ABSTRACT
Reactions of reactive halogen species (Clâ¢, Brâ¢, and Cl2â¢-) with trace organic contaminants (TrOCs) have received much attention in recent years, and their k values are fundamental parameters for understanding their reaction mechanisms. However, k values are usually unknown. In this study, we developed machine learning (ML)-based quantitative structure-activity relationship (QSAR) models to predict k values. We tested five algorithms, namely, random forest, neural network, XGBoost, support vector machine (SVM), and multilinear regression, using molecular descriptors (MDs) and molecular fingerprints (MFs) as inputs. The optimal algorithms were MD-XGBoost for Cl⢠and Brâ¢, and MF-SVM for Cl2â¢-, respectively, with R2test values of 0.876, 0.743, and 0.853. We found that electron-withdrawing/donating groups tended to interfere with the reactivity of Cl2â¢- more than Cl⢠and Brâ¢. This explains why MFs are better inputs for predictive models of Cl2â¢-, whereas MDs are more suitable for Cl⢠and Brâ¢. Furthermore, we interpreted the models using SHAP analysis, and the results indicated that our models accurately predicted k values both statistically and mechanistically. Our models provide useful tools for obtaining unknown k values and help researchers understand the inherent relationships between the models.
Subject(s)
Algorithms , Halogens , Machine Learning , Neural Networks, Computer , Random Forest , Quantitative Structure-Activity RelationshipABSTRACT
The degradation kinetics and mechanism of the antiepileptic drug oxcarbazepine (OXC) by UV-activated persulfate oxidation were investigated in this study. Results showed that UV/persulfate (UV/PS) process appeared to be more effective in degrading OXC than UV or PS alone. The OXC degradation exhibited a pseudo-first order kinetics pattern and the degradation rate constants (k obs) were affected by initial OXC concentration, PS dosage, initial pH, and humic acid concentration to different degrees. It was found that low initial OXC concentration, high persulfate dosage, and initial pH enhanced the OXC degradation. Additionally, the presence of humic acid in the solution could greatly inhibit the degradation of OXC. Moreover, hydroxyl radical (OHâ¢) and sulfate radical (SO4 (-)â¢â¢) were identified to be responsible for OXC degradation and SO4 (-)⢠made the predominant contribution in this study. Finally, major intermediate products were identified and a preliminary degradation pathway was proposed. Results demonstrated that UV/PS system is a potential technology to control the water pollution caused by emerging contaminants such as OXC.