ML-accelerated Prediction of Thermoelectric Semiconductors from Locally Sourced Wolframite

Thermoelectric materials are of practical interest for the conversion of waste heat into electricity and the development of efficient refrigeration systems. Tungsten-containing wolframite minerals are a promising class of semiconductors for thermoelectric applications. This project is focused on maximizing the thermoelectric efficiency of wolframite materials AWO(4) as a function of chemical composition (A = Zn, Mn, Fe, Co, Ni) using ML-accelerated atomistic simulations.