Advances in material phase change modelling approach for EU-DEMO limiter’s plasma-facing components

Abstract

Within the EU-DEMO first wall protection framework, work on limiter’s plasma-facing component design has started under plasma disruptive events (Richiusa et al., 2022). Starting from the rationale behind the TARTIFL&TTE software (Richiusa et al., 2022), this companion paper describes the progress on the engineering modelling of the plasma-facing material phase change under high heat flux, with the aid of COMSOL Multiphysics® software. The aim is to develop a reliable technique which can be used by designers for predicting how much of the solid armour undergoes phase change. This helps satisfy requirements for actively cooled components, such as at which armour depth safely locating the cooling system, and if its design can safely handle the heat transfer in the resulting component configuration after the disruption is extinguished. A few changes to the driving idea in Richiusa et al. (2022) will be also highlighted.The multiphysics software allows us to implement a 1D model which can be extended to 2D/3D geometries subjected to both uniform and non-uniform heat flux. It also offers the capability of conjugated heat transfer in solids and liquids by coupling the two different domains. Although this multiphysics approach is also investigated, no effort in melting layer motion modelling is done. Therefore, the equivalent way of validating this approach while reducing its computational time is working with one single solid domain, within which any liquid phase changes are tracked by an apparent heat capacity formulation. The vapour domain is not modelled. The material removal due to the evaporative mass flux is modelled by means of moving mesh frames which push the recessing liquid interface backwards according to gas kinetics-driven boundary conditions. The melt pool is not removed during the transient. Mass balance considerations drive the liquid-to-vapour interface velocity.The 3D Multiphisics implementation (3D-TARTIFL&TTE), is here supported by a benchmark activity and an application to the limiter’s plasma-facing armour, whose preferred chosen thickness is 20 mm.