Korean researchers create AI-based microplastic detection kit

The study uses 3D-plasmonic gold nanopocket (3D-PGNP) nanoarchitecture on a paper substrate for simultaneous MP filtration and detection. The paper-based 3D-PGNP is integrated with a syringe filter device, and then MP-containing solutions are injected through the syringe. MPs were successfully identified without pretreatment thanks to surface-enhanced Raman scattering (SERS).

3D-PGNP significantly improves the sensitivity, confirmed by finite-difference time-domain simulation. Then, the SERS mapping images obtained from a portable Raman spectrometer are transformed into digital signals via machine learning technique to identify and quantify the MP distribution.

SERS is an optical technology that amplifies the Raman scattering signals of molecules absorbed on plasmonic materials.

Overcoming limitations

Recent research using the SERS technique for MP detection focuses on the sensitive quantification of MP concentration, detection of various MP sizes and classification of the MP composition in the sample matrix.

However, most of these studies use SERS substrate with nanogaps, which limits the detection of large-sized MPs in the micro to submicrometer range because they’re hard to place at the hotspot region.

Therefore, the development of substrate-type SERS materials that can capture and generate harmonised MP SERS signals can overcome conventional limits. Multiple hotspots need to be generated at the SERS substrate-MP interface for sensitive MP detection. This can be achieved by having nanoarchitectures with large-sized pores than can capture and surround the MPs.