WorksPhysics and Astronomy
Ultralow Thermal Conductivity and Thermoelectric Properties of Bi4GeTe7 with an Intrinsic van der Waal Heterostructure
Abstract
Ternary chalcogenides, having large crystalline unit cell and van der Waal stacking of layers, are expected to be poor thermal conductors and good thermoelectric (TE) materials. We are reporting that layered Bi4GeTe7, with alternating quintuplet-septuplet layers of Bi2Te3 and Bi2GeTe4, has an ultralow thermal conductivity, \k{appa}total 0.42 Wm-1K-1 because of high degree of anharmonicity as estimated from large Gruneisen parameter (γ 4.07) and low Debye temperature (θd 135 K). The electron dominated charge transport has been realized from the Seebeck coefficient, S - 82 uV/K, at 380 K, and Hall carrier concentration of ne ~ 9.8 x 1019 cm-3 at 300 K. Observation of weak antilocalization (WAL), due to spin-orbit coupling (SOC) of heavy Bi and Te, advocate Bi4GeTe7 to be a topological quantum material also. The cross-sectional transmission electron microscopy images show the inherent stacking of hetero-layers, which are leading to a large anharmonicity for poor phonon propagation. Thus, being a poor thermal conductor with a TE figure of merit, ZT ~ 0.24, at 380 K, the Bi4GeTe7 is a good material for TE applications.
Thermoelectric materials are important for harvesting waste thermal energy by directly converting heat into useful electricity.[1] The efficiency of TE materials are governed by the dimensionless figure of merit, , where is Seebeck coefficient, is electrical conductivity, is absolute temperature and denotes the total thermal conductivity which has contribution from electrons () as well as lattice ().[1] The involved physical parameters, except , are intricately interdependent and thus limiting the usage of TE materials for commercial applications. After understanding the challenges of optimizing the charge carriers in TE research, recently, materials with intrinsically ultralow have grabbed the scientific attention since it allows to modulate the with negligible rise in , independently.[2, 3, 4, 5] In general, materials having high degree of anharmonicity shows ultralow due to damping effects of phonon-phonon interactions, which plays an important role in engineering of their thermal properties.[5]
In TE materials with large unit cells, complex and layered crystal structures, the presence of large number of defects give a high degree of crystal anharmonicity hence reducing .[6, 3, 5] By adopting various methodologies such as nanostructuring, second phase nanoinclusions, mass contrast, soft phonons and phonon-glass/liquid-electron-crystal (PGEC or PLEC), a further reduction of can also be achieved.[3, 7, 8, 9, 10, 11, 12, 13] In addition, the heterostructured TE materials have shown a promise for significant enhancement in through (i) reduction of from enhanced phonon scattering and (ii) enhancement of from selective filtering of low energy charge carriers, simultaneously.[14] In literature, a has been reported in the thin film heterostructure of BiTe/SbTe through controlled manipulation of electrons and phonons.[15] A significant enhancement of the through low energy charge carrier filtering have been reported for BiTe/Te heterostructured thin films.[16] However, all these heterostructured TE compounds are artificially grown or chemical modified, thus required expensive and tedious ways for large scale productions.
Owing to their layered crystal structure and high degree of anharmonicity, the ternary chalcogenides, such as , , , , , and possesses ultralow . [5, 17, 18]With the narrow band gaps and layered crystal structures, the family of chalcogenides is known for the excellent properties and first generation topological insulators () also. [1, 7] The exotic surface states are arising because of the involvement of SOC of heavier metal like Bi/Sb/Pb/Sn, and Te. [19, 20, 21, 22] In this regard, is an important as well as material with inherent alternate repeating quintuplet-septuplet heterostructured crystal structure. While, the magneto transport studies show the presence of () and supports possibility of topological surface states, the existence of high degree of crystal anharmonicity and large Grüneisen parameter, , leading to an ultralow value of at 380 K. Our results provide subtle understanding of the structural aspects for as a potential material.
Polycrystalline was grown by a solid-state melting route using high purity chunks of Bi (99.99 %), Ge (99.999%) and Te (99.99%) in the stoichiometric ratio. The vacuum sealed ( mbar) quartz tube, with pure elements, was heated to K for h followed by water quenching. The obtained ingot was finely ground and pelletized into a mm dia. pellet, which was further vacuum sealed and subjected to an annealing at K for h. The annealing was done to improve the consolidation and homogenization of the compound. The annealed pellet was cut into a rectangular bar () for thermoelectric transport measurements.
Structural and phase analysis of the powder was carried out with a rotating anode Rigaku Smartlab X-ray diffractometer using radiation. High resolution transmission electron microscopy (HRTEM) was done using FEI make Tecnai G2 20 S-TWIN transmission electron microscope. Morphology and elemental mapping were identified using field-emission scanning electron microscopy (FESEM, Nova Nano SEM 450). Temperature dependent thermoelectric and magnetotransport properties were performed in a physical properties measurement system (PPMS, DynaCool, Quantum Design).
X-ray diffraction pattern of the finely ground powder of the annealed (Fig. 1(a)) have been indexed to Rhombohedral geometry (space group , 164).[23] Rietveld refinement confirms as the primary phase () with the inherent is identified as the secondary phase ().[24] The observation of secondary phase is quite likely due to the shorter annealing time during the synthesis.[25, 26] The refined lattice parameters for are Å and Å and the corresponding unit cell volume, Å.[27] A schematic in Fig. 1(b), depicts the natural arrangements of alternating quintuplets () and septuplets (), which are separated by van der Waal gaps. The cross sectional HRTEM image in Fig. 1(c) clearly showcase the quintuplet-septuplet heterostructure arrangements in . The global layered nature and hierarchical layered stacking is shown in SEM image (Fig. 1(d)).[5, 13] Elemental mapping performed on polished surface of the pellet (Figure 2), suggests absence of isolated secondary phase of in .

Figure 2. (a) FESEM image of top view of the pellet, and (b-f) the homogenous distribution of various elements observed in the mapping performed on polished surface of pellet.
The large unit cell is also a topological quantum material because of the presence of heavy elements like Bi and Te, which provides adequate spin-orbit-coupling for observation of signature metallic surface states in magnetotransport studies.[19] Hall analysis suggest electrons as the dominating charge carriers with a concentration, at 300 K. The magnetoresistivity, , is shown in Figure 3(a), where a sharp dip near the zero-field region providing a clue for SOC driven WAL, below 5K.[20, 28, 29, 30, 31] To understand further, we fitted the the magnetoconductivity data with Hikami-Larkin-Nagaoka (HLN) equation,

Figure 3. Magnetic field dependence of (a) longitudinal magnetoresistivity (), and (b) HLN fitted magnetoconductivity, and temperature dependence of (c) , (d) and (Inset ).
where is electronic charge, is a dimensionless fitting parameter called prefactor, is applied magnetic field, is phase coherence length and represents digamma function.[32, 33] From the fitting (Figure 3(b)), the estimated phase coherence length decreases from ~ 161 nm (2.5 K) to ~ 67 nm (10 K), which suggest the inelastic scattering of electrons from phonons. Generally, single conducting surface channel correspond to the ,[34] however a significant contribution of bulk conductivity reduces prefactor, from ~ - 22 (at 2.5 K) to ~ - 0.11 (at 10 K). Thus, the observation of WAL provides an evidence of involvement of bulk as well as topological surface states in .
To understand the potential for TE applications, temperature dependence of the involved physical parameters, for ZT, have been studied in detail. The positive temperature coefficient of resistivity in , indicates a metallic behavior or a narrow band gap degenerate nature of semiconductors, (Figure 3(c)), with a residual resistivity ratio () of ~ 1.87.[13] A relatively higher for , in comparison with ternary chalcogenides shows a higher crystal complexity involved and an additional resistance to the electron flow in the material.[5, 35] The , in Figure 3(d), further confirms the electron dominated charge transport with µV/K and correspondingly, the power factor (inset of Figure 3(d)), WmK at 380 K, has been determined.
The shows ultralow values, throughout the temperature range, with a WmK at 380 K (Figure 4(a)), which assures that has a potential for better TE material. The contribution of and in , is estimated using the modified Wiedemann Franz law: .[5, 36] Thus WmK and WmK, at 380 K, are extracted from temperature dependence of the Lorenz number, as shown in the inset of Figure 4(a). Here the interesting crystal structure of , with quintuplet-septuplet heterostructure having van der Waal’s gaps, has sufficient hindrance for phonon propagation as reflected from the ultralow values of . The room temperature comparison of ultralow values of of various layered chalcogenides such as , , , , and is shown in Figure 4(b), which showcases the importance of structural complexities of quintuplet-septuplet heterostructure in . [5, 13, 35]

Figure 4. (a) Temperature dependence of , , , and inset shows ), and (b) comparative graph for , and of various layered chalcogenides.[5, 13, 35] Temperature dependence of (c) (Inset: vs ) and (d) .
In TE materials, the higher crystalline anharmonicity drives phonon relaxation processes by damping the exchange of energy between thermal and mechanical vibrations, and thus resulting in ultralow . [5, 6, 37] In order to understand the origin of ultralow , the anharmonicity can be assessed from the heat capacity data by estimating the Debye temperature () and the Gruneisen parameter (), of the materials. The low temperature heat capacity () data is shown in Figure 4(c), and the linear fit (inset of Figure 4(c)), of , [38] gives the characteristic constants (~ ) related to electronic contributions and (~ ) related to lattice contributions in . The ~ 135.7 K has been estimated using, , where is the number of atoms per formula unit and is the molar gas constant. [17] The Grüneisen parameter, , is calculated using the relation: , where is the averaged mass of atoms of the material, is the volume per atom, represents the total number of atoms in the primitive unit cell and is a constant (the units of is in Å, is in and is in amu). [5, 17, 39] Here, possesses a high value of ~ 4.07 (at 380 K), which is better than champion and comparable to the TE materials known for ultralow thermal conductivity, such as , and . [3, 5] The low and large signify that has a high degree of anharmonicity. An additional insight on the nature of phonon propagation in can be drawn from the calculation of average phonon velocity, (). In general, materials with stiffer bonds tends to have higher and hence higher , whereas smaller corresponds to lower . For BiGeTe, km s, has been estimated using the equation , where and h represents Boltzmann constant and Planck’s constant, respectively, while denotes number of atoms per unit volume of the material. The for BiGeTe is considerably smaller when compared to ultralow materials such as BiGeTe ( km s) and BiTe ( km s).[5, 17] Therefore, intrinsically ultralow in BiGeTe can be understood as a result of anharmonicity arising due to large unit cell, intrinsic heterostructures, presence of heavy Bi atoms, large and small . Furthermore, the calculated at 380 K is (Figure 4(d)), which is relatively higher than the similar ternary layered chalcogenides.[5, 35] The results establish that BiGeTe has potential for good TE Material. Moreover, other members of Bi based ternary layered chalcogenides with higher structural complexities such as (Bi/Pb)GeTe, (Bi/Pb)GeTe and (Bi/Pb)GeTe can be expected to show ultralow and host topological surface states which upon proper tuning of the electrical properties can be good TEs and TIs.
In summary, the structural aspects of BiGeTe has been discussed for its magneto transport, TE transport properties. Here, the possibility of topological surface states have been understood by the nm (2.5 K) and an existence of WAL in magneto transport studies. With µV/K at 380 K, and cm at 300 K, the BiGeTe has an electron dominated charge transport like degenerate semiconductors. The intrinsic van der Waal heterostructure of quintuplet-septuplet layers provide a large anharmonicity and phonon scattering for an ultralow WmK at 380 K. Hence, BiGeTe is an important topological quantum material and promising TE material with high and an overall at 380 K.
NKS would like to acknowledge CSIR-SRF and AK acknowledges SERB for providing the fellowship. AS would like to acknowledge DST-SERB (Grant No. CRG/2018/002197). Authors thank IIT Mandi for the research facilities.

Figure 1. (a) Rietveld refinement of powder XRD pattern, (b) Schematic of the quintuplet-septuplet hetero-structure with vdW gaps of , (c) cross-sectional TEM image showcasing the
Figure 1 (PDF p. 12). stacked layers at atomic scales, and (d) FESEM image of fractured crystal showcasing global pictures of the layers.
References
- [1]D. M. Rowe, Thermoelectrics Handbook : Macro to Nano (CRC/Taylor & Francis, 2005).
- [2]P. Acharyya, T. Ghosh, K. Pal, K. Kundu, K. Singh Rana, J. Pandey, A. Soni, U. V. Waghmare and K. Biswas, Journal of the American Chemical Society 142 (36), 15595-15603 (2020).
- [3]S. Acharya, J. Pandey and A. Soni, ACS Applied Energy Materials 2 (1), 654-660 (2019).
- [4]J. Pandey, S. Mukherjee, D. Rawat, S. Athar, K. S. Rana, R. C. Mallik and A. Soni, ACS Applied Energy Materials 3 (3), 2175-2181 (2020).
- [5]N. K. Singh and A. Soni, Applied Physics Letters 117 (12), 123901 (2020).
- [6]J. P. Heremans, Nature Physics 11 (12), 990-991 (2015).
- [7]G. S. Nolas, J. Sharp and H. J. Goldsmid, Thermoelectrics: Basic Principles and New Materials Developments. (Springer, 2001).
- [8]M. G. Kanatzidis, Chem Mater 22 (3), 648-659 (2010).
- [9]S. Bathula, M. Jayasimhadri, B. Gahtori, N. K. Singh, K. Tyagi, A. K. Srivastava and A. Dhar, Nanoscale 7 (29), 12474-12483 (2015).
- [10]A. Soni, Y. Shen, M. Yin, Y. Zhao, L. Yu, X. Hu, Z. Dong, K. A. Khor, M. S. Dresselhaus and Q. Xiong, Nano Letters 12 (8), 4305 (2012).
- [11]A. Soni, Z. Yanyuan, Y. Ligen, M. K. K. Aik, M. S. Dresselhaus and Q. Xiong, Nano Letters 12 (3), 1203 (2012).
- [12]N. K. Singh, S. Bathula, B. Gahtori, K. Tyagi, D.Haranath and A. Dhar, Journal of Alloys and Compounds 668, 152-158 (2016).
- [13]N. K. Singh, J. Pandey, S. Acharya and A. Soni, Journal of Alloys and Compounds 746, 350-355 (2018).
- [14]Y. Zhang and G. D. Stucky, Chem Mater 26 (1), 837-848 (2014).
- [15]R. Venkatasubramanian, E. Siivola, T. Colpitts and B. O'Quinn, Nature 413 (6856), 597-602 (2001).
- [16]H.-J. Shang, F.-Z. Ding, Y. Deng, H. Zhang, Z.-B. Dong, W.-J. Xu, D.-X. Huang, H.-W. Gu and Z.-G. Chen, Nanoscale 10 (43), 20189-20195 (2018).
- [17]Q. Guo, J.-B. Vaney, R. Virtudazo, R. Minami, Y. Michiue, Y. Yamabe-Mitarai and T. Mori, Inorganic Chemistry 57 (9), 5258-5266 (2018).
- [18]C. Shekhar, C. E. ViolBarbosa, B. Yan, S. Ouardi, W. Schnelle, G. H. Fecher and C. Felser, Physical Review B 90 (16), 165140 (2014).
- [19]C. Fu, Y. Sun and C. Felser, Apl Mater 8 (4), 040913 (2020).
- [20]M. Neupane, S.-Y. Xu, L. A. Wray, Petersen, R. Shankar, N. Alidoust, Chang Liu, A. Fedorov, H. Ji, J. M. Allred, Y. S. Hor, T.-R. Chang, H.-T. Jeng, H. Lin, A. Bansil, R. J. Cava and a. M. Z. Hasan, Physical Review B 85, 235406 (2012).
- [21]E. Papalazarou, L. Khalil, M. Caputo, L. Perfetti, N. Nilforoushan, H. Deng, Z. Chen, S. Zhao, A. Taleb-Ibrahimi, M. Konczykowski, A. Hruban, A. Wołos, A. Materna, L. Krusin-Elbaum and M. Marsi, Physical Review Materials 2 (10), 104202 (2018).
- [22]V. L. Kuznetsov, L. A. Kuznetsova and D. M. Rowe, Journal of Physics D: Applied Physics 34 (5), 700 (2001).
- [23]V. L. Kuznetsov, L. A. Kuznetsova and D. M. Rowe, Journal of Applied Physics 85 (6), 3207-3210 (1999).
- [24]J. Rodríguez-Carvajal, Physica B: Condensed Matter 192 (1), 55-69 (1993).
- [25]L. A. Kuznetsova, V. L. Kuznetsov and D. M. Rowe, Journal of Physics and Chemistry of Solids 61 (8), 1269-1274 (2000).
- [26]P. P. Konstantinov, L. E. Shelimova, E. S. Avilov, M. A. Kretova and J. P. Fleurial, Journal of Solid State Chemistry 146 (2), 305-312 (1999).
- [27]O. G. Karpinsky, L. E. Shelimova, M. A. Kretova and J. P. Fleurial, Journal of Alloys and Compounds 265 (1), 170-175 (1998).
- [28]J. Chen, X. Y. He, K. H. Wu, Z. Q. Ji, L. Lu, J. R. Shi, J. H. Smet and Y. Q. Li, Physical Review B 83 (24), 241304 (2011).
- [29]S.-P. Chiu and J.-J. Lin, Physical Review B 87 (3), 035122 (2013).
- [30]L. Fu, C. L. Kane and E. J. Mele, Physical Review Letters 98 (10), 106803 (2007).
- [31]B. Bhattacharyya, B. Singh, R. P. Aloysius, R. Yadav, C. Su, H. Lin, S. Auluck, A. Gupta, T. D. Senguttuvan and S. Husale, Scientific Reports 9 (1), 7836 (2019).
- [32]S. Hikami, A. I. Larkin and Y. Nagaoka, Progress of Theoretical Physics 63 (2), 707-710 (1980).
- [33]N. K. Singh, D. Rawat, D. Dey, A. Elsukova, P. O. Å. Persson, P. Eklund, A. Taraphder and A. Soni, arXiv:2110.06587 [cond-mat.mtrl-sci] (2021).DOI
- [34]P. Sahu, J.-Y. Chen, J. C. Myers and J.-P. Wang, Applied Physics Letters 112 (12), 122402 (2018).
- [35]F. von Rohr, A. Schilling and R. J. Cava, Journal of Physics: Condensed Matter 25 (7), 075804 (2013).
- [36]H. S. Kim, Z. M. Gibbs, Y. L. Tang, H. Wang and G. J. Snyder, Apl Mater 3 (4), 041506 (2015).
- [37]C. W. Li, J. Hong, A. F. May, D. Bansal, S. Chi, T. Hong, G. Ehlers and O. Delaire, Nature Physics 11 (12), 1063-1069 (2015).
- [38]E. S. R. Gopal, Specific Heats at Low Temperatures. (Plenum Press, 1966).DOI
- [39]D. Bessas, I. Sergueev, H. C. Wille, J. Perßon, D. Ebling and R. P. Hermann, Physical Review B 86 (22), 224301 (2012).