Environ. Sci. Technol. 2010, 44, 3381–3387
Modeling the Mass Transfer of Hydrophobic Organic Pollutants in Briefly and Continuously Mixed Sediment after Amendment with Activated Carbon SARAH E. HALE AND DAVID WERNER* School of Civil Engineering and Geosciences, Newcastle University, Newcastle upon Tyne, NE1 7RU, England, United Kingdom
Received November 25, 2009. Revised manuscript received March 17, 2010. Accepted March 29, 2010.
Activated carbon (AC) amendment is currently being investigated as an in situ remediation technique for sediments contaminated by persistent organic pollutants. Understanding the mass transfer of pollutants from weaker binding sites on sediment particles, to stronger binding sites inside AC particles, is important for the evaluation of this strategy. Here we study the mass transfer of polycyclic aromatic hydrocarbons (PAHs) from River Tyne sediment to polyethylene (PE) passive samplers in the presence and absence of AC under two mixing regimes. Continuously mixing and a brief initial mixing period to incorporate AC to the system, followed by unmixed conditions in settled sediments, were compared. The reduction in total PAH concentration in the PE sampler was greater than 99% after 12 months AC contact for both conditions. A numerical model based on concepts used to simulate wellmixed AC-sediment slurries was further developed to describe the briefly mixed system. These models could predict upper and lower limits for the expected remediation effectiveness for variable AC-sediment mixing regimes. It appears that mixing mode has a small impact on the treatment effectiveness for River Tyne sediment which has a strongly bound, slowly released pollutant source. However, a greater impact is anticipated for contaminated sediments containing more available pollutants.
Introduction Numerous laboratory studies have demonstrated significant reductions in the free aqueous concentration and biouptake of polychlorinated biphenyls (PCBs) (1, 2), polycyclic aromatic hydrocarbons (PAHs) (3), and the organochlorine pesticide dichlorodiphenyltrichloroethane (DDT) (4) after the amendment of field-contaminated sediments with activated carbon (AC). AC is a strong sorbent and the benefits of AC amendment are now being studied under field conditions, for example at Hunters Point, CA (5, 6). This remediation approach relies on the mass transfer of sediment pollutants to the AC in order to reduce the pollutant free aqueous concentration, availability and uptake by sediment dwelling organisms. The AC-sediment mixing regime has been cited as causing lower short-term treatment efficiencies for field AC * Corresponding author phone: 0044 191 222 5099; fax: 0044 191 222 6502; e-mail:
[email protected]. 10.1021/es903582n
2010 American Chemical Society
Published on Web 04/14/2010
amendments compared to batches used in the laboratory due to differences in the hydraulic regime between conditions (5, 6). In briefly mixed sediment, the addition of AC may result in spatially heterogeneous pollutant concentrations in sediment porewater, and the average reduction in sediment porewater pollutant concentration could take years to approach the reduction observed in well-mixed sediment systems (7). Contrary to this observation, both a brief initial mixing of sediment and AC, and applying AC as a layer on top of contaminated sediment were sufficient to substantially reduce the uptake of PCBs by the oligochaete Lumbriculus variegatus in bench-scale experiments (8). Intensive bioturbation resulting from a high worm density in these biouptake studies may have assisted the mass transfer of PCBs from the sediment particles to the AC. The aim of this study is to further our understanding of the impact of mixing regime on the short-term effectiveness of AC amendment to contaminated sediment. To this end we compare a completely mixed batch system with one mixed briefly to incorporate AC into sediment, then left undisturbed. We use sediment from the River Tyne polluted with PAHs as an experimental case study and passive sampling techniques are employed to monitor changes in PAH availability. The numerical modeling tool described by Werner et al. (7) is further developed to discuss expected differences in pollutant mass transfer for a completely mixed AC-sediment system, compared to sediment which remains settled after the incorporation of AC, for variable compound and sediment properties. The two AC-sediment contact scenarios investigated should represent reasonable boundaries for mixing conditions achieved in field-contaminated sediments.
Materials and Methods Sediment Characterization. Sediment was obtained from the Walker Riverside Country Park, Newcastle upon Tyne, UK (National Grid Reference for the approximate center of the site is NZ 291 631). The sediment was characterized by carrying out a density separation and Soxhlet extraction to determine the solid phase PAH concentrations, a Tenax bead extraction to measure desorption from the bulk sediment in order to determine PAH release rates, and a petrographic analysis to determine the sediment matrix composition. In addition, free aqueous PAH concentrations were determined by using polyethylene (PE) passive samplers. Details of these methods are provided in the Supporting Information (SI) (pages S3-S5). Passive Samplers. Passive samplers were produced by cutting 26 µm thick PE plastic bags (VWR International Ltd., Leicestershire, UK) into small pieces (0.30 ( 0.01 g) and precleaning them with dichloromethane, methanol and distilled water in series, each for 24 h. Sorption, release, and diffusion characteristics of phenanthrene, anthracene and pyrene have been reported for these samplers elsewhere (9), and were in good agreement with values predicted by empirical relationships presented by Booij et al. (10). Activated Carbon Characterization. A bitumen AC was used in all experiments and its properties are given in the SI on page S5. The AC was ground and sieved to particle sizes of less than 45 µm and 63-125 µm. For the smallest fraction the AC was passed through a 45 µm mesh sieve and collected on a base tray. Grain size analysis of this fraction showed 45% of particles were 10-20 µm in size and the remaining 55% were 20-45 µm in size. To determine the activated carbon-water partitioning coefficient (KAC), PE samplers (0.30 ( 0.01 g) were preloaded with PAHs and possibly sediment organic matter by exposing them to sediment for 20 weeks. VOL. 44, NO. 9, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3381
Following this period, samplers were cleaned and then added to triplicate 40 mL amber glass vials with Teflon lined lids containing powdered AC (3 mg, particle size
