biotreatment of pah-contaminated soils/sedimentsa

Biotreatment of PAH-contaminated Soils/Sediments"

EDWARD J. BOUWER," WEIXIAN ZHANG,' LIZA P. WILSON: AND NEAL D. DURANT"

bDepartment of Geography and Environmental Engineering The Johns Hopkins University

Baltimore, Maryland 21218 'Department of Civil und Enuironmental Engineering

Lehigh University Bethlehem, Pennsylvania 181 05

'American Association for the Advancement of Science Fellow National Center ,for Environmentul Assessment

US. Environmentul Protection Agency 401 M. Street S. W. (8620) Washington, D. C. 20460

Polycyclic aromatic hydrocarbons (PAHs) are anthropogenic and naturally occurring chemicals that humans have exploited for many purposes. Two common sources of PAH-contaminated soil and groundwater are creosote and coal tars. Crcosotc is used in wood preservatives, and coal tars are waste products of coal or oil gasification. Many species of bacteria are known to biodegrade the more soluble PAHs, making in situ bioremediation an important technology for the cost-effective treatment of contaminated soils and groundwater.lTria1-and-error methods of imple- menting this complex process at a field-scale are inefficient and costly. Therefore, it is important to conduct studies of microbial reactions and to develop reliable engineering models that can analyze in situ options prior to field testing. Specifically, it is important to establish the appropriate chemical conditions required for biodegradation of the contaminants and to assess the relative importance of mass transfer (bioavailability) versus kinetic (biodegradation control) effects.

This paper addresses the importance of chemical conditions and mass transfer effects to in situ bioremediation of PAHs, particularly as they apply to a coal-tar- contaminated site presented as a case study. The site was formerly used as a manufactured gas plant (MGP), and is presently contaminated with mono- and polycyclic aromatic hydrocarbons. As part of an ongoing bioremediation feasibility study, an experiment was conducted to assess the ability of the sediment bacteria to mineralize I4C-labeled forms of primary coal tar constituents (benzene, naphthalene, and phenanthrene) in laboratory microcosms that simulate site conditions. Modeling and experimental studies were combined to determine the influence of mass transfer rate (desorption rate) on the biodegradation performance. The first half of this paper describes the experimental techniques and results of the biodegra-

This work was supported in part by Cooperative Agreement ECD-8907039 between the National Science Foundation and Montana State University and in part by the Hazardoub Substance Managemcnt Research Center and the New Jersey Commission on Science and Technology headquartered at the New Jersey Institute of Technology.

103

104 ANNALS NEW YORK ACADEMY OF SCIENCES

TABLE 1. Groundwater Quality at the Site of a Former Manufactured Gas Plant Analvte Range Tvuical Background

PH 3 to 7 6 6.3

Biological oxygen demand (mg/L) <12 to 110 25 Redox potential (mV) -192 to 470 -20 0

Chemical oxygen demand (mglL) <50 to 610 200 <50

BTEX (mg/L) ND to 43 1 ND Naphthalene (mg/L) ND to 49b 0.5 ND PAH-naphthalene' (mglL) ND to 1 0.05 ND Total organic carbon <0.5 to 43 10

Dissolved oxygen (mg/L) <0.1 to 5 0.5 - 2 d

Ammonia (mg NIL) <0.05 to 83 20 Fe(I1) (mg/L) 0.1 to 2350 50 16 Sulfate (mg/L) <5to7300 500 4 Carbon dioxide (mglL) 10 to700 200

Nitrate (mg/L) <0.5 to 4.6 <0.5 <0.5

ND = not detected. " Not measured.

Upper bound value exceeds solubility due to presence of NAPL in aqueous sample. EPA priority pollutant PAH minus naphthalene. Dissolved oxygen measured by probe; may read higher than Winkler titration method or

Micro. dissolved oxygen probe.

dation evaluation for the primary coal tar constituents dissolved in the groundwater at the site. In the second half of the paper we develop a theoretical model to describe the sorption-limited biodegradation that is characteristic of many hydrophobic aromatic compounds in sediment-water systems.

SITE DESCRIPTION

The MGP site is contaminated with coal tar and petroleum hydrocarbons including benzene, toluene, ethylbenzene, xylene, naphthalene, acenaphthylene, acenaph- thene, fluorene, phenanthrene, anthracene, fluoranthrene, and pyrene. Ranges for the aqueous concentration of these compounds, electron acceptors and other groundwater quality parameters measured in groundwater collected from various locations throughout the site are summarized in TABLE 1. Most of the contamination is contained within two sand and gravel aquifers at the 72 acre site (FIGS. 1 and 2). Sulfate and ferric iron are the primary electron acceptors available in the subsurface, with only low concentrations of oxygen andlor nitrate detected at most locations.

METHODS

Aquifer Material Sampling

Aquifer sediment samples were collected from five separate boreholes (FIGS. 1 and 2) in order to meet the following objectives: 1) to compare microbial activity from both contaminated and uncontaminated sediments; and (2) to investigate the

BOUWER et al.: PAH CONTAMINATION 105

spatial distribution of microbial activity across the site. Sediment cores were obtained with a split spoon sampler and mud-rotary drilling. Cores were aseptically subsampled with a truncated sterile 60 mL syringe. Rhodamine-WT dye was added to the mud (5 mglL) in order to trace mud intrusion into aquifer cores; sediments containing the dye as measured by fluorometry were assumed to be contaminated with nonindigenous bacteria and therefore were excluded from use in biological assays. Additional details on the sampling procedure are provided by Barranco et al.' and Durant et al.'

Microcosm Assay

In this experiment, aquifer sediments were combined with site groundwater to assay the ability of the aquifer bacteria to mineralize coal tar constituents in the absence of nutrient and electron acceptor amendments. The study entailed the addition of various radiolabeled (I") test compounds to batch sediment-water microcosms and the entrapment of I4CO2 (using KOH) produced from mineralization. l4C-benzene, I4C-naphthalene, I4C-phenanthrene, and 14C-fluorene were selected for investigation because they are common constituents in the coal-tar contaminated aquifer. The sediments were also evaluated for their ability to mineralize I4C-acetate and I4C-p-hydroxybenzoate. Acetate was of interest because it was pre- sumed that 14C-acetate mineralization would provide an effective means of screening for microbial activity. I4C-p-hydroxybenzoate was assayed to determine whether

I l100o 1 I I

/ 1000 u I I

I Ground

North _____)

Feet

0 400 7

FIGURE 1. Sampling borehole locations A, A', B, C, and D, groundwater flow directions, and naphthalene plume configurations (pg/L) in the upper aquifer formation (15-21 m). (From Durant et aL3)


North ____)

Feet

0 400 -

FIGURE 2. Sampling borehole locations A, A', B, C , and D, groundwater flow direction, and naphthalene plume configuration (pg/L) in the lower aquifer formation (24-28 m). (From Durant et aLJ)

mineralization of this putative intermediate of aromatic hydrocarbon degradation would indicate a presence of bacteria capable of degrading benzene and PAHs.

Each microcosm consisted of a 15 rnL serum bottle that contained 5 g of sediment and 3 mL of filter sterilized groundwater. Groundwater was obtained from monitoring wells adjacent to the sediment sampling locations and from well screen depths which corresponded with the sediment sample depth. Each microcosm received 0.07 pCi of radiolabeled and 1 ppm of unlabeled substrate. A sterile 2 mL glass vial containing 0.5 mL of 1 N KOH was placed inside each microcosm to capture the I4CO2 generated from the mineralization of the aromatic substrate. The microcosms were sealed with Teflon'"-faced, rubber-butyl septa and aluminum crimp caps. The microcosms were incubated at 22°C for 4 to 6 weeks. Anaerobic microcosms were prepared and incubated inside an anaerobic glove box (Coy) with an atmosphere consisting of 97% : 3% N, : H, . The majority of sediments were incubated under aerobic conditions because the monitoring well data corresponding to most of the borehole locations indicated a prevalence of microaerophilic conditions. Anaerobic microcosms were prepared for sediments where groundwater data indicated anaerobic conditions.

The microcosms were sampled using a syringe to extract and replenish the KOH in the I4CO2 trap through the septa which were then resealed with silicone cement. 14C activity was quantified by scintillation counting on a Beckman LS 3801 Liquid Scintillation Counter. For each sample and test compound, triplicate active microcosms were prepared as well as duplicate killed control microcosms (with 1 mL of 0.5 N HgCl, and 1 mL of 2 X N NaN, added to the microcosm supernatant). The difference in the amount of I4C trapped in KOH in the control microcosm and

ROUWEK er al.: PAH CONTAMINATION 107

the active microcosms represents the amount of mineralization of the test compound due to biotic processes. Biotic mineralization of the target compounds was assessed as the difference between mineralization observed in the average of the active microcosms less the average of the associated controls. The assay is a conservative cstimate of biodegradation capability because it only measures “COZ captured by the KOH, and does not quantify the portion of “C converted to cells and intermedi- ates. The oxygen concentrations were not measured in the experiments. However. stoichiometric calculations indicate that sufficient oxygen was available in the headspace to support the oxidation of thc target PAH substratcs.

KESULTS OF THE MICROCOSM ASSAY

The results of the mineralization assay suggest that bacteria capablc of aerobically biodegrading benzene and PAHs are prescnt at a variety of locations throughout the subsurface of the former MGP site (TABLE 2). Although the majority of sediments analyzed could not degrade these compounds during four-week incuba- tions, benzene-, naphthalene-, and phcnanthrcne-degrading bacteria wcrc dctcctcd at a variety of depths. Benzene mineralization rangcd from 4-42%, naphthalene mineralization ranged from 8-55%, and phenanthrene mineralization rangcd from 4-23% in the active microcosms (295% confidcncc level). No significant removal o f fluorene was observed. Thc most extensive biodegradation of benzene, naphthalene and phcnanthrcnc was observed in a sample from 3 m in borehole C. The in sitzc dissolved oxygcn concentration at this location and sample dcpth was among the highest measured for this site. High oxygen enhances the growth of bacteria as well as their ability to biodegrade aromatic ring compounds. The presence of oxygen and substrate allows for the induction of oxygenase enzymes which facilitate the breakdown of thc highly stablc bcnzcnc ring. The high levels of oxygen in this portion of the site probably encourage the proliferation of morc mctabolically activc microbial populations.

The most widespread mineralization of naphthalcnc was observed throughout boreholes A and A’. Relatively high levels of mineralization of p-hydroxybenzoate, a structural analogue of naphthalcnc biodegradation intcrmcdiatcs, was also observed. Both of these boreholes exhibited high viable and total cell counts,’ sug- gesting a link between biomass density and biodegradation capacity.

Two samples from 13.1 m and 22.2 m from borehole A‘ exhibited statistically significant naphthalene mineralization (8% 2 1 and 13% 2 2, respectively) under anacrobic conditions. Although the operative electron acceptor was not identified, significant levels of nitrate (13 mglL) and sulfate (24 mglL) were present in the groundwater used to prepare these microcosms. Aerobic acetate mineralization in these samples was very high (57% 2 4 and 52% 2 4, respectively) indicating the occurrence of facultative anaerobic bacteria. Other researchers have observed anaerobic biodegradation of

Statistically significant ( p < 0.05) I4C-acetate mineralization was observed in nearly half the sediments tested (TARLI: 2). The presence of bacteria capable of mineralizing “C-acetate. however, did not always correspond with the presence of bacteria capable of mineralizing the aromatic compounds. While ’“C-acetate appears to be an effective indicator of microbial activity in aquifer sediments, our results suggest that it cannot be used to screen for the presence of bacteria capable of mineralizing benzene and PAHs. Mineralization of p-hydroxybenzoate may be a more effective tool for screening sediments for this purpose, but our results were inconclusive.


TABLE 2. Microbial Mineralization of Test Compounds over Four Weeks

Site Depth (m) Type (standard deviation in parentheses) Sample Sediment Percent Mineralized

A

A'

B

C

2.7 8.8

14.9 18.0 18.0 MIC 21.0 21.0 MIC 24.7 24.7 MIC

2.7 5.8

13.1 13.1 AN 16.8 22.3 22.3 AN

1.1 13.7 14.9 18 18 MIC 21 22.2 22.2 MIC

3.3 12.5 16.7 AN 17.4 22.2 27.0

silty sand sandy gravel sand coarse sand

gravel

gravel

silty clay fine sand sandy gravel

sand gravel

sandy silt gravel sand coarse sand

gravel saprolite

sand silty clay clayey silt claycy silt sand gravel

p-Hydroxy

5 6

27

10 30 19 20 19

Fluorene 0 0

-

1(2) - - - -

p-Hydroxy

6

6 2 0 5 0 0

Benzene

-

42 0 0 3 0 - - . .

None of the samples from Borehole D exhibited significant mineralization, and were thercfore excluded from presentation in this table. AN = Anaerobic; MTC = microaerophilic; - = no data, p-Hydroxy = p-hydroxybenzoate, Naph = naphthalene, Phen = phenanthrene. p-Hydroxybenzoate was not assayed in A' and C, fluorene was only assayed in A', and benzene was only assayed in C.

The extent of mineralization for aromatic substrates was typically limited (<50%), and this observation may be due to a number of factors, including volatilization in the microcosm headspace (especially true for benzene), a poor availability of oxygen in the microcosm supernatant, and poor substrate bioavailability due to sorption to the aquifer solids (especially true for naphthalene and phenanthrene). Limited degradation of PAHs is common for sediment-water microcosm studies.6

BOUWER el al.: PAH CONTAMINATION 109

Hot methanol extractions of some randomly selected sediments from our mineralization assay confirm that a significant portion of the I4C-phenanthrene was sorbed, or otherwise not bioavailable, in the microcosms.' Ancillary experiments on aerobic 'T-naphthalene in the presence and absence of sediment have also observed sediment to significantly limit the extent of mineralization owing to sorption and reduced bioavailability?

The results of the intrinsic mineralization studies demonstrate that active microorganisms exist in the subsurface sediments at the site under both aerobic and anaerobic conditions. Some sediments contain bacteria capable of mineralization of PAHs, and in the case of naphthalene, biodegradation was observed under both aerobic and anaerobic conditions. This degradation was feasible without the use of nutrient (N and P) amendments. Surprisingly, the PAH biodegradation capacity of the sediments could not be correlated with PAH concentrations in groundwater collected proximal to the sediment sampling locations. Some sediments from contaminated zones failed to degrade any of the 14C-substrates during the four-week incubation. Significant naphthalene mineralization was observed in sediments from the upper and middle portions of borehole A', despite an absence of aromatic contamination in the in situ groundwater for these depths. This result suggests that these sediments were exposed to naphthalene previously, or that naphthalene- degrading bacteria from another contaminated zone were transported to the Bore- hole A' location. Either explanation is plausible considering that PAHs had been used at this site for most of this century.

INFLUENCE OF MASS TRANSFER ON BIOAVAILABILITY IN AQUIFER SEDIMENTS

Introduction to Bioavailability

The laboratory mineralization assay demonstrated that the indigenous site bacteria arc capablc of aerobic mineralization of benzene, toluene, naphthalene, and phenanthrene, and limited anaerobic mineralization of naphthalene. Data collected in this work and elsewhere,"-'* however, suggest that regardless of the chemical conditions (e.g., electron acceptor and nutrients), sediments can limit the extent of biodegradation of naphthalene and phenanthrene by sorption and reduced bioavailability. This is especially true in aquifers, where hydrophobic compounds are exposed to sediments for much longer periods than those examined in this work.

Prolonged exposure to sediments allows hydrophobic compounds such as naphthalene and phenanthrene to sorb strongly to intraparticle organic carbon. There- fore, the coal tar constituents at the site are likely less available for microbial utilization because they reside on the surface or in the pores of soil particles. The remainder of this paper describes a theoretical and experimental study that was conducted to assess the influence of mass transfer rate on the overall biodegradation performance.

There must be a close association between a microorganism and contaminant for biodegradation to occur. The contaminant must be available for uptake and utilization by the microorganism.'" Microorganisms and hydrophobic organic pollutants are distributed among the solid, liquid, and gas phases within the subsurface. In systems where a pure organic phase (NAPL) is present, bioavailability will be controlled, in part, by the rate at which compounds dissolve from the NAPL. In locations where NAPL is absent, sorptive processes will ultimately control bioavailability.


Many organic contaminants of interest tend Lo sorb onto soil such that only a small fraction of the compound may actually be in the bulk water phase. Over long contact time, the sorbing pollutants slowly diffuse into the inorganic and organic matrix and may also form bound residues. Most evidence indicates the uptake of compounds by bacteria proceeds via the liquid phase. Consequently, a process such as sorption or volatilization that reduces the solution concentration tends to reduce the biotransformation rate. Furthermore, the accumulation of contaminants in fis- sures and cavities within subsurface solids renders them inaccessible to microorganisms and their enzymes. These processes decrease the bioavailability. The important conclusion from the influence of mass transfer in terms of reduced bioavailability is that the overall reaction rate (in the absence of NAPL) is controlled by the desorption rate and not by the activity of the degrading microorganisms. The practical effect of such slow diffusion from within soil aggregates and other kinetic limitations to desorption is a decrease of the rate of removal of the contaminant, thereby increasing the time required to achieve clean-up.

Sorption of contaminants tends to separate the direct contact betwecn microorganisms and contaminants which is necessary for biodegradation to occur. For soils, sediments, and groundwater aquifers where the solid fraction and surface area are high, sorption is the primary process limiting bioavailability. Effects of sorption on biodegradation in the subsurface can be classified into concentration effects and desorption rate limitations as discussed below.

Concentration Effects

A direct impact of sorption on biodegradation is the reduction of organic coin- pounds in the bulk water phase. As sorbed chemicals are effectively protected from direct biodegradation, microbial growth must rely on organic substrate(s) in the bulk water phase. The rate of bulk water uptake and metabolism of the organic compound is generally given by the Monod relationship:

d C - kXC dt K , + C

where C is the bulk water concentration of the organic compound, t is time, k is the maximum rate of biodegradation, X is the concentration of microorganisms in water, and K , is the substrate concentration when the growth rate is at half the maximum. Sorption reduces the bulk water concentration and as a result will prolong the time to degrade a given amount of organic pollutant as compared with a system free of sediments or soil where sorption will not occur. At very low contaminant concentrations, often in the range of kg to ng per liter, insufficient energy and carbon may be available for microbial growth and maintenance. Rittmann and McCarty” defined a critical concentration (C,ll,n) at which microbial growth is just balanced by decay:

where Y is the biomass yield coefficient and b is the microbial decay coefficient. If sorption diminishes the concentration below C,,, , biodegradation will decrease or even stop with time because thcre will be net decay of biomass.

On the other hand, at high concentrations, many organic compounds become toxic to microorganisms. One expression for the biodegradation rate is the Haldane


equation which is given below:22

where Kiis the inhibition constant. Here sorption can reduce the bulk liquid concentration to lessen the toxic inhibition and increase microbial growth and biodegradation.

Desorption Rate Limitations

Owing to geometrical and mass transfer restrictions, most bacteria are present in the external surface of soil particles and in the bulk water. Only limited biological activities exist within the intraparticle porest4 Decontamination of soils and sediments may involve 3 steps: 1) dissolution from the organic phase, 2) desorption of previously sorbed contaminant, and 3) biotic or abiotic transformation of the contaminant in the aqueous phase. In the absence of NAPL, the apparent biodegradation rate in a solid-water system can be controlled either by the desorption rate or biodegradation rate.

The influence of mass transfer on biodegradation can be characterized by a single parameter

- = - dC BfkhC dt (4)

Here B,, termed the Bioavailability Factor, is used to account for the impact of exchange of the organic compound between soil and water, or sorption, on biodegradation rates; kh is the first-order rate coefficient for biodegradation. The use of kh assumes that the organism concentration changes little with time which is likely for large-scale subsurface contamination with low active organism concentrations and slow groundwater movement. The objective of the following analysis is to derive Bj as function of fundamental sorption and biodegradation mechanisms.

In a soil-water slurry, the mass balance for an organic compound can be writ- ten as:

d - dt [VC - ms] = - v&)

Here V is the volume of the bulk water, m is thc mass of the solid material, and S is the organic concentration in the solid phase. The left-hand side of this equation represents the change of organic compound with time; Ac, is the sink term which occurs only in the bulk water. If biodegradation is the only mechanism for removal of the organic compound from the system and is assumed to follow first-order kinetics, then


The ratio of m to V is termed the soil/water ratio, Rb,w

Substitution of Equations 6 and 7 into Equation 5 yields Equation 8 below:

The Equation 8 mass balance in a soil-water system can be rearranged to yield Equation 9.

The following expression results when Equation 9 is solved for dC/dt:

Bf in Equation 4 is thus defined as:

The B, can be quantified by selecting an explicit expression for dS/dC. There are two possible cases: i) fast sorptionldesorption and ii) slow sorptionldesorption.

i) Fast sorption. As sorption is fast compared to biodegradation, sorption will be at equilibrium so that the following equation applies:

dS dC - _ - Kd

K,, is the soil-water distribution coefficient assuming a linear sorption isotherm. The corresponding B, using Equation 11 becomes:

1 1 + &R,,

B -

Here Bfis analogous to the retardation factor used to model movement of a sorbing solute in porous media.I5

ii) Slow sorption. When the rate of biodegradation is much faster than the sorption rate, microorganisms can effectively degrade available organic compound as soon as it desorbs from the solid phase. Consequently, the rate of biodegradation is approximately equal to the rate of desorption:

dS dt m - = VkbC

BOUWER et aL: PAH CONTAMINATION 113

One approximation for the mass exchange between soil and water (sorptionldesorption) is a first-order reaction:

Here k,,, is the first-order sorption rate coefficient. The “driving force” for sorption is (KdC - S) in terms of the solid phase concentration. Combining Equations 14 and 15, the following expression results for dS/dC:

The corresponding B, for slow sorption (substitution of Equation 16 into Equation 11) becomes:

The four independent fundamental parameters, khr k,, K d , and R,,,, can be combined into a dimensionless parameter, Q, which is termed the Thiele Modulus:

The expression for B, incorporating the Thiele Modulus becomes:

1 B - - 1 + Kd Rsiw( 1 + a’)

@ is a parameter for the kinetic effect of sorption on biodegradation, an analogy to the widely used parameter in chemical engineering for the effect of diffusion (mass transfer) on catalytic reactions in porous media.’”lg The Thiele Modulus (a) can be rearranged as shown in the following expression to reflect the ratio of characteristics times for sorption and biodegradation:

(21) Characteristic Time for Sorption

Characteristic Time for Biodegradation

The practical value of the B, model stems from its simplicity and ease of use. The B, provides an estimate of the impact of sorption on biodegradation to be used as a substitute for extensive numerical calculations. Therefore, the overall rate of biodegradation in a soil-water slurry is determined by the two factors: B, and kh (Equation 4). Bi is evaluated from the extent and rate of sorption; kb is determined by the metabolic capability of microorganisms and environmental factors such as


temperature. As applied here, the computation of kb assumes that an excess of electron acceptor is present (k, biodegradation is not limited by electron acceptor availability). The time to reduce the bulk water concentration in half is given by:

A Bf value of 1.0 means there is no influence of sorption on the biodegradation, and this situation corresponds to a well-mixed liquid culture system without sorbent. As the Bfdecreases below 1.0, biodegradation will be significantly limited by sorption and the persistence of the chemical increases.

The relationship between the Thiele Modulus (@) and the value of Br as a function of the KC1 for the organic contaminant is illustrated in FIGURE 3. The curves in FIGURE 3 were obtained using Equation 19 and R,q,w = 5 kg/L. Measured Kd values for many organic compounds of interest in the subsurface are greater than 0.1 L/kg.20 From the curves in FIGURE 3, Kd values >0.5 L/kg will yield B, values that are less than 0.3 for biodegradation in the subsurface. Consequently, organic compounds which can be readily biodegraded in liquid culture with half-lives ranging from a few hours to a few days (large k b ) can be very persistent in soils and groundwater aquifers (curves with B, < 0.3 in FIG. 3). The Thiele Modulus (@) incorporates the effect of desorption rate on the biodegradation rate. As @increases, the B, decreases which indicates that the overall biodegradation rate will be kinet- ically limited by the rate of desorption. Low bioavailability (small B,) can be a major factor responsible for slow in situ bioremediation.

Application of Bioavailability Model to MGP Site Aquifer Sediments

The modeling concepts described above were applied to estimate the extent of mass transfer control on biodegradation of naphthalene and phenanthrene for

0:1 i i o

Thiele Modulus (Q)

FIGURE 3. Influence of Thiele Modulus (a) and soil-water distribution coefficient (Kd) on the Bioavailability Factor (B,) . The curves were calculated from Equation 19 with RT,% = 5 kgIL.

BOUWER er al.: PAH CONTAMINATION 115

TABLE 3. Calculated Values for the Thiele Modulus (@) and B f for Naphthalene and Phenanthrene in the Presence of MGP Site Aquifer Sediments

~~ ~ ~

Compound R,,,, kg/L Q, B/ Naphthatene“ 5 0.94 0.28 Phenanthreneb 5 0.64 0.11

Measured parameters for naphthalene: K,, = 0.27 L/kg; k , = 0.0281 h ’; kh = 0.0337 h-’. ” Measured parameters for phenanthrene: K , = 1.19 L/kg; k,,, = 0.0103 h-I; kb = 0.0251 h-l.

“average” sediments at the MGP site. Laboratory studies were conducted with composite MGP site sediment samples to evaluate the intrinsic biodegradation rates (kh) , the sorption partition coefficients (K,,), and thc sorption rates (k,) for naphthalene and phenanthrene. The organic carbon content for the sediment composite was 0.004%. Details of the sediment characteristics and batch methods to measure k b , K d , and k , are presented elsewhere.*’ From the values of k,, , Kd. and k , determined by Zhang,2l the calculated values of the Thiele Modulus (@) and B, for naphthalene and phenanthrene appear in TABLE 3. The Bf values in TABLE 3 illustrate the importance of contaminant sorption. Given this model and composite sediment used, the aquifer sediments at the MGP site will slow naphthalene biodegradation by about 3.6 times (l/Bf = 1/0.28) and phenanthrene by slightly less than 10 times (1/B, = l / O . l l ) compared to biodegradation rates in the absence of aquifer sediment.

CONCLUDING REMARKS

Aquifer sediments recovered from a variety of locations at a coal-tar contaminated site contain bacteria capable of aerobically mineralizing benzene, toluene, naphthalene, and/or phenanthrene without an acclimation period. These bacteria appear to be distributed independent of depth in the aquifer, and occur both near and distal of contaminant sources. Acetate mineralization was an effective screening tool for aerobic and anaerobic microbial activity in the sediments, but could not be correlated to the presence of bacteria capable of degrading aromatic hydrocarbons. Limited anaerobic biodegradation of naphthalene was observed in certain sediments, but additional work is required to identify the electron acceptor in this reaction.

The results of mineralization experiments suggest that enhanced aerobic bioremediation would be an effective means of reducing the contaminant mass at the MGP site. Our work confirms that the aromatic hydrocarbon-degrading bacteria are present, and that in situ biodegradation is likely occurring. Empirical data and numerical simulations suggest, however, that the rate of bioremediation will be limited by contaminant sorption to the aquifer sediments. Modeling analyses (@ and Bf) and batch experiments demonstrated that naphthalene and phenanthrene exhibit reduced extents and rates of biodegradation in the presence of aquifer sediments. Consequently, the mass transfer rate becomes an important control over the biodegradation rate. The practical effect of such slow diffusion from within soil aggregates and other kinetic limitations to desorption is a decrease of the rate of removal of the contaminant, thereby increasing the time required to achieve clean- up and the amount of chemicals that must be supplied to sustain andlor enhance microbial activity.


SUMMARY

The importance of chemical conditions and mass transfer effects to in situ bioremediation of PAHs is presented using a case study. In situ bioremediation is being evaluated as a means for remediating a coal-tar contaminated aquifcr at the site of a former manufactured gas plant. Two objectives of this work have been to evaluate the potential for the indigenous bacteria to biodegrade coal tar constituents and to identify factors controlling biodegradation rates. Aquifer sediments collected from a variety of locations across the site contain bacteria capable of aerobically mineralizing some of the principal aromatic compounds in the groundwater plume (benzene, naphthalene, and phenanthrene). Parallel mineralization assays incubated under aerobic and anaerobic conditions strongly suggest that O2 availability is a primary factor controlling the rate and extent of biodegradation. Data indicate that sorption may have also significantly affected biodegradation rates by limiting the bioavailability of the aromatic compounds. A mass transfer-limited numerical model was developed to explore the effect of sorption and bioavailability on biodegradation rates. In this model biodegradation rates are proportional to aqueous concentration, which is directly reduced by sorption. Both biotransformation and bacterial growth are described as being controlled by the rate of desorptive mass transfer. The influence of sorption on biodegradation is quantified by defining a Bioavailability Factor, B,. A Thiele Modulus which indicates the ratio of characteristic times for sorption and biodegradation is helpful for determining the extent of mass transfer control during biodegradation of the aromatic compounds. This approach is pre- ferred to equilibrium partitioning models, which may overestimate biodegradation rates by failing to consider the effect of rate-limited desorption on bioavailability.

1.

2.

6.

7.

8.

9.

REFERENCES

NATIONAL RESEARCH COUNCIL. 1993. In Situ Bioremediation: When Docs It Work? National Academy Press. Washington, DC.

BARRANCO, F. T., J. L. KOCORNIK, I. D. MACFARLANE, N. D. DUKANT & L. P. WILSON. 1994. Subsurface sampling techniques used for a microbiological investigation. In Bioremediation of Chlorinated and Polycyclic Aromatic Hydrocarbon Compounds. R. E. Hinchee, A. Leeson, L. Semprini & S. K. Ong, Eds.: 474-478. Lewis. Ann Arbor, MI.

DURANT, N. D., L. P. WILSON & E. J. BOUWER. 1995. Microcosm studies of subsurface PAH-degrading bacteria from a former manufactured gas plant. J. Contaminant Hydrol. 17: 213-237.

MIHELCIC, J. R. & R. G. LUTHY. 1988. Degradation of polycyclic aromatic hydrocarbons under various redox conditions in soil-water systems. ADPI. Environ. Microbiol. 54: _ _ 1182-1 187.

GRBIC-CALK D.. S. M. HENRY. E. M. GODSY. E. EDWARDS & K. P. MAYER. 1991. Anaerobic degradation of aromatic hydrocarbons and aerobic degradation of trichloro- ethylene by subsurface microorganisms. In Organic Substances and Sediments in Water. Vol. 3. R. A. Baker, Ed.: 239-266. Lewis. Chelsea, MI.

MADSEN, E. L., J. L. SINCLAIR & W. C. GHIORSE. 1991. In situ biodegradation: Microbio- logical patterns in a contaminated aquifer. Science 252: 830-833.

DURANT, N. D. 1993. Screening for natural subsurface biotransformation of coal tar constituents at a former manufactured gas plant. Master’s Thesis, The Johns Hopkins University, Baltimore, MD. pp. 96-97.

ZHANG, W., E. BOUWER, L. WILSON & N. DURANT. 1995. Biotransformation of aromatic hydrocarbons in subsurface biofilms. Water Sci. Technol. 31: 1-14.

ROBINSON, K. G., W. S. FARMER & J. T. NOVAK. 1990. Availability of sorbed toluene in soils for biodegradation by acclimated bacteria. Water Rcs. 24 345-350.


10.

1 1 .

12.

13.

14.

15.

16.

17. 18.

19.

20.

21.

22.

MIHELCIC, J. R., D. R. LUEKING, R. MITZELL & J. M. STAPLETON. 1993. Bioavailability

ALEXANDER, M. 1995. How toxic are toxic chemicals in soil? Environ. Sci. Technol.

BURGOS, W. D., J . T. NOVAK & D. F. BERRY. 1996. Reversible sorption and irreversible binding of naphthalene and a-naphthol to soil: Elucidation of processes. Environ. Sci. Technol. 30: 1205-1211.

RITTMANN, B. E. & P. L. MCCARTY. 1980. Model of steady-state biofilm kinetics. Biotech- nol. Bioeng. 22: 2343-2357.

J O N ~ S , W. L., J. D. DOCKEKY, C. R. VOCEL & P. J. STLJRMAN. 1993. Diffusion and reaction within porous packing media: A phenomenological model. Biotechnol. Bioeng. 41:

FREEZE, R. A. & J. A. CHERRY. 1979. Groundwater. Prentice-Hall, Inc. Englewood

SATTERFIELU, c . N. 1070. Mass Transfer in Heterogeneous Catalysis, M.1.T. Press. Cam-

LEVENSPIEL, 0. 1972. Chemical Reaction Engineering, 2nd Edit. Wiley. New York. ARIS, R. 1975. The Mathematical Theory of Diffusion and Reaction in Permeable Cataly-

sis. Clarendon Press. Oxford, UK. RUTHVEN. D. M. 1984. Principles of Adsorption and Adsorption Processes. John Wiley &

Sons. New York. BALL, W. P. & P. V. ROBERTS. 1991, Long-term sorption of halogenated organic chemicals

by aquifer materials: I. Equilibrium Studies. Environ. Sci. Technol. 25: 1223-1237. ZHANG, W. 1995. Effect of Sorption on Bioavailability of Hydrophobic Organic Contami-

nants: Experimental and Model Studies, Ph.D. Dissertation, Department of Geography and Environmental Engineering, The Johns Hopkins University, Baltimore, MD.

ANDREWS, J. F. 1968. A mathematical model for the continuous culture of microorganisms utilizing inhibitory substrates. Biotechnol. Bioeng. 10 707-723.

of sorbed- and separate-phase chemicals. Biodegradation 4: 141-154.

29: 2113-2717.

947-956.

Cliffs, NJ.

bridge, MA.

biotreatment of pah-contaminated soils/sedimentsa

Documents