Assessing the Bioavailability of PAHs in Field ... - ACS Publications

Feb 7, 2004 - University of Illinois at Chicago, Chicago, Illinois 60607, and United States Environmental Protection Agency,. Cincinnati, Ohio 45268...
1 downloads 0 Views 207KB Size
Environ. Sci. Technol. 2004, 38, 1786-1793

Assessing the Bioavailability of PAHs in Field-Contaminated Sediment Using XAD-2 Assisted Desorption L I L E I , † M A K R A M T . S U I D A N , * ,† AMID P. KHODADOUST,‡ AND HENRY H. TABAK§ University of Cincinnati, Cincinnati, Ohio 45221-0071, University of Illinois at Chicago, Chicago, Illinois 60607, and United States Environmental Protection Agency, Cincinnati, Ohio 45268

An XAD-2 assisted desorption assay was evaluated to assess its functionality in determining the bioavailability of polycyclic aromatic hydrocarbons (PAHs) in an aged fieldcontaminated sediment. In the study, various dosages of XAD-2 resin were added to abiotic sediment-water slurry systems to adsorb the PAHs from the aqueous phase thus accelerating the desorbability of these contaminants from the sediment. A parallel experiment on the biodegradation of these PAHs by microorganisms indigenous to the sediment was also conducted. Both the desorbability of the PAHs in the XAD-2 assisted desorption assay and their biodegradability decreased with time and eventually approached constant values. The two procedures showed very similar residual concentrations of PAHs for compounds with less than five benzene rings. This suggests that the XAD-2 assisted desorption assay shows promise in measuring the bioavailability of PAHs in field-contaminated sediments and could be used for predicting the end point of PAH bioremediation.

Introduction Polycyclic aromatic hydrocarbons (PAHs) are contaminants often found at petroleum-impacted sites. Their potency as carcinogens has caused increased environmental concern. A total of 13 unsubstituted PAHs have been regulated as priority pollutants under the 1977 Clean Water Act amendments (1), while past disposal of those PAHs has been managed under the Comprehensive Environmental Response, Compensation and Liability Act of 1980 (CERCLA). As a promising remediation technology for PAH contamination, bioremediation features a relatively low cost and minimal disturbance to the environment. In aged contamination cases, many studies have confirmed that the degradation rate of organic contaminants, including PAHs, often decreases with time, and the degradation eventually ceases with a considerable residual concentration of contaminants left (2-4). This phenomenon involves the bioavailability of the contaminant, which can be defined as the extent the contaminant can be readily accessible to microorganisms * Corresponding author phone: (513)556-2599; fax: (513)556-7277; e-mail: [email protected]. † University of Cincinnati. ‡ University of Illinois at Chicago. § U.S. Environmental Protection Agency. 1786

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 6, 2004

for biodegradation to occur (5). In contaminated sediments, the bioavailability of PAHs to microorganisms is believed to be a major constraint limiting the extent of remediation (6). Therefore, determination of the bioavailable fraction of PAHs at a contaminated site helps in predicting the potential for bioremediation. It can also provide a procedure for assessing the actual environmental impact of PAH contamination. Studies of the mineralization kinetics of the organic contaminants associated with solid phases could improve the understanding of bioavailability of those contaminants to some degree. Guerin and Boyd (7) reported that the extent of degradation of naphthalene varies when it is sorbed to various natural or synthetic sorbents and different naphthalene-degrading bacteria are used. This approach, however, sheds information on bioavailability that is limited to the abilities of the available microbial culture in biodegrading PAH compounds and does not provide a measure for true exposure. Nonvigorous extraction has also been reported to reflect the bioavailability of hydrophobic organic contaminants (HOCs) to some extent (8-10). However, a careful study on solvent selection, the degree of agitation, and duration of contact may be necessary for each contaminant and geosorbent (soils or sediments) combination. In real world contamination scenarios, PAHs with various numbers of benzene rings coexist with each other under the impact of indigenous microbial consortia that contain numerous species. With these constrains in mind, this research was undertaken to develop an alternative method using nonionic polymeric adsorbent XAD-2 to assess the bioavailability of PAHs in field-contaminated sediments. Amberlite XAD resins are widely used in monitoring and characterizing organic matters by adsorbing organic matters mainly through hydrophobic bonding, although the exact mechanism of adsorption is unknown. Various XAD resins have been used for the characterization of natural organic matter (11, 12), for differentiation between hydrophobic and hydrophilic fractions in activated sludge exopolymeric substances (13), and for preconcentration of pesticides (14) and estrogens in the effluents of sewage treatment plants (15). Among different types of XAD resins, the styrenedivinylbenzene XAD-2 resin has been used for the isolation of humic substances (16) and preconcentrating trace amount of PAH in seawater (17) because of its strong affinity to aromatic compounds. In our study, various amounts of XAD-2 resin were added to abiotic sediment-water slurries to accelerate desorption of PAHs from the sediment phase, making a prompt quantification of the desorbable PAH fraction in the sediment possible. Measurements were made of XAD-2 mediated desorption of PAHs from the contaminated sediment. Also determined were PAH removal efficiencies via aerobic biodegradation in laboratory slurry systems using microorganisms indigenous to the sediment. The residual level of PAHs determined from both the abiotic desorption study and the biological degradation study was correlated.

Experimental Procedures Sediment Samples. The sediment used in the research was dredged from the East River, NY, near Ricker’s Island. The overlaying water at the sampling site was also collected for preparing sediment-water slurries. Because of the vicinity of the sampling site to an estuarine area and the backflush of seawater, the water used in the experiments shares many characteristics of seawater, such as the high content of chloride and a comparable sulfate level (approximately 21 mM) (18). Pebbles, shells, and vegetable matter were removed 10.1021/es030643p CCC: $27.50

 2004 American Chemical Society Published on Web 02/07/2004

TABLE 1. Initial PAH Concentrations in the East River Sediment in Desorption Study PAHs

no. of rings

concentrations (mg/kg)

naphthalene 2-methyl naphthalene

2-ring

18.15 21.70

acenaphthylene acenaphthene dibenzofuran fluorine phenanthrene anthracene

3-ring

8.67 53.28 7.41 10.05 76.86 57.45

fluoranthene pyrene benzo[a]anthracene chrysene

4-ring

83.30 141.51 69.99 51.93

benz[b+k]fluoranthene benzo[e]pyrene benzo[a]pyrene

5-ring

62.49 21.86 54.59

indeno[1,2,3-c,d]pyrene benzo[g,h,i]perylene

6-ring

19.59 18.37

from the sediments before storing them in sealed plastic carboys at 4 °C. The sediment is known for chronic PAH contamination, although the origin of the contamination is unknown and probably due to multiple sources. It contains 32.1% sand, 62.9% silt, and 5% clay and thus has the silt loam texture according to the soil classification systems developed by the U.S. Department of Agriculture. Its organic carbon content (w/w) and cation exchange capacity are 6% and 37.9 meq/100 g, respectively, which would have resulted in strong adsorption and sequestration of PAHs into the sediment during the long aging period in the field. The sediment characterization data at the beginning of the XAD-2 assisted desorption study are presented in Table 1. Biodegradation Study. Slurries containing 10 g of East River sediment and 50 mL of overlaying water were prepared in 125 mL serum bottles. The sediment here served as both the inoculum and the medium containing substrates. A total of 2 g of crushed limestone was added to each bottle to maintain a pH within the range of 6-7.5 for the duration of the experiment. The bottles were prepared inside a hood where an atmosphere of approximately 75% O2 and 25% N2 was maintained. At the end of the first day of the experiment, the bottles were reopened in the hood and exposed to the same atmosphere again. This was necessary to satisfy the initial oxygen demand attributable to the high sulfide content in the sediment and to ensure sufficient oxygen for PAH degradation throughout the experiment. Once sealed with Teflon-coated butyl rubber stoppers and aluminum crimpseals, the bottles were continuously mixed in a rotating tumbler at room temperature. Killed controls to which NaN3 and Na2MoO4 were added were prepared to monitor abiotic losses of PAHs. XAD-2 Assisted Desorption Study. A polystyrene resin Amberlite XAD-2 (Supelco, Bellefonte, PA) was used as the sorbent, which has a particle size of 20-60 mesh, a surface area of 300 m2/g, and a pore diameter of 90 nm. XAD-2 mediated desorption systems were prepared with abiotic slurries containing 6 g of sediment and 120 mL of overlaying water, where NaN3, Na2MoO4, and HgCl2 were used to suppress biological activities. To ensure that the XAD-2 amount used had enough adsorption capacity, 6, 12, 24, or 48 g of XAD-2 were added respectively to the bottles, corresponding to a XAD-2 to sediment ratio (w/w) of 1, 2, 4, and 8. The XAD-2 was air-dried prior to use. A mixture of KH2PO4 and NaOH was added as a buffer to control the pH between 6 and 7.5. Crushed limestone was not used for pH

control because of the possible interference with the adsorption process due to the large surface area of the limestone power. Bottles were loaded into a rotating tumbler for mixing at room temperature. Controls with no XAD-2 amendment were also used to assess PAH losses due to processes other than desorption. PAH Quantification. For both the biodegradation and XAD-2 assisted desorption studies, triplicate samples were removed periodically and analyzed for PAHs in the sediment and liquid phases. In the XAD-2 assisted desorption study, sediment was separated from XAD-2 resin by centrifuging the slurry at 5000 rpm for 30 min and then harvesting XAD-2 from the top while the sediment settled to the bottom of the centrifuge tube. Although the specific density of XAD-2 resin (1.02 g/mL) is close to that of water, the XAD-2 resin, as applied, floated on the top of slurries due to the hydrophobic property of its surface. PAHs were extracted using a mixture of 1:1 methanol and dichloromethane for 24 h with continuously mixing. The extract was quantified using an HP 5890 Series II Gas Chromatography (GC) (Hewlett-Packard, Palo Alto, CA), which is equipped with a flame ionization detector (FID), a 7673 automatic sampler, and a 30 m long DB-5 fused capillary column (J&W, Folsom, CA) with a film thickness of 0.5 µm and a internal diameter of 0.53 mm. The column temperature started at 100 οC for 1 min, increased to 280 °C at 4 °C/min, and then increased to 300 °C at 24 °C/min and was held at 300 °C for 12 min. The GC/FID quantification procedure used was similar to U.S. EPA 8100 method for PAH analysis (19). The external standard method was selected for quantification, and 2-fluorobiphenyl was used as the surrogate and showed a recovery of 70-80%. Mass Spectroscopic analysis of the chromatography peaks was also performed to confirm the GC results on individual PAH compounds. Most of the triplicate samples that were extracted and quantified as to PAH concentrations showed standard deviations within 10%. Because of the insignificance of PAHs in the liquid phase (less than 3% as compared with that in the sediment phase even for naphthalene, the PAH with the highest water solubility), a decrease in PAH concentrations in the sediment was used as the measure of the mass desorbed in the desorption study, and the mass degraded by indigenous bacteria in the biodegradation study. Experiments were considered complete when the concentrations of PAHs reached constant values between two consecutive readings (i.e., varied randomly by less than 10%).

Results When sufficient oxygen was present, considerable degradation of all 2-, 3-, and 4-ring PAHs was observed, while 5-ring PAHs exhibited only limited degradation. The 6-ring PAHs, on the other hand, did not degrade at all. The concentrations of the PAHs in the killed controls remained essentially unchanged throughout the experiment, except that a few of the lower ring PAHs, such as naphthalene and 2-methylnaphthelene, showed a slight decrease in their concentrations. These exceptions could be attributed to the relative high water solubility and volatility of these compounds. As shown in Figure 1, the degradation of 2- and 3-ring PAHs started immediately and leveled off by week 2, while the majority of degradation of 4- and 5-ring compounds continued till week 9. It took 24 weeks for all PAHs to reach a constant residual level in the sediment. Figure 1 also indicates that compounds with simpler structures and lower molecular weights generally reached lower residual concentrations than higher ring PAHs, if they were initially presented in the sediment at comparable concentrations. Figures 2-6 present comparisons of the degradation and desorption time courses for representative 2-6-ring PAHs, which include the 2-ring compound naphthalene; 3-ring VOL. 38, NO. 6, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

1787

FIGURE 1. PAH concentration time courses in aerobic degradation.

FIGURE 2. Comparison of the biodegradation time courses (a) and desorption time courses (b) of 2-ring PAH naphthalene (I) and 3-ring PAH acenaphthene (II). compound acenaphthene; 4-ring compounds fluoranthene, pyrene, benzo[a]anthracene, and chrysene; 5-ring compounds benzo[b+k]fluoranthene, benzo[a]pyrene, and benzo[e]pyrene; and 6-ring compound benzo[g,h,i]pyrelene. As shown in Figure 2(I), the controls for naphthalene in both the biodegradation and the desorption experiments decreased slightly, which was probably due to the volatiliza1788

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 6, 2004

tion of naphthalene as discussed earlier. However, in terms of naphthalene being able to come off the solid phase, use of its residual level in the sediment is still valid in assessing its bioavailability. In both experiments, naphthalene reached a residual level of about 15 mg/kg, although it is a lower ring PAH and supposedly degrades and desorbs easily. This may be attributed to the fact that during the aging process, a large

FIGURE 3. Comparison of the biodegradation time courses (a) and desorption time courses (b) of 4-ring compounds fluoranthene (I) and pyrene (II). fraction of naphthalene that was originally on the sediment may have undergone biodegradation anaerobically, leaving what was left mainly irreversibly adsorbed to the sediment. This observation suggests that vigorous extraction alone could not reflect bioavailable amounts of PAHs in aged contamination situation. As shown in Figures 2(II) -6, the concentration of all other PAHs in the abiotic controls remained essentially unchanged in both the biodegradation and the desorption experiments, indicating minimal losses other than their degradation or desorption, respectively. The essentially constant concentration levels of PAHs in desorption controls proved that spontaneous desorption is extremely slow for most PAHs and that XAD-2 is very effective in accelerating the process. Figure 2(II) shows the typical pattern of biodegradation and desorption for most of the 3- and 4-ring compounds that exhibited significant biodegradation. Shown in Figure 2(II) is the biodegradation of acenaphthene, which proceeded rapidly at the beginning in the aerobic slurries and then ceased with a considerable residual concentration on the sediment of approximately 16 mg/kg. This observation is in agreement with the many findings reported for other aged contaminations. The desorption of acenaphthene reached equilibrium within one week, while biodegradation proceeded over nine weeks. The desorption profiles corresponding to different doses of XAD-2 amendments collapsed together after relatively short periods of time, suggesting that all XAD-2 doses were sufficient to provide the required adsorption capacity. Final residual levels of the PAH in both studies were very comparable. Similar characteristics were evident in the biodegradation and desorption time courses of the 4-ring PAHs fluoranthene, pyrene, benzo[a]anthracene, and chrysene as shown in Figures 3 and 4.

In the XAD-2 assisted desorption study, it was also noticed that lower ring PAHs generally reached desorption equilibrium faster than higher ring compounds, similar to what was observed in the biodegradation study. For all the XAD-2 to sediment ratios used, the desorption rate of all 2- and 3-ring PAHs and most 4-ring PAHs approached zero in 2 weeks, while Chrysen (4-ring) and all the 5- and 6-ring PAHs continued to slowly desorb from the sediment until week 8 in systems with XAD-2 to sediment ratios less than 8. This could be attributed to the higher affinity of higher ring PAHs to the sediment. In Figure 4(II), it is noticeable that the larger ratios of XAD-2 amendment resulted in a faster initial desorption rate of chrysene, although all systems with various XAD-2 additions reached approximately the same residual levels by week 8. Similar patterns were observed for all the 5- and 6-ring PAHs, shown in Figures 5 and 6. While all the XAD-2 dosages provided sufficient adsorption capacity, higher XAD-2 to sediment ratios did accelerate the rate of desorption of these higher ring PAHs but made no difference with the lower ring ones. With all XAD-2 additions, the concentrations of these higher ring compounds in the aqueous phase were essentially zeros throughout the experiments, making the driving force for their desorption from the sediment approximately the same. Therefore, in the coupled processes of PAH desorption from sediment and adsorption to XAD-2, it probably was the adsorption that limited the process and resulted in slower desorption of those higher ring compounds from the sediment. This could suggest the preferential adsorption of lower ring PAHs to the XAD-2 resin when XAD-2 was not present in such abundance that easily accessible adsorption sites were available for all PAHs at the same time. In systems with less XAD-2 amendments, higher ring PAHs had to utilize the sites left by lower ring ones, which may not VOL. 38, NO. 6, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

1789

FIGURE 4. Comparison of the biodegradation time courses (a) and desorption time courses (b) of 4-ring compounds benzo[a]anthracene (I) and chrysene (II). be as easily available for adsorption and thus resulted in lower adsorption rates of these compounds. However, the adsorption capacity provided by various XAD-2 amendments probably exceeded the capacity required to obtain desorption equilibrium for all PAHs, which explains why the same residual levels of PAHs were measured for all the XAD-2 to sediment ratios used. As shown in Figure 5(I), the degradation rate of benzo[b+k]fluoranthene was much slower when compared with that of lower ring PAH compounds. Its desorption rate, on the other hand, was still considerably fast as it was able to reach equilibrium by week 8. While benzo[b+k]fluoranthene showed only limited degradation with a residual level of about 50 mg/kg in the biodegradation experiment, much more was desorbed from the sediment in the desorption study reaching a final residual level on the sediment of about 35 mg/kg. This discrepancy indicates that the desorbable amount of the compound could only be partially degraded by the indigenous microbial community under the experimental conditions practiced. Similar observations were obtained for another 5-ring compound benzo[a]pyrene, as shown in Figure 6(I). Shown in Figures 5(II) and 6(II) are the results of the biodegradation and desorption experiments for the 5-ring PAH benzo[e]pyrene and the 6-ring PAH benzo[g,h,i]pyrelene, respectively. No appreciable biodegradation was observed for either compound, although they still showed considerable XAD-2 facilitated desorption. This suggests that these compounds might be bioavailable in terms of their accessibility to the aqueous phase but are not susceptible to biodegradation under the experimental conditions followed. On the basis of the observed experimental results, residual concentrations of 2-4-ring PAHs in the sediment phase obtained from the biodegradation and the XAD-2 assisted 1790

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 6, 2004

desortpion studies were comparable. In other words, the XAD-2 assisted desorption study seemed promising in assessing the availability of those compounds. These PAHs are reportedly readily biodegradable as was exhibited in the experiments conducted in this study. On the other hand, 5and 6-ring PAHs showed only limited or negligible degrees of biodegradation, while considerable levels of desorption were measured for these compounds. This is in agreement with the fact that the bioremediation of those higher ring PAHs is often limited by their resistance to degradation rather than their bioavailability. To better evaluate the XAD-2 assisted desorption as a bioavailability assay for PAHs whose bioavailability plays a determining role in their bioremediation, the residual sediment phase concentrations of all the 2-, 3-, and 4-ring PAHs in biodegradation and desorption experiments were correlated as shown in Figure 7. Since various XAD-2 doses resulted in essentially the same extents of desorption for all the PAH compounds, the results in the system with 24 g of XAD-2 addition are used in preparing Figure 7. As shown in Figure 7, all the data points are closely distributed around the 45° line, and their linear regression using a zero-intercept fit has a slope of 1.04 and a square of the correlation coefficient (R2) of 0.8753. That is, the measures of desorption and degradation are closely correlated to each other and essentially identical. Since the desorption process, as described herein, is much faster and much easier to control than the biodegradation experiment, it can be very useful in predicting how much of a PAH contamination is bioavailable.

Discussion The study was designed to investigate the feasibility of using XAD-2 mediated desorption to assess the bioavailability of

FIGURE 5. Comparison of the biodegradation time courses (a) and desorption time courses (b) of 5-ring compounds benzo[b+k]fluoranthene (I) and benzo[e]pyrene (II). mixtures of PAHs in an aged contaminated sediment by correlating the mass of PAHs adsorbed onto XAD-2 and the amount degraded by indigenous microorganisms. There are two necessary preconditions that make the desorption study of simplified abiotic systems possible for assessing PAH bioavailability. First, the influence on bioavailability exerted by microorganisms should be negligible. Second, the slow desorption or release of contaminants from porous media should be the degradation-limiting factor in the coupled desorption/degradation processes. The first precondition is very likely true in our study since indigenous sediment microorganisms are typically slowgrowing organisms with relatively low activity. The characteristics of the microbial community affect the correlation between the mass of desorbable and the mass of degradable PAHs in several ways. If the indigenous bacteria were present in the sediment in a larger quantity, the biodegradation time course would reach the steady level of PAH residuals sooner. It was still the bioavailable amount of PAHs that determines the end point of biodegradation. However, the correlation between the degradation potential and the measurable desorption of PAHs would change with changes in bacterial characteristics other than quantity. For example, the presence of bacteria that are capable of secreting biological surfactants probably would result in increases in the extent of biodegradation that can then exceed the extent of desorption. When bacteria more capable of degrading higher ring PAHs are used, the amounts measured with XAD-2 assisted desorption could possibly correlate better with the biodegradation potentials for those higher ring compounds. The second precondition is also valid for most PAHs with less than five rings, whose degradation has been reported extensively. The experimental results showed that for all 2-,

3-, and 4-ring PAHs, a close approximation of the end point of their bioremediation could be obtained by measuring their desorbable amounts. For PAHs with higher benzene ring numbers, their recalcitrance to biodegradation may become the bioavailability-controlling factor. So XAD-2 assisted desorption tended to overestimate the extent of the biodegradation of some 5- and 6-ring compounds as observed. Therefore, XAD-2 assisted desorption seems to be a good measure of the true exposure that a contaminant/contaminants would ultimately exert on the environment. The desorbable amount would well reflect the bioavailability as well as the bioremediation potential of 2-4-ring PAHs. For higher ring PAHs, the desorbable amount measured would be a better measure of environmental impact of that contamination, depending on the degradative capacity of the microorganisms present in the site. The XAD-2 assisted desorption is different from mild extractions as proposed by Kelsey et al. in assessing the bioavailability of organic contaminants (10). Mild extraction procedures use nonvigorous extraction to mimic the available fraction of contaminants, where the soils/sediments-liquid phase distribution coefficients of contaminants are changed as compared with the natural soils/sediments-water environment. The optimal solvent and agitation combination obtained by exhaustive studies for one contaminant/medium does not necessarily apply to other contaminants or media. On the other hand, the soils/sediments-liquid phase distribution coefficients of contaminants stay unchanged in the XAD-2 assisted desorption system. The desorption is accelerated by the presence of XAD-2. XAD-2 acts as a contaminant sink by quickly adsorbing the desorbed fraction. Thus, the concentration of contaminants in the aqueous phase is minimized, and the driving force for them to desorb VOL. 38, NO. 6, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

1791

FIGURE 6. Comparison of the biodegradation time courses (a) and desorption time courses (b) of 5-ring PAH benzo[a]pyrene (I) and 6-ring PAH benzo[g,h,i]pyrelene (II).

FIGURE 7. Correlation between 2-4-ring PAH residuals after biodegradation and XAD-2 assisted desorption studies. Results of the XAD-2 assisted desorption study were taken from the system with 24 g of XAD-2 amendment. Standard deviations of triplicate samples in biodegradation and desorption experiments are shown as vertical and horizontal error bars, respectively. from soils/sediments is maximized. Theoretically, XAD-2 assisted desorption could measure exactly how much contaminants would come off the soils/sediments and become available for degradation. This is why for various 2-, 3-, and 4-ring PAHs the desorbable amounts measured were comparable to their extent of biodegradation at a given XAD-2 to sediment ratio. In a separate study, we compared XAD-2 assisted desorption with mild extractions using various 1792

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 6, 2004

alcohol-water cosolvents for assessing the bioavailability of PAHs. The result showed that the XAD-2 assisted desorption was considerably better in simulating the bioavailability of lower ring PAHs that were initially present in the sediment in small amounts (20). For those compounds, the change in their sediment-water distribution coefficients resulted in more significant differences between extraction and desorption. Without changing the solid-liquid-phase distribu-

tion coefficients of contaminants, XAD-2 assisted desorption also has the possibility of convenient application for assessing the bioavailability of other hydrophobic organic contaminants in various porous media, as long as sufficient XAD-2 adsorption capacity is provided. This is to be confirmed with additional experiments on a variety of contaminants-media combinations. It is an accepted fact that bioavailability plays an important role in determining the degree of cleanup needed for aged contaminations. To our knowledge, however, this study is the first reported approach that directly compares the bioavailability of organic contaminants to their desorption with the assistance of sorbents. XAD series resins appeared to be suitable candidate sorbents for the study because of their high efficiency, capacity, and chemical stability. XAD-2 has had applications in the adsorption, concentration, and analysis of polyaromatic compounds and is capable of adsorbing compounds of up to 20 000 in molecular weight. Another advantage for the use of this resin can be attributed to its low contaminant bleed as compared to acrylic-ester resins, such as XAD-8 (21). The relatively mild agitation of approximately 18 rpm used in the XAD-2 assisted desorption experiments helped PAHs reach desorption equilibrium sooner by ensuring increased contact between the contaminated sediments, solution, and resin. The affect of agitation on the sediment/aqueous phase distribution equilibrium points should be limited, if any. The amounts of desorbed PAHs measurable with XAD-2 assisted desorption can be a good bioavailability indicator for ex-situ bioremediations, where mixing is often provided. For in-situ bioremediation cases where mass transfer limitations play an important role, the predicted availability of contaminants may not be observed for years. It could be viewed as the maximum possible removal attainable by bioremediation and washout (where no extractant or surfactant is used or present in a significant amount). Such predictions will err on the conservative side in yielding an estimate of the environmental impact of a contamination. As observed, the kinetics of PAH desorption improves with higher adsorbent mass, while the desorption equilibrium remains unchanged. Desorption reached equilibrium within about 2 weeks using a XAD-2 to sediment mass ratio of 8. However, it was more difficult to separate XAD-2 from the sediment at this ratio due to the large amount of XAD-2. For contaminated sediments with the comparable amount of available PAHs as shown in the study, a XAD-2 to sediment mass ratio of 4 would be recommended to take advantage of faster kinetics without compromising the ease of separation. Sample preacidification is a common practice used to optimize the adsorption capacity of XAD resin. Studies reported 2-3 times as much organic matter was extracted by XADs2 from preacidified samples than from pH neutral ones (21, 22). Further investigations are needed to determine is sample preacidification would improve the desorption kinetics of the XAD-2 assisted desorption on the per mass basis. Other polystyrene based XAD resins, XAD-4, XAD-16, and XAD-2010, which feature a larger surface area than XAD2, could also be tested as alternative sorbents that potentially may result in faster kinetics of desorption.

Acknowledgments This research was supported by the National Risk Management Research Laboratory of U.S. EPA. We thank Dr. Neal Sellers for his assistance in our PAH analysis.

Literature Cited (1) Watts, R. J. Hazardous Wastes: Sources, Pathways, Receptors; John Wiley & Sons: New York, 1998. (2) Steinberg, S. M.; Pignatello, J. J.; Sawhney, B. L. Persistence of 1,2-dibromoethane in soils: entrapment in intraparticle micropores. Environ. Sci. Technol. 1987, 21, 1201. (3) National Research Council. In Situ Bioremediation: When Does it Work?; National Academy Press: Washington, DC, 1993. (4) Hatzinger, P. B.; Alexander, M. Effect of Aging of Chemicals in Soil on Their Biodegradability and Extractability. Environ. Sci. Technol. 1995, 29, 537. (5) Valdes, J. L. Bioremediation; Kluwer Academic Publishers: Norwell, MA, 2000. (6) Luthy, R. G.; Aiken, G. R.; Brusseau, M. L.; Cunningham, S. D.; Gschwend, P. M.; Pignatello, J. J.; Reinhard, M.; Traina, S. J.; Weber, W. J., Jr.; Westall, J. C. Sequestration of Hydrophobic Organic Contaminants by Geosorbents. Environ. Sci. Technol. 1997, 31, 3341. (7) Guerin, W. F.; Boyd, S. A. Bioavailability of naphthalene associated with natural and synthetic sorbents. Water Res. 1997, 31, 1504. (8) Tang, J.; Robertson, B. K.; Alexander, M. Chemical-Extraction Methods to Estimate Bioavailability of DDT, DDE, and DDD in Soil. Environ. Sci. Technol. 1999, 33, 4346. (9) Tang, J.; Alexander, M. Mild extractability and Bioavailabilty of Polycyclic Aromatic Hydrocarbons in Soil. Environ. Toxicol. Chem. 1999, 18, 2711. (10) Kelsey, J. W.; Kottler, B. D.; Alexander, M. Selective Chemical Extractants To Predict Bioavailability of Soil-Aged Organic Chemicals. Environ. Sci. Technol. 1997, 31, 214. (11) Maurice, P. A.; Pullin, M. J.; Cabaniss, S. E.; Zhou, Q.; NamjesnikDejanovic, K.; Aiken, G. R. A comparison of surface water natural organic matter in raw filtered water samples, XAD, and reverse osmosis isolates. Water Res. 2002, 36, 2357. (12) Cho, J.; Amy, G.; Pellegrino, J.; Yoon, Y. Characterization of clean and natural organic matter (NOM) fouled NF and UF membranes and foulants characterization. Desalination 1998, 118, 101. (13) Jorand, F.; Boue´-Bigne, F.; Block, J. C.; Urbain, V. Hydrophobic/ hydrophilic properties of activated sludge exopolymeric substances. Water Sci. Technol. 1998, 37 (4-5), 307. (14) Baun, A.; Nyholm, N. Monitoring pesticides in surface water using bioassays on XAD-2 preconcentrated samples. Water Sci. Technol. 1996, 33, 339. (15) Kuch, H. M.; Ballschmiter, K. Determination of endogenous and exogenous estrogens in effluents from sewage treatment plants at the ng/l-level. Fresenius J. Anal. Chem. 2000, 366, 392. (16) Lepane, V. Comparison of XAD-2 resins for the isolation of humic substances from seawater. J. Chromatogr. A 1999, 845, 329. (17) Utvik, T. I. R.; Durell, G. S.; Johnsen, S. Determining produced water originating polycyclic aromatic hydrocarbons in North Seawaters: comparison of sampling techniques. Mar. Pollut. Bull. 1999, 38 (11), 977. (18) Stumm, W.; Morgan, J. J. Aquatic Chemistry; John Wiley and Sons: New York, 1981. (19) U.S. Environmental Protection Agency, Office of Solid Waste, Economic, Methods, and Risk Analysis Division. Test Methods for Evaluating Solid Waste, Physical/Chemical Methods; U.S. Government Printing Office: Washington, DC, 1996; EPA publication SW-846. (20) Lei, L. Ph.D. Dissertation. University of Cincinnati, 2003. (21) Slauenwhite, D. E.; Wangersky, P. J. Extraction of marine organic matter on XAD-2: Effect of sample acidification and development of an in situ pre-acidification technique. Mar. Chem. 1996, 54, 107. (22) Fu, T.; Pocklington, R. Quantitative adsorption of organic matter from seawater on solid matrixes. Mar. Chem. 1983, 13, 255.

Received for review September 25, 2003. Revised manuscript received December 30, 2003. Accepted December 30, 2003. ES030643P

VOL. 38, NO. 6, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

1793