Search

Disruptions are a serious problem in tokamaks, in which thermal and magnetic energy confinement is lost. This paper uses data from the DIII-D experiment, theory, and simulations to demonstrate that resistive wall tearing modes (RWTMs) produce the thermal quench (TQ) in a typical locked mode shot. Analysis of the linear RWTM dispersion relation shows the parameter dependence of the growth rate, particularly on the resistive wall time. Linear simulations of the locked mode equilibrium show that it is unstable with a resistive wall and stable with an ideally conducting wall. Nonlinear simulations demonstrate that the RWTM grows to sufficient amplitude to cause a complete thermal quench. The RWTM growth time is proportional to the thermal quench time. The nonlinearly saturated RWTM magnetic perturbation amplitude agrees with experimental measurements. The onset condition is that the q = 2 rational surface is sufficiently close to the resistive wall. Collectively, this identifies the RWTM as the cause of the TQ. In ITER, RWTMs will produce long TQ times compared to present-day experiments. ITER disruptions may be significantly more benign than previously predicted.

New high speed localized measurements of the pedestal magnetic field during the edge localized mode (ELM) cycle of a DIII-D High confinement mode (H-mode) discharge indicate a temporally and spatial complex redistribution of the edge current density profile, jedge. The measurement technique extracts the magnetic field magnitude, B, via the spectral separation of Stark-split neutral beam radiation in the pedestal. Single spatial channel measurements from a novel spatial heterodyne spectrometer are validated in discharges with core current profile changes. The technique measures Stark-splitting changes that imply B changes as small as 1 mT with high time resolution (50 μs). At normalized poloidal flux [Formula: see text], B appears saturated in the inter-ELM period and then rapidly decreases in edge, B increases at the ELM crash. The behavior is consistent with a rapid collapse of jedge at the ELM crash and subsequent pedestal recovery. In some discharges, at [Formula: see text], changes in B are observed throughout the ELM cycle. In others, B recovers and is relatively stable until a few ms leading up to the next crash. Measurements of B during the H-mode transition show a large increase at [Formula: see text] with little change at [Formula: see text], consistent with the formation of the edge bootstrap current density peak. The [Formula: see text] spectrum is complicated by predicted changes to the Stark component intensities with density at the L–H transition.

In this paper, we report the key role that toroidal rotation and the related E × B shear physics played in the very high energy confinement quality ( H98y2>1.5) of super H-mode experiments on DIII-D. Experiments show that the energy confinement quality decreases when toroidal rotation decreases due to the decreased externally controlled torque per particle. Meanwhile, the total pedestal pressure in the experiments remains very high during the rotation and confinement quality change. TGYRO transport modeling suggests the contribution from rotation in the E × B shear is responsible for the confinement quality in excess of standard H-mode ( H98y2∼1). CGYRO gyrokinetic simulations reveal the governing physics in the core plasma of super H-modes: significant up-shift of nonlinear the ITG critical gradient is observed when applying E × B shear physics in the modeling based on experimental data. The effects of other physical parameters and contribution from pedestal height, which may play minor roles in this study, are also discussed.

Experiments have been executed in the DIII-D tokamak to extend suppression of Edge Localized Modes (ELMs) with Resonant Magnetic Perturbations (RMPs) to ITER-relevant levels of beam torque. The results support the hypothesis for RMP ELM suppression based on transition from an ideal screened response to a tearing response at a resonant surface that prevents expansion of the pedestal to an unstable width [Snyder et al., Nucl. Fusion 51, 103016 (2011) and Wade et al., Nucl. Fusion 55, 023002 (2015)]. In ITER baseline plasmas with I/aB = 1.4 and pedestal ν * ∼ 0.15, ELMs are readily suppressed with co- I p neutral beam injection. However, reducing the beam torque from 5 Nm to ≤ 3.5 Nm results in loss of ELM suppression and a shift in the zero-crossing of the electron perpendicular rotation ω ⊥ e ∼ 0 deeper into the plasma. The change in radius of ω ⊥ e ∼ 0 is due primarily to changes to the electron diamagnetic rotation frequency ω e * . Linear plasma response modeling with the resistive MHD code m3d-c1 in...