ERI: A microfluidic approach for monitoring microplastics in seawater

As the global microplastic pollution intensifies and related health concerns grow, monitoring microplastics in seawater through effective sampling and identification has become imperative. Traditional methods of sampling are often time-consuming and ineffective at isolating tiny microplastics (<100 micrometers), which in fact constitute the majority of microplastics in the oceans. Moreover, while vibration spectroscopy has proven effective in microplastic identification, interpreting results through traditional library matching heavily relies on expert knowledge due to significant spectral variations caused by plastic degradation. Given the pressing need to enhance microplastic monitoring down to sizes as small as 1 micrometer in both sampling and identification aspects, this project aims to develop new methodologies for effectively monitoring small-sized microplastics in seawater in a label-free and reliable manner. The outcomes of this research have significant implications for environmental and ecological monitoring, including abundance assessments, source tracking, and analysis of microplastic degradation in the seas. These findings will also drive participation from researchers, engineers, and other stakeholders, heightening public awareness about these global issues. Ultimately, the data from microplastic monitoring will contribute to restoring the safety of our waters, allowing aquatic ecosystems to thrive again. To meet these goals, this project aims to develop a microfluidic microplastic monitoring tool that leverages both acoustofluidic particle manipulation and machine learning-assisted Raman spectral identification. The microfluidic device will provide a streamlined monitoring process capable of sampling, concentrating, and sorting microplastics in seawater down to 1 micrometer by adjusting the frequency and layout of surface acoustic waves, while also identifying seawater microplastics regardless of their degradation status through machine learning classification coupled with Raman spectroscopy. Specifically, this project will explore the use of both traveling and standing surface acoustic waves for sampling and sorting microplastics in microfluidic chips. The samples will include pristine microplastics, artificially degraded variants, and environmental microplastics, with a focus on those sourced from surface seawater in Rhode Island. A microfluidic particle trapping component will also be developed to trap microplastics at predetermined locations in a microfluidic chip, enhancing the reliability of Raman spectra acquisition by facilitating an automated process. Lastly, machine learning classification will be employed to interpret the Raman spectral data of seawater microplastics, even when they are agglomerated. The results from this project will contribute fundamental knowledge towards addressing the global challenge of plastic pollution, strengthening collaboration among academia, industry, and fisheries. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.