Hybrid Perovskite–CIGS Solar Cells with Machine Learning-Driven Performance Prediction

Abstract

Hybrid perovskite–Cu(In,Ga)Se₂ (CIGS) solar cells represent a promising pathway for next-generation photovoltaics due to their complementary optoelectronic properties, tunable bandgaps, and potential for low-cost tandem architectures. While perovskite layers offer high absorption coefficients and tunability, CIGS provides long-term stability and proven large-scale deployment. However, integrating these two materials into a high-performance hybrid solar cell introduces challenges in interface engineering, defect management, and process optimization. This paper presents a systematic framework that combines experimental studies with machine learning (ML)-driven performance prediction models to optimize hybrid perovskite–CIGS solar cells. We first provide a comprehensive review of the state-of-the-art in perovskite/CIGS tandem research and highlight key material and structural bottlenecks. Then, a hybrid architecture is proposed with ML-assisted predictive modeling to optimize bandgap alignment, defect passivation, and carrier transport. The system architecture integrates a neural network regression model trained on experimental and simulated datasets, enabling accurate prediction of power conversion efficiency (PCE) under varying fabrication and environmental conditions. The methodology leverages device physics simulation data, experimental results, and feature engineering from material descriptors. Experimental validation demonstrates that ML-guided optimization achieves improved fill factors and higher open-circuit voltage compared to conventional design approaches. The findings indicate that ML-augmented design can accelerate the development cycle of hybrid perovskite–CIGS solar cells, reduce experimental trial-and-error, and push device efficiencies closer to theoretical limits. The paper concludes with a discussion of future directions, including federated learning for collaborative optimization and scalable fabrication strategies to enable commercial deployment in the U.S. renewable energy sector.