An article published on March 22, 2017 in the peer-reviewed journal Packaging Technology and Science discusses the accuracy of migration testing performed for polyethylene terephthalate (PET) bottles used to store beverages. Carina Gehring and Frank Welle from the Fraunhofer Institute for Process Engineering and Packaging (IVV), Freising, Germany, compared the results of accelerated migration tests with real migration after one year of storage at room temperature.

Traditionally, migration tests covering long-term storage at or below room temperature have been performed for 10 days at 40°C. Newly, the European regulation on plastic food contact materials (FCMs) (EU) 10/2011 specifies that when real contact times are expected to be longer than 30 days, accelerated migration tests at elevated temperature should be applied. These are performed for a maximum of 10 days at 60°C. The exact conditions are calculated with the use of the Arrhenius equation on the basis of a default activation energy of ~80 kJ/mol.

However, as Gehring and Welle point out, in PET the activation energy is known to depend on the molecular size of the migrant. The activation energy of ~80 kJ/mol is characteristic only of small molecules like acetaldehyde (CAS 75-07-0), while for larger molecules like 2-aminobenzamide (CAS 88-88-6), the activation energy of diffusion is predicted to be ~134.4 kJ/mol. In this case, migration after one year at 23°C was calculated to be equivalent to 18.9 days at 40°C and only 0.89 days at 60°C. These calculations were largely confirmed previously, where a limited set of experiments was performed for 2-aminobenzamide (FPF reported), and in the current study, where the authors covered a broader range of potential migrants at several combinations of food simulants and migration conditions.

The authors concluded that for PET bottles, the accelerated migration test conditions of 10 days at 60°C “in most cases significantly overestimate real migration at the end of shelf life.” They suggested that for migrants such as PET monomers, which have a low concentration in PET and relatively high specific migration limits (SMLs), this is not critical, because they are usually not detected even in the accelerated migration tests. However, “for the migration testing of substances with a low SML that is easy to exceed, the choice of realistic migration test conditions is decisive,” the authors emphasized. For example, in the accelerated migration test 2-aminobenzamide’s SML of 50 µg/kg “will be exceeded for all commercially available bottle sizes” even in the non-swelling food simulants such as 3% acetic acid or 10% ethanol. In the simulants with higher ethanol content, the migration will be further increased due to polymer swelling. Therefore, the authors argue that “to test the specific migration of . . . critical substances for long-term applications, migrant-specific and polymer-specific diffusion parameters . . . should . . . be considered [instead of default values] when designing accelerated migration tests.”


Gehring, C., and Welle, F. (2017). “Migration testing of polyethylene terephthalate: Comparison of regulated test conditions with migration into real food at the end of shelf life.Packaging Technology and Science (published on March 22, 2017).