Development of a Method for the Analysis of Primary Aromatic Amines by LC‐MS/MS and LC‐HRMS and their Stability under Different Migration Conditions
E. Koch, H. J. Fröhlich, A. I. Ostermann, M. D. Wälisch, S. Richter, D. Joch, A. Roloff, T. BruhnPrimary aromatic amines (PAA) can occur in food contact materials (FCM) as a contamination from azo dyes synthesis or as a result of azo bond cleavage under reducing conditions. Besides colorants, other sources of PAA include e. g. residues of isocyanates in polyurethane‐based adhesives, which can be hydrolyzed to PAA in aqueous environments, or co‐monomer addition in polyamide manufacturing. Thus, plastic kitchenware, but also paper and board FCM, can be relevant sources of PAA exposure for consumers. When PAA migrate from FCM into food, this may raise toxicological concerns as some congeners of this diverse substance class are classified as carcinogenic to humans according to Regulation (EC) No 1272/2008 (CLP Regulation). Therefore, migration into food is restricted according to Regulation (ELI) No 10/2011 for plasticware and the BfR Recommendation XXXVI for paper and board FCM to be non‐detectable, with a specified limit of detection (LOD) of 0.002 mg/kg food per individual PAA. PAA migration from plasticware is typically tested using 3 % acetic acid, as this is considered to be the worst‐case food simulant. However, previous studies reported considerable degradation of PAA under acidic conditions [1, 2] which might result in an underestimation of PAA migration.
In this work we present a multi‐analyte method for the determination of 42 PAA using liquid chromatography‐mass spectrometry (LC‐MS) on a biphenyl analytical column (150 times 2.1 mm ID, 3 pm particle size) within a total run time of 18 min. The method is suitable for both, triple quadrupole (QqQ) MS instruments as well as high‐resolution Orbitrap systems. For most analytes, both setups yielded LODs below 2 ng/mL as defined in Regulation (ELI) No 10/2011. The tandem‐MS system was typically more sensitive, achieving detection limits of < 0.1 ng/mL. The method was successfully applied for the direct analysis of PAA in cold water extracts of paper FCM and was characterized by high inter‐day precision (< 10%) and recoveries (95‐105%) for most analytes. Additionally, we compare migration of PAA from polyamide kitchenware utensils into 3 % acetic acid and water, and present data on the stabilities of PAA standards during storage in solvents and cold water extracts. The results will help to further optimize the strategy for migration testing of PAA from FCM including the definition of realistic worst‐case conditions.