Physics Analysis of Diver tor Modifications in ITER

In the course of the ITER design review it was found that the divertor dome as originally designed imposes severe requirements on the accuracy of the plasma position control system. Taking into account previous modelling results (1), an alternative design with a somewhat smaller dome shifted away from the separatrix, (F53, Fig. 1), which tolerates larger excursions of the separatrix branches, was proposed, and a series of B2-Eirene (SOLPS4.3) code runs was performed to determine the implications of these modifications for the divertor performance. Moreover, further analysis of the ITER magnetic system and current experimental data revealed a concern that the reference x-point position might not be attainable if the plasma current profile were flatter, li B 0.65, than the reference case (0.85) and that then the separatrix would strike the dome. Accordingly, a dedicated design effort developed a divertor configuration accommodating a wider range of magnetic equilibria, achieved by a further reduction of the dome and an outward shift of the inner target. The inner divertor is then somewhat shorter and the wetted area there smaller, which could in turn have a detrimental effect on the power loading (particularly with impurity seeding, see below). In order to explore this effect quickly, we selected a rather extreme design variant where the inner target was displaced outwards by as much as 20 cm, (F55, Fig. 1), and examined two configurations: the reference equilibrium with li = 0.85 and a flat-current equilibrium with li = 0.63. The modelling parameters were the same as in (1). Fluid equations are solved for the transport of ions and electrons (B2), and the neutral transport part (Eirene) employs a non-linear Monte-Carlo modelling taking into account the neutral-neutral collisions together with elastic collisions between the neutrals and ions and the most essential molecular interactions with ions. The Monte-Carlo algorithm now allows multi-processor parallel execution in the MPI environment. The plasma consists of D (representing both D and T), He, and C ions and atoms and D2 molecules. Molecular fuel is puffed in at the top and pumped out from the bottom with a moderate outflow of D ions from the core to simulate the core fuelling. The helium ion source from the core is proportional to the fusion power, the targets are carbon and the walls are assumed to be carbon-covered, and carbon is released from the surfaces by physical and chemical sputtering.