4.3.2. Effect of Design Parameters on Sensor Performance

Solution in terms of fE.The solution of Eq. (15) can be rewritten in terms of the enzyme loading factor fE:

Therefore,

Sensor Output for Diffusion Control. When fE > 25 (criterion for diffusion control), i.e.,

Key Design Parameter. Note that the keydesign parameter is fE:

To obtain diffusion control, how does one adjust Vmax, d, and Ds ? Which variable is most effective?

Response Time. The response time of the sensor will be proportional to the diffusion time of the product d:

Enzyme Reserve. The variation of the enzyme loading is a means of determining the minimum amount of enzyme required for maximum sensitivity. Furthermore, this test should reveal the magnitude of the enzyme reserve of diffusion controlled sensors.

Loading Test. Fig. 4.4. shows the results of a loading test of GOD entrapped in a gelatin layer of 30 µm thickness between two dialysis membranes of 15 µm thickness each. The stationary currents for 0.14 mmol/L glucose (lower part of the linear measuring range) and for 5 mmol/L glucose (saturation) increase linearly with enzyme loading from 46 mU/cm2 to 1 U/cm2. At higher GOD loading a saturation value is attained. To calculate the enzyme loading factor, fE, the following values have been used:

Parameters used

• d = 30 m
• KM = 10 mM/L
• DS = 1.63x10-6 cm2
• A = 0.22 mm2
• Polarization: +0.6V

As is evident from Fig. 4.4, the transient from the linear region to saturation occurs at fE values between 7 and 20. This agrees with the theoretically predicted value and indicates that above 1 U/cm2 the function of the GOD electrode is controlled by internal diffusion.

B. Concentration dependence of signal output

Linear Range . The linear measuring range of biosensors extends over 2-5 decades of concentration The lower detection limit of simple amperometric enzyme electrodes is about 100 nmol/L whereas potentiometric sensors may only be applied down to 100 µmol/L. This shows that the sensitivity is affected not only by the enzyme reaction but also by the transducer.

Oxygen Effect. The linear range extends to 2 mmol/L glucose in the measuring cell. In this region, saturation of the measuring solution by oxygen increases

the measuring signal by only 10%. At low glucose concentration the cosubstrate concentration (ca. 200 mol/L at air saturation) influences the enzyme reaction only slightly. By contrast, in the saturation region above 2 mmol/L glucose the current rises by a factor of 4.5. At the same time the linear range is extended by oxygen saturation (see Fig. 4.5).

C. Effect of pH on sensor output

With a high enzyme excess in the membrane, pH variations should have only a minor influence on the measuring process. Therefore the pH profiles in the linear measuring range and under diffusion control should be substantially less sharp than those of the respective enzyme in solution. The results obtained with a GOD-gelatin membrane (Fig. 4.6) agree with this assumption. With 0.14 mmol/L glucose the curve is almost as flat as that of the H2O2 signal. On the other hand, with 10 mmol/L a pronounced maximum is found. At this saturating concentration, the signal depends on the enzyme activity and therefore distinctly on pH. The pH optimum of immobilized GOD is about 0.9 pH units more alkaline than that of the soluble enzyme. Obviously the formation of gluconic acid within the enzyme membrane causes a local pH decrease, shifting the optimum to higher pH in the solution.

D. Effect of temperature on sensor output

The rate of enzyme reactions rises with temperature up to a certain optimum. Above that, the effect of thermal inactivation dominates over that of the increase of the collision frequency. Enzyme stabilization by immobilization is frequently reflected by an increase of the temperature optimum for substrate conversion. If kinetic and diffusion control are superimposed, the higher activation energy results in a predominant acceleration of the enzyme reaction with rising temperature. Thus, the slower enhancement of the diffusion rate makes mass transfer the limiting factor. Therefore, the activation energy determined at lower temperatures is ascribed to the enzyme reaction, and that at higher temperatures to diffusion. Besides this, the temperature profile is affected by temperature-dependent conformational changes of the enzyme and decreasing solubility of the cosubstrate. The glucose sensor with the GOD-gelatin membrane exhibits a temperature optimum of about 40_C . Below the optimum the Arrhenius plot (Fig. 4.7) gives parallel straight lines for different glucose concentrations and enzyme loading. The difference between the activation energy of H2O2 diffusion, 33.5 kJ/mol, and that of GOD-catalyzed glucose oxidation, 25.5 kJ/mol, is probably too small to give rise to two separate linear regions. That is why purely diffusion controlled GOD

electrodes are not significantly different from kinetically controlled ones with regard to activation energy.