Modification of microbial electrolysis cells (MECs) by altering the sizes of the anode and the cathode, or their interaction with the liquid stream affects the performance of the MEC, but it remains difficult to quantify how much each change will contribute to the overall performance. Through this study, a method for quantifying the effects of design decisions on internal resistance was applied for the first time to MECs. The anode surface area (15 cm2) was the most limiting factor when phosphate buffer concentrations were greater than 50 mM. Increasing the anode surface area by layering carbon cloth was found to be an effective way to relieve this hindrance. The cathode surface area contributed less to the internal resistance, due to the use of a highly effective Molybdenum Phosphide (MoP) catalyst. Additionally, inducing fluid motion between the electrodes enhanced performance, whereas varying the spacing of the electrodes (all spacings < 1 cm) was found, surprisingly, to have little effect.
The maximum cathodic current density achieved with this design was 47 A/m2, with a hydrogen (H2) production rate of 3.7 L/ L-liquid volume/day. Highest performance was obtained with an anode to cathode surface area ratio of just larger than 3.5 : 1, which was the optimal ratio predicted through the internal resistance dissection procedure. The internal resistance relation was used to adequately predict the electrical performance of two MECs from other studies. This method gives an in-depth understanding of the distribution of internal resistance, which enables design of MECs and their operational conditions for high performance.
MECs with volumes greater than one liter have been recently developed and tested, with a few demonstrating high current density and H2-production rates. However, many of the designs shown to give the best performance are difficult to scale-up further. In the present study a highly scalable 10 L MEC was designed and evaluated. The shape of the reactor and the orientation of the electrodes encourage high fluid mixing, and the sheet metal electrode frames provide both structural support and distributed electrical connection. Solid-core insulated copper wires were found to be suitable for bringing the concentrated electrical current to and from the electrode frames. With a comparable surface area to volume ratio (30 m2/m3), the current density (970 A/m3) was higher than all MEC reactors of the same size range, to the best of our knowledge. The observed current density was accurately predicted based on the internal resistance results of small scale MEC testing (0.15L), demonstrating that the electrical current scaled directly from the smaller scale. A volumetric H2-production rate of 5.9 L/L/d was achieved. Further analysis provided some evidence that location of an electrode pair next to a reactor wall decreased current density (~ 20%), as did separating the electrodes with J-Cloth and narrower electrode spacing, rather than without a separator (~30%).