Quantitative structure-activity relationship for the oxidation of organic contaminants by peracetic acid using GA-MLR method

Abstract

Peracetic acid (PAA) is considered as an effective and powerful oxidant for eliminating organic contaminants in wastewater treatment. The second-order rate constant (k_app) for the reaction of PAA with organic contaminants is practically important for evaluating their removal efficiency in wastewater treatment, but only limited numbers of k_app values are available. In this study, 70 organic compounds with various structures were selected, and the k_app of PAA with each organic compound was used to develop two quantitative structure-activity relationship (QSAR) models based on three kinds of descriptors including constitutional, quantum chemical, and the PaDEL descriptors. The genetic algorithm (GA) was applied to select the molecular descriptors, then the models developed by multiple linear regression (MLR). The most important descriptors that explain the reactivity of organic compounds with PAA are the E_HOMO for the model with the constitutional and quantum chemical descriptors. The maxHdsCH and minHdCH₂ are two most important descriptors for the model with only PaDEL descriptors. The developed models can be used to predict k_app for a wide range of organic contaminants. The accuracy of the developed models was proved by the internal, external validation and the Y-scrambling technique. The developed QSAR models using the GA-MLR method can be used as a screening tool for predicting the elimination of organic contaminants by PAA and increasing the understanding of chemical pollutant fate.

Keywords: Degradation rate constants; Genetic algorithm-multiple linear regression (GA-MLR); Organic contaminants; Peracetic acid (PAA); Quantitative structure-activity relationship (QSAR).