Janus Halogenated Silicene for Photocatalytic and Photovoltaic Applications: A Machine Learning Combined with First-Principles Study

In this paper, we integrate machine learning (ML) with two-dimensional material screening to develop novel halogenated silicene-based materials with considerable potential for photocatalytic and photovoltaic applications. We proposed a data-driven framework that incorporates first-principles electronic structure calculations and 286 self-contained databases trained by supervised learning to accurately predict the structural stability and electronic properties of halogenated silicene-based compounds. A total of 110 potential candidates for photocatalytic water splitting were identified, and five materials with high predicted stability exhibited ideal band edges and demonstrated high carrier mobility rates. Among the 321 heterojunctions composed of halogenated silicene compounds with potential type-II band alignments, we randomly screened a Si8HFCl5I/Si8FClBrI5 heterojunction and validated that it can achieve a high photoelectric conversion efficiency of 21.04%.