Description
Edge plasma transport in tokamaks is intrinsically three-dimensional in the presence of non-axisymmetric magnetic perturbations such as ripple. This work presents the first fully 3D flux-surface-aligned geometry implemented in SOLEDGE3X, enabling the self-consistent treatment of non-axisymmetric flux surfaces.
SOLEDGE3X is a drift-reduced fluid code for edge plasma transport and turbulence, capable of operating either in a 2D mean-field transport mode or in a first-principles 3D turbulent regime. It is widely used to study impurity transport, heat exhaust and divertor physics. While three-dimensional wall geometries have been previously considered, its application has so far been restricted to axisymmetric meshes. Previous attempts to treat non-axisymmetric magnetic effects relied on a perturbative approach that retained an axisymmetric mesh, with only limited validity when the perturbation amplitude approaches the magnitude of the poloidal magnetic field.
To overcome this restriction, fully non-axisymmetric geometries are constructed from a given 2D equilibrium using two complementary methods: a three-dimensional reconstruction of the poloidal flux that preserves iso-ψ surfaces, and a field-line tracing approach to recover flux-surface-aligned coordinates. Both approaches produce comparable 3D meshes, whose quality is assessed through the divergence of the magnetic field and the radial contravariant magnetic component. Both metrics decrease toward zero as the toroidal resolution increases or the perturbation amplitude decreases.
These developments are applied to simulations of magnetic ripple in WEST, including a core energy source and a recycling fluid neutral model. First simulations indicate a toroidal modulation of parallel heat and particle fluxes due to flux expansion. Radial profiles shift near the separatrix, with much larger deviations in the far scrape-off layer. At the divertor targets, ripple-induced variations of the incidence angle lead to modified recycling patterns and localised heat flux peaks on the wall, which are also observed experimentally. These results demonstrate that a fully 3D flux-surface-aligned treatment is required for consistent modelling of edge transport and plasma–wall interaction in realistic non-axisymmetric tokamak configurations.