Design and Development of Mixed Perovskite Solar Cells with High Efficiency and Stability through DFT and AI-Based Design Approaches

Abstract

Compared to different solar technologies, Perovskite-based solar cells are preferred by many for their high PCE and cost-effectiveness. Still, building a market-ready solution requires handling various important issues related to stability, how the device is designed, and efficiency. The study presents comprehensive approaches involving Density Functional Theory (DFT), device modeling, and Machine Learning (ML) to improve and evaluate mixed perovskite materials. DFT was utilized to study the electronic structure, energy gap, and defect properties of perovskites. By using SCAPS-1D simulations, different optimization factors were studied. Furthermore, different ML algorithms were trained to find key device properties. The training involved both experiments and simulations to learn from the data and predict how each material would work, allowing for fast screening of various perovskite compositions. Because of this framework, researchers can identify new, efficient materials and learn more about how different compositions affect solar cell performance. This strategy uses DFT modeling, numerical simulation, and data analysis together to improve the speed of developing better perovskite solar cells.