Chemiresistive Gas Sensors Made with PtRu@SnO₂ Nanoparticles for Machine Learning-Assisted Discrimination of Multiple Volatile Organic Compounds

Volatile organic compounds (VOCs) constitute key pollutants in the environment, and exposure to them is associated with negative health impacts. The vigilant monitoring of these pernicious VOCs is imperative for their timely detection and for curtailing the likelihood of both immediate and prolonged exposure, thus safeguarding against the deterioration of environmental quality. In this study, porous PtRu nanoalloys are successfully synthesized via a hydrothermal method and innovatively integrated with SnO 2 nanoparticles to significantly enhance the performance of gas sensors. Density functional theory (DFT) calculations substantiated the pivotal role of PtRu nanoalloys in amplifying the sensitivity of SnO 2 to acetone. A primary challenge in VOC surveillance is achieving the selectivity required for sensors to accurately identify specific compounds. By employing machine learning algorithms, with a particular emphasis on particle swarm optimization-support vector machine (PSO-SVM), we attained a classification accuracy of 100% in distinguishing between acetone, ethanol, methanol, and formaldehyde. This study demonstrates the potential for creating advanced sensors with selective detection of VOCs.