Retrospective Monitoring of Organotin Compounds in Marine Biota

(TBT) and triphenyltin (TPT) as well as their degradation products were quantified. Biota samples from North Sea and. Baltic Sea areas were analyzed b...
11 downloads 0 Views 115KB Size
Download PDF
Environ. Sci. Technol. 2003, 37, 1731-1738

Retrospective Monitoring of Organotin Compounds in Marine Biota from 1985 to 1999: Results from the German Environmental Specimen Bank HEINZ RU ¨ DEL,* PETER LEPPER, AND JU ¨ RGEN STEINHANSES Fraunhofer Institute for Molecular Biology and Applied Ecology (Fraunhofer IME), 57377 Schmallenberg, Germany CHRISTA SCHRO ¨ TER-KERMANI German Federal Environmental Agency (Umweltbundesamt), 14195 Berlin, Germany

In archived samples from the German Environmental Specimen Bank, organotin compounds including tributyltin (TBT) and triphenyltin (TPT) as well as their degradation products were quantified. Biota samples from North Sea and Baltic Sea areas were analyzed by gas chromatography/ atomic emission detection-coupling after extraction and Grignard or ethylborate derivatization. TBT and TPT were detected in nearly all samples. A decrease of TPT contamination was observed in bladder wrack, common mussels, and eelpout muscle tissues in the period 19851999. In this period, TPT concentrations in North Sea mussels decreased from 98 to 7 ng/g (as organotin cation concentration in wet tissue). Concentrations of TBT remained relatively constant with 17 ( 3 ng/g for mussels from a site with nearby marine traffic and 8 ( 2 ng/g for a more remote area. The results reflect that TBT is still used as a biocide in antifouling paints whereas the use of TPT as a co-toxicant in such preparations had been ceased in the 1980s. The fact that the use of TBT in antifouling paints was banned in 1991 for small boats within the European Community seems not to have resulted in a decrease of TBT levels in marine biota.

Introduction The German Environmental Specimen Bank (ESB) is part of the ecological observation program in Germany that is financed by the Federal government (1, 2). It serves to recognize impending faulty developments in ecosystems, to identify the type and extent of possible damage, to supply knowledge for priority setting of political measures, and to work out fundamental concepts for precautionary policy. Detailed information on the ESB and analytical data are available from the German Federal Environmental Agency via the Internet (3). The greatest potential of the ESB is the possibility to conduct retrospective monitoring. The archive permits specimens to be retrieved for analysis at a later date, in case unexpected questions arise or more sensitive analytical methods become available. * Corresponding author phone: +49 2972 302 301; fax: +49 2972 302 319; e-mail: [email protected]. 10.1021/es026059i CCC: $25.00 Published on Web 03/26/2003

 2003 American Chemical Society

Here, we report the analysis of the organotin compounds tributyltin (TBT), dibutyltin, (DBT), monobutyltin (MBT), triphenyltin (TPT), diphenyltin (DPT), monophenyltin (MPT), tricyclohexyltin (TCT), dioctyltin (DOT), monooctyltin (MOT), and tetrabutyltin (TTBT) in a set of archived ESB samples from the period 1985-1999. The aim of this study was to elucidate retrospectively the levels of organotin compounds in marine organisms during the 1990s. The main objective was to investigate whether imposed reduction measures resulted in decreased levels of TBT and its degradation products. A further objective was to assess whether the detected levels of organotin compounds are of ecotoxicological relevance for exposed organisms. Organotin compounds have been proven to be of high toxicological relevance and are considered as dangerous for the environment. Detailed information on these compounds was presented in a review by Fent (4) and in a monograph edited by de Mora (5). Most important in relation to exposure and effect strength are the butyl- and phenyltin compounds. Triorganotin compounds are especially toxic because they disturb the cell energy metabolism (4, 6). Representative effect concentrations for marine organisms are 0.36 µg/L TBT chloride () 0.32 µg/L TBT calculated as organotin cation) for growth inhibition of marine algae Skeletonema costatum (EC50 after 72 h exposure; 7) and 0.97 µg/L TBT chloride () 0.86 µg/L TBT) for lethal effects on common mussels Mytilus edulis (LC50 after 66 d; 8). The 6-d no observed effect concentration (NOEC) for the marine copepod Acartia tonsa was 0.011 µg/L bis(tri-n-butyltin)oxide (TBTO; ) 0.011 µg/L TBT); for the mollusk Mercenaria mercenaria, a NOEC of 0.010 µg/L TBTO () 0.010 µg/L TBT) for growth inhibition during a 14-d period from fertilization to metamorphosis was detected (as cited in ref 9). Additionally, it was demonstrated that some organotin compounds act as endocrine disrupters, at least in sensitive mollusks. The lowest effect concentration for imposex effects (i.e., the development of females with male reproductive organ and impaired reproductive capacities) in dogwhelks (Nucella lapillus) was approximately 0.5 ng/L TBT-Sn () 1.2 ng/L TBT; 10). Female ramshorn snails (Marisa cornuarietis) developed imposex with 10% of the maximum effect (EC10) after 4 months at approximately 12 ng/L TPT-Sn () 35 ng/L TPT) and reduced fecundity at approximately 6 ng/L TPT-Sn () 18 ng/L TPT; 11). For humans, the lowest documented toxicological end point for a TBT compound is a depression of the immune system of the thyroid gland by TBTO. The estimated tolerable daily intake (TDI) value derived by the World Health Organization (WHO) for TBTO is 0.25 µg per kg body weight (9). For TPT acetate or hydroxide, the respective TDI is 0.4 µg per kg body weight, as derived by the German Federal Institute for Health Protection of Consumers and Veterinary Medicine (12). In 1996, the annual world production of organotin compounds was approximately 40 000 t (13). The major use (76%) is as stabilizers in plastics, for example, DBT in poly(vinyl chloride). Furthermore, approximately 18% are used as biocides, mainly as active ingredients in antifouling paints (approximately 4000 t/yr; 13). A major portion of this fraction reaches the marine environment during normal use. In antifouling preparations, mainly TBT is used. However, until approximately 1985, TPT also was used in antifouling preparations as a co-toxicant (14). Persistence of organotin compounds in the environment is governed by moderate to fast aerobic biotic degradation processes, slow anaerobic biotic degradation, slow abiotic VOL. 37, NO. 9, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

1731

TABLE 1. Marine Sampling Areas and Sample Types of the Environmental Specimen Bank Analyzed in This Studya organism; sampled tissue

functional/ trophic level

Fucus vesiculosus (common primary producer bladder wrack, brown algae); forked thallus

Mytilus edulis (common mussel); soft body without shell

sampled material (per sampling)

sampling pooled samples from 6 samplings/yr (every 2 months)

approx 3-5 kg

consumer (primary; pooled samples from 6 approx 2500 filter feeder) samplings/yr (every 2 mussels months) for North Sea or from 2 samplings/yr (every 6 months) for Baltic Sea

sampling areas North Sea I

sampling sites Eckwarderho¨ rne

North Sea II List Ko¨ nigshafen (since 1992) or List, south of harbor (until 1992) North Sea I Eckwarderho¨ rne

Zoarces viviparus (eelpout); muscle tissue

consumer (secondary)

sampled in May or June

North Sea II List Ko¨ nigshafen (since 1992) or List, south of harbor (until 1992) Baltic Sea Darsser Ort approx 100 fishes North Sea I Jadebay

Larus argentatus (herring gull); egg content

consumer (top predator)

sampled in May or June

approx 30 eggs

Baltic Sea North Sea I

Darsser Ort Mellum

North Sea III Trischen a

Organisms represent different levels of the marine trophic system.

degradation by photolysis, and fast but reversible adsorption/ desorption processes (15-17). The environmental half-lives of organotin compounds are in the range of days in the water phase, but they are persistent in anaerobic sediments (15, 16). Organotin compounds are ubiquitously distributed in marine organisms, for example, in the North Sea (18), Baltic Sea (19), Bering Sea, Gulf of Alaska and Japan Sea (20), and Pacific Ocean (21-23). In a monitoring study in the Mediterranean Sea performed in 1988, Tolosa et al. (24) reported for the first time the occurrence of TPT in marine biota as well as in seawater and sediment. Generally, high concentrations are found particularly in the proximity of sources, for example, commercial ports, shipyards, and major shipping routes. Recently, TBT and unexpectedly high levels of TPT were detected in deep-sea organisms from the Mediterranean Sea. This finding confirms a possible long-range transport of organotin compounds toward deep-sea ecosystems (25). Reported bioconcentration factors (BCF) for TBT in marine mussels (M. edulis) are between 5000 and 10 400 (26, 27) and between 9400 and 11 000 for several marine fish species (Pagrus major, Rudarius ercodes, Mugil cephalus; 28). BCFs for TPT are 36 000 in M. edulis (27) and 3100-4100 in P. major and R. ercodes (29). Organotin traces are detectable in different tissues. A correlation between TBT or TPT concentrations and lipid content could not be established (30, 25). In the early 1990s, restrictions for the use of TBT in antifouling paints were introduced. First, measures were imposed by France in 1982 when negative effects of TBT on oyster populations, such as shell deformations, were observed (31). More countries followed with bans of TBT-based antifouling coatings for small boats (United Kingdom in 1987, United States in 1988, European Community in 1990; 32). Nowadays, these restrictions for small boats are implemented in most developed countries, and a general phasing out of organotin compounds was agreed on at the International Maritime Organisation (IMO) meeting in October 2001 (33). The resolution called for a global prohibition on the application of organotin compounds that act as biocides in antifouling systems on ships by January 1, 2003, and a complete prohibition by January 1, 2008. 1732

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 9, 2003

Experimental Section Sampling of ESB Material. Within the framework of the ESB in Germany, representative environmental samples are collected, homogenized, and archived in form of subsamples of approximately 10 g. Sample organisms are from different trophic levels of terrestrial, limnic, and marine ecosystems being characteristic for Germany. Immediately after being sampled, the material is stored in the gas phase above liquid nitrogen at temperatures below -150 °C. The inert gas layer of evaporated nitrogen and the low temperature rule out, as far as possible, changes in state or loss of chemical characteristics over a period of decades. Currently, about 850 different samples with up to 200 subsamples are stored in the ESB (135 000 samples in total). The ESB is a long-term project, which requires a high degree of continuity in all work steps. Therefore, all important procedures are performed according to ESB standard operating procedures (SOPs; 34). The applied methods have been described in detail previously: sampling (35, 36), cryomilling (37), homogenization (37), particle size characterization (37), and cryostorage (38). Table 1 lists the three North Sea and one Baltic Sea sampling sites and the respective organisms analyzed in this study. The North Sea sampling areas (Figure 1) are located in the Wadden Sea, which is an unique ecosystem with high biodiversity and productivity, especially in the sediment. It consists of sand and mud flats along the North Sea coast that fall dry during low tide. The Baltic Sea is a low-salinity water body with an area of approximately 400 000 km2. The ESB sampling area is located seawards off the peninsula of Zingst and Darss near the island of Ru ¨ gen (Figure 1). The sampling sites are situated in or adjacent to national parks or biosphere reserves and are not directly influenced by major shipping routes, larger harbors, or shipyards. Only the North Sea site I (Figure 1) is partly influenced by industry, dumping of dredged sediment from the Weser into the Jadebay and landing of oil at a nearby harbor (Wilhemshaven). Analytical Methods. For the quantification of the following organotin compounds either the Grignard or the ethylborate method was applied: TTBT (CAS Registry No. 1461-25-2), TBT chloride (CAS Registry No. 1461-22-9), DBT dichloride

FIGURE 1. Map of sampling areas at the German coast; sampling sites are marked by circles. North Sea sites: I, Jadebay, Eckwarderho1 rne, Mellum; II, List (Sylt); III, Trischen. (CAS Registry No. 683-18-1), MBT trichloride (CAS Registry No. 1118-46-3), TPT chloride (CAS Registry No. 2279-76-7), DPT dichloride (CAS Registry No. 1135-99-5), MPT trichloride (CAS Registry No. 1124-19-2; this substance could only be determined by applying the ethylborate method), TCT chloride (CAS Registry No. 3091-32-5), DOT dichloride (CAS Registry No. 3542-36-7), and MOT trichloride (CAS Registry No. 3091-25-6). Tripropyltin chloride, diheptyltin dichloride, and monoheptyltin trichloride were applied as internal standards to the fresh material during digestion in order to check the derivatization for each alkylation level. Standards were purchased from ABCR, Acros, Aldrich, or Witco GmbH and had purities of 95-99%. For the analytical methods, standard operating procedures were elaborated, which were followed for all samples (39). The Grignard method was optimized on base of protocols from refs 22 and 40. Routinely 1 g of the biological material was digested with 5 mL of tetramethylammonium hydroxide solution (25% in water) at 60 °C for 1 h. After the addition of 10 mL of purified water (deionized and afterward doubledistilled in a quartz system), the acidification with 5 mL of glacial acetic acid, and the addition of 6 g of sodium chloride, an extraction using 10 mL of hexane with 0.05% tropolone (2-hydroxy-2,4,6-cycloheptatrienon) as complexant was performed by stirring the mixture for 1 h at >1000 rpm. After centrifugation for phase separation, the organic layer was dried with 2 g of sodium sulfate and mixed with 500 µL of Grignard reagent (n-pentylmagnesium bromide in diethyl ether; prepared from freshly distilled n-pentylbromide and magnesium turnings) for derivatization. Five milliliters of ice water and 1 mL of sulfuric acid (50%) were added after 10 min, and the mixture was shaken thoroughly. The organic layer was dried with 2 g of sodium sulfate and concentrated to 1-2 mL by rotary evaporation. The organic phase was cleaned up by chromatography over layers of 2 g of sodium sulfate, 3 g of silica gel (100-200 mesh, 3% water; from ICN), and 3 g of Florisil (reagent grade, 60-100 mesh, 2.5% water; from Aldrich or GALAB) in a glass column (30 cm × 1 cm diameter). After being conditioned with 30 mL of n-hexane, the sample was applied. The analytes were eluted with 40 mL of a mixture of 5% (vol/vol) acetone in n-hexane. The eluate was concentrated by rotary evaporation to 1 mL and afterward by nitrogen purging to approximately 100 µL. For the ethylborate protocol, the tetramethylammonium hydroxide digest was prepared as described for the Grignard

method. After the addition of 10 mL of purified water, the acidification with 5 mL of 10% hydrochloric acid, and the addition of 6 g of sodium chloride, an extraction with 20 mL of n-hexane containing 0.05% (wt/vol) sodium diethyldithiocarbamate was performed by strongly shaking the mixture. After 1 h, the phases were separated by centrifugation for 15 min. The organic layer was mixed with 10 mL of ethanol, 10 mL of purified water, 2 mL of acetate-buffered solution (pH 4.5), 5 g of sodium chloride, and 500 µL of the ethylborate reagent (prepared by dissolving 5 g of sodium tetraethylborate, 98% from ACBR, in 50 mL of tetrahydrofurane). The mixture was shaken thoroughly for 30 min, and then the organic layer was separated by centrifugation and dried by the addition of 2 g of sodium sulfate. The organic phase was cleaned up by chromatography over 2 g of sodium sulfate (and additionally 3 g of Florisil for egg sample material) in a glass column, as described for the Grignard protocol. Quantification was performed by capillary gas chromatography and atomic emission detection at the tin specific wavelengths of 270.651 and 303.419 nm. Results from both methods were equivalent. However, MPT could only be analyzed by the ethylborate method. Data are reported as nanogram per gram of organotin cation (if not otherwise mentioned). Quality Assurance. Beside each set of actual samples, appropriate blank samples were analyzed in order to identify possible contamination problems. Limits of determination (LODs) were estimated from the precision of the calibration, based on the concept of the respective German standard (41). LODs varied depending on compound and matrixes (actual values are listed in the tables). For all sample materials, fortification experiments were performed with additions of organotin compounds corresponding to concentration increases of 25-40 ng/g (equivalent to typical contamination levels). The recoveries were in the range of 87-121% for TBT and 74-137% for TPT (n ) 12; all matrixes). For organotin compounds, only one biological reference material was commercially available. The “mussel tissue CRM 477” (from BCR; 42) with certified values for TBT (2.20 mg/ kg), DBT (1.54 mg/kg), and MBT (1.50 mg/kg) was used for method validation. Recoveries for the Grignard (ethylborate) method were 104% (97%) for TBT, 70% (84%) for DBT, and 152% (158%) for MBT. The lower recovery values for DBT and higher values for MBT may reflect a degradation of DBT (possibly a stability problem as observed for the phenyltin compounds in the same material; 42). However, since concentrations levels of the reference material were by factors of 100-1000 above the concentrations in the specimens analyzed here, the results were only of limited value.

Results and Discussion Monitoring Results. For this investigation bladder wrack (Fucus vesiculosus, 16 samples, 1985-1996), common mussel (M. edulis, 31 samples, 1985-1999), eelpout (Zoarces viviparus, 9 samples, 1994-1998), and sea gull eggs (Larus argentatus, 6 samples, 1994-1998) from North Sea and Baltic Sea areas (Table 1) were analyzed. TBT and TPT were found in nearly all samples whereas DBT, DPT, MBT, and MOT were detected only in few samples. The concentrations of the further analyzed organotin compounds TTBT, DOT, and TCT were below the LOD in all samples. In North Sea bladder wrack TBT, TPT, DBT, and DPT were detected in samples from Eckwarderho¨rne but not in those from List near the island of Sylt (Table 2; period 19851996). At this site with few marine traffic, only TBT was detected. At Eckwarderho¨rne, observed TPT levels (8-14 ng/g wet wt) were relatively high in relation to TBT values (3-5 ng/g) in the 1980s but decreased to figures below the LOD in the 1990s. In two samples with TPT levels g10 ng/g, DPT was detected too. The TBT concentrations showed no VOL. 37, NO. 9, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

1733

TABLE 2. Organotin Compounds in Bladder Wrack and Common Mussels from North Sea and Baltic Seaa TBT (ng/g)

year

DBT (ng/g)

DPT (ng/g)

TPT (ng/g)

∑Sn (ng/g of Sn)

TBT (ng/g)

Bladder Wrack North Sea I (Eckwarderho1 rne)