Speaker
Description
Modelling reactive transport (RT) in fractured crystalline rocks plays a crucial role in controlling the migration and retention of radionuclides in geological repositories. Conventional RT models typically describe the retardation of radionuclides using bulk parameters such as the distribution coefficient (Kd), which is assumed to depend primarily on average mineral composition. While this approach is computationally efficient, it neglects the fact that groundwater interact sequentially with different mineral surfaces along the flow paths within fractures. As a consequence, the order in which the different minerals are encountered along the flow path may influence the evolution of solution chemistry and the retardation of radionuclides. As this contribution shows, mineral sequence effects, defined as changes in the reactive transport behaviour, can arise solely from the order of mineral-solution interaction, even when the same minerals are present in the same amount along the flow path. The central hypothesis is that sequence effects persist under realistic subsurface conditions due to nonlinear reaction behaviour, finite reaction kinetics and limited transport times. Indeed, natural systems are characterised by slow flow velocities, heterogeneous mineralogy and kinetically limited reactions, which prevent complete equilibrium and should allow sequence-dependent effects to persist. To investigate these effects, a compartment-based reactive transport framework was developed where each mineral surface is represented as an individual compartment. A solution with fixed initial composition passes sequentially through these compartments, and geochemical interactions within each compartment is simulated using ideal equilibrium reactions modelled with PHREEQC. This stepwise modelling approach allows the cumulative impact of mineral sequence on sorption and retardation to be analysed under controlled conditions. It is shown that nonlinear reaction mechanisms such as surface site saturation, pH-dependent sorption, and competitive binding processes can lead to sequence-dependent retention not properly described by Kd values of average mineralogy. An important finding is that the transport of different radionuclides and other relevant interacting elements, can lead to different sequence-dependent retention.