Historical trends in occurrence and atmospheric inputs of halogenated volatile organic compounds in untreated ground water used as a source of drinking water

https://doi.org/10.1016/j.scitotenv.2003.09.007Get rights and content

Abstract

Analyses of samples of untreated ground water from 413 community-, non-community- (such as restaurants), and domestic-supply wells throughout the US were used to determine the frequency of detection of halogenated volatile organic compounds (VOCs) in drinking-water sources. The VOC data were compiled from archived chromatograms of samples analyzed originally for chlorofluorocarbons (CFCs) by purge-and-trap gas chromatography with an electron-capture detector (GC-ECD). Concentrations of the VOCs could not be ascertained because standards were not routinely analyzed for VOCs other than trichloromonofluoromethane (CFC-11), dichlorodifluoromethane (CFC-12) and 1,1,2-trichloro-1,2,2-trifluoroethane (CFC-113). Nevertheless, the peak areas associated with the elution times of other VOCs on the chromatograms can be classified qualitatively to assess concentrations at a detection limit on the order of parts per quadrillion. Three or more VOCs were detected in 100% (percent) of the chromatograms, and 77.2% of the samples contained 10 or more VOCs. The maximum number of VOCs detected in any sample was 24. Modeled ground-water residence times, determined from concentrations of CFC-12, were used to assess historical trends in the cumulative occurrence of all VOCs detected in this analysis, as well as the occurrence of individual VOCs, such as CFC-11, carbon tetrachloride (CCl4), chloroform and tetrachloroethene (PCE). The detection frequency for all of the VOCs detected has remained relatively constant from approximately 1940 to 2000; however, the magnitude of the peak areas on the chromatograms for the VOCs in the water samples has increased from 1940 to 2000. For CFC-11, CCl4, chloroform and PCE, small peaks decrease from 1940 to 2000, and large peaks increase from 1940 to 2000. The increase in peak areas on the chromatograms from analyses of more recently recharged water is consistent with reported increases in atmospheric concentrations of the VOCs. Approximately 44% and 6.7% of the CCl4 and PCE detections, respectively, in pre-1940 water, and 68% and 62% of the CCl4 and PCE detections, respectively, in water recharged in 2000 exceed solubility equilibrium with average atmospheric concentrations. These exceedences can be attributed to local atmospheric enrichment or direct contaminant input to ground-water flow systems. The detection of VOCs at concentrations indicative of atmospheric sources in 100% of the samples indicates that untreated drinking water from ground-water sources in the US recharged within the past 60 years has been affected by anthropogenic activity. Additional inputs from a variety of sources such as spills, underground injections and leaking landfills or storage tanks increasingly are providing additional sources of contamination to ground water used as drinking-water sources.

Introduction

Approximately 50% (percent) of the US population and 97% of the US population in rural areas relies on ground water as a source of domestic supply (Mlay, 1990). Ground water, however, is subject to many potential sources of contamination in urban, suburban and rural areas. Contamination of ground water with volatile organic compounds (VOCs) is of particular concern because many VOCs are known or suspected carcinogens. Federal drinking-water standards exist for only a subset of VOCs (US Environmental Protection Agency (USEPA) primary and secondary drinking water regulations). Other VOCs that are known to occur in drinking water are unregulated but are included in the USEPA's Contaminant Candidate List (CCL) for potential regulation under the Safe Drinking Water Act (SDWA), and still other VOCs have yet to be included in either category.

VOCs are used extensively in industrial applications and consumer products, and are released into the environment during their production, distribution, storage, handling and use. In many ground-water flow systems, VOCs are mobile and do not easily degrade (Dinicola et al., 2000, Harte et al., 1991, Jeffers et al., 1989, Stackelberg et al., 2000). In some highly contaminated ground-water flow systems, degradation of VOC parent compounds to numerous daughter compounds is common (Davis et al., 2003, Dinicola et al., 2000, Barrio-Lage et al., 1986, Bouwer and McCarty, 1983a, Bouwer and McCarty, 1983b, Gupta et al., 1996, Parsons et al., 1984). Consequently, it is not uncommon to observe detectable concentrations of multiple VOCs in ground water (Westrick, 1990, Bender et al., 1999, Moran et al., 1999, Squillace et al., 1999, Lapham et al., 2000, Rowe et al., 2001, Zogorski et al., 2001, Shapiro et al., 2002, Squillace et al., 2002, Stackelberg et al., 2000). The co-occurrence of VOCs in ground water used for domestic supply, even at concentrations below the maximum contaminant level (MCL), may have a negative long-term cumulative and synergistic effect on human health (e.g. Carpenter et al., 1998, Carpenter et al., 2002, Colburn et al., 1996, Lappé, 1991, Squillace et al., 2002, Thomas, 1990).

Recent surveys have focused on characterizing the extent of VOC contamination in ground water (Westrick, 1990, Bender et al., 1999, Moran et al., 1999, Squillace et al., 1999, Lapham et al., 2000, Rowe et al., 2001, Zogorski et al., 2001, Shapiro et al., 2002, Moran et al., 2002, Squillace et al., 2002, Stackelberg et al., 2000). In a survey of drinking-water wells, including domestic and public supply wells, primarily in rural areas, Squillace et al. (2002) detected at least one VOC or pesticide or anthropogenic nitrate in 70% of wells, with a minimum detection limit of 10−3 micrograms per liter (μg/l), that is parts per billion (ppb), for concentrations of VOCs and pesticides. With a detection limit of 10−6 μg/l (parts per quadrillion), Shapiro et al. (2002) reported that 98% of the samples from the untreated-drinking-water sites surveyed contained at least one VOC. With a lower detection limit, Shapiro et al. (2002) showed that the percentage of the samples from these sites containing at least trace concentrations of one VOC was substantially larger than previously recognized (see, e.g. Mlay, 1990, Squillace et al., 1999, Squillace et al., 2002). The inverse relation between analytical reporting limits and frequency of pesticide detection reported by Kolpin et al. (1995) is apparently also applicable to the occurrence of VOCs at drinking-water sites. Lapham et al. (2000) also noted this relation for VOCs at much higher concentrations, where detection frequency (for wells in which one or more VOCs were detected) increased notably (from 22 to 54%) as the reporting limit decreased (from 2 to 0.2 μg/l).

Surveys of trace concentrations of VOCs are potentially important to ground-water resource managers because they offer an uncensored assessment of VOC occurrence. Detection of trace concentrations of VOCs in ground water indicates the source of water is susceptible to contamination (e.g. Shelton et al., 2001), and the potential may exist for more severe deterioration in ground-water quality. In addition the detection of VOCs at trace concentrations provides recognition of the types of VOCs that may need to be monitored with greater frequency.

This report evaluates the occurrence of selected halogenated VOCs at a detection limit of 10−6 μg/l in water samples taken from community-, non-community- (such as restaurants), and domestic-supply wells throughout the US. A subset of the data compiled by Shapiro et al. (2002) is used for this report with additional interpretation of trace concentrations of VOCs from the chromatograms used in the compilation of those data on VOC occurrence. The data for this report can be found in Shapiro et al. (2003). This report also uses modeled ground-water residence times, determined from concentrations of dichlorodifluoromethane (CFC-12), to assess historical trends in the detection frequency and the magnitude of VOC concentrations for all the VOCs combined, as well as individual VOCs, such as carbon tetrachloride (CCl4), chloroform and tetrachloroethene (PCE). The VOC detections are assessed in terms of the percentage that is due to equilibrium with average atmospheric concentrations vs. the percentage that is due to additional enhancements resulting from local atmospheric enrichment or direct contaminant input to ground-water flow systems. The occurrence of VOCs is also evaluated with respect to factors that may affect ground-water vulnerability, such as the type of well (community, non-community, or domestic), well depth, hydrogeologic conditions, local land use and ground-water redox conditions as inferred from measured concentrations of dissolved oxygen and methane.

Section snippets

Data selection

The data compiled by Shapiro et al. (2002) were obtained from archived chromatograms that were originally used to measure concentrations of CFC-12, trichloromonofluoromethane (CFC-11) and 1,1,2-trichloro-1,2,2-trifluoroethane (CFC-113) in water samples by purge-and-trap gas chromatography with an electron-capture detector (GC-ECD) at the US Geological Survey (USGS) Chlorofluorocarbon (CFC) Laboratory in Reston, VA. The water samples were collected over the period from 1997 to 2000, using metal

Detection frequency of VOCs on archived chromatograms

All CFC and VOC data from archived chromatograms for the 413 samples are compiled in Shapiro et al. (2003). Details about the sites can also be found in Shapiro et al. (2003). The six most frequently detected VOCs, other than CFC-11, CFC-12 and CFC-113, on the CFC chromatograms include the chlorinated solvents 1,1-dichloroethylene (DCE), 1,1,1-trichloroethane (TCA), CCl4, chloroform, trichloroethylene (TCE) and PCE (Shapiro et al., 2002). In Shapiro et al. (2003), the retention time of a

Historical trends in VOC occurrence in ground water and in atmospheric sources of VOCs to ground water

Previous investigators have found a high frequency of detection of anthropogenic compounds in ground water recharged in the past 50 years (e.g. Kolpin et al., 1995, Böhlke and Denver, 1995, Busenberg and Plummer, 1992, Plummer and Friedman, 1999, Plummer and Busenberg, 2000; http://water.usgs.gov/lab/cfc accessed on July 29, 2003; Kolpin et al., 1995, Fogg et al., 1999, Shelton et al., 2001, Stackelberg et al., 2000). For example, Kolpin et al. (1995) found a correlation between tritium

Assessment of aquifer vulnerability using VOC data

Numerous investigators have considered the correlation between the occurrence of VOCs in ground water and factors such as the hydrogeologic and physical setting, local land use, well type and geochemical conditions (e.g. Lapham et al., 2000, Moran et al., 2001, Shelton et al., 2001, Squillace and Moran, 2000, Squillace et al., 1999, Squillace et al., 2002, Stackelberg et al., 2000). For example, Squillace et al. (2002) found a higher detection frequency for VOCs in unconfined aquifers than in

Summary and conclusions

Archived chromatograms from purge-and-trap GC-ECD analyses of drinking-water samples were used to determine the frequency of detection of halogenated VOCs in water samples from 413 community-, non-community- (such as restaurants), and domestic-supply wells throughout the US. A subset of the data on VOC occurrence compiled by Shapiro et al. (2002) is used in this report with additional interpretation of trace concentrations of VOCs from the chromatograms used in the compilation. When the

Acknowledgements

The authors gratefully acknowledge Jerry Casile (USGS, Reston, Va.) and Julian Wayland (USGS, Reston, Va.) for CFC analyses, and Allen M. Shapiro (USGS, Reston, Va.), Patricia Toccalino (Oregon Health and Science University, Beaverton, Ore.) and Michael Moran (USGS, Rapid City, S.D.) for their insightful reviews of the manuscript. This project was conducted in cooperation with the US Environmental Protection Agency Office of Ground Water and Drinking Water.

References (65)

  • E.J. Bouwer et al.

    Transformations of 1- and 2-carbon halogenated aliphatic organic compounds under methanogenic conditions

    Appl Environ Microbiol

    (1983)
  • E.J. Bouwer et al.

    Transformations of halogenated organic compounds under denitrification conditions

    Appl Environ Microbiol

    (1983)
  • Bender DA, Zogorski JS, Halde MJ, Rowe BL. Selection procedure and salient information for volatile organic compounds...
  • J.K. Böhlke et al.

    Combined use of groundwater dating, chemical and isotopic analyses to resolve the history and fate of nitrate contamination in two agricultural watersheds, Atlantic coastal plain, Maryland

    Water Resour Res

    (1995)
  • B.W. Bruce et al.

    Contrasting water quality from paired domestic/public supply wells, Central High Plains

    J Am Water Resour Assoc

    (2001)
  • J.L. Bullister et al.

    The solubility of carbon tetrachloride in water and seawater

    Deep-Sea Res 1

    (1998)
  • E. Busenberg et al.

    Use of chlorofluorocarbons (CCl3F and CCl2F2) as hydrologic tracers and age-dating tools: the alluvium and terrace system of Central Oklahoma

    Water Resour Res

    (1992)
  • J.H. Butler et al.

    A record of atmospheric halocarbons during the twentieth century from polar firn air

    Nature

    (1999)
  • Butler JH, Elkins JW, Montzka SA, Thompson TM, Swanson TH, Clarke AD, Moore FL, Hurst DF, Romashkin PA, Yvon-Lewis SA,...
  • D.O. Carpenter et al.

    Understanding the human health effects of chemical mixtures

    Environ Health Perspect

    (2002)
  • D.O. Carpenter et al.

    Human health and chemical mixtures: an overview

    Environ Health Perspect

    (1998)
  • T. Colburn et al.
  • Dinicola RS, Cox SE, Bradley PM. Natural Attenuation of Chlorinated Volatile Organic Compounds in Ground Water at Area...
  • Fogg GE, LaBolle EM, Weissmann GS. Groundwater vulnerability assessment: hydrogeologic perspective and example from...
  • G.W. Gribble

    The natural production of chlorinated compounds

    Environ Sci Technol

    (1994)
  • Hall BD, Butler JH, Clarke AD, Dutton GS, Elkins JW, Hurst DF, King DB, Kline ES, Lind J, Lock LT, Mondeel D, Montzka...
  • J. Harte et al.
  • E.J. Hoekstra et al.

    Natural formation of chloroform and brominated trihalomethanes in soil

    Environ Sci Technol

    (1998)
  • V.A. Isidorov

    Organic chemistry of the earth's atmosphere

    (1990)
  • P.M. Jeffers et al.

    Homogeneous hydrolysis rate constants for selected chlorinated methanes, ethanes, ethenes and propanes

    Environ Sci Technol

    (1989)
  • W.C. Keene et al.

    Composite global emissions of reactive chlorine from anthropogenic and natural sources: reactive chlorine emissions inventory

    J Geophys Res

    (1999)
  • M.A.K. Khalil et al.

    Natural emission of chlorine-containing gases: reactive chlorine emissions inventory

    J Geophys Res

    (1999)
  • Cited by (27)

    • Odorants in the organophosphorus pesticide contaminated soils:identification, prioritization, and quantification using GC-HRMS based on target and non-target screening

      2022, International Journal of Mass Spectrometry
      Citation Excerpt :

      Odorous gases, derived from organic decomposition in the soil, frequently cause odor nuisances and pose potential health risks to receptors during the site remediation [2]. Although efforts are being made to identify halogenated volatile organic compounds (VOCs) [3–6] and aromatic compounds [7–9] for health risk assessment in the soil and groundwater pollution sites, there are still difficulties to find odorants and explain odor resource. Traditionally, gas chromatography, gas chromatography coupled to mass spectrometry have been commonly used as a reliable tool for analysis to investigate the pollution status of volatile odorants in ambient air.

    • Revealing chlorinated ethene transformation hotspots in a nitrate-impacted hyporheic zone

      2019, Water Research
      Citation Excerpt :

      Chlorinated ethenes (CEs) such as trichloroethene (TCE) and its metabolites cis-1,2-dichloroethene (cDCE) and vinyl chloride (VC) are among the most common volatile organic contaminants (VOCs) detected in groundwater (Shapiro et al., 2004; Rivett et al., 2012).

    • A toolkit for groundwater mean residence time interpretation with gaseous tracers

      2013, Computers and Geosciences
      Citation Excerpt :

      In spite of their advantages, there are some problems and limitations with their use and interpretation (more details in Oster et al., 1996; Busenberg and Plummer, 2000; Harnisch et al., 2000; Koh et al., 2007; Sebol et al., 2007; Newman et al., 2010). From 1930 until the mid-1990s, CFC atmospheric-mixing concentrations increased (Bullister, 1989; Shapiro et al., 2004), and began to decrease as a result of international treaties (Cunnold et al., 1997). Consequently, simple parameter models can produce ambiguous estimations from measured tracer concentrations (Plummer et al., 2003).

    • Tracing groundwater with low-level detections of halogenated VOCs in a fractured carbonate-rock aquifer, Leetown Science Center, West Virginia, USA

      2013, Applied Geochemistry
      Citation Excerpt :

      Storage life of these blanks was found to be several months. Samples fused in borosilicate glass ampoules have storage lives of at least 10 a (Shapiro et al., 2004a,b). All the waters analyzed in this study are young and contain a large number of atmospheric derived halogenated VOCs.

    • Emerging organic contaminants in groundwater: A review of sources, fate and occurrence

      2012, Environmental Pollution
      Citation Excerpt :

      Fluorinated organic compounds have a large range of uses including as refrigerants, fire retardants, in packaging and as protective coatings and there is growing interest in there environmental fate and occurrence (de Voogt et al., 2006). A study in the UK detected perfluorinated compounds in 26% of groundwater monitoring sites (n = 219) at concentrations > 0.3 μg/L (EA, 2007), groundwater contamination from chloroflurocarbons is widely reported (e.g. Shapiro et al., 2004; Morris et al., 2005). Two life-style compounds were found in more than four case studies, caffeine (n = 14) and the nicotine metabolite cotinine (n = 4).

    • Pollutant plume delineation from tree core sampling using standardized ranks

      2012, Environmental Pollution
      Citation Excerpt :

      Chloroethenes (CE) are among the most encountered contaminants at industrial sites, as they were widely used as cleaning and degreasing agents (Doherty, 2000). CE are of particular interest and concern here because they are known to be carcinogenic (Shapiro et al., 2004), neurotoxic, and associated with kidney failure in human beings (James et al., 2009). In the sub-surface, CE are characterized by high densities, low solubilities, and resistance to bio-degradation and chemical degradation (Ma and Burken, 2002).

    View all citing articles on Scopus
    View full text