2022 Volume 78 Issue 7 Pages III_185-III_194
Disinfection byproducts (DBPs) can be generated from the reaction of natural organic matters (NOM) with halogen during disinfection process. In this study, we used ultra-high resolution mass spectrometry to identify the molecular formulas of deuterium-labeled Xn-DBPs (where Xn is a halogen such as chlorine, bromine, or iodine) to clarify the NOM precursors and reaction pathways of DBPs. The relative contributions of electrophilic addition and substitution reactions to the formation of DBP was found to be dependent on the halogen species involved in the reaction, and the acidity of the hypohalogenic acid species was the important factor affecting the reaction pathway. Furthermore, a binary classification machine learning model (random forest) was constructed to predict the formation of DBP from the molecular properties of NOM. The model showed high prediction accuracy and suggested that the carbon oxidation state of the precursor is an important factor in the formation of iodine containing DBP.