Embed Size (px)
Citation preview
RSC Advances
PAPER
Publ
ishe
d on
14
Aug
ust 2
014.
Dow
nloa
ded
by I
ndia
n In
stitu
te o
f T
echn
olog
y N
ew D
elhi
on
12/0
5/20
15 0
5:53
:04.
View Article OnlineView Journal | View Issue
Enhancing therm
CSIR-Network of Institutes for Solar Energy,
CSIR-National Physical Laboratory, Dr K.S.
E-mail: [email protected]; dakkmisra@
† Electronic supplementary informa10.1039/c4ra04889j
Cite this: RSC Adv., 2014, 4, 34552
Received 23rd May 2014Accepted 31st July 2014
DOI: 10.1039/c4ra04889j
www.rsc.org/advances
34552 | RSC Adv., 2014, 4, 34552–345
oelectric properties of a p-typeMg3Sb2- based Zintl phase compound by Pbsubstitution in the anionic framework†
A. Bhardwaj and D. K. Misra*
Mg3Sb2-based Zintl compounds have recently attracted attention as a potential candidate for
thermoelectric applications due to their low thermal conductivity and promising thermoelectric
performance (i.e. ZT ¼ 0.6 at 773 K in Mg3Sb2�xBix). We have reported previously that isoelectronic Bi3�
substitution of Sb3� leads to a moderate increase in the electrical conductivity, enhanced Seebeck
coefficient and reduced thermal conductivity. Herein, we report a large enhancement of the electrical
conductivity while maintaining the Seebeck coefficient by substituting Pb4� on Sb3� sites in Mg3Sb2�xPbx(0 # x # 0.3) alloys. Transport measurements reveal that optimum doping of 10 at% Pb4� on Sb3�
enhances the ZT to 0.84 at 773 K which is comparable to bismuth telluride and selenide industrial
materials which are toxic and expensive. The enhancement in ZT is attributed to a decrease in lattice
thermal conductivity and simultaneously an increase in the power factor resulting from the significant
increase in the electrical conductivity. We observe that Pb4� substitutions on Sb3� sites in Mg3Sb2�xPbx(0 # x # 0.3) increase the hole carrier concentration. Electronic transport data of Mg3Sb2�xPbx (0 # x #
0.3) alloys have been analyzed using a single parabolic band model and have been compared to Mg3Sb2.
The relatively high figure of merit and affordable material ingredients coupled with one step synthesis
process makes these materials a promising cost effective solution as thermoelectric materials.
1. Introduction
In recent years there has beena tremendous amount of effort toidentify high performance thermoelectric materials based oninexpensive and relatively abundant constituents for applica-tion in power generation and refrigeration in various electronicdevices.1,2 Many applications of thermoelectric devices can beenvisioned. However, their widespread implementation isrestricted because of the poor performance of thermoelectricmaterials. Thermoelectric material performance is primarilydetermined by a dimensionless quantity, the so called ther-
moelectric gure of merit, ZT ¼ a2s
kT, where a represents the
Seebeck coefficient, s is the electrical conductivity, T is theabsolute temperature and k (i.e. k ¼ kel + klattice) is the thermalconductivity accounting both the electrical (kel) and lattice (kl)contributions.3
Thus high ZT materials must possess large a, high s and lowk values and is rather difficult to optimize all these parameterssimultaneously in the bulk conventional crystalline materials.
Materials Physics & Engineering Division,
Krishnan Marg, New Delhi-110012, India.
gmail.com
tion (ESI) available. See DOI:
60
However, several reports show that a partial decouplingbetween these parameters can be achieved by adapting severalstrategies namely (1) doping mechanism4–7 (2) nanostructuring/nanocomposite8–12 and (3) ideally engineered materials13–17 withthe electronic properties of a crystal and low thermal conduc-tivity of a glass, the so called phonon glass electron crystal(PGEC). Following these concepts, in the last decade, manymaterials such as Bi2Te3,12,18–21 AgPbmSbTe2+m (LAST),14 TeAg-GeSb,15 PbTe,16,22 SiGe17,23 and Zintl phase compounds5,7,24–38
have been widely investigated for their high ZT. Among severalclasses of compounds, the Zintl phase is a unique class ofmaterials that possesses the structural characteristics neededfor phonon-glass electron-crystal properties (PGEC)39 leading toa high thermoelectric gure of merit (ZT).40,41 They combinedistinct regions of covalent bonding ideal for electron-crystalproperties and ionically bonded cations that can be easilysubstituted for precise tuning of electronic properties. Thisleads to the desired “electron crystal” behavior.40,42 The complexstructures and disorder creates phonon-glass propertiesresulting in low lattice thermal conductivity, making many Zintlcompounds natural “phonon glasses”.43,44 As a consequences ofthese behaviors, a varieties of Zintl phase compounds such asYb14MnSb11,24 Zn4Sb3,25 Yb9Mn4.2Sb9,26 lled skutterudites,27,28
clathrates,29,30 YbZn2Sb2,31 YbCd2Sb2,32 BaZn2Sb2,33 BaGa2Sb2,34
Eu5In2Sb6,35 Yb5In2Sb6,36 EuZn2Sb2,37 AZn2Sb238 etc. with highZT materials compared to other state-of-the-art thermoelectric
This journal is © The Royal Society of Chemistry 2014
Paper RSC Advances
Publ
ishe
d on
14
Aug
ust 2
014.
Dow
nloa
ded
by I
ndia
n In
stitu
te o
f T
echn
olog
y N
ew D
elhi
on
12/0
5/20
15 0
5:53
:04.
View Article Online
materials have been explored. However, the use of expensiverare earth elements and toxic in majority of Zintl phase mate-rials limits their use in large scale industrial application forpower generation and refrigeration.
Among several Zintl phase compounds, Mg3Sb2 basedmaterials involve a rare-earth free, non-toxic, cheap and abun-dant constituent which makes them economic and environ-mentally friendly utilization in power generation. Moreover,these compounds have been recognized as a potential candi-date for lithium battery,45 photoconduction,46 hydrogen trans-mitting47 and structural applications.48 A Zintl phase compoundof Mg3Sb2 crystallizes both in cubic bixbyite as well as inhexagonal structures.49 The cubic bixbyite structure is a-phaseof La2O3-type which is the high temperature phase, structuredwith 80 atoms unit cell with 48 Mg and 32 Sb atoms. At lowtemperature, below �1200 K, Mg3Sb2 crystallizes to Mn2O3 typeb-phase with unit cell composed of 5 atoms (3 Mg and 2 Sb). Theb-phase hexagonal structure (space group P�3m1 no. 164)consists of two inequivalent Mg sites, denoted as Mg(I) andMg(II) which are ionic and covalent in nature respectively. Thus,the bonding characteristic of Mg3Sb2 is in between metallic andionic.50,51 Similar to the structure of CaAl2Si2 structure, thestoichiometric Mg3Sb2 compound consists of interspersedMg2Sb2
2� layers (the tetrahedral position in the lattice) andMg2+ cation layers (the octahedral position in the lattice).52
The thermoelectric properties of Mg3Sb2 have been studiedby several groups.53–56 However, due to difficulties in synthe-sizing hexagonal single phase of Mg3Sb2 by using conventionalsynthesis routes (e.g. ball milling and furnace reaction followedby hot-pressing) presence of measurable content of oxygen atgrain boundaries,56 lack of micro structural details and pres-ence of impurity phases of C and MgC2 contaminated bygraphite die used in hot-pressing55 and more importantlyinconsistencies in the transport properties (e.g. ZTz 0.55 at 600K by Kajikawa et al.55 and ZTz 0.21 at 875 K by Condron et al.56)no signicant progress has been made towards its viability forthermoelectric applications. However, recently, Singh et al.57
theoretically investigated the detailed electronic structure andtransport properties of Zinlt phase Mg3Sb2 and a series of alloys(AeMg2) Pn2 (Ae ¼ Ca, Sr, Ba; Pn ¼ As, Sb, Bi) compounds inrelation to their thermoelectric performance. They claimed thatthe several promising compositions in this family are not fullyoptimized in terms of carrier concentration. The differences inelectronic structure studied theoretically by them, suggestobserving its accessibility by the experiments. Furthermore,several unanswered questions of fundamental nature as noticedfrom the previous studies,53–56 combined with the relatively highabundance of constituents make this class of materials quiteinteresting and addresses the detail investigation.
Recently, we have studied the thermoelectric properties ofsingle phase p-type Mg3Sb2 and its derivative of isoelectronic Bidoped; Mg3Sb2�xBix (0# x# 0.4) alloys.7 An enhanced ZTz 0.6at 773 K for p-type Mg3Sb1.8Bi0.2 was realized by Bi3� substitu-tion on Sb3�. Unfortunately, despite of high Seebeck coefficientand low thermal conductivity of Mg3Sb2, a moderate value ofelectrical conductivity was noted.7 Herein, we report anenhancement in the electrical conductivity while maintaining
This journal is © The Royal Society of Chemistry 2014
the Seebeck coefficient by substituting Pb4� on Sb3� site inMg3Sb2�xPbx (0 # x # 0.3) alloys. The optimum doping of 10at% Pb4� on Sb3� enhances the ZT to 0.84 at 773 K which is 40%and >200% larger over the Mg3Sb1.8Bi0.2 (ZT z 0.6 at 773 K)7
and parent Mg3Sb2 (ZT z 0.26 at 773 K)7 compound respec-tively. We observe that Pb4� substitutions on Sb3� site in Mg3-Sb2�xPbx (0 # x # 0.3) yields a best control of carrierconcentration for drastically improving the ZT. The hightemperature electronic and thermal transport measurementscombined with a single parabolic bandmodel have been used tocharacterize the thermoelectric properties of Mg3Sb2�xPbx (0 #
x # 0.3) alloys.
2. Experimental procedures
Stoichiometric amounts of high purity elements magnesium(Mg; 99.99%, Alfa Aesar), antimony (Sb; 99.99%, Alfa Aesar), andlead (Pb; 99.99%, Alfa Aesar) for synthesizing Mg3Sb2�xPbx (0#
x # 0.3) samples were blended in mechanical milling andsubsequently grounded in an agate mortar. The blendedpowders were then subjected to spark plasma sintering (SPS) attemperature 1073 K and a pressure of 50 MPa for holding timeof 10 minutes. The consolidated pellets were 12.7 mm diameterand 2.5 mm thick. The SPS were carried out by evacuating andashing the SPS chamber with Ar gas several times and nallySPS were performed in high vacuum to avoid the oxidation. SPSeliminates adsorptive gas and impurities existing on the surfaceof the powder particles which results to very clean samples. Thepresent synthesis strategy employing SPS involves simultaneousmelting and consolidation of the stoichiometric Mg3Sb2�xPbx(0 # x # 0.3) at high temperature and pressure followingcooling to form relatively very small grains of single Mg3-Sb2�xPbx composition phase.
The gross structural characterization of Mg3Sb2�xPbx (0 # x# 0.3) samples was carried out by powder X-ray diffractometer(Rigaku Mini Flex II) using a graphite monochromator andCuKa radiation with wavelength lz 1.5406 A along with CuKa2lter and rotating anode equipped with powder 2q diffractom-eter ranging from 20 to 80�. The microstructure and composi-tional analysis were investigated by eld emission scanningelectron microscopy (FE-SEM; Model: SUPRA40 VP, operating at30 kV) equipped with energy dispersive spectroscopy (EDS) andtransmission electron microscopy (TEM; Technai G2 T30; W-Twin) operating at 300 KV. The TEM specimens were preparedin three steps and described elsewhere.7
The polished SPSed pellets were used directly for thermaldiffusivity measurements parallel to the pressing direction.Specic heat was determined by a DSC instrument (822e MettlerToledo). The thermal conductivity of the sample was calculatedusing the relation k ¼ a � Cp � r, where k is the thermalconductivity, a the thermal diffusivity, r the geometrical pelletdensity and Cp the specic heat capacity. The thermal diffu-sivity, specic heat and density measured for all samples, arepresented in ESI S1, S2 and T1.† The polished bars of about 3 �2� 10mmwere cut from the consolidated disks and are used tomeasure the electrical conductivity and Seebeck coefficient in adirection perpendicular to the pressing direction. These bar
RSC Adv., 2014, 4, 34552–34560 | 34553
RSC Advances Paper
Publ
ishe
d on
14
Aug
ust 2
014.
Dow
nloa
ded
by I
ndia
n In
stitu
te o
f T
echn
olog
y N
ew D
elhi
on
12/0
5/20
15 0
5:53
:04.
View Article Online
samples were also used for the thermal conductivity measure-ment to verify the isotropy of thermal properties.
3. Results and discussion3.1 X-ray diffraction analysis
All the samples were identied for phase purity prior to anytransport properties measurement and X-ray diffraction patternperformed on the SPSed samples are shown in Fig. 2(a) for eachcomposition of Mg3Sb2�xPbx (0 # x # 0.3) alloys. In composi-tions with x < 0.30, all the reections can be indexed tob-Mg3Sb2 (JCPDS-00-003-0375) and no secondary phase isobserved (Fig. 2(a)). However, the sample with nominalcomposition of Mg3Sb1.7Pb0.3, reveals Pb as additional phasetogether with b-Mg3Sb2 whichmight have precipitated due to itslimited solubility in Mg3Sb2. The cell constants were estimatedby the POLSQ FORTRAN program58 and were found increasingwith increasing Pb concentration, being consistent with theVegard's law as shown in Fig. 2(b). Thus, all the samples ofMg3Sb2�xPbx (0 # x # 0.2) are solid solution phase. It is note-worthy that the increase in the lattice parameters is observed tobe consistent with the increasing ionic radii of Pb ions, whilethe broadening of the peaks is attributed to local strain due toheavier Pb substitution in the Sb sub lattice of Mg3Sb2 or due todisorder scattering. No additional peaks other than b-Mg3Sb2peaks in the XRD spectrum of Mg3Sb2�xPbx (0 # x # 0.2) wereobserved, conrming unambiguously the solubility of Pb in theanionic Sb sub-lattice. A solubility of 10 at% Pb in anionic Sbsite in Mg3Sb2 observed may be expected similar to otherreports.59–61
3.2 Scanning electron microscopy
In order to conrm the phase purity at microscopic scale,homogeneities and compositional analysis of b-Mg3Sb2�xPbx (0# x # 0.3), Field emission-Scanning electron microscopy (FE-SEM) investigation was carried out. The homogeneities of
Fig. 1 Schematic diagram of layered crystal structure of Mg3Sb2�xPbxshowing the anionic framework of [Mg2 (Sb/Pb)2]
2� with double layerand Mg2+ cations between the layers.
34554 | RSC Adv., 2014, 4, 34552–34560
samples Mg3Sb2�xPbx (0 # x # 0.2) were assessed by averagingthe compositions at 4 different regions of each sample obtainedby Energy dispersive X-ray analysis (EADX). The average value ofcomposition is shown in Table 1 marked as SEM-EDAXcompositions. All the samples show macroscopically homoge-neous as revealed by SEM-EADX analysis given in Table 1. Fig. 3represents SEM investigation of Mg3Sb2 and Mg3Sb1.8Pb0.2alloys. Morphological evidences as presented in Fig. 3(a) and (d)for Mg3Sb2 and Mg3Sb1.8Pb0.2 respectively show almost similarappearance of microstructures. Energy dispersive spectrummapping combined with SEM-EDAX was recorded to see qual-itatively the presence of any minor impurity other than Mg & Sband the results are shown in Fig. 3(b). It can clearly be seen thatonly Mg and Sb, were present in their respective concentrations.Fig. 3(c) represent SEM-EADX spectrum of Mg3Sb2 samplerecorded from a region marked by rectangle (Fig. 3(a)), clearlywitnessed the presence of only Mg and Sb. Quanticationresults estimated from the spectrum (inset Fig. 3(c)) clearlyreveals the Mg3Sb2 composition. Similar investigation such asSEM- imaging, SEM-mapping and SEM – EADX for Mg3Sb1.8-Pb0.2 have been presented in Fig. 3(d)–(f) respectively showinghomogeneities in terms of composition (see Table 1), all theelement; Mg, Sb & Pb in their respective concentration as shownin EDAX-mapping (Fig. 3(e)) and SEM-EDAX spectrum andquantication (Fig. 3(f)) with composition very close to thenominal composition of Mg3Sb1.8Pb0.2 alloy.
3.3 Transmission electron microscopy investigation
The detail microstructures at lattice scale of Mg3Sb2 and Mg3-Sb1.8Pb0.2 alloys have been carried out by high resolutiontransmission electron microscopy (HR-TEM). In general brighteld electron micrographs obtained from the specimen ofMg3Sb2 exhibits a polycrystalline structure throughout thevolume of the material (Fig. 4(a)). A corresponding selected areaelectron diffraction pattern (SAEDP) as shown in Fig. 4(b)reveals a set of Debye rings with ne sharp spots overlapping onindividual rings. The analysis of these rings reveals that thematerial consists of a single phase Mg3Sb2 with lattice planes,with lattice planes, hkl: 11�20, 20�20, 11�24, 21�33 having interplaner spacing of 0.229, 0.198, 0.142, 0.127 nm respectively,of hexagonal crystal structure with lattice parameter a ¼ 4.58 A,c ¼ 7.24 A (space group P�3ml). At low magnication (Fig. 4(a)),themicrograph shows grains with different sizes ranging from 8nm to 80 nm and signicant variation in grey contrast. Theelemental composition of Mg3Sb2 sample estimated fromenergy dispersive spectroscopy analysis (EDAX) attached withTEM (Fig. 4(c)) reveals the composition very close to thenominal composition of Mg3Sb2 alloy. Several lattice scaleimages were recorded to understand the presence of differentorientations of the crystallographic planes and their interfaceboundaries at atomic level. Fig. 4(d) presents high resolutiontransmission electron micrograph (HRTEM) image obtainedfrom sample showing several grains orientated in direction ofdifferent planes of Mg3Sb2 and several joint interface bound-aries. The micrograph (Fig. 4(d)) clearly reveals that the indi-vidual grains are truly crystalline with stacking of different
This journal is © The Royal Society of Chemistry 2014
Fig. 2 (a) X-ray diffraction (XRD) pattern of Mg3Sb2�xPbx (0 # x # 0.3) alloys. The XRD pattern reveals a single b-Mg3Sb2 type Zintl phase solidsolution upto x ¼ 0.2. The Pb peaks were noted for x ¼ 0.3 together with b-Mg3Sb2. (b) Plot for Vegard's law following linear trend of cellparameters with increasing Pb concentration showing complete solid solution phase formation of Mg3Sb2�xPbx (0 # x # 0.2) alloys.
Table 1 Contains the Hall measurement data and SEM-EDAX composition of Mg3Sb2�xPbx (0 # x # 0.3) alloys
Nominal compositionActual composition(SEM-EDAX)
Hall coefficient(RH) � 10�2 cm3 C�1
Carrier conc.n (1020 cm�3)
Mobility m(cm2 V�1 s�1)
Mg3Sb2 Mg60.29Sb39.71 5.02 1.2 11Mg3Sb1.95Pb0.05 Mg59.82Sb38.96Pb1.12 2.08 2.9 11.4Mg3Sb1.90Pb0.10 Mg59.96Sb37.98Pb2.06 1.67 3.6 11.7Mg3Sb1.80Pb0.20 Mg60.24Sb35.30Pb4.46 1.04 5.8 12.1Mg3Sb1.70Pb0.30 Mg60.34Sb33.65Pb6.01 0.39 15.5 12.9
Fig. 3 FE-SEM micrographs of (a) Mg3Sb2 parent compound showing polycrystalline nature of sample (b) elemental EDAX mapping of Mg3Sb2showing qualitatively the presence of constituent elements Mg and Sb (c) SEM-EADX of Mg3Sb2 confirming the composition, very close toMg3Sb2 as shown in the inset (d) SEM image of Mg3Sb1.8Bi0.2 revealing almost similar type morphology of Mg3Sb2 (e) elemental EDAX mappingimages, demonstrating the presence of all three constituent elements; Mg, Sn and Pb (f) SEM-EADX of Mg3Sb1.8Pb0.2 showing the samecomposition as the nominal composition (see inset).
Paper RSC Advances
Publ
ishe
d on
14
Aug
ust 2
014.
Dow
nloa
ded
by I
ndia
n In
stitu
te o
f T
echn
olog
y N
ew D
elhi
on
12/0
5/20
15 0
5:53
:04.
View Article Online
planes and with random orientation with respect to each other.A bright eld electron micrograph (Fig. 4(e)) corresponding tothe specimen of Mg3Sb1.8Pb0.2 shows a polycrystalline structure
This journal is © The Royal Society of Chemistry 2014
similar to Mg3Sb2 with relatively smaller grain sizes rangingfrom 5 nm to 60 nm. A corresponding SAEDP (not shown here)conrms b-Mg3Sb2 hexagonal structure. The lattice scale
RSC Adv., 2014, 4, 34552–34560 | 34555
Fig. 4 (a) TEM image obtained from the specimen of Mg3Sb2 showinghighly densified grains (b) SAED pattern corresponding to Mg3Sb2,revealing b-Mg3Sb2-type hexagonal structure, (c) EDS-TEM patternsrecorded fromMg3Sb2 confirming its exact composition. (d) The latticescale image of Mg3Sb2 exhibiting the presence of different orientationsof the crystallographic planes and their interface boundaries. (e)Bright field electron micrograph recorded from the specimen ofMg3Sb1.8Pb0.2 showing densely packed grains. (f) The lattice scaleimage of Mg3Sb1.8Pb0.2 demonstrating distorted lattices. HRTEMimage shown in the inset (f) exhibits misfit type of dislocation at theinterfaces marked by arrow.
RSC Advances Paper
Publ
ishe
d on
14
Aug
ust 2
014.
Dow
nloa
ded
by I
ndia
n In
stitu
te o
f T
echn
olog
y N
ew D
elhi
on
12/0
5/20
15 0
5:53
:04.
View Article Online
images reveals the randomly distributed grains with the interplaner spacing of the planes 112 (0.193 nm), 102 (0.27 nm) andmany other planes of Mg3Sb2 hexagonal crystal structure(Fig. 4(f)). Interestingly, the lattices associated with grains ofMg3Sb1.8Pb0.2 are observed to be little distorted together withsome mist-type of dislocations at the interfaces as marked byarrows in Fig. 4(f). The distortions in the lattices at microscopiclevel may be originated due to local strain due to substitution ofheavy metal Pb at Sb site in the structure of Mg3Sb2.
3.4 Electronic transport properties
To determine the effect of Pb4� substitution on Sb3� site on thethermoelectric properties of Mg3Sb2�xPbx, high temperatureelectrical conductivity, Seebeck coefficient and thermalconductivity were measured. Fig. 5(a) displays the temperaturedependence of electrical conductivity s(T), for different dopingconcentration of Pb in Mg3Sb2 alloy. Regardless of thetemperature, the electrical conductivity increases withincreasing Pb concentration. The s(T) increases monotonicallywith temperature for samples upto x # 0.2 over the entiretemperature range of 323 K to 773 K, showing semiconductingbehavior. Interestingly, for the case of Mg3Sb1.7Pb0.3, s(T)
34556 | RSC Adv., 2014, 4, 34552–34560
decreases with rising temperature up to �673 K showing ametallic behavior and further, above 673 K, it saturates indi-cating semi-metallic characteristics. Thus increasing Pbconcentration in Mg3Sb2 leads to a transition from semicon-ducting to metallic behavior. The room temperature measure-ments of the Hall coefficient (RH) were used to determine aHall carrier concentration (nH ¼ 1/RHe) for all x-values inMg3Sb2�xPbx ((0 # x # 0.3)). As observed in Fig. 5(b), carrierconcentration increases linearly with x up to x¼ 0.2 and furthera large increment of nH for x ¼ 0.3 was noted. The linearincrease in carrier concentration with x upto x ¼ 0.2 indicatesthat the Pb4� is indeed substituting on Sb3� site and a true solidsolution exists across series Mg3Sb2�xPbx (0# x# 0.2) which isalso conrmed by Vegard's law (Fig. 1(b)). The room tempera-ture electrical conductivity, and carrier concentration nH areused to calculate the room temperature mobility (m) by a rela-tion s ¼ nem and results are shown in Fig. 5(c). It is noted fromthe graph (Fig. 4(c)), that measured Hall nobilities at roomtemperature are around 11–13 cm2 V�1 s�1, and are not changedsignicantly, leading to the assumption that the carrierconcentration primarily dominates the electronic transport.
Fig. 5(d) and (e) represents the temperature dependent See-beck coefficient and effect of room temperature charge carrierson the Seebeck coefficient of Mg3Sb2�xPbx alloys respectively.The Seebeck coefficients of all the samples are positive, indi-cating holes as the majority carrier type, consistent with thepositive carrier concentration obtained from room temperatureHall measurement as shown in Table 1. A decrease in the See-beck coefficient with increasing carrier concentration(increasing x) is observed (Fig. 5(e)), following the n�2/3
dependence in the eqn (1) for a degenerate semiconductor withenergy independent scattering. The equation for the a(T)dependence on the carrier concentration can be described inthe heavy doping regime as62
a ¼ 8p2KB2
3eh2m*T
�p3n
�23 (1)
where m* is the effective mass, T, the temperature and n, thecarrier concentration. The linear dependence of a withtemperature is found only at low temperature as is assumed inthis model. The temperature dependent Seebeck coefficient ofMg3Sb2�xPbx (Fig. 5(d)) initially increases with temperature andattain a peak at 650 K. With further increasing temperature,beyond 650 K, the thermal excitation of electrons in presentcase begins to reduce the thermopower (Fig. 5(d)). We speculatethat the thermally excited electrons do not cause a reduction inthe Seebeck coefficient within the measurement range below650 K. However, a detail high temperature Hall effect over theentire temperature range should be carried out to understandthe exact mechanism of electronic transport. Thus, we observethat both s and a increases with increasing temperature. Theelectrical conductivity usually depends on both carrierconcentration (n) and mobility (m). With regard to increase a
and s, we infer that “n” may be decreasing but m would beincreasing with temperature. The simultaneous increase insigma and S is not usually expected for semiconductors andwould require band structure information as well as high
This journal is © The Royal Society of Chemistry 2014
Fig. 5 (a) Temperature dependence of the electrical conductivity of Mg3Sb2�xPbx (b) room temperature carrier concentration with increasing x(c) room temperature Hall mobility with increasing x (d) temperature dependence of the Seebeck coefficient, a(T); (e) Pisarenko plot at 300 Kshowing the dependence of Seebeck coefficient on the carrier concentration (solid line). The experimental data lie on or near the curve (singleparabolic band model) generated for m* ¼ 1.96 me for x ¼ 0.2, suggesting all the samples have almost equal effective masses (f) temperaturedependent power factor, (sa2(T)).
Fig. 6 Temperature dependence behaviour of the total thermal
Paper RSC Advances
Publ
ishe
d on
14
Aug
ust 2
014.
Dow
nloa
ded
by I
ndia
n In
stitu
te o
f T
echn
olog
y N
ew D
elhi
on
12/0
5/20
15 0
5:53
:04.
View Article Online
temperature Hall measurement to have betterunderstanding.
The effective mass (m*) of Mg3Sb2�xPbx (0 # x # 0.2) werecalculated from carrier concentrations (n) using eqn (1) andslope of Seebeck coefficient versus temperature plot. We noticedthat effective mass remains almost unchanged across the solidsolution of Mg3Sb2�xPbx (0# x# 0.2) which is not shown here.Further, assuming only one type of carrier at lower temperaturewith the assumption of the acoustic phonon scattering (l¼ 0), asingle parabolic model may be expected. The effective mass m*
¼ 1.96 me was used to calculate the Pisarenko relation at roomtemperature (Fig. 5(e)). The effective mass m* ¼ 1.96 me hasbeen calculated from the experimental Seebeck and Hall carrierconcentration for x ¼ 0.20 at room temperature using eqn (1).
As evident from the Pisarenko plot shown in Fig. 5(e), theexperimental Seebeck coefficient a and nH fall on or near thecurve corresponding tom*¼ 1.96me for x¼ 0.2, suggesting thata single parabolic band may be in good agreeing model forelectronic transport of Mg3Sb2�xPbx system at low temperatureswhere only one type of carrier exists. Based on the above anal-ysis at low temperature, one can infer that substitution of Pbdoesn't signicantly alter the effective mass or mobility of theholes and thereby conrming the electronic transport is mainlymonitored by the charge carrier densities.
The temperature dependence behavior of power factor ofMg3Sb2�xPbx (0 # x # 0.3) is plotted in Fig. 5(f). Regardless oftemperature, the power factor increases due to large increase inthe electrical conductivity with increasing Pb concentration.However, with increasing temperature, the power factor for allthe samples except x ¼ 0.3 increases with rising temperatureand maximized at 673 K. The highest power factor is optimized
This journal is © The Royal Society of Chemistry 2014
for Mg3Sb1.8Pb0.2 at 673 K which is 45% larger than the parentMg3Sb2 compound. We believe that Pb4� substitution on Sb3�
site yields a best control over electrical conductivity and Seebeckcoefficient in order to optimize high power factor.
3.5 Thermal transport properties
Fig. 6 shows the temperature dependence of thermal conduc-tivity k(T) of Mg3Sb2�xPbx alloys. As mentioned in the experi-mental section that the thermal conductivity are measured in adirection parallel to the pressing direction which is perpen-dicular to the direction in which the electronic transport is
conductivity k(T) of Mg3Sb2�xPbx (0 # x # 0.3).
RSC Adv., 2014, 4, 34552–34560 | 34557
RSC Advances Paper
Publ
ishe
d on
14
Aug
ust 2
014.
Dow
nloa
ded
by I
ndia
n In
stitu
te o
f T
echn
olog
y N
ew D
elhi
on
12/0
5/20
15 0
5:53
:04.
View Article Online
measured. Additionally, the thermal conductivity measure-ments were performed on the rectangular bar specimens whichwere used for electrical transport. We nd that the variation ofabout 5% in thermal conductivity, measured in parallel andperpendicular direction. This difference is not signicant andlies within the equipment error suggesting near isotropic natureof sample. The isotropic nature could be attributed to the nano-sized powder of Mg3Sb2 and Mg3Sb1.8Pb0.2 similar to thebehavior in layered structure of Bi2Te3 nano-sized powderreported12,21,63 where a variations of 5–10% in k have beendescribed and attributed to isotropic nature of Bi2Te3 sample innanosized powder sample. Off course micron size powderalways exhibit anisotropy which has been reported in layeredstructure of Bi2Te3 material.64 Interestingly, the total thermalconductivity decreases with increasing Pb concentration inMg3Sb2 alloy regardless to the temperature as shown in Fig. 6(a).Moreover, the total thermal conductivity, k decreases withincreasing temperature indicating 1/T type behavior, which iscommonly occurred in the bulk crystalline solids. The lowestvalue of thermal conductivity is 0.23 W m�1 K�1 for x ¼ 0.3 wasobserved which is attributed to mass uctuations and grainboundary scattering due to increasing the Pb concentration. Wehave also veried the variation in thermal conductivity data bymeasuring the samples with thickness (�2.5 mm and 1 mm) tosee the effect of heat dissipation on the thermal conductivity.However, only a little variation of 4–5% in kwas observed, whichis presented in ESI S3.†
The temperature dependence of thermoelectric gure ofmerit, ZT of Mg3Sb2�xPbx alloys is shown in Fig. 7. Themaximum ZT z 0.84 at 773 K was optimized for Mg3Sb1.8Pb0.2which is about >200% larger than the ZT value observed inparent Mg3Sb2 compound and 40% larger over Mg3Sb1.8Bi0.2 ofour previous report. The enhanced ZT is resulted from thesignicant increase in the electrical conductivity at littleexpense of Seebeck coefficient with further simultaneousdecrease in the total thermal conductivity.
Fig. 7 Temperature dependence of thermoelectric figure of merit ofMg3Sb2�xPbx (0 # x # 0.3).
34558 | RSC Adv., 2014, 4, 34552–34560
4. Conclusions
A single solid solution Zintl phase of Mg3Sb2�xPbx (0# x# 0.2)was ascertained by characterizing the specimens employingXRD, FE-SEM and TEM investigation. Substitution of Pb4� onSb3� site introduces holes as charge carriers, ascribed by Hallmeasurement data, and results into large p-type electricalconductivity. It was noted that the Pb substitution yields a bestcontrol over thermoelectric properties and does not signi-cantly alter the effective mass or mobility and thereby conningthe electronic properties to be mainly monitored by chargecarrier densities. Electronic transports data of Mg3Sb2�xPbx(0# x # 0.2) alloys have been analyzed using a single parabolicband model at low temperature. The peak ZT z 0.84 at 773 Khas been achieved in Mg3Sb1.8Pb0.2 alloy which is about >200%larger than the ZT value observed in parent Mg3Sb2 compound.The enhancement in ZT is due to increase in power factorresulted primarily by large increase in the electrical conductivityand with simultaneous decrease in the thermal conductivity.Additional enhancement in ZT could be expected to increase thepower factor and reducing the lattice thermal conductivity bysuitable doping. The present ZT value is comparable to withbismuth tellurides and selenides industrial materials which aretoxic and expensive. Further, Mg3Sb2-based Zintl compoundsbeing free from expensive rare earth elements, makes thesematerials cost-effective, environment friendly and abundant forthe use of power generation.
Acknowledgements
This work was nancially supported by CSIR-TAPSUN (CSIR-NWP-54) programme. The authors thank Prof. R. C. Budhaniand Dr Jiji Pullikotil for useful discussions and comments. Weacknowledge Dr Ajay Dhar (HOD, Metals and Alloys Group) forproviding facilities and for his critical comments. Dr SunilPandey (NIMS University, Jaipur) is highly acknowledged forproviding the room temperature Hall measurement equipment.One of the authors AB greatly acknowledges UGC-CSIR fornancial support. We thank Radheshyam and Naval K. Upad-hyay for technical and experimental support.
References
1 B. Raton, CRC Handbook of Thermoelectrics, CRC, Florida,USA, 1995.
2 H. J. Goldsmid, Thermoelectric Refrigation, Temple PressBooks Ltd., London, 1964.
3 G. J. Snyder and E. S. Toberer, Nat. Mater., 2008, 7, 105.4 Y. Pei, A. D. LaLonde, S. Iwanaga and G. J. Snyder, EnergyEnviron. Sci., 2011, 4, 2085–2089.
5 J. J. Pulikkotil, D. J. Singh, S. Auluck, M. Saravanan,D. K. Misra, A. Dhar and R. C. Budhani, Phys. Rev. B:Condens. Matter Mater. Phys., 2012, 86, 155204.
6 H. Wang, Y. Pei, A. D. LaLonde and G. J. Snyder, Adv. Mater.,2011, 23, 1366–1370.
This journal is © The Royal Society of Chemistry 2014
Paper RSC Advances
Publ
ishe
d on
14
Aug
ust 2
014.
Dow
nloa
ded
by I
ndia
n In
stitu
te o
f T
echn
olog
y N
ew D
elhi
on
12/0
5/20
15 0
5:53
:04.
View Article Online
7 A. Bhardwaj, A. Rajput, A. K. Shukla, J. J. Pulikkotil,A. K. Srivastava, A. Dhar, G. Gupta, S. Auluck, D. K. Misraand R. C. Budhani, RSC Adv., 2013, 3, 8504.
8 Y. Pei, J. L. Falk, E. S. Toberer, D. L. Medlin and G. J. Snyder,Adv. Funct. Mater., 2011, 21, 241–249.
9 D. K. Misra, A. Bhardwaj and S. Singh, J. Mater. Chem. A,2014, 2, 11913–11921.
10 A. Bhardwaj, D. K. Misra, J. J. Pulikkotil, S. Auluck, A. Dharand R. C. Budhani, Appl. Phys. Lett., 2012, 101, 133103.
11 A. J. Minnich, M. S. Dresselhaus, Z. F. Ren and G. Chen,Energy Environ. Sci., 2009, 2, 466.
12 B. Poudel, Q. Hao, Y. Ma, Y. Lan, A. Minnich, B. Yu, X. Yan,D. Wang, A. Muto, D. Vashaee, X. Chen, J. Liu,M. S. Dresselhaus, G. Chen and Z. Ren, Science, 2008, 320,634.
13 Y. Pei, H. Wang and G. J. Snyder, Adv. Mater., 2012, 24, 6124.14 K. F. Hsu, S. Loo, F. Guo, W. Chen, J. S. Dyck, C. Uher,
T. Hogan, E. K. Polychroniadis and M. G. Kanatzidis,Science, 2004, 303, 818.
15 J. Androulakis, K. F. Hsu, R. Pcionek, H. Kong, C. Uher,J. J. D'Angelo, A. Downey, T. Hogan and M. G. Kanatzidis,Adv. Mater., 2006, 18, 1170.
16 K. Ahn, K. Biswas, J. He, I. Chung, V. Dravid andM. G. Kanatzidis, Energy Environ. Sci., 2013, 6, 1529.
17 G. Joshi, H. Lee, Y. Lan, X. Wang, G. Zhu, D. Wang,R. W. Gould, D. C. Cuff, M. Y. Tang, M. S. Dresselhaus,G. Chen and Z. F. Ren, Nano Lett., 2008, 8, 4670.
18 S. Yu, J. Yang, Y. Wu, Z. Han, J. Lu, Y. Xie and Y. Qian, J.Mater. Chem., 1998, 8, 1949.
19 J. Shen, T. Zhu, X. Zhao, S. Zhang, S. Yanga and Z. Yina,Energy Environ. Sci., 2010, 3, 1519.
20 M. E. Anderson, S. S. N. Bharadwaya and R. E. Schaak, J.Mater. Chem., 2010, 20, 8362.
21 S. Sumithra, N. J. Takas, D. K. Misra, W. M. Nolting,P. F. P. Poudeu and K. L. Stokes, Adv. Energy Mater., 2011,1, 1141.
22 J. R. Sootsman, H. Kong, C. Uher, J. J. D'Angelo, C. I. Wu,T. P. Hogan, T. Caillat and M. G. Kanatzidis, Angew. Chem.,Int. Ed., 2008, 47, 8618.
23 R. Basu, S. Bhattacharya, R. Bhatt, M. Roy, S. Ahmad,A. Singh, M. Navaneethan, Y. Hayakawa, D. K. Aswal andS. K. Gupta, J. Mater. Chem. A, 2014, 2, 6922.
24 S. R. Brown, S. M. Kauzlarich, F. Gascoin and G. J. Snyder,Chem. Mater., 2006, 18, 1873.
25 G. J. Snyder, M. Christensen, E. Nishibori, T. Caillat andB. B. Iversen, Nat. Mater., 2004, 3, 458.
26 S. K. Bux, A. Zevalkink, O. Janka, D. Uhl, S. Kauzlarich,J. G. Snyder and J. P. Fleurial, J. Mater. Chem. A, 2014, 2, 215.
27 B. C. Sales, D. Mandrus and R. K. Williams, Science, 1996,272, 1325.
28 J. W. Graff, X. Zeng, A. M. Dehkordi, J. He and T. M. Tritt, J.Mater. Chem. A, 2014, 2, 8933–8940.
29 G. S. Nolas, J. L. Cohn, G. A. Slack and S. B. Schujman, Appl.Phys. Lett., 1998, 73, 178.
30 H. Kleinke, Chem. Mater., 2010, 22, 604.31 F. Gascoin, S. Ottensmann, D. Stark, S. M. Haile and
G. J. Snyder, Adv. Funct. Mater., 2005, 15, 1860.
This journal is © The Royal Society of Chemistry 2014
32 X. J. Wang, M. B. Tang, H. H. Chen, X. X. Yang, J. T. Zhao,U. Burkhardt and Y. Grin, Appl. Phys. Lett., 2009, 94, 092106.
33 X. J. Wang, M. B. Tang, J. T. Zhao, H. H. Chen and X. X. Yang,Appl. Phys. Lett., 2007, 90, 232107.
34 S. J. Kim and M. G. Kanatzidis, Inorg. Chem., 2001, 40, 3781.35 S. M. Park, E. S. Choi, W. Kang and S. J. Kim, J. Mater. Chem.,
2002, 12, 1839.36 S. J. Kim, J. R. Ireland, C. R. Kannewurf andM. G. Kanatzidis,
J. Solid State Chem., 2000, 155, 55.37 H. Zhang, J. T. Zhao, Y. Grin, X. J. Wang, M. B. Tang,
Z. Y. Man, H. H. Chen and X. X. Yang, J. Chem. Phys., 2008,129, 164713.
38 G. S. Pomrehn, A. Zevalkink, W. G. Zeier, A. van deWalle andG. J. Snyder, Angew. Chem., Int. Ed., 2014, 53, 3422.
39 G. A. Slack, in CRC Handbook of Thermoelectrics, CRC, NewYork, 1995.
40 S. M. Kauzlarich, Chemistry, Structure, and Bonding of ZintlPhases and Ions, VCH Publishers, New York, 1996.
41 S. M. Kauzlarich, S. R. Brown and G. J. Snyder, Dalton Trans.,2007, 2099–2107.
42 A. M. Mills, R. Lam, M. J. Ferguson, L. Deakin and A. Mar,Coord. Chem. Rev., 2002, 233, 207.
43 G. A. Slack, Solid State Physics, Academic Press, New York,1979.
44 E. S. Toberer, A. Zevalkink and G. J. Snyder, J. Mater. Chem.,2011, 21, 15843.
45 H. Honda, H. Sakaguchi, I. Tanaka and T. Esaka, J. PowerSources, 2003, 123, 216.
46 P. Singh and K. K. Sarkar, Solid State Commun., 1985, 55, 439.47 N. Nishimiya, A. Suzuki and S. Ono, Int. J. Hydrogen Energy,
1982, 7, 741.48 J. C. Viala, F. Barbeau, F. Bosselet and M. Peronnet, J. Mater.
Sci. Lett., 1998, 17, 757.49 M. M. Ripoll, A. Haase and G. Brauer, Acta Crystallogr., Sect.
B: Struct. Sci., Cryst. Eng. Mater., 1974, B30, 2006.50 J. H. Slowik, Phys. Rev. B: Solid State, 1974, 10, 416.51 L. M. Watson, C. A. W. Marshall and C. P. Cardoso, J. Phys. F:
Met. Phys., 1984, 14, 113.52 C. Zheng, R. Hoffmann, R. Nesper and H. G. von Schnering,
J. Am. Chem. Soc., 1986, 108, 1876.53 J. H. Bredt and L. F. Kendall, Proceedings—IEEE/AIAA, 1966.54 D. M. Verbrugge and J. B. J. Van Zytveld, J. Non-Cryst. Solids,
1993, 736, 156.55 T. Kajikawa, N. Kimura and T. Yokoyama, Proceedings of the
22nd International Conference on Thermoelectrics, 2003, p.305.
56 C. L. Condron, S. M. Kauzlarich, F. Gascoin and G. J. Snyder,J. Solid State Chem., 2006, 179, 2252.
57 D. J. Singh and D. Parker, J. Appl. Phys., 2013, 114, 143703.58 D. Keazler, D. Cahen and J. Lbers, POLSQ FORTRAN program,
Northwestern University Evanston, IL, 1984.59 X. Shi, Y. Pei, G. J. Snyder and L. Chen, Energy Environ. Sci.,
2011, 4, 4086.60 G. J. Tan, L. W. Liu, C. H. Chi, X. L. Su, S. Y. Wang, Y. G. Yan,
X. F. Tang, W. Wong-Ng and C. Uher, Acta Mater., 2013, 61,7693–7704.
RSC Adv., 2014, 4, 34552–34560 | 34559
RSC Advances Paper
Publ
ishe
d on
14
Aug
ust 2
014.
Dow
nloa
ded
by I
ndia
n In
stitu
te o
f T
echn
olog
y N
ew D
elhi
on
12/0
5/20
15 0
5:53
:04.
View Article Online
61 A. F. May, E. F. Larsen and G. J. Snyder, Phys. Rev. B: Condens.Matter Mater. Phys., 2010, 81, 125205.
62 M. Cutler, J. F. Leavy and R. L. Fitzpatrick, Phys. Rev., 1964,133, A1143.
34560 | RSC Adv., 2014, 4, 34552–34560
63 Y. Ma, Q. Hao, B. Poudel, Y. Lan, B. Yu, D. Wang, G. Chenand Z. F. Ren, Nano Lett., 2008, 8, 2580–2584.
64 J. Jiang, L. D. Chen, S. Bai, Q. Yao and Q. Wang, Scr. Mater.,2005, 52, 347–351.
This journal is © The Royal Society of Chemistry 2014
