In search of experimental evidence for the biogeobattery

Recent work has suggested that the electrical self-potential (SP) geophysical technique may be used to non-invasively map redox conditions associated with contaminant plumes or bioremediation schemes. The proposed mechanism linking SP response and redox involves the generation of a current source and sink in the subsurface whereby electrons are transferred between anoxic and oxic environments via a conductive biofilm and/or biominerals, creating a biogeobattery. Here, we seek direct experimental evidence for the biogeobattery hypothesis, which has so far remained elusive, using a flow-through column experimental set-up specifically designed for this purpose. We successfully created contrasting redox zones, with an oxic section containing clean sand transitioning into an Fe(III)-reducing section. In experiments, Fe(III)-reduction was mediated both by a natural microbial community and by a pure culture of the model organism Shewanella oneidensis MR-1 in two separate column experiments. Visual observations and electron microscopy showed that ferrihydrite was sequentially transformed to goethite and magnetite; despite this, no SP signal was generated in either column. Electron microscopy suggested that in the pure culture column, S. oneidensis MR-1 cells did not form a continuous, interconnected biofilm but rather interacted with the iron (oxyhydr)oxide surfaces as individual cells. We thus conclude that (i) there is still no unambiguous direct proof of the biogeobattery hypothesis and (ii) development of Fe(III)-reducing conditions does not necessarily lead to SP signal generation. Overall, this study shows that key redox variations are not necessarily accompanied by a large SP response, suggesting that SP cannot be used in isolation to monitor subsurface biogeochemical conditions.