Accurate Methane Detection in Combustible Gas Mixtures by Using SnO2-Ag-ZnO Gas Sensors with Rapid Responses

Abstract

Efficient monitoring of methane is crucial for avoiding gas explosions in industrial processes. Metal oxide methane sensors exhibit promising gas detection performance, which has been well studied recently. However, conventional metal oxide sensors suffer from high operating temperatures, limited selectivity in multigas interference scenarios, and insufficient compatibility between hardware efficiency and algorithmic complexity for real-life applications. Here, we developed methane sensors based on SnO2-Ag-ZnO composite materials. Experimental results demonstrate that Ag-doping reduces the optimal operating temperature and enhances the methane response 1.79-fold, compared with that of pure SnO2 sensors. Introducing ZnO further amplifies gas adsorption and reaction activity by heterojunction effects. Furthermore, the SqueezeNet transfer learning model was applied to analyze the gas response signals, achieving 91.6% accuracy in the classification task of combustible gas mixtures. This research provides a comprehensive solution for monitoring methane in complex gas mixture environments.