VOCs Identification Method Based on One Single ZnTTP Sensor

Abstract

In this work, an innovative approach based on the combination of hierarchical ascending classification (HAC) and principal component analysis (PCA) is proposed to discriminate individual vapors such as methanol, toluene, acetone, isopropanol, and ethanol using dynamic responses of one single sensor based on 5, 10, 15, 20 tetratoloylphenylporphyrinato Zn (II) (ZnTTP) at room temperature. Experimental data of 20 measurements as individuals (i.e., four different concentrations of each vapor) and extraction of new mathematical features associated with dynamic behavior were the basis of two subsets entered into the gas identification method. With PCA, four distinct groups were observed for each subset. The first three principal components have been added as data features for each previous corresponding subset of the HAC process. The classification becomes clearer and unambiguously separates the different clusters in good compliance with PCA. The obtained results have demonstrated the possibility to recognize the studied vapors by exploiting a single gas sensor with an opportunity to apply the developed PCA/HAC-based platform in cost-effective and highly selective gas monitoring systems.