Validation of an automated screening method for persistent organic contaminants in fats and oils by GC×GC-ToFMS

• Published in: Food chemistry Vol. 211; pp. 645 - 653
• Main Authors: López, Patricia, Tienstra, Marc, Lommen, Arjen, Mol, Hans G.J.
• Format: Journal Article
• Language: English
• Published: England Elsevier Ltd 15-11-2016
• Subjects: Animal Feed - analysis; Animals; Fat; Flame Retardants - analysis; Food Contamination - analysis; Gas Chromatography-Mass Spectrometry - methods; GC × GC-ToFMS; Limit of Detection; Plant Oils - analysis; Polybrominated Biphenyls - analysis; Polycyclic Aromatic Hydrocarbons - analysis; Qualitative analysis; Reproducibility of Results; Screening; Validation; Validation; Fat; Screening; Qualitative analysis; GC×GC-ToFMS
• ISSN: 0308-8146; 1873-7072
• DOI: 10.1016/j.foodchem.2016.05.041

•Straightforward fully-automated screening method for POPs in fatty feed sources was developed.•Automation considerably reduced the time needed for data treatment.•Screening detection limits lower than maximum regulatory limits flagged suspicious samples.•Data were acquired in full-scan mode, which added a retrospective feature if necessary.
An screening method, comprised of straightforward sample treatment based on silica clean-up, GC×GC-ToFMS detection and automated data processing with the non-proprietary free downloadable software MetAlignID, has been successfully validated with respect to false negatives for the sum PCB 28, 52, 101, 138, 153 and 180), for the sum of BDE 28, 47, 99, 100, 153, 154 and 183, for the four markers of PAHs and for a number of emerging brominated flame retardants. A screening detection limit (SDL) equal to or lower than the maximum regulatory level was always achieved. MetAlignID considerably decreased the time needed for data treatment from 20 to 5min/file.
Automated identification of the signature mass spectral patterns was applied to identify chlorinated- and brominated-containing substances with more than two halogen atoms, and PAH derivates. Although the success rate was variable and needs to be further improved, the tool was considered to be of added value.