Anode and Electrolyte Selection to Improve Hydrogen Recovery in Microbial Electrolysis Cells

Open Access
- Author:
- Zikmund, Emily Ann
- Graduate Program:
- Environmental Engineering
- Degree:
- Master of Science
- Document Type:
- Master Thesis
- Date of Defense:
- December 11, 2017
- Committee Members:
- Bruce E. Logan, Thesis Advisor/Co-Advisor
John M. Regan, Committee Member
Christopher Gorski, Committee Member
Patrick Joseph Fox, Committee Member - Keywords:
- Microbial Electrolysis Cell
Hydrogen
Ohmic Resistance
Electrode Spacing
Felt anodes - Abstract:
- A microbial electrolysis cell (MEC) is a bioelectrochemical system that generates hydrogen from organic matter. A biofilm of microbes grows on an anode and oxidizes the organic matter, and then the electrons are transferred to a conductive cathode to electrochemically produce hydrogen gas. Conductivities of wastewaters vary, influencing the solution resistance, and therefore affecting the extent of treatment, current, and gas production. In order to scale up MECs, it will be necessary to reduce the spacing between the electrodes (to reduce the reactor width per pair of electrodes), but to still produce the same amount of hydrogen per projected area. Brush anodes have shown greater power densities than felt anodes in microbial fuel cells (MFCs), as brush anodes can be placed closer to the cathodes than felt anodes without power being adversely affected. However, oxygen crossover is not a factor for current generation in MECs. In this study, flat carbon felt anodes were compared to carbon brush anodes to determine if current densities and hydrogen production rates were affected by the type of anode. The use of flat anodes could reduce the width of the anode chamber, increasing the overall hydrogen production rate per volume of reactor for the same current. In addition, the impact of solution conductivity was examined to see if higher electrolyte conductivities could improve hydrogen production rates in MECs, as they do for current in MFCs. MFCs with carbon brush in a 2 cm long chamber or carbon felt anodes in a 4 cm long chamber were batch fed with sodium acetate in a 50 mM phosphate buffer solution (PBS). Following acclimation to this medium, the MFCs with the felt anodes produced a maximum power density of 0.80 ± 0.02 W/m2, which was lower than that of the brush anodes (1.69 ± 0.10 W/m2). These power densities were consistent with previous studies for both types of electrodes. The anodes were transferred and acclimated to two chamber MECs with anion exchange membranes and platinum carbon cathodes in the same medium until similar current was produced for at least 3 cycles. The brush anode MECs had a higher hydrogen production rate of 0.38 ± 0.02 m3H2/m3d, with an average current density (current for 90% of the total charge transferred, I90) over the cycle of 4.2 ± 0.5 A/m2. The MECs with felt anodes had a lower hydrogen production rate of 0.32 ± 0.02 m3H2/m3d consistent with the lower current (I90 = 3.4 ± 0.1 A/m2). The extent of acetate removal for a cycle, based on the change in chemical oxygen demand (COD) was also greater for the brush anode MECs (89 ± 2%) than the felt anodes (71 ± 2%). The sharp decline in the current during a fed batch cycle of the MECs with the felt anodes suggested a diffusion limitation of substrate to the anodes, which could have reduced hydrogen production compared to brush anode MECs. To test this hypothesis, the anolytes of the felt MECs were stirred to improve mass transfer to the anode biofilm. Stirring increased the hydrogen production rate of the felt anode MECs by 8% to 0.41 ± 0.04 m3H2/m3d. The I90 and COD removal also increased to 5.1 ± 0.4 A/m2 and 92 ± 1%. By improving mass transfer into the felt anode in an MEC with mixing, the current density, COD removal, and hydrogen production rate became comparable to that obtained with brush anode MECs. When the solution conductivity was increased to 100 mM PBS in MFCs, the power density improved for the felt anodes (1.00 ± 0.04 W/m2), but it was still lower than that produced using brush anodes (1.25 ± 0.12 W/m2). The MFCs with brushes showed a decline in power compared to the 50 mM tests, likely as a result of the higher phosphate concentration or oxygen intrusion adversely affecting the microbial community. When the anodes from these reactors were transferred into MECs, the decreased solution resistance did not improve the hydrogen production rate or average current density for the MECs with brush (0.37 ± 0.10 m3H2/m3d, 3.9 ± 1.6 A/m2), but did for the felt anodes (0.40 ± 0.04 m3H2/m3d, 3.7 ± 0.4 A/m2). Increasing the solution conductivity reduces solution and membrane resistance, but only improved hydrogen production rates for the felt anode MECs.