Final Report: Influence of Nonionic Surfactants on the Bioavailability of Chlorinated Benzenes for Microbial Reductive Dechlorination

EPA Grant Number: R825404
Title: Influence of Nonionic Surfactants on the Bioavailability of Chlorinated Benzenes for Microbial Reductive Dechlorination
Investigators: Pavlostathis, Spyros G. , Chang, Eric , Karagunduz, Ahmet , Marti, Charlotte A. , Pennell, Kurt D. , Yeh, Daniel H
Institution: Georgia Institute of Technology
EPA Project Officer: Chung, Serena
Project Period: November 12, 1996 through November 11, 1999
Project Amount: $333,348
RFA: Environmental Fate and Treatment of Toxics and Hazardous Wastes (1996) RFA Text |  Recipients Lists
Research Category: Hazardous Waste/Remediation , Land and Waste Management , Safer Chemicals

Objective:

The primary goal of this research was to conduct a systematic assessment of the benefits as well as potential limitations resulting from the use of surfactants in combination with microbial reductive dechlorination. Chlorinated benzenes, especially hexachlorobenzene (HCB), were the target compounds. The specific objectives of the research were to: (1) assess the effect of surfactants on methanogenesis and reductive dechlorination; (2) assess the anaerobic biodegradability of selected surfactants and its potential correlation to their structural features; (3) quantify the effects of surfactants on the solubilization, desorption, and reductive dechlorination of chlorinated benzenes in batch systems; (4) evaluate surfactant-enhanced desorption, biotransformation, and transport of hexachlorobenzene in column systems; and (5) develop and evaluate mathematical models to describe the coupled transport, sorption, and biotransformation of both surfactant and chlorinated benzenes.

Summary/Accomplishments (Outputs/Outcomes):

Findings from this study demonstrate that the compatibility of surfactants with the biological system considered is highly system-specific and often difficult to forecast. The study also underlines the importance of accounting for contaminant phase distribution and the impact of surfactant losses (sorption and phase separation) on contaminant availability. To arrive at an optimum system where surfactants could successfully enhance the bioavailability of strongly sequestered hydrophobic organic compounds (HOCs), further detailed delineation of the complex physical, chemical, and biological interactions among surfactant, microorganisms, contaminant, and solid-phase is required.

The experimental plan was divided into six tasks. Task I involved selection of surfactants and other materials. Task II focused on the compatibility of selected surfactants with the anaerobic processes of methanogenesis and reductive dechlorination using previously enriched, dechlorinating cultures. In Task III, the anaerobic biodegradability of selected nonionic surfactants was tested to select biodegradable surfactants that could serve as electron donors for reductive dechlorination. Task IV focused on the surfactant-enhanced desorption of HCB, as well as the quantification of surfactant sorption in soils. Batch and column experiments were conducted to evaluate and quantify the simultaneous adsorption-desorption and transport of HCB in the presence of nonionic surfactants in Task V. Experimental measurements obtained in batch and column systems were then utilized to develop and evaluate mathematical models capable of describing the coupled transport, sorption, and biotransformation of both surfactant and HCB (Task VI).

Task I: Selection of Materials

A total of 16 surfactants were evaluated in this study. Fourteen of these surfactants represented two main classes of nonionic surfactants: linear polyoxyethylene (POE) alcohols (Brij 30, and 35; Witconol SN-70, 90, and 120), and food grade POE sorbitan esters (Tween 20, 21, 40, 60, 61, 65, 80, 81, and 85). Within each class, the selected surfactants represent variations in carbon chain length and type, number of EO groups, and hydrophile-lipophile balance (HLB) numbers. Surfactants from these two classes are generally considered to be nontoxic. In addition, two surfactants known to be biologically inhibitory, nonionic octylphenol ethoxylate (Triton X-100) and the anionic sodium dodecyl sulfate (SDS), also were included in the preliminary surfactant screening tests for comparison purposes. Because actual field applications would involve surfactants in a commercially available form, these surfactants were used as received from the suppliers with no purification. All surfactants were obtained from the Aldrich Chemical Co. (Milwaukee, WI) with the exceptions of Tween 21, 61, 65, and 81 that were obtained from ICI Americas, Inc. (Wilmington, DE) and the Witconol SN series that was obtained from the Witco Corporation (Houston, TX).

Task II: Assessment of the Effect of Selected Surfactants on the Methanogenesis and Reductive Dechlorination Processes

Using reduced growth media and the Bayou d'Inde contaminated estuarine sediment as inoculum, an actively dechlorinating, methanogenic mixed culture consortium was enriched and maintained at 22?C by feeding HCB and glucose. Reductive dechlorination activity in the culture was sustained with glucose as the electron donor and monitored via liquid/liquid extraction of chlorinated benzenes with iso-octane and gas chromatography. HCB was sequentially transformed to dichlorobenzene via the major pathway of HCB ?> pentachlorobenzene (PeCB) ?> 1,2,3,5-tetrachlorobenzene (TeCB) ?> 1,3,5-trichlorobenzene (TrCB) ?> 1,3-dichlorobenzene (DCB). The culture was fed on a weekly basis with its contents replaced with fresh media every 14 days to result in a hydraulic retention time of 84 days. After a 1-year period of continuous operation, the mixed culture became sediment-free and was used for all of the biotransformation experiments. A two-tier approach was followed in conducting the biotransformation studies: first, the effect of surfactants on methanogenesis was evaluated; second, the potential impact of the surfactants on reductive dechlorination was assessed. Because the presence of soil or sediment solids would introduce complexities associated with desorption, mass transfer, and bioavailability, the experimental systems for this phase of the research contained only liquid culture, with microbial biomass as the only solid-phase.

Effect of Surfactants on Methanogenesis

The first phase in the biological screening of surfactants was designed to assess their effect on the methanogenic activity of the mixed culture. A serum bottle assay was conducted for all 16 surfactants, which were evaluated at an initial surfactant concentration of 200 mg/L. Two experimental systems were prepared for each surfactant: (1) surfactant plus glucose, to assess the effect of surfactants on methanogenesis; and (2) surfactant only, to assess the anaerobic biodegradability of the surfactants when present as the only carbon source. The bottles (70 mL total volume, 50 mL liquid culture) were prepared in triplicate for each series and incubated in the dark at 22?C (shaken daily). Over the 82-day incubation period, total gas and methane production from the surfactant-amended series were measured from the bottle headspace and compared to those from a reference series (containing glucose and no surfactant) and a seed blank series (no carbon source added). The main findings of this assay were: (1) the linear POE alcohols, Triton X-100, and SDS greatly inhibited the utilization of glucose for methanogenesis by the mixed-culture; (2) although none of the Tween surfactants negatively affected the extent of methanogenesis compared to the reference series, they resulted in lower initial rates of methanogenesis; and (3) methane production in all series containing only Tween surfactants were statistically higher than the seed blank control ( 0.05). In terms of the impact on the Tween surfactants on the rate of methanogenesis, the following series was observed (listed in order of least to greatest reduction in rate): T61, T65, T60, T40, T85, T20, T81, T80, and T21. The extent of surfactant anaerobic biodegradation?expressed as the fraction of initial surfactant chemical oxygen demand (COD) converted to methane?ranged from 0.29 ? 0.04 to 0.46 ? 0.01 (mean ? standard deviation).

Effect of Surfactants on Reductive Dechlorination

The second phase of the biotic surfactant screening involved an evaluation of the effect of the nine Tween surfactants, which did not inhibit methanogenesis, on the reductive dechlorination of HCB by the mixed culture. The experiments were conducted in 28 mL serum tubes sealed with Teflon-lined rubber septa and aluminum crimps. Each tube contained HCB dissolved in methanol, glucose, and Tween surfactants at concentrations of either 10, 50, 200, or 1,000 mg/L. At each sampling time, the contents of each tube were extracted with iso-octane followed by an analysis for HCB and its dechlorination products, and excess gas production was measured. As the initial surfactant concentration was increased, both the rate and extent of dechlorination decreased. At the initial surfactant concentration of 1,000 mg/L, reductive dechlorination was completely inhibited in all surfactant-amended series, except for those amended with Tween 61 and Tween 65. However, despite the inhibition to dechlorination at the initial surfactant concentration of 1,000 mg/L, partial biodegradation of the surfactants was observed in all series at this concentration after an acclimation period. As in the methanogenic screening assay, the least inhibitory surfactants toward HCB dechlorination were Tween 60, 61, and 65. Consistent with the findings from the methanogenesis assays, several non-Tween surfactants (Triton X-100, Brij 30 and Brij 35) were found to be inhibitory towards HCB reductive dechlorination at initial surfactant concentrations as low as 200 mg/L.

Task III: Assessment of Surfactant Anaerobic Biodegradability and Use of Surfactants as the Sole Electron Donor To Sustain HCB Reductive Dechlorination

Anaerobic Biodegradability of Surfactants

Because Tween 60, 61, and 65 exhibited the least toxicity toward both methanogenesis and reductive dechlorination, these three stearic-acid based surfactants were selected for further evaluation. A detailed assessment of the anaerobic biodegradability of the three surfactants was conducted in serum bottles (160 mL total volume, 120 mL liquid culture) using the mixed methanogenic culture at an initial surfactant concentration of 500 mg/L. The following parameters were measured: total gas and methane production, pH, oxidation-reduction potential (ORP), particulate organic carbon, volatile fatty acids (VFAs), and total COD. The three surfactants were only partially degradable under anaerobic conditions based on total COD destruction after 42 days (53, 62, and 62 percent for Tween 60, 61, and 65, respectively). The degradability of the surfactants generally corresponded to their fatty acid portion COD, suggesting that the observed surfactant degradation was mostly attributed to the degradation of the surfactant fatty acid moieties. The degradable portion of the surfactants was converted readily to methane, and VFAs did not accumulate during the incubation. However, when methanogenesis was inhibited by the addition of a specific inhibitor of methanogenesis (50 mM of 2-bromoethanesulfonic acid, BESA), significant levels of acetate accumulated. With the exception of Tween 61, the percent surfactant COD conversion to acetate observed in the BESA-amended series (37.8, 37.7 and 45.3 percent for Tween 60, 61, and 65, respectively) was in close agreement with the COD to methane conversion in the non-BESA-amended series (34.8, 56.7 and 48.3 for Tween 60, 61, and 65, respectively). These findings suggest that the biotransformation and observed partial biodegradation of the Tween surfactants occurred via the following steps: cleavage of the fatty acid moiety through hydrolysis, ?-oxidation of the fatty acid moiety to acetate, and conversion of acetate to methane.

Use of Surfactants as the Sole Electron Donor

Having established that Tween 60, 61, and 65 surfactants were at least partially biodegradable under anaerobic conditions, the ability of these three surfactants to sustain methanogenesis and dechlorination as the sole carbon and electron source, as well as the effect of long-term exposure of the enriched culture to these surfactants, were investigated. Three 1.6 L subcultures were developed in which, over a long period, the ultimate sole electron donor was either Tween 60, 61 or 65. A reference culture, which was never fed surfactant, was used for purposes of comparison. Over a 1-year acclimation period, the carbon source in the surfactant-fed cultures was gradually changed from glucose to surfactant only, while the surfactant concentration was increased from an initial concentration of 100 mg/L to 400 mg/L. The transition from glucose to Tween surfactants as the sole, external electron donor did not adversely affect methanogenesis and HCB dechlorination. Despite long-term exposure to the surfactants, the extent and pathway of HCB dechlorination were unaltered as 1,3-DCB continued to be the main final product. These findings indicate that reductive dechlorination of HCB supported by the biodegradation of Tween surfactants is feasible. Although the surfactants were only partially degraded, the degradation products were not inhibitory. This conclusion was based on observed consistent HCB dechlorination and methane production by the three surfactant-fed cultures over 1 year of continuous feedings with these surfactants. Potentially, Tween 60, 61, and 65 could be used to simultaneously increase the bioavailability of sorbed contaminants while serving as the carbon and electron source for microbial reductive dechlorination.

Task IV: Batch Investigations of Surfactant-Enhanced Desorption and Dechlorination of Hexachlorobenzene

The objectives of this task were to: (1) measure the solubility of HCB in micellar surfactant solutions; (2) assess the sorption and desorption of HCB alone, surfactant alone, and HCB + surfactant; and (3) to quantify the coupled desorption and transformation of HCB in the presence of surfactants. Due to the tendency for Tween 60, 61, and 65 to separate into a surfactant-rich phase, especially under high ionic strength conditions, many of the assays in this and subsequent tasks focused on Tween 80, which is structurally similar to Tween 60 (containing oleic acid rather than stearic acid) but exhibits greater aqueous solubility and a much lower tendency to form surfactant rich-phases.

HCB Solubilization

The micellar solubilization of HCB in aqueous solutions of Tween 60 and Tween 80 was measured in completely mixed batch reactors over a surfactant concentration range of 200 to 2,500 mg/L. Each 35-mL glass reactor contained excess HCB, which was deposited as a thin film on the bottom of the reactor. Analysis of HCB in aqueous surfactant solutions was achieved using a direct injection gas chromatography technique. This analytical method yielded more accurate and reproducible results than conventional solvent extraction methods, which resulted in the formation of persistent macroemulsions between the organic solvent and aqueous phases. A least-squares linear regression procedure (SYSTAT ver. 5.0) was used to obtain the weight solubilization ratio values (WSR) of 5.96 X 10-4 g HCB/g Tween 60 (r2 = 0.996) and 7.0 X 10-4 g HCB/g Tween 80 (r2 = 0.991) after 11 days of mixing. Based on the average molecular weight of Tween 60 and Tween 80, WSR values were converted to molar solubilization ratio values (MSR) of 2.74 X 10-3 mole HCB/mole Tween 60 and 3.2 X 10-3 mole HCB/mole Tween 80. The solubilization of HCB in micellar solutions of Tween 80 also was found to be rate-limited based on measurements made after 2, 7, 11, and 30 days of mixing. Approximately 70 percent of the solubilization capacity was achieved after 2 days of mixing, and equilibrium solubility values were observed after mixing for approximately 10 days. These data indicate that even under ideal mixing conditions, micellar solubilization of HCB by Tween 80 is strongly rate-limited.

Surfactant Sorption

Sorption data were obtained for two of the Tween surfactants (Tween 60 and Tween 80) and a reference nonylphenol ethoxylate (Tergitol NP-10). To provide a range in organic carbon (OC) content, three soils were employed for the sorption studies: F-70 Ottawa sand (0 percent OC), Appling soil (0.75 percent OC), and Webster soil (3.35 percent OC). Sorption measurements were conducted in batch reactors consisting of either 25 mL glass centrifuge tubes or 30 mL polypropylene centrifuge tubes. The initial concentration of surfactant solutions added to each reactor typically ranged from 200 to 2,000 mg/L, but in cases where the sorptive capacity of the soil was large, the upper concentration limit was increased to 20,000 mg/L. The contents of the reactors were mixed for periods ranging from 1 day to 4 weeks to measure the adsorption rates and to establish equilibrium sorption capacities. Following mixing, the solid phase was separated by centrifugation, and the resulting supernatant was analyzed for surfactant using a high pressure liquid chromatography (HPLC) unit equipped with a diode array detector (DAD) and an evaporative light scattering detector (ELSD).

HCB Sorption

The sorption of HCB by Appling soil was initially measured in the absence of surfactant. Batch experiments were conducted in 25 mL Corex glass centrifuge tubes sealed with Teflon/aluminum-lined caps. The aqueous phase contained 0.005 M CaCl2 as a background electrolyte and 500 mg/L NaN3 to minimize biological activity. The sorption data yielded linear sorption isotherms, with the observed distribution coefficient (KD) increasing from 385 L/kg to 527 L/kg over the range of mixing times evaluated. These values correspond to organic carbon distribution coefficients (KOC), expressed on a Log scale, of 2.71 to 2.84. The HCB distribution coefficients obtained for 1-day and 3-day experiments were similar, but substantially lower than KD values obtained after 15 and 22 days of mixing. These data indicate that HCB sorption by Appling soil exhibited an initial period of rapid uptake, followed by a prolonged period of relatively slow adsorption.

The effect of surfactant on HCB sorption by Appling soil was evaluated as a function of Tween 80 concentration and mixing time. For each initial surfactant concentration, the initial concentration of HCB was varied to allow for the determination of individual HCB sorption isotherms, rather than single point measurements. In these experiments, the final aqueous phase concentrations of both HCB and Tween 80 were measured independently by GC and HPLC analysis, respectively. For all systems investigated, the presence of Tween 80 at aqueous phase concentrations above the CMC reduced the sorption of HCB by Appling soil. Specifically, as the surfactant concentration was increased, the observed HCB distribution coefficient (KD) decreased. For a given surfactant loading, the HCB distribution coefficient increased with time, indicating that HCB sorption is rate-limited in the presence of surfactant. Because the aqueous phase concentration of Tween 80 decreased with mixing time due to sorption by the solid phase, these data illustrate the importance of considering the effects of multiple rate-limitations on the distribution of hydrophobic organic compounds between the free aqueous, micellar, and sorbed phases.

Effect of Surfactants on the Bioavailability of HCB in Historically Contaminated Sediments

The influence of Tween 60, 61, and 65 on the bioavailability of sorbed-phase HCB in contaminated sediments was evaluated in a series of batch experiments. The Bayou d'Inde sediment, which had been contaminated with HCB and other chlorinated benzene congeners for more than 40 years, was used as the solid phase. To improve the homogeneity of the sediment, the sample was air-dried and ground to pass a 500- m mesh screen prior to use. For homogenized sediment samples containing either Tween 60, 61, 65 or glucose, HCB was readily desorbed and dechlorinated to 1,3-DCB. It was concluded that the homogenization process drastically increased the intrinsic desorption rate of HCB to the point that HCB availability no longer limited biological processes. Consequently, the homogenized sediment was considered unsuitable for evaluating the potential of surfactants to enhance the bioavailability of sorbed-phase contaminants.

Task V: Column Studies of Surfactant-Enhanced Desorption, Transformation and Transport of Hexachlorobenzene

The objective of this task was to measure the coupled sorption, transport, and transformation of HCB and surfactant in one-dimensional soil columns. A matrix of one-dimensional soil column experiments was conducted to assess the transport of three surfactants, Tween 60, Tween 80, and and Triton X-100, in Appling soil and Ottawa sand. The length of the packed soil columns was approximately 10 cm, with a bulk density of 1.85 g/cm3, a porosity of 0.33, and an aqueous pore volume of 18 to 20 mL. Following complete water saturation, a nonreactive tracer study was conducted using a pulse injection of a solution containing 1,000 mg/L potassium iodide (KI) and 500 mg/L NaN3. Effluent KI concentrations were determined by HPLC/UV analysis. The resulting breakthrough curves (BTCs) were fit to a dimensionless form of the advective dispersive reactive (ADR) transport equation using the CFITIM3 program. Overall mass balance calculations demonstrated that only 41 percent of the injected surfactant mass had been recovered after flushing with 7 pore volumes of background solution. Subsequent attempts to recover the remaining sorbed-phase surfactant with either ethanol solutions and pH adjustments were not successful. The mass remaining in the soil column corresponded to a sorbed-phase concentration of approximately 1 mg Tween 80/g Appling soil. This value is well below that corresponding to monolayer coverage of Appling soil by Tween 80 (~2.5 mg/g). These data indicate that: (1) desorption of Tween 80 was strongly rate-limited; and (2) the complete recovery of sorbed-phase surfactant from natural soils may be extremely difficult. This finding is important from a remediation perspective because it demonstrates that significant amounts of surfactant may remain in the subsurface following surfactant treatment. In the case of a nontoxic, biodegradable surfactant this behavior may actually be beneficial because surfactant will remain in the system for longer periods of time than previously thought. If the surfactant is toxic or recalcitrant, however, there is a risk that surfactant could persist for extended periods of time in the environment. It should be noted that under these conditions, the actual concentration of surfactant in the aqueous phase may be quite low, and therefore difficult to detect.

Task VI. Mathematical Modeling of HCB and Surfactant Phase Distribution and Transport in Porous Media.

The objective of this task was to develop and evaluate mathematical models that could be used to simulate the coupled transport, sorption, and transformation of surfactant and HCB in porous media. During the initial phase of this study, a mathematical model was developed to predict the equilibrium phase distribution of HCB in the presence of sorbed-phase surfactant. For this modeling effort, the overall or apparent solubility of a HOC in the presence of surfactant was represented as the amount of HOC associated with surfactant monomers plus the amount associated with surfactant micelles. To simulate the coupled sorption and transport of both surfactant and HCB in a porous medium, a mathematical model was developed that incorporates rate-limited, nonlinear (Langmuir) surfactant sorption, rate-limited micellar solubilization of HCB, and rate-limited HCB sorption, and transformation.

Simulations of Tween 60 and Tween 80 transport for equilibrium sorption conditions yielded breakthrough curves (BTCs) that were characterized by a steep leading edge and prolonged tailing. Such behavior is characteristic of solutes exhibiting Langmuir sorption, and can be explained by the strong affinity (steep slope of the sorption isotherm) at low concentrations. Sorption of HCB by the solid phase was represented by a Freundlich isotherm, SHCB = KFCN. However, HCB sorption isotherms were linear, and therefore the value of the exponent, N, was set to 1.0. Rate-limited sorption of both surfactant and HCB was accounted for with a linear driving force expression, which is equivalent to a mobile-immobile water or two-region representation of nonequilibrium sorption processes.


Journal Articles on this Report : 2 Displayed | Download in RIS Format

Other project views: All 13 publications 3 publications in selected types All 2 journal articles
Type Citation Project Document Sources
Journal Article Yeh DH, Pennell KD, Pavlostathis SG. Toxicity and biodegradability screening of nonionic surfactants using sediment-derived methanogenic consortia. Water Science and Technology 1998;38(7):55-62. R825404 (1999)
R825404 (Final)
not available
Journal Article Yeh D, Pavlostathis SG, Pennell KD. Effect of tween surfactants on methanogenesis and microbial reductive dechlorination of hexachlorobenzene. Environmental Toxicology and Chemistry 1999;18(7):1408-1416. R825404 (1997)
R825404 (1999)
R825404 (Final)
not available

Supplemental Keywords:

bioavailability, chlorinated compounds, sorption, surfactants, transport, bioremediation, modeling., Scientific Discipline, Toxics, Waste, Ecosystem Protection/Environmental Exposure & Risk, Bioavailability, Environmental Chemistry, Chemistry, HAPS, Fate & Transport, Bioremediation, 33/50, fate and transport, microbiology, bioremediation model, surfactant-aided desportion, sorption kinetics, benzene, kinetic studies, chemical transport, hazardous waste, biotechnology, geochemistry, hazardous waste cleanup, biotransformation, chemical releases, contaminant release, environmental toxicant, mobility, waste chemicals, Benzene (including benzene from gasoline), chlorinated benzenes, nonionic surfactants, microbial reductive dechlorination, chlorinated solvents, transport

Progress and Final Reports:

Original Abstract
  • 1997 Progress Report
  • 1998 Progress Report