Assessing Microbial Uptake of Petroleum Hydrocarbons in

Jun 8, 2010 - Golder Associates, Ltd., 2390 Argentia Road,. Mississauga, ON, L5N 5Z7, Canada. Received January 9, 2010. Revised manuscript received...
2 downloads 0 Views 179KB Size
Download PDF
Environ. Sci. Technol. 2010, 44, 5092–5097

Assessing Microbial Uptake of Petroleum Hydrocarbons in Groundwater Systems Using Natural Abundance Radiocarbon J A S O N M . E . A H A D , †,§ L E A N N E B U R N S , ‡ SILVIA MANCINI,‡ AND G R E G F . S L A T E R * ,† School of Geography and Earth Sciences, McMaster University, 1280 Main Street West, Hamilton, ON, L8S 4K1, Canada, and Golder Associates, Ltd., 2390 Argentia Road, Mississauga, ON, L5N 5Z7, Canada

Received January 9, 2010. Revised manuscript received May 11, 2010. Accepted May 19, 2010.

Carbon sources utilized by the active microbial communities in shallow groundwater systems underlying three petroleum service stations were characterized using natural abundance radiocarbon (14C). Total organic carbon (TOC) ∆14C values ranged from -314 to -972‰ and petroleum-extracted residues (EXTRES) ranged from -293 to -971‰. Phospholipid fatty acids (PLFAs)sbiomarkers for active microbial populationssranged from -405 to -885‰ and a comparison of these values with potential carbon sources pointed to significant microbial assimilation of 14C-free fossil carbon. The most 14C-depleted PLFAs were found in the samples with the highest concentrations of total petroleum hydrocarbons (TPHs). A radiocarbon mass balance indicated up to 43% of the carbon in microbial PLFAs was derived from TPHs, providing direct evidence for biodegradation at two of three sites. At lower levels of TPHs ∆14C values of PLFAs were generally similar to or more enriched than all other carbon in the system indicating microbial utilization of a more 14C-enriched carbon source and no resolvable evidence for microbial incorporation of petroleumderived carbon. Results from this study suggest that it is possible to delineate petroleum biodegradation in groundwater systems using these techniques even in complex situations where there exists a wide range in the ages of natural organic matter (i.e., EXT-RES).

Introduction Contamination of soils and groundwater by petroleum hydrocarbons is a chronic environmental problem across the globe. Leaking underground storage tanks are a major cause of subsurface petroleum contamination, particularly at service stations (1). The ability of indigenous subsurface microorganisms to biodegrade petroleum hydrocarbons is well documented in the literature (2, 3). However, while exploiting natural attenuation processes at contaminated sites can significantly reduce the cost of remediation, * Corresponding author e-mail: [email protected]; tel: +1 (905) 525-9140, x26388; fax: +1 (905) 546-0463. † McMaster University. ‡ Golder Associates, Ltd. § Now at Geological Survey of Canada, Natural Resources Canada, 490 rue de la Couronne, Que´bec, QC, G1K 9A9, Canada. 5092

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 13, 2010

confirming intrinsic biodegradation in situ often poses a significant challenge. The use of compound-specific stable carbon and hydrogen isotopes has shown considerable success as a tool to monitor in situ biodegradation of petroleum hydrocarbons, particularly in aquifers contaminated by the relatively water-soluble monoaromatic hydrocarbons benzene, toluene, ethylbenzene, and xylenes (BTEX) (4). Stable isotope enrichment factors determined in controlled laboratory experiments can be used to estimate the extent of biodegradation of individual BTEX compounds across a plume (4). However, at many petroleum-contaminated sites complex mixtures of aliphatic (e.g., straight-chained and branched alkanes) and mono-, di-, and polyaromatic hydrocarbons are present beyond the BTEX compounds. Significant isotopic fractionation effects are generally not associated with the biodegradation of the larger organic molecules that comprise this total petroleum hydrocarbon (TPH) fraction. For instance, Mazeas et al. (5) found no discernible isotopic fractionation for n-alkanes (>C16) or phenanthrene compounds during the course of crude oil biodegradation experiments. Additionally, the bulk of TPHs are apolar and hydrophobic, leading to substantial sorption onto soil organic matter and mineral surfaces and heterogeneous distributions that can make it difficult to quantify mass loss across a site based on concentration measurements. The use of molecular level natural abundance radiocarbon analysis can overcome these limitations and provide valuable insight into microbial biodegradation of TPHs (6-9). This technique is based on the fact that petroleum carbon is millions of years old and no longer contains significant detectable 14C whereas recently fixed natural organic matter (NOM) has higher, close to modern, levels of 14C. Determination of 14C contents of cellular membrane components such as phospholipid fatty acids (PLFAs) and comparison to potential carbon sources (TPHs, NOM) allows direct identification of microbial metabolism and uptake of petroleumderived carbon by the active microbial population. At sites contaminated by significant amounts of petroleum hydrocarbons, measurement of the 14C content of PLFAs has provided direct evidence of biodegradation of petroleumderived hydrocarbons (6, 8, 9). In other cases, the preferential utilization of relatively more modern NOM has been reported, both in petroleum contaminated sediments (7) and in agricultural soils containing significant amounts of fossil carbon (10, 11). However, in all of these cases the background NOM was primarily modern, and the contamination found at or close to the ground/sediment surface (ca. 2 mm diameter) were removed from aquifer matrix material used for TOC, TLE (including TPHs), and EXT-RES characterization, and samples were homogenized using a mortar and pestle. Samples used for PLFA analyses were not homogenized after removing visible plant debris and larger pebbles (>ca. 4 mm diameter) due to the large volumes of material required (0.6-2.1 kg). More detailed information on sampling protocol and study sites including geology and history of site disturbance, and a work flow diagram illustrating the sequential treatment of aquifer matrix material and organic fractions is provided in the Supporting Information (SI). TPHs. Approximately 5 g of matrix material was ovendried at 50 °C for 48 h and spiked with an internal standard (5R-cholestane) prior to extraction in 1:1 hexane:acetone using a microwave accelerated reaction system (MARS, CEM Corporation). The TLEs derived from these extractions were filtered using solvent-rinsed (hexane, dichloromethane (DCM), methanol) glass fiber filters (GF/F, Whatman) to remove particulates and treated with activated copper to remove elemental sulfur. TLEs were then separated into two fractions (F1, 1:1 hexane/DCM; F2, methanol) by gravity column chromatography using precombusted (450 °C for 8 h) fully activated silica gel (70-230 mesh, VWR). TPHs, including polycyclic aromatic hydrocarbons (PAHs) were eluted in F1. Samples were evaporated to 1 mL under ultrahigh purity (UHP) N2 and spiked with an external recovery standard (o-terphenyl). Recoveries of 5R-cholestane ranged from 82 to 109% (average 99 ( 11%). Concentrations of TPHs were determined by integrating the total area of unresolved complex mixture (UCM) on an Agilent gas chromatograph mass spectrometer (GC/MS) equipped with a 30 m × 0.25 mm i.d. DB-XLB column (J&W Scientific). The GC oven temperature program was 80 °C, ramped to 270 at 10 °C/ min, with a final hold time of 15 min. Concentrations of the

16 EPA Priority PAHs measured in these samples comprised a minor part of TPHs at the three sites (0.4-4.8%) and generally mirrored TPH concentration trends; thus only TPHs are reported here (total PAH concentrations are presented in SI). Microbial PLFAs. To obtain sufficient mass of PLFAs required for radiocarbon analyses, between 586 and 2078 g of wet matrix material was extracted by the modified Bligh and Dyer method (14) using 2:1 methanol/DCM (7). Samples were filtered and phase separated and the organic fraction was subsequently separated into three fractions (DCM, acetone, methanol) by gravity column chromatography using precombusted (450 °C for 8 h) fully activated silica gel (70-230 mesh, VWR). The phospholipid fraction (dissolved in methanol) was evaporated to dryness under ultra-high purity (UHP) N2 and reacted to fatty acid methyl esters (FAMEs) via the mild alkaline methanolysis reaction (14, 15). A secondary silica gel chromatography step (hexane/DCM 4:1, DCM, methanol) was used to separate FAMEs, which eluted in DCM. Identification and quantification of FAMEs, and confirmation of sample purity (i.e., the fact that no non-FAME compounds were present) utilized the same GC/MS and column described above. The GC oven temperature program was 40 °C for 1 min, ramped to 130 at 20 °C/min, ramped to 160 at 4 °C/ min, then ramped to 300 at 8 °C/min, with a final hold time of 5 min. FAMEs were identified using a bacterial reference standard (Bacterial Acid Methyl Esters CP Mix, Matreya Inc.), mass-fragmentation patterns, and retention times, and quantified using external standards (FAMEs of various chain lengths with C12, C14, C16, C18, and C20). Percentage TOC, δ13C, and ∆14C Analyses. Stable (δ13C) and radiocarbon (∆14C) isotope signatures of TOC, EXT-RES, and TLEs were determined using matrix material samples that were oven-dried at 50 °C for 48 h but not spiked with internal standards. EXT-RES is defined as the residual organic carbon remaining in the matrix material following solvent extraction as per White et al. (16). TLEs were filtered through glass fiber filters and treated with activated copper to remove elemental sulfur as described previously. δ13C and ∆14C of FAMEs were measured on the large sample extracts described above and corrected for the isotopically characterized (13C and 14C) methyl group added during methanolysis. Due to the low concentrations of PLFAs typically associated with aquifer matrix material (17) and also observed in this study, the isolation and collection of sufficient masses (>50 µg) of individual PLFAs for accelerator mass spectrometry (AMS) analysis for some samples would have required prohibitively large sample sizes (e.g., up to ∼20 kg for S3a and S3b). However, with the notable exception of a study involving a mixed autotrophic/heterotrophic community (6), previous results have generally shown little variation between PLFAs from the same sample site (7-10). Thus, here we report the radiocarbon content of the bulk PLFA fractions (determined as FAMEs) from the matrix material. GC/MS analysis confirmed that these fractions contained only FAMEs (see above and SI). Large-scale process blanks extracted using identical solvent volumes, conditions, and procedures (i.e., phase separation, primary and secondary silica gel column chromatography, methanolysis) as samples yielded no detectable amounts of background lipid contamination, thus eliminating the need for background 13C or 14C correction. Percentages of total organic carbon (% TOC) in matrix material were determined using a Costech elemental analyzer following decarbonation using HCl. δ13C were measured using either a VG PRISM or VG OPTIMA isotope ratio mass spectrometer (IRMS), and 14C content was determined by AMS at the National Ocean Sciences Accelerator Mass Spectrometry Facility (NOSAMS) facility at Woods Hole Oceanographic Institution after conversion of the CO2 to graphite (18). δ13C and ∆14C of TOC and EXT-RES were VOL. 44, NO. 13, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

5093

TABLE 1. Sampling and Geochemical Parameters for the Eight Aquifer Matrix Material Samples from the Three Different Sites

depth, mbgs % TOC TPHs, mg/kg PLFAs, µg/kg cells g-1 (× 106) no. of PLFAs

S1a

S1b

S1c

S2a

S2b

S2c

S3a

S3b

0.6 0.10 1 83 12 22

1.8 0.38 349 63 9 37

3.5 0.34 1230 194 30 54

∼2 0.26 612 117 20 41

∼2 0.16 26 101 10 27

∼2 1.95 16 773 100 59

4.8 0.15 1 31 4 22

4.8 0.12 4 36 5 23

measured on decarbonated samples. Radiocarbon measurements were normalized to δ13C values of -25‰ and are reported as ∆14C according to international convention (19). In this context, petroleum has a “14C-free” value of -1000‰ while carbon photosynthesized from the atmosphere over the past couple of decades is closer to the current tropospheric value of approximately 50-100‰ (20). The uncertainly for δ13C incorporating both accuracy and reproducibility of the analysis was (0.5‰ and for ∆14C was (10‰ (TOC, EXTRES, and TLEs) and (20‰ (PLFAs). Based on replicate sample analyses the coefficient of variance for % TOC measurements ranged between 2 and 25%. The relatively high level of variability can be attributed to the high sand content in some samples and thus a greater uncertainly associated with weighing out small sample masses required for % TOC analyses.

Results TPH Concentrations. The concentrations of TPHs (Table 1) reflected varying levels of subsurface contamination in the aquifer matrix material reported in this study, ranging from 1 to 1230 mg/kg (Site 1), 16 to 612 mg/kg (Site 2), and 1 to 4 mg/kg (Site 3). The highest concentrations of TPHs were found at S1c (1230 mg/kg), and the lowest concentrations were found at S1a (1 mg/kg), S3a (1 mg/kg), and S3b (4 mg/ kg). The levels of TPHs were comparatively lower than those reported in the subsurface underlying other fuel dispensing facilities or oil storage stations (e.g., 21, 22). However, concentrations for the Fraction 2 hydrocarbon range (nC10-nC16) in the three most contaminated samples exceeded the Canadian Council of Ministers of the Environment Tier 1 guidelines for commercial land-use surface soils (260 mg/kg) (23). While the concentrations of TPHs in contaminated soils are controlled by a number of variables, it is likely that the low levels of organic carbon (0.12-1.95%; Table 1) found at these sites limited sorption onto subsurface material. For instance, Shen and Jaffe (24) reported higher adsorption/ partitioning of TPHs and PAHs onto humic acid-coated montmorillonite, aluminum oxide, and kaolinite compared to pure clays. Full scan GC/MS chromatograms for the fractions containing TPHs are provided in SI. PLFA Concentrations and Microbial Cell Densities. The total PLFA concentrations ranged from 63 to 194 µg/kg at Site 1, 101 to 773 µg/kg at Site 2, and 31 to 36 µg/kg at Site 3 (Table 1). The highest total PLFA concentration (773 µg/ kg) was reported in the least contaminated sample collected at Site 2 (S2c, TPHs 16 mg/kg), whereas the lowest total PLFA concentration (31 µg/kg) was found in the least contaminated sample collected from Site 3 (S3b, TPHs 1 mg/kg). Using an average generic conversion factor of 4 × 104 cells pmol-1 of PLFA (17), this corresponded to cell densities (cells g-1) of 9 × 106 to 3 × 107 at Site 1, 2 × 107 to 1 × 108 at Site 2, and 4 to 5 × 106 at Site 3 (Table 1). These cell densities are within the range of those previously reported for aquifer microbial communities (17). The total number of individual PLFAs extracted from each sample (i.e., PLFA diversity) and microbial abundance (i.e., 5094

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 13, 2010

FIGURE 1. δ13C and ∆14C signatures for TOC, EXT-RES, TLEs, and PLFAs at the three different sites. The uncertainly for δ13C incorporating both accuracy and reproducibility of the analysis was (0.5‰ and for ∆14C was (10‰ (TOC, EXT-RES, and TLEs) and (20‰ (PLFAs). cell densities) were greatest in samples with the highest % TOC and/or TPH. The highest PLFA diversity and microbial abundance was found at S2c, the least contaminated sample from Site 2 and the sample with the highest % TOC reported in this study. The lowest PLFA diversity and microbial abundance was found in the two samples from Site 3. The broad positive relationship between microbial diversity and abundance as noted by Fierer et al. (25) points to a much lower level of microbial activity at Site 3 that was perhaps the result of lower carbon and nutrient availability associated with increasing depth (Site 3 samples were collected at 4.8 mbgs versus 0.6-3.5 mbgs at the other two sites). As there were no noticeable systematic variations in PLFA classes within and between sites that would provide insight into microbial community structures and processes these data are not presented here. A description of PLFA distributions and full scan GC/MS chromatograms for the fractions containing PLFAs are provided in SI. Stable and Radiocarbon Isotopes. δ13C signatures of TOC (-27.7 ( 1.1‰, n ) 8), EXT-RES (-27.5 ( 1.2‰, n ) 8), and TLEs (-28.4 ( 0.9‰, n ) 7) at all three sites fell within a relatively narrow range and were characteristic of C3 plantderived natural organic matter and petroleum (Figure 1). The δ13C values for PLFAs were in general several ‰ more 13 C-depleted than the other three fractions (-31.5 ( 3.2‰, n ) 8), although a significantly 13C-depleted δ13CPLFA signature of -38.7‰ was found at S3b. While little fractionation (30‰ and 20‰, respectively. Samples plotting to the left of the y-axis are characterized by high TPH contamination (i.e., the EXT-RES is younger than TOC after removal of substantial amounts of solvent-extractable 14 C-free TPH). As shown in Figure 2, the three most contaminated samples reported in this study (S1b, S1c, and S2a) all plot to the left of the y-axis while the others plot near the y-axis, indicating that TOC is less dominated by TPHs. One sample reported here (S1a) plotted to the right of the y-axis, indicating a younger age for TOC compared to EXTRES. This was an unexpected observation since 14C contents of EXT-RES in contaminated soils and sediment are generally similar to or more 14C-enriched than corresponding ∆14CTOC values (16, 31). Although great care was taken to remove all plant debris and root fragments, it would not require a

FIGURE 2. Differences in ∆14C values between TOC and EXT-RES plotted versus the differences in ∆14C values between PLFAs and EXT-RES. Different symbols are used for Site 1 (closed circles), Site 2 (closed triangles), Site 3 (closed squares), and for contaminated sediments from Wild Harbor, West Falmouth, MA (open circles). In the latter, the ∆14C values of 16:0 FAMEs were used as a proxy for total PLFAs (7) and correspond to the 14C contents of TOC and EXT-RES measured in similar sediment depth intervals (16). The plot has been divided into four qualitative quadrants that illustrate the relationship between TPH concentration and microbial assimilation of TPHs. significant amount of modern carbon material to affect the radiocarbon age of TOC in less contaminated, heterogeneous, low % TOC matrix material such as S1a. S1a was also the shallowest sample collected in this study (0.6 mbgs) and therefore the most likely to contain recently photosynthesized plant material. Conversely, it is also possible that the EXTRES for S1a contained a slightly higher proportion of nonsolvent-extractable sedimentary fossil carbon, resulting in an older age for EXT-RES compared to TOC. Notwithstanding this one sample, samples contaminated with TPHs with an expected ∆14C of -1000‰ are expected to fall to the left of the y-axis, with the extent of offset indicating the relative presence of TPHs as compared to the NOM. Samples falling below the x-axis in Figure 2 indicate significant uptake of fossil carbon by the microbial community (i.e., PLFAs are 14C-depleted with respect to EXTRES), whereas samples plotting above the x-axis indicate the preferential microbial assimilation of carbon more 14Cenriched than either the corresponding NOM or TPHs (i.e., PLFAs are younger than EXT-RES). Three samples from this study plot unambiguously within the lower left quadrant of Figure 2, providing direct resolution of fossil carbon assimilation associated with high levels of TPHs. The three samples that fall in this quadrant were S1b, S1c, and S2asthe three most contaminated samples (TPHs 349-1230 mg/kg) as described above. The two least contaminated samples from Site 2 (S2b and S2c) plot near the origin, pointing to the microbial assimilation of NOM as reflected in the 14C contents of EXT-RES. The two relatively uncontaminated samples from Site 3 and S1a plot significantly above the x-axis, indicating the preferential uptake of a carbon source that is relatively more 14C-enriched than anything else measured in the system. Rethemeyer et al. (10) reported 14C contents of monounsaturated PLFAs in a rural agricultural soil from the south of Germany that were similar to atmospheric 14C levels VOL. 44, NO. 13, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

5095

and varied little between 0 and 45 cm depth. Similarly Cowie et al. (9) reported that the majority of PLFA from 10 cm depth in an uncontaminated soil had modern 14C values. The modern 14C signal observed in these microbial PLFA was attributed to the incorporation of fresh, labile soil OM that had been transported downward into the subsurface as dissolved organic matter (DOM). As most of the CO2 produced during decomposition is derived from relatively short-lived components that do not reflect the bulk age of soil OM (32), it is not surprising to observe instances where PLFAs are significantly more 14C-enriched than the background NOM. This may also explain why it is unlikely that many samples would fall into the lower right quadrant corresponding to low TPH/fossil carbon uptake (Figure 2), since in heterogeneous, uncontaminated, carbon-limited systems it is likely that the preferential uptake of more labile, modern carbon dominates the ∆14CPLFA signal. That is, if we consider sedimentary fossil carbon to be a highly refractory, nonlabile carbon source (33), then microbial communities would likely assimilate what little fresh modern OM remains. An isotopic mass balance was used to estimate the percentage incorporation of fossil carbon into microbial PLFAs as per eq 1: ∆14CPLFA ) fNOM(∆14CNOM) + ffossil(∆14Cfossil)

(1)

where ∆14CPLFA, ∆14CNOM, and ∆14Cfossil represent the 14C contents of PLFAs, NOM (i.e., EXT-RES), and fossil carbon, respectively. The ∆14C value for fossil carbon is assumed to be -1000‰ (i.e., 14C-free). The radiocarbon contents of PLFAs comprise contributions from both NOM and fossil carbon fractions (fNOM + ffossil ) 1). Rearranging eq 1 and solving for ffossil we found that that between 1 and 43% of the carbon in microbial PLFAs was derived from fossil carbon sources. It was possible to solve for ffossil in only five of the samples: S1b (21%), S1c (43%), S2a (33%), S2b (1%), and S2c (13%). For S1a, S3a, and S3b eq 1 yielded negative fractions, indicating a negligible fossil carbon contribution (i.e., ffossil ) 0) to PLFAs in these samples and hence no resolvable evidence for microbial incorporation of petroleum hydrocarbons. As shown in Figure 2, four of the five samples with solvable ffossil fall below the x-axis whereas the three samples with negligible ffossil fall above the x-axis. Since the difference between PLFAs and EXT-RES for S2b is not discernible within analytical error the estimated 1% fossil carbon incorporation is not considered meaningful. It is possible that the fossil carbon contribution to PLFAs contained both TPH and sedimentary fossil carbon fractions, as both are sources of 14C-free carbon. However, a statistically significant positive linear relationship (r2 ) 0.88, n ) 8; P < 0.001) between TPH concentrations and ffossil supports the argument that 14C-depletion in ∆14CPLFA signatures was associated with the microbial assimilation of petroleum hydrocarbons rather than sedimentary fossil carbon. Although evidence for microbial mineralization of sedimentary fossil carbon has been demonstrated (29) it is expected that the bulk of this material in soils and sediments is highly refractory and resistant to biodegradation (33). Our findings suggest that TPHs, or at least certain fractions within TPHs, are a relatively more labile carbon source than sedimentary fossil carbon in these shallow groundwater systems. Some of the most labile compounds found in crude and refined oils are n-alkanes, which may be preferentially biodegraded over other petroleum constituents under both aerobic (6) and anaerobic conditions (34). Implications for Natural Attenuation of Petroleum Hydrocarbons. Wakeham et al. (8) estimated that between 6 and 10% of the carbon in microbial PLFA in heavily contaminated marsh sediments (TPHs ∼ 6 g/kg) from southeastern Georgia, USA was derived from petroleum. 5096

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 13, 2010

However, the ∆14CPLFA values (+4 to +261‰) reported by Wakeham et al. (8) were significantly more 14C-enriched than those reported here, indicating a predominantly modern microbial carbon source. Slater et al. (7) observed little difference between the radiocarbon contents of PLFAs and NOM in heavily contaminated (TPHs up to ∼9 g/kg) saltmarsh sediments from Wild Harbor, West Falmouth, MA, pointing to no measurable metabolism of petroleum residues. Using the ∆14C signatures for 16:0 FAMEs as a proxy for bulk PLFAs and the 14C contents of TOC and EXT-RES measured in similar sediment depth intervals (16), these samples are plotted alongside our data in Figure 2. Despite the significant TPH contamination in these sediments, most samples plot near the origin, with only one sample falling into the upper left quadrant corresponding to high TPH presence but modern carbon uptake. The lack of a more distinct difference between TOC and EXT-RES can be attributed to differences in the TPH contribution to % TOC between the two environments. After converting TPHs to carbon equivalents by assuming a 15% contribution of hydrogen to hydrocarbons, White et al. (16) estimated that between 0 and 6.5% of % TOC was derived from petroleum. The highest estimations coincided with the greatest differences between TOC and EXT-RES, as confirmed by a radiocarbon mass balance. Using the same GC approach as White et al. (16), the contribution of petroleum to % TOC in the aquifer matrix material was much higher (0.1-30.8%), providing an explanation as to why the most contaminated samples reported here plot significantly to the left of the y-axis in Figure 2. In turn, the greater TPH component in % TOC was likely the result of the significantly lower % TOC in aquifer matrix material (0.10-1.95%) compared to West Falmouth sediments (8.1-12.8% (16)). The higher TPH component in conjunction with lower % TOC may also explain why microbial PLFAs from this study were significantly more 14C-depleted than those reported by Slater et al. (7) and Wakeham et al. (8), where in the latter study % TOC was as high as 13.6%. In highly productive environments such as salt marshes, the abundance of fresh, labile OM may limit the assimilation of recalcitrant petroleum hydrocarbons. In contrast, the smaller amounts of labile OM in the matrix material reported here and on the surface of intertidal rocks (6) present more favorable conditions for microbial uptake of TPHs, although nutrient limitation and variability in terminal electron-accepting processes will also play important roles in subsurface biodegradation (35, 36). As suggested by Slater et al. (7), interpreting the extent of petroleum biodegradation in shallow groundwater systems using natural abundance radiocarbon measurements of microbial biomarkers will require an understanding of other OM present. Nonetheless, results from this study suggest that it may be possible to delineate petroleum biodegradation in groundwater systems using the radiocarbon characterization approach described here in complex situations regardless of the large range in NOM ages. However, the carbon-limited matrix material examined in this study cannot be considered exclusively representative of subsurface environments present at hydrocarbon contaminated sites; further research should address the response of indigenous microbial populations under a range of groundwater conditions.

Acknowledgments We thank Jennie Kirby, Martin Knyf, Jenifer Hansen, and Nagissa Mahmoudi for assistance with laboratory analyses, and Bradley Aaron and Laura Jones for collection of field samples. Tony Missiuna, Edmund Rodrigues, and David Smyth at Golder Associates provided logistical support for site selection. This work was funded by the Natural Sciences & Engineering Research Council of Canada (NSERC). Support was also provided by the Geological Survey of Canada, Natural

Resources Canada. Special thanks to the staff at NOSAMS, Woods Hole, MA, for carrying out radiocarbon analyses.

Supporting Information Available Details on sampling protocols and study sites, a description of microbial PLFA distributions, a work flow diagram illustrating the sequential treatment of aquifer matrix material and organic fractions, a data summary table, and full scan GC/MS chromatograms for the fractions containing TPHs and PLFAs. This information is available free of charge via the Internet at http://pubs.acs.org.

Literature Cited (1) Atlas, R. M.; Cerniglia, C. E. Bioremediation of petroleum pollutants - Diversity and environmental aspects of hydrocarbon biodegradation. BioScience 1995, 45, 332–338. (2) Leahy, J. G.; Colwell, R. R. Microbial degradation of hydrocarbons in the environment. Microbiol. Rev. 1990, 54, 305–315. (3) Zhang, C. L.; Bennett, G. N. Biodegradation of xenobiotics by anaerobic bacteria. Appl. Microbiol. Biotechnol. 2005, 67, 600– 618. (4) Meckenstock, R. U.; Morasch, B.; Griebler, C.; Richnow, H. H. Stable isotope fractionation analysis as a tool to monitor biodegradation in contaminated acquifers. J. Contam. Hydrol. 2004, 75, 215–255. (5) Mazeas, L.; Budzinski, H.; Raymond, N. Absence of stable carbon isotope fractionation of saturated and polycyclic aromatic hydrocarbons during aerobic bacterial biodegradation. Org. Geochem. 2002, 33, 1259–1272. (6) Slater, G. F.; Nelson, R. K.; Kile, B. M.; Reddy, C. M. Intrinsic bacterial biodegradation of petroleum contamination demonstrated in situ using natural abundance, molecular-level 14C analysis. Org. Geochem. 2006, 37, 981–989. (7) Slater, G. F.; White, H. K.; Eglinton, T. I.; Reddy, C. M. Determination of microbial carbon sources in petroleum contaminated sediments using molecular 14C analysis. Environ. Sci. Technol. 2005, 39, 2552–2558. (8) Wakeham, S. G.; McNichol, A. P.; Kostka, J. E.; Pease, T. K. Natural-abundance radiocarbon as a tracer of assimilation of petroleum carbon by bacteria in salt marsh sediments. Geochim. Cosmochim. Acta 2006, 70, 1761–1771. (9) Cowie, B. R.; Greenberg, B. M.; Slater, G. F. Determination of microbial carbon sources and cycling during remediation of petroleum hydrocarbon impacted soil using natural abundance 14 C analysis of PLFA. Environ. Sci. Technol. 2010, 44, 2322– 2327. (10) Rethemeyer, J.; Kramer, C.; Gleixner, G.; John, B.; Yamashita, T.; Flessa, H.; Andersen, N.; Nadeau, M. J.; Grootes, P. M. Transformation of organic matter in agricultural soils: radiocarbon concentration versus soil depth. Geoderma 2005, 128, 94–105. (11) Rethemeyer, J.; Kramer, C.; Gleixner, G.; Wiesenberg, G. L. B.; Schwark, L.; Andersen, N.; Nadeau, M. J.; Grootes, P. M. Complexity of soil organic matter: AMS C-14 analysis of soil lipid fractions and individual compounds. Radiocarbon 2004, 46, 465–473. (12) Bugna, G. C.; Chanton, J. P.; Stauffer, T. B.; MacIntyre, W. G.; Libelo, E. L. Partitioning microbial respiration between jet fuel and native organic matter in an organic-rich long timecontaminated aquifer. Chemosphere 2005, 60, 177–187. (13) Hendry, M. J.; Wassenaar, L. I. Origin and migration of dissolved organic carbon fractions in a clay-rich aquitard: 14C and δ13C evidence. Water Resour. Res. 2005, 41, 10. (14) White, D. C.; Davis, W. M.; Nickels, J. S.; King, J. D.; Bobbie, R. J. Determination of the sedimentary microbial biomass by extractible lipid phosphate. Oecologia 1979, 40, 51–62. (15) Guckert, J. B.; Antworth, C. P.; Nichols, P. D.; White, D. C. Phospholipid, ester-linked fatty acid profiles as reproducible assay for change in prokaryotic community structure of estuarine sediments. FEMS Microbiol. Ecol. 1985, 31, 147–158.

(16) White, H. K.; Reddy, C. M.; Eglinton, T. I. Isotopic constraints on the fate of petroleum residues sequestered in salt marsh sediments. Environ. Sci. Technol. 2005, 39, 2545–2551. (17) Green, C. T.; Scow, K. M. Analysis of phospholipid fatty acids (PLFA) to characterize microbial communities in aquifers. Hydrogeol. J. 2000, 8, 126–141. (18) McNichol, A. P.; Osborne, E. A.; Gagnon, A. R.; Fry, B.; Jones, G. A. TIC, TOC, DIC, DOC, PIC, POC - unique aspects in the preparation of oceanographic samples for 14C-AMS. Nucl. Instrum. Methods Phys. Res., Sect. B 1994, 92, 162–165. (19) Stuiver, M.; Polach, H. A. Discussion: Reporting of 14C data. Radiocarbon 1977, 19, 355–363. (20) Levin, I.; Kromer, B. The tropospheric 14CO2 level in mid-latitudes of the Northern Hemisphere (1959-2003). Radiocarbon 2004, 46, 1261–1272. (21) Iturbe, R.; Flores, R. M.; Torres, L. G. Subsoil contaminated by hydrocarbons in an out-of-service oil distribution and storage station in Zacatecas, Mexico. Environ. Geol. 2003, 44, 608–620. (22) Ringelberg, D.; Richmond, M.; Foley, K.; Reynolds, C. Utility of lipid biomarkers in support of bioremediation efforts at army sites. J. Microbiol. Methods 2008, 74, 17–25. (23) Canadian Council of Ministers of the Environment. Canadawide standards for petroleum hydrocarbons (PHC) in soil. Available at http://www.ccme.ca/. (24) Shen, L.; Jaffe, R. Interactions between dissolved petroleum hydrocarbons and pure and humic acid-coated mineral surfaces in artificial seawater. Mar. Environ. Res. 2000, 49, 217–231. (25) Fierer, N.; Schimel, J. P.; Holden, P. A. Variations in microbial community composition through two soil depth profiles. Soil Biol. Biochem. 2003, 35, 167–176. (26) Hayes, J. M. Fractionation of carbon and hydrogen isotopes in biosynthetic processes. In Stable Isotope Geochemistry; Mineralogical Society of America: Washington, DC, 2001; Vol. 43, pp 225-277. (27) Londry, K. L.; Jahnke, L. L.; Marais, D. J. D. Stable carbon isotope ratios of lipid biomarkers of sulfate-reducing bacteria. Appl. Environ. Microbiol. 2004, 70, 745–751. (28) Teece, M. A.; Fogel, M. L.; Dollhopf, M. E.; Nealson, K. H. Isotopic fractionation associated with biosynthesis of fatty acids by a marine bacterium under oxic and anoxic conditions. Org. Geochem. 1999, 30, 1571–1579. (29) Petsch, S. T.; Eglinton, T. I.; Edwards, K. J. C-14-dead living biomass: Evidence for microbial assimilation of ancient organic carbon during share weathering. Science 2001, 292, 1127–1131. (30) Cowie, B. R.; Greenberg, B. M.; Slater, G. F. Determination of microbial carbon sources and cycling during remediation of petroleum hydrocarbon impacted soil using natural abundance 14 C analysis of PLFA. Environ. Sci. Technol. 2010, 44, 2322– 2327. (31) White, H. K.; Reddy, C. M.; Eclinton, T. I. Radiocarbon-based assessment of fossil fuel-derived contaminant associations in sediments. Environ. Sci. Technol. 2008, 42, 5428–5434. (32) Trumbore, S. Age of soil organic matter and soil respiration: Radiocarbon constraints on belowground C dynamics. Ecol. Appl. 2000, 10, 399–411. (33) Dickens, A. F.; Gelinas, Y.; Masiello, C. A.; Wakeham, S.; Hedges, J. I. Reburial of fossil organic carbon in marine sediments. Nature 2004, 427, 336–339. (34) Jones, D. M.; Head, I. M.; Gray, N. D.; Adams, J. J.; Rowan, A. K.; Aitken, C. M.; Bennett, B.; Huang, H.; Brown, A.; Bowler, B. F. J.; Oldenburg, T.; Erdmann, M.; Larter, S. R. Crude-oil biodegradation via methanogenesis in subsurface petroleum reservoirs. Nature 2008, 451, 176–U6. (35) Breedveld, G. D.; Sparrevik, M. Nutrient-limited biodegradation of PAH in various soil strata at a creosote contaminated site. Biodegradation 2000, 11, 391–399. (36) Vroblesky, D. A.; Chapelle, F. H. Temporal and spatial changes of terminal electron-accepting processes in a petroleum hydrocarbon-contaminated aquifer and the significance for contaminant biodegradation. Water Resour. Res. 1994, 30, 1561–1570.

ES100080C

VOL. 44, NO. 13, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

5097