Related Experiment Video
Updated: May 12, 2026
07:57
Quantitative Detection of Trace Explosive Vapors by Programmed Temperature Desorption Gas Chromatography-Electron Capture Detector
Published on: July 25, 2014
Deep Learning-Assisted Sensor Array Based on Host-Guest Chemistry for Accurate Fluorescent Visual Identification of
Wenxing Gao1, Zhibin Wang2, Qiang Li1
1State Key Laboratory of Solidification Processing, School of Materials Science and Engineering, Northwestern Polytechnical University, Xi'an 710072, China.
Analytical Chemistry
|May 17, 2025
Summary
This study introduces a novel deep learning vision platform using multicolor fluorescent gold nanoclusters for rapid and precise explosive detection. The system achieves 100% accuracy in identifying seven explosives, enhancing security and monitoring capabilities.
Area of Science:
- Nanotechnology
- Analytical Chemistry
- Artificial Intelligence
Background:
- Conventional analytical techniques struggle with accurate and rapid discrimination of multiple explosives.
- Explosive detection is critical for national security, environmental protection, and public health.
Purpose of the Study:
- To develop a deep learning-assisted artificial vision platform for highly accurate and rapid discrimination of multiple explosives.
- To leverage multicolor fluorescent gold nanoclusters (CD-AuNCs) and advanced algorithms for enhanced explosive detection.
Main Methods:
- Utilized cyclodextrin-protected multicolor fluorescent gold nanoclusters (CD-AuNCs) with four distinct emission wavelengths.
- Employed host-guest interactions for selective explosive binding and unique fluorescence fingerprint generation.
- Integrated a dense convolutional network (DenseNet) algorithm with smartphone-captured multicolor fluorescence data (RGB values) for analysis.
Main Results:
- Achieved 100% recognition accuracy for seven explosives at a concentration of 200 microM.
- Demonstrated that fluorescence enhancement is due to ligand rigidification and quenching by photoinduced electron transfer.
- Successfully captured multicolor fluorescence responses via smartphone, enabling rapid visual classification.
Conclusions:
- The developed platform offers a powerful tool for on-site explosive monitoring with high precision and speed.
- The strategy provides a versatile approach for intelligent detection of diverse analytes, showing potential for real-world security and environmental applications.
