Trapping and chemical characterization of sub-microplastics using Raman optical tweezers with machine learning

The accumulation of plastics in the environment and their fragmentation into micro- and nanoplastics represent a growing ecological and public health concern. Accurate detection and classification of these particles remain challenging due to their small size and the complexity of environmental matrices. Here, we present an advanced analytical platform that combines optical tweezers Raman spectroscopy (OTRS) with machine learning-driven principal component analysis (PCA) to enable precise identification of microplastics at the single-particle level. By analyzing 35 Raman spectra, we demonstrate the platform’s ability to effectively distinguish common plastics—such as polyethylene (PE), polypropylene (PP), and polystyrene (PS)—from organic matter without extensive sample preparation. This approach not only enhances sensitivity and specificity but also supports high-throughput, automated analysis, offering a scalable solution for real-time environmental monitoring. Our findings highlight the potential of this integrated method to improve microplastic surveillance and inform mitigation strategies in polluted ecosystems.