Using hysteresis to predict the charge recombination properties of perovskite solar cells

Abstract

The mixed halide perovskites have become famous worldwide due to their rapid development of power conversion efficiency (PCE) and unique photoelectric properties. Hysteresis is one of the unique characteristics of perovskite solar cells (PSCs) in the current–voltage (IV) measurement, which sometimes causes the misjudgment of the true efficiency of the PSCs. In this work, the hysteresis relationship between the scanning way and efficiency in the IV measurement of PSCs is numerically studied through a time-dependent drift-diffusion model. For each PSC, there is an optimal scan rate in the IV measurement, which leads to the highest PCE. Furthermore, a neural network is constructed and trained in order to predict the PSC characteristic parameters through its hysteresis relationship. Through further experimental verification, the interface recombination and ionic migration information can be obtained from hysteresis. Overall, our work deepens the understanding of the complex PSC electrical performance caused by hysteresis, showing that hysteresis is a double-edged sword for the hybrid perovskite.