The effects of an open and closed divertor on particle exhaust during edge-localized mode suppression by resonant magnetic perturbations in DIII-D

Abstract

This paper compares the effects of divertor geometry on particle exhaust characteristics during the suppression of ELM using resonant magnetic perturbations (RMPs) on DIII-D. The subject is timely, particularly for ITER, because the combination of techniques to control or mitigate ELMs and control particle exhaust can provide confidence in the ability of an external pumping system to fully remove the particle exhaust. The differences between an open and closed divertor magnetic topology show a strong coupling of the perturbed strikepoint to the pumping manifold in closed divertor configurations, which can increase the particle exhaust by a factor of four. There is also an observed dependence on q 95 in this configuration, which is a common feature of RMP ELM suppression. Neutral density in both the active and non-active divertors is seen to increase during the RMP in the ISS configuration, and edge plasma conditions (i.e. n e,sep and midplane profile of D α) are seen to increase in the closed divertor configuration. Finally, the pumping exhaust is also shown to have a strong dependence on local measurements of the recycling flux. These observations, when taken as a whole, point to a substantial change in the plasma edge conditions, i.e. near the LCFS, throughout the poloidal cross-section of the vacuum vessel. This is coincident with the application of the RMP affecting the pumping capability of the system.

On gas flow effects in 3D edge transport simulations for DIII-D plasmas with resonant magnetic perturbations

“…Unfortunately, the total recycling flux is not measured in the experiment, but e.g. in [28] it is estimated from the core density, the confinement time and some generic "recycling coefficient". The two former quantities are related to the upstream particle flux, while the latter essentially gives an amplification factor for the divertor recycling …”

“…Experimental observations are in the range of Γ pump /Γ rec = 2 − 9 % depending on the plasma shape and edge safety factor [28]. If we include the experimentally observed wall pumping of 2 − 3 % (no explicitly taken into account in the simulations) in an effective pumping loss fraction of Γ pump /Γ rec ≈ 5 − 6 % for an ISS plasma at q 95 = 3.6, then figure 5 allows to estimate ε pump ≈ 0.4 − 0.5.…”

Decoupled recovery of energy and momentum with correction ofn = 2 error fields

“…Like n = 1, n = 2 fields can cause significant rotation braking and drive tearing instabilities [14,15]. However unlike n = 1, n = 2 fields also significantly enhance particle transport (termed density pumpout) [16,17] and degrade energy confinement [18]. Plasmas also operate farther from n = 2 stability limits and so field amplification is weaker for n = 2 than n = 1 [11,19].…”

ELM suppression in helium plasmas with 3D magnetic fields

“…This high-power phase was then followed by a low-power, low toroidal rotation phase, matching ITER's non-active 4 He plasma conditions. In these discharges, 11 Torr-liters of 4 He was injected during the first 1100 ms and at 1200 ms, approximately 100 ms after reaching the I p plateau, P NBI was increased from 3.7 MW to 8.4 MW resulting in the addition of 36 Torr-liters of 2 H fueling over the 2600 ms duration of the 2 H-NBI heating phase as determined from global particle balance calculations [17]. In addition, at 1100 ms ECRH was added with P ECRH = 2.9 MW resulting in P TOT = 11.6 MW, including P OH = 0.3 MW.…”