Virtual Sensor Array Based on Piezoelectric Cantilever Resonator for Identification of Volatile Organic Compounds

Area of Science:

• Materials Science
• Chemical Sensing
• Biotechnology

Background:

• Piezoelectric cantilever resonators are promising for real-time volatile organic compound (VOC) sensing.
• A significant challenge in VOC sensing is eliminating cross-sensitivity between different compounds.

Purpose of the Study:

• To propose a virtual sensor array (VSA) for accurate identification of VOCs and VOC mixtures.
• To demonstrate the VSA's capability in diagnosing plant diseases by detecting VOC biomarkers.

Main Methods:

• A graphene oxide (GO) film was deposited onto an aluminum nitride (AlN) piezoelectric cantilever with multiple electrode groups.
• The VSA utilized frequency shifts of multiple resonance peaks and impedance changes as multidimensional responses.
• Machine learning algorithms were employed to analyze the unique VOC fingerprints generated by the VSA.

Main Results:

• The VSA achieved high accuracies of 95.8% for identifying different VOCs and 87.5% for VOC mixtures.
• The VSA successfully identified emissions from healthy and late blight-infected plants with 89% accuracy.
• The study demonstrated low limits of detection (LODs) for VOCs due to high amplitude resonance peaks.

Conclusions:

• The proposed VSA effectively overcomes cross-sensitivity issues in VOC detection.
• The VSA shows significant potential for early diagnosis of infectious plant diseases through VOC biomarker analysis.
• Multidimensional responses combined with machine learning offer a robust platform for complex chemical sensing applications.