Perfluoroalkyl and polyfluoroalkyl substances (PFASs) are extremely stable highly fluorinated aliphatic compounds, widely used in chemical and industrial applications since 1950s. They are ubiquitously detected in the environment and diet is the main route for human exposure. To ensure human safety, it is necessary to follow the whole food production chain, including animal feed. Still PFASs are not regulated as undesirable substances in feed, although several studies have shown they can be transferred from feed to livestock and thereafter to food. This paper describes the development, optimization and the full validation of a sensitive and reliable analytical protocol enabling the quantification of 19 PFASs in animal feeds by liquid chromatography - mass spectrometry (LC-MS/MS). The method was optimized, assessing the possible matrix interferences, and submitted to comprehensive validation (55 independent spiking experiments). Validation experiments were conducted on blank fish feed samples (natural levels of PFASs < 0.10 ng g−1) spiked at five different concentrations (0.10, 0.50, 1.0, 5.0 and 10 ng g−1). Apparent recoveries (R%) were generally between 88 and 111%; R% < 80% were obtained only at the lower validation levels for those molecules not having the corresponding labelled analogues. Relative standard deviations in repeatability conditions (RSDr) and within-lab reproducibility conditions (RSDwR) were lower or equal to 11% and 22% respectively. Limits of quantification were set for most of the analytes at 0.10 ng g−1 (LOQs) and verified with repeated analysis on fortified samples (0.10 ng g−1). Limits of detection (LODs) were calculated as 1/3 LOQ. Finally the method was applied to 23 feed samples of different type (compound feed, vegetable feed material, complementary feed, pre-mixture and mineral feed) and all of them did not show PFAS above LOQ.
Read full abstract