Computation-driven materials search for thermoelectric applications

The advancement of computational tools for material property predictions enables broad search of novel materials for various energy-related applications. However, challenges still exist in accurately predicting the mean free paths (MFPs) of electrons and phonons in a high-throughput frame for thermoelectric property predictions, which largely hinders the computation-driven search for novel materials. In this work, this need is eliminated under the small-grain-size limit, in which these MFPs are restricted by the grain sizes within a bulk material. The maximum ZT is anticipated when the grain size is reduced to the majority electron MFPs. Based on phonon dispersions and electronic band structures predicted by first-principles calculations, a ZT formulation for general nanograined bulk materials under the small-grain-size limit is proposed and is demonstrated with representative oxides.