Prediction of impurities in hydrogen fuel supplies using a thermally-modulated CMOS gas sensor: Experiments and modelling

We report here on the results of a study on the response of copper oxide based low-power MEMS thermally modulated gas sensor to low ppm levels of hydrogen sulphide (H
₂
S) in a hydrogen environment. It was found that by using this material with a method of transient frequency analysis, this resistive gas sensor can operate reliably in a harsh environment including hydrogen atmosphere and high humidity levels. We implemented a Bayesian method for data analysis to predict the concentration of H2S in hydrogen supplies used in fuel cells.