Environ. Sci. Technol. 2007, 41, 1352-1358
Origin of a Mixed Brominated Ethene Groundwater Plume: Contaminant Degradation Pathways and Reactions B R A D L E Y M . P A T T E R S O N , * ,† E L I Z A B E T H C O H E N , †,‡ HENNING PROMMER,† DAVID G THOMAS,§ STUART RHODES,| AND ALLAN J. MCKINLEY‡ CSIRO Land and Water, Private Bag No. 5, Wembley, WA 6913, Faculty of Life and Physical Sciences, University of Western Australia, Crawley, WA 6907, Golder Associates, P.O. Box 1914, West Perth, WA 6872, and Rio Tinto Technical Services, Ground Floor, 120 Christie Street, St. Leonards, NSW 2065, Australia
On the basis of a combination of laboratory microcosm experiments, column sorption experiments, and the current spatial distribution of groundwater concentrations, the origin of a mixed brominated ethene groundwater plume and its degradation pathway were hypothesized. The contaminant groundwater plume was detected downgradient of a former mineral processing facility, and consisted of tribromoethene (TriBE), cis-1,2-dibromoethene (c-DBE), trans1,2-dibromoethene (t-DBE), and vinyl bromide (VB). The combined laboratory and field data provided strong evidence that the origin of the mixed brominated ethene plume was a result of dissolution of the dense non-aqueousphase liquid 1,1,2,2-tetrabromoethane (TBA) at the presumed source zone, which degraded rapidly (half-life of 0.2 days) to form TriBE in near stoichiometric amounts. TriBE then degraded (half-life of 96 days) to form c-DBE, t-DBE, and VB via a reductive debromination degradation pathway. Slow degradation of c-DBE (half-life >220 days), t-DBE (half-life 220 days), and VB (half-life >220 days) coupled with their low retardation coefficients (1.2, 1.2, and 1.0 respectively) resulted in the formation of an extensive mixed brominated ethene contaminant plume. Without this clearer understanding of the mechanism for TBA degradation, the origin of the mixed brominated ethene groundwater contamination could have been misinterpreted, and inappropriate and ineffective source zone and groundwater remediation techniques could be applied.
Introduction There are many limitations in interpreting the history of a groundwater plume on the basis of measurements of a plume’s current spatial distribution, since modeling contaminant groundwater transport with reversed time is an * Corresponding author phone: +61-8-93336276; fax: +61-8-9333 6211; e-mail:
[email protected]. † CSIRO Land and Water. ‡ University of Western Australia. § Golder Associates. | Rio Tinto Technical Services. 1352
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 4, 2007
ill-posed problem (1-3). Factors affecting a plume’s current spatial distribution include dispersive properties of the transport medium, contaminant source inputs, contaminant dissolution, retardation, and transformation processes. This is especially the case for plumes with a mixture of uncommon contaminants. Without a clear understanding of the origin of a mixed contaminant groundwater plume, inappropriate and ineffective source zone and groundwater remediation techniques could be applied. The detection of a mixed brominated ethene groundwater plume (4), consisting of tribromoethene (TriBE), cis-1,2dibromoethene (c-DBE), trans-1,2-dibromoethene (t-DBE), and vinyl bromide (VB), prompted a laboratory and field investigation to elucidate the origin of the mixed contaminant groundwater plume. While there are currently no World Health Organization (WHO) drinking water guidelines for these compounds, limited studies (5) on the human toxicity of these compounds have been undertaken, with some studies indicating that TBA and its immediate daughter products are possible carcinogens, mutagens, and central nervous system depressants. The potential source of this mixed brominated ethene groundwater plume was a former mineral processing facility (in operation from 1980 to 2000) that used 1,1,2,2-tetrabromoethane (TBA) for density separation of minerals. TBA has previously been found to degrade rapidly under near neutral pH conditions, with a half-life of 220 >220
FIGURE 2. cis/trans-DBE mixture microcosm experiment plotted as (A) concentrations and (B) a ratio of current to initial concentrations. Also shown are half-life curves fitted to the experimental data. Errors bars are the standard deviation of triplicate results.
FIGURE 1. TriBE natural attenuation microcosm experiment showing (A) TriBE reduction and (B) degradation product (c-DBE and t-DBE) formation. No VB production was observed. Also shown are model fits to the experimental data. Errors bars are the standard deviation of triplicate results. gave half-lives of 75, 1.0, 0.2, 0.2, and 75 days. These data indicated little TBA degradation in distilled water at neutral pH with or without NaCl addition. However, under nonneutral pH conditions or neutral pH conditions with Fe2+ addition, rapid TBA degradation was observed. These data suggested that abiotic acid/base- or Fe2+-catalyzed dehydrohalogenation (10) was the possible mechanism for TBA degradation. Further microcosm experiments with (i) soil and distilled water and (ii) groundwater only (data not shown) indicated the conditions for rapid degradation were associated with both the site soil and groundwater. TriBE Degradation. Anaerobic TriBE natural attenuation microcosm experiments (Figure 1) showed relatively slow TriBE degradation (half-life 96 ( 15 days, Table 1) during the first 105 days of the experiment from a TriBE starting concentration of 60 µM, with the production of low concentrations of c-DBE (2.4 µM) and t-DBE (0.8 µM) by day 105. No VB was detected (220 and 220 ( 70 days for c-DBE and t-DBE, respectively (Figure 2). While these degradation rates were low and difficult to assess on the basis of the loss of parent compounds only, degradation of at least t-DBE could be confirmed by (i) the relative reduction of t-DBE compared to total DBE (c-DBE + t-DBE) (from 20.8 ( 0.4% to 18.0 ( 1.0%), (ii) the production of low concentrations of VB (0.20 µM), and (iii) the production of bromide (2 mg L-1, 25 µM) determined at the end of 220 days of the cis/trans-DBE microcosm experiment. The higher rate of degradation of t-DBE compared to c-DBE was consistent with the TriBE microcosm experiment which showed a relative decrease in the net production rate of t-DBE versus c-DBE with time (Figure 1). While the bromide produced (25 µM) from the cis/trans-DBE microcosm was low (twice the analytical detection level), this concentration would be consistent with DBE to VB transformation (80% bromide recovery) as a result of the cis/trans-DBE reduction (30 µM). For transformation of cis/trans-DBE to acetylene, the bromide mass balance would be about 40%. VB production (0.20 µM) from the cis/trans-DBE microcosm was low compared to the reduction of cis/trans-DBE (30 µM), suggesting either (i) cis/trans-DBE transformation to acetylene, rather than VB, or (ii) the rates of VB formation were similar to the rates of VB degradation or (iii) a combination of both. Acetylene, ethene, or ethane production could not be confirmed as concentrations of these com-
pounds at the end of the experiment were below detection. Sterile control DBE microcosm experiments (data not shown) showed no DBE degradation (c-DBE half-life >220 days, t-DBE half life >220 days). VB Degradation. The VB-only anaerobic natural attenuation and control microcosm experiments showed no substantial degradation (half lives >220 days for both natural attenuation and control microcosms; see Table 1). Also, no production of bromide (
