ISSN 2756-3251
Advanced Journal of Environmental Science and Technology | Vol. 17 (11), pp. 211-225, November, 2026
DOI: 10.46882/2026.AJEST.171115
Research Article
Rapid Automated Microplastic Identification in Drinking Water Treatment Plants Using Raman Microspectroscopy and Convolutional Neural Networks
Min-Ji Park¹*, David O'Connor², Fatima Al-Khatib³
¹Department of Environmental Engineering, Yonsei University, Seoul, South Korea
²School of Chemistry, University College Dublin, Dublin, Ireland
³Department of Chemical Engineering, Jordan University of Science and Technology, Irbid, Jordan
Corresponding Author: Min-Ji Park — [email protected]
Abstract
Manual spectral matching of microplastics down to 1 μm in treated drinking water is time-consuming and prone to fluorescence interference from residual organic matter. We trained a one-dimensional residual convolutional neural network (1D-ResNet) on 48,000 raw Raman spectra spanning 14 weathered polymer types and biofilms to automate particle classification across three municipal drinking water treatment plants. The model classified microplastics down to 1.2 μm at 97.8% accuracy and processed 1,000 particles in 6 minutes—45 times faster than conventional library correlation. Conventional coagulation-sand filtration removed 94.6 ± 1.4% of raw water microplastics, while granular activated carbon reduced residual 1 to 5 μm fragments to 2.4 particles/L.
Keywords: drinking water microplastics; Raman microspectroscopy; convolutional neural networks; automated polymer identification; water treatment efficiency; emerging contaminants
Manuscript Timeline: Received 16 October, 2026; Revised 18 November, 2026; Accepted 20 November, 2026; Published 20 November, 2026.
Citation: Park, M.-J., O'Connor, D., and Al-Khatib, F. (2026). Rapid Automated Microplastic Identification in Drinking Water Treatment Plants Using Raman Microspectroscopy and Convolutional Neural Networks. Advanced Journal of Environmental Science and Technology, 17(11), pp. 211-225.
Subscribe to read the full article: https://www.internationalscholarsjournals.org/subscribe-to-read