First application of the massively-parallel Monte Carlo code ERO2.0 for plasma-wall interaction and 3D local impurity transport at JET ILW

摘要

Introduction Estimating erosion for plasma-facing components (PFCs) is one of the key issues for ITER. Effective sputter yields can be obtained experimentally e.g. by estimating flux ratios with S/XB ratios [1]. The interpretation of such experiments and extrapolation to ITER conditions is not straightforward, because the effective yields result from a complex interplay of plasma conditions, wall geometry and impurity transport. This makes modelling tools like the Monte-Carlo code ERO necessary. However, ERO was originally designed for simulation volumes of~(10 cm)~3, typically covering only a few adjacent wall tiles. This limitation is overcome by the new version ERO2.0. With a flexible 3D representation of wall geometries and plasma parameters, as well as increased performance due to massive parallelisation, ERO2.0 can simulate larger volumes with more PFC components. In this contribution, we re-visit recent ERO modelling from [1] for Beryllium (Be) erosion of the JET Inner-Wall Guard Limiter IWGL in octant 7X, tiles 6-8. The new code version allows the following improvements: 1) increased simulation volume in toroidal direction, 2) consideration of tiles from the neighboring IWGL limiters as particle sources, and 3) a more detailed model for magnetic shadowing of the wall. We focus on the effect of these improvements on Be self-sputtering.