Microcystin-LR degradation kinetics during chlorination: Role of water quality conditions

• Published in: Water research (Oxford) Vol. 185; p. 116305
• Main Authors: Huang, Kun, MacKay, Allison A.
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 15.10.2020
• Subjects: Dissolved organic matter; Drinking water treatment; Free chlorine; Microcystin-LR; Predictive models; Drinking water treatment; Free chlorine; Predictive models; Dissolved organic matter; Microcystin-LR
• ISSN: 0043-1354; 1879-2448
• DOI: 10.1016/j.watres.2020.116305

•1,3,5-trimethoxybenzene can quench samples containing N-chlorinated MCLR product.•pH and temperature affect MCLR degradation efficiency during chlorination.•Free chlorine consumption correlated well with dissolved organic matter properties.•A model is built to predict MCLR decay in natural waters during chlorination. Microcystin-LR (MCLR) produced during certain cyanobacteria blooms can contaminate drinking water sources and pose a threat to public health. Previous studies of MCLR degradation by free chlorine may have artifacts from using strong reducing agents to quench chlorination reactions, and they also have not explored the influence of water quality characteristics such as pH, alkalinity, temperature and dissolved organic matter (DOM). Using a novel quencher, 1,3,5-trimethoxybenzene (TMB), the apparent MCLR degradation rate constants were found to be higher than those obtained with thiosulfate (S2O32−), a traditionally used strong reducing quencher. Thiosulfate converted N-chlorinated MCLR degradation products back to the parent MCLR, thereby underestimating MCLR loss over time. The second-order rate constants for HOCl (kHOCl) and OCl− (kOCl-) during chlorination of MCLR were determined to be 72 ± 13 and 28 ± 1.8 M−1s−1, respectively, allowing for determination of the apparent MCLR rate constants (kapp,MCLR) for any known pH condition. The MCLR reaction with free chlorine was strongly affected by temperature and the presence of DOM, while changes in ionic strength and alkalinity had little effect. Free chlorine in the presence of DOM, originating from both terrestrial and microbial sources, exhibited two-stage decay. The initial chlorine demand in the first 15 s of reaction can be determined by the dissolved organic carbon (DOC) concentration (initial chlorine demand = 1.8 × DOC), and the second-order rate constants for the later slower decay correlated well with SUVA254 (kapp,DOM = 0.73 × SUVA254 - 0.41). The results yielded a practical model to predict the decay of MCLR during chlorination of waters with varied water quality characteristics. [Display omitted]