Combining Brain Power with Microbial Oxidising Power

Microbial Fuel Cells (MFCs) are a two chambered system used to generate an electrical current by utilizing the oxidising power of microorganisms. Bacteria are kept in an anode chamber and separated from the cathode by a polymeric proton exchange membrane. The cathode is usually aqueous and generated by bubbling oxygen through water in order to dissolve the oxygen. MFCs have been around for a relatively long time (early 20^th Century) and have suggested applications in electricity generation in developing countries, however due to the extremely low current (many cannot achieve higher than 2milliamps) research has slowed in this area. Feng et al (2013) have demonstrated an alternative use for MFCs when using them in conjunction with Artificial Neuron Networks (ANNs).

ANNs are mathematical models utilizing several interconnected artificial “neurons” (digital neurons, not physical real neurons!), which are capable of identifying complex non-linear relationships. Through the use of intermediary stations or nodes, several different inputs can be sorted into the correct output based on a series of rules or parameters. This is likened to the vast network of neurons in a brain. ANNs are mostly used in computing for inconsistent or conflicting inputs, but are also used in robotics for pattern recognition. Biologically ANNs have applications in genetic computer models to predict how alterations in single genes will effect whole populations over a period of time.

In this study, Feng et al (2013) explored the idea of combining MFC biosensing with ANNs in order to identify specific chemicals present in the water column for use in the water purification industry.  They find accurate identification of all substrates used as a test of the method.

As a summary of this paper, I will admit that it is poorly worded and VERY confusing! However looking past that, it is easy to see the methodology used here should be deemed a success and the potential application of this in the water purification industry is significant. Accurate identification of pollutants or compounds can lead to more efficient preventative measures and identification of the source of the contamination leading to healthier water and thus healthier people.

 REF: Feng, Y., Barr, W. & Harper W.F. Jr. (2013) Neural network processing of microbial fuel cell signals for the identification of chemicals present in water. Journal of Environmental Management, 120, 84-92.

Accessed From: http://www.sciencedirect.com/science/article/pii/S030147971300056X