Document 0g9qXpbwZzvd7xvzeYEe3EnXM

4 R i l 3,6 - REPORT 96-HOUR ACUTE TOXICITY STUDY IN ZEBRA-FISH WITH T-4127 (STATIC) Sl * u S im e o n NOTOX Project 334338 NOTOX Substance 113607 - Page 1 of 38 - OCDJ 1 G~k "'I T-4127 STATEMENT OF GLP COMPLIANCE NOTOX Project 334338 NOTOX B.V., 's-Hertogenbosch, The Netherlands The study described in this report has been correctly reported and was conducted in compliance with the most recent edition of: The OECD Principles of Good Laboratory Practice which are essentially in conformity with: The United States Food and Drug Administration. Title 21 Code of Federal Regulations Part 58. The United States Environmental Protection Agency (FIFRA). Title 40 Code of Federal Regulations Part 160. The United States Environmental Protection Agency (TSCA). Title 40 Code of Federal Regulations Part 792. Study Director Drs. M. Bogers Management: Ing. E.J. van de Waart, M.Sc. Head of Genetic & Ecotoxicology T-4127 QUALITY ASSURANCE STATEMENT NOTOX Project 334338 NOTOX B.V., 's-Hertogenbosch, The Netherlands This report was audited by the NOTOX Quality Assurance Unit to ensure that the methods and results accurately reflect the raw data. The dates of Quality Assurance inspections and audits are given below. During the on-site inspections procedures applicable to this type of study were inspected. DATES OF QAU INSPECTIONS/AUDITS on-site inspection(s) (Process) May 13-31,2002 (Analytical Support) July 08-15, 2002 (Ecotoxicology) protocol inspection (s) (Study) December 06, 2001 report audit(s) (Study) April 28-29, 2003 REPORTING DATES June 04, 2002 July 17, 2002 December 06, 2001 April 29, 2003 Head of Quality Assurance C.J. Mitchell B.Sc. -3- _ m T-4127 NOTOX Project 334338 SUMMARY 96-Hour Acute Toxicity Study in Zebra-fish with T-4127. The study procedures described in this report were based on the EEC directive 92/69; Part C: methods for the determination of ecotoxicity, Publication No. L383, December 1992, C.1. "Acute toxicity for fish", and the OECD guideline No. 203: "Fish Acute Toxicity Test", Adopted 17 July, 1992. T-4127 is a salt consisting of Perfluorooctanesulfonylmethylamide (anionic part) and Triphenylbenzylphosphonium (cationic part). Based on the response of the anionic part of the test substance water solubility of T-4127 was determined to be between 0.6 and 0.9 mg/l at 19.5 0.6C. The study started with a limit test exposing seven fish per concentration to a filtered and an unfiltered solution, both prepared at loading rates of 100 mg/l and both magnetically stirred for three days. A blank-control was also included. Analysis of the samples taken during the limit test showed that the average exposure concentration calculated over the first 24 hours was 0.76 mg/l in the filtered solution and 0.28 mg/l in the unfiltered solution. The average exposure concentration calculated over the 96-hour test period was 0.51 mg/l in the filtered solution and 0.16 mg/l in the unfiltered solution. During the first 24 hours of exposure all fish died in the filtered solution, while only 1 out of seven died in the unfiltered solution and no more fish died during the remaining test period. Hence the biological effects confirmed a higher actual concentration in the filtered solution. A final test was performed exposing seven fish per concentration to 4.5, 10, 20, 45 and 100% of a Water Accommodated Fraction (WAF) prepared at a loading rate of 100 mg/l and a blank control for a maximum of 96 hours. All glassware used in the final test was pre-treated with dichlorodimethylsilane. Although vapour pressure was relatively low, based on the Henry constant evaporation was still a possible cause for decrease of test concentrations. Therefore the study was performed in air-tight vessels. Samples taken during the final test were added directly to isopropanol at a ratio of 1:5 to prevent any additional loss. The analytical program based on measurement of the anionic part showed that the actual concentrations could not be maintained at more than 80 % of the initial concentration, in spite of all precautions taken to prevent adsorption to glass and volatilization. The measured concentrations could not be quantified for the whole test period with the exception of the two highest concentrations which are relevant for determination of the LC50. In the control no mortality was observed. Further, all test conditions (pH, oxygen concentration and temperature) remained within the ranges prescribed by the protocol. A water phase of T-4127 originating from a loading rate of 100 mg/l, induced 100% mortality in fish. This corresponded with 0.76 mg/l for 24 hours (limit test) and 0.44 mg/l for 48 hours (final test) when based on the anionic part in the water phase. No mortality or less than 50% mortality occurred at average concentrations of 0.095 mg/l (final test) and 0.16 mg/l (limit test). Hence, the 96h-LC50 of zebra-fish exposed to T-4127 was < 1 mg/l (between 0.16 and 0.44 mg/l) based on quantification of the anionic part. Limiting the analytical support on only the anionic part of T-4127, validity of the test could not be ensured. There is a realistic possibility that the actual concentrations of the cationic part were higher and more stable than those of the anionic part. However, the sponsor did not agree with further research and decided to accept the highly toxic labelling (R50) based on the current results. > T-4127 NOTOX Project 334338 PREFACE Sponsor Study Coordinator Study Monitor Testing Facility Aquatic Toxicology: Study Director Technical coordinator Analytical Chemistry: Principal Scientist Study Plan 3M Corporate Toxicology 3M Center, Building 220-2E-02 P.O. Box 33220 ST. PAUL, MINNESOTA 55133-3220 U.S.A. Mrs. M. Mitchell Mrs. Dr. S. Beach 3M Environmental Technology and Safety Services 935 Bush Avenue, Building 2-3E-09 ST. PAUL, MINNESOTA 55144 U.S.A. NOTOX B.V. Hambakenwetering 7 5231 DD 's-Hertogenbosch The Netherlands Drs. M. Bogers Ing. E.J.H. Mutsaards Ir. M.J.C. Brekelmans Start of project: December 05, 2001 Start of first exposure: July 01,2002 Completion of last exposure: November 15, 2002 Completion of analysis: November 15, 2002 Completion of project: May 09, 2003 TEST SUBSTANCE Identification Description Batch Composition Test substance storage Stability under storage conditions Expiry date Stability in water T-4127 Dark amber waxy solid D-2491 lot 2 95 - 99% Fluoroelastomer curative <2% N-Methyl Perfluorooctanesulphonamide <1% Methyl Alcohol <1% Ethyl Alcohol <1% Isopropyl Alcohol At room temperature in the dark Stable 31 March 2003 Not indicated The sponsor is responsible for all test substance data unless determined by NOTOX. . ^. A A A /# T-4127 NOTOX Project 334338 PURPOSE The purpose of the study is to evaluate the test substance for its ability to generate acute toxic effects in Danio rerio during an exposure period of 96 hours and, if possible, to determine the LC50 at all observation times. GUIDELINES The study procedures described in this report were based on the EEC directive 92/69; Part C: methods for the determination of ecotoxicity, Publication No. L383, December 1992, C.1. "Acute toxicity for fish", and the OECD guideline No. 203: "Fish Acute Toxicity Test", Adopted 17 July, 1992. DEFINITIONS Fish were considered to be dead when no reaction was observed after touching the caudal peduncle and visible breathing movements are absent. The LC50 is the concentration killing 50% of the fish after a defined period of exposure. TEST SYSTEM Species Source Mean length Mean weight Characteristics Reason for selection Total fish used Zebra-fish (Danio rerio, Teleostei, Cyprinidae). Ornamental Fish Hatchery Atlanta, Hellevoetsluis, the Netherlands. Limit test: 2.5 0.2 cm Final test: 2.7 0.2 cm Limit test: 0.24 0.06 g Final test: 0.36 0.12 g Healthy fish supplied with a health certificate. This system has been selected as an internationally accepted species. 63 - R- ^OQObn7 T-4127 NOTOX Project 334338 HOLDING Quarantine/Acclimatisation Medium Measurements Feeding Control of sensitivity Validity of batch At least 12 days after delivery. ISO-medium, formulated using Milli-Ro water (tapwater purified by reverse osmosis; Millipore Corp., Bedford, Mass., USA) with the following composition: C a 2+ 80 mg/l Mg2* 12 mg/l N a * 15 mg/l K+ 3 mg/l c r 145 mg/l S42' 49 mg/l h c o 3' 47 mg/l Hardness is 250 mg CaCCM pH, nitrate and nitrite concentration and ammonia concentration: once a week. Temperature: every day. Daily with Trouvit. The results of the most recent performed test are appended to the report. In the batch of fish used for the test, mortality during the seven days prior to the start of the test was less than 5%. PREPARATION OF TEST SOLUTIONS The standard test procedures required generation of test solutions, which contain completely dissolved test substance concentrations or stable and homogeneous mixtures or dispersions. The testing of concentrations that disturbed the test system were prevented (e.g. film of the test substance on the water surface). T-4127 is a salt consisting of Perfluorooctanesulfonylmethylamide (anionic part) and Triphenylbenzylphosphonium (cationic part). Based on the response of the anionic part of the test substance water solubility of T-4127 was determined to be between 0.6 and 0.9 mg/l at 19.5 0.6C (NOTOX Project 340842, using the column elution method). Limit test Two stock solutions were prepared by weighing 500 mg of T-4127 on a glass surface (twice). After placing these carriers in measuring flasks 5 litre of test medium was added to each flask to obtain a loading rate of 100 mg/l. Both mixtures were magnetically stirred for three days. After the stirring period the mixtures were clear but contained test substance particles, which settled at the bottom of the flasks. One of the mixtures was used as such while the second mixture was filtered through a paper filter (Schleicher and Schuell 604) to remove the larger undissolved test substance particles (ca. > 5//m). The final test solution was clear and colourless. .? "7 A A A A ^ A T-4127 NOTOX Project 334338 Final test All glassware used In the final test was pre-treated with dichlorodimethylsllane to prevent loss of test substance due to adsorption to glass surfaces. A stock solution was prepared by weighing 500 mg of T-4127 on a glass surface. After placing the carrier in a measuring flask 5 litre of test medium was added to obtain a loading rate of 100 mg/l. Simultaneously, another measuring flask was prepared in the same way but without adding test substance (blank-control). After the stirring period, the solution prepared at a loading rate of 100 mg/l was transferred to a separation funnel and left to stabilise for 23A hours. Subsequently the Water Accommodated Fraction (WAF) was separated from the centre of the separation funnel. Lower test concentrations were prepared by dilution of the WAF. All final test solutions were clear and colourless. Part of the test solutions prepared were used for testing with Daphnia magna (NOTOX Project 334349). LIMIT TEST A limit test was performed exposing seven fish per concentration to a filtered and an unfiltered solution prepared at a loading rate of 100 mg/l and a blank-control. Sampling: Frequency Volume Storage at t=0 h, t=24 h and t=96 h. 10 ml from the approximate centre of the test vessel. Not applicable, all samples were analysed on the day of sampling. FINAL TEST: TEST CONCENTRATIONS T-4127 Control 4.5, 10, 20, 45 and 100% of a WAF prepared at a loading rate of 100 mg/l. Test medium without test substance or other additives (Blank-control). TEST PROCEDURE AND CONDITIONS Test duration 96 hours Test type Static without interim renewal of test solutions. Test vessels 10 litres, all-glass. Pre-treated with dichlorodimethylsilane. Although vapour pressure was relatively low, based on the Henry constant evaporation was still a possible cause for decrease of test concentrations. Therefore the study was performed in air-tight vessels. Test medium ISO-medium, aerated until the dissolved oxygen concentration had reached saturation and the pH had stabilised. After aeration the hardness was 250 mg CaC03 per litre and the pH was 8.0. T-4127 NOTOX Project 334338 Number of fish Loading Illumination Aeration Feeding Introduction of fish Euthanasia 7 fish per concentration and control. 0.28 g fish/litre, i.e. 7 fish per 9 litres of test medium. 16 hours photoperiod daily The test media were not aerated during the test. No feeding from 48 hours prior to the test and during the total test period. Within 11/2 hours after preparation of the test media. At the end of the test the surviving fish were rapidly killed by exposing them to ca. 1.2% ethylene glycol monophenylether in water. SAMPLING FOR ANALYSIS OF TEST CONCENTRATIONS Duplicate samples were taken from the stock solution directly after preparation of the WAF as a specification of the origin of samples taken during the test. During the final test duplicate samples were taken from all concentrations and the blank-control at the start and after 48 hours of exposure. At the end of the test period duplicate samples were taken from all dilutions (4.5, 10, 20 and 45%) of the WAF and the blank-control for analysis. The method of analysis is described in the appended Analytical Report. Sampling: Volume T reatment Storage 5 ml from the approximate centre of the test vessels. Samples were added directly to isopropanol at a ratio of 1:5 to prevent any additional loss. Not applicable, samples were analysed on the day of sampling. MEASUREMENTS AND RECORDINGS Mortality and other effects Fish length and weight At 2Vz, 24, 48, 72 and 96 hours following the start of exposure. Dead fish were removed when observed. Ten fish of the batch used for the test, were weighed and measured prior to the start of the test. Dissolved oxygen content, pH and temperature. Daily in all vessels, beginning at the start of the test (day 0). DATA HANDLING Determination of the average exposure concentrations The average exposure concentrations over the first 24 or 48 hours of exposure were calculated as t/ C (=0 x Ct=x , being the geometric means of the concentrations of T-4127 measured in the samples taken at the start (Ct=0) and after 24 or 48 hours of exposure (Ct=x). T-4127 NOTOX Project 334338 The average exposure concentrations over the 96-hour test period were calculated as: 2 4 x ^ C t=0 x C t=24 +72xV C t=24 x C t_g6 96 48x^/Ct=o x C t=48 + 48x.^Ct=4g x C t=96 96 Being the geometric means of the concentrations of T-4127 measured in the samples taken at the start (Ct=0), after 24 or 48 hours (Ct=24 or Ct=48) and the end of the test (C,=96). RESULTS Limit test: Measured concentrations The limit test included samples prepared at a nominal loading rate of 100 mg/l either filtered through 5 jm or unfiltered. Samples were diluted with methanol in 1:1 (v:v) ratio and analyzed using the LCQduo LCMSMS (injection volume 100//I). Concentrations in the filtered sample decreased from 1.0 mg/l to 0.33 mg/l during the test whereas concentrations in the unfiltered sample decreased from 0.45 mg/l to 0.08 mg/l, see also Table 7 of the appended Analytical Report. The average exposure concentration calculated over the first 24 hours was 0.76 mg/l in the filtered solution and 0.28 mg/l in the unfiltered solution. The average exposure concentration calculated over the 96-hour test period was 0.51 mg/l in the filtered solution and 0.16 mg/l in the unfiltered solution. Mortality and other effects The mortality data are presented in Table 1. During the first 24 hours of exposure all fish died in the filtered solution, while only 1 out of seven died in the unfiltered solution and no more fish died during the remaining test period. Hence the biological effects confirmed a higher actual concentration in the filtered solution. Table 1: Incidence of mortality and total mortality during the limit test. Loading rate Initial Cum ulative m ortality Total T-4127 (m g/l) num ber M ortality 3yh 24h 48h 72h 96h o f fish (% ) Blank-control 7 0 0 0 0 0 0 100; filtered 7 o1 7 7 7 7 100 100; unfiltered 7 o2 13 14 1s 15 ' 2 fish were slow and snapping at the surface, while the other 5 were slow and swimming at the bottom. 2 1 fish was immobile 3 All surviving fish were slow and 1 fish was also swimming at the surface. 4 4 fish were swimming at the surface. 5 All fish were discoloured. 100 Experimental conditions The results of measurement of pH and oxygen concentrations are presented in Table 2. The results of measurement of the temperature in the various test solutions is presented in Table 3. T-4127 NOTOX Project 334338 Table 2: pH-values and dissolved oxygen concentrations (mg/l) during the limit test. Loading rate T-4127 (mg/l) Blank-control 100; filtered 100; unfiltered Day 0 pH 0 2 8.0 8.8 8.1 8.4 7.9 8.1 Day 1 pH 0 2 7.8 8.3 7.7 6.4 7.6 6.8 Day 2 pH 0 2 7.7 7.9 7.3 7.3 6.5 Day 3 pH 0 2 7.7 7.8 -- 7.5 6.7 Table 3: Temperatures (C) measured during the limit test. 04 0 1 a. Day 4 7.7 7.5 -- 7.5 6.4 Loading rate T-4127 (m g/l) Blank-control 100; filtered 100; unfiltered Day 0 22.9 2 2 .8 23.1 Day 1 21.8 21.9 21.8 Day 2 21.7 21.9 21.8 Day 3 21.8 - 21.9 Day 4 21.8 - 21.8 Final test: Measured concentrations The results of analysis of the samples taken during the study are described in Table 8 of the appended Analytical Report. During the final test the exposure consisted of a range of dilutions of a water fraction prepared at 100 mg/l containing 4.5 to 100% of this water fraction. Samples of 5 ml taken from these dilutions were added to 25 ml isopropanol and analyzed using the API300 LCMSMS (injection volume 10//I). Recovery samples were prepared in ISO-medium at 0.01 mg/l, 0.1 mg/l and 0.6 mg/l. The 0.01 mg/l concentration proved to be below LOD. Recoveries at 0.1 mg/l and 0.6 mg/l were relatively high (117% and 111 % respectively). Samples were taken in duplicate from the test solutions and each pretreated sample was injected in triplicate. The results are listed in Table 4. Note that the decrease in concentration was more than 20% in spite of all precautions and measures taken. T-4127 NOTOX Project 334338 Table 4: Mean of duplicate samples and their deviation. T-4127 % of a WAF (100 m g/l) 4.5 10 20 45 100 4.5 10 20 45 100 4.5 10 20 45 M easured concentration (m g/l) < LOD 0.064 0.16 0.24 0.56 < LOD < LOD 0.044 0.073 0.30 < LOD < LOD < LOD 0.049 Oh 48 h 96 h 0.024 0.029 0.13 0.20 0.75 < LOD < LOD 0.038 0.083 0.31 < LOD < LOD < LOD 0.040 Mean concentration (m g/l) 0.024 0.046 0.144 0.222 0.655 Mean 0.041 0.078 0.301 Mean 0.045 D e v ia tio n 55 12 12 20 D e v ia tio n 11 9 3 D e v ia tio n 14 Duplicate samples showed a significant difference in concentration at t=0. At t=48 hours and at t=96 hours, concentrations of duplicate samples were similar. Figure 1 includes the curves for the exponential decrease of the test concentrations with identical slopes for the concentrations measured in 45 and 100% of the water fraction. Figure 1: Concentration curves for the three highest test concentrations. an n n n n > o T-4127 NOTOX Project 334338 Mortality and other effects The mortality data are presented in Table 5. Table 6 specifies the clinical effects observed at different test concentrations. All fish died at the highest treatment corresponding with an average concentration of 0.44 mg/l during the first 48 hours, whereas no fish died in the test solution containing 45% of the water fraction corresponding with an average concentration of 0.095 mg/l. Table 5: Incidence of mortality and total mortality during the final test. T-4127 % of a WAF (100 m g/I) Blank-control 4.5 10 20 45 100 Initial number of fish 7 7 7 7 7 7 2 '/2h 0 0 0 0 0 0 Cum ulative m ortality 24h 48 h 72h 000 000 000 000 000 577 Table 6: Clinical effects observed during the final test. 96 h 0 0 0 0 0 7 T-4127 % of a WAF (100 m g/l) 10 Tim e of recording (hours) 24 and 72 Specification of effects Swimming at the surface 45 24 Slow swimming Swimming at the surface 48 Slow swimming and swimming at the surface 72 and 96 Slow swimming 100 24 Immobile Total M ortality (% ) 0 0 0 0 0 100 R e la tiv e number 7/7 2/7 7/7 7/7 7/7 2/2 Experimental conditions The results of measurement of pH and oxygen concentrations are presented in Table 7. The results of measurement of the temperature in the various test solutions is presented In Table 8. Table 7: pH-values and dissolved oxygen concentrations (mg/l) during the final test. T-4127 % of a WAF (100 m g/l) Blank-control 4.5 10 20 45 100 Day 0 pH 0 2 8.0 8.7 8.0 8.4 7.9 8.4 7.8 8.4 7.9 8.2 7.8 7.6 Day 1 pH 0 2 7.4 6.6 7.5 7.6 7.5 7.7 7.4 7.4 7.3 6.3 7.2 6.0 CM X Q. CM O X a Day 2 7.5 6.6 7.5 7.5 7.5 7.4 7.5 7.1 7.2 5.7 -- Day 3 7.6 6.5 7.6 7.3 7.6 7.3 7.5 6.6 7.4 5.5 -- Day 4 pH 0 2 7.5 6.7 7.5 7.2 7.5 7.1 7.5 6.2 7.4 5.4 -- T-4127 NOTOX Project 334338 Table 8: Temperatures (C) measured during the final test. T-4127 % of a WAF (100 m g/l) Blank-control 4.5 10 20 45 100 DayO 22.0 21.8 21.8 21.9 22.1 21.8 Day 1 22.3 22.4 22.4 22.3 22.3 22.3 Day 2 22.3 22.3 22.3 22.3 22.1 - Day 3 22.2 22.3 22.2 22.1 22.1 - Day 4 22.5 22.6 22.5 22.4 22.3 - ACCEPTABILITY OF THE TEST 1. No mortality was observed in the control group. . 2. All test conditions (pH, oxygen concentration and temperature) remained within the ranges prescribed by the protocol. 3. The analytical program based on measurement of the anionic part showed that the actual concentrations could not be maintained at more than 80 % of the initial concentration, in spite of all precautions taken to prevent adsorption to glass and volatilization. The measured concentrations could not be quantified for the whole test period with the exception of the two highest concentrations which are relevant for determination of the LC50. CONCLUSIONS Under the conditions of the present tests a water phase of T-4127 originating from a loading rate of 100 mg/l, induced 100% mortality in fish. This corresponded with 0.76 mg/l for 24 hours (limit test) and 0.44 mg/l for 48 hours (final test) when based on the anionic part in the water phase. No mortality or less than 50% mortality occurred at average concentrations of 0.095 mg/l (final test) and 0.16 mg/l (limit test). Hence, the 96h-LC50 of zebra-fish exposed to T-4127 was < 1 mg/l (between 0.16 and 0.44 mg/l) based on quantification of the anionic part. Limiting the analytical support on only the anionic part of T-4127, validity of the test could not be ensured. There is a realistic possibility that the actual concentrations of the cationic part were higher and more stable than those of the anionic part. However, the sponsor did not agree with further research and decided to accept the highly toxic labelling (R50) based on the current results. T-4127 NOTOX Project 334338 REFERENCE TEST 96-hour acute toxicity study in the zebra-fish with PCP; NOTOX Project 352441 (Batch 02-01 ZA) The study procedures described in this report were based on the EEC directive 92/69, Part C.1. "Acute toxicity for fish"; and the OECD guideline No. 203: "Fish Acute Toxicity Test", Adopted 17 July, 1992. Start: 17-06-2002 End: 21-06-2002 This reference test was carried out to check the sensitivity of the test system as used by NOTOX. The reference substance was PENTACHLOROPHENOL (PCP, SIGMA, Art. P-9441, Batch 103H3488). Concentrations: 0.06, 0.10, 0.15, 0.22 and 0.33 mg/l in ISO-medium. Control: ISO-medium without test substance. Incidence of mortality observed in the reference study: Concentration PCP (mg/l) N o m in al Control 0.06 0.10 0.15 0.22 0.33 Initial Number Of fish 5 5 5 5 5 5 Cum ulative number of dead fish recorded at various tim e points after start of exposure 3h 24h 48h 72h 96h 0 00 0 0 0 00 0 0 0 000 0 0 000 0 000 0 0 0 24 44 Total M o rtality (%) 0 0 0 0 0 80 During the test the pH, oxygen concentration and the temperature of the medium were within the optimal ranges for fish. Under the conditions of the present test PENTACHLOROPHENOL induced no lethal effects in zebra-fish at or below 0.22 mg/l. The 96h-LC50for zebra-fish exposed to PCP was 0.29 mg/l with 0% mortality at 0.22 mg/l and 80% mortality at 0.33 mg/l. The 24h-LC50 was 0.35 mg/l with 0% mortality at 0.22 mg/l and 40% mortality at 0.33 mg/l. The range of the 96h-LC50for zebrafish is generally between 0.18 mg/l and 0.30 mg/l (with a 95% confidence interval between 0.28 and 0.39 mg/l) based on historical data of reference tests performed by NOTOX. The response observed in zebra-fish originating from the present batch falls within this range. The raw data and report from this study are kept in the NOTOX archives. The test described above was performed under GLP-conditions with a QA-check. ANALYTICAL REPORT 96 HOUR ACUTE TOXICITY STUDY IN ZEBRA-FISH WITH T-4127 (STATIC); DETERMINATION OF THE CONCENTRATIONS NOTOX Project 334338 NOTOX Substance 113607 - Page 1 of 23 - T-4127 REPORT APPROVAL PRINCIPAL SCIENTIST: NOTOX Project 334338 Ir. M.J.C. Brekelmans (Analytical Chemistry) Date: cS, zcol /\/\A A 4 fi T-4127 NOTOX Project 334338 PREFACE Analytical study Start: Completed: 1 July 2002 15 November 2002 PURPOSE The purpose of the analytical study was to determine the test concentrations and to validate the analytical methods used. REAGENTS _______________ Milli-Q water Tap water purified by reversed osmosis and subsequently passed over activated carbon and ionexchange cartridges; Millipore, Bedford, MA, USA Methanol P.a., Merck, Darmstadt, Germany ISO-medium see main report Ammonium acetate Fractopur, Merck, Darmstadt, Germany 2.0 mM ammonium acetate 154 mg ammonium acetate in 1 litre of Milli-Q water Isopropanol (2-propanol) p.a., Merck, Darmstadt, Germany Internal standard solution 100 pi of a solution of 1204 mg/l perfluorooctane sulfonate in methanol filled up to 10 ml with 50/50 (v/v) ISO-medium/methanol. Isopropanol containing internal standard 140 pi of a solution of 2816 mg/l perfluorooctane sulfonate in methanol filled up to 2000 ml with isopropanol. INTERNAL STANDARD Identification number Name Description Batch number Purity Expiry Date Certified Storage conditions Supplier AS517 Perfluorooctane sulfonate (FC-95) White solid (determined at NOTOX) Lot 217 90.49% 21 February 2003 (determined at NOTOX) Yes At room temperature in the dark 3M The sponsor is responsible for all test substance data unless determined by NOTOX. T-4127 NOTOX Project 334338 SAMPLE PRETREATMENT PROCEDURE Analyses in July 2002 Each sample (10 ml) was quantitatively transferred to a 20 ml volumetric flask. A volume of 200 pi internal standard solution was added to the samples taken at t=24 and t=96 hours, after which the flasks were filled up to the mark using methanol (internal standard concentration 120 pg/l). No internal standard was added to the flasks containing the t=0 samples. Analyses in November 2002 In order to prevent from adsorption, it was decided to change sample pre-treatment. Based on additional information supplied by the sponsor, each sample (5 ml) was transferred quantitatively into a 50 ml volumetric flask and 25 ml of isopropanol containing internal standard (internal standard concentration 164pg/l) was added. ANALYTICAL METHODS T-4127 is a salt consisting of Perfluorooctanesulfonylmethylamide (anionic part) and Triphenylbenzylphosphonium (cationic part). Upon request of the sponsor, quantitative analyses were only based on the anionic part of T-4127 using High Performance Liquid Chromatography with Mass Spectrometric detection (LCMSMS). The analytical method used in July 2002 was not sensitive enough to measure concentrations in pretreated samples from the final test. Therefore, a second method was used in November 2002. Conditions for the analytical method used in July 2002 Column Mobile phase Flow Column temperature Autosampler temperature Injection volume Detection T-4127, anionic part Internal standard Betasil C18, 50 x 2.0 mm; dp= 5 pm (Thermo Hypersil Keystone) Gradient Eluens A: 2.0 mM ammonium acetate Eluens B: 100 % methanol Time (min) Eluens A (%) Eluens B (%) 0,00 90 10 1.00 90 10 5.50 5 95 10.00 5 95 '11.00 90 10 13.00 90 10 300 pl/min ambient temperature ambient temperature 100 pi LCQ Duo mass spectrometer (Thermo Finnigan, San Jose, CA, USA) ESI negative mass-mass detection Position 3 Collision energy: 30% Isolation width: 1.5 M/z 512.0 -> 388.2, 419.0 ESI negative mass detection Position 3 M/z 499.2 T-4127 NOTOX Project 334338 Conditions for the analytical method used in November 2002 Column Column temperature Mobile phase Flow Column temperature Autosampler temperature Injection volume Detection Interface Monitored masses Betasil C18, 50 x 2.0 mm; dp=5 pm (Thermo Hypersil, Keystone) 25C Time 2.0 mM Ammonium acetate Methanol (minutes) (%) (%) 0 90 10 1 90 10 5.5 5 95 7.5 5 95 8.0 90 10 11 90 10 300 pl/min 25 C 4 C 10 pi SCIEX MSMS system API-300 mass spectrometer (Applied Biosystems, Toronto, Canada) Turbo ionspray at 450C; N2flow rate of 7000 ml/min.; operated in negative ion mode MRM test substance m/z 512.1 - > 168.8 MRM internal standard m/z 499.0 - > 99.0 Standard solutions Standard solutions of T-4127 were prepared in methanol. Calibration solutions used in July 2002 On each day of analysis, calibration solutions in 50/50 (v/v) methanol/ISO-medium were made up from two standard solutions. Internal standard solution was added to a final concentration of 120 pg/l. Calibration solutions used in November 2002 On each day of analysis, calibration solutions in 1/5 (v/v) ISO-medium/isopropanol containing internal standard (internal standard conceritration 164 pg/l) were made up from two standard solutions. VALIDATION OF THE ANALYTICAL METHOD The LCMSMS methods used were only partly validated. Upon request of the sponsor, validation was not completed. The parameters validated are given below. T-4127 NOTOX Project 334338 Validation of the analytical method used in July 2002 Specificity Blank ISO-medium was pretreated as specified in `sample pretreatment procedure' 1and subsequently injected in duplicate into the HPLC system. The resulting chromatograms were critically evaluated for interfering peaks by comparison with chromatograms of a test substance solution in the same medium. Interfering peaks are required to be s 30% of the LOQ. Linearity From two standard solutions (1020 and 1060 mg/l), eight dilutions were prepared in 50/50 (v/v) methanol/ISO-medium. This resulted in a concentration range of 0.0509 - 1.00 m g/l1, Each of these solutions was injected in triplicate. Responses were plotted against the concentrations. A linear regression program was used to calculate the regression line from the responses and concentrations. The correlation coefficient is required to be at least 0.99. Stability of the solutions Solutions of 0.0509 mg/l, 0.106 mg/l and 1.00 mg/l T-4127 1were injected two times (in triplicate) over a 9.6, 8.9 and 6.1-hour time period, respectively. The maximum deviation of the responses was calculated for each concentration. Validation of the analytical method used in November 2002 Specificity Blank ISO-medium was pretreated as specified in 'sample pretreatment procedure' 2and subsequently injected in duplicate into the HPLC system. The resulting chromatograms were critically evaluated for interfering peaks by comparison with chromatograms of a test substance solution in the same medium. Interfering peaks are required to be s 30% of the LOQ. Linearity From two standard solutions (782 and 874 mg/l), eight dilutions were prepared in 1/5 (v/v) ISOmedium/isopropanol containing internal standard. This resulted in a concentration range of 0.00500 - 0.100 mg/l 2. Each of these solutions was injected in duplicate. Responses were plotted against the concentrations. A linear regression program was used to calculate the regression line from the responses and concentrations. The correlation coefficient is required to be at least 0.99. Accuracy and Precision - repeatability Blank ISO-medium (5 ml) was spiked using 50 pi of a 1.00 mg/l, 10.0 mg/l or 60.1 mg/l solution of the test substance in methanol, yielding concentration levels of 0.00991 mg/l, 0.0991 mg/l and 0.595 mg/l test substance. At all concentration levels, six samples were prepared. These samples were treated as described in `sample pretreatment procedure' and analysed in triplicate. The recovery was calculated for each sample. The mean recovery (n=6 in triplicate) at each concentration level is required to be in the range 70-110%. 1 Internal standard was added to a final concentration 120 ng/l. 2 Internal standard was added to a final concentration 164 ng/l. T-4127 NOTOX Project 334338 Limit of quantitation (LOQ) The LOQ was determined as the lowest concentration at which the mean recovery is in the range 70-110% and the coefficient of variation of the recovery is s 20%. Limit of detection (LOP) A 0.00500 mg/l solution of the test substance in 1/5 (v/v) ISO-medium/isopropanol containing internal standard 2was injected in duplicate. In each chromatogram, the test substance peak height was measured as well as the noise level of the system (both in cps). The LOD was calculated from the mean peak height and the mean noise level. DATA HANDLING -VALIDATION OF THE ANALYTICAL METHOD Response: Peak area test substance R= Peak area internal standard Mean: x = - ^ Xi where: Xi : measured value n : number of measurements Standard deviation: Coefficient of variation: Maximum deviation: Linearity Validation of analytical method (July 2002) Sn- i=1 (rt-1 ) (standard deviation / mean value) * 100% [(highest - lowest)/mean] * 100% where 'mean' is the mean value of the highest and the lowest value. A regression program was used to calculate the regression line or curve from the responses and concentrations. Regression analysis was performed using the least squares method. A (1/concentration) weighting factor was used in November 2002. Regression curve: Y = f X2 + g X+ h 2 Internal standard was added to a final concentration 164 ng/l. T-4127 NOTOX Project 334338 Validation of analytical method (November 2002) Recovery: Limit of quantification (LOQ): Limit of detection (LOD): Regression line: Y = a X +b where: Y: response X: concentration a: slope b: intercept f, g, h: regression constants (Concentration analysed 3/ Concentration prepared) * 100% The lowest concentration of T-4127 tested with an acceptable recovery and coefficient of variation. The limit of detection is defined as the concentration of T-4127 with a signal (peak height) of three times the noise level (S/N=3). Limit of detection= ((3 * noise level)/ signal) * cone, where: noise level (N ): height of the noise [cps] signal (S): height of the test substance peak [cps] cone.: concentration of test substance [mg/l] DATA HANDLING - SAMPLE ANALYSIS Calibration Response: Calibration curve: (July 2002) Peak area test substance R = --------------------------------------- Peak area internal standard A regression program was used to calculate the regression curve from the responses and concentrations. Regression analysis was performed using the least squares method. If necessary, a (1/concentration) or (1/concentration2) weighting factor was used. R =a* C+b R = f * C 2+ g*C + h R : response calibration solution C : concentration of test substance in calibration solution [mg/l] a, b, f, g, h: regression coefficients On each day of analysis, a calibration curve or line was constructed using eight concentrations injected in triplicate. The coefficient of correlation was > 0.99. See `DATA HANDLING - SAMPLE ANALYSIS'. T-4127 NOTOX Project 334338 Calibration curve: (November 2002) Samples Recovery of recovery samples: Concentration relative to nominal: Concentration analysed: (Quadratic regression July 2002) The response was correlated with the concentration of test substance, using linear regression analysis (least squares method). R = a*C+b R = response calibration solution C = concentration in the calibration solution [mg/l] a = slope [l/mg] b = intercept On each day of analysis, two calibration solutions were used for quantification. Both calibration solutions were injected (in duplicate) before and after a maximum of six samples. Using the four responses, a calibration curve was constructed. Concentration analysed -------------------------------- * 100 [%] Concentration prepared Concentration analysed * 100 [%] Concentration nominal c _ d ~ - 9 + Vg2 - 4 * f * ( h - R) [mg/l] Concentration analysed : (Linear regression July 2002) Concentration analysed: (November 2002) C- = dH**-R---~--b [mg/l] a R d a, b, f ,g ,h : response sample [units] : dilution factor : regression constants rC = dH**-R---~--b- [mg/l] a R : response sample [units] d : dilution factor a : slope [units*l/mg] b : intercept [units] T-4127 NOTOX Project 334338 RESULTS -VALIDATION OF THE ANALYTICAL METHOD The calculations for the validation tests were performed using not-rounded concentrations and responses. Therefore, some differences might be observed when calculating the statistical parameters using the values as mentioned in the tables. Validation of the analytical method in July 2002 Specificity Figures 1 and 2 show chromatograms of a blank solution (50/50 (v/v) methanol/ISO-medium containing internal standard) and of a 1.00 mg/l T-4127 solution, respectively. It was clear that blank chromatograms did not contain any interfering peaks at the position of the test substance. Linearity The results are summarized in Table 1. The regression line is shown in Figure 3. Table 1 Linearity. Concentration [mg/l] Response 1 [units] 0.0509 0.0714 0.106 0.204 0.403 0.612 0.806 1.00 0.0109/0.0102/0.0098 0.0141 / 0.0149/ 0.0148 0.0215/0.0207/0.0226 0.0408 / 0.0407 / 0.0444 0.0701 / 0.0726 / 0.0697 0.0979/0.1038/0.1030 0.125/0.130/0.118 0.146/0.153/0.156 Triplicate measurements. These results show that there is a relationship (Y = 0.00244+ 0.1846 X - 0.03718 X2; R2 = 0.9959) between response and concentration in the concentration range of 0.059 - 1.00 mg/l though a deviation of more than 10% of the calibration points from the calculated line was observed at various concentrations. T-4127 Stability of the solutions The results are summarised in Tables 2-4. Table 2 Stability of a 0.0509 mg/l T-4127 solution. Elapsed time [hours] Response 0.0 0.0109 0.2 0.0102 0.5 0.0098 9.1 0.0107 9.4 0.0105 9.6 0.0095 NOTOX Project 334338 Maximum deviation [%] 14.2 Table 3 Stability of a 0.106 mg/l T-4127 solution. Elapsed time [hours] Response 0.0 0.0215 0.2 0.0207 0.5 0.0226 8.4 0.0217 8.7 0.0228 8.9 0.0204 Maximum deviation [%] 11.0 Table 4 Stability of a 1.00 mg/l T-4127 solution. Elapsed time [hours] Response 0.0 0.1456 0.2 0.1530 0.5 0.1559 5.6 0.1583 5.9 0.1426 6.1 0.1573 Maximum deviation [%] 10.3 These results show that the solutions were stable (maximum deviation <20%) over at least a 9.6-hour time interval at a concentration of 0.0509 mg/l, at least a 8.9-hour time interval at a concentration of 0.106 mg/l and at least a 6.1-hour time interval at a concentration of 1.00 mg/l. T-4127 NOTOX Project 334338 Validation of the analytical method in November 2002 Specificity Figures 4 and 5 show chromatograms of a blank solution (1/5 (v/v) ISO-medium/isopropanol containing internal standard) and of a 0.500 mg/l T-4127 solution, respectively. It was clear that blank chromatograms did not contain any interfering peaks at the position of the test substance. Linearity The results are summarized in Table 5. The regression line is shown in Figure 6. Table 5 Linearity. Concentration [mg/l] Response 1 [units] 0.00500 0.00699 0.0100 0.0199 0.0400 0.0600 0.0782 0.100 0.0742 / 0.0727 0.101 / 0.0944 0.127/0.126 0.272 / 0.260 0.563 / 0.558 2 0.780 / 0.803 0.891 / 0.952 1.23/1.16 Slope Intercept with Y-axis Weighting factor R 12.1 0.0134 1/concentration 0.9979 Duplicate measurements. Outliers; not used in calculation of the calibration curve. From these results, it was concluded that there is a linear relationship between response and concentration in the concentration range of 0.00500 - 0.100 mg/l if a (1/concentration) weighting factor is used. T-4127 NOTOX Project 334338 Accuracy and Precision - repeatability ISO-medium spiked with T-4127 to concentrations of 0.00991 mg/l, 0.0991 and 0.595 mg/l (and 164 jj.g/1 internal standard), was pretreated as described in `sample pretreatment procedure'. Chromatograms of pretreated 0.00991 mg/l, 0.0991 and 0.595 mg/l samples are shown in Figures 7, 8 and 9. The results are summarised in Table 6. Table 6 Accuracy and Precision. Date of Preparation (dd-mm-yy) Date of Analysis (dd-mm-yy) Cone. prepared [mg/l] Cone. analysed [mg/l] 15-Nov-02 15-Nov-02 0.00991 <LOD <LOD <LOD <LOD <LOD <LOD Recovery1 [%] Mean Recovery [%] Coefficient of variation [%] n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. 15-Nov-02 15-Nov-02 0.0991 0.114 0.111 0.118 0.117 0.118 0.118 115 117 2.4 112 119 118 119 119 15-Nov-02 15-Nov-02 0.595 0.644 2 0.641 2 0.668 2 0.670 2 0.658 2 0.662 2 108 108 112 113 111 111 111 1.8 1 Triplicate measurements. Maximum deviation o f triplicate measurements was <20% 2 Calculated by extrapolation. Responses were <110% of the high calibration level. Though mean recoveries were all (slightly) higher than 110%, the analytical method was considered acceptable for samples in ISO-medium in the concentration range of 0.0991 - 0.595 mg/l because the coefficient of variation was 20 % . Limit of quantification (LOQ) The LOQ for the samples in ISO-medium is reported to be 0.0991 mg/l though recovery was 117%. T-4127 NOTOX Project 334338 Limit of detection (LOP) From two chromatograms of a 0.00500 mg/l solution of the test substance, the mean noise level (N) was determined to be 22.5 cps. From the same chromatograms, the test substance signal (S), i.e. the mean height of the test substance peak, was determined to be 87.5 cps. Using these values, the limit of detection (S/N=3) was calculated to be 0.0039 mg/l at an injection volume of 10 pi. Because samples from the ecotoxicological study were diluted by a factor of 6, the limit of detection for these samples is 0.023 mg/l. RESULTS - SAMPLE ANALYSIS Samples July 2002 Table 7 shows the analytical results for the samples from the test performed in July 2002*. Table 7 Concentrations of T-4127 in test medium in July 2002 (Full test 1). Time of sampling [hours] Date of sampling [dd-mm-yy] Date of analysis [dd-mm-yy] Nominal [mg/l] Concentration Analysed1 [mg/l] Relative to nominal [%] 0 01-Jul-02 02-Jul-02 4 24 02-Jul-02 02-Jul-02 96 05-Jul-02 05-Jul-02 0 100 2 100 0 100 2 100 0 100 2 100 n.d. 1.030 0.447 n.d. 0.562 0.178 n.d. 0.332 0.082 3 n.a. 1.030 0.447 n.a. 0.562 0.178 n.a. 0.332 0.082 1 Mean of triplicate analysis. The maximum deviation between the responses calculated for each sample w as < 20%. 2 5 pm filtered solution prepared at a nominal concentration of 100 mg/l. 3 Calculated by extrapolation. Responses were >8Q% of the lowest calibration level. * Samples were pretreated on 01-Jul-02 and had to be stored overnight in a refrigerator together with the calibration solutions due to analytical problems. Inadvertently no Internal standard was added to these samples and calibration solutions. All relative values were calculated using not-rounded concentrations. Therefore, some differences might be observed when calculating the relative values using the concentrations as mentioned in the table. T-4127 NOTOX Project 334338 Samples November 2002 Table 8 shows the analytical results for the samples from the test performed in November 2002. Table 8 Concentrations of T-4127 in test medium in November 2002 (final test). Time of sampling [hours] Date of sampling [dd-mm-yy] Date of analysis [dd-mm-yy] Nominal [mg/l] Concentration Analysed1 [mg/l] Relative to nominal [%] 0 2 11-Nov-02 11-Nov-02 100 100 0 2 11-Nov-02 11-Nov-02 0 0 0.034 0.034 0.076 0.076 0.15 0.15 0.34 0.34 0.76 0.76 48 13-Nov-02 13-Nov-02 0 0 0.034 0.034 0.076 0.076 0.15 0.15 0.34 0.34 0.76 , 0.76 0.770 0.750 n.d. n.d. 0.021 3 0.024 34 0.064 0.029 3 0.16 0.13 0.24 0.20 0.56 0.75 n.d. n.d. < LOD < LOD 0.0223 0.021 3'4 0.044 0.038 0.073 0.083 0.30 0.31 n.a. n.a. 61 69 84 38 103 87 70 59 74 98 n.a. n.a. - - 29 27 29 25 21 24 39 40 T-4127 NOTOX Project 334338 (Table 8 Continued) Time of sampling [hours] Date of sampling [dd-mm-yy] Date of analysis [dd-mm-yy] Nominal [mg/l] 96 15-NOV-02 15-Nov-02 0 0 0.034 0.034 0.076 0.076 0.15 0.15 0.34 0.34 Concentration Analysed12 [mg/l] n.d. n.d. < LOD < LOD < LOD < LOD 0.021 3 0.019 3,4 0.049 4 0.040 4 Relative to nominal [%] n.a. n.a. - - 14 13 14 12 1 Mean of triplicate analysis. The maximum deviation between the responses calculated for each sample was < 20% unless otherwise indicated. 2 Combined with N O TO X project 334349. 3 Calculated by extrapolation. Responses were between 50 and 100% of the lowest calibration level. These concentrations were not reported as < LOD because they were only slightly below the limit of detection. 4 Maximum deviation between responses was 23%, 36%, 32% , 22% and 22%, respectively, n.d. Not detected. The limit of detection (LOD) was 0.023 mg/l. n.a. Not applicable. T-4127 Samples July 2002 NOTOX Project 334338 Figure 1 HPLC chromatogram of a blank (50/50 (v/v) methanol/ISO-medium containing internal standard). Relative Abundance T-4127 RT: 0 .0 0 -1 3 .0 0 100^ 90" 80i 70"E Internal standard 60-E 50i 40-E 30-E 2(H 10-z o4 J NOTOX Project 334338 NL: 1.46E7 m/z= 498.70-499.70 F: c ESI Full ms [ 200.00-600.00] MS 334338a 29 NL: 2.36E6 mlz= 387.70-388.70+ 418.50-419.50 F :-cE S I Full ms2 512.00@ 30.00 [140.00-500.00] MS 334338a_29 Relative Abundance Figure 2 HPLC chromatogram of 1.00 mg/l T-4127 in 50/50 (v/v) methanol/ISO-medium containing internal standard. T-4127 NOTOX Project 334338 T-4127 Y =0.00244219+0.184632*X-0.0371766*XA2 RA2 = 0.9959 W: Equal Area Ratio 0.0 0.2 0.4 0.6 0.8 1.0 mg/l Figure 3 Regression line for solutions in 50/50 (v/v) methanol/ISO-medium containing internal standard: Responses against concentrations. T-4127 Samples November 2002 NOTOX Project 334338 Figure 4 HPLC-chromatogram of a blank (1/5 (v/v) ISO-medium/isopropanol containing internal standard). Figure 5 HPLC-chromatogram of 0.500 mg/l T-4127 in 1/5 (v/v) ISO-medium/isopropanol containing internal standard. T-4127 i ~ n 3a r g g g.m _ D tn x* m o i m t NOTOX Project 334338 Figure 6 Regression line for solutions in 1/5 (v/v) ISO-medium/isopropanol containing internal standard: Responses against concentrations. T-4127 NOTOX Project 334338 Figure 7 HPLC-chromatogram of 0.00991 mg/l T-4127 in 1/5 (v/v) ISO-medium/isopropanol containing internal standard. T-4127 NOTOX Project 334338 Figure 8 HPLC-chromatogram of 0.0991 mg/l T-4127 in 1/5 (v/v) ISO-medium/isopropanol containing internal standard. Figure 9 HPLC-chromatogram of 0.595 mg/l T-4127 in 1/5 (v/v) ISO-medium/isopropanol containing internal standard.