Multigas Selective Identification Based on a Single Chemiresistive Gas Sensor via Dynamic Light-Pulse Modulation and...

Area of Science:

• Materials Science
• Chemical Sensing
• Nanotechnology

Background:

• Metal oxide semiconductor (MOS) gas sensors are cost-effective for gas detection but lack selectivity for similar molecules.
• Current solutions like sensor arrays increase complexity and cost.
• Distinguishing structurally analogous volatile organic compounds (VOCs) remains a challenge.

Purpose of the Study:

• To develop a dynamic light-pulse modulation strategy to enhance gas-sensing selectivity using a single MOS sensor.
• To achieve selective identification of structurally similar VOCs by expanding signal dimensionality.
• To overcome the cross-sensitivity limitations of conventional MOS gas sensors.

Main Methods:

• Fabrication of an Ag-modified ZnO nanocomposite gas sensor.
• Application of dynamic light-pulse modulation to extract multidimensional features from transient resistance signals.
• Integration of dynamic and steady-state parameters into a feature set.
• Utilizing a support vector machine classifier for VOC identification.

Main Results:

• The Ag-modified ZnO sensor showed analogous responses to formic acid, ethanol, and acetic acid.
• Multidimensional feature extraction via light-pulse modulation enabled differentiation.
• High identification accuracy (97.8%) was achieved for structurally similar VOCs across a wide concentration range.
• The strategy successfully overcame cross-sensitivity issues.

Conclusions:

• Dynamic light-pulse modulation combined with machine learning offers a pathway to enhance single MOS sensor selectivity.
• This approach enables the development of compact and intelligent gas recognition systems.
• The proposed strategy addresses limitations in distinguishing structurally similar VOCs.