Three studies published in February 2023 screened plastic flakes, pellets, or products used in recycling and finished recycled products for their chemical composition using targeted and non-targeted chemical analysis. They provided insight into the chemicals’ identity, the samples’ safety as well the strength and limitations of four analytical methods for chracterizing post-consumer recycled plastics.
In an article published on February 14, 2023, in the journal Science of the Total Environment, Leah Chibwe and co-authors from Environment Climate Change Canada, Ontario, Canada, performed targeted and non-targeted analyses to evaluate the chemicals present in 21 plastic flakes and pellets used for the production of recycled goods.
The authors obtained samples made of low-density polyethylene (LDPE), high-density polyethylene (HDPE), polypropylene (PP), polyethylene terephthalate (PET), and high impact polystyrene (HIPS) of different colors from Canadian recycling companies in Ontario and British Columbia. They extracted the chemicals with dichloromethane (DCM) and analyzed brominated/chlorinated flame retardants (Br/Cl-FRs) with gas chromatography negative ion mass spectrometry (GC-ECNI-MS) and organophosphorus ester flame retardants and plasticizers (OPEs) and perfluoroalkyl acids (PFAAs) with LC high-resolution Orbitrap mass spectrometry (LC-HRMS). For non-targeted analysis, they applied comprehensive two-dimensional gas chromatography MS (GC×GC-ToFMS). To obtain a multielement profile, inductively coupled plasma−mass spectrometry (ICP-MS) was used.
The authors, reported more than 280 compounds being present over all 21 pellets and flakes. Of the three groups analyzed in the targeted approach, OPEs were detected with the highest frequency (up to 100% of the samples) followed by Br/Cl-FRs (up to 76% of the samples). Concentrations reached up to 4.7 ng/g and 2.15 ng/g, respectively. OPEs commonly detected include 2-ethylhexyl diphenyl phosphate (EHDPP, CAS 1241-94-7, 3.850 ng/g), tributyl phosphate (TNBP, CAS 126-73-8, 620 ng/g,), and 4-tert-butylphenyl diphenyl phosphate (TBDPP, 280 ng/g). Previously, researchers have demonstrated the extensive contamination of foodstuff with OPEs (FPF reported) and uncovered plastic food packaging as a source of these OPEs (FPF reported). The Food Packaging Forum’s (FPF’s) October 2022 Fact Bite on OPEs illustrates the number of OPEs that have been detected in different food contact materials.
In their nontargeted analysis, Chibwe et al. identified 217 chemicals by using authentic standards (Level 1) or library matching (Level 2). 2-Hexyl hydroxy benzoate (CAS 302776-68-7) was present in all samples and also in the highest concentration (up to 1030 ng/g). Looking at more than 60 elements present in the pellets and flakes, the authors reported total concentrations between 0.0005 and 2.98 mg/kg with calcium, sodium, and iron in the highest concentrations. Neither the color nor the type of product (flake vs pellet) affected the concentrations of the analyzed compounds.
Overall the study “highlights that while recycling addresses sustainability goals, additional screening of goods and products made from recycled plastics is needed to fully document potentially hazardous chemicals and exposure.”
During the recycling process also non-intentionally added substances (NIAS) get into the product, such as reaction or degradation products of additives, or contaminants of material recollection. Food-contact plastic might also be cross-contaminated with chemicals from materials not approved for food contact. In an article published on February 9, 2023, in the journal Recycling, Christian Rung from the Fraunhofer Institute for Process Engineering and Packaging IVV, Freising, Germany, and co-authors evaluated post-consumer recyclates for their absorbed chemicals, with a focus on NIAS, and their toxicity.
They acquired 179 products previously used in food and non-food contact application made of PET, LDPE, HDPE, PP, or polystyrene (PS) from companies across Europe and extracted them with DCM. Subsequently, they applied extracts to headspace gas chromatography with flame ionization detection (GC-FID) and headspace gas chromatography with mass spectrometry (GC-MS) and compared the spectra with the NIST library to identify the compound. Furthermore, the scientists used Cramer classification and the Carcinogenicity ISS model (carcinogen vs. non-carcinogen) to assess the toxicity of the compounds.
The researchers detected 205 chemicals across the 179 extracts of which they identified 175. On average the lowest number of substances (8) were found in PET while 20 were detected in PE samples. Concerning chemical diversity they stated that “PET samples were more likely to contain the same substances.” Assessing the toxicity of identified substances showed that 51 had a Cramer class II or III with PS having the highest and PET having the lowest number of substances with this classification. Rung and co-authors reported polyolefins to “contain more substances classified as toxic than PET, potentially due to their higher diffusivity.” However, also in PS, they detected many chemicals although it has a much lower diffusivity than polyolefins. A comparison with virgin (non-recycled) polymers showed that the toxicity is higher for post-consumer products. Oxidized Irgafos 168 which is a degradation product of the processing stabilizer Irgafos 168 (CAS 31570-04-4), was the substance most frequently detected across the tested products.
The scientists concluded that “PET is currently the only polymer that complies with EFSA’s [European Food Safety Authority’s] requirements for a circular economy.” Furthermore, they emphasized that “research gaps must be closed, and new innovation must be developed” to enable a safe circular economy for polyolefins and PS. Here, they think that due to the high diffusivity of polyolefins, it might be difficult to reach reliable safety standards.
The new European regulation on recycled plastics in food contact allows non-authorized technologies on the market while necessary data is collected. This data shall be generated with suitable methods which are not further specified (FPF reported and here). Andrea Hochegger and co-authors from the University of Technology Graz, Austria, and the LECO European Application and Technology Center, Berlin, Germany, compared the quality of GC-based techniques for the analysis of recycled materials. Their article was published on February 23, 2023, in the journal Analytical and Bioanalytical Chemistry.
For that, the scientists purchased two buckets made of recycled polypropylene (PP) with no given information about restrictions in usage in local markets in Graz, cryo-milled and extracted them with cyclohexane for one hour at 60 °C ultrasound. Total extracts were applied to conventional GC-FID for a fast quantification and to GC with electron ionization mass spectrometry (HS-SPME-GC-MS) to identify the extractable chemicals using conventional mass spectrometry databases. They compared these methods with comprehensive two-dimensional GC including online-coupled high-performance liquid chromatography (HPLC)-GC-FID to quantify saturated and aromatic hydrocarbons and GC time of flight mass spectrometry (GC × GCTOF-MS) for a comprehensive chemical characterization of the samples. They described that the latter analytical method “has huge advantages over conventional GC, such as an increased peak capacity, higher chromatographic resolution, higher sensitivity, and structured chromatograms.” Generally, the researchers followed the approach proposed by Nerin et al. for the non-targeted screening of NIAS migrating from food contact materials (FPF reported).
The authors summarized the strengths and limitations of the four methods evaluated. Accordingly, they described GC-FID as a quick and easy method to determine the total amount of volatile compounds while online-coupled HPLC-GC-FID allows to further separate and quantify saturated and aromatic hydrocarbons. The limitation of FID is that it does not enable a “confident” identification of the chemicals which MS approaches allow for. HS-SPME-GC-MS is useful for the fast analysis of volatile compounds but has limited sample capacity. According to the researchers, GC × GCTOF-MS “is much more powerful” since it provides more comprehensive information such as a greater separation and detection of substances with chain lengths from C10 to C50. Overall, Hochegger and co-authors think “it is questionable, if a simple 1D GC–MS run, identifying the most prominent 20 peaks, is suitable to generate the required data” for the analysis of recycled materials. The application of their extract to Ames tests supported their concern since both samples showed DNA reactivity but only GC × GCTOF-MS, and not 1D GC, suggested the presence of substances of concern. Their analysis let the authors further emphasize the necessity “for detailed and comprehensive methods of analysis for post-consumer recycled plastics.”
References
Chibwe, L. et al. (2023). “Target and Nontarget Screening of Organic Chemicals and Metals in Recycled Plastic Materials.” Environmental Science & Technology. DOI: 10.1021/acs.est.2c07254
Hochegger, A. et al. (2023). “One-dimensional and comprehensive two-dimensional gas chromatographic approaches for the characterization of post-consumer recycled plastic materials.” Analytical and Bioanalytical Chemistry. DOI: 10.1021/acs.est.2c07254
Rung, C. et al. (2023). “Identification and Evaluation of (Non-)Intentionally Added Substances in Post-Consumer Recyclates and Their Toxicological Classification.” Recycling. DOI: 10.3390/recycling8010024
