1609

Environmental Toxicology and Chemistry, Vol. 21, No. 8, pp. 1609–1617, 2002q 2002 SETAC

Printed in the USA0730-7268/02 $9.00 1 .00

EFFECT OF SORPTION AND DESORPTION RESISTANCE ON AEROBICTRICHLOROETHYLENE BIODEGRADATION IN SOILS

SANGJIN LEE,† WILLIAM M. MOE,† KALLIAT T. VALSARAJ,‡ and JOHN H. PARDUE*††Department of Civil & Environmental Engineering, Louisiana State University, Baton Rouge, Louisiana 70803, USA

‡Gordon A. and Mary Cain Department of Chemical Engineering, Louisiana State University, Baton Rouge, Louisiana 70803, USA

(Received 17 May 2001; Accepted 4 February 2002)

Abstract—Biodegradation of trichloroethylene (TCE) by toluene-degrading bacteria was measured under aerobic conditions inaqueous and soil-slurry batch microcosms. For soil-phase experiments, a freshly contaminated soil and a soil containing only thedesorption-resistant fraction of TCE were tested. In both cases, presence of soil resulted in biodegradation rates substantially lowerthan those determined in the absence of soil. In aqueous-phase experiments, an appreciable increase in the rate and extent of TCEbiodegradation was observed in microcosms when toluene was added multiple times. The TCE degradation rates were clearlycorrelated with toluene dioxygenase (TOD) enzyme activity over time, thus providing an indication of the cometabolic pathwayemployed by the microbial population. In soil-slurry experiments containing freshly contaminated soil, a TCE degradation rate ofapproximately 150 mg TCE/kg/h was observed during the first 39-h period, and then the TCE degradation rate slowed considerablyto 0.59 and 0.84 mg TCE/kg/h for microcosms receiving one and two additions of toluene, respectively. The TCE degradation ratesin soil-slurry microcosms containing the desorption-resistant fraction of TCE-contaminated soil were approximately 0.27 and 0.32mg TCE/kg/h in microcosms receiving one and two additions of toluene, respectively. It is clear from these results that mass transferinto the aqueous phase limited bioavailability of TCE in the contaminated soil.

Keywords—Bioavailability Cometabolism Soil remediation Trichloroethylene Toluene dioxygenase enzyme

INTRODUCTION

Trichloroethylene, a suspected carcinogen and U.S. Envi-ronmental Protection Agency priority pollutant, is a commongroundwater and soil contaminant in the United States. Varioustypes of aerobic bacteria, including methanotrophs [1–3], tol-uene oxidizers [4,5], phenol oxidizers [6], and ammonia ox-idizers [1], have been shown to cometabolically degrade TCE.Growth of these bacteria on substrates other than chlorinatedaliphatic compounds involve nonspecific oxygenase enzymesthat are able to fortuitously degrade these compounds. In tol-uene-degrading microorganisms, for example, TCE can be de-graded by means of both dioxygenase [4,5,7,8] and mono-oxygenase [9] enzyme systems. The relationship between com-etabolic TCE transformation and toluene consumption hasbeen described qualitatively by many researchers over the pastdecade [5,10–15].

Most previous studies have been performed in the absenceof soil, and little is known about how sorption of TCE to soilaffects its degradation rate by cometabolic reaction pathways.Although degradation of organic pollutants in soil and wateris often the direct result of microbial activity, sorption of theorganic compounds to soil or sediments may reduce avail-ability of the organic molecules to microorganisms and therebyslow biodegradation. This phenomenon has been observed fora variety or organic contaminants in soil. Zhang and Bouwer[16] studied the bioavailability of benzene, toluene, and naph-thalene in a soil–water slurry and noted that the rate of bio-degradation decreased with increasing organic compound hy-drophobicity, soil-to-water ratio, soil particle size, and soilorganic carbon content, suggesting that the contaminant bio-

* To whom correspondence may be addressed (jpardue@lsu.edu).

availability was affected by soil characteristics and the rate ofcontaminant desorption.

Contaminant bioavailability depends on physicochemicalprocesses such as adsorption/desorption, diffusion, and dis-solution [17,18]. A substantial body of evidence indicates thata fraction of the contaminants are inaccessible for biodegra-dation in soils and sediments contaminated for prolonged pe-riods of time [19–25]. This aging, as it is usually referred to,may result from a number of processes, including chemicaloxidation reactions incorporating contaminants into soil or-ganic matter, slow diffusion into very small pores and ab-sorption into organic matter, or the formation of films aroundnonaqueous-phase liquids with a high resistance toward or-ganic-to-water mass transfer [26].

Several researchers have confirmed that biodegradation canbe limited by the slow rate of desorption of organic compounds[19–21]. In earlier papers, we reported that desorption kineticsfor TCE showed a biphasic desorption pattern with one fractionthat is readily desorbed and a second fraction that was resistantto desorption [27]. The rate and extent of desorption weredependent on soil and sorbate properties, such as soil organiccarbon content, cation exchange capacity, specific surface area,and water solubility, as was also reported by other researchers[28–31].

For the soil phase, little information on the role of toluene-degrading bacteria and degree of bioavailability of TCE incontaminated soil is available, to the best of our knowledge.Studies described herein were conducted to determine the ef-fect of toluene on TCE degradation in both aqueous and soil-slurry phases. For the aqueous phase, TCE degradation wasinvestigated in the presence and absence of toluene with dif-ferent concentrations of toluene as a primary substrate. For

1610 Environ. Toxicol. Chem. 21, 2002 S. Lee et al.

soil-slurry experiments, a freshly contaminated soil and a soilthat was artificially aged containing only the desorption-re-sistant fraction of TCE (soil that was repeatedly extracted toremove the readily desorbing fraction) were used to studycontaminant bioavailability. Research described herein is partof ongoing research conducted in support of remediation ac-tivities at a Louisiana Superfund site known as the Petro Pro-cessors Inc. (PPI) site located in north Baton Rouge (LA,USA). Results of the studies described herein have implica-tions for determining an acceptable endpoint for site cleanupgoals.

EXPERIMENTAL METHODS

Chemicals

Trichloroethylene (99.5% purity, spectrophotometric grade)was purchased from Aldrich Chemical (Milwaukee, WI, USA)and was used as supplied. Trichloroethylene solutions wereprepared by dissolving aliquots of TCE in deionized water toobtain the desired contaminant concentration. Toluene (99.8%purity, nano grade, Fisher Scientific, Pittsburgh, PA, USA) wasused as a primary growth substrate. Both 2-propanol (99.8%purity, nano grade, Fisher Scientific) and methanol (99.9%purity, Fisher Scientific) were used for extraction of TCE fromsoil. Hydrogen peroxide (30% purity, Fisher Scientific) wasused as an oxygen supply.

Soils

Soil was collected from an uncontaminated region of theBrooklawn site (one of the two Superfund sites collectivelyknown as the PPI site) located in north Baton Rouge. Largelumps of soil were broken apart, and the material was ovendried at 558C for 24 h. The soil was then pulverized andhomogenized by passage through a sieve with 150-mm open-ings (U.S. standard sieve number 100). Samples of this soilwere methanol extracted, analyzed, and found to be free ofboth TCE and toluene.

Soil classification was performed according to standardmethods in the Louisiana State University Soil Science lab-oratory (Baton Rouge, LA, USA). The soil is classified as siltysoil, with approximately 82% (by mass) silt, 10% clay, 8%sand, and 1.35% organic carbon. All soil samples were au-toclaved before use in subsequent experiments.

Biodegradation experiments were conducted in three dif-ferent sets using aqueous-phase TCE solution, soil freshly con-taminated with TCE, and soil containing only the desorption-resistant fraction of TCE contamination. Preparation of theaqueous TCE solution and two types of TCE-contaminatedsoil are described here.

For aqueous-phase experiments, an aqueous stock solutionspiked with TCE (;6 mg/L) was prepared and added to thevials for degradation experiments to a final concentration ofapproximately 1 mg/L.

For experiments using freshly contaminated soil, 5 g ofdry, uncontaminated soil were added to 40-ml-capacity vials.An aqueous solution spiked with TCE (;6 mg/L) was addedto the vials, the vials were mixed vigorously, and then thevials were filled to capacity without headspace and cappedwith Teflont-lined silicone rubber septa. The vials were equil-ibrated in a tumbler for 3 d at 80 rpm. The vials were thencentrifuged to obtain a clear soil–water interface. The soilobtained was used in microcosm studies directly after deter-mining the soil concentration of TCE. For experiments with

soil containing only the desorption-resistant fraction of TCE,soil was further prepared following the isopropanol cosolventextraction method developed by Liu et al. [32]. In this process,equal volumes of isopropanol and electrolyte solution (0.01M NaCl, 0.01 M CaCl2) were used. Contaminated soil wasmixed vigorously with cosolvent solution in a tumbler for 24h, and then the cosolvent was separated from the soil viacentrifugation. The separated soil was rinsed with electrolytesolution two times to remove the residual isopropanol, thenmixed with cosolvent solution again and placed in tumbler foranother 24 h. An additional rinse with electrolyte solution wasconducted, and TCE concentration in the aqueous phase wasmeasured. The soil containing only the desorption-resistantfraction was obtained after centrifuging followed by rinsingwith electrolyte solution two more times.

The TCE concentration in soil was measured by a hot ex-traction method [33,34] to determine the mass of TCE boundin soil. In this process, 10 ml of methanol were added to soilremaining in the vials described previously. Vials were shakenvigorously for approximately 1 min before placement in awater bath maintained at 758C for 20 h. The supernatant, aftercentrifuging, was analyzed for the test contaminant. Spike-recovery experiments demonstrated .90% recovery of TCEusing this method (data not shown).

Glassware

Degradation experiments were conducted using 40-ml-ca-pacity U.S. Environmental Protection Agency certified cleanvials (VWR Scientific, Sugar Land, TX, USA) that were sealedwith Teflon-lined silicone rubber septa. These vials could becentrifuged at low speeds and thus eliminate the potential vol-atilization losses from transferring contents into a separatecentrifuge tube. An initial study was conducted to quantifyabiotic losses following puncturing of the septa. No significantabiotic losses were found in vials with septa punctured twotimes by the needle of a 10 ml-size gastight syringe (Hamilton,Baton Rouge, LA, USA). However, as a precautionary mea-sure, Teflon tape attached with laboratory tape was placed overpunctured septa in degradation experiments to minimize pos-sible losses of TCE through the puncture.

TCE and toluene analysis

Aqueous-phase TCE and methanol extracts containing TCEwere analyzed using a purge-and-trap attached to an HP 6890gas chromatograph (Hewlett-Packard, Avondale, PA, USA)equipped with a capillary column and mass selective detector(Model HP 5972A) and an autosampler (Tekmar, Model 2016,Tekmar-Dohrmann, Cincinnati, OH, USA). The injection portwas operated in electronic pressure control split inlet mode,and flow was purged to the split vent at 100 ml/min for 2 minwith N2 gas. Pressure was set at 18 psi, and the flow rate ofN2 carrier gas to the column was 2.0 ml/min. A capillarycolumn (30.0 m 3 250 mm 3 0.25 mm, Model HP 19091S-433) filled with 5% phenyl methyl siloxane was used. Theoven temperature was held at 2808C for 1.0 min, increasedat a rate of 208C/min to 758C/min in ramp 1, increased at arate of 108C/min to 808C/min in ramp 2, and then increasedat a rate of 208C/min to 2208C/min in ramp 3.

Toluene concentrations were measured using an HP 5890Agas chromatograph (GC) equipped with a HP 5971 mass spec-trometer (MS) detector. Samples were introduced into the GCby an autosampler and purge-and-trap method. The purge-and-trap of aqueous samples was performed at ambient temperature

Effect of sorption on trichloroethylene biodegradation Environ. Toxicol. Chem. 21, 2002 1611

Fig. 1. Gas-sparged bioreactor for cultivating toluene-degrading bac-teria.

using helium as a carrier gas. The GC employed a 30-m cap-illary column, with a 0.25- to 0.32-mm internal diameter and1.0-mm film thickness. The injector temperature was between2008C and 2258C, the transfer line temperature was between2508C and 3008C, and oven temperature was isothermal at808C. The total cycle time per sample was 32 min.

Culture of toluene-degrading microorganisms

A schematic diagram of the sparged-gas reactor used togrow toluene-degrading microorganisms is provided in Figure1. Compressed air from the laboratory air tap flowed throughan activated carbon filter and then a pressure regulator beforethe air passed through a glass tube equipped with a septum-filled injection port. A KD Scientific model 1000 syringe pump(Boston, MA, USA) delivered toluene from a glass gastightsyringe (Hamilton, Reno, NV, USA) through a needle thatpierced the septum into the injection port and into the airstreamat a rate of 0.2 ml toluene/h. The toluene-contaminated airthen passed through an aeration stone submerged in a 4.0-Lglass kettle reactor (Pyrex, Acton, MA, USA) with a workingliquid volume of 3.0 L. A flow meter (Manostat, New York,NY, USA) measured and regulated the airflow rate at 200 ml/min.

The reactor was filled with 3.0 L of nutrient solution con-taining the following constituents added (per liter) to distilledwater: 5.3 g Na2HPO4·12H2O, 1.4 g KH2PO4, 0.2 gMgSO4·7H2O, 1.0 g (NH4)2SO4, and 5 ml of a trace elementsolution as described by Zeikus [35]. The trace element so-lution contained the following constituents added to 1.0 L ofdistilled water: 0.2 g FeCl3·4H2O, 0.1 g MnCl2·4H2O, 0.17 gCoCl2·6H2O, 0.1 g CaCl2·2H2O, 0.1 g ZnCl2, 0.02 g CuCl2,0.01 g H3BO3, 0.01 g NaMoO4·2H2O, 1 g NaCl, and 0.02 gNa2SeO3. The reactor was maintained at ambient room tem-perature (;238C). A preliminary experiment conducted beforeinoculation of microbes indicated that the toluene concentra-tion in the reactor was 45 mg/L in the absence of microbialpopulations.

The reactor was inoculated with 50 ml of activated sludgewith a total suspended solids (TSS) concentration of 350 mg/L obtained from an ongoing experiment that utilized a cultureoriginally obtained from a municipal wastewater treatment fa-cility in Baton Rouge. The reactor was maintained withoutnutrient addition or sludge wasting for a period of 15 d. Start-ing on day 16 of reactor operation, 300 ml of mixed liquorsuspended solids (MLSS) was removed from the reactor on adaily basis, and 300 ml of nutrient solution was added to

produce a mean cell residence of 10 d. The toluene concen-tration in the reactor was found to be approximately 15 mg/L in the presence of toluene-degrading bacteria on day 30 ofoperation.

On the days when microbes were taken from the reactorfor use in experiments described in the subsequent sections,a sample was withdrawn from the reactor, and measurementswere made to determine TSS, volatile suspended solids (VSS),oxygen uptake rate, and protein content. Both TSS and VSSwere measured following standard methods [36]. Oxygen up-take rate was measured using a YSI Model 5300 biologicaloxygen monitor (Yellow Springs Instrument, Yellow Springs,OH, USA).

Protein content

Biomass samples were collected for protein analysis byplacing 10 ml of microbial suspension in a 15-ml sterile cen-trifuge tube (Nalge Nunc, Rochester, NY, USA). The tube wascentrifuged at 7,740 g for 20 min, and then the supernatantwas decanted. Next, protein was extracted from the biomassby adding 10 ml of 0.5 N NaOH and heating to 908C for 10min following a method similar to that of Herbert et al. [37]and Daniels et al. [38]. A 1.0-ml sample was removed fromthe centrifuge tube and placed in a 1.5-ml tube that was cen-trifuged at 12,100 g for 10 min to remove particulates. Proteinconcentration was measured using a modified Lowry methodusing preprepared reagents with Bovine serum albumin usedas a standard (Bio-Rad Laboratory, Richmond, CA, USA) fol-lowing the manufacturer’s instructions. Absorbance was mea-sured at 750 nm in a Shimadzu Model ultraviolet-1201 spec-trophotometer (Shimadzu, Columbia, MD, USA).

Toluene dioxygenase enzyme assay

For experiments conducted in the absence of soil, cells werecollected by centrifugation and washed once in potassiumphosphate buffer (8.72 g/L of KH2PO4, pH 7.2). Cell pelletswere resuspended in extraction fluid containing potassiumphosphate buffer (100 ml) containing 10% glycol, 10% eth-anol, 0.355 g of NADH, and 0.154 g of dithiothreitol. Thencells were disrupted by sonification (Branson Sonifier Model450, Danbury, CT, USA) and centrifuged for 10 min in anEppendorf centrifuge (10,000 g, 48C) to remove cellular debris.The clear supernatant solution was used as a source of crudecell extract in the enzyme assay described in the following.

Toluene dioxygenase enzyme activity was determined byadapting the indole oxidation assay developed by Jenkins etal. [39]. All enzyme assays were carried out at room temper-ature (;248C). Reaction mixtures containing (1 L) 4.36 g ofpotassium phosphate buffer, pH 7.2, 0.0278 g of FeSO4·7H2O,and 0.234 g of NADH were prepared. A colorimetric reactionwas initiated by addition of 5 ml of indole solution with aconcentration of 11.7 g/L. Oxidation of indole to cis-indoledihydrodiol and in turn to indigo was monitored as a functionof time by measuring the increase in absorbance at 400 nmagainst a blank containing all compounds except indole. Therate of indole oxidation was related to the TOD enzyme ac-tivity. Results of TOD enzyme activity measurements made atvarious time intervals are reported relative to the concentrationat time zero by dividing the enzyme activity by the initialenzyme activity value.

Biodegradation studies

Biodegradation experiments were conducted in three dif-ferent settings, including aqueous-phase studies, freshly con-

1612 Environ. Toxicol. Chem. 21, 2002 S. Lee et al.

Table 1. Summary of microcosm preparation for biodegradation studies in aqueous and soil-slurry phases

Aqueous phase

Toluene spike withlow concentration

A B C

Toluene spikewith high

concentration

D E

Soil phase

Freshly contaminated

F G H

Desorption resistant

I J K

Volume of TCE solution (;6 mg/L) added(ml)

Volume of cell suspension added (ml)Volume of nutrient solution added (ml)Volume of H2O2 added (ml)Toluene concentrations spiked (mg/L)Volume (ml) of NaN3 solution (1 g/L)

addedAmount of dry soil added (g)

1010100.050

2—

1010100.054.5b

——

1010100.054.5c

——

1010100.05

27b

——

1010100.05

27c

——

41a

10100.050

25.0

41a

10100.054.5b

—5.0

41a

10100.054.5c

—5.0

41a

10100.050

25.0

41a

10100.054.5b

—5.0

41a

10100.054.5c

—5.0

a The volume of trichloroethylene (TCE) solution (10 mg/L) added to each vial containing soil at the start of the sorption process.b Toluene was spiked only one time.c Toluene was spiked two times.

Fig. 2. (A) Trichloroethylene (TCE) and toluene degradation in aque-ous-phase microcosms when toluene was added in low (;4.5 mg/L)concentrations (treatment A, killed control; treatment B, one tolueneaddition; and treatment C, two toluene additions). (B) Toluene addedin high (;27 mg/L) concentration (treatment D, one toluene addition,and treatment E, two toluene additions). Arrow indicates time of tol-uene addition.

taminated soil studies, and studies using soil containing onlythe desorption-resistant fraction of the contaminant. Table 1summarizes the preparation of microcosm studies in the aque-ous and soil phases. As shown in the table, 11 different treat-ment conditions, arbitrarily named A through K, were used.

Aqueous-phase experiments were comprised of two cate-

gories: those that received low (4.5 mg/L) and high (27 mg/L) concentration toluene spikes. Treatments A was preparedas a killed control with addition of sodium azide as a biocide.Treatments B and C were spiked with toluene concentrationsof 4.5 mg/L, while treatments D and E received 27 mg/Ltoluene. Treatments B and D were spiked with toluene onlyonce (at time zero). Treatments C and E were spiked withtoluene at time zero as well as 3.8 h into the experiment. Forall noncontrol samples, 0.05 ml of 30% H2O2 was added tomicrocosms during preparation as an oxygen supply.

For the freshly contaminated soils studies, TCE was addedto treatments to reach a final concentration of approximately7 mg/kg in the soil phase (see previous description of soilpreparation). Treatment F was prepared as a killed control withaddition of sodium azide as a biocide. Meanwhile, 4.5 mg/Lof toluene was spiked in treatments G and H at time zero, andthen a second toluene spike was added to treatment H 120 hafter the experiment began.

For studies containing only desorption-resistant TCE, treat-ment I was prepared as a killed control with addition of sodiumazide as a biocide. Meanwhile, 4.5 mg/L of toluene were spikedin treatments J and K at time zero, and then a second toluenespike was added to treatment K 120 h after the experimentbegan. Soil containing only desorption-resistant TCE was pre-pared as described previously.

For all microcosm studies, 300 ml of toluene-degradingbacteria were removed from the reactor and placed in a beaker.The microbial suspension had a TSS of 315 mg/L, a VSS of260 mg/L, an oxygen uptake rate of 16.6 mg O2/L/h, and aspecific oxygen uptake rate of 63.8 mg O2/g VSS/h. The proteinconcentration was 49 mg/L, which corresponds to a biomasscontent of 19% protein based on VSS. The microbial suspen-sion originating from the culture reactor was aerated for 30min to volatilize residual toluene. Then 10.0 ml of the micro-bial suspension (2.6 mg of biomass measured as VSS) wereadded to each vial, and the vials were capped. The vials wereput on tumbler immediately and were maintained there untilvials were sacrificed at designated time intervals. The TCEbiodegradation in the soil phase was calculated after account-ing for TCE losses measured in control microcosms.

RESULTS AND DISCUSSION

Biodegradation of TCE in aqueous-phase experimentsFigure 2A depicts TCE and toluene concentrations as a

function of time in aqueous-phase degradation experiments.

Effect of sorption on trichloroethylene biodegradation Environ. Toxicol. Chem. 21, 2002 1613

As clearly shown in the figure, the TCE concentration wasessentially constant in the treatment A microcosms, wheresodium azide was added as an inhibitor. This provides an in-dication that the observed decrease in TCE in other micro-cosms was due to microbial activity rather than abiotic pro-cesses.

For microcosms receiving treatment B (toluene added onlyat time zero), toluene degradation proceeded rapidly with theconcentration decreasing from an initial concentration of ap-proximately 4.5 mg/L at time zero to a concentration of near0 mg/L after 1 h. Trichloroethylene decreased from an initialconcentration of approximately 1 mg/L at time zero to ap-proximately 0.56 mg/L at 2 h. From time 2 h to 7 h (whenthe experiment was terminated), the TCE concentration de-creased steadily, though at a rate somewhat less than duringthe first 2 h. After a total of 7 h of incubation, approximately0.3 mg/L of TCE remained in treatment B microcosms.

For microcosms receiving treatment C (same as treatmentB but with a second addition of toluene at time 3.8 h), thetoluene concentration increased from a concentration of ap-proximately zero to a concentration of approximately 4.7 mg/L during the 15-min time period following the second additionof toluene. During the next hour, the toluene concentrationdecreased rapidly to near 0 mg/L. Immediately following thesecond toluene addition, the TCE concentration decreased ata much faster rate, comparable to that during the first 2 h ofthe study. Unlike treatment B microcosms, where residual TCEwas present at the end of the experiment, the TCE concentra-tion in treatment C microcosms could no longer be detectedafter approximately 6 h of total incubation time. The secondaddition of toluene (present in treatment C but not in treatmentB) had an obvious positive effect on the rate and extent ofTCE degradation.

As shown in Figure 2B, a similar pattern was observed inaqueous-phase microcosms that received a higher concentra-tion (;27 vs 4.5 mg/L) toluene spike. For microcosms re-ceiving treatment D (toluene added only at time zero), tolueneconcentration decreased rapidly from an initial level of ap-proximately 27 mg/L at time zero to near 0 mg/L after 2 h.Likewise, TCE degradation was rapid during the first 2 h,decreasing from an initial concentration of approximately 1.07mg/L to a concentration of approximately 0.2 mg/L after 2 h.Shortly after the toluene concentration approached zero, therate of TCE degradation became relatively constant but at arate much less than that observed during the first 2 h. Ap-proximately 0.1 mg/L of TCE was still present in treatment Dmicrocosms when the experiment was terminated at time equalto 10 h. This observed result is very similar to that fromtreatment B, which was identical to treatment D in all respectsexcept for the initial toluene concentration.

Similar to treatment C microcosms, treatment E micro-cosms received toluene additions at 0 and 3.8 h. Followingthe second addition of toluene, the concentration of tolueneincreased rapidly to 27 mg/L within 30 min after addition. Ittook approximately 2 h for toluene to degrade to a concen-tration near zero. Similar to the pattern observed for TCEdegradation in treatment C, the TCE degradation rate increasedfollowing the second toluene addition, and TCE concentrationreached approximately zero after 6 h. The TCE biodegradationrates for the first 2 h of treatment and 2 h following the secondaddition were comparable. As was observed in treatment C,the second addition of toluene had an obviously positive im-pact on the rate and extent of TCE degradation.

Biodegradation rates were calculated to quantify the effectsof toluene on TCE degradation. Because biomass concentra-tions were not measured after time zero, degradation rates arereported in terms of concentration of contaminant per unit timerather than in terms of specific contaminant removal rate. Thetoluene biodegradation rates in treatment B and C microcosms(initial toluene concentration of 4.7 mg/L) were approximately5.97 and 6.16 mg toluene/L/h, respectively, during the periodimmediately following toluene addition, and these concentra-tions lasted approximately 1 h. Toluene biodegradation ratesin treatment D and E microcosms (initial toluene concentrationof 27 mg/L) were approximately 13.2 and 11.8 mg toluene/L/h, respectively, during the period immediately following ad-dition of toluene.

The TCE biodegradation rates were approximately 270 and313 mg TCE/L/h for the first 2 h following toluene additionin experiments with low (4.5 mg/L) and high (27 mg/L) initialtoluene concentration (see Fig. 2). Shortly after toluene wasdepleted from the system, the TCE degradation rate slowedconsiderably to 52 and 35 mg TCE/L/h, respectively. Signif-icant differences in TCE biodegradation rates were observedin the presence of toluene in microcosms when a second tol-uene spike was added (treatments C and E). Biodegradationrates in treatments C and E were 231 and 265 mg/L/h duringthe one and a half hours after the second toluene spike. Mean-while, the TCE biodegradation rates in microcosms not re-ceiving a second addition of toluene (treatments B and D)remained at approximately 52 and 35 mg TCE/L/h, respec-tively, during the same time period.

Interestingly, the initial concentration and total mass oftoluene supplied had less of an effect on extent of TCE deg-radation in the microcosms than did the time interval duringwhich toluene was supplied. This is clearly evident when onecompares the TCE degradation in treatment C and D micro-cosms. Treatment C microcosms received only one-sixth ofthe toluene concentration of Treatment D (4.5 vs 27 mg/L)and only one-third of the toluene mass (9 vs 27 mg/L), but itstill degraded more of the TCE. This observation suggests thatthe periodic addition of low levels of toluene can effectivelymaintain sufficient enzyme levels to yield a high rate and extentof cometabolic TCE degradation while minimizing the massof toluene that must be added. This concept, previously sug-gested but not experimentally verified by Fan and Scow [11],is further supported by TOD enzyme activity measurementsdescribed in the following. The TCE concentration in treat-ments B and D decreased to the method detection level afterthe second addition of toluene; however, if the initial TCEconcentration had been higher, it is expected that the rate ofTCE degradation would slow again shortly after all the toluenewas degraded in the system.

The increase in TCE degradation rate following tolueneaddition and the subsequent slowing of TCE degradation ratefollowing toluene depletion can be readily explained by chang-es in the TOD enzyme activity. Toluene dioxygenase enzymeactivity as a function of time for microorganisms subjected totreatments B and C is depicted in Figure 3A. As clearly shownin the figure, the TOD enzyme activity was initially high fol-lowing addition of toluene at time zero. After toluene wasdepleted from the system (;1 h; see Fig. 2), the enzyme ac-tivity decreased rapidly to a level only a fraction of its originalvalue. The time period during which TOD enzyme activitydecreased corresponds to the time period during which theTCE degradation rate slowed appreciably. The decreased TCE

1614 Environ. Toxicol. Chem. 21, 2002 S. Lee et al.

Fig. 3. (A) Relative toluene dioxygenase enzyme activity over timein microcosms with one and two additions of toluene (treatments Band C). Enzyme activity was normalized by dividing enzyme activityby the activity at time zero. (B) Protein concentration as a functionof time in treatment B and C microcosms. TOD 5 toluene dioxy-genase.

Fig. 4. Biodegradation of trichloroethylene (TCE) in soil-slurry mi-crocosms containing freshly contaminated soil (treatment F, killedcontrol; treatment G, one toluene addition; and treatment H, two tol-uene additions).

degradation rate was likely due to decreased levels of TODenzyme available to catalyze cometabolic reactions.

Addition of toluene at a time of 3.7 h in treatment C mi-crocosms caused a rapid increase in TOD enzyme activity toa level approximately the same as its initial value. At the sametime, a rapid increase in the rate of TCE degradation alsooccurred (see Fig. 2). Compared with the treatment that didnot receive a second toluene spike (treatment B), the treatmentwith a second toluene spike (treatment C) had more than fivetimes higher TOD enzyme activity 30 min after toluene ad-dition and more than 20 times higher activity 55 min aftertoluene addition. At the same time, treatment C microcosmshad a much higher TCE biodegradation rate. Enzyme concen-tration decreased again after all toluene from the second ad-dition was biodegraded. Trichloroethylene degradation by tol-uene-degrading bacteria was clearly correlated with the TODconcentrations over time. Although enzymes other than TOD(e.g., toluene mono-oxygenase) have been reported to come-tabolically degrade TCE [9], and the presence of other enzymeswas not measured in the studies described herein, changes inTOD activities reasonably account for the observed changesin TCE degradation rates over time. Furthermore, it is expectedthat enzymes from inducible toluene degradation pathways notcontaining TOD would follow a similar pattern of increase anddecrease in the experimental system tested.

Protein concentration measured as a function of time intreatment B and C microcosms are depicted in Figure 3B.

Although data were somewhat scattered, a general pattern ofincreasing protein concentration over time was observed. Thiswas likely caused by biomass growth in the microcosms astoluene was degraded. No appreciable difference was observedbetween microcosms that received a second addition of toluene(treatment C) and those that did not (treatment D). This wasnot unexpected considering the relatively small amount of bio-mass increase expected in the system.

Biodegradation of TCE in freshly contaminated soil

Figure 4 depicts TCE concentration as a function of timein microcosms that contained soil freshly contaminated byTCE. As clearly shown in the figure, the TCE concentrationwas essentially constant in the treatment F microcosms, wheresodium azide was added as a biocide. Similar to the results oftreatment A microcosm studies, this indicates that the observeddecrease in TCE concentration in other microcosms was dueto microbial activity rather than abiotic processes.

Treatment G (toluene added only at time zero) and H (tol-uene added at time zero and 120 h) microcosm studies wereconducted for approximately 480 h. In both treatments, theTCE concentration decreased from an initial concentration ofapproximately 7.0 mg/kg (mass TCE per mass dry solids) toa concentration of approximately 1.2 mg/kg after 39 h. Thiscorresponds to a TCE degradation rate of approximately 150mg TCE/kg/h during the 39-h period following addition oftoluene at time zero. The TCE degradation rates then slowedconsiderably in both treatments. The TCE biodegradation ratesin treatment G and H microcosms were approximately 0.59and 0.84 mg TCE/kg/h, respectively, for the period from 39to 480 h.

The second addition of toluene to microcosms that receivedtreatment H resulted in a distinct pattern of TCE removal asdepicted in the inner plot in Figure 4. Compared to treatmentG, the TCE concentration observed in treatment H soil extractsincreased appreciably during the 30 h following the secondaddition of toluene. A likely explanation for this is the roleof toluene as a cosolvent through which the presence of tolueneenhances desorption of TCE from soil particles. This phenom-enon is supported by previous findings that aromatic com-

Effect of sorption on trichloroethylene biodegradation Environ. Toxicol. Chem. 21, 2002 1615

Fig. 5. Desorption of trichloroethylene (TCE) in soil during the co-solvent (isopropanol) extraction process.

Fig. 6. Biodegradation of trichloroethylene (TCE) in soil-slurry mi-crocosms containing only the desorption-resistant fraction of TCE(treatment I, killed control; treatment J, one toluene addition; andtreatment K, two toluene additions).

pounds may compete with TCE for sorption sites on soil par-ticles [11].

Following the period during which an increase in TCE con-centration was observed in soil extracts, the TCE concentrationin treatment H microcosms decreased to a level less than thatin treatment G microcosms treated with a toluene spike onlyone time. Although the differences were relatively small, theextent of TCE degradation in treatment H was greater than intreatment G. When the experiment was terminated after 480h, the TCE concentrations were 0.94 and 0.83 mg/kg in mi-crocosms that received treatments G and H, respectively. Thiscorresponds to overall TCE removal of approximately 86.6and 88.1%, respectively. In both cases, the rate and extent ofTCE degradation was much less than that observed in micro-cosm experiments conducted in the absence of soil. It is in-teresting to note that approximately 96% of the TCE removaloccurred during the first 39 h of the treatment, and only about4% of the removal occurred during the subsequent 441 h.

Because toluene concentration was not measured either inthe aqueous phase of the slurry or from soil extracts, it is notknown whether toluene added to the treatment G and H mi-crocosms was present in solution or whether it was sorbed tothe soil. Likewise, because TOD enzyme measurements werenot conducted on extracts from these microcosms, it is un-known whether toluene induced TOD activity. Thus, from ex-periments described herein, it cannot be concluded unequiv-ocally whether the slow degradation rate was caused becausetoluene was not bioavailable to induce enzymes necessary forcometabolism or whether the slow degradation rate was causedbecause TCE was not bioavailable. In either case, lack ofbioavailability limited the TCE degradation rate in the con-taminated soil. This is consistent with previous reports thatphysical occlusions in soil due to entrapment in soil microporeslimit bioavailability of organic compounds [20]. This processcan be caused by the rate-limited diffusion in intraparticlewater or diffusion through the solid-phase organic matter insoil particles [40]. Limited access of microorganisms to ma-trix-bound organic compounds thereby limits the biodegra-dation rate. This is further discussed in the following section.

Error bars depicted in Figures 4 and 6 represent one stan-dard deviation from the triplicate analyses. Error may havebeen caused by analytical error associated with the methanolextraction process, GC analysis, or soil content that was notcompletely homogenized.

Biodegradation of TCE in soil containing the desorption-resistant fraction

For biodegradation studies with soil containing only thedesorption-resistant fraction of TCE, soil was prepared by sub-jecting TCE-contaminated soil, prepared in the same manneras that used for experiments described in the previous section,to a cosolvent extraction process. The TCE concentration inthe soil was measured following each extraction step, and re-sults are depicted in Figure 5. As shown in the figure, the TCEconcentration decreased from its initial level of approximately4.9 mg/kg in the freshly contaminated soil to a concentrationof 1.1 mg/kg after one extraction cycle. After the extractionprocess was complete (total of six desorption cycles), approx-imately 500 mg/kg TCE remained in the soil. The methoddescribed by Liu et al. [32] has been demonstrated to producesoil that exhibits the same contaminant partitioning behavioras soil subjected to a large number of sequential desorptionsteps in the laboratory. Following the isopropanol extraction,

the observed contaminant partitioning is indistinguishablefrom that of the desorption-resistant compartment as describedin papers by Kan et al. [30,31].

Figure 6 depicts TCE concentrations as a function of timein microcosms that contained soil with only the desorption-resistant fraction of TCE. As shown in the figure, the TCEconcentration was essentially constant in the treatment I mi-crocosms, where sodium azide was added as a biocide. Similarto the results of treatments A and F, results indicate that theobserved decrease in TCE concentration in other microcosmswas due to microbial activity rather than abiotic processes.

Biodegradation of TCE observed in microcosms receivingtreatments J (toluene added only once, at time zero) and K(toluene added twice, at time zero and 120 h) is shown inFigure 6. The TCE concentration decreased from an initiallevel of approximately 500 mg/kg at time zero to near 280 mg/kg after 75 h. Similar to the pattern observed in treatment Hmicrocosms, the TCE concentration in soil extracts in treat-ment K microcosms increased during the period followingaddition of toluene compared to the microcosms that did notreceive a second toluene addition (treatment J). Subsequently,the TCE concentration decreased from approximately 350 to210 mg/kg within 200 h after the second toluene addition. Bycomparison, nearly twice as much TCE remained in the treat-ment J microcosms. As was shown in microcosms receivingtreatment H, a cosolvent effect was observed. When the ex-

1616 Environ. Toxicol. Chem. 21, 2002 S. Lee et al.

periment was terminated after a total of 480 h, approximately260 and 220 mg/kg TCE remained in the soil receiving treat-ments J and K, respectively. This corresponds to removal of48 and 56%, respectively. As with the freshly contaminatedsoil, the second addition of toluene produced a small but no-ticeable increase in the overall extent of TCE degradation.

Because data were somewhat scattered, the average TCEdegradation rate over the entire experimental duration (480 h)was calculated for comparison purposes rather than TCE deg-radation rates for subintervals of the treatment process. TheTCE degradation rates in soil-slurry microcosms containingthe desorption-resistant fraction of TCE-contaminated soilwere approximately 0.27 and 0.32 mg TCE/kg/h, respectively,in microcosms receiving one and two additions of toluene(treatments J and K, respectively).

The rate and extent of TCE degradation was much lowerin microcosms that contained TCE-contaminated soil com-pared to microcosms that contained only aqueous-phase TCE.Additionally, the rate of TCE degradation was lower in mi-crocosms that contained only the desorption-resistant fractionof TCE compared to microcosms that contained freshly con-taminated soil. Similar reports of limited biodegradation in thesoil phase caused by the limited rate of organic compounddesorption from soil into the aqueous phase have been reportedpreviously [16]. Thus, TCE mass transfer from soil to theaqueous phase could play a significant role in limiting bio-degradation [21]. Volatile organic compounds present in soilsare subject to sorption, volatilization, and solubilization andare distributed among the soil solution, atmosphere, and solidphase. Microbial populations in soil are concentrated in thewater films that fill pores and coat soil particles, and only theaqueous phase of a chemical is generally considered to bedirectly available for uptake by microorganisms.

The lower rate and extent of biodegradation in soil-slurrymicrocosms (compared to aqueous-phase microcosms) can beattributed to bioavailability limitations caused by desorptionrate limitations of TCE, toluene, or both. The higher TCEdegradation rates observed in microcosms containing freshlycontaminated soils compared to those containing only the de-sorption-resistant fraction of TCE suggest that bioavailabilityof TCE rather than toluene caused the decreased degradationrate. On the other hand, the increased degradation rates ob-served with multiple additions of toluene suggest that limitedbioavailability of toluene needed to induce enzymes necessaryfor cometabolism may be a factor. Interpretation of these re-sults is somewhat complicated by the role that toluene mayplay as a cosolvent, thereby increasing solubilization and/orextractability of TCE. From a practical perspective, furtherresearch is needed to ascertain whether addition of toluene tothe soil resulted in residual nonbioavailable toluene contam-ination in the soil.

In terms of engineering applications, to obtain high TCEremoval efficiencies using cometabolic process, appropriatestrategies are needed for maximizing oxygenase enzyme pro-duction. According to the results in this study, multiple ad-ditions of toluene could stimulate TOD enzyme productionand subsequently TCE biodegradation. Consequently, additionof toluene may be a useful tool for enhancing TCE removaland degradation at contaminated sites.

CONCLUSIONS

Biodegradation of TCE by toluene-degrading bacteria wasmeasured under aerobic conditions in aqueous and soil-slurry

batch microcosms. In aqueous-phase experiments, TCE bio-degradation rates rapidly increased during the first 2 h follow-ing addition of either low (4.5 mg/L) or high (27 mg/L) initialtoluene concentrations. Shortly after toluene was depleted fromthe system, TCE degradation rates slowed considerably. Anappreciable increase in the rate and extent of TCE biodegra-dation was observed in microcosms when toluene was addedmultiple times, with the TCE degradation rate clearly corre-lated with TOD enzyme activity over time.

For soil-phase experiments, a freshly contaminated soil anda soil containing only the desorption-resistant fraction of TCEwere tested. In both cases, presence of soil resulted in bio-degradation rates substantially lower than those determined inthe absence of soil.

In soil-slurry experiments containing freshly contaminatedsoil, a TCE degradation rate of approximately 150 mg TCE/kg/h was observed during the first 39-h period, and then theTCE degradation rate slowed considerably to 0.59 and 0.84mg TCE/kg/h for microcosms receiving one and two additionsof toluene, respectively. The TCE degradation rates in soil-slurry microcosms containing the desorption-resistant fractionof TCE-contaminated soil were approximately 0.27 and 0.32mg TCE/kg/h in microcosms receiving one and two additionsof toluene, respectively. It is clear from these results that masstransfer into the aqueous phase limited bioavailability of TCEin the contaminated soil. These observations are in agreementwith results reported in the literature for other organic con-taminants, and they support the general theory that the longera contaminant is aged with soil, the lower the fraction of con-taminant will be bioavailable.

REFERENCES

1. Moran BN, Hickey WJ. 1997. Trichloroethylene biodegradationby mesophilic and psychrophilic ammonia oxidizers and meth-anotrophs in groundwater microcosms. Appl Environ Microbiol63:3866–3871.

2. Chu KH, Alvarez-Cohen L. 1998. Effect of nitrogen source ongrowth and trichloroethylene degradation by methane-oxidizingbacteria. Appl Environ Microbiol 64:3451–3457.

3. Fox BG, Borneman JG, Wackett LP, Lipscomb JD. 1990. Hal-oalkene oxidation by the soluble methane monooxygenase fromMethylosinus trichosporium OB3b: Mechanistic and environ-mental implications. Biochemistry 29:6419–6427.

4. Heald S, Jenkins RO. 1994. Trichloroethylene removal and ox-idation toxicity mediated by toluene dioxygenase of Pseudomo-nas putida. Appl Environ Microbiol 60:4634–4637.

5. Cox CD, Woo HJ, Robinson KG. 1998. Cometabolic biodegra-dation of trichloroethylene (TCE) in the gas phase. Water SciTechnol 37:97–104.

6. Lee CY, Cheng SZ. 1998. Trichloroethylene biodegradation byphenol-oxidizing cultures grown from various conditions. J En-viron Sci Health B 33:705–721.

7. Wackett LP, Gibson DT. 1988. Degradation of trichloroethyleneby toluene dioxygenase in whole-cell studies with Pseudomonasputida F1. Appl Environ Microbiol 54:1703–1708.

8. Li S, Wackett LP. 1992. Trichloroethylene oxidation by toluenedioxygenase. Biochem Biophys Res Commun 185:443–451.

9. Shields MS, Montgomery SO, Cuskey SM, Chapman PJ, Prit-chard PH. 1991. Mutants of Pseudomonas cepacia G4 defectivein catabolism of aromatic compounds and trichloroethylene. ApplEnviron Microbiol 57:1935–1941.

10. Fliermans CB, Phelps TJ, Ringelberg D, Mikell AT, White DC.1988. Mineralization of trichloroethylene by heterotrophic en-richment cultures. Appl Environ Microbiol 54:1709–1714.

11. Fan S, Scow KM. 1993. Biodegradation of trichloroethylene andtoluene by indigenous microbial populations in soil. Appl EnvironMicrobiol 59:1911–1918.

12. Leahy JG, Byrne AM, Olsen RH. 1996. Comparison of factorsinfluencing trichloroethylene degradation by toluene-oxidizingbacteria. Appl Environ Microbiol 62:825–833.

Effect of sorption on trichloroethylene biodegradation Environ. Toxicol. Chem. 21, 2002 1617

13. Mars AE, Prins GT, Wietzes P, Koning WD, Janssen DB. 1998.Effect of trichloroethylene on the competitive behavior of toluene-degrading bacteria. Appl Environ Microbiol 64:208–215.

14. Bielefeldt AR, Stensel HD, Strand SE. 1995. Cometabolic deg-radation of TCE and DCE without intermediate toxicity. J EnvironEngineering 121:791–797.

15. El-Farhan YH, Scow KM, Fan S, Rolston DE. 2000. Kinetics oftrichloroethylene cometabolism and toluene biodegradation:Model application to soil batch experiments. J Environ Qual 29:778–786.

16. Zhang W, Bouwer EJ. 1997. Biodegradation of benzene, tolueneand naphthalene in soil-water slurry microcosms. Biodegradation8:167–175.

17. Ogram AV, Jessup RE, Ou LT, Rao PSC. 1985. Effect of sorptionon biological degradation rates of 2,4-dichlorophenoxy acetic acidin soils. Appl Environ Microbiol 49:582–587.

18. Luthy RG, Dzombak DA, Peters CA, Roy SB, Ramaswami A,Nakles DV, Nott BR. 1994. Remediating tar-contaminated soilsat manufactured gas plant sites. Environ Sci Technol 28:266A–277A.

19. Steinberg SM, Pignatello JJ, Sawhney BL. 1987. Persistence of1,2-dibromoethane in soils: Entrapment in intraparticle micro-pores. Environ Sci Technol 21:1201–1208.

20. Pignatello JJ. 1989. Sorption dynamics of organic compounds insoils and sediments. In Sawney BL, Brown K, eds, Reactionsand Movement of Organic Chemicals in Soils. Soil Science So-ciety of America and American Society of Agronomy, Madison,WI, pp 31–80.

21. Robinson KG, Farmer WS, Novak JT. 1990. Availability of sorbedtoluene in soils for biodegradation by acclimated bacteria. WaterRes 24:345–350.

22. Scribner SL, Benzing TR, Sun S, Boyd SA. 1992. Desorptionand bioavailability of aged simazine residues in soil from a con-tinuous corn field. J Water Qual 21:115–120.

23. Hatzinger PB, Alexander M. 1993. Effect of aging of chemicalsin soil on their biodegradability and extractability. Environ SciTechnol 29:539–545.

24. Alexander M. 1995. How toxic are toxic chemicals in soil? En-viron Sci Technol 29:2713–2717.

25. Pignatello JJ, Xing B. 1996. Mechanisms of slow sorption oforganic chemicals to natural particles. Environ Sci Technol 30:1–11.

26. Bosma TNP, Middeldorp PJM, Schraa G, Zehnder AJB. 1997.Mass transfer limitation of biotransformation: Quantifying bio-availability. Environ Sci Technol 31:248–252.

27. Lee S, Kommalapati RR, Valsaraj KT, Pardue JH, Constant WD.2002. Rate-limited desorption of volatile organic compounds from

soils and implications for the remediation of a Louisiana Super-fund site. Environ Monit Assess 75:93–111.

28. Pavlostathis SG, Mathavan GN. 1992. Desorption kinetics of se-lected volatile organic compounds from field contaminated soils.Environ Sci Technol 26:532–538.

29. Valsaraj KT, Kommalapati RR, Robertson ED, Constant WD.1999. Partition constants and adsorption/desorption hysteresis forvolatile organic compounds on soil from a Louisiana Superfundsite. Environ Monit Assess 58:225–241.

30. Kan AT, Fu G, Hunter MA, Tomson MB. 1997. Irreversible ad-sorption of naphthalene and tetrachlorobiphenol to Lula and Sur-rogate sediments. Environ Sci Technol 31:2176–2185.

31. Kan AT, Fu G, Hunter MA, Chen W, Ward CH, Tomson MB.1998. Irreversible sorption of natural hydrocarbons to sediments:Experimental observations and model predictions. Environ SciTechnol 32:892–902.

32. Liu W, Kan A, Tomson M, Chen W, Hughes J, Reible D. 1999.Protocol for assessment of biological available pollutant concen-tration in soil. Extended Abstracts, 218th American ChemicalSociety National Meeting, New Orleans, LA, August 23–27, Vol39, pp 175–177.

33. Sawhney BL, Pignatello JJ, Steinberg SM. 1987. Determinationof 1,2-dibromoethane (EDB) in field soils: Implications for vol-atile organic compounds. J Environ Qual 17:149–152.

34. Pignatello JJ. 1990. Slowly reversible sorption of aliphatic hal-ocarbons in soils: 1. Formation of residual fractions. EnvironToxicol Chem 9:1107–1115.

35. Zeikus JG, Hegge PW, Anderson MA. 1979. Thermoanaerobiumbrockii gen. Nov. and sp. nov., a new chemoorganotrophic, cal-doactive, anaerobic bacterium. Arch Microbiol 122:41–48.

36. American Public Health Association, American Water Works As-sociation, Water Pollution Control Federation. 1998. StandardMethods for the Examinations of Water and Wastewater, 20thed. American Public Health Association, Washington, DC.

37. Herbert D, Phipps PJ, Strange RE. 1971. Chemical analysis ofmicrobial cells. In Norris JR, Ribbons DW, eds, Methods in Mi-crobiology, Vol 5B. Academic, New York, NY, USA, pp 209–344.

38. Daniels L, Hanson RS, Phillips JA. 1994. Chemical analysis. InGerhardt P, ed, Methods for General and Molecular Biotech-nology. American Society for Microbiology, Washington, DC, pp512–554.

39. Jenkins RO, Dalton H. 1985. The use of indole as a spectropho-tometric assay substrate for toluene dioxygenase. FEMS Micro-biol Lett 30:227–231.

40. Wu SC, Gschwend F. 1986. Sorption kinetics of hydrophobicorganic compounds to natural sediments and soils. Environ SciTechnol 20:717–725.