Speaker
Description
The ITER Research Plan re-baseline [1], especially the change of first wall (FW) material from beryllium to tungsten (W) has required the re-validation of the initial ramp-up phase in limiter configuration. The direct plasma contact on W surfaces, without the screening of eroded atoms provided by diverted operation, is expected to lead to high radiative ractions of about f rad = 70-80% as found in simulations for ITER with the SOLPS-ITER code [2, 3]. The W is generated by self-sputtering and self-regulated due to the high radiative power, leading to moderate last closed flux surface (LCFS) temperatures. These findings are in contrast with earlier W limiter start-upexperiments on ASDEX Upgrade (AUG) [4], where W sputtering was found to be mainly by lowand medium Z impurities, e.g. oxygen (O), with W self-sputtering only a minor contributor.
New W limiter experiments in support of the ITER re-baseline were performed in 2024-5 on EAST, AUG and WEST [2,5]. Here we report on further, complementary studies on AUG in which low density, strongly ECRH heated and well diagnosed plasmas were executed on W outboard limiters following a fresh boronization. This allows easier initial start-up, but, because the boron (B) layer erodes rapidly in the limiter contact area, is equivalent to limiting on a W surface. The primary aim was to access ITER-like conditions with high temperatures at the LCFS. In these plasmas the W source on the limiter is observed to increase even though the plasma density, the heating and also the low Z impurity sources and concentrations are rather constant. Radiation losses increase together with the W source and slowly saturate on timescales of a few 100 ms. This is an indication of a regime where the W self-sputtering dominates the radiation build up, a supposition supported by dedicated, time dependent SOLPS-ITER simulations. The latter, well constrained by the multiple experimental measurements, predict that these AUG plasmas should have a high W self-sputtering source and moreover find that the prompt re-deposition fraction is the principal mechanism allowing for good agreement between code and experiment regarding the 2D radiation distribution.
References
[1] A. Loarte et al., PPCF 67 (2025) 065023
[2] R.A. Pitts et al., NME 42 (2025), 101854
[3] Y. Zhang et al., NF 65 (2025) 056035
[4] A. Kallenbach, et al., NF 49 (2009) 045007
[5] J. Hobirk et al., 2025 paper presented at 34th IAEA FEC (Chengdu, China), EX-35811, https://conferences.iaea.org/event/392/papers/35811/files/13761-IAEAPaper25v3.pdf