Machine-learning driven approach for exploration of properties of antimony chalcogenide perovskite based double absorber with back surface field layer
摘要
To address the toxicity and stability concerns of lead-based perovskites, this study investigates a lead-free, antimony chalcogenide-based double-absorber solar cell with the structure WSe2/Sb2S3/Sb2Se3/WS2. Numerical simulations were performed using SCAPS-1D, followed by machine learning-based efficiency prediction using Support Vector Regression (SVR), Random Forest (RF), Stacked SVR + RF, and Extreme Gradient Boosting (XGBoost). The optimized configuration, with 0.2 µm Sb2S3 (shallow acceptor density: 1016 cm−3), 0.8 µm Sb2Se3 (shallow donor density: 1019 cm−3), achieved a power conversion efficiency (PCE) of 28.39%, VOC of 0.97 V, JSC of 33.32 mA/cm2, and fill factor of 87.91%. All layers were modelled with a bulk defect density of 1015 cm−3 and an interfacial defect density of 1010 cm−2. Among the ML models, XGBoost demonstrated the best performance with an MSE of approximately 0.003 and R2 of 0.9996. SHAP analysis identified Sb2Se3 donor concentration as the most impactful feature, while Sb2S3 thickness had the least effect. This study showcases the potential of combining SCAPS-1D simulation with interpretable ML models for accelerated design and optimization of lead-free solar cells.