Diffusion and advection of radionuclides through a cementitious backfill with potential to be used in the deep disposal of nuclear waste

This work focuses on diffusion and advection through cementitious media, the work arises from two research contracts undertaken at Loughborough University: Experiments to Demonstrate Chemical Containment funded by UK NDA and the SKIN project, funded by the European Atomic Energy Community's Seventh Framework Programme. Diffusion will be one of the most significant mechanisms controlling any radionuclide migration from a nuclear waste, deep geological disposal facility. Advection may also occur, particularly as the immediate post closure groundwater rebound and equilibration proceeds but is expected to be limited by effective siting and management during the operational phase of the facility. In this work advection is investigated at laboratory scale as a possible shorter timescale technique for providing insight into the much slower process of diffusion. Radial techniques for diffusion and advection have been developed and the developmental process is presented in some detail. Both techniques use a cylindrical sample geometry that allows the radionuclide of interest to be introduced into a core drilled through the centre of the test material. For diffusion the core is sealed and submerged in a container of receiving solution which is sampled and analysed as the radionuclide diffuses into it. For advection, a cell has been designed that allows inflow via the central core to pass through the sample in a radial manner and be collected as it exits from the outer surface. The radionuclide of interest can be injected directly into the central core without significant disturbance to the advective flow. Minor improvements continue to be made but both techniques have provided good quality, reproducible results. The majority of the work is concentrated on a potential cemetitious backfill known as NRVB (Nirex Reference Vault Backfill) this is a high porosity, high calcium carbonate content cementitious material. The radioisotopes used in this work are 3H (in tritiated water), 137Cs, 125I, 90Sr, 45Ca, 63Ni, 152Eu, 241Am along with U and Th salts. In addition the effect of cellulose degradation products (CDP) on radioisotope mobility was investigated by manufacturing solutions where paper tissues were degraded in water, at 80⁰C, in the absence of air and at high pH due to the presence of the components of NRVB. All diffusion experiments were carried out under a nitrogen atmosphere. All advection experiments were undertaken using an eluent reservoir pressurised with nitrogen where the system remained closed up to the point of final sample collection. Results for tritiated water and the monovalent ions of Cs and I were produced on a timescale of weeks to months for both diffusion and advection. The divalent ions of Sr, Ca and Ni produced results on a timescale of months to years. Variations of the experiments were undertaken using the CDP solutions. The effects of CDP were much more apparent at radiotracer concentration than the much higher radiotracer with non-active carrier, concentration. In the presence of CDP Cs, I and Ni were found to migrate more quickly; Sr and Ca were found to migrate more slowly. Additional Sr experiments were undertaken at elevated ionic strength to evaluate the effect of the higher dissolved solids content of the CDP solutions. Some of the results for HTO, Cs, I and Sr have been modelled using a simple numerical representation of the system in GoldSim to estimate effective diffusivity and partition coefficient. The diffusion model successfully produced outputs that were comparable to literature values. The advection model is not yet producing good matches with the observed data but it continues to be developed and more processes will be added as new results become available. Autoradiography has been used to visualise the radionuclide migration and several images are reproduced that show the fate of the radiotracers retained on the NRVB at the end of the experiments. As the experimental programme progressed it was clear that results could not be produced in a suitable timescale for Eu, Am U and Th. These experiments have been retained and will be monitored every six months until either diffusion is detected or the volume of receiving liquid is inadequate to ensure the NRVB is saturated.