Date of Award

Winter 1983

Project Type


Program or Major


Degree Name

Doctor of Philosophy


Analytical systems used in a clinical setting should offer high sample throughput and eae of operation, while at the same time providing selective, sensitive, accurate and precise determinations of clinically important analytes. Described in this study are two flow systems based on chemiluminescence (CL) detection that provide the features important in a practical clinical method: a flow injection analysis (FIA) system and a two phase flow system. Both systems take advantage of the specificity of enzyme reactions.

FIA systems for the determination of Cr(III) and glucose, based on luminol CL were developed. The Cr(III) system used EDTA as a masking agent to eliminate interferences from other metal ions, making it selective for Cr(III). The FIA system for glucose determination in biological fluids was based on the use of glucose oxidase, which was covalently immobilized to controlled porosity glass beads. The immobilized enzymes were packed into a column incorporated into the FIA system. Optimum operating conditions, detection limits and linear ranges were determined. The system was capable of sampling rates as high as 240 samples/hour. There was no significant differences at the 95% confidence level between CL-FIA results for blood glucose determination and results obtained by a commercially available reference method. The immobilized glucose oxidase was quite stable and could be reused for many assays.

The two phase flow system work employed a custom-designed flow cell that featured enzymes immobilized by entrapment by a membrane. Analyte in the flow stream could diffuse under a concentration gradient across the membrane into the reagent cell. Theory describing the performance of the cell was developed. The system was characterized using the peroxidase-catalyzed luminol reaction, firefly bioluminescence (BL) and bacterial BL. The kinetics of the luminol and firefly reactions were varied in order to observe the effect of reaction rate on response time. Rise and fall times to steady state emission were measured, as were the relative intensities for the various reaction conditions. Rise times as fast as 40 seconds were observed for the firefly reaction when 10 mg/mL of apyrase was added to the system. The bacterial BL system exhibited much slower response than the modified firefly systems, but the bacterial system was more stable over time. Response time in the bacterial system was a function of NADH concentration. The performance of these systems for ATP and NADH determination was investigated, and approximate detection limits and linear ranges were determined.