Search

This study investigates the interaction of natural organic matter with iron oxide (goethite) on chlorine decay, disinfection by-product (DBP) formation, and DBP compound speciation [total trihalomethanes (TTHM4) and haloacetic acids (HAA5)]. Batch experiments were conducted with goethite, multiple finished drinking waters, variable chlorine dose, and fixed pH 8. The overall objective was to assess natural organic matter (NOM) adsorption onto goethite and its effect on chlorine decay and DBP formation. Chlorine consumption always increased in the presence of goethite and is attributed to an increase in the reactivity and/or modification of adsorbed NOM. Adsorbed NOM also led to an overall increase in TTHM4, however, HAA5 formation was suppressed during the first 2 h. Chloroform was identified as the increasing species and dichloracetic acid was identified as the suppressed species. This study clearly shows that goethite, which is the predominant iron oxide of pipe deposits, alters both chlorine decay and DBP formation and should be considered when assessing water treatment plant operations and DBP monitoring site selection. CE Database subject headings: Halogen organic compounds; Water distribution systems; Disinfection; Chlorine; Iron.

The purpose of this study is to develop a kinetic model that links [0.sub.3] decomposition reactions from the TFG ozone decay model with recognized [Br.sup.-] oxidation reactions, secondary * OH reactions, and [H.sub.2][0.sub.2] reactions in order to improve [0.sub.3] decay and bromate formation prediction capabilities under multiple water quality conditions. The model was compared with experimentally measured ozone decomposition and final bromate concentration data sets provided by two researchers. The data sets included varying pH (6.5-8.5), initial hydrogen peroxide (0-1 mM), and initial bromide concentration (0.1-1 mM). Model verification was carried out by sensitivity analysis of the rate constants and then optimization of the most sensitive rate constants using the method of least squares. Model predicted ozone decay data was analyzed and compared with measured ozone decay data using R-squared statistic for linear regression model. The model predicted final bromate concentration is analyzed by comparing it with the residual [DELTA](%) between experimental and model results. The TFG model was effectively tested for multiple data sets and it was found that model prediction was a success both for ozone decay (regression coefficients >0.95 for all experimental conditions but one) and bromate prediction with residual of less than 100% for all experimental conditions except low peroxide dose ( CE Database subject headings: Ozonization; Models; Water treatment.