Fs, and Non-orthoPCBs, in Water and Bottom Sediments

Apr 14, 2009 - Norwegian Institute for Water Research (NIVA), Branch Office. South, 4879 Grimstad, Norway. Received July 3, 2008. Revised manuscript ...
0 downloads 0 Views 105KB Size
Download PDF
Environ. Sci. Technol. 2009, 43, 3442–3447

PCNs, PCDD/Fs, and Non-orthoPCBs, in Water and Bottom Sediments from the Industrialized Norwegian Grenlandsfjords ¨ HR, R . I S H A Q , * ,† N . J . P E R S S O N , Y . Z E B U AND D. BROMAN Department of Applied Environmental Science (ITM), Stockholm University, SE-106 91 Stockholm, Sweden K. NÆS Norwegian Institute for Water Research (NIVA), Branch Office South, 4879 Grimstad, Norway

Received July 3, 2008. Revised manuscript received February 12, 2009. Accepted March 9, 2009.

Chlorinated toxic planar aromatic compounds were analyzed in the heavily industrialized Grenlandsfjords, which is a system of silled fjords in southern Norway. Surface water samples contained 7.4-160 ng/m3 polychlorinated dibenzo-p-dioxins and furans (PCDD/Fs), 14-410 ng/m3 polychlorinated naphthalenes (PCNs), and 0.31-2.4 ng/m3 non-orthochlorinated biphenyls (non-orthoPCBs). The concentrations of PCDD/Fs were about 300 times higher than in the Baltic Sea. Highest level of the compounds was found near a magnesium production plant. HeptaCDFs and penta-CNs dominated in the inner-fjord waters, and tetra-CB 77 was the major non-orthoPCB congener. Sediment samples had PCDD/F concentration of 25-730 ng/g dw. Highest concentration was detected close to the magnesium plant. Octa-CDF dominated in the fjord sediments, especially near the magnesium plant, indicating a discharge-specific contamination with this congener. The isomer composition of PCDD/Fs and of PCNs, was unchanged when comparing samples from different layers of a sediment core from the deep anoxic water. This concludes that essentially zero degradation had occurred during ∼50 years in this environment.

Introduction Polychlorinated naphthalenes (PCNs) and polychlorinated biphenyls (PCBs) are complex toxic, ubiquitous environmental pollutants that were originally synthesized and produced to be used for industrial purposes in applications such as wood, paper, and fabric impregnators; as engine oil additives; capacitor fluids; transformer coolants; paint; and sealant plasticizers (1, 2). In addition, PCNs are formed during combustion processes such as waste incineration (3, 4), in chloroalkali production (5), copper roasting (6), and appear as impurities in PCB formulations (7). Polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDDs and PCDFs) are also toxic, ubiquitous contaminants that are unintentionally released into the environment as byproducts. PCDD/Fs are formed via thermal processes such * Corresponding author phone: +46-8-785 51 32; fax: +46-8-651 57 50); e-mail: [email protected]. † Current address: County Administrative Board of Stockholm, Box 22067, SE-104 22 Stockholm, Sweden. 3442

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 10, 2009

as solid-waste incineration (8), chemical production processes such as metal smelting (9), pulp and paper production (10), and in the production of chlorinated organic compounds such as PCBs, pentachlorophenol (PCP), and herbicides among others (11, 12). The occurrence and fate of PCBs and PCDD/Fs have been analyzed in many environmental matrices including water, sediments, and biota (13, 14). However, although reported data on the distribution of PCNs in the environment is still limited, there are a number of studies that demonstrate their occurrence in sediments, air, and mammals (15-17). In this study, we present the occurrence and distribution of PCDD/Fs, PCNs and three non-orthoPCBs in water and PCDD/Fs in sediment samples collected in the polluted area of the Grenlandsfjords, Norway, which, for a long period of time, has been subjected to high discharges primarily from magnesium production. To our knowledge, we report the first PCN data in water. This paper focuses on spatial distribution and patterns of these compounds in this area. The pattern is also used to evaluate whether any degradation has occurred in the sediment. Furthermore, congener-based toxic equivalents (TEQs) are calculated and discussed.

Experimental Section Study Region. The Grenlandsfjords are five jointed fjords in the south of Norway (59° 5′ N, 9° 38′ E). The innermost fjord, the Frierfjord, has since the 1950s been substantially polluted by PCDD/F discharges from a magnesium production plant, Heroya (Supporting Information (SI) Figure 1) operating from 1951 until 2002. Large amounts of PCDD/Fs and polychlorinated naphthalenes (PCNs), among others were formed during the chlorination of magnesium oxide to yield waterfree magnesium chloride (18-20). During the last 25 years, the emissions of PCDD/Fs have declined drastically, from kilos down to 1-2 g toxic equivalents annually (gTEQ/yr) (21). However, concentrations of PCDD/F are still high in water, sediment and biota (22-24). Sample Collection and Handling. The locations for the water samples were chosen to map two expected major spatial trends in concentrations. First a horizontal decline with increasing distance to Heroya, and second, a vertical trend in the water column. Since water in silled Norwegian fjords often has a restricted renewal, we hypothesized that different vertical water layers may have different contaminant concentrations. The samples were taken in 1999 and 2000 (SI Figure 1 and Table S1). Two reference water samples were collected far from the discharge location; one upstream in River Farelva (W2-5) and one upstream in the coastal currents of Skagerrak Sea (W2-1). The locations for the sediment samples were chosen to map both spatial and time trends in concentrations. Sediment samples were taken in October 1989 and May 2000. Collection of particulate and dissolved fractions from the water column was carried out primarily according to previously described methodology (25-27). Briefly, 600-1400 L was pumped through a GF/F filter for collection of suspended particles. The filtrate was passed through polyurethane foam (PUF) sorbent to collect the dissolved and the filter-passing fraction. The PUF is considered to be a reliable sorbent giving high sampling efficiency and low break-through for the studied compounds (25, 28). Sediment samples were collected with a Niemisto¨ gravity corer in 1989 (29) and with a gravity Kajak corer 2-4 May 2000 (25). Sample S7 (from 2000) was sliced down to 29 cm in six sections to examine vertical congener patterns (SI Table S1). 10.1021/es8011595 CCC: $40.75

 2009 American Chemical Society

Published on Web 04/14/2009

TABLE 1. PCDD/F, PCN and Non-Ortho PCB Concentrations (ng/m3) in Water Samples surface water samples

ΣPCDD/Fs

ΣPCNs

ΣPCB

W2-7 (Frierfjord) W3-8 (Frierfjord) W3-12 (Frierfjord) W2-4 (Frierfjord) Frierfjord average W3-11 (Skienelva) W3-10 (Skienelva) W2-1 (Farelva) W2-3 (Brevik sill) W3-2 (Brevik sill) Brevik sill average W2-2 (Brevikfjord) W3-1 (Brevikfjord) Brevikfjord average W2-5 (Skagerrak)

16 11 15 14 14 161 7,4 0,35 19 7,6 13 31 14 22 0,67

26 18 44 27 29 406 27 0,49 29 14 21 17 23 20 0,93

0,38 0,53 1,2 0,38 0,62 2,4 0,43 0,13 0,50 0,31 0,40 0,36 0,38 0,37 0,075

Extraction and Clean-up. The analysis was done largely according to previously described methodology (16, 25, 26, 30). Briefly, the GF/F filters and PUFs were Soxhlet-extracted wet for 48 h and the surficial sediments for 24 h with toluene. Before extraction, 13C-labeled standards were added to the samples (see SI Experimental for more details). Procedural blanks were extracted and analyzed with every set of six samples to control the quality of analysis and account for interferences. The water extracts were first cleaned up on silica column (10% water) and then on another column with three layers of modified silica (10% water, 33% KOH, and 40% H2SO4). The sediment extracts were cleaned up on one multilayer column. Sulfur reduction was achieved with copper. The extracts were fractionated on HPLC first using an aminopropylsilica column. A fraction containing dicyclic aromatic compounds (target compounds), was further fractionated on a 2-(1-pyrenyl)ethyldimethylsilylated silica column. Target compounds were back flushed from the column and further cleaned up on a small silica column (10% H2O, w/w) (see SI Experimental for details). GC/MS Analysis. Analysis of non-orthoPCBs and PCNs was achieved by injection on a PTE 5 column and of PCDD/Fs by injection on an SP 2331 column after adding 13 C-labeled recovery standards. A high-resolution gas chromatograph, HP6890 coupled to a magnetic sector Autospec Ultima HRMS was used and the detection was carried out in SIM mode and electron impact (EI) at 34 eV at a resolution of 10 000. Response factors were established with an external standard. Interfering peaks of target compounds detected were subtracted from each sample, i.e. all concentrations were blank subtracted (see SI Experimental for details).

Results and Discussion The concentrations (pg or ng/m3 water and ng/g dry sediment) of 17 common 2,3,7,8-substituted PCDD/Fs, 47 PCN congeners and three non-orthoPCBs (IUPAC -77, -126, -169) in addition to total concentrations of tetra- to octachlorinated isomer groups in water and sediment samples are presented in Table 1 and SI Tables S1 and S2. Total organic carbon, TOC (mg/g dry weight, dw) in the sediments is shown in SI Table S2. The TOC varied between 17 and 83 mg/g dw. The recoveries of the 13C-labeled non-orthoPCB and PCDD/F standards were generally good ranging between 40 and 80%, respectively. The detection limit varied between the samples and compound congeners ranging typically in water between 0.11 and 5.6 pg/m3 for PCDD/Fs, 0.12-4.4 pg/m3 for PCNs and 0.27-4.6 pg/m3 for non-orthoPCBs. All analyzed PCDD/Fs in the sediment samples were detected. All

congener concentrations below detection level were considered zero when presented in the graphs. Water Concentrations. ΣPCDD/F, ΣPCN, and ΣnonorthoPCB concentrations of total phase (sum of filter and PUF fractions) surface water were quite uniform within the Grenlandsfjords having 7.4-160 ng/m3, 14-410 ng/m3 and 0.31-2.4 ng/m3, respectively (Table 1 and SI Table S2). For all contaminant groups, highest concentrations were detected in sample W3-11, which was collected in the strong southflowing surface water current of River Skienselva, upstream of the magnesium plant. The reason for the elevated concentrations found in W3-11 is not fully understood, but may be provoked by mixing of the compensation current into the water column. This current or salt wedge which extends from Frierfjord to the locks of Skien is part of the estuarine circulation (20). A high content of particulate organic carbon (OC) was measured in the W3-11 sample, 700 ( 143 µg/L compared to 144 ( 2 µg/L upstream by the City of Porsgrunn (W3-10), and 86 ( 2 µg/L further upstream in River Farelva (W3-9) reflecting high share of resuspended OC-rich sediments. The reference samples (W2-5 from upstream in Farelva, and W2-1 in Skagerrak Sea) showed, as expected, lowest total concentrations for PCDD/Fs, PCNs, and non-orthoPCBs in Farelva (0.67; 0.93; 0.075 ng/m3), and in Skagerrak (0.35; 0.49; 0.13 ng/m3, respectively). The average concentrations of ΣPCDD/Fs, ΣPCNs, and ΣnonorthoPCBs varied only little between the samples inside the Frierfjord (14, 29, 0.62 ng/m3, respectively) and in the Brevikfjord (22, 20, 0.37 ng/ m3, respectively, Table 1). This can be explained by a net transport of contaminants from Frierfjord out over the Brevik sill. This transport is today mainly driven by resuspension from the shallow and intermediate sediments in the Frierfjord (20). The Grenlandsfjords’ ΣPCDD/F concentrations were elevated by nearly 300 times when compared to surface water concentrations measured at offshore and coastal locations in the Baltic Sea which were 0.24-0.55 ng/m3 (average 0.34 ng/m3) (26). Guerzoni et al. reported 29 pg/L as highest PCDD/F water concentration in Porto Maghera (0.4-29 pg/L) which is significantly lower than the Grenlandsfjords’ concentrations (31). PCDD/Fs concentrations in Farelva and Skagerrak were more in accordance with water samples in the Baltic, than with the samples in the Grenlandsfjords although Farelva concentrations were twice as high compared to Skagerrak. No comparison was enabled for PCNs in surface water since we know of no literature data at present. The vertical water gradients (SI Experimental) showed a significant increase with depth (∼2 and ∼40 m) in ΣPCDD/F (a factor 6, average for the two samples closest to Heroya and 7, for the location west of Heroya) and in ΣPCN concentrations (a factor 14 and 6, respectively). This increase of contaminant concentration with depth reflects most probably resuspension of contaminated sediments. This is supported by the DIG-model described by Persson et al. which predicted larger PCDD/F flux from the Frierfjord sediments than from the source (magnesium production which ceased in 2002) at the time of sampling (25). However, the Σnon-orthoPCB concentrations in the same vertical gradients did not exhibit any significant variations with depth. We hypothesize that the PCB contamination source is other than magnesium production giving relatively lower concentrations for the non-orthoPCBs than for the PCDD/Fs and the PCNs. However, this hypothesis needs to be further supported by more PCB data. PCDD/F Pattern in Water. Polychlorinated dibenzofurans (PCDFs) dominated in Grenlandsfjords’ surface water comprising 85-88% to the ΣPCDD/F, where the hepta (especially 1,2,3,4,6,7,8-HpCDF) and octa-CDF were most prominent contributing to 25 and 22% of ΣPCDD/Fs, respectively. OCDF dominated the individual congeners comprising 40% (average VOL. 43, NO. 10, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

3443

TABLE 2. PCDD/F Concentrations (ng/g dw) in Sediment Samples sampling year

1989 1989 1989 1989 1989 1989 1989 1989 1989 1989 2000 2000 2000 2000 2000 2000 2000 2000 2000 1997

sample id ng/g d.w. I5 G5 G9 G2 13 G12 G8 G11 G13 G14 S1 S2 S3 S5 S6 S7 S8 S9 S10 ΣTCDF 116 48 25 43 81 21 18 39 16 24 8,8 0,90 11 10 18 11 12 10 40 ΣPeCDF 357 54 28 55 147 27 32 68 41 39 15 2,2 23 18 32 18 20 16 70 ΣHxCDF 575 301 212 88 231 62 58 140 94 80 20 4,1 25 23 40 23 26 20 97 ΣHpCDF 214 379 223 98 103 45 70 168 127 105 30 4,3 47 46 64 40 49 29 179 OCDF 1141 576 247 219 601 271 220 427 418 612 36 11 52 48 70 40 47 36 277 ΣPCDF 2403 1358 735 502 1163 426 398 842 697 860 111 22 158 145 223 131 155 111 663 ΣTCDD 28 15 7,9 3,5 15 2,2 1,4 3,0 1,0 1,5 1,0 0,11 0,91 1,1 1,1 1,2 1,2 1,0 2,4 ΣPeCDD 40 26 14 17 27 4,3 4,6 13 6,2 4,2 1,5 0,28 2,2 1,8 3,2 1,7 2,1 1,5 6,7 ΣHxCDD 38 7,1 9,5 7,5 21 7,4 5,1 6,6 5,5 9,9 2,4 0,46 3,7 2,9 4,9 2,7 3,2 2,3 13 ΣHpCDD 21 32 21 10 11 4,6 9,8 19 16 14 2,9 0,73 3,9 4,0 5,7 3,6 3,6 2,5 18 OCDD 18 38 28 12 11 10 12 27 30 26 3,8 1,0 5,0 4,6 6,5 3,7 4,2 3,4 31 ΣPCDD 145 117 79 50 85 28 33 68 59 56 12 2,6 16 15 21 13 14 11 71 total PCDD/F 2548 1475 815 553 1248 454 431 910 755 915 122 25 173 160 245 144 169 121 733 a

Results from sampling in 1997.

within the fjords) rising to 46% in W3-10 and W3-11 in River Skienelva upstream Heroya (average). Oehme et al. (18) also reported PCDF dominance in wastewater from Heroya. The reference samples (Farelva and Skagerrak) differed in their homologue composition being more dominated by tetraand penta-CDFs which together comprised 32-40% of the ΣPCDD/Fs and were also richer in PCDDs (40% for both locations). Within the PCDD group 1,2,3,4,6,7,8-HpCDD and OCDD were elevated (SI Figure 2 and 3). These two congeners dominated the PCDD/F profile together with 1,2,3,4,6,7,8HpCDF in water and air samples from the Baltic (26, 32). The pattern found in the Baltic had mostly OCDD (36%) followed by Hp-CDDs, Hp-CDFs and TeCDFs (12-14%) (26). Air samples from the UK also showed OCDD dominance together with hepta-CDD and tetra-CDF which were related to combustion processes (33). The congener composition in the reference samples is influenced by different sources, whereas OCDF in the Grenlandsfjord samples seems to be strongly related to magnesium production (SI Figure 6). PCN Patterns in Water. Penta-CNs (40%) followed by tetra- and hexa-CNs (20% each) contributed most to the ΣPCN concentrations in Grenlandsfjords’ water samples. Farelva and Skagerrak showed different homologue distribution where tetra-CNs comprised more than half of ΣPCNs in Skagerrak (SI Figure 3). Tetra- and penta-CNs dominated also in particulate matter from the Bothnian Sea and in Swedish air samples (34, 35). In W3-10 and W3-11 more pentaand hexa-CNs and less tetra-CNs contributed to the PCN homologue (45 and 26%, respectively) indicating influence from different industrial sources upstream in River Skienelva or from combustion. The most prominent peak in the Grenlandsfjords was the one containing the coeluting congeners PeCN-52/60 (17%) followed by HpCN-73 (12%) and the coeluted congeners HxCN-66/67 (10%) (SI Figure 5). W3-11 had a similar profile but with higher proportion of the HpCN-73. These congeners (i.e., the first GC-eluting peaks in their homologue groups) were also pronounced in MWI-flyash and chloroalkali plant samples (5, 36). W3-10 had a more “diluted” profile than the mean fjord with dominance by PeCN-52/60 (SI Table S2). This may have been a result of urban contamination from the city of Porsgrunn or from other industrial sources due to mixing of the salt wedge from Heroya with river water. In Farelva PeCN-52/60, -54, -59 and HpCN-73 were elevated and Skagerrak had larger contribution from the tetra-CNs -33/34/37, -44/47 and -38/48 compared to the other samples (SI Figure 5). The tetra-congeners mentioned above together with PeCN-52/60 have been detected in air samples collected in Sweden (35) and Chicago (37) suggesting that the reference samples contain combustion-associated congeners but are “diluted” by diffuse contamination sources. PCN analyses (selected congeners and totals) of different samples collected 3444

S12a 4,6 7,9 14 16 0,34 44 0,40 0,83 1,7 2,1 2,0 7,0 51

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 10, 2009

from Heroya before shutting down the production, showed elevated dominance of HpCN-73 and HxCN-66/67 in flue gas samples taken in 1995 and of HpCN-73, PeCN-52 and HxCN-66/67 in wastewater from 1997. However, in samples of treated process water, the higher chlorinated congeners were less prominent and the congener pattern was more uniformly distributed between tetra- to hepta- chlorinated congeners (38). The non-orthoPCB composition was uniform in all water samples; TCB-77 comprised 85-92% of the Σnon-orthoPCBs (mean 88%) followed by 6.6-10.4 (mean 9%) for PeCB-126. HxCB-169 was only detected in few samples (SI Table S2). Sediment Concentrations. The surficial sediment samples collected 2000 contained 25-730 ng ΣPCDD/F per g dw. Highest concentrations (733 ng/g dw) were, as expected, found in S10, close to Heroya. Furthermore, the total concentrations, varied by only factor 2 (120-240 ng/g dw) between all other locations except S2 (25 ng/g dw, Table 2 and SI Table S3). At present we have no explanation for the lower concentrations in S2 which was collected in a shallower depth than the rest of the locations. The ΣPCDD/F mean sediment concentrations in pooled samples from 1989 were approximately a factor 5 higher than the samples from year 2000, showing a decrease in the discharge from the source during the years. The concentration in the 1989-samples were 430-2500 ng/g dw, and were highest at I5 and G5 (2500 and 1500 ng/g dw, respectively) (Table 2 and SI Table S3). The location of I5 is not directly influenced from the main contaminant discharge, but has received input for secondary effluent streams, whereas G5 is close to Heroya. Transport of contaminants from the Frierfjord sediments to the outer parts cannot be excluded according to Persson et al. (20) but is not confirmed by the concentration in sample S12. Isosaari et al. and Kjeller et al. reported ΣPCDD/F concentrations of 0.61 and 1.4 ng/g dw, respectively from Baltic Sea background sediments (39, 40). Several studies have reported high concentrations of PCDD/Fs in contaminated sediments. Salo et al. reported extremely high concentrations (42 µg/g dw) in chlorophenol polluted sediments from the River Kymijoki in Finland (41). In Tittabawassee River sediments collected downstream Midland, PCDD/F concentrations varied between 1.6-19 ng/g dw (mean 6.2 ng/g dw). Concentrations as high as 56 ng/g dw were found in floodplain soils (42). In sediments from the lagoons and channels around Venice 0.3-14 ng PCDD/F/g dw were found (31). Birch reported concentrations ranging between 32 and 4352 pg WHO-TEQ/g in Australian sediments from Port Jackson (43) which correspond to a factor 2-9 and 10-130 less than the 2000 and 1989 year-sediments in this study, respectively. Hence, the concentrations in the Grenlandsfjords seem to be in the high end of reported polluted sediments from industrialized areas worldwide.

PCDD/F Patterns in Sediment. In the Grenlandsfjords’ sediments, the PCDD/F group was dominated by PCDFs, which constituted more than 90% of the total PCDD/F concentrations (Table 1 and SI Table S3). The distribution of PCDD/F homologues between the two sets of samples (1989 and 2000) differed. In the pooled samples from 1989, the homologue distribution varied greatly between the samples. OCDF dominated, constituting 30-67% of the total PCDD/Fs followed by Hp- or HxCDFs. In the samples from 2000, the homologue distribution was more uniform and OCDF dominated the PCDD/Fs followed by Hp- and HxCDFs. Sediments closer to the magnesium plant seem to be richer in higher chlorinated PCDF congeners. These congeners could probably be used as indicators for this type of contamination in accordance with the water samples. The distribution of PCDD homologues in samples from 2000 followed the same pattern as the PCDFs’ being rich in higher chlorinated homologues while the PCDD pattern varied greatly between the samples from 1989 (Table 2 and SI Table S3). The homologue distribution of Baltic background sediments were different compared to the sediments of Grenlandsfjords with OCDDs and HpCDDs being the most abundant showing mixed patterns of combustion products and transformed PCP (39, 40). Ambient air from the UK was also rich in particle-associated OCDD and HpCDD (33). Buekens et al. (44) reported favorable formation of TeHxCDF and OCDD in annealing electrostatic precipitator (ESP) sample at 300 °C for 2 h in air, which is also different to the magnesium production pattern. Pattern Evaluation. It has been suggested that PCDD/Fs can degrade in anoxic sediments, for instance by microbial mediated dehalogenation pathways (45). Based on estimates of Gibbs free energies of formation for PCDDs, Huang et al. (46) suggested that dehalogenation would proceed at different rates for different congeners. This means that the congener composition can be expected to change over time. We suggest that the Grenlandsfjords is an excellent study site to evaluate such composition changes. This is because the emission term was strongly dominated by an unchanged magnesium production process point source for over 50 years. In order to examine whether the composition of the pollutants in the sediment have changed over time, we looked at different layers of a sediment core. The congener composition in different layers of a sectioned bottom-sediment core S7 (SI Table S1) was investigated for similarities by a correlation analysis of a set of congeners that was present in all layers (16 PCDD/Fs and 32 PCNs). The congener composition in the deeper sediment layers (7-8, 11-12, 27-28 cm) was identical to that in the surface layer (2-3 cm) (r2 > 0.83 and p < 10-6 for PCDD/Fs, and r2 > 0.96 and p < 10-22 for PCNs). This strong correlation indicates that essentially zero degradation of the congeners has occurred in the sediment since the emissions started in 1951. However, in the very deepest layer (28-29 cm), the congener composition correlation was weaker (r2 ) 0.76 and p ) 10-5 for PCDD/Fs, and r2 ) 0.11 and p ) 0.06 for PCNs) to that in the surface layer. The two deepest layers (27-28 and 28-29 cm) were actually dated to be deposited before 1950 when no pollution was expected at that point (47). The concentrations were 10-fold lower in the very deepest layer and may be a result of pore water and bioturbation-mediated mixing from the higher polluted layers. There is a weak indication of some reduction of higher chlorinated and enrichment of lower chlorinated PCDD/F congeners that seem to have occurred in the deep layer (28-29 cm). However, it is hard to establish whether this is a result of an anaerobic microbial dechlorination similar to mechanisms reported by Buerskens et al. (45) or that there was a contamination of this layer from overlaying sediment layers. However, Beurskens et al. pointed out that laboratory-conducted

experiment results could be different from naturally occurred mechanisms since factors like lower microbial population densities, lower activities due to influence from nutrients, cosubstrates and temperatures and lower bioavailability due to stronger sorption of chlorinated aromatics to aged sediments. Nevertheless, microbial dechlorination mechanisms in deeper layers cannot be excluded. The sediments from 1989 and 2000 were very similar in their congener distribution; the most abundant congeners were OCDF (47%) and 1,2,3,4,6,7,8-HpCDF (20%). However, the proportions of the same congeners in the sediments from 1989 were 66 and 11%, respectively (SI Figure 6). The Baltic background sediments were dominated by OCDD and HpCDDs originating from transformed pentachlorophenol with some influence by combustion-related PCDD/Fs with a composition that differs greatly from magnesium production. High portion of OCDD was also found in settling particulate matter in the vicinity of a steel plant and metal smelter area in the Baltic (48). Toxic Potential. The relative toxic potential for the studied compound classes was estimated by calculating toxic equivalents relative to 2,3,7,8-TCDD (TEQs). Toxic equivalent factors (TEFs), based on ethoxyresorufin-O-deethylase (EROD) induction in rat liver cells, were obtained from Hanberg et al. (49) for PCN congeners -66/67, -63, -64, -69, -71, and -73) and from Villeneuve et al. (50) for PCN congeners -54, -56, -57, and -70. WHO-TEFs (51) were used to calculate 2,3,7,8substituted PCDD/F and non-orthoPCBs congener TEQs. Total TEQs of PCDD/Fs in surface water from within the Grenlandsfjords ranged between 0.13 and 2.9 ng/m3 where the highest values were found on the outmouth of Skienselva (W3-11) north of the magnesium plant as shown in SI Table S4. Lowest water TEQs were found in the reference samples at Farelva and Skagerrak. PCDD/F was the contaminant group responsible for the highest toxicity potential in the water samples (total phase, i.e., sum of filter and PUF) contributing to 87-96% of the total toxicity potential in the surface water within the Grenlandsfjords where the furans constituted 74-80% of the total potential. However, in the reference samples from Farelva and Skagerrak and W3-9, the contribution of PCDD/F toxicity was slightly less (87 and 77%, 81%, respectively). In fact, these samples showed higher TEQ influence from non-orthoPCBs (10, 21, and 16%, respectively) compared to the locations closer to the discharge area. PCNs contributed to 3-12% of the total TEQs and were highest in samples close to Heroya (i.e., W3-10, W3-11, and W3-12, SI Table S4). Among the PCNs, the hexaCN and the heptaCN congeners were contributing most to the toxic potential in Grenlandsfjord surface water ranging between 41-62% and 38-63%, respectively. Regarding non-orthoPCBs, PeCB-126 possessed the highest toxic potential (SI Table S3). In the 1989 sediments TEQs ranged between 4.2 and 40 ng/g dw (mean 13 ng/g dw) and in the 2000 sediments TEQs ranged between 0.31 and 10 ng/g dw (mean 3.3 ng/g dw, SI Table S5). Total PCDD/F TEQs were highest at I5 and S10 giving a 4-fold decrease during the 1990s. In the sediment samples, only PCDD/Fs were analyzed and the PCDF group contributed to the highest toxicity potential, 82-91%. The hexa- and penta-CDF congeners were the most potent in all sediment samples. From our study, it can be concluded that industrial activities from primarily a magnesium production plant have generated very high concentrations of toxic, persistent organic pollutants in water and sediments in the Grenlandsfjords. The concentrations with the rather unique pattern of PCDD/Fs are decreasing over time (30), but the decrease rate is most likely caused mainly by transport processes (dissipation to other areas and new sedimentation of more uncontaminated material) along with decreased and ceased emissions, as shown in a modeling study where degradation VOL. 43, NO. 10, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

3445

in sediments was assumed to be zero (26). The unchanged pattern of PCDD/Fs and PCNs in the deep anoxic sediments that accumulated during the period of the magnesium production further supports that no degradation takes place in such environments. Our documentation can hopefully serve as a basis for future studies of the persistency of these pollutants.

Supporting Information Available File 1 (multiple Excel worksheets) contains the following: Table S1, sample information; Table S2, Congener concentrations (pg/m3) of PCDD/Fs, PCNs, and non-orthoPCBs in water samples; Table S3, PCDD/F concentrations (ng/g dw) in sediment samples; Table S4, TEQ concentrations (pg/m3) in water (total water, i.e., filter and PUF summed up); Table S5, TEQ concentrations (pg/g dw) in surface sediment samples; File 2 (Word document) contains the following: Detailed description of experimentals; Figure 1, map showing the location of the sampling sites for surface water, water-column depth profile and bottom sediments in the Grenlandsfjords, Norway; Figure 2, PCDD/F and PCN concentrations (ng/m3) and contaminant distribution between filter and PUF fractions in water samples; Figure 3, homologue composition of PCDFs, PCDDs, and PCNs in water samples; Figure 4, homologue composition of PCDFs, PCDDs in sediment samples; Figure 5, congener profiles of PCNs in water samples; Figure 6, congener profiles of PCDD/Fs in sediment and water samples. This material is available free of charge via the Internet at http://pubs.acs.org.

Literature Cited (1) Brinkman, U. A. T.; Reymer, H. G. M. Polychlorinated naphthalenes. J. Chromatogr. 1976, 127, 203–243. (2) de Voogt, P.; Brinkman U. A. T. Production, Properties and Usage of Polychlorinated Biphenyls. In Halogenated Biphenyls, Terphenyls, Naphthalenes, Dibenzodioxins and Related Products; Kimbrough, R. D., Jensen, A. A., Eds.; Elsevier Science: New York, 1989. (3) Imagawa, T.; Lee, C. W. 2001. Correlation of polychlorinated naphthalenes with polychlorinated dibenzofurans formed from waste incineration. Chemosphere. 2001, 44, 1511–1520. (4) Schneider, M.; Stieglitz, L.; Will, R.; Zwick, G. Formation of polychlorinated naphthalenes on fly ash. Chemosphere. 1998, 37, 2055–2070. (5) Ja¨rnberg, U.; Asplund, L.; de Wit, C.; Grafstro¨m, A. K.; Haglund, P.; Jansson, B.; Lexe´n, K.; Strandell, M.; Olsson, M.; Jonsson, B. Polychlorinated biphenyls and polychlorinated naphthalenes in Swedish sediment and biota: Levels, patterns, and time trends. Environ. Sci. Technol. 1993, 27, 1364–1374. (6) Theisen, J.; Maulshagen, A.; Fuchs, J. Organic and inorganic substances in the copper slag “kieselrot”. Chemosphere. 1993, 26, 881–896. (7) Haglund, P.; Jakobsson, E.; Asplund, L.; Athanasiadou, M. Bergman, Å. Determination of polychlorinated naphthalenes in polychlorinated biphenyl products via capillary gaschromatography-mass spectrometry after separation by gel permeation chromatography. J. Chromatogr. 1993, 634, 79–86. (8) Marklund, S. Dioxin Emissions and Environmental Imissions. A Study of Polychlorinated Dibenzodioxins and Dibenzofurans in Combustion Processes. Ph.D. Thesis, Umeå University, Sweden, 1990. (9) Lexe´n, K.; de Wit, C.; Jansson, B.; Kjeller, L.-O.; Kulp, S-E.; Ljung, K.; So¨derstro¨m, G.; Rappe, C. Polychlorinated dibenzo-p-dioxin and dibenzofuran levels and patterns in samples from different Swedish industries analyzed within the Swedish dioxin survey. Chemosphere. 1993, 27, 163–170. (10) Swanson S. E. Dioxins in the Bleach Plant. A Study of the Occurrence and Formation of Polychlorinated Dibenzofurans and Polychlorinated Dibenzo-p-Dioxins in the Chlorine Bleaching of Wood Pulp. Ph.D. Thesis, Umeå University, Sweden, 1988. (11) Fletcher, C. L.; McKay, W. A. Polychlorinated dibenzo-p-dioxins (PCDDs) and dibenzofurans (PCDFs) in the aquatic environment: A literature review. Chemosphere 1993, 26, 1041–1069. (12) Carroll, W. F., Jr.; Berger, T. C.; Borelli, F. E.; Garrity, P. J.; Jacobs, R. A.; Ledvina, J.; Lewis, J. W.; McCreedy, R. L.; Smith, T. P.; Tuhovak, D. R.; Weston, A. F. Characterization of emissions of 3446

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 10, 2009

(13) (14)

(15) (16)

(17)

(18)

(19)

(20)

(21)

(22)

(23)

(24)

(25)

(26)

(27)

(28)

(29)

(30)

(31)

dioxins and furans from ethylene dichloride, vinyl chloride monomer and polyvinyl chloride facilities in the United States. Consolidated report. Chemosphere 2001, 43, 689–700. Srogi, K. Levels and congener distributions of PCDDs, PCDFs and dioxin-like PCBs in environmental and human samples: a review. Environ. Chem. Lett. 2008, 6, 1–28. Kannan, N. Non- and Mono-Ortho Chlorinated Biphenyls. In the Handbook of Environmental Chemistry, 3 Part K. New Types of Persistent Halogenated Compounds; Paasivirta, J. Ed.; Springer: Berlin, 2000. Falandysz, J. Polychlorinated naphthalenes: An environmental update. Environ. Pollut. 1998, 101, 77–90. Ishaq, R.; Karlson, K.; Na¨f, C. Tissue distribution of polychlorinated naphthalenes (PCNs) and non-ortho chlorinated biphenyls (non-ortho CBs) in harbour porpoises (Phocoena phocoena) from Swedish waters. Chemosphere 2000, 41, 1913– 1925. Lundgren, K.; Tysklind, M.; Ishaq, R.; Broman, D.; van Bavel, B. Polychlorinated naphthalene levels, distribution, and biomagnification in a benthic food chain in the Baltic Sea. Environ. Sci. Technol. 2002, 36, 5005–5013. Oehme, M.; Mano¨, S.; Bjerke, B. Formation of polychlorinated dibenzofurans and dibenzo-p-dioxins by production processes for magnesium and refined nickel. Chemosphere. 1989, 18, 1379– 1389. Musdalslien, U. I.; Stendal, N. A.; Johansen, J. G.; Oehme, M. Pilot plant tests with a wet electrostatic precipitator for reducing PCDD/F in corrosive off-gas from magnesium production. Chemosphere 1991, 23, 1097–1108. Persson, N. J.; Cousins, I.; Molvær, J.; Broman, D.; Næs, K. Modelling the long-term fate of polychlorinated dibenzo-pdioxins and furans (PCDD/Fs) in the Grenland Fjords, Norway. Sci. Total Environ. 2006, 369, 188–202. Næs, K.; Persson, N. J.; Saloranta, T.; Andersen, T.; Berge, J. A.; Hylland, K.; Ruus, A.; Tobiesen, A.; Bergstad, O. A.; Knutzen, J. A. Report 4876; Norwegian Institute for Water Research, 2004. Schlabach, M.; Knutzen, J.; Bjerkeng, B.; Becher, G. Tracing of the PCDF/PCDD contamination from the Frierfjord area along the Norwegian south coast. Organohalogen Compd. 1998, 36, 505–508. Knutzen, J.; Bjerkeng, B.; Næs, K.; Schlabach, M. Polychlorinated dibenzofurans/dibenzo-p-dioxins (PCDF/PCDDs) and other dioxin-like substances in marine organisms from the Grenland fjords, S. Norway, 1975-2001: present contamination levels, trends and species specific accumulation of PCDF/PCDD congeners. Chemosphere. 2003, 52, 745–760. Oehme, M.; Bartonova, A.; Knutzen, J. Estimation of polychlorinated dibenzofuran and dibenzo-p-dioxin contamination of a coastal region using isomer profiles in crabs. Environ. Sci. Technol. 1990, 24, 1836–1841. ¨ .; Bucheli, T. D.; Ishaq, R.; Næs, K.; Persson, N. J.; Gustafsson, O Broman, D. Soot-carbon influenced distribution of PCDD/F in the marine environment of the Grenlandsfjords, Norway. Environ. Sci. Technol. 2002, 36, 4968–4974. Broman, D.; Na¨f, C.; Rolff, C.; Zebu ¨ hr, Y. Occurrence and dynamics of polychlorinated dibenzo-p-dioxins and dibenzofurans and polycyclic aromatic hydrocarbons in the mixed surface layer of remote coastal and offshore waters of the Baltic. Environ. Sci. Technol. 1991, 25, 1850–1864. Næs, K.; Axelman, J.; Na¨f, C.; Broman, D. Role of soot carbon and other carbon matrices in the distribution of PAHs among particles, DOC, and the dissolved phase in the effluent and recipient waters of an aluminium reduction plant. Environ. Sci. Technol. 1998, 32, 1786–1792. ¨ .; Bucheli, T. D.; Ishaq, R.; Næs, K.; Persson, N. J.; Gustafsson, O Broman, D. Distribution of PCNs, PCBs and other POPs together with soot and other organic matter in the marine environment of the Grenlandsfjords, Norway. Chemosphere. 2005, 60, 274– 283. Naes, K.; Oug, E. The significance of sediments for the contamination status in the Frierfjord and adjacent areas, Report 2570; Norwegian Institute of Water Research. 1991, (in Norwegian). Bandh, C.; Ishaq, R.; Broman, D.; Na¨f, C.; Ro¨nquist-Nii, Y.; Zebu ¨ hr, Y. Separation for subsequent analysis of PCBs, PCDD/ Fs, and PAHs according to aromaticity and planarity using a two-dimensional HPLC system. Environ. Sci. Technol. 1996, 30, 214–219. Guerzoni, S.; Rossini, P.; Sarretta, A.; Raccanelli, S.; Ferrari, G.; Molinaroli, E. POPs in the Lagoon of Venice: budgets and pathways. Chemosphere. 2007, 67, 1776–1785.

(32) Broman, D.; Na¨f, C.; Zebu ¨hr, Y. Long-term high- and low volume air sampling of polychlorinated dibenzo-p-dioxins and dibenzofurans and polycyclic aromatic hydrocarbons along a transect from urban to remote areas on the Swedish Baltic coast. Environ. Sci. Technol. 1991, 25, 1841–1850. (33) Lohmann, R.; Ockenden, W. A.; Shears, J.; Jones, K. C. Atmospheric distribution of polychlorinated dibenzo-p-dioxins, dibenzofurans (PCDD/Fs), and non-ortho biphenyls (PCBs) along a north -south Atlantic transect. Environ. Sci. Technol. 2001, 35, 4046–4053. (34) Lundgren, K.; Tysklind, M.; Ishaq, R.; Broman, D.; van Bavel, B. Flux estimates and sedimentation of polychlorinated naphthalenes in the northern part of the Baltic Sea. Environ. Sci. Technol. 2003, 126, 93–105. (35) Egeba¨ck, A.-L.; Wideqvist, U.; Ja¨rnberg, U.; Asplund, L. Polychlorinated naphthalenes in Swedish background air. Environ. Sci. Tehcnol. 2004, 42 (19), 4913–4919. (36) Ja¨rnberg, U.; Asplund, L.; de Wit, C.; Egeba¨ck, A.-L.; Wideqvist, U.; Jakobsson, E. Distribution of polychlorinated naphthalene congeners in environmental and source-related samples. Arch. Environ. Contam. Toxicol. 1997, 32, 232–245. (37) Harner, T.; Bidleman, T. F. Polychlorinated naphthalenes in urban air. Atmos. Environ. 1997, 31, 4009–4016. (38) Personal communication at the Norwegian Institute of Water Research (NIVA). (39) Isosaari, P.; Kankaanpa¨a¨, H.; Mattila, J.; Kiviranta, H.; Verta, M.; Salo, S.; Vartianen, T. Spatial distribution and temporal accumulation of polychlorinated dibenzo-p-dioxins, dibenzofurans, and biphenyls in the Gulf of Finland. Environ. Sci. Technol. 2002, 36, 2560–2565. (40) Kjeller, L.-O.; Rappe, C. Time trends in levels, patterns, and profiles for polychlorinated dibenzo-p-dioxins, dibenzofurans, and biphenyls in a sediment core from the Baltic proper. Environ. Sci. Technol. 1995, 29, 346–355. (41) Salo, S.; Verta, M.; Malve, O.; Korhonen, M.; Lehtoranta, J.; Kiviranta, H.; Isosaari, P.; Ruokoja¨rvi, P.; Koistinen, J.; Vartiainen, T. Contamination of River Kymijoki sediments with polychlorinated dibenzo-p-dioxins, dibenzofurans and mercury and their transport to the Gulf of Finland in the Baltic Sea. Chemosphere. 2008, 73, 1675–1683. (42) Hilscherova, K.; Kannan, K.; Nakata, H.; Hanari, N.; Yamashita, N.; Bradley, P. W.; Mccabe, J. M.; Taylor, A. B.; Giesy, J. P.

(43)

(44) (45)

(46)

(47) (48)

(49)

(50)

(51)

Polychlorinated dibenzo-p-dioxin and dibenzofuran concentration profiles in sediments and flood-plain soils of the Tittabawassee River, Michigan. Environ. Sci. Technol. 2003, 37, 468–474. Birch, G. F.; Harrington, C.; Symons, R. K.; Hunt, J. W. The source and distribution of polychlorinated dibenzo-p-dioxin and polychlorinated dibenzofurans in sediments of Port Jackson, Australia. Mar. Pollut. Bull. 2007, 54, 295–308. Buekens, A.; Stieglitz, L.; Hell, K.; Huang, H.; Segers, P. Dioxins from thermal and metallurgical processes: recent studies for the iron and steel industry. Chemosphere. 2001, 42, 729–735. Buerskens, J. E. M.; Toussaint, M.; de Wolf, J.; van der Steen, J. M. D.; Slot, P. C. S.; Commandeur, L. C. M.; Parsons, J. R. Dehalogenation of chlorinated dioxins by an anaerobic microbial consortium from sediment. Environ. Toxicol. Chem. 1995, 14, 939–943. Huang, C.-L.; Harrison, K. H.; Madura, J.; Dolfing, J. Gibbs free energies of formation of PCDDs: evaluation of estimation methods and application for predicting dehalogenation pathways. Environ. Toxicol. Chem. 1996, 15, 824–836. DHI. 210Pb-dating of three sediment cores from the Frierfjord, Norway, DHI report 91340, 2001, (in Danish). Na¨f, C.; Broman, D.; Pettersen, H.; Rolff, C.; Zebu ¨ hr, Y. Flux estimates and pattern recognition of particulate polycyclic aromatic hydrocarbons, polychlorinated dibenzo-p-dioxins, and dibenzofurans in the waters outside various emission sources on the Swedish Baltic coast. Environ. Sci. Technol. 1992, 26, 1444–1457. Hanberg, A.; Waern, F.; Asplund, L.; Haglund, E.; Safe, S. Swedish dioxin survey: Determination of 2,3,7,8-TCDD toxic equivalent factors for polychlorinated biphenyls and naphthalenes using biological tests. Chemosphere. 1990, 20, 1161–1164. Villeneuve, D. L.; Kannan, K.; Khim, J. S.; Falandysz, J.; Nikiforov, V. A.; Blankenship, A. L.; Giesy, J. P. Relative potencies of individual polychlorinated naphthalenes to induce dioxin-like responses in fish and mammalian in vitro bioassays. Arch. Environ. Contam. Toxicol. 2000, 39, 273–281. Van den Berg, M.; Birnbaum, L.; Bosveld, A. T. C.; Brunstro¨m, B.; Cook, P.; Feeley, F.; Giesy, J. P.; Hanberg, A.; Hasegawa, R.; Kennedy, S. W.; et al. Toxic equivalency factors (TEFs) for PCBs, PCDDs, PCDFs for humans and wildlife. Environ. Health Perspect. 1998, 106 (12), 775–792.

ES8011595

VOL. 43, NO. 10, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

3447