Sorption of Organic Compounds to Fresh and Field-Aged Activated

Dec 1, 2011 - ... Institute (NGI), P.O. Box 3930, Ullevål Stadium, N-0806 Oslo, Norway. ‡ ... and Environmental Engineering, University of Maryland...
1 downloads 0 Views 2MB Size
Download PDF
Article pubs.acs.org/est

Sorption of Organic Compounds to Fresh and Field-Aged Activated Carbons in Soils and Sediments Amy M. P. Oen,† Barbara Beckingham,‡ Upal Ghosh,*,‡ Marie Elmquist Kruså,† Richard G. Luthy, Thomas Hartnik,∥ Thomas Henriksen,⊥ and Gerard Cornelissen*,†,#,∇

§

†

Norwegian Geotechnical Institute (NGI), P.O. Box 3930, Ullevål Stadium, N-0806 Oslo, Norway Chemical, Biochemical, and Environmental Engineering, University of Maryland Baltimore County, Baltimore, Maryland 21250, United States § Civil and Environmental Engineering, Stanford University, Stanford, California 94305-4020, United States ∥ Soil and Environment Division, Norwegian Institute for Agricultural and Environmental Research (Bioforsk), Fredrik A. Dahls vei 20, N-1432 Ås, Norway ⊥ Lindum Ressurs og Gjenvinning AS, Lerpeveien 155, N-3036 Drammen, Norway # Applied Environmental Sciences (ITM), Stockholm University, 10691 Stockholm, Sweden ∇ Plant and Environmental Sciences (IPM), University of Life Sciences (UMB), Ås, Norway ‡

S Supporting Information *

ABSTRACT: Activated carbon (AC) amendment to polluted sediment or soil is an emerging in situ treatment technique that reduces freely dissolved porewater concentrations and subsequently reduces the ecological and human health risk of hydrophobic organic compounds (HOCs). An important question is the capacity of the amended AC after prolonged exposure in the field. To address this issue, sorption of freshly spiked and native HOCs to AC aged under natural field conditions and fresh AC amendments was compared for one soil and two sediments. After 12−32 months of field aging, all AC amendments demonstrated effectiveness for reducing pore water concentrations of both native (30−95%) and spiked (10−90%) HOCs compared to unamended sediment or soil. Values of KAC for field-aged AC were lower than freshly added AC for spiked HOCs up to a factor of 10, while the effect was less for native HOCs. The different behavior in sorbing native HOCs compared to freshly spiked HOCs was attributed to differences in the sorption kinetics and degree of competition for sorption sites between the contaminants and pore-clogging natural organic matter. The implications of these findings are that amended AC can still be effective in sorbing additional HOCs some years following amendment in the field. Thus, a certain level of long-term sustainability of this remediation approach is observed, but conclusions for decade-long periods cannot be drawn solely based on the present study.

■

INTRODUCTION An innovative in situ technique to reduce the freely dissolved concentrations of hydrophobic organic compounds (HOCs) is to amend contaminated sediment or soil with activated carbon (AC).1−4 The HOCs sorb strongly to the AC particles such that porewater concentrations and consequently the uptake in benthic organisms and risks associated with food-web bioaccumulation are reduced. The type of AC used for the amendment (which influences the sorption affinity for the contaminants of concern), the amount of AC applied (especially as a fraction of native organic carbon), and the ability to effectively distribute the amendment in the field are important factors determining the potential success of AC amendment. Predicting the necessary dose of AC and effectiveness in reducing porewater concentrations based on fundamental sorption properties of clean AC is challenging due © 2011 American Chemical Society

to the attenuation of sorption in the sediment matrix. Attenuation of AC sorption properties is likely caused by fouling and competition with natural sorbates in sediment. Therefore, understanding effective sorption properties of AC exposed to a soil/sediment matrix is necessary to develop models that can better predict reductions in porewater concentrations of HOCs in soils/sediments. Activated carbon is produced from carbonaceous materials such as coal, peat, or agricultural residues. The differences in starting materials and the type of activation utilized result in AC particles with different pore size distributions and surface Received: Revised: Accepted: Published: 810

August 12, 2011 November 19, 2011 December 1, 2011 December 1, 2011 dx.doi.org/10.1021/es202814e | Environ. Sci. Technol. 2012, 46, 810−817

Environmental Science & Technology

Article

properties.5 Commonly, AC particles have a highly microporous structure resulting in very high internal surface areas.6−8 Pure AC particles are clean of natural organic matter (NOM) on the surface. The sorption of HOCs to the AC particle is fully dominated by adsorption to the AC surface, mainly the internal pore surfaces.9−11 While investigations of AC amendment to soils/ sediments have reported reduction of porewater HOC concentrations of >70% (Figure 2 in Ghosh et al.6) these reductions are less than what is predicted based on the much greater sorption capacity of clean AC compared to natural organic matter.4 Mixing into sediments or soils has been observed to reduce or “attenuate” HOC sorption to AC by sorptive competition and/or pore blockage.4,12−17 In the present paper the word “attenuation” thus means “diminished sorption”, and not “risk reduction”. With exposure to soils/ sediments, the binding of HOCs to AC particles may change as a result of the following three processes. First, physical, chemical, or biological changes of the AC particle itself might occur. However, these are expected to be unlikely or very slow under natural field conditions because of the high stability of the AC matrix.18 Second, clogging/blockage of sorption sites may occur through binding of colloids or natural organic matter (NOM),12 other organic contaminants,13 or possibly by oil14,19 and leading to decreased AC sorption for new HOCs. However, this process would not impact the sorption of already sorbed HOCs. Third, increasing penetration of clogged/blocked pores over time could occur, due to slow diffusion through pore-blocking NOM and oil,20 probably leading to gradually increased sorption to AC over time. Hale et al.20 recently showed that HOC sorption to AC-amended sediment actually increased over time, and after 26 months laboratory-aged AC-amended sediment sorbed equally strongly as clean AC in the absence of sediment. The impact that competing or pore-clogging sorbates, e.g. NOM, has on sorption of target analytes may depend on the molecular size, AC pore size distribution, and relative concentrations of competing NOM and/or chemicals21 and is therefore likely to vary by carbon type and sediment or soil location. In this study we combine results from three separate investigations with widely varying site characteristics (marine mudflat sediment, freshwater river sediment, and terrestrial soil) to interpret the extent of attenuation of HOC sorption to AC under different matrix effects. Although there were differences in the specific experimental protocols used by the three separate sets of investigators, the general approach of the present study was to collect soil and sediments previously amended by AC in the field and investigate their sorption characteristics by utilizing passive samplers in the laboratory. We compare the sorption of such field-aged AC to that of freshly amended AC with respect to i) matrix type (sediment and soil) and ii) native vs spiked sorbates.

been described previously (Hunters Point sediment,2,3 Grasse River sediment,22 and urban soil19). In brief, the field-aged sediment from the tidal mud flat at Hunters Point was sampled from test plot D which was amended with AC to about 30 cm sediment depth in January 2006 using a barge with a rotavator.2 For the sediment used in the current study from Grasse River, AC was applied in September 2006 using a tine sled device supported by a barge-mounted crane that injected a slurry of AC into surficial sediments under 4.6 m of water depth. The urban soil was mechanically mixed with AC in November 2007 using an excavator. The material was thereafter placed in 5 × 5 × 1 m plots with concrete walls supported by a kaolinite clay layer to prevent leakage of water to the sides. Different types of ACs were used in the various field trials: coconut shell-based granular AC in the sampled treatment area in Grasse River, coal-based AC at Hunters Point, and two coalbased ACs of different particle sizes (pulverized and granular) in the urban soil (Table S1). The AC content in the field-aged sample from the Grasse River sediment was determined with a wet-chemical oxidation method.23 The AC content reported for the Hunters Point sediment and the urban soil were determined from their respective TOC contents after amendment.2 Cho et al.2 also used the wet-chemical oxidation method for Hunters Point and found that the amount of AC dose estimated gave similar values (within about 10%) for both methods. Although the experiments with the three different environmental matrices vary in their respective execution, they provide useful materials for comparison that can also be extended to other systems. Experimental Design for Grasse River Sediments. To study freshly amended AC, clean AC was added to reference sediment from one of the sampling sites (UTA-14)22 collected in 2006 prior to activated carbon amendment and stored in the dark at 4 °C until use. Field-aged AC was studied for sediment sampled in 2008 from the same site (2-years postamendment). Thus three sediment-AC combinations were examined: nonamended reference sediment, reference sediment with fresh AC, and sediment containing field-aged AC. The clean AC additions were at levels of 0, 2, 4, and 8% (dw) in duplicate. In glass vials, wet sediment (3.2 g dw) was combined with Grasse River water (30 mL), NaN3 (1 mg), and spiking solution (20 μL acetone, containing 2000 ng of PCB-14, PCB-65, and PCB166). After one day of mixing 200 mg of polyoxymethylene (POM, 55 μm thick, prepared by slicing block cylinders of POM on a lathe equipped with a high precision razor blade,24 Table S2 in the SI) was added to the slurry and subsequently rolled end-over-end for 30 d, an adequate time to achieve equilibrium sorption in the POM sampler as documented in previous verification experiments.15,24 Duplicate samples were also prepared with a lower spike level (250 ng) for unamended (0% AC) and field-aged sediment. After the equilibration period, the POM was retrieved, wiped clean, and tripleextracted (3 × 24 h) with hexane:acetone (1:1, vol). The pooled extract was solvent exchanged to hexane, cleaned up by shaking overnight with an addition of granular copper followed by column chromatography with 3% deactivated silica gel, and analyzed for congener-level PCBs by gas chromatography with electron capture detection following a modified EPA method, including quality control procedures.25 An additional extraction of the POM samplers found no measurable PCBs remaining after the first sequential extractions. Sediment was analyzed for PCBs as described in Beckingham and Ghosh.26 Instrument detection limits for individual PCBs translate to 0.2−4 μg/kg

■

EXPERIMENTAL SECTION Tested Environmental Matrices. Two sediments (Hunters Point, San Francisco, CA and Grasse River, Massena, NY, both in the USA) and one urban soil consisting of excavated material from a construction site (Lindum AS, Drammen, Norway) were investigated. The two sediments are PCBcontaminated, whereas the urban soil is PAH-contaminated (Table S1 in the Supporting Information, SI). The respective AC amendment techniques used at the different sites and the specific sediment/soil characteristics for each sediment have 811

dx.doi.org/10.1021/es202814e | Environ. Sci. Technol. 2012, 46, 810−817

Environmental Science & Technology

Article

for POM. Surrogate standard recoveries for POM blank samples (N = 2) were 96 ± 1.4% for PCB-14 and 90 ± 0.6% for PCB-65, and concentration of PCBs in blanks were below detection limits. Experimental Design for Hunters Point Sediments. Clean AC was added to nonamended field-mixed reference sediment (Plot C)2,3 as well as to field-aged AC-amended sediment sampled in August 2008 (2.5-years postamendment). Four sediment-AC combinations were examined: nonamended reference sediment, reference sediment with fresh AC, sediment containing field-aged AC, and sediment with both field-aged AC and an extra addition of fresh AC (+2% dw). Spiking solution containing PCB-29, PCB-69, and PCB-103 (in acetone) was added to sediments (4 g dw) in glass flasks with water (40 mL), NaN3 (1 mg), and 100 mg of POM (17 μm thick, lathe cut at NGI24) at different concentrations giving isotherms with 5, 25, 50, 250, 500, and 5000 ng of each spiked PCB. The experimental setup and subsequent PCB analysis was similar to the one described above with a few differences including a longer equilibration time (90 d mixing on a roller at 2 rpm) and use of a POM sampler that was 1/3 as thick (Table S2). Detection limits for PCBs were 0.05 μg/kg for sediment and 0.2 μg/kg for POM. Surrogate standard recoveries were 84−96% for PCB-14 and 88−108% for PCB-65 across all analyses. Experimental Design for Urban Soil Samples. ACamended urban soil was sampled in November 2008, 12 months after amendment, by collecting 1 kg of soil from three different places per treatment field and thoroughly mixing the samples. Fresh AC was added to field-aged AC-amended soil in incremental amounts, giving final AC concentrations of 1−9% (dw) granular activated carbon (GAC) and 8−15% (dw) pulverized activated carbon (PAC; Table S2). Three soil-AC combinations were examined for each tested AC type: unamended reference soil, soil containing field-aged AC, and soil containing field-aged AC with an extra addition of fresh AC. Spiking solution (100 μL methanol) contained the following compounds: anthracene-d10, fluoranthene-d10, chrysene-d12 (10−17 μg per labeled PAH), PCB-12, PCB-14, and PCB-32 (1−3 μg per PCB). The experimental setup was similar to the others (Table S2), with 55 μm POM added to a slurry of soil (5 g dw), water (40 mL), and NaN3 (1 mg) in glass flasks and mixed for 30 d. POM was extracted by shaking in heptane and then concentrated to approximately 0.5 mL before cleaning the extract on a silica column topped with 5 mm of sodium sulfate and concentrated to 0.2 mL before the PAHs were analyzed on a Agilent 6850 Gas Chromatograph coupled to an Agilent 5973 mass spectrometer as described by Cornelissen et al.15 Before each series, four external calibration standards (1− 1000 μg/mL of analyte) containing constant amounts of internal standards (100 μg/mL) were run to check for linearity. Detection limits for single PAHs were 0.02 μg/kg for soil and 0.1 μg/kg for POM. Surrogate standard recoveries were 80− 120% for deuterated phenanthrene, deuterated benzo(a)antracene, and PCB-52. All PAH abbreviations are in the footnote of Table S1 in the SI.

■

nF,AC CS = fTOC KTOCCPW + fAC KACCPW

KAC =

CS − fTOC KTOCCPW nF , AC fAC CPW

(1)

(2)

where CS is HOC concentration in sediment or soil (ng/kg dry weight), CPW is porewater HOC concentration determined with POM passive samplers (ng/L), f TOC and fAC are the fractions of TOC and AC (either freshly added or field-aged), respectively. KTOC and KAC are the sorbent-water distribution coefficients for TOC (L/kg) and AC, respectively [(ng/kg)/(ng/L)nF]. KTOC was calculated as the measured distribution coefficients for unamended sediment normalized to sediment organic carbon content (KTOC = CS/CPW f OC). Porewater concentrations were calculated from measured POM concentration using KPOM values in Cornelissen et al.27 for POM-55 μm and Oen et al.28 for POM-17 μm and the relationship KPOM = CPOM/CPW. In the present study, the Freundlich exponent of AC sorption (nF,AC) was assumed to be 0.8 based on recent modeling results by Werner et al.,29 studies by McDonough et al.,16 and best fit values to the sorption isotherms for the spiked PCBs at Hunters Point (Figure SI in the SI). Here a nF,AC value of 0.8 is used under the assumption that the main sorption mechanisms are similar for various PAHs and PCBs in the same AC material. Furthermore, a sensitivity analysis indicated that varying nF,AC between 0.6 and 1.0 led to a variation in KAC of generally around 0.5 log unit but in some cases up to one log unit (Table S3 in the SI). However, KAC values before and after field-aging can still be compared, as long as the same nF,AC is used in both cases. Equation 2 was used to determine the KAC,fresh (fresh AC added to untreated Grasse River and Hunters Point sediment) and KAC,field‑aged (sediment or soil without any added fresh AC). To calculate the KAC,fresh for the urban soil samples and Hunters Point sediment (fresh AC added to field-aged), a slight modification was made to eq 2 because the samples contained both fresh AC and field-aged AC portions. Thus, we used eq 3 to calculate KAC,fresh.

KAC,fresh = ((Cs − fTOC KTOCCPW nF,AC − fAC,field − aged KAC,field − agedCPW )/ n ,F,AC (fAC,fresh CPW ))

(3)

Here, the fAC,field‑aged is the concentration of aged AC in the sample, KAC,field‑aged is the sorption coefficient calculated with eq 2 for the aged-AC-only samples, and the fAC,fresh is the amount of freshly amended AC.

■

RESULTS AND DISCUSSION Grasse River Sediment. In Figure 1, horizontal lines indicate where CPW of each respective PCB is lowered by 95% compared to CPW in nonamended sediment. Dosing Grasse River untreated sediment with 8% fresh GAC, or 1.4 times native TOC, reduced CPW of spiked compounds by 76−99% compared to nonamended sediment. The CPW was reduced by >95% for PCB-14 (di) at all fresh AC doses (2−8%) but for PCB-65 (tetra) only at the highest AC dose (8%; or 1.4 times native TOC), while the CPW for PCB-166 (hexa) was reduced by less than 95% (33−76%) at all AC doses. The field-aged

CALCULATION OF KAC FOR FRESH AND FIELD-AGED AC

Sorption of HOCs to AC in an AC-amended sediment was calculated with a nonlinear Freundlich isotherm 812

dx.doi.org/10.1021/es202814e | Environ. Sci. Technol. 2012, 46, 810−817

Environmental Science & Technology

813

9.18g 0.2 0.2 0.3 0.1 ± ± ± ± 6.6 6.6 6.7 6.9 4.3 0.2 0.2 0.2 0.2 ± ± ± ± 7.4 7.4 7.6 7.8 6.3 0.3 0.3 0.3 0.2 ± ± ± ± 6.8 6.7 6.9 7.3 2.0

7.5 ± 0.2 6.5 ± 0.3

6.7 6.5 6.5 7.7 4.8 6.9 ± 0.2 7.0 ± 0.2 7.3 ± 0.4 7.4 ± 0.4 2.0−8.0

PCB-(52 + 49) PCB-101 PCB-136 PCB-153 PCB-180 AC (%) 6.8 7.1 ± 0.04

PCB-(52 + 49) PCB-101 PCB-132 PCB-153 PCB-180 AC (%)

Fresh (U) indicating freshly added AC to untreated sediment, fresh (FA) indicating freshly added AC to field-aged AC-treated sediment and soil. bAverage values for three AC (%) amendment levels. Only one measured value at field-added concentration. dAverage values for five concentrations of the extra-spiked compounds, five replicates of the native compounds. eAverage values for five AC (%) amendment levels. fExtra-spiked compounds after field aging. gLiterature values for a peat-based AC.34 hLiterature values for a coal-based AC.16 ILiterature values for a coal-based AC.35

8.3 ± 0.2 9.1 ± 0.2 8.6 ± 0.3 9.3 ± 0.3 0.9−7.8

6.8 ± 0.3 8.36g, 7.82h 6.3 ± 0.3

8.16g, 8.23h 8.44g 7.0 ± 0.2 7.2 ± 0.3 7.2 ± 0.3 -

fresh (FA) GACd fresh (U) GACd

6.9 ± 0.1 7.2 ± 0.1 7.0 ± 0.1 PCB-29f PCB-69f PCB-103f PCB-18 PCB-28 6.2 6.6 6.7 6.7 6.6 0.1 0.4 0.2 0.1 0.04 ± ± ± ± ± 7.4 8.0 7.6 7.0 7.1

c

8.1 ± 0.5 8.7 ± 1.1 8.5 ± 0.8 9.5 ± 0.7 1.0−7.5 8.1 9.0 8.7 9.3 1.0

7.9 ± 0.5 6.9

0.4 0.4 0.5 0.5 0.5 ± ± ± ± ± 9.3 9.4 9.8 7.8 8.2 8.8 8.7 8.6 7.4 7.2 0.2 0.2 0.2 0.3 0.3 ± ± ± ± ± 9.1 9.0 9.1 7.3 7.5

PCB-12h PCB-14h PCB-32h ANT-d10h FLUOd10h CHRSd12h PHEN FLUO PYR B(a)PYR AC (%) 6.3 ± 0.3 6.1 ± 0.2 5.7 ± 0.1 -

fresh (FA) GACe compound literature values field-aged GACd

PCB

compound field-aged GACc fresh (U) GACb compound

PCB-14f PCB-65f PCB-166f PCB-18 PCB-28

a

8.3 9.2 8.7 9.3 7.9

7.2

7.8I 8.7I 8.5I 8.6I

8.96g 9.06g 9.2 9.0 9.2 7.2 7.5

literature values fresh (FA) PACe field-aged GACc

urban soil

a a

Hunters Point

a

Grasse River AC-amended sediment samples contained 4.8% AC (or 0.8 times native TOC). Reductions for spiked compounds in field-aged sediment were 82, 35, and 12% for PCB-14, -65 and -166, respectively, with similar results for both the high and low spiking level. The similar reductions at the two spiking levels indicated that competition among PCB molecules for sorption sites was not a dominant mechanism for attenuation in Grasse River sediments at the level of the total spike which was 10−80% of native total PCBs. For native PCBs in the field-aged AC sediment, reductions of porewater concentrations were 94% for dichlorobiphenyls, 73% for tetrachlorobiphenyls, and 29% for hexachlorobiphenyls, with an overall total native PCB porewater concentration reduction of 91% compared to unamended sediment. Values of log KAC (calculated assuming nF,AC = 0.8) for freshly added AC were higher than those of field-aged AC for the spiked PCBs (0.9 to 1.4 log unit) and for the majority of native PCBs (0.2 to 0.8 log unit) (Table 1). In addition, KAC,field‑aged/KAC,fresh values were calculated and indicated that sorption of native PCBs was attenuated by up to a factor of 5 and sorption of spiked PCBs by up to a factor of 10 (Table S4). Therefore, exposure of AC in the field decreased sorption coefficients for PCBs compared to fresh AC. However, it is important to note that observations of porewater concentration reductions measured in the field already incorporate sorption attenuation, and a much lower dose of AC would be required in the field if AC sorption was not attenuated. The log KAC values for native PCB-18 and PCB-(52 + 49) can be compared directly to a study by McDonough et al.16 for sorption of PCBs to clean AC or biofilm/DOM-loaded AC. For fresh AC amended to sediment, values are about an order of magnitude lower than determined for their coal-based AC in clean water, yet both fresh and field-aged AC results are more comparable (within about 0.2 to 0.5 log units) to values determined for biofilm or DOM-loaded AC. We did not find that sorption attenuation increased with increasing hydrophobicity, in contrast to observations reported by others.30 However, our observations are in line with McDonough et al.16 who did not find the difference in PCB sorption coefficients for DOM/biofilm-loaded AC versus AC in

a

Figure 1. Spiked PCB freely dissolved porewater concentrations measured for Grasse River sediment containing fresh or field-aged amended granular AC. Horizontal lines indicate the concentration where 95% of the added spike is removed compared to the reference sediment without AC. Error bars show ±1 standard deviation.

Grasse River

Table 1. Determined log KAC (nF,AC = 0.8) Values for AC in Different Sediment and Soil Systems and for Native and Extra-Spiked Compounds

a

field-aged PACc

PAH

Article

dx.doi.org/10.1021/es202814e | Environ. Sci. Technol. 2012, 46, 810−817

Environmental Science & Technology

Article

Figure 2. Spiked PCB freely dissolved porewater concentrations measured for Hunters Point sediment with fresh, field-aged, or a combination of fresh and field-aged granular AC. The PCB concentrations spiked to the sediment were 5000 ng (A), 500 ng (B), 50 ng (C), or 5 ng (D). Horizontal lines indicate the concentration where 95% of the added spike is removed compared to the reference sediment without AC.

increasing the spiking concentration to 500 and 5000 ng, the decrease in CPW dropped to 52% for PCB-103 (penta), whereas the percentage was almost the same for the two smaller PCB congeners with 98% (tri PCB-29) and 81% (tetra PCB-69). When making an extra addition of 2% fresh AC to the Hunters Point sediment samples already containing field-aged AC (a 50% increase in dose), CPW decreased further (Figure 2). PCB sorption with Hunters Point sediment and AC is thus both affected by spiking concentration and field aging. Log KAC for Hunters Point sediment with fresh AC ranged from 6.9 to 7.2 for spiked PCBs and 6.9 to 7.8 for the native PCBs (Table 1, Table S4 for KAC,field‑aged/KAC,fresh ratios). All log KAC values for native PCB-(52 + 49) are lower than values determined by McDonough et al.16 except for freshly added AC to field-aged sediment with a log KAC value (7.5) lower than that determined for AC in clean water but higher than the values determined for biofilm or DOM-loaded AC. Values of log KAC were higher when calculated using eq 3 with fresh AC added to field-aged AC in sediment, which indicates some uncertainty in this calculation approach. For field-aged AC samples, the log KAC decreased by 0.6−1.1 log units for the spiked PCBs and 0.1−0.4 log units for the native PCBs compared to fresh AC added to untreated sediment. Although attenuation is observed in the field-aged sediments, the reductions in CPW are still significant (>80% reduction as seen in Figure 2 and observed in previous field studies28,33) for the field-aged AC compared to nonamended reference Hunters Point sediment. For both Hunters Point and Grasse River fieldaged AC, similarly diminished sorption for the spiked PCBs was observed (up to 1.4 log unit in KAC); however, the decrease in sorption coefficients compared to values for fresh AC are greater for most of the native PCBs in the Grasse River sediment. The greater attenuation for the field-aged Grasse

clean water to be related to hydrophobicity. A lack of trend of sorption attenuation with hydrophobicity may indicate that the NOM in Grasse River sediment competes more directly with less hydrophobic PCBs or acts to block the smaller pores of the coconut shell carbon that are more amenable to sorption for smaller PCB molecules. On the other hand, the phenomenon may also be observed if the fresh AC did not reach equilibrium sorption due to the slower sorption kinetics of more hydrophobic PCBs; for instance, percent reduction in aqueous concentration for 4% fresh AC compared to unamended decreased with increasing hydrophobicity, from 97% for PCB18 to 33% for PCB-180. Increased sorption effectiveness with time (2 to 3 years postamendment) has been reported for more hydrophobic PCBs in field-amended Grasse River sediments.22 Also, sorption coefficients of native PCBs to AC in the present study did not increase linearly with increased hydrophobicity. Dissimilar hydrophobicity dependence of HOC sorption to AC versus octanol or TOC has been observed earlier, including in the same study by McDonough et al.16 which reported a concave relationship between log KAC and PCB congener octanol−water partitioning coefficients.31,32 The interplay of KAC in sorption kinetics and equilibrium may also contribute to the lack of attenuation phenomenon with hydrophobicity. Hunters Point Sediment. The field-aged Hunters Point sediment contained 4.3% GAC, or 6 times native TOC (Table S2), and the extra-spiked PCB concentration ranged from 0.04 to 17 mg/kg. Dosing Hunters Point reference sediment with 2% fresh GAC, or three times native TOC, reduced CPW by 87−98% compared to nonamended sediment at low spiked PCB concentrations (5 ng per PCB congener; Figure 2). For the field-aged AC-amended sediments, CPW reductions for freshly spiked PCBs of 80−97% were observed. When 814

dx.doi.org/10.1021/es202814e | Environ. Sci. Technol. 2012, 46, 810−817

Environmental Science & Technology

Article

Figure 3. Spiked PAH freely dissolved porewater concentrations measured for urban soil with either field-aged or freshly amended pulverized AC (PAC; A) or granular AC (GAC; B). Horizontal lines indicate the concentration where 95% of the added spike is removed compared to the reference sediment without AC. Where standard deviations are shown, triplicate measurements were performed.

sediment for PAC treated soils (7.9%) but was at least double the content of the GAC treated soils (0.9%). Also, to separate the contribution of fresh and field-aged AC to sorption, eq 3 supposes that native HOCs redistribute between field-aged and freshly added AC during the 30 d equilibration. However, as shown in Figure S3 (in the SI), reduction in CPW for native compounds in PAC-treated field-aged soil compared to unamended reference soil was already >90% and amendment of additional PAC did not result in much further reduction. Perhaps most of the rapidly desorbing fraction of soil-associated HOCs has been transferred to AC in the field, and thus there is little additional improvement realized with the addition of more fresh AC to the same sample, at least that can be observed in the time period of the laboratory equilibration. Interpretative Discussion. In the present study, we explore the differences in sorption characteristics of fresh AC and AC exposed to both soil and sediment that have different concentrations of NOM. The following discussion is a hypothesis on how the observations could be explained. However, more experiments on the particle-size scale would be necessary to test the validity of these hypotheses. Such experiments were outside the scope of the present study. When AC is amended to a soil/sediment, HOCs and NOM start to sorb to the AC surface. Initially, the outer surfaces and pores of AC will be easily accessible for HOCs, but as NOM accumulates on the surface/pores, it may become more difficult for fresh HOCs to migrate into the pore structure of the AC as the mass transfer would be slower through the sorbed/clogged NOM matrix (Figure 4: B1, C1). Since the large NOM molecules are unlikely to penetrate into the micropore structure, a steady state of NOM adsorption would take place in a short period on the AC surface with different degrees of attenuation of AC capacity which will depend on abundance of sorbable NOM, ratio of AC dose to available NOM, and type of NOM. In a relatively low NOM environment, over time the HOCs will start to penetrate through the NOM layer, finally reaching the deeper AC pores. In this situation field aging leads to less attenuation of AC capacity over time, which will result in either equal or greater apparent sorbing capacities of the field-aged AC material compared to nonaged soil/sediment-amended AC (Figure 4:B2; KAC, field‑aged > KAC,fresh). This may explain the case for native PAHs in the urban soil, which had relatively low organic carbon content compared to AC doses applied. With a

River sediments compared to the Hunters Point sediments could be due to the shorter laboratory contact time and the nearly ten times higher TOC present in the Grasse River sediment. Urban Soil. In the urban soil, incremental amounts of fresh GAC and PAC types were added to samples containing fieldaged AC of the same type (Table S1). The field-aged urban soils without any freshly added AC contained either 1% GAC or 8% PAC. The field-aged PAC (8%, or 3.2 times native TOC) mixed in soil reduced CPW by >99% for freshly extra-spiked PAHs and PCBs compared to the nonamended soil (Figure 3A and Figure S2A in the SI). Previous laboratory tests of the unamended urban soil mixed with 2% fresh PAC also showed 95−99% reduction in CPW for 15 PAHs.19 The field-aged GAC (1%, or 0.4 times native TOC) reduced the CPW of spiked PCB12 (di) and -14 (di) and the ANT-d10 CPW by 90−95%, yet chrysene and PCB-32 (tri) reductions were lower (83 and 76%, respectively) (Figure 3B and Figure S2B in the SI). By adding a low dose (1%) of fresh GAC to the soil already containing fieldaged GAC (a 100% increase in dose), the CPW further decreased, to >94% reduction for all compounds (Figure 3B). The log KAC was calculated to range between 6.9 and 9.2 for the spiked compounds and between 8.1 and 9.3 for native compounds in both PAC and GAC amended soil (Table 1, Table S4 for KAC,field‑aged/KAC,fresh ratios). The log KAC values for spiked PCB-(12) are all higher (0.5 to as much as 2 log units) than the values determined by McDonough et al.16 The values determined for spiked PAHs were about 1.5 log units lower for the spiked PAH compounds when comparing with peat-based GAC in clean water.34 However, values for native phenanthrene and pyrene for the two field-aged AC types tested were equal to or greater than what was calculated in another study for a coalbased AC material.35 Thus, KAC for native compounds was more similar to literature values for clean AC sorption of PAHs. Values of log KAC for field-aged and fresh AC were similar for native compounds, but KAC values were lower for spiked compounds by 0.3−0.5 log units for GAC and 0.1−0.7 log units for PAC. Therefore, sorption of spiked compounds was attenuated in field-aged activated carbon treated soils in comparison to fresh AC. Performance of field-aged GAC and PAC for sorption of native PAHs appears similar, but comparison is difficult due to the calculation approach. Dose of fresh AC (1−8%) added was less than or approximately equal to content of field-aged AC 815

dx.doi.org/10.1021/es202814e | Environ. Sci. Technol. 2012, 46, 810−817

Environmental Science & Technology

Article

different nature of organic matter in sediments compared to soil, iv) pore clogging has a higher impact on PCBs due to the larger molar volume and less-planar structure of PCBs compared to PAHs,38,39 and/or v) kinetics of chemical desorption from particles and adsorption to AC are different in soil and sediment. In addition to the impact of competing NOM on sorption of HOCs in the field, it will also be important to further explore these other potential causes for the differences observed. In conclusion, attenuation of HOC sorption to field-aged activated carbon was observed to differ for the soil and sediments tested which was influenced by the abundance of NOM relative to the activated carbon present. Sorption may not be as impacted for native HOCs that are rapidly desorbed from sediment and translocated to activated carbon particles within the time frame of competition with NOM or other competing, pore clogging moieties. The most important implication of these findings is that aged amended AC can still be effective for sorbing additional HOCs some years following amendment in the field but not as much as fresh AC.

Figure 4. Illustration shows a side-view of an AC particle with its interior micropores (after ref 12). The sorption to field-aged AC of spiked compounds differs when the AC surface is clean (A) and has low NOM (B) or high NOM (C). In specific environmental situations, the sorbing capacity of field-aged AC may be higher than fresh AC (B2) or lower (C2). The black circles show spiked compounds, whereas the curly lines represent the NOM.

■

ASSOCIATED CONTENT

* Supporting Information S

high abundance of NOM or competing sorbates it is possible that a thicker layer of sorbed NOM will develop, and the full sorption capacity of field-aged AC material in such a system would take longer to achieve (Figure 4:C1−C2; KAC,field‑aged < KAC,fresh). This may describe the case for Hunters Point and Grasse River sediments. Hunters Point sediment is contaminated with PAHs (8 mg/kg) and other organic compounds,25 while Grasse River sediment has a higher TOC content. However, for both sediments the native compounds showed lower attenuation than the spiked compounds (and also for the soil, which showed higher sorption coefficients for field-aged than fresh AC). This possibly points to kinetic factors in the competition between HOCs and NOM for binding to AC: native compounds, copresent with AC during field aging, have had more time to outcompete NOM for AC sorption sites, and thus their sorption to field-aged AC is relatively strong compared to that of extra-spiked PCBs. Therefore in these experiments we have likely measured sorption of spiked and native compounds to both fresh and field-aged AC under quasi steady-state conditions. To reach thermodynamic sorption equilibrium may take months, years, or decades depending on many factors influencing kinetics, including the diffusion path lengths and the sorption strength and size of particles in the system.36 Results to date from field experiments conducted at Hunters Point and Grasse River indicate enhancement of AC performance with time.22,28,33,37 Cho et al.37 observed that PCB uptake in passive samplers was lower after 5 years compared to 1 month after treatment with AC. Furthermore, initial modeling conducted by Cho et al.37 is encouraging, and the results suggest continued reductions of HOCs in pore water concentrations over a period of decades. The KAC determined for the urban soil amendments (both PAC and GAC) was about one to 2 orders of magnitude higher than for AC amended to sediments (both field-aged and freshly added AC). This indicates that either i) the AC in the urban soil was an inherently stronger adsorbent than the ACs applied in the sediments, ii) more fouling occurred in the 2 years of sediment aging than the 1 year of soil aging, iii) AC sorption is attenuated to a higher degree in these particular sediments than in this particular soil, for instance, due to a higher content or

Additional information. This material is available free of charge via the Internet at http://pubs.acs.org.

■ ■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected] (G.C.).

ACKNOWLEDGMENTS The Research Council of Norway funded the Norwegian part of this study through the project “Active” via a grant in their KMB program (“knowledge-building projects with user involvement”; project number 192936). Lindum Ressurs and Gjenvinning A/S is gratefully acknowledged for funding as well as excellent field assistance. Project funding was provided through the Strategic Environmental Research and Development Program (SERDP, ER-1552), Alcoa, NIEHS Superfund Research Program, and the Norwegian Geotechnical Institute’s research stipend. Rahel Brändli (formerly of NGI) is thanked for help in trial establishment.

■

REFERENCES

(1) Ghosh, U.; Luthy, R. G.; Cornelissen, G.; Werner, D.; Menzie, C. In-situ sorbent amendments: a new direction in contaminated sediment management. Environ. Sci. Technol. 2011, 45 (4), 1163− 1168. (2) Cho, Y. M.; Ghosh, U.; Kennedy, A. J.; Grossman, A.; Ray, G.; Tomaszewski, J. E.; Smithenry, D. W.; Bridges, T. S.; Luthy, R. G. Field application of activated carbon amendment for in-situ stabilization of polychlorinated biphenyls in marine sediment. Environ. Sci. Technol. 2009, 43 (10), 3815−3823. (3) Cho, Y. M.; Smithenry, D. W.; Ghosh, U.; Kennedy, A. J.; Millward, R. N.; Bridges, T. S.; Luthy, R. G. Field methods for amending marine sediment with activated carbon and assessing treatment effectiveness. Mar. Environ. Res. 2007, 64 (5), 541−555. (4) Werner, D.; Ghosh, U.; Luthy, R. G. Modeling polychlorinated biphenyl mass transfer after amendment of contaminated sediment with activated carbon. Environ. Sci. Technol. 2006, 40 (13), 4211− 4218. (5) Yoon, T. H.; Benzerara, K.; Ahn, S.; Luthy, R. G.; Tyliszczak, T.; Brown, G. E. Nanometer-scale chemical heterogeneities of black carbon materials and their impacts on PCB sorption properties: Soft 816

dx.doi.org/10.1021/es202814e | Environ. Sci. Technol. 2012, 46, 810−817

Environmental Science & Technology

Article

X-ray spectromicroscopy study. Environ. Sci. Technol. 2006, 40 (19), 5923−5929. (6) Ghosh, U.; Luthy, R. G.; Cornelissen, G.; Werner, D.; Menzie, C. A. In-situ Sorbent Amendments: A New Direction in Contaminated Sediment Management. Environ. Sci. Technol. 2011, DOI: dx.doi.org/ 10.1021/es102694h. (7) McDougall, G. J. The Physical Nature and Manufacture of Activated Carbon. J. South Afr. Inst. Min. Metall. 1991, 91, 109−120. (8) Pelekani, C.; Snoeyink, V. L. Competitive adsorption in natural water: Role of activated carbon pore size. Water Res. 1999, 33, 1209− 1219. (9) Fukushima, M.; Oba, K.; Tanaka, S.; Nakayasu, K.; Nakamura, H.; Hasebe, K. Evaluation of pyrene from activated carbon into an aqueous system containing humic acid. Environ. Sci. Technol. 1997, 31 (8), 2218−2222. (10) Cornelissen, G.; Elmquist, M.; Groth, I.; Gustafsson, O. Effect of sorbate planarity on environmental black carbon sorption. Environ. Sci. Technol. 2004, 38 (13), 3574−3580. (11) Jonker, M. T. O.; Smedes, F. Preferential sorption of planar contaminants in sediments from Lake Ketelmeer, The Netherlands. Environ. Sci. Technol. 2000, 34 (9), 1620−1626. (12) Pignatello, J. J.; Kwon, S.; Lu, Y. F. Effect of natural organic substances on the surface and adsorptive properties of environmental black carbon (char): Attenuation of surface activity by humic and fulvic acids. Environ. Sci. Technol. 2006, 40 (24), 7757−7763. (13) Cornelissen, G.; Gustafsson, O. Effects of added PAHs and precipitated humic acid coatings on phenanthrene sorption to environmental Black carbon. Environ. Pollut. 2006, 141 (3), 526−531. (14) Kwon, S.; Pignatello, J. J. Effect of natural organic substances on the surface and adsorptive properties of environmental black carbon (char): Pseudo pore blockage by model lipid components and its implications for N-2-probed surface properties of natural sorbents. Environ. Sci. Technol. 2005, 39 (20), 7932−7939. (15) Cornelissen, G.; Breedveld, G. D.; Kalaitzidis, S.; Christanis, K; Kibsgaard, A.; Oen, A. M. P. Strong sorption of native PAHs to pyrogenic and unburned carbonaceous geosorbens in sediments. Environ. Sci. Technol. 2006, 40, 1197−1203. (16) McDonough, K. M.; Fairey, J. L.; Lowry, G. V. Adsorption of polychlorinated biphenyls to activated carbon: Equilibrium isotherms and a preliminary assessment of the effect of dissolved organic matter and biofilm loadings. Water Res. 2008, 42, 575−584. (17) Zimmerman, J. R.; Werner, D.; Ghosh, U.; Millward, R. N.; Bridges, T. S.; Luthy, R. G. al. Effects of dose and particle size on activated carbon treatment to sequester polychlorinated biphenyls and polycyclic aromatic hydrocarbons in marine sediments. Environ. Toxicol. Chem. 2005, 24 (7), 1594−1601. (18) Hilber, I.; Bucheli, T. D. Activated carbon amendment to remediate contaminated sediments and soils. Global NEST J. 2010, 12 (3), 305−317. (19) Brandli, R. C.; Hartnik, T.; Henriksen, T.; Cornelissen, G. Sorption of native polyaromatic hydrocarbons (PAH) to black carbon and amended activated carbon in soil. Chemosphere 2008, 73 (11), 1805−1810. (20) Hale, S. E.; Tomaszewski, J. E.; Luthy, R. G.; Werner, D. Sorption of dichlorodiphenyltrichloroethane (DDT) and its metabolites by activated carbon in clean water and sediment slurries. Water Res. 2009, 43, 4336−4346. (21) Ebie, K.; Li, F.; Azuma, Y.; Yuasa, A.; Hagashita, T. Pore distribution effect of activated carbon in adsorbing organic micropollutants from natural water. Water Res. 2001, 35, 167−179. (22) Beckingham, B.; Ghosh, U. Field scale reduction of PCB bioavailability with activated carbon amendment to river sediments. Environ. Sci. Technol. In press 2011. (23) Grossman, A.; Ghosh, U. Measurement of activated carbon and other black carbons in sediments. Chemosphere 2009, 75 (4), 469− 475. (24) Cornelissen, G.; Pettersen, A.; Broman, D.; Mayer, P.; Breedveld, G. D. Field testing of equilibrium passive samplers to

determine freely dissolved native PAH concentrations. Environ. Toxicol. Chem. 2008, 27 (3), 499−508. (25) Ghosh, U.; Zimmerman, J. R.; Luthy, R. G. PCB and PAH speciation among particle types in contaminated harbor sediments and effects on PAH bioavailability. Environ. Sci. Technol. 2003, 37 (10), 2209−2217. (26) Beckingham, B.; Ghosh, U. Comparisons of field and laboratory exposures of Lumbriculus variegatus to polychlorinated biphenylimpacted river sediments. Environ. Toxicol. Chem. 2010, 29 (10), 2851−2858. (27) Cornelissen, G.; Arp, H. P. H.; Pettersen, A.; Hauge, A; Breedveld, G. D. Assessing PAH and PCB emissions from the relocation of harbour sediments using equilibrium passive samplers. Chemosphere 2008, 72, 1581−1587. (28) Oen, A. M. P.; Janssen, E.; Cornelissen, G.; Breedveld, G. D.; Eek, E.; Luthy, R. G. In-situ measurement of PCB pore water concentration profiles in activated carbon-amended sediment using passive samplers. Environ. Sci. Technol. 2011, 45 (9), 4053−4059. (29) Werner, D.; Hale, S. E.; Ghosh, U.; Luthy, R. G. Polychlorinated biphenyl sorption and availability in field-contaminated sediments. Environ. Sci. Technol. 2010, 44 (8), 2809−2815. (30) Hale, S. E.; Kwon, S.; Ghosh, U.; Werner, D. Polychlorinated biphenyl sorption to activated carbon and the attenuation caused by sediment. Global NEST J. 2010, 12, 318−326. (31) Lohman, R. The emergence of black carbon as a super-sorbent in environmental chemistry: The end of octanol? Environ. Forensics 2003, 4 (3), 161−165. (32) Arp, H. P.; Breedveld, G. D.; Cornelissen, G. Estimating the in situ sediment-porewater distribution of PAHs and chlorinated aromatic hydrocarbons in anthropogenic impacted sediments. Environ. Sci. Technol. 2009, 43 (15), 5576−5585. (33) Tomaszewski, J.; Luthy, R. G. Field Deployment of Polyethylene Devices to Measure PCB Concentrations in Pore Water of Contaminated Sediment. Environ. Sci. Technol. 2008, 42 (16), 6086− 6091. (34) Jonker, M. T. O.; Koelmans, A. A. Sorption of polycyclic aromatic hydrocarbons and polychlorinated biphenyls to soot and soot-like materials in the aqueous environment mechanistic considerations. Environ. Sci. Technol. 2002, 36 (17), 3725−3734. (35) Hale, S. E.; Werner, D. Modeling the mass transfer of hydrophobic organic pollutants in briefly and continuously mixed sediment after amendment with activated carbon. Environ. Sci. Technol. 2010, 44 (9), 3381−3387. (36) Sun, X.; Werner, D.; Ghosh, U. Modeling PCB mass transfer and bioaccumulation in a freshwater oligochaete before and after amendment of sediment with activated carbon. Environ. Sci. Technol. 2009, 43 (4), 1115−1121. (37) Cho, Y. M.; Werner, D.; Choi, Y. J.; Luthy, R. G. Long-term monitoring and modeling of the mass transfer of polychlorinated biphenyls in sediment following pilot-scale in-situ amendment with activated carbon. J. Contam. Hydrol. 2011, DOI: doi: 10.1016/ j.jconhyd.2011.09.009. (38) Jonker, M. T. O.; Barendregt, A. Oil is a sedimentary supersorbent for polychlorinated biphenyls. Environ. Sci. Technol. 2006, 40 (12), 3829−3835. (39) Koelmans, A. A.; Meulman, B.; Meijer, T.; Jonker, M. T. O. Attenuation of polychlorinated biphenyl sorption to charcoal by humic acids. Environ. Sci. Technol. 2009, 43 (3), 736−742.

817

dx.doi.org/10.1021/es202814e | Environ. Sci. Technol. 2012, 46, 810−817