English version. This Technical Specification (CEN/TS) was approved by CEN on 16 December 2004 for provisional application.

Transcription

1 TECHNICAL SPECIFICATION SPÉCIFICATION TECHNIQUE TECHNISCHE SPEZIFIKATION CEN/TS February 2005 ICS English version Materials and articles in contact with foodstuffs - Plastics substances subject to limitation - Part 26: Determination of 1- octene and tetrahydrofuran in food simulants Matériaux et objets en contact avec les denrées alimentaires - Substances dans les matières plastiques soumises à des limitations - Partie 26 : Détermination du 1- octène et du tétrahydrofurane dans les simulants d'aliments Werkstoffe und Gegenstände in Kontakt mit Lebensmitteln - Substanzen in Kunststoffen, die Beschränkungen unterliegen - Teil 26: Bestimmung von 1-Octen und Tetrahydrofuran in Prüflebensmitteln This Technical Specification (CEN/TS) was approved by CEN on 16 December 2004 for provisional application. The period of validity of this CEN/TS is limited initially to three years. After two years the members of CEN will be requested to submit their comments, particularly on the question whether the CEN/TS can be converted into a European Standard. CEN members are required to announce the existence of this CEN/TS in the same way as for an EN and to make the CEN/TS available promptly at national level in an appropriate form. It is permissible to keep conflicting national standards in force (in parallel to the CEN/TS) until the final decision about the possible conversion of the CEN/TS into an EN is reached. CEN members are the national standards bodies of Austria, Belgium, Cyprus, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta, Netherlands, Norway, Poland, Portugal, Slovakia, Slovenia, Spain, Sweden, Switzerland and United Kingdom. EUROPEAN COMMITTEE FOR STANDARDIZATION COMITÉ EUROPÉEN DE NORMALISATION EUROPÄISCHES KOMITEE FÜR NORMUNG Management Centre: rue de Stassart, 36 B-1050 Brussels 2005 CEN All rights of exploitation in any form and by any means reserved worldwide for CEN national Members. Ref. No. CEN/TS :2005: E

2 Contents page Foreword... 3 Introduction Scope Normative references Method A Determination of 1-octene in food simulants Method B Determination of tetrahydrofuran in food simulants Bibliography

3 Foreword This document (CEN/TS :2005) has been prepared by Technical Committee CEN/TC 194 Utensils in contact with food, the secretariat of which is held by BSI. This part of EN has been prepared within the Standards, Measurement and Testing project, MAT1-CT , Development of Methods of Analysis for Monomers and has been prepared by Subcommittee (SC 1) of TC 194 "Utensils in contact with food" as one of a series of test methods for plastics materials and articles in contact with foodstuffs. This standard is intended to support Directives 2002/72/EC [1], 89/109/EEC [2], 82/711/EEC [3] and its amendments 93/8/EEC [4] and 97/48/EC [5], and 85/572/EEC [6]. At the time of preparation and publication of this part of EN the European Union legislation relating to plastics materials and articles intended to come into contact with foodstuffs is incomplete. Further Directives and amendments to existing Directives are expected which could change the legislative requirements which this standard supports. It is therefore strongly recommended that users of this standard refer to the latest relevant published Directive(s) before commencement of a test or tests described in this standard. This part of EN should be read in conjunction with EN Further parts of EN 13130, under the general title Materials and articles in contact with foodstuffs - Plastics substances subject to limitation, have been prepared, and others are in preparation, concerned with the determination of specific migration from plastics materials into foodstuffs and food simulants and the determination of specific monomers and additives in plastics. The parts of EN are as follows. Part 1: Guide to test methods for the specific migration of substances from plastics to foods and food simulants and the determination of substances in plastics and the selection of conditions of exposure to food simulants Part 2: Determination of terephthalic acid in food simulants Part 3: Determination of acrylonitrile in food and food simulants Part 4: Determination of 1,3-butadiene in plastics Part 5: Determination of vinylidene chloride in food simulants Part 6: Determination of vinylidene chloride in plastics Part 7: Determination of monoethylene glycol and diethylene glycol in food simulants Part 8: Determination of isocyanates in plastics Part 9: Determination of acetic acid, vinyl ester in food simulants Part 10: Determination of acrylamide in food simulants Part 11: Determination of 11-aminoundecanoic acid in food simulants Part 12: Determination of 1,3-benzenedimethanamine in food simulants 3

4 Part 13: Determination of 2,2-bis(4-hydroxyphenyl)propane (Bisphenol A) in food simulants Part 14: Determination of 3,3-bis(3-methyl-4-hydroxyphenyl)-2-indoline in food simulants Part 15: Determination of 1,3-butadiene in food simulants Part 16: Determination of caprolactam and caprolactam salt in food simulants Part 17: Determination of carbonyl chloride in plastics Part 18: Determination of 1,2-dihydroxybenzene, 1,3-dihydroxybenzene, 1,4-dihydroxybenzene, 4,4 -dihydroxybenzophenone and 4,4 dihydroxybiphenyl in food simulants Part 19: Determination of dimethylaminoethanol in food simulants Part 20: Determination of epichlorohydrin in plastics Part 21: Determination of ethylenediamine and hexamethylenediamine in food simulants Part 22: Determination of ethylene oxide and propylene oxide in plastics Part 23: Determination of formaldehyde and hexamethylenetetramine in food simulants Part 24: Determination of maleic acid and maleic anhydride in food simulants Part 25: Determination of 4-methyl-pentene in food simulants Part 26: Determination of 1-octene and tetrahydrofuran in food simulants Part 27: Determination of 2,4,6-triamino-1,3,5-triazine in food simulants Part 28: Determination of 1,1,1-trimethylolpropane in food simulants Parts 1 to 8 are European Standards. Parts 9 to 28 are Technical Specifications. WARNING All chemicals are hazardous to health to a greater or lesser extent. It is beyond the scope of this Technical Specification to give instructions for the safe handling of all chemicals, that meet, in full, the legal obligations in all countries in which this Technical Specification may be followed. Therefore, specific warnings are not given and users of this Technical Specification should ensure that they meet all the necessary safety requirements in their own country. According to the CEN/CENELEC Internal Regulations, the national standards organizations of the following countries are bound to announce this CEN Technical Specification: Austria, Belgium, Cyprus, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta, Netherlands, Norway, Poland, Portugal, Slovakia, Slovenia, Spain, Sweden, Switzerland and United Kingdom. 4

5 Introduction 1-octene, C 8 H 16 or CH 2 =CH-(CH 2 ) 5 CH 3, PM/Ref. No 22660, and tetrahydrofuran (THF), C 4 H 8 O, PM/Ref. No 25150, are monomers used in the manufacture of certain plastics materials and articles intended to come into contact with foodstuffs. After manufacture, residual monomer can remain in the polymer and may migrate into foodstuffs coming into contact with that plastics article. NOTE However, the following should be taken into account at carrying out a migration test for 1-octene. From migration experiments carried out for 10 d for 40 C it was recognized that irreproducible loss of 1-octene, 11 % to 69 %, due to volatilization, can arise when using aqueous food simulants. Method A describes the determination of 1-octene in food simulants. Method B describes the determination of tetrahydrofuran in food simulants. The methods have been pre-validated by collaborative trials with two laboratories. 5

6 1 Scope This document, part of EN 13130, specifies analytical procedures for the determination of 1-octene and THF in food simulants water, 3 % w/v aqueous acetic acid, 15 % v/v aqueous ethanol and olive oil. The level of 1-octene and THF monomer determined is expressed as milligrams of monomer per kilogram of food simulant. The methods are appropriate for the quantitative determination of 1-octene in the range of 2 mg/kg to 30 mg/kg in food simulants and of THF in the range of 0,06 mg/kg to 1,5 mg/kg in food simulants. NOTE The method should also be applicable to other aqueous food simulants as well as to the other fatty food simulants e.g. sunflower oil, corn oil or a mixture of synthetic triglycerides. 2 Normative references The following referenced documents are indispensable for the application of this document. For dated references, only the edition cited applies. For undated references, the latest edition of the referenced document (including any amendments) applies. EN :2004, Materials and articles in contact with foodstuffs Plastics substances subject to limitation Part 1: Guide to test methods for the specific migration of substances from plastics to foods and food simulants and the determination of substances in plastics and the selection of conditions of exposure to food simulants. 3 Method A Determination of 1-octene in food simulants 3.1 Principle The level of 1-octene in a food or a food simulant is determined by headspace gas chromatography (HSGC) of the food simulant sample, sealed in headspace glass vials. Headspace gas chromatography is carried out applying automatic injection and flame ionization detection. Quantification is achieved using iso-octane as internal standard with calibration against food simulant samples fortified with known amounts of 1-octene. Confirmation of 1-octene levels is carried out by combined gas chromatography/mass spectrometry (GC/MS). 3.2 Reagents NOTE All reagents should be of recognized analytical quality unless otherwise stated Analytes octene, CH 2 =CH-(CH 2 ) 5 CH 3, purity greater than 97,5 % (GC) Iso-octane, (CH 3 ) 2 -CH-CH 2 -C(CH 3 ) 3, purity greater than 99 % (GC) Chemical N,N-dimethylacetamide (DMAA), CH 3 -CO-N(CH 3 ) 2, purity greater than 99 %. 6

7 3.2.3 Solutions Stock solutions of 1-octene with a defined concentration of approximately 2 mg/ml in DMAA Weigh a 50 ml volumetric flask, including cap, filled with about 45 ml of DMAA (3.2.2), to an accuracy of 0,1 mg. Add approximately 100 mg of 1-octene (about 150 µl), then reweigh to an accuracy of 0,1 mg. Make up to 50 ml with DMAA. Close and shake. Calculate the exact concentration of the stock solution in milligrams of 1-octene per millilitre of solution. Repeat the procedure to provide a second stock solution. NOTE These stock solutions can be stored at + 4 C, with the exclusion of light, for up to 3 months Standard solutions of 1-octene in DMAA with a defined concentration of approximately 200 µg/ml Place 1,0 ml of the 1-octene stock solution ( ) in a 10 ml volumetric flask and make up to the mark with DMAA (3.2.2). Close and mix thoroughly. Calculate the exact concentration of the standard solution in milligrams of 1-octene per millilitre of solution. Repeat the procedure using the second stock solution prepared in to provide a second standard solution. NOTE These standard solutions can be stored at + 4 C and with the exclusion of light for up to 3 months Stock solution of internal standard iso-octane in DMAA at a defined concentration of approximately 2 mg/ml Weigh a 50 ml volumetric flask, including cap, filled with about 45 ml of DMAA (3.2.2), to an accuracy of 0,1 mg. Add approximately 100 mg of iso-octane (about 150 µl), then reweigh to an accuracy of 0,1 mg. Make up to the mark with DMAA close and mix. Calculate the exact concentration of the stock solution in milligrams of iso-octane per millilitre of solution. NOTE This stock solution can be stored at + 4 C, with the exclusion of light, for up to 3 months Internal standard solution of iso-octane in DMAA at a defined concentration of approximately 120 µg/ml Place 3,0 ml of internal standard stock solution ( ) in a 50 ml volumetric flask and make up to the mark with DMAA. Close and mix thoroughly. Calculate the exact concentration of the internal standard solution in milligrams of iso-octane per millilitre of solution. NOTE This internal standard can be stored at + 4 C, with the exclusion of light, for up to 3 months. 3.3 Apparatus NOTE An instrument or item of apparatus is listed only where it is special or made to a particular specification, the usual laboratory glassware and equipment being assumed to be available. 7

8 3.3.1 Gas chromatograph, equipped with a flame ionization detector (FID) and fitted with an automatic headspace sampler Gas chromatographic column, capable of the separation of DMAA from 1-octene and isooctane, such that the peaks of 1-octene and iso-octane do not overlap by more than 1 % peak area with other compounds. NOTE The following are examples of GC columns known to be suitable for 1-octene analysis: a) 50 m x 0,53 mm i.d. fused silica capillary column, coated with a phenyl-methyl silicone phase, film thickness 2,5 µm. For guidance, the operating conditions established for the column described above, were the following: Headspace sampler: Sample thermostatting time: 80 min Sample temperature: 70 C Transfer line temperature: 130 C Needle temperature: 85 C Pressure equilibration time: 3 min Injection time: 6 sec Gas chromatograph: Injector: 150 C Detector: 220 C Oven program: 110 C (11 min), 10 C/min to 150 C (5 min) Carrier gas: Nitrogen at 250 kpa Linear velocity: 35 cm/sec Injection mode: total injection FID gases: to be optimized according to the manufacturer's specification Alternatively, the following system has been found to be suitable: b) Column: 30 m x 0,2 mm i.d. fused silica capillary column, coated with a 5 % diphenyl and 95 % dimethyl silicone phase, film thickness 0,33 µm Headspace sampler: Sample thermostatting time: 60 min Sample temperature: 100 C Gas chromatograph: Carrier gas: Helium at 90 kpa and 1 ml/min Oven program: 75 C (8 min), 25 C/min to 250 C (5 min) Injection mode: split (ratio 1:5) 8

9 3.3.3 Glass sample vials, 20 ml or another size suitable for the particular autosampler employed, with polytetrafluroethylene (PTFE) coated butyl or silicone rubber septa and aluminium crimp-cap closures Volumetric flasks, 50 ml and 10 ml Microsyringes, 250 µl and 100 µl Volumetric pipettes, 1 ml and 3 ml. 3.4 Samples General Laboratory samples of the food simulant to be analyzed shall be obtained as described in EN Samples shall be kept refrigerated at 4 C in closed containers. Analyte-free samples of relevant food simulants of the same type as those to be analyzed shall also be prepared for calibration purposes. Take into account the possible loss of analyte due to volatilization in aqueous food simulants (see NOTE in the Introduction) Test sample preparation Aqueous food simulants Place 1,0 ml of the food simulant obtained from the migration experiment into a sample vial (3.3.3) using a volumetric pipette (3.3.6), close immediately the vial with a septum and cap. Add 200 µl isooctane internal standard solution ( ) followed by 200 µl DMAA (3.2.2) to the food simulant by injection through the septum using the 250 µl microsyringe (3.3.5). Prepare each test sample at least in duplicate Olive oil Weigh 1,0 g ± 0,01 g of the food simulant, as obtained from the migration experiment into a sample vial (3.3.3). Close the vial immediately with a septum and cap. Add 200 µl iso-octane internal standard solution ( ) followed by 200 µl DMAA (3.2.2) to the olive oil test sample by injection through the septum using the 250 µl microsyringe (3.3.5) and mix thoroughly. Prepare each test sample at least in duplicate Blank sample preparation Aqueous food simulants Place 1,0 ml of aqueous 1-octene-free food simulant into a sample vial (3.3.3) using a 1,0 ml volumetric pipette (3.3.6), then cap immediately. Add 200 µl iso-octane internal standard solution ( ) followed by 200 µl of DMAA (3.2.2) through the septum using the 250 µl microsyringe (3.3.5). Prepare each blank sample at least in duplicate. 9

10 Olive oil Weigh 1,0 g ± 0,01 g of 1-octene-free olive oil into a sample vial (3.4.3). Cap immediately. Add 200 µl iso-octane internal standard solution ( ) followed by 200 µl of DMAA (3.2.2) through the septum using the 250 µl microsyringe (3.3.5) and mix thoroughly. Prepare each blank sample at least in duplicate Preparation of calibration samples General Calibration shall be obtained from at least five levels. The concentration range of calibration solutions solution shall span from approximately 2,0 mg to 30 mg 1-octene per kilogram of food simulant Aqueous food simulants Place 1,0 ml of aqueous 1-octene-free food simulant into a series of six sample vials (3.3.3) using a 1,0 ml volumetric pipette, then cap immediately. Add to each vial 200 µl of iso-octane internal standard solution ( ) through the septum using the 250 µl microsyringe (3.4.5) and mix thoroughly. Add through the septum into the vial the volumes of 1-octene standard solution ( ) and DMAA (3.2.2) given in Table 1, using appropriate microsyringes (3.3.5). Table 1 Volumes of 1-octene and DMAA in calibration samples Calibration sample Addition of 1-octene standard solution ( ) in µl Addition of DMAA (3.2.2) in µl Approximate concentration of 1- octene in the calibration sample mg/kg Prepare each calibration sample at least in duplicate. Calculate the exact concentrations of the calibration samples in micrograms of 1-octene added per millilitre of aqueous food simulant, which corresponds to milligrams per kilogram of aqueous food simulant. 10

11 NOTE Commission Directive 2002/72/EC [1] states that the specific gravity of all simulants should conventionally be assumed to be 1. Milligrams of substance released per litre of simulant will thus correspond numerically to milligrams of substance released per kilogram of simulant and, taking into account of the provisions laid down in Directive 82/711//EEC [3], to milligrams of substance released per kilogram of foodstuff. Repeat this procedure using the second standard solution prepared in to provide a second set of calibration samples Olive oil Weigh 1,0 g ± 0,01 g of 1-octene-free olive oil into a series of five sample vials (3.3.3) and cap immediately. Add to each vial 200 µl of iso-octane internal standard solution ( ) through the septum using the 250 µl microsyringe (3.3.5). Add through the septum into the vials the same volumes of standard solution ( ) and DMAA (3.2.2) as given in the table under to obtain the same concentrations as given there. Mix thoroughly. Prepare each calibration sample at least in duplicate. Calculate the exact concentrations of the calibration samples in micrograms of 1-octene added per gram of olive oil, which corresponds to milligrams per kilogram of olive oil food simulant. Repeat this procedure using the second standard solution prepared in to provide a second set of calibration samples. 3.5 Procedure Headspace gas chromatographic analysis (HSGC) General Examine the baseline stability and response linearity of the detector before starting measurements. Maintain the same operating conditions throughout the measurement of all samples and calibration solutions. Inject each solution at least in duplicate, i.e. fill two headspace vials with the same solution and analyze each vial once. NOTE Under the conditions given in (a), the retention times of 1-octene and iso-octane were 4,2 min and 6,3 min, respectively. Approximately 20 min are enough to perform a complete analysis programme Sample treatment Analyze the test samples, blanks and calibration samples prepared in to as they are without any further sample treatment Execution of the determination Equilibrate sample, calibration and blank vials, prepared in accordance with 3.4, in the thermostatted manifold, kept at 70 C, of the automated headspace sampler for 80 min before sampling and commencing the analysis programme (see NOTE to 3.3.2). Identify the 1-octene and iso-octane peaks on the basis of their retention time and measure the respective peak heights or read the computer print-out of the peak areas. Calculate the ratio between the 1-octene peak area and the iso-octane peak area to obtain the peak area ratio (PAR). 11

12 3.5.2 Calibration Measure the calibration samples (3.4.3) as described in Construct or calculate the calibration curve plotting PAR values against the concentration of 1-octene in milligrams per kilogram of food simulant. The calibration curves shall be rectilinear and the correlation coefficient shall be 0,996 or better. The two sets of calibrant solutions made from independently prepared stock solutions shall be crosschecked by generating two calibration curves which on the basis of peak area measurement shall agree to ± 5 % of one another. Inject each solution at least in duplicate Evaluation of data NOTE The following calculations assume that for all measurements exactly the same volume, i.e. 1,0 ml of aqueous food simulant or 1,0 g of olive oil, has been used and that in all cases the same volume of internal standard solution, i.e. 200 µl of internal standard solution, has been added. Following the method described, no interferences have been observed. If a 1-octene-free sample shows an interference in the iso-octane region of the chromatogram exceeding 10 % of the area of iso-octane in the calibration samples (3.4.3), and if the analysis of replicate blank samples reveals that this interference varies by more than ± 20 % in absolute size, external calibration shall be used. If the analysis of the zero point calibration sample (3.4.3) shows a peak in the 1-octene region corresponding to less than 1,5 mg/kg when calculated in accordance with / and the absolute area of duplicates does not vary by more than 10 %, the PAR of the zero point calibration sample shall be subtracted from the ratios of the test sample and the calibration samples and the data plotted as in If the interference corresponds to more than 1,5 mg/kg, the method of standard addition shall be used. 3.6 Expression of results Calculation of analyte level Graphical determination Calculate the average of PAR-values obtained from the test samples in accordance with and read the 1-octene concentration of the test sample from the calibration graph (3.5.2) Calculation from the regression parameters If the regression line equation is: ( a ) + b y[par] = x where y [PAR] a x is the peak area ratio of 1-octene/iso-octane; is the slope of the regression line; is the concentration of 1-octene in the food simulant in milligrams per kilogram; 12

13 b is the intercept of the regression line, then the concentration of 1-octene in the food stimulant is given by: C octene, fs = y b a where C octene,fs is the concentration of 1-octene in the food simulant, in milligrams per kilogram. Both procedures yield directly the 1-octene concentration in the food simulant in milligrams per kilogram. The method applying calculation from the regression parameters is the preferred method Calculation of the specific migration of 1-octene Depending on the fill volume of the test material and on the surface area/food simulant ratio, the concentration in the laboratory sample as determined in accordance with may need mathematical transformation to calculate the specific migration value to be compared with the specific migration limit (SML). For guidance see EN :2004, Clause Precision Validation This method was pre-validated by a collaborative trial with two laboratories. In each laboratory withinlaboratory precision experiments were performed using the four official EC food simulants for establishment of precision data at the restriction criterion as well as migration testing using a linear low density polyethylene (LLDPE) film sample in contact with 15 % v/v aqueous ethanol and olive oil, respectively Repeatability and reproducibility Evaluation of the within-laboratory precision experiment results according to ISO 5725, at a concentration of 15 mg/kg yielded the following performance characteristics at the 95 % probability level (lab 1/lab 2): Repeatability: r = 0,94/0,31 mg 1-octene per kilogram of water; r = 0,85/0,28 mg 1-octene per kilogram of 3 % w/v aqueous acetic acid; r = 1,90/0,42 mg 1-octene per kilogram of 15 % v/v aqueous ethanol; r = 3,56/2,49 mg 1-octene per kilogram of olive oil Detection limit The within-laboratory detection limit (WDL) of 1-octene, based on the calibration curve method according to DIN 32645, was found to be in the range of 0,20 mg/kg to 2,0 mg/kg, depending on the simulant used. Thus the method is capable of quantitative detection at a minimum value of 2,0 mg 1-octene per kilogram. 13

14 3.7 Confirmation Requirement for confirmation If the specific migration of 1-octene into the food simulant, calculated according to the procedure given in EN , from the analyte level calculated according to exceeds a restriction, i.e. a specific migration limit (SML) of 15 mg/kg, the result of the determination shall be confirmed by the method described in The confirmation is qualitative in the sense that it should demonstrate the correct identity of the measured analyte and the absence of interferences. For the purposes of quantification the result as calculated according to shall be taken as the true value Confirmation by combined gas chromatography/mass spectrometry (GC/MS) In the scanning mode, re-analyze the test sample(s) (3.4.1) and a calibration sample with concentration at the specific migration limit in the m/z range 40 to 120. Compare the mass spectra obtained from 1-octene both in the test sample and the calibration sample. The relative abundances in the spectra shall not differ by more than 5 %. In the selected ion monitoring mode, re-analyze the test sample(s) (3.4.1), the calibration sample (3.4.3) and the blank sample. The ions monitored should be m/z 55, 70, 83, 112 for 1-octene and 41, 56, 57, 99 for iso-octane. The peak ratios 55/70 and 83/112 for the 1-octene peaks in the test sample have to be the same as the ratios of the 1-octene peaks in the calibration sample. The peaks attributed to 1-octene and iso-octane shall maximize within one-half peak (measured at half-height) or within 2 % of the absolute retention time, of standards, whichever is the smaller. NOTE Measured ratios at the SML-concentration were found to be 1,10 to 1,05 (for 55/70 m/z ratio) and 1,55 to 1,80 (for 83/112 m/z ratio). 3.8 Test report The test report shall include at least the following, where applicable: a) identification of the sample; b) name of the laboratory; c) name of responsible analyst; d) date of report; e) date of analysis; f) analyte; g) a reference to this method; h) sample details, such as: 14 1) type of food/food simulant/material/article; 2) origin and denotation of the sample; 3) date and method of obtaining the laboratory sample;

15 4) storage conditions; i) results expressed in milligrams of 1-octene per kilogram of food simulant. Results shall be reported as the average value from two or more determinations satisfying the repeatability criterion of ; j) details of confirmation procedure, if any; k) reasons for modifications introduced into the test method, if any. 4 Method B Determination of tetrahydrofuran in food simulants 4.1 Principle The level of THF in a food or a food simulant is determined by headspace gas chromatography (HSGC) of the food simulant sample, sealed in headspace glass vials. Headspace gas chromatography is carried out applying automatic injection and a flame ionization detection. Quantification is achieved using tetrahydropyran (THP) as internal standard with calibration against food simulant samples fortified with known amounts of THF. Confirmation of THF levels is carried out by combined gas chromatography/mass spectrometry (GC/MS). 4.2 Reagents NOTE All reagents should be of recognized analytical quality unless otherwise stated Analytes Tetrahydrofuran (THF), C 4 H 8 O, purity greater than 99,5 % (GC) Tetrahydropyran (THP), C 5 H 10 O, purity greater than 99 % (GC) Chemicals N,N-dimethylacetamide (DMAA), CH 3 -CO-N(CH 3 ) 2, purity greater than 99 % Distilled water Solutions NOTE The stock and standard solutions prepared as described in to can be stored + 4 C, with the exclusion of light, for not longer than one week Stock solutions of THF in water with a defined concentration of approximately 2,2 mg/ml, for aqueous food simulants Weigh a 100 ml volumetric flask, including cap, filled with about 99 ml of water ( ), to an accuracy of 0,1 mg. Add approximately 220 mg of THF ( , about 250 µl), then re-weigh to an accuracy of 0,1 mg. Make up to the mark with water. Close and shake. Calculate the exact concentration of the stock solution in milligrams of THF per millilitre of solution. Repeat the procedure to provide a second stock solution. 15

16 Standard solutions of THF in water with a defined concentration of approximately 22 µg/ml, for aqueous food simulants Place 1,0 ml of the THF stock solution ( ) in a 100 ml volumetric flask containing already about 90 ml water and make up to the mark with water ( ). Close and mix thoroughly. Calculate the exact concentration of the standard solution in milligrams of THF per millilitre of solution. Repeat the procedure using the second stock solution prepared in to provide a second standard solution Stock solution of internal standard THP in water at a defined concentration of approximately 1 mg/ml, for aqueous food simulants Weigh a 100 ml volumetric flask, including cap, filled with about 99 ml of water ( ), to an accuracy of 0,1 mg. Add approximately 100 mg of THP ( , about 150 µl), then reweigh to an accuracy of 0,1 mg. Make up to the mark with water, close and mix. Calculate the exact concentration of the stock solution in milligrams of THP per millilitre of solution Internal standard solution of THP in water at a defined concentration of approximately 70 µg/ml Place 7,0 ml of internal standard stock solution ( ) in a 100ml volumetric flask containing already about 90 ml water and make up to the mark with water. Close and mix thoroughly. Calculate the exact concentration of the internal standard solution in milligrams of THP per millilitre of solution Stock solutions of THF in DMAA with a defined concentration of approximately 1,0 mg/ml for olive oil Weigh a 50 ml volumetric flask, including cap, containing about 45 ml of DMAA ( ), to an accuracy of 0,1 mg. Add approximately 50 mg of THF ( , about 60 µl), then reweigh to an accuracy of 0,1 mg. Make up to 50 ml with DMAA. Close and shake. Calculate the exact concentration of the stock solution in milligrams of THF per millilitre of solution. Repeat the procedure to provide a second stock solution Standard solutions of THF in DMAA with a defined concentration of approximately 50 µg/ml, for olive oil Place 1,0 ml of the THF stock solution ( ) in a 20 ml volumetric flask and make up to the mark with DMAA ( ). Close and mix thoroughly. Calculate the exact concentration of the standard solution in milligrams of THF per millilitre of solution. Repeat the procedure using the second stock solution prepared in to provide a second standard solution. 16

17 Stock solution of internal standard THP in DMAA at a defined concentration of approximately 1 mg/ml, for olive oil Weigh a 50 ml volumetric flask, including cap, filled with about 45 ml of DMAA ( ), to an accuracy of 0,1 mg. Add approximately 50 mg of THP ( , about 60 µl), then reweigh to an accuracy of 0,1 mg. Make up to the mark with DMAA, close and mix. Calculate the exact concentration of the stock solution in milligrams of THP per millilitre of solution Internal standard solution of THP in DMAA at a defined concentration of approximately 75 µg/ml Place 1,5 ml of internal standard stock solution ( ) in a 20 ml volumetric flask and make up to the mark with DMAA. Close and mix thoroughly. Calculate the exact concentration of the internal standard solution in milligrams of THP per millilitre of solution. 4.3 Apparatus NOTE An instrument or item of apparatus is listed only where it is special or made to a particular specification, the usual laboratory glassware and equipment being assumed to be available Gas chromatograph, equipped with a flame ionization detector (FID) and fitted with an automatic headspace sampler Gas chromatographic column, capable of the separation of DMAA from THF and THP, such that the peaks of THF and THP do not overlap by more than 1 % peak area with other compounds. NOTE The following are examples of GC columns known to be suitable for THF analysis: a) 60 m x 0,53 mm i.d. fused silica capillary column, coated with a phenyl-methyl silicone phase, film thickness 5,0 µm. For guidance, the operating conditions established for the column described above, were the following: Headspace sampler: Sample thermostatting time: 60 min Sample temperature: 70 C for aqueous food simulants 90 C for olive oil Transfer line temperature: 110 C Needle temperature: 90 C Pressure equilibration time: 3 min Injection time: 9 sec It is important to check for the different sample temperatures with regard to the respective food simulant used. The higher temperature for olive oil is necessary to achieve the required sensitivity. Gas chromatograph: Injector: 150 C Detector: 250 C Oven program: 60 C (17min) Carrier gas: Nitrogen at 200 kpa Linear velocity: 83 cm/sec Injection mode: total injection FID gases: to be optimized according to the manufacturer's specification 17

18 Alternatively, the following system was found to be suitable: b) Two fused silica capillary columns connected in series with a glass connector, each column: 30 m x 0,53 mm i.d. coated with a methyl silicone phase, film thickness 5,0 µm Gas chromatographic conditions: Oven program: for aqueous simulants for olive oil Carrier gas: 60 C (12 min), 5 C/min to 70 C (3,5 min) 60 C (12 min), 5 C/min to 70 C (2 min), 20 C/min to 150 C (10 min) Helium at 6 ml/min Glass sample vials, 20 ml or another size suitable for the particular autosampler employed, with polytetrafluoroethylene (PTFE)-coated butyl or silicone rubber septa and aluminium crimp-cap closures. NOTE It has been observed that THF could be employed for the manufacture of PTFE-coated butyl rubber septa. This leads to severe and irreproducible interferences during THF analysis. Therefore, if necessary, it is recommended to purge the septa by boiling them in distilled water for at least 15 min followed by drying in an oven just before use. Unused purged septa should be purged again before next use ml and 100 ml volumetric flasks Microsyringes, 25, 50, 100, 250 and 500 µl Volumetric pipettes, 1,0 ml, 1,5 ml, 2,5 ml, 5,0 ml, 6,0 ml, 7,0 ml and 10,0 ml. 4.4 Samples Test sample preparation General Laboratory samples of the food simulant to be analyzed shall be obtained as described in EN Samples shall be kept refrigerated at 4 C in closed containers. Analyte-free samples of relevant food simulants of the same type as those to be analyzed shall also be prepared for calibration purposes. Take into account the possible loss of analyte due to volatilization in aqueous food simulants (see NOTE in the Introduction) Aqueous food simulants Place 10,0 ml of the food simulant obtained from the migration experiment (see EN ) into a sample vial (4.3.3) using a 10,0 ml volumetric pipette (4.3.6), close immediately the vial with a septum and cap. Add 100 µl THP internal standard solution ( ) to the food simulant by injection through the septum using the 100 µl microsyringe (4.3.5). Prepare each test sample at least in duplicate. 18

19 Olive oil Weigh 10,0 g ± 0.01 g of the food simulant, as obtained from the migration experiment (see EN ) into a sample vial (4.3.3). Close immediately the vial with a septum and cap. Add 100 µl THP internal standard solution ( ) to the olive oil test sample by injection through the septum followed by injection of 300 µl DMAA ( ) using the 100 µl and 250 µl microsyringe (4.3.5), respectively, and mix thoroughly. Prepare each test sample at least in duplicate Blank sample preparation Aqueous food simulants Place 10,0 ml of aqueous THF-free food simulant into a sample vial (4.3.3) using a 10,0 ml volumetric pipette (4.3.6), then cap immediately. Add 100 µl of THP internal standard solution ( ) through the septum using the 100 µl microsyringe (4.3.5). Prepare each blank sample at least in duplicate Olive oil Weigh 10,0 g ± 0,010 g of THF-free olive oil into a sample vial (4.3.3). Cap immediately. Add 100 µl of THP internal standard solution ( ) and 300 µl of DMAA ( ) through the septum using the appropriate microsyringe (4.3.5) and mix thoroughly. Prepare each blank sample at least in duplicate Preparation of calibration samples General Calibration shall be obtained from at least five levels. The concentration range of calibration solutions solution shall span from approximately 0,09 to 1,5 mg THF per kilogram food simulant Aqueous food simulants Place 45 ml of aqueous THF-free food simulant into a series of five 50 ml volumetric flasks. Add into the flasks the volumes of THF standard solution ( ) as given in Table 2 and make then up to the mark with the food simulant. Calibration sample Table 2 Volumes of THF in calibration samples Addition of THF standard solution (3.3.2) ml 1 0,2 0,09 2 0,5 0,22 3 1,5 0,66 4 2,5 1,10 5 3,5 1,54 Approximate concentration of THF in the calibration sample µg/ml 19

20 Transfer from each solution 10,0 ml into a vial (4.3.3) using a 10,0 ml volumetric pipette (4.3.6), then cap immediately. Add into each vial 100 µl of THP internal standard solution ( ) through the septum using the 100 µl microsyringe (4.3.5) and mix thoroughly. Prepare each calibration sample at least in duplicate. Calculate the exact concentrations of the calibration samples in micrograms of THF per 1 ml aqueous food simulant corresponding to mg/kg food simulant. Repeat this procedure using the second standard solution prepared in to provide a second set of calibration samples Olive oil Weigh 10,0 g ± 0,010 g of THF-free olive oil into a series of five sample vials (4.3.3) and cap immediately. Add into each vial 100 µl of THP internal standard solution ( ) through the septum using the 250 µl microsyringe (4.3.5). Add through the septum into the vials the volumes of standard solution ( ) and DMAA ( ) given in Table 3 to obtain the approximate concentrations given there. Mix thoroughly. Prepare each calibration sample at least in duplicate. Table 3 Volumes of THF and DMAA in calibration samples No. of calibration sample Addition of THF standard solution (3.3.6) µl Addition of DMAA (3.2.1) µl Approximate concentration of THF in the calibration sample µg/g , , , , ,50 Calculate the exact concentrations of the calibration samples in micrograms of THF added per 1 g olive oil corresponding to milligrams per kilogram olive oil food simulant. Repeat this procedure using the second standard solution prepared in to provide a second set of calibration samples. 4.5 Procedure Headspace gas chromatographic analysis (HSGC) General Examine the baseline stability and response linearity of the detector before starting measurements. Maintain the same operating conditions throughout the measurement of all samples and calibration solutions. Inject each solution at least in duplicate, i.e. fill two headspace vials with the same solution and analyze each vial once. 20

21 NOTE Under the conditions given in (a), the retention time of THF and THP were 8,3 min and 13,8 min, respectively. About 20 min are enough to perform a complete analysis programme Sample treatment Analyze the test samples, blanks and calibration samples prepared in to as they are without any further sample treatment Execution of the determination Equilibrate sample, calibration and blank vials, prepared in accordance with 4.4, in the thermostatted manifold, kept at 70 C or 90 C (see 4.3.2), of the automated headspace sampler for 60 min before sampling and commencing the analysis programme (operating conditions, for instance, see 4.3.2). Identify the THF and THP peaks on the basis of their retention time and measure the respective peak heights or read the computer print-out of the peak areas. Calculate the ratio between the THF peak area and the THP peak area to obtain the peak area ratio (PAR) Calibration Measure calibration samples (4.4.3) in accordance with Construct or calculate the calibration curve plotting PAR values against the concentration of THF in milligrams per kilogram of food simulant. NOTE Commission Directive 2002/72/EC [1] states that the specific gravity of all simulants should conventionally be assumed to be 1. Milligrams of substance released per litre of simulant will thus correspond numerically to milligrams of substance released per kilogram of simulant and, taking into account of the provisions laid down in Directive 82/711//EEC [3], to milligrams of substance released per kilogram of foodstuff. The calibration curves shall be rectilinear and the correlation coefficient shall be 0,996 or better. The two sets of calibrant solutions made from independently prepared stock solutions shall be crosschecked by generating two calibration curves which on the basis of peak area measurement shall agree to within ± 5 % of one another Evaluation of data NOTE The following calculations assume that for all measurements exactly the same volume, i.e. 10,0 ml of aqueous food simulant or 10,0 g of olive oil, has been used and that in all cases the same volume of internal standard solution, i.e. 100 µl of internal standard solution, and , respectively, has been added GC interferences Following the method described, no interference have been observed. (See, however, NOTE under 4.3.3). If a THF-free simulant sample shows an interference in the THP region of the chromatogram exceeding 10 % of the area of THP in the calibration samples (4.4.3), and if the analysis of replicate blank samples reveals that this interference varies by more than ± 20 % in absolute size, external calibration shall be used. 21

22 If the analysis of the zero point calibration sample (4.4.3) shows a peak in the THF region corresponding to less than 0,07 mg/kg when calculated according to and the absolute area of duplicates does not vary by more than 10 %, the PAR of the zero point calibration sample shall be subtracted from the ratios of the test sample and the calibration samples and the data plotted as described in If the interference corresponds to more than 0,07 mg/kg, the method of standard addition shall be used. 4.6 Expression of results Calculation of analyte level Graphical determination Calculate the average of PAR-values obtained from the test samples in accordance with and read the THF concentration of the test sample from the calibration graph (4.5.2) Calculation from the regression parameters If the regression line equation is: ( a ) + b y[par] = x where y[par] a x b is the peak area ratio of THF/THP; is the slope of the regression line; is the concentration of THF in the food simulant in milligrams per kilogram; is the intercept of the regression line, then the concentration of THF in the food simulant is given by: C THF, fss = y b a where C THF,fs is the concentration of THF in the food simulant in milligrams per kilogram. Both procedures yield directly the THF concentration in the food simulant in milligrams per kilogram. The method applying calculation from the regression parameters the preferred method Calculation of the specific migration of THF Depending on the fill volume of the test material and on the surface area/food simulant ratio, the concentration in the laboratory sample as determined in accordance with may need mathematical transformation to calculate the specific migration value to be compared with the specific migration limit (SML). For guidance see EN :1999, Clause

23 4.6.3 Precision Validation This method was pre-validated by a collaborative trial with two laboratories. In each laboratory a within-laboratory precision experiment using the four official EC food simulants for establishment of precision data at the restriction criterion as well as migration testing using a THF-containing (as a comonomer) polymer sample being in contact with 15 % v/v aqueous ethanol and olive oil, respectively Repeatability Laboratory samples of the food simulant to be analyzed shall be obtained as described in EN at a concentration of 0,6 mg/kg yielded the following performance characteristics at the 95 % probability level (lab 1/lab 2): Repeatability: r = 0,06/0,02 mg THF per kilogram of water; r = 0,09/0,02 mg THF per kilogram of 3 % w/v aqueous acetic acid); r = 0,14/0,04 mg THF per kilogram of 15% v/v aqueous ethanol; r = 0,06/0,03 mg THF per kilogram of olive oil Detection limit The within-laboratory detection limit (WDL) of THF, based on the calibration curve method according to DIN 32645, was found to be in the range of 0,04 mg/kg to 0,07 mg/kg, depending on the simulant used. Thus the method is capable of quantitative detection at a minimum value of 0,07 mg THF per kilogram. 4.7 Confirmation Requirement for confirmation If the specific migration of THF into the food simulant, calculated according to the procedure given in EN from the analyte level calculated according to exceeds a restriction, i.e. a specific migration limit of 0,6 mg/kg, the result of the determination shall be confirmed by the method described in The confirmation is qualitative in the sense that it should demonstrate the correct identity of the measured analyte and the absence of interferences. For the purposes of quantification the result as calculated according to shall be taken as the true value Confirmation by combined gas chromatography/mass spectrometry (GC/MS) In the scanning mode, re-analyze the test sample(s) (5.1) and a calibration sample with a concentration at the specific migration limit (4.4.3) in the m/z range 15 to 87. Compare the mass spectra obtained from THF both in the test sample and the calibration sample. The relative abundances in the spectra shall not differ by more than 5 %. In the selected ion monitoring mode, re-analyze the test sample (s) (4.4.1), the calibration sample (4.4.3) and the blank sample. The ions monitored should be m/z 41, 42 and 72 for THF and 41, 45 and 56 for THP. The peak ratios 42/41 and 41/72 of the THF peaks in the test sample have to be the same as the ratios of the THF peaks in the calibration sample. The peaks attributed to THF and THP shall maximize within one-half peak with (measured at half-height) or within 2 % of the absolute retention time, of standards, whichever is the smaller. 23

24 NOTE Measured ratios at the SML-concentration were found to be 2,0 to 2,2 (for 42/41 m/z ratio) and 1,85 to 2,0 (for 41/72 m/z ratio). 4.8 Test report The test report shall include at least the following, where applicable: a) identification of the sample; b) name of the laboratory; c) name of responsible analyst; d) date of report; e) date of analysis; f) analyte; g) a reference to this method; h) sample details, such as: 1) type of food/food simulant/material/article; 2) origin and denotation of the sample; 3) date and method of obtaining the laboratory sample; 4) storage conditions; i) results expressed in milligrams of THF per kilogram of food simulant. Results shall be reported as the average value from two or more determinations satisfying the repeatability criterion of ; j) details of confirmation procedure, if any; k) reasons for modifications introduced into the test method, if any. 24

25 Bibliography [1] Commission of the European Communities, Commission Directive 2002/72/EC of 6 August 2002 relating to plastics materials and articles intended to come into contact with foodstuffs, Official Journal of the European Communities, 15 August 2002, no. L220, p18. [2] Commission of the European Communities, Council Directive of 21 December 1988 on the approximation of the laws of the Member States relating to materials and articles intended to come into contact with foodstuff (89/109/EEC), Official Journal of the European Communities, 11 February 1989, no. L 40, p 38. [3] Commission of the European Communities, Council Directive of 18 October 1982 laying down the basic rules necessary for testing migration of the constituents of plastics materials and articles intended to come into contact with foodstuffs (82/711/EEC), Official Journal of the European Communities, 23 October 1982, no. L 297, p 26. [4] Commission of the European Communities, Commission Directive of 15 March 1993 amending Council Directive 82/711/EEC laying down the basic rules necessary for testing migration of the constituents of plastics materials and articles intended to come into contact with foodstuffs (93/8/EEC), Official Journal of the European Communities, 14 April 1993, no. L 90, p 22. [5] Commission of the European Communities, Commission Directive of 97/48/EC of 29 July 1997 amending Council Directive 82/711/EEC laying down the basic rules necessary for testing migration of the constituents of plastics materials and articles intended to come into contact with foodstuffs, Official Journal of the European Communities, 12 August 1997, no. L 222, p 10. [6] Commission of the European Communities, Council Directive of 19 December 1985 laying down the list of simulants to be used for testing migration of constituents of plastics materials and articles intended to come into contact with foodstuffs (85/572/EEC), Official Journal of the European Communities, 31 December 1985, no. L372, p14. [7] ISO 5725, Accuracy (trueness and precision) of measurement methods and result. [8] DIN 32645, Chemical analysis; decision limit; detection limit and determination limit; estimation in case of repeatability; terms, methods, evaluation. 25