Is machine learning redefining the perovskite solar cells?

Development of novel materials with desirable properties remains at the forefront of modern scientific research. Machine learning (ML), a branch of artificial intelligence, has recently emerged as a powerful technology in optoelectronic devices for the prediction of various properties and rational design of materials. Metal halide perovskites (MHPs) have been at the centre of attraction owing to their outstanding photophysical properties and rapid development in solar cell application. Therefore, the application of ML in the field of MHPs is also getting much attention to optimize the fabrication process and reduce the cost of processing. Here, we comprehensively reviewed different applications of ML in the designing of both MHP absorber layers as well as complete perovskite solar cells (PSCs). At the end, the challenges of ML along with the possible future direction of research are discussed. We believe that this review becomes an indispensable roadmap for optimizing materials composition and predicting design strategies in the field of perovskite technology in the future. (c) 2021 Science Press and Dalian Institute of Chemical Physics, Chinese Academy of Sciences. Published by ELSEVIER B.V. and Science Press. All rights reserved.

Automated summary

Machine learning has emerged as a powerful technology in the field of metal halide perovskites for the prediction of material properties and rational design. Its applications in optimizing fabrication processes and reducing costs are gaining significant attention. This review provides a comprehensive overview of the use of machine learning in designing absorber layers and complete perovskite solar cells, discussing challenges and future research directions.