Related Experiment Video
Updated: Jun 16, 2026
Protocol for Microplastics Sampling on the Sea Surface and Sample Analysis
Published on: December 16, 2016
Detection of Microplastics Based on a Liquid-Solid Triboelectric Nanogenerator and a Deep Learning Method
Tao Huang1, Wei Sun1, Lei Liao2
1Key Lab New Processing Technology for Nonferrous Metals & Materials Ministry of Education, Guangxi Key Lab of Optical and Electronic Materials and Devices, College of Materials Science and Engineering, Guilin University of Technology, Guilin, Guangxi 541004, China.
A novel liquid-solid triboelectric nanogenerator (LS-TENG) sensor combined with deep learning can detect and classify microplastics in liquids. This method accurately quantifies microplastic content and identifies types, offering a significant advancement in environmental monitoring.
Area of Science:
- Environmental Science
- Materials Science
- Sensor Technology
Background:
- Microplastics are pervasive pollutants with detrimental effects on ecosystems and human health.
- Traditional methods for microplastic detection face challenges due to their small size and varied properties.
- Accurate identification and quantification of microplastics are crucial for environmental risk assessment.
Purpose of the Study:
- To develop a novel method for detecting and classifying microplastics in liquids.
- To utilize a liquid-solid triboelectric nanogenerator (LS-TENG) coupled with a deep learning model for microplastic analysis.
- To establish a quantitative and qualitative detection system for microplastics in aqueous environments.
Main Methods:
- A liquid-solid triboelectric nanogenerator (LS-TENG) was employed as a sensor to detect microplastics in liquids.
- Different types of microplastics (polyethylene, polypropylene, PVC, PET, polystyrene) were introduced to assess their impact on LS-TENG output signals.
- A convolutional neural network (CNN) deep learning model was trained using LS-TENG voltage signals for microplastic classification.
Main Results:
- The type and concentration of microplastics significantly influenced the output voltage of the LS-TENG sensor.
- A linear relationship was observed between microplastic mass fraction (0.025-0.25 wt %) and LS-TENG sensor voltage output.
- The deep learning model achieved high accuracy in identifying different types of microplastics based on sensor signals.
Conclusions:
- The LS-TENG sensor, in conjunction with deep learning, provides a viable method for quantitative detection and classification of microplastics in liquids.
- This approach offers a promising advancement for environmental monitoring and the study of microplastic pollution.
- The findings highlight the potential of LS-TENG technology in addressing challenges in microplastic analysis.
More Related Videos
05:31Sampling, Sorting, and Characterizing Microplastics in Aquatic Environments with High Suspended Sediment Loads and Large Floating Debris
Published on: July 28, 2018
09:10Multimodal Analysis of Microplastics in Drinking Water using a Silicon Nanomembrane Analysis Pipeline
Published on: June 13, 2025
