Machine Learning-Augmented COSMO-SAC Model for Accurate Screening of Green Deep Eutectic Solvents in Sustainable CO2 Capture

Supporting Information: names and abbreviations of HBAs and HBDs (Tables S1-S2); CO₂ solubility predictions from the original COSMO-SAC model (Table S3); CO₂ solubility predictions based on different QM/COSMO methods (Table S4); performance of different Δ-Models on CO₂ solubility predictions for the test set and training set (Tables S5-S6); ANN model predictions for CO₂ solubility using rigorous splitting methods and non-rigorous splitting methods (Tables S7-S8).