Low temperature heat transport in crystalline bismuth telluride

Thermoelectric (TE) alloys based on Bi₂Te₃ are under intense development for current large-scale industrial applications. A quantitative understanding of their lattice thermal conductivity $ K_{lat} $ is trailing mainly for two reasons: (i) lack of reliable $ K_{lat} $ data and, surprisingly, (ii) lack of precise knowledge of the thermal properties of the parent compound, even in its crystalline form. In this work, we examine the existing data on crystalline Bi₂Te₃ and point out why data below its Debye temperature $ \Theta_D $ are necessary in order to develop a better understanding of heat transfer in the compound and alloys under current development. The measured temperature dependence of $ K_{lat} $ from 2–300 K is compared to a simplified heat transport model based on the relaxation time approximation, as well as to more recent and elaborate calculations. Most data on polycrystalline TE materials currently under development only cover a limited temperature range above their Debye temperature $ \Theta_D $. Yet, comparisons with hot extruded bulk polycrystalline Bi–Sb–Te–Se alloys with compositions close to Bi₂Te₃ stress the importance of obtaining data below $ \Theta_D $ in order to advance our understanding of the thermal properties of these materials so important for TE applications below 500 K. Attempting to understand and examine the thermal transport of solids based on the approximations involved in the pioneering work of the late 1950s is not justified except below their Debye temperature.