Automating workflows for the determination of 3-MCPD-, 2-MCPD-, and glycidol-esters as well as mineral oil hydrocarbon residues in edible oils and fats to overcome the challenges associated with manual sample preparation

Fatty acid esters of 3—and 2-monochloropropanediol (3-MCPD-e, 2-MCPD-e) and glycidol (Gly-e) are process contaminants formed, for example, when edible oils and fats are refined. After ester cleavage during digestion in the human body, they pose a relevant health risk. Mineral oil hydrocarbons (MOH) contamination, including mineral oil saturated hydrocarbons (MOSH) and mineral oil aromatic hydrocarbons (MOAH), originates from sources such as packaging, processing aids, and machine oils, posing health risks due to accumulation in human tissues and potential carcinogenic properties. Determination of these contaminants in fats and oils is therefore of high importance for consumer safety.

Determination of these compounds is, however, quite complex, requiring many manual sample preparation steps and large amounts of organic solvents. To increase efficiency, reliability, and sustainability, automation of sample preparation is highly desirable. Using a wide array of sample preparation modules to automate individual workflows, we established methods for MOSH/MOAH determination according to ISO 16995 and glycidol- and 3-MCPD ester determination according to ISO 18363-4 (Zwagerman-Overman method), demonstrating the advantages and potential pitfalls of automated sample preparation.
For example, in the context of MOSH/MOAH determination, a complete sample preparation method was developed, including fully automated saponification of oil and fat samples in combination with performic acid epoxidation and online cleanup using activated aluminum oxide. Studies on the reproducibility of the method showed very good results for different matrices. Since manual integration of the resulting unresolved complex mixtures is very subjective and can lead to huge deviations between different users integrating the same chromatogram, automating the data integration was found to enhance throughput and reproducibility of the data evaluation allowing a potential to standardize MOH analysis across laboratories and improving food safety monitoring.