The microbial fuel cell (MFC) is a technology in which microorganisms employ an electrode (anode) as a solid electron acceptor for anaerobic respiration. This result in direct transformation of chemical energy into electrical energy, which in turn means that organic wastewater can be used as fuel. Amongst the various organic wastes that are employed as fuel for MFCs, urine is of interest since it represents 75% of the nitrogen present in domestic wastewaters and yet only 1% of the total volume. However, here is a persistent problem for the scaling up of MFCs. The smaller the surface of electrode to volume ratio of an MFC is, the higher its power density is. Hence, to reach usable power levels, a plurality of units needs to be assembled in stack, which implies configuring both the hydraulic circuitry and the serial/parallel electrical connection patterns. Because of this plurality, the units need to have a simple design for the whole system to be cost-effective. The goal of this work is to address how to build these multiple MFCs in stack.

We report a novel membrane-less stack design using manifold ceramic plates semi-submerged with anodes and cathodes sharing urine solution. The top half of each plate were semi-submerged were covered in cathodes while the anodes, on the lower half of each plate, were fully submerged. The MFCs in each box were connected in parallel and multiple modules were configured in series and placed. This allowed  self-stratification of the collective environment (urine column) under the natural activity of the microbial consortia thriving in the system. For size comparisons, the module footprints were enlarged from 900 mL to 5000 mL and, importantly, this scaling-up did not negatively affect power density (» 19 W/m^3), a factor that has proven an obstacle in previous studies. However, we should note that this observation is limited for the dimensions we tested. This was achieved by maintaining a plurality of microenvironments within the collective system and resulted in a simple, robust system fuelled by urine. The developed system was then used to treat urine of single individual (» 2.5L per 24h), in order to serve as the energy source charging mobile phones. Six boxes, each comprising 20 MFCs connected in parallel, were hydraulically cascaded and connected in series. Best performances were obtained in pulse-feed batch mode. At equilibrium, the stack of six boxes was feed with 600 mL every 6h at a flow rate of 2 L/min and was producing between 110 and 125 mW. Under these conditions, a smart phone (Samsung Galaxy S, battery of 1600 mAh) was fully charged in 68 h and 80 h, when turned off or on respectively. It has to be noted that these performances were stable for a period of 8 months. We concluded that our scaling-up approach within the tested range was successful to convert chemical energy in urine to electricity and this was demonstrated by charging cell phones.