Boosting the performances of Mg3SbBr3-Based perovskite solar cell with machine learning analysis over 27 % utilizing effective transport layers

Abstract

Inorganic A₃BX₃ perovskites, characterized by their remarkable semiconducting properties, have garnered considerable interest in the field of solar cell research. The present study involves evaluating the efficiency of innovative Mg₃SbBr₃ absorber-based solar cells by examining the effect of various electron transport layers (ETLs), such as SnS₂, and In₂S₃. The analysis of Al/FTO/ETL(SnS₂/In₂S₃)/Mg₃SbBr₃/Ni solar cells was carried out using SCAPS-1D simulation software. To maximize performance, essential aspects such as thickness variation, doping density, interface defect, defect density, and operating temperature were thoroughly examined. Solar cell architectures attain power conversion efficiencies (PCE) of 27.59 %, and 23.12 % using SnS₂, and In₂S₃ ETL layers, respectively. The corresponding open-circuit voltages (V_OC) are 0.6786 V, and 0.6250 V, while the short-circuit current densities (J_SC) are 48.674 mA/cm², and 48.666 mA/cm². The fill factors (FF) are 83.52 %, and 76.03 %, respectively. A machine learning (ML) model was subsequently created in order to forecast the solar cells' performance metrics. With an accuracy rate of around 97.75 %, ML predicted the performance matrix of the best optimal solar cell under investigation. These results demonstrate SnS₂ ETL's potential for Mg₃SbBr₃-based photovoltaic applications with outstanding performance.