Related Experiment Video
Updated: Jun 11, 2026
08:27
Sampling and Identification of Microplastics in Groundwater
Published on: November 7, 2025
Microplastics identification framework: Integration of microplastic-derived dissolved organic matter fingerprints and
Ruxin Yang1, Jinjiang Duan2, Jianhao Song2
1School of Environmental Science and Engineering, Southwest Jiaotong University, Chengdu 611700, China; Key Laboratory of Synergetic Control and Joint Remediation for Soil & Water Pollution, Ministry of Ecology and Environment, Chengdu, 610059, China.
Water Research
|June 9, 2026
Summary
This study introduces a new method to identify microplastic (MP) types in water using their dissolved organic matter (DOM) fingerprints and a machine learning model. This approach helps understand MP sources and ecological risks.
Area of Science:
- Environmental Chemistry
- Polymer Science
- Analytical Chemistry
Background:
- Microplastics (MPs) are pervasive in aquatic ecosystems, with their ecological impact tied to polymer type.
- Identifying MP polymer composition is crucial for risk assessment, but current methods are limited.
- Dissolved organic matter (DOM) released by MPs contains polymer-specific chemical fingerprints.
Purpose of the Study:
- To develop an accessible and interpretable framework for identifying polymer-related MPs using MP-DOM fingerprints.
- To integrate an interpretable machine learning model with MP-DOM fingerprints for MP identification.
- To differentiate between aliphatic, aromatic, and biodegradable MPs based on their DOM signatures.
Main Methods:
- Derived MP-DOM fingerprints from dissolved organic carbon (DOC) levels, UV-Visible absorbance, and fluorescence indices.
- Analyzed aliphatic (polyethylene, polypropylene), aromatic (polystyrene, polyethylene terephthalate), and biodegradable (polylactic acid) MPs.
- Utilized functional group analyses and molecular-scale simulations to understand DOM composition and release pathways.
- Developed and optimized a random forest machine learning model using seven MP-DOM fingerprints.
Main Results:
- MP-DOM fingerprints showed distinct characteristics for aliphatic (weak optical signatures), aromatic (enhanced aromaticity), and biodegradable (high DOC release, low aromaticity) MPs.
- The optimized random forest model achieved high classification performance (AUC=0.953) in controlled conditions.
- The model maintained good performance (AUC=0.903) in environmental samples (river water) after background correction, with accurate identification of biodegradable MPs.
Conclusions:
- MP-DOM fingerprints provide a feasible basis for polymer-related MP identification in aquatic environments.
- The developed framework offers an accessible approach to support source-related interpretation of MPs.
- This method can aid in assessing the ecological risks associated with different types of microplastics.
