Neural Network-Assisted Dual-Functional Hydrogel-Based Microfluidic SERS Sensing for Divisional Recognition of...

Area of Science:

• Materials Science and Engineering
• Analytical Chemistry
• Nanotechnology

Background:

• Existing Raman detection systems require enhancement in sensitivity, integration, and practicality.
• Microfluidic devices offer precise control over sample handling and reaction conditions.
• Hydrogels provide a versatile matrix for incorporating nanoparticles and controlling fluid flow.

Purpose of the Study:

• To develop a deep learning-based, dual-functional, subregional microfluidic integrated hydrogel surface-enhanced Raman scattering (SERS) platform.
• To improve the sensitivity, integration, and practicality of Raman detection systems.
• To enable simultaneous detection and classification of multiple analytes.

Main Methods:

• Synthesis of homogeneous silver nanoparticles (Ag NPs) via a one-step reduction method.
• Embedding Ag NPs within N-isopropylacrylamide/poly(vinyl alcohol) (Ag NPs-NIPAM/PVA) hydrogels.
• Fabrication of a four-channel microfluidic platform with hydrogel switches and adjustable detection regions for controlled Ag NP gap modulation and Raman enhancement.

Main Results:

• Achieved a low detection limit of 10^-10 mol/L for Rhodamine 6G (R6G) with an enhancement factor of 10⁷.
• Demonstrated high precision with relative standard deviations below 10% for characteristic peaks.
• Successfully performed simultaneous subarea detection and classification of four real molecules (thiram, pyrene, anthracene, dibutyl phthalate) using fully connected neural network technology.

Conclusions:

• The developed dual-functional microfluidic-integrated hydrogel SERS platform offers enhanced sensitivity and practicality for chemical detection.
• The platform enables precise subregional sampling and simultaneous multi-analyte detection.
• Integration with deep learning improves the predictability and applicability for molecule classification and identification.