Theory Weekly Highlights for September 2006

September 29, 2006

A third mechanism, in addition to the novel Kinetic Alfven Wave (KAW) and energetic particle redistribution mechanisms described earlier (see the October 28 2005 and February 10 2006 highlights at Theory Weekly Highlights for October 2005 and Theory Weekly Highlights for February 2006 respectively) is being considered as a candidate for providing the negative central current drive that maintains the axis safety factor 0 > 1 in DIII-D hybrid discharges. The new scheme also relies on the development of an m/n = 2/2 sideband parallel electric field excited by the rotating 3/2 island. The electric field sideband is excited by the 3/2 magnetic perturbations due to the magnetic curvature drift of the ions. This mechanism is a purely toroidal effect and contributes to the negative current drive over a larger range of q0 values than the KAW mode conversion mechanism.

September 22, 2006

Jeff Candy (GA) and Mark Fahey (ORNL-NCCS) are the co-PI's for a recently announced OASCR funded SciDAC Scientific Application Partnership (SAP) to develop a steady-state gyrokinetic transport (SSGKT) code for ITER scenario and performance studies. The SAP is attached to the larger SciDAC project led by John Cary (Tech-X Corp.) for integrated tokamak modeling (see The SSGKT code will rest on a novel computational method using a master code to run multiple copies of the turbulence code GYRO in parallel (one copy at each of 10-20 tokamak radii).

September 15, 2006

Modeling of the ORNL Mark IV massive gas injector (MGI) configuration for disruption mitigation using the 2D compressible fluid dynamic code FLUENT predicts rise times of 11 ms to reach the exit pressure and flow flattop from the time of the valve trigger for DIII-D, in agreement with measured exit pressure rise data. Calculations for long times > 10 ms agree with test stand measurements of the “apparent steady” exit flow rate of 3.28 x 1024/s for argon gas. All exit flow variables have a finite 'rise time' resulting from a transient period while the gas flow adjusts asymptotically to its “apparent steady” rate. We can now confidently predict how many particles are injected during the first 8 ms period over which thermal quench begins and ends. In a typical discharge, e.g. #122516, the “mixing efficiency” is 9.2%, which is the fraction of injected particles assimilated by the plasma due to large-scale MHD mixing processes. By providing meaningful interpretations of MGI experiments on DIII-D, these calculations are contributing to understanding of the unsteady gas flow in the jet delivery tube that is required in order to make projections for disruption mitigation on ITER.

September 08, 2006

A new refined calculation for fast ion stabilization of the internal kink using the Porcelli model in two DIII-D L-mode discharges with giant and small sawteeth yielded results more consistent with the observed discharge behavior. The refinement used a new more realistic formula for the ideal contribution to &delta-W obtained by the Lausanne group from fitting numerically calculated growth rates to equilibrium shape parameters. In the earlier analysis, the fast ion contribution was sufficient to stabilize the ideal internal kink for the discharge with giant sawteeth (see the July 28 highlight at Theory Weekly Highlights for July 2006). For the discharge exhibiting small sawteeth, however, the analytic Bussac model for the internal kink used in that analysis predicted stability even before the RF is on. With the new analysis, both discharges are predicted to be ideally unstable, consistent with both the ideal numerical stability calculations from GATO and the observed sawtoothing behavior of the discharges.

Alan Turnbull attended the Workshop on Theory of Fusion Plasmas in Varenna Italy from August 28 through September 1 and presented the sawtooth stabilization analysis.

Jon Kinsey and Jeff Candy attended the European TTF meeting in Marseilles.

September 01, 2006

A 5D δf Eulerian gyrokinetic code (EGK), which uses (R, μ, vpar) coordinates, has been developed. Here, μ is the magnetic moment and vpar the parallel velocity. The code was successfully benchmarked with the flux tube-based GS2 gyrokinetic code in the linear, collisionless, electrostatic limit, including trapped electron dynamics, for ITG/TEM physics and tests of the collisionless damping of the zonal flow potential. EGK is being used to study various numerical dissipation algorithms for the parallel velocity space. Upgrades to include nonlinear dynamics and a radial grid are in progress. As part of the Edge Simulation Laboratory project, EGK will be extended to a full-f formulation to study turbulence and transport in the edge/scrape-off region where order ρ*

These highlights are reports of research work in progress and are accordingly subject to change or modification