Related Experiment Video
Updated: Jan 13, 2026
03:33
Author Spotlight: Development and Application of SERS Flexible Substrates Using Synthesized AgNPs
Published on: November 17, 2023
3.2K
Synergistically Enhanced Ta2O5/AgNPs SERS Substrate Coupled with Deep Learning for Ultra-Sensitive Microplastic
Chenlong Zhao1, Yaoyang Wang1, Shuo Cheng1
1School of Physics and Electronics, Shandong Normal University, Jinan 250014, China.
Materials (Basel, Switzerland)
|January 10, 2026
Summary
A novel composite substrate enhances microplastic detection using Surface-Enhanced Raman Scattering (SERS). A deep-learning model accurately identifies microplastics even in noisy environments, aiding environmental monitoring.
Area of Science:
- Materials Science
- Environmental Science
- Analytical Chemistry
Background:
- Microplastic pollution poses a significant environmental threat, necessitating sensitive and accurate detection methods.
- Existing detection techniques often struggle with sensitivity, specificity, and performance in complex environmental matrices.
- Surface-Enhanced Raman Scattering (SERS) offers high sensitivity but requires optimized substrates and robust data analysis.
Purpose of the Study:
- To develop a high-performance Ta2O5/AgNPs composite SERS substrate for sensitive microplastic detection.
- To engineer a robust deep-learning model for accurate microplastic classification in challenging conditions.
- To provide a comprehensive solution for environmental monitoring and risk assessment of microplastic pollution.
Main Methods:
- Fabrication of a Ta2O5/AgNPs composite SERS substrate with modulated morphology and band-gap engineering.
- Integration of electromagnetic (EM) and chemical (CM) enhancement mechanisms for improved SERS performance.
- Development of a multi-scale deep-learning model incorporating wavelet transform, CNN, and Transformers for spectral analysis.
- Evaluation of microplastic capture efficiency and SERS detection limits for various polymers (PS, PET, PMMA).
Main Results:
- Achieved an ultra-low detection limit of 10^-13 M for Rhodamine 6G (R6G) with excellent linearity.
- Demonstrated superior capture capability for microplastics (PS, PET, PMMA) using a 3D 'pseudo-Neuston' network structure.
- The deep-learning model achieved 98.7% classification accuracy for microplastics under high-noise conditions, outperforming traditional methods.
Conclusions:
- The engineered Ta2O5/AgNPs composite SERS substrate offers high sensitivity and capture efficiency for microplastics.
- The proposed deep-learning model provides robust and accurate microplastic identification in complex, noisy environments.
- This integrated approach presents a significant advancement for environmental monitoring and microplastic pollution risk assessment.
