About Raman Automation for Small Microplastics

Automated microplastic detection using particle detection software and µ-Raman spectroscopy addresses the labor-intensive bottleneck of identifying and classifying small particles (20–53 µm) in environmental samples. Developed by Garth A. Covernton, Ludovic Hermabessiere, Ungku Zoë Anysa Ungku Fa’iz, and Chelsea M. Rochman at the University of Toronto, this R-based tool couples image analysis with Raman spectral identification to streamline the workflow from sample preparation through polymer classification.

The method uses particle detection software to locate candidate microplastics in sample images, then automates the acquisition and analysis of Raman spectra for each particle. The tool includes preprocessing routines, spectral matching algorithms, and polymer classification outputs. Validation testing on known polymer particles in the target size range demonstrated the method’s accuracy for routine microplastic analysis in environmental and drinking-water samples.

Key Features

• Automated particle detection and Raman spectral acquisition for 20–53 µm microplastics, reducing manual microscopy time
• R source code with particle localization, spectral preprocessing, and polymer assignment modules
• Validation data from known polymer standards and environmental test samples
• Compatible with standard Raman spectrometer hardware and image acquisition systems
• Reproducible workflow enabling interlaboratory method comparison and adoption

Development and Validation

The tool was developed at the University of Toronto by researchers with expertise in microplastics spectroscopy and analytical chemistry. Covernton et al. (2026) published the method validation and automated workflow in ChemRxiv, describing the accuracy and precision of particle detection and Raman identification for small microplastics. The work addresses a critical gap in microplastics methods: most Raman-based workflows focus on particles >100 µm, whereas many environmental samples contain substantial mass and particle counts in the 20–53 µm fraction. The validation study included both synthetic polymer particles and real environmental samples to establish performance metrics for accuracy and false-positive rates.

Access and Data Availability

The complete Python codebase, validation datasets, and method documentation are available via Zenodo (DOI: 10.5281/zenodo.19488852) as a compressed archive (GCov/Raman-Automation-for-Small-Microplastic-Particles-ChemRxiv.zip). The code is also hosted on GitHub for version control and collaborative development. Both repositories include the source code, validation data, and supplementary material from the ChemRxiv preprint.

No license terms are explicitly stated in the available metadata, so users should contact the authors or check the GitHub repository for licensing details before use or modification. The tool requires R and standard scientific computing libraries; specific dependencies are documented in the code repository. The method is designed for researchers with access to a Raman microscope and basic image analysis capability.

Community and Support

The GitHub repository accepts issue reports and pull requests for bug fixes and feature suggestions. Questions about the method or validation can be directed to the corresponding author. This tool complements existing Raman-based microplastics methods, such as the Standard Operating Procedures for Extraction and Measurement by Raman Spectroscopy of Microplastic Particles in Drinking Water, by automating the spectroscopy step for smaller size fractions.

This resource is part of the Plastiverse.org ecosystem, which curates tools and knowledge for microplastics research.