Discovery of Two-Dimensional Multinary Component Photocatalysts Accelerated by Machine Learning.

Searching for novel and high-performance two-dimensional (2D) materials is an important task for photocatalytic applications. Although multinary compounds exhibit more diversity in structure and properties in comparison to binary 2D materials, they are comparatively under-studied. Herein, using a machine-learning (ML) technique and high-throughput screening, we develop an efficient approach to accurately predict 2D multicomponent photocatalysts. Over 4000 monolayers are examined, and 75 multinary compounds are identified for photocatalytic applications. Considering our predictions, we find that the ternary and quaternary compounds A 2 P 2 X 6 and ABP 2 X 6 with A = Cu/Zn/Ge/Ag/Cd, B = Ga/In/Bi, and X = S/Se exhibit superior properties, making them promising candidates for overall water splitting. Thus, our work provides an efficient way to explore novel photocatalysts, which could stimulate further theoretical and experimental investigations on 2D multinary compounds for application in photocatalytic water splitting.