Subsurface contamination is common at nuclear sites and it is likely that radioactive wastes will be managed in the long-term via burial in a deep Geological Disposal Facility (GDF). The migration of radionuclides in the geosphere from such sites is a major societal concern. In particular, long-lived, redox-active radionuclides (in the case of this thesis: 99Tc and Np) can migrate over large distances due to their high solubility under oxic conditions. Bioremediation has been proposed as a mechanism to limit the migration of 99Tc and Np in the environment. Here, an electron donor is supplied to the subsurface and soluble Tc(VII) and Np(V) are reduced to poorly soluble Tc(IV) and Np(IV), respectively. Reduction occurs via direct microbial action (termed bioreduction) or through radionuclide reaction with the by-products of microbial metabolism (primarily Fe(II)). Given the ubiquity of microorganisms and Fe in the geosphere, similar reactions can be expected in the deep subsurface surrounding a GDF. Once reduced, the long-term stability of the Tc(IV) and Np(IV) phases will significantly impact migration rates. Oxidative dissolution of Tc(IV)- and Np(IV)-bearing solids has been demonstrated in the literature and can be pervasive, thus questioning the efficacy of bioreduction. However, these studies have been conducted over short time-scales and during a single period of oxidation.Given the long half-life of 99Tc and Np and the ephemeral nature of redox conditions in the subsurface, there is a need to better understand 99Tc and Np biogeochemistry during longer time-scales and across multiple redox cycles. In this thesis, microcosm experiments have been used to address this knowledge gap. Sediment and groundwater used in the microcosms were representative of the Sellafield Ltd. nuclear site. For Tc, three successive redox cycles (reduction followed by oxidation with O2) over 2 years, gradually reduced the extent of Tc remobilisation during oxidation, and molecular scale characterisation of solids revealed that sediment associated Tc was always present as Tc(IV). Further, over time sequential extractions and EXAFS revealed an increased significance of Tc-Fe bonding in the sediment at the expense of TcO2. Despite this, a small but significant fraction of Tc(IV) was also found to be stable in solution during the experiments and XAS and TEM analysis suggested this was Tc(IV) associated with magnetite colloids. In other experiments completed with higher concentrations of bioavailable Fe (added as ferrihydrite to sediments, and in pure culture experiments with Geobacter sulfurreducens), the significance of Tc-Fe bonding was again highlighted, and potential Tc(IV) incorporation into biogenic magnetite was also documented. In experiments with Np, virtually all of the Np(V) added to oxic groundwater was removed to the sediment commensurate with microbially mediated Fe(III) reduction. Further, in systems with elevated bioavailable Fe, Np removal from solution was more extensive. Taken together, the data for Tc and Np reveals critical links between redox-active radionuclides and Fe cycling in sediments over periods of years and across multiple redox cycles. Furthermore, these processes help to predict the long-term fate of radioactive contamination at the Sellafield Ltd. nuclear site and have implications for contaminated land worldwide.