Search

As part of a scoping study for an ITER‐sized tokamak, more than 50,000 simulations have been performed of the edge plasma with SOLPS5.0‐B2 using aggressive charge state bundling, fluid neutrals, and constant‐in‐time boundary conditions. These simulations have 40, 80, 100, 125, and 250 MW crossing into the simulation domain from the core region, and a range of D/T (deuterium/tritium) and impurity gas puffs giving a variation of electron density and Zeff. Most of the simulations are steady‐state, but about 10% show oscillations where the range of peak power flux densities to the outer target exceed 1% of the average value, and about 1% where the normalized range exceeds 10%. These oscillating cases present a challenge in determining whether a particular case has converged, or needs to be continued. The oscillations seem to have a physical origin because: the frequency of oscillations changed by less than a factor of two despite the time‐step for one case being varied by a factor 100; doubling the grid‐resolution resulted in similar oscillations; at least in some cases, a physically plausible limit‐cycle is present. [ABSTRACT FROM AUTHOR]

Heat exhaust is a challenge for ITER and becomes even more of an issue for devices beyond ITER. The main reason for this is that the power produced in the core scales as R3 while relying on standard exhaust physics results in the heat exhaust scaling as R1 (R is the major radius). ITER has used SOLPS (B2-EIRENE) to design the ITER divertor, as well as to provide a database that supports the calculations of the ITER operational parameter space. The typical run time for such SOLPS runs is of the order 3 months (for D+C+He using EIRENE to treat the neutrals kinetically with an extensive choice of atomic and molecular physics). Future devices will be expected to radiate much of the power before it crosses the separatrix, and this requires treating extrinsic impurities such as Ne, Ar, Kr and Xe - the large number of charge states puts additional pressure on SOLPS, further slowing down the code. For design work of future machines, fast models have been implemented in system codes but these are usually unavoidably restricted in the included physics. As a bridge between system studies and detailed SOLPS runs, SOLPS offers a number of possibilities to speed up the code considerably at the cost of reducing the fidelity of the physics. By employing a fluid neutral model, aggressive bundling of the charge state of impurities, and reducing the size of the grids used, the run time for one second of physics time (which is often enough for the divertor to come to a steady state) can be reduced to approximately one day. This work looks at the impact of these trade-offs in the physics by comparing key parameters for different simulation assumptions. (© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim) [ABSTRACT FROM AUTHOR]

A density scan is performed in SOLPS5.0 for two divertor configurations of MAST Upgrade: (i) a short divertor silimiar to configurations in present-day tokamaks, (ii) the Super-X divertor. In the simulation, a clear roll-over of the ion flux and plasma density at the target is observed as the plasma detaches. The separatrix density at which the transition occurs is estimated. In addition, we investigate the ability of the long-legged divertor to reduce the target power loads and enhance the radiated power at higher collisionalities. We also study how distribution of power losses in the divertor region changes with the modification of geometry and we separately analyze neutral radiation and carbon ion line radiation. (© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim) [ABSTRACT FROM AUTHOR]