Comparison of microwave-assisted saponification and extraction methods for MOH determination in meat

Mineral oil hydrocarbons (MOH) are a complex mixture of liposoluble environmental and processing contaminants of petrogenic origin. They may pose different toxicological risks to humans depending on their structure (i.e., saturated (MOSH) or aromatic (MOAH)) [1]. The increasing interest in more detailed information on the composition of MOH is driving a shift from LC-GC-FID, considered the routine technique, to more advanced techniques, notably LC-GCxGC-FID. Moreover, due to the complexity of MOH and food matrices, an efficient sample preparation method must be performed. One of the steps commonly included in the sample preparation for MOH analysis is saponification, which allows the complete extraction of MOH from the lipids present in the matrix [2]. Nonetheless, this step impacts the distributions of some internal standards used for quantification due to a different partition in the solvent phases [3]. This study aimed to compare two different microwave-assisted saponification and extraction (MASE) methods for the analysis of MOH in various types of meat.

Three different types of unprocessed meat (raw bacon, pig rib, and beef rib) were chosen. The samples undergo two different MASE procedures: i) KOH saturated solution in methanol (MASE 1) followed by a washing step with 40 mL of water plus 3 mL of methanol, and the vessels were kept at −18 °C for 30 min before recovering the hexane phase [4]; or ii) a KOH 2 M in EtOH/H2O (1/1 v/v) (MASE 2), followed by the addition of 20 mL of H2O and storing the vessels 20 min in the fridge before recovering the hexane phase. The methods were evaluated in terms of internal standards distribution and recovery. The results confirmed that the partition of the standards tri-tert-butyl benzene (TBB) and 2-methyl naphthalene (2-MN), which are commonly used to quantify MOAH, depended on the MASE method applied. The differences in recovery between both standards were higher when MASE 1 was applied, showing a lower recovery of 2-MN compared to TBB (ratio TBB/2-MN of 1.24 on average). When MASE 2 was applied, the ratio TBB/2-MN (ratio of 1.06 on average) was closer to their expected proportion in the internal standard (ratio of 1), which guarantee a better representativeness of the MOAH extraction and prevent differences between the results when quantifying with TBB or 2-MN. The ratio between the two standards was consistent among the different meat types tested.
The accuracy of both MASE methods for MOSH was higher than 90% in both spiked levels (added levels of ≈1.6 and 3.6 mg/kg). For MOAH (added at the same levels as MOSH), the accuracy was lower when MASE 1 was applied (73 – 80%) in comparison to MASE 2 (95 – 97%).