Oxygen and Chlorine Isotopic Fractionation during ... - ACS Publications

U.S. Geological Survey, Reston, Virginia 20192, and ... ϵ37Cl) also has significant implications for forensic .... role in environmental forensics fo...
1 downloads 0 Views 237KB Size
Environ. Sci. Technol. 2007, 41, 2796-2802

Oxygen and Chlorine Isotopic Fractionation during Perchlorate Biodegradation: Laboratory Results and Implications for Forensics and Natural Attenuation Studies N E I L C . S T U R C H I O , * ,† JOHN KARL BO ¨ HLKE,‡ ABELARDO D. BELOSO, JR.,† SHERYL H. STREGER,§ LINNEA J. HERATY,† AND PAUL B. HATZINGER§ University of Illinois at Chicago, Chicago, Illinois 60607, U.S. Geological Survey, Reston, Virginia 20192, and Shaw Environmental, Inc., Lawrenceville, New Jersey 08648

Perchlorate is a widespread environmental contaminant having both anthropogenic and natural sources. Stable isotope ratios of O and Cl in a given sample of perchlorate may be used to distinguish its source(s). Isotopic ratios may also be useful for identifying the extent of biodegradation of perchlorate, which is critical for assessing natural attenuation of this contaminant in groundwater. For this approach to be useful, however, the kinetic isotopic fractionations of O and Cl during perchlorate biodegradation must first be determined as a function of environmental variables such as temperature and bacterial species. A laboratory study was performed in which the O and Cl isotope ratios of perchlorate were monitored as a function of degradation by two separate bacterial strains (Azospira suillum JPLRND and Dechlorospirillum sp. FBR2) at both 10 °C and 22 °C with acetate as the electron donor. Perchlorate was completely reduced by both strains within 280 h at 22 °C and 615 h at 10 °C. Measured values of isotopic fractionation factors were 18O ) -36.6 to -29.0‰ and 37Cl ) -14.5 to -11.5‰, and these showed no apparent systematic variation with either temperature or bacterial strain. An experiment using 18O-enriched water (δ18O ) +198‰) gave results indistinguishable from those observed in the isotopically normal water (δ18O ) -8.1‰) used in the other experiments, indicating negligible isotope exchange between perchlorate and water during biodegradation. The fractionation factor ratio 18O/37Cl was nearly invariant in all experiments at 2.50 ( 0.04. These data indicate that isotope ratio analysis will be useful for documenting perchlorate biodegradation in soils and groundwater. The establishment of a microbial fractionation factor ratio (18O/ 37Cl) also has significant implications for forensic studies. * Corresponding author phone: (312)355-1182; fax: (312)413-2279; e-mail: [email protected]. Corresponding author address: Dept. of Earth and Environmental Sciences, University of Illinois at Chicago, 845 West Taylor St., MC-186, Chicago, IL 60607-7059. † University of Illinois at Chicago. ‡ U.S. Geological Survey. § Shaw Environmental, Inc. 2796

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 8, 2007

Introduction Perchlorate has emerged during the past decade as a widespread groundwater contaminant in the United States, with at least 35 states reporting detections in groundwater or drinking water (1). Sources include military and commercial application of perchlorate salts as oxidizers in propellants, flares, munitions, matches, blasting agents, and other materials. In addition, recent studies show that perchlorate derived from the use of Chilean nitrate fertilizers and that present in natural deposits within the U.S. are also likely to contribute to groundwater and drinking water occurrences (2-5). The total scope of perchlorate contamination in the United States is not clear as most private wells and small public water supplies are not routinely tested for the anion. However, tests of 3356 public water supplies conducted under the Unregulated Contaminant Monitoring Rule (UCMR) during the period 2001-2003 revealed that 136 (4.1%) had perchlorate at levels exceeding 2 µg/L, and 55 (1.6%) exceeded 6 µg/L (6). Research conducted during the past several years has revealed that there is a diverse group of bacteria capable of degrading perchlorate under anoxic conditions by using the molecule as a terminal electron acceptor (7-10). Moreover, these organisms appear to be widely distributed in nature, being present in most soils, groundwaters, and surface waters examined to date (8, 11, 12). Biological degradation of perchlorate results in complete reduction of the chlorine atom to chloride, an innocuous product. This occurs through an initial two-step reduction of perchlorate (ClO4-) to chlorate (ClO3-) and then chlorite (ClO2-), mediated by a perchlorate reductase enzyme (11, 13, 14). The chlorite then undergoes a disproportionation reaction catalyzed by chlorite dismutase to yield chloride (Cl-) and oxygen (O2) (8, 11). The microbiological research concerning perchlorate biodegradation has been rapidly applied for treatment of the contaminant in the field. Seven full-scale bioreactor systems (19 reactors total) are currently in operation in the U.S. removing perchlorate from groundwater and wastewater (1). In addition, in situ treatment of perchlorate-contaminated aquifers using electron-donor addition to stimulate naturally occurring bacteria has proven to be a viable option in several field trials; at least two full-scale systems have been constructed for this purpose (1, 12, 15, 16). In contrast to the rapid development and application of active bioremediation systems, there has been relatively little research to examine the potential for intrinsic biodegradation (i.e., natural attenuation) of perchlorate. This is true despite the fact that this process is likely to be a significant natural sink for the molecule in anoxic environments where co-contaminants or natural compounds serve as electron donors. Natural attenuation is a major field of study for other contaminants subject to biodegradation, including chlorinated solvents and petroleum hydrocarbons, and is an accepted remedy for site remediation (i.e., monitored natural attenuation; MNA), when specific criteria are verified (17, 18). Compound-specific stable isotope analysis has proven to be an important tool to support MNA because, when properly applied, this technique allows contaminant biodegradation to be distinguished from dilution and other apparent loss mechanisms (e.g., refs 19 and 20). For this tool to have practical application for perchlorate in field settings, the extent of isotopic fractionation during perchlorate biodegradation must be evaluated and quantified. In addition to documenting MNA, biological fractionation also has important implications for perchlorate forensics. 10.1021/es0621849 CCC: $37.00

 2007 American Chemical Society Published on Web 02/28/2007

Measurements of δ18O, δ17O, and δ 37Cl in perchlorate have recently been used to differentiate natural perchlorate derived from Chilean sources from a variety of synthetic perchlorates (2, 21, 22). Moreover, the same isotopic ratios quantified in natural and synthetic perchlorate salts have also been observed in perchlorate collected from groundwater (2, 22), suggesting that stable isotope analysis can play an important role in environmental forensics for this molecule. However, isotopic fractionation of perchlorate during biodegradation can complicate forensic source determination. Quantification of the relative fractionation of Cl and O isotopes during biodegradation, and the consistency of this ratio, provides critical and useful information for distinguishing such fractionation from other processes that may affect bulk isotopic signatures, such as mixing of two or more different sources. Recent studies have quantified the isotopic fractionation of Cl during microbial perchlorate reduction by two different strains of A. suillum (23, 24). A large and consistent kinetic isotope effect for chlorine was observed in both studies. However, there is relatively little information concerning the corresponding isotopic fractionation of oxygen by this species, and no information is available concerning species differences or the influence of environmental variables on the microbial fractionation of either 37Cl or the oxygen isotopes. In this paper, we characterize and compare the O and Cl isotopic fractionations accompanying anaerobic growth of two different common genera of perchloratedegrading bacteria, Azospira and Dechlorospirillum, and we assess the influence of temperature on isotopic fractionation. A strong and consistent correlation between O and Cl isotopic fractionation, which was independent of bacterial species or temperature, was observed. These data provide a quantitative basis for applying compound-specific stable isotope analysis for detecting perchlorate biodegradation, assessing MNA in field settings, and verifying perchlorate forensic data.

Experimental Section Culture, Growth Conditions, and Sample Collection. Two perchlorate-degrading cultures were used for isotopic fractionation studies. Azospira suillum JPLRND was initially isolated from a groundwater sample collected in southern California, and Dechlorospirillum sp. FBR2 was obtained from a bioreactor treating groundwater. The cultures were purified using traditional enrichment and plating methods and identified by 16S rRNA gene sequencing (Acculabs, Newark, DE). Each culture was initially cultivated in low chloride basal salts medium (BSM; 23) at pH ) 7.3 under anoxic conditions with 17 mM acetate as the electron donor and 10 mM perchlorate as the electron acceptor. When the cultures were turbid, they were centrifuged and resuspended in fresh BSM; then, 5-mL volumes were transferred into 4-L bottles containing 3 L of autoclaved BSM that was subsequently amended with 10 mM perchlorate and 17 mM acetate. Isotopic fractionation was quantified in six separate treatments. Three bottles were inoculated with strain JPLRND, two received strain FRB2, and one remained uninoculated as a sterile control. One bottle with each bacterium was incubated at room temperature (∼22 °C) and one each at 10 °C. The sterile control was incubated at 10 °C. The third bottle with strain JPLRND was amended with 18O-enriched water (to give bulk δ18O ) +198‰) and incubated at room temperature to test for oxygen isotope exchange between perchlorate and water during perchlorate degradation. The water used in all other experiments had a more typical δ18O value of -8.1‰. All bottles were sealed with butyl rubber stoppers to prevent oxygen intrusion. Subsamples were taken periodically in an environmental chamber under a nitrogen headspace, and these samples were immediately filtered through sterile 0.2 µm pore-size cellulose acetate filter units

(Corning Inc., Corning, NY) and stored at 4 °C. The sampling times and volumes were determined by measuring perchlorate concentration via ion-selective electrode (Thermo Scientific, Waltham, MA) and calculating the filtrate volume required to obtain 10 mg of perchlorate for isotopic analysis. Liquid subsamples used for isotopic analysis of perchlorate were shipped to the University of Illinois at Chicago overnight on ice and stored at 4 °C until perchlorate extraction was performed. Determination of Anion Concentrations. Perchlorate concentrations in filtered solutions used for isotopic analysis were determined by ion chromatography (IC) according to EPA Method 314.0. The analysis was conducted using a Dionex ICS-2000 ion chromatograph fitted with a 2 mm Dionex AS16 column and AG16 guard column and eluent generator. Samples were analyzed using suppressed conductivity with an external water source through a 2-mm Dionex ASRS Ultra II suppressor. Perchlorate was eluted from the column using a 60-mM potassium hydroxide solution. Accuracy and reproducibility of ClO4- concentrations in spiked samples were typically within (10%, and the minimum detection limit (MDL) was calculated as 0.085 µg/L. The reporting limit was 1 µg/L. Concentrations of chlorate, chlorite, and chloride were determined by using IC according to EPA method 300.0. A Dionex DX-120 equipped with a 4-mm Dionex AS18 column and AG18 guard column was operated under a recirculating suppressed conductivity mode. The eluent source was sodium hydroxide at 32.8 mM. The reporting limit for these analytes was 0.1 mg/L with MDLs between 0.01 and 0.02 mg/L. Acetate was analyzed by using a modified EPA 300.0 method developed by Shaw Environmental (Lawrenceville, NJ). Samples were run through a Dionex DX600 chromatograph equipped with a 4 mm Dionex AS11HC column and AG11HC guard column and detection was by recirculating suppressed conductivity. A sodium hydroxide gradient (0.25-60 mM) was used to elute the acetate from the column. Separation of Perchlorate for Isotopic Analysis. Amounts of ClO4- extracted from experimental solutions for the isotope ratio measurements ranged from 10 to 80 mg. Perchlorate was extracted from these solutions using a highly selective bifunctional anion-exchange resin (Purolite A530E). The ClO4- was eluted in a solution of FeCl3 and HCl, purified by additional cation exchange, oxidation, and evaporation steps, and then precipitated as CsClO4 by addition of CsCl as described previously (21, 25). Tests of the ion exchange extraction and purification procedures using reagents with known isotopic compositions indicated that no significant isotopic fractionation was caused by these procedures (22). Determination of Oxygen Isotope Ratios in Perchlorate. Oxygen isotope ratios (Table 1) were measured at the Reston Stable Isotope Laboratory (RSIL) of the U.S. Geological Survey. Oxygen isotope ratios are reported as δ18O, defined as the difference between the 18O/16O amount ratio of a sample and that of Vienna Standard Mean Ocean Water (VSMOW)

δ18O ) [(18O/16O)sample/(18O/16O)VSMOW - 1]

(1)

and δ18O is normally reported in parts per thousand (per mil, ‰) following multiplication by 1000 of both sides of eq 1. For δ18O determinations, CsClO4 samples were loaded into silver capsules and dropped into a carbon reactor held at 1325 °C to produce CO, which was transferred in a He carrier stream through a molecular-sieve gas chromatograph to a Finnigan Delta Plus XP isotope-ratio mass spectrometer and analyzed in continuous-flow mode by monitoring peaks at m/z 28 and 30. Derived values of δ18O were calibrated with respect to the VSMOW-SLAP scale by analyzing the perchlorate samples along with reference materials NBS-127 sulfate (δ18O ) +8.6‰), IAEA-N3 nitrate (δ18O ) +25.6 ‰), VOL. 41, NO. 8, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

2797

TABLE 1. Chemical and Isotopic Data for Biodegradation Experiments culture JPL

JPL (HW)

temp (°C)

time (h)

[CH3COO](mM)

[ClO4](mM)

[ClO3](mM)

[Cl](mM)

22

0 65 166 260 269 280 0 65 166 260 269 280 0 65 166 184 195 208 215 0 447 480 505 519 527 652 0 447 652 685 698 709 714 723 0 166 260 280 481

16.9 17.5 16.9 7.20 4.22 1.21 19.3 19.0 18.1 8.37 5.75 4.69 18.0 17.5 13.8 10.2 6.63 1.23 1.17 20.7 17.1 12.9 8.27 5.59 2.71 0.37 16.7 15.3 10.6 5.76 3.07 1.00 0.45 0.017 19.5 19.2 19.2 19.2 21.4

10.5 10.6 9.80 3.94 1.91