Date of Award
Fall 2022
Document Type
Dissertation
Degree Name
Doctor of Philosophy (PhD)
Department
Electrical and Computer Engineering
First Advisor
Josse, Fabien
Second Advisor
Bender, Florian
Third Advisor
Schneider, Susan
Abstract
Direct liquid-phase sensing of aromatic hydrocarbons such as benzene, toluene, ethylbenzene, and m-, p-, o-xylene (BTEX) is of significant interest in environmental monitoring applications due to their severe impact on human health, and their prevalence in the products of the petrochemical industry. Moreover, each of these chemicals, including each xylene isomer, is utilized for petrochemical applications for which there are not currently viable replacements or cost-effective substitutions. As a result, isomer-specific identification is an important process-control capability for a number of industrial chemical separations, a family of processes that amounts to approximately 15% of total global energy consumption. In this dissertation, the detection and quantification of individual hydrocarbons within BTEX mixtures and potential interferent compounds at concentrations from 100-1200 ppb by volume is reported for a specifically designed polymer-plasticizer coating deposited on a shear-horizontal surface acoustic wave (SH-SAW) device. In addition, the isomer-specific detection and quantitation of m-, p-, and o-xylene and ethylbenzene, dissolved singly and as mixtures in aqueous solutions is also reported. The selected polystyrene-ditridecyl phthalate-blend coating was designed to maximize affinity for the target analytes by utilizing Hansen solubility parameters, with special consideration for the dipole moment and polarizability of the target analytes and coating components. Using the measured sensing parameters (sensitivities and time constants) extracted from multivariable sensor responses for each target analyte, multi-analyte sensor responses were processed with exponentially weighted recursive-least-squares estimation (EW-RLSE) to identify (with near 100% accuracy) and quantify (with ±5-11% accuracy) each target analyte, including the xylene isomers. This highly accurate estimation of target analyte concentrations has been achieved by combining the specifically tailored, high-sensitivity coating with an SH-SAW platform and by using the EW-RLS estimator. This sensor system is capable of estimating unknown parameters accurately even in the presence of significant measurement noise and interferent compounds. To better understand the potential of this sensor system for application in real-world environments, the RMS noise of several coated devices with similar coating compositions and thicknesses are also studied. Limits of detection are calculated for each coating/analyte pair, which indicates highly sensitive analyte detection in aqueous environments at very low ppb levels.