A multi-target QSAR model for simultaneously predicting the rate constants of reactions across multiple oxidation...

The chemical degradation of volatile organic compounds (VOCs) in the troposphere is primarily governed by reactions with hydroxyl radicals (•OH), ozone (O₃), and nitrate radicals (NO₃•). The corresponding reaction rate constants-k_OH, k_O3 and k_NO3-are essential indicators for evaluating the atmospheric lifetime and exposure risk of VOCs. Most previous studies on predicting these rate constants have focused on individual oxidation systems. In contrast, this study compiled a comprehensive dataset of 462 experimentally determined rate constants (182 k_OH, 83 k_O3, and 197 k_NO3) and established the connection among the three oxidation systems based on differences in oxidation potential (E₀). A multi-target quantitative structure-activity relationship (QSAR) model was developed to simultaneously predict k_OH, k_O3 and k_NO3 values. The optimal model (R² = 0.896, RMSE = 0.756, and Q² = 0.898) incorporates three variables: E₀, E_HOMO (energy of the highest occupied molecular orbital), and f (-)_x (electrophilic Fukui index). Comparative analysis with previously reported QSAR models further confirms that E_HOMO is the most critical quantum chemical parameter influencing the degradation rates of VOCs. The proposed model has undergone rigorous internal and external validation, including applicability domain analysis, demonstrating its strong predictive accuracy and robustness.