Predictive mixed-gas detection using rGO/In₂O₃ nanocomposite sensors assisted by machine learning

Area of Science:

• Materials Science
• Nanotechnology
• Sensor Technology

Background:

• Chemiresistive gas sensors face challenges in analyte selectivity and sub-ppm detection limits.
• Hybrid materials combining reduced graphene oxide (rGO) and metal oxides offer enhanced sensitivity at ultralow concentrations.

Purpose of the Study:

• To develop a highly sensitive and selective gas sensor using rGO/In2O3 nanocomposites.
• To implement a machine-intelligent framework for simultaneous gas identification and concentration prediction in mixed environments.

Main Methods:

• Synthesized rGO via a modified Hummers' method and incorporated it into nanocrystalline In2O3.
• Fabricated rGO/In2O3 thin films using spin coating and post-deposition annealing.
• Employed a machine-intelligent framework analyzing dynamic response curves for gas analysis.

Main Results:

• Optimized rGO/In2O3 sensor demonstrated stability and a 100 ppb detection limit for H2S.
• Machine learning framework achieved 99.7% accuracy in distinguishing gas clusters.
• Accurate prediction of H2S, NH3, and CO concentrations in mixed environments was achieved.

Conclusions:

• The rGO/In2O3 hybrid material enhances gas sensing performance.
• The machine-intelligent framework enables robust, simultaneous multi-gas detection and quantification.
• This integrated platform advances smart, ultra-low-level gas sensing for complex real-world applications.