Two-Dimensional SERS Sensor Array for Identifying and Visualizing the Gas Spatial Distributions of Two Distinct Odor...

Area of Science:

• Chemical Sensing
• Spectroscopy
• Data Analysis

Background:

• Odor source spatial distribution offers insights into composition, size, and location.
• Surface-enhanced Raman scattering (SERS) gas sensors provide high sensitivity, specificity, and rapid detection.
• Visualizing gas distribution is crucial for odor source analysis.

Purpose of the Study:

• To develop a SERS gas sensor array for visualizing spatial gas distribution.
• To localize and estimate the size of odor sources.
• To identify and differentiate odor compounds using a convolutional neural network.

Main Methods:

• Utilized a SERS gas sensor array for spectral data acquisition.
• Employed non-negative matrix factorization for feature and concentration extraction.
• Applied Gaussian fitting for odor source localization and size estimation.
• Implemented a convolutional neural network for spectral recognition.

Main Results:

• Successfully visualized the spatial distribution of benzaldehyde and 4-ethylbenzaldehyde.
• Accurately localized and estimated the size of odor sources.
• Achieved 98.21% accuracy in identifying odor compounds and mixtures using a convolutional neural network.

Conclusions:

• The SERS gas sensor array effectively visualizes odor spatial distribution.
• The developed system enables odor source localization, size estimation, and identification.
• Potential applications include environmental monitoring, healthcare, and odor source analysis.