Abstract:
The identification and monitoring of volatile organic compounds (VOCs) have gained an
increasing concern in recent years because of their harmful effects on humans and the
environment. This has greatly promoted the development of chemical sensors to discern these latters. The selective detection of volatile organic compounds by means of chemical sensors is one of the open challenges in chemical sensing. In practice, most chemical sensors suffer from cross-sensitivity problem. To overcome this drawback, the identification and quantification of volatile organic compounds (VOC) by using a non-selective sensor requires a combination of sensors, followed by pattern recognition methods. In this context, a device widely studied is the quartz crystal microbalance (QCM) Based on this concept, the QCM-based sensors coated with different monomers were deposited by the technique of plasma enhanced chemical vapor deposition. A brief summary on physico-chemical and morphological analysis of sensitive layers, with infrared spectroscopy (FTIR) and atomic force microscopy (AFM) is also presented. The sensor sensitivity was evaluated by recording the frequency shift of the quartz (∆f) when exposed to different concentrations of VOCs, such as; ethanol, benzene and chloroform and their binary mixture. This has allowed the creation of a database consisting of different fingerprints of VOCs and their mixtures with different combinations of QCM to evaluate the performance of
each sensor combination by a qualitative and quantitative study using different methods of multivariate analysis, namely; hierarchical ascendant clustering (HAC), principal component analysis (PCA), factorial discriminate analysis (FDA), support vector machines (SVM) and artificial neural networks (ANN).
The results obtained showed that the combination of all the sensors as well as the
combination of sensors coated with different monomers are the most effective for the
identification and quantification of individual VOCs as well as the evaluation of a VOC in the mixtures, with different methods of multivariate analysis.
