• E-mail
• Print
• Comment
• Font Size
• Digg
• del.icio.us
• Discuss article

Modeling the Kinetics of Contaminant Biodegradation By a Mixed Microbial Consortium Based on Carbon Mass Balance

Posted on: Tuesday, 1 March 2005, 03:00 CST

The fate of contaminant carbon was monitored during aerobic biodegradation in the presence of a mixed indigenous microbial consortium in order to calibrate a microbial-growth-based biokinetic model. The methodology simultaneously monitored mineralization, substrate depletion and microbial population evolution in biomass extract spiked with ^sup 14^C-labeled hexadecane. Hexadecane depletion and hexadecane-degrader population were monitored using sacrificed microcosms by centrifuging the extract so that the supernatant and the residue contained residual hexadecane and microbial population, respectively. This methodology allowed verification of the carbon mass balance (average ^sup 14^C-carbon recovery of 90.33 1.62% for biotic microcosms) and calibration of a biokinetic model. Four biokinetic parameters and three yield coefficients were identified (Haldane kinetic parameters: μ^sub S^ = 1.3639 d^sup -1^, Ks = 0.4295 mg-C, K^sub I^ = 6.6457 mg-C; decay kinetic parameter: μ^sub d^ = 1.3.10^sup -2^ d^sup -1^; substrate/biomass, carbon dioxide/ biomass during growth and carbon dioxide/biomass during decay yield coefficients: Y^sub S^ = 1.5948 mg-C/mg-C, Y^sup g^^sub P^ = 0.4554 mg-C/mg-C, Y^sup d^^sub P^ = 1.3263 mg-C/mg-C) and compared with the literature data. The methodology can facilitate the identification of biodegradation models by decoupling the intrinsic ability of microorganisms to degrade contaminant from restrictions imposed by limiting conditions.

Keywords Intrinsic contaminant biodegradation, indigenous microorganisms, carbon mass balance, biokinetic model.

Introduction

Characterizing and modeling of the biodegradation of contaminants by the indigenous microbial population in complex media such as soil is a key issue in the optimization of treatment bioprocesses. However, this is a difficult task due to the number and the complexity of the phenomena involved: contaminant properties, soil properties, nutrient and oxygen supply as well as microbial population evolution, etc. One way to cope with this difficulty is to decouple the phenomena involved so as to better understand the phenomena individually. In particular, it is useful to characterize the intrinsic capacity of the indigenous microorganisms to degrade a given contaminant in the absence of limiting conditions, such as a restricted supply of nutrients or oxygen. This characterization can be achieved by putting the contaminant and the indigenous microorganisms in contact, within a controlled and non-limiting environment (a biomass extract with adequate supply of nutrients and oxygen). The intrinsic capacity of the indigenous microorganisms to degrade the contaminant can then be characterized by monitoring the three main phenomena involved in contaminant biodegradation under the experimental conditions; contaminant depletion, contaminant mineralization (conversion of the contaminant carbon into carbon dioxide) and microbial growth (conversion of the contaminant carbon into cellular material). This requires simultaneous monitoring of the contaminant, contaminant mineralization and the degrading microbial population. Note that monitoring only two of these three states leads to an incomplete characterization of the phenomena and therefore restricts the choice of model structures. For example, if the biomass is not measured, only non-growth-based models such as first-order and Michaelis-Menten kinetics (Feng et al., 2000; Park et al., 2001) can be used. Similar experimental protocols reported in the literature to characterize contaminant biodegradation and to calibrate biokinetic models can be classified into two categories: protocols based on the monitoring by chemical analysis and protocols based on the monitoring of radioactive atoms. The former are the most commonly used (Misra and Pavlostathis, 1997; Yerushalmi and Guiot, 1998; Wick et al., 2001; Woo et al., 2001). On the one hand, no analytical tool permits to monitor directly the degraders mass in a mixed microbial consortium. Only indirect measurements such as culture-based bacterial counts are available to give qualitative information on the evolution of the microbial population. This means that the estimation of mass balance based on chemical analysis requires assumptions to convert indirect measurements into bacterial mass concentrations. For example, to determine degrader mass concentration from bacterial counts, the cellular mass weight must be known and this is very difficult in the presence of a mixed microbial consortium. On the other hand, the main advantages of the protocols based on the monitoring of radio labeled atoms are their high level of accuracy and (when using ^sup 14^C) their capacity to monitor the fate of the contaminant carbon. In particular, these protocols allow distinction between contaminant biodegradation and other biological changes (e.g., mineralization of non-contaminant organic matter, mortality of non-degrading microorganisms, etc). ^sup 14^C-labeled compounds are widely used to monitor contaminant mineralization by trapping and quantifying the contaminant carbon converted into carbon dioxide (Yerushalmi and Guiot, 1998; Feng et al., 2000; Park et al., 2001). However, ^sup 14^C-labeled compounds have not been widely used to monitor the fate of the contaminant carbon mainly due to the difficulty to simultaneously monitor the residual contaminant and the degrading microorganisms. Poeten and collaborators used ^sup 14^C-labeled PAH compounds to monitor the CO2 produced, the unreacted PAH in the dissolved phase or sorbed to sediment and the non-polar intermediate products (Poeton et al., 1999).

The experimental methodology developed in this paper uses a protocol based on monitoring the fate of the ^sup 14^C-labeled contaminant and also aims to corroborate the fate of the ^sup 14^C with contaminant depletion, contaminant mineralization and the population evolution of the degraders. The protocol was designed to permit the estimation of carbon mass balance to monitor the population evolution of the contaminant degraders and to calibrate a microbial-growth-based biokinetic model capable of describing the biodegradation of a given contaminant by a mixed microbial consortium. An application of this protocol concerns modeling of biodegradation in soils. Phenomenological models of biodegradation in soils are often faced with identification problems because of the number and the complexity of the phenomena involved. Characterization of the intrinsic contaminant biodegradation allows decoupling of the intrinsic capacity of the microorganisms to degrade the contaminant from the influence of limiting conditions and will therefore facilitate the identification of models for biodegradation in soils. In this paper, the experimental procedure is applied to a biomass extract, in order to achieve this decoupling.

This paper consists of three parts: a description of the methodology, an analysis of carbon fate (including a mass balance) and the calibration of a biokinetic model.

Materials and Methods

Preparation of the Biomass Extract

A surface soil, classified as loamy sand according to USDA criteria, was used for this study. The following extraction procedure based on a mechanical detachment of soil biomass has been chosen for its efficiency with loamy-sand soil structures (Ramsay, 1984).

In a 150 ml glass bottle, 10 10^sup -2^ g of non-polluted soil was mixed with 95 ml of sterilized mineral medium and thirty sterilized glass beads (3 mm). The mineral medium had a pH of 7.02 after autoclaving and contained the following quantities of salts per liter of pure water; MgSO^sub 4^ * 7 H2O, 1.0 g; KCl, 0.7 g; KH^sub 2^PO^sub 4^, 2.0 g; K^sub 2^HPO^sub 4^, 3.7 g; NH^sub 4^NO^sub 3^, 1.0 g. The slurry was mixed for 30 min on a Wrist Action shaker (Burrell Scientific, Pittsburgh, PA). The extract was then transferred into sterile centrifugation tubes (Fisher Scientific) and centrifuged for 30 sec at 1400 g (IEC 21000R, Needham Heights, USA). The supernatant was transferred into a sterilized glass bottle and was used as the biomass extract for the biodegradation assays.

Biodegradation Assays in Microcosms

Hexadecane was used as the contaminant. Each microcosm consisted of 10 ml of the biomass extract in an autoclaved 120 ml serum bottle (Fisher Scientific) containing a 5 ml glass tube. The glass tube contained 2 ml of 1 N KOH as a CO2 trap. Abiotic control microcosms were prepared with 0.2 g sodium azide (NaN^sub 3^). All serum bottles were closed using an aluminum seal with septa (Supelco) sterilized prior to use by treatment with ethanol (70% vol./vol.) and exposure to UV-light (30 min). Serum bottles were injected with a hexadecane solution containing 200,000 dpm of ^sup 14^C-labeled hexadecane. Prepared microcosms were incubated in the dark at environmental temperature under aerobic conditions. A continuous agitation was provided at approximately 80 rpm using an incubator shaker (Lab Line).

Microcosm characteristics are summarized in Table 1. Two initial contaminant levels (S^sub 0^) were tested: 8.2 mg-C and 16.4 mg-C of total hexadecane. Given that each microcosm contains biomass from 1 g of soil, these two contaminant levels correspond to two typical soil contamination levels: about 8,000 and 16,000 mg-C/kg of soil respectively. In orde\r to verify the reproducibility of the biodegradation tests (discussed in the Results and Discussion section), the microcosms used for the monitoring of the carbon dioxide production were performed in triplicates. The concentration of ^sup 14^C-labeled hexadecane (4.25 10^sup -3^ mg-C/microcosm) was negligible when compared to the concentration of a non- radioactive hexadecane.

Monitoring of Carbon Dioxide, Hexadecane Depletion and Biomass Evolution

The overall methodology is schematically represented in Figure 1.

Monitoring of Carbon Dioxide Production. Nine microcosms were prepared for continuous measurement of total CO2 and ^sup 14^C- labeled CO2: one abiotic control microcosm and three biotic microcosms for each of the two contaminant levels and a single non- contaminated biotic microcosm. KOH solution was renewed in all prepared microcosms after each analysis of ^sup 14^C-labeled CO2 and total CO2, respectively.

Table 1

Summary of microcosms

The quantity of ^sup 14^C-labeled CO2 trapped was determined by retrieving 1 ml of KOH and adding 10 ml of scintillation cocktail (Opti-phase Hisafe-3, Wallac, Montreal, Canada). The mixture was analyzed by a liquid scintillation counter (Wallac, model 1409, Montreal, Canada). Pure 1N KOH solution was used as a blank.

Total carbon dioxide was determined by acid-base titration at the same time intervals as for measurement of ^sup 14^C-labeled CO2. The quantity of total CO2 trapped was determined by retrieving the 1 ml of KOH remaining and adding 1 ml of 1N barium chloride solution (BaCl^sub 2^*H2O) to precipitate carbonates as well as three drops of phenolphtaleine solution as an indicator. The mixture was titrated with standardized hydrochloric acid (HCl, 0.1N).

Although the oxygen in the microcosms was theoretically sufficient to support aerobic microbial activity, 2 ml of air was added to all microcosms at the time of each CO2/^sup 14^C-CO2 measurement in order to overcome the air depression caused by oxygen consumption and carbon dioxide entrapment.

Monitoring of Biomass Evolution and Contaminant Depletion. Forty microcosms were prepared for one-time measurement of ^sup 14^C- labeled biomass and residual ^sup 14^C-labeled hexadecane concentration. For each sampling event, five microcosms were sacrificed; one abiotic and one biotic microcosm for each contaminant level as well as one non-polluted biotic microcosm. The sampling events were timed to correspond to suitable stages of the biodegradation process as determined by production of ^sup 14^CO^sub 2^.

The 10-ml solution of the sacrificed microcosm was placed into a 25 ml Corex centrifuge tube (Fisher Scientific) and centrifuged at ambient temperature for 10 min at 4500 rpm. Hexane (2 ml) was added to the Corex centrifuge tube to rinse the inner wall. After a second centrifugation with the same settings, the supernatant was carefully removed with minimum disturbance of the sediment. The serum bottle and intermediary equipment were rinsed three times with hexane and the rinsing solution was added to the supernatant. The supernatant was then mixed with scintillation cocktail and analyzed with the liquid scintillation counter. The non-polluted microcosm was used as a blank.

Figure 1. Schematic representation of the experimental procedures.

^sup 14^C-labeled biomass was determined from the radioactivity measurement in the residue which had to be treated specially due to the incorporation of the radioactivity within the cells and to allow color interferences during the scintillation counts. A sterilized lysis solution of 4 ml (50 mM TRIS, 50 mM EDTA, 50 mM NaCl, pH = 8) and 2 g of sterilized glass beads (3 mm) were added to the residue obtained after the two centrifugations described above. The mixture was vortexed for 5 min to completely disperse the residue. Lysozyme powders (Sigma) (40 mg) were added to the dispersed solution which was incubated for 10 min at room temperature. The tube was vortexed for 2 min. A volume of 8 ml of the sterilized lysis solution was added to the tube, which was vortexed again for 2 min. The solution was transferred into 3 scintillation vials in order to dilute it and to cope with color interferences. The Corex centrifuge tubes and intermediary equipment were rinsed three times with the sterilized lysis solution, which was then transferred into a scintillation vial. The four mixtures were combined with scintillation cocktail and were analyzed with the liquid scintillation counter. The non- contaminated microcosm was used as a blank.

Hexadecane mineralization and population evolution of hexadecane degraders were determined from the ^sup 14^C-labeled hexadecane partitioning under the assumption that the radioactive and non- radioactive hexadecane have the same fate.

Alternative Monitoring of Biomass Evolution. The technique of biomass monitoring described in the previous section was qualitatively corroborated using the Most Probable Number (MPN) method (Mills et al., 1978) on 24 microcosms that do not contain ^sup 14^C-labeled hexadecane with the assumption that cellular mass does not change throughout the biodegradation test. For each sacrifice, the entire 10 ml was used for the dilution, and triplicate assays were performed. Nutrient broth and hexadecane in a mineral medium (described above) were used to measure the total heterotrophic and hexadecane-degrader populations, respectively.

Results and Discussion

Fate of Contaminant Carbon

The three fractions of radioactivity recovered in the CO2 trap, both in the supernatant and in the residue, are plotted as a function of time in Figure 2.

The CO2 curve presenting the average values of the triplicate data shows that the biodegradation test was repeatable. The error bars during the period of high CO2 production rates in the exponential phase which occurs between the periods of low CO2 production rates during the lag phase and the plateau phase were significantly larger.

Mass balance on the ^sup 14^C, calculated at each sacrifice led to an average recovery of 90.33 1.62% for biotic microcosms and 94.70 1.72% for abiotic microcosms. This indicates that the loss of radioactivity during the manipulations is less than 10% and that the data are acceptable for the purpose of quantitative interpretation regarding the fate of the contaminant carbon.

Curve characteristics are similar for both low and high initial pollution values. For the abiotic microcosms, the radioactivity level in the CO2 trap and the residue remained close to zero, indicating that no physico-chemical oxidation and no hexadecane adsorption has occurred on the residue. For the biotic microcosms, after a six-day lag phase, the radioactivity in the supernatant decreased rapidly corresponding to a rapid increase in the radioactivity of the CO2 trap and in the residue. As the radioactivity in the supernatant approached zero, the radioactivity in the CO2 trap increased at a low and constant rate and the radioactivity in the residue decreased at a low and constant rate. These observations indicate that the radioactivity fractions in the CO2 trap, in the supernatant and in the residue characterize the hexadecane mineralization and the population evolution of hexadecane- degrading microbes. After the lag phase, the hexadecane was rapidly consumed corresponding to an exponential growth of the hexadecane- degrader population and an exponential increase in CO2 production. When most of the hexadecane was consumed, the hexadecane degraders died at a constant rate corresponding to a constant CO2 production rate. Radioactivity in the supernatant of the abiotic flasks remained constant at approximately 100% during the experiment, indicating that hexadecane was not degraded abiotically.

The MPN data were used to confirm this interpretation. Results are shown in Figure 3. In the microcosms without hexadecane, the total heterotrophic and hexadecane-degrader populations remained constant (around 10^sup 8^ MPN index/10 ml and 10^sup 5^ MPN index/ 10 ml, respectively) when compared to the ones with hexadecane. In the polluted microcosms, microbial growth occurred during the first 15 days for both initial contaminant levels, followed by a population decline. For the low initial contaminant level, the total heterotrophic population was significantly higher than the hexadecane-degrader population. For the high initial contaminant level, the difference between the total heterotrophic and hexadecane- degrader populations was less pronounced and the hexadecane- degrader population became predominant.

Figure 2. ^sup 14^C partitioning in different phases as a function of time ([black circle] supernatant-biotic; [white circle] supernatant abiotic; [black triangle down] residue-biotic; [white triangle down] residue-abiotic; CO2-biotic; [white square] CO2- abiotic).

The mass balance and MPN data indicate that the fate of the radiolabeled contaminant carbon can be used to characterize the intrinsic capacity of a mixed microbial consortium to degrade a given contaminant and to calibrate a biokinetic model.

Calibration of a Biokinetic Model

Figure 3. Monitoring of the total heterotrophic and hexadecane- degrader populations in microcosms with or without hexadecane using the MPN technique ([black circle] hexadecane-degrader-with hexadecane; [white circle] hexadecane-degrader-without hexadecane; [black triangle down] total heterotrophic-with hexadecane; [white triangle down] total heterotrophic-without hexadecane; * data corresponding to minima due to insufficient dilution).

Consequently, seven parameters must be identified: four biokinetic parameters (μ^sub S^, K^sub S^, K^sub I^, and μ^sub d^) and three yield coefficients (Y^sub S^, Y^sup g^^sub P^ and Y^sup d^^sub P^). Parameter identification was performed using a non-linear least-squares method based on the Marquardt- Levenberg algorithm (Isqcurvefit in Matlab). Data from carbon dioxide production, hexadecane depletion a\nd microbial population evolution were all used for the curve fitting leading to two sets of parameters corresponding to the two initial contaminant levels. The initial value of the hexadecane-degrader mass, X^sub 0^, could not be determined directly and was assumed to be equal to 4.25 * 10^sup - 2^ mg-C based on the initial biomass dry weight. As the parameter identification procedure produces a non-unique solution, the solution must be constrained by initializing the parameter identification procedure with realistic parameter values: μ^sub S^ close to 1; K^sub S^ < K^sub I^; μ^sub d^ [much less than] μ^sub S^, Y^sub S^, Y^sup g^^sub P^ and Y^sup d^^sub P^ close to 1. The results of the parameter identification procedure are presented in Table 2. Figure 4 illustrates the curve fitting for parameter identification, and demonstrates that the model satisfactorily describes: (1) CO2 production due to hexadecane mineralization and decay of hexadecane-degraders; (2) hexadecane depletion; and (3) population evolution of hexadecane-degraders.

As the identification procedure depends on the initial value of X (X^sub 0^), which was indirectly estimated using the biomass dry weight and the bacterial counts, the sensitivity of the model is examined with regard to X^sub 0^. Simulations have been performed after decreasing and increasing X^sub 0^ by 10% and 50%. A 10% increase and decrease of X^sub 0^ has no significant (no visible) impact on curves presented in Figure 4. Results for a 50% increase or decrease of X^sub 0^, which corresponds to a quite large change, are presented in Figure 4. It is clear that despite the large changes in X^sub 0^, CO2 production, hexadecane depletion and biomass evolution are influenced by small variations. This sensitivity study shows that the model is not very sensitive with respect to X^sub 0^, which strengthens the assumption made in the initialization of X.

Some comments concerning the values of the parameters can be made. The characteristic growth rate, μ^sub S^, is of the same order of magnitude as those generally observed for bacteria (Bailey and Ollis, 1986). Secondly, the affinity constant, K^sub S^ and the inhibitory constant, K^sub I^, are much higher than the solubility of hexadecane in water at 25C, which is equal to 3.58*10^sup -3^ mg/ L (Schwarzenbach et al., 1993). Indeed, when expressed in mg/L, K^sub S^ and K^sub I^ are equal to 50.55 mg/L and 782.25 mg/L, respectively. This can be explained by the ability of microorganisms to increase the apparent solubility of hexadecane by the specific biological phenomena like the production of extracellular enzymes or of the biosurfactants as suggested by several studies (Pignatello et al., 1983; Harms and Zehnder, 1995; Dziel et al., 1996). To prevent the affinity constant and the inhibitory constant from being greater than the contaminant solubility, one should incorporate into the biokinetic model both the physico-chemical and biological phenomena involved in the contaminant mass transfer from the free phase to the aqueous phase. Thirdly, identification of the yield coefficients led to a CO2/hexadecane yield coefficient (Y^sup g^^sub P^/Y^sub S^) equal to 0.29 mg CO2-C (mg hexadecane-C)^sup -1^, which is in accordance with the values by Graham et al., who reported yield coefficients ranging from 0.18 to 0.52 mg CO2-C (mg hexadecane- C)^sup -1^, depending on the C:N:P ratio (Graham et al., 1999).

Table 2

Results of parameter identification

Figure 4. Parameter identification ([black circle] Experimental data for S^sub 0^ - 8.2 mg-C; [white circle] Experimental data for S^sub 0^ = 16.4 mg-C; - Simulated values; Model sensitivity with respect to X^sub 0^: curves superposed for a 10% change, ... ..50% change).

Conclusion

An experimental procedure based on the fate of the contaminant radiolabeled carbon has been developed to selectively and simultaneously monitor contaminant, carbon dioxide and biomass during biodegradation assays, thereby permitting calculation of a carbon mass balance. This method was applied to a biomass extract that allowed the identification of four kinetic parameters (three Haldane-type parameters and one mortality parameter), and three yield coefficients, which correlates contaminant depletion, mineralization and microbial population evolution. The identification procedure has been illustrated using a simple biokinetic model but more complex ones can also be identified using the same methodology. Because ^sup 14^C-labeled monitoring is very sensitive, lower substrate and biomass concentrations can be used, depending on the field conditions studied. More sacrifices can also be performed to increase the accuracy of the parameter identification. The procedure allows (1) increasing the biokinetic model reliability with the use of a carbon mass balance, and (2) decoupling the intrinsic biodgradation kinetics under non-limiting conditions from biodegradation under limiting conditions (substrate, nutrients oxygen, soil structure). Thus, if the parameters describing intrinsic contaminant biodegradation by the indigenous soil microorganisms are known, only the parameters accounting for the limiting conditions remain to be identified.

The experimental procedure described in this paper is very simple, requiring only a centrifuge and a scintillation counter and is independent of the biomass extraction procedure. It can be applied to any situation requiring the characterization of the contaminant biodegradation in the presence of a mixed microbial consortium.

Acknowledgements

The authors thank Dr. Yves Dudal and Dr. Dan Walker for the reading of the manuscript and for their helpful comments on this manuscript and acknowledge the financial support from the partners of the NSERC Industrial Chair in Site Remediation and Management: Alcan, EDF/GDF, Bell Canada, Cambior, Canadian Pacific Railway, Centre Expertise Analyse Environnementale du Qubec (CEAEQ), City of Montreal, Total Fina Elf, Hydro-Qubec, Natural Science and Engineering Research Council (NSERC), Petro-Canada and Solvay.

References

Bailey, J.E. and Ollis, D.F. 1986. Biochemical Engineering Fundamentals. New York, McGraw-Hill. Dziel, E., Paquette, G., Villemur, R., Lpine, R, and Bisaillon, J.-G. 1996. Biosurfactant production by a soil pseudomonas strain growing on polycylic aromatic hydrocarbons. Appl. and Environ. Microbiol. 62(6), 1908- 1912.

Feng, Y., Park, J.-H., Voice, T.C., and Boyd, S.A. 2000. Bioavailability of soil-sorbed biphenyl to bacteria. Environ. Sci. Technol. 34(10), 1977-1984.

Graham, D.W., Smith, V.H., Cleland, D.L., and Law, K.P. 1999. Effects of nitrogen and phosphorus. supply on hexadecane biodegradation in soil systems. Water Air Soil Poll. 111, 1-18.

Harms, H. and Zehnder, A.J.B. 1995. Bioavailability of sorbed 3- chlorodibenzofuran. Appl. Environ. Microbiol. 61, 27-33.

Mills, A.L., Breuil, C., and Colwell, R.R. 1978. Enumeration of petroleum-degrading marine and estuarine microorganisms by the most probable number method. Can. J. Microbiol. 24, 552-557.

Misra, G. and Pavlostathis, S.G. 1997. Biodegradation kinetics of monoterpenes in liquid and soil-slurry systems. Appl. Microbiol. Biotechnol. 47, 572-577.

Park, J.-H., Zhao, X., and Voice, T.C. 2001. Biodegradation of non-desorbable naphtalene in soils. Environ. Sci. Technol. 35(13), 2734-2740.

Pignatello, J.J., Martinson, M.M., Steiert, J.G., Carlson, R.E., and Crawford, R.L. 1983. Biodegradation and photolysis of pentachlorophenol in artificial freshwater streams. Appl. Environ. Microbiol. 46, 1024-1031.

Poeton, T.S., Stensel, H.D., and Strand, S.E. 1999. Biodegradation of polyaromatic hydrocarbons by marine bacteria: effect of solid phase on degradation kinetics. Water Res. 33(3), 868- 880.

Ramsay, A.J. 1984. Extraction of bacteria from soil: efficiency of shaking or ultrasonication as indicated by direct counts and autoradiography. Soil Biol. Biochem. 16(5), 475-481.

Schwarzenbach, R.P., Gschwend, P.M., and Imboden, D.M. 1993. Environmental Organic Chemistry. New York, John Wiley & Sons, Inc.

Wick, L. Y., Colangelo, T., and Harms, H. 2001. Kinetics of mass transfer-limited bacterial growth on solid PAHs. Environ. Sci. Technol. 35(2), 354-361.

Woo, S. H., Park, J. M., and Rittmann, B. E. 2001. Evaluation of the interaction between biodegradation and sorption of phenanthrene in soil-slurry systems. Biotechnol. and Bioeng. 73(1), 12-24.

Yerushalmi, L. and Guiot, S.R. 1998. Kinetics of biodegradation of gasoline and its hydrocarbon constituents. Appl. Microbiol. Biotechnol. 49, 475-481.

OLIVIER SCHOEFS, LUCIE JEAN, ANDREE ELEERT, MICHEL PERRIER, AND RJEAN SAMSON

Department of Chemical Engineering, cole Polytechnique de Montral, Montreal, Quebec, Canada

Address correspondence to Olivier Schoefs, Chemical Engineering Department, University of Technology of Compiegne, PO Box 20.529, 60205 Compiegne Cedex, France. E-mail: olivier.schoefs@utc.fr

Copyright CRC Press 2005

Source: Soil & Sediment Contamination

More News in this Category