UNDERSTANDING TUNGSTEN ACCUMULATION DURING ICRH OPERATION ON WEST

International audience; y The transport of impurities in a tokamak with metallic walls can lead to accumulation events and radiative collapses, and this is particularly true when neoclassical transport is enhanced by sources of impurity poloidal asymmetry [1, 2]. On WEST, where the RF heating scheme (torque-free without core particle sources) is relevant for ITER operation, these events are rare, but still radiative collapses are sometimes observed [3, 4]. The physics of collisional tungsten transport in this case involves 3 main channels: i) a direct channel by which the hydrogen temperature anisotropy driven by ICRH drives a poloidal asymmetry of the electrostatic potential, ii) a non-linear channel by which the tungsten accumulation mitigates the ICRH drive by increasing the parallel temperature of the fast ions and reducing the temperature anisotropy in a self-limited process, and iii) an indirect channel related with the ICRH induced toroidal rotation. The role of Finite Orbit Width (FOW) effects is found to be instrumental : it strongly reduces the hydrogen temperature anisotropy, the favorable Hydrogen Temperature Screening (HTS) effect, and the electron heat source. The tungsten peaking as a function of toroidal rotation and ICRH power shows 2 possible branches consistent with bolometry inversion, one with a peaking driven by ICRH and one with a peaking driven by rotation. The condition for a radiative collapse in the core brings an additional constraint, and indicates that the electron heat source is probably much lower than computed, suggesting important losses of fast ions in the ripple. The radiative collapse would then be due to a limited electron heat source and a tungsten peaking driven by the induced rotation. This peaking is initially moderate, and it is reinforced during the collapse as the temperature screening effect reverses