Quantitative gas identification using HMDSO coated QCM sensors and multilayer neural networks

Abstract

Quartz crystal microbalance (QCM) electrodes were coated from a mixture of hexamethyldisiloxane (HMDSO) and O2 for the detection of volatile organic compounds (VOCs). The sensitivity of the coated QCM-based sensors was evaluated by monitoring the frequency shifts of the quartz exposed to different concentrations of volatile organic compounds (VOCs), such as; ethanol, benzene and chloroform. The sensors responses data have been used in subsequent data processing step for quantitative gas identification. It was shown that, with varying oxygen proportion in the mixture, the sensitivity and the selectivity of the QCM sensor towards on a given analyte change. Several multivariable analysis techniques, with varying degrees of success for automated identification of VOCs such as; linear method: principal component analysis (PCA) and nonlinear method: Artificial neural networks multilayer perceptions (ANNMLP) have been used to explore the data in this study. The collected data processed by principal component analysis has identified ethanol compared to the other two gases. The artificial neural networks multilayer perception after training has allowed the identification of 100% of these three types of VOCs and estimated their concentration, which demonstrates the success of nonlinear methods for identification and quantification of gas using this type of sensors.