Using Point of Use Sampling Devices and High Resolution Mass Spectrometry Techniques for Characterizing Drinking Water Exposures

Oral Presentation

Prepared by M. Strynar, R. McMahen, S. Newton
US EPA, 109 TW Alexander Dr., Durham, NC, 27711, United States


Contact Information: strynar.mark@epa.gov; 919-541-3706


ABSTRACT

Safe drinking water supplies are critical for public health, and it has been demonstrated that chemicals present in the water supply can increase the risk for disease and adverse health outcomes, especially over long-term exposure periods. Research has also indicated that conventional drinking water treatment processes are unable to remove many trace organic contaminants from finished drinking water, and this drinking water can be a very important pathway for human exposure to many chemicals. As a result, monitoring programs are critical for the maintenance of safe drinking water supplies. Many of these programs rely on targeted methods that only screen for a limited number of chemicals, highlighting the need for methods that screen finished drinking water for a broader suite of chemical contaminants. In an effort to more fully describe human exposure through drinking water, point of use sampling devices were employed to collect time-integrated drinking water samples in a pilot study of nine North Carolina homes. After extraction, high resolution mass spectrometry techniques were used to screen the samples for an inventory of approximately 30,000 chemicals. There were 241 and 189 chemicals tentatively identified in positive and negative mode, respectively, which represented 17% of total peak area and 3% of total number of features. After removing redundant occurrences, this list was reduced to a set of 260 unique formulas across both modes and nine samples, which could equate to 758 unique chemicals. Matches include chemicals such as perfluorodecanoic acid, atrazine, tris(1,3-dichloroisopropyl)phosphate, azithromycin, fipronil sulfone, progesterone, and nonylparaben, although confirmation with analytical standards is needed. The unique formulas that matched to the database were prioritized according to factors such as detection frequency, abundance, and mass defect. There were 22 formulas found in at least 33% of the samples, five of which contained at least one halogen. The total list of matched formulas was compared with the list of chemicals monitored for as part of the USEPA National Primary Drinking Water Regulations that apply to public water systems. The methods used to carry out drinking water analysis will be presented, along with results that tentatively identify drinking water contaminants that may be missed using the conventional monitoring methods.