Optimizing Inorganic Cs₄CuSb₂Cl₁₂/Cs₂TiI₆ Dual-Absorber Solar Cells: SCAPS-1D Simulations and Machine Learning

Area of Science:

• Materials Science
• Renewable Energy
• Photovoltaics

Background:

• Perovskite solar cells (PSCs) show high efficiency but face limitations from the Shockley-Queisser limit in single-junction designs.
• Lead-free perovskite derivatives like Cs4CuSb2Cl12 (CCSC) and Cs2TiI6 (CTI) offer environmental benefits and intrinsic stability.

Purpose of the Study:

• To theoretically investigate a dual-absorber PSC architecture using CCSC and CTI for enhanced performance and sustainability.
• To identify optimal electron/hole transport layers (ETLs/HTLs) and optimize device parameters for high power conversion efficiency (PCE).

Main Methods:

• Computational screening of 26 ETL/HTL candidates, identifying SrTiO3 (STO) and CuSCN as optimal.
• Theoretical optimization of defect densities, band gap, layer thickness, and electrode materials.
• Incorporation of practical constraints like radiative recombination and resistance for realistic efficiency calculations.
• Comparative analysis of machine learning algorithms, highlighting eXtreme Gradient Boosting (XGBoost).

Main Results:

• Initial simulated PCE of 16.27% with STO and CuSCN ETLs/HTLs.
• Optimized device achieved a simulated PCE of 30.86%.
• Realistic simulation incorporating practical constraints yielded a PCE of 26.60%.
• XGBoost demonstrated superior predictive power, identifying CTI defect density as critical for PCE.

Conclusions:

• The CCSC/CTI dual-absorber PSC architecture presents a promising route for stable, efficient, and environmentally friendly photovoltaics.
• Theoretical optimization and realistic constraint incorporation are crucial for predicting practical device performance.
• CTI defect density is a key factor for optimizing PCE in these devices, guiding future experimental efforts.