Authentication of recycled plastic content in water bottles using volatile fingerprint and chemometrics

The environment is threatened by the continuously increasing volume of plastic residue. Plastic recycling is an interesting alternative to mitigate this problem. However, recycled plastic products may have pollutants from their recycling process, collecting system and/or previous life which may hurt consumers health, thus making it key to authenticate and characterize recycled materials. An innovative non-targeted methodology by means of static headspace gas chromatography-mass spectrometry (SHS-GC-MS) has been developed to measure the volatile organic profile of virgin polyethylene terephthalate (PET) and with diverse content of recycled PET samples. A home-made MS database, with 161 organic compounds characteristics from plastic materials based on the literature, was made. Seventeen of those compounds were found in the studied samples and identified by matching their MS spectra with MS database libraries. These compounds are mainly aldehydes (pentanal, hexanal, heptanal, octanal, nonanal and decanal), and benzene derivatives (styrene, p-xylene, benzaldehyde, methylbenzene, and 1,2-dichlorobenzene) which we found to be the common in the samples of recycled PET. The combination of the dataset consisting in the peak area of the detected species by SHS-GC-MS and the use of chemometrics shown to be a valuable methodology for the discrimination between virgin PET samples and those with different recycled PET content based on their volatile profile. In addition, a novel strategy applying a statistical model based on partial least squares (PLS) regression was proposed, for the first time, to quantify the recycled plastic content in the PET samples.

Automated summary

The environment is threatened by increasing plastic residue, and plastic recycling is an interesting solution. A new method using SHS-GC-MS was developed to measure the volatile organic profile of virgin and recycled PET samples. The study identified common compounds in recycled PET samples and proposed a statistical model to quantify the recycled plastic content in PET samples for the first time.