Document gbyb4EKLmDrKdVOnb94ZQa5VN

TNO Quality of Life TNO Report Final I V7054 Migration study of.111.11111M toowait-nattiurwQtonschappolijk ondecekf Natherlands Organisation Car Applied Scientific Research Analytical Research Utrechtseweg 48 P.O.Box 360 3700 AJ Zeist The Netherlands www.tno.nl P +31 30 694 41 44 F +31 30 694 48 94 @tno.n1 Date 6 October, 2006 Authors Sponsor TNO project number TNO study code Sponsor's study code Status Previous versions Number of pages Number of tables Number of figures Number of annexes Number of appendices Solvay Solexis 031,10279/01 V7054 Final 26 4 10 1 All rights reserved. No part of this publication may be reproduced and/or published by print, photoprint, microfilm or any other means without the previous written consent of TNO. En case this report was drafted on instructions, the rights and obligations of contracting parties are subject to either the Standard Conditions for Research Instructions given to TNO, or the relevant agreement concluded between the contracting parties. Submitting the report for inspection to parties who have a direct interest is permitted. e 2006 TNO TNO Report I V7054 I Final 2126 6 October, 2006 Summary At the request of Solvay Solexis a study was performed to determine the residual content of in a f!uoro olymer sample. is a chlorofluoropolyether acid used as an emulsifier/dispersing agent during the polymerisation process of fluoropolymers. ical method for the determination of the residual content of - in a fluoropolymer consisted ofa hot extraction with methanol/water (80/20; v/v) and analyses of the extract with the use of liquid chromatography in combination with mass spectrometry. In the blank (extraction solvent without the use of sample material) a background signal of approximately <l 0% of the lowest calibration standard was measured. This low background signanfluence on the final results because the detected concentrations o were below the LOD. Recoveries measured were adequate viz. between 93 and 103% at the concentration level of42.7 ng/rnl extract which corresponds with 79 g per kg fluoropolymer. d was suitable for the determination of the residual content of ...-i in fluoropolymer at the concentu extract which corresponds with approximately 20-370 g -- per kg fluoropolymer. The repeatability and linearity were satisfactory, The overall migration of the fluoropolymer in 95% ethanol and iso-octane under migration conditions of respectively 6 hours at 60C and 4 hours at 60C was < lg/ 6 dm2, Further characterization of the migration substances with NMR did not show any significant reaction product. TNO Report I V7054 I Final 6 October, 2006 3126 Contents Summary-2 1 Introduction-4 2 Procedure-S 2.1 Sample materials- 5 2.2 Determination of the residual content- 5 2,3 Calibration curve - 6 2.4 Recovery experiment by standard addition to the test sample -6 2.5 Overall migration -7 2.5.1 95% ethanol -7 2.5.2 !so-octane-7 2.6 Characterization of the migrating substances by NMR- 7 3 Results-8 3.1 Actual content of in the test sample-8 3.2 Detection limit -8 3.3 Recovery-8 3.4 Repeatability and linearity- 9 3.5 Overall migration -9 3.6 Characterization ofmigrating substances with NMR-9 4 Conclusions -10 5 6 Tables-12 7 Figures- 15 8 Appendix- 26 TNO Report I V7054 I Final 4 /26 6 October, 2006 1 Introduction Solexis has re uested to detennine the residual content of CAS no. 329238-24-6) in a fluoropolymer, is used as an emulsifier/dispersing agent during the polymerisation process of fluoropolymers. These fluoropolymers are processed to produce a range of articles for repeat use like parts for food processiug equipment, gs on cooking utensils. After the manufacture, can migrate into foodstuffs coming into contact with that product. With the described analytical method the residual content of rcehsliodruoafllucoornotpeonltyectahnerbaeciudssedintothecaflicnuallatfelutohreopwoolyrsmtecrapsreodmuicgtraitsiodnetoenfn ined. The - from fluoropolyrners into foodstuffs. Additional the overall migration from the fluoropolymer sample into 95% ethanol and iso-octane was detennined. The migration substances were characterized by NMR. TNO Report I V7054 I Final 6 October, 2006 5 /26 2 Procedure The study consists of a hot extraction with 0/20; v/v) of a fluoropolymer sample produced with the use of--- The extraction rmed for 8 hours. The extraction solvent is analysed for containing ... ..... by the use ofliquid chromatography in combination with mass spectrometry. The extraction procedure was repeated with a fresh amount of methanol/water. This second extract was also tested for containing residues o. The maximum amount of used in the polymerization recipe is A recovery study is performed by adding to the sample material. Repeatability ofthe analysis is tested by performing the analysis in triplicate. The overall migration of fluoropolymer was determined by the incubation of the fluoropolymer into 95% ethanol and iso-octane. After incubation de solvent was evaporated and the amount ofresidue was determined gravimetrically. The residue was characterized by NMR. The analysis ofthe residual content was performed in May 2006. TI1e overall migration and characterization ofthe migrating substances was performed in August-September 2006. 2.1 Sample materials Solvay Solexis, Bollate (Mi) Italy, provided sample material for the testing. The samples were provided with a unique TNO sample code. The samples were received in 6 and stored at room temperature. The reference substance - was received in april 2006. TNO san1 le ID 0939 -02-0202 0939-02-0951 remark Used for reference Sheets approximately 13.5xl3.5x0.04cm propane, I,I,1,2, 2,3,3-heptafluoro-3-[(trifluoroethenyl)oxy]-, polymer with tetrafluoroethene and trifluoro(trifluoromethoxy)ethene. 2,2 Determination of the residual content The sheets of fluoi'opolymer were cut into pieces ofea. l x3 cm. A test-sample of approximately 27 g i s transferred to a conical flask of 100 ml. To the sample 50 ml ofthe extraction solvent methanol/water (80/20; v/v) was added. The hot extraction was performed for 8 hours. After the extraction an aliquot of the extract was transferred to an HPLC vial. The remaining extraction solvent was decanted and the extraction was repeated with a fresh amount of 50 ml methanol/water (80/20;v/v). TNO Report I V7054 I Final 6 October, 2006 6 /26 From the collected first and second extracts 50 I was injected into the HPLC and analysed for containing as described in Appendix 1. The extraction was repeated by using methanol/water (80 ammonia. By addition of ammonia the less volatile ammonium salt was formed and so the possibility ofevaporation of the .... during the extraction procedure was reduced. Blank values were obtained by treatment of50 ml ofextraction solvent in the same way as described above (without adding fluoropolymer). All experiments were carried out in triplicate. Typical chromatograms obtained for the determination of in methanol/water (blank), in methanol/water containing 0.1 % anunonia and in the first extract ofa fluoropolymer samples are presented in Figures I and 2. 2.3 Calibration curve A stock solution of , containing 0.534 g/ml, was prepared by dissolving 26.70 mg into 50 ml methanol/water (80/20;v/v). The stock solution was diluted by transferring I ml to a volumetric flaks of20 ml and addition of methanol/water (80/e of 20 ml. The diluted stock solution contained 26.7 ng/ml of-- From the diluted stock solution 0, 10, 20, 40, 80, 120, 160 and 200 I were transferred to a set ofvolumetric flask of25 ml. The volumetric flasks were filled up to the mark with methanol/water (80/20;v/v) and mixed thoroughly. The solutions thus obtained contain approximately 0, 10.68, 21.36, 42.76, 85.44, 128.16, 170.88, 213.60 ng of per ml. The calibration standards were analysed by LC-MS as described in Appendix I ofthis report. A typical LC-MS chromatogram obtained for the calibration standard of I 0.68 ng/ml in methanol/water (80/20;v/v) is presented in Figure 3. 1s obtained, the peak heights of the specific MRM ions of -- were measured. From these data, the calibration curve was constructed and the correlation coefficient was calculated. A graphical plot ofthe calibration curve and the raw data and statistical parameters are presented in Figures 5 and 6. 2.4 Recovery experiment by standard addition to the test sample The sheets of fluoropolymer were cut into pieces of ea. lx3 cm. A test sample of approximately 27 g was transferred to a conical flask of I 00 ml. To the test sample 50 ml of the extraction solvent methanol/water (80/20;v/v) and 80 l of diluted stock solution of were added. In he sample extract contains (0.08(ml)*26.70(ng/ml)*1000/50(ml)=) 42.72 ng --- per ml. This corresponds with 42.72(ng/ml)*50(ml)/27(g oftest-sample) = 79 ng per g fluoropolymer (=g/kg). The sample was treated as described in item 2.2. TNO Report I V7054 I Final 6 October, 2006 7126 The recovery experiment was repeated by using methanol/water (80/20) containing 0.1% ammonia. Typical chromatograms obtained for a standard addition of to the test sample is presented in Figure 4a without the addition of ammonia, and Figure 4b with the addition of ammonia to the extraction solvent. 2.5 Overall migration 2,5.1 95% ethanol A test sample o fluoropolymer (2 sheets with the dimension 13x13.4 cm one-sided, corresponding with contact area of 3.484 dm2) was immersed in 100 ml 95% ethanol. After incubation period 6 hours at 60C the samples were removed and after evaporating of the solvents under controlled conditions, the residues were estimated gravimetrically, The test was performed in duplicate. From the results the overall migration in mg/6 dm2 was determined. 2.5.2 !so-octane A test sample of fluoropolymer (2 sheets with the dimension 13x13.4 cm one-sided, corresponding with contact area of 3.484 dm2) was immersed in 100 ml iso-octane. After incubation period of 4 hours at 60C the samples were removed and after evaporating offthe solvents under controlled conditions, the residues are estimated gravimetrically. The test was performed in duplicate. From the results the overall migration in mg/6 dm2 was determined. 2.6 Characterization of the migrating substances by NMR The residue obtained under item 2.5 was characterized by NMR. Prior to NMR spectroscopy, the residue was dissolved in I ml deuterated chloroform (CDC13) containing 0.05 % tetramethylsilane (TMS). An aliquot of 600 1 of the solution was transferred to 5 mm NMR tubes. ID NMR spectra were recorded on a Bruker AVANCE 600 MHz spectrometer operating at a resonance frequency of 600.13 MHz a probe temperature of 27C. ID 1H-NMR experiments were acquired using the pulse sequence 'zg'. Free induction decays (F!Ds) were collected into 64K data points using a spectral width of 12376 Hz. 45 Degree pulses were used with an acquisition time of 2.65 s and a relaxation delay of 5.0 s. The spectra were acquired by accumulation of 256 F!Ds. Spectra were recorded using a spectral width of 8000 Hz. Chemical shifts of all spectra are expressed in ppm by reference to TMS (0,00 ppm). Resolution enhancement and noise reduction of the spectra was performed using an exponential window function with a line broadening of 0.3 Hz and a manual baseline correction was applied, TNO Report I V7054 I Final 8/26 6 October, 2006 3 Results 3.1 Actual content of in the test sample The results obtained for the actual content of offlented in Table l. The .-i consists ofdifferent oligomers there own s ecific MRM (Multiple-Reaction-Monitorin )ions. in the test sample From the results it was concluded that although traces of were detected in the first extracts ofthe fluoropolymer sample, all concentrations were lower than the LOD. The responses monitored in the second extracts of the sample did not significant differ from the blank (extract without fluoropolymer). In the blank a background signal of approximately < 10% of the lowest calibration standard was measured. This low background signal has no significant influence on the final results because the detected concentrations o were below the LOD. Furthermore there was no significant difference between the samples extracted with and without the addition of ammonia. Therefore it was concluded that no was evaporated by using methanol/water as the extraction solvent. 3.2 Detection limit The concentration of the lowest calibration solution, 10,68 ng/ml, was set as the detection limit. A detection limit ofI0.68 ng/ml corresponds to a residual content of20 ng per kg fluoropolymer. The sum of peak heights ofall detected MRM ions were used for the construction ofthe calibration curve. 3.3 Recovery The results obtained for the detennination of the recovery of from fluoropolymer is presented in Table 2. TNO Report I V7054 I Final 6.October, 2006 9 126 T for the responses observed in the samples without the addition 0 The results obtained for the recovery samples extracted with and without ammonia did not show significant differences. Recoveries measured were adequate viz. between 93 and 103% at the concentration level of 42.7 ng/ml extract which corresponds with (42.7*50/27) 79 ng/g (=79 flg/kg) of fluoropolymer sample. 3.4 Repeatability and linearity Repeatability of the method was demonshe standard deviation of the recovery of the standard addition of to the test sample. The measured standard deviation was 7% for the extraction with methanol/water (80/20;v/v) and 8% for the extraction with methanol/water (80/Z0;v/v) containing 0.1% of ammonia. The calibration curve appeared to be linear in the range of 10.68 to 213,6 ng/ml for the sum of peak heights plotted against the concentration (Fig 6). The correlation coefficient was 0.9995. 3.5 Overall migration Results obtained for the overall migration are presented in Table 3. From the results it is concluded that although the duplicate analyses differ significant - due to the low amount measured - the total amount of migration substances in 95% ethanol and iso octane are < l mg/6 dm2. 3.6 Characterization of migrating substances with NMR 1H -NMR spectra of the 95% ethanol migration residue obtained from (fluoropolymer) sample and a blank sample are shown in Figures 7 and 8, respectively. The spectra of the iso-octane migration residue obtained for the sample of fluoropolymer and a blank sample are shown in Figures 9 and 10, respectively. Tentative 1H chemical shifts were assigned by means of ACD/HNMR prediction software (ACDlabs, Toronto) and are listed in Table 4. Signals present in both, the fluoropolymer extracts and in their blanks originate from compound introduced during the extraction procedure and the NMR sample preparation; i.e. residual solvent (7.26 ppm), acetic acid (2.10 ppm_water (1.56 ppm), TMS (0.00 ppm). Due to the absence of protons in the fluoropolymer and monomers there off are not visible for 1H-NMR. The signals present in the extract samples and absent in their blanks are assigned as proton containing migrants. Due to the low intensities en pure resolution of the signals further assignment was not possible. Final conclusions is that no significant peaks are detected due to reaction products of the fluoropolymer. TNO Report I V7054 I Final 6 October, 2006 10 /26 4 Conclusions The detection limit for the determination of in samples of 0 ,g/kg. From this result the worst case migration (M) of ....._ to foodstuff can be calculated. Therefore the following formula has to be applied: M=(Q*A*Ld*D)/1000 Assuming that: Q= cone in mg/kg in fluoropolymer= 0.020 mg/kg, A= area of the food contact material in dm2, conventionally set at 600 cm2, Ld= thickness of food contact material = 0.04 cm and D=fluoropolymer density is 2 g/cm3 The worst case migration into foodstuff is M=(0.020*600*0.04*2)/1000=0.001 mg/kg. It was found that the residual content of in the fluoropolymer was <20 ,g/kg corresponding with a worst case migration of < 1 ,g per kg foodstuff. The results obtained from the recovery experiments of demonstrate that the method is accurate and repeatable for the determination of chlorofluoropolyether acids. The mean recoveries at the concentration level of 42.7 ng/ml extract were 997% for the extraction with methanol/water and 1018% for the extraction with methanol/water containing 0.1 % ammonia. The overall migration of the fluoropolymer in 95% ethanol and iso-octane under migration conditions of respectively 6 hours at 60C and 4 hours at 60C was < l,g/ 6 dm2 Further characterization of the migration substances with NMR did not show any significant reaction product. TNO Report I V7054 I Final 6 October, 2006 5 Signatures 1 1 /26 , . Analytical Research Department TNO Report I V7054 I Final 6 October, 2006 12 /26 6 Tables Table 1 Residual content of TNO ID of test Cone, in first sample' MeOH/H,O extrad (ng/ml) 0939-02-0951 A < 1 0 0939-02-0951 B < 1 0 0939-02-0951 C < 1 0 in the fluoro ol er test sample Cone in second Cone. in g per MeOHIH20 kg of (ng/ml) fluoropolymer test sam le2 < 10 < 20 <10 < 20 <10 < 20 0939-02-0951 D < 1 0 < 10 < 20 0939-02-0951 E < 1 0 < 10 < 20 0939-02-0951 F < 1 0 < 10 < 20 1 LOD = 1 0 ng/ml; concentrations in second extracts were <LOD. 2 Residual content is calculated by using the following formula: Concentration of - in extract (ng/ml)*50(ml)/27(gram of test sample) ng/g (=g/kg) 3 Samples A,B,C are extracted with methanol/water (80/20;v/v) and the samples D,E,F are extracted with methanol/water (80/20;v/v) containing 0.1% ammonia. Table 2 Recove TNO ID sample' of Cone. added (ng/ml) 0939-02-095I A 42.7 0939-02-0951 B 42.7 0939-02-095I C 42.7 93 45.6 1 07 41.3 97 Avera e 99 SD 7 0939-02-0951 D 42.7 0939-02-0951 E 42.7 0939-02-0951 F 42.7 39.2 92 45.8 1 07 44.2 104 Avera e 1 01 SD 8 1 Recoveries are corrected for the 'blank sample'=sample without addition of --; LOD = 1 0 ng/ml; results of second extractions: <l O ng/ml - Samples A,B,C are extracted with methanol/water (80/20;v/v) and the samples D,E,F are extracted with methanol/water (80/20;v/v) containing 0.1 % ammonia. TNO Report I V7054 I Final 6 October, 2006 1 3 /26 Table 3 Overall migration Sample Simulant Sample 0939/02/0951 95% ethanol Sample 0939/02/095 1 !so-octane Time/temp 6 hours at 60C mean 4 hours at 60C mean Overall migration in ml!I 6 dm2 0.27 0.68 0.48 0.45 0.95 0.7 TNO Report I V7054 I Final 6 October, 2006 1 4 126 Table 4 1H NMR chemical shift. of compounds present in A) 95% ethanol extract of the fluoropolymer, B) 95% ethanol extract of a blank , C) iso-octane extract of fluoropolymer and D) iso-octane extract of a blank samples, recorded in CDC], at 27 C. Chemical Shift (ppm) Compound 8.1 I 7.43 7.26 7.09 7.3 -6.9 6.6 -6.4 5.3 - 5.2 4.69 4.4 - 4.2 4.2 - 4. 1 4.1 - 4.0 3.96 3.8 - 3.3 3.67 3.49 2.86 2.58 2.5 - 2.2 2.10 2.0 - 1 .5 1 .5 - 1 .3 1.3 - 1 .2 0.9 0.8 0. 10 0.00 -0. IO n.d.** satellite of chloroform chloroform satellite of chloroform miITTant mio-rant mie:rant miITTant mierant miITTant mim-ant + contaminant n.d. n.d. n.d n.d contaminant in iso-octanol contaminant in iso-octanol miITTand + other comnound acetic acid water n.d. n.d. n.d. satellite ofTMS internal reference TMS satellite ofTMS Functional Group n.d. CH CH CH n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. CH- , Alkvl RCH,CH1 Alkvl RCH,CH, Albl, CH, CH3 CH3 CH, Presence in samolc A B C D X - X X X X X X X X X X X X X X - - X X - X X - - - X X - - X X - - X X - X - X - X X X X X X - -X -X X X X X X X X X - - X X --X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X In ppm relative to the signal of internal TMS at 0.00 ppm n.d., not determined. Functional groups are not detemrined due to multiple possibilities TNO Report I V7054 I Final 6 October, 2006 7 Figures Figure l a Representative LC-MS chromate (methanol/water=80/20; v/v 1 5 /26 bl opw A bfa61ll6$9 Sm (Mn. 2x3) 100 ;.{ 0 - 0.00 2.00 4.00 Ofo6\8689 Sm (ln, 2x3) , ' 0.00 2.00 4.00 Ofa5 186!19 Sm (Mn. 2x3) '.'J"' ; o,l...,:; 0.00 2.0o 4.00 bfaS 18689 Sm (.In. 2,:;3) 'i0.00 2.00 4.00 b!a618689 Sm (Mn. 2x3) ".t100-', 0.00 200 4,00 6.21 75 6.00 8.00 VG62B 10.00 10.70 150 12.00 14.00 -- 16.00 08-Jun-2006 22:27:07 MRM of 6 CnaMels ES 958.9 > 201 1 . 69,;13 Are;1 18.00 '"20'.oo MRM ol 6 Ctiannels ES792.9 > 201 2.QJpJ ' A<eo -,../4,. 6.00 8.00 10.00 1 2 .00 14.00 16.00 T.- -r--.......- 18.00 20.00 MRM of 6 Channels ES- 9.28 434 ,49 626.9 > 201 4.34e3 f73 Ama - 6.00 8.00 10.00 - 12.00 14,00 16.00 18.00 20.00 MRI.I of 6 CMnnels F.S 937 532 9 ,. 201 129 2, 18e3 Area 6.00 8.00 6.35 1041 10.00 ...' ,,..,..,... 12.00 14.00 16.00 1 -,,......._,.., 18.00 20.00 I\IRM of 6 Channels ES- 460 9 > 20 1 6.37e3 Areo 6.00 8.00 10.00 12.00 14.00 16.00 18.00 2,0,.00 TNO Report I 1/7054 I Final 6 October, 2006 1 6 /26 Figure lb Representative LC-MS chromato am ofblank extraction solven1 bi opw 8 NH4 VG628 09-Jun-2006 07:15:20 bfa618714 Sm (Mn. 2x3) t,IRM of 6 Chanricls ES- 1 '.:'] 9SB.9 ,. ;m _._,'-"-A------JV,,.,..,,,AA,J..,. .._-,.,..._,...__J 82S: "'L "' 0.00 .2. :oo-,T.r oo--,T.oo---,.o-o --,,r.oro--r,,,.,Torr14'.oo'T-,T.r .O- O - 18.00 20:00 bla.618714 Sm {Mn, 2x3) MRM ofG Channels (S. 100 1067 121 792.9 > 20? 1 . 59e:1 * ,-...,..___J,...__,.,L._______,._AJ-..Jv.\...-...N,.,.,,,_,. Arca ,1..._T, -T,--,------~----T 0.00 2.00 4.00 6.00 8.00 - 10.00 -------- 12.00 14,00 ------- , 16.00 18.00 20.00 bfa6187l4 Sm (Mrl. 2x3) MRM of 6 Channels ES- 626.9 > 201 6.3&3 Arca 20.00 MRI.I of 6 Cbanm!1$ ES 313 3.17c3 9 Acea 'i .;;:::;;:;;:;:;;:::;;;:;;;:;::'::;:;:'.:::;:;::::;;::::;:';"".:,:::;;::::::::;:::::;:;:;:::;:;:'.:;:;::::;::::;:::::;=;;:;:;:::;:;;1:5:3;;2.:.;_: -.,. 201 ,?00-.. o.oo 2.00 4.oo s.oo a.oo 10.00 12.ooOO 1a'.oo -20\X> bfa618714 Sm (Mn. 2x3) MRM ol 6 Channels ES 1001 8 30 460,9 > 201 9 56e3 _1 :.\i-'>'T=,r"'-="l"="r'=i==r"""i=s;,.,..,C'-'=="r~===,p=-f- ="r="r'='i'"""'"'-1 Area 0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00 bf.361871,i Sm {Mn, 2)(3) MRI.I of 6 Channels ES- ,oo- 29-1.9 201 :l -::: 0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 16.00 20.00 TNO Report I V7054 I Final 6 October, 2006 17 /26 Figure 2a Representative LC-MS chromatogram of a first extract of a fluoropolymer -----------_,__,____J.L-._. _______ 0939,02/0951 J bfa618593 Sm (Mn. ::I) 100] VG628 08.Jun-2006 23:51:27 tlRM or$ Cnanne1$ ES958.9 > 201 a --'1 :; o+.-..,.,T-----------;---- t -~T ------ 0.QQ 2.0o 4.oo s.oo a.oo 10.00 12.00 14.oo 1s.oo 1a.oo 20.00 ol.i6!8693 Sm (Mn, 2.3) MRM of 6 Channels ES 10 68 792.\1 > 20: ':JlL=====, ~==========='"L ==="-'===========l;\;:; "" 0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00 ofa.51B693 Sm (Mn. 2x3) MRM of 6 Char111els ES- 1n9 626.9 > 201 ':L.-===-======","-.=====,====~-.==---= '- 0.00 2.00 4.00 6.00 8.00 10.00 bfa518693 Sm (Mn. 2x31 910.3505 :L========= ""================"' 0.00 2.00 4.00 6.00 bfa518693 Sm (Mn. 2x3) ;L ,_,, ~""=-. - 0 ======_,.....,...<i'+-r 0.00 2.00 4.00 6.00 tlla61 663 Sm \Mn. 2x3) == 6.00 10.00 169 4 1 8.00 10.00 12.00 14.00 16.00 12.00 14.00 16.00 '!' ===== 11.53 12.00 14.00 16.00 18.00 20.00 MRM of 6 Channels ES 532.9 > 201 6,86e3 Arc.1 18.00 20.00 MRM oi 6 Channels ES 460,9 >- 201 1 . 1A0rec.lJ a 16.00 20.00 MRM of 6 Channels ES TNO Report I V7054 I Final 6 October, 2006 1 8 /26 Figure 26 Representative LC-MS chromatogram ofa first extract of a fluoropolymer sample. l 0939/02/0951 D VG628 09.Jun-2006 02:19:01 llfa618700 Sm (Mn. 2x3) l\1RM o! 6 Cti.:innels E;S- 1 9.84 59 958.!:I 201 1.50e3 --L Areo r--r~,..,,-.--------- "r-- . ... , 0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 llt.i<ilt:1700 Sm (Mn. 7x'.l) 16.00 18-. 00 .2..0.-,.00 2.00 4.00 b1.:i518700 Sm {Mn. 2)(3} ""l _9.30 8.00 10,00 12.00 "'L 9.29 6.00 8.00 10.00 12.00 8.29 1507- 14.00 16.00 18.00 20.00 MRM of 6 CMnnels ES 626.9 ,. 201 1 04e-l Acea 16.00 1 8.00 20.00 MRM of 6 Channels ES- 460,9 > ?.01 9 39e3 Area bf;'l618700 Sm (Mn, 2x3) MRM of 6 Channels ES 29J.9 > 201 I 03e3 'l: :::: T::' 0,00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00 TNO Report I V7054 I Final 6 October, 2006 Figure 3 Representative LC-MS chromato (LOD). 1 9 /26 VG628 08-Jun-2006 19:17:23 , ___,,__'-------_,___,,.,.__ t /, , Mf{Mof6Channels ES, 9589 >201 01e, ,.... .,,.i Are;i 10.71 MAM of 6 ChaMeis ES 792.9 > 201 'i====-=========='''''.n'-, _ --======,......' %::: 0.00 2.00 4,00 ofa618600 Sm (Mn. 2x:3) 100 4 - - o4 0,00 2.00 4.00 bfi'!618680 Sm (Mn, 23} 'i . - , ,& , ofa6186e0 Sm (Mn. 2x3l 6.00 8.00 10.00 9,31 4795 6,00 8.00 10,00 9.31 3039 1 a 30 12.00 12.00 14.00 14.00 16.00 18.00 20.00 t,lR/\1 of 6 Cn<mn;.:1$ ES 626.9 ,. 201 3 14!,7< 16.00 16.00 MRM or 6 Chant1el1 ES 532_9 "' 201 2.24e<l Area 1 1 - MR.M of 6 Channel$ ES- 460.9 > 201 ':1 . ''"l ,r .';: : o.oo 2.00 4.o s.Oo s.bo m'.oo 12'.oo 14'.oo 16'.oo ,a'.oo 20.00 bl.1618680 Sm /Mn, 2d) MRM of fi Channels ES 100 294.9 > 201 l--- u_,_;,': 0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16,00 18.00 20,00 TNO Report I V70M I Final 6 October, 2006 20 /26 Figure 4a Represenative LC-MS chromatogram of the first extract after standard addition of 42.7 ng/ml to 0939/02/0951 A bfa618500 Sm (Mn, 2x3) VG62B ':1] ""l r- ' 0.00 2.00 4.0o Ofa61 8690 Sr:i {Mn. 2;,:3) 100] -l ' ' ' ' ' oo.loo 2.0o 4.00 bfafilf:!600 Sm (Mn, 2x3) ':oi]J.------'-=--=-' -".,'l,,,:,,,--' --------,,,'._.,-,- 0,00 2.00 4.00 , ,..l bfo0"\8690Srn{Mn. 2x3) 10.00 10.68 s.bo a.00 '1"07.00 8992898 1 6.00 8.00 10.00 9.30 e.Oo 8,00 8-29 10,00 12.00 ..1..4....0-0+- . 16.00 20,00 MRM or fi Cnannels ES- 79;:'..9 > 20 l :.13e4 Area 12.00 14.00 16.00 1 8.00 20.00 MRM of n Channels ES 626 IP 201 5 94-l . Aien 12.00 14.00 16.00 1a'.oo 20'.oo MRhl oi 6 Ct1anneIs ES- 532.9 > 201 3.4De4 Area 12.00 14'.oo 16:oo 20.00 MRM nf G Ch:mncls. ES . ' ' , iJ-- ~--4 .00~--6. 00 -'- 8.0 "0"'-1- 0.0- 0 -12 .00-- 14'.0- 0 1- 6,0- 0 ---r'_'_ 9B\ TNO Report I V7054 I Final 6_October, 2006 21 /26 Figure 4b Represenative LC-MS chromatogram of the first extract after standard addition of 42.7 ng/ml to the sample of fluoropolymer. Exlr'action solvent methanol/water=80/20 v/v containin 0.1 % of ammonia. 0939/02/0951 D spiked VG628 09-Jun-2006 03:22:36 ta518703 Sm (Mn, 2x3} MRM of 6 Ch.:lni1els ES 100 951;! 9 " 201 . .J v...-,._.,.__._.,..,_ A:; o+----- r-----""-- 0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16,00 ofa61 tr/03 Sm (Mn, 2x3) 1066 1s:oo 20.00 MRM ol 6 Channels c;s. 792.9 >201 Area ,L""'-.,.------,.--...--,--".,'._,l,a;,-.--,_,. ,.____ ' , o.oo 2,00 4.oo a.o a.00 10.00 12.00 14.00 16.00 _,'_..,........,227e4 18.00 20.00 bfa518703 Sm {Mn. 2x:Jj MRM of 6 Channels ES- " :0 930 626.9 :, 201 6.19,;4 " 1 Are,;1 l ' "oo----,.o-o --,.o-,---,-.o-o---,-.o- o '=,,.:,o.-o,...,._1_2r:o- o -14.- 00 --16-.00--- 18.0-0 -20'.00 Ofa6187D Sm {Mn, 2x3) MRM of 6 Channels ES 532.9 :> 201 ioi 3.42e4 Area , l g+_O= Q =-,.-0---..00--,.0-0--,-.0-IJ -,-e,o"'.oO=-,,-.o--14.00 16.-001- 8.00-20.00 bfai31!l703 Sm (Mn, 23) MRM of O Channels ES 8.30 460.9 ::- 201 :1 1141'1 ',': 7.45c-1 Arca j ( 1 , - ,_ bfc1618703 Sm (Mn, 2:',) MRM of 6 Chnnnels ES 1 2 4.9 > 201 :I! \..A,_,-,Vo,N<'w\JJ.ivJv,J,v,"h..wW 9 A; 0.00 2.00 4.00 6.00 8.00 TNO Report I V7054 I Final 6 October, 2006 22 /26 tilii.1e raw data and statistical parameters of the calibration curve of- (ng/ml) (area) analvte 0,00 1 783,00 1 ,783E+03 10,68 9079,00 9,079E+03 21 ,36 1 7965,00 1 ,797E+04 42,76 30303 3,030E+04 85,44 57995 5,800E+04 1 28,16 86417 8,642E+04 1 70,88 1 09608 1 ,096E+05 213,6 1 42900 1 ,429E+05 0,00 4600 4,600E+03 10,68 8535 8,535E+03 21 ,36 1 5325 1 ,533E+04 42,76 29834 2,983E+04 85,44 58566 5,857E+04 128,16 86496 8,650E+04 1 70,88 1 1 3441 1 , 1 34E+05 213,6 1 45209 1 .452E+05 Statistical evaluation Calculated statistical carameters Intercept of uocer confidence bound (Yu""' Intercept of lower confidence bound CY1ow1 Standard error of estimate Y Est IS.,\ Standard error ofthe crocedure S,o Slope a Standard errror ofslope a Intercept ofXi (Y, = 0) Intercept b (Xi = 0l Standard error of interceot b Correlation coefficient r Response lowest calibration value Resoonse hiahest calibration value Decrees of freedom (di = n-2\ tvalue 99%, one-tailed: 0.01-,n Within laboratorv detection limit rwou Linear Regression fix) = a * x + b 7202, 1 2 8 8 -2487,2 1 1 9 1 727,1798 2,6403 654,1558 5,7991 3,6038 2357,4585 651 ,4300 0,9995 1 783,00000 145209,00000 14 2,62449 7,406 TNO Report I V7054 I Final 6 October, 2006 Figure 6 Typical calibration curve of 23 /26 ! " \i)OEc..,; ! o,o,:rn..;,.,. l.ll!IEc(M u.OOEi ,_ - - - - -- - - - - - ------- - - - --- --- 50 100 - , 200 Concentraton of {nghn!) ---1 TNO Report I V7054 I Final 6 October, 2006 24 126 Figure 7 600 MHz 1H-NMR spectrum of the migrating substances of the fluoropolymer into 95% ethanol, desolved in CDC], containing 0.05% TMS. I 9 8 6 5 4 3 2 0 ppm Figure 8 600 MHz 1H-NMR spectrum of the blank 95% ethanol after migration, desolved in CDC13 containing 0.05% TMS. 9 8 .I I l .J 6 5 4 3 2 a ppm TNO Report I V7054 I Final 6 October, 2006 25 /26 Figure 9 600 MHz 1H-NMR spectrum of the mib,rating substances from the fluoropolymer into Isa-octane, desolved in CDC13 containing 0.05% TMS. ' ' I, 'I 9 7 6 5 4 2 0 ppm Figure 10 600 MHz 1H-NMR spectrum of the blank sample iso-ocatane desolved in CDC13 containing 0.05% TMS. ' 9 ' I" I ' .. I 6 5 3 2 0 ppm TNO Report I V7054 I Final 6 October, 2006 26 126 8 Appendix BO I Detem1ination of chlorofluoropolyether acids in fluoropolymer TNO Report I V7054 BO! I Final 1/7 6 October, 2006 Determination of chlorofluoropolyether acids in f1uoropoIymer 1 Introduction is a chlorofluoropolyether acid used as an emulsifier/dispersing agent during the polymerisation process of fluoropolymers. These fluoropolymers are processed to produce a range of articles for repeat use like parts for food processing equipment, tubes, tapes, coatings on cooking utensils. After the manufacture, chlorofluoropolyether acids can migrate into foodstuffs coming into contact with that product. The analytical method described the determination of the residual content of chlorofluoropolyether acids in the final fluoropolymer product. 2 Scope This method describes the determination of chlorofluoropolyether acids in methanol/water extracts obter samples produced by the use of chlorofluoropolyether acids. The method is appropriate for the quantitative determination of chlorofluoropolyether acids in approximate analyte concentration range of IO to 200 ng/ml methanol/water. 3 Principle The fluoropolymer sample is extracted with methanol/water (80/20;v/v) containing 0.1% ammonia. The extract is analysed for the content of chlorofluoropolyether acids by LC-MS. Quantification is done by means of an external standard calibration. 4 Reagents Reagents and solvents shall be of analytical quality unless otherwise stated. 4.1 Ana/ytes 4. 1. 1 Chemical description: chloro-fluoropolyethers acid ~90% CAS Name: 1-Propene, l,1 ,2,3,3,3-hexafluoro-, telomers with chlorotrifluoroethene oxidized reduced, hydrolized Trade Name: Molecular Weight: CAS No.:329238-24-6 70Structural Formula: 4.2 Chemicals 4. 2 . 1 Water deionized, purified with Bamstead NanoPure Infinity Purifier), HPLC quality water 4.2.2 Ammonium acetate (Merck art 11 16) TNO Report I V7054 B01 I Final 2/7 6 October, 2006 4.2.3 4.2.4 4.2.5 4.2.6 4.3 4.3. 1 4.3.2 4.3.3 4.3.4 4.3,5 Methanol, HPLC grade (Rathburn art RH1 0 1 9) Ammonia 25% (Merck art 31.05.04) Ethanol, for spectroscopy (Merck art 1.000980) Acetic acid (glacial) 1 00% (Merck art 1 .00063) Solutions Standard stock solution of (80/20; v/v) (approximate 0.6 mg/ml), m methanol/water the nearest 0. 1 mg, approximately9mg of - (4. 1 . 1 ) in a 50 ml volumetric flask and dissolve in methanol/water (80/20; v/v). Fill the volumetric flask up to the mark with methanol/water. Close and mix thoroughly. Calculate the actual concentration in mg Repeat the procedure lo obtain a second standard stock solution. NOTE: The two primary standard solutions of analyte shall be checked against one another. The response factor, i.e. detector response divided by concentration of analyte solution, of the two primary standard solutions shall not differ more than 5%. If there is agreement within 5 %, subsequent diluted standard solutions are made from only one of the primary standard solutions. If the levels of the two independently prepared stock solutions do not correspond to within 5% then both stock solutions shall be discarded, and new solutions shall be prepared, Diluted stock solution Transfer to a volumetric flaks of 20 ml, 1 ml of the stock solution of 0.6 mg/ml (4,3.1 ) and add methanol/water (80/20) to a final volume of 20 ml. The diluted stock solution contains 30 ng/ml of 0.1% Ammonia in methanol Mix 4 ml ammonia 25% (4.2.4) with 996 ml methanol (4.2.3). Extraction solvent (0.1% ammonia in methanol:water ; 8:2) Mix 800 ml 0.1 % ammonia in methanol (4.3.3) with 200 ml water (4.2.1). Extraction solvent is used with and without the addition of ammonia. Ammonium acetate solution (! mol/1) Dissolve 77,08 g ammonium acetate (4.2.3) in water (4.2.2) and adjust to a volume of 1000 ml with water. TNO Report I V7054 BOI I Final 3/7 6 Oclober, 2006 4.3.6 Eluent A: 5mM Ammonium acetate solution Mix 5 ml of 1 M ammonium acetate solution (4.3.3) with 995 ml water (4.2.2). 4.3.7 Eluent B: l0mM Ammonium acetate in methanol Mix 1 0 ml of l M ammonium acetate solution (4.3.4) with 990 ml methanol (4.2.3). 5 Apparatus NOTE:An instrument or item of apparatus is listed only where it is special or made to a particular specification, usual laboratory glassware and equipment being assumed to be available. LC-MS system whose main components include a High Performance Liquid Chromatograph preferably with an automatic injector or 100 l injection loop, and a mass spectrometer (MS) detector operated in negative electrospray ionisation mode (ES!-). HPLC column, capable of producak of - and capable to separate from interfering peaks originating from sample matrix or extracting solvent. The following chromatographic conditions have been found to be suitable: NOTE: Depending on the type of equipment used for the determination, the appropriate operating conditions are to be established. Liquid Chromatography The LC-equipment operates under control of the MS data system using LC method files. HPLC chromatograph: Waters Alliance 2690 separations module. Use eluent A (4.3.6) and B (4.3.7). Column: Aqua 5 Cl8 125A column; dimensions 150 x 3 mm Linear gradient: Time (min) 0 2 2.10 1 3.00 1 3.10 14.00 Eluent A (%) 40 40 10 10 40 40 Eluent B (%) 60 60 90 90 60 60 Flow rate: 0,3 ml/min Injection volume: 50 l Autosampler temperature: 20C TNO Report I V7054 BO I I Final 4/7 6 October, 2006 Column temperature: 30C Runtime: 20 min Mass Spectrometry The mass spectrometer is controlled using the MassLynx 4.0 software MS: Waters Ultima triple quadrupole MS in the negative electrospray ionisation mode. MS conditions Ionisation: Interface: Scanmode: m/z values detected: Start time: End time: Collision energy (eV) 0.5 min 20 20-42 6 Sample preparation 6. 1 Test samplepreparation Cut the sample into pieces of approximately lx3 cm. Weight into a conical flask approximately 27 g of sample material. Add 50 ml of the extraction solvent methanol/water (80/20;v/v). Reflux the sample for 8 hours and transfer an aliquoate of the extract to an HPLC vial. Discard the remaining extraction solvent and repeat the extraction with a fresh amount of 50 ml methanol/water (80/20;v/v). Inject 50 ,1 ofthe first and second extract into the HPLC and analyse the extracts for containing Fluorolink 7800:7850. 6.2 Blank sample preparation Treat extraction solvent in the same way as described in 6.1. 6.3 Recove1y experiments TNO Report I V7054 BO I I Final 5/7 6 October, 2006. Add to t g of the sample material 80 I of standard solution (30 g/ml see item 6.4). This corresponds with an addition of 85 g per kg sample material. Treat the spiked sample as described in section 6.1. The final extract thus obtained contains approximately 48 ng of per ml extract. 6.4 Calibration sample preparation Transfer, into a series ofeight 25 ml volumetric flasks 0, 10, 20, 40, 80, 120, 160 and 200 I ofthe diluted standard stock solution (4,3.2). Fill the Transfer to a volumetric flaks of 20 ml, I ml of the stock solution of 0.6 mg/ml (4.3.1) and add methanol/water (80/20) to a final volume of20 ml. The diluted stock solution contains 30 g/ml of up to the mark with methanol/water (80/20;v/v) and mix thoroughly. The solutions thus obtained contain approximately 0, 12, 24, 48, 96, 144, 192 and 240 ng of per ml. 7 Procedure 7.1 LC-MS analysis NOTE: When starting measurements, baseline stability and response linearity of the detector should be examined, together with verification of the detection limit. The same operating conditions of the LC-MS system shall be maintained through out the analysis of all test samples and solutions set out in section 6. Each test sample shall be analysed at least in triplicate. 7.2 Sample treatment Extracts of the samples, recovery samples, blanks, as well as calibration samples prepared in section 6 are analysed by LC-MS. Identify the peak on the basis of the retention time and the specific MRM ions measure the peak height. 7.2. l Calibration solutions Inject each of the calibration samples (6.4) into the HPLC column. Measure the peak height of the six MRM ions of in the chromatogram obum of these peak heights against the concentration of__. in the calibration samples in ng per ml methanol/water as calculated in section 6.4. NOTES: The calibration curve should be linear. The correlation coefficient should be 0,996 or better, If correlation requirement is not met, new standard solutions shall be prepared from the original stock solutions, Analysis of the standard solutions and construction of the calibration graph shall be repeated. The two sets of calibrant solutions made from independently prepared stock solutions should be cross-checked and should agree to 5 % of one another on the basis of peak ratio measurement. TNO Report I V7054 BO I I Final 6/7 6 October, 2006 7.2.2 Test samples and blank solutions Inject SO l of the solutions obtained according to section 6.1 and 6.2 into the LC-MS system, using the conditions given in 5.16. 7.3 Quantification 7.3,1 Calculating of concentration in the test samples Graphical determination: Using the peakd from the test samples according to 7.2.2, read the---, concentration of the methanol/water (80/20;v/v) extract from the calibration graph (7.2.1) in ng/ml. Calculation from the regression parameters: NOTE: An EXCEL computer programme was developed which preferably should used to construct a calibration curve, and to calculate the correlation coefficient and other statistical data (according DIN 32645). Use the measured peak area as obtained in 7.2.2 in the following formula. If the regression line equation is y [log (peak area)] = a x [g/ml] + b, then the CFLU,extr is CFLU,tr = (y-b)/a concentration m the methanol/water Both procedures yield directly the the methanol/water solution in ng/ml. concentration in NOTE: The method applying calculation from the regression parameters should be the preffered one. 7.3.2 Calculation of the specific Calculate the residual content of fluoropolymer sample as follows: RC (ng/g) = C,w,oxo(ng/ml) * V / G in which: CrLU.extr = concentration of solution as obtained in 7.3.1 V = volume ofisopropanol solution (ml) G = gram oftest sample in methanol/water TNO Report I V7054 BOI I Final 717 6 October, 2006 8 Validation 8. 1 Recovery and repeatability Recovery and repeatability (SD of the recovery; n=3) of the method were for the standard addition to the sample material at the concentration level of 42.7 ng/ml: 997% for the extraction with methanol/water (80/20;v/v) and 1018% for the extraction with methanol/water containing 0.1% of ammonia. 8.2 Detection limit The detection limit, based on the calibration curve was the lowest calibration point i.e. 1 0.68 ng/ml. 9 Test report The test report shall contain, as a minimum, the following: an identification name ofthe laboratory name ofthe person responsible for the analysis date ofthe report date ofthe analysis chemical name of the analyte a reference to the method of analysis the performance characteristics of the method sample details, such as: type of food/food simulant/material/article date ofreceipt of the sample, its origin and its denotation * date of preparation of the laboratory sample conditions of storage of sample and laboratorium sample results expressed in /lg/kg fluoropolymer. results should be reported as the average value from two or more determinations satisfying the repeatability criterion in section 8.3. reasons for modification ofthe method of analysis, if applied.