Stability and storage problems in organotin speciation in environmental samples J. L. Go�mez-Ariza,*a I. Gira�ldez,a E. Morales,a F. Ariese,b W. Cofinob and Ph. Quevauvillerc aDepartamento de Quý�mica y Ciencia de los Materiales, Escuela Polite�cnica Superior, Universidad de Huelva, La Rabida, Huelva, Spain. E-mail: ariza@uhu.es bVrije Universiteit, Instituut voor Milieuvraagstukken, de Boelelaan 1115, NL-1081 HV Amsterdam, The Netherlands cEuropean Commission, Standards, Measurements and Testing Programme, 200 rue de la Loi, B-1049 Brussels, Belgium Received 16th October 1998, Accepted 23rd February 1999 The stability of both tributyltin (TBT) and triphenyltin (TPT) in water, sediment, oysters and cockles was studied over a period of 18 months using several storage conditions. Butyltins were stable in unacidified sea-water stored in polycarbonate bottles in the dark at 4 °C for 7 months, but half of the TBT concentration was lost after 540 d.A comparable preservation time was achieved for butyltins stored on C18 cartridges at room temperature. However, phenyltins extracted from sea-water were stable for only 60 d stored on cartridges and even more pronounced losses (about 90% after 540 d) occurred when they were stored in either polycarbonate or Pyrex glass bottles.Losses of organotins were observed in sediments after air drying and pasteurization treatments using a freeze-dried sediment as a comparator, whereas both butyltin and phenyltin species remained stable in sediments stored at -20 °C for the 18 months tested, irrespective of the treatment used for stabilization.Air drying followed by pasteurization was shown to be superior to other treatments for the stabilization of organotin compounds in sediments stored at higher temperatures, but 30% of TBT was lost after 540 d at 25 °C. Finally, butyltins were stable in both frozen cockles and oysters in the dark over a 7 month period and in freeze-dried samples stored at 4 °C for 5 months, but TBT losses of about 70% were observed after 540 d.Only a few studies have been specifically devoted to Introduction organotin stability in environmental samples during storage.7 The main inputs of tributyltin (TBT) and triphenyltin (TPT) No losses of diVerent organotin compounds in acidified disinto the marine environment are by release from marine tilled water were observed for at least 20 d in 1 l brown glass antifouling paints containing these species, which are then bottles stored at 25 °C.8 TBT and monobutyltin (MBT) were dissolved in sea-water and partially adsorbed on suspended found to be stable in natural filtered sea-water for 4 months solids or bioaccumulated by living organisms.Several eVects when stored at 4 °C in the dark, but the stability of dibutyltin of these compounds in organisms exposed to extremely low (DBT) was more doubtful.9 Otherwise, freezing followed by levels of both TBT and TPT are well documented: shell oven drying at 50 °C has been shown to be suitable to preserve thickening in oysters1 and mussels,2 imposex in gastropods,1 the stability of butyltins in sediments for 4 months9 and both poor growth in bivalves,1 etc.Therefore, monitoring of tin freezing and storage at 4 °C preserved butyltins for 1 year.10 compounds is required by EC legislation, and comparability However, degradation of TPT was observed after 3 months.10 of data produced by diVerent laboratories is necessary. Finally, good stability of butyltins was achieved for mussel The diVerent toxicities of organotin compounds4 necessitate samples stored for 44 months at -20 °C in the dark.11,12 the use of sophisticated approaches for analytical speciation.5 The aim of this study was to assess the stability of TBT, These procedures generally involve several analytical steps.DBT, MBT, TPT, diphenyltin (DPT) and monophenyltin Therefore, the validation of these techniques and the improve- (MPT) in natural samples (sea-water, sediments, oysters and ment of quality control in tin determination require the use cockles) during storage under diVerent conditions such as pH, certified reference materials (CRMs).These materials are temperature and type of container, in order to select the best products of very high added value.Therefore, rigorous control preservation conditions for both environmental studies and of their stability over long periods of time is necessary. Oven- for the preparation of CRMs. or freeze-drying and gamma irradiation pre-treatments have been used to reduce chemical and microbiological changes in Experimental solid materials and for achieving long-term stability.6 The accuracy of organotin determinations in natural samples Reagents, standards and apparatus not only depends on the measurement step, but also sampling, The reagents used in the experiments were of analytical-reagent storage and sample preparation aVect the reliability of the grade and obtained from Merck (Darmstadt, Germany) and results.Several processes such as contamination, volatilization, Aldrich (Milwaukee, WI, USA).C18 cartridges (Sep-Pak adsorption and degradation due to microbial activity or UV Classic cartridges, 600 mg of sorbent) were obtained from irradiation may alter the initial composition of the sample. Waters (Milford, MA, USA). Pesticide grade solvents were Many of the above-mentioned eVects are highly dependent on temperature, which has to be monitored during sample storage.purchased from Romil (Loughborough, UK). Organotin com- J. Environ. Monit., 1999, 1, 197–202 197pounds as chlorides were obtained from Aldrich (purity higher Organotin concentrations were deduced from calibration curves derived from derivatized calibrant solutions using peak than 95%, evaluated by FAAS) and were used without further purification, but analysis did not reveal detectable impurities.areas. The calibration curves were linear for Sn amounts less than 1.4 ng. All results produced in these experiments are Water used in the experiments was doubly distilled and deionized and gave blank readings in all the analyses. The expressed as Sn, ng l-1 for water and ng g-1 dry mass for sediments and animal tissues. glassware used for experiments was previously soaked in saturated Na2Cr2O7 in H2SO4 for 24 h, rinsed carefully with The analytical quality was monitored by preparing a calibration graph each week and by injecting a pentylated doubly distilled water and then with methanol and dried.organotin calibrant solution with all the tin species every day Stock standard calibrant solutions of each organotin species to test the instrument signal.Quality control diagrams were were prepared gravimetrically in diethyl ether at about constructed and rejection thresholds for analytical samples 100 mg l-1 (as Sn). They were stored refrigerated in the dark were set at three standard deviations of the mean, generated and diluted in hexane to give working calibrant solutions.An for 10 sample solutions analyzed at the beginning of the internal standard, dimethyldipentyltin (DPeT), was prepared experiments. The same calibrant solution stored at -20 °C from dimethyltin chloride (Aldrich) by Grignard derivawas used for all calibrations. This solution was verified with tization.13 a fresh calibrant solution prepared every 6 months and was A gas chromatograph (Star 360, Varian, San Fernando, found to be stable.Two Certified Reference Materials, a CA, USA) fitted with an SPI injector, a glass capillary column sediment (CRM 462) and a mussel sample (CRM 477), both (SPB-1; Supelco, Bellefonte, PA, USA) (15 m×0.53 mm id, obtained from BCR (Brussels, Belgium) were used for vali- film thickness 1.5 mm) and a pulsed flame photometric detector dation of the procedures.Recovery tests were performed for operating with a 610 cut-oV interference filter was used. The each kind of sample every month and ranged from 85±8% injector temperature was programmed as follows: initial tem- (for MPT in water) to 99±6% (for TBT in biota). All perature 40 °C held for 0.1 min, 300 °Cmin-1 ramp to 250 °C experiments for each organotin preservation study were per- and a final isotherm for 10 min.The injelume was formed by the same operator. The detection limits (evaluated 1 ml. The oven temperature started at 75 °C held for 0.85 min, as 3s of the blank) in water, biota and sediment analysis followed by a heating ramp of 10 °Cmin-1 to 250 °C and a (including the extraction step) are given in Table 1. Samples final isotherm for 5 min.Helium served as the carrier gas at a were analyzed at least three times with relative standard flow rate of 9.5 cm3 min-1. The detector was operated at deviations (RSD) in the range 4–10% for all the organotin 300 °C. compounds evaluated in water and biota samples and for TBT and DBT in sediment. However, higher RSD values, ranging from 2 to 37%, were obtained for both MBT and phenyltins Organotin determination calibration and quality control in sediments, particulary in the pasteurized sample. These poor Water samples were extracted using both liquid–liquid and results may be caused by the presence of an interfering solid phase extractions following methods described in the chromatographic peak close to MBT and by the low levels of literature.14,15 Briefly, a 1000 ml portion of sample acidified phenyltins in the sediment samples, being close to the detecwith 10 ml of HBr was extracted by shaking vigorously in the tion limits.dark with 300 ml of a 0.07% m/v solution of tropolone in pentane for 10 min. The organic extract was dried with anhy- Sample collection and preservation drous Na2SO4 and reduced in volume to 0.5 ml by rotary Two water samples collected in Cadiz harbor (southwest evaporation.The extract was derivatized by pentylation. Solid Spain) on diVerent days were used to study the stability of phase extraction was carried out as follows: for cartridge organotin compounds in polycarbonate and Pyrex glass bottles conditioning, 10 ml of methanol followed by 10 ml of distilled and on C18 cartridges, respectively.Previous analyses carried water were passed through a C18 cartridge with the aid of a out using liquid–liquid extractions revealed that phenyltin vacuum pump at a flow rate of 5 ml min-1. The water sample compound concentrations were below the detection limits in (1000 ml ) was passed through the cartridge at a flow rate of these samples.Therefore, the samples had to be spiked with 8 mlmin-1. Then the cartridge was dried under a stream of known amounts of phenyltin chlorides in order to assess the nitrogen for 5 min. Organotins were eluted using 2 ml of preservation of these species. One sample was spiked with 380, 1% v/v HBr and 0.1% tropolone solution in methanol. Prior 800 and 960 ng l-1 (as Sn) of MPT, DPT and TPT, respect- to the derivatization step, the solvent was removed under a ively, and two subsamples were stored in 2.5 l bottles.However, stream of nitrogen and the residue was solubilized in 1 ml these concentrations are high compared with those usually of hexane. found in environmental samples. These subsamples were stabil- Sediment and biota samples were digested using 50 ml of ized using diVerent treatments: (i) storage in polycarbonate water–hydrogen bromide mixture (1+1) and extracted with bottles at 4 °C both without acid addition and filtering and 50 ml of 0.04% m/v tropolone solution in dichloromethane for (ii) acidification with 0.5% HBr and storage in Pyrex glass 2 h.The organic phase was dried with anhydrous sodium bottles at 4 °C in the dark.The other sample was spiked to sulfate and reduced to 1 ml. The solid phase was washed with final concentrations of 380, 700 and 770 ng l-1 (as Sn) of 5 ml of hexane and was then added to the dichloromethane MPT, DPT and TPT, respectively, and was passed through extract and reduced again to 1 ml to substitute the solvent for the cartridges, which were then stored at 25 °C in the dark.final derivatization.16,17 The organotins were derivatized with 4 ml of 1 mol l-1 pentylmagnesium bromide solution in diethyl ether for 1 h at room temperature. The excess of Grignard reagent was Table 1 Detection limits for organotin species in water, biota and removed with 4 ml of 0.5 mol l-1 sulfuric acid and the ether sediment samples, with values given as ng Sn l-1 for water and ng extract was reduced to 1 ml and passed through a Florisil Sn g-1, dry mass basis, for biota and sediment column (4 g).The organotin compounds were eluted with Sample TBT DBT MBT MPT DPT TPT pentane, concentrated to dryness under a stream of nitrogen and dissolved in a suitable solution containing the internal Water 4.6 5.3 5.3 12 12 11 standard (ranging between 0.05 and 1.0 ml ).They were quant- Biota/sediment 0.30 0.32 0.33 1.4 1.6 1.3 ified by GC-FPD. 198 J. Environ. Monit., 1999, 1, 197–202stored in 250 ml brown glass bottles with plastic screw-caps. The initial concentrations of organotins were determined by five replicate analyses. No phenyltin compounds were found in the samples. Phenyltins were not added to the samples because of the physical and/or chemical binding of these analytes to the matrix, hence their analytical behavior may diVer considerably from bioaccumulated organotins.The subsequent analyses at 15, 30, 60, 90, 150, 210, 270, 365 and 540 d were made in triplicate. The samples were manually homogenized prior to each analysis. All samples were stored in the dark since butyltin and phenyltin photodegradation has been demonstrated.18 Statistical treatment The data were analyzed statistically for diVerences using factorial analysis of variance (ANOVA).Prior to analysis, all the data were tested for homogeneity of variance using the Barlett and Levene tests. Student’s t-test was applied to test diVerent hypotheses. An a value of 0.05 was adopted as the critical level for all statistical testing giving a 95% confidence level (CSS: STATISTICA).Results and discussion Organotin stability in water samples The results for diVerent concentrations of both butyl- and phenyltin species as a function of the storage time are shown in Fig. 1. No significant changes in concentration of TBT were Fig. 1 Stability of organotin species in sea-water stored in the dark observed during the first 210 d for samples stored in both under the following conditions: (a) and (b) unacidified samples in polycarbonate bottles and adsorbed on C18 cartridges polycarbonate bottles at 4 °C; (c) and (d) acidified samples in Pyrex (ANOVA, p=0.33 and 0.24, respectively).However, a slight glass bottles at 4 °C; (e) and (f ) on C18 cartridges at room temperature.decrease in TBT concentration was observed after 90 d in Symbols and error bars represent means and standard deviation of samples stored in Pyrex glass bottles (t-test, p=0.002). organotin concentrations in triplicate analyses. DiVerent results were obtained for TPT, whose preservation was more diYcult to achieve than that of TBT, showing a decrease in concentration from the first month of storage both A sediment sample was collected in a side arm of the North Sea Canal (northwest of Amsterdam, The Netherlands), homo- in polycarbonate and in Pyrex glass bottles (t-test, p<0.008).The decrease in both TBT and TPT was not followed by an genized and divided into three subsamples. Each subsample was subject to diVerent stabilization procedures at the Institute increase in the concentration of dialkyltin or monoalkyltin species, which indicated that losses of TBT and TPT may not for Reference Materials and Measurements (IRMM): (i) freeze-drying, (ii) air drying at 40 °C and (iii) air drying at necessarily be due to degradation, and adsorption on the bottle surface is also possible.Better preservation of this 40 °C followed by pasteurization at 120 °C.Some of these processes may alter the composition of the sample. However, species was obtained in cartridges and no changes in concentration were observed during the first 60 d (ANOVA, p= since this sample will be used as a candidate certified reference material, the main purpose was to obtain the maximum 0.57). Therefore, the use of cartridges for organotin determination has undoubted advantages: (i) it requires a smaller stability of the sample during a long-term period. Each subsample was divided into four aliquots and stored at four diVerent volume of solvent than traditional liquid–liquid extraction; (ii) it involves simple manipulations which are not time temperatures in the dark: (i) 40±0.1, (ii) 25±1, (iii) 4±0.5 and (iv) -20 °C.Analyses of the organotins at 30, 60, 90, consuming and allow for in situ treatment of samples; and (iii) the cartridges can be used for storage of organotins for 2 120, 180, 360 and 540 d were performed in triplicate on diVerent days. The purpose was to test the stability of organot- months at room temperature. The stability of butyltins in both acidified, filtered natural ins with regard to eVects of storage temperature and drying procedure in a sediment sample.samples and non-acidified synthetic aqueous solutions stored at 4 °C in Pyrex glass bottles for 4–5 months has been The study of the stability of butyltins in biota during storage was performed on Crassostrea gigas oyster samples and reported.9,19,20 The disagreement between these longer preservation times and those obtained in this work could be explained Cerastoderma edulis cockle samples.They were collected in three estuarine areas of southwest Spain: Piedras River (oys- by the use of unfiltered samples with 30–35 mg l-1 of suspended matter, which has been proved to decrease the stability ters), Carreras River (oysters) and San Pedro River (two samples of cockles, collected downstream and upstream).of butyltin compounds stored in Pyrex glass bottles.9 Moreover, we spiked the samples with high phenyltin concen- Oysters and cockles were depurated in clean 0.45 mm filtered sea-water for 24 h and then the animal tissues were removed trations and the losses obtained in the present study may be lower than those obtained in samples with more realistic from their shells and drained.The edible part of each sample was homogenized using a mechanical mixer and divided into environmental levels because the biological activity may be slowed, aVecting the stability of both butyl- and phenyltins, two parts. One half was stored at -20 °C and the other was lyophilized (48 h at -60 °C), ground in a Teflon-coated and the adsorption on the container walls is also higher at lower concentrations.grinding mill and stored at 4 °C. The material was finally J. Environ. Monit., 1999, 1, 197–202 199Fig. 3 Stability of organotin species in air-dried sediments stored in Fig. 2 Stability of organotin species in freeze-dried sediments stored the dark at the following temperatures: (a) and (b) 40 °C; (c) and (d) in the dark at the the following temperatures: (a) and (b) 40 °C; (c) 25 °C; (e) and (f ) 4 °C; (g) and (h) -20 °C.Symbols and error bars and (d) 25 °C; (e) and (f ) 4 °C; (g) and (h) -20 °C. Symbols and represent means and standard deviation of organotin concentrations error bars represent means and standard deviation of organotin in triplicate analyses. concentrations in triplicate analyses.Organotin stability in sediment samples natural sediments,10 these treatments may be used to achieve long-term stabilized reference materials and as a consequence The integrity of the organotin levels in the sediment sample during the stabilization treatment was evaluated by triplicate they were considered for further experiments. Figs. 2–4 summarize the trends of the contents of the analyses of each sample.The results are given in Table 2. Freeze-drying treatment was selected as a comparator to check organotin species as a function of the storage time for the diVerent storage conditions tested. All organotin species were the eVects of both air-drying and pasteurization processes on organotin levels as it does not produce changes in either shown to be stable in sediment samples stored at -20 °C for the 540 d checked and the stability was not aVected by the use butyltin or phenyltin species in sediments.9,10 Air drying treatment caused a decrease of 27.5% in the TBT content, of diVerent drying treatments (ANOVA, p>0.05).This con- firms previous studies demonstrating the stability of these with a simultaneous increase in MBT content.Pasteurized samples displayed a larger decrease of the TBT content (54.8%) compounds in frozen sediments.9,10,20 Moreover, losses or interconversion of species were prevented in both freeze-dried and a simultaneous increase in both DBT and MBT (26.2 and 110%, respectively). These results indicated that both air and pasteurized sediments stored at 4 °C in the dark for the 540 d tested (ANOVA, p>0.07 for both samples), showing drying and pasteurization degraded TBT to DBT and MBT. Similar results were obtained for phenyltin species in pasteur- the longest stability reached with both treatment methods in comparison with air-dried samples. ized samples, in which a decrease of TPT and a corresponding increase of DPT contents were observed (68.3 and 25.9%, Temperature aVected the stability of the organotin compounds.A significant decrease in the TBT content was respectively). Although a CRM should be representative of real samples, one of the most important requirements for the observed after 3 months in both air-dried and freeze-dried sediments stored at 25 °C (t-test, p<0.01) and after 1 month production of this kind of material is its long-term stability.Consequently, some compromise has to be found between the when stored at 40 °C (t-test, p<0.01). A less pronounced influence of the temperature on the organotin stability was matrix resembling natural samples and stability. Although neither air drying nor pasteurization is not advisable for observed in pasteurized sediments, in which marked TBT losses were only observed after 1 year of storage at 40 °C.It environmental studies as they change the composition of the Table 2 Influence of the stabilization treatment on the organotin concentrations in a candidate CRM sediment Concentration (ng g-1 as Sn, dry mass basis)±one standard deviation Treatment TBT DBT MBT MPT DPT TPT Freeze-drying 418±14 225±12 58.5±5.3 32.9±3.1 10.8±0.5 22.0±0.5 Air drying 303±12 241±11 105±35 44.3±7.3 17.3±0.7 27.2±2.2 Air drying+pasteurization 189±9.7 284±2.4 123±10 20.4±1.5 13.6±1.7 6.98±1.54 200 J.Environ. Monit., 1999, 1, 197–202Fig. 4 Stability of organotin species in air-dried sediments followed by pasteurization stored in the dark at the following temperatures: (a) Fig. 5 Stability of butyltin species in: (a) frozen C.gigas; (b) frozen and (b) 40 °C; (c) and (d) 25 °C; (e) and (f ) 4 °C; (g) and (h) -20 °C. C. edulis; (c) C. gigas freeze-dried at 4 °C; (d) C. edulis freeze-dried at Symbols and error bars represent means and standard deviation of 4 °C. All samples were stored in the dark. Symbols and error bars organotin concentrations in triplicate analyses. represent means and standard deviation of organotin concentrations by triplicate analyses.is advisable to prepare the CRM in such way that the samples in the dark, the three butyltin species being stable for only 5 remain stable at room temperature, avoiding special storage months (ANOVA, p>0.15). This behavior indicated that TBT conditions because of the large amount of CRM to be prodegraded by a stepwise debutylation mechanism to DBT, cessed.Moreover, they should not be aVected by short expo- MBT and inorganic tin. A higher stability has been obtained sure to extreme conditions during shipment of the material to for freeze-dried mussels stored under experimental conditions customers. Therefore, air drying followed by pasteurization similar to those used in this study, no changes in butyltin was considered to be the optimum treatment.contents being found after 44 months of storage.12 Possibly the diVerent nature of the biological organism tested could Organotin stability in biota samples account for these diVerences. Prior to studying the stability of the butyltin compounds in biota samples, the possibility of losses of these compounds Conclusions during the lyophilization process was evaluated.Analyses were performed on fresh and freeze-dried samples and the results Only a few preservation studies concerning aqueous, sediment and biota samples containing organotin compounds have been were compared. Five replicate analyses were performed for each sample and no significant diVerences were found (by reported in the literature.Storage of unfiltered and nonacidified sea-water in polycarbonate bottles at 4 °C in the dark using the regression line test, the correlation coeYcient was 0.997 and the calculated slope and intercept did not diVer is suitable to achieve good stability for butyltins and phenyltins for 7 and 3 months, respectively, which allows a considerable significantly from the values of 1 and 0, respectively), which indicated that lyophilization is an excellent procedure to dry period of time between sampling and the final determination.However, better preservation was achieved using C18 cartridges the samples without losses or interconversions of butyltin compounds. even at ambient temperature with the additional advantage of the small space necessary to store the samples, which is The results for butyltin contents as a function of the storage time are shown in Fig. 5. According to these results, butyltin important if a large number of samples have to be analyzed. Both butyl- and phenyltin species were stable in air-dried, compounds were stable in both cockles and oysters stored at -20 °C in the dark over a 7 month period (ANOVA, p>0.11).freeze-dried and pasteurized sediments stored at -20 °C for at least 540 d, which allows the use of the three treatments to However, a decrease in the TBT content of about 14% was detected after 270 d followed by a increase in DBT content (t- obtain suitable stability for a candidate CRM. However, although the organotin content changed during pasteurization, test, p<0.007).A longer storage period produced a decrease in both TBT and DBT levels and a corresponding increase of the long-term stability of this kind of sample was less aVected by relatively high temperatures (40 °C), which is important MBT and finally, after 540 d, a decrease in all the butyltin compounds was observed (t-test, p<0.001). Comparable when extreme conditions are present during CRM shipment.Finally, the maximum time of storage for oyster and cockle results were obtained for freeze-dried samples stored at 4 °C J. Environ. Monit., 1999, 1, 197–202 2017 M. Abalos, J. M. Bayona, R. Compan�o� , M. Granados, C. Leal samples at either-20 or 4 °C after freeze-drying was 6 months, and M. D. Prat, J. Chromatagr., 1997, 788, 1. which allows adequate preservation for both environmental 8 K.Bergmann, U. Ro�hr and B. Neidhart, Fresenius’ J. Anal. studies and interlaboratory quality control analysis. However, Chem., 1994, 349, 815. other types of biological organisms such as mussels are more 9 Ph. Quevauviller and Q. F. X. Donard, Fresenius’ J. Anal. Chem., useful for preparing CRMs owing to the long-term stability 1991, 339, 6. 10 J. L. Go� mez-Ariza, E. Morales, R. Beltra�n, I. Gira�ldez and of butyltin species. M. Ruiz-Bený�tez, Quý�m. Anal., 1994, 13, s76-s79. 11 A. M. Caricchia, S. Chiavarini, C. Cremisini, R. Morabito and R. Scerbo, Anal. Chim. Acta, 1994, 286, 329. Acknowledgement 12 Ph. Quevauviller, R. Morabito, L. Ebdon, W. Cofino, H. Muntau and M. J. Campbell, EUR Report, EN 17921, European The authors express their thanks to the Measurements and Commission, Brussels, Belgium, 1997. Testing Programme (BCR) Project MAT1-CT94–071 and to 13 J. L. Go�mez-Ariza, E. Morales and M. Ruiz-Benitez, Analyst, DGICYT (Direccio�n General de Investigacio�n Cientý�fica y 1992, 117, 641. Te�cnica), Grant No. Pb95–0731. 14 J. L. Go�mez-Ariza, E. Morales and M. Ruiz-Benitez, Appl. Organomet. Chem., 1992, 6, 279. 15 J. L. Go� mez-Ariza, R. Beltran, E. Morales, I. Gira�ldez and M. Ruiz-Benitez, Appl. Organomet. Chem., 1994, 8, 553. References 16 J. L. Go�mez-Ariza, E. Morales, I. Gira�ldez and R. Beltra�n, Inter. 1 C. Alzieu, Mar. Environ. Res., 1991, 32, 7. J. Environ. Anal. Chem., 1997, 66, 1. 2 M. D. Stephenson, D. R. Smith, J. Goetz, G. Ichikawa and M. 17 J. L. Go� mez-Ariza, E. Morales, I. Gira�ldez, R. Beltran and Martin, in Proceedings of the Organotin Symposium of the Oceans J. A. Pozas-Escobar, Fresenius’ J. Anal. Chem., 1997, 357, 1007. ’86 Conference, Washington, DC, Marine Technology Society, 18 J. A. Navý�o, C. Cerrillos, M. A. Pradera, E. Morales and 1986, p. 1246. J. L. Go�mez-Ariza, Langmuir, 1998, 14, 388. 19 C. A. Dooley and V. Homer, Naval Oceans Systems Technical 3 E. Oberdo� rster, D. Rittschof and P. McClellan-Green, Mar. Report No. 197, 1983, San Diego. Pollut. Bull., 1998, 36, 144. 20 W. R. Blair, G. H. Olson, F. E. Brinckman, R. C. Paule and D. A. 4 J. Widdows and D. S. Page, Mar. Environ. Res., 1993, 35, 233. Becker, Natural Bureau of Standards, Gaithersburg, MD, 1986. 5 F. Pannier, A. Astruc and M. Astruc, Appl. Organomet. Chem., 1994, 8, 595. 6 Ph. Quevauviller, Mikrochim. Acta, 1996, 123, 3. Paper 8/08043