Accumulation of heavy impurities in the tokamak core is detrimental for its performance and can lead to disruption of the plasma. In small to medium-sized tokamaks the effective neoclassical transport in the pedestal is typically oriented radially inward. In larger tokamaks—e.g., ITER—where the temperature gradient is higher and density gradients are lower due to the need to operate in a radiative divertor regime, the neoclassical transport is predicted to be outwards. The models are derived for axisymmetric quasi-steady-state plasmas. Applied 3D magnetic fields, i.e., Resonant Magnetic Perturbation (RMPs) as they are used to suppress Edge Localized Modes (ELMs), have experimentally been observed in AUG to enhance the outflow of heavy impurities in the pedestal. There is no model that can predict neoclassical heavy impurity transport in these ergodized 3D magnetic fields self-consistently. In this contribution, we present our kinetic tungsten transport simulation for an ASDEX Upgrade plasma with applied RMPs. Our model based on Hoelzl et al. [Nucl. Fusion 61, 065001 (2021)], van Vugt et al. [Phys. Plasmas 26, 042508 (2019)], and Korving et al. [Phys. Plasmas 30, 042509 (2023)] utilizes a full-orbit pusher, ionization, recombination, effective line, and continuum radiation and neoclassical collisions with the background plasma. The effective collisional radiative rates are from the OpenADAS database, the neoclassical collision operator uses the framework of Homma et al. [J. Comput. Phys. 250, 206–223 (2013)] and Homma et al. [Nucl. Fusion 56, 036009 (2016)]. We show that the adopted collision operator produces neoclassical transport within a satisfactory degree of accuracy. A sufficiently high RMP current causes an increase in tungsten diffusion in the pedestal by a factor of 2. We compare the average radial transport between axisymmetric and 3D RMP scenarios in the pedestal region. RMPs enhance the pedestal permeability for impurities, which results in enhanced transport. In addition to the enhanced transport, some of W is found to be trapped in 3D potential wells in the scrape-off layer. Due to the lack of suitable diagnostics for W in the pedestal, we investigate and suggest that argon can be an adequate substitute in experiments for model validation and further understanding impurity transport in scenarios with applied 3D magnetic fields. With the newly developed neutral model [Korving et al., Phys. Plasmas 30, 042509 (2023)], we can combine the interaction in the divertor with the 3D RMPs to model the tungsten transport from the divertor toward the core of the plasma.
Read full abstract