Optimisation of Microbial Electrochemical Technologies Through Computational Modelling

Abstract

Mathematical modelling can reduce the cost and time required to design complex systems. Microbial Electrochemical Technologies (MET) have shown high levels of potential for treat ment of wastewater alongside recovering valuable resources. However, development of large MET reactors is expensive and time-consuming. A range of models have been developed to replicate and predict laboratory-scale reactors, but few have focused on pilot-scale and larger. Following an initial literature review, we identify a suitable modelling strategy and make adjustments ready for linking with a mass transportation system. Following initial model development, we develop a mass transportation system and link it with the original MET model. This combined model simulates a single substrate and single bacterial species in a pilot-scale MET. This model is benchmarked against experimental data gathered from a pilot-scale system, showing good agreement in overall behaviour. Signi cant spatial variations in both substrate concentration and current density occurred, with large substrate gradients adjacent to the anodes. This work proved that minor adjustments to channel width increased overall reactor e ciency, highlighting optimal retention times and cartridge arrangements. Building on the single population and substrate model, additional microbial species and sub strate are introduced to better replicate an MET system integrated within a wastewater setting. The improved model was then used to optimise anode layouts, through the use of fermenting bio lms. This led to minimal performance improvements in current generation (< 10%) but can reduce the cost of reactor construction by halving anode material required. Finally, optimisation of mass delivery methods was conducted by implementing a pulsed ow to increase mixing. This variable ow method improved current generation by a maximum of 17% for the widest channels tested. This decreased to an improvement of 0.5% for high concentrations of chemical oxygen demand (COD) and narrow channels. Similar improve ments were also present for COD removal rates. This work highlighted the importance of disrupting central ow patterns in large-scale METs and should be applied to all constructed reactors going forward. The work conducted in this thesis has highlighted the bene t mathematical models can have on the testing of large-scale MET systems. Therefore this has proved pilot-scale modelling work should be continually applied to the development of METs to reduce nancial overheads and timeline for deployment within the wastewater sector.