Enzymatic Biological Fuel Cell    |     home
Introduction/Background   |   Design   |   References   |   Results   |   Downloads   |   Pictures   |   Members   |   Links
Results

Conventional Fuel Cell

Varying Concentration of Electrolyte Solution
The relationship between the concentration of the potassium hydroxide electrolyte solution and power density was determined by comparing the power density
Figure 1. Voltage vs. Time for varying concentrations of potassium hydroxide solution.
produced by an ethanol fuel cell for varying concentration of the electrolyte solution over time. The experiments were run at room temperature with electrolyte concentrations of 1M, 2M, 4M, and 6M. The power density obtained from the different concentrations of potassium hydroxide showed steady state values of 18.25 ± 0.18 mW/m2 for 1M, 19.66 ± 0.20 mW/m2 for 2M, 20.56 ± 0.21 mW/m2 for 4M, and 25.26 ± 0.25 mW/m2 for 6M. These results indicate that power density increased with increasing electrolyte concentration (Figure 1). This conclusion was expected since an increase in potassium hydroxide concentration would also increases the rate that the fuel is oxidized, thus increasing the cell's performance.

Varying Type of Fuel: Methanol vs. Ethanol
     To determine how the type of fuel affects the power density, experiments were performed at room temperature using either ethanol or methanol fuel in 1M potassium hydroxide electrolyte solution. The results indicate that ethanol produced a higher power density of 18.25 ± 0.18 mW/m2 as compared to the value of 16.57 ± 0.17 mW/m2 obtained from methanol (Figure 2). This result is contrary to what is stated in the Mini Fuel Cell manual that was included in the Fuel Cell kit that was utilized for the experiments. The manual indicated that
Figure 2. Power Density vs. Time for Methanol and Ethanol Fuel
since ethanol fuel undergoes only one stage of oxidation instead of the three stages of oxidation associated with methanol, it should produce a smaller voltage and thus a smaller power density (Sweet, 2003).
      However, from the Gibbs Equation, DG°= -nFx°, where DG° is the change in gibbs free energy at standard conditions, n is the number of electrons associated with the redox reaction, F is the Faraday's constant, and x° is the cells potential at standard temperature and pressure, the cell's potential for ethanol and methanol were calculated to be 1.975V and 1.21V respectively.  This equates to a theoretical power density of 176.56mW/m2 for ethanol and 66.28mW/m2 for methanol assuming that the anode area and resistance are equal to that used in our experiment. Although these values are significantly larger than those obtained through our experiment, they indicate that ethanol should produce a larger power density than methanol.  This was observed in our experimental results. Discrepancies between the values of power density obtained from theory and our results may arise from impurities in the fuel or electrolyte, which can considerably lower the cell's performance (Bockris et al. 1969).

Varying Temperature of the Fuel Cell
     To determine the effect of temperature on the performance of the fuel cell, two experiments were performed, one at room temperature and the other at 45 °C.
Figure 3. Power Density vs. time with varying temperature of electrolyte and fuel.
For the experiment at 45 °C, a water bath was utilized to raise and maintain the temperature of the 1M potassium hydroxide electrolyte solution and ethanol fuel. The results suggest that increasing the temperature decreases the power density. For the experiment at 25°C, a power density of 18.25 ± 0.18 mW/m2 was obtained while the experiment at 45°C yeilded a power density of 16.84 ± 0.17 mW/m2 (Figure 3).
     Through the Gibbs equation, which was defined above, the theoretical voltage produced by the fuel cell for 25°C and 45 °C is 1.975V and 2.08V respectively. This corresponds to a power density of 176.56mW/m2 for the cell at room temperature and 195.85mW/m2 for the cell at 45°C assuming that the anode area and resistance are equal to that used in our experiment. These numbers indicate that the fuel cell 45°C should produce a higher power density than the fuel cell 25°C, which was not observed in our results. This could be due to the harsh conditions that the fuel cell was subjected to in this experiment. Since the fuel cell was not designed to operate at temperatures higher than 25°C, the seal between the cathode and anode degraded during the experiment, which allowed the cathode to be partly covered by the electrolyte solution. This lowered the amount of oxygen available for reduction thus lowering the performance of the cell.


Consistency in results:
Figure 4. Power Density vs. Time for 1M and 6M electrolyte solutions performed twice for each case in order to determine if results are reproducible.

     To determine if the results obtained were reproducible, two sets of experiments were performed twice using ethanol fuel with either a 1M or 6M potassium hydroxide electrolyte solution at 25°C. For the experiment that utilized the 1M-electrolyte solution the average difference in the power density of the two runs was 0.10 mW/m2.  For the 6M electrolyte case this value increased to 1.97 mW/m2. Since these values are only a small fraction of the actual power density measured, for the 1M-electrolyte case for the 6M electrolyte case, the results indicate that the data is reproducible (Figure 4).











Biological Fuel Cell

pH of the Anode Solution
     Experiments were performed to determine the optimal pH for the solution contained in the anode compartment. In order to increase the pH from the initial value of 0.7
Figure 5. Graph of power density vs. time for varying values of pH.
associated with the lactic acid, a 1M solution of sodium bicarbonate was utilized. Three values of pH were tested, 4, 7 and 8, in order to observe the affects of an acidic, neutral and basic solution. As can be seen from figure 5, the fuel cell utilizing the solution with a pH of 7 produced the largest power density with an average value of 4.68 ± 0.05 W/m2 while the solution with a pH of 4 and 8 produced a power density of 3.00 ± 0.03 W/m2 and 1.24 ± 0.01W/m2 respectively. Since the acidic solution produced a power density 2.41 times higher than the basic solution, the effect of extraneous hydroxide ions as compared to hydronium ions appears to be more detrimental to the power density produced from the fuel cell. This is reasonable since the formation of PQQH2 is hindered by the excess hydroxide ions in the second reaction that occurs in the anode. The decomposition of this product is needed to produce the two electrons that travel to the cathode (see Table 1). Although the first reaction at the anode is hindered by the surplus hydronium ions, the effect does not appear to be as significant as that seen with the hydoxide ions.


Temperature of the PQQ solution
Figure 6. Graph of rower density vs. time for varying temperatures for the PQQ solution. The pH of the anode solution was kept at a value of 4 for both trials.

Since the PQQ is refrigerated between uses, the effect of utilizing a solution of low temperature as opposed to a room temperature solution was explored in order to see if it was necessary to warm the mixture before the PQQ bonding step. As can be seen from the figure 6, the effect of using the room temperature mixture as opposed to the approximately 4°C mixture is minimal since both mixtures equilibrated to a power density of approximately 3.00 ± 0.03 W/m2. So it is possible to utilize the chilled mixture in bonding the PQQ onto the crystamine modified anode. The hump in the data from the fourth to fifth hour for the solution at room temperature may have occurred from movement of the cathode or the fuel cell.


Addition of Calcium Ions
     The effect of calcium ions on the power density produced from the fuel cell were observed by performing four experiments that varied the concentration of calcium ions from 0 mM to 5 mM. The effect of adding calcium ions appears to be disadvantageous since all of the experiments that contained calcium ions produced a power density that was lower than the trial that did not (Figure 7).
Figure 7. Graph for power density vs. time for varying concentration of calcium ions. The pH of the anode solution was kept at 7 for all runs.
The power density associated with a calcium concentration of 0mM, 1mM, 3mM, and 5 mM are 4.68 ± 0.05 W/m2, 2.01 ± 0.03 W/m2, 1.70 ± 0.02 W/m2, and 2.58 ± 0.03 W/m2 respectively. Theoretically this result appears to be counterintuitive since the calcium ions should stabilize the reduce complex of PQQ thus pushing the redox potential towards the oxidation of NADH. Since this extra push will lead to the increased formation of PQQH2 and thus the production of more electrons, the performance of the cell should be increased. A possibility for this ineffectiveness may have arisen from inadequate mixing in the anode compartment.

Number of Membranes

As indicated previously, the dialysis membrane is incapable of restricting the flow of NAD+ out of the anode compartment due to its low molecular weight. To determine how the flow of NAD+ out of the anode compartment can be limited, an experiment that increased the number of dialysis membranes was performed. As indicated in figure 8, increasing the number of dialysis membranes from one to two, decreased the average power density from a value of 4.68 ± 0.05 W/m2 to 1.38 ± 0.01 W/m2. Thus increasing the number of dialysis membrane does not restrict the flow of NAD+ adequately enough counter the effects of ohmic resistance and concentration gradients.

Figure 8. Graph of power density vs. time for varying number of dialysis membranes. The pH of the solution was kept constant at a value of 7 for both experiments.