Skip to Main Content
In this paper, we have studied the selectivity and sensitivity of an array of porous silicon sensors having different porosity and pore morphology for sensing organic vapors like methanol, ethanol and isopropyl alcohol. The output of the array when exposed to the individual and mixture form of vapors give a unique fingerprints of the vapors analyzed by multilayer perceptron based pattern recognition technique with principal component data analysis. Experimental results show that the response of the array to nonpolar molecules is negligible while the response to the polar vapor molecules is significant. The pattern recognition engine estimates the concentration of the identified vapor with almost 100% accuracy.