Food Chromatography Approaches for Mineral Oil Hydrocarbon Analysis

Mineral oil hydrocarbons (MOH), including mineral oil saturated hydrocarbons (MOSH) and mineral oil aromatic hydrocarbons (MOAH), have become an important focus in food safety due to their widespread presence in edible oils, dairy products, cereals, infant formula, chocolate, and packaged foods. These contaminants can enter food through environmental exposure, food processing equipment, and migration from packaging materials. Because MOH compounds consist of highly complex mixtures of structurally similar hydrocarbons, chromatographic analysis remains challenging, often producing unresolved “hump-shaped” profiles in liquid chromatography-gas chromatography coupled toflame ionization detection (LC–GC–FID) analysis.

Recent research highlighted current chromatographic and sample preparation strategies used for MOH determination in food matrices, including solvent extraction, saponification, activated alumina clean-up, and epoxidation techniques designed to remove interfering compounds such as triglycerides, olefins, and n-alkanes. The resulting article, published in Microchemical Journal,¹ also discusses the toxicological relevance of MOSH and MOAH fractions, particularly the bioaccumulation potential of MOSH in the C20–C35 range and the carcinogenic concerns associated with certain polyaromatic MOAH compounds.
LCGC International spoke to Ana Jano, Ana M. Ares, Floriatan Santos Costa, Jorge A. Custodio-Mendoza, José Bernal, and Adrián Fuente-Ballesteros, co-authors of the aforementioned paper, about their work.

How does the complexity of unresolved mixtures in liquid chromatography–gas chromatography coupled to flame ionization detection (LC–GC–FID) analysis impact the accurate quantification of mineral oil saturated hydrocarbons (MOSH) and mineral oil aromatic hydrocarbons (MOAH) in diverse food matrices?

The main challenge comes from the fact that MOSH and MOAH are not single compounds, but highly complex unresolved mixtures composed of different hydrocarbons. In LC–GC–FID analysis, insufficient chromatographic separation can lead to overlapping signals, commonly observed as broad unresolved humps in the chromatogram. This makes it difficult to distinguish the mineral oil hydrocarbons (MOH) from naturally occurring interferences present in food matrices. As a consequence, poor separation directly affects quantification accuracy, because the detector response may include contributions from non-target compounds. This is particularly critical for MOAH, since their toxicological relevance strongly depends on their molecular structure, especially the presence of polycyclic aromatic systems with 3-7 fused rings. Therefore, achieving efficient chromatographic separation and adequate sample clean-up is essential not only for accurate quantification, but also for correctly assessing potential health risks associated with these contaminants.

What chromatographic strategies can be employed to improve the separation of MOAH fractions containing 3–7 fused aromatic rings from interfering food components?

One of the most important approaches is the use of selective clean-up procedures prior to chromatographic analysis. For example, epoxidation is commonly applied to remove naturally occurring olefins that may coelute with MOAH fractions. Alumina or silica-based clean-up steps can also reduce matrix interferences such as long-chain n-alkanes, lipids, and polar compounds.

From the chromatographic perspective, comprehensive two-dimensional gas chromatography (GC×GC), coupled either to flame ionization (FID) or mass spectrometry (MS), has become a promising alternative to conventional LC–GC–FID. GC×GC provides higher separation power and allows better discrimination of aromatic compounds according to volatility and polarity. This is useful for resolving these MOAH, which are toxicologically more relevant and may otherwise remain hidden within unresolved chromatographic humps. Additionally, optimizing stationary phases, gradient conditions, and LC fractionation procedures can further improve selectivity and reduce coelution with food-derived compounds. In some recent studies, LC fractionation on silica has also been proposed as an alternative to epoxidation, minimizing MOAH losses while improving separation efficiency.²

How do different food matrices (e.g., edible oils vs. dairy products) influence the efficiency of chromatographic detection of MOSH?

Different food matrices influence chromatographic detection because each matrix contains its own characteristic compounds that may interfere with MOSH separation and quantification. Even if the analytical method achieves good separation between MOSH and MOAH fractions, matrix-specific components can still coelute near the MOSH region and complicate interpretation of the chromatogram.

In general, fatty matrices tend to accumulate MOH more easily because of the lipophilic nature of these contaminants, but they also present greater analytical challenges due to the abundance of coextractable compounds. For example, edible oils are highly lipidic matrices rich in triglycerides and naturally occurring hydrocarbons, while dairy products may additionally contain sterols, waxes, and other degradation products. These compounds can overload the chromatographic system, distort the unresolved hump, or mask parts of the MOSH signal. As a result, more extensive sample preparation steps, such as saponification, alumina clean-up, or enrichment procedures, are often required depending on the matrix complexity. Therefore, the efficiency of chromatographic detection depends not only on the instrumental separation itself, but also on how effectively the sample preparation removes matrix interferences beforehand.

What are the main limitations of current chromatographic methods in distinguishing MOSH from structurally similar compounds such as polyolefin oligomeric saturated hydrocarbons (POSH) in processed foods?

One of the main limitations is that POSH and MOSH share very similar physicochemical properties, since both mainly consist of saturated branched hydrocarbons. Because of this similarity, they frequently coelute during LC separation and generate overlapping unresolved humps in LC–GC–FID chromatograms, making accurate identification and quantification particularly difficult. However, in some situations, their chromatographic behavior may help differentiate them. For example, POSH originating from polypropylene often produce characteristic clusters of peaks and unresolved humps that differ from the broader MOSH patterns typically associated with petrogenic sources. On the other hand, POSH derived from polyethylene are much more difficult to distinguish because their chromatographic profiles can closely resemble those of MOSH. Certain features may still provide clues: polyethylene-derived POSH in the C12-C18 range usually show predominantly even-numbered n-alkanes, whereas MOSH generally contain both even- and odd-numbered homologues throughout the same range.³

Despite these observations, conventional on-line LC–GC–FID methods still do not provide sufficient selectivity to fully separate MOSH from POSH when both are present simultaneously. As a result, reliable quantitative discrimination between these two fractions remains one of the major analytical challenges in MOH analysis, especially in processed foods where packaging-related contamination is common.³

How can sample preparation techniques be optimized to reduce matrix effects prior to chromatographic analysis of mineral oil hydrocarbons (MOH) in complex foods like chocolate or infant formula?

For complex matrices such as chocolate or infant formula, sample preparation is crucial because these foods contain high amounts of lipids, proteins, and naturally occurring compounds that may interfere with chromatographic analysis. Matrix effects can be minimized by combining selective extraction and clean-up strategies adapted to the specific composition of the food. Typically, an initial solvent extraction with hexane or hexane mixtures is followed by saponification to remove triglycerides and reduce lipid content. Additional purification steps, such as alumina clean-up, can eliminate long-chain n-alkanes and other interfering hydrocarbons, while epoxidation may be applied to remove naturally occurring olefins that coelute with MOAH fractions. In highly fatty matrices, enrichment and reconcentration steps are also important to improve sensitivity without overloading the chromatographic system. The optimization of these procedures depends strongly on the matrix complexity. For example, infant formula often requires extensive triglyceride removal, whereas chocolate may additionally contain compounds such as waxes, emulsifiers, or cocoa-derived lipophilic substances that interfere with separation.

In what ways does the carbon number distribution (e.g., C20–C35 fraction) of MOSH affect chromatographic resolution and detector response?

The carbon number distribution influences chromatographic separation because hydrocarbons with similar chain lengths and boiling points tend to elute very close to each other. As the carbon numbers become more similar, chromatographic resolution decreases, contributing to the formation of broad unresolved humps characteristic of MOSH analysis.

The C20-C35 fraction is important because it is considered the range most likely to accumulate in human tissues and organs. Therefore, achieving reliable separation in this interval is particularly relevant from a toxicological perspective. In terms of detector response, FID generally provides a nearly uniform response per mass unit for hydrocarbons, which is an advantage for MOSH analysis. However, poor chromatographic resolution in specific carbon ranges can still hinder accurate integration and quantification, especially when interferences from the food matrix overlap with the MOSH signal.

How does the choice of extraction solvent (e.g., hexane vs. dichloromethane mixtures) influence chromatographic performance and sensitivity in MOH analysis?

Since MOSH and MOAH are highly apolar compounds, non-polar solvents such as hexane are commonly used as the primary extraction medium. However, more complex matrices or slightly more polar aromatic fractions may require solvent mixtures, such as hexane with dichloromethane or isopropanol, to improve extraction efficiency. The solvent choice affects chromatographic performance because excessive coextraction of lipids, pigments, olefins, or other food-derived compounds can overload the chromatographic system and increase coelution problems. Cleaner extracts generally lead to better chromatographic resolution and easier integration of MOSH and MOAH humps. Sensitivity is also influenced, since more efficient extraction of the target analytes combined with lower background interference improves signal-to-noise ratios and facilitates detection at low concentrations.

What role do clean-up steps such as epoxidation and alumina column fractionation play in enhancing chromatographic selectivity for MOAH determination?

As has been mentioned before, clean-up procedures are important for improving chromatographic performance, because food matrices contain many naturally occurring compounds that may interfere with MOAH detection. Epoxidation is mainly used to remove unsaturated compounds such as olefins, including naturally occurring substances like squalene, sterenes, or carotenoids, which can coelute with the MOAH fraction and distort the unresolved chromatographic hump. By converting double bonds into epoxides, these compounds become more polar and are retained differently during chromatographic separation, reducing interference in the MOAH region. However, procedures such as epoxidation may induce partial MOAH losses or variability in recoveries. Alumina column fractionation, on the other hand, is mainly applied to remove long-chain n-alkanes and other saturated hydrocarbons that interfere with MOSH determination. Activated alumina selectively retains these compounds, improving hump definition and preventing chromatographic overloading.

How can green analytical chemistry principles be integrated into chromatographic workflows for MOH analysis without compromising sensitivity and reproducibility?
Integrating green analytical chemistry principles into MOH analysis is challenging because these methods already require complex separations, extensive clean-up procedures, and very low limits of quantification as they are. Nevertheless, several strategies can improve sustainability without compromising analytical reliability. One approach is miniaturization, which reduces solvent consumption, sample size, and waste generation while maintaining extraction efficiency. Increased automation is also important because it decreases manual handling, improves reproducibility, and reduces procedural complexity. Replacing highly hazardous solvents with safer alternatives may contribute as well, although changing individual reagents alone is usually insufficient unless the entire workflow is optimized simultaneously. The application of analytical metrics and the principles of green or white analytical chemistry in these analyses still represents a challenge and a field that remains largely unexplored.

What are the key challenges preventing harmonization of chromatographic methods for MOH determination across different food safety laboratories, and how might they be addressed?

Although organizations such as the Joint Research Centre (JRC) from the European Commission have provided guidance for MOH analysis,⁴ complete harmonization between laboratories has not yet been achieved. One major challenge is the large variability in food matrices, which often requires matrix-specific sample preparation and clean-up procedures. Different laboratories may therefore apply different extraction solvents, saponification conditions, enrichment methods, or clean-up strategies depending on the type of food analyzed. Another important issue is the difficulty in interpreting unresolved chromatographic humps and distinguishing MOH from naturally occurring or packaging-derived interferences. Variability in integration criteria, epoxidation protocols, and confirmation methods can lead to differences in quantification between laboratories.

References

1. Jano, A.; Ares, A. M.; Santos Costa, F. et al. Recent Advances in Sample Preparation for the Determination of Mineral Oil Saturated (MOSH) and Aromatic Hydrocarbons (MOAH) in Food Matrices. Microchem J 2026, 225, 118083. DOI: 10.1016/j.microc.2026.118083
2. Gorska, A.; Bauwens, G.; Beccaria, M. et al. Purification of Mineral Oil Aromatic Hydrocarbons and Separation Based on the Number of Aromatic Rings Using a Liquid Chromatography Silica Column. An Alternative to Epoxidation. J Chromatogr A 2025, 1743, 465684. DOI: 10.1016/j.chroma.2025.465684
3. Biedermann-Brem, S.; Kasprick, N.; Simat, T. et al. Migration of Polyolefin Oligomeric Saturated Hydrocarbons (POSH) into Food. Food Addit. Contam. Part A Chem Anal Control Expo Risk Assess 2012, 29, 449–460. DOI: 10.1080/19440049.2011.6411
4. Bratinova, S.; Robouch, P.; Hoekstra, E. et al. Guidance on Sampling, Analysis and Data Reporting for the Monitoring of Mineral Oil Hydrocarbons in Food and Food Contact Materials – 2nd Edition; JRC Publications. 2023. DOI: 10.2760/963728