Lead halide perovskites with superior optoelectrical properties are emerging as a class of excellent materials for applications in solar cells and light-emitting devices. However, perovskite films often exhibit abundant intrinsic defects, which can limit the efficiency of perovskite-based optoelectronic devices by acting as carrier recombination centers. Thus, an understanding of defect chemistry in lead halide perovskites assumes a prominent role in further advancing the exploitation of perovskites, which, to a large extent, is performed by relying on first-principles calculations. However, the complex defect structure, strong anharmonicity, and soft lattice of lead halide perovskites pose challenges to defect studies. In this perspective, on the basis of briefly reviewing the current knowledge concerning computational studies on defects, this work concentrates on addressing the unsolved problems and proposing possible research directions in future. This perspective particularly emphasizes the indispensability of developing advanced approaches for deeply understanding the nature of defects and conducting data-driven defect research for designing reasonable strategies to further improve the performance of perovskite applications. Finally, this work highlights that theoretical studies should pay more attention to establishing close and clear links with experimental investigations to provide useful insights to the scientific and industrial communities.
Keywords: defects; machine learning; nonradiative recombination; perovskite; phase degradation.
© 2024 The Authors. Advanced Science published by Wiley‐VCH GmbH.