Environ. Sci. Technol. 2010, 44, 6752–6759
Field Measurement of Diffusional Mass Transfer of HOCs at the Sediment-Water Interface E S P E N E E K , * ,† G E R A R D C O R N E L I S S E N , †,§ A N D G I J S D . B R E E D V E L D †,‡ Norwegian Geotechnical Institute, P.O. Box 3930, Ullevaal Stadion, NO-0806 Oslo, Norway, Department of Geosciences, University of Oslo, P.O. Box 1047, Blindern, NO-0316 Oslo, Norway, and Department of Applied Environmental Sciences (ITM), Stockholm University, 10691 Stockholm, Sweden
Received March 13, 2010. Revised manuscript received June 27, 2010. Accepted July 14, 2010.
The sediment to water diffusive flux of PAHs and PCBs was measured under field conditions with a novel infinite-sink benthic flux chamber that deployed semipermeable membrane devices (SPMD) as a sorbing material. Fluxes were measured before and after in situ capping of sediments in Oslo Harbour with clean clay. The fluxes of native pyrene and PCB 52 from uncapped contaminated sediment measured with the flux chamber were 0.3-1.6 µg m-2 d-1 and 2-8 ng m-2 d-1, respectively. Fluxes from the capped sediment were reduced by 93-97%. The in situ measured fluxes were compared to fluxes independently calculated from freely dissolved concentrations in pore water and overlying water, measured using equilibrium passive samplers, diffusive boundary layer (DBL) thickness, measured by an alabaster dissolution method and literature values of diffusion coefficients. Measured fluxes from the uncapped sediment agreed well with calculated fluxes, the median of the ratio of the measured flux over the calculated flux was 0.9 with an inter quartile range of 0.5-1.6.
Introduction Seabed sediments are commonly considered as a sink for many contaminants in the marine environment including organic contaminants like polycyclic aromatic hydrocarbons (PAHs) and polychlorinated biphenyls (PCBs). However, when emissions from other sources are reduced, sediments might become a substantial source of contaminant transport into the marine environment. Recent studies deploying equilibrium passive samplers have shown substantially higher dissolved concentrations of PAHs in the porewater of contaminated sediments than in the overlying water (1, 2). Due to this concentration gradient PAHs can diffuse from pore water through the diffusive boundary layer (DBL) into overlying water. As the sediment has been recognized as a potential contaminant source, strategies have been developed to better manage these sediments (3, 4). To isolate contaminated sediments and to minimize the sediment-to-water flux, capping of contaminated sediments with clean materials has * Corresponding author: e-mail:
[email protected]. † Norwegian Geotechnical Institute. ‡ University of Oslo. § Stockholm University. 6752
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 17, 2010
been performed in many sediment remediation projects (5). Thus the quantification of the sediment-water flux is important for two reasons: 1) the flux can be used as an assessment of the risk associated with contaminated sediments, and 2) it can be used to evaluate the effectiveness of remediation by capping. Flux mechanisms commonly regarded as potentially important to sediment water flux are 1) molecular diffusion through the DBL, 2) molecular diffusion through the surface sediment, 3) colloidal diffusion, 4) flushing of overlying water through pores in the sediment (bioirrigation), 5) moving parcels of solid matter or pore water (bioturbation), 6) resuspension of particles, 7) wave pumping, 8) subaqueous groundwater discharge (SGD), and 9) advection caused by gas ebullition (6, 7). The variety of mechanisms that can contribute to sediment-water contaminant flux requires that assessment of the total flux needs to follow several lines using different methods. Sediment-to-water fluxes of compounds dissolved in the sediment pore water can be determined by three methods: 1) calculation from sediment-water activity gradients, 2) measurement with in situ flux chambers, or 3) measurements with retrieved sediment cores. Earlier efforts have been made to measure contaminant flux in situ in areas with advective transport of water through the sediment-water interface (8). To our knowledge there are no reports of in situ measured flux of hydrophobic organic contaminants (HOCs) from low permeable fine grained sediments where transfer across the sediment-water interface by diffusion is important. The common approach has rather been approach 1), i.e. to calculate the flux from the concentration gradient of the HOCs over the sediment water interface applying Fick’s first law (9, 10). Flux calculations have usually been based on measured total sediment concentrations and generic sediment/water distribution coefficients (Kd) (10). However, it has been demonstrated that these generic Kd values are often inappropriate for many organic contaminants, as actual field-Kd values can vary from sediment to sediment by orders of magnitude, depending on the presence of carbonaceous geosorbents (11, 12). Unless specific knowledge of the sitespecific Kd in the sediment exists, estimated flux levels can thus be off by orders of magnitude. Deploying equilibrium passive samplers, pore water concentrations can now be measured directly in the sediment as well as in the overlying water (13, 14). Therefore, on the basis of measured pore water concentrations, sediment-water fluxes can be determined more reliably, since the use of generic Kd-values is avoided. An alternative approach is the second one, i.e. to directly measure the sediment-water HOC flux. A method based on this approach was developed in the present study. In this method the amount of HOCs that enters the overlying water was quantified with help of infinite-sink benthic flux chambers, analogous to methods previously used to determine oxygen and nutrient fluxes (15, 16). Advantages of this method include the following: 1) The flux can be determined in situ with a minimum alteration of environmental conditions in the sediment and at practically any required depth. 2) The measured flux reflects the conditions at the sedimentwater interface which are the conditions that control the sediment-water diffusive flux. The aims of the present study were 1) to develop and test a benthic infinite-sink flux chamber to measure flux of PAHs and PCBs from sediments under field conditions; 2) to compare in situ sediment-water fluxes for capped and uncapped sediments; and 3) to compare fluxes measured in situ by the flux chamber to fluxes derived from measured 10.1021/es100818w
2010 American Chemical Society
Published on Web 08/12/2010
FIGURE 1. a) Outline of the infinite-sink flux chamber; b) theoretical concentration gradients across DBL inside and outside the chamber; and c) outline of the alabaster chamber for measurement of DBL thickness. values of the pore water concentrations (Cpw) and thickness of the DBL at the sediment surface.
Materials and Methods Chamber Design. In classical flux chamber designs (16) the chamber encloses a volume of water above the sediment surface. The current chamber design (Figure 1a) introduces a novel method for the collection of HOCs deploying: a triolein-filled (1 mL) semipermeable membrane devices (SPMD) (Exposmeter, Sweden, 0.92 m length, 534 cm2 surface area, 75-90 um thick low density polyethylene) a strongly sorbing phase acting as an “infinite-sink” collecting HOCs diffusing from the seabed. Sediment-SPMD HOC flux is controlled by the mass-transfer at the two interfaces sediment-water and water-SPMD (Figure 1b). The chamber itself was cylindrical (L 250 mm, area 0.049 m2, inner volume 3.4 L) with a rim that extending 135 mm out of the chamber wall and preventing the top 70 mm of the chamber from penetrating the sediment as well as ensuring that the sediment-SPMD distance was identical at each deployment. The 50 mm part of the chamber below this rim penetrated the sediment surface to prevent water flow into the chamber. Gentle placement and minimal resuspension of the surface sediments was ensured by a 50 mm top side opening allowing free water flow through the chamber during placement. This opening was closed immediately after placement by a 2.5-kg lid equipped with a neoprene seal with
