The Role of Microbial Reductive Dechlorination of TCE at a ...

International Journal of Phytoremediation: Vol. 5, No. 1, pp. 73–87 (2003)

1522-6514/03/$.50© 2003 by CRC Press LLC

The Role of Microbial Reductive Dechlorination of TCEat a Phytoremediation Site

E. Michael Godsy,1* Ean Warren,2 and Victor V. Paganelli31U.S. Geological Survey, Water Resources Division, Menlo Park, CA 94025;email: [email protected]; Tel: (650) 329-4504; Fax: (650) 329-4463; 2U.S.Geological Survey, Water Resources Division, Menlo Park, CA, 94025; email:[email protected]; Tel: (650) 329-4554; Fax: (650) 329-4463; 3U.S.Geological Survey, Water Resources Division, Menlo Park, CA, 94025; email:[email protected]; Tel: (650) 329-4503; Fax: (650) 329-4463.

* Corresponding author: E. Michael Godsy, U.S. Geological Survey, Water ResourcesDivision, 345 Middlefield Road, MS 480, Menlo Park, CA 94025

ABSTRACT

In April 1996, a phytoremediation field demonstration site at the Naval AirStation, Fort Worth, Texas, was developed to remediate shallow oxic ground water(< 3.7 m deep) contaminated with chlorinated ethenes. Microbial populations weresampled in February and June 1998. The populations under the newly plantedcottonwood trees had not yet matured to an anaerobic community that could dechlo-rinate trichloroethene (TCE) to cis-1,2-dichloroethene (DCE); however, the micro-bial population under a mature (~22-year-old) cottonwood tree about 30 m southwestof the plantings had a mature anaerobic population capable of dechlorinating TCE toDCE, and DCE to vinyl chloride (VC). Oxygen-free sediment incubations withcontaminated groundwater also demonstrated that resident microorganisms werecapable of the dechlorination of TCE to DCE. This suggests that a sufficient amountof organic material is present for microbial dechlorination in aquifer micronicheswhere dissolved O2 concentrations are low. Phenol, benzoic acid, acetic acid, and acyclic hydrocarbon, compounds consistent with the degradation of root exudates andcomplex aromatic compounds, were identified by gas chromatography/mass spec-trometry (GC/MS) in sediment samples under the mature cottonwood tree. Else-where at the site, transpiration and degradation by the cottonwood trees appears tobe responsible for loss of chlorinated ethenes.

KEY WORDS: chlorinated solvents, phreatophytes, anaerobic microorganisms,bioremediation.

240103.pgs 3/17/03, 1:48 PM73

74

Godsy et al.

I. INTRODUCTIONThe terrace alluvial aquifer that underlies the Naval Air Station (NAS) in Fort

Worth, Texas, is contaminated with TCE. Dissolution and transport of TCE, itsdegradation products, and associated chlorinated compounds have created a plumeof contaminated ground water that extends beneath the NAS and Carswell GolfCourse. An Air Force-led phytoremediation demonstration project (Figure 1) isunderway to determine if cottonwood trees are effective in the removal and degra-dation of TCE from a Type 3 behavior, shallow oxic ground water (U.S. EPA, 1998).Type 3 behavior describes a low native organic carbon, oxic aquifer where TCE andDCE do not further degrade to the daughter products, vinyl chloride (VC) and ethene(C2H4).

This report discusses the results of a reconnaissance for microbial activity con-ducted in the alluvial aquifer under Carswell Golf Course during February and June1998. The focus of this portion of the U.S. Geological Survey research was todetermine the nature of the microbial community in soil under the cottonwood treesand to determine if the microbial community had evolved into one that would supportthe reductive dechlorination of TCE and its daughter products.

A. Site DescriptionThe study area is located on the northwest side of Fort Worth, Texas (Figure 1).

The project area is 10 hectares on the northeast corner of Carswell Golf Course. Theland surface slopes gently southeast and more steeply to the southwest from RoaringSprings Road toward Farmers Branch Creek where the slope is nearly flat. FarmersBranch is incised through the terrace alluvium, and cross-bedding can be seen in thecut bank near the site.

Two cottonwood tree plots, approximately orthogonal to ground-water flow, wereprepared and planted in April 1996. The plots are 15 × 75 m each, consisting of 6rows of 50 equally spaced plantings. The westernmost plot was planted with lesscostly cottonwood whips (cuttings). The second plot was planted with 2.5 to 4 cmcaliper (trunk diameter) cottonwood trees.

Concentrations of TCE in ground water decrease from >600 µg/L in the back-ground near Roaring Springs Road to approximately 50 µg/L near the maturecottonwood at Farmers Branch Creek; however, DCE remains fairly constant through-out the background area and under the plantings with an average value of 160 µg/L.DCE increases to 250 µg/L only at Farmers Branch Creek near the mature tree. Thisconcentration gradient is perpendicular to the ground-water flow, suggesting that thecottonwood tree plots are responsible for the TCE and DCE decrease throughsorption, volatilization, and possibly degradation. The ratio of TCE/DCE remainsconstant throughout the concentration range of TCE, with the exception of the groundwater near the mature cottonwood tree. The TCE/DCE ratio at this site is much lower(~ 4 vs. 0.2) and VC has been detected (Lee et al., 2000), suggesting Type 2 behavior(U.S. EPA, 1998). This type of behavior is described as having sufficient nativeorganic carbon (from root exudates of the mature cottonwood) and the absence ofdissolved O2 to facilitate the microbial reductive dechlorination TCE to DCE, DCEto VC, and VC to C2H4.

240103.pgs 3/17/03, 1:48 PM74

75

Role of Microbial Reductive Dechlorination of TCE

FIGURE 1. Drawing of the phytoremediation site at the Carswell Golf Course, Fort Worth,Texas with associated sampling sites. TCE concentrations profiles (µg/L) forFebruary 1998 (Lee et al. 2000) are also shown.

240103.pgs 3/17/03, 1:48 PM75

76

Godsy et al.

B. HydrogeologyThe terrace alluvium consists of poorly sorted, cross-bedded unconsolidated

gravel, sand, silt, and clay. These sediments were deposited over an eroded Creta-ceous surface composed of massive limestone (the Goodland Limestone) or a shellconglomerate (the Walnut Formation). This Cretaceous surface forms the top of theGoodland-Walnut confining unit, which is the base of the terrace alluvial aquifer(Kuniansky et al., 1996). The terrace alluvial deposits range from 2 to 4.5 m thickat the site. Saturated thickness in the unconfined aquifer varies, but normally rangesfrom 0.3 to 1.5 m thick in the silt, sand, and gravel just above the confining unit.

C. Biochemical ReactionsUnder anoxic conditions, microbially mediated reductive dechlorination of TCE

and its daughter products occurs when microorganisms transfer electrons fromreduced organic compounds to the chlorinated compounds. TCE is reduced to DCE;DCE to VC; and VC to C2H4. Hydrogen (H2) can be used as a cofactor carrying theelectrons. The chlorine atom is removed from the compound as a chloride ion asfollows:

The microbial community’s ability to reductively dechlorinate TCE depends onconditions present in an aquifer. Ideally, microorganisms will pair oxidation (elec-tron donor) and reduction (electron acceptor) reactions that yield the most energy(Vogel et al., 1987; Stumm and Morgan, 1981). Thus with the same organic sub-

strate, microorganisms will preferentially use O2, NO3- , Fe3+ , SO4

2-, and CO2 in that

order. Theoretically, TCE can act as an electron acceptor under conditions normallythought of as iron reducing; however, there is evidence that TCE dechlorinationoccurs at a significant rate only under methanogenic conditions, and, furthermore,that stronger reducing conditions are required for the dechlorination of the daughterproducts DCE and VC (Friedman and Gossett, 1989; DeBrunin et al., 1992; DiStefanoet al., 1991).

II. MATERIALS AND METHODSA. Core Sample Collection

Thirteen boreholes were sampled during February and June 1998. The sedimentsamples were collected in clear polycarbonate liners using a Geoprobe® approxi-mately 30 cm below the water table (3 to 4 m below land surface). The liners werecut and, with oxygen-free argon flowing over the sediment, about 10 g of sedimentwas scooped out and placed in 25 × 142 mm anaerobic isolation roll streak tubes

240103.pgs 3/17/03, 1:48 PM76

77


(Bellco Glass Inc., Vineland, N.J.) filled with 20 mL of mineral salts solution thatwas prereduced and anaerobically sterilized (PRAS), as described by Holdeman andMoore (1972) and contained 10 mg/L Tween 80® (a nonionic surfactant added toremove microorganisms from the sediment) (Bekins et al., 1999). The tubes werethen sealed and allowed to stand for 2 h to permit penetration of the Tween 80 intothe sample. The tubes were opened with oxygen-free argon flowing over the surfaceand sonicated (10W for 30 s) to dislodge microorganisms into the mineral salts usinga Branson Sonifer, Model 200, equipped with a microtip (Branson Corp., Danbury,Conn.). The samples were stored no more than 4 h on ice before microbial enumera-tion.

B. Water Sample CollectionAfter the core-barrel was removed from the well bore, a 2-m section of 2.5-cm

PVC pipe with a 15-cm slotted screen that was previously cleaned with a detergentsolution, rinsed with tap water, rinsed with deionized water, disinfected with 70%ethanol, and finally rinsed with deionized water was lowered into the bore hole.Water samples were removed using a peristaltic pump equipped with Teflon®-linedtubing that had been cleaned previously as above. Water samples were collected afterachieving stable pH and temperature values. Generally, 5 to 10 casing volumes werediscarded before samples were collected.

Water samples for cation analyses were first collected in a 60-mL syringe. Thewater samples were then filtered directly into 40-mL volatile organic analysis (VOA)vials using a 0.2-µm nylon filter after 5 mL of the filtered water sample wasdiscarded. Samples were preserved with three drops of 1 M HCl. Samples for anionanalyses were collected as above and filtered directly into 40-mL VOA vials, withouta chemical preservative. Both cation and anion samples were chilled at 4˚C untilovernight shipment to the laboratory for analyses.

Water samples for dissolved gas analyses of methane (CH4), C2H4, ethane (C2H6),and CO2 were collected in 60-mL syringes and injected into evacuated 30-mL serumbottles until atmospheric pressure was reached. The samples were preserved with 50mg/L Hg as HgCl2. Methane and CO2 were analyzed by GC-TCD as described byGodsy et al. (1992). From the same sample bottle, CH4, C2H4, and C2H6 wereanalyzed by GC-FID using a Baseline 1010 chromatograph equipped with a 2 m 1/8″aluminum column packed with 60/80 mesh Molecular Sieve 5A. Helium was usedas the carrier gas at 30 mL/min. The column and injection port were maintained at75˚C.

Twenty mL water samples for microbial enumeration were collected with adisposable 60-mL sterile syringe and injected into a sterile 30-mL glass serum bottle.The serum bottles contained an O2-free N2 atmosphere sealed with a butyl rubberstopper and secured with an aluminum crimp seal. All water samples were stored onice for no more than 4 h before microbial analyses on site.

C. Ion Chromatography MethodsCations were determined by Ion Chromatography (IC) using a Dionex IC System

equipped with a conductivity detector, a Cation Self-Regenerating Suppressor in

240103.pgs 3/17/03, 1:48 PM77

78

Godsy et al.

Auto-Suppression External Water Mode, and a CS12 analytical column with a guardcolumn and a cation trap column. An autosampler with a 50-µL sample loop wasused. A H2SO4 acid gradient elution profile (Dionex document 034657, revision 1)was used. Cation concentrations were determined using a 5-point external calibrationcurve.

Anions and volatile fatty acids were determined by Ion Chromatography (IC)using a Dionex IC System equipped with a conductivity detector, an Anion Self-Regenerating Suppressor in Auto-Suppression External Water Mode, and an AS15analytical column maintained at 30˚C with a guard column and an anion trap column.An autosampler with a 50-µL sample loop was used. A NaOH gradient elution profile(Dionex document 034791, revision 01) was used. Anion concentrations were deter-mined using a 5-point external calibration curve.

D. Microbial Number DeterminationsMicrobial concentrations in both water and sediment samples were determined

using a five-tube Most Probable Number (MPN) analysis. Samples were seriallydiluted by an order of magnitude into PRAS mineral salts solution. Aliquots of thedilutions were inoculated into six different media, designed to promote growth andthe enumeration of aerobic and anaerobic heterotrophic, denitrifying, iron-reducing,sulfate-reducing, and methanogenic microorganisms as described by Bekins et al.(1999).

E. GC/MS Analyses of Water and Sediment SamplesWater samples were analyzed for low molecular weight, hydrophilic acid com-

pounds by first acidifying the water sample to a pH<2 with H2SO4. The acidifiedwater sample was extracted into an equal volume of diethylether by shaking themixture for ~2 min. The diethylether layer was removed and passed through a pipettepacked with anhydrous CaCl2 to remove any residual water. The dried diethyletherextract was analyzed for acidic compounds by injecting 1 µL into a Finnigan GCQGC/MS equipped with a 30-m DB-WAX fused silica carbowax column. The injec-tion port was maintained at 250˚C in the splitless mode for 0.8 min after injection,then switched to split mode for the rest of the run. The column oven was programmedto hold at 35˚C for 5 min, then ramp at 10˚C/min to 240˚C, and then hold for 5 min.

Sediment samples were dried to a free-flowing condition with the addition ofNa2SO4. A volume of diethylether equal to the weight of the sediment was then addedand the mixture was sonicated using a Branson Sonifier equipped with a standard tipto extract the acidic compounds. The sonicator was set to a 50% duty cycle for 30min at a power setting of 30 watts. The diethylether was then centrifuged to removeany particulate matter before injection into the GC/MS.

III. RESULTSIn order to determine if the trees were influencing geochemical and microbial

environments, the site was divided into areas based on the presence or absence oftrees and contamination: upgradient of planted trees; within the whips and calipers;

240103.pgs 3/17/03, 1:48 PM78

79


underneath mature cottonwoods; downgradient from the tree stands; and a back-ground site without contamination.

A. Microbiologically Significant CompoundsNitrate, Fe2+, SO4

2-, CH4 are microbiologically significant compounds. Nitrate

and SO42- are used as terminal electron acceptors (TEAs), and Fe2+ and CH4 are the

products of the reduction of Fe3+ and CO2. The presence and concentrations of theseconstituents indicates which TEA process prevails. Nitrate concentrations around thecottonwoods are 6 to 11 mg/L (Figure 2). Star radial diagrams in Figure 2 present thesubstituent concentrations on the individual radial arms. The shaded interior areaallows for quick visual differentiation of the chemistry at each location. Upgradientfrom the trees (BUSGSTA001), the concentration was 20 mg/L, whereas the concen-tration near the creek (BUSGSTA005) was 26 mg/L.

All the boreholes have similar SO42- concentrations of 20 to 40 mg/L except

BUSGSTA005, which was much higher (218 mg/L). Water samples from this sitealso have approximately twice the fluoride, four times the chloride, three times thesodium, three times the nitrate, one-third the potassium, twice the magnesium, andtwice the calcium than other sites in the area (Figure 3). The marked differences inthe chemistry indicate that the water from this borehole comes from a differentsource, or has been affected by golf-course fertilization. There were very lowconcentrations of reduced iron found in the boreholes; only BUSGSTA007A andBUSGSTA011 had measurable levels (0.06 mg/L) of reduced iron. Only traceamounts of CH4 (< 0.02 mg/L) were found in any of the boreholes with the exceptionof BUSGSTA007A that contained 0.4 mg/L.

B. Microbial PopulationsAt the background site, BUSGSTA001, low numbers of aerobes, fermenters,

and denitrifiers (102/g and /mL) were present both attached to the sediment (Figure4) and in the water sample (Figure 5). It is interesting to note that these numbersare indistinguishable within the 95% confidence interval. When comparing micro-bial numbers determined by a 5-tube MPN using 10-fold dilutions, the 95%confidence interval is a factor of 3.3 (Meynell and Meynell, 1970). Orders ofmagnitude differences in MPNs are statistically different using the Student’s t test,and the 95% confidence intervals do not overlap (Jones, 1979). Because the aquiferis generally oxic, with the exception of areas around the larger mature trees, thesimilar numbers may reflect a population of facultative organisms living aerobi-cally and utilizing the native carbon on the sediments or dissolved in the aquifer.When cultured in the various media, these organisms may grow aerobically, eitheras heterotrophic fermenters, or as denitrifying organisms. The small numbers ofiron reducers may reflect the ability of some denitrifying organisms to utilize Fe3+

as an electron acceptor. These observations are corroborated by the very lownumbers of sulfate reducers (~10/g and /mL) and the complete lack of methanogenicmicroorganisms.

240103.pgs 3/17/03, 1:48 PM79

80

Godsy et al.

FIGURE 2. Star radial diagrams of Carswell phytoremediation site showing the concentra-tions of microbiologically significant compounds at each of the borehole sites.Refer to legend for key to the arms’ labels and concentration values. All valuesare in mg/L. Values less than method detection levels (MDL) are plotted at thelowest tick mark. Those samples without values are not plotted.

240103.pgs 3/17/03, 1:48 PM80

81


FIGURE 3. Star radial diagrams of Carswell phytoremediation site showing the concentra-tions of inorganic compounds at each of the borehole sites. Refer to legend forkey to the arms’ labels and concentration values. All values are in mg/L. Valuesless than method detection levels (MDL) are plotted at the lowest tick mark.

240103.pgs 3/17/03, 1:48 PM81

82

Godsy et al.

FIGURE 4. Star radial diagrams of the Carswell phytoremediation site showing MPNs ofsix physiological microbial types present on the sediment at each of the boreholesites. Refer to legend for key to the arms’ labels and MPN values. All values arein numbers per g dry weight sediment. Values less than MDL are plotted at thelowest tick mark. Those samples without values are not plotted.

240103.pgs 3/17/03, 1:48 PM82

83


FIGURE 5. Star radial diagrams of the Carswell phytoremediation site showing MPNs of sixphysiological types in the water at each of the borehole sites. Refer to legend forkey to the arms’ labels and MPN values. All values are in numbers per mL ofwater. Values less than MDL levels are plotted at the lowest tick mark. Thosesamples without values are not plotted.

240103.pgs 3/17/03, 1:48 PM83

84

Godsy et al.

Within the area of the phreatophyte remediation sites (BUSGSTA002, 003, 008,009, 010, and 011), the microbial populations are similar to the background sites,with the exception of the increased numbers of anaerobic microorganisms and thepresence of methanogenic microorganisms. The numbers of aerobes and fermentersgenerally increased by several orders of magnitude for the February 1998 sampling(BUSGSTA002 and 003), but remain essentially the same in numbers as the back-ground for the June 1998 sampling (BUSGSTA008, 009, 010, and 011). The majordifference is the smaller numbers or lack of these organisms in the water. Theobservations that the organisms are generally attached to the sediment suggest thatthe populations are stressed and may not be metabolically active due to the extremelylow rainfall amounts during this time. Peyton et al. (1995) found that growth andmetabolic activity in response to increased substrate resulted in an increase in thepercentage of free-living microorganisms when compared to the attached microor-ganisms. The numbers of iron-reducing bacteria were approximately two orders ofmagnitude greater than background for the February sampling and much smaller ornonexistent for the June sampling. The sulfate-reducing bacteria were not signifi-cantly different from background for the February sampling, but as in the case ofaerobes and fermenters, smaller numbers or lack of sulfate reducers in water sampleswere seen for the June samples. The presence of attached methanogenic microorgan-isms in this area (BUSGSTA002) suggests that reducing conditions were developingbelow the phreatophyte remediation areas.

BUSGSTA006 and 007A are in the area of the mature cottonwood tree and exhibitequal or higher microbial populations than background for all physiological types ofmicroorganisms. The relative high percentage of the population in the water samplesis consistent with a microbial population that is active. The presence of methanogensat these sites is also indicative of an anaerobic microbial population that is activelydegrading organic compounds.

The microbial population downgradient in the direction of groundwater flow(BUSGSTA004) appears to be affected by the phreatophytes in the remediationareas. The population is indicative of reducing conditions shown by the higherconcentrations of iron-reducing and sulfate-reducing bacteria, and the presence ofmethanogenic microorganisms. These populations appear to be similar to the popu-lations present at BUSGSTA006 and 007A. The microbial population present atBUSGSTA005 is similar in numbers and composition to that at BUSGSTA004, 006,and 007A suggesting an active anaerobic ecosystem is present at this site.

C. Organic Compound AnalysisThe 11 water samples and sediment samples from BUSGSTA001 and BUSGSTA007

were analyzed for low-molecular-weight, hydrophilic acid compounds by GC/MS.Only BUSGSTA007, the sediment sample from beneath the mature cottonwood tree,contained any identifiable acidic compounds by the GC/MS method. Phenol, benzoicacid, and acetic acid were identified and confirmed by comparison of mass spectra andretention time with known standards. A cyclic hydrocarbon with side chain(s) that mostlikely had undergone partial biodegradation was also detected but could not be iden-tified. The identified compounds are the most common intermediates observed in

240103.pgs 3/17/03, 1:48 PM84

85


anaerobic ecosystems where complex organics are undergoing biodegradation (Godsyet al., 1996) and are consistent with the complex organic root exudates at this site.These compounds are most likely acting as electron donors for the reductivedehalogenation of the chlorinated ethenes present at the site.

Sediment samples throughout the field demonstration site that were stored in ananaerobic glove box for approximately 12 months to facilitate O2 removal werecapable of dechlorinating TCE to DCE when anaerobically incubated with contami-nated ground water. This observation strongly suggests that the necessary microbialpopulation and sufficient organic compounds are present on the sediments and thepopulation will become active when environmental conditions are conducive forreductive dechlorination.

IV. DISCUSSIONMicrobiological and geochemical analyses of borehole water and sediment samples

from the terrace alluvial aquifer at the phytoremediation demonstration area haveshown that an anaerobic microbial ecosystem exists near a mature cottonwood treeapproximately 30 m downgradient from the caliper trees. Near this tree, a markedlyanaerobic microbial ecosystem consisting of elevated numbers of iron- and sulfate-reducing bacteria and culturable methanogenic microorganisms has developed. Aconcomitant change in geochemistry at this site is evidenced by increased oxygenconsumption, iron reduction, and measurable CH4 production. Microbial reductivedechlorination of TCE is apparent by the increased appearance of intermediatedegradation products DCE and VC (Lee et al., 2000) in the field and confirmed bythe reductive dechlorination of TCE in laboratory microcosms by the resident micro-bial community in both the sediment and ground water under the cottonwood tree.

As of 1998, the planted cottonwood trees were beginning to affect the microbiol-ogy and geochemistry at the phytoremediation demonstration site. Dissolved oxygenconcentrations are decreasing at the whip plantings and concomitantly, Fe2+ anddissolved H2 concentrations are increasing (Lee et al., 2000). The increase in theanaerobic bacteria in general and the appearance of methanogenic microorganismsin this area suggests that the areas under the cottonwood plantings are evolvingtoward an anaerobic ecosystem; however, the δ13C values for TCE and the constantconcentrations of DCE suggests that volatilization and transpiration account for theloss of TCE (Lee et al., 2000) to date. As the ecosystem evolves toward a maturemethanogenic ecosystem, microbial reductive dechlorination will become a moreprominent process in the total loss of TCE from the system.

TCE is subject to microbial reductive dechlorination according to the equationspresented. The efficiency of dechlorination, however, appears to differ undermethanogenic, sulfate-reducing, iron-reducing, and nitrate-reducing conditions. Thedechlorination of TCE to DCE is favored under mildly reducing conditions such asNO3

- or iron reduction (Vogel et al., 1987), whereas the dechlorination of DCE toVC or of VC to ethene requires the more strongly reducing conditions ofmethanogenesis (DeBrunin et al., 1992; DiStefano et al., 1991; Freedman andGossett, 1989).

240103.pgs 3/17/03, 1:48 PM85

86

Godsy et al.

The anions used as electron acceptors (NO3- and SO4

2-) do not significantlydecrease during travel through the demonstration area, even under the mature cotton-wood tree. The observation that all possible electron acceptors exist in this evolvinganaerobic ecosystem suggests that the system may be carbon limited. Dissolvedorganic carbon (DOC) concentrations in the phytoremediation area range from 0.8 to1.8 mg/L with typical concentrations of 0.8 to 1.1 mg/L; however, samples frombeneath the mature cottonwood tree range from 1.7 to 1.8 mg/L DOC (Lee et al.,2000). The concentrations indicate that the root system at this location is producinglabile organic carbon that enters the ground water. Analyses of acid, base, and neutralfractions of hydrophobic and hydrophilic organic compounds composing the dis-solved organic carbon show that the principle compound classes at the site arehydrophobic neutral and acid, and hydrophilic acid. Concentrations of these fractionsare less than 0.4 mg/L for most samples. Water samples taken from wells beneath themature cottonwood contained the greatest concentrations of hydrophilic acids, rang-ing from 0.6 to 0.9 mg/L (Lee et al., 2000). GC/MS analyses of the sediment samplestend to confirm these results. Phenol, benzoic acid, and acetic acid were identifiedonly in samples from under the mature cottonwood tree.

The presence of iron-reducing bacteria and appearance of Fe2+ in the ground waterresulting from microbial reduction of Fe3+

on the sediments is in corroboration withthe dissolved H2 concentrations (Lee et al., 2000), suggesting that iron reduction isthe main anaerobic process occurring to date. Decaying root tissue, root exudates ofhydrophilic acids, and naturally occurring sediment organic material appear to be thesource of oxidizable organic carbon used by the microorganisms as electron donorsfor the process of reductive dechlorination. Maturation of the newly planted trees andthe root systems at the demonstration site should similarly increase the labile organiccarbon content of the shallow aquifer and allow for stronger reducing conditions todevelop.

ACKNOWLEDGMENTSThe Environmental Security Technology Certification Program of the Department

of Defense, the Superfund Innovative Technology Evaluation Program, and theAeronautical System Center/Environmental Management Directorate at Wright-Patterson AFB funded this research.

REFERENCESBekins, B.A., Godsy, E.M., and Warren, E. 1999. Distribution of microbial types in an aquifer

contaminated by crude oil. Microb. Ecol. 37, 263–275.

DeBrunin, W.P., Kotterman, M.J.J., Posthumus, M.A., Shraa, G., and Zehnder, A.J.B. 1992.Complete biological reductive transformation of tetrachloroethene to ethane. Appl. Environ.Microbiol. 58, 1996–2000.

DiStefano, T.D., Gossett, J.M., and Zinder, S.H. 1991. Reductive dechlorination of highconcentrations of tetrachloroethene to ethene by an anaerobic enrichment culture in theabsence of methanogenesis. Appl. Environ. Microbiol. 57, 2287–2292.

240103.pgs 3/17/03, 1:48 PM86

87


Friedman, D.L. and Gossett, J.M. 1989. Biological reductive dechlorination of tetrachloro-ethylene and trichloroethylene to ethylene under methanogenic conditions. Appl. Environ.Microbiol. 55, 2144–2151.

Godsy, E.M., Goerlitz, D.F., and Grbic´-Galic, D. 1992. Methanogenic biodegradation ofcreosote contaminants in natural and simulated ground-water ecosystems. Ground Water30, 232-242.

Godsy, E.M., Goerlitz, D.F., and Grbic´-Galic, D. 1996. Pathways of methanogenic biodegra-dation of creosote-derived aromatic compounds. In: U.S. Geological Survey Toxic Sub-stances Hydrology Program—Proceedings of the Technical Meeting, Colorado Springs,Colorado, September 20-24, 1993, pp. 835-842, (Morganwalp, D.W. and Aronson, D.A.,Eds.). U.S. Geol. Survey Water-Resources Inv. Rept. 94-4015, vol. 2.

Holdeman, L.V. and Moore, W.E.C. 1972. Anaerobe Laboratory Manual (2nd edition). Vir-ginia Polytechnic Institute and State University, Blacksburg, Virginia.

Jones, J.G. 1979. A Guide to Methods for Estimating Microbial Numbers and Biomass inFresh Water: Freshwater Biological Association, Scientific Publication No. 39. Cumbria,LA220LP, England, 112 p.

Kuniansky, E.L., Jones, S.A., Brock, R.D., and Williams, M.D. 1996. Hydrogeology at AirForce Plant 4 and vicinity and water quality of the Paluxy Aquifer, Fort Worth, Texas.U.S. Geol. Survey Water-Resources Inv. Rept. 96-4091, 41 p.

Lee, R.W., Jones, S.A., Kuniansky, E.L., Harvey, G.J., Sherwood-Lolar, B., and Slater, G.F.2000. Phreatophyte influence on reductive dechlorination in a shallow aquifer containingTCE. Int. J. Phytorem. 2, 193–211.

Meynell, G.G. and Meynell, E. 1970. Theory and Practice in Experimental Bacteriology.Cambridge University Press, London N.W.1, 347 p.

Peyton, B.M., Skeen, R.S., Hooker, B.S., Lundman, R.W., and Cunningham, A.B. 1995.Evaluation of bacterial detachment rates in porous media. Appl. Biochem. Biotech. 51,785–797.

Stumm, W. and Morgan, J.J. 1981. Aquatic Chemistry. John Wiley & Sons, Inc., New York,780 p.

U.S. Environmental Protection Agency. 1998. Technical Protocol for Evaluating NaturalAttenuation of Chlorinated Solvents in Ground Water, Office of Research and Develop-ment, Washington, DC, EPA/600/R-98/128.

Vogel, T.M., Criddle, C.S., and McCarty, P.L. 1987. Transformation of halogenated aliphaticcompounds. Environ. Sci. Technol. 21, 722–736.

240103.pgs 3/17/03, 1:48 PM87

The Role of Microbial Reductive Dechlorination of TCE at a ...

Documents

Transcript of The Role of Microbial Reductive Dechlorination of TCE at a ...