Qimin Yan

Sponsor: DOE

Synthesis of motif and symmetry for accelerated learning, discovery, and design of electronic structures for energy conversion applications

The goal of the proposed work is to develop the data-driven approach based on structure motifs and orbital symmetries to discovery and design inorganic semiconductor compounds with optimal electronic structures for energy conversion applications. Objectives are: (i) to develop a framework toward the universal description of structure motifs as well as crystal/orbital symmetries in inorganic compounds; (ii) to accelerate the motif and symmetry based discovery and design of oxide and layered semiconductor compounds with optimal electronic properties for energy conversion applications; iii) to enable the effective learning of structure motifs and orbital symmetries through the combined use of national language processing, graph theory, and deep learning.