Advanced oxidation process (AOPs) have emerged as powerful technologies and are widely used in drinking water treatment for the destruction of endocrine disrupting chemicals (EDCs) identified by the U.S. Environmental Protection Agency and the World Health Organization. These EDCs could be organic or inorganic contaminants including pesticides, industrial solvents and chemicals, pharmaceutical and personal care products, and petroleum organics present in drinking water supplies. This study focues on the use of different AOPs for the destruction of organic EDCs, and these processes include ozone, hydrogen peroxide and ultraviolet radiations, and/or combinations thereof, including direct photolysis, ozonation, hydrogen peroxide-ultraviolet radiation (HP-UV) oxidation, and hydrogen peroxide-ozone (HP-ozone) oxidation. These AOPs were studied for the destruction of the chlorinated pesticides alachlor and heptachlor to concentrations below their regulated maximum contaminant levels (MCLs) of 0.4 mg/L and 2 mg/L, respectively. The study investigated the reaction mechanisms and pathways as well as formation of intermediates. A new aspect of the study addressed in this presentation involves the kinetic modeling of free-radical reactions, and examining the predictive capability of the modeling approach in various operating and process condtions.

Reaction mechanisms and pathways were proposed for some of the AOPs and target contaminants, and the formation of intermediates and products were studied. More importantly, the adverse effects of free-radical scavengers including natural organic matter (NOM) such as humic acid, and the carbonate species on the reaction kinetics were investigated. The efficient and cost-effective application of these processes required a good understanding of the associated complex kinetics, reaction mechanisms, and free-radical reactions. Modeling the process kinetics of these AOPs in the presence of NOM had far-reaching implications in their decomposition into smaller molecules and generation of disinfection byproducts (DBPs) such as aledehydes and ketones.

Ozonation and HP-ozone oxidation of heptachlor were compared under different conditions such as pH, initial pesticide concentration, and influence of free-radical scavengers such as carbonate species and NOM. The results showed that direct ozonation facilitated decomposition of the pesticide, while the hydroxyl radical was not effective as primary oxidant. The influence of hydrogen peroxide and NOM on the free radical reaction kinetics were studied. The HP-ozone oxidation processes was found to be more effective than ozonation only under certain process conditions. In both cases, oxidation products for heptachlor were formaldehyde, acetaldehyde, glyoxal, glyoxylic acid and heptachlor epoxide, albeit subtle variations in reaction mechanisms.

The parameters influencing the process destruction efficiencies of these AOPs were investigated through a systematic approach of combining experimental data acquisition with a kinetic modeling technique. A free-radical reaction kinetic model was developed for the HP-UV process to predict the concentrations of all principal species, including alachlor, NOM, hydrogen peroxide, carbonate species, and intermediate radicals. The proposed model, that addresses all the important aspects of previous models, incorporates additional features such as gradual pH decrease during the oxidation period (attributed to NOM mineralization), and used a lumped parameter approach for quantification of NOM. The alachlor decomposition efficiency was experimentally evaluated as a function of pH, NOM concentrations, total carbonate concentration, hydrogen peroxide concentrations, and the UV intensity. The modeling approach provided insight into the complex kinetics and reaction mechanisms involved in the HP-UV system under the influence of NOM, and the oxidation products of alachlor. This approach could be extrapolated to other AOPs for optimization of process parameters for a given water quality matrix to achieve efficient and cost-effective .

Furthermore, the study evaluated the adverse effects of NOM and carbonate species as free-radical scavengers on the decomposition kinetics. It was shown that larger hydrogen peroxide dosage achieved very marginal improvements in alachlor decomposition efficiency in the presence of NOM and carbonate species. The investigation further demonstrated that the HP-UV process was very effective in the destruction of alachlor in the presence of NOM, and that it could be applied for the destruction of other ECDs as well. Model sensitivity studies also provided a qualitative evaluation of the influence of different process variables on decomposition efficiency of target contaminant(s).

The AOPs under investigation were very effective in destruction of alachlor and heptachlor, meeting their MCL requirements at short reaction times and low oxidant dosages. The proposed AOP reaction mechanisms could adequately explain the product formations. The addition of hydrogen peroxide to UV radiation accelerated the decomposition of alachlor due to faster generation of hydroxyl radicals. The free-radical scavenging of carbonate species and NOM adversely affected the oxidation efficiencies. In UV-based AOPs, the radiation shielding due to NOM also affected the oxidation kinetics of target contaminants.

Key words: Alachlor, heptachlor, advanced oxidation processes, direct ultraviolet photolysis, ultraviolet radiation-hydrogen peroxide oxidation, ozonation, hydrogen peroxide-ozone oxidation, hydroxyl radicals, oxidation products, kinetic modeling.