Microfluidic electrochemical sensors for the selective detection of toxic compounds

This thesis details the use and optimization of several different analyte selective, liquid based detection chemistries, as well as the micro-fluidic device platforms which enable these sensors to be operated in a practical, effective manner. The majority of the work in this thesis involves an aqueous based chemistry utilizing an oxime reagent for the detection of acetylcholinesterase inhibitors. However, two other liquid chemistries based on the principles of chemical amplification were also studied in great detail. Research first focused on improving the life time of the oxime reagent in solution. The oxime reagent decomposition in solution was studied electrochemically as well as by solid phase micro-extraction (SPME). Through these data, both an empirical degradation kinetic rate law and likely decomposition mechanisms were established. This information then allowed a degradation suppression scheme to be developed. By dissolving the oxime in an organic solvent, it could be prevented from dissociating, thereby greatly suppressing degradation. The concentrated organic oxime solution could then be combined with a basic aqueous buffer just prior to being used in the detection process, thus suppressing degradation without significantly affecting electrochemical performance. Ethanol was chosen as the optimal organic solvent as it was able to increase the oxime life time in solution 10 fold, while causing minimal decreases in sensor sensitivity and response speed. Finally, in order to mix the aqueous and organic solutions on chip, a serpentine channel micro-mixer was created with the same dimensions and material as the original micro-sensor to allow for seamless integration. The micro-fluidic platform previously developed for this oxime based chemistry yielded a sensor able to rapidly and selectively detect acetylcholinesterase inhibitors. A large variety of possible gaseous sensor interferent compounds were evaluated on this oxime based micro-sensor and none were found to produce any response from the device. Further characterization of this micro-sensor showed that although its response speed was fundamentally limited by the reaction rate of oxime with the analyte, this limitation had a negligibly affect on sensor performance if the device was coupled to a chromatography column. The rapid, selective nature of the oxime micro-sensor allowed for quick evaluation of individual compounds separated by the GC column; only responding to acetylcholinesterase inhibitors even if they are co-eluted with interferent compounds. Finally, this oxime micro-sensor was further miniaturized and created via a silicon micro-fabrication process. This allowed it to be integrated with 7 single walled carbon nano-tube (SWNT) based sensors on a single chip to form a complete micro-sensor array. The other liquid chemistries studied in this thesis are classified as Non-Biological Inhibition Based Sensing (NIBS) mechanisms. These detection schemes utilize the inhibition of a catalyst with the target analyte to dramatically slow the rate of a main reaction, thus chemically amplifying the presence of a single molecule by reducing the production of hundreds or thousands of molecules. Two such examples of this NIBS mechanism have been previously developed; one for the detection of toxic sulfides and one for the detection of the riot control agent Adamsite. The final chapter in this thesis is a critical evaluation of these two NIBS mechanisms which includes a more detailed study of the chemistry involved, possible issues which hamper the detection of the target analytes and recommendations for future research using these types of sensing mechanisms