The European Union Reference Laboratory for Polycyclic Aromatic Hydrocarbons (EU-RL PAHs), operated by the Institute for Reference Materials and Methods (IRMM) of the Joint Research Centre (JRC), organised a method validation study by inter-laboratory comparison (ILC-MVS) for evaluating the effectiveness of a method based for the determination of the four marker PAHs (benz[a]anthracene, benzo[a]pyrene, benzo[b]fluoranthene, and chrysene in different food commodities within the scope of Commission Regulation (EC) No 1881/2006. The test sample is homogenised, a test portion is mixed with desiccant, sand and the labelled internal standard mixture. It is then extracted with n-hexane or, alternatively, with cyclohexane, by pressurised liquid extraction (PLE). Soxhlet extraction has proven to give equivalent results compared to PLE, provided that a sufficient number of extraction cycles are performed (at least 7 hours of extraction, with about 6 cycles/h). Co-extracted water is separated from the organic phase of the extract; then the organic extract is evaporated to small volume, filtered and purified by size exclusion chromatography (SEC), using a mixture of ethyl acetate and cyclohexane as eluent. The extraction step is skipped for edible oils. For this matrix a portion of the sample is diluted with a mixture of ethyl acetate and cyclohexane, then the labelled internal standard mixture is added and the sample is directly processed by SEC. After SEC, 200 μL of toluene are added as a keeper to the collected SEC fraction, which is then evaporated to about 200 μL, and further cleaned up by solid phase extraction on silica, using cyclohexane as eluent. The cleaned-up sample extract is evaporated again to 200 μL. Finally an injection standard is added to the sample prior to measurement by GC-MS. The final extract is preferably injected into a programmable temperature vaporizer inlet. However split/splitless injection may be applied alternatively. The chromatographic separation is achieved on a specific capillary column which allows the separation of the four target PAHs from the other EU priority PAHs and from triphenylene. The analytes are ionised by electron ionization (EI) at 70 eV. The target PAHs are recorded in Single Ion Monitoring mode, and quantified by comparison with the labelled analogues. A total of 10 materials (corresponding to 20 expected results) including edible oil, meat, smoked fish, bivalve molluscs, cereals and infant formula, containing the 15+1 EU priority PAHs (inclusive of the four marker ones) at different levels, comparable to maximum levels laid down in legislation, were sent to 18 laboratories from 9 EU Member States, and a laboratory in Switzerland. Recovery was calculated by participants from the labelled standards. All samples were sent as blinded duplicates but fish and meat which were sent as open duplicates. 12 laboratories reported results, of which 1 laboratory did not apply the SOP and was therefore excluded from the evaluation. Relative standard deviations for reproducibility (RSDR) ranged from 7% for BaP in oil to 55% for BaP in mussels (BaP content was 0.9 μg/kg). The values for RSDr values ranged from 2% for BaP and BbF in oil to 17% for BbF in wheat flour and CHR in infant formula. Robust mean values for recoveries ranged from 60% for CHR in fish to 86% for BaP in mussels' tissue. The Commission Regulation (EC) No 333/2007 lays down performance criteria that must be met by a method to determine BaP in food. These criteria were extended also to BaA, BbF and CHR and have been met by this method for all materials but for RSDR for BaP in mussels.
