Development of three simulation tools for open-surface liquid metal magnetohydrodynamic flows in plasma-facing components using OpenFOAM

Abstract

The use of open-surface liquid metal (LM) flows in plasma-facing components (PFC), namely the divertor and the first wall (FW) in large tokamak nuclear fusion reactors, promises several advantages compared to solid designs, such as reduced erosion and impurity retention. The current understanding of the flow dynamics governing open-surface liquid metals for PFC applications is incomplete. In this study, we develop and test three computational models to simulate open-surface magnetohydrodynamic (MHD) flows based on the OpenFOAM computational fluid dynamics (CFD) platform. The first model is 3D and is based on the volume of fluid (VoF) method for the interface capturing and on the electric potential formulation for the MHD effects. The second model is 2D. It also uses the VOF method but relies on the integration of the MHD effect in the plane perpendicular to the magnetic field; it allows for reducing one dimension in the direction of the magnetic field. The third model is 1D, it captures the interface using a height function, which simplifies the dimension perpendicular to the substrate, and also leverages the integration of the MHD effects in the plane perpendicular to the magnetic field. Our models couple several physical phenomena and involve special numerical techniques. The three models are validated against existing experimental results for LM flows in a horizontal non-conducting chute under a co-planar magnetic field. Further simulations of the same chute with electrically conducting walls are also presented in this work, showing the capability of the 3D model to couple different regions. While the 3D model identified well the complex flow behavior and the increase in the MHD drag, the simplified 1D and 2D models failed at providing accurate estimates of flow velocity and film height. The increased drag results in the formation of an MHD-induced hydraulic jump after certain distance from the inlet nozzle. The development of these models is a necessary step towards a better understanding of LM flows, which enables a successful design of technologies capable of protecting the PFC of a fusion reactor.