Electrochimica Acta 82 (2012) 165–174
Contents lists available at SciVerse ScienceDirect
Electrochimica Acta journal homepage: www.elsevier.com/locate/electacta
Bioelectrochemical systems: Microbial versus enzymatic catalysis Stefano Freguia a,∗ , Bernardino Virdis a,∗∗ , Falk Harnisch a,b , Jurg Keller a a b
The University of Queensland, Advanced Water Management Centre, Level 4, Gehrmann Building (60), Brisbane, QLD 4072, Australia Institute of Environmental and Sustainable Chemistry, TU Braunschweig, Hagenring 30, 38106 Braunschweig, Germany
a r t i c l e
i n f o
Article history: Received 13 February 2012 Received in revised form 29 February 2012 Accepted 4 March 2012 Available online 11 March 2012 Keywords: Microbial bioelectrochemical system Biofuel cell Bioelectrocatalysis Oxygen reduction reaction Enzyme
a b s t r a c t Bioelectrochemical systems rely upon bioelectrocatalysts of microbial as well as enzymatic nature. Both bioelectrocatalysts serve the same purpose of enabling or accelerating a given electrochemical reaction but they often differ signiﬁcantly in their properties as well as degree of scientiﬁc understanding. This review characterizes and compares the advantages and disadvantages of both classes of bioelectrocatalysts – discussed on three selected bioelectrochemical reactions: (i) the anodic glucose oxidation, (ii) the cathodic oxygen reduction and (iii) anodic conversion of light to electricity. The article will not only focus in the respective bioelectrocatalyst application but will also highlight the current level of understanding and address future challenges and research needs. © 2012 Elsevier Ltd. All rights reserved.
Contents 1. 2.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Glucose oxidation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Enzymes and enzyme arrays for the bioelectrochemical glucose oxidation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. Microbially catalyzed glucose oxidation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The cathodic oxygen reduction reaction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Enzyme based ORR-cathodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Oxygen reduction: microbial catalysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Solar bioanodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Enzymatic solar bioanodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2. Microbial solar bioanodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions and outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1. Introduction Bioelectrochemical systems rely upon the utilization of biological entities to catalyze electrochemical processes, i.e. the transformation from chemical to electrical energy, or vice versa. The underlying effect of “bioelectrocatalysis” is characterized as the enablement of this interconversion of energy at lower
∗ Corresponding author. Tel.: +61 7 3346 3221; fax: +61 7 3365 4726. ∗∗ Corresponding author. Tel.: +61 7 3346 3218; fax: +61 7 3365 4726. E-mail addresses: [email protected]
(S. Freguia), [email protected]
(B. Virdis). URLs: http://www.awmc.uq.edu.au (S. Freguia), http://www.awmc.uq.edu.au (B. Virdis). 0013-4686/$ – see front matter © 2012 Elsevier Ltd. All rights reserved. doi:10.1016/j.electacta.2012.03.014
165 167 167 167 169 169 170 171 171 171 172 173 173
overpotentials than would be possible without the biological moiety . Bioelectrocatalysts can be classiﬁed according to their nature in three classes: (i) enzymes, (ii) enzyme arrays and organelles and (iii) microbial cells. In this paper, we consider individual enzymes and enzyme arrays/organelles as a combined group as they have similar characteristics in the context of this comparison. Furthermore, depending on the catalyzed reactions, they can be distinguished as anode catalysts, i.e. conveying electrons from an electron donor to an electrode, and cathode catalysts, i.e. conveying electrons from an electrode to an electron acceptor. All of these catalysts have been used in bioelectrochemical systems and will be exemplarily discussed within the following article. The practicability of different catalysts for selected target reactions and underlying applications will be highlighted.
S. Freguia et al. / Electrochimica Acta 82 (2012) 165–174
Bioelectrochemical systems (BESs) have been mostly developed for one of four distinct purposes: (i) electrical power generation , (ii) selective pollutant removal (bioremediation ), (iii) production of valuable chemicals  and (iv) biosensing . Considering the majority of possible electrochemical conversions, the (sub)reactions involved are often so complex that only biological moieties are able to catalyze them, with pure electrocatalysts typically failing at the task. An example of such a case is anodic glucose oxidation, which cannot be achieved with any metal or chemical catalysts, while it is easily performed by enzymes or in bacterial cells . In other cases, the reactions do proceed even in the absence of catalysts or in the presence of inorganic catalysts. In these cases bioelectrocatalysts may provide beneﬁts in terms of reaction kinetics, thermodynamics and reactant selectivity or may simply represent a more economic or sustainable alternative . An example of such a reaction is cathodic oxygen reduction: it does occur without catalysts on carbon or steel electrodes, albeit at high overpotentials and low rates . The addition of an inorganic catalyst such as platinum greatly boosts the rate of reduction, but at a high cost and high environmental footprint. Here the use of enzymes or whole microbial cells as catalysts may enable the reaction to proceed at similar rates as platinum, but at a fraction of the economic and environmental costs . Until quite recently, bioelectrochemical reactions were mostly studied with puriﬁed enzymes. Cultures of bacteria that are known to produce the desired enzyme are grown and harvested, then the cells are lysed and the enzyme extracted and puriﬁed . Subsequently, the enzymes are immobilized onto the electrode surface and exploited for the target reaction. Enzymatic bioelectrocatalysis is possible via direct electron transfer (DET) between enzyme and electrode if direct electrical contact can be established. More often, a suitable soluble (or surface-bound) redox mediator needs to be added to the system to shuttle electrons between enzyme and electrode. The latter mechanism is termed mediated electron transfer (MET). Enzymes can achieve very high turnover rates and signiﬁcant current densities, thus being very attractive for electrochemical applications. This has justiﬁed a vast amount of research in the area over the last 60 years . However, enzymatic electrocatalysis is affected by a number of limitations: • Enzyme production and immobilization: The enzyme production and puriﬁcation process is very complex, often requiring genetically modiﬁed organisms to maximize production . The immobilization of the enzyme on the electrode is also challenging to achieve both reactivity and electron conductivity to the electrode. Of critical importance is to maintain the native (active) conformation of the enzyme in the puriﬁcation and immobilization stage to ensure its activity and selectivity toward the substrate. • Lifetime limitation: Enzymes are affected by limited lifetime which imposes – depending on the actual operation environment – very frequent electrode replacements . Irreversible damage of active enzyme moieties occur as a consequence of localized non-physiological pH and temperature conditions, high voltages and chemical reactions with compounds in the media, plus range of other factors, which are not all well understood. • Limited electron transfer: The high speciﬁcity of enzymatic reactions is often limited to only a single reaction step, i.e. the transfer of mostly one or two electrons, per enzymatic step. This implies that the use of only one type of enzyme may produce just an intermediate compound, which can then accumulate and create a product inhibition effect. For example, oxidation of glucose to carbon dioxide (total transfer of 24 electrons per glucose molecule) involves around 10 enzymes. In case only some of the enzymes are present in minor concentrations the reaction can
be signiﬁcantly limited due to the reduced driving forces once a reaction product accumulates. For the whole reaction to proceed, each of the enzymes in the overall pathway needs to be present and able to effectively interact with the downstream and upstream enzymes in the reaction chain. This is obviously extremely difﬁcult to achieve using electrodes based on individually immobilized enzymes, gained from enzyme puriﬁcation and immobilization. An alternative to enzymes as bioelectrocatalysts is the use of whole microbial cells. As enzymes are contained within cells, the catalysis is expected to still be effective as long as a path for electron transfer from the cytoplasm to the extracellular electrode is found. Whole cell bioelectrocatalysis was ﬁrst proposed based on a ﬁnding reported in 1911 , but it is only in the last ten years that research in this area has picked up signiﬁcant momentum, due to the discovery of a number of ways by which bacteria are able to establish extracellular electron transfer (EET) with electrodes. These mechanisms include (i) soluble, self-synthesized, redox mediators such as ﬂavins , phenazines  and quinones , (ii) cytochromes conveniently located in the cell outer membrane  and (iii) a special type of pili known as nanowires  which have been shown to exhibit electric conductivity. For a detailed description on these mechanisms the reader is referred to several recent reports e.g. [19–22]. Using whole bacterial cells as the catalysts of bioelectrochemical processes avoids three main constraints that can limit enzymatic electrochemistry. First, no enzyme production and puriﬁcation process is required, as enriched cultures of microorganisms naturally generate (or may be engineered to generate) all enzymes for a particular pathway and can regulate their production to sustain the metabolism in response to the environmental conditions. Secondly, enzymes are not affected by stability limitations when they are in live cells, since bacterial cells do generally create conditions well suited for their enzyme activity and they have an active regeneration process of these enzymes within the cells whereby the enzymes are re-synthesized dependent on the metabolic needs and natural degradation rates. Finally and most importantly, living cells do contain typically all of the required enzymes for a complete biochemical pathway (or even several different ones in parallel), thus they are not limited to a single reaction step. Instead, they are able to sustain complex reactions that may require dozens of enzymes in a well balanced and controlled interaction. Despite these advantages over enzymatic systems, whole cell bioelectrochemistry also suffers from a number of limitations. These include: • Nutrient and energy requirements: Bacteria need a consistent supply of nutrients and energy in order to grow and thus replace dead cells that are sloughed off the bioﬁlm. Yet, this self-regeneration of the bioelectrocatalyst can also be seen as an intrinsic advantage. • Lack of substrate speciﬁcity: Even in pure cultures, bacteria rarely rely on only one metabolic pathway, thus are typically able to utilize more than one type of substrate (electron donor for anodes, electron acceptor for cathodes). This makes microbial bioelectrochemical systems much less suitable for compound-speciﬁc biosensing applications. • Lack of product speciﬁcity: Especially when utilizing mixed microbial cultures, competition between different bioreactions will inevitably lead to loss of electrons to unwanted by-products. Competition has the effect of reducing the Coulombic efﬁciency and affecting electrode potential (due to mixed potential formation), and in the case of production of speciﬁc compounds it dilutes and pollutes the desired product, generating additional down-stream processing needs.
S. Freguia et al. / Electrochimica Acta 82 (2012) 165–174
Despite the vast array of possible bioelectrochemical reactions that have been studied, some have been subjected to particularly intense investigation, due to their importance in a number of applications. In the following, we will present a review of signiﬁcant research results related to three exemplary important bioelectrochemical reactions, namely glucose oxidation (Section 2), oxygen reduction (Section 3) and anodic conversion of light to electricity (Section 4). For all three cases, we will highlight the advantages and disadvantages of catalyzing the reactions with either enzymes or whole bacterial cells. 2. Glucose oxidation In the quest for improved sustainability, electricity generation from renewable, but complex bio-organic feedstocks is one of the key ambitions in today’s society. Consequently, the development of fuel cells that could harvest the energy content of biopolymers such as fats, proteins and most importantly carbohydrates including starch and celluloses are high on the global research agenda. Commonly, these highly complex and diverse materials cannot be oxidized by organic or inorganic chemical catalysts (see e.g. ). However, nature provides a whole arsenal of highly specialized enzymes and microbial cells able to oxidize these natural feedstocks to harvest the chemical energy contained therein. Biofuel cells seem to be well placed to exploit this chemical energy and directly convert it to electric energy using the large array of biochemical pathways available in the diverse range of enzymes and microorganisms. However, so far only few complex substrates have been targeted using enzymes/enzyme arrays (see e.g. [24,25]). In contrast, the exploitation of complex substrates like celluloses [26,27] and proteins  in microbial fuel cells already highlights the versatility of whole cell bioelectrocatalysis. As glucose, C6 H12 O6 , is one of the most important carbohydrate monomers, the following discussion of enzymatic and microbial bioelectrocatalysts will be based on the oxidation of this exemplary molecule. 2.1. Enzymes and enzyme arrays for the bioelectrochemical glucose oxidation The best-known example of enzymes oxidizing glucose are the glucose dehydrogenases (GDH). This enzyme family, being oxidoreductases with the enzyme classiﬁcation EC 18.104.22.168 can be found in numerous species ranging from Bacillus subtilis to the Homo sapiens  (source: UniProt KB database, query: “ec:22.214.171.124” last access 28.10.2011). The GDH commonly catalyses the oxidation of glucose to gluconolactone, thereby liberating two electrons, see Eq. (1) . C6 H12 O6 → C6 H10 O6 + 2e−
E 0 = −0.364 V
In nature, this enzyme plays an important role in the pentosephosphate pathway and the electrons are generally liberated to redox shuttles like (i) NAD(P)+ forming NAD(P)H + H+ ; or (ii) pyrroloquinoline quinone (PQQ) (e.g. [30,31]). Therefore the ﬁrst attempts of connecting the enzymatic activity of the GDH to the electrode were related to the electrochemical recycling of their cofactors. Here the electrochemical oxidation of NAD(P)H and FADH2 plays a key role (see e.g. ). However, besides progress in the pure electrochemical co-factor regeneration one main task to be solved is the “direct wiring” of the enzymatic bioelectrocatalyst to the electrode. Here several approaches have been followed, including tailoring of the electrode surface at nano-scale , immobilization of the natural and artiﬁcial redox mediators at the electrode surface using binders, or polymerization approaches (see e.g. ).
Fig. 1. Schematic representation of a multi-enzyme array of 4 enzymes A to D catalyzing the oxidation of a substrate (Sub.) via 3 intermediates (IM 1, IM 2 and IM 3) to the ﬁnal product (Prod.). Each step is catalyzed by a speciﬁc enzyme, which transfer electrons to the electrode (as indicated by the straight arrows).
The “direct wiring” of the enzyme to the electrode surface has several advantages, most prominently the increase of the kinetic and thermodynamic performance of the enzyme electrode . Furthermore, the enzymatic glucose oxidation has, like all other single-enzyme bioelectrocatalytic reactions, the limitation that only a single (bio)chemical reaction can be catalyzed. As Eq. (1) shows, only two electrons are liberated, in contrast to the complete oxidation of glucose to carbon dioxide yielding 24 electrons. Therefore, the energetic efﬁciency of the enzymatic glucose oxidation is (besides the strongly negative standard potential of the reaction) very limited. To overcome the limited electron-harvesting efﬁciency of single enzymes, arrays of enzymes catalyzing multiple steps – sometimes denominated as metabolons – were proposed (see e.g. [35,25,36]). Fig. 1 depicts the principle of a multi-enzyme array allowing the oxidation of a substrate by consecutive reaction steps to a product molecule. Two main ﬁelds of application for oxidation enzymes, like GDH or the glucose oxidase (see Fig. 2B), are envisaged: anodes of biofuel cells or biosensors, particularly for medical applications. Biosensing arrays, based on bioelectrocatalytic reactions, are not in regular clinical use yet , however they may enable real time monitoring of blood sugar level in future and thus direct control of insulin dosing for diabetes patients. So far, enzymatic bioelectrocatalysis of glucose oxidation is only exploited in test-strips and here the interference of other blood constituents plays a crucial role . However, the establishment of these exemplary devices may open the door for a multitude of different electrochemical biosensors for medical applications. For environmental monitoring, however, the main concerns for the application of electrochemical biosensors are the number of possible side-reactions and the diversity and concentration range of denaturing agents that far exceed the conditions existing in the medical sphere. As discussed above, this will not only limit the selectivity of the sensor, but also its lifetime. Enzymatic biofuel cells  can use many of the same substrates as in microbial fuel cells (see below). However, since they employ a single or limited number of puriﬁed enzymes to catalyze the oxidation of the electron donor, the conversion of the fuel is usually only partial. 2.2. Microbially catalyzed glucose oxidation Contrary to single enzymes interacting with electrodes, whole living organisms (or better, the chain of enzymes carrying out the
S. Freguia et al. / Electrochimica Acta 82 (2012) 165–174
Fig. 2. (A) Pathways of oxidation of glucose in respiratory metabolism of anode respiring bacteria. Glucose is ﬁrstly converted by glycolysis into pyruvate and acetyl Co-A, which enters the TCA cycle where it is converted into CO2 , ATP and reducing equivalents NADH, NADPH, FADH2 , which transfer electrons ﬁrstly to the electron transport chain, and then to the electrode via direct electron transfer (DET) or mediated electron transfer (MET). (B) Structure of glucose oxidase from Asperigillus niger indicated as cyan ribbons, the ﬂavin cofactor is indicated with yellow sticks (PDB code 1CF3), structure solved by Wohlfahrt et al. .
conversions) have the ability to catalyze a wider range of reactions. As a result, they are far less speciﬁc toward substrates compared to pure enzymatic biofuel cells since a more complex network of possible biochemical reactions exists within a microbial cell . They can also operate under broader physiological conditions of temperature  and pH [42,43] than single enzymes. Whole living organisms act as miniature, self-reproducing bioreactors, allowing for extended oxidation of the substrates. Typical microorganisms used for anodic conversions include pure cultures of various bacterial strains, mixed cultures as well as yeast cells, see e.g.  for an overview. Studies utilizing single substrates have shown that pure or mixed culture microbial communities are able to fully oxidize volatile fatty acids (e.g., acetate, propionate, butyrate, valerate), alcohols (e.g., ethanol and methanol) as well as more complex substrates (e.g., glucose, sucrose, cellulose, starch, xylose  and even real wastewater ), with the a maximum achieved current density of 3.0 mA cm−2 . In most cases, more diverse and interrelated microbial populations may be required to break down the complex substrate into easily degradable compounds for electroactive organisms located at the electrode. This process network is only partly understood, see e.g. [47,48], and is likely similar to the hydrolysis and fermentation steps in anaerobic digestion. In comparison to anaerobic processes, however, a wider range of substrates can be directly converted by electroactive organisms at the anode than can be utilized by methanogens (only acetate or hydrogen). Two key metabolic pathways that help microorganisms perform complete oxidation of larger substrates such as carbohydrates include glycolysis and the tricarboxylic acid cycle (TCA cycle, also known as Krebs cycle). Thereby, glycolysis has to be considered as a ﬁrst pre-step for breakdown of the glucose into smaller units that can be converted on electrocatalytic electrodes, see e.g. , or that are channeled into the TCA for direct bioelectrocatalysis. The TCA cycle is a highly interconnected metabolic pathway consisting of eleven enzymatic steps with nine different enzymes involved (starting from pyruvate) and a similar number of intermediate metabolites. Importantly, most of the conversion steps are regulated by some form of product inhibition, which limits the kinetic
rates based on the accumulation of the product from the enzymatic reaction. Therefore, the complete pathway can achieve a much higher turnover rate when operating in a well integrated and balanced way than each of the individual steps on its own can achieve. Here the use of “metabolons”  or whole organelles  was proposed as alternative bioelectrocatalyst, however these moieties are quite complex to extract and immobilize, and are not self-reproducing. The exact metabolic process network ultimately depends on the substrates provided and their concentrations . Microbial cells contain all the necessary enzymatic machinery to undertake these metabolic processes, acting in fact as catalyst for the complete (or at least, extended) conversion of the substrates. This can achieve complete oxidation of substrates like glucose to CO2 and water. Therefore, the use of whole cells is advantageous where the maximal electron harvesting efﬁciency is important, such as in fuel cell applications. Electrode reducing (anodic) microorganisms convert the chemical energy of organic molecules into electricity via a process that is considered metabolically similar to other anaerobic respiration processes (e.g. using iron(III)-minerals) , with the important difference that the ﬁnal electron acceptor is the electrode. The main products of this conversion are water, CO2 , and ATP. In this microbial respiration process, small carbon substrates like acetate can enter the TCA cycle directly through pyruvate or acetyl-CoA and their oxidation is directly linked to the respiratory chain. Larger molecules like glucose, however require an additional enzymatic process called glycolysis as ‘pre-treatment’, which involves several enzymes catalyzing the conversion of glucose into pyruvate and acetyl-CoA. Yet, some of the microorganisms able to use anaerobic respiration (including electrodes as acceptors) do not contain the glycolysis enzymes and therefore rely on the “pre-digestion” of carbohydrates by other, e.g. fermenting, microorganisms. Pure culture glucose oxidation at bioanodes has been reported in microbial fuel cells at both circumneutral  and acidic conditions , however, fermentative bacteria (which do not require an external electron acceptor such as an anode) would have a competitive advantage over anodic electroactive microorganisms in mixed culture
S. Freguia et al. / Electrochimica Acta 82 (2012) 165–174
electrochemical systems, resulting in the establishment of syntrophic relationships between fermenters and anodic organisms . Fig. 2A shows an exempliﬁcation of the complete glucose oxidation by an anode-reducing microorganism. Glucose is ﬁrstly converted into pyruvate and acetyl-CoA by glycolysis, with the concomitant generation of ATP and reduced NADH, which are then added to the pool of reducing equivalents of NADH, NADPH, and FADH2 generated through the TCA cycle (Fig. 2A). These are reduced molecules that represent the primary electron donors for the following electron transport chain, which is made up of a series of membrane associated electron carriers that include ﬂavoproteins, iron–sulfur proteins, quinones and multi-heme cytochromes. The electron carriers are arranged in the electron transport chain in a way that electrons are transferred from one complex to the next at a higher potential. This electron translocation generates also a proton motive force across the membrane with up to 10 protons generated for each electron pair derived of NADH. The proton motive force drives ATP generation via phosphorylation by proton-translocating ATP-synthase, which assures that energy conservation is maintained. Electrons from the electron transport chain are then conveyed extracellularly toward the electrode, either through direct or mediated electron transport. In principle, the anodic compartment of a bioelectrochemical system is an anaerobic reactor where the presence of a solid-state electron acceptor provides a very selective environment for organisms that are able to perform extracellular electron transfer (EET) [55–57]. However, competitive processes such as fermentation and methanogenesis are also likely to get established when mixed cultures are used. The effect of the presence of these processes is a consumption of the substrate without electrons being delivered to the anode. In addition, the fact that these competitive processes do not rely on the electrode, makes them advantageous over current generation by anodophillic organisms as they can occur at a distance from and without any connection to the electrode and therefore make use of the whole reactor volume rather than only the electrode surface. Fermentation occurs when an external electron acceptor is not available and the substrate is at ﬁrst oxidized to intermediate metabolites that then act as electron acceptor for the production of fermentation products. Due to the lack of an electron acceptor at high potential, fermentation only yields a limited amount of energy in the form of ATP through substrate level phosphorylation. Fermentative organisms are in general advantaged over electroactive organisms due to the fast turnover rates of fermentation processes compared to respiration, which requires a high expression of glycolytic enzymes. However, fermentation is only possible for carbohydrates and proteins and can usually not be used by the majority of electroactive microorganisms. Consequently, fermentation may not always represent a limitation for mixed culture electroactive consortia, especially when they produce fermentation by-products such as acetic acid, lactic acid and hydrogen that can be readily degradable substrates for anode-respiring organisms  or are used by direct oxidation at suitable electrodes [27,59]. Moreover, due to the thermodynamic requirement of reducing the H2 partial pressure during fermentation, syntrophic relationships between fermenters and electroactive anodic bacteria may be established , but other metabolite-based synergistic interactions can also be established . Methanogenesis refers to the generation of methane by Archaea. Substrates for methanogenesis are limited to acetate (acetoclastic methanogenesis) or H2 and CO2 (hydrogenotrophic methanogenesis), all of which could be commonly found in bioelectrochemical systems. Given the absence of oxygen in a microbial fuel cell, which would inhibit methanogens growth, methanogenesis is likely to develop in such systems and competes for substrate with
anode-respiring organisms and can represent a substantial Coulombic loss with up to 50% of the electron equivalents from glucose converted to methane . Also microbial bioelectrochemical sensors, based on anodic bioﬁlm reactions, have been proposed for biological oxygen demand (BOD) monitoring [61,62] and for toxicity sensing. For the latter in particular, it should be noted that bioﬁlms can be signiﬁcantly less sensitive to several toxins than suspended cultures and this approach might therefore not be suitable for a toxicity biosensor development . 3. The cathodic oxygen reduction reaction The oxygen reduction reaction (ORR) is the desired cathode reaction of all fuel cells being chemical, enzymatic or microbial ones (see e.g. [23,64]). It provides typically the most attractive electron acceptor reaction to combine with the substrate (organics) oxidation reaction at the anode in order to create a net-exergonic reaction. Eqs. (2)–(4) provide the principle reactions of the (bio)electrochemical oxygen reduction and the respective standard potentials at neutral pH solutions . O2 + 4e− + 4H+ → 2H2 O −
O2 + 2e + 2H → H2 O2 −
E 0 = 0.818 V E
H2 O2 + 2e + 2H → 2H2 O
= 0.257 V
= 1.375 V
At the macroscopic level the water producing ORR can either proceed as a 4e− /4H+ reaction (Eq. (2)), or as a two-step process as shown in Eqs. (3) and (4). These two 2e− /2H+ reactions can be combined to result in the identical net-reaction (i.e. Eq. (2)) and netpotentials. However, as it has been shown for numerous catalyst materials, and as further discussed below, the intermediate H2 O2 (formed as shown by Eq. (3)) does not necessarily react completely to the ﬁnal product water (Eq. (4)). This loss of a fraction of the intermediate H2 O2 results in the lowering of the electrode potential due to mixed potential formation, an effect that is similar to a limited selectivity of the catalyst . Furthermore, the strong oxidizing agent hydrogen peroxide can severely limit the stability of a biological moiety. Recently, however, the production of H2 O2 on carbon based materials was exploited for the production of bleaching solutions  creating a valuable reagent recovery option for particular industries such as paper and pulp manufacturing. A common element of enzymatic and microbial ORR cathodes is their operational environment, which is characterized by several properties clearly differing from that of ORR cathodes in chemical fuel cells (e.g. ). These include most prominently solutions as reaction environment, which is characterized by (i) only a low ionic strength and (ii) a limited oxygen solubility, (iii) operation at room temperature, (iv) the presence of other molecules than the reactants, which altogether challenge the development and assessment of these catalysts. Table 1 summarizes selected studies on the oxygen reduction reaction for conditions present in bioelectrochemical systems. 3.1. Enzyme based ORR-cathodes Several redox enzymes catalyzing the ORR have been described, including cytochrome oxidases, bilirubin oxidases, laccases or peroxidases (e.g. horseradish peroxidase) . These enzymes were often employed in enzyme based biofuel cells  but also in microbial fuel cells possessing an enzymatic ORR cathode . In the latter case by using a laccase of the fungi Trametes versicolor one of the highest open circuit voltages (OCV) in MFCs of 1.1 V has been achieved , whereas in enzyme based fuel cells the maximum reported OCV is 0.95 V .
S. Freguia et al. / Electrochimica Acta 82 (2012) 165–174
Table 1 Comparison of performance of selected enzymatic and microbial ORR cathodes. Catalyst
Open circuit cathode potential (V vs. SHE)
Steady state current density (mA cm−2 )
Laccase from Trametes versicolor Blue copper oxidase ORR mixed bioﬁlm ORR mixed bioﬁlm
pH 5 McIlvaine buffer
Carbon aerogel-modiﬁed glassy carbon Pyrolytic graphite edge Carbon ﬁber mat Graphite granules
Tsujimura et al. 
+0.85 +0.10 +0.54
0.60 0.10 0.48
>60 days >2 months >9 months
Blanford et al.  Rabaey et al.  Freguia et al. 
pH 4 acetate buffer pH 7 phosphate buffer pH 7 phosphate buffer
In general, the ORR at enzyme electrodes is characterized by a high thermodynamic performance, i.e. an open circuit voltage close to the theoretical value, due to the high enzymatic selectivity/low tendency to side-reactions and thus no mixed potential formation. In contrast to their superb thermodynamic characteristics for open circuit condition, the performance of enzyme based ORR cathodes during polarization are governed by several limiting factors. Obviously, the highest ORR cathode potential can be achieved by enzymes based on direct electron transfer (DET). Here the electrons are transferred from the cathode directly to the redox enzyme (and subsequently to the terminal electron acceptor O2 ) and thus the potential of the electrode solely depends on the formal potential of the active site and is not lowered by a further mediator-related potential drop. However, in general DET does not allow high electron transfer rates and is therefore characterized by relatively poor reaction kinetics. As a consequence, mediator systems are used to accelerate the kinetics, with a considerable increase of the ORR current density as a consequence. Geometric current densities up to 0.5 mA cm−2 (using a laccase isolated from T. hirsute ) have been reported. However, the exploitation of mediator substances often decreases the electron transfer thermodynamics signiﬁcantly. This is due to the fact that this involvement of an additional electron shuttling partner elongates the overall redox-cascade, thus creating a further potential drop. Furthermore, frequently used substances like 2,2 -azinobis(3-ethylbenzthiazoline-6-sulfonate)(ABTS) and tethered osmium complexes  possess redox potentials signiﬁcantly lower than the formal potential of the oxygen reduction reaction , leading to high thermodynamic losses. The major drawback of enzyme ORR-cathodes, however, is their limited longevity with the most robust known systems showing a decrease of 43% of their initial activity within 60 days , but much faster activity decreases are often observed, such as a 50% loss of the activity within a week . Particularly denaturing environmental conditions such as pH < 4 or T > 50 ◦ C are known to lead to total loss of performance within 1 h . Therefore, biosensors or biofuel cells based on enzymatic ORR seem to be suitable only for short-term applications. 3.2. Oxygen reduction: microbial catalysis Whole bacterial cells are able to create routes for high-potential oxygen reduction on a variety of materials and were ﬁrst described by Mollica and co-workers over 30 years ago . The fact that seawater bioﬁlms grown on metal coupons have a boosting effect on oxygen reduction on metals has been known for decades due to the observation of enhanced corrosion rates in the presence of bioﬁlms. This phenomenon is associated with an increase of open circuit potential (corrosion potential) known as ennoblement . The mechanisms of such an increase of the onset potential of oxygen reduction have not yet been clearly elucidated. Investigators have proposed a number of hypotheses, including production of hydrogen peroxide within the bioﬁlm  or involvement of extracellular enzymes  and direct electron transfer .
More recently, researchers have demonstrated the ability of bacterial communities to catalyze oxygen reduction to water on graphite electrodes in freshwater and wastewater environments, implying that this feature is widespread among bacteria and not medium/material-dependent. In two independent studies, Freguia et al.  and Clauwaert et al.  discovered that bacterial bioﬁlms growing on graphite granules were capable of enhancing oxygen reduction rates signiﬁcantly compared to abiotic graphite, with a reduction of cathodic overpotential by up to 0.3 V, leading to a cathodic open circuit potential as high as +0.54 V vs. standard hydrogen electrode (SHE). This makes this biocatalyst nearly as effective as platinum or multi-copper oxidases in terms of overpotential reduction . Community analysis revealed phylogenetically diverse populations on these ORR electrodes, see e.g.. Further work by Rabaey et al.  proved the unequivocal link between the presence of bacteria on the surface of carbon electrodes and enhanced rates of cathodic oxygen reduction. Simultaneously, an independent study by Faimali et al.  revealed the tight interdependence between observed cathodic currents and the number of bacteria deposited on electrode surfaces in marine bioﬁlms. More recently, pure cultures have also been employed successfully at oxygen cathodes, one example being the acidophilic species Acidithiobacillus ferrooxidans, which was shown to catalyze oxygen reduction with onset potential of +0.29 V vs. SHE . However, the mechanisms of extracellular electron transfer and the associated metabolic pathways involved in the microbially mediated cathodic oxygen reduction are still largely unclear. So far only very few studies, e.g. , have shed some light on possible mechanisms, revealing that quinone shuttling and direct electron transfer via outer membrane heme groups are used by different species grown in pure cultures, both yielding catalytic currents with onsets in the range of 0–0.1 V vs. SHE. More work is warranted to elucidate the mechanisms of electron transfer at the higher potentials observed in mixed cultures. Microbial oxygen cathodes are living factories of enzymes that catalyze oxygen reduction with the ability to continuously selfregenerate the catalytic activity. These bacterial cathodes have been operated successfully with no interruptions for at least nine months without any loss of performance, as long as a supply of nutrients was maintained. Moreover, oxygen reducing cathodic bioﬁlms do not suffer from competing bioreactions as much as glucose oxidizing bioﬁlms: if oxygen is the only electron acceptor and the electrode the only electron donor, the only product found is water with no appreciable Coulombic losses to e.g. hydrogen peroxide. Even in the presence of alternative electron acceptors such as nitrate, the bacteria will typically switch to nitrate reduction only after complete depletion of any oxygen present in the system. The only important drawback of microbially catalyzed cathodes is that bacteria switch to organotrophy if they are given a chance. A polarized cathode is a far poorer electron donor than organic substrates such as sugars and fatty acids. The same bacteria responsible for the observed cathode catalysis turn to acetate as electron donor as soon as it is added to the system . This causes oxygen depletion by organotrophic metabolism and consequential loss of electrochemical performance. However, the activity is typically regenerated after acetate depletion, as long
S. Freguia et al. / Electrochimica Acta 82 (2012) 165–174
as the organotrophic period is not long enough to cause shifts in the microbial community composition. Some short periods of organotrophic growth might even be favorable for active biomass growth, however this has not been investigated in detail so far. Application of microbial ORR cathodes is promising as alternative to chemically catalyzed oxygen cathodes for biosensors and biofuel cells, as they are self-regenerating, less likely prone to deactivation (e.g. from sulﬁde) and need almost no maintenance. However, further fundamental understanding and improved engineering solutions to achieve higher performance are necessary. 4. Solar bioanodes Photosynthesis is the biological process by which phototrophic organisms (i.e. plants, algae and phototrophic bacteria) harvest the energy of light and efﬁciently convert it to a ﬂow of electrons, which ultimately results in the reduction of carbon dioxide to organic chemicals and biomass. The electron donor in the process is water for oxygenic phototrophs, which make up the vast majority of photosynthetic organisms, but it can also be sulﬁde or organic carbon for anoxygenic phototrophs . Photosysnthesis in living organisms is a membrane-bound process. The biological membranes associated with photosynthesis are called thylakoids and rely on two membrane-spanning protein complexes, namely Photosystem 1 (PS1) and Photosystem 2 (PS2), both comprising a number of enzymes and co-factors, whose purpose is to transport electrons from the primary electron donor (water) to the terminal electron acceptor (NADP+ ) while producing ATP from the proton-motive force that derives from the electron transport cascade. In PS2, the chlorophyll-based molecule P680 undergoes a light-induced redox potential step from +1.25 V in its normal state to −0.58 V (SHE) in the excited state . This enables the oxidation of water to oxygen and the simultaneous generation of an electron ﬂow over conventional biological electron carriers. In PS1, the light excitation of P700 induces a potential change from +0.43 V to −1.3 V vs. SHE . These highly reducing conditions enable electrons to cascade over to NADPH and from there onto the Calvin cycle for CO2 reduction to carbohydrates and biomass. This ﬂow of electrons at highly reduced conditions is thermodynamically very attractive as a potential anodic process. If electron transfer can be established between any electron carrier in the photosynthetic transfer chain and an electrode (Fig. 3), a solar-powered anodic process is generated, which can be coupled to either oxygen reduction at a cathode to harvest electrical power, or to other cathodic processes for the production of fuels or chemicals . The light-harvesting efﬁciency of such solar bioanodes is dependent on the exit point of the electrons, decreasing for every step in the electron transfer chain due to utilization of energy for the generation of proton motive force. Based on the absorption spectrum of P680 and P700, 45% of incident sunlight can be absorbed, and a maximum of 30% can be stored as chemical energy in NADPH or organic derivatives . Only a maximum of 10% of incident sunlight, and typically less than 5%, can be converted to biomass. For this reason, solar bioanodes may potentially be able to convert sunlight to usable energy more efﬁciently than processes that rely on biomass growth and usage, such as biofuel production from microalgae. However, this still needs to be further investigated and demonstrated both scientiﬁcally and practically. In the following we will focus the discussion on the direct transfer of the photosynthetically generated electrons to the electrode. Alternative, indirect approaches based on the oxidation of photosynthetic hydrogen  and on MFCs using organic acids produced by plants  will be not discussed in detail. Such solar anodic process can be pursued by either immobilizing photosynthetic enzymes, including organelles, onto electrodes, or by allowing whole cells of phototrophic microorganisms to interact
with electrodes. In both cases the challenge is to establish a suitable electron transfer path between the enzyme redox centers and the electrode. 4.1. Enzymatic solar bioanodes Anodic current from the photocatalytic oxidation of water by electrode-immobilized PS2 was ﬁrst achieved by Carpentier et al. , in the presence of dichlorobenzoquinone (DCBQ) or ferricyanide as redox mediators. From the very start, it became clear that the main difﬁculty in establishing these photoanodes lied within the electron transfer from enzyme to electrode. Unlike some of the enzymes used for oxygen reduction such as laccases, the redox centers in photosystem complexes are located too far from the protein external surface to enable direct electron transfer. Both PS2 and PS1 have been immobilized on anode surfaces with some degree of success, but always required a method to artiﬁcially convey electrons to the electrode surface. Amako et al.  utilized PS2 with 2,6-dimethylbenzoquinone as mediator and obtained a light-induced current of approximately 8 A mW−1 (current density normalized to the power density of incident light within the XY spectral window). Badura et al.  achieved similar current output from PS2 by using 2,6-dichloro-1,4-benzoquinone as mediator. Other mediators used with PS2 include duroquinone  and mercapto-p-benzoquinone , both of which produced more modest currents. More recently, PS2 was embedded in polymeric coatings made of osmium complexes, yielding a promising 17 A mW−1 : this would correspond to a current density of 0.35 mA cm−2 in a typical direct sunlight intensity of 200 W m−2 . When PS1 is immobilized on anode surfaces, not only a mediator, but also a suitable electron donor (typically ascorbate) is needed to drive the process. As with PS2, the most promising results have been obtained with Os complexes as polymeric redox mediators, with a maximal current density of 17 A mW−1 obtained thus far . Other than the need for mediators, the main limitation of PS1 and PS2-based enzymatic systems for solar energy harvesting is the short lifetime of some of the enzymes involved, which leads to a rapid loss of activity . Therefore, the study of these systems is primarily driven by scientiﬁc interest at this stage. An interesting alternative is the use of inorganic compounds mimicking the PS1/PS2 structure and function with no stability limitation . 4.2. Microbial solar bioanodes Due to the intrinsic difﬁculties in establishing enzymatic solar anodes based on immobilized photosystems, research has focused also on the use of whole cells of photosynthetic organisms. This approach would solve the problem of short-lived enzymatic activity thanks to the rapid in vivo self-repair mechanisms, which enable the replacement of subunits and cofactors at a high turnover rate, typically with physiological half times of 20–30 min . Most of the work done on microbial solar anodes was based on cyanobacteria as catalysts. Extracellular electron transfer in prokaryotic phototrophs such as cyanobacteria is expected to occur faster than in eukaryotic cells (such as microalgae) due to the smaller number of membranes that separate the photosystems from the external electrode. Indeed, in eukaryotic phototrophs the photosystems are embedded in organelles known as chloroplasts, which in turn are enclosed by the cell membrane and wall, making it harder to transport electrons to the extracellular environment. In a study by Pisciotta et al. , cyanobacteria were shown to produce light-driven electrical currents even in the absence of externally added mediators, at a power output (normalized to incident light intensity) of up to 0.7 mA mW−1 . However, the very small
S. Freguia et al. / Electrochimica Acta 82 (2012) 165–174
Fig. 3. Intracellular and extracellular electron transfer paths in oxygenic phototrophic organisms such as cyanobacteria.
Table 2 Comparison of literature results with different enzymatic and whole-cell approaches for solar bioanodes. Catalyst
PS2 from spinach chloroplast PS2 from Thermosynechococcus elongatus PS2 from Thermosynechococcus elongatus PS1 from Thermosynechococcus elongatus Synechococcus sp. Synechococcus sp. PCC7942 Pond photosynthetic consortium
Incident light intensity (W m−2 ) 140 10
Current density (mA cm−2 )
Normalized current densitya (A mW−1 )
Amako et al. Badura et al. 
Badura et al. 
Os complexes/methyl viologen
Badura et al. 
2-Hydroxy-1,4-naphthoquinone 2,6-Dimethyl-1,4-benzoquinone None
40 15 0.15
0.3 0.8 0.01
90 500 700
Yagishita et al.  Tsujimura et al.  Pisciotta et al. 
Current density normalized to power density of incident light.
light intensity used (100 lux, equivalent to 0.15 W m−2 ) limited absolute current production to 10 A cm−2 . Reportedly, the use of higher light intensities resulted in bacterial cell photo-damage. More results have been obtained with the addition of mediators. Yagishita et al.  reported photo-induced currents up to 90 A mW−1 using an anode based on Synechococcus sp. and 2hydroxy-1,4-naphthoquinone as mediator at 1 mM concentration. Tsujimura et al.  utilized a similar system based on Synechococcus and 2,6-dimethyl-1,4-benzoquinone (0.5 mM), obtaining encouraging current densities up to 0.52 mA mW–1 , which would translate to an impressive ∼6 mA cm−2 at direct sunlight conditions. Unlike glucose oxidation anodes and oxygen cathodes, it appears that microbial solar anodes can achieve far better rates and selectivity than enzymatic ones. This could be ascribed to the complexity of photosystem complexes compared to the single-enzyme systems used for glucose and oxygen electrodes. However, signiﬁcant limitations are still present. Firstly, the concentrations of mediators required to produce signiﬁcant currents are very high. As quinones experience photo-oxidation and biodegradation, periodic replacement of lost mediators would be needed, implying signiﬁcant operating costs of such a system. Secondly, hydroquinones can be directly reoxidised by the oxygen produced by the cyanobacteria. This effect results in diminishing current densities at high light intensities. For instance, the current peaked at a light intensity of 50 W m−2 in the study of Yagishita et al.  due to the parasitic
reaction between oxygen and hydroquinone. Finally, cyanobacteria experience inhibition at high light intensity, possibly due to the toxic effect of oxygen, with onset at approximately 200 W m−2 according to , but at much lower values (<1 W m−2 ) according to . The comparison between several enzymatic and microbial solar anodes is summarized in Table 2.
5. Conclusions and outlook As shown in this article, enzyme and whole-cell-based bioelectrochemical processes are scientiﬁcally highly interesting and their study has a signiﬁcant impact on our understanding of nature. From the application side, however, the suitability of a given bioelectrocatalyst largely depends on (i) the target reaction; (ii) the reaction environment; and (iii) the desired selectivity, performance and longevity. In general, the stronger substrate selectivity and quantitative conversions achieved in enzymatic systems make these particularly suitable for biosensing and speciﬁc bioelectrochemical conversion applications. On the other hand, whole-cell-based microbial bioelectrocatalysts seem most suitable in applications where extensive and diverse bioelectrochemical conversions of chemical compounds are required, e.g. complete oxidation of complex substrates at bioanodes.
S. Freguia et al. / Electrochimica Acta 82 (2012) 165–174
There also appear to be some differences in the way enzymatic and microbial bioelectrocatalysis processes are studied. Enzymatic systems are usually investigated in a systematic, largely electrochemistry-based manner, often focusing more on the scientiﬁc understanding than the practical application aspects. In contrast, the microbial processes are typically studied with a more empirical and practically oriented methodology, so far with less emphasis on the fundamental understanding. There are always exceptions to these general observations, but there are likely signiﬁcant beneﬁts in a much stronger exchange of ideas and methodologies between these different approaches. Indeed, it seems in many aspects that these two ﬁelds have developed in parallel (or even in competition) to some degree over the last few years, while more intensive interactions and active collaborations could likely yield major advances in the development of the broader bioelectrochemical process ﬁeld. Such partnerships are to be encouraged and they will require true inter-disciplinary exchanges of chemical and biological sciences with biotechnology and process engineering disciplines. Given the highly versatile and scientiﬁcally attractive nature of bioelectrocatalytic processes, their true potential is not yet fully recognized, let alone well exploited. For instance, enabling a better selectivity of mixed and pure-culture microbial systems will further improve the advantages of these systems in terms of energy yield as well as robustness and longevity. This could be achieved with better understanding of the underlying bioelectrochemical enzyme systems and particularly their regulatory and metabolic control mechanisms. Novel approaches such as metabolomics and ﬂuxomics would potentially help to create a far better understanding of the chemical and electron transformation processes in these complex metabolic system. Ultimately, the use of genetically engineered organisms will allow to directly inﬂuence and manipulate the selected conversion processes within these cells, which will enable even better control of the cell-internal reactions. To achieve this stage, the integration of leading scientiﬁc knowledge in both electrochemical enzyme and microbial processes will be essential, further highlighting the need for closer collaborations across the different disciplines and research ﬁelds in future. Acknowledgements This work is supported by the Australian Research Council through projects DP0985000 (B.V.), DP110100539 (J.K.) and DP120104415 (S.F. and J.K.). F.H. thanks the Fonds der Chemischen Industrie (FCI) for support. References  A.J. Bard, G. Inzelt, F. Scholz, Electrochemical Dictionary, 1st ed., Springer, Berlin, Heidelberg, 2008.  K. Rabaey, G. Lissens, S.D. Siciliano, W. Verstraete, Biotechnol. Lett. 25 (2003) 1531.  F. Aulenta, A. Catervi, M. Majone, S. Panero, P. Reale, S. Rossetti, Environ. Sci. Technol. 41 (2007) 2554.  K. Rabaey, R.A. Rozendal, Nat. Rev. Microbiol. 8 (2010) 706.  A. Kumlanghan, J. Liu, P. Thavarungkul, P. Kanatharana, B. Mattiasson, Biosens. Bioelectron. 22 (2007) 2939.  S.K. Chaudhuri, D.R. Lovley, Nat. Biotechnol. 21 (2003) 1229.  F. Harnisch, U. Schroder, Chem. Soc. Rev. 39 (2010) 4433.  S. Freguia, K. Rabaey, Z. Yuan, J. Keller, Electrochim. Acta 53 (2007) 598.  S. Tsujimura, Y. Miura, K. Kano, Electrochim. Acta 53 (2008) 5716.  W. Choosri, R. Paukner, P. Wuhrer, D. Haltrich, C. Leitner, World J. Microbiol. Biotechnol. 27 (2011) 1349.  A.T. Yahiro, S.M. Lee, D.O. Kimble, Biochim. Biophys. Acta 88 (1964) 375.  P.W.M. van Dijck, G.C.M. Selten, R.A. Hempenius, Regul. Toxicol. Pharm. 38 (2003) 27.  M.C. Potter, Proc. R. Soc. Lond. Ser. B: Biol. Sci. 84 (1911) 260.  M. Masuda, S. Freguia, Y.F. Wang, S. Tsujimura, K. Kano, Bioelectrochemistry 78 (2010) 173.  K. Rabaey, N. Boon, M. Hofte, W. Verstraete, Environ. Sci. Technol. 39 (2005) 3401.
 S. Freguia, M. Masuda, S. Tsujimura, K. Kano, Bioelectrochemistry 76 (2009) 14.  H. Richter, K.P. Nevin, H.F. Jia, D.A. Lowy, D.R. Lovley, L.M. Tender, Energy Environ. Sci. 2 (2009) 506.  G. Reguera, K.D. McCarthy, T. Mehta, J.S. Nicoll, M.T. Tuominen, D.R. Lovley, Nature 435 (2005) 1098.  B.E. Logan, Nature reviews, Microbiology 7 (2009) 375.  D.R. Lovley, Curr. Opin. Biotechnol. 19 (2008) 564.  K. Rabaey, L. Angenent, U. Schröder, J. Keller, in: P. Lens (Ed.), Integrated Environmental Technology Series, IWA Publishing, London, 2010.  M. Rosenbaum, F. Aulenta, M. Villano, L.T. Angenent, Bioresour. Technol. 102 (2011) 324.  S. Srinivasan, Fuel Cells From Fundamentals to Applications, Springer Science & Business Media, New York, 2006.  R. Ludwig, W. Harreither, F. Tasca, L. Gorton, Chemphyschem 11 (2010) 2674.  M.J. Moehlenbrock, T.K. Toby, A. Waheed, S.D. Minteer, J. Am. Chem. Soc. 132 (2010) 6288.  H. Rismani-Yazdi, A.D. Christy, B.A. Dehority, M. Morrison, Z. Yu, O.H. Tuovinen, Biotechnol. Bioeng. 97 (2007) 1398.  J. Niessen, U. Schroder, F. Harnisch, F. Scholz, Lett. Appl. Microbiol. 41 (2005) 286.  J. Heilmann, B.E. Logan, Water Environ. Res. 78 (2006) 531.  M. Sakamoto, K. Takamura, Bioelectrochem. Bioenerg. 9 (1982) 571.  B. Persson, L. Gorton, G. Johansson, A. Torstensson, Enzyme Microb. Technol. 7 (1985) 549.  N. Yuhashi, M. Tomiyama, J. Okuda, S. Igarashi, K. Ikebukuro, K. Sode, Biosens. Bioelectron. 20 (2005) 2145.  P.M. Allen, W.R. Bowen, Trends Biotechnol. 3 (1985) 145.  B. Willner, E. Katz, I. Willner, Curr. Opin. Biotechnol. 17 (2006) 589.  C. Kohlmann, W. Markle, S. Lutz, J. Mol. Catal. B-Enzym. 51 (2008) 57.  S. Xu, S.D. Minteer, ACS Catal. 2 (2012) 91.  F. Tasca, L. Gorton, M. Kujawa, I. Patel, W. Harreither, C.K. Peterbauer, R. Ludwig, G. Noll, Biosens. Bioelectron. 25 (2010) 1710.  J. Wang, Chem. Rev. 108 (2008) 814.  L. Heinemann, Diabet. Technol. Ther. 12 (2010) 847.  S.C. Barton, J. Gallaway, P. Atanassov, Chem. Rev. 104 (2004) 4867.  M.T. Madigan, J.M. Martink, J. Parker, Brock Biology of Microorganisms, 8th ed., Prentice Hall International, 1999.  S.A. Patil, F. Harnisch, B. Kapadnis, U. Schroder, Biosens. Bioelectron. 26 (2010) 803.  P.T. Moseley, J. Power Sources 95 (2001) 218.  S.A. Patil, F. Harnisch, C. Koch, T. Hubschmann, I. Fetzer, A.A. CarmonaMartinez, S. Muller, U. Schroder, Bioresour. Technol. 102 (2011) 9683.  L. Huang, B.E. Logan, Appl. Microbiol. Biotechnol. 80 (2008) 655.  J.J. Fornero, M. Rosenbaum, L.T. Angenent, Electroanalysis 22 (2010) 832.  S.L. Chen, H.Q. Hou, F. Harnisch, S.A. Patil, A.A. Carmona-Martinez, S. Agarwal, Y.Y. Zhang, S. Sinha-Ray, A.L. Yarin, A. Greiner, U. Schroder, Energy Environ. Sci. 4 (2011) 1417.  A. Venkataraman, M. Rosenbaum, J.B.A. Arends, R. Halitschke, L.T. Angenent, Electrochem. Commun. 12 (2010) 459.  A. Venkataraman, M.A. Rosenbaum, S.D. Perkins, J.J. Werner, L.T. Angenent, Energy Environ. Sci. 4 (2011) 4550.  J. Niessen, F. Harnisch, M. Rosenbaum, U. Schroder, F. Scholz, Electrochem. Commun. 8 (2006) 869.  R. Arechederra, S.D. Minteer, Electrochim. Acta 53 (2008) 6698.  D. Sokic-Lazic, S.D. Minteer, Biosens. Bioelectron. 24 (2008) 939.  M.E. Hernandez, D.K. Newman, Cell. Mol. Life Sci. 58 (2001) 1562.  M. Malki, A.L. De Lacey, N. Rodrfguez, R. Amils, V.M. Fernandez, Appl. Environ. Microbiol. 74 (2008) 4472.  S. Freguia, K. Rabaey, Z. Yuan, J. Keller, Environ. Sci. Technol. 42 (2008) 7937.  F. Harnisch, C. Koch, S.A. Patil, T. Hubschmann, S. Muller, U. Schroder, Energy Environ. Sci. 4 (2011) 1265.  K. Rabaey, N. Boon, S.D. Siciliano, M. Verhaege, W. Verstraete, Appl. Environ. Microbiol. 70 (2004) 5373.  C.I. Torres, R. Krajmalnik-Brown, P. Parameswaran, A.K. Marcus, G. Wanger, Y.A. Gorby, B.E. Rittmann, Environ. Sci. Technol. 43 (2009) 9519.  M.A. Rosenbaum, H.Y. Bar, Q.K. Beg, D. Segre, J. Booth, M.A. Cotta, L.T. Angenent, Bioresour. Technol. 102 (2011) 2623.  F. Harnisch, U. Schroder, M. Quaas, F. Scholz, Appl. Catal. B-Environ. 87 (2009) 63.  S. Freguia, K. Rabaey, J. Keller, Proceedings of the 11th Anaerobic Digestion Congress, 23–27 September 2007, Brisbane, Australia, 2007.  B.H. Kim, I.S. Chang, G.M. Gadd, Microbial fuel cells as biochemical oxygen demand (BOD) and toxicity sensors, in: K. Rabaey, L. Angenent, U. Schroder, J. Keller (Eds.), Bio-electrochemical Systems: From Extracellular Electron Transfer to Biotechnological Application, IWA Publishing, London, 2010.  D. Odaci, S. Timur, A. Telefoncu, Bioelectrochemistry 75 (2009) 77.  S. Patil, F. Harnisch, U. Schroder, Chemphyschem: Eur. J. Chem. Phys. Phys. Chem. 11 (2010) 2834.  R.A. Bullen, T.C. Arnot, J.B. Lakeman, F.C. Walsh, Biosens. Bioelectron. 21 (2006) 2015.  R.A. Rozendal, F. Harnisch, A.W. Jeremiasse, U. Schroder, in: K. Rabaey, L. Angenent, U. Schroder, J. Keller (Eds.), Bio-electrochemical Systems: From Extracellular Electron, Transfer to Biotechnological Application, IWA Publishing, London, 2010.  F. Harnisch, S. Wirth, U. Schroder, Electrochem. Commun. 11 (2009) 2253.
S. Freguia et al. / Electrochimica Acta 82 (2012) 165–174
 R.A. Rozendal, E. Leone, J. Keller, K. Rabaey, Electrochem. Commun. 11 (2009) 1752.  F. Zhao, F. Harnisch, U. Schrorder, F. Scholz, P. Bogdanoff, I. Herrmann, Environ. Sci. Technol. 40 (2006) 5193.  S. Tsujimura, Y. Kamitaka, K. Kano, Fuel Cells 7 (2007) 463.  C.F. Blanford, R.S. Heath, F.A. Armstrong, Chem. Commun. (2007) 1710–1712.  K. Rabaey, S.T. Read, P. Clauwaert, S. Freguia, P.L. Bond, L.L. Blackall, J. Keller, Isme J. 2 (2008) 519.  S. Freguia, K. Rabaey, Z. Yuan, J. Keller, Water Res. 42 (2008) 1387.  S.C. Barton, H.H. Kim, G. Binyamin, Y.C. Zhang, A. Heller, J. Am. Chem. Soc. 123 (2001) 5802.  J.A. Cracknell, K.A. Vincent, F.A. Armstrong, Chem. Rev. 108 (2008) 2439.  O. Schaetzle, F. Barriere, U. Schroder, Energy Environ. Sci. 2 (2009) 96.  A. Zebda, C. Gondran, A. Le Goff, M. Holzinger, P. Cinquin, S. Cosnier, Nat. Commun. (2011) 2.  C. Vaz-Dominguez, S. Campuzano, O. Rudiger, M. Pita, M. Gorbacheva, S. Shleev, V.M. Fernandez, A.L. De Lacey, Biosens. Bioelectron. 24 (2008) 531.  M.J. Cooney, V. Svoboda, C. Lau, G. Martin, S.D. Minteer, Energy Environ. Sci. 1 (2008) 320.  N. Mano, H.H. Kim, Y.C. Zhang, A. Heller, J. Am. Chem. Soc. 124 (2002) 6480.  A. Mollica, A. Trevis, 4th International Congress on Marine Corrosion, Antibes, France, 1976, p. 351.  V. Scotto, R. Dicintio, G. Marcenaro, Corros. Sci. 25 (1985) 185.  I. Dupont, D. Feron, G. Novel, Int. Biodeterior. Biodegrad. 41 (1998) 13.  V. L’Hostis, C. Dagbert, D. Feron, Electrochim. Acta 48 (2003) 1451.  M. Mehanna, R. Basseguy, M.L. Delia, A. Bergel, Electrochem. Commun. 11 (2009) 568.  P. Clauwaert, D. Van der Ha, N. Boon, K. Verbeken, M. Verhaege, K. Rabaey, W. Verstraete, Environ. Sci. Technol. 41 (2007) 7564.  B. Erable, I. Vandecandelaere, M. Faimali, M.L. Delia, L. Etcheverry, P. Vandamme, A. Bergel, Bioelectrochemistry 78 (2010) 51.  M. Faimali, E. Chelossi, F. Garaventa, C. Corra, G. Greco, A. Mollica, Electrochim. Acta 54 (2008) 148.  S. Carbajosa, M. Malki, R. Caillard, M.F. Lopez, F.J. Palomares, J.A. Martin-Gago, N. Rodriguez, R. Amils, V.M. Fernandez, A.L. De Lacey, Biosens. Bioelectron. 26 (2010) 877.
 S. Freguia, S. Tsujimura, K. Kano, Electrochim. Acta 55 (2010) 813.  B. Virdis, K. Rabaey, R.A. Rozendal, Z.G. Yuan, J. Keller, Water Res. 44 (2010) 2970.  A. Badura, T. Kothe, W. Schuhmann, M. Rogner, Energy Environ. Sci. 4 (2011) 3263.  A. Melis, Plant Sci. 177 (2009) 272.  M. Rosenbaum, U. Schroder, Electroanalysis 22 (2010) 844.  D.P.B.T.B. Strik, R.A. Timmers, M. Helder, K.J.J. Steinbusch, H.V.M. Hamelers, C.J.N. Buisman, Trends Biotechnol. 29 (2011) 41.  R. Carpentier, S. Lemieux, M. Mimeault, M. Purcell, D.C. Goetze, Bioelectrochem. Bioenerg. 22 (1989) 391.  K. Amako, H. Yanai, T. Ikeda, T. Shiraishi, M. Takahashi, K. Asada, J. Electroanal. Chem. 362 (1993) 71.  A. Badura, B. Esper, K. Ataka, C. Grunwald, C. Woll, J. Kuhlmann, J. Heberle, M. Rogner, Photochem. Photobiol. 82 (2006) 1385.  J. Maly, J. Krejci, M. Ilie, L. Jakubka, J. Masojidek, R. Pilloton, K. Sameh, P. Steffan, Z. Stryhal, M. Sugiura, Anal. Bioanal. Chem. 381 (2005) 1558.  J. Maly, J. Masojidek, A. Masci, M. Ilie, E. Cianci, V. Foglietti, W. Vastarella, R. Pilloton, Biosens. Bioelectron. 21 (2005) 923.  A. Badura, D. Guschin, B. Esper, T. Kothe, S. Neugebauer, W. Schuhmann, M. Rogner, Electroanalysis 20 (2008) 1043.  A. Badura, D. Guschin, T. Kothe, M.J. Kopczak, W. Schuhmann, M. Rogner, Energy Environ. Sci. 4 (2011) 2435.  M. Edelman, A.K. Mattoo, Photosynth. Res. 98 (2008) 609.  M. Katterle, V.I. Prokhorenko, A.R. Holzwarth, A. Jesorka, Chem. Phys. Lett. 447 (2007) 284.  E.M. Aro, M. Suorsa, A. Rokka, Y. Allahverdiyeva, V. Paakkarinen, A. Saleem, N. Battchikova, E. Rintamaki, J. Exp. Bot. 56 (2005) 347.  J.M. Pisciotta, Y. Zou, I.V. Baskakov, PLoS One (2010) 5.  T. Yagishita, S. Sawayama, K.I. Tsukahara, T. Ogi, Solar Energy 61 (1997) 347.  S. Tsujimura, A. Wadano, K. Kano, T. Ikeda, Enzyme Microb. Technol. 29 (2001) 225.  G. Wohlfahrt, S. Witt, J. Hendle, D. Schomburg, H.M. Kalisz, H.J. Hecht, Acta Crystallogr. D 55 (1999) 969.