Au-Ag OHCs-based SERS sensor coupled with deep learning CNN algorithm to quantify thiram and pymetrozine in tea

Area of Science:

• Analytical Chemistry
• Materials Science

Background:

• Accurate pesticide residue detection in food is critical for public health and regulatory compliance.
• Traditional methods like High-Performance Liquid Chromatography (HPLC) can be time-consuming and require complex sample preparation.

Purpose of the Study:

• To develop a rapid, sensitive, and accurate method for detecting pesticide residues in tea using Surface-Enhanced Raman Scattering (SERS).
• To investigate the efficacy of novel Au-Ag octahedral hollow cages (Au-Ag OHCs) as SERS substrates.
• To apply intelligent algorithms for quantitative analysis of specific pesticides.

Main Methods:

• Fabrication of Au-Ag octahedral hollow cages (Au-Ag OHCs) using Cu2O templates.
• Utilizing SERS to amplify Raman signals of pesticide molecules.
• Applying Convolutional Neural Network (CNN), Partial Least Squares (PLS), and Extreme Learning Machine (ELM) for quantitative analysis.

Main Results:

• Au-Ag OHCs demonstrated enhanced SERS performance due to their unique structure.
• CNN algorithms achieved high accuracy in quantifying thiram (R=0.995) and pymetrozine (R=0.977).
• The developed SERS method showed no significant difference compared to HPLC analysis in real tea samples.

Conclusions:

• The proposed Au-Ag OHCs-based SERS technique provides a sensitive and rapid platform for pesticide residue detection in tea.
• Intelligent algorithms, particularly CNN, are effective for quantitative analysis in SERS.
• This method offers a viable alternative to conventional techniques for food safety monitoring.