This thesis investigates the potential utilisation of microbial electrolytic cells (MECs), driven by direct electron transfer (DET), mediated electron transfer (MET), or both, for the treatment of nitrate-polluted surface water. Nitrate pollution has been a widespread problem, both globally and locally, coinciding with the development of intensive agricultural and livestock production practices. Conventional nitrate abatement occurs via either ‘separation’ or ‘reduction-based’ processes. The former involves the use of reverse-osmosis (RO), electrodialysis, and ion exchange technologies, which are both energy intensive and produce highly concentrated nitrate effluents (i.e. brine). In contrast, reduction-based processes have classically leveraged heterotrophic microbial denitrification processes. While reduction-based processes are preferred, the need for carbon substrate supplementation (e.g. methanol, acetate) necessitates further downstream processing to prevent further environmental impacts. Autotrophic denitrification, which involves the use of inorganic carbon (such as CO2), could serve as an alternate solution to conventional reduction-based processes for nitrate abatement. Furthermore, the slower growth rates of autotrophic microorganisms have the potential of establishing a more biologically stable system compared to their heterotrophic counterparts. Autotrophic MECs have been previously investigated for surface and groundwater nitrate remediation because of their potential to minimise environmental impacts (e.g. brine formation, accumulation of denitrification intermediates and organic carbon substrates) and competitive energy consumption.

Dual-chambered MECs have been previously used to establish autotrophic denitrification systems. In these systems, the cathode acts as an electron donor for both denitrification and the reductive assimilation of inorganic carbon (e.g. CO2). Dual-chambered MECs are thus able to overcome the absence of electron donors within environmentally sourced waters (thereby preventing further chemical dosing) while also offering a carbon capture solution for the remediation of CO2. Furthermore, autotrophic denitrifying systems are characterised as being ‘biologically stable’ due to the slower bacterial growth rates compared with their heterotrophic counterparts. However, the nitrate removal rates achieved by this strategy are slower than those achieved by other conventional nitrate abatement technologies. Extensive studies have been conducted on reactor configurations to improve nitrate removal rates, typically involving the use of larger electrode surface areas (such as granular activated carbon – GAC). The central tenant of this investigation is that by improving the denitrification performance of biofilm electrodes, more compact bio-electrochemical denitrifying (BED) systems could be developed to address issues of overpotentials and pH fluctuations, which are common in larger-scale BED technologies.

In this thesis, electroactive, denitrifying microbial communities were enriched using different strategies in four parallel MECs from locally obtained sediments. Three BED reactors (BEDs 1, 3, and 4) were inoculated using microbial communities sourced from Okeover stream sediments, whereas one reactor (BED2) was inoculated using previously enriched Lowcliffe sediment samples. Three reactors (BEDs 1, 2, and 3) were operated with a cathodic poised potential of -1100 ± 2 mV (vs Ag/AgCl) while BED4 was operated using a cathodic poised potential of -700 ± 2 mV (vs Ag/AgCl). The cathodic poised potential of -1100 ± 2 mV has been demonstrated to generate hydrogen on the cathodic surface and was used for the enrichment of a hydrogenotrophic (or H2-MET) denitrifying biofilm. Conversely, -700 ± 2 mV has been widely used in prior studies for the enrichment of a DET-driven electroactive biofilm (EAB).

During the nitrate loading experiments, BEDs 2 and 3 exhibited the fastest nitrate removal rates (e.g. 4.94 ± 0.04 and 8.62 ± 0.13 g NO3--N∙m-2∙d-1 respectively), revealing the potential of utilising both H2-MET and DET for nitrate remediation. Subsequent microbial community analysis conducted on the EABs of the four reactors revealed the enrichment of hydrogenotrophic denitrifiers (Hydrogenophaga spp., Simplicispira spp., and Dechloromonas spp.) in BEDs 1, 2, and 3, and the enrichment of Candidatus Nitrotoga spp. in BED4. Next, the cathodic poised potential was varied to investigate the effect of operating poised potentials on nitrate removal rates, which demonstrated increased nitrate removal rates at lower poised potentials (< -1100 ± 2 mV vs Ag/AgCl). Finally, the treatment of nitrate-supplemented surface water revealed several operational challenges regarding the application of this method, which were later addressed with suggestions for future studies. These findings usher in further development of BED systems utilising both MET and DET while uncovering the potential of utilising such systems to improve our understanding of electroactive denitrifying communities.