Neural-Network-Assisted Impedance Analysis for Humidity and Ammonia Detection Using MXene and PtSnO₂ Sensors with...

Area of Science:

• Materials Science
• Chemical Sensors
• Nanotechnology

Background:

• Semiconducting gas sensors face challenges like humidity interference, baseline drift, and poor selectivity.
• Direct current measurements are particularly susceptible to environmental variations, limiting sensor accuracy.

Purpose of the Study:

• To develop a high-performance gas sensing system mitigating humidity interference.
• To enable simultaneous and accurate detection of ammonia (NH3) and relative humidity (RH).

Main Methods:

• Integration of Platinum Tin Oxide (PtSnO2) and MXene-based sensors.
• Multi-frequency impedance measurements for sensor characterization.
• Utilizing a multilayer perceptron model for data analysis and prediction.

Main Results:

• Achieved tunable gas responses, low noise, and enhanced baseline stability.
• Demonstrated minimal humidity cross-sensitivity and extended dynamic range.
• Accurate deconvolution and prediction of NH3 and RH concentrations, outperforming commercial sensors.

Conclusions:

• The combined PtSnO2 and MXene sensor system effectively addresses humidity interference in gas sensing.
• The developed analytical framework offers a robust solution for advanced gas detection.
• This methodology is adaptable for various material systems and applications, including industrial safety and biomedical diagnostics.