Predicting reaction kinetics of reactive bromine species with organic compounds by machine learning: Feature...

Reactive bromine species (RBS) such as bromine atom (Br^•) and dibromine radical (Br₂^•-) are important oxidative species accounting for the transformation of organic compounds in bromide-containing water. This study developed quantitative structure-activity relationship (QSAR) models to predict second order rate constants (k) of RBS by machine learning (ML) and conducted knowledge transfer between RBS and reactive chlorine species (RCS, e.g., Cl^• and Cl₂^•-) to improve model performance. The ML-based models (RMSE_test = 0.476 -0.712) outperformed the multiple linear regression-based models (RMSE_test = 0.572 -3.68) for predicting k of RBS. In addition, the combination of molecular fingerprints (MFs) and quantum descriptors (QDs) as input features improved the performance of ML-based models (RMSE_test = 0.476 -0.712) compared to those developed by MFs (RMSE_test = 0.524 -0.834) or QDs (RMSE_test = 0.572 -0.806) alone. E_HOMO and E_gap were identified to be the most important features affecting k of RBS based on SHAP analysis. A unified model integrating the datasets of four reactive halogen species (RHS, e.g., Br^•, Br₂^•-, Cl^• and Cl₂^•-) was further developed (R²_test = 0.802), which showed better predictive performance than the individual models (R²_test = 0.521 -0.776). Meanwhile, the model performance changed differently by employing knowledge transfer among RHS, which was improved for Br^•/Cl^•, mixed for Br^•/Br₂^•- and Cl^•/Cl₂^•-, but worse for Br₂^•-/Cl₂^•-. This study provides useful tools for predicting k of RHS in aqueous environments.