Intra-class Classification of Amines using Impedance Spectroscopy with a SnO2-Ppy Composite based Sensor

Amines play a significant role in everyday life, and their detection remains a crucial focus in research and development. Conducting polymer-based gas sensors have been widely reported for amine detection, primarily relying on conductivity-based sensing techniques. However, these methods often lack selectivity. To overcome this limitation, electrochemical impedance spectroscopy offers a more effective alternative. In this paper, the impedance response of the tin-oxide poly-pyrrole sensor was observed for four different amines namely ammonia, Di-methyl Amine, Tri-methyl amine and Tri-ethyl amine at a frequency range of 100 Hz to 8 MHz. Equivalent electrical parameters were derived from the impedance spectroscopy data visually represented by Nyquist Plot with overall error less than 3%. Based on the equivalent electrical parameter derived, sensing phenomenon of the material is explained. Multivariate analysis of the sensor states that sensor behaves differently at different frequencies in terms of impedance. Principal Component Analysis was used for the classification of 20 classes including five concentrations of each of four amines using different set of frequencies as features. Further, a classification model using SVM classifier trained with 40% training data was used to classify remaining test data and the performance of the model was evaluated using various evaluation parameters. The model demonstrates classification accuracy greater than 90% for classification of 20 classes. Also, optimized frequencies were selected by feature selection method via wrapper approach.