Speaker
Description
Glow-discharge boronisation (GDB) is a standard wall-conditioning technique in fusion devices to ensure reliable plasma operation. With ITER’s re-baseline decision to adopt tungsten (W) as the main-chamber material, GDB is foreseen as a key method to achieve favourable plasma–wall conditions in its full-W environment. To enable reliable extrapolation to ITER, dedicated studies on existing full-W devices are required. Since ITER’ first operations campaign will employ a partial-anode configuration, investigations at ASDEX Upgrade (AUG) and WEST have examined the uniformity and properties of boron (B) layers from partial versus standard GDBs.
Plasma restart on a pristine, unboronised W wall proved difficult in both AUG and WEST. In contrast, sustained plasma discharges were achieved immediately after non-uniform GDBs employing only part of the available hardware—two of four anodes in AUG and three of six gas inlets in WEST—showing that even partial B coverage ensures reliable start-up.
To validate predictive simulations of GDB performance for ITER, quartz microbalances (QMBs) were installed at seven AUG locations to monitor in-situ layer growth, while witness samples on manipulators in AUG and WEST enabled post-mortem ion-beam analysis of layer thickness, composition, and density.
The B layer thickness, oxygen (O) content, and retained deuterium (D) fraction—typically 5–50% relative to B—depended strongly on substrate material and on activation of nearby glow anodes (AUG) or gas inlets (WEST). In WEST, samples near active inlets showed higher B thickness and D fraction but lower O content. In contrast, in AUG, higher B deposition occurred when adjacent anodes were inactive, indicating local erosion during GDB may play a role.
Both the in-situ QMB measurements in AUG and the sample analyses from the two manipulators in WEST revealed a considerably smaller toroidal variation of the B deposition than predicted by modelling, likely explaining the successful restart even after a non-uniform GDB. This discrepancy may be attributed to the sticking probability of the B-containing precursor molecules, measured to be approximately 0.3 at AUG, whereas the initial modelling made no assumption on sticking.
Apart from the witness samples exposed during GDBs, AUG wall samples installed for the entire campaign were analysed to assess long-term B migration during plasma operation and the additional tritium inventory that could be retained in ITER as a result of successive GDBs.
The results indicate that ITER’s planned GDB system should ensure reliable plasma start-up. Future work will address the persistence of GDB effects in more detail.