Date of Award
Fall 2012
Document Type
Thesis
Degree Name
Master of Science (MS)
Department
Electrical and Computer Engineering
First Advisor
Josse, Fabien J.
Second Advisor
Schneider, Susan C.
Third Advisor
Lee, Chung Hoon
Abstract
The shear horizontal surface acoustic wave (SH-SAW) device was investigated as a micro-chemical sensor for the direct, rapid and in-situ monitoring of chemical contaminants in groundwater and wastewater. The chemical contaminants of interest are organo-phosphate-based compounds (parathion, parathion-methyl, and paraoxon). The polymers used as partially chemically selective coatings are 2,2’-diallylbisphenol A - 1,1,3,3,5,5-hexamethyltrisiloxane (BPA-HMTS), and 2,2’-diallylbisphenol A- polydimethylsiloxane (BPA-PDMS), polyepichlorohydrin (PECH). The experimental results indicate that the analyte/polymer interaction is a result of mass loading and viscoelastic changes in the polymer. BPA-HMTS, BPA-PDMS, and PECH have the highest to lowest sensitivity towards parathion, parathion-methyl, and paraoxon respectively. The polymers tested at a thickness of 0.5μm show a higher (up to 20 kHz) response than the 0.25μm-thick polymers due to a more pronounced effect of mass loading and modulus change. The response times of the polymers from fast to slow are PECH, BPA-PDMS, and BPA-HMTS, and the order for the analytes is paraoxon, parathion-methyl, and parathion respectively. An array of sensors consisting of devices with different coatings and thicknesses is designed to increase sensor selectivity, which provides a specific pattern for each analyte. Based on visual pattern recognition, analyte identification is performed, using radial plots with the frequency shift and response times of BPA-HMTS, BPA-PDMS, and PECH at a thickness of 0.5μm as input parameters. The concentration of the unknown sample is determined using an algorithm that computes the unknown sample’s distance from the points of the known pattern in the three-dimensional space of the steady-state frequency shifts of the three 0.5μm-thick polymer coatings. The probability of the unknown pattern being identical to the known pattern can be determined by a level of confidence. The sensor is tested for stability, sensitivity, and reproducibility via subjection to a real life application simulation. The sensor is exposed to different interferents such as pH levels of 6.2, 7.0, and 8.0 and red clay. The sensor response to the analytes is shown to be unaffected by the presence of these interferents.