Assessing the Fate of Organic Contaminants During Water Treatment Using TOF Mass Spectrometry and Sample Profiling

Poster Presentation

Prepared by T. Anumol, S. Snyder, S. Merel, M. Sgroi
University of Arizona, 1657 E Helen Street, BIO5 Institute, Office: 0400A22, Tucson, AZ, 85719, United States


Contact Information: tanumol@email.arizona.edu; 412-656-0490


ABSTRACT

Water scarcity due to climate change and growing urban population constrains many major cities to look for new water resources. Therefore, the possibility of reusing wastewater is increasingly considered worldwide. However, the major concern resides in removing known and unknown contaminants with potential impact on human health. Targeted analysis with triple quadruple mass spectrometers indicate that advanced oxidation processes (AOPs) might play a major role in water treatment because of their ability to provide efficient and unspecific removal of organic contaminants through hydroxyl radicals. However, little is known on the fate of unknown contaminants and the formation of byproducts. The present study aims to address these points using time of flight mass spectrometry to generate the chemical profile of water samples before and after different AOP treatments.

Wastewater effluent was collected at a local facility and circulated through a pilot allowing simulation of O3, UV, O3/UV, UV/H2O2 and O3/UV/H2O2 treatment. After solid phase extraction, compounds were analyzed by LC-QTOF with C18 column and electrospray ionization.

The analysis of water extracts confirmed that a vast majority of compounds in the sample were unknowns. Indeed, more than a thousand compounds were detected in each extract. Sample profiling using Mass Profiler Professional (MPP) software allowed for deconvolution of the large dataset and pointed out the compounds removed by each AOP but also the by-products formed. Such differences could be easily visualized through a heat map indicating the relative abundance of each compound in each extract. In addition, sample profiling revealed similarities between samples allowing building clusters of compounds with similar behavior. This would allow for selection of ‘indicator’ compounds to mimic the behavior of a wider class of organic contaminants.