Review on global migration, fuel retention and modelling after TEXTOR decommission

Abstract

ABSTRACT Before decommissioning of the TEXTOR tokamak in 2013, the machine was conditioned with a comprehensive migration experiment where MoF6 and 15N2 were injected on the very last operation day. Thereafter, all plasma-facing components (PFCs) were available for extensive studies of both local and global migration of impurities – Mo, W, Inconel alloy constituents, 15 N, F – and fuel retention studies. Measurements were performed on 140 limiter tiles out of 864 throughout the whole machine to map global transport. One fifth of the introduced molybdenum could be found. Wherever possible, the findings are compared to results obtained previously in other machines. This review incorporates both published and unpublished results from this TEXTOR study and combines findings with analytical methods as well as modelling results from two codes, ERO and ASCOT. The main findings are: • Both local and global molybdenum transport can be explained by toroidal plasma flow and E ⇀ × B ⇀ drift. The suggested transport scheme for molybdenum holds also for other analysed species, namely tungsten from previous experiments and medium-Z metals (Cr-Cu) introduced on various occasions. • Analytical interpretation of several deposition profile features is possible with basic geometrical and plasma physics considerations. These are deposition profiles on the collector probe, the lower part of the inner bumper limiter, the poloidal cross-section of the inner bumper limiter, and the poloidal limiter. • Any deposition pattern found in this TEXTOR study, including fuel retention, has neither poloidal nor toroidal symmetry, which is often assumed when determining deposition profiles on global scale. • Fuel retention is highly inhomogeneous due to local variation of plasma parameters – by auxiliary heating systems and impurity injection – and PFC temperature. • Local modelling with ERO yields good qualitative agreement but too high local deposition efficiency. • Global modelling with ASCOT shows that the radial electric field and source form have a high impact on global deposition patterns, while toroidal flow has little influence. Some of the experimental findings could be reproduced. Still, qualitative differences between simulated and experimental global deposition patterns remain. The review closes with lessons learnt during this extensive TEXTOR study which might be helpful for future scientific exploitation of other tokamaks to be decommissioned.