Document 6Bd3K00gM6nJVNgY97GD7yoo3

AR226-3337 Migration QtAgEQ frQjn^IFE It was decided to develop a new method of extraction of APFO from PTFE samples (extraction during 8 h at reflux with 9 5 % ethanol at pH - 10 on finely ground material) as well as a new method of detection of APFO using capillary gas chromatography. This method allows the detection down to 1 pg, which corresponds to a detection limit of 10 ppb in the polymeric sample. The methods of extraction and titration are described in detail in attachments I and (I. It was then decided to determine the residual APFO in five samples produced by the four manufacturers of the common interest group. The samples were chosen in such a way that the worst case situation Is covered (maximum Initial content of APFO , softest post treatment in terms o f temperature, time, washing etc., ensuring the potentially highest residue). The five samples were tested by three different companies to make sure that the developed methods are sufficiently robust. The results are summarised in the attached table I. The table show9 cleariy that the highest residual amount found was 30 ppb. Using the convention of 6 dm2 in contact with 1 kg of food and considering a maximum wall thickness of 6 mm, the amount of PTFE in contact with 1 kg of food becomes 60 cm x 10 cm x 0,6 cm x 2.16 g/cm3 = 777,6 g The residue of APFO in this material is 30 ppb or 3 0 , 1 0 '9 x 777,6 = 23328 10 ' 9 g or 23328 10 " 6 mg or 0,023328 mg Assuming a 100 % migration, which is probably a large overestimation for a 6 mm thick sample, we see that the migration would be around 0 ,0 2 5 mg/kg food or 25 ppb. In other words, the maximum potential migration in food simulant is 0,025 mg/kg, which is twice below the limit set for a reduced technical dossier. IahlaJ Samples 1 2 3 4 5 APFO residues in ppb as determined by company A 15-30 23 29 31 31 B 10-20 10-19 11-12 16-20 9-14 C < 10 <10 <10 <10 < 10