Thermoelectric properties of GeSe · Thermoelectric properties of GeSe Xinyue Zhang, Jiawen Shen,...
Transcript of Thermoelectric properties of GeSe · Thermoelectric properties of GeSe Xinyue Zhang, Jiawen Shen,...
Available online at www.sciencedirect.com
ScienceDirect
J Materiomics 2 (2016) 331e337www.ceramsoc.com/en/ www.journals.elsevier.com/journal-of-materiomics/
Thermoelectric properties of GeSe
Xinyue Zhang, Jiawen Shen, Siqi Lin, Juan Li, Zhiwei Chen, Wen Li, Yanzhong Pei*
Key Laboratory of Advanced Civil Engineering Materials of Ministry of Education, School of Materials Science and Engineering, Tongji Univ., 4800 Caoan Rd.,
Shanghai 201804, China
Received 27 July 2016; revised 29 August 2016; accepted 5 September 2016
Available online 13 September 2016
Abstract
The recently reported superior thermoelectric performance of SnSe, motivates the current work on the thermoelectric properties of poly-crystalline GeSe, an analog compound with the same crystal structure. Due to the extremely low carrier concentration in intrinsic GeSe, variousdopants are utilized to substitute either Ge or Se for increasing the carrier concentration and therefore for optimizing the thermoelectric powerfactor. It is shown that Ag-substitution on Ge site is the most effective, which enables a hole concentration up to ~1018 cm�3. A further isovalentsubstitution by Pb and Sn leads to an effective reduction in the lattice thermal conductivity. A peak figure of merit, zT of ~0.2 at 700 K can beachieved in Ag0.01Ge0.79Sn0.2Se, a composition with the highest carrier concentration. The transport properties can be well described by a singleparabolic band model with a dominant carrier scattering by acoustic phonons at high temperatures (>500 K). This further enables a prediction onthe maximal zT of ~0.6 at 700 K and the corresponding carrier concentration of ~5 � 1019 cm�3.© 2016 The Chinese Ceramic Society. Production and hosting by Elsevier B.V. This is an open access article under the CC BY-NC-ND license(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Keywords: Thermoelectric; GeSe
1. Introduction
Thermoelectric materials, which can directly convert heatinto electricity based on either Seebeck or Peltier effect, hasdrawn increasing attention in the research community con-cerning the energy crisis [1]. The thermoelectric performanceof a material depends on the dimensionless figure of merit,zT ¼ S2T/r(kE þ kL), where S, r, T, kE, and kL represent theSeebeck coefficient, electrical resistivity, absolute tempera-ture, electronic thermal conductivity and lattice thermal con-ductivity, respectively. Materials possessing a high zT require ahigh power factor (S2/r) and a low thermal conductivity(kE þ kL). However, the strongly coupled S, r, and kE, lead tothe difficulty for obtaining high zT value [2].
Since S, r, and kE mentioned above are all closely related tothe carrier concentration (n), an optimization of n is
* Corresponding author.
E-mail address: [email protected] (Y. Pei).
Peer review under responsibility of The Chinese Ceramic Society.
http://dx.doi.org/10.1016/j.jmat.2016.09.001
2352-8478/© 2016 The Chinese Ceramic Society. Production and hosting by Elsevi
creativecommons.org/licenses/by-nc-nd/4.0/).
fundamentally required for realizing the maximal available zTin a given material [3e5]. To further enhance the maximal zT,proven strategies are typified either by enhancing the powerfactor (S2/r) through band engineering [6] including bandconvergence [7e11], band nestification [12], band effectivemass [13], or by minimizing the lattice thermal conductivity(kL), the only one independent material property. Several ap-proaches have been taken to reduce the lattice thermal con-ductivity, through alloying [9,14], nanostructuring [15e18],lattice anharmonicity [19,20], liquid-like transport behaviorscompounds [21] and more recently a low sound velocity [22].Providing the carrier concentration is optimized, these strate-gies have been demonstrated to be effective for advancing thethermoelectric performance in many materials.
As a new thermoelectric material, single-crystal SnSe hasrecently been reported to show an extraordinarily high ther-moelectric figure of merit at high temperatures (zT of 2.6 at973 K along the b axis), due to the ultralow thermal conduc-tivity [20]. With a further optimization in the carrier concen-tration by sodium doping on Sn site, single-crystal SnSe was
er B.V. This is an open access article under the CC BY-NC-ND license (http://
Table 1
Room temperature transport properties of GeSe doped with Cu, Na, La, and
As.
p-Type doping r (mU$cm) S (mV$K�1) k (W$m�1 K�1)
Ge1-xCuxSe x ¼ 2% 1.51 � 108 591.25 1.54
Ge1-xNaxSe x ¼ 2% 6.81 � 105 626.93 1.48
x ¼ 4% 2.99 � 107 631.24 1.40
x ¼ 6% 3.39 � 107 500.80 1.43
n-Type doping r (mU$cm) S (mV$K�1) k (W$m�1 K�1)
Ge1-xLaxSe x ¼ 2% 2.03 � 108 668.85 1.52
x ¼ 6% 4.24 � 108 441.10 1.03
GeSe1-xAsx x ¼ 2% 3.25 � 108 670.66 1.63
x ¼ 4% 3.99 � 107 614.29 1.55
332 X. Zhang et al. / J Materiomics 2 (2016) 331e337
reported to show a large increase in zT at moderate tempera-tures and therefore a high average zT (zTave) of ~1.3, ascompared with the zTave of only ~0.2 for pristine single-crystalSnSe [23,24]. Afterward, polycrystalline SnSe has also beenreported to have a low lattice thermal conductivity and a highzT (~0.8) [25e28].
These literatures send a message that high zT value can beprobably realized in the same orthorhombic bulk materials. Itis therefore important to explore new thermoelectric materialshaving the same or similar crystal structure with SnSe, sinceits low lattice thermal conductivity and high zT are probablyinherent to materials with this type of crystal structure orsimilar, particularly for those having constituent elements withsimilar chemical properties to SnSe.
Actually, all the layer-like IVeVI compounds, such asGeS, GeSe, SnS, and SnSe, crystallize in the GeS-typestructure at room temperature. This structure belongs to theorthorhombic family with a space group of D16
2h (Pnma)[29,30]. Among these IVeVI compounds, SnS and GeSeshow not only the same crystal structure but also very similarchemical composition with SnSe. The thermoelectric prop-erties of SnS have recently reported to show a decent zT of0.6 [31]. However, thermoelectric properties of GeSe have sofar been rarely reported, and most of the available workfocused on the theoretical calculations [32,33]. A minimumlattice thermal conductivity of 0.39 W m�1 K�1 is predictedfor GeSe, which is lower than that of SnS and GeS [33].Meanwhile, a large value of Seebeck coefficient and powerfactor can be achieved by proper p-type or n-type doping[33]. In another work [32], the predicted peak zT of 2.5 at800 K for GeSe along b axis due to a multiband effect, iseven higher than that of SnSe. All these finding indicatesGeSe should potentially be a superior thermoelectricmaterial.
This motivates the current work focusing on the measuredtransport properties of GeSe. The chemical doping for tuningthe carrier concentration, the transport properties and therelevant physics of polycrystalline of GeSe are discussed. Theresistivity of undoped GeSe is very large due to the extremelylow carrier concentration, as compared with conventionalthermoelectric materials [14,34]. This leaves a primary pur-pose to improving its electrical properties by increasing itscarrier concentration by doping. Therefore, various elements(Cu, Ag and Na for p-type, and Bi, Sb, La, As and I for n-type)are used to achieve a high enough carrier concentration. Itturns out that, among the above mentioned elements, only Agenables a carrier concentration up to ~1018 cm�3. Addition-ally, isovalent alloying with Sn and Pb on Ge site, is found toachieve an effectively reduction in the lattice thermal con-ductivity approaching the theoretical minimal estimated by theCahill model [35]. As a result of both doping and alloying, themeasured maximal zT is about 0.2 at 700 K. The transportproperties can be well described by a single parabolic bandmodel with a dominant carrier scattering of acoustic scatteringat high temperatures. This predicts a maximal zT of 0.6 at700 K for polycrystalline GeSe with an optimize carrier con-centration of ~5 � 1019 cm�3.
2. Experiment
Various dopants, including sodium (Na), iodine (I), anti-mony (Sb), lanthanum (La), arsenic (As), copper (Cu) andsilver (Ag), were used to dope GeSe to tune the carrier con-centration. Polycrystalline GeSe samples and alloys with PbSeand SnSe, with and without doping, were synthesized bymelting the stoichiometric quantities of high purity elements(>99.99%) sealed in silica ampoule at 1123 K for 7 h,quenching in cold water and then annealing at 853 K for 3days. The obtained ingots were hand ground into fine powdersfor X-ray diffraction (XRD) to identify the phase composition,and for hot press by induction heating at 800 K for 30 minunder a uniaxial pressure of ~70 MPa. The resulting pelletswith a density higher than 96% of the theoretical densityvalue, were ~12.0 mm in diameter and ~1 mm in thickness.
The electrical transport properties including resistivity,Seebeck coefficient and Hall coefficient of the pellet sampleswere simultaneously measured, under vacuum in the temper-ature range of 300e700 K. The resistivity and Hall coefficientwere measured using the van der Pauw technique under areversible magnetic field of 1.5 T. The Seebeck coefficient wasobtained from the slope of the thermopower versus tempera-ture gradients [36]. Thermal diffusivity (l) measured by a laserflash technique (the Netzsch LFA457 system) was used tocalculate the thermal conductivity, k ¼ dlCp, where d is thedensity measured by a mass/volume method and Cp isassumed to be a DulongePetit limit of heat capacity and to betemperature independent as well. The measurement uncer-tainty for S, r and k is 5% approximately.
Sound velocities were measured on the pellet samples atroom temperature, using pulse-receiver (Olympus-NDT)equipped with an oscilloscope (Keysight). Water and Shear gel(Olympus) were used as couplant during the measurements oflongitudinal and transverse sound velocity (vL and vS)respectively.
3. Results and discussion
The room temperature transport properties for dopedpolycrystalline GeSe with various dopants, excluding Ag, arelisted in Table 1. These elements are considered as ineffective
333X. Zhang et al. / J Materiomics 2 (2016) 331e337
dopants since the resistivity and Seebeck coefficient are toolarge to enable a high thermoelectric performance. It is foundthat Ag is the most effective dopant enabling a carrier con-centration up to ~1018 cm�3, probably due to the small elec-tronegative and atomic size differences between Ag and Ge.Therefore Ag-doped samples are focused on for the followingdiscussion.
The crystal structure of orthorhombic GeSe has been wellinvestigated previously [30]. The crystal structure of GeSe inthe room-temperature phase (Pnma) is shown in Fig. 1a, whichis a distorted NaCl-type structure. The primitive cell containseight atoms and two adjacent double layers. Each atom isconnected with three nearest neighbors through covalent bondin a single layer, creating zigzag chains along the b axis. It isknown that GeSe undergoes a first-order phase transition fromorthorhombic structure to NaCl-type structure at ~920e930 Kand remains cubic up to the melting point [37].
The XRD patterns for Ge1-xAgxSe (x ¼ 0%, 0.2%, 0.5%,1%, 1.5%, 3%) samples and Ge1-xAgxSe alloyed with PbSeand SnSe (x ¼ 1%) are shown in Fig. 1b. All the peaks can bewell indexed to orthorhombic GeSe structure without anyobservable peaks of impurities. The diffraction peaks of PbSe-and SnSe-alloyed samples shift toward the lower angle, indi-cating the increase of lattice constant.
Fig. 2 shows the temperature dependent Hall carrier con-centration (a) and Hall mobility (b) for Ge1-xAgxSe. It can beseen that Ag-doping can effectively increase the Hall carrierconcentration (nH) up to ~1018 cm�3. It is also shown that theHall carrier concentration of Ge1-xAgxSe quickly saturates atx ~0.002, meaning that nH does not increase with a furtherincrease in high nominal doping levels. It is seen thatalloying with Sn on Ge site can still enable a hall carrierconcentration of ~1018 cm�3 while alloying with Pb signifi-cantly reduces nH.
The Hall mobility (mH) for all Ag-doped samples show avery similar decrease with increasing temperature viamHfT�1.5 when temperature is above 500 K, indicating a
Fig. 1. (a) Crystal structure of orthorhombic GeSe, in which the blue and red at
Ge0.99Ag0.01Se-based alloys with PbSe and SnSe.
dominant mechanism of charge carriers scattering by acousticphonons at these temperatures. However, at lower tempera-tures (300e500 K), the Hall mobility rises with increasingtemperature via mHfT2.5 approximately, revealing additionalscattering mechanism which is believed to be due to grain-boundary potential barrier scattering. This observed increasein the Hall mobility with increasing temperature has been seenin SnSe, SnS and PbTe based materials [31,38,39].
According to the recent band structure calculation for GeSe[32], the energy difference between the first and secondvalence band maximum is small. Therefore it is in principleeasy to achieve a two-band conduction behavior for enhancingthe thermoelectric performance. It should keep in mind thatthe Fermi level needs to be deep in the valence band, in orderto involve the beneficial contribution from the second valenceband. This usually means that a heavily doping is required,which is unfortunately not achieved in this work due to lowdoping effectiveness of the dopants considered. In fact, theobtainable carrier concentration by Ag-doping is not highenough to position the Fermi level deep enough into thevalence band to enable a two-band conduction.
The band gap of 1.1 eV in GeSe [40,41] is much larger thanthat of conventional thermoelectric materials (0.2e0.5 eV forBi2Te3, PbTe and PbSe) [42e44], leading to a much weakerinteraction between the valence and conduction bands.Therefore, it is reasonable to approximate the band as para-bolic [45].
Therefore, the Seebeck coefficient at 700 K can be verywell described by a single parabolic band (SPB) model in theHall carrier concentration range obtained in this work(1017~1018 cm�3) as shown in Fig. 3a. It should be noted thatthe SPB model here assumes a dominant acoustic scattering,as evident from Fig. 2b. In other words, polycrystalline GeSein this work can be effectively approximated as a single bandconduction due to its low carrier concentration (only up to~1018 cm�3), although the band calculation indeed shows amultiband structure.
oms represent Ge and Se respectively. (b) XRD patterns for Ge1-xAgxSe and
Fig. 2. (a) Temperature dependent Hall carrier concentration for Ge1-xAgxSe and Ge0.99Ag0.01Se-based alloys with PbSe and SnSe. (b) Temperature dependent Hall
mobility for Ge1-xAgxSe.
Fig. 3. Hall carrier concentration dependent Seebeck coefficient (a) and Hall mobility (b) for Ge1-xAgxSe and its alloys at 700 K. The curves represent the
prediction based on the SPB model with a constant density-of-states effective mass (m*) of 1.04 me and a deformation potential coefficient (Edef) of ~17 eV.
334 X. Zhang et al. / J Materiomics 2 (2016) 331e337
The approximation of a single parabolic band conductioncan be further supported by the carrier concentrationdependent Hall mobility as shown in Fig. 3b. It is seen thatthe Hall mobility can be nicely predicted by the same SPBmodel as well, for all the Ag-doped GeSe samples obtainedin this work. When GeSe is further alloyed with PbSe (bluesymbols, concentrations of 8%, 10% and 14%) or SnSe (olivesymbols, concentrations of 10% and 20%), the mobility de-creases due to the additional scattering by substitutionaldefects.
The measured temperature Seebeck coefficient and elec-trical resistivity for Ge1-xAgxSe are shown in Fig. 4a and b,respectively. Seebeck coefficient for all the samples is posi-tive, indicating a p-type conduction. Seebeck coefficient de-creases with the increasing Ag content. Due to the low carrierconcentration, resistivity of the samples here is high, ascompared with traditional thermoelectrics. With increasingAg-doping level, the resistivity decreases by orders ofmagnitude. When further alloying with PbSe, both resistivityand Seebeck coefficient increase, because of the reduced Hallcarrier concentration as shown in Fig. 2a. The decrease inSeebeck coefficient and resistivity at T > 600 K is believed tobe due to the bipolar conduction.
The DulongePetit limit of heat capacity (Cp) and density(d ) of all samples are listed in Table 2.
Temperature dependent total thermal conductivity and itslattice component for Ge1-xAgxSe and its alloys are shown inFig. 4c and d. The lattice thermal conductivity (kL) is obtainedby subtracting the electronic thermal conductivity (kE) fromthe total thermal conductivity. The electronic thermal con-ductivity is estimated by kE ¼ LT/r according to the Wiede-manneFranz law, where L is the Lorenz factor determined bythe single parabolic band (SPB) model discussed above. Bothk and kL of all the samples decline as temperature rises. Itshould be noted that the electronic contribution to the totalthermal conductivity is less than 1% for all the samples in thiswork, even at temperatures that the bipolar conduction hap-pens. This is again due to the low carrier concentration.
The lattice thermal conductivity decreases as 1/T approxi-mately, indicating a dominant phonon scattering by Umklappprocesses in intrinsic GeSe and Ag-doped compounds. Ascompared with polycrystalline SnSe [27], GeSe obtained hereshows a very similar lattice thermal conductivity in theimportant temperature range of 500e700 K for thermoelectricapplications, which can be understood by the same crystalstructure between these two compounds.
Fig. 4. Temperature dependent Seebeck coefficient (a), resistivity (b), total thermal conductivity (c), thermal diffusivity (inset) and lattice thermal conductivity (d)
for Ge1-xAgxSe and its alloys. The thermal conductivity of polycrystalline SnSe [27] is also included for comparison.
Table 2
DulongePetit limit of heat capacity (Cp) and density (d ) of Ge1-xAgxSe and
Ge0.99Ag0.01Se-based alloys with PbSe and SnSe.
Samples Cp(J$g�1 K�1) d (g$cm�3)
x ¼ 0 0.3291 5.52
x ¼ 0.2% 0.3290 5.47
x ¼ 0.5% 0.3287 5.6
x ¼ 1% 0.3284 5.58
x ¼ 1.5% 0.3280 5.56
x ¼ 3% 0.3268 5.62
Ge0.89Ag0.01Pb0.1Se 0.3016 5.77
Ge0.79Ag0.01Sn0.2Se 0.3096 5.69
335X. Zhang et al. / J Materiomics 2 (2016) 331e337
The lattice thermal conductivity decreases with increasingconcentration of Ag-doping in the whole temperature range.Through a further alloying with PbSe or SnSe, the latticethermal conductivity is significantly reduced down to ~0.4 W/m-K at 700 K. With the measured average longitudinal(vL ¼ 3210 m/s) and transverse (vS ¼ 1952 m/s) sound ve-locities for polycrystalline samples obtained in this work, thetheoretical minimal lattice thermal conductivity (kL
min) is esti-mated to be ~0.4 W/m-K at high temperatures, based on theCahill model [35]:
kminL ¼�p6
�1 =
3
kBn2 =
3ð3vDÞ�
T
QD
�2 ZQD=T
0
x3ex
ðex � 1Þ2 dx ð1Þ
where kB is Boltzmann constant, n is the number density ofatoms and QD is the Debye temperature, QD ¼ vD (ħ/kB)(6p2n)1/3. And vD is average sound velocity determined by
vD ¼
0BBBB@
11
13
�1
v3L
þ 2
v3S
�
1CCCCA
1=3
ð2Þ
Noted that the sound velocity measurements here show wellagreement with previous calculations for single crystal GeSe[33].
This estimation on minimal lattice thermal conductivityshows an excellent agreement with the calculations [33]. Mostimportantly, the samples obtained here show a very compa-rable lattice thermal conductivity with the theoretical minimalvalue.
Fig. 5. Temperature dependent figure of merit, zT (a) for Ge1-xAgxSe and its alloys, and the predicted Hall carrier concentration dependent zT at 700 K (b).
336 X. Zhang et al. / J Materiomics 2 (2016) 331e337
Temperature-dependent figure ofmerit, zT is shown in Fig. 5a.Amaximal zTof 0.2 is achieved in Ge0.79Ag0.01Sn0.2Se.With themeasured average lattice thermal conductivity for these mate-rials, the SPB model enables a prediction on the carrier con-centration dependent zTat a given temperature. Shown in Fig. 5bis a case for 700 K. First of all, the model prediction agrees wellwith the experimental data. It is shown that a maximal zTof ~0.6is predicted, if the carrier concentration can be further increasedto ~5 � 1019 cm�3.
zT achieved so far in this compound is low, however, theband calculation [32] indicated a multiband structure, whichshould in principle enable a significantly enhanced zT if thematerial can be doped to have a carrier concentration of~1020 cm�3. It should be, once again, noted that such amultiband effect has not been taken into account in the pre-diction here, because it is unnecessary to include thiscomplexity in the carrier concentration range studied here.However, it is still believed that GeSe has a potential as apromising thermoelectric material, through a further carrierconcentration increase and/or band engineering.
4. Summary
In summary, Ag acts as an effective dopant on Ge site forincreasing the hole concentration, which results in anenhancement on power factor and therefore figure of merit, zT.A further alloying with Pb and Sn on Ge site can reduce thethermal conductivity effectively in the entire temperaturerange. The obtained lattice thermal conductivity is actuallyapproaching the theoretical minimum as estimated by theCahill model. The combined approach of both doping andalloying leads to a maximal zT of 0.2 at 700 K for GeSe. Thehigh temperature carrier concentration dependent transportproperties here can be well understood by a single parabolicband conduction with acoustic scattering, which further en-ables a prediction of a peak zT of 0.6 at 700 K. According tothe model prediction, the carrier concentration obtained in thiswork stays not fully optimized, suggesting available room forfurther improvements, especially by doping and band engi-neering approaches.
Acknowledgment
This work is supported by the National Natural ScienceFoundation of China (Grant No. 51422208, 11474219 and51401147) and the national Recruitment Program of GlobalYouth Experts (1000 Plan).
References
[1] Bell LE. Cooling, heating, generating power, and recovering waste heat
with thermoelectric systems. Science 2008;321:1457e61.[2] Snyder GJ, Toberer ES. Complex thermoelectric materials. Nat Mater
2008;7:105e14.
[3] Pei Y, Gibbs ZM, Balke B, Zeier WG, Snyder GJ. Optimum carrier
concentration in n-type PbTe thermoelectrics. Adv Energy Mater 2014;4:
1400486.
[4] Pei Y, May AF, Snyder GJ. Self-tuning the carrier concentration of PbTe/
Ag2Te composites with excess Ag for high thermoelectric performance.
Adv Energy Mater 2011;1:291e6.[5] Androulakis J, Todorov I, Chung DY, Ballikaya S, Wang GY, Uher C,
et al. Thermoelectric enhancement in PbTe with K or Na codoping from
tuning the interaction of the light- and heavy-hole valence bands. Phys
Rev B 2010;82:115209.
[6] Pei Y, Wang H, Snyder GJ. Band engineering of thermoelectric materials.
Adv Mater 2012;24:6125e35.
[7] Pei Y, Shi X, LaLonde A, Wang H, Chen L, Snyder GJ. Convergence of
electronic bands for high performance bulk thermoelectrics. Nature
2011;473:66e9.
[8] Liu W, Tan X, Yin K, Liu H, Tang X, Shi J, et al. Convergence of
conduction bands as a means of enhancing thermoelectric performance
of n-type Mg2Si1-xSnx solid solutions. Phys Rev Lett 2012;108:
166601.
[9] Li W, Chen Z, Lin S, Chang Y, Ge B, Chen Y, et al. Band and scattering
tuning for high performance thermoelectric Sn1�xMnxTe alloys.
J Materiomics 2015;1:307e15.
[10] Jian Z, Chen Z, Li W, Yang J, Zhang W, Pei Y. Significant band engi-
neering effect of YbTe for high performance thermoelectric PbTe.
J Mater Chem C 2015;3:12410e7.
[11] Pei Y, Wang H, Gibbs ZM, LaLonde AD, Snyder GJ. Thermopower
enhancement in Pb1-xMnxTe alloys and its effect on thermoelectric ef-
ficiency. NPG Asia Mater 2012;4:e28.
[12] Chattopadhyay D, Queisser HJ. Electron scattering by ionized impurities
in semiconductors. Rev Mod Phys 1981;53:745e68.
[13] Peng H, Wang CL, Li JC. Enhanced thermoelectric properties of
AgGaTe2 utilizing carrier concentration adjusting. Phys B Condens
Matter 2014;441:68e71.
337X. Zhang et al. / J Materiomics 2 (2016) 331e337
[14] Long D, Myers J. Ionized-impurity scattering mobility of electrons in
silicon. Phys Rev 1959;115:1107e18.
[15] Pei Y, Lensch-Falk J, Toberer ES, Medlin DL, Snyder GJ. High ther-
moelectric performance in PbTe due to large nanoscale Ag2Te pre-
cipitates and La doping. Adv Funct Mater 2011;21:241e9.[16] Biswas K, He J, Blum ID, Wu C-I, Hogan TP, Seidman DN, et al. High-
performance bulk thermoelectrics with all-scale hierarchical architec-
tures. Nature 2012;489:414e8.[17] Poudel B, Hao Q, Ma Y, Lan YC, Minnich A, Yu B, et al. High-ther-
moelectric performance of nanostructured bismuth antimony telluride
bulk alloys. Science 2008;320:634e8.
[18] Hsu KF, Loo S, Guo F, Chen W, Dyck JS, Uher C, et al. Cubic
AgPbmSbTe2þm: bulk thermoelectric materials with high figure of
merit. Science 2004;303:818e21.
[19] Morelli DT, Jovovic V, Heremans JP. Intrinsically minimal thermal con-
ductivity in cubic I-V-VI2 semiconductors. PhysRevLett 2008;101:035901.
[20] Zhao L-D, Lo S-H, Zhang Y, Sun H, Tan G, Uher C, et al. Ultralow
thermal conductivity and high thermoelectric figure of merit in SnSe
crystals. Nature 2014;508:373e7.[21] Liu H, Shi X, Xu F, Zhang L, Zhang W. Copper ion liquid-like ther-
moelectrics. Nat Mater 2012;11:422e5.
[22] Li W, Lin S, Ge B, Yang J, Zhang W, Pei Y. Low sound velocity
contributing to the high thermoelectric performance of Ag8SnSe6. Adv
Sci 2016:1600196.
[23] Zhao L-D, Tan G, Hao S, He J, Pei Y, Chi H, et al. Ultrahigh power factor
and thermoelectric performance in hole-doped single-crystal SnSe. Sci-
ence 2016;351:141e4.[24] Peng K, Lu X, Zhan H, Hui S, Tang X, Wang G, et al. Broad temperature
plateau for high ZTs in heavily doped p-type SnSe single crystals. En-
ergy Environ Sci 2016;9:454e60.[25] Li Y, Shi X, Ren D, Chen J, Chen L. Investigation of the anisotropic
thermoelectric properties of oriented polycrystalline SnSe. Energies
2015;8:6275e85.
[26] Sassi S, Candolfi C, Vaney JB, Ohorodniichuk V, Masschelein P,
Dauscher A, et al. Assessment of the thermoelectric performance of
polycrystalline p-type SnSe. Appl Phys Lett 2014;104:212105.
[27] Chen C-L, Wang H, Chen Y-Y, Day T, Snyder GJ. Thermoelectric
properties of p-type polycrystalline SnSe doped with Ag. J Mater Chem
A 2014;2:11171.
[28] Wei TR, Tan G, Zhang X, Wu CF, Li JF, Dravid VP, et al. Distinct impact
of alkali-ion doping on electrical transport properties of thermoelectric
p-type polycrystalline SnSe. J Am Chem Soc 2016;138:8875e82.
[29] Taniguchi M, Johnson RL, Ghijsen J, Cardona M. Core excitons and
conduction-band structures in orthorhombic GeS, GeSe, SnS, and SnSe
single crystals. Phys Rev B 1990;42:3634e43.[30] Okazaki A. The crystal structure of germanium selenide GeSe. J Phys
Soc Jpn 1958;13:1151e5.
[31] Tan Q, Zhao L-D, Li J-F, Wu C-F, Wei T-R, Xing Z-B, et al. Thermo-
electrics with earth abundant elements: low thermal conductivity and
high thermopower in doped SnS. J Mater Chem A 2014;2:17302e6.
[32] Hao S, Shi F, Dravid VP, Kanatzidis MG, Wolverton C. Computational
prediction of high thermoelectric performance in hole doped layered
GeSe. Chem Mater 2016;28:3218e26.
[33] Ding G, Gao G, Yao K. High-efficient thermoelectric materials: the case
of orthorhombic IV-VI compounds. Sci Rep 2015;5:9567.
[34] Pei Y, LaLonde A, Iwanaga S, Snyder GJ. High thermoelectric figure of
merit in heavy-hole dominated PbTe. Energ Environ Sci 2011;4:2085e9.
[35] Cahill DG, Watson SK, Pohl RO. Lower limit to the thermal conductivity
of disordered crystals. Phys Rev B 1992;46:6131.
[36] Zhou ZH, Uher C. Apparatus for Seebeck coefficient and electrical re-
sistivity measurements of bulk thermoelectric materials at high temper-
ature. Rev Sci Instrum 2005;76:023901.
[37] Wiedemeier H, Siemers P. The thermal expansion and high temperature
transformation of GeSe. Z Anorg Allg Chem 1975;411:90e6.
[38] Li Y, Li F, Dong J, Ge Z, Kang F, He J, et al. Enhanced mid-temperature
thermoelectric performance of textured SnSe polycrystals made of sol-
vothermally synthesized powders. J Mater Chem C 2016;4:2047e55.
[39] Martin J, Wang L, Chen L, Nolas GS. Enhanced Seebeck coefficient
through energy-barrier scattering in PbTe nanocomposites. Phys Rev B
2009;79:115311.
[40] Vaughn II DD, Patel RJ, Hickner MA, Schaak RE. Single-crystal
colloidal nanosheets of GeS and GeSe. J Am Chem Soc 2010;132:
15170e2.[41] Elkorashy AM. Photoconductivity in germanium selenide single crystals.
Phys Status Solidi B 1989;152:249e59.
[42] Goldsmid HJ. Introduction to thermoelectricity. Heidelberg: Springer;
2009.
[43] Ravich YI, Efimova BA, Smirnov IA. Semiconducting lead chalcogen-
ides. New York: Plenum Press; 1970. p. 377.
[44] Wang H, Pei Y, LaLonde AD, Snyder GJ. Heavily doped p-type PbSe
with high thermoelectric performance: an alternative for PbTe. Adv
Mater 2011;23:1366e70.
[45] Wolfe CM. Electron mobility in high-purity GaAs. J Appl Phys 1970;41:
3088.
Professor Yanzhong Pei, Tongji University, Email:
[email protected], Yanzhong Pei is a profes-
sor at Tongji University, China. He has been working
on advanced thermoelectric semiconductors for
about a decade, from synthesizing the materials to
understanding the underlying physics and chemistry.
He holds a B.E. from Central South University in
China, a Ph. D from Shanghai Institute of Ceramics,
CAS and postdoctoral research experience from
Michigan State University and the California Insti-
tute of Technology.