Compression Eliminates Charge Traps by Stabilizing Perovskite Grain Boundary Structures: An Ab Initio Analysis with Machine Learning Force Field.

Abstract

Grain boundaries (GBs) play an important role in determining the optoelectronic properties of perovskites, requiring an atomistic understanding of the underlying mechanisms. Strain engineering has recently been employed in perovskite solar cells, providing a novel perspective on the role of perovskite GBs. Here, we theoretically investigate the impact of axial strain on the geometric and electronic properties of a common CsPbBr3 GB. We develop a machine learning force field and perform ab initio calculations to analyze the behavior of GB models with different axial strains on a nanosecond time scale. Our results demonstrate that compressing the GB efficiently suppresses structural fluctuations and eliminates trap states originating from large-scale distortions. The GB becomes more amorphous under compressive strain, which makes the relationship between the electronic structure and axial strain nonmonotonic. These results can help clarify the conflicts in perovskite GB experiments.