Elsevier

Food Chemistry

Volume 333, 15 December 2020, 127488
Food Chemistry

Analytical Methods
Mild mixed-solvent extraction for determination of total mineral oil hydrocarbon contaminants in milk powder products

https://doi.org/10.1016/j.foodchem.2020.127488Get rights and content

Highlights

  • n-Hexane/isopropanol (3:1) extract total mineral oils in milk powder products.

  • The mixed-solvent extraction avoid the loss of low-boiling-point hydrocarbons.

  • Major part of the mineral oil contamination is derived from sources before packaging.

Abstract

A mild mixed-solvent of n-hexane/isopropanol is proposed for extracting total mineral oil hydrocarbons (MOH) from commercial milk powder products. Unlike acid-hydrolysis, the mixed-solvent extraction was performed at ambient temperature and the low-boiling-point hydrocarbons were retained to the greatest extent. After extraction, total MOH was determined by on-line liquid chromatography-gas chromatography with a flame ionization detector (LC-GC-FID). The validation of the proposed extraction method revealed a recovery efficacy of 83.0–107.5% and a limit of quantification of 0.5 mg/kg. Then, the total MOH in ten commercial milk powders was determined and mineral oil saturated hydrocarbons (MOSH)/polyolefin oligomeric saturated hydrocarbons (POSH) were found to be within the range of 0.61–5.46 mg/kg. The comparison of the total and surface MOSH/POSH indicated that a major part of the contamination was derived from sources before packaging. The present study provides a robust method for the extraction and determination of total MOH in milk powders.

Introduction

In the past few decades, mineral oil contaminants in various foods have been of great concern due to their potential adverse health effects (European Food Safety Authority (EFSA), 2012, Barp et al., 2014, Barp et al., 2017a, Barp et al., 2017b). According to numerous previous studies, the sources of mineral oils are probably packaging materials, processing aids, machine lubricants, which may migrate into food during harvesting, manufacturing, transportation and storage processes (Biedermann and Grob, 2010, Dima et al., 2011, Moret et al., 2003, Droz and Grob, 1997, Moret et al., 1997). Chemically, mineral oils are extremely complex mixtures from petrogenic hydrocarbons and synthetic fuels of C10–C50, consisting of mineral oil saturated hydrocarbons (MOSH) and mineral oil aromatic hydrocarbons (MOAH). MOSH include paraffins and naphthenes, and MOAH are mostly alkyl-substituted aromatics containing 1–5 benzene rings (EFSA, 2012).

Milk powder products play an important role in providing rich nutrients to humans, especially to some newborn babies, who may obtain almost all their nutrients from infant formula. Accordingly, mineral oils in infant formula have attracted much public attention. In 2017, a non-profit organization released on its website the mineral oil contamination levels in some commercial infant formula products in the Chinese market (Zhang et al., 2019). In 2019, another non-profit campaigning organization reported the baby milk products containing mineral oil hydrocarbons (MOH), especially MOAH, which is considered to be potentially carcinogenic due to their similar structure as that of polyaromatic hydrocarbons (PAHs) (Foodwatch, 2019). However, a subsequent report from the European Food Safety Authority (EFSA) pointed out that MOAH were not detected in the same batches of infant formula by German authorities. Such discrepancies could be due to the complexity of the analytical methods used for determining mineral oils in milk powder products. In those analyses, on-line coupled high-performance liquid chromatography-gas chromatography with a flame ionization detector (LC-GC-FID) was used for the determination and comprehensive two-dimensional GC coupled with mass spectrometry (GC × GC–MS) for the verification of the results (Foodwatch, 2019, Arcella et al., 2019).

To date, on-line LC-GC-FID has been widely used to determine MOSH and MOAH levels in various foods since it integrates sample preparation, implements automation and minimizes the introduction of sample contamination (Biedermann, Munoz, & Grob, 2017). The MOSH and MOAH fractions in the mineral oil extracts are purified and separated by LC using a silica gel column and then subjected to GC-FID analysis (Biedermann and Grob, 2012a, Biedermann and Grob, 2012b). The advantages of LC-GC-FID are high sensitivity, good repeatability, and little introduction of external pollution. However, due to the limited selectivity of FID, which cannot distinguish between mineral oils and sample matrix interferences, the confirmatory GC × GC–MS method is sometimes necessary for LC-GC-FID chromatograms. Besides, the quantitative LC-GC-FID and qualitative GC × GC–MS analysis of mineral oils, when necessary the methods of aluminum oxide and epoxidation are correspondingly used to remove natural odd number n-alkanes and biogenic alkenes in food samples (Bratinova & Hoekstra, 2019).

Nevertheless, the extraction methods are equally important in purification and detection techniques for mineral oils in complex food samples. Different extraction procedures have been used depending on the food composition and the contamination source. For milk powder samples, only superficial mineral oils are extracted using n-hexane overnight or at 60 ℃, which is generally used to extract mineral oils in most of foods (Biedermann and Grob, 2012a, Zhang et al., 2019). Usually, a substantial part of mineral oil contamination in milk powder is derived from raw materials and the production process of milk powder products. The mineral oils as well as oils and fats in milk powder are wrapped in protein and they cannot be extracted by n-hexane. Accordingly, Biedermann and Grob (2012a) introduced acid hydrolysis to release mineral oils from milk protein and then extracted them with n-alkane. The acid hydrolysis is usually performed at 80 ℃ for 30 min and the harsh reaction condition may cause volatilization loss of low-boiling mineral oils and internal standards used for quantification, such as bicyclohexyl (Cycy) and 2-methylnaphthalene (2-MN), thus leading to deviations of the quantitative results. This may be the reason of the discrepancies among the results from several comparisons (Koster et al., 2020, Foodwatch, 2019, Arcella et al., 2019).

To address these issues, the aim of this study is to develop a method for determining total mineral oils in milk powders using a mild mixed solvent for their extraction combined with LC-GC-FID analysis. The results of the two extraction methods, using acid hydrolysis followed by n-hexane and mixed solvents, were compared. Additionally, the detection data of the total and surface mineral oils in the same milk powder products available in Chinese markets were also compared.

Section snippets

Reagents and standards

n-Hexane (pesticide residue analysis grade), dichloromethane (pesticide residue analysis grade), and ethanol (HPLC grade) were purchased from Thermo Fisher Scientific (Fair Lawn, NJ, USA); isopropanol (pesticide residue analysis grade) was obtained from Meridian Medical Technologies Inc. (Columbia, MD, USA). The standards, including n-undecane (C11), n-tridecane (C13), bicyclohexyl (Cycy), 5α-cholestane (Cho), n-pentylbenzene (5B), 1,3,5-tri-tert-butylbenzene (TBB), 1-methylnaphthalene (1-MN),

Selectivity of the extraction solvents

The methods for the extraction of mineral oils present in milk powders are almost the same as those for fats, as these hydrocarbons are similar in polarity to fats and are miscible. To some extent, the degree of extraction of mineral oils from milk powder samples can be determined by the extraction yield of fat. The experimental data from Bierdermann-Brem, Kasprick, Simat and Grob (2012) indicated that only 10% of fats declared on the label of the formula products were extracted at ambient

Conclusion

A mild liquid–liquid solvent extraction method using a n-hexane/isopropanol (3:1) solvent mixture is proposed to prevent volatility loss during the analysis of total mineral oils in milk powder products. The extraction was performed using the mixed solvents to extract twice at room temperature with high efficiency for various powdered milk products. After optimization, 2 g of samples were dissolved in 8 mL of water, extracted twice with 40 mL and 20 mL of n-hexane/isopropanol (3:1),

CRediT authorship contribution statement

Lingling Liu: Methodology, Investigation, Writing - original draft. Bingning Li: Validation, Investigation, Software. Jie Ouyang: Investigation, Formal analysis, Visualization, Writing - review & editing. Yanwen Wu: Funding acquisition, Conceptualization, Supervision, Writing - review & editing, Writing - original draft.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

The authors gratefully acknowledge the support from the Beijing Natural Science Foundation (No. 2182020) and Beijing Academy of Science and Technology-Reform and Development Project (BJAST-RD, No. PY2020JC38).

References (29)

  • M. Biedermann et al.

    Aromatic hydrocarbons of mineral oil origin in foods: Method for determining the total concentration and first results

    Journal of Agricultural and Food Chemistry

    (2009)
  • M. Biedermann et al.

    Is recycled newspaper suitable for food contact materials? Technical grade mineral oils from printing inks

    European Food Research and Technology

    (2010)
  • S. Bierdermann-Brem et al.

    Migration of polyolefin oligomeric saturated hydrocarbons (POSH) into food

    Food Additives and Contaminants: Part A

    (2012)
  • View full text