Machine Learning?Enabled Smart Gas Sensing Platform for Identification of Industrial Gases

Both ammonia and phosphine are widely used in industrial processes, yet they noxious exhibit detrimental effects on human health. Despite the remarkable progress sensors development, there still some limitations, for instance, requirement of high operating temperatures, that most solely dedicated to individual gas monitoring. Herein, an ultrasensitive, highly discriminative platform is demonstrated detection identification at room temperature using a graphene nanosensor. Graphene exfoliated successfully functionalized by copper phthalocyanine derivate. In combination with efficient machine learning techniques, developed nanosensor demonstrates excellent performance even ultralow concentrations: 100 ppb NH3 (accuracy—100.0%, sensitivity—100.0%, specificity—100.0%) PH3 (accuracy—77.8%, sensitivity—75.0%, specificity—78.6%). Molecular dynamics simulation results reveal derivate molecules attached surface facilitate adsorption owing hydrogen bonding interactions. The smart sensing paves path design selective, sensitive, miniaturized, low-power consumption, nondedicated, system toward wide spectrum gases.