Novel characterisation of microplastics and other contaminant particles using new scanning electron microscopy technologies

Scanning electron microscopy with combined backscattered electron and X-ray imaging (SEM-BEX) represents a new way to conduct direct visual assessment and elemental characterisation of particles in environmental samples up to 18× faster than standard scanning electron microscopy techniques. SEM-BEX provides all element maps combined with back-scattered electron microscopy, which allows the detection of the elemental composition of individual particles, such as microplastics and others, in a semi-automated fashion. Detection of characteristic elements leads to further quantification of specific particles. This provides improved versatility compared to the elemental scans provided by standalone energy-dispersive X-ray spectroscopy (EDX) techniques and strongly increases multi-sample throughput speeds. The applications of this new technology for environmental contamination research include estimations on the morphology and distribution of microplastics and other particles alongside their interaction with micro-organisms and toxicity assessments by tracing the transport of trace metals through the environment on both contaminant (microplastics) and natural (suspended sediment) vectors. This study used filtered samples from the Cát Bà Islands of Viet Nam to assess the applicability of SEM-BEX to environmental contamination research, our results indicate that microplastics and other particles can be physically and chemically characterised across all samples down to a minimum particle size of 5 µm, in addition, we show that SEM-BEX is particularly powerful for identifying transparent fragments that are otherwise missed by optical studies. Trace metals were also detected, including Cr, Ti, and Hg, which might be due to pigment composition in paints or plastics, or adsorbed onto particles from the environment. Ultimately, SEM-BEX has broad applications as a rapid screening tool for environmental assessments in identifying contamination hot-spots before conducting particle-specific analysis (such as Raman spectroscopy). Further potential also exists to accelerate screening using artificial intelligence machine learning.