A scientific, peer-reviewed article published online on January 29, 2013 in Food Chemistry analyzed migration and toxicity of PET and glass packaging materials (Bach et al. 2013). Starting with a list of 17 organic substances of interest, the authors first performed chemical analysis of bottled waters from two different French brands, stored for 10 days at 40, 50 or 60°C. Each water brand was packaged in PET or glass, and one brand was carbonated, the other still water. Notably, phthalates were not identified in the bottled waters.

The authors identified the PET-specific migrants antimony, formaldehyde, acetaldehyde and bis(2-hydroxyethyl)terephthalate (CAS 959-26-2), an intermediate of PET production. Antimony concentrations  were less than the WHO drinking water guideline value (20 ppb) but at temperatures higher than 20°C above the Japanese recommendation for drinking water (2 ppb) for some of the samples. Levels of formaldehyde and acetaldehyde were below the EU specific migration limits, but formaldehyde exceeded the French quality standard for bottled water (5 ppb).

In addition, a substance with potential endocrine disrupting properties was identified in PET and glass bottled waters: 2,4-di-tert-butylphenol (CAS 96-76-4) has been shown to be anti-androgenic and is listed for further evaluation under REACH. The substance itself is not authorized in the EU for use as plastics additive, but it is assumed to be a break-down product (NIAS; non-intentionally added substance) of a commonly used polyolefine antioxidant, Irgafos 168. Its source in the bottled water is likely to be the plastic closure.

Furthermore, the authors tested extracts of the bottled waters for toxicological effects. No damage to DNA was observed, neither were any of the analyzed samples cytotoxic or estrogenic. In addition, samples were tested for their anti-androgenic properties: in competition with the natural male sex hormone sample extracts were assessed for their ability to block androgen action. All but one sample were found to be negative. The positive result was observed for an extract of glass bottled water. The authors speculate that the effect may be due to an interaction between the androgen receptor and unidentified compounds present in the water extract.


Bach, C., et al. “Effect of temperature on the release of intentionally and non-intentionally added substances from polyethylene terephthalate (PET) bottles into water: chemical analysis and potential toxicity.” Food Chemistry(online 29 January 2013).