A\nData-Driven Platform for Two-Dimensional Hybrid\nLead-Halide Perovskites

Abstract

The\nexceptional properties of two-dimensional hybrid organic–inorganic\nlead-halide perovskites (2D HOIPs) have led to a rapid increase in\nthe number of low-dimensional materials for optoelectronic engineering\nand solar energy conversion. The flexibility and controllability of\n2D HOIPs create a vast structural space, which presents an urgent\nissue to effectively explore 2D HOIPs with better performance for\npractical applications. However, the traditional RP-DJ classification\nmethod falls short in describing the influence of structure on the\nelectronic properties of 2D HOIPs. To overcome this limitation, we\nemployed inorganic structure factors (SF) as a classification descriptor,\nwhich considers the influence of inorganic layer distortion of 2D\nHOIPs. And we investigated the relationship between SF, other physicochemical\nfeatures, and band gaps of 2D HOIPs. By using this structural descriptor\nas a feature for a machine learning model, a database of 304920 2D\nHOIPs and their structural and electronic properties was generated.\nA large number of previously neglected 2D HOIPs were discovered. With\nthe establishment of this database, experimental data and machine\nlearning methods were combined to develop a 2D HOIPs exploration platform.\nThis platform integrates searching, download, analysis, and online\nprediction, providing a useful tool for the further discovery of 2D\nHOIPs.