Suspected-target screening for the assessment of plastic-related chemicals in honey

Abstract

Plastic-related chemicals (PRCs) are commonly reported in a variety of foods as a result of (i) the initial contamination of the plant or the animal used to produce food, (ii) contamination during food processing or (iii) chemical migration from food contact materials. These contaminants may affect the organoleptic properties of the food or represent a food safety hazard for the consumers. To date, the non-targeted screening for PRCs in honey has not been reported. The aim of this study was to apply a non-targeted method to screen PRCs in honey and investigate their occurrence among honey products from the market. A total of 662 putative compounds were detected and the identification of two of these compounds, namely 2-ethyhexyladipate (DEHA) and tris (2-butoxyethyl) phosphate (TBOEP), was further confirmed with standards as proof-of-concept for the approach. Their concentrations and estimated daily intakes were calculated, resulting in a total of 0.001 and 0.006 μg.kg−1bw.day−1 for DEHA and TBOEP, respectively. With regards to the comparison of the chemical burden in honey samples sold in either glass or plastic jars, various data treatment approaches were used which resulted in different lists of relevant contaminants. As expected based on the low frequency of detection of the 662 potential PRCs found in honey, unique entity analysis with 100% detection and volcano plot with p < 0.05 found none and 2 compounds, respectively. When applying a data treatment approach based on the differential frequency of detection, however, the number of relevant compounds increased to 13 in glass and 40 in plastic, among which 6 were unique to honey samples sold in plastic jars and 3 were unique to honey samples sold in glass jars. The different results obtained with the various data treatment approaches suggest that the relatively low frequency of contaminants in food needs to be accounted for when selecting the appropriate data treatment tool to compare groups of samples.