Probing chlorine reactivity of dissolved organic matter for disinfection by-product (DBP) formation: Relations with specific ultraviolet absorbance (SUVA) and development of the DBP reactivity profile

The main objective of this study was to systematically probe the chlorine reactivity of DOM for DBP formation and speciation in selected natural waters. DOM was fractionated employing various physicochemical separation processes in two different approaches: bulk water fractionation (granular activated carbon (GAC) and XAD-8 resin adsorption, and alum coagulation) and isolation/fractionation (ultrafiltration (UF) and resin adsorption chromatography (RAC)). Several DOM fractions (50–100) were obtained for each water tested. The impact of GAC surface chemistry on DOM uptake and on the subsequent DBP formation was also investigated in detail.; GAC and XAD-8 adsorption and alum coagulation fractionated DOM in waters based on specific ultraviolet absorbance (SUVA). By increasing the adsorbent or coagulant dose in small increments and preferentially removing high-SUVA fractions from water, it was possible to probe the reactivity of different fractions in a SUVA distribution that appears to exist in natural waters. Knowing the SUVA distribution and its relationship to reactivity seems to be more important and informative than the source water aggregate SUVA value. The use of bulk water fractionation is a new and promising approach for characterizing DBP reactivity in natural waters. DOM removal and subsequent reductions in DBP formation can be maximized by selecting hydrophobic GACs with minimal surface acidity and mesoporous characteristics.; A new experimental approach, called the DBP reactivity profile, has been developed to monitor and predict the reactivity of DOM fractions in natural waters for DBP formation as well as to optimize and evaluate different technologies for DOM removal and DBP control. This approach was based on the observation that there was a single and strong correlation between the SUVA values and DBP (THMs and HAA₉) formation of DOM fractions, independent of the physicochemical separation process used to obtain the fractions. In addition, since low- or non-UV absorbing components of DOM are also captured to different degrees in the fractions, DBP reactivity profiles also allow monitoring of their reactivities. The trends in the reactivity profiles indicated that the UV-absorbing components of DOM are the major reactive sites responsible for DBP formation. The extent of brominated THMs and HAA₉ formation was found to be larger in lower-SUVA (i.e., less aromatic), hydrophilic-dominant, and smaller-molecular weight DOM fractions.; Experimental and theoretical reconstitution of source waters from reverse osmosis (RO), RAC or UF isolates/fractions showed that these processes did not statistically modify the original characteristics and DBP reactivity of DOM at the source waters. The contribution of each RAC or UF fraction to total THMs or HAA₉ yields was linearly cumulative, indicating that fractions had no synergistic effects on total DBP formation. Although hydrophobic fractions had larger DBP yields than those of hydrophilic fractions and source waters, hydrophilic fractions exhibited appreciable DBP yields.