A kinetic transport code for energetic particles (EPs) has been developed in collaboration with a Peking University graduate student He Sheng. The EPtran code evolves the EP distribution function in radial, energy, and pitch angle space to steady state with classical slowing down, pitch angle scattering, as well as radial transport of the injected (NBI or fusion alpha) EPs from high-n ITG/TEM microtubulence and EP driven, low-n Alfven eigenmodes (AEs). The EP-AE transport is treated with a critical EP density (or equivalent pressure) gradient model. EPtran simulations have been compared to simulations with the ALPHA EP density transport code, which assumes the EP velocity space distribution is not significantly distorted from the classical slowing down distribution. Results of the recent ALPHA code DIII-D NBI transport validation study (reported in the April 24 2015 Highlight at Theory Weekly Highlights for April 2015) have been recovered by the EPtran code. The simulated and experimental EP pressure profile match. About half the birth EPs are lost from the mid-core. The FIDA data suggests the hotter EP’s have stronger AE critical gradient transport. The EPtran simulations have helped to understand why, because, unlike the ALPHA code, the EPtran kinetic code can distinguish between EP particle and energy transport.
The between-shot EFIT01 equilibrium reconstructions have now been set to use the “jta_f” rather than “jt” as the default SNAP file for between-shot EFIT analysis. While both SNAP representations allow the reconstructed equilibrium to have finite edge current density, the “jta_f“ representation is designed to have one additional degree of freedom with a fitting parameter to move vertically so that the fitting is less susceptible to vertical oscillation issues. “jta_f” was previously used by RT-EFIT in PCS for plasma control, but the new “jta_f” SNAP file analysis has been tested for between-shot DIII-D plasma operation, with all results generally well converged and with good magnetic chi squared.
The US-Japan Joint Institute for Fusion Theory (JIFT) Workshop was held at General Atomics from July 22–24. The topic of this year’s workshop was “Theory and Simulation of 3D Physics In Toroidal Plasmas - Comparison to Experiments.” Approximately 25 US and Japanese scientists attended the workshop and presented advances in the development, application, and validation of models of the effects of 3D fields in tokamaks, RFPs, stellarators, and hybrid devices. The agenda of the workshop is posted at https://fusion.gat.com/conferences/usjapan2015/agenda.html . Presentations will be posted here upon approval.
Nonlinear GYRO simulations of a beam-heated DIII-D discharge have identified a small suppressive effect on energetic particle (EP) transport caused by plasma rotation shear. EP transport by EP-driven Alfvén eigenmodes (AEs) appears to be characterized by the onset of “stiff” (clamped critical gradient) transport above a critical driving EP density gradient. The transition to stiff transport depends on interaction between the EP drive and the suppression of zonal flows driven primarily by ITG/TEM microturbulence. The present results represent the first systematic study of the effect of equilibrium flow shear in this process. Flow shear generally suppresses both ITG and AE growth, affecting both sides of the predator-prey system that governs the onset of stiff EP transport. At the moderately low level of flow shear, saturated microturbulent transport appears unaffected but the AE stiff transport critical gradient is pushed about 15% higher at the location of greatest AE instability compared to the same case with no rotation. A signature of this effect in the linear spectrum has yet to be identified. The possibility that stronger shear compared to ITG drive could have the opposite effect by suppressing microturbulence (and thus suppressive ITG-driven zonal flows) is still under investigation.
In order to better understand ELM suppression experiments, gyrokinetic simulations with the GENE code have been performed to examine turbulent transport in the presence of flute-like perturbations of the flux surfaces from non-axisymmetric fields. This revealed some surprising new insights into the basic character of the linear ITG mode, namely that the slab ITG (Landau resonance driven), can play an important role alongside the toroidal ITG (curvature drift resonance driven), which is usually thought of as dominant in tokamaks. Across all scales, the linear ITG obeys a critical balance relationship, where modes at long perpendicular scales have extended parallel mode structure. At long perpendicular wavelengths (< ~0.2 x the ion gyroradius), the ITG mode is almost completely insensitive to changes in the magnetic geometry, indicating a slab-like instability mechanism. At shorter perpendicular wavelengths (>~0.2 x the ion gyroradius) the mode responds to modulation of the magnetic curvature, indicating a toroidal instability. However, even the shorter wavelength “toroidal-like” mode is still quite sensitive to the Landau resonance - mode localization in the parallel direction can be either catalytic or catastrophic for the instability, depending on the degree of localization.
A new version of the GATO code is ready for public release. In addition to including the option to read equilibria from the CHEASE code (see Highlight from February 14, 2014 at Theory Weekly Highlights for February 2014), this version is more robust, includes new diagnostics on the input current density, and fixes for several minor issues. The latter included an issue in the reading of CORSICA i-file equilibria that appears to have only minor consequences but induced a failure when the additional current density diagnostic was added. The new version has additional diagnostics used to check alternative calculations of the coordinate system Jacobian and non-orthogonality, and a calculation of the field lines in the poloidal-toroidal plane, useful for diagnosing the magnetic field line pitch as stability changes. Also included is an option to deform an arbitrary input wall smoothly onto the plasma boundary when the wall is contracted. This option results in a more systematic wall scan, though at the expense of inducing a slight kink in the growth rate curve at unit scale factor. The new version will be released publicly once checkout has been completed.
These highlights are reports of research work in progress and are accordingly subject to change or modification