Investigating dietary exposure to nanoplastics and microplastics

Plastics are an emerging environmental contaminant of concern and can enter human food and beverages in the agricultural and industrial production stages, as well as during domestic food preparation. Plastic particles in the micro (< 5 mm and ≥ 1 μm) and nano (<1 μm) size range represent an important human health risk due to potential for uptake and absorption into the body following consumption. Despite the growing body of research on microplastic (MP) and nanoplastic (NP) contamination in food over the past decade, significant knowledge gaps remain in the contamination of widely consumed food products, and quantitative assessments based on concentration measurements are limited, leading to considerable uncertainty in dietary exposure estimates. This thesis aimed to advance understanding of dietary exposure to these contaminants in the Australian population, and to assess the suitability of pyrolysis-gas chromatography-mass spectrometry (Py-GC-MS) to quantify this exposure. In Chapter 2 a literature review was conducted to critically examine existing literature estimating total annual dietary exposure to NP and MP, and to provide an overview of current data on NP and MP contamination of common food and beverage products from the major food groups. The literature review revealed that drinking water was the greatest contributor to NP and MP dietary exposure, and that estimates of total dietary exposure to NP and MP varied greatly, ranging from 11,440 to >29 billion particles, or 2.1 x 10-4 to 431 g/person/year. However, these ranges largely reflect limitations of the current data. Estimates are based on an amalgamation of selectively chosen studies using incomparable methodologies and variable quality control procedures and only cover a narrow range of food products that do not adequately represent the human diet. Mass concentration exposure estimates are derived from calculations that convert particle number to mass, but there is disagreement in the literature as to how to do this reliably. Estimates are therefore highly reliant on assumptions and extrapolations and are based on studies that do not represent a typical human diet and are subject to significant knowledge gaps. To obtain more reliable data, further studies are needed that investigate a broader range of commonly consumed foods using consistent laboratory methods, and that measure mass-based concentrations directly. Based on the knowledge gaps identified in Chapter 2, a study evaluating the utility of Py-GC-MS to measure mass-based concentrations of NP and MP contamination of an Australian food basket and to estimate dietary exposure for the Australian population was conducted as part of Chapter 3. This investigation found accelerated solvent extraction combined with Py-GC-MS did not achieve required sensitivity for detecting and quantifying NP and MP in food products due to insufficient sample size, background contamination, high method detection limits, and matrix interferences. However, size-fractionated filtration of an entire beverage sample and subsequent Py-GC-MS analysis of the whole filter was found to be a sensitive method and NP and MP exposure from drinking water, tea, coffee, beer and wine was subsequently estimated at 1.7 - 2.0 mg/person/year, with polypropylene being the major contributor. Beverages were found to be an important contributor to dietary exposure and were therefore investigated further in Chapters 4 and 5. Chapter 4 aimed to quantify the mass concentration of NP and MP in Australian carbonated beverages and investigate the extent to which the fingerprint of local water sources contributes to this contamination. Bottled carbonated beverages from two Australian states were analysed and compared to bottled water from the same location, and soda fountain carbonated beverages were compared to tap water from the same fast-food restaurant. Analysis by Py-GC-MS found no relationship between the water and corresponding carbonated beverages. NP and MP concentrations were greater in carbonated beverages, with nylon the most prevalent polymer in bottled drinks and polyethylene terephthalate in the soda fountain beverages. The potential for PET oligomers to interference with PET polymer analysis during Py-GC-MS was also demonstrated for the first time. Drinking water is one of the most significant contributors to dietary exposure to NP and MP, yet little is known about the extent of this exposure in the Australian population. The study in Chapter 5 provides the most comprehensive mass-based assessment of MP in Australian tap water to date. Samples were collected from faucets across Australia and from a single location across a two-week period. MP were detected in all the Australia-wide tap water samples, with substantial spatial and temporal heterogeneity observed, indicating ubiquity of exposure to the Australian population. Chapter 6 is the final chapter of this thesis, and it synthesises the previous chapters into key findings, implications and recommendations for future research into dietary exposure to NP and MP.