An article published on April 21, 2017 in the peer-reviewed journal Food Additives & Contaminants: Part A focused on the migration of oligomers from polyethylene terephthalate (PET). Maria Hoppe and colleagues from the Fraunhofer Institute for Process Engineering and Packaging IVV, Freising, Germany, first determined the diffusion coefficients for several oligomers experimentally, and then used their data to verify the two existing migration models.

Diffusion coefficients were estimated from the data obtained in a kinetic migration study where PET bottle material was submerged in 50% ethanol at 80°C, and oligomer release was measured every hour for 15 hours. Such severe test conditions were chosen in order to obtain “reliably measurable migration values” that could be properly compared with migration models, despite the low migration rates expected for PET oligomers due to their “relatively high molecular weight and the low basic diffusivity in the PET polymer.”

The obtained results allowed calculating diffusion coefficients for five PET oligomers (from the 11 detected). These included first series dimer and trimer, second series dimer and trimer, and third series dimer. First series oligomers comprise an equal number of terephthalic acid and ethylene glycol units. In second series oligomers, one monoethylene glycol unit is replaced by a diethylene glycol unit, while third series oligomers carry two such replacements. Previously, experimentally determined diffusion characteristics were available only for the first series trimer.

Two migration models were compared, the one by Piringer and the one by Welle. The Piringer model assigns a default activation energy parameter for all possible migrants in a given polymer (i.e. 100 kJ/mol for PET). This model tends to underestimate migration for substances with the molecular weight lower than 80-90 Da, and overestimate it for larger molecules. Welle’s approach uses a correlation between molecular volume and activation energy to estimate the activation energy for each migrant. This model underestimated the oligomer migration into ethanolic simulants at elevated temperatures, but “as long as swelling effects due to the presence of ethanol can be excluded, the Welle equation appears to be applicable,” the authors concluded. They further emphasized that “the type and nature of food should be carefully considered” when applying the Welle model for prediction of PET oligomers migration into foods.

As a next step, the authors suggested studying the migration of polyester oligomers into real foods in order to “explore the precise applicability of the Welle equation.” The results of this study may be relevant not only for PET but also for other polyester related oligomers.


Hoppe, M., et al. (2017). “Migration of oligomers from PET: determination of diffusion coefficients and comparison of experimental versus modelled migration.Food Additives & Contaminants: Part A (published April 21, 2017).

Frank Welle (2013). “A new method for the prediction of diffusion coefficients in poly(ethylene terephthalate).Journal of Applied Polymer Science 129:1845-1851.

Piringer, O., et al. (1998). “Migration from food packaging containing a functional barrier: Mathematical and experimental evaluation.Journal of Agricultural and Food Chemistry 46:1532-1538.