Thin layer bioelectroanalysis

The suitability of the octagonally coordinated complex, ruthenium (III) hexamine, as a soluble electron mediator for glucose oxidase in the electrochemical capillary fill device (CFD) was investigated. Electrochemical studies revealed highly reversible electrochemistry at gold electrodes with E1/2=-0.2 V and a diffusion coefficient of ~1.7x10*9 m2 s-1 (2 M KCI, pH 7). The second order rate constant, ke, for the reaction between reduced glucose oxidase and oxidized mediator could not be determined by voltammetric sweep techniques, indicating a value for ke far inferior to the ferrocenes. A numerical analysis of the difference between the current produced in the presence of the homogeneous catalytic reaction under pseudo first order conditions and the diffusion current yielded a value of kefor ruthenium (III) hexamine of -80 M-1 s'1. The maximum pH-dependent response for the reaction between ruthenium (III) hexamine and glucose oxidase occurred at pH 9. There was no increase in the rate of electron transfer at elevated pH values intrinsic to the ruthenium (Ill/ll) couple, from studies of the heterogeneous rate constant & employing the Marcus relationship. The increased reactivity at pH 9 was thought to be due to enhanced thermodynamic viability, shown by an increase in the difference in between the half wave potentials of the FAD/FADHg & H u^^^(N H 2)^ couples. Exhaustive coulometric determination of glucose in a CFD using ruthenium (III) hexamine and glucose oxidase resulted in a response with a linear range up to 30 mM glucose (r2=0.999, cv<5%, n=100) and a limit of detection of 0.2 mM. Strategies for a reduction in the total analysis time were explored. Numerical models were constructed to simulate the behaviour of the catalytic phase of the analysis. Modelling the flux of glucose and reduced mediator to, from and within the catalytic layer showed that the low value for kg was not flux limiting and suggested that a reduction in the duration of the catalytic phase could not be achieved by increasing the concentration of enzyme in the layer. The models suggested that a dramatic improvement in response time could be achieved by reducing the gap width of the CFD and emphasised the importance of a high fractional diffusion coefficient in the catalytic layer. Correlative parameters based on a cell dimension and diffusion coefficient dependent time constant and the Km of the enzyme generally applicable to catalytic devices of this type were constructed. The use of a series of potential pulses separated by potential relaxation periods in the CFD to derive analytical information was investigated theoretically using analytical models and a series of numerical models, and practically using manufactured devices. The method allowed a reduction in analysis time in the glucose CFD using ferricyanide as mediator from 60s to 10 s with an increased limit of detection (to 0.1 mM from 0.5 mM) and a greater precision (cv from 8% to 4%) without adjustments to the reagent layer. The models were also used to solve the flux equations for lateral diffusion, demonstrating the redundancy of the guard electrode using this method of current signal analysis. The amenability of the CFD design to accommodate complex chemistries was demonstrated for the Llactate dehydrogenase/NAD/diaphorase/ferricyanide system.