Investigating the optoelectronic properties and photovoltaic performance of Na2AuGaBr6 based double perovskite solar cells via numerical simulation and AI techniques

This study offers a thorough analysis of Density Functional Theory (DFT) and SCAPS-1D to assess the optoelectronic performance of cubic Na2AuGaBr6 double perovskites, highlighting its potential for advanced optoelectronic and photovoltaic applications. Device simulations were methodically conducted utilizing Na2AuGaBr6 as the active absorber material, in conjunction with various electron transport layers (ETLs) including TiO2, ZnO, WS2, C60, IGZO, and In2S3, as well as hole transport layers (HTLs) such as CuI, CFTS, NiO, CuSbS2, V2O5, Sb2S3, MoTe2, and CuO, to ascertain the optimal device configuration. Comprehensive parametric optimization was performed by analyzing the effects of different left (Co, Ni, Au, Pt, Pd, and Se) and right (Ca, Ba, Mg, Ag, Al, and Cr) metal contacts, band alignment, layer thicknesses, interface and bulk defect concentrations, as well as temperature on the overall photovoltaic performance. Of the 48 simulated device structures, the Al/FTO/WS2/Na2AuGaBr6/V2O5/Ni configuration demonstrated superior performance, achieving a power conversion efficiency (PCE) of 28.96%. Additionally, advanced machine learning (ML) and deep learning (DL) models were utilized to forecast and corroborate device performance trends. Of the eleven methods evaluated, the Gradient Boosting model exhibited remarkable predictive accuracy, attaining a R2 of 0.954 and a negligible mean absolute percentage error (MAPE) of 0.0218. These findings affirm the significant promise of Na2AuGaBr6-based perovskites for lead-free, high-efficiency solar systems and establish ML and DL-assisted modeling as an efficient method for performance improvement and material design in photovoltaic research.