Simulations of material damage to divertor and first wall armour under ITER transient loads by modelling and experiments

Operation of ITER at high fusion gain is assumed to be the H-mode. A characteristic feature of this regime is the transient energy release (TE) from the confined plasma onto plasma facing components (PFCs), which can play a determining role in lifetime of these components. The expected fluxes on the ITER PFCs during transients are: Type I ELM Q = 0.5 - 4 MJ/m2 in timescales t = 0.3 - 0.6 ms, and thermal quench Q = 2 - 13 MJ/m2 with t = 1 - 3 ms. CFC and tungsten macrobrush armour are foreseen as PFCs for ITER divertor and Be - as FW armour. During the intense TE in ITER the evaporation (CFC, W, Be) and surface melting and melt splashing (W and Be) are seen as the main mechanisms of PFC erosion. A noticeable erosion of CFC PAN fibres and rather intense crack formation for the W targets were observed in plasma gun experiments at rather small heat loads at which the melt damage to W armour is not substantial. The expected erosion of the ITER PFCs TE can be properly estimated by numerical simulations validated against erosion experiments at the plasma gun facilities QSPA-T, MK- 200UG and QSPA-Kh50. Within collaboration between EU fusion programme and Russian Federation, CFC and W macrobrush targets manufactured in EU were exposed to multiple ITER TE-like loads with Q = 0.5 - 2.2 MJ/m2 and t = 0 .5 ms at the QSPA-T. The measured erosion was used to validate the modelling codes developed in FZK (PEGASUS, MEMOS, and others), which are then applied to model the erosion of the divertor and main chamber ITER PFCs under expected transient loads in ITER. Numerical simulations performed for the expected ITER-like loads predicted: a significant erosion of the CFC target for Q > 0.5 MJ/m2 was caused by the inhomogeneous structure of the CFC; the W macrobrush structure is effective in preventing gross melt layer displacement. Optimization of macrobrush geometry to minimize melt splashing is done. Different mechanisms of melt splashing are compared with the results obtained in QSPA experiments. The crack formation at W surface was modelled using the code PEGASUS and validated against the experiments. Simulation of carbon dust production has been performed using the PEGASUS code and validated against MK-200UG experiments. Simulations carried out for Be armour demonstrated that the Lorentz force generates the violent melt motion thus becoming the most dangerous cause of melt splashing. (author)