Mineral Oil Residues in Food - Part 2 - Automated Removal of Natural Interferences by Online Epoxidation

Significance of the Topic

Mineral oil hydrocarbons, including saturated (MOSH) and aromatic (MOAH) fractions, are common contaminants in food with concentrations up to several thousand mg/kg. Accurate routine analysis is essential to assess consumer exposure and ensure compliance with health regulations.

Objectives and Overview

The study presents an automated epoxidation approach integrated into an online LC-GC-FID workflow to selectively remove natural olefinic interferences such as squalene, sterenes, and carotenoids, improving the reliability of MOAH quantification.

Methodology

Sample preparation involves weighing oil samples into vials with n-hexane and internal standards, followed by automated addition of meta-chloroperbenzoic acid solution at 40 °C for epoxidation. Post-reaction neutralization with sodium sulfite, phase separation by centrifugation, and injection protocols are fully automated. The LC step employs a silica column with a n-hexane/dichloromethane gradient to fractionate MOSH and MOAH. Each fraction is transferred via a Chronect interface to dedicated high-temperature GC columns with flame ionization detection.

Used Instrumentation

• Shimadzu LC-20AD with D2 UV detector
• Shimadzu GC-2010 Dual FID
• Axel Semrau Chronect LC-GC interface
• PAL RTC autosampler with integrated agitator and centrifuge
• Allure Silica column (250 × 2.1 mm, 5 µm)
• Restek MXT Siltek guard columns and MXT-1 analytical columns

Main Results and Discussion

Automated epoxidation achieved over 90 % removal of squalene, eliminating severe chromatographic distortion in the MOAH fraction. Post-epoxidation chromatograms display well-resolved internal standard peaks, confirming improved quantification accuracy. The workflow supports simultaneous separation of MOSH and MOAH fractions in a single run with high throughput.

Benefits and Practical Applications

• Enhanced selectivity enabling accurate MOAH quantification in complex matrices
• Fully automated sample cleanup reduces manual labor and variability
• High sample throughput suitable for routine quality control laboratories

Future Trends and Opportunities

Future developments may include integration of mass spectrometric detection for compound-specific identification, adaptation to additional biogenic interferents, and implementation of green chemistry principles for reagent and solvent consumption. Advanced data processing and machine learning could further streamline interpretation.

Conclusion

The LC-GC-FID system with online epoxidation effectively removes olefinic interferences, enabling reliable routine determination of mineral oil hydrocarbons in food. Automated operation enhances throughput and reproducibility.

References

• EFSA Panel on Contaminants in the Food Chain CONTAM Scientific Opinion on Mineral Oil Hydrocarbons in Food DOI 10.2903/j.efsa.2012.2704
• Biedermann M Aromatic Hydrocarbons of Mineral Oil Origin in Foods J Agric Food Chem 2009 57 8711–8721
• Nestola M Determination of mineral oil aromatic hydrocarbons in edible oils and fats by online LC GC FID Journal of Chromatography A 1505 2017 69–76