Photodegradation of organic contaminants by photolysis and H2O2/UV processes

Abstract

Organic contaminants are an emerging concern in environmental sciences. The use of ultraviolet-initiated advanced oxidation processes (AOPs) is becoming an attractive alternative for the degradation of harmful organic contaminants that are not easily removed using conventional water treatment processes. Although several UV-initiated AOPs are available, the focus of the research undertaken in this thesis is the hydrogen peroxide-ultraviolet irradiation (H2O2/UV) process, in addition to direct photolysis. The main goal of this thesis is to develop kinetic models for the prediction of organic contaminants degradation and to elucidate the relevance of the computational predictive models that can be used to fill the lack of knowledge on photochemical properties for environmental issues. Quantitative structure property relationship (QSPR) models were used to predict the photochemical properties of organic compounds from their structures. These modeling techniques make use of existing experimental data to predict unknown chemical properties. The conceptual basis of QSPR models is that similar structures are expected to exhibit similar chemical properties. In this thesis computer simulations of combined QSPR and kinetic models were carried out to investigate the various photochemical parameters and the removal potential of organic contaminants. First a kinetic model, which predicts organic contaminants degradation by low-pressure UV irradiation in aqueous media, was developed and verified based on experimental results obtained from the literature. Pollutants with different chemical natures and belonging to various organic contaminant classes, including pharmaceuticals, pesticides and disinfection byproducts were selected as model compounds. The specific objective was to assess the predictive ability of the kinetic model with different organic contaminants and under various process and experimental conditions. The combined approach QSPR / kinetic modeling was then used to predict the photodegradation of commercially unavailable halogenated disinfection byproducts. Second a dynamic chemical kinetic model was developed to predict DBCP and Alachlor photodegradation by H2O2/UV process in aqueous media in a completely mixed batch reactor under various process and experimental conditions. The validated kinetic model was then combined with QSPR model to predict the photodegradation of p-Cresol by H2O2/UV process in aqueous media considering the formation and fate of its intermediate compounds. The validation of the combined approach was conducted by testing the prediction of the p-Cresol degradation under different initial concentrations of p-Cresol and H2O2. In this thesis, the stiff differential equations that describe the organic contaminants photodegradation were implemented and solved within the Matlab environment. QSPR models were constructed with QSARINS software by multiple linear regression and genetic algorithm analysis with the use of Dragon generated molecular descriptors and quantum descriptors performed with Gaussian software on the basis of density functional theory method.