Surface water contains natural organic matter (NOM) that reacts with disinfectants creating disinfection byproducts (DBPs), some of which are USEPA regulated contaminants. Characterizing NOM can provide insight with respect to DBP formation and water treatment process adaptation to climate change as the nature of NOM varies. This study collected NOM from the Ohio River over 15 months (April 2010 to July 2011) in order to assess seasonal variability in NOM characteristics. The NOM was characterized using fluorescence spectroscopy, UV254, TOC, high performance liquid chromatography – size exclusion chromatography (HPLC-SEC), and elemental analysis. NOM was concentrated, freeze-dried (lyophilized), and validated with the source NOM creating a standardized lyophilized NOM that may be used in water treatment process evaluations investigating utility adaptation to seasonal changes. Additionally, NOM was concentrated at multiple concentration factors, lyophilized, and reconstituted allowing for the determination of optimal NOM concentration and reconstitution conditions. The NOM was characterized using UV254, TOC, HPLC-SEC, fluorescence spectroscopy, and DBP formation.
Raw Ohio River water NOM was concentrated in the following order: ultrafiltration (UF), cation ion exchange, reverse osmosis (RO), sulfate removal, and lyophilization. Lyophilization allows for long-term storage of NOM while providing the ability to reconstitute at various NOM concentrations compared to liquid material with a short shelf-life. Lyophilized NOM was used for elemental analysis while UF effluent, concentrate, and reconstituted lyophilized NOM were employed for all other analyses. A single RO concentration factor (150X) was used during the 15-month study while 50X, 100X, 150X, 200X, and 250X were used to determine the optimal RO concentration factor versus reconstitution factor. Parallel factor
analysis (PARAFAC) determined the locations of principle components within fluorescence excitation-emission matrices (EEMs). DBP formation from chlorination was analyzed for the multiple concentration factor concentrate and reconstituted NOM at 1X source, concentrate, and 250X TOC concentrations.
TOC and UV254 results demonstrated seasonal variation of NOM concentration during the 15-month study. However, PARAFAC determined that the nature of the NOM components were humic-like and constant. The humic characterization was further supported by the humification index (0.79-0.90) determined from corrected fluorescence EEMs. In addition, elemental analysis revealed mid-range oxygen to carbon ratios that are 0.62-0.87, also indicative of humic NOM. Humic substances are composed of naturally-occurring biologically-decayed plant material that can affect the alkalinity, pH, and other treatment characteristics of surface water. Since the humic nature of the NOM was similar over the 15-month study, the primary difference was NOM concentration. Reconstituted lyophilized NOM was validated against source NOM allowing its use as a standardized NOM material in treatment process studies that evaluate effects of changing NOM character and concentration. This temporal library of well-characterized, drinking water relevant NOM is the first of its kind in the drinking water industry and will be a valuable research tool for the drinking water community. The multiple concentration factor study revealed that NOM may be concentrated at RO factors from 50X to 250X, lyophilized, and reconstituted to 1X source, concentrate, and 250X conditions without changing the characterization of the NOM or DBP formation.