RBFNN-Based PPy/GaN sensor array with wide dynamic range and sub-ppb detection for accurate ammonia identification in...

Area of Science:

• Materials Science
• Nanotechnology
• Chemical Engineering

Background:

• Ammonia (NH3) detection in human breath is crucial for early disease diagnosis.
• Polymer composite nanomaterials offer high-performance NH3 gas sensing capabilities.
• Chronic kidney disease (CKD) diagnosis can be improved by breath analysis.

Purpose of the Study:

• To synthesize and fabricate polypyrrole (PPy) and gallium nitride (GaN) nanostructures for NH3 gas sensing.
• To evaluate the gas sensing performance, including detection range, moisture resistance, and stability, of PPy/GaN sensors.
• To validate the sensor's sensitivity and accuracy in detecting NH3 in human exhaled breath using machine learning.

Main Methods:

• Metal-organic chemical vapor deposition (MOCVD) and in situ oxidative polymerization were used to synthesize PPy/GaN nanostructures.
• Fabrication of PPy/GaN gas sensors.
• Systematic analysis of gas sensing performance, including detection range (100 ppb-1000 ppm) at room temperature.
• Validation using a sensor array for human breath analysis and machine learning algorithms for prediction.

Main Results:

• The PPy/GaN-1 sensor exhibited an ultra-wide NH3 detection range with excellent moisture resistance and long-term stability.
• Optimal synergy between GaN and PPy due to uniform film distribution contributed to enhanced performance.
• High-precision prediction of low-concentration gases (1.17 ppm error) was achieved using machine learning on breath samples.

Conclusions:

• The developed PPy/GaN gas sensor demonstrates significant potential for sensitive and stable NH3 detection in human breath.
• This technology can contribute to the development of non-invasive early warning systems for diseases like CKD.
• The combination of nanomaterials, advanced fabrication, and machine learning offers a promising approach for breath analysis.