Distributed Sensor System for Quantification of Individual Components in a Multiple Gas Mixture

A distributed gas sensor system consists of a measurement cell, with a row of gas sensors, and a catalyst in proximity to the sensors. The test gas flows along the row of sensors. The gas is then, if possible, catalytically altered along the flow direction and monitored at different stages of catalytic conversion with the sensors. Here, a system consisting of combinations of palladium and platinum metal oxide semiconductor field effect transistor (MOSFET) sensors, and palladium and platinum catalysts, was studied, especially regarding the problem of quantifying individual components in multiple-gas mixtures. Quaternary mixtures of hydrogen, ammonia, ethyne (C2H2), and ethanol were analyzed with the system. Using selected sensor signals from the different sensor/catalysts combinations, at different stages of catalytic conversion, high-accuracy artificial neural network prediction models were obtained for all four gas components. It was further shown that the two sensor types gave different information, and the two catalysts affected the analyzed gas mixtures differently. This rather simple distributed system, consisting of only two different sensor types and two catalysts, thus gave enough information to solve the rather elaborate problem of quantifying individual components in a complex mixture. The possible amount of information to be gained from such a system seems large. It can easily be extended to other sensors/catalysts and be incorporated into, e.g., an electronic nose system.