- Email: [email protected]
Occurrence of organotin compounds in house dust in Berlin (Germany)
Chemosphere 58 (2005) 1377–1383 www.elsevier.com/locate/chemosphere
Occurrence of organotin compounds in house dust in Berlin (Germany) H. Fromme a
a,*
, A. Mattulat b, T. Lahrz c, H. Ru¨den
d
Bayerisches Landesamt fu¨r Gesundheit und Lebensmittelsicherheit, Veterina¨rstrasse 2, D-85764 Oberschleißheim, Germany b PiCA-Pru¨finstitut Chemische Analytik GmbH, 12489 Berlin, Germany c Institut fu¨r Lebensmittel, Arzneimittel und Tierseuchen Berlin, 10557 Berlin, Germany d Institut fu¨r Hygiene und Umweltmedizin, Charite´—Universita¨tsmedizin Berlin, 12203 Berlin, Germany Received 19 January 2004; received in revised form 2 August 2004; accepted 29 September 2004
Abstract In a study in the year 2000 on the occurrence of hazardous environmental contaminants house dust samples from 28 Berlin apartments were measured for the presence and concentrations of six organotin compounds, monobutyltin (MBT), dibutyltin (DBT), tributyltin (TBT), monooctyltin (MOT), dioctyltin (DOT) and triphenyltin (TPT). The concentrations of MBT and DBT determined ranged considerably from 0.01 mg kg 1 to 1.5 mg kg 1 (median: 0.05 mg kg 1) and 0.01 to 5.6 mg kg 1 (median: 0.03 mg kg 1), respectively. Maximum levels of TBT and MOT were only 0.08 mg kg 1 and 0.04 mg kg 1. The maximum total value of the organotins was 7.2 mg kg 1 (median: 0.11 mg kg 1). MBT was found in 86% and DBT in 82% of the samples above the limit of quantification, TBT and MOT only in 50% and DOT in 43%. The focus of ecotoxicology is on the risks arising from organotin compounds (especially butyltins) when used as biocides in antifouling paints. TBT acts as an endocrine disrupter in animals, inducing masculinization (imposex) in female gastropods of different species by increasing testosterone levels. The most critical organ site in experimental animals is the cellular immune system, where lymphocyte depletion in the thymus and peripheral lymphoid tissues takes place. Our study does not provide data on the basis of which population exposure could be estimated; house dust containing harmful organotins could, however, under some conditions, become a relevant intake possibility for young children. 2004 Elsevier Ltd. All rights reserved. Keywords: Organotin; Butyltin; Octyltin; Phenyltin; DBT; TBT; House dust; Indoor
1. Introduction Organotins are a large class of compounds which differ markedly in their properties and have been used for a
Rn SnX3 n ;
*
Corresponding author. Fax: +49 89 315 60 425. E-mail address: [email protected] Fromme).
variety of purposes since their discovery and initial use in the 1920s. They are organic derivatives of tetravalent tin and characterized by the presence of at least one covalent carbon–tin bond. They have the general structure
(H.
where R is an alkyl or aryl group, Sn the central tin atom in the oxidation state +4 and X a singly charged
0045-6535/$ - see front matter 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2004.09.092
1378
H. Fromme et al. / Chemosphere 58 (2005) 1377–1383
anion or an anionic organic group. Most importantly, methyl, butyl, octyl and phenyl moieties are contained in the organic substituents of all compounds of any practical relevance, whereas the anionic moiety usually comprises chloride, fluoride, oxide, hydroxide, carboxylate or thiolate. Under normal conditions their physicochemical properties appear as only slightly volatile, insoluble in water and with a strong tendency to absorb particulate matter (KEMI, 2000). Of the various organotins, the mono-, di- and trisubstituted compounds are the most widely employed, the first two (notably mono- and di-methyltin, butyltin and octyltin compounds) being used extensively as heat and light stabilizers in the manufacture of polyvinyl chloride (PVC) polymers and as industrial catalysts for polyurethane and silicone elastomers. Thus they are present in water pipes, food packing materials (e.g., dioctyltin), glass coatings (e.g., butyltin trichloride), polyurethane foams and many other consumer products. Another less extensive use of organotins is in industrial and agricultural biocides (fungicides, bactericides, miticides, insecticides), surface disinfectants and preservatives for wood, paper, textiles, paints and some electrical equipment. On a world wide basis tri-substituted organotins (notably tributyltin oxide and tributyltin methacrylate) are used as antifouling agents in paints for boats and ships, thereby accessing the aquatic environment (WHO, 1990; US EPA, 2002). It can be estimated that their most frequent uses are as PVC stabilizers (approximately 16 000 ton annually in Europe) and as catalysts (1300–1650 ton), thus accounting in each case for approximately 67% and 8% of total organotin consumption. About 2–4% of PVC production is ascribable to this type of stabilizer. Biocidal uses account for nearly 20% of the market and all other uses (including latex paint preservatives, anthelminthics and coccidiostats) for 5% (US EPA, 2002). The prominent toxicological feature of the organotins is their immunotoxicity, an effect produced by di- and trialkyltins as well as triphenyltins. The acute toxicity after dermal or oral application of most organotins is low to moderate (depending on routes and species, for DBT 58–5480 mg kg 1 bw), though local effects possibly occurring are eye and skin irritation or sensitization at the site of contact (WHO, 1990; ATSDR, 1992; WHO, 1999a,b; KEMI, 2000). The focus of ecotoxicology is on risks from the use of organotin compounds (especially butyltins) as biocides in antifouling paints. TBT acts as an endocrine disrupter in animals and induces masculinization (imposex) by increasing testosterone levels in different species of female gastropods (Fent, 1996). In experimental animals TBT produced endocrine tumors, although no evidence of any TBT mutagenicity is known (Wester et al., 1990). In vitro studies found that butyltins inhibit human cytochrome P450 aromatase and 5a-reductase activity and might therefore have
a distinct impact on human hormone metabolism (Doering et al., 2002).
2. Material and methods The present study was conducted for the identification of hazardous chemicals in the indoor environment of apartments in a particular urban area as part of a wider investigation (Fromme et al., 2004a,b). The aim was to measure the most significant organotins produced in industrial processes and present in house dust. That is, monobutyltin (MBT), dibutyltin (DBT), tributyltin (TBT), monooctyltin (MOT), dioctyltin (DOT), and triphenyltin (TPT). It is well known that house dust is a receptor of and repository for chemicals in the home and thereby a potential source of exposure to them, especially for small children and babies, who spend significant portions of their time on floors and engaging in frequent mouth contact activities (Butte and Heinzow, 2002). The concentrations of organotin compounds in household dust were determined for a total of 28 residences in the year 2000. The apartments were built between 3 and 104 years before and had areas of 51–250 m2.
3. Sampling procedure and analysis The house dust was taken from the bags of the vacuum cleaners in regular use for cleaning the apartments. Inhomogeneity was avoided by only taking the fine dust from the space between the inner and outer paper layers of the bags. Any larger dust particles, fibres and debris, which might otherwise have distorted the results, were thus excluded. The dust was put directly to clean aluminium foil and then transferred to glass bottles for storage. Information reported in the literature indicates that this fine dust contains significant quantities of persistent organic pollutants and is therefore a good indicator of human exposure levels (Lewis et al., 1999). The determination of organotin compounds was carried out according to German standard methods DIN 38407-13 (DIN, 1999). For analysis 0.5–1.0 g of dust samples were spiked with internal standards (monoheptyltin, diheptyltin, tripropyltin from Promochem) and extracted with ethanol using an ultrasonic method. An aliquot was buffered with sodium acetate to pH 4.5 and alkylated with sodium tetraethylborate (STEB). Subsequently, the tetra-alkylated tin compounds were extracted with n-hexane and analysed by gaschromatography/mass spectrometry on an HP 6890/5973 GC/MS system with a CTC autosampler Combi PAL. Separation was performed using a 30 m · 0.25 mm i.d. DB-5 fused silica capillary column with a 0.5 lm film thickness. The injector (Gerstel, cold injection system
H. Fromme et al. / Chemosphere 58 (2005) 1377–1383
blanks with every sample sequence. There was no reference material available for organotins in house dust.
Table 1 Ions used for quantification and qualification of organotin compounds Compound
Mass used for quantification (m/z)
Qualifier masses (m/z)
MBT DBT TBT MOT DOT TPT
235 263 291 291 375 351
147, 179, 179, 179, 263, 197,
MHT DHT TPrT
277 347 249
179, 275 249, 345 193, 235
1379
4. Results and discussion
179 205 289 289 291 349
Table 2 shows the statistical parameters of organotin measurements in house dust. MBT and DBT were found in 86% and 82% of the samples above the detection limit, TBT and MOT only in 50% and DOT only in 43%. The concentrations of MBT and DBT determined ranged, respectively from 0.01 to 1.5 mg kg 1 (median: 0.045 mg kg 1) and 0.01 mg kg 1 to 5.6 mg kg 1 (median: 0.03 mg kg 1). Median levels of TBT and MOT were only 0.01 mg kg 1. The maximum value of the total organotins (using only values above the detection limit) was 7.2 mg kg 1 (median: 0.1 mg kg 1). A considerable amount of data exists on the distribution of organotins (especially TBT) in the aquatic environment, though much less information is available regarding the distribution of these compounds in other locations, the indoor environment for example. It must be understood that house dust is a very heterogeneous mixture whose composition depends on many different factors, such as the type and construction of the building, the use of the rooms in question, types of furnishing and flooring materials present, heating and ventilation systems, etc. It is not surprising that results published in the literature (see Table 3 and Fig. 1) vary widely. From 1998 to 2000 Kersten and Reich (2003) examined dust from 50 apartments in Hamburg and found median concentrations of 1.4 mg kg 1 (MBT) and 0.2 mg kg 1 (DBT). In studies conducted by the Greenpeace Research Laboratories in parliament buildings of 8 European countries and in private apartments, notably in the UK, total levels of organotins of between 0.14 mg kg 1 and 5.81 mg kg 1 were discovered (Santillo et al., 2001, 2003). In apartments these compounds ranged from 1.6 mg kg 1 to 5.1 mg kg 1, slightly higher than the range previously reported by the same study group for dust from parliament buildings (0.5–3.5 mg kg 1;
Internal standards: monoheptyltin (MHT), diheptyltin (DHT) and tripropyltin (TPrT).
CIS 4) was operated in splitless mode, started at 100 C and heated to 280 C at a rate of 12 C/s (held for 5 min) with 2 ll injection volume. The GC oven temperature was kept at 50 C for 2 min and then increased to a temperature of 230 C at a rate of 10 C/min, 280 C at a rate of 4 C/min and 345 C at a rate of 10 C/min, then sustained for 15 min. The transferline was held at 320 C. The MS was operated in the single ion monitoring (SIM) mode with 6–15 ions per group and a dwell time of 30 ms. Identifications were based on the GC retention times relative to those of corresponding internal standards and the relative abundance of the ions monitored. Quantifications were based on comparisons of the integrated ion responses of the target ions which those of the corresponding internal standards and by using average response factors of the target analytes generated from standard calibration. The ions used for quantification and qualification are listed in Table 1. The limit of quantification of the target compounds was 0.01 mg kg 1 for each analyte, based on a minimum signal/noise ratio of 5:1. The recovery was >70% at a concentration of 0.01 mg kg 1, tested by standard addition to dust samples. The methods quality was checked by analysis
Table 2 Statistical characteristics of organotin compounds in the house dust of apartments (mg kg 1)
N N > d.l. Mean Standard deviation Median 90th Percentile 95th Percentile Maximum
MBT
DBT
TBT
MOT
DOT
Total organotina
28 24 0.16 0.32 0.05 0.35 0.70 1.50
28 23 0.51 1.31 0.03 1.03 3.28 5.60
28 14 0.02 0.02 0.01 0.05 0.07 0.08
28 14 0.01 0.01 0.01 0.03 0.03 0.04
28 12 0.02 0.07 0.01 0.02 0.03 0.36
28 26 0.77 1.66 0.11 1.79 4.15 7.18
d.l.: detection limit; values under the d.l. are used with half of the d.l. a Sum of values above the detection limit.
1380
Table 3 Concentrations of organotin compounds in household dust and floorings in mg kg DBT
TBT
MOT
DOT
TPT
Total organotinsa
<0.1–24.0 (0.53) 0.18–2.39 (1.05) 0.012–81.8 (0.41) 0.81–2.80 (1.35) 0.08–1.20 (0.65) 0.1–18.0 (1.4) 0.005–1.50 (0.05)
<0.1–2200 (3.40) 0.17–0.89 (0.46) 0.005–130.3 (0.25) 0.16–1.30 (0.52) 0.03–0.33 (0.22) 0.01–5.6 (0.2) <0.005–5.60 (0.03)
<0.1–15.0 (0.46) 0.004–0.19 (0.022) 0.005–2.03 (0.009) 0.02–0.76 (0.05) 0.004–0.16 (0.03) <0.001–0.2 (0.03) <0.005–0.08 (0.008)
–
0.015–0.83 (0.26) 0.005–0.98 (0.02) 0.08–1.30 (0.35) 0.02–0.91 (0.14) 0.01–2.8 (0.2) <0.005–0.04 (0.008)
0.004–0.14 (0.06) 0.005–2.12 (0.02) 0.02–0.55 (0.06) 0.003–3.60 (0.05) <0.001–0.2 (0.01) <0.005–0.36 (0.005)
<0.001
1.3–2222 (33.1) 0.49–3.48 (1.78) 0.06–212 (0.91) 1.58–5.05 (2.43) 0.14–5.81 (1.36) –
<0.1–70.0 (5.5) <0.5–38.0 (6.1) 0.33–48.8 (2.62)
<0.1–920.0 (120.0) <0.5–210 (12.0) 37.7–569 (200)
<0.1–3.40 (<0.1) <0.5–1.2 (0.25) 0.13–17.9 (4.39)
<100–30.0e
<0.1–6.50f
–
<0.5–91 (7.1) <0.05–0.98 (0.26)
<0.5–276 (36) 0.11–10.2 (1.56)
<0.5
Allsopp et al. (2001) Allsopp et al. (2001)
6
<0.0003–0.009
<0.0003–0.03
<0.0003–0.007
<0.0004
<0.0004
<0.0003
0.002–0.042
2
0.50/1.14
0.10/7.20
2.70/47.50
<0.0004/ <0.0004
<0.0004/ <0.0004
<0.0003/ <0.0003
3.32/56.0
a
33 15 134 100c 10 50 28
33 35 5
Only values above d.l. Only 1 sample >d.l. c Partly pooled samples from households of UK regions. d Only 2 samples >d.l. e Only 3 samples >d.l. f Only 10 samples >d.l. g Carpets treated against bacteria, moulds, fungi and dust mites. b
0.009b <0.001–0.07 (<0.001) <0.001–0.04d <0.001–0.02 (0.004) <0.005
<0.05
0.01–7.18 (0.10)
1.5–997 (245) 0.2–381 (78) 51–592 (256)
Apartments in south Germany Parliament buildings in Europe Apartments and large buildings, Netherlands Samples from all UK regions Samples from 5 European countries Apartments in Hamburg, Germany Apartments in Berlin, Germany
PVC floorings in apartments, Germany PVC floorings, Germany PVC floorings from various suppliers, UK Carpets from various suppliers, UK Carpets from various suppliers, UKg
H. Fromme et al. / Chemosphere 58 (2005) 1377–1383
Flooring materials Hagenau (2001) Mattulat (2003) Allsopp et al. (2000)
(ranges; values in brackets are medians)
MBT
N Household dust Hagenau (2001) Santillo et al. (2001) Greenpeace (2001) Santillo et al. (2003) Santillo et al. (2003) Kersten and Reich (2003) Our study
1
H. Fromme et al. / Chemosphere 58 (2005) 1377–1383
1381
Median min.
m g/kg
0.01
0.1
m a x.
1.0
10
100
1000
Hagenau 2001 Santillo et al. 2001
Santi llo et al. 2003 (UK r egions) Santillo et al. 2003 (Europe) Kersten & Reich 2003 Greenpeace 2001
Our study
Fig. 1. Comparison of MBT (light bars) and DBT (dark bars) levels in household dust samples described in the scientific literature.
Santillo et al., 2001). The most abundant organotin was MBT followed by DBT and MOT. In another study, performed by Greenpeace Netherlands (Greenpeace, 2001), dust samples were collected in large buildings and 134 apartments all over the Netherlands. The median concentration was 0.41 mg kg 1 (MBT) and 0.25 mg kg 1 (DBT). The total organotin content varied in this investigation between 0.06 mg kg 1 and 212 mg kg 1. Very high concentrations of organotins were reported by Hagenau (2001), who measured 33 samples with concentrations from 1.3 to 2222 mg kg 1 in German apartments. An explanation for these larger amounts might be that he used a vacuum cleaner with a filter holder for sampling. Hagenau (2001) also found a significant correlation between organotin levels in household dust and PVC materials. The examinations carried out to date have shown generally wide variations in indoor organotin levels, a fact which may merely reflect the variability of their presence in the form of stabilizer additives in plastics. Previous studies (Allsopp et al., 2000, 2001; Hagenau, 2001; Mattulat, 2003) also discovered the widespread presence of organotins in carpets and floor coverings with levels of up to 1000 mg kg 1 (Table 3). It is to be expected that other products (plastic crib rail teethers, clothing, floor tiles, wall paper, handbags, mattress covers, and so on) are also very likely sources of exposure (Harmon, 2001; Stringer et al., 2001). Particularly for determining substances in house dust it has become clear that sampling procedures and the preparation before conducting the analyses do definitely
influence the results. The sampling strategies and methods of the aforementioned studies differ widely. In some studies samples were collected using either exactly the same model of vacuum cleaner throughout (Greenpeace, 2001; Hagenau, 2001; Santillo et al., 2003 [UK study]) or with the participants vacuum cleaners (Kersten and Reich, 2003). While Hagenau (2001) used specific filter holders with glass fibre filters, others collected the dust with ordinary dust filter bags. In the investigation by Santillo et al. (2001) and in our study samples were taken from the bags of vacuum cleaners in regular use for cleaning the apartments. Furthermore, Hagenau (2001) and Santillo et al. (2001) described no specific sample preparation, removing only large and recognizable particles and fibres, while Kersten and Reich (2003) sieved the dust to the fraction of <63 lm, and Santillo et al. (2003) sieving their samples through a 2 mm gauge sieve. In our study only the fine dust between the inner and outer paper bag layers was used. To conclude, the variable results can be explained mainly by the different sampling methods employed, the different particle fraction of dust used for analysis and the variability of the indoor sources (duration of contact of the dust with the source, source characteristics). The results of house dust measurements can, therefore, be used only as an approximate indicator of the amounts of contaminants at a particular location. The most critical organ site in experimental animals is the cellular immune system. Organotins like DBT, TBT, DOT and TPT have a similar (therefore additive) effect on mammalian species, causing lymphocyte
1382
H. Fromme et al. / Chemosphere 58 (2005) 1377–1383
depletion in the thymus and peripheral lymphoid tissues. Toxicity at the thymus appears to provoke selective inhibition of the proliferation of immature thymocytes. The principal relevant study, in which rats were fed tributyltin oxide, is a long term study of immunotoxicity inducing the critical outcome of immunosuppression (reduced IgE and increase in Trichinella spiralis larvae in the muscles) (Vos et al., 1990). Based on this study the WHO calculated a no-observed-adverse-effect-level (NOAEL) of 0.025 mg kg 1 bw and using an uncertainty factor of 100 a tolerable daily intake (TDI) of 0.25 lg kg 1 bw (WHO, 1999b). As a result of the same study the US EPA established an oral reference dose (RFD) of 0.3 lg kg 1 bw and day after benchmark dose analyses (US EPA, 1997). With increased intake rates of house dust the exposure of children is a cause for concern in view of the sensitivity of their developing immune system and their playing habits. To gain a rough idea of the toxicological importance of our results we calculated a worst case exposure scenario. We assumed a dust ingestion rate of 100 mg per day for a child of 15 kg body weight (Stanek and Calabrese, 1995, 2000). Taking our maximum value of total organotin compounds as 7.2 mg kg 1 and a 100% absorption of organotins in the gastro-intestinal tract, a daily intake of 0.048 lg kg 1 bw would result. Comparing this somewhat conservative assessment with the aforementioned TDI and the NOAEL values, a margin of safety (MOS) of between 5 and 500 emerged. Using the maximum value measured in the Greenpeace study, there was an MOS (in relation to the NOAEL) of 20. Our results confirm the widespread contamination of house dust by organotins with sometimes unusually high amounts. These substances are absorbed by the dust as a result of leaching or abrasion from a wide variety of products used in the indoor environment. On the basis of the estimated maximum human exposure levels and the safety margins calculated the ingestion of organotins with house dust does not seem to be an important exposure pathway. We conclude that the results of our study and the limited toxicological data available do not allow for a reliable risk assessment. It can nevertheless not be ruled out that the presence of harmful organotins in households could, under certain conditions, become a relevant source of the intake of organotins by children.
Acknowledgements The authors would like to thank Dr. Richard Flower for his help in preparing this manuscript. The study was funded by the Gesellschaft fu¨r Umweltmedizin und Toxikologie e.V.
References Allsopp, M., Santillo, D., Johnston, P., 2000. Hazardous chemicals in PVC flooring. Greenpeace Research Laboratories Technical Note 14/2000. Available from:. Allsopp, A., Santillo, D., Johnston, P., 2001. Hazardous chemicals in carpets. Greenpeace Research Laboratories Technical Note 01/2001. Available from: . ATSDR (Agency for Toxic Substances and Disease Registry), 1992. Toxicological profile for Tin. TP-91/27. Butte, W., Heinzow, B., 2002. Pollutants in house dust as indicators of indoor contamination. Rev. Environ. Contam. Toxicol. 175, 1–46. DIN (Deutsches Institut fu¨r Normung e.V.), 1999. DIN 3840713. Verfahren zur Bestimmung ausgewa¨hlter Organozinnverbindungen mittels Gaschromatographie. Beuth Verlag, Berlin. Doering, D.D., Steckelbroeck, S., Doering, T., Klingmu¨ller, D., 2002. Effects of butyltins on human 5a-reductase type 1 and type 2 activity. Steroids 67, 859–867. Fent, K., 1996. Ecotoxicology of organotin compounds. Crit. Rev. Toxicol. 26, 1–117. Fromme, H., Lahrz, T., Piloty, M., Gebhardt, H., Oddoy, A., Ru¨den, H., 2004a. Polycyclic aromatic hydrocarbons inside and outside of apartments in an urban area. Sci. Total Environ. 326, 143–149. Fromme, H., Lahrz, T., Piloty, M., Gebhart, H., Oddoy, A., Ru¨den, H., 2004b. Occurrence of phthalates and musk fragrances in indoor air and dust of apartments and kindergartens in Berlin (Germany). Indoor Air 14, 188–195. Greenpeace Nederland, 2001. Toxins in household dust. Research into POPs in a Hundred Dutch households. Hagenau, W., 2001. Zinnorganische Verbindungen in PVC— Fußodenbela¨gen und im Hausstaub. Diplomarbeit, Fachhochschule Weihenstephan. Harmon, M.-E., 2001. This vinyl house. Hazardous additives in vinyl consumer products and home furnishings. Greenpeace, USA. Available from: . KEMI (Swedish National Chemicals Inspectorate), 2000. Organotin stabilisers in PVC. Assessmemt of risks and risk reduction measures. Stockholm, Sweden. Kersten, W., Reich, T., 2003. Schwer flu¨chtige organische Umweltchemikalien in Hamburger Haussta¨uben. Gefahrstoffe-Reinh. Luft 63, 85–91. Lewis, R.G., Fortune, C.R., Willis, R.D., Camann, D.E., Antley, J.T., 1999. Distribution of pesticides and polycyclic aromatic hydrocarbons in house dust as a function of particle size. Environ. Health Perspect. 107, 721–726. Mattulat, 2003. Hausinterne Studie. Bisher unvero¨ffentlichter Bericht. PiCA-Pru¨finstitut Chemische Analytik GmbH, Berlin. Santillo, D., Johnston, P., Brigden, K., 2001. The presence of brominated flame retardants and organotin compounds in dusts collected from parliament buildings from eight countries. Greenpeace Research Laboratories Technical Note 03/ 2001. Available from: .
H. Fromme et al. / Chemosphere 58 (2005) 1377–1383 Santillo, D., Labunska, I., Davidson, H., Johnston, P., Strutt, M., Knowles, O., 2003. Consuming Chemicals: Hazardous Chemicals in house dust as an indicator of chemical exposure in the home. Greenpeace Research Laboratories Technical Note 01/2003. Available from:. Stanek, E.J., Calabrese, E.J., 1995. Daily estimates of soil ingestion in children. Environ. Health Perspect. 103, 276–285. Stanek, E.J., Calabrese, E.J., 2000. Daily soil ingestion estimates for children at a Superfund site. Risk Anal. 20, 627–635. Stringer, R., Johnston, P., Erry, B., 2001. Toxic chemicals in a childÕs world: an investigation into PVC plastic products. Greenpeace Research Laboratories, University of Exeter, UK. Available from: . US EPA (US-Environmental Protection Agency), 1997. Toxicological review. Tributyltin oxide. Integrated Risk Information System. Washington, DC. Available from: .
1383
US EPA (US-Environmental Protection Agency), 2002. Ambient aquatic life water quality criteria for tributyltin (TBT)— Draft. EPA-822-B-02-001. Vos, J.G., DeKlerk, A., Krajnc, E.I., Van Loveren, V., Rozing, J., 1990. Immunotoxicity of bis(tri-n-butyltin) oxide in the rat: effects on thymus-dependent immunity and on nonspecific resistance following long-term exposure in young versus aged rats. Toxicol. Appl. Pharmacol. 105, 144–155. Wester, P.W., Krajnc, E.I., van Leeuwen, F.X.R., Loeber, J.G., van der Heifden, C.A., Vaessens, H.A.M.G., 1990. Chronic toxicity and carcinogenicity of bis(tri-n-butyltin)oxide (TBTO) in the rat. Food Chem. Toxicol. 28, 179–196. WHO (World Health Organization), 1990. IPCS. Environmental health criteria 116. Tributyltin Compounds. Geneva. WHO (World Health Organization), 1999a. Triphenyltin compounds. Concise International Chemical Assessment Document 13. Geneva. WHO (World Health Organization), 1999b. Tributyltin oxide. Concise International Chemical Assessment Document 14. Geneva.
Occurrence of organotin compounds in house dust in Berlin (Germany) H. Fromme a
a,*
, A. Mattulat b, T. Lahrz c, H. Ru¨den
d
Bayerisches Landesamt fu¨r Gesundheit und Lebensmittelsicherheit, Veterina¨rstrasse 2, D-85764 Oberschleißheim, Germany b PiCA-Pru¨finstitut Chemische Analytik GmbH, 12489 Berlin, Germany c Institut fu¨r Lebensmittel, Arzneimittel und Tierseuchen Berlin, 10557 Berlin, Germany d Institut fu¨r Hygiene und Umweltmedizin, Charite´—Universita¨tsmedizin Berlin, 12203 Berlin, Germany Received 19 January 2004; received in revised form 2 August 2004; accepted 29 September 2004
Abstract In a study in the year 2000 on the occurrence of hazardous environmental contaminants house dust samples from 28 Berlin apartments were measured for the presence and concentrations of six organotin compounds, monobutyltin (MBT), dibutyltin (DBT), tributyltin (TBT), monooctyltin (MOT), dioctyltin (DOT) and triphenyltin (TPT). The concentrations of MBT and DBT determined ranged considerably from 0.01 mg kg 1 to 1.5 mg kg 1 (median: 0.05 mg kg 1) and 0.01 to 5.6 mg kg 1 (median: 0.03 mg kg 1), respectively. Maximum levels of TBT and MOT were only 0.08 mg kg 1 and 0.04 mg kg 1. The maximum total value of the organotins was 7.2 mg kg 1 (median: 0.11 mg kg 1). MBT was found in 86% and DBT in 82% of the samples above the limit of quantification, TBT and MOT only in 50% and DOT in 43%. The focus of ecotoxicology is on the risks arising from organotin compounds (especially butyltins) when used as biocides in antifouling paints. TBT acts as an endocrine disrupter in animals, inducing masculinization (imposex) in female gastropods of different species by increasing testosterone levels. The most critical organ site in experimental animals is the cellular immune system, where lymphocyte depletion in the thymus and peripheral lymphoid tissues takes place. Our study does not provide data on the basis of which population exposure could be estimated; house dust containing harmful organotins could, however, under some conditions, become a relevant intake possibility for young children. 2004 Elsevier Ltd. All rights reserved. Keywords: Organotin; Butyltin; Octyltin; Phenyltin; DBT; TBT; House dust; Indoor
1. Introduction Organotins are a large class of compounds which differ markedly in their properties and have been used for a
Rn SnX3 n ;
*
Corresponding author. Fax: +49 89 315 60 425. E-mail address: [email protected] Fromme).
variety of purposes since their discovery and initial use in the 1920s. They are organic derivatives of tetravalent tin and characterized by the presence of at least one covalent carbon–tin bond. They have the general structure
(H.
where R is an alkyl or aryl group, Sn the central tin atom in the oxidation state +4 and X a singly charged
0045-6535/$ - see front matter 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2004.09.092
1378
H. Fromme et al. / Chemosphere 58 (2005) 1377–1383
anion or an anionic organic group. Most importantly, methyl, butyl, octyl and phenyl moieties are contained in the organic substituents of all compounds of any practical relevance, whereas the anionic moiety usually comprises chloride, fluoride, oxide, hydroxide, carboxylate or thiolate. Under normal conditions their physicochemical properties appear as only slightly volatile, insoluble in water and with a strong tendency to absorb particulate matter (KEMI, 2000). Of the various organotins, the mono-, di- and trisubstituted compounds are the most widely employed, the first two (notably mono- and di-methyltin, butyltin and octyltin compounds) being used extensively as heat and light stabilizers in the manufacture of polyvinyl chloride (PVC) polymers and as industrial catalysts for polyurethane and silicone elastomers. Thus they are present in water pipes, food packing materials (e.g., dioctyltin), glass coatings (e.g., butyltin trichloride), polyurethane foams and many other consumer products. Another less extensive use of organotins is in industrial and agricultural biocides (fungicides, bactericides, miticides, insecticides), surface disinfectants and preservatives for wood, paper, textiles, paints and some electrical equipment. On a world wide basis tri-substituted organotins (notably tributyltin oxide and tributyltin methacrylate) are used as antifouling agents in paints for boats and ships, thereby accessing the aquatic environment (WHO, 1990; US EPA, 2002). It can be estimated that their most frequent uses are as PVC stabilizers (approximately 16 000 ton annually in Europe) and as catalysts (1300–1650 ton), thus accounting in each case for approximately 67% and 8% of total organotin consumption. About 2–4% of PVC production is ascribable to this type of stabilizer. Biocidal uses account for nearly 20% of the market and all other uses (including latex paint preservatives, anthelminthics and coccidiostats) for 5% (US EPA, 2002). The prominent toxicological feature of the organotins is their immunotoxicity, an effect produced by di- and trialkyltins as well as triphenyltins. The acute toxicity after dermal or oral application of most organotins is low to moderate (depending on routes and species, for DBT 58–5480 mg kg 1 bw), though local effects possibly occurring are eye and skin irritation or sensitization at the site of contact (WHO, 1990; ATSDR, 1992; WHO, 1999a,b; KEMI, 2000). The focus of ecotoxicology is on risks from the use of organotin compounds (especially butyltins) as biocides in antifouling paints. TBT acts as an endocrine disrupter in animals and induces masculinization (imposex) by increasing testosterone levels in different species of female gastropods (Fent, 1996). In experimental animals TBT produced endocrine tumors, although no evidence of any TBT mutagenicity is known (Wester et al., 1990). In vitro studies found that butyltins inhibit human cytochrome P450 aromatase and 5a-reductase activity and might therefore have
a distinct impact on human hormone metabolism (Doering et al., 2002).
2. Material and methods The present study was conducted for the identification of hazardous chemicals in the indoor environment of apartments in a particular urban area as part of a wider investigation (Fromme et al., 2004a,b). The aim was to measure the most significant organotins produced in industrial processes and present in house dust. That is, monobutyltin (MBT), dibutyltin (DBT), tributyltin (TBT), monooctyltin (MOT), dioctyltin (DOT), and triphenyltin (TPT). It is well known that house dust is a receptor of and repository for chemicals in the home and thereby a potential source of exposure to them, especially for small children and babies, who spend significant portions of their time on floors and engaging in frequent mouth contact activities (Butte and Heinzow, 2002). The concentrations of organotin compounds in household dust were determined for a total of 28 residences in the year 2000. The apartments were built between 3 and 104 years before and had areas of 51–250 m2.
3. Sampling procedure and analysis The house dust was taken from the bags of the vacuum cleaners in regular use for cleaning the apartments. Inhomogeneity was avoided by only taking the fine dust from the space between the inner and outer paper layers of the bags. Any larger dust particles, fibres and debris, which might otherwise have distorted the results, were thus excluded. The dust was put directly to clean aluminium foil and then transferred to glass bottles for storage. Information reported in the literature indicates that this fine dust contains significant quantities of persistent organic pollutants and is therefore a good indicator of human exposure levels (Lewis et al., 1999). The determination of organotin compounds was carried out according to German standard methods DIN 38407-13 (DIN, 1999). For analysis 0.5–1.0 g of dust samples were spiked with internal standards (monoheptyltin, diheptyltin, tripropyltin from Promochem) and extracted with ethanol using an ultrasonic method. An aliquot was buffered with sodium acetate to pH 4.5 and alkylated with sodium tetraethylborate (STEB). Subsequently, the tetra-alkylated tin compounds were extracted with n-hexane and analysed by gaschromatography/mass spectrometry on an HP 6890/5973 GC/MS system with a CTC autosampler Combi PAL. Separation was performed using a 30 m · 0.25 mm i.d. DB-5 fused silica capillary column with a 0.5 lm film thickness. The injector (Gerstel, cold injection system
H. Fromme et al. / Chemosphere 58 (2005) 1377–1383
blanks with every sample sequence. There was no reference material available for organotins in house dust.
Table 1 Ions used for quantification and qualification of organotin compounds Compound
Mass used for quantification (m/z)
Qualifier masses (m/z)
MBT DBT TBT MOT DOT TPT
235 263 291 291 375 351
147, 179, 179, 179, 263, 197,
MHT DHT TPrT
277 347 249
179, 275 249, 345 193, 235
1379
4. Results and discussion
179 205 289 289 291 349
Table 2 shows the statistical parameters of organotin measurements in house dust. MBT and DBT were found in 86% and 82% of the samples above the detection limit, TBT and MOT only in 50% and DOT only in 43%. The concentrations of MBT and DBT determined ranged, respectively from 0.01 to 1.5 mg kg 1 (median: 0.045 mg kg 1) and 0.01 mg kg 1 to 5.6 mg kg 1 (median: 0.03 mg kg 1). Median levels of TBT and MOT were only 0.01 mg kg 1. The maximum value of the total organotins (using only values above the detection limit) was 7.2 mg kg 1 (median: 0.1 mg kg 1). A considerable amount of data exists on the distribution of organotins (especially TBT) in the aquatic environment, though much less information is available regarding the distribution of these compounds in other locations, the indoor environment for example. It must be understood that house dust is a very heterogeneous mixture whose composition depends on many different factors, such as the type and construction of the building, the use of the rooms in question, types of furnishing and flooring materials present, heating and ventilation systems, etc. It is not surprising that results published in the literature (see Table 3 and Fig. 1) vary widely. From 1998 to 2000 Kersten and Reich (2003) examined dust from 50 apartments in Hamburg and found median concentrations of 1.4 mg kg 1 (MBT) and 0.2 mg kg 1 (DBT). In studies conducted by the Greenpeace Research Laboratories in parliament buildings of 8 European countries and in private apartments, notably in the UK, total levels of organotins of between 0.14 mg kg 1 and 5.81 mg kg 1 were discovered (Santillo et al., 2001, 2003). In apartments these compounds ranged from 1.6 mg kg 1 to 5.1 mg kg 1, slightly higher than the range previously reported by the same study group for dust from parliament buildings (0.5–3.5 mg kg 1;
Internal standards: monoheptyltin (MHT), diheptyltin (DHT) and tripropyltin (TPrT).
CIS 4) was operated in splitless mode, started at 100 C and heated to 280 C at a rate of 12 C/s (held for 5 min) with 2 ll injection volume. The GC oven temperature was kept at 50 C for 2 min and then increased to a temperature of 230 C at a rate of 10 C/min, 280 C at a rate of 4 C/min and 345 C at a rate of 10 C/min, then sustained for 15 min. The transferline was held at 320 C. The MS was operated in the single ion monitoring (SIM) mode with 6–15 ions per group and a dwell time of 30 ms. Identifications were based on the GC retention times relative to those of corresponding internal standards and the relative abundance of the ions monitored. Quantifications were based on comparisons of the integrated ion responses of the target ions which those of the corresponding internal standards and by using average response factors of the target analytes generated from standard calibration. The ions used for quantification and qualification are listed in Table 1. The limit of quantification of the target compounds was 0.01 mg kg 1 for each analyte, based on a minimum signal/noise ratio of 5:1. The recovery was >70% at a concentration of 0.01 mg kg 1, tested by standard addition to dust samples. The methods quality was checked by analysis
Table 2 Statistical characteristics of organotin compounds in the house dust of apartments (mg kg 1)
N N > d.l. Mean Standard deviation Median 90th Percentile 95th Percentile Maximum
MBT
DBT
TBT
MOT
DOT
Total organotina
28 24 0.16 0.32 0.05 0.35 0.70 1.50
28 23 0.51 1.31 0.03 1.03 3.28 5.60
28 14 0.02 0.02 0.01 0.05 0.07 0.08
28 14 0.01 0.01 0.01 0.03 0.03 0.04
28 12 0.02 0.07 0.01 0.02 0.03 0.36
28 26 0.77 1.66 0.11 1.79 4.15 7.18
d.l.: detection limit; values under the d.l. are used with half of the d.l. a Sum of values above the detection limit.
1380
Table 3 Concentrations of organotin compounds in household dust and floorings in mg kg DBT
TBT
MOT
DOT
TPT
Total organotinsa
<0.1–24.0 (0.53) 0.18–2.39 (1.05) 0.012–81.8 (0.41) 0.81–2.80 (1.35) 0.08–1.20 (0.65) 0.1–18.0 (1.4) 0.005–1.50 (0.05)
<0.1–2200 (3.40) 0.17–0.89 (0.46) 0.005–130.3 (0.25) 0.16–1.30 (0.52) 0.03–0.33 (0.22) 0.01–5.6 (0.2) <0.005–5.60 (0.03)
<0.1–15.0 (0.46) 0.004–0.19 (0.022) 0.005–2.03 (0.009) 0.02–0.76 (0.05) 0.004–0.16 (0.03) <0.001–0.2 (0.03) <0.005–0.08 (0.008)
–
0.015–0.83 (0.26) 0.005–0.98 (0.02) 0.08–1.30 (0.35) 0.02–0.91 (0.14) 0.01–2.8 (0.2) <0.005–0.04 (0.008)
0.004–0.14 (0.06) 0.005–2.12 (0.02) 0.02–0.55 (0.06) 0.003–3.60 (0.05) <0.001–0.2 (0.01) <0.005–0.36 (0.005)
<0.001
1.3–2222 (33.1) 0.49–3.48 (1.78) 0.06–212 (0.91) 1.58–5.05 (2.43) 0.14–5.81 (1.36) –
<0.1–70.0 (5.5) <0.5–38.0 (6.1) 0.33–48.8 (2.62)
<0.1–920.0 (120.0) <0.5–210 (12.0) 37.7–569 (200)
<0.1–3.40 (<0.1) <0.5–1.2 (0.25) 0.13–17.9 (4.39)
<100–30.0e
<0.1–6.50f
–
<0.5–91 (7.1) <0.05–0.98 (0.26)
<0.5–276 (36) 0.11–10.2 (1.56)
<0.5
Allsopp et al. (2001) Allsopp et al. (2001)
6
<0.0003–0.009
<0.0003–0.03
<0.0003–0.007
<0.0004
<0.0004
<0.0003
0.002–0.042
2
0.50/1.14
0.10/7.20
2.70/47.50
<0.0004/ <0.0004
<0.0004/ <0.0004
<0.0003/ <0.0003
3.32/56.0
a
33 15 134 100c 10 50 28
33 35 5
Only values above d.l. Only 1 sample >d.l. c Partly pooled samples from households of UK regions. d Only 2 samples >d.l. e Only 3 samples >d.l. f Only 10 samples >d.l. g Carpets treated against bacteria, moulds, fungi and dust mites. b
0.009b <0.001–0.07 (<0.001) <0.001–0.04d <0.001–0.02 (0.004) <0.005
<0.05
0.01–7.18 (0.10)
1.5–997 (245) 0.2–381 (78) 51–592 (256)
Apartments in south Germany Parliament buildings in Europe Apartments and large buildings, Netherlands Samples from all UK regions Samples from 5 European countries Apartments in Hamburg, Germany Apartments in Berlin, Germany
PVC floorings in apartments, Germany PVC floorings, Germany PVC floorings from various suppliers, UK Carpets from various suppliers, UK Carpets from various suppliers, UKg
H. Fromme et al. / Chemosphere 58 (2005) 1377–1383
Flooring materials Hagenau (2001) Mattulat (2003) Allsopp et al. (2000)
(ranges; values in brackets are medians)
MBT
N Household dust Hagenau (2001) Santillo et al. (2001) Greenpeace (2001) Santillo et al. (2003) Santillo et al. (2003) Kersten and Reich (2003) Our study
1
H. Fromme et al. / Chemosphere 58 (2005) 1377–1383
1381
Median min.
m g/kg
0.01
0.1
m a x.
1.0
10
100
1000
Hagenau 2001 Santillo et al. 2001
Santi llo et al. 2003 (UK r egions) Santillo et al. 2003 (Europe) Kersten & Reich 2003 Greenpeace 2001
Our study
Fig. 1. Comparison of MBT (light bars) and DBT (dark bars) levels in household dust samples described in the scientific literature.
Santillo et al., 2001). The most abundant organotin was MBT followed by DBT and MOT. In another study, performed by Greenpeace Netherlands (Greenpeace, 2001), dust samples were collected in large buildings and 134 apartments all over the Netherlands. The median concentration was 0.41 mg kg 1 (MBT) and 0.25 mg kg 1 (DBT). The total organotin content varied in this investigation between 0.06 mg kg 1 and 212 mg kg 1. Very high concentrations of organotins were reported by Hagenau (2001), who measured 33 samples with concentrations from 1.3 to 2222 mg kg 1 in German apartments. An explanation for these larger amounts might be that he used a vacuum cleaner with a filter holder for sampling. Hagenau (2001) also found a significant correlation between organotin levels in household dust and PVC materials. The examinations carried out to date have shown generally wide variations in indoor organotin levels, a fact which may merely reflect the variability of their presence in the form of stabilizer additives in plastics. Previous studies (Allsopp et al., 2000, 2001; Hagenau, 2001; Mattulat, 2003) also discovered the widespread presence of organotins in carpets and floor coverings with levels of up to 1000 mg kg 1 (Table 3). It is to be expected that other products (plastic crib rail teethers, clothing, floor tiles, wall paper, handbags, mattress covers, and so on) are also very likely sources of exposure (Harmon, 2001; Stringer et al., 2001). Particularly for determining substances in house dust it has become clear that sampling procedures and the preparation before conducting the analyses do definitely
influence the results. The sampling strategies and methods of the aforementioned studies differ widely. In some studies samples were collected using either exactly the same model of vacuum cleaner throughout (Greenpeace, 2001; Hagenau, 2001; Santillo et al., 2003 [UK study]) or with the participants vacuum cleaners (Kersten and Reich, 2003). While Hagenau (2001) used specific filter holders with glass fibre filters, others collected the dust with ordinary dust filter bags. In the investigation by Santillo et al. (2001) and in our study samples were taken from the bags of vacuum cleaners in regular use for cleaning the apartments. Furthermore, Hagenau (2001) and Santillo et al. (2001) described no specific sample preparation, removing only large and recognizable particles and fibres, while Kersten and Reich (2003) sieved the dust to the fraction of <63 lm, and Santillo et al. (2003) sieving their samples through a 2 mm gauge sieve. In our study only the fine dust between the inner and outer paper bag layers was used. To conclude, the variable results can be explained mainly by the different sampling methods employed, the different particle fraction of dust used for analysis and the variability of the indoor sources (duration of contact of the dust with the source, source characteristics). The results of house dust measurements can, therefore, be used only as an approximate indicator of the amounts of contaminants at a particular location. The most critical organ site in experimental animals is the cellular immune system. Organotins like DBT, TBT, DOT and TPT have a similar (therefore additive) effect on mammalian species, causing lymphocyte
1382
H. Fromme et al. / Chemosphere 58 (2005) 1377–1383
depletion in the thymus and peripheral lymphoid tissues. Toxicity at the thymus appears to provoke selective inhibition of the proliferation of immature thymocytes. The principal relevant study, in which rats were fed tributyltin oxide, is a long term study of immunotoxicity inducing the critical outcome of immunosuppression (reduced IgE and increase in Trichinella spiralis larvae in the muscles) (Vos et al., 1990). Based on this study the WHO calculated a no-observed-adverse-effect-level (NOAEL) of 0.025 mg kg 1 bw and using an uncertainty factor of 100 a tolerable daily intake (TDI) of 0.25 lg kg 1 bw (WHO, 1999b). As a result of the same study the US EPA established an oral reference dose (RFD) of 0.3 lg kg 1 bw and day after benchmark dose analyses (US EPA, 1997). With increased intake rates of house dust the exposure of children is a cause for concern in view of the sensitivity of their developing immune system and their playing habits. To gain a rough idea of the toxicological importance of our results we calculated a worst case exposure scenario. We assumed a dust ingestion rate of 100 mg per day for a child of 15 kg body weight (Stanek and Calabrese, 1995, 2000). Taking our maximum value of total organotin compounds as 7.2 mg kg 1 and a 100% absorption of organotins in the gastro-intestinal tract, a daily intake of 0.048 lg kg 1 bw would result. Comparing this somewhat conservative assessment with the aforementioned TDI and the NOAEL values, a margin of safety (MOS) of between 5 and 500 emerged. Using the maximum value measured in the Greenpeace study, there was an MOS (in relation to the NOAEL) of 20. Our results confirm the widespread contamination of house dust by organotins with sometimes unusually high amounts. These substances are absorbed by the dust as a result of leaching or abrasion from a wide variety of products used in the indoor environment. On the basis of the estimated maximum human exposure levels and the safety margins calculated the ingestion of organotins with house dust does not seem to be an important exposure pathway. We conclude that the results of our study and the limited toxicological data available do not allow for a reliable risk assessment. It can nevertheless not be ruled out that the presence of harmful organotins in households could, under certain conditions, become a relevant source of the intake of organotins by children.
Acknowledgements The authors would like to thank Dr. Richard Flower for his help in preparing this manuscript. The study was funded by the Gesellschaft fu¨r Umweltmedizin und Toxikologie e.V.
References Allsopp, M., Santillo, D., Johnston, P., 2000. Hazardous chemicals in PVC flooring. Greenpeace Research Laboratories Technical Note 14/2000. Available from:
H. Fromme et al. / Chemosphere 58 (2005) 1377–1383 Santillo, D., Labunska, I., Davidson, H., Johnston, P., Strutt, M., Knowles, O., 2003. Consuming Chemicals: Hazardous Chemicals in house dust as an indicator of chemical exposure in the home. Greenpeace Research Laboratories Technical Note 01/2003. Available from:
1383
US EPA (US-Environmental Protection Agency), 2002. Ambient aquatic life water quality criteria for tributyltin (TBT)— Draft. EPA-822-B-02-001. Vos, J.G., DeKlerk, A., Krajnc, E.I., Van Loveren, V., Rozing, J., 1990. Immunotoxicity of bis(tri-n-butyltin) oxide in the rat: effects on thymus-dependent immunity and on nonspecific resistance following long-term exposure in young versus aged rats. Toxicol. Appl. Pharmacol. 105, 144–155. Wester, P.W., Krajnc, E.I., van Leeuwen, F.X.R., Loeber, J.G., van der Heifden, C.A., Vaessens, H.A.M.G., 1990. Chronic toxicity and carcinogenicity of bis(tri-n-butyltin)oxide (TBTO) in the rat. Food Chem. Toxicol. 28, 179–196. WHO (World Health Organization), 1990. IPCS. Environmental health criteria 116. Tributyltin Compounds. Geneva. WHO (World Health Organization), 1999a. Triphenyltin compounds. Concise International Chemical Assessment Document 13. Geneva. WHO (World Health Organization), 1999b. Tributyltin oxide. Concise International Chemical Assessment Document 14. Geneva.