Room-Temperature Trace NO₂ Monitoring System Based on Two-Dimensional Heterostructures and Integrated with Deep...

Area of Science:

• Materials Science
• Chemical Sensing
• Environmental Monitoring

Background:

• Accurate detection of trace nitrogen dioxide (NO2) is vital for air quality management and early diagnosis of respiratory illnesses.
• High-precision gas detection at low concentrations remains a significant research challenge.
• Material optimization and deep learning are key strategies for improving sensor accuracy.

Purpose of the Study:

• To develop a remote NO2 monitoring system for low-concentration detection.
• To leverage Bi2S3/WO3 heterostructures and deep learning for enhanced gas sensing performance.
• To enable real-time, high-precision NO2 detection at room temperature.

Main Methods:

• Fabrication of Bi2S3/WO3 heterostructures for gas sensing.
• Integration of wireless communication modules for remote monitoring.
• Application of a 1D-CNN/LSTM deep learning model with data augmentation for accurate regression.

Main Results:

• The Bi2S3/WO3 sensor demonstrated high response (17.9 to 5 ppm NO2), sensitivity (3.84/ppm), and rapid response/recovery times (27/110 s).
• The 1D-CNN/LSTM model achieved high accuracy (R2 = 0.9826) for ppb-level NO2 detection after data augmentation.
• The system facilitated real-time monitoring, multichannel operation, and intelligent alarming.

Conclusions:

• Bi2S3/WO3 heterostructures significantly enhance NO2 sensing performance through improved charge transfer and active oxygen species generation.
• The 1D-CNN/LSTM deep learning model effectively overcomes data scarcity issues, boosting low-concentration detection accuracy.
• The integrated system offers a promising strategy for high-precision, real-time remote NO2 monitoring.