior ob- subtracting electronic thermal conductivity (je) from total
served in all the BSTC and BSTF compositions. thermal conductivity (j). Electronic thermal conductivity

(je) was calculated by using Wiedemann–Franz law, je Furthermore, we studied the dielectric behavior of
5 LTr, where L is the Lorentz number, r is the electrical BSTC ceramics to check if the material is relaxor indeed.
conductivity, and T is the temperature. Figure 6(b) shows The temperature dependence of dielectric constant (e9)
the lattice thermal conductivity of BSTC ceramics. and loss tangent (tan d) at various frequencies for all the
However, electronic thermal conductivity (je) was found BSTC compositions in the temperature range from 100 to
to be very small in this BSTC system. Thus, thermal 350 K is shown in Fig. 6(d). These materials showed
conductivity of BSTC ceramics is primarily controlled by relaxor behavior at low temperature, i.e., with increasing
phonon transport (lattice thermal conductivity). Such frequency dielectric permittivity, e9decreased and the
glass-like low thermal conductivity is very beneficial to temperature of dielectric maxima was shifted toward
increase the figure of merit (ZT) and the performance of higher temperatures. Also, these BSTC compositions
the TE device. The maximum value of ZT was obtained were found to follow the modified Curie–Weiss law as
as ;0.3 at 1220 K in BSTC with x 5 0.2 composition, as described by the following equation55:
shown in Fig. 6(c).
However, the origin of such glass-like thermal con- 1 1 ðT Tm Þc
¼ ; ð11Þ
ductivity in these BSTC double perovskites could not be e em C1
explained by just considering phonon scattering due to
small grain size and multiple cation sites. Moreover, the where c is called a degree of relaxation and its values lie
thermal conductivity of the BSTC sample was found to between 1 and 2. The value c 5 1 shows that the Curie–
be continuously increasing with temperature till 1200 K, Weiss law is valid for normal ferroelectric behavior.
as shown in Fig. 6. This is opposite to what was observed However, c 5 2 expresses the case of quadratic de-
for other TE materials, which showed a decrease in pendence suitable for an ideal relaxor ferroelectric.56 The
thermal conductivity with increase in tempera- c value obtained for this BSTC composition is 1.54,
ture.13,17,50,51 However, the similar behavior of increas- which suggests the ferroelectric relaxation behavior in
ing thermal conductivity with temperature was found in these ceramics. Relaxor ferroelectrics are known to
the literature for the spin glass system.52,53 Interestingly, behave like dipolar glass57,58 where dipolar interactions
such thermal conductivity behavior was also reported in of polar nanoregions (PNR) are thought to be responsible
the relaxor ferroelectric material, which is known to for relaxor behavior. However, at very high temperature,
behave like dipolar spin glass.54 thermal fluctuations are so large that there are no such
FIG. 6. Temperature dependence (a) thermal conductivity, (b) lattice thermal conductivity, (c) ZT of BaxSr2xTiCoO6 ceramics with x 5 0.0, 0.1,
0.15, and 0.2, (d) dielectric constant (e) and tan d of Ba0.2Sr1.8TiCoO6 ceramics and the inset shows the Vogel–Fulcher fitting. Reproduced from
Ref. 22 with the permission of AIP Publishing.

well-defined dipole moments and it exhibits glass-like ‘Phonon-Glass’ and ‘Electron Crystal’ behavior in
behavior above Burns temperature (TD).59–61 The Burns oxides, which was considered as the bottleneck in the
temperature (TD) value for BSTC with x 5 0.2 compo- research of oxide TEs. Here we showed that using relaxor
sition was calculated to be 312 K at 100 kHz, which ferroelectrics above Burns temperature can produce
suggests that BSTC double perovskite becomes a glassy glass-like thermal conductivity without affecting electri-
state above 312 K. The relaxation time (s) for these cal conductivity much. To the best of our knowledge, this
relaxors62–64 can be estimated by Vogel–Fulcher law, is the first material to demonstrate that decoupling of
which was originally derived for spin glass as described ‘Phonon Glass’ and ‘Electron Crystal’ behavior is possi-
by the following equation: ble in oxides.
Furthermore, we compared the thermal conductivity of
s ¼ s0 exp½Ea =kB ðTm TVF Þ ; ð12Þ BSTC double perovskites with wide range of oxides such
as double perovskites, single perovskites, layered oxide-
where s0 is the pre-exponential factor, Ea is the activation like Ruddlesden–Popper phases and nonperovskite
energy for polarization fluctuations of an isolated cluster, oxides, as summarized in Table II. It is evident that the
kB is the Boltzmann constant, Tm is the temperature thermal conductivity of the BSTC samples was found to
corresponding to dielectric maxima, and TVF is the be comparable with some double perovskites. However,
characteristic Vogel Fulcher freezing temperature. The it demonstrated lower thermal conductivity compared to
temperature dependence of relaxation time of BSTC other layered perovskite and nonperovskites. The trend in
double perovskite was found to be well fitted with the temperature-dependent thermal conductivity of BSTC
Vogel Fulcher relation, as shown in the inset of Fig. 6(d) ceramics was found to be similar to that what was
for Ba0.2Sr1.8TiCoO6 composition. It further justifies that reported for the relaxors ferroelectrics and spin
these double perovskites behave like spin glass as a result glasses.54,65 However, we could not find any high-
of relaxor ferroelectricity. Moreover, these relaxors above temperature studies on spin glasses as their thermal
Burns temperature (TD) behave like glass, which is the conductivity was studied primarily for other purposes
reason behind such low glass-like thermal conductivity such as magnetoresistance. Although some nonperovskite
observed in these double perovskite. Inducing dipolar oxides showed a slightly increasing trend in thermal
glassy phase as a result of relaxor ferroelectricity in conductivity. However, there is no proper explanation for
double perovskites can be a good strategy for decoupling this behavior available in the literature. More theoretical
TABLE II. Thermal conductivity of different oxides.
Thermal conductivity (W/mK) at low Thermal conductivity (W/mK) at high

Composition Structure temperature temperature Ref.
Ba0.2Sr1.8TiCoO6 0.11 at 300 K 0.35 at 1220 K 22

Sr2RuYO6 ;0.44 at 300 K ;0.22 at 1270 K 13
Sr2RuErO6 ;0.30 at 300 K ;0.16 at 1270 K 13
Sr2FeMoO6 ;0.64 at 300 K ;0.20 at 1050 K 51
Double perovskite
Sr1.7Ba0.3FeMoO6 ;0.76 at 300 K ;0.20 at 1050 K 51
Ca2FeMoO6 ;4.1 at 300 K ;3.10 at 1020 K 17
Ca1.7Sr0.3FeMoO6 ;3.7 at 300 K ;2.2 at 1020 K 17
Pb2FeMoO6 ;1.4 at 300 K ... 66
Pr0.9Sr0.1Mn0.9Fe0.1O3 0.68 at 300 K ;0.94 at 1000 K 67
Ho0.9Ca0.1CoO3 ;0.85 at 350 K ;1.24 at 850 K 68
La0.5Na0.5Ti0.8Nb0.2O3 2.1 at 300 K 2.5 at 950 K 69
La0.6K0.4TiO3 2.45 at 300 K 2.36 at 950 K 33
Perovskite
Pb(Mg1/3Nb2/3)O3 (relaxor ferroelectric) ;0.5 at 80 K ... 54
(Pb0.915La0.085) (Zr0.65Ti0.35)O3 (relaxor ;0.1 at 21 K ... 54
ferroelectric)
Pb(Mn0.5Ta0.5)O3 (relaxor ferroelectric) ;0.08 at 15 K ... 54
Sr0.65Ba0.35Nb2O6 (relaxor ferroelectric) ;0.7 at 88 K1 ... 54
SrlxBax(Nb2O6) x 5 0.55 (relaxor ;0.8 at 70 K ... 65
ferroelectric) Nonperovskite oxide
Nd1xSr11xCoO4 x 5 0.33 (layred cobalt 1.5 at 300 K ... 70
oxide)
20%Nb doped SrO(SrTiO3) ;5.0 at 300 K ;2 at 1000 K 71
(Sr1xLax)n11TinO31n x 5 0.05, n 5 2 Ruddlesden-popper ;4.8 at 300 K ;2.58 at 1000 K 72
(Sr1xEux)3Ti2O7 x 5 0.05 phases ;4.20 at 300 K ;2.4 at 1000 K 73
SrO(SrTiO3)n n 5 1 ;4.45 at 300 K ;1.9 at 1000 K 74
10 J. Mater. Res., 2018

as well as experimental analyses are required to un- behavior. However, the p–n type switching temperature
derstand the heat transport behavior in these double was found to be decreased with increasing Ba concen-
perovskites. tration in BSTM composition in contrary to LSTM,
where the p–n transition temperature was increased due
to La doping, as shown in Fig. 7. In the n-type region,
VI. METAL-LIKE ELECTRICAL CONDUCTIVITY IN
Seebeck coefficient (S) for all the compositions was
OXIDES
found to be increased with temperature. Although these
Both STC and STF materials discussed above showed STM-based double perovskites showed metallic behavior
good Seebeck coefficient. However, these materials (dr/dT , 0) in the temperature-dependent conductivity
suffered from low electrical conductivity (;102 S/m) graph, they were not found to follow the degenerate
causing low power-factor obtained in these ceramics. Fermi gas model in the Seebeck coefficient due to
Increasing the electrical conductivity of oxides is a daunt- localized Anderson transition.75,76 Generally, these kinds
ing task because oxides are inherently insulator in nature. of perovskites with high carrier concentration
Recently, we developed Sr2TiMoO6 (STM)-based TE (.1018 cm3) are expected to show metal–insulator
materials24 which demonstrated metal-like electrical con- 1=3
transition, as predicted by Mott criteria, ne a0 ;0:25,
ductivity (;105 S/m). We also studied the effect of where ne is the electronic carrier concentration and a0 is
lanthanum and barium doping on AxSr2xTiMoO6 based the Bohr radius related with the carrier.77 As per the Mott
TE materials. Figure 7 shows TE properties like electrical criterion if the value exceeds 0.25, the conduction
conductivity (r), Seebeck coefficient (S), and power mechanism transforms from semiconductor to metallic.78
factor (S2r) of AxSr2xTiMoO6 (A 5 La, Ba) ceramic Since STM based double perovskites possess very high
samples in the temperature range from 300 to 1273 K. carrier concentration (1022 cm3), they are expected to
Very high carrier concentration on the order of 1022 cm3 exhibit metallic behavior satisfying Mott criterion. How-
was found in the Hall measurement of these double ever these STM-based compounds comprise transition
perovskites. The highest value of electrical conductivity, metals like Ti and Mo, which are known to exist in
1.64 105 S/m, at 517 K was obtained for La0.25Sr1.75- multiple oxidation states. As a result, there can be
TiMoO6 composition, as shown in Fig. 7(a). a significant variation in the local electric fields and
The Seebeck coefficient (S) of these materials showed lattice distortions causing random electric field around
temperature driven p-type to n-type conduction switching the defect sites, resulting the charge carriers in the system
FIG. 7. TE properties: (a) electrical conductivity and (b) Seebeck coefficient of La- and Ba-doped STM ceramics with curve fitting and XPS
analysis of (c) Mo and (d) Ti in LSTM with x 5 0.25.
J. Mater. Res., 2018 11

localized, which is known as Anderson localization.75,76 Furthermore, we explained this p–n switching ob-
As a result, these compounds do not follow the degenerate served in the temperature-dependent Seebeck coefficient
semiconductor model. To understand the charge transport using an analytical model. In the case of coexisting
mechanism in these complex oxides, it is essential to know electrons and holes, the total Seebeck coefficient ‘S’
the source of charge careers in these materials. was considered to be the sum of contribution from both
From the temperature-dependent Seebeck coefficient types of charge carrier as given below.
measurement of these samples showing p–n type switch-
ing, it is imperative to have the coexistence of both S ¼ ðSþ rþ þ S r Þ=r : ð17Þ
positive and negative charge carriers in these compounds.
Source of negative charge carriers can be explained by Furthermore, the following equation can be derived:
defect reaction (13) and (14) which suggests that La13
inclusion in the place of Sr12 and Mo16 in Ti14 site are S ¼ aT þ ðbDEh =kT þ cÞ expðDEh =kT Þ ; ð18Þ
the reason behind the generation of electrons. Besides, 2 2
the electron generation due to oxygen vacancies is p k 1
where; a ¼ ;
inherent defect reaction in oxides especially at high 3 e EF EC
temperature as depicted by Eq. (10). However, both Mo
and Ti present in STM double perovskites can exist at 13 r0 k
oxidation state as well. The cause of positive charge b¼ ;
Ae
carriers in these materials was considered to be 13
oxidation state of Mo and Ti occupying B-site of the r0 k
double perovskite, which has formal valence 14 as c¼B ;
Ae
expressed by Eq. (16).
where DEh is the activation energy for generation of
2SrO 1 positive charge carriers, r0 and B are constant; kB is the
La2 O3 ! 2LaSr 0
þ 2e þ 2O
O þ O2 [ ; ð13Þ Boltzmann constant, e is a charge of the electron, T is the
2
temperature, EF is the Fermi energy, and EC is the energy
TiO2 1 at the bottom of the conduction band. Detailed derivation
MoO3 ! Mo 0
Ti þ 2e þ 2OO þ O2 [ ; ð14Þ of the proposed analytical model can be found in our
2
earlier publication.25 Figure 7(b) shows a least fitting
TiO2 1 curve for the temperature dependence of the Seebeck
O
O ! O2 [ þ VO þ 2e
0
; ð15Þ coefficient of STM-based ceramic samples by using Eq.
2
(18). The proper fit to experimental data with our model
2TiO2 [Eq. (18)] further validates that these samples have
M2 O3 ! 2M0 Ti þ V
O þ 3OO ; ð16Þ coexistence of high mobility electrons and low mobility
oxygen vacancies. Nevertheless, these STM-based ce-
where M 5 Mo, Ti. ramic samples demonstrated that metal-like electrical
Above defect reactions suggest that STM-based double conductivity (;105 S/m) is possible to attain in ceramics
perovskites can have the presence of electrons as well as in the wide range of temperature from 300 to 1223 K.
positively charged oxygen vacancies. However, oxygen
vacancies are known to have lower mobility as compared
to electrons. Furthermore, the multiple oxidation states of VII. OPTIMIZATION OF POWER FACTOR BY
Mo and Ti cations was confirmed by X-ray photoelectron COMPOSITIONAL MODIFICATION
spectroscopy (XPS; PHI 5000 Versaprobe II, ULVAC- The large Seebeck coefficient obtained in STC and
PHI, Inc., Chigasaki, Japan) as shown in Figs. 7(c) and 7 STF-based double perovskites. However, these materials
(d) for Mo 3d and Ti 2p spectra of LSTM x 5 0.25. The suffered from low electrical conductivity. On the other
XPS spectra confirmed the existence of Mo and Ti with hand, we observed metal-like electrical conductivity
multiple oxidation states. The obtained values of binding (105 S/m) in Sr2TiMoO6-based double perovskite.
energies are very similar to the other reports in the However, their Seebeck coefficient is not very promising
literature.79–82 XPS data suggest that the existence of to be considered as good TE material.
Mo in 16 and 15 states enabled to reduce Ti even into Hence we further engineered a recipe of designing
12 state, although in a little fraction of total Ti in the novel high-temperature TE materials by combining these
system. Mo31, Ti31, and Ti21 were attributed to being two different types of double perovskites to use their
the reason behind the positively charged oxygen vacan- respective good property coefficients, similar to what was
cies. On the other hand, higher valence cations such as used in composites over the years. However, in our case,
Mo51 and Mo61 generated the electrons. two chosen materials have the same crystal structure and
12 J. Mater. Res., 2018

similar lattice constant is resulting single-phase solid- delocalize these carriers resulting in effective charge trans-
solution. We investigated the TE properties of Sr2Ti- port in the double perovskite. To explain the conduction
Fe0.5Mo0.5O6 (STFM), which can be considered as the mechanism, small polaron hopping conduction seems more
combination of Sr2TiFoO6 having large Seebeck coeffi- suitable for these systems.83–85 Equation (19) expressed the
cient and Sr2TiMoO6 double perovskite, which exhibited small polaron hopping conduction mechanism.86
metal-like electrical conductivity. Furthermore, we also
synthesized Sr2TiCo0.5Mo0.5O6 (STCM), which can be r0 EHop
r¼ exp ; ð19Þ
regarded as the combination of Sr2TiCoO6 and Sr2Ti- T kB T
MoO6 double perovskites.28 These engineered double
perovskite materials showed higher electrical conductiv- where EHop is the activation energy of polarons, T is the
ity as compared to Sr2TiFeO6 and Sr2TiCoO6. Besides, temperature, and kB is the Boltzmann constant. If the
they demonstrated a negative Seebeck coefficient imply- small polaron hopping conduction is followed, then there
ing n-type TEs. Furthermore, we modified these materials should be a linear relation between ln (rT) and 1/kBT.
by doping of Ba and La at Sr-site. Figure 8 shows the TE Double perovskite oxides followed the small polaron
properties such as electrical conductivity, Seebeck co- hopping conduction mechanism in the semiconductor
efficient and power factor of LaxSr2xTiCo0.5Mo0.5O6 region of the conductivity curve as evident from the
(LSTCM) and BaxSr2TiFe0.5Mo0.5O6 (BSTFM) good linear fit shown in Fig. 9 for various double
ceramics. It can be seen from Fig. 8 that doping of La perovskite systems. Activation energy ‘EHop’ is equal to
in the place of Sr in STCM increased the electrical the slope of the curve between ln (rT) and 1/kBT.
conductivity but decreased the Seebeck coefficient. On Table III presents the calculated EHop values for double
the other hand, Ba-doping increased the Seebeck co- perovskite oxides.
efficient at the expense of the conductivity values in Thermopower is fundamentally a transport entropy of
STFM. It shows that trade-off between the Seebeck charge carrier that discloses the further information
coefficient and electrical conductivity can be done to regarding charge transport mechanism. For these perov-
achieve optimum power factor in these double perovskite oxides in which small polaron hopping model drives
skites by manipulating simply the doping concentration. conduction mechanism, thermopower can be explained
by Heikes formula. Chaikin and Beni modified the
formula to address the spin interactions of the electron
VIII. CHARGE TRANSPORT MECHANISM IN
better as given by the following equation87:
DOUBLE PEROVSKITES

In these types of transition metal-based perovskites, band kB 2c
S ¼ ln ; ð20Þ
model may not be a proper way to understand the e c
conduction mechanism. In these materials, charge carriers
are expected to be localized at defect sites. So activation where ‘c’ is the fractional small polaron concentration
energy should be required to overcome the energy barrier to arising from multivalent cations occupying similar lattice
FIG. 8. TE properties: (a) electrical conductivity, (b) Seebeck coefficient, and (c) power factor of La and Ba doped STCM and STFM-based
ceramics.
J. Mater. Res., 2018 13

sites. Estimated values of c for double perovskite oxides further validates the proposed model for charge transport
are given in Table II. It is evident that as the activation mechanism in these double perovskites.
energy decreases and the fractional polaron concentra-
tion, ‘c’ values increase in these double perovskites, the
electrical conductivity of the material increases, which IX. COLOSSAL CHANGE IN THERMOPOWER
Understanding the relationships between the structural
and chemical compatibility, solid solubility, thermal
stability, and lattice strain in perovskite structure can
help in developing a broad range of novel functional
materials. In modern electronics, development of novel
inorganic semiconductor-based p–n junction has become
pivotal for application in switches,88 memory devices,89
transistors,90 sensors91 etc.
In the current investigation, in the course of searching
n-type oxide TE materials, we developed some novel
SrTiO3-based double perovskites, such as BaxSr2xTi0.8-
Fe0.8Nb0.4O6 (BSTFN), and LaxSr2xTiFeO6 (LSTF)
which exhibited metal–semiconductor transition induced
p–n type conduction switching with colossal change in
thermopower. These BSTFN and LSTF ceramics were
FIG. 9. Small polaron hopping (SPH) fitting of electrical conductivity synthesized by the solid-state method with different
of different double perovskite ceramic samples.
amounts of Ba and La in the Sr site of Sr2TiFeO6-based
double perovskites. Both the compositions, BSTFN and
TABLE III. The calculated value of EHop required for generation of LSTF, exhibited semiconductor (dr/dT . 0) to metal-
the charge carrier. like (dr/dT , 0) transition and temperature driven p–n
EHop Fractional small polaron Electrical type conduction switching with enormous change in
Compositions (eV) concentration (c) conductivity (S/m) thermopower, as shown in Fig. 10. Also, the variation
of thermopower of these double perovskites is depicted in
STF 0.393 0.26 400
BSTF x 5 0.396 0.23 347
Table IV.
0.1 The conduction switching of these compositions was
STC 0.226 0.71 2011 first tried to be understood by the source of charge
BSTC x 5 0.380 0.48 834 carriers. The source of positive charge carriers in Sr2Ti-
0.15 FeO6 (STF) was explained by defect reaction expressed
STCM 0.08 0.99 3791
LSTCM 0.075 0.97 4608
by Eq. (10) using Kroger Vink notation, where the
substitution of Ti41 ion with Fe31 gave rise to holes as
FIG. 10. The electrical conductivity and thermopower variation with temperature for (a) BSTFN (x 5 025) and (b) LSTF (x 5 0.15) double
perovskites. Reproduced from Ref. 20 with permission from PCCP Owner Societies.
14 J. Mater. Res., 2018

majority charge carriers. The n-type behavior of the compensated by both types of charge carriers resulting in
Sr2Ti0.8Fe0.8Nb0.4O6 (STFN) ceramics at higher temper- low values of S at room temperature, where it does not
ature arose due to the doping of Nb51 in the place of Ti41 produce much electrical conductivity. However, the holes
ion at the B-site of the double perovskite structure as contributed by defect reaction Eq. (10) act as the
expressed by Eq. (21), suggesting electrons as the charge dominant source of charge carriers from room tempera-
carriers. ture up to Tp–n, resulting in a positive thermopower value
in the S versus T graph in both the ceramics, whereas
2TiO2 1
Nb2 O5 ! 2NbTi þ 2e0 þ 4O
O þ O2 [ : ð21Þ beyond Tp–n, Eqs. (13), (15), and (21) dominated the
2 conductivity behavior of these samples. The dominant
defect reactions determine the major carriers in the
To verify the oxidation state of the B site cations, we system. Nevertheless, in these double perovskite systems,
further performed X-ray photoelectron spectroscopy of a p-type-to-n-type semiconductor phase transformation
the BSTFN compositions and it confirms the presence of occurred via an intermediate metallic phase. Hence, it
Fe in 31 and 21 states,92,93 Nb in 51 and 41 states94,95 was hypothesized that p–n-type switching facilitated
and Ti in 41 & 31 states,96,97 as shown in Figs. 11(a)– large electron restructuring in LSTF and BSTFN, which
11(c), respectively. was responsible for colossal change in thermopower.
It is clear from the XPS analysis that pure STF, which Further investigation to trap those structural changes in
behaved previously as a p-type semiconductor exhibited high temperature XRD is currently underway.
conduction switching (p to n-type) due to the incorpora- Previously such p–n type conduction switching assis-
tion of Nb in the B site of these double perovskite. ted large change in thermopower was reported in the
Similarly, the n-type behavior of LSTF double perov- literature for few chalcogenide-based compounds, only in
skites can be described by Eqs. (13) and (15) and it is the last five years. But for the very first time it was
apparent that both electrons and holes are present in the observed in perovskite-based oxide materials and also the
BSTFN and LSTF systems. Hence the thermopower was change in thermopower was larger than the previously
reported results, as shown in the comparison20,21,98–102 bar
chart in Fig. 12. Such p–n switching coupled with a large
TABLE IV. Change in thermopower and its corresponding tempera- change in thermopower in a single material as a result of
ture obtained from the temperature-dependent thermopower graph for only temperature change, opens up a new avenue for
BSTFN and LSTF ceramics. developing novel high-temperature diodes, thyristors,
Composition Change in thermopower Temperature of p–n switching sensors, switches, thermistors, etc. The target areas for
(x) (DS) (lV/K) (Tp–n) (K) these devices are envisaged in space, chemical plants,
automobile industries, and consumer electronics like cell
0.0 625 940.6
0.1 718.2 1035.6 phones, in which overheating or overcharging can cause
BSTFN the explosion of batteries due to thermal runaway. Co-
0.15 988.8 983.1
0.25 1146.2 935.3 lossal change in thermopower on the order of a few
0.1 2516.0 1139.9 millivolts per Kelvin, integrated with p–n-type conduction
0.15 2191.4 947.2
LSTF switching, may lead to the development of novel thermal
0.2 1342.0 941.8
0.25 1163.6 942.7 runaway monitors. Moreover, it can also be used for
chemical-to-thermal-to-electrical energy conversion on the
FIG. 11. The respective binding energy of Fe, Nb, and Ti in BSTFN double perovskite obtained by X-ray photoelectron spectroscopy.
J. Mater. Res., 2018 15

ZT values of these oxides are lower compared to that of

chalcogenides and intermetallics, although oxides

showed good thermal stability at high temperature. More
efforts are needed to improve the ZT values of these
double perovskite oxides to realize the commercialization
of these materials for TE power generation.
ACKNOWLEDGMENT
This work is supported by the grant from Science and
Engineering Research Board, DST (SERB-DST), India
(Grant No. SB/S3/ME/008/2015).
REFERENCES
1. J. He, M.G. Kanatzidis, and V.P. Dravid: High performance bulk
thermoelectrics via a panoscopic approach. Mater. Today 16, 166
(2013).
FIG. 12. Comparison of the maximum change in thermopower (DS) 2. R. Ovik, B. Long, M. Barma, M. Riaz, M. Sabri, S. Said, and
coupled with p–n type resistance switching behavior observed in R. Saidur: A review on nanostructures of high-temperature
different materials. thermoelectric materials for waste heat recovery. Renewable
Sustainable Energy Rev. 64, 635 (2016).
3. K. Ullah, R. Saidur, H. Ping, R. Akikur, and N. Shuvo: A review
of solar thermal refrigeration and cooling methods. Renewable
basis of the recently developed thermopower wave Sustainable Energy Rev. 24, 499 (2013).
concept. 4. S.F. Tie and C.W. Tan: A review of energy sources and energy
management system in electric vehicles. Renewable Sustainable
Energy Rev. 20, 82 (2013).
X. CONCLUSION 5. M. Scheele, N. Oeschler, K. Meier, A. Kornowski, C. Klinke,
In the current work, a wide range of double perovskite and H. Weller: Synthesis and thermoelectric characterization of
Bi2Te3 nanoparticles advanced. Funct. Mater. 19, 3476 (2009).
oxide-based high-temperature TE materials were synthe- 6. B. Poudel, Q. Hao, Y. Ma, Y. Lan, A. Minnich, B. Yu, X. Yan,
sized by the solid-state reaction method. The crystal D. Wang, A. Muto, and D. Vashaee: High-thermoelectric
structure of these materials was found to be cubic with performance of nanostructured bismuth antimony telluride bulk
Pm 3m space group. A dense microstructure structure alloys. Science 320, 634 (2008).
with grain size distribution scaling from nm to lm range 7. T. Caillat, J-P. Fleurial, and A. Borshchevsky: Preparation and
thermoelectric properties of semiconducting Zn4Sb3. J. Phys.
was achieved in these double perovskites. It was shown Chem. Solids 58, 1119 (1997).
that by manipulating cations and processing parameters, 8. K. Kadel, L. Kumari, W. Li, J.Y. Huang, and P.P. Provencio:
one can optimize electrical conductivity and Seebeck Synthesis and thermoelectric properties of Bi2Se3 nanostructures.
coefficient in these double perovskite oxides to achieve Nanoscale Res. Lett. 6, 1 (2011).
very good TE power factor. However, shift in semi- 9. R. Brebrick and A. Strauss: Anomalous thermoelectric power as
evidence for two-valence bands in SnTe. Phys. Rev. 131, 104
conductor to metal transition temperature, TS-M by the (1963).
doping played a key role in determining the operational 10. T. Takeuchi, T. Kondo, T. Takami, H. Takahashi, H. Ikuta,
temperature of these TE materials with maximum ZT U. Mizutani, K. Soda, R. Funahashi, M. Shikano, and
values. Some of these double perovskites oxides also M. Mikami: Contribution of electronic structure to the large
demonstrated temperature driven p–n switching assisted thermoelectric power in layered cobalt oxides. Phys. Rev. B 69,
125410 (2004).
colossal change in Seebeck coefficient, which can be 11. N. Van Nong, N. Pryds, S. Linderoth, and M. Ohtaki: Enhance-
used for developing novel multifunctional devices. Con- ment of the thermoelectric performance of p‐type layered oxide
duction mechanism of these materials was explained by Ca3Co4O9 1 d through heavy doping and metallic nanoinclu-
the small polaron hopping model. These double perov- sions. Adv. Mater. 23, 2484 (2011).
skites also showed low-temperature relaxor ferroelectric 12. G.J. Snyder and E.S. Toberer: Complex thermoelectric materials.
Nat. Mater. 7, 105 (2008).
behavior inducing dipolar glassy phase. We demonstrated 13. M.H. Aguirre, D. Logvinovich, L. Bocher, R. Robert,
that using these relaxors above the Burns temperature, S.G. Ebbinghaus, and A. Weidenkaff: High-temperature thermo-
one can use the glassy phase of relaxor to obtain very low electric properties of Sr2RuYO6 and Sr2RuErO6 double perov-
thermal conductivity without affecting its electrical con- skites influenced by structure and microstructure. Acta Mater. 57,
ductivity. In other words, inducing glassy phase as a result 108 (2009).
14. T. Kawano, J. Takahashi, T. Yamada, and H. Yamane: Prepa-
of relaxor ferroelectricity in double perovskites can be ration, crystal structure and high-temperature thermoelectric
a right approach for decoupling “Phonon-Glass Electron properties of double perovskite-type rare-earth cobalt ruthenium
Crystal” behavior in these oxides. Nevertheless, oxides. J. Ceram. Soc. Jpn. 115, 792 (2007).
16 J. Mater. Res., 2018

15. A.E. Smith, A.W. Sleight, and M. Subramanian: Electrical and 33. L. Daniels, S. Ling, S. Savvin, M. Pitcher, M. Dyer, J. Claridge,
magnetic properties of new rhodium perovskites: La2MRhO6, B. Slater, F. Corà, J. Alaria, and M. Rosseinsky: A and B site
M 5 Cr, Fe, Cu. Mater. Res. Bull. 45, 460 (2010). doping of a phonon-glass perovskite oxide thermoelectric. J.
16. T. Sugahara and M. Ohtaki: Structural and semiconductor-to- Mater. Chem. A 6, 15640–15652 (2018).
metal transitions of double-perovskite cobalt oxide 34. N. Nong, C-J. Liu, and M. Ohtaki: High-temperature thermo-
Sr2xLaxCoTiO6d with enhanced thermoelectric capability. electric properties of late rare earth-doped Ca3Co4O9 1 d. J.
Appl. Phys. Lett. 99, 062107 (2011). Alloys Compd. 509, 977 (2011).
17. T. Sugahara, N. Van Nong, and M. Ohtaki: Structure and 35. F. Zhang, Q. Lu, and J. Zhang: Synthesis and high temperature
thermoelectric properties of Ca2xSrx FeMoO6 (0 # x # 0.3) thermoelectric properties of BaxAgyCa3xyCo4O9 compounds.
double-perovskite oxides. Mater. Chem. Phys. 133, 630 (2012). J. Alloys Compd. 484, 550 (2009).
18. T. Sugahara, M. Ohtaki, and K. Suganuma: La doped effects on 36. D. Sarma, E. Sampathkumaran, S. Ray, R. Nagarajan,
structure and thermoelectric properties of Sr2MnMoO6 double- S. Majumdar, A. Kumar, G. Nalini, and T.G. Row: Magnetore-
perovskite oxides. J. Asian Ceram. Soc. 1, 282 (2013). sistance in ordered and disordered double perovskite oxide, Sr2
19. P. Roy, V. Waghmare, and T. Maiti: Environmentally friendly FeMoO6. Solid State Commun. 114, 465 (2000).
BaxSr2xTiFeO6 double perovskite with enhanced thermopower 37. S. Kumar, G. Giovannetti, J. van den Brink, and S. Picozzi:
for high temperature thermoelectric power generation. RSC Adv. Theoretical prediction of multiferroicity in double perovskite
6, 54636 (2016). Y2NiMnO6. Phys. Rev. B 82, 134429 (2010).
20. P. Roy, V. Waghmare, K. Tanwar, and T. Maiti: Large change in 38. S. Vasala and M. Karppinen: A2B9B0O6 perovskites: A review.
thermopower with temperature driven p–n type conduction Prog. Solid State Chem. 43, 1 (2015).
switching in environment friendly BaxSr2xTi0.8Fe0.8Nb0.4O6 39. G. Demazeau, B. Siberchicot, S. Matar, C. Gayet, and
double perovskites. Phys. Chem. Chem. Phys. 19, 5818 (2017). A. Largeteau: A new ferromagnetic oxide La2MnIrO6: Synthesis,
21. P. Roy and T. Maiti: Colossal change in thermopower characterization, and calculation of its electronic structure. J.
with temperature-driven p–n-type conduction switching in Appl. Phys. 75, 4617 (1994).
LaxSr2xTiFeO6 double perovskites. J. Phys. D: Appl. Phys. 40. O.N. Meetei, O. Erten, A. Mukherjee, M. Randeria, N. Trivedi,
51, 065104 (2018). and P. Woodward: Theory of half-metallic double perovskites. I.
22. M. Saxena, P. Roy, M. Acharya, I. Bose, K. Tanwar, and Double exchange mechanism. Phys. Rev. B 87, 165104 (2013).
T. Maiti: Enhanced thermoelectric figure-of-merit in environ- 41. P. Roy, I. Bose, and T. Maiti: Synthesis and characterization of
Sr2TiMO6 (M 5 Fe, Co) double perovskites for high temper-
mentally benign BaxSr2xTiCoO6 double perovskites. Appl.
ature thermoelectric applications. Integr. Ferroelectr. 174, 34
Phys. Lett. 109, 263903 (2016).
(2016).
23. M. Acharya and T. Maiti: Effect of bismuth doping on
42. C.J. Howard, B.J. Kennedy, and P.M. Woodward: Ordered
thermoelectric properties of Sr2TiCoO6. Ferroelectrics 532
double perovskites—A group-theoretical analysis. Acta Crystal-
(2018). (in press).
logr., Sect. B: Struct. Sci. 59, 463 (2003).
24. M. Saxena, K. Tanwar, and T. Maiti: Environmental friendly
43. M.T. Anderson, K.B. Greenwood, G.A. Taylor, and
Sr2TiMoO6 double perovskite for high temperature thermoelec-
K.R. Poeppelmeier: B-cation arrangements in double perov-
tric applications. Scr. Mater. 130, 205 (2017).
skites. Prog. Solid State Chem. 22, 197 (1993).
25. M. Saxena and T. Maiti: Metal-like electrical conductivity in
44. G. King and P.M. Woodward: Cation ordering in perovskites. J.
LaxSr2x TiMoO6 oxides for high temperature thermoelectric
Mater. Chem. 20, 5785 (2010).
power generation. Dalton Trans. 46, 5872 (2017). 45. V. Goldschmidt: The laws of crystal chemistry. Naturwissen-
26. M. Saxena and T. Maiti: Effect of Ba-doping on high temper-
schaften 14, 477 (1926).
ature thermoelectric properties of Sr2TiMoO6 double perov- 46. J. Rodríguez-Carvajal: An Introduction to the Program FullProf
skites. J. Alloys Compd. 710, 472 (2017). 2000 (Laboratoire Léon Brillouin, CEA-CNRS: Saclay, France,
27. M. Saxena and T. Maiti: Evaluation of Ba doped Sr2TiFe0.5Mo0.5O6 2001).
double perovskites for high temperature thermoelectric power 47. B.K. Maji, H. Jena, R. Asuvathraman, and K.G. Kutty: Electrical
generation. Scr. Mater. 155, 85 (2018). conductivity and thermal expansion behavior of MMoO4 (M 5
28. M. Saxena and T. Maiti: Compositional modification of Ca, Sr, and Ba). J. Alloys Compd. 640, 475 (2015).
Sr2TiCoO6 double perovskites by Mo and La for high 48. D. Logvinovich, R. Aguiar, R. Robert, M. Trottmann,
temperature thermoelectric applications. Ceram. Int. 44, 2732 S. Ebbinghaus, A. Reller, and A. Weidenkaff: Synthesis, Mo-
(2018). valence state, thermal stability and thermoelectric properties of
29. K. Tanwar, M. Saxena, and T. Maiti: Enhancement of thermo- SrMoO3xNx (x . 1) oxynitride perovskites. J. Solid State
electric power factor of Sr2CoMoO6 double perovskite by Chem. 180, 2649 (2007).
annealing in reducing atmosphere. J. Appl. Phys. 122, 164902 49. G.A. Slack and M. Rowe: CRC Handbook of Thermoelectrics
(2017). (CRC Press, Boca Raton, Florida, 1995); p. 407.
30. P. Roy, V. Pal, and T. Maiti: Effect of spark plasma sintering 50. N. Wang, H. He, X. Li, L. Han, and C. Zhang: Enhanced
(SPS) on the thermoelectric properties of SrTiO3: 15 at.% Nb. thermoelectric properties of Nb-doped SrTiO3 polycrystalline
Ceram. Int. 43, 12809 (2017). ceramic by titanate nanotube addition. J. Alloys Compd. 506, 293
31. S. Ohta, T. Nomura, H. Ohta, and K. Koumoto: High-temper- (2010).
ature carrier transport and thermoelectric properties of heavily 51. T. Sugahara, M. Ohtaki, and T. Souma: Thermoelectric proper-
La-or Nb-doped SrTiO3 single crystals. J. Appl. Phys. 97, ties of double-perovskite oxide Sr2xMxFeMoO6 (M 5 Ba, La).
034106 (2005). J. Ceram. Soc. Jpn. 116, 1278 (2008).
32. R. Funahashi, A. Kosuga, N. Miyasou, E. Takeuchi, S. Urata, 52. D. Ackerman, D. Moy, R. Potter, A. Anderson, and W. Lawless:
K. Lee, H. Ohta, and K. Koumoto: Thermoelectric properties of Glassy behavior of crystalline solids at low temperatures. Phys.
CaMnO3 system. In Thermoelectrics, 2007. ICT 2007. 26th Rev. B 23, 3886 (1981).
International Conference (IEEE, Jeju Island, South Korea, 53. C. Huang: Some experimental aspects of spin glasses: A review.
2007); p. 124. J. Magn. Magn. Mater. 51, 1 (1985).
J. Mater. Res., 2018 17

54. J. De Yoreo, R. Pohl, and G. Burns: Low-temperature thermal 76. P. Bernasconi, I. Biaggio, M. Zgonik, and P. Günter: Anisotropy
properties of ferroelectrics. Phys. Rev. B 32, 5780 (1985). of the electron and hole drift mobility in KNbO3 and BaTiO3.
55. K. Uchino and S. Nomura: Critical exponents of the dielectric Phys. Rev. Lett. 78, 106 (1997).
constants in diffused-phase-transition crystals. Ferroelectrics 44, 77. P.A. Cox: Transition Metal Oxides: An Introduction to Their
55 (1982). Electronic Structure and Properties (Oxford University Press,
56. G. Smolenskii and A. Agranovskaya: Dielectric polarization New York, New York, 2010).
and losses of some complex compounds. Zh. Tekh. Fiz. 28, 78. S. Lee, G. Yang, R.H. Wilke, S. Trolier-McKinstry, and
1380–1382 (1958). C.A. Randall: Thermopower in highly reduced n-type ferroelec-
57. D. Viehland, S. Jang, L.E. Cross, and M. Wuttig: Freezing of the tric and related perovskite oxides and the role of heterogeneous
polarization fluctuations in lead magnesium niobate relaxors. J. nonstoichiometry. Phys. Rev. B 79, 134110 (2009).
Appl. Phys. 68, 2916 (1990). 79. K-Y. Yang, K-Z. Fung, and M-C. Wang: X-ray photoelectron
58. D. Viehland, J. Li, S. Jang, L.E. Cross, and M. Wuttig: Dipolar- spectroscopic and secondary ion mass spectroscopic examina-
glass model for lead magnesium niobate. Phys. Rev. B 43, 8316 tions of metallic-lithium-activated donor doping process on
(1991). La0.56Li0.33TiO3 surface at room temperature. J. Appl. Phys.
59. G. Burns and F. Dacol: Ferroelectrics with a glassy polarization 100, 056102-3 (2006).
phase. Ferroelectrics 104, 25 (1990). 80. I.C. Lekshmi, A. Gayen, and M. Hegde: The effect of strain on
60. G. Burns and F. Dacol: Glassy polarization behavior in ferro- nonlinear temperature dependence of resistivity in SrMoO3 and
electric compounds Pb(Mg13Nb23)O3 and Pb(Zn13Nb23)O3. Solid SrMoO3xNx films. Mater. Res. Bull. 40, 93 (2005).
State Commun. 48, 853 (1983). 81. Y. Lu and C. Clayton: An XPS study of the passive and
61. A. Bhalla, R. Guo, L. Cross, G. Burns, F. Dacol, and transpassive behavior of molybdenum in deaerated 0.1 M HCl.
R.R. Neurgaonkar: Measurements of strain and the optical Corros. Sci. 29, 927 (1989).
indices in the ferroelectric Ba0.4Sr0.6Nb2O6: Polarization effects. 82. J-G. Choi and L. Thompson: XPS study of as-prepared and
Phys. Rev. B 36, 2030 (1987). reduced molybdenum oxides. Appl. Surf. Sci. 93, 143 (1996).
62. S. Shtrikman and E. Wohlfarth: The theory of the Vogel–Fulcher 83. N.F. Mott and L. Friedman: Metal–insulator transitions in VO2,
law of spin glasses. Phys. Lett. A 85, 467 (1981). Ti2O3, and Ti2xVxO3. Philos. Mag. 30, 389 (1974).
63. R. Pirc and R. Blinc: Vogel–Fulcher freezing in relaxor ferro- 84. E. Gorham-Bergeron and D. Emin: Phonon-assisted hopping due
electrics. Phys. Rev. B 76, 020101 (2007). to interaction with both acoustical and optical phonons. Phys.
64. T. Maiti, R. Guo, and A. Bhalla: Structure‐property phase Rev. B 15, 3667 (1977).
diagram of BaZrxTi1xO3 system. J. Am. Ceram. Soc. 91, 1769 85. N. Tsuda, K. Nasu, A. Fujimori, and K. Siratori: Electronic
Conduction in Oxides (Springer Science & Business Media, New
(2008).
65. E. Fischer, W. Hässler, and E. Hegenbarth: Glass‐like behaviour York, New York, 2013).
86. A. Nag and V. Shubha: Oxide thermoelectric materials: A
in the thermal conductivity of Sr1xBax(Nb2O6) single crystal.
structure–property relationship. J. Electron. Mater. 43, 962
Phys. Status Solidi A 72, K169 (1982).
(2014).
66. X. Yuan and M. Xu: Thermal transport properties of poly-
87. P. Chaikin and G. Beni: Thermopower in the correlated hopping
crystalline Pb2FeMoO6. Mater. Res. Express 5, 066102 (2018).
regime. Phys. Rev. B 13, 647 (1976).
67. H. Nakatsugawa, M. Saito, and Y. Okamoto: High-temperature
88. R. Waser and M. Aono: Nanoionics-based resistive switching
thermoelectric properties of perovskite-type
memories. Nat. Mater. 6, 833 (2007).
Pr0.9Sr0.1Mn1xFexO3 (0 # x # 1). J. Electron. Mater. 46,
89. U. Chon, H.M. Jang, M. Kim, and C. Chang: Layered perov-
3262 (2017). skites with giant spontaneous polarizations for nonvolatile
68. J-W. Moon, Y. Masuda, W-S. Seo, and K. Koumoto: Ca-doped
memories. Phys. Rev. Lett. 89, 087601 (2002).
HoCoO3 as p-type oxide thermoelectric material. Mater. Lett. 48, 90. S. Mathews, R. Ramesh, T. Venkatesan, and J. Benedetto:
225 (2001). Ferroelectric field effect transistor based on epitaxial perovskite
69. L. Daniels, S. Savvin, M. Pitcher, M. Dyer, J. Claridge, S. Ling, heterostructures. Science 276, 238 (1997).
B. Slater, F. Corà, J. Alaria, and M. Rosseinsky: Phonon-glass 91. J.W. Fergus: Perovskite oxides for semiconductor-based gas
electron-crystal behaviour by a site disorder in n-type thermo- sensors. Sens. Actuators, B 123, 1169 (2007).
electric oxides. Energy Environ. Sci. 10, 1917 (2017). 92. T. Yamashita and P. Hayes: Analysis of XPS spectra of Fe21 and
70. R. Ang, Y. Sun, and W. Song: The magnetic, electrical and Fe31 ions in oxide materials. Appl. Surf. Sci. 254, 2441 (2008).
thermal transport studies in the layered cobalt oxide 93. E. Paparazzo: XPS and auger spectroscopy studies on mixtures
Nd1xSr11xCoO4 (x 5 0.25 and 0.33). J. Phys. D: Appl. Phys. of the oxides SiO2, Al2O3, Fe2O3, and Cr2O3. J. Electron
40, 5206 (2007). Spectrosc. Relat. Phenom. 43, 97 (1987).
71. Y. Wang, K.H. Lee, H. Ohta, and K. Koumoto: Thermoelectric 94. H-R. Lee, S-J. Kim, I-S. Yang, and J-H. Choy: Low-dimensional
properties of electron doped SrO(SrTiO3)n (n 5 1, 2) ceramics. J. and extended metal—metal bonded networks in transition metal
Appl. Phys. 105, 103701 (2009). compounds: Ba2Nb5O9, Ba2xYxNb5O9. J. Solid State Chem.
72. Y.F. Wang, K.H. Lee, H. Ohta, and K. Koumoto: Fabrication 108, 253 (1994).
and thermoelectric properties of heavily rare-earth metal-doped 95. C.J. Carmalt, T.D. Manning, I.P. Parkin, E.S. Peters, and
SrO(SrTiO3)n (n 5 1, 2) ceramics. Ceram. Int. 34, 849 (2008). A.L. Hector: Formation of a new (1T) trigonal NbS2 polytype
73. R. Sun, D. Li, L. Li, J. Zhang, Q. Wang, and X. Qin: Preparation via atmospheric pressure chemical vapour deposition. J. Mater.
and thermoelectric properties of rare-earth-metal-doped SrO Chem. 14, 290 (2004).
(SrTiO3)n oxides. Procedia Eng. 27, 103 (2012). 96. A. Dementjev, O. Ivanova, L. Vasilyev, A. Naumkin,
74. K.H. Lee, S.W. Kim, H. Ohta, and K. Koumoto: Ruddlesden- D. Nemirovsky, and D. Shalaev: Altered layer as sensitive initial
popper phases as thermoelectric oxides: Nb-doped SrO(SrTiO3)n chemical state indicator. J. Vac. Sci. Technol., A. 12, 423 (1994).
(n 5 1, 2). J. Appl. Phys. 100, 063717 (2006). 97. C. Lu, A. Kuang, and G. Huang: X-ray photoelectron spectros-
75. P.W. Anderson: Absence of diffusion in certain random lattices. copy study on composition and structure of sol–gel derived
Phys. Rev. 109, 1492 (1958). PbTiO3 thin films. J. Appl. Phys. 80, 202 (1996).
18 J. Mater. Res., 2018

98. S.N. Guin, J. Pan, A. Bhowmik, D. Sanyal, U.V. Waghmare, and H. Eckert: Reversible switching between p-and n-type conduc-
K. Biswas: Temperature dependent reversible p–n–p type con- tion in the semiconductor Ag10Te4Br3. Nat. Mater. 8, 101 (2009).
duction switching with colossal change in thermopower of 101. R. Littleton, IV, T.M. Tritt, J. Kolis, D. Ketchum, N.D. Lowhorn,
semiconducting AgCuS. J. Am. Chem. Soc. 136, 12712 (2014). and M. Korzenski: Suppression of the resistivity anomaly and
99. C. Xiao, X. Qin, J. Zhang, R. An, J. Xu, K. Li, B. Cao, J. Yang, corresponding thermopower behavior in the pentatelluride sys-
B. Ye, and Y. Xie: High thermoelectric and reversible pnp tem by the addition of Sb: Hf1XZrXTe5YSbY. Phys. Rev. B 64,
conduction type switching integrated in dimetal chalcogenide. J. 121104 (2001).
Am. Chem. Soc. 134, 18460 (2012). 102. O. Osters, M. Bawohl, J-L. Bobet, B. Chevalier, R. Decourt, and
100. T. Nilges, S. Lange, M. Bawohl, J.M. Deckwart, M. Janssen, T. Nilges: A conceptional approach to materials for resistivity
H-D. Wiemhöfer, R. Decourt, B. Chevalier, J. Vannahme, and switching and thermoelectrics. Solid State Sci. 13, 944 (2011).
J. Mater. Res., 2018 19

Double Perovskite Sr2B B O6 Oxides For H

Uploaded by

Copyright:

Available Formats

Double Perovskite Sr2B B O6 Oxides For H

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Double Perovskite Sr2B B O6 Oxides For H

Uploaded by

Copyright:

Available Formats

INVITED REVIEW

Double perovskite (Sr2B9B0O6) oxides for high-temperature

(Received 1 June 2018; accepted 20 September 2018)

I. INTRODUCTION transforms thermal energy (waste heat) into electrical

2 J. Mater. Res., 2018

Crystal structure/ Electrical conductivity Seebeck coefﬁcient Power factor (S2rmax)

J. Mater. Res., 2018 3

A. Effect of processing atmosphere in 10% H2 þAr 1

4 J. Mater. Res., 2018

J. Mater. Res., 2018 5

6 J. Mater. Res., 2018

J. Mater. Res., 2018 7

8 J. Mater. Res., 2018

J. Mater. Res., 2018 9

TABLE II. Thermal conductivity of different oxides.

Thermal conductivity (W/mK) at low Thermal conductivity (W/mK) at high

Ba0.2Sr1.8TiCoO6 0.11 at 300 K 0.35 at 1220 K 22

10 J. Mater. Res., 2018

J. Mater. Res., 2018 11

12 J. Mater. Res., 2018

J. Mater. Res., 2018 13

14 J. Mater. Res., 2018

J. Mater. Res., 2018 15

ZT values of these oxides are lower compared to that of

chalcogenides and intermetallics, although oxides

16 J. Mater. Res., 2018

J. Mater. Res., 2018 17

18 J. Mater. Res., 2018

J. Mater. Res., 2018 19

You might also like