Quantitative detection of formaldehyde and ammonia gas via metal oxide-modified graphene-based sensor array combining with neural network model

Abstract

This paper reports metal oxide (MOx)-decorated graphene-based sensor array combining with back-propagation (BP) neural network toward the detection of indoor air pollutant exposure. Tin dioxide (SnO2) nanospheres and copper oxide (CuO) nanoflowers-decorated graphene were used as candidates for formaldehyde and ammonia gas sensing, respectively. The as-synthesized sensing materials were characterized in terms of their nanostructural, morphological and compositional features by SEM, Raman spectra, and XRD. The sensor array was fabricated via one-step hydrothermal route and layer-by-layer (LbL) self-assembly technique on the substrate with interdigital microelectrodes. The sensing properties of MOx/graphene composite toward the mixture gas of ammonia and formaldehyde, such as dynamic response, sensitivity, response/recovery time, and stability, were investigated at room temperature. And furthermore, this work successfully achieved the recognition and quantitative prediction of components in the gas mixture of formaldehyde and ammonia through the combination of MOx/graphene-based sensor array and neural network-based signal processing technologies.