Related Experiment Video
Updated: Sep 25, 2025
03:33
Author Spotlight: Development and Application of SERS Flexible Substrates Using Synthesized AgNPs
Published on: November 17, 2023
2.6K
Machine learning for rapid quantification of trace analyte molecules using SERS and flexible plasmonic paper
Reshma Beeram1, Dipanjan Banerjee1, Linga Murthy Narlagiri1
1Advanced Centre of Research in High Energy Materials (ACRHEM), University of Hyderabad, Hyderabad 500046, Telangana, India. soma_venu@uohyd.ac.in.
Analytical Methods : Advancing Methods and Applications
|April 27, 2022
Summary
This study introduces a low-cost, flexible SERS substrate for quantifying trace analytes like crystal violet and picric acid. Machine learning models achieved high accuracy, enabling rapid and affordable chemical analysis.
Area of Science:
- Analytical Chemistry
- Materials Science
- Spectroscopy
Background:
- Surface-enhanced Raman scattering (SERS) faces challenges in reproducibility and signal stability for trace analyte detection.
- Quantifying analytes with SERS is difficult due to low reproducibility and signal blinking, especially at trace levels.
Purpose of the Study:
- To develop a simple, cost-effective, and flexible SERS substrate for accurate quantification of trace analytes.
- To utilize machine learning for enhancing the precision of SERS-based quantification.
Main Methods:
- Fabrication of a hydrophobic plasmonic filter paper-based SERS substrate using silicone oil coating and gold nanoparticles synthesized via femtosecond laser ablation.
- Characterization of gold nanoparticles using UV-Vis, TEM, and SEM.
- Application of machine learning techniques, including Principal Component Analysis (PCA) for dimensionality reduction and Support Vector Regression (SVR) for quantification, using over 900 SERS spectra.
Main Results:
- Achieved high quantification accuracy with an R-squared error of 0.9629 for crystal violet (CV) and 0.9472 for picric acid (PA).
- Demonstrated the substrate's effectiveness with trace analytes (CV and PA).
- Validated the rapid quantification capability with a computation time of less than 10 seconds.
Conclusions:
- The developed hydrophobic plasmonic filter paper SERS substrate offers an affordable and rapid solution for trace analyte quantification.
- The integration of machine learning with SERS data significantly improves quantification accuracy and overcomes SERS limitations.
- This method holds promise for portable and efficient chemical analysis in various applications.
