Analysis of photocatalytic water splitting and CO2 reduction activity of halide perovskites: A machine learning approach
Applied Catalysis B: Environmental, cilt.385, 2026 (SCI-Expanded, Scopus)
- Yayın Türü: Makale / Tam Makale
- Cilt numarası: 385
- Basım Tarihi: 2026
- Doi Numarası: 10.1016/j.apcatb.2025.126275
- Dergi Adı: Applied Catalysis B: Environmental
- Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Chemical Abstracts Core, Chimica, Compendex, Environment Index, INSPEC
- Anahtar Kelimeler: Halide perovskites, Photocatalytic CO2 reduction, Photocatalytic H2 evolution, Machine learning, Random Forest
- Boğaziçi Üniversitesi Adresli: Evet
Özet
Machine learning analysis was performed to evaluate the photocatalytic activity and stability of halide perovskites in water splitting and CO2 reduction using data from literature. Random forest (RF) regression models were developed for the prediction of bandgap, hydrogen evolution rate for HER and total electron consumption rate for CO2R while SHAP (SHapley Additive exPlanations) was applied to identify the influential descriptors. All RF models were quite satisfactory; R2 of 0.86 (RMSE=0.18) for testing was achieved for bandgap prediction. HER model provided R2= 0.83 (RMSE=0.52) for testing while these values were R2= 0.69 (RMSE=0.39) and R2= 0.64 (RMSE=0.49) for gas and liquids phase CO2R respectively. SHAP interpretation revealed that halide composition, synthesis temperature, cocatalyst, and preparation route strongly influenced the results. Antisolvent precipitation and hot injection methods enhanced CO2 conversion, whereas crystallization and solvothermal syntheses cocatalysts boosted H2 production. Solvent-assisted syntheses and mixed halide systems (Br–I, Cl–I) significantly improved stability.