Dynamic modelling and optimisation of microbial fuel cells and microbial electrolysis cells.

摘要

The first contribution of this thesis is the development of microbial fuel cell (MFC) and microbial electrolysis cells (MEC) models capable of describing the dynamics of substrate consumption, microorganism's growth, and electricity (MFCs) or H2 (MECs) generation. By using ordinary differential equations to describe biomass growth and substrate consumption in the anodic compartments, a fast numerical solution was found for both models. First a MFC model describing the acetate competition between electricigenic and acetoclastic methanogenic microorganisms was developed. The MEC model foundation was based on the concepts presented in the MFC model. By including fermentative and hydrogenotrophic methanogenic microorganisms, the MEC model was able to predict hydrogen production from wastewater degradation. Model parameters were estimated for both models with experimental results obtained in continuous flow, gas diffusion cathode MFCs and MECs. Only model parameters with small confidence intervals were selected to be estimated. Moreover, using independent experimental data sets, both models were validated and were successful in describing experimental results at diverse operating conditions.;A further contribution of this thesis is the analysis of both models for process optimisation. Preliminary analysis demonstrated the influence of operating conditions on product generation for both models. Interestingly, the external resistance and the applied voltage (manipulated variables for MFCs and MECs respectively) were shown to significantly influence the biofilm microbial composition. This aspect was further analysed for the MFC model with a steady state analysis of the biofilm composition. It was shown that depending on the selection of the external resistance, the MFC biofilm could present three scenarios: (i) the coexistence of both microbial populations; or the exclusion of one of the microbial population with (ii) only electricigens, or (iii) only methanogens present. Following these results, a comparison between the substrate consumption of the three scenarios was performed, showing that coexistence always leads to lower substrate consumption. The treatment capacity of MFCs was then optimised by reactor staging. The optimum treatment capacity of a unit with two staged reactors was shown to depend on the influent concentration and effluent requirement. Finally, experiments using acetate-fed MFCs were presented to qualitatively confirm the effects of external resistance on the biofilm composition.;The last contribution of this thesis is the presentation of a unified MFC/MEC model, which includes electricigenic and methanogenic microorganisms in the anode compartment, while the electrochemical balance accounts for the cathode differences between the MFC and MEC. The model is first analysed in terms of biofilm composition, which is shown to depend on the reactor's operating current. Once more, biofilm coexistence was present only for a defined interval of operating current. An optimisation study was performed to maximise electricity (MFC) or H2 production (MEC), while respecting a treatment requirement. By defining power productivity functions for both reactors, analytical optimum current expressions were found and were shown to be the same for MFCs and MECs. Furthermore, these expressions were dependent on the reactor's internal resistance. Finally, an alternative MEC productivity function was defined and analysed. This productivity was a function of the H2 production efficiency and its unique analytical optimum solution was shown to be independent of the reactor's internal resistance. (Abstract shortened by UMI.)