Polyethylene terephthalate recycling for food-contact applications: testing, safety and technologies: a global perspective

Scientific Guidance on the criteria for the evaluation and on the preparation of applications for the safety assessment of post-consumer mechanical PET recycling processes intended to be used for manufacture of materials and articles in contact with food

In the context of entry into force of Regulation (EU) 2022/1616, EFSA updated the scientific guidance to assist applicants in the preparation of applications for the authorisation or for the modification of an existing authorisation of a 'post-consumer mechanical PET' recycling process (as defined in Annex I of Regulation (EU) 2022/1616) intended to be used for manufacturing materials and articles intended to come into contact with food. This Guidance describes the evaluation criteria and the scientific evaluation approach that EFSA will apply to assess the decontamination capability of recycling processes, as well as the information required to be included in an application dossier. The principle of the scientific evaluation approach is to apply the decontamination efficiency of a recycling process, obtained from a challenge test with surrogate contaminants, to a reference contamination level for post-consumer PET, set at 3 mg/kg PET for a contaminant resulting from possible misuse. The resulting residual concentration of each surrogate in recycled PET is then compared to a modelled concentration in PET that is calculated using generally recognised conservative migration models, such that the related migration does not give rise to a dietary exposure exceeding 0.0025 μg/kg body weight (bw) per day. This is the lowest threshold for toxicological concern (TTC) value, i.e. for potential genotoxicity, below which the risk to human health would be negligible. The information to be provided in the applications relates to: the recycling process (i.e. collection and pre-processing of the input, decontamination process, post-processing and intended use); the determination of the decontamination efficiency by the challenge test; the self-evaluation of the recycling process. On the basis of the submitted data, EFSA will assess the safety of the mechanical PET recycling process.

Evaluation of chemicals leached from PET and recycled PET containers into beverages

The use of recycled polyethylene terephthalate (rPET) containers, a recent shift in the beverage industry, poses new potential human health concerns including contamination from the original container; use of additives, detergents, and catalysts during recycling; and improper recycling practices. The purpose of this analysis was to evaluate available data regarding: (1) chemicals leached from PET and rPET in bottle form; (2) concentration of these chemicals; and (3) trends between rPET percent and concentration of chemicals leached. This analysis identified 211 scientific articles related to recycled plastic and leachables. Three articles met the inclusion criteria: (1) plastic was in bottle form; (2) plastic was made of PET or rPET; and (3) the study analyzed both PET and rPET using the same methods. This evaluation demonstrated that only nine compounds - benzene, styrene, acetaldehyde, 2-methyl-1,3-dioxolane, furan, bisphenol A (BPA), 2-buta-none, acetone, and limonene - have been studied. Notably, the leachable concentration of benzene, styrene, and BPA increased as the percent of recycled content increased from 0 to 100%. However, 2-methyl-1,3-dioxolane and furan implied a reverse trend, where the leachable concentration decreased as the percent of recycled content increased from 0 to 100%. The concentrations of 2-butanone, acetone, and limonene did not follow any suggested trend. Evidently, recycling PET can lead to changes in the leachables profile. This analysis further identified key areas of research, including testing a variety of liquid types, that need to be addressed to adequately conduct a human health risk assessment.

Comprehensive environmental performance of bottle-to-bottle recycling of PET bottles based on deposit-refund system in China

Polyethylene terephthalate (PET) is a widely used packaging material and has high value in recycling. However, under China's dominant informal recycling system, most PET bottles are downcycled into fibers. The deposit-refund system (DRS) is considered a feasible mechanism to facilitate the high-value recycling of PET bottles. To comparatively evaluate the environmental performance [reduction of greenhouse gas (GHG) and pollutant emissions] under different scenarios using life cycle assessments, including the current system based on informal recycling, an improved system with a larger contribution from the source separation of municipal solid waste, and evolving systems with DRS application, five scenarios were set up. The DRS can reduce GHG emissions and the comprehensive environmental impact by 0.538 kg CO2 /kg PET bottles and 1.73 × 10-3 PE/kg PET bottles, respectively, compared to informal recycling. It can be concluded that the DRS-based recycling approach and the bottle-to-bottle recycling provide the substantial emission reduction potential of GHGs and pollutants.

Occurrence of volatile contaminants in recycled poly(ethylene terephthalate) by HS-SPME-GC×GC-QTOF-MS combined with chemometrics for authenticity assessment of geographical recycling regions

A comparison was performed on various methods detecting the volatile contaminants (VCs) in recycled poly(ethylene terephthalate) (rPET) flakes, the results demonstrated that head-space solid phase micro-extraction combined with comprehensive two-dimensional gas chromatograph-tandem quadrupole-time-of-flight mass spectrometry (HS-SPME-GC×GC-QTOF-MS) was a sensitive, effective, accurate method, and successfully applied to analyze 57 rPET flakes collected from different recycling plants in China. A total of 212 VCs were tentatively identified, and the possible source were associated with plastic, food, and cosmetics. 45 VCs are classified as high-priority compounds with toxicity level IV or V and may pose a risk to human health. Combined chemometrics for further analysis revealed that significant differences among these three geographical recycling regions. 6, 7, and 6 volatile markers were chosen based on VIP values and S-plot among plant1 plant 2 and plant 3, respectively. The markers differed significantly between recycled rPET samples in three geographical recycling regions based on chemometrics analysis. The initial classification rate and cross-validation accuracy were 100% on the identified VCs. These significant differences demonstrate that a systematic study is needed to obtain a comprehensive data on the contamination of rPET for food contact applications in China.

Determination of potential volatile compounds in polyethylene terephthalate (PET) bottles and their short- and long-term migration into food simulants and soft drink

Head space (HS)-GC-MS was used to analyze possible migration of volatile compounds from polyethylene terephthalate (PET) bottles for soft drinks, and a total of six compounds were identified. Next, a rapid, simple, and accurate simultaneous method was established using purge-and-trap (PT)-GC-MS, to quantify their amounts in the liquid contents after short- and long-term storage in PET bottles. Starting with brand-new PET bottles, the maximum migration of 2-methyl-1,3-dioxolane into distilled water and 50 % aqueous ethanol after 2 years at 25 °C were 2.3 and 19 ng/mL, respectively. In commercially available bottled mineral water sold inside and outside Japan, we were able to detect 2-methyl-1,3-dioxolane in the same way. While nonanal was also detected in some products, 2-methyl-1,3-dioxolane was confirmed as the main volatile compound. Finally, the human exposure to 2-methyl-1,3-dioxolane was estimated based on the per capita intake of soft drinks in Japan and the migration amount in this study.

Occurrence, potential release and health risks of heavy metals in popular take-out food containers from China

Global consumption of take-out food increased rapidly and the chemicals in their containers become a potential source of human exposure. However, available information on heavy metals in the containers is extremely limited and the associated health risks remain poorly understood. We investigated Cd, Cr, Pb, Sb, Mn, Ni and Co in the popular take-out food containers from China and found the concentrations of these metals were moderate in comparison to the concentrations reported in other food contact materials. The metal concentrations in sampled containers by material type differed significantly, and higher concentrations of Cd, Pb, Sb, Mn and Co were observed in expanded polystyrene samples. The metals in the containers likely originated from intentional addition and/or contamination of materials. The potential release of the metals from containers was simulated and found the median leaching rates of Cd, Pb, Sb, Ni and Co in the range of 0.36-4.80% under typical conditions, which depended largely on the material types. Based on the observed leaching rates, we estimated that the summed carcinogenic risks of Cd, Pb, Ni and Co were unacceptable under specific exposure frequency, although the total non-carcinogenic risks from metal intake were low.

Identification of Potential Migrants in Polyethylene Terephthalate Samples of Ecuadorian Market

Polyethylene terephthalate (PET) is the plastic packaging material most widely used to produce bottles intended for contact with food and beverages. However, PET is not inert, and therefore, some chemical compounds present in PET could migrate to food or beverages in contact, leading to safety issues. To evaluate the safety of PET samples, the identification of potential migrants is required. In this work, eight PET samples obtained from the Ecuadorian market at different phases of processing were studied using a well-known methodology based on a solvent extraction followed by gas chromatography–mass spectrometry analysis and overall migration test. Several chemical compounds were identified and categorized as lubricants (carboxylic acids with chain length of C12 to C18), plasticizers (triethyl phosphate, diethyl phthalate), thermal degradation products (p-xylene, benzaldehyde, benzoic acid), antioxidant degradation products (from Irgafos 168 and Irganox), and recycling indicator compounds (limonene, benzophenone, alkanes, and aldehydes). Additionally, overall migration experiments were performed in PET bottles, resulting in values lower than the overall migration limit (10 mg/dm²); however, the presence of some compounds identified in the samples could be related to contamination during manufacturing or to the use of recycled PET-contaminated flakes. In this context, the results obtained in this study could be of great significance to the safety evaluation of PET samples in Ecuador and would allow analyzing the PET recycling processes and avoiding contamination by PET flakes from nonfood containers.

Exploratory analysis of the presence of 14 carbonyl compounds in bottled mineral water in polyethylene terephthalate (PET) containers

Carbonyl compounds (CCs) can migrate from bottles to mineral water because of plastic degradation. An exploratory analysis of the presence a significant number of CCs (14) in bottled mineral water with and without gas in polyethylene terephthalate (PET) containers was performed using ultra-fast liquid chromatography coupled to mass spectrometry (UFLC-MS). The data from the analysis was submitted to chemometric treatment (principal component analysis, PCA). Formaldehyde, acetaldehyde, and benzaldehyde were found in all samples (0.07-125 ng mL^-1). Acrolein and acetone were present in 81% and 75% of the samples, respectively. The concentration of acrolein in carbonated water was up to 3.8 times greater than that measured in non-carbonated water (0.07-0.44 ± 0.01 ng mL^-1). PCA analysis showed that gasification can influence the composition of CCs present in mineral water and that the plastic material of the bottles is a likely source of CCs. In addition, benzaldehyde levels may be associated with the use of recycled materials.

Contamination Levels in Recollected PET Bottles from Non-Food Applications and their Impact on the Safety of Recycled PET for Food Contact

PET beverage bottles have been recycled and safely reprocessed into new food contact packaging applications for over two decades. During recollection of post-consumer PET beverage bottles, PET containers from non-food products are inevitably co-collected and thereby enter the PET recycling feed stream. To explore the impact of this mixing on the safety-in-use of recycled PET (rPET) bottles, we determined the concentrations of post-consumer substances in PET containers used for a range of non-food product applications taken from the market. Based on the chemical nature and amounts of these post-consumer substances, we evaluated their potential carry-over into beverages filled in rPET bottles starting from different fractions of non-food PET in the recollection systems and taking worst-case cleaning efficiencies of super-clean recycling processes into account. On the basis of the Threshold of Toxicological Concern (TTC) concept and Cramer classification tools, we present a risk assessment for potential exposure of the consumer to the identified contaminants as well as unidentified, potentially genotoxic substances in beverages. As a result, a fraction of 5% non-food PET in the recycling feed stream, which is very likely to occur in the usual recollection systems, does not pose any risk to the consumer. Our data show that fractions of up to 20%, which may sporadically be contained in certain, local recollection systems, would also not raise a safety concern.

Development of Sustainable and Cost-Competitive Injection-Molded Pieces of Partially Bio-Based Polyethylene Terephthalate through the Valorization of Cotton Textile Waste

This study presents the valorization of cotton waste from the textile industry for the development of sustainable and cost-competitive biopolymer composites. The as-received linter of recycled cotton was first chopped to obtain short fibers, called recycled cotton fibers (RCFs), which were thereafter melt-compounded in a twin-screw extruder with partially bio-based polyethylene terephthalate (bio-PET) and shaped into pieces by injection molding. It was observed that the incorporation of RCF, in the 1⁻10 wt% range, successfully increased rigidity and hardness of bio-PET. However, particularly at the highest fiber contents, the ductility and toughness of the pieces were considerably impaired due to the poor interfacial adhesion of the fibers to the biopolyester matrix. Interestingly, RCF acted as an effective nucleating agent for the bio-PET crystallization and it also increased thermal resistance. In addition, the overall dimensional stability of the pieces was improved as a function of the fiber loading. Therefore, bio-PET pieces containing 3⁻5 wt% RCF presented very balanced properties in terms of mechanical strength, toughness, and thermal resistance. The resultant biopolymer composite pieces can be of interest in rigid food packaging and related applications, contributing positively to the optimization of the integrated biorefinery system design and also to the valorization of textile wastes.

Determination of non-volatile components of a biodegradable food packaging material based on polyester and polylactic acid (PLA) and its migration to food simulants

Bioplastic materials are increasingly used due to its benefits for the environment preservation. Among them, food packaging materials based on polylactic acid (PLA) are among the most employed. In this work, a sample treatment methodology based on dissolution/precipitation has been optimized, selecting finally dichloromethane/ethanol as solvent/antisolvent system. The extracts obtained were analysed by UPLC-MS(QTOF), that allowed the identification of the main PLA non-volatile components. The recovery results were between 100.9 to 114.0%. The methodology was applied to the analysis of pellets and films of a PLA-polyester blend sample. A total of 37 different compounds were detected, where the four compounds with the highest intensity in pellet samples were cyclic oligomers coming from the polyester part of the blend and composed by adipic acid (AA), phthalic acid (PA) and butanediol (BD). Migration experiments to 3 food simulants were also performed: ethanol 95% (v/v), ethanol 10% (v/v) and acetic acid 3% (w/v). The results showed that in addition to those compounds previously detected in the film, new compounds coming from the reaction of PLA components with food simulants were present in migration solutions.

Sensor array for the detection of organic and inorganic contaminants in post-consumer recycled plastics for food contact

Post-consumer recycled (PCR) plastic material is made by collecting used plastic products (e.g., bottles and other plastic packaging materials) and reprocessing them into solid-state pellets or flakes. Plastic recycling has positive environmental benefits, but may also carry potential drawbacks due to unwanted organic and inorganic contaminants. These contaminants can migrate into food packaging made from these recycled plastic materials. The purpose of this research was to identify economically viable real-time monitoring technologies that can be used during the conversion of virgin and recycled resin feedstocks (i.e., various blends of virgin pellets and recycled solid-state pellet or mechanically ground flake) to final articles to ensure the safety, quality and sustainability of packaging feedstocks. Baseline analysis (validation) of real-time technologies was conducted using industry-standard practices for polymer analysis. The data yielded supervised predictive models developed by training sessions completed in a controlled laboratory setting. This technology can be employed to evaluate compliance and aid converters in commodity sourcing of resin without exceeding regulatory thresholds. Furthermore, this technology allowed for real-time decision and diversion strategies during the conversion of resin and flake to final articles or products to minimise the negative impact on human health and environmental exposure.

Predicting diffusion coefficients of chemicals in and through packaging materials

Most of the physicochemical properties in polymers such as activity and partition coefficients, diffusion coefficients, and their activation with temperature are accessible to direct calculations from first principles. Such predictions are particularly relevant for food packaging as they can be used (1) to demonstrate the compliance or safety of numerous polymer materials and of their constitutive substances (e.g. additives, residues…), when they are used: as containers, coatings, sealants, gaskets, printing inks, etc. (2) or to predict the indirect contamination of food by pollutants (e.g. from recycled polymers, storage ambiance…) (3) or to assess the plasticization of materials in contact by food constituents (e.g. fat matter, aroma…). This review article summarizes the classical and last mechanistic descriptions of diffusion in polymers and discusses the reliability of semi-empirical approaches used for compliance testing both in EU and US. It is concluded that simulation of diffusion in or through polymers is not limited to worst-case assumptions but could also be applied to real cases for risk assessment, designing packaging with low leaching risk or to synthesize plastic additives with low diffusion rates.

Decontamination efficiency of a new post-consumer poly(ethylene terephthalate) (PET) recycling concept

The aim of the study was to investigate and evaluate the cleaning efficiency of a new recycling concept for post-consumer poly(ethylene terephthalate) (PET). The so-called Flake To Resin (FTR) recycling process produces PET pellets or preforms from conventionally recycled PET flakes for the application in new PET packaging in direct food contact. The investigated process can be considered as 'super-clean' recycling process and was developed to introduce conventional recycled post-consumer (PCR) PET flakes up to an amount of 50% into the pellet and preform production. Within the study the cleaning efficiency of the investigated FTR process was determined by a challenge test. The experimental results obtained from three challenge tests with different input concentrations of the surrogates and different amounts of post-consumer PET flakes show that all applied surrogates are very efficiently removed by the investigated recycling process. The cleaning efficiencies for all surrogates are above 99.9%. In the final product of the process no surrogates could be determined above the detection limits (0.5 mg kg(-1)) even if the initial concentrations were in the percentage range. From a migrational point the final product, which are either PET pellets or preforms, was similar to virgin PET. Only the PET typical substances acetaldehyde and ethylene glycol had slightly higher concentrations than found in a conventional PET virgin sample used as reference. The generally accepted migration limit of 10 microg kg(-1) for the surrogates in the final products of the challenge tests is established for all kinds of foodstuffs.

UPLC–MS as a powerful technique for screening the nonvolatile contaminants in recycled PET

The possibility of using recycled polyethylene terephthalate as a food contact material is being seriously considered, but the potential migration of nonvolatile compounds from it must be assessed to ensure that it is safe to do so. In the study presented here, four samples of recycled PET were each exposed to three food simulants under the harsh extraction conditions stipulated by European legislation regarding migration tests. The nonvolatile compounds that migrated from them were determined by ultra performance liquid chromatography-mass spectrometry using three different cone voltages, and both positive and negative ionization modes. A total of 36 chemical compounds were detected, some of which were identified, including common additives such as N,N'-di-beta-naphthyl-p-phenylenediamine (antioxidant) and 2,4-di-tert-butyl-6-(5-chloro-2H-benzotriazol-2-yl)phenol (light stabilizer) as well as degradation compounds such as ethylene terephthalate dimers and trimers. In addition, specific migration values of three common components of polyethylene terephthalate (diethylene glycol, terephthalic acid, and isophthalic acid) were determined and found to occur at levels of <1 mg/kg-much lower than the specific migration limits stipulated by European legislation.

Migration modelling from food-contact plastics into foodstuffs as a new tool for consumer exposure estimation

One important aspect within the European Union's public healthcare is the exposure of consumers to undesirable chemicals in the diet. Food-contact materials (FCM) are one potential contamination source and therefore of particular interest for food exposure assessment. On the other hand, scientific investigations concerning the migration potential and behaviour of food-packaging materials have demonstrated that diffusion in and migration from FCM are foreseeable physical and, in principle, mathematically describable processes. Because of this situation and the current state-of-the-art in migration science, a research project was initiated within the 5th Framework Programme of the European Commission. This project, with the acronym 'FOODMIGROSURE' (European Union Contract No. 'QLK1-CT2002-2390') started on 1 March 2003, was due to last 3 years and had the participation of nine European project partners (see the project website: www.foodmigrosure.org). The aim of the project was to extend currently existing migration models (which have been demonstrated to be applicable for less complex matrices such as food simulants) to foodstuffs themselves. In this way, the project aims to provide a novel and economic tool for estimation of consumer exposure to chemicals migrating from food-contact plastic materials under any actual contact conditions. In addition, the project aims to increase knowledge of the mechanisms of diffusion of organic compounds in foodstuffs and provide data on the partitioning effects between FCM and foods. Today the latter aspect is increasingly regarded as a fundamental influence parameter for migration into foods. Based on the project achievements, a much better scientific basis is available to allow scientifically appropriate amendments of European Union Directive 85/572/EEC as well as to support further developments with the so-called Plastics Directive 2002/72/EC. The paper introduces the project and presents an overview of the project work progress.

Typical diffusion behaviour in packaging polymers – application to functional barriers

When plastics are collected for recycling, possibly contaminated articles might be recycled into food packaging, and thus the contaminants might subsequently migrate into the food. Multilayer functional barriers may be used to delay and to reduce such migration. The contribution of the work reported here is to establish reference values (at 40 degrees C) of diffusion coefficients and of activation energies to predict the functional barrier efficiency of a broad range of polymers (polyolefins, polystyrene, polyamide, PVC, PET, PVDC, [ethylene vinyl alcohol copolymer], polyacrylonitrile and [ethylene vinyl acetate copolymer]). Diffusion coefficients (D) and activation energies (Ea) were measured and were compiled together with literature data. This allowed identification of new trends for the log D=f(molecular weight) relationships. The slopes were a function of the barrier efficiency of the polymer and temperature. The apparent activation energy of diffusion displayed two domains of variation with molecular weight (M). For low M (gases), there was little variation of Ea. Focusing on larger molecules, high barrier polymers displayed a larger dependence of Ea with M. The apparent activation energy decreased with T. These results suggest a discontinuity between rubbery and glassy polymers.

Study of barrier properties and chemical resistance of recycled PET coated with amorphous carbon through a plasma enhanced chemical vapour deposition (PECVD) process

Many studies have been carried out in order to make bottle-to-bottle recycling feasible. The problem is that residual contaminants in recycled plastic intended for food packaging could be a risk to public health. One option is to use a layer of virgin material, named functional barrier, which prevents the contaminants migration process. This paper shows the feasibility of using polyethylene terephthalate (PET) recycled for food packaging employing a functional barrier made from hydrogen amorphous carbon film deposited by Plasma Enhanced Chemical Vapour Deposition (PECVD) process. PET samples were deliberately contaminated with a series of surrogates using a FDA protocol. After that, PET samples were coated with approximately 600 and 1200 Angstrons thickness of amorphous carbon film. Then, the migration tests using as food simulants: water, 10% ethanol, 3% acetic acid, and isooctane were applied to the sample in order to check the chemical resistance of the new coated material. After the tests, the liquid extracts were analysed using a solid-phase microextraction device (SPME) coupled to GC-MS.

Volatile and non-volatile radiolysis products in irradiated multilayer coextruded food-packaging films containing a buried layer of recycled low-density polyethylene

The effects of gamma-irradiation (5-60 kGy) on radiolysis products and sensory changes of experimental five-layer food-packaging films were determined. Films contained a middle buried layer of recycled low-density polyethylene (LDPE) comprising 25-50% by weight (bw) of the multilayer structure. Respective films containing 100% virgin LDPE as the buried layer were used as controls. Under realistic polymer/food simulant contact conditions during irradiation, a large number of primary and secondary radiolysis products (hydrocarbons, aldehydes, ketones, alcohols, carboxylic acids) were produced. These compounds were detected in the food simulant after contact with all films tested, even at the lower absorbed doses of 5 and 10 kGy (approved doses for food preservation). The type and concentration of radiolysis products increased progressively with increasing dose. Generally, there were no significant differences in radiolysis products between samples containing a buried layer of recycled LDPE and those containing virgin LDPE (all absorbed doses), indicating the good barrier properties of external virgin polymer layers. Volatile and non-volatile compounds produced during irradiation affected the sensory properties of potable water after contact with packaging films. Taste transfer to water was observed mainly at higher doses and was more noticeable for multilayer structures containing recycled LDPE, even though differences were slight.

Identification of chemicals, possibly originating from misuse of refillable PET bottles, responsible for consumer complaints about off-odours in water and soft drinks

Mineral water and soft drinks with a perceptible off-odour were analysed to identify contaminants originating from previous misuse of the refillable polyethylene terephthalate (PET) bottle. Consumers detected the off-odour after opening the bottle and duly returned it with the remaining content to the producers. The contaminants in question had thus been undetected by the in-line detection devices (so-called 'sniffers') that are supposed to reject misused bottles. GC-MS analysis was carried out on the headspace of 31 returned products and their corresponding reference products, and chromatograms were compared to find the possible off-odour compounds. Substances believed to be responsible for the organoleptic change were 2-methoxynaphthalene (10 bottles), dimethyl disulfide (4), anethole (3), petroleum products (4), ethanol with isoamyl alcohol (1) and a series of ethers (1). The mouldy/musty odour (5 bottles) was caused by trichloroanisole in one instance. In some cases, the origins of the off-odours are believed to be previous consumer misuse of food products (liquorice-flavoured alcohol, home-made alcohol containing fusel oil) or non-food products (cleaning products, petroleum products, oral moist snuff and others). The results also apply to 1.5-litre recyclable PET bottles, since the nature and extent of consumer misuse can be expected to be similar for the two bottle types.

Analysis of Formaldehyde, Acetaldehyde and Oligomers in Recycled Polyethylene Terephthalate

Formaldehyde (FA), acetaldehyde (AA) and oligomers in recycled polyethylene terephthalate (PET) were analyzed by HPLC. All of the physically recycled PET contained detectable levels of FA, AA and oligomers, and the levels were almost the same as in used bottles. Most superclean-like and chemically recycled PET contained lower levels than new pellets. These compounds showed no decrease upon physical recycling, but showed a marked decrease upon superclean-like recycling. In PET sheets made using physically recycled PET, FA was decreased, though AA was increased by the sheeting process as same as new one. FA, AA and oligomers originated from PET resin and their levels in recycled products were almost equivalent to those in new products. It was concluded that there is no particular safety concern about their presence in recycled PET.

Migration of model contaminants from PET bottles: influence of temperature, food simulant and functional barrier

To simulate post-consumer recycled plastics, selected model contaminants were incorporated into PET bottles using a time saving method. Migration into 3% acetic acid, a cola-type beverage and 95% ethanol was followed during 1 year of storage at 20 and 40 degrees C. Aroma compounds previously found in post-consumer PET material were used as model contaminants. Benzaldehyde was found to migrate to the highest extent. Storage at 40 degrees C affected the bottle material and this might be one reason for the high migration values of these bottles. Migration into ethanol was up to 20 times higher than into 3% acetic acid or a cola-type beverage. Bottles with a functional barrier resisted migration into food simulants even when filled with 95% ethanol and stored for 1 year at 40 degrees C. Differential scanning calorimetry measurements showed that ethanol was interacting with the plastic material. This resulted in a lower glass transition temperature of bottles stored with ethanol compared with bottles stored empty or with other food simulants.

European survey on post-consumer poly(ethylene terephthalate) (PET) materials to determine contamination levels and maximum consumer exposure from food packages made from recycled PET

Typical contamination and the frequency of misuse of poly(ethylene terephthalate) (PET) bottles are crucial parameters in the risk assessment of post-consumer recycled (PCR) PET intended for bottle-to-bottle recycling for direct food contact applications. Owing to the fact that misuse of PET bottles is a rare event, sustainable knowledge about the average concentration of hazardous compounds in PCR PET is accessible only by the screening of large numbers of samples. In order to establish average levels of contaminants in PET source materials for recycling, PET flakes from commercial washing plants (689 samples), reprocessed pellets (38) and super-clean pellets (217) were collected from 12 European countries between 1997 and 2001. Analysis of these materials by headspace gas chromatography revealed average and maximum levels in PCR PET of 18.6 and 86.0 mg kg-1 for acetaldehyde and 2.9 and 20 mg kg-1 for limonene, respectively. Acetaldehyde and limonene are typical compounds derived from PET itself and from prior PET bottle contents (flavouring components), respectively. Maximum levels in PCR PET of real contaminants such as misuse chemicals like solvents ranged from 1.4 to 2.7 mg kg-1, and statistically were shown to result from 0.03 to 0.04% of recollected PET bottles that had been misused. Based on a principal component analysis of the experimental data, the impact of the recollecting system and the European Union Member State where the post-consumer PET bottles had been collected on the nature and extent of adventitious contaminants was not significant. Under consideration of the cleaning efficiency of super-clean processes as well as migration from the bottle wall into food, it can be concluded that the consumer will be exposed at maximum to levels < 50 ng total misuse chemicals day-1. Therefore, PCR PET materials and articles produced by modern superclean technologies can be considered to be safe in direct food applications in the same way as virgin food-grade PET.

Contaminants and levels of occurrence in washed and shredded poly(ethylene terephthalate) from curbside collection. II: Validation of extraction procedures, particle size sampling and crystallinity

Part I of this study showed that washed and dried, shredded poly(ethylene terephthalate) (PET) (flake) obtained from curbside collection when Soxhlet extracted contained 26 semivolatile contaminants below the US FDA threshold of 215 ppb and six above this level. This paper reports the validation of the Soxhlet extraction technique by comparison with total dissolution with trifluoroacetic acid (TFA). The work was carried out for two of the three particle size ranges obtained by grinding the PET flake (300-425 and 425-700 microm) and for the unground flake. Further validation was undertaken by comparison of contaminant levels determined by total dissolution with TFA and sonication with dichloromethane (DCM) using flake ground to the 0-300 microm size range. The levels of contaminants increased with decreasing particle size range, but X-ray diffraction measurements of degrees of crystallinity were similar for each PET particle size range, thus showing that the differences in contaminant levels were not due to variable percentages of the amorphous material from the tops and bottoms of shredded bottles, relative to the amounts of crystalline PET from the mid-sections of the bottles. Hence, it was postulated that the variations in contaminant levels were due to selective grinding of the more highly contaminated surfaces, whilst the larger particles incorporated the less contaminated interior material. The grinding was also strongly selective with respect to the amorphous flake. Analysis of the segregated amorphous and crystalline flake phases indicated that many contaminants were similarly absorbed into both phases, whilst some were preferred by the amorphous PET and others were preferred by the crystalline PET.

Contaminants and levels of occurrence in washed and shredded poly(ethylene terephthalate) from curbside collection. Part 1: Extraction conditions

The aim was to determine which contaminants were present in washed and dried shredded poly(ethylene terephthalate) (PET, flake) obtained from curbside collection and whether the concentrations were above the US FDA threshold of 215 ppb. Thirty-two semi-volatile contaminants were extracted from the treated flake by Soxhlet extraction using dichloromethane as a PET swelling solvent and gas chromatography-mass spectrometry (GC-MS) was used for identification and quantification. Soxhlet extraction of flake ground to 0-300 pm was effectively completed in 24 h, whereas sonication reduced the extraction time to 3 h. In contrast, Soxhlet extractions of flake ground to a larger particle size range (300-425 pm) were completed in 4 h, possibly due to less aggregation in the extraction thimble. The levels of 26 contaminants were below 215 ppb, but six were not. Dodecanoic acid was present at about 1200 ppb, 2-butoxyethanol was approximately 1000 ppb, limonene, benzophenone and methyl salicylate were above 800 ppb, and 2-methyl-naphthalene near 215 ppb.

Evaluation of some screening methods for the analysis of contaminants in recycled polyethylene terephthalate flakes

A range of different analytical techniques were used to test recycled polyethylene terephthalate (PET) flakes for potential chemical contaminants. The techniques used were headspace gas chromatography coupled to mass spectrometry (GC-MS), liquid extraction followed by GC-MS, supercritical-fluid extraction followed by GC-MS, and migration testing followed by elemental analysis using inductively coupled plasma mass spectroscopy (ICP-MS). The PET samples were 50 representative samples taken from 600 that had been collected throughout Europe and which had been screened for potential contaminants using a single technique of high-temperature GC-MS. Six of the 50 samples tested had been spiked with a selection of model contaminants, three samples were virgin PET flakes and two of the samples were from supercleaning processes. All samples were analysed 'blind' in this exercise. The qualitative results showed that most of the contaminants came from the first use, being flavour-aroma compounds from soft drinks. The quantitative analysis found concentrations under a few mg kg(-1) in the polymer, except for the spiked samples. Element migrations were low and only calcium, silicon and sodium had median migrations above 50 microg l(-1). This in-depth analysis of recycled PET flakes did not identify any significant contaminants that had not already been detected by the high-temperature static GC-MS screening method, thus demonstrating its utility.