Accelerated discovery of defect tolerant organo-halide perovskites

Abstract

One of the most effective ways to increase power conversion efficiency and maintain stability in hybrid organic–inorganic perovskite (HOIP) semiconductor materials is to avoid defect formation. A rapid screening of defects and trap-states of many HOIP candidate materials is required. For this, we have implemented an “accelerated” material discovery strategy to predict the electronic structures and defects of HOIP-candidate materials by using artificial intelligence and density functional theory. We rationalize material classification and structure–property relationships. With this strategy, we were able to propose previously undiscovered HOIP materials that are defect-tolerant and active candidates for perovskite solar cells. We found that HOIPs are more sensitive to halide interstitial defects rather than halide vacancy defects. Semiconductor HOIPs typically form relatively unstable/shallow halide vacancy defects and stable/deep halide interstitial defects.