Discrimination of volatile compounds through an electronic nose based on ZnO SAW sensors

This work reports on the performance of an array of surface acoustic wave (SAW) sensors, using piezoelectric zinc oxide deposited by sputtering over silicon substrates, developed in order to detect low concentrations of different volatile compounds, such as octane, toluene, and methylethylketon. The array, developed in dual configuration, is composed by seven sensors coated by spray coating technique with diverse polymer thin films and a reference uncoated. Several pattern recognition techniques as principal component analysis (PCA), probabilistic neural networks (PNN) and partial least squares (PLS) have been used to explore the data distribution, classify the samples and predict the gas concentration. A good classification success rate and prediction have been achieved for the different tested gas types and concentrations.