Environ. Sci. Technol. 2009, 43, 1977–1985
Spatial and Temporal Distributions of Geobacter lovleyi and Dehalococcoides spp. during Bioenhanced PCE-NAPL Dissolution BENJAMIN K. AMOS,† E R I C J . S U C H O M E L , †,| K U R T D . P E N N E L L , †,§ A N D ¨ F F L E R * ,†,‡ FRANK E. LO School of Civil and Environmental Engineering and School of Biology, Georgia Institute of Technology, 311 Ferst Drive, Atlanta, Georgia 30332-0512, and Department of Neurology, Emory University School of Medicine, 615 Michael Street, Atlanta, Georgia 30322
Received September 30, 2008. Revised manuscript received December 2, 2008. Accepted December 23, 2008.
The spatial and temporal distributions of multiple reductively dechlorinating bacteria were simultaneously assessed in a onedimensional sand column containing a tetrachloroethene (PCE) nonaqueous phase liquid (NAPL) source and associated plume zones. The column was uniformly inoculated with a PCEto-ethene dechlorinating microbial consortium that contained Dehalococcoides spp., Dehalobacter spp., and Geobacter lovleyi strain SZ. Geobacter and Dehalococcoides populations grew and colonized the column material, including the mixedNAPL (0.25 mol/mol PCE in hexadecane) source zone. In contrast, Dehalobacter cells did not colonize the porous column material, and planktonic Dehalobacter cell titers remained below the detection limit of ca. 2.6 × 102 cells/mL throughout the experiment. Significant PCE dechlorination was observed and resulted in bioenhanced NAPL dissolution up to 21-fold (maximum) and 5.2-fold (cumulative) relative to abiotic dissolution. cis-1,2-Dichloroethene (cis-DCE) was the primary dechlorination product, although vinyl chloride (VC) was also formed throughout the experiment. Ethene production occurred after significant depletion of PCE from the NAPL and when cis-DCE concentrations dropped below 6 µM. Data obtained after increasing the column residence time from 1.1 to 2.8 days and introducing a VC pulse to the column indicated that both the residence time and cis-DCE inhibition limited significant VC and ethene production. Although both Geobacter and Dehalococcoides cells were present and active in the mixed-NAPL source zone and plume region, Geobacter cell numbers were typically more than 1 order of magnitude higher than Dehalococcoides cell numbers, which is consistent with the production of predominantly cis-DCE. Analysis of both liquid- and solid-phase samples indicated that Geobacter * Corresponding author phone: (404) 894-0279; fax: (404) 8948266; e-mail:
[email protected]. † School of Civil and Environmental Engineering, Georgia Institute of Technology. ‡ School of Biology, Georgia Institute of Technology. § Emory University School of Medicine. | Present address: Geosyntec Consultants, 475 14th Street, Suite 400, Oakland, CA 94612. 10.1021/es8027692 CCC: $40.75
Published on Web 02/10/2009
2009 American Chemical Society
cells grew and remained attached to the porous medium within the source zone but were largely planktonic in the plume region. In contrast, Dehalococcoides cell were attached throughout the column, and Dehalococcoides cell titers increased by 1 to 2 orders of magnitude over the length of the column, correlating to increases in VC concentrations. The results from this study highlight that bioenhanced dissolution is governed by a complex interplay between resident dechlorinators, contaminant concentrations, and other aquifer-specific characteristics (e.g., hydrology).
Introduction Tetrachloroethene (PCE) and trichloroethene (TCE) are pervasive groundwater contaminants capable of forming dense nonaqueous phase liquids (DNAPLs) in subsurface environments (1). Slow contaminant dissolution from aquifer formations containing DNAPL (i.e., DNAPL source zones) significantly impacts environmental and human health via the formation of persistent contaminant plumes. DNAPL source zone remediation has been the subject of intensive research, and several promising in situ remediation technologies (e.g., surfactant flushing, chemical oxidation, and thermal treatment) have been developed and employed that reduce source zone contaminant mass and hence longevity (1, 2). Nevertheless, mitigating near-term environmental risks associated with DNAPL source zones continues to be technically challenging, and alternate cost-effective and efficient solutions that attain current regulatory objectives are desirable (1, 3). Over the past decade, the microbial reductive dechlorination process has emerged as a promising source zone treatment technology, either by reducing source zone longevity (i.e., bioenhanced dissolution (4-11)) or by acting as a polishing step to detoxify residual contaminants and control contaminant mass flux after more aggressive treatments (e.g., surfactant flushing (2, 12, 13)). In microbial reductive dechlorination, specialized bacteria couple growth and energy metabolism to the detoxification of chlorinated ethenes to benign ethene (14, 15). Several bacterial isolates transform PCE to cis-1,2-dichloroethene (cis-DCE) (e.g., Sulfurospirillum spp. (16, 17), Geobacter lovleyi strain SZ (18), and Dehalobacter spp. (19, 20)), but only some members of the Dehalococcoides group have been shown to dechlorinate beyond cis-DCE to vinyl chloride (VC) and ethene (summarized in ref 15). Biostimulation and bioaugmentation approaches to promote microbial reductive dechlorination have gained acceptance as viable plume containment technologies (21-23). Source zone bioremediation was initially discounted as a practical remedial option because of concerns over the toxicity of high contaminant concentrations to dechlorinating bacteria (24). Although recent studies demonstrated that dechlorination is inhibited at aqueous PCE concentrations exceeding 540 µM (90 ppm) (4, 25), reductive dechlorination at PCE concentrations approaching or at saturation levels has been reported (26-32). In fact, 1.5- to 14-fold enhancements in NAPL dissolution, relative to abiotic dissolution, have been observed in bioaugmented laboratory- and pilot-scale systems (4-11). During bioenhanced dissolution, microbial activity within source zones increases the mass transfer driving force (i.e., the boundary layer concentration gradient) for NAPL dissolution by lowering dissolved-phase contaminant concentrations. Source zone bioremediation, therefore, may be a cost-effective approach to deplete source zone contaminant VOL. 43, NO. 6, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1977
mass, reduce overall remediation times, and control contaminant plume formation. Despite recent demonstrations of bioenhanced contaminant dissolution, the activity and distribution of relevant microbes within PCE-DNAPL source zones are poorly understood. Adamson et al. (5) evaluated changes in microbial diversity and the presence of Dehalococcoides strains in a limited number of samples from a 3-D pilot-scale aquifer tank. Denaturing gradient gel electrophoresis indicated reduced microbial diversity in the DNAPL-containing aquifer tank as compared to the bioaugmentation inoculum, but Dehalococcoides spp. remained present within the tank. Sleep et al. (9) performed a study to evaluate microbial community dynamics and composition during bioenhanced PCE-DNAPL dissolution in a 2-D aquifer cell. After bioaugmentation, PCE was dechlorinated to mainly cis-DCE and the number of Dehalococcoides cells detected in the effluent increased almost 3 orders of magnitude. Interestingly, the highest number of Dehalococcoides organisms was detected in samples taken nearest the DNAPL source zone, suggesting colonization of the source zone. Another study demonstrated that Sulfurospirillum multivorans colonized and dechlorinated PCE to cis-DCE within a mixed-NAPL source zone (4). Although these studies provided insights into microbial activity and distribution during source zone bioremediation, each study only evaluated the contribution of a single dechlorinating population (e.g., Dehalococcoides or S. multivorans) to bioenhanced dissolution. Laboratory studies have demonstrated that complete detoxification of PCE to ethene is most efficiently catalyzed by multiple dechlorinating populations, and bioaugmentation consortia contain PCEto-cis-DCE dechlorinators and Dehalococcoides strains to promote DCE and VC dechlorination to ethene (15, 33-35). The distributions of multiple dechlorinators relative to the source zone, however, have not been investigated. Additionally, the reasons why dechlorination stalled at toxic intermediates (i.e., cis-DCE and/or VC) during source zone bioremediation, even when PCE-to-ethene dechlorinating microbial consortia were used, are unclear (5-11, 36). To better understand the role of multiple dechlorinators during bioenhanced dissolution and to gain insight into reported dechlorination stalls, we performed a continuous-flow column experiment to simultaneously assess the spatial and temporal distributions of three key dechlorinators (Dehalococcoides spp., Dehalobacter sp., and G. lovleyi strain SZ) within a PCE-NAPL source zone and the associated plume area.
Materials and Methods Materials. PCE, which has an equilibrium aqueous-phase solubility of ca. 1200 µM (200 mg/L) (37) and a liquid density of 1.625 g/mL (38), was purchased from Fisher Scientific. Hexadecane (HD; 99% purity), which has an equilibrium aqueous-phase solubility of ca. 0.016 µM (0.0036 mg/L) and a liquid density of 0.77 g/mL (39), was obtained from SigmaAldrich. Standard curves were prepared as described (40, 41) with PCE (g99.9%; Sigma), TCE (g99.5%; Sigma), cis-DCE (99.9%; Supelco), trans-1,2-dichloroethene (trans-DCE, 99.9%; Supelco), VC (g99.5%; Fluka Chemical), and ethene (99.5%; Scott Specialty Gases). Federal Fine Ottawa sand (30-140 mesh; U.S. Silica Company), characterized by a mean grain size of 0.32 mm, intrinsic permeability of 4.2 × 10-11 m2 (42), particle density of 2.65 g/mL, low organic carbon content (
