Sensitive and selective gas/VOC detection using MOS sensor array for wearable and mobile applications

We demonstrate a monolithic, micro-fabricated metal-oxide semiconductor (MOS) sensor array comprising of different metal oxides, allowing for independent temperature controls and readouts from individual pixels in a multiplexed fashion. The sensor pixels are designed on an ultrathin membrane to minimize heat dissipation, thereby significantly lowering the overall power consumption (<10μW). The use of such an array at varying temperatures results in multidimensional data, which we use to train neural network machine learning algorithms to determine unique fingerprints of individual gases within a homogenous mixture. Our findings indicate that a multiplicity of MOS elements together with the ability to vary and measure at various temperatures can identify individual gases or VOCs within mixtures. The small form-factor and microfabrication approach of our sensor array also lends itself to CMOS integration, paving the way for a sensitive and selective gas/VOC platform for wearable and portable applications.