Development and validation of a CFD model predicting the backfill process of a nuclear waste gallery

Nuclear waste material may be stored in underground tunnels for long term storage. The example treated in this article is based on the current Belgian disposal concept for High-Level Waste (HLW), in which the nuclear waste material is packed in concrete shielded packages, called Supercontainers, which are inserted into these tunnels. After placement of the packages in the underground tunnels, the remaining voids between the packages and the tunnel lining is filled-up with a cement-based material called grout in order to encase the stored containers into the underground spacing. This encasement of the stored containers inside the tunnels is known as the backfill process.A good backfill process is necessary to stabilize the waste gallery against ground settlements. A numerical model to simulate the backfill process can help to improve and optimize the process by ensuring a homogeneous filling with no air voids and also optimization of the injection positions to achieve a homogeneous filling. The objective of the present work is to develop such a numerical code that can predict the backfill process well and validate the model against the available experiments and analytical solutions.In the present work the rheology of Grout is modelled as a Bingham fluid which is implemented in OpenFOAM - a finite volume-based open source computational fluid dynamics (CFD) tool box. Volume of fluid method (VOF) is used to track the interface between grout and air. The CFD model is validated and tested in three steps. First, the numerical implementation of the Bingham model is verified against an analytical solution for a channel flow. Second, the capability of the model for the prediction of the flow of grout is tested by means of a comparison of the simulations with experimental results from two standard flowability tests for concrete: the V-funnel flow time and slump flow tests. As a third step, the CFD model is compared with experiments in a transparent Plexiglas experimental test setup performed at Delft University of Technology, to test the model under more practical and realistic conditions. This experimental setup is a 1:12.5 scaled version of the setup of the full-scale mock-up test for backfilling of a waste gallery with emplaced canisters used in the European 6th framework project ESDRED (Bock et al., 2008). Furthermore, the plexiglas setup is used to study the influence of different backfill parameters.The CFD results for a channel flow shows good comparison against the analytical solution, demonstrating the correct implementation of the Bingham model in OpenFOAM. Also, the CFD results for the flowability tests show very good comparison with the experimental results, thereby ensuring a good prediction of the flow of grout. The simulations of the backfill process show good qualitative comparison with the plexiglas experiment. However, occurrence of segregation and also varying rheological properties of the grout in the plexiglas experiment results in significant differences between the simulation and the experiment.