Rapid discrimination of recycled and virgin poly(ethylene terephthalate) based on non-targeted screening of semi-volatile organic compounds using a novel method of DSI/GCxGC-Q-TOF-MS coupled with various chemometrics

Increasing interest has been given to the use of recycled poly(ethylene terephthalate) (PET) as food contact materials. However, the use of recycled PET without decontamination in food packages causes serious safety hazards for consumers, thus representing a major limit hurdling the recycling and reuse of PET in many coun-tries. Herein, a feasible study was employed for the discrimination of 105 total types of virgin and recycled PET samples based on 267 semi-volatile organic compounds (SVOCs) tentatively identified by direct sample intro-duction/comprehensive two-dimensional gas chromatography and quadrupole time-of-flight mass spectrometry (DSI/GCxGC-Q-TOF-MS) coupled with various chemometrics. 100% prediction accuracy was accomplished using orthogonal partial least squares discrimination analysis (OPLS-DA), and principal component analysis (PCA) combined with nonparametric tests. Cross-discrimination based on misclassified samples with various chemometrics enabled identifying large sample sizes and improving prediction accuracy. The possible origins of the markers containing 46 substances could be food, medicine, cosmetics, pesticides, fuel, plastic and industrial sources, where most of them are highly toxic chemicals, indicative of the urgent requirement of employing closed-loop recycling and sorting of plastic wastes. More importantly, our strategy could also act as a basis for discriminating against other plastics, making it promising as a novel, efficient and versatile approach for sorting of plastic wastes. Finally, this report also demonstrates the combination of various chemometrics and chroma-tography tool is reliable and powerful to effectively address PET adulteration, thereby facilitating food contact uses of rPET bottles, which is also beneficial for the health and safety of consumers.

Automated summary

This study successfully discriminated different types of PET samples and identified 267 substances using chemometrics and chromatography tools. The cross-discrimination method helped identify large sample sizes and improve prediction accuracy. The markers in the samples came from food, medicine, cosmetics, pesticides, fuel, plastic, and industrial sources, containing highly toxic chemicals, suggesting the need for closed-loop recycling and sorting of plastic wastes.