Theoretical Study of Ba₂X₆ (X = S, Se, Te) for Thermoelectric Applications Based on First-Principles Calculations and Machine Learning

Machine learning techniques such as support vector regression (SVR) and artificial neural network (ANN) algorithms are employed to screen binary barium family (BamXn) compounds with low lattice thermal conductivity. To this aim, the electronic structures and thermoelectric (TE) properties of Ba2X6 (X = S, Se, Te) are studied utilizing the first-principles density functional theory (DFT) together with the semiclassical Boltzmann transport theory (SBTT). The results indicate that Ba(2)Se(6 )and Ba2Te6 exhibit low lattice thermal conductivity (similar to 0.26 and 0.31 W m(-1) K-1) at room temperature. The electronic structure calculations show that three Ba2X6 (X = S, Se, Te) compounds are indirect band gap semiconductors with gap values of 1.37, 0.84, and 0.92 eV, respectively. The TE calculations show that the maximum ZT value of Ba2Te6 can be up to 3.51 at 800 K and 1.4 at 300 K. Our results indicate that all Ba2X6 (X = S, Se, Te) host good TE performances covering wide temperature ranges and have potential applications. In addition, the developed methods in this work may help to further search other materials of high TE performances.