Article pubs.acs.org/est
Aerosol−Water Distribution of PCDD/Fs and PCBs in the Baltic Sea Region Anna Sobek,*,† Hans Peter H. Arp,‡ Karin Wiberg,§ Jenny Hedman,∥ and Gerard Cornelissen*,†,‡,⊥ †
Department of Applied Environmental Science (ITM), Stockholm University, SE-10691 Stockholm, Sweden Norwegian Geotechnical Institute (NGI), Oslo, Norway § Department of Aquatic Sciences and Assessment, Swedish University of Agricultural Sciences (SLU), SE-750 07, Uppsala, Sweden ∥ Department of Contaminant Research, Swedish Museum of Natural History, SE-104 05 Stockholm, Sweden ⊥ Department of Plant and Environmental Sciences, Norwegian University of Life Sciences (UMB), P.O.Box 5003, N-1432, Ås, Norway ‡
S Supporting Information *
ABSTRACT: Atmospheric deposition is a major pathway of PCDD/Fs to the Baltic Sea. We studied the aerosol−water distribution for aerosols collected close to the Baltic Sea in order to investigate the availability of pollutants sorbed to aerosols deposited on water. Aerosols were analyzed for both total concentration (Soxhlet extraction) and the freely dissolved water concentration (extraction with 17-μm polyoxymethylene equilibrium passive samplers). Concentrations of PCDD/F and sum PCB-7 in aerosols were 65−1300 pg/g dw TEQ and 22−100 ng/g dw, respectively. Organic carbon (OC)-normalized aerosol−water distribution ratios (Kaer‑water,OC) were consistently lower (factor 2−60) than previously determined sediment organic carbon−water distribution ratios (Ksed,OC). Hence PCDD/Fs and PCBs entering the Baltic Sea through aerosol deposition seem to be more available for desorption to the water phase than PCDD/Fs and PCBs sorbed to sediment. Further, we investigated whether aerosol−water distribution may be predicted from the air−aerosol partitioning constant multiplied by the Henry’s Law constant. This proposed model for aerosol−water distribution underestimated measured values for PCBs by factors of 1−17 and for PCDD/Fs by more than a factor 10. These findings can be used to improve future fate modeling of PCBs and PCDD/Fs in marine environments and specifically the Baltic Sea.
■
fw,5 implying that herring and other fatty fish caught in the Baltic Sea cannot be traded within the EU. Atmospheric transport from regions in the south and east and subsequent deposition is probably the major input pathway of PCDD/Fs to the Baltic Sea,6−8 with model estimates for the Baltic proper and Bothnian Sea of 133 g TEQ per year as opposed to riverine inputs of 32 g TEQ per year.6 Dry deposition with aerosols is suggested to be the dominating pathway from the atmosphere to water for PCDD/Fs, especially for the highly chlorinated congeners, whereas air− water exchange seems to be more important for PCBs, as shown by a previous study in the Atlantic Ocean.9 Even though transport via aerosols forms a significant pathway of PCBs and PCDD/Fs to oceans, studies on the partitioning between deposited aerosols and water have until
INTRODUCTION Polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans (PCDD/Fs) are widespread environmental contaminants. Due to their toxicity, persistence in the environment, and high bioaccumulation potential they have received much attention over the last decades. PCDD/Fs are introduced to the environment as byproducts of various industrial processes, in particular from combustion.1 Actions to reduce emissions of PCDD/Fs since their peak in the 1970s have been effective as evidenced by observations of decreasing concentrations in the environment and in humans.2,3 The Baltic Sea is a semienclosed brackish sea, with 9 countries sharing its coastline. High levels of hydrophobic organic contaminants (HOC) in Baltic Sea biota have been a matter of great concern during the last decades.4 Today biota concentrations of many HOCs, such as polychlorinated biphenyls (PCBs), are decreasing, whereas levels of PCDD/ Fs remain stable or even increase.4 Concentrations of PCDD/ Fs in lipid-rich fish of the Baltic Sea exceed the EU quality standard for food and feed of 4 toxic equivalents (TEQ) pg/g © 2012 American Chemical Society
Received: Revised: Accepted: Published: 781
July 14, 2012 December 5, 2012 December 7, 2012 December 7, 2012 dx.doi.org/10.1021/es3028567 | Environ. Sci. Technol. 2013, 47, 781−789
Environmental Science & Technology
Article
Table 1. Modeled and Measured Distribution Ratios (log Kaer‑water; OC- or WIOC-Normalized) for PCBs in Aerosols aerosol−water distribution ratio SPARCa log Kaer‑water‑WIOC (L/kgWIOC) PCB PCB PCB PCB PCB PCB PCB PCB PCB PCB PCB PCB PCB PCB PCB PCB PCB
28 52 77 81 101 105 114 118 126 138 153 156 157 167 169 180 189
5.8 6.4 6.5 6.5 7.0 7.1 7.1 7.1 7.1 7.7 7.7 7.7 7.7 7.7 7.7 8.4 8.4
± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±
1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5
PP-LFERb log Kaer‑water‑WIOC (L/kgWIOC) 6.1 6.5 6.8 6.8 7.0 7.2 7.2 7.2 7.4 7.6 7.5 7.7 7.7 7.7 7.9 8.1 8.3
± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±
measuredc log Kaer‑water‑OC (L/kgOC)
1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0
6.2 6.9 7.2 7.4 7.5 7.5 7.7 7.8 7.8 7.5 8.2 8.0 10.0 8.6 8.6 8.9 9.5
± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±
0.4 0.1 0.1 0.1 0.0 0.2 0.1 0.0 0.1 0.6 0.1 0.5 0.0 0.0 0.1 0.1 0.8
measuredd log Kaer‑water‑WIOC (L/kgWIOC) 6.6 7.2 7.5 7.7 7.8 7.8 8.0 8.1 8.1 7.8 8.5 8.3 10.3 8.9 8.9 9.2 9.8
± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±
0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5
a See eq 9; the error is based on comparisons with gas−particle distribution (1 log unit)33,34 and assumed uncertainty regarding the fraction of OC that is WIOC (0.5 log unit). bSee eqs 5, 7, and 8; the error is based on comparisons with gas−particle distribution (0.5 log unit)10 and assumed uncertainty regarding the fraction of OC that is WIOC (0.5 log unit). cErrors reflect analytical reproducibility. dThe error is based on assumed uncertainty regarding the fraction of OC that is WIOC (0.5 log unit).
constant10 as this would be a possible way of estimating aerosol−water distribution ratios for other HOCs and environments. Because aerosol deposition is a significant input pathway for PCDD/Fs and PCBs to the Baltic Sea, knowledge about the availability of these compounds after they enter the water column will have implications for future actions to be taken to reduce levels of PCDD/Fs in fish and the food web of the Baltic Sea.
now been lacking. Some tentative conclusions can, however, be drawn based on current knowledge on sorption mechanisms of PCBs and PCDD/Fs. In a screening study on the dominating sorption phase for diverse organic contaminants in aerosols, it was found that most HOCs, including PCBs, favor partitioning into the hydrophobic domain of aerosol organic matter, termed the water insoluble organic matter (WIOM) fraction.10,11 The one notable exception was PAHs for some aerosol samples. The reason WIOM failed to describe partitioning of PAHs to aerosols may be that the PAHs, being coformed with atmospheric soot, were occluded in nonexchangeable domains of the aerosols, or strongly (nonlinearly) sorbed to black carbon (BC) components of the aerosols.10 PCDD/Fs are, like PAHs, formed from combustion processes. Therefore it is possible that PCDD/Fs in some cases are irreversibly occluded in crystalline soot structures and micropores of the aerosols, as opposed to PCBs and other HOCs that are not coformed during combustion.10,12 For sediments, many studies have shown that PCDD/Fs and PAHs sorb strongly to BC, with sorption coefficients that may be several magnitudes higher than those describing sorption to amorphous organic carbon.12−14 However, compilations of large data sets for sediments in background environments have found that assuming BC-dominated sorption can lead to more imprecise characterizations and greater uncertainty than when assuming organic carbon (OC)-dominated sorption.15 In this study we investigated aerosol−water distribution of native PCDD/Fs and PCBs for aerosols collected in Stockholm, Sweden, close to the Baltic Sea. The aim was to assess the availability of PCDD/Fs and PCBs once the aerosols have entered the water column by using passive samplers, and to relate the aerosol−water distribution to earlier reported sediment−pore water distribution in Baltic Sea top-layer sediments. Further, we wanted to investigate whether the aerosol−water distribution ratios may be predicted from air− aerosol distribution ratios multiplied by the Henry’s Law
■
MATERIALS AND METHODS Materials. All solvents used were of glass-distilled purity (Merck, Darmstadt, Germany). The PCDD/F calibration and internal standards and the calibration standards of PCB 77, 81, 126, and 169 were obtained from Wellington Laboratories (Ontario, Canada). The remaining PCB calibration standards were from AccuStandard (New Haven, CT, USA), while all PCB internal standards were from Cambridge Isotope Laboratories (Andover, MA, USA). The internal standards included seventeen 13C-labeled PCDD/Fs (all 2,3,7,8-substituted) and fourteen 13C-labeled PCBs. Quantification was made by the isotope dilution method and by using the native (12C) calibration standards. Seventeen PCDD/Fs (all 2,3,7,8substituted) and seventeen PCBs were quantified (Tables 1 and 2). Sampling of Aerosols. A high-volume air sampler equipped with GFF filters (MG 160, 293 mm diameter, Munktell, Sweden) was deployed at 10 m height at Stockholm University, around 1 km from the shore of the Baltic Sea (59.4° N; 18.1° E). Sampling was done continuously from Feb 22, 2010 through Jan 3, 2011 (Supporting Information (SI) Table S5), and the filters were changed regularly (n = 12). Each of the filters consisted of approximately 0.4−0.6 g of aerosol sampled in 10 000−40 000 m3 of air (average 22 110 m3). Large volumes were sampled since the objective was to collect as much sample as possible for the distribution ratio experiments described below. The flow rate in large-volume samples is not constant 782
dx.doi.org/10.1021/es3028567 | Environ. Sci. Technol. 2013, 47, 781−789
Environmental Science & Technology
Article
Chemical Analysis of the Aerosols and the Available Fraction (POMs). Filter-aerosol samples were weighed (0.04− 0.23 g of aerosol) into pre-Soxhlet-toluene-extracted cellulose extraction thimbles, after which internal standards were added. Extraction was carried out in Soxhlet−Dean−Stark by reflux with 100 mL of toluene for at least 15 h. Tetradecane (50 μL) was added as a keeper. After reflux, a solvent exchange from toluene to n-hexane was carried out. The POM strands were extracted by horizontal shaking in 250 mL of n-hexane for 96 h (180 rpm; 20 ± 1 °C) as described by Cornelissen et al.21 The cleanup of aerosol and POM extracts generally followed the methodology outlined by Danielsson et al.22 with modifications described by Josefsson et al.23 For the aerosol filters, three open liquid chromatographic columns were used, including a multilayer silica column (KOH-silica, neutral silica, 40% H2SO4-silica, and Na2SO4), a column with activated carbon (AX21) mixed with Celite, and a miniaturized multilayer silica column. Clean-up and fractionation of the POM samples required only the second and third columns. After cleanup, four 13 C-labeled PCDD/F and two 13C-labeled PCB recovery standards were added, and the final volume was reduced to 40 μL (tetradecane). Instrumental analysis was performed using a GC/HRMS system (Agilent 6890N GC coupled to Waters Autospec Ultima MS) equipped with a 60 m × 0.25 mm DB-5 column (J&W Scientific, Folsom, CA, USA).24 TOC and BC Determination. TOC in the aerosol samples was determined with catalytic combustion elemental analysis at 1030 °C after micro acidification (1 M HCl) to remove inorganic carbonates.25 BC content was determined using the same method on small samples (15 mg) combusted at 375 ± 1 °C for 18 h under abundant oxygen access.25 Determination of KPOM‑17. Earlier studies assumed that POM-17 and POM-55 have the same POM−water distribution ratios.21 This was tested in the present work for PCDD/Fs by precalibration of POM−water distribution ratios, KPOM‑17:18
Table 2. Modeled and Measured Distribution Ratios (log Kaer‑water; OC- or WIOC-Normalized) for PCDD/Fs in Aerosols aerosol−water distribution ratio a
SPARC log Kaer‑water‑WIOC (L/kgWIOC) 2378 TeCDD 12378 PeCDD 123478 HxCDD 123678 HxCDD 123789 HxCDD 1234678 HpCDD OCDD 2378 TeCDF 12378 PeCDF 23478 PeCDFa 123478 HxCDF 123678 HxCDF 234678 HxCDFb 123789 HxCDFc 1234678 HpCDF 1234789 HpCDF OCDF
6.6 7.2 7.9 7.9 7.9 8.6 9.2 6.6 7.3 7.2 7.9 7.9 7.9 7.9 8.6 8.6 9.2
± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±
1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5
measuredb log Kaer‑water‑OC (L/kgOC) 7.9 8.2 9.1 9.5 9.1 10.4
