Ca1 xprxmno3

JOURNAL OF RARE EARTHS, Vol. 31, No. 9, Sep. 2013, P. 885

Effects of praseodymium doping on thermoelectric transport properties of
CaMnO3 compound system
ZHANG Feipeng (ᓴ亲吣)1,2,*, NIU Baocheng (⠯ֱ៤)1, ZHANG Kunshu (ᓴസк)1, ZHANG Xin (ᓴ ᗏ)2,
LU Qingmei (䏃⏙ṙ)2, ZHANG Jiuxing (ᓴЙ݈)2
(1. Institute of Physics, Henan University of Urban Construction, Pingdingshan 467036, China; 2. Key Laboratory of Advanced Functional Materials, Chinese
Ministry of Education, College of Materials Science and Engineering, Beijing University of Technology, Beijing 100124, China)
Received 17 May 2013; revised 25 June 2013

Abstract: The rare earth Pr doped Ca1–xPrxMnO3 (x=0, 0.06, 0.08, 0.1, 0.12, and 0.14) compound bulk samples were prepared to
study the effect of Pr doping on thermoelectric transport properties of CaMnO3 compound system. The doped samples exhibited single phase composition within the experimental doping range, with condensed bulk microstructure and small porosities. The electrical
resistivity was remarkably reduced for doped samples, on account of the enhanced carrier concentration; the absolute value of Seebeck coefficient was deteriorated mainly due to enhanced electron carrier concentration. The electrical performances of the doped
samples reflected by resistivity and Seebeck coefficient fluctuations were optimistically tuned, with an optimized power factor
value of 0.342 mW/(m·K2) at 873 K for x=0.08 sample, which was very much higher comparing with that of the un-doped sample.
The lattice thermal conduction was really confined, leading to distinctly repressed total thermal conductivity. The thermoelectric performance was noticeably improved by Pr doping and the dimensionless figure of merit ZT for the Ca0.92Pr0.08MnO3 compound was
favorably optimized with the maximum value 0.16 at 873 K.
Keywords: CaMnO3 compound; Pr doping; thermoelectric properties; rare earths

More and more efforts have been paid to the field of
thermoelectric (TE) materials in the past decades owing
to their clean energy conversion between heat and electricity through Seebeck effect and Peltier effect. The efficiency of energy transformation is positively correlated
to the materials’ dimensionless figure of merit ZT formulated by:
ZT=Į2T/ȡț
(1)
where Į, ȡ, T and ț are the Seebeck coefficient, the electrical resistivity, the absolute temperature and the total
thermal conductivity, respectively[1–3]. For crystal phase
materials, the total thermal conductivity ț is usually regarded as composing of the carrier thermal conductivity
component țc and the lattice thermal conductivity component țL:
ț=țc+țL
(2)

Good TE materials should have high Seebeck coefficient Į, low electrical resistivity ȡ and total thermal conductivity ț simultaneously. Nevertheless, these parameters are not independent of each other; they are closely
correlated to transport mechanism, for instance, the sort
of charge carriers, the carrier density, mobility, carrier
effective mass, phonon mean free path, vibration and
phonon modes. A good combination of transport parameters is needed in order to achieve a considerable ZT

value for applicable TE materials[4]. They are also sensitive to materials’ microstructures and textures, thereafter
the materials preparation techniques. The hotspot systems are tellurides, silicides, sulphide, half-Heusler
alloys, clathrates, skutterudites and oxides[4–10]. The oxides-based TE materials have many advantages over alloys-based materials such as atmospheric stability, easy
fabrication, cheapness, high temperature stability, etc.[4].
The n-type CaMnO3 compound shows semiconductorlike conductivity (dȡ/dT<0), high Seebeck coefficient Į
(|ĮRT|§350 ȝV/K) and high temperature stability (1500
K)[4]. It has been regarded as one of the most promising
n-type TE oxide materials, and it has also been receiving
much interest in the past decade in terms of its structural,
topological, physical, magnetic properties and TE performance[4,11,12].
The structure of the CaMnO3 is based on the framework of corner sharing O–Mn–O octahedron in which
the Mn is surrounded by six O and the Ca is surrounded
by twelve anions within the cavity. This framework allows it the ability of incorporating dopants with different
sizes and valences[13]. It is verified from the theoretical
investigation that the doping type, namely n-type or
p-type, together with the dopant atomic mass, plays a key
role in determining the optimization of TE performance

Foundation item: Project supported by National Natural Science Foundation of China (50801002), Beijing Municipal Natural Science Foundation
(2112007) and Basic and Advanced Technology Research Project of Henan Province (132300410071)
* Corresponding author: ZHANG Feipeng (E-mail: ; Tel.: +86-375-2089151)
DOI: 10.1016/S1002-0721(12)60374-3

886

of CaMnO3 [11–14]. There are many reports concerning
doping-induced behaviors for CaMnO3 system, the doping categories include Ca site doping, Mn site doping and
double doping for both Ca and Mn sites. The rare earth
(RE) doping provides electron carriers in the system, the
carrier concentration should be increased, and the carrier
mean free path should not be intensively lowered due to
the comparable ionic radii between Ca and RE, this is
favorable for reducing electrical resistivity[13,14]. At the
same time, the enhanced effective mass introduced by
heavy dopants should contribute to the maintaining of
Seebeck coefficient[11,12,14]. Thirdly, the heavy elements
RE doping affords confined phonon mean free path,
thereby this is favorable for lattice thermal conductivity
suppression[14]. It is thus mostly hopeful to tune the TE
transport parameters by RE elements doping, to obtain a
good combination of these parameters. In this paper, the
rare earth Pr doped Ca1–xPrxMnO3 (x=0, 0.06, 0.08, 0.1,
0.12, and 0.14) compound bulk samples were fabricated;
the effect of Pr doping on the TE transport properties of
n-type CaMnO3 compound from 373 K up to 973 K was
investigated in detail.

1 Experimental
As reported within our former works[15], the CaMnO3based compound powder with grain size in nanometer
magnitude order was synthesized by citrate acid sol-gel
reaction method. In the present paper, the Pr doped
Ca1–xPrxMnO3 (x=0, 0.06, 0.08, 0.1, 0.12, and 0.14) compound powders were synthesized by the same method,
and then the bulk samples were prepared by ceramic

preparation technique. Stoichiometric ratios of highly
pure nitrates of Ca, Pr, and Mn were dissolved in distilled water; the citric acid was added in the aqueous solution. The solution was continuously mixed at 360 K in
order to form the precursor gel. The gel was dried at 473 K
for 12 h in air to evaporate the excessive water. Then the
dried gel was ground and calcined at 1173 K for 8 h to
remove excess organic compounds and to get the
Ca1–xPrxMnO3 (0”x”0.14) powder. Then the powder was
finely ground and pressed into platelets. Finally the
pressed platelets were heated slowly to the temperature
of 1473 K in air at the heating rate of 10 K/min, and the
samples were maintained at 1473 K for 12 h, then the
samples were subjected to furnace cooling from that
temperature to room temperature.
The phase constitutions of bulk samples were analyzed
by X-ray diffraction (XRD) at room temperature on a
Rigaku diffractometer with Cu KĮ radiation in a 2ș range
of 20º–85º, with steps of 0.02°(2ș) and a time per step of
1 s. The microscopic image of the bulk samples was obtained with the scanning electron microscopy (SEM) using secondary electron mode by Nova NanoSEM operated at 18 kV. The electrical resistivity and Seebeck co-

JOURNAL OF RARE EARTHS, Vol. 31, No. 9, Sep. 2013

efficient were measured in He atmosphere from room
temperature up to 1000 K using a conventional DC standard four-probe method on ULVAC ZEM-2 system. The
specific heat capacity Cp and thermal diffusivity Ȝ were
measured in Ar atmosphere by the laser flash technique
on ULVAC-RIKO TC-7000 system. The total thermal
conductivity was then calculated by ț=dCpȜ, the density
d was measured by Archimedes method. The electron
carrier concentration n and the mobility ȝ were measured
via n=1/RHe on Accent HL5500 Hall System and the RH

is Hall coefficient.

2 Results and discussion
2.1 Phase composition and bulk microstructure
Fig. 1 presents the XRD patterns for rare earth Pr
doped Ca1–xPrxMnO3 (x=0, 0.06, 0.08, 0.1, 0.12, and 0.14)
compound bulk samples. All diffraction peaks are indexed by comparing with standard JCPDS card No.
50-1746 for orthorhombic CaMnO3, no impurity phases
are found. This confirms the formation of single phase
CaMnO3 compound. In addition, 2ș shift to higher angles
could be observed (Fig. 2), indicating the decreased lattice parameters with increasing Pr doping content. The
rare earth element Pr exhibits trivalent Pr3+ or tetravalent
Pr4+ state within the compound; this phenomenon is con-

Fig. 1 XRD patterns for Ca1–xPrxMnO3 (0”x”0.14) compounds

Fig. 2 2ș shift for Ca1–xPrxMnO3 (0”x”0.14) compounds


ZHANG Feipeng et al., Effects of praseodymium doping on thermoelectric transport properties of CaMnO3 …

sidered to be caused by ionic radii differences between
Pr3+/Pr4+ and Ca2+. Moreover, the width of the diffraction
peaks for Pr doped Ca1–xPrxMnO3 tends to get slightly
broader as the doping content increases (Fig. 2), which
indicates the crystalline grain size decreasing. Fig. 3 shows
the cross-section SEM images for the Ca1–xPrxMnO3 (x=0,
0.08, 0.1, 0.12) compound bulk samples, and condensed
bulk samples are formed as seen in the figures. Table 1
presents the measured relative density of the Pr doped

bulk samples, increased bulk density with increasing the
doping content x can be observed. It can also be observed
from the SEM images that the bulk samples have enhanced inter-grain connections with increasing the Pr
doping content x. Furthermore, decreased crystalline
grain size could also be seen with increasing the Pr doping concentration, this is in agreement with the above estimations. This phenomenon is in accordance with that of
Yb and Fe doped Ca1–xMxMnO3 (M=Yb, Fe) system[15,16].
Although further work should be done, it is estimated
that the grain nucleation rate can be enhanced, while the
grain growth rate remained moderately invariable with
increasing the doping concentration, so the crystalline
grain size is decreased. The enhanced grain connection
should contribute to carrier transport, while the modified
grain size would be positive for increasing the phonon
scattering, likewise decreasing the thermal conductivity.
2.2 Electrical transport properties

887

The electrical resistivity behavior is a combination
phenomenon of carrier transport parameters. The analyzing of resistivity would shed light on carrier transport
process. The electrical resistivity ȡ as a function of temperature and doping content x for rare earth Pr doped
Ca1–xPrxMnO3 (x=0, 0.06, 0.08, 0.1, 0.12, and 0.14) compound bulk samples is presented in Fig. 4. The un-doped
CaMnO3 compound shows a high resistivity ȡ, in agreement with its anti-ferromagnetic insulator nature. Favorably, the ȡ of rare earth Pr doped samples is remarkably reduced along with increasing doping concentration x, and
the ȡ is moderately saturated at x=0.14. The conductivity
enhancement of the titled system can be mainly ascribed to
the increased carrier concentration owing to the bivalent
Ca2+ doped by trivalent Pr3+ or tetravalent Pr4+ that provides electron carriers to the system[17]. The electrical resistivity ȡ of CaMnO3 system could be approximately expressed as a function of carrier density n and mobility ȝ:
ȡ§1/nqȝ
(3)
where q is the elementary charge of a carrier[18]. Table 1

shows the measured room temperature carrier concentration n and mobility ȝ for several samples. It is obvious
that the carrier concentration n is increased with increasing the doping content x, the mobility ȝ is also enhanced with increasing the doping content x. For instance,
the n is increased from 1.02×1019 cm–3 of the un-doped
sample to 2.82×1019 cm–3 of x=0.14 sample. The increased electron carrier concentration and the en-

Fig. 3 Cross-section SEM images for Ca1–xPrxMnO3 compounds
(a) x=0; (b) x=0.08; (c) x=0.1; (d) x=0.12


888

JOURNAL OF RARE EARTHS, Vol. 31, No. 9, Sep. 2013

Fig. 4 Resistivity ȡ as a function of temperature for Ca1–xPrxMnO3
(0”x”0.14) compounds

Fig. 5 ln(ȡ/T) vs. 1/T for Ca1–xPrxMnO3 (0”x”0.14) compounds

Table 1 Measured carrier density (n) and mobility (ȝ) for
Ca1–xPrxMnO3 (x=0, 0.06, 0.1, 0.14) compounds
x

n/(1019 cm–3)

ȝ/(cm2/Vs)

0.14

2.82

16.9

96.7

0.1

1.68

10.1

96.0

0.06

1.29

7.2

95.2

0

1.02

2.8

95.1

Relative density/%

hanced mobility are regarded to be responsible for the
resistivity decreasing.
According to transport theory proposed by Mott[12,19],
the electrical resistivity ȡ as a function of temperature for
the titled compound system could be simulated by the
small polaron model expressed as:
ȡ(T)=CT exp(Ea/kT)
(4)
where Ea is the activation energy of the polarons, k the
Boltzman constant, T the absolute temperature, and the
structure dependent variable C can be given by:
C

C0

exp(2ȞR )
Ne 2 a 2 f 1  f
Ȟp

(5)

where N the number of ion sites per unit cell volume
(Mn sites), e the electron charge, a an average inter-site
distance for polaron hopping, Ȟthe electron wave function decay constant, Ȟp the optical phonon frequency, f
the fraction of available sites occupied by small polarons and the Co a constant. For this transport model, the
activation energy of polaron carriers Ea is regarded as
the determining factor influencing carrier transport
process and the fluctuations of resistivity, the structure
dependent variable C can be neglected in analyzing the
carrier transport process. By calculating the equation,

the Ea can be obtained. Fig. 5 shows the plots of ln(ȡ/T)
and 1/T for all samples. It can be seen that the plots for
all samples lie on the straight lines in the whole temperature region; this verifies the applicability of the
transport model. By deducing the slopes of the linear fit
of ln(ȡ/T) and 1/T, the activation energy Ea of the polarons can be obtained. As calculated, the Ea is decreased
from 0.09 eV of the un-doped CaMnO3 to 0.03 eV

of the Ca0.86Pr0.14MnO3 sample. It is true that the carriers
are more easily activated to surpass the band gap with
increasing Pr doping content, and thus the conduction
capability is ultimately enhanced[13].
Fig. 6 presents the Seebeck coefficient D as a function
of temperature for rare earth Pr doped Ca1–xPrxMnO3
(x=0, 0.06, 0.08, 0.1, 0.12, and 0.14) compound bulk
samples. The D values for all samples are negative, confirming that the electron carriers are dominant for the titled compound. The dependence of Į on Pr doping content x for all the samples shows the same behaviors with
that of the electrical resistivity ȡ, namely the Į value is
decreased by increasing the doping content x. The fluctuation of Seebeck coefficients D for CaMnO3 system in
the studied temperature range can be explained by employing Marsh and Parris’s theory for strong coupling
systems[20]:
Į=–(kB/q)·ln[(3–r–y)/(r–1+y)]
(6)
where q the elementary charge of an electron, kB the
Boltzman constant, r the average number of eg electrons
per trivalent Mn3+ and y the electron concentration. According to the simplified formula, the absolute Seebeck
coefficient Į can be reduced by increasing the electron
concentration y value. The electron concentration y
would be increased by increasing the doping content x.
Therefore the Seebeck coefficient is decreased by

Fig. 6 Seebeck coefficient (Į) as a function of temperature for

Ca1–xPrxMnO3 (0”x”0.14) compounds


ZHANG Feipeng et al., Effects of praseodymium doping on thermoelectric transport properties of CaMnO3 …

889

increasing x. It can also be seen that the Seebeck coefficient of the Pr doped samples is not linearly dependent of
doping content x. Since the Seebeck coefficient D is also
related to the carrier effective mass and carrier scattering
mechanism[1,14], although further work is needed, it is estimated that the modulated scattering effects as well as
the carrier effective mass are responsible for the phenomenon.
2.3 Thermal properties
Fig. 7 presents the total thermal conductivity ț as a
function of temperature for rare earth Pr doped
Ca1–xPrxMnO3 (x=0, 0.06, 0.08, 0.1, 0.12, and 0.14) compound bulk samples. It can be seen that the total thermal
conductivity ț for all samples is weakly dependent of the
temperature, indicating that the main thermal transport
process is undertaken by lattice thermal conduction. In
order to evaluate the influence of rare earth Pr doping on
lattice thermal conductivity, the lattice thermal conductivity kL is deduced. The lattice thermal conductivity kL is
obtained by subtracting carrier thermal conductivity
component kc from total thermal conductivity k. The carrier contribution kc is calculated by applying the Wiedemann-Franz[18] law:
kc=LT/ȡ
(7)
–8
2 2
where L is the Lorenz constant 2.45×10 V /K and T is
absolute temperature. Fig. 8 gives the calculated lattice
thermal conductivity for all samples. It can be seen that

the total thermal conductivity depression is mainly due to
the lattice thermal conductivity reduction. The suppression of total thermal conductivity can be explained by the
reason that the phonon scattering is made much more effective through heavy element Pr doping, the vibration
modes that carry heat efficiently is lowered by the
enlarged number of species in the unit cell[21], therefore
the lattice thermal conductivity and the total thermal
conductivity are reduced. This is in accordance with the
theoretical study[14,21]. Additionally, the grain boundary
area and quantity should be increased as a result of the
reduced crystalline grain size; this would be in favor of

Fig. 8 Lattice thermal conductivity as a function of temperature
for Ca1–xPrxMnO3 (0”x”0.14) compounds

the grain boundary phonon scattering enhancement
which is partially responsible for the thermal conductivity suppression.
2.4 Thermoelectric figure of merit, ZT
Fig. 9 presents the TE dimensionless figure of merit
ZT as a function of temperature for rare earth Pr doped
Ca1–xPrxMnO3 (x=0, 0.06, 0.08, 0.1, 0.12, and 0.14) compound bulk samples. The ZT values increase with increasing temperature indicating that the titled compound
is more suitable for high temperature energy conversion
fields. The slopes of temperature dependence curves for
doped sample are all larger than that of the un-doped
samples, showing rapidly enhanced TE properties with
elevating the applied temperature. Secondly, the ZT value
of doped sample are remarkably higher than that of the
un-doped sample, these confirm that the TE transport
property could be really improved by rare earth elements
doping at a lower doping content. It is worth noting that
the x=0.08 sample has the highest ZT value with the peak

0.16 at 873 K which is much higher than that of the
un-doped compound system.

Fig. 9 Dimensionless figure of merit ZT as a function of temperature for Ca1–xPrxMnO3 (0”x”0.14) compounds
Fig. 7 Total thermal conductivity as a function of temperature
for Ca1–xPrxMnO3 (0”x”0.14) compounds

3 Conclusions
In summary, the effects of rare earth Pr doping within


890

low concentration on phase compositions and TE transport properties of the Ca1–xPrxMnO3 (x=0, 0.06, 0.08, 0.1,
0.12, and 0.14) compound systems were studied. All
samples were found to be single phased within the experimental doping range and they exhibited condensed
bulk microstructure accompanied by enhanced inter-grain connection and decreased grain size. The electrical resistivity was remarkably decreased along with
increasing the Pr doping content, mainly due to carrier
density enhancement. All the samples followed the polaron transport model well and the energy for polarons to
hop was decreased. The Seebeck coefficient experienced
a decreasing trend, in accordance with the electrical resistivity behavior as a function of doping content x. The
total thermal conductivity was reduced because of noticeable lattice thermal conductivity suppression. On account of tuning TE transport parameters independently,
the Ca0.92Pr0.08MnO3 sample showed the largest dimensionless figure of merit ZT with a peak value of 0.16 at
873 K, greatly higher than that of the un-doped CaMnO3
compound. The present investigation suggested that the
thermoelectric properties of the titled compound system
could be effectively improved via tuning the transport
parameters independently, by rare earth doping at a
lower content through simple sample preparation procedure.

References:
[1] Li J F, Liu W S, Zhao L D, Zhou M. High-performance
nanostructured thermoelectric materials. NPG Asia Mater.,
2010, 2: 152.
[2] Liu H Q, Zhao X B, Zhu T J, Gu Y J. Thermoelectric properties of YbxCo4Sb12 system. J. Rare Earths, 2012, 30:
456.
[3] Liu H Q, Gao H L, Gu Y J, Zhao X B. Effect of Y-filling
on thermoelectric properties of CoSb3. J. Rare Earths,
2011, 29: 596.
[4] Fergus J F. Oxide materials for high temperature thermoelectric energy conversion. J. Eur. Ceram. Soc., 2012, 32:
525.
[5] Poudel B, Hao Q, Ma Y, Lan Y, Minnich A, Yu B, Yan X,
Wang D, Muto A, Vashaee D, Chen X, Liu J, Dresselhaus
M S, Chen G, Ren Z. High-thermoelectric performance of
nanostructured bismuth antimony telluride bulk alloys.
Science, 2008, 320: 634.
[6] Hauser D, Savelli G, Plissonnier M, Montès L, Simon J.
Growth of heavily doped monocrystalline and polycrystalline SiGe-based quantum dot superlattices. Thin Solid
Films, 2012, 520: 4259.
[7] Zhu T J, Xiao K, Yu C, Shen J J, Yang S H, Zhou A J,

JOURNAL OF RARE EARTHS, Vol. 31, No. 9, Sep. 2013
Zhao X B, He J. Effects of yttrium doping on the thermoelectric properties of Hf0.6Zr0.4NiSn0.98Sb0.02 half-Heusler alloys. J. Appl. Phys., 2010, 108: 044903.
[8] Saiga Y, Suekuni K, Du B, Takabatake T. Thermoelectric
properties and structural instability of type-I clathrate
Ba8Ga16Sn30 at high temperatures. Solid State Commun.,
2012, 152: 1902.
[9] Peng J Y, Liu X Y, Fu L W, Xu W, Liu Q Z, Yang J Y.
Synthesis and thermoelectric properties of In0.2+xCo4Sb12+x
composite. J. Alloys Compd., 2012, 521: 141.

[10] Zhang R Z, Hu X Y, Guo P, Wang C L. Thermoelectric
transport coefficients of n-doped CaTiO3, SrTiO3 and BaTiO3: A theoretical study. Phys. B, 2012, 407: 1114.
[11] Sanmathi C S, Takahashi Y, Sawaki D, Klein Y, Retoux R,
Terasaki I, Noudem J G. Microstructure control on thermoelectric properties of Ca0.96Sm0.04MnO3 synthesized by
co-precipitation technique. Mater. Res. Bull., 2010, 45:
558.
[12] Wang Y, Sui Y, Wang X J, Su W H. Seebeck coefficient
of LnxCa1–xMnO3 perovskites in paramagnetic state. Appl.
Phys. A, 2011, 104: 135.
[13] Zhang F P, Lu Q M, Zhang X, Zhang J X. First principle
investigation of electronic structure of CaMnO3 thermoelectric compound oxides. J. Alloys Compd., 2011, 509:
542.
[14] Zhang F P, Zhang X, Lu Q M, Zhang J X, Liu Y Q. Electronic structure and thermal properties of doped CaMnO3
system. J. Alloys Compd., 2011, 509: 4171.
[15] Zhang F P, Zhang X, Lu Q M, Zhang J X, Liu Y Q.
Preparation, characterization and high temperature transport properties of Fe doped Ca1-xFexMnO3 oxides. 30th International Conference on Thermoelectrics. New Jersey:
IEEE Piscateway, 2011.
[16] Lu Q M, Zhang B X, Zhang F P, Zhang J X. Rietveld refinement and thermoelectric properties of ytterbium doped
Ca1–xYbxMnO3 (x=0–0.2) compounds. J. Chin. Soc. Rare
Earths (in Chin.), 2010, 28: 471.
[17] Wang Y, Sui Y, Su W H. High temperature thermoelectric
characteristics of Ca0.9R0.1MnO3 (R=La, Pr, …, Yb). J.
Appl. Phys., 2008, 104: 093703.
[18] Zhao L D, Zhang B P, Li J F, Zhang H L, Liu W S. Enhanced thermoelectric and mechanical properties in textured n-type Bi2Te3 prepared by spark plasma sintering.
Solid State Sci., 2008, 10: 651
[19] Mott N F, Davis E A. Electronic Processes in Non-crystalline Materials. Oxford: Clarendon Press, 1971.
[20] Schatz E R. Transition Metal Compounds. New York:
Gordon and Breach, 1963.
[21] Shi X, Hong H, Li C P, Uher C, Yang J, Salvador J R,
Wang H, Chen L, Zhang W. Low thermal conductivity

and high thermoelectric figure of merit in n-type BaxYby
Co4Sb12 double-filled skutterudites. Appl. Phys. Lett., 2008,
92: 182101.