Brendan Lyons visited the Princeton Plasma Physics Laboratory for the week of July 16th to attend the 6th Annual Theory and Simulation of Disruptions Workshop. He gave a contributed oral presentation entitled “Overview of Theory and Simulation of Disruption Mitigation Research at General Atomics” on behalf of the Theory and Computational Science Group. Progress on coupling M3D-C1 to KPRAD for impurity-injection modeling was presented along with an axisymmetric benchmark of thermal-quench dynamics between M3D-C1 and NIMROD. In addition, results from recent 3-D NIMROD modeling of shattered-pellet injection on ITER and an analytic assessment of the damping of fast waves in post-disruptive plasmas were shown. After the workshop, Dr. Lyons spent several days collaborating with Stephen Jardin and Nathaniel Ferraro of PPPL to improve pellet and impurity modeling in M3D-C1, and with Roman Samulyak of Stony Brook University on coupling a Lagrangian particle code for pellet ablation to M3D-C1.

Analysis of JET-like H-mode hybrid scenario experimental parameters with an ITER-like wall has been performed with NEO and CGYRO to understand the mechanisms leading to observed tungsten accumulation. The tungsten neoclassical pinch dominates in the core for r/a < 0.3, as it is significantly enhanced, by about an order of magnitude, due to sonic toroidal rotation effects. In the outer core, the gyrokinetic ITG drive is also strongly affected by the rotation. At small collisionality, the tungsten nonlinear turbulent particle flux is radially inward and dominates over the neoclassical transport. As the collision frequency increases, the turbulent flux reverses to a more favorable outward flow of particles away from the core. But the neoclassical transport becomes competitive with the turbulence, such that with moderate collisionality, a net strong inward pinch returns, indicating possible detrimental core accumulation. Ultimately, for the modeling of tungsten, both the neoclassical and turbulent transport must be considered.

The influence of energetic particles and plasma resistivity on the n = 1 internal kink and fishbone modes in tokamak plasmas was numerically investigated, using the full toroidal, resistive MHD-kinetic hybrid stability code MARS-K. The results show that energetic particles can either stabilize or destabilize the ideal internal kink mode, depending on the radial profiles of the particles’ density and pressure. Resistive fishbones with and without an ideal wall are investigated. It is found that, in the presence of energetic particles as well as plasma resistivity, two branches of unstable roots exist, for a plasma which is ideally stable to the internal kink instability. One is the resistive internal kink mode. The other is the resistive fishbone mode. These two-branch solutions show similar behaviors, independent of whether the initial ideal kink stability is due to an ideal wall stabilization for high-beta plasmas, or due to a stable equilibrium below the Bussac pressure limit. For a realistic toroidal plasma, the resistive internal kink is the dominant instability, which grows much faster than the resistive fishbone. The plasma resistivity destabilizes the resistive internal kink while stabilizes the resistive fishbone. Systematic comparison with an analytic model qualitatively confirms the MARS-K results. The results are in agreement with analytic theories by White et al. and Bussac et al. Compared to analytic models based on the perturbative approach, MARS-K offers an improved physics model via self-consistent treatment of coupling between the fluid and kinetic effects due to energetic particles.

An effort is underway to make various OMFIT modules and classes COCOS compliant. Different codes and machines in the tokamak physics community use a variety of coordinate conventions, complicating code-coupling in integrated-modeling efforts. COCOS (see O. Sauter and S. Y Medvedev, Comput. Phys. Commun. 184 (2013) 293) is a system for standardizing the nomenclature between possible tokamak COordinate COnventionS (e.g., the directions of the toroidal and poloidal angles). By identifying the COCOS identifier of each code in an integrated-modeling workflow, a series of standardized transformations can be performed ensuring that data is exchanged properly between the codes. Thus far, COCOS has been used to ensure such consistency when working with the EFIT, ONETWO, and TGYRO modules within OMFIT, as well as for exchanging data between several common file formats, including gEQDSK, GACODE input.profiles, and ONETWO's statefile. In addition, OMFIT allows for the computation of flux-surface averages in arbitrary COCOS and leverages the OMAS library [1] to transfer its data to the ITER IMAS data storage while ensuring compliance with the ITER coordinate convention number 11.