Date of Graduation


Document Type


Degree Name

Doctor of Philosophy in Chemistry (PhD)

Degree Level



Chemistry & Biochemistry


Julie Stenken

Committee Member

Ingrid Fritsch

Second Committee Member

Christa Hestekin

Third Committee Member

Colin Heyes

Fourth Committee Member

David Paul

Fifth Committee Member

Steve Tung


MEMS, Microdialysis, Microfabrication, Microfluidics, PDMS, Separations


A new microdialysis sampling method and microfluidic device were developed in vitro. The method consisted of using up to four microdialysis sampling probes connected in series to evaluate the relative recovery (RR) of different model solutes methyl orange, fluorescein isothiocyanate (FITC)-dextran average mol. wt. 4,000 (FITC-4), FITC-10, FITC-20, and FITC-40. Different flow rates (0.8, 1.0, and 1.5 µL/min) were used to compare experimentally observed relative recoveries with theoretical estimations. With increasing the number of probes in series, the relative recovery increases and ~100% (99.7% ± 0.9%) relative recovery for methyl orange was obtained. For larger molecules such as fluorescein isothiocyanate (FITC)-dextran average mol. wt. 4,000 (FITC-4), FITC-10, FITC-20, and FITC-40, RR of 66.3% ± 0.0%, 39.4% ± 0.6%, 18.7% ± 0.1%, and 7.7% ± 0.1%, respectively, were obtained using four microdialysis sampling probes in series at 0.8 µL/min. Using theoretical estimations, the number of microdialysis probes in series needed to achieve 99% RR was determined for each solute. The theoretical estimations started deviating from experiments at mol. wt. 10,000 (FITC-10). For example, the deviations from experiments for FITC-10, FITC-20, and FITC-40 were +52%, +149%, and +179% respectively. On the other hand, methyl orange and FITC-4 theoretical estimations were closer to the experiments (-1%, underestimation, and +15%, overestimation). The method developed for this dissertation was miniaturized in a polydimethylsiloxane (PDMS) microfluidic device having a flat polyethersulfone membrane and seven micro-channels connected in series. Push-pull experiments determined that the optimal setting for this microfluidic device prototype during the collection of methyl orange was 0.2 µL/min-1.0 µL/min. The relative recovery of methyl orange using this setting was 78.8% ± 2.5%. This result indicated that a working microfluidic device prototype was developed. Further optimizations need to be performed to reach the same level as the microdialysis probes in series method. All the work conducted to achieve the development and miniaturization of the microdialysis probes in series approach is presented in this dissertation.