In collaboration with Peking University graduate student He Sheng, progress was made in verifying low-n energetic particle driven Alfven eigenmode (AE) TGLF linear growth and frequency rates against published GYRO rates for the fusion alpha particle GA standard case. This project is part of Ph.D. thesis work by Sheng, who is visiting GA for a year. By using a high resolution TGLF wavefunction, the growth rates for both the TAE and EPM modes were found to be in good agreement. Verifying the linear growth rates in the recent GYRO DIII-D AE-EP critical gradient validation study, which again requires wavefunction optimization, show that the “universal” TGLF equations normally used for the high-n ITG/TEM modes, will be sufficiently accurate.
NIMROD has been used to validate a novel two-staged approach to disruption mitigation using an impurity shell-pellet in the first stage to deposit a small quantity of impurities preferentially into the central plasma region. The impurities can then permit a (mostly) radiative thermal collapse while minimizing edge MHD activity and thermal loads on the divertor. In a DIII-D simulation with 1.36 mmoles of Ar deposited in the core, the outermost flux surfaces were indeed found to remain intact until the thermal quench was nearly complete, producing a much higher ratio of radiated-to-conductive heat loss than that of massive gas injection. Fast core cooling generates a concentric skin current ring around the magnetic axis and a current bump in response to the lowered inductance. The relaxation of this current density profile leads to an initially fast current drop followed by a slower phase. Such an unusual current quench (CQ) may lead to surprises for runaway electron (RE) amplification, and RE suppression in the CQ using the second stage: Massive Particle Injection (MPI) with Deuterium gas or shattered pellets. Although the DIII-D simulation uses an idealized impurity deposition model, investigations are underway to develop an understanding of the basic processes involved in more realistic integrated modeling scenarios for ITER using the two-staged shell pellet-MPI approach.
M.R. Cianciosa visited GA from ORNL to work with A. Wingen, R.S. Wilcox, and E.A. Unterberg of ORNL and GA Theorists, on reconstruction of a 3D helical equilibrium. This is the first time a 3D reconstruction of a tokamak discharge has been done. The reconstruction utilized a newly developed parallel implementation of the VMEC code coupled to the V3FIT 3D equilibrium reconstruction code. Utilizing supercomputing resources, 3D equilibrium reconstruction can now be done at the resolution necessary to resolve 3D effects in tokamaks. Using SXR and MSE diagnostics, an equilibrium showing a 5 cm n = 1 helical displacement of the magnetic axis (helical core or “snake”) was reconstructed for a high performance hybrid mode discharge (shot #164661) from a recent flux pumping experiment on DIII-D. Modeled signals of the rotating helical core show good agreement with the measured SXR and MSE signals. The helical core brings the central q profile closer to 1 in comparison to a 2D reconstruction, potentially allowing for redistribution of poloidal flux (i.e., flux pumping) and elimination of sawtooth instabilities. Besides ongoing studies in these applications, the stability of the helical core, which can be considered a saturated internal kink instability, will be investigated with respect to operating parameters such as off-axis current drive, core pressure and plasma rotation.
The new gyrokinetic solver CGYRO has been used to study the complex spectrum of modes in the pedestal region. Analysis was done for a typical DIII-D H-mode, including kinetic ions and electrons and Miller shaped flux surface geometry. Even though beta is small in the pedestal (~0.1%), electromagnetic effects, including both transverse and compressional magnetic perturbations, were found to be important. A transition from a dominant mode with KBM-like symmetry to a mode with ITG-like symmetry occurs near the top of the pedestal. A scan over beta found that, at the experimental normalized pressure parameter value, alpha, the mode appears to be a KBM-ITG hybrid. Despite the large driving ion temperature gradient, the low-k modes are surprisingly stabilized within the pedestal, due to the large effective alpha value. At high k_perp, the ETG modes are also stabilized in the pedestal, despite the large driving electron temperature gradient. However, since the ETG modes are not strongly affected by electromagnetic effects, this was found to be due to the large electron density gradient. We plan to use CGYRO assessment of the onset of the linear KBM with full collisional effects to develop an improved KBM/RBM model for the EPED pedestal structure model.
These highlights are reports of research work in progress and are accordingly subject to change or modification