|Home | About | Journals | Submit | Contact Us | Français|
Geobacter sulfurreducens can form electrically conductive biofilms, but the potential for conductivity through mixed-species biofilms has not been examined. A current-producing biofilm grown from a wastewater sludge inoculum was highly conductive with low charge transfer resistance even though microorganisms other than Geobacteraceae accounted for nearly half the microbial community.
The discovery of long-range electron transport through electronically conductive biofilms offers new possibilities in microbe-electrode interactions and bioelectronics (11, 12, 17, 22) and has revealed the potential for microorganisms to make direct electrical connections for interspecies electron transfer (9, 19, 27). Most biofilms that have been studied are insulating (1, 2, 17, 20). The possibility of electrically conductive biofilms was first suggested based on the findings that (i) Geobacter sulfurreducens produced thick (40 to 50 μm) biofilms when growing on anode surfaces; (ii) biofilm cells not in contact with the anode contributed to current production as much as cells in direct contact; and (iii) the production of thick current-producing biofilms was dependent on the presence of conductive pili (25). Subsequent studies modeling current production in biofilms in which Geobacter species predominated found that it was necessary to include an empirically fitted conductivity value in the model in order to accurately predict observed current densities (6, 29).
Direct measurements of conductivity in current-producing biofilms of Geobacter sulfurreducens revealed high conductivities, rivaling those of synthetic conducting polymers (17). Multiple lines of evidence indicated that, as previously proposed (25), conductivity could be attributed to a network of pili (17). Surprisingly, the pili have metal-like conductivity (17). Metal-like conductivity is a new paradigm for long-range electron transport in biological systems (13, 22), and it has been suggested that electron hopping between c-type cytochromes in biofilms, a more traditional mechanism of electron transfer, might account for electron transport through G. sulfurreducens biofilms (26). However, many experimental findings refute the electron-hopping hypothesis (13, 16).
Conductivity through biofilms is essential for high current densities in microbial fuel cells (MFCs), because it permits microorganisms not in direct contact with the anode to contribute to current production (11, 15, 25, 28). Conductive networks may also make it possible for microorganisms to directly exchange electrons in syntrophic partnerships (19, 27), which may be a more efficient mode of syntrophic interaction than interspecies hydrogen transfer (9).
The anodes of microbial fuel cells generating current from wastewater or organic matter in aquatic sediments can be colonized by a diversity of microorganisms (7, 10). In order to evaluate the conductivity of a mixed-species current-producing biofilm, an inoculum of anaerobic digester sludge from the Pittsfield, MA, wastewater treatment plant was prepared as described earlier (21) and immediately inoculated into previously described (17) “ministack” microbial fuel cells that contained two gold anodes (6.45-cm2 total geometric area) separated by a 50-μm nonconducting gap. Anodes were connected by a 560-Ω load to a carbon-cloth cathode which was immersed in a 50 mM FeCN solution. External potential was not applied to the anode for the MFC operation, ensuring true fuel cell mode. Acetate (10 mM) served as the electron donor, and the incubation temperature was 37°C. All results were confirmed by repeated measurements on multiple biofilms.
The production of current in the microbial fuel cells (Fig. 1a) was associated with the growth of a biofilm that covered the two anodes and converged, bridging the nonconducting gap (Fig. 1b; see also Fig. 3a). When electrical conductance across the gap was measured as previously described (17), there was significant biofilm conductance (Fig. 1c). Biofilm conductivity (Fig. 1d), calculated with conformal mapping as previously described (17), was comparable to that previously reported for current-producing biofilms of strain KN400 (17) (see supplemental material for details). As previously described, the effluent from the anode chamber was passed to another chamber which was identical with the exception that the two gold electrodes were not connected to the cathode (17). No biofilm grew in the control chamber, and conductance between the two electrodes was low (Fig. 1c). The demonstrated high electrical conductivity of mixed-species-derived biofilms provides an explanation for their capacity for high current densities (0.9 ± 0.45 A/m2) comparable to those obtained with G. sulfurreducens biofilms grown in the same type of ministack microbial fuel cells (0.7 A/m2) under similar conditions (17, 21).
Charge transfer resistance represents an energy barrier at the electrode interface (15, 28). In addition to promoting long-range electron transport through biofilms, high biofilm conductivity can lower the charge transfer resistance (Rct) at the biofilm/anode interface because electrons reaching the biofilm/anode interface after traveling through a biofilm with higher conductivity have greater energy than electrons transported through biofilms of lower conductivity (15, 28). This higher energy reduces the energy barrier at the biofilm/anode interface that lowers the charge transfer resistance. This possibility was evaluated by measuring the charge transfer resistance using electrochemical impedance spectroscopy (15, 23). In this configuration, the two sides of the split anode were connected and used as the working electrode. The reference electrode was Ag/AgCl, placed in the anode chamber, and the counter-electrode was a carbon cloth, placed in the cathode chamber. The anode was disconnected from the cathode, and all of the impedance measurements were performed at the open-circuit potential of the anode (−550 mV versus Ag/AgCl) (14, 15). For all comparisons between mixed-species and G. sulfurreducens biofilms, the amplitude excitation was 0.1 V of alternating current (ac) (4, 5). Linearity of the ac signal was ensured by measuring impedance over an amplitude range of 0.001 V to 0.1 V at the open-circuit potential (see Fig. S2 in the supplemental material). The charge transfer resistance was evaluated from the measured impedance spectra by fitting (see Table S1 in the supplemental material) the previously described (15, 18, 23) equivalent circuit model (see Fig. S3 in the supplemental material). As expected, the charge transfer resistance of the mixed-species biofilms was much lower than that seen in uninoculated controls (Fig. 2a; see also Fig. S4 in the supplemental material). Charge transfer resistance of the mixed-species biofilms declined with increased maturity, presumably reflecting enhanced electrical contact with the anode (Fig. 2b). In comparison to other measurements of charge transfer resistance made under similar conditions of open-circuit potential, the charge transfer resistance of the mature mixed-species biofilms grown on gold electrodes (1.45 ± 0.32 kΩ·cm2) was higher than the 0.48 kΩ·cm2 previously reported for another mixed-species biofilm grown on carbon electrodes (24) but comparable with that seen with biofilms of G. sulfurreducens strain KN400 (1.1 kΩ·cm2) and much lower than the 204 kΩ·cm2 reported for biofilms of Shewanella oneidensis (18).
Current-producing mixed-species biofilms had two distinct layers (Fig. 3a)—a top, outer brown layer that was loosely attached to the anode (Fig. 3b) and a bottom, inner pink layer that was strongly attached to the anode (Fig. 3c). In order to identify the microorganisms which were conferring conductivity to mixed-species biofilms, clone libraries of 16S rRNA genes were constructed from the initial inoculum, as well as for the inner pinkish biofilm, which was closely attached to the electrode, and for the outer brownish biofilm, which was loosely attached to the electrode. At day 54, the outer and inner layers of the biofilm were individually sampled with a micropipette for community analysis. As previously described (3, 19), genomic DNA was extracted and 16S rRNA gene sequences were amplified with PCR, cloned, and sequenced. The detailed experimental procedure for community analysis is provided in the supplemental material, and the results are presented in Fig. 3c. The proportion of Geobacteraceae in the initial inoculum was 8%, whereas the proportions of Geobacteraceae in the inner and the outer biofilm zones were ca. 50% and 10%, respectively.
The finding that mixed-species biofilms can possess electrical conductivity comparable to that of pure culture biofilms of G. sulfurreducens with low charge transfer resistance provides an explanation for the capacity of mixed-species biofilms to produce the thick biofilms necessary for high current densities. Modeling studies have previously demonstrated that invoking a highly conductive biofilm could explain the effective function of high-current-density multispecies biofilms in which Geobacter species predominated (6, 29, 30). The conductivity of the mixed-species biofilms was an order of magnitude higher than that of multispecies methanogenic aggregates derived from wastewater digesters (19) and 2 orders of magnitude higher than that of dual-species Geobacter aggregates (27). There may be stronger selection for higher conductivity in current-producing biofilms than in the previously described conductive aggregates, because electrons released from microorganisms near the outer surface of current-producing biofilms need to be transported much farther than in cell aggregates, in which electrons need to be transported only to nearby cells.
It is not possible to determine from the data available whether microorganisms other than Geobacter species contributed to the conductivity of the mixed-species biofilms. Biofilms of Escherichia coli and Pseudomonas aeruginosa grown on the two-electrode device described here were not conductive (17), and other microbial biofilms were also found to have poor conductivity (1, 2, 20). Other current-producing microorganisms such as Shewanella oneidensis (8) and Thermincola strain JR (31) do not form thick biofilms when producing current, suggesting that they are incapable of forming highly conductive biofilms. Thus, these results indicate that biofilms containing high proportions of organisms other than Geobacter species may be conductive, but whether the other organisms contribute to biofilm conductivity warrants further investigation.
This research was supported by the Office of Naval Research (grant no. N00014-10-1-0084 and N00014-12-1-0229), the Office of Science (BER), and the U.S. Department of Energy (award no. DE-SC0004114 and Cooperative Agreement no. DE-FC02-02ER63446) as well as the NSF Center for Hierarchical Manufacturing (grant no. CMMI-1025020).
We thank Trevor Woodard for assistance with wastewater sludge and Pravin Shrestha for help with community analysis.
Published ahead of print 15 June 2012
Supplemental material for this article may be found at http://aem.asm.org/.