Chlorine used in the disinfection of drinking water can react with bromide and naturally-occurring organic matter present in water sources, forming disinfection by-products (DBPs). DBPs are of concern due to their toxicity. Most previous studies examining the kinetics of DBP precursor chlorination and bromination were conducted in “clean” systems comprised of synthetic waters. The lack of data obtained in natural waters that go through the disinfection process presents the question of whether reaction rates calculated for “clean” systems are applicable to natural waters. This research aims to quantify reaction rates of dimethenamid (a widely-used herbicide) in four natural water sources (Chesapeake Bay, Atlantic Ocean, Susquehanna River, and Loch Raven Reservoir). These rates will be compared to those previously found in “clean” systems to determine under what solution conditions reaction rates of “clean” systems can be applied to natural waters, including those used for drinking water. To determine rates in the natural waters, reactions were performed using each of the four water sources in which pH, chloride and bromide concentration were treated as independent variables. Aqueous chlorine and bromine were quantified using 1,3,5-trimethoxybenzene as a halogen trap. Reactors were analyzed using gas chromatography with mass spectrometry. The formation rates of the products, bromo- and/or chlorodimethenamid, were used to calculate second-order rate constants. Even though bromination typically occurs faster, chlorodimethenamid was the major product in the Susquehanna River and Loch Raven Reservoir (excluding reactors in which additional bromide was added) due to the low natural concentration of bromide relative to the concentration of free available chlorine spiked in to start the reaction. In the Chesapeake Bay and Atlantic Ocean reactors, however, bromodimethenamid was the major product due to the higher bromide levels of those waters. In these two water sources, halogenation reactions were too fast to quantify using the techniques employed in this study. Results show that reactions rates calculated for “clean” systems overpredict those observed in all the natural waters, particularly for water sources of higher salinity. This implies data from “clean” systems cannot be assumed to apply to natural waters.

Bromination and chlorination kinetics in natural waters: How useful are data collected in “clean” systems?

255th National Meeting (2018)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.