Analytics of Substances in Food Packaging Material
The Challenge of Tracking NIAS in Food Contact Materials
Non-intentionally added substances (NIAS) can be present in plastic materials used in contact with food and thus have the potential to be transferred to the food. NIAS might be impurities in the substances that are intentionally used for the production of a plastic material, reaction intermediates formed during the polymerisation process or decomposition or reaction products. Examples of NIAS typically found in food contact materials and their possible sources are given in Table 1. In contrary to intentionally added substances (IAS) such as monomers, additives or production aids, the identity of such substances is very often not known, which makes their analyses and food regulatory assessment challenging. In the following the possibilities and difficulties in tackling the problem of NIAS migration are discussed.
Food Regulatory Background
Whereas intentionally added substances such as monomers and additives or polymer production aids need to be authorized in the positive list of the European Plastics Regulation (EU) No 10/2011, NIAS need not be listed. However, migration of such substances must comply with the general safety requirements of Article 3 of the European Framework Regulation (EC) No 1935/2004 and shall be assessed in accordance with internationally recognised scientific principles of risk assessment (Article 19 of Regulation (EU) No 10/2011). For non-evaluated NIAS a limit of 10 µg/kg food is typically applied.
The fact that the identity of NIAS is usually not or incompletely known to the manufacturer of a food contact material as well as the required detection limit of < 10 µg/kg food (simulant) pose a challenge to their analytical determination. For unknown substances, non-target screening tests need to be applied. The analytical method of choice should be able to detect, identify and quantify a wide variety of substances differing in their physical and chemical properties. The mode of detection should be independent from the detected molecule, i.e. the response of the selected detector should be the same or similar in order to enable a (semi-)quantification of the detected substances.
Depending on their volatility, gas chromatographic and liquid chromatographic techniques can be applied for the analysis of NIAS, either in the material itself or in migrates.
Table 2 gives an overview over the possibilities and limitations of different non-target screening methods.
By headspace gas chromatography (GC), highly volatile substances can be determined in the material. Substances with a molecular weight of up to approx. 300 g/mol can be detected with this method. Identification can be achieved by coupling with mass spectrometry (MS) and comparison of the obtained mass spectra with databases. Extensive databases exist for GC-MS as the ionisation and fragmentation of the analytes is reproducible and resulting spectra can be compared directly.
For quantification a flame ionisation detector (FID) can be used. The principle of headspace-GC is the partitioning of compounds between the gas phase and the liquid phase, which strongly depends on the volatility and thus the molecular weight and polarity of a substance. The signal intensity thus decreases with increasing molecular weight. Quantification can be achieved by multiple headspace extraction (MHE), using the corresponding substance or a structurally related substance with similar polarity for calibration.
Substances of lower volatility are analysed by GC-FID/MS. Typically, extracts of a material are investigated as the concentration of unknown substances is higher than in migrates, which facilitates their identification. With this method, the compounds detected can be semi-quantified using a universal standard as the FID signal intensity correlates with the number of carbon atoms per time unit.
Non-volatile compounds are determined by high performance liquid chromatography (HPLC). Here the identification of substances is more difficult as no extensive mass spectra databases are available. The reason for this is that different ionisation modes can be applied depending on the different properties of the analytes, which leads to different fragmentation patterns. Furthermore, the setting of the analytical instrument and the conditions in the ion source also influence the fragmentation pattern, which makes a comparison of obtained mass spectra between different laboratories difficult.
For the quantification of known analytes determined by HPLC, a wide variety of detectors (e.g. UV, fluorescence, MS, ELSD) specialized on the structural elements of the analytes to be quantified is available. However, these are impracticable for the quantification of unknown substances as they only respond to certain structural elements such as the presence of a chromophore. As universal detector for HPLC, the charged aerosol detector (CAD) is most suitable. The principle behind this mode of detection is a nebulisation of the mobile phase and subsequent drying of the aerosol and charging of the resulting particles. These charged particles give a signal which correlates to the quantity of analyte.
In many cases, instead of non-target analyses multi-analyte methods that are focussed on a defined group of substances or certain structural elements can be applied. An example are primary aromatic amines (PAAs) which can occur in coloured or printed materials as residual educts from the synthesis of azo pigments or as degradation products of azo pigments generated by reductive cleavage. PAAs can also occur as reaction products of isocyanates formed by degradation of polyurethane adhesives. Such an LC-MS/MS multi-analyte method focussing on PAAs deriving from azo pigments and isocyanates typically used in food contact materials is currently being developed at Fraunhofer IVV.
Food contact materials are a complex mixture of known, intentionally added substances (IAS) and non-intentionally added substances (NIAS) with different physical and chemical properties which can migrate into food. It must be noted that no single analytical method is able to detect all possible migratable NIAS. Different screening approaches need to be applied for different groups of substances. A combination of several screening approaches will yield the maximum of information. The more is known about IAS, the better the chances to identify NIAS because then the screening methods applied can be adapted to the physical and chemical properties of the groups of substances expected. Multi-analyte methods can also be useful in some cases.
 Annika Ebert, Roland Franz, Carina Gehring, Diana Kemmer, Frank Welle; Testing Migration from Food Packaging Materials: Testing & Quality Assurance In book: Food Packaging Materials, pp.251-302; 07-2017; 2017; DOI: 10.1201/9781315374390-12
 ILSI Europe, Guidance on best practices on the risk assessment of non intentionally added substances (NIAS) in food contact materials and articles, 2015; D/2015/10.996/39