EFFECT OF CARBON FIBER BRUSH ANODE SURFACE AREA ON SINGLE CHAMBER MFC START UP TIME AND INTERNAL RESISTANCE
Open Access
- Author:
- Hutchinson, Adam Joseph
- Graduate Program:
- Environmental Engineering
- Degree:
- Master of Science
- Document Type:
- Master Thesis
- Date of Defense:
- April 06, 2011
- Committee Members:
- Bruce Ernest Logan, Thesis Advisor/Co-Advisor
Bruce Ernest Logan, Thesis Advisor/Co-Advisor - Keywords:
- carbon fiber brush anodes
electrochemical impedance spectroscopy
Microbial Fuel Cell
internal resistance
surface area - Abstract:
- The use of carbon brush anodes in single chamber air cathode microbial fuel cells (MFCs) has significantly increased power generation. It has been shown that the additional surface area and porous structure of these anodes are the characteristics that increase power generation. Here it was shown that this large amount of anode surface area is much more beneficial for MFC operation at startup than it is once MFCs have been operated for extended time periods. The startup time for MFCs using full brush anodes and phosphate buffer solution (PBS) concentrations of 50 mM, 25 mM, and 8 mM was 8 ± 1 days. Startup time increased when 75% of the brush anode was removed to 13 days in MFCs using 50 mM PBS, 14 days using 25 mM PBS, and 21 days using 8 mM PBS. Once MFC operation and the microbial community has been established, up to 75% of the anode brush material can be removed without observing a significant decrease in power generation. MFCs using a full brush anode produced a maximum power density of 1050 ± 140 mW/m2, and when 75% of the anode was removed the power density decreased to 770 ± 70 mW/m2. Power density, total internal resistance, and distribution of total internal resistance were examined as functions of solution conductivity, electrode spacing, and amount of brush anode material. Acetate-fed MFCs with 50 mM substrate buffer solution, a spacing of 0.4 cm between the front of brush anode and the cathode, and a full brush anode, produced the largest power density of 1280 mW/m2. MFCs with 8 mM PBS, 1.4 cm electrode spacing, and 75% of brush anode removed produced the lowest power density of 387 mW/m2. Each time power density was determined internal resistance was also determined using two different methods: the slope of the polarization curve; and summation of the component resistances from EIS results. Power density was plotted versus one over internal resistance to examine if there was systematic variation in measured resistances. A plot of all data showed that there was no bias in either method determining internal resistance related to the power density produced. Given a certain internal resistance the power density can be estimated with 95% confidence using the linear relationship. Electrochemical impedance spectroscopy (EIS), conducted with the MFC under an external load, was used to identify the specific contributions of solution, charge transfer, and diffusion resistances to overall internal resistance. The results show that charge transfer resistance was the same at both electrodes within a standard deviation for MFCs with different solution conductivity, electrode spacing, and amount of anode material. Anode charge transfer resistance (ARCT) was 21 ± 5 Ω for MFCs that produced the highest power densities (50 mM PBS, 0.4 cm electrode spacing, 0% brush anode removed), and 26 ± 5 Ω for MFCs that produced the lowest power density (8 mM PBS, 1.4 cm electrode spacing, and 75% of the brush anode removed). Similarly these variations did not affect the cathode charge transfer resistance (CRCT), as evidenced by CRCT values of 19 ± 5 Ω for MFCs that produced highest power densities and 21 ± 5 Ω for the MFCs that produced the lowest power densities. The primary sources of internal resistance were solution (RS) and diffusion (RD), which together increased by 100 Ω from highest to lowest power density conditions. Internal resistance values determined by EIS were in agreement with values determined by the method of using the stable region of the polarization slope. The agreement of the summed EIS and polarization slope values suggests that EIS is useful for identifying distinct sources of internal resistance, and that it does not miss any critical components of internal resistance. Taken together, these results show that carbon fiber brushes with less material than current models could be used to achieve high power densities and the primary obstacles to decreasing internal resistance are in the solution conductivity which impacts solution and diffusion resistances.