"Plasma Surface Interaction Issues of an All-Metal ITER"

We assess key plasma surface interaction issues of an all-metal plasma facing component (PFC) system for ITER, in particular a tungsten divertor surface, and a beryllium or tungsten first wall. Such a system eliminates problems with carbon divertor erosion and T/C codeposition, and for an all-tungsten system would better extrapolate to post-ITER devices. The issues studied are sputtering, transport, and formation of mixed surface layers, tritium codeposition, core plasma contamination, ELM response, and He on W irradiation effects. Code package OMEGA computes PFC sputtering erosion/redeposition in an ITER full power D-T plasma edge regime with convective transport. The HEIGHTS package analyzes divertor plasma transient response. PISCES and other data are used with code results to assess PFC performance. Predicted outer wall sputter erosion rates are acceptable for Be (0.3 nm/s) or bare (stainless steel/Fe) wall (0.05 nm/s) for the low duty factor ITER, and are very low (0.002 nm/s) for W. Most wall-sputtered Be is redeposited on the wall itself or baffle region, with about 10% transported to the divertor target. T/Be codeposition in redeposited wall material could be significant (~2 gT per 400 s ITER pulse). Core plasma contamination potential from wall sputtering appears acceptable for Be (~2%), and negligible for W (or Fe) due to near-surface ionization of sputtered W (Fe) atoms and subsequent strong redeposition. A tungsten divertor likewise appears acceptable from the self-sputtering and plasma contamination standpoints, and would have negligible T/W codeposition. Be can grow on/near the strike point region of a W divertor, but for the predicted maximum surface temperature of ~800°C, deleterious Be/W alloy formation may be avoided. ELM’s are a serious challenge to the divertor, but this is true for all materials. We identify acceptable ELM parameters for W. We conclude that an all-metal PFC system is likely a much better choice for ITER D-T operation than a system using carbon, but critical R&D issues remain, e.g., in areas of transient surface erosion (of all materials), W surface integrity with energetic He etc. bombardment, and in predictive plasma/surface interaction modeling generally. Steps are suggested to ameliorate problems and reduce uncertainties, e.g., via a 300 or 400°C baking capability for T/Be reduction, and using a deposited tungsten first wall test section.