Development of a Novel Sensor for Flow-Through Microplastics Detection

Beckett Colson and Anna Michel, Woods Hole Oceanographic Institution, Applied Ocean Physics and Engineering, Woods Hole, MA, United States
Consistent, objective, and quantitative methods are needed to understand the fate and effects of microplastics in aquatic environments. Manual identification of individual particles filtered from water samples is the predominant method, which is time-consuming and can be subject to confirmation bias and lack of inter-lab comparability. We present a novel technique for real-time flow-through microplastic quantification via impedance spectroscopy. Impedance spectroscopy is used to infer the dielectric properties of particles, which differ between inorganic and organic materials. In the biomedical field, impedance spectroscopy flow cytometry has been well demonstrated to distinguish between living red blood cells, dead cells and plastic spheres. In a flow-through system, parallel plate electrodes are used to monitor the change in electrical impedance as particles pass. Particle size can be inferred from the change at low frequencies (10-100 kHz) and particle type can be inferred from higher frequencies (>100kHz). Differential electrodes allow the technique to be robust to significant variability in water conductivity, temperature, and particle concentration. Lab results indicate successful differentiation of biological material from plastic particles of a variety of sizes. This technology has the potential to be developed into a hand-held, real-time, and low cost (<$500) microplastic sensor. This instrument will produce quantitative measurements of microplastic particle count, size distribution, and plastic to biological material ratio. Example applications for this instrument include: broadening sampling area coverage via shipboard flow-through installation, supporting mitigation efforts by monitoring wastewater plant effluent, and expanding in situ measurements to areas where collecting and filtering water samples is infeasible.