REVIEWS APPLIED AND INDUSTRIAL MICROBIOLOGY Microbial electrosynthesis — revisiting the electrical route for microbial production Korneel Rabaey and René A. Rozendal Abstract | Microbial electrocatalysis relies on microorganisms as catalysts for reactions occurring at electrodes. Microbial fuel cells and microbial electrolysis cells are well known in this context; both use microorganisms to oxidize organic or inorganic matter at an anode to generate electrical power or H2, respectively. The discovery that electrical current can also drive microbial metabolism has recently lead to a plethora of other applications in bioremediation and in the production of fuels and chemicals. Notably, the microbial production of chemicals, called microbial electrosynthesis, provides a highly attractive, novel route for the generation of valuable products from electricity or even wastewater. This Review addresses the principles, challenges and opportunities of microbial electrosynthesis, an exciting new discipline at the nexus of microbiology and electrochemistry. Biocatalyst Electrical power can now be produced in a sustainable term microbial electrosynthesis was used to describe the (REF. 6) A catalyst of biological origin, way by, for example, wind turbines and photovoltaic electricity-driven reduction of CO2 using whole which can be an enzyme, an cells. This prospect of sustainable energy produc- microorganisms as electrocatalysts. In line with the organelle or even a whole cell. tion has made electrical energy the key for our future definition of conventional (that is, non-microbial) elec- 1 Bioelectrosynthesis transportation and chemical production needs (using trosynthesis, we expand microbial electrosynthesis here The use of biocatalysts to battery-driven vehicles and electrosynthesis, respec- to mean ‘the microbially catalysed synthesis of chemi- achieve electricity-driven tively). For electrosynthesis, adequate electrocatalysts cal compounds in an electrochemical cell’, which, in synthesis. are needed to catalyse the electrode-driven chemical addition to the electricity-driven reduction of CO2, also reactions. Owing to their higher specificity and versatil- includes the electricity-driven reduction or oxidation of Overpotential biocatalysts The difference between ity relative to existing chemical catalysts, are other organic feedstocks. In this Review, we describe the the thermodynamically increasingly being considered for these electrosynthetic known pathways of extracellular electron transfer (EET) determined potential and processes. Bioelectrosynthesis relies on the interaction in bacteria and discuss the opportunities for microbial the experimentally observed between biocatalysts and electrodes2,3 and mainly uses electrosynthesis. potential of a half reaction; in an electrolytic cell, this enzymes or organelles that are physically immobilized on corresponds to an energy the electrode surface. However, although enzymes Bioelectrochemical systems: the basics loss, such that more energy and organelles can provide a high reaction specifi- Microbial electrosynthetic processes are conducted in is required to carry out the city and controllability, the use of whole microorganisms so-called bioelectrochemical systems (BESs), which reaction than is expected. in bioelectrosynthetic processes has several advantages, consist of an anode, a cathode and, typically, a mem- including self-regeneration of the catalyst, adaptation of brane separating the two (FIG. 1). An oxidation process the catalyst quantity to the required conversion activ- occurs at the anode (for example, acetate oxidation or ity, flexibility in substrate use and higher versatility than water oxidation), whereas a reduction process occurs Advanced Water enzymes or organelles for product formation or conver- at the cathode (for example, O2 reduction or H2 evolu- Management Centre, sion pathways. The disadvantage of microorganisms tion). The electrodes are surrounded by an electrolyte Gehrmann Building, is that they consume part of the substrate or donor for — the fluid around the electrode containing the reac- The University of Queensland, growth — albeit possibly only intermittently — and, tants and/or products — which is generally an aqueous Brisbane, Queensland 4072, as such, they are not true catalysts. However, like true solution or wastewater (as a feed source). BESs can be Australia. Correspondence to K.R. catalysts, whole microorganisms have been shown to operated in ‘microbial fuel cell’ mode, in which they 4 7 e-mail: [email protected] decrease the overpotentials at both anodes and cath- deliver power , in short-circuit mode, in which the doi:10.1038/nrmicro2422 odes5, resulting in improved performance. Recently, the anode and cathode are connected without a resistor, or 706 | OCTOBER 2010 | VOLUME 8 www.nature.com/reviews/micro © 2010 Macmillan Publishers Limited. All rights reserved REVIEWS hydroxyl ions may become diffusion limited14. This also Power production leads to a decrease in the power output or an increase in e– e– the power requirement. or Short circuit Drawing electrons from microorganisms e– or In 1910, M. C. Potter wrote that “The disintegration of organic compounds by microorganisms is accom- Power supply e– e– panied by the liberation of electrical energy” (REF. 15). or This finding, made using Saccharomyces cerevisiae, was perhaps the first observation of what we now know as Renewable energy – – (solar or wind) e e EET, the process by which microorganisms can trans- port electrons into and out of the cell from or towards an insoluble electron donor or acceptor. The primary focus of most research on EET has been (and still is) e– e– the transfer from organic electron donors towards min- Microbially Chemically or erals and electrodes. Community analyses of microbial catalysed catalysed Chemically or Microbially C+ catalysed catalysed fuel cell anodes reveal a high species diversity, including e– 16,17 CO both Gram-positive and Gram-negative organisms . 2 O2 Product Product – However, the current models for EET are built around e H O thode 2 Anode Ca only Gram-negative isolates, as most Gram-positive + H – A Electron Electron isolates have not shown a strong capacity for EET thus Organics H 18–20 2 acceptor acceptor far . Two key mechanisms for electron transfer can be discerned: these are direct and indirect transfer. Based on the innate capabilities of organisms isolated from microbial fuel cells, it seems that in microbial populations C+ A– C+ multiple strategies are in operation simultaneously21, or or maximizing the use of available resources. A– C+ A– Here we define direct EET as ‘not requiring the diffusion of a mobile component to and from the cell No membrane for electron transport’. Direct transfer has been widely studied in Geobacter sulfurreducens22,23 and Shewanella Figure 1 | A high-level overview of the concepts associated with bioelectrochemical oneidensis str. MR-1 (REF. 24), and there are several excel- systems. A plethora of choices can be made regarding the membrane,Nature Revie thews nature| Microbiolog of they catalysts at both the anode and the cathode, and the source of the reducing power. This lent reviews regarding the putative mechanisms of direct 17,25–27 leads to a highly versatile technology that can carry out a diverse range of processes. transfer in these species . Briefly, direct transfer typ- ically involves at least a series of periplasmic and outer- membrane complexes. For S. oneidensis, the apparent in ‘microbial electrolysis cell’ mode, in which power is terminal cell-bound complex is MtrC, a decahaem cyto- invested to increase the kinetics of the reactions and/or chrome located on the outside of the membrane and to drive thermodynamically unfavourable reactions8. In capable of donating electrons in a broad potential range28. theory, much energy could be derived from microbial Electrons are transported from the periplasm to MtrC conversion reactions and limited energy would need through a transmembrane electron transfer module con- to be invested to drive a microbial electrolysis process sisting of MtrA, the transporting protein, incorporated (BOX 1), but in reality the energy gained or invested is inside MtrB, a sheath protein28. For G. sulfur reducens, a considerably less or more, respectively. To understand similar dependency on membrane-bound cytochromes this, one needs to consider the losses in the BESs (for has been well documented29. In recent years the involve- an in depth discussion, see REFS 9–12). First, the oxi- ment of pili or pilus-like appendages (called nanowires dation or reduction reaction at the electrode will incur in this context) was established30. These seem to be essen- so-called activation overpotential, causing a voltage loss tial for high levels of current production in G. sulfur- due to imperfect catalysis at the electrode. The addition reducens31, in conjunction with OmcZ, a matrix-located of a chemical or biological catalyst decreases this activa- cytochrome32. It has been suggested that nanowires tion overpotential but will never eliminate it. Second, also establish electron transport between different when electrons flow through an electrical circuit, ions microorganisms in a community33. simultaneously need to move through the electrolyte to The second, indirect method for EET involves the restore the charge balance between anode and cathode. production or use of so-called electron shuttles, which The electrolyte has a certain conductivity (for waste- transport the electrons