In a new peer-reviewed scientific study published on March 26, 2013 in the journal Food Chemistry, Spanish researchers report migration of antimony from polyethylene terephthalate (PET) plastic (Sánchez-Martínez et al. 2013). Using migration testing rules for plastics established in the EU’s Plastics Regulation (EU 10/2011), the scientist found levels of elemental antimony (Sb) in food simulants well below the EU’s specific migration limit of 40 µg/kg food, the European drinking water limit of 5 µg/L and the WHO’s drinking water standard of 20 µg/L (WHO 2003).

Migration into different types of food simulant (distilled water, 3% acetic acid, 10% ethanol and 20% ethanol) ranged from 0.56 to 1.3 µg/L. Highest levels were reported for 20% ethanol, a food simulant used to assess migration into foods with a low fat content, as well as for foods containing up to 20% alcohol (for example wines and beers). Repeated use migration was also assessed in the study, showing a reduction of migration below detection limit (0.003 µg/L) for all food simulants except 3% acetic acid. For the latter, migration markedly increased with repeated use to 5.7 µg/L, above the European drinking water limit. The authors speculate that the increased migration is likely due to acid-induced polymer degradation. This finding is particularly relevant for refillable PET bottles used for acidic beverages, though the tested PET was not intended for repeated use. In addition, citric acid may enhance Sb migration compared to acetic acid due to formation of stable complexes (Hansen et al. 2010). This issue was not addressed in the present study.

Antimony trioxide (Sb2O3) is used in the manufacture of PET as a catalyst. It has been shown to migrate from PET and migration is enhanced at higher temperatures and lower pH (corresponding to higher acidity). The toxicity of Sb is not fully established yet. Since Sb is chemically related to arsenic it is also suspected to be carcinogenic, with the trivalent form Sb(III) assumed to be more toxic than the oxidized Sb(V). Sb2O3 is a priority pollutant in Europe and the US. In the present study Sb(V) was the mainly observed migrant into food simulants. The authors state that “the often addressed question regarding toxic effects caused by antimony from PET bottles appears to be groundless”, based on the assumed lesser toxicity of Sb(V) compared to Sb(III).


Sánchez-Martínez, M., et al. “Migration of antimony from PET containers into regulated EU food simulants.” Food Chemistry (online 26 March 2013).

Hansen, C., et al. (2010). “Elevated antimony concentrations in commercial juices.” J Environ Monit12(4): 822-824.

Our opinion:

Further research focused on understanding Sb toxicity may be warranted before safety concerns related to Sb migration from PET can be disregarded. Further, the influence of citric acid on Sb migration should be studied to verify the suitability of 3% acetic acid as food simulant for fruit drinks and lemonades.