Quantitative Modeling of Microbial Fuel Cells for Water Purification

Although the 2015 Millennium Development Goal (MDG) for access to safe drinking water has largely been achieved, over 700 million people cannot access safe drinking water [1].  More importantly, the MDG for water sanitation is is far from achieved: 2.5 million people, primarily in sub-saharan Africa, lack access to water sanitization. 

One solution for these populations is low-cost, mobile water purification based on microbial fuel cells (MFCs). MFCs are capable of producing energy from organic water contamination by oxidizing the contaminants, thereby producing electricity and reducing the chemical oxygen demand (COD) a key measure of water contamination, by more than 90% [2]. However, such a system would need to meet stringent cost and durability requirements; a self-powered purification system would require an approximately 10-fold increase in typical MFC power density and lifetime. Moreover, bench scale MFCs of 0.1 L size would would need to be scaled up to over 10 L for practical applications.  These requirements can only be met by through careful engineering design and optimization. 

Here we describe a numerical model of the the steady-state and transient operation of a microbial fuel cell for water purification. The model is based on laboratory-scale experiments conducted on MFCs using the Dissimilatory metal reducing bacteria Shewanella oneidensis MR-1 [3]. The model accounts for growth of bacterial biofilms, transport of reactants and products within the films, and related consumption of organic contaminants to electricity production. 

Particular attention is paid to biofilm composition, and electron transport mechanisms with the biofilm, including electron transport via dissolved mediators, pili and nano filaments, and surface exchange. These considerations allow the the model to applied to a variety of metal-reducing species. After calibration with laboratory-scale MFC data, the model is used to predict performance of 10L-scale MFCs, including the relationship between decontamination rate, water flow rate, and power generation. Sensitivity studies may be directly compared with principal component analysis of experimental results, indicating the degree to which the model can be used for design and optimization of real-world devices.

1. "Millennium Development Goal drinking water target met”, http://www.who.int/mediacentre/news/releases/2012/drinking_water_20120306/en/
2. M. M. Ghangrekar and V. B. Shinde, Bioresour. Technol., 98, 2879 (2007). doi:10.1016/j.biortech.2006.09.050
3. E. Baranitharan, M. R. Khan, D. M. R. Prasad and J. Bin Salihon, Water, Air, Soil Pollut., 224, 1533 (2013). doi:10.1007/s11270-013-1533-1
4. S. Babanova, O. Bretschger, J. Roy, A. Cheung, K. Artyushkova and P. Atanassov, Phys. Chem. Chem. Phys., 16, 8956 (2014). doi:10.1039/c4cp00566j