Food contact: conformity assessment - Part 1

T he prediction of food contamination by packaging substances is a recent concern. Although the models available today are still imperfect, the use of predictive approaches is strongly encouraged by European and North American regulatory authorities. There are two priorities here:

• facilitate the free circulation of packaged foods, packaging and their components (polymers, additives) on their respective large domestic markets;

• protect consumers from potential chemical contaminants in packaging.

In line with this, article 14 of the European "plastics" directive 2002/72/EC [16] introduces the possibility of demonstrating the conformity of plastic materials using predictive approaches. These same approaches can contribute to meeting the obligations imposed by the 2004/1935/EC framework directive [18] : obligation to trace materials in contact with food, obligation to carry out a risk assessment and a risk management decision for the introduction of a new substance.

This dossier reviews the mechanisms responsible for contamination of foodstuffs by contact materials, including packaging, and the means of predicting them qualitatively or quantitatively. The content is aimed upstream of the packaging sector, at – polymer producers, additive manufacturers and formulators – who want to improve product safety right from the design phase. It is also aimed upstream at food processors and manufacturers, who need to demonstrate the conformity of materials placed on the market.

Due to the disproportionate amount of knowledge shared between users of materials or substances and the manufacturers themselves, this dossier focuses on the development of recommendations, which enable the elaboration of robust contamination scenarios despite the non-exhaustiveness of available information (e.g. detailed formulation of packaging materials, physico-chemical properties of materials, quantities present...). Based on a molecular description of transfers, predictive models of transport coefficients are also introduced to extend predictive approaches of contamination risk to a large number of substances, materials and storage or shaping temperature conditions. Techniques for robust interval or probabilistic modeling of packaged food contamination and consumer exposure are detailed in the