Divertor tungsten tile melting and its effect on core plasma performance

For the 2007 and 2008 run campaigns, Alcator C-Mod operated with a full toroidal row of tungsten tiles in the high heat flux region of the outer divertor; tungsten levels in the core plasma were below measurement limits. An accidental creation of a tungsten leading edge in the 2009 campaign led to this study of a melting tungsten source: H-mode operation with strike point in the region of the melting tile was immediately impossible due to some fraction of tungsten droplets reaching the main plasma. Approximately 15 g of tungsten was lost from the tile over ∼100 discharges. Less than 1% of the evaporated tungsten was found re-deposited on surfaces, the rest is assumed to have become dust. The strong discharge variability of the tungsten reaching the core implies that the melt layer topology is always varying. There is no evidence of healing of the surface with repeated melting. Forces on the melted tungsten tend to lead to prominences that extend further into the plasma. A discussion of the implications of melting a divertor tungsten monoblock on the ITER plasma is presented.