Integrating Metal Phenolic Networks-Mediated Separation and Machine Learning-Aided SERS for High-Precision Quantification and Classification of Nanoplastics

The increasing accumulation of nanoplastics across ecosystems poses a significant threat to both terrestrial and aquatic life. Surface-enhance Raman scattering (SERS) is an emerging technique used for nanoplastic detection. However, the identification and classification of nanoplastics using SERS have challenges regarding sensitivity and accuracy, as nanoplastics are sparsely dispersed in the environment. Metal-phenolic networks (MPNs) have the potential to rapidly concentrate and separate various types and sizes of nanoplastics. SERS combined with machine learning may improve prediction accuracy. Herein, for the first time, we report the integration or MPNs-mediated separation with machine learning-aided SERS methods for the accurate classification and high-precision quantification of nanoplastics which is tailored to include the complete region of characteristic peaks across diverse nanoplastics in contrast to the traditional manual analysis of SERS spectra on a singular characteristic peak. Our customized machine learning system (e.g., outlier detection, classification, qualification) allows for the identification of detectable nanoplastics (accuracy 81.84%), accurate classification (accuracy > 97%) and the sensitive quantification of various types of nanoplastics (PS, PMMA, PE, PLA) down to ultra-low concentrations (0.1 ppm) as well as the accurate classification (accuracy > 92%) of nanoplastics mixtures to sub-ppm level. The effectiveness and novelty of this approach are substantiated by its ability to discern between different nanoplastics mixtures and detect nanoplastics samples in natural water systems.