- Email: [email protected]
Use of tracer tests to evaluate the impact of enhanced-solubilization flushing on in-situ biodegradation
Journal of Contaminant Hydrology 64 (2003) 191 – 202 www.elsevier.com/locate/jconhyd
Use of tracer tests to evaluate the impact of enhanced-solubilization flushing on in-situ biodegradation S.R. Alter a, M.L. Brusseau a,b,*, J.J. Piatt c, A. Ray-Maitra b, J.-M. Wang d, R.B. Cain a a Hydrology and Water Resources Department, University of Arizona, 429 Shantz, Tucson, AZ 85721, USA Soil, Water and Environmental Science Department, University of Arizona, 429 Shantz, Tucson, AZ 85721, USA c Carroll College, 100 North East Avenue, Waukesha, WI 53186, USA d Department of Environment Engineering, Hung Kuang Institute of Technology, Taichung, 433 Taiwan, ROC
b
Received 13 March 2002; received in revised form 26 August 2002; accepted 27 September 2002
Abstract Tracer tests were conducted to evaluate the effect of a complexing sugar flush (CSF) on in-situ biodegradation potential at a site contaminated by jet fuel, solvents, and other organic compounds. Technical-grade hydroxypropyl-h-cyclodextrin was used during the CSF study, which was conducted in a hydraulically isolated cell emplaced in a surficial aquifer. In-situ biodegradation potential was assessed with the use of tracer tests, which were conducted prior to and immediately following the CSF study. Ethanol, hexanol, and benzoate were used as the biodegradable tracers, while bromide was used as a nonreactive tracer. The results indicate that the biodegradation of benzoate was similar for both tracer tests. Conversely, the biodegradation of ethanol (23% increase) and hexanol (41% increase) was greater for the post-CSF tracer test. In addition, analysis of core samples collected from within the test cell indicates that the population density of aerobic jet-fuel degraders increased in the vicinity of the injection wells during the CSF. These results indicate that the cyclodextrin flush did not deleteriously affect the indigenous microbial community. This study illustrates that tracer tests can be used to evaluate the impact of remediation activities on in-situ biodegradation potential. D 2003 Elsevier Science B.V. All rights reserved. Keywords: Ground water quality; Bioremediation; Contaminant transport
* Corresponding author. Soil, Water and Environmental Science, University of Arizona, 429 Shantz, Tucson, AZ 85721, USA. 0169-7722/03/$ - see front matter D 2003 Elsevier Science B.V. All rights reserved. doi:10.1016/S0169-7722(02)00203-6
192
S.R. Alter et al. / Journal of Contaminant Hydrology 64 (2003) 191–202
1. Introduction Remediation technologies based on microbial processes have become popular and widespread. Bioremediation offers several advantages compared to standard pump-andtreat technologies. However, its effectiveness can be limited by numerous factors. One such factor is the presence of large quantities of contaminant mass (e.g., source zones), which often constrains bioremediation applications for highly contaminated sites. Successful cleanup of such sites may require combining bioremediation with a technology that can remove large quantities of contaminant mass. One source-zone remediation technology of current interest is the use of enhanced-solubilization flushing agents in conjunction with pump and treat. Types of flushing agents include surfactants, complexing agents, natural organic matter, and cosolvents. The flushing agents increase the apparent solubility of hydrophobic organic compounds, thereby increasing their rate of removal from the subsurface. It is possible that the bioavailability of the contaminants may be increased due to this enhanced solubilization, thereby altering the magnitudes and rates of biodegradation. Such changes in biodegradation properties could impact the dynamics of bioremediation. In addition, it is to be expected that some of the flushing agent will remain in the treated zone after the flushing event. The potential impact of residual flushing agent on microbial processes is a question of concern. For these reasons, it is important to evaluate the potential impact of enhanced-flushing operations on microbial processes for systems wherein they will be used in conjunction with bioremediation. Cyclodextrin was used as a flushing agent for a pilot-scale field study conducted at Hill Air Force Base (AFB) in Layton, UT. Cyclodextrin has a toroidal molecular structure composed of a hydrophobic interior and a hydrophilic exterior. This structure provides a cavity into which a hydrophobic organic compound can partition, which is the basis for its use as an enhanced-solubilization agent (e.g., Wang and Brusseau, 1993). The work presented herein investigates the effect of the cyclodextrin flushing on in-situ biodegradation potential, which was characterized through the use of tracer tests.
2. Materials and methods 2.1. Background Various methods have been proposed to characterize biodegradation potential for a given site of interest. These methods include conducting laboratory experiments using core samples collected from the field, using contaminant-concentration and geochemical field data to evaluate potential occurrence and rates of attenuation processes, and conducting insitu microcosm studies (e.g., Gilham and Miller, 1990; McAllister and Chiang, 1994; Borden et al., 1995; Nielsen et al., 1996; EPA, 1998). Laboratory-based approaches for assessing biodegradation potential are attractive for several reasons. First, mass-balance calculations are possible for laboratory experiments. In contrast, the amount and distribution of contaminants is unknown for most field sites. Second, in laboratory studies, abiotic processes can generally be separated from biotic processes, another issue that is often difficult to overcome for field sites. Third, as opposed to field conditions, environ-
S.R. Alter et al. / Journal of Contaminant Hydrology 64 (2003) 191–202
193
mental factors such as pH, temperature, redox potential, and nutrient levels can be controlled and their impacts investigated in laboratory experiments. Despite these advantages, the use of laboratory studies is constrained by the fact that they may not adequately represent field conditions (e.g., Chapelle et al., 1996). Recently, so-called ‘‘biotracer’’ tests have been proposed as a method to characterize biodegradation potential (Istok et al., 1997; Brusseau et al., 1999). This method involves conducting a tracer test using a nonreactive tracer along with biodegradable compounds. Breakthrough curves for the biotracers and nonreactive tracer are constructed using data collected for various sampling locations and analyzed to obtain information regarding transport and mass recovery, from which the biodegradation potential can be assessed. The breakthrough curve measured at a specific monitoring location reflects the impact of processes occurring within the swept zone associated with the injection and monitoring points. Thus, the biodegradation parameters obtained from a biotracer test represent an integrated measurement of the monitored zone influenced by the tracer test. The ability to obtain a representative profile of biodegradation activity across a large spatial scale is a distinct advantage of the biotracer method. Another major advantage of the biotracer test is that a known mass of substrate is injected into the subsurface under controlled conditions. This provides the opportunity for conducting mass balances, and thus for quantifying the magnitude and rate of biodegradation of the tracers. 2.2. Site description The tracer tests were conducted in Operable Unit 1 (OU1) at Hill AFB in Layton, UT. The subsurface at this site is contaminated by a multiple-component, immiscible organic liquid. The contamination originates from multiple sources, including chemical disposal pits, fire training areas, landfills, and waste oil pits. The immiscible liquid is comprised primarily of petroleum hydrocarbons (i.e., jet fuel), chlorinated solvents, and polynuclear aromatic hydrocarbons. Past water-table fluctuations have smeared the immiscible liquid such that it is distributed from above the current water table throughout the saturated zone extending to the confining unit. The initial cell-averaged immiscible-liquid saturation was approximately 12%, based on the results of a pre-complexing sugar flush (CSF) partitioning tracer test and organic-carbon analyses of selected cores (Cain et al., 2000). The tracer tests were conducted as part of a series of experiments designed to test innovative remediation technologies (Bedient et al., 1999). The tests were conducted in a hydraulically isolated section of the shallow aquifer that consists of fine-to-coarse sand interbedded with gravel and clay stringers. Depth to groundwater ranged from approximately 5 to 5.5 m. A 60-m-thick clay layer, located at approximately 8 m below ground surface, forms a bottom boundary for the aquifer. The porosity of the test area is approximately 20% (Cain et al., 2000). As discussed in detail by McCray and Brusseau (1998), the CSF was found to be effective in removing organic contaminants at the Hill AFB test site. The cell-wide mass removal was calculated to be approximately 42% based on the results of partitioning tracer tests and analysis of core samples collected before and after cell remediation. The final cell-averaged immiscible-liquid saturation was approximately 7%, based on the results of a post-CSF partitioning tracer test (Cain et al., 2000).
194
S.R. Alter et al. / Journal of Contaminant Hydrology 64 (2003) 191–202
2.3. Tracers Ethanol, hexanol, and benzoate were used as the biodegradable tracers. It is well documented that these compounds can be degraded under both aerobic and anaerobic conditions. These compounds have low volatilities and should experience minimal hydrolysis under the system conditions. The results of preliminary laboratory experiments have shown that sorption of these tracers by the site aquifer material is minimal. Therefore, mass loss as a result of abiotic processes is unlikely. Bromide was used as a nonreactive tracer. As noted above, an immiscible-liquid phase of organic contaminants is present in the test cell. Bromide, benzoate, and ethanol do not partition measurably to this phase. Conversely, hexanol does partition measurably into the immiscible-liquid phase. The immiscible-liquid/water partition coefficient measured for hexanol in batch studies was 2.5 (Cain et al., 2000). The transport and biodegradation of hexanol may have been influenced by mass transfer between the aqueous and immiscible-liquid phases. For example, the bioavailability of hexanol may have been constrained compared to the nonpartitioning tracers. However, preliminary analysis of the transport of tracers that partitioned to the immiscible liquid but experienced minimal biodegradation indicates that this inter-phase mass transfer process can be considered to be essentially instantaneous with respect to hydraulic residence times prevalent in the system (unpublished data). Thus, while quantitative analysis of biodegradation rates for hexanol would require consideration of its partitioning behavior, it is possible to compare overall magnitudes of hexanol biodegradation to those obtained for ethanol and benzoate. Preliminary experiments were conducted in the laboratory using core material collected from the site to evaluate the biodegradation potential of the indigenous microbial community for the biotracers. Batch studies using 14C-labelled compounds confirmed that the biotracers could be biodegraded by the indigenous microbial community under both aerobic and anaerobic conditions. In addition, a batch substrate-loss study was conducted using all three biotracers to examine the relative affinity of the indigenous community for the biotracers. The results showed that hexanol was preferred relative to ethanol, which was preferred relative to benzoate. The more rapid biodegradation of the alcohols compared to benzoate is not surprising, given their respective molecular structures (e.g., Maier, 1999). Miscible-displacement studies were conducted using site core material to evaluate the transport and biodegradation behavior of the tracers, and to compare the behavior of the biotracers to that of representative site contaminants. The results indicated that the general transport and biodegradation behavior of the biotracers was similar to that of representative site contaminants. For example, the rate and magnitude of biodegradation of benzoate during transport was very similar to that of toluene. The results of these laboratory experiments are discussed in more detail in Brusseau et al. (1999) and Wang (1999). 2.4. Field methods The tracer tests were conducted in a cell in which several other studies have been conducted, including the CSF study (McCray and Brusseau, 1998) and partitioning tracer
S.R. Alter et al. / Journal of Contaminant Hydrology 64 (2003) 191–202
195
tests to evaluate the performance of the CSF (Cain et al., 2000). The tracer tests were conducted in conjunction with the partitioning tracer tests, and were conducted prior to and immediately after the CSF. The cell was created by driving steel sheet piling into the clay unit located at an approximate depth of eight meters below ground surface. Interlocking joints were grouted to prevent leakage, which was shown to be minimal in tests conducted prior to initiation of the experiments. A line of four fully screened injection wells and a line of three fully screened extraction wells, both normal to the direction of flow and located approximately 4 m apart, were installed at opposite ends of the cell (Fig. 1). The injection and extraction wells were used to create a steady state flow field prior to the injection of the tracers in each study. Peristaltic pumps (Cole Parmer Master Flex I/P with Tygon LFL tubing) with separate heads for each well were used to generate flow. Equal injection and extraction flow rates of approximately 4.5 l/min were maintained, with the flow divided evenly among the three extraction wells and four injection wells. During each experiment, tracer-free water was first flushed through the cell to create steady-state conditions. The tracer pulses were then injected, followed by the injection of tracer-free water again. Constant flow rates were maintained during the flow establishment, tracer injection, and water flushing portions of each experiment. Tracer injection concentrations were 300, 1000, 925, and 325 mg/l for bromide, ethanol, hexanol, and benzoate, respectively. Approximately 1300 l of tracer solution was injected for each experiment. The tracer tests were approximately 12 days in duration. A three-dimensional, multi-level sampling array (12 locations, 5 depth intervals) connected to a vacuum system was used to collect depth-specific water samples. Samples
Fig. 1. Diagram illustrating cell configuration.
196
S.R. Alter et al. / Journal of Contaminant Hydrology 64 (2003) 191–202
were drawn by vacuum through a 3.2-mm stainless steel tube connected to a 2.5-cm long, 1.25-cm diameter, 40-Am HPLC pump filter buried in situ. During this study, samples were collected from depths of approximately 6.25 and 6.9 m bgs. Depth-integrated samples were collected from the three extraction wells. Extraction wells were screened from approximately 5 to 8.4 m bgs. A total of 18 sampling locations were used for this study. Samples were collected with minimal headspace in polyethylene or glass vials containing a 1-ml aliquot of 1 N NaOH, which served as a preservative to prevent biodegradation. All samples were stored at 4 jC until analysis. Bromide concentrations were determined using an ion-specific electrode. The analytical quantification limit for bromide was approximately 0.4 mg/l. Ethanol and hexanol were analyzed using a gas chromatograph (Shimadzu GC-14A) equipped with a flame ionization detector. Direct liquid injection of 1-Al aliquots was performed on an SPB-624 capillary column (Supelco), 0.53 mm, 30-m long. The analytical quantification limit for the alcohols was approximately 0.5 mg/l. Benzoate concentrations were determined using a high-pressure liquid chromatograph (Waters) with a UV –VIS spectrophotometer. The analytical quantification limit for benzoate was approximately 0.5 mg/l. Due to analytical difficulties (overlapping peaks), benzoate breakthrough curves were obtained for only 8 of the 18 total sampling locations. Effluent samples were periodically collected for analysis of dissolved oxygen. 2.5. Microbial evaluation A microbial characterization was conducted to evaluate the impact of the CSF on the indigenous microbial community. Core samples were collected from the cell after the CSF and tracer tests were completed. The samples were used to assess the indigenous microbial community with respect to both general aerobic microbial populations and to aerobic jetfuel degrader populations. Core samples (approximately 1.2 m) were collected from 4 to 8 m bgs. Samples were collected aseptically from three locations corresponding to the injection end of the cell, the middle of the cell, and the extraction end of the cell. The soil samples were stored in a cooler until plating occurred. Plating was completed within 24 hours of sample collection. Two grams of soil for each sample was placed in 9 ml of autoclaved, deionized water (Basrnstead, Nanopure water, Dubuque, IA) and vortexed for 1 min. The samples were then serially diluted to a range that would produce statistically robust microbial counts. The dilutions were plated on plates containing R2A (Becton and Dickinson, Cockeysville, MD) or mineral salts medium (MSM) combined with 0.05% JP8 jet fuel (added after autoclaving). MSM contains 0.4% Na2HPO4, 0.1% KH2PO4, 0.1% NH4Cl, 0.2% MgSO4, 0.005% yeast extract, 0.0005% ammonium iron (III) citrate, 0.001% CaCl2, and 15 g of agar. The R2A media provides a measure of heterotrophic bacteria while the MSM/JP8 media provides a measure of bacteria capable of degrading jet-fuel components. Samples were plated in triplicate, along with appropriate controls. The plates were incubated at room temperature for 1 week and were then counted. Prior to the second tracer test, a 10% cyclodextrin solution was pumped through the aquifer for 10 days (a total of 65,500 l), followed by flushing with water free of cyclodextrin (about 40,000 l). This pumping activity was associated with the CSF pilot test noted above. During this test, approximately 100% of the injected cyclodextrin was recovered, indicating
S.R. Alter et al. / Journal of Contaminant Hydrology 64 (2003) 191–202
197
that degradation of this compound was negligible. This pumping introduced aerated water into the cell for a significant period of time (greater then 15 days). The oxygen concentration in the injected water was approximately 8 mg/l, while it was 1– 2 mg/l in the effluent pumped from the extraction wells. In addition, the cyclodextrin solution was shown to have enhanced the solubilization and removal of organic constituents (McCray and Brusseau, 1998). This enhanced-solubilization process may have increased the bioavailability of the organic contaminants, as was demonstrated in prior laboratory experiments (Wang et al., 1998). Thus, it is likely that the injection of the aerated cyclodextrin solution stimulated microbial activity within the cell, which in turn resulted in an increase in biomass near the injection wells that will be discussed below.
3. Results and discussion Representative breakthrough curves obtained from the tracer tests are presented in Fig. 2. Breakthrough curves obtained before and after the CSF study exhibit the same general trends. The arrival fronts for benzoate and ethanol are generally coincident with the arrival front for bromide. Conversely, hexanol transport, due to partitioning to the immiscibleliquid phase, is retarded with respect to the other tracers. Bromide has a higher peak than that of the biodegradable tracers, and exhibits longer elution tails. This behavior is attributed to biodegradation and resultant mass loss of the biotracers. Mass recoveries were calculated for the tracers using moment analysis. Mass loss of the biodegradable tracers was observed for all sampling locations. Given that oxygen concentrations were limiting in the extraction-well effluent, these results suggest that both aerobic and anaerobic degradation processes contributed to biotracer mass loss. This is consistent with the results of the preliminary batch experiments noted above. The mass loss of the biotracers varied with sampling location. The greatest losses were observed for sampling locations near the injection wells and for those with the largest associated hydraulic residence times. Detailed analysis of the spatial variability of biotracer biodegradation for these tracer tests is presented by Sandrin (2001). The mean mass recoveries for bromide, the nonreactive tracer, were close to 100% for both the pre- and post-tracer tests. This indicates there was minimal loss of tracer mass due to hydraulic factors. Conversely, the mean mass recoveries for the biodegradable tracers were significantly less than 100% (see Table 1). The average recovery of benzoate for the tracer test conducted before the CSF study, as calculated from eight MLS locations, was approximately 82%. The average recoveries of ethanol and hexanol for the pre-tracer test, calculated using only the data for the same eight locations for which benzoate data were measured, were approximately 78% and 82%, respectively. The average recoveries for ethanol and hexanol, calculated using data from all sampling locations, were approximately 76% and 83%, respectively. There is minimal difference in the average ethanol and hexanol recoveries calculated using the data collected for the eight locations for which benzoate data are available and those calculated using data collected from all 18 locations. This suggests that the fewer number of measurements available for benzoate does not significantly impact the calculated average mass recoveries. Interestingly, the recoveries for all three biotracers are similar.
198
S.R. Alter et al. / Journal of Contaminant Hydrology 64 (2003) 191–202
Fig. 2. Breakthrough curves representative of those obtained from the tracer tests. These data were collected at the 6.9 m bgs sampling plane, from locations 11 (top), 13 (middle), and 33 (bottom), during the post-CSF tracer test.
S.R. Alter et al. / Journal of Contaminant Hydrology 64 (2003) 191–202
199
Fig. 2 (continued).
The average recoveries of benzoate, ethanol, and hexanol for the tracer test conducted after the CSF were approximately 79%, 64%, and 54%, respectively, for the eight MLS locations for which benzoate data are available. The average percent difference in mass recoveries for the pre- and post-tracer tests was approximately 4% for benzoate, which is considered within experimental uncertainty, indicating the CSF had minimal impact on benzoate biodegradation potential. Conversely, the differences between pre- and post-CSF average mass recoveries for ethanol and hexanol were approximately 23% and 41%, respectively. Thus, the CSF appeared to enhance the biodegradation potential for ethanol and hexanol. Table 1 Comparison of pre- and post-CSF recoveries for biotracers Tracer
Pre-CSF study
Post-CSF study
Benzoate 8 locations
Percent difference
82
79
4
Ethanol 8 locations All locations
78 76
64 59
18 23
Hexanol 8 locations All locations
82 83
54 49
34 41
Note: the values reported in the table are considered to have an associated measurement uncertainty of V 10% based on analytical methods.
200
S.R. Alter et al. / Journal of Contaminant Hydrology 64 (2003) 191–202
Hexanol exhibited the greatest difference in recovery between pre- and post-CSF experiments, while ethanol and benzoate exhibited smaller differences. The pre-CSF mass recovery of hexanol was similar to those of ethanol and benzoate, while the post-CSF recovery for hexanol is significantly lower than the other two. This behavior could be related to the fact that hexanol is the only one of the three biotracers that partitions to the immiscible-liquid phase. Partitioning to the immiscible liquid may have reduced the bioavailability of hexanol, thus reducing its magnitude of biodegradation. Given that there was less immiscible-liquid present during the post-CSF biotracer test (due to removal during the CSF), the retention of hexanol was less for the post tracer test compared to the tracer test conducted prior to the CSF. Thus, it is possible that the bioavailability of hexanol increased after the CSF, allowing for greater biodegradation. This phenomenon would not have affected the other tracers, as they did not partition measurably to the immiscible liquid. The magnitude of the change in hexanol biodegradation with respect to the other tracers may also in part be due to the fact that the indigenous community appears to have a greater affinity for hexanol than for benzoate and ethanol, as shown by the results of the batch laboratory experiments noted above. It is possible the perturbations associated with the CSF, such as the solubilization of immiscible-liquid constituents and the introduction of oxygen, may have had the greatest impact on the more readily degradable compound. 3.1. Microbial evaluation The biomass concentrations obtained for core samples collected from the cell are similar for both heterotrophic and jet-fuel degraders. However, the biomass concentrations are significantly higher for the core collected near the injection wells, approximately 107 CFU/g soil, versus approximately 105 for the cores collected from the center of the cell and near the extraction wells (see Fig. 3). The biomass concentrations were most likely
Fig. 3. Distribution of biomass concentrations within the cell; aerobic heterotrophic degraders and aerobic jet-fuel degraders.
S.R. Alter et al. / Journal of Contaminant Hydrology 64 (2003) 191–202
201
relatively uniform prior to the CSF, as indicated by the concentrations measured for the shallow depth intervals above the water table, approximately 4 to 5.2 m bgs, which are uniformly low (see Fig. 3). This section of the subsurface was not affected by the CSF and, therefore, the data collected for this zone can be used to represent general baseline conditions within the cell. The results presented in Fig. 3 suggest that biomass concentrations increased during the CSF. This growth, which occurred primarily in the region near the injection wells, is most likely due to the impact of the CSF on microbial activity, as noted above.
4. Conclusions Results of the tracer tests indicate that a complexing sugar flush did not impede microbial activity and associated in-situ biodegradation potential. The results of the microbial evaluation demonstrate that viable, active aerobic microbial populations were present following the CSF. This is consistent with the results obtained from the tracer test conducted after the CSF, wherein mass loss of the biodegradable tracers was observed. Overall, the magnitude of biodegradation of the biotracers was greater after the CSF study. This, in combination with the evidence of active microbial populations present after the study, suggests that the CSF remediation technology may have enhanced biodegradation potential, at least for some of the lower molecular weight hydrocarbons. Finally, the usefulness of the ‘‘biotracer’’ method for characterizing field-scale, in-situ biodegradation potential and evaluation of the effects of remedial actions on biodegradation has been illustrated in this study.
Acknowledgements We thank William Blanford, John McCray, Adria Bodour, Qinhong Hu, Gwynn Johnson, Gale Famisan, Fiona Jordan, and other former UA students for their help. We also would like to thank Lynn Wood and Carl Enfield of the US Environmental Protection Agency for their support. This research was supported by projects funded by the US Environmental Protection Agency, one as part of the Joint Bioremediation Program and one in conjunction with the DOD SERDP program.
References Bedient, P., Holder, A., Enfield, C.G., Wood, L., 1999. Enhanced remediation demonstration at Hill Air Force Base: introduction. In: Brusseau, M., Sabatini, D., Gierke, J., Annable, M. (Eds.), Innovative Subsurface Remediation Field Testing of Physical, Chemical, and Characterization Technologies. American Chemical Society, Washington, DC, pp. 36 – 48. Borden, R.C., Gomez, C.A., Becker, M.T., 1995. Geochemical indicators of intrinsic bioremediation. Ground Water 33, 180 – 189. Brusseau, M., Piatt, J., Wang, J., Hu, M., 1999. A biotracer test for characterizing the in-situ biodegradation potential associated with subsurface systems. In: Brusseau, M., Sabatini, D., Gierke, J., Annable, M. (Eds.),
202
S.R. Alter et al. / Journal of Contaminant Hydrology 64 (2003) 191–202
Innovative Subsurface Remediation, Field Testing of Physical, Chemical, and Characterization Technologies. American Chemical Society, Washington, DC, pp. 240 – 250. Cain, R.B., Johnson, G.R., McCray, J.E., Blanford, W., Brusseau, M.L., 2000. Partitioning tracer tests for evaluating remediation performance. Ground Water 38, 752 – 761. Chapelle, F., Bradley, P., Lovley, D., Vroblesky, D., 1996. Measuring rates of biodegradation in a contaminated aquifer using field and laboratory methods. Ground Water 34, 691 – 698. Gilham, R.W., Miller, D.J., 1990. A device for in situ determination of geochemical transport parameters: 2. Biochemical reactions. Ground Water 28, 858 – 862. Istok, J.D., Humphrey, M.D., Schroth, M.H., Hyman, M.R., O’Reilly, K.T., 1997. Single-well, ‘‘push-pull’’ test for in-situ determination of microbial activies. Ground Water 35, 619 – 631. Maier, R., 1999. Microorganisms and organic pollutants. In: Maier, R., Pepper, I.L., Gerba, C.P. (Eds.), Environmental Microbiology. Academic Press, San Diego, pp. 363 – 402. McAllister, P.M., Chiang, C.Y., 1994. A practical approach to evaluating natural attenuation of contaminants in ground water. GWMR, Spring, 161 – 173. McCray, J.E., Brusseau, M.L., 1998. Cyclodextrin-enhanced in situ flushing of multiple-component immiscible organic liquid contamination at the field scale: mass removal effectiveness. Environ. Sci. Technol. 32, 1285 – 1293. Nielsen, P.H., Bjerg, P.L., Nielsen, P., Smith, P., Christensen, T.H., 1996. In situ and laboratory determined firstorder degradation rate constants of specific organic compounds in an aerobic aquifer. Environ. Sci. Technol. 30, 31 – 37. Sandrin, S., 2001. PhD dissertation, University of Arizona. United States Environmental Protection Agency, 1998. Technical protocol for evaluating natural attenuation of chlorinated solvents in groundwater. EPA/600/R-98/128. Wang, J.-M., 1999. Intrinsic and enhanced biodegradation of polyaromatic hydrocarbons in aqueous and soil systems. PhD dissertation, University of Arizona. Wang, J.-M., Marlowe, E.M., Miller, R.M., Brusseau, M.L., 1998. Cyclodextrin-enhanced biodegradation of phenanthrene. Environ. Sci. Technol. 32, 1907 – 1912. Wang, X., Brusseau, M.L., 1993. Solubilization of some low-polarity organic compounds by hydroxypropyl-hcyclodextrin. Environ. Sci. Technol. 27, 2821 – 2825.
Use of tracer tests to evaluate the impact of enhanced-solubilization flushing on in-situ biodegradation S.R. Alter a, M.L. Brusseau a,b,*, J.J. Piatt c, A. Ray-Maitra b, J.-M. Wang d, R.B. Cain a a Hydrology and Water Resources Department, University of Arizona, 429 Shantz, Tucson, AZ 85721, USA Soil, Water and Environmental Science Department, University of Arizona, 429 Shantz, Tucson, AZ 85721, USA c Carroll College, 100 North East Avenue, Waukesha, WI 53186, USA d Department of Environment Engineering, Hung Kuang Institute of Technology, Taichung, 433 Taiwan, ROC
b
Received 13 March 2002; received in revised form 26 August 2002; accepted 27 September 2002
Abstract Tracer tests were conducted to evaluate the effect of a complexing sugar flush (CSF) on in-situ biodegradation potential at a site contaminated by jet fuel, solvents, and other organic compounds. Technical-grade hydroxypropyl-h-cyclodextrin was used during the CSF study, which was conducted in a hydraulically isolated cell emplaced in a surficial aquifer. In-situ biodegradation potential was assessed with the use of tracer tests, which were conducted prior to and immediately following the CSF study. Ethanol, hexanol, and benzoate were used as the biodegradable tracers, while bromide was used as a nonreactive tracer. The results indicate that the biodegradation of benzoate was similar for both tracer tests. Conversely, the biodegradation of ethanol (23% increase) and hexanol (41% increase) was greater for the post-CSF tracer test. In addition, analysis of core samples collected from within the test cell indicates that the population density of aerobic jet-fuel degraders increased in the vicinity of the injection wells during the CSF. These results indicate that the cyclodextrin flush did not deleteriously affect the indigenous microbial community. This study illustrates that tracer tests can be used to evaluate the impact of remediation activities on in-situ biodegradation potential. D 2003 Elsevier Science B.V. All rights reserved. Keywords: Ground water quality; Bioremediation; Contaminant transport
* Corresponding author. Soil, Water and Environmental Science, University of Arizona, 429 Shantz, Tucson, AZ 85721, USA. 0169-7722/03/$ - see front matter D 2003 Elsevier Science B.V. All rights reserved. doi:10.1016/S0169-7722(02)00203-6
192
S.R. Alter et al. / Journal of Contaminant Hydrology 64 (2003) 191–202
1. Introduction Remediation technologies based on microbial processes have become popular and widespread. Bioremediation offers several advantages compared to standard pump-andtreat technologies. However, its effectiveness can be limited by numerous factors. One such factor is the presence of large quantities of contaminant mass (e.g., source zones), which often constrains bioremediation applications for highly contaminated sites. Successful cleanup of such sites may require combining bioremediation with a technology that can remove large quantities of contaminant mass. One source-zone remediation technology of current interest is the use of enhanced-solubilization flushing agents in conjunction with pump and treat. Types of flushing agents include surfactants, complexing agents, natural organic matter, and cosolvents. The flushing agents increase the apparent solubility of hydrophobic organic compounds, thereby increasing their rate of removal from the subsurface. It is possible that the bioavailability of the contaminants may be increased due to this enhanced solubilization, thereby altering the magnitudes and rates of biodegradation. Such changes in biodegradation properties could impact the dynamics of bioremediation. In addition, it is to be expected that some of the flushing agent will remain in the treated zone after the flushing event. The potential impact of residual flushing agent on microbial processes is a question of concern. For these reasons, it is important to evaluate the potential impact of enhanced-flushing operations on microbial processes for systems wherein they will be used in conjunction with bioremediation. Cyclodextrin was used as a flushing agent for a pilot-scale field study conducted at Hill Air Force Base (AFB) in Layton, UT. Cyclodextrin has a toroidal molecular structure composed of a hydrophobic interior and a hydrophilic exterior. This structure provides a cavity into which a hydrophobic organic compound can partition, which is the basis for its use as an enhanced-solubilization agent (e.g., Wang and Brusseau, 1993). The work presented herein investigates the effect of the cyclodextrin flushing on in-situ biodegradation potential, which was characterized through the use of tracer tests.
2. Materials and methods 2.1. Background Various methods have been proposed to characterize biodegradation potential for a given site of interest. These methods include conducting laboratory experiments using core samples collected from the field, using contaminant-concentration and geochemical field data to evaluate potential occurrence and rates of attenuation processes, and conducting insitu microcosm studies (e.g., Gilham and Miller, 1990; McAllister and Chiang, 1994; Borden et al., 1995; Nielsen et al., 1996; EPA, 1998). Laboratory-based approaches for assessing biodegradation potential are attractive for several reasons. First, mass-balance calculations are possible for laboratory experiments. In contrast, the amount and distribution of contaminants is unknown for most field sites. Second, in laboratory studies, abiotic processes can generally be separated from biotic processes, another issue that is often difficult to overcome for field sites. Third, as opposed to field conditions, environ-
S.R. Alter et al. / Journal of Contaminant Hydrology 64 (2003) 191–202
193
mental factors such as pH, temperature, redox potential, and nutrient levels can be controlled and their impacts investigated in laboratory experiments. Despite these advantages, the use of laboratory studies is constrained by the fact that they may not adequately represent field conditions (e.g., Chapelle et al., 1996). Recently, so-called ‘‘biotracer’’ tests have been proposed as a method to characterize biodegradation potential (Istok et al., 1997; Brusseau et al., 1999). This method involves conducting a tracer test using a nonreactive tracer along with biodegradable compounds. Breakthrough curves for the biotracers and nonreactive tracer are constructed using data collected for various sampling locations and analyzed to obtain information regarding transport and mass recovery, from which the biodegradation potential can be assessed. The breakthrough curve measured at a specific monitoring location reflects the impact of processes occurring within the swept zone associated with the injection and monitoring points. Thus, the biodegradation parameters obtained from a biotracer test represent an integrated measurement of the monitored zone influenced by the tracer test. The ability to obtain a representative profile of biodegradation activity across a large spatial scale is a distinct advantage of the biotracer method. Another major advantage of the biotracer test is that a known mass of substrate is injected into the subsurface under controlled conditions. This provides the opportunity for conducting mass balances, and thus for quantifying the magnitude and rate of biodegradation of the tracers. 2.2. Site description The tracer tests were conducted in Operable Unit 1 (OU1) at Hill AFB in Layton, UT. The subsurface at this site is contaminated by a multiple-component, immiscible organic liquid. The contamination originates from multiple sources, including chemical disposal pits, fire training areas, landfills, and waste oil pits. The immiscible liquid is comprised primarily of petroleum hydrocarbons (i.e., jet fuel), chlorinated solvents, and polynuclear aromatic hydrocarbons. Past water-table fluctuations have smeared the immiscible liquid such that it is distributed from above the current water table throughout the saturated zone extending to the confining unit. The initial cell-averaged immiscible-liquid saturation was approximately 12%, based on the results of a pre-complexing sugar flush (CSF) partitioning tracer test and organic-carbon analyses of selected cores (Cain et al., 2000). The tracer tests were conducted as part of a series of experiments designed to test innovative remediation technologies (Bedient et al., 1999). The tests were conducted in a hydraulically isolated section of the shallow aquifer that consists of fine-to-coarse sand interbedded with gravel and clay stringers. Depth to groundwater ranged from approximately 5 to 5.5 m. A 60-m-thick clay layer, located at approximately 8 m below ground surface, forms a bottom boundary for the aquifer. The porosity of the test area is approximately 20% (Cain et al., 2000). As discussed in detail by McCray and Brusseau (1998), the CSF was found to be effective in removing organic contaminants at the Hill AFB test site. The cell-wide mass removal was calculated to be approximately 42% based on the results of partitioning tracer tests and analysis of core samples collected before and after cell remediation. The final cell-averaged immiscible-liquid saturation was approximately 7%, based on the results of a post-CSF partitioning tracer test (Cain et al., 2000).
194
S.R. Alter et al. / Journal of Contaminant Hydrology 64 (2003) 191–202
2.3. Tracers Ethanol, hexanol, and benzoate were used as the biodegradable tracers. It is well documented that these compounds can be degraded under both aerobic and anaerobic conditions. These compounds have low volatilities and should experience minimal hydrolysis under the system conditions. The results of preliminary laboratory experiments have shown that sorption of these tracers by the site aquifer material is minimal. Therefore, mass loss as a result of abiotic processes is unlikely. Bromide was used as a nonreactive tracer. As noted above, an immiscible-liquid phase of organic contaminants is present in the test cell. Bromide, benzoate, and ethanol do not partition measurably to this phase. Conversely, hexanol does partition measurably into the immiscible-liquid phase. The immiscible-liquid/water partition coefficient measured for hexanol in batch studies was 2.5 (Cain et al., 2000). The transport and biodegradation of hexanol may have been influenced by mass transfer between the aqueous and immiscible-liquid phases. For example, the bioavailability of hexanol may have been constrained compared to the nonpartitioning tracers. However, preliminary analysis of the transport of tracers that partitioned to the immiscible liquid but experienced minimal biodegradation indicates that this inter-phase mass transfer process can be considered to be essentially instantaneous with respect to hydraulic residence times prevalent in the system (unpublished data). Thus, while quantitative analysis of biodegradation rates for hexanol would require consideration of its partitioning behavior, it is possible to compare overall magnitudes of hexanol biodegradation to those obtained for ethanol and benzoate. Preliminary experiments were conducted in the laboratory using core material collected from the site to evaluate the biodegradation potential of the indigenous microbial community for the biotracers. Batch studies using 14C-labelled compounds confirmed that the biotracers could be biodegraded by the indigenous microbial community under both aerobic and anaerobic conditions. In addition, a batch substrate-loss study was conducted using all three biotracers to examine the relative affinity of the indigenous community for the biotracers. The results showed that hexanol was preferred relative to ethanol, which was preferred relative to benzoate. The more rapid biodegradation of the alcohols compared to benzoate is not surprising, given their respective molecular structures (e.g., Maier, 1999). Miscible-displacement studies were conducted using site core material to evaluate the transport and biodegradation behavior of the tracers, and to compare the behavior of the biotracers to that of representative site contaminants. The results indicated that the general transport and biodegradation behavior of the biotracers was similar to that of representative site contaminants. For example, the rate and magnitude of biodegradation of benzoate during transport was very similar to that of toluene. The results of these laboratory experiments are discussed in more detail in Brusseau et al. (1999) and Wang (1999). 2.4. Field methods The tracer tests were conducted in a cell in which several other studies have been conducted, including the CSF study (McCray and Brusseau, 1998) and partitioning tracer
S.R. Alter et al. / Journal of Contaminant Hydrology 64 (2003) 191–202
195
tests to evaluate the performance of the CSF (Cain et al., 2000). The tracer tests were conducted in conjunction with the partitioning tracer tests, and were conducted prior to and immediately after the CSF. The cell was created by driving steel sheet piling into the clay unit located at an approximate depth of eight meters below ground surface. Interlocking joints were grouted to prevent leakage, which was shown to be minimal in tests conducted prior to initiation of the experiments. A line of four fully screened injection wells and a line of three fully screened extraction wells, both normal to the direction of flow and located approximately 4 m apart, were installed at opposite ends of the cell (Fig. 1). The injection and extraction wells were used to create a steady state flow field prior to the injection of the tracers in each study. Peristaltic pumps (Cole Parmer Master Flex I/P with Tygon LFL tubing) with separate heads for each well were used to generate flow. Equal injection and extraction flow rates of approximately 4.5 l/min were maintained, with the flow divided evenly among the three extraction wells and four injection wells. During each experiment, tracer-free water was first flushed through the cell to create steady-state conditions. The tracer pulses were then injected, followed by the injection of tracer-free water again. Constant flow rates were maintained during the flow establishment, tracer injection, and water flushing portions of each experiment. Tracer injection concentrations were 300, 1000, 925, and 325 mg/l for bromide, ethanol, hexanol, and benzoate, respectively. Approximately 1300 l of tracer solution was injected for each experiment. The tracer tests were approximately 12 days in duration. A three-dimensional, multi-level sampling array (12 locations, 5 depth intervals) connected to a vacuum system was used to collect depth-specific water samples. Samples
Fig. 1. Diagram illustrating cell configuration.
196
S.R. Alter et al. / Journal of Contaminant Hydrology 64 (2003) 191–202
were drawn by vacuum through a 3.2-mm stainless steel tube connected to a 2.5-cm long, 1.25-cm diameter, 40-Am HPLC pump filter buried in situ. During this study, samples were collected from depths of approximately 6.25 and 6.9 m bgs. Depth-integrated samples were collected from the three extraction wells. Extraction wells were screened from approximately 5 to 8.4 m bgs. A total of 18 sampling locations were used for this study. Samples were collected with minimal headspace in polyethylene or glass vials containing a 1-ml aliquot of 1 N NaOH, which served as a preservative to prevent biodegradation. All samples were stored at 4 jC until analysis. Bromide concentrations were determined using an ion-specific electrode. The analytical quantification limit for bromide was approximately 0.4 mg/l. Ethanol and hexanol were analyzed using a gas chromatograph (Shimadzu GC-14A) equipped with a flame ionization detector. Direct liquid injection of 1-Al aliquots was performed on an SPB-624 capillary column (Supelco), 0.53 mm, 30-m long. The analytical quantification limit for the alcohols was approximately 0.5 mg/l. Benzoate concentrations were determined using a high-pressure liquid chromatograph (Waters) with a UV –VIS spectrophotometer. The analytical quantification limit for benzoate was approximately 0.5 mg/l. Due to analytical difficulties (overlapping peaks), benzoate breakthrough curves were obtained for only 8 of the 18 total sampling locations. Effluent samples were periodically collected for analysis of dissolved oxygen. 2.5. Microbial evaluation A microbial characterization was conducted to evaluate the impact of the CSF on the indigenous microbial community. Core samples were collected from the cell after the CSF and tracer tests were completed. The samples were used to assess the indigenous microbial community with respect to both general aerobic microbial populations and to aerobic jetfuel degrader populations. Core samples (approximately 1.2 m) were collected from 4 to 8 m bgs. Samples were collected aseptically from three locations corresponding to the injection end of the cell, the middle of the cell, and the extraction end of the cell. The soil samples were stored in a cooler until plating occurred. Plating was completed within 24 hours of sample collection. Two grams of soil for each sample was placed in 9 ml of autoclaved, deionized water (Basrnstead, Nanopure water, Dubuque, IA) and vortexed for 1 min. The samples were then serially diluted to a range that would produce statistically robust microbial counts. The dilutions were plated on plates containing R2A (Becton and Dickinson, Cockeysville, MD) or mineral salts medium (MSM) combined with 0.05% JP8 jet fuel (added after autoclaving). MSM contains 0.4% Na2HPO4, 0.1% KH2PO4, 0.1% NH4Cl, 0.2% MgSO4, 0.005% yeast extract, 0.0005% ammonium iron (III) citrate, 0.001% CaCl2, and 15 g of agar. The R2A media provides a measure of heterotrophic bacteria while the MSM/JP8 media provides a measure of bacteria capable of degrading jet-fuel components. Samples were plated in triplicate, along with appropriate controls. The plates were incubated at room temperature for 1 week and were then counted. Prior to the second tracer test, a 10% cyclodextrin solution was pumped through the aquifer for 10 days (a total of 65,500 l), followed by flushing with water free of cyclodextrin (about 40,000 l). This pumping activity was associated with the CSF pilot test noted above. During this test, approximately 100% of the injected cyclodextrin was recovered, indicating
S.R. Alter et al. / Journal of Contaminant Hydrology 64 (2003) 191–202
197
that degradation of this compound was negligible. This pumping introduced aerated water into the cell for a significant period of time (greater then 15 days). The oxygen concentration in the injected water was approximately 8 mg/l, while it was 1– 2 mg/l in the effluent pumped from the extraction wells. In addition, the cyclodextrin solution was shown to have enhanced the solubilization and removal of organic constituents (McCray and Brusseau, 1998). This enhanced-solubilization process may have increased the bioavailability of the organic contaminants, as was demonstrated in prior laboratory experiments (Wang et al., 1998). Thus, it is likely that the injection of the aerated cyclodextrin solution stimulated microbial activity within the cell, which in turn resulted in an increase in biomass near the injection wells that will be discussed below.
3. Results and discussion Representative breakthrough curves obtained from the tracer tests are presented in Fig. 2. Breakthrough curves obtained before and after the CSF study exhibit the same general trends. The arrival fronts for benzoate and ethanol are generally coincident with the arrival front for bromide. Conversely, hexanol transport, due to partitioning to the immiscibleliquid phase, is retarded with respect to the other tracers. Bromide has a higher peak than that of the biodegradable tracers, and exhibits longer elution tails. This behavior is attributed to biodegradation and resultant mass loss of the biotracers. Mass recoveries were calculated for the tracers using moment analysis. Mass loss of the biodegradable tracers was observed for all sampling locations. Given that oxygen concentrations were limiting in the extraction-well effluent, these results suggest that both aerobic and anaerobic degradation processes contributed to biotracer mass loss. This is consistent with the results of the preliminary batch experiments noted above. The mass loss of the biotracers varied with sampling location. The greatest losses were observed for sampling locations near the injection wells and for those with the largest associated hydraulic residence times. Detailed analysis of the spatial variability of biotracer biodegradation for these tracer tests is presented by Sandrin (2001). The mean mass recoveries for bromide, the nonreactive tracer, were close to 100% for both the pre- and post-tracer tests. This indicates there was minimal loss of tracer mass due to hydraulic factors. Conversely, the mean mass recoveries for the biodegradable tracers were significantly less than 100% (see Table 1). The average recovery of benzoate for the tracer test conducted before the CSF study, as calculated from eight MLS locations, was approximately 82%. The average recoveries of ethanol and hexanol for the pre-tracer test, calculated using only the data for the same eight locations for which benzoate data were measured, were approximately 78% and 82%, respectively. The average recoveries for ethanol and hexanol, calculated using data from all sampling locations, were approximately 76% and 83%, respectively. There is minimal difference in the average ethanol and hexanol recoveries calculated using the data collected for the eight locations for which benzoate data are available and those calculated using data collected from all 18 locations. This suggests that the fewer number of measurements available for benzoate does not significantly impact the calculated average mass recoveries. Interestingly, the recoveries for all three biotracers are similar.
198
S.R. Alter et al. / Journal of Contaminant Hydrology 64 (2003) 191–202
Fig. 2. Breakthrough curves representative of those obtained from the tracer tests. These data were collected at the 6.9 m bgs sampling plane, from locations 11 (top), 13 (middle), and 33 (bottom), during the post-CSF tracer test.
S.R. Alter et al. / Journal of Contaminant Hydrology 64 (2003) 191–202
199
Fig. 2 (continued).
The average recoveries of benzoate, ethanol, and hexanol for the tracer test conducted after the CSF were approximately 79%, 64%, and 54%, respectively, for the eight MLS locations for which benzoate data are available. The average percent difference in mass recoveries for the pre- and post-tracer tests was approximately 4% for benzoate, which is considered within experimental uncertainty, indicating the CSF had minimal impact on benzoate biodegradation potential. Conversely, the differences between pre- and post-CSF average mass recoveries for ethanol and hexanol were approximately 23% and 41%, respectively. Thus, the CSF appeared to enhance the biodegradation potential for ethanol and hexanol. Table 1 Comparison of pre- and post-CSF recoveries for biotracers Tracer
Pre-CSF study
Post-CSF study
Benzoate 8 locations
Percent difference
82
79
4
Ethanol 8 locations All locations
78 76
64 59
18 23
Hexanol 8 locations All locations
82 83
54 49
34 41
Note: the values reported in the table are considered to have an associated measurement uncertainty of V 10% based on analytical methods.
200
S.R. Alter et al. / Journal of Contaminant Hydrology 64 (2003) 191–202
Hexanol exhibited the greatest difference in recovery between pre- and post-CSF experiments, while ethanol and benzoate exhibited smaller differences. The pre-CSF mass recovery of hexanol was similar to those of ethanol and benzoate, while the post-CSF recovery for hexanol is significantly lower than the other two. This behavior could be related to the fact that hexanol is the only one of the three biotracers that partitions to the immiscible-liquid phase. Partitioning to the immiscible liquid may have reduced the bioavailability of hexanol, thus reducing its magnitude of biodegradation. Given that there was less immiscible-liquid present during the post-CSF biotracer test (due to removal during the CSF), the retention of hexanol was less for the post tracer test compared to the tracer test conducted prior to the CSF. Thus, it is possible that the bioavailability of hexanol increased after the CSF, allowing for greater biodegradation. This phenomenon would not have affected the other tracers, as they did not partition measurably to the immiscible liquid. The magnitude of the change in hexanol biodegradation with respect to the other tracers may also in part be due to the fact that the indigenous community appears to have a greater affinity for hexanol than for benzoate and ethanol, as shown by the results of the batch laboratory experiments noted above. It is possible the perturbations associated with the CSF, such as the solubilization of immiscible-liquid constituents and the introduction of oxygen, may have had the greatest impact on the more readily degradable compound. 3.1. Microbial evaluation The biomass concentrations obtained for core samples collected from the cell are similar for both heterotrophic and jet-fuel degraders. However, the biomass concentrations are significantly higher for the core collected near the injection wells, approximately 107 CFU/g soil, versus approximately 105 for the cores collected from the center of the cell and near the extraction wells (see Fig. 3). The biomass concentrations were most likely
Fig. 3. Distribution of biomass concentrations within the cell; aerobic heterotrophic degraders and aerobic jet-fuel degraders.
S.R. Alter et al. / Journal of Contaminant Hydrology 64 (2003) 191–202
201
relatively uniform prior to the CSF, as indicated by the concentrations measured for the shallow depth intervals above the water table, approximately 4 to 5.2 m bgs, which are uniformly low (see Fig. 3). This section of the subsurface was not affected by the CSF and, therefore, the data collected for this zone can be used to represent general baseline conditions within the cell. The results presented in Fig. 3 suggest that biomass concentrations increased during the CSF. This growth, which occurred primarily in the region near the injection wells, is most likely due to the impact of the CSF on microbial activity, as noted above.
4. Conclusions Results of the tracer tests indicate that a complexing sugar flush did not impede microbial activity and associated in-situ biodegradation potential. The results of the microbial evaluation demonstrate that viable, active aerobic microbial populations were present following the CSF. This is consistent with the results obtained from the tracer test conducted after the CSF, wherein mass loss of the biodegradable tracers was observed. Overall, the magnitude of biodegradation of the biotracers was greater after the CSF study. This, in combination with the evidence of active microbial populations present after the study, suggests that the CSF remediation technology may have enhanced biodegradation potential, at least for some of the lower molecular weight hydrocarbons. Finally, the usefulness of the ‘‘biotracer’’ method for characterizing field-scale, in-situ biodegradation potential and evaluation of the effects of remedial actions on biodegradation has been illustrated in this study.
Acknowledgements We thank William Blanford, John McCray, Adria Bodour, Qinhong Hu, Gwynn Johnson, Gale Famisan, Fiona Jordan, and other former UA students for their help. We also would like to thank Lynn Wood and Carl Enfield of the US Environmental Protection Agency for their support. This research was supported by projects funded by the US Environmental Protection Agency, one as part of the Joint Bioremediation Program and one in conjunction with the DOD SERDP program.
References Bedient, P., Holder, A., Enfield, C.G., Wood, L., 1999. Enhanced remediation demonstration at Hill Air Force Base: introduction. In: Brusseau, M., Sabatini, D., Gierke, J., Annable, M. (Eds.), Innovative Subsurface Remediation Field Testing of Physical, Chemical, and Characterization Technologies. American Chemical Society, Washington, DC, pp. 36 – 48. Borden, R.C., Gomez, C.A., Becker, M.T., 1995. Geochemical indicators of intrinsic bioremediation. Ground Water 33, 180 – 189. Brusseau, M., Piatt, J., Wang, J., Hu, M., 1999. A biotracer test for characterizing the in-situ biodegradation potential associated with subsurface systems. In: Brusseau, M., Sabatini, D., Gierke, J., Annable, M. (Eds.),
202
S.R. Alter et al. / Journal of Contaminant Hydrology 64 (2003) 191–202
Innovative Subsurface Remediation, Field Testing of Physical, Chemical, and Characterization Technologies. American Chemical Society, Washington, DC, pp. 240 – 250. Cain, R.B., Johnson, G.R., McCray, J.E., Blanford, W., Brusseau, M.L., 2000. Partitioning tracer tests for evaluating remediation performance. Ground Water 38, 752 – 761. Chapelle, F., Bradley, P., Lovley, D., Vroblesky, D., 1996. Measuring rates of biodegradation in a contaminated aquifer using field and laboratory methods. Ground Water 34, 691 – 698. Gilham, R.W., Miller, D.J., 1990. A device for in situ determination of geochemical transport parameters: 2. Biochemical reactions. Ground Water 28, 858 – 862. Istok, J.D., Humphrey, M.D., Schroth, M.H., Hyman, M.R., O’Reilly, K.T., 1997. Single-well, ‘‘push-pull’’ test for in-situ determination of microbial activies. Ground Water 35, 619 – 631. Maier, R., 1999. Microorganisms and organic pollutants. In: Maier, R., Pepper, I.L., Gerba, C.P. (Eds.), Environmental Microbiology. Academic Press, San Diego, pp. 363 – 402. McAllister, P.M., Chiang, C.Y., 1994. A practical approach to evaluating natural attenuation of contaminants in ground water. GWMR, Spring, 161 – 173. McCray, J.E., Brusseau, M.L., 1998. Cyclodextrin-enhanced in situ flushing of multiple-component immiscible organic liquid contamination at the field scale: mass removal effectiveness. Environ. Sci. Technol. 32, 1285 – 1293. Nielsen, P.H., Bjerg, P.L., Nielsen, P., Smith, P., Christensen, T.H., 1996. In situ and laboratory determined firstorder degradation rate constants of specific organic compounds in an aerobic aquifer. Environ. Sci. Technol. 30, 31 – 37. Sandrin, S., 2001. PhD dissertation, University of Arizona. United States Environmental Protection Agency, 1998. Technical protocol for evaluating natural attenuation of chlorinated solvents in groundwater. EPA/600/R-98/128. Wang, J.-M., 1999. Intrinsic and enhanced biodegradation of polyaromatic hydrocarbons in aqueous and soil systems. PhD dissertation, University of Arizona. Wang, J.-M., Marlowe, E.M., Miller, R.M., Brusseau, M.L., 1998. Cyclodextrin-enhanced biodegradation of phenanthrene. Environ. Sci. Technol. 32, 1907 – 1912. Wang, X., Brusseau, M.L., 1993. Solubilization of some low-polarity organic compounds by hydroxypropyl-hcyclodextrin. Environ. Sci. Technol. 27, 2821 – 2825.