Integrated approach of migration prediction using numerical modelling associated to experimental determination of key parameters.

The principle of utilizing a computing program describing precisely the migration of additives from a polymer into a food simulant is presented. The model has been validated with a UV absorber in polypropylene migrating into glyceryl tripelargonate, a pure triglyceride whose behaviour and average molecular weight are similar to Myglyol (a synthetic mixture of C8-C12 triglycerides). Six parameters were used to fit the simulant sorption and additive extraction kinetics, and these were determined by independent experiments. The possibility of eliminating any of the parameters is also discussed. This work provides the first consistent set of experimental data that can be used to overestimate the diffusion coefficients of additives both in virgin (without contact with solvent) and in swollen (fat contact) polymer. The influence of mobility increase brought out by temperature or swelling are compared. The effects were more important for high molecular weight compounds.