Citation:

Armstrong, M., T. Mcdonald, Y. Sey, G. Onstad, W. Mitch, AND J. Simmons. Do prevailing XAD extraction methods used to generate extracts from disinfected water adequately link extract toxicology to disinfected water chemistry? Gordon Reseach Conference on DBPs; STEAC, Mount Holyoke, MA, July 30 - August 05, 2017.

Impact/Purpose:

The EPA, specifically the Office of Water and the Regional Offices, needs to be able to relate the observed toxicity of disinfected water to the chemicals formed or transformed during the disinfection process. Use of XAD resins to extract disinfection byproducts (DBPs) from water and concentrate them for toxicological assessment has been the state of the science for many years. Preparation of XAD resin extracts is labor and time intensive and small sample volumes result in a matrix not compatible with chemical analysis. Thus, the standard practice is to measure the DBPs in the water before extraction. Then, the toxicology results are interpreted in light of the DBPs in the water before extraction. The ability of the XAD resin method for extraction of the large number of chemical classes of DBPs that are formed, with varying physical and chemical properties, has not been examined. Thus, we conducted an experiment to determine the efficiency of DBP removal by XAD resin extraction methodology.

Description:

Motivation: It is common to use XAD resins to extract disinfection byproducts (DBPs) from disinfected water. The resulting extract is used in toxicological assays to study the effects of DBP mixtures and has been considered representative of the original disinfected water. However, many DBPs of toxicological concern may elude capture via the resins due to volatility, incompatibility with the resins, or from the elution and concentration process. As such, the resulting extract may not be representative of the original water DBP mixture. The purpose of this study was to determine the efficacy of XAD resins for extracting DBPs. Methods: Water was collected from a drinking water treatment plant before entering disinfection tanks. The water was spiked in our laboratory with bromide and iodide, disinfected by either chlorine or chloramine, and extracted by XAD resins. The disinfected water, both before quenching with acid, and after XAD extraction, was sampled for DBP quantification. We show concentration-response data (the same source water spiked at four bromide and iodide concentrations) for eight DBP classes generated by chlorination and chloramination, totaling 51 DBPs quantified. Results: Comparing DBP concentrations in the chlorinated water going on the resin to the water leaving the resin, total organic halides (TOX) and individual haloacetic acid (HAA) analyte removal (i.e., extracted+lost) ranged from 50-64% and 0-65%, respectively. Chloramination reactions generated total organic nitrosamines (TONO) (up to 19.1 ng/L as NDMA) and with chloraminated samples, HAAs and TONO were either generated, or better detected in the water leaving the resin (7-100% and 14-18% increase, respectively). Trihalomethanes (THMs) were removed 100% in chloraminated samples; in chlorinated samples, removal ranged from 24-100% with most THM analytes removed >52%. In future work, we will attempt to quantify DBPs in the extracts to complete the mass balance, determining which DBPs were extracted and which were lost in the process, or from volatility. Conclusions: Depending on the disinfectant and DBP class, concentrations in sample water before and after XAD resin extraction vary widely; suggesting XAD resin extraction may not preserve DBP concentrations and ratios of the original water. To generate extracts of DBP mixtures that accurately represent their origins, the prevailing XAD extraction technique needs modification. (This abstract does not represent EPA policy.)

Record Details: