Various phenoxazine, phenothiazine, phenazine, indophenol and bipyridilium derivatives were tested for their effectiveness as redox mediators in microbial fuel cells containing Alcaligenes eutrophus, Bacillus subtilis, Escherichia coli, or Proteus vulgaris as the active biological agent, and glucose or succinate as the oxidisable substrate. A ferricyanide-Pt cathode was used. The open-circuit cell e.m.f.'s increased in the order of increasing negative formal redox potentials at pH 7(Ez) of the redox compounds. Several of the redox agents worked well as mediators, maintaining steady currents over several hours, and thionine was found to be particularly effective in maintaining relatively high cell voltages when current was drawn from the cell. A number of the compounds tested did not function well, either because they were incompletely or slowly reduced by the microorganisms or because of their instability. P. vulgaris, with thionine as mediator and glucose as substrate, showed the best performance in a fuel cell. This system was examined in some detail under various conditions of external load to establish the effects of organism concentration, mediator concentration, and substrate addition. Coulombic outputs from these cells were calculated by integration of the current-time plots. Coulombic yields of 30-60% were obtained, on the basis of (theoretical) complete oxidation of added substrate to C02 and water.

Redox mediators promote electron transfer in microbial fuel cells. The reduction of a range of redox mediators by bacteria was studied in some detail in order to identify effective mediator—organism combinations. Rates of reduction of mediator dyes by bacteria were measured spectrophotometrically at 30°C under anaerobic conditions for standardised concentrations of organism, substrate and dye. The kinetics of dye reduction showed two general patterns: a simple, exponential curve or a complex curve with an initial linear rate followed by a faster exponential rate of reduction. Dye‐reduction rates were greater than rates of oxygen consumption (QO2) for several combinations of organism and redox dye. Thionine, brilliant cresyl blue, methylene blue and benzyl viologen were tested in combination with Alcaligenes eutrophus, Azotobacter chroococcum, Bacillus subtilis, Escherichia coli, Proteus vulgaris, Pseudomonas aeruginosa and Pseudomonas putida, using glucose and succinate as substrates. Rates of reduction of alizarin brilliant blue, 2,6‐dichlorophenolindophenol, gallocyanine, new methylene blue, N,N‐dimethyl‐disulphonated thionine, phenazine ethosulphate, resorufin, safranine‐O, phenothiazinone and toluidine blue‐O were also measured with Pr. vulgaris only. For E. coli, both QO2 and the rate of thionine reduction increased with increasing temperature in the range 25 to 37°C, but for Pr. vulgaris thionine reduction rates did not correlate with temperature in this way. Dye‐reduction rates and QO2 for Az. chroococcum were dependent on the components of the washing solution and/or the temperature at which cell suspensions were prepared. The results are discussed in relation to the use of these dyes as electron‐transfer mediators in microbial fuel cells.

The pattern of glucose metabolism was studied in a thionine-mediated Proteus culgaris fuel cell by using lT-labelled glucose. Added glucose was rapidly taken up by the bacteria and converted to CO', acetate, lactate and a fourth product, tentatively identified as propionate. When the glucose-dependent electric current from the fuel cells had been completely discharged, about 50% of glucose carbon was found in C 0 2 , 30% in acetate, 10% incorporated in the bacteria, and the residual 10% distributed as small amounts in various soluble products. Thus, although there was transient accumulation of lactate and 'propionate', these were largely reutilized. Coulombic yield from glucose oxidation was about 50% and correlated with the amount of CO, produced. Glucose metabolism in the fuel cell was intermediate between conventional aerobic and anaerobic conditions. Under anaerobic (conventional) conditions acetate, lactate and 'propionate' were produced and not reutilized. Under aerobic conditions, acetate and lactate were only transiently produced and the rate of glucose uptake was lower. The major limitation on coulombic yield from glucose oxidation was the production of acetate.

No abstract