Speaker
Description
The ITER divertor design has been guided by extensive scoping studies focused on baseline burning conditions at $Q_{DT} = 10$. They were conducted with the SOLPS-4.3 plasma boundary code without drifts and currents, assuming fuel injection from the top of the machine and pumping directly underneath the dome umbrella. The resulting simulation database was used to optimize the divertor operational window limited by target heat loads, fuel throughput, helium (He) exhaust and core-edge integration requirements. In 2015 the ITER Organization launched a new SOLPS-ITER version of the code which, in subsequent years, has been systematically upgraded, providing a much more complete edge plasma description in comparison with SOLPS-4.3. Together with switching to gas injection from the sub-divertor volume (now the scheme favoured at ITER), this resulted in a shift of the operational window to much greater divertor pressures ($p_n$) and higher average separatrix neon (Ne) concentrations ($c_{Ne}$), posing questions of compatibility with the throughput requirements and core plasma performance.
However, when also including more complex sub-divertor geometry (accounting for realistic neutral bypass conductances around the divertor cassette body), and a proper description of pumping system capabilities [1], the results again change radically. We present here a completely new SOLPS-ITER $Q_{DT} = 10$ simulation database in which, rather unexpectedly at first sight, the operational window reverts to something rather close to the initial SOLPS-4.3 database. This is identified as due primarily to two factors: 1) a strong influx of neutrals from the sub-divertor space to the far-SOL reducing the peak target heat flux and decreasing the value of $p_n$ required to meet the $10\ MW/m^2$ constraint; 2) the proper assessment of pressure at the pumping duct translating to a wider range of pn compatible with throughput requirements.
Accounting for the neutral bypasses also leads to a reduction of the plasma temperature in the divertor baffle regions. This is extremely beneficial given the results of simulations with the new wide-grid SOLPS-ITER capability, also presented here, which show these locations to be the origin of significant quantities of sputtered W impurity which is rather poorly screened compared with that released from vertical target areas. Finally, taking into account the improved divertor description and the new code capabilities, the important question for reactor divertor designs beyond ITER of the impact of the dome umbrella on the global divertor performance has also been explored.
[1] N. Vasileiadis et al., Fusion Eng. Des. 151 (2020) 111383