Evaluation of artificial neural network algorithms for predicting the effect of the urine flow rate on the power performance of microbial fuel cells.

Microbial fuel cell (MFC) power performance strongly depends on the biofilm growth, which in turn is affected by the feed flow rate. In this work, an artificial neural network (ANN) approach has been used to simulate the effect of the flow rate on the power output by ceramic MFCs fed with neat human urine. To this aim, three different second-order algorithms were used to train our network and then compared in terms of prediction accuracy and convergence time: Quasi-Newton, Levenberg-Marquardt, and Conjugate Gradient. The results showed that the three training algorithms were able to accurately simulate power production. Amongst all of them, the Levenberg-Marquardt was the one that presented the highest accuracy (R = 95%) and the fastest convergence (7.8 s). These results show that ANNs are useful and reliable tools for predicting energy harvesting from ceramic-MFCs under changeable flow rate conditions, which will facilitate the practical deployment of this technology.

A New Method for Modulation, Control and Power Boosting in Microbial Fuel Cells.

Microbial fuel cells (MFCs) are energy transducers, which through the metabolic reactions of facultative anaerobic microorganisms, transform the energy in organic matter directly into electricity. Extrinsic parameters such as hydraulic retention time, fuel quality (type and concentration) and physicochemical environment of electrodes and biofilms (e.g., temperature, pH, salinity, and redox), can all influence system efficiency. This work proposes that MFCs can be "fine-tuned" by adjustment of any of the physicochemical conditions including redox potential; in this context, an entirely novel method was investigated as a practical means of tuning, modulating and monitoring the redox potential within the electrode chambers. The method uses additional electrodes - known as 3rd and 4th-pins for anode and cathode chambers, respectively - which can be used in individual units, modules, cascades or stacks, for optimising the production of a large variety of chemicals, as well as biomass, water and power. The results have shown that the power output modulation resulted in an up to 79% and 33% increase, when connected via 3rd and 4th pins, respectively. Apart from power improvement, this study also demonstrated a method of open circuit potential (OCP) sensing, by using the same additional electrodes to both monitor and control the MFC signal in real time.

Microalgae as substrate in low cost terracotta-based microbial fuel cells: Novel application of the catholyte produced.

In this work, the by-product generated during the operation of cylindrical MFCs, made out of terracotta material, is investigated as a feasible means of degrading live microalgae for the first time. In addition to the low cost materials of this design, the reuse of the solution produced in the cathode renders the technology truly green and capable of generating bioenergy. In this study, the effect of a light/dark cycle or dark conditions only on the digestion of live microalgae with the catholyte is investigated. The results show that a combination of light/dark improves degradation and allows algae to be used as substrate in the anode. The addition of 12.5mL of a 1:1 mix of catholyte and microalgae (pre-digested over 5days under light/dark) to the anode, increases the power generation from 7µW to 44µW once all the organic matter in the anode had been depleted.

Self sufficient wireless transmitter powered by foot-pumped urine operating wearable MFC.

The first self-sufficient system, powered by a wearable energy generator based on microbial fuel cell (MFC) technology is introduced. MFCs made from compliant material were developed in the frame of a pair of socks, which was fed by urine via a manual gaiting pump. The simple and single loop cardiovascular fish circulatory system was used as the inspiration for the design of the manual pump. A wireless programmable communication module, engineered to operate within the range of the generated electricity, was employed, which opens a new avenue for research in the utilisation of waste products for powering portable as well as wearable electronics.

Intermittent load implementation in microbial fuel cells improves power performance.

This study reports on the response of small-scale MFCs to intermittent loading, in terms of power output over time. The aim was to understand the evolution with time of power output under different duty cycles, in conditions close to practical implementation. Inexpensive ceramic membranes were compared to cation exchange membranes, under continuous flow and with a pre-digester connected. Results show that at the minute-scale, all the duty cycles investigated, produced 78% higher power bursts from the MFCs (500µW) than when under continuous loading (280µW). These results were recorded from MFCs employing ceramic membranes, whereas the difference in performance for MFCs employing commercially available cation-exchange-membranes was insignificant. When normalising to daily energy production, only specific duty cycles produced more power than continuous loading. Furthermore, the introduction of a pre-digester increased the MFC power outputs 10-fold, thus confirming that separating fermentation from electro-active respiration, significantly enhances the system performance.

Bi-directional electrical characterisation of microbial fuel cell.

The electrical performance of microbial fuel cells in steady-state is usually investigated by standard characterisation methods that reveal many important parameters e.g. maximum power. This paper introduces a novel "bi-directional" method to study how the acquisition parameters (i.e. sweep rate and sweep regime) can influence measurements and consequently performance estimations. The investigation exhibited considerable differences (hysteresis) between the forward and backward characterisation regimes, indicating a difficulty to reach steady-state under certain conditions. Moreover, it is found that fast sweep rates (time-step of 2 min) can lead to an overestimation of the short-circuit currents, while prolonged operation with high external loads leads to maximum power overestimation and extended conditioning at high currents can result in its underestimation.