Multi-sensor array used as an "electronic tongue" for mineral water analysis

An "electronic tongue" based on a monolithically integrated array of chemical sensors is presented. The device is composed by six independent ion selective field effect transistors (ISFETs), an interdigitated platinum electrode (IDS) and a silicon diode used as a temperature sensor. K+, Na+, Ca2+ and Cl- ISFET based sensors were obtained by depositing different photocurable membranes onto their gates. IDS was used to measure conductivity and redox potential. The multi-sensor was used to classify 13 different brands of mineral water. Several pattern recognition methods such as hierarchical clustering analysis (HCA) and principal component analysis (PCA) were applied to the multi-sensor output data sets. They showed the possibility of this electronic tongue to distinguish well between different classes (HCA) as well as different brands (PCA) of waters. Eventually, soft independent modelling class analogy (SIMCA) algorithm was used to construct the classification model based on experimental data sets. The model applied to the test data provided a 100% correct prediction of the water brand, giving possibility to perform fast and cheap classification or quality analyses. (c) 2006 Elsevier B.V. All rights reserved.