In collaboration with Eero Hirvijoki (PPPL), the NIMROD code was coupled to the MCCC (Monte-Carlo Collision Code) for improved modeling of runaway electron (RE) test particle orbits. Whereas the previous model included a drag term due to small angle RE collision with the background plasma, the coupled code accounts for both small-angle and large-angle collisions. This allows pitch-angle scattering to be calculated, which is expected to be particularly important for disruption mitigation scenarios with high-Z impurities. While each RE orbit is followed many times around the torus within a single NIMROD time step, MCCC is invoked once per NIMROD time step to update the momentum vector due to collisions. This approximation, justified by the low collision rate of the REs, allows the coupled code to run without significant slowdown compared to the previous model. Following initial tests of the model, disruption mitigation cases run using the previous model will be repeated to compare directly. The improved collision model will also facilitate development of RE generation models in NIMROD.
Approximate expressions for the static Dirac-Hartree-Fock-Slater (DHFS) screening potentials first used by F. Salvat, were used to calculate the elastic scattering logarithm for argon and neon. The scattering logarithm for elastic collisions of fast electrons with impurities is required for Fokker-Planck calculations of the attenuation of fast electrons scattering from pellet ablation clouds. It is the counterpart to the ionic Coulomb logarithm and is derived from the differential scattering cross section (DSC) based on the first Born approximation for high impact energies. While Salvat originally proposed a three pole Yukawa potential for the Thomas Fermi (TF) atomic potential, a simpler model uses an approximation with a single Yukawa potential and one free parameter, s. This may be estimated in a number of ways. In the calculations, the value of s was adjusted to get the approximate TF based DSC to coincide with the experimental and more refined calculations for grazing angles. Different impurities require different values of s. It is notable that the approximate TF atomic potential is far more accurate at small angles than the actual TF atomic potential since in the approximate TF atomic potential, the density decreases exponentially, which models the DHFS calculation better compared to the (1/r)**6 decay of the TF model.
OMFIT has been extended to allow users to interact with the distributed version control system ‘git’ from within the OMFIT framework itself. This is an important development, which streamlines the process of contributing to the physics modules and lowers the initial barrier to contributing to the OMFIT code-base. By leveraging GUIs, users can now contribute to the development of physics modules with minimal knowledge of the `git` version control system. The new capability allows users to import/export modules from and to multiple code repositories and development branches while handling the underlying complexity that can naturally arise with multiple people working on the same module simultaneously.
The paper “Gyrokinetic Predictions of Multiscale Transport in a DIII-D ITER Baseline Discharge” by C. Holland, N. T. Howard, and B. A. Grierson has been accepted for publication in Nuclear Fusion. This article details new multiscale gyrokinetic simulations of turbulent transport performed with the GYRO code, for parameters measured in a ITER baseline discharge performed with dominant electron heating and low injected torque on DIII-D, which were presented at the 2016 IAEA FEC meeting. These simulations demonstrate that even at low torque, transport in these plasmas is robustly multiscale, with approximately equal amounts of electron thermal transport being driven by ion and electron scale fluctuations. More surprisingly, although the rotation level is low, the simulations show an extreme sensitivity to the rotational shear because the turbulence instabilities are near-marginal with correspondingly small growth rates. The results of these simulations will be used to test and improve the fidelity of TGLF for ITER relevant parameters.
These highlights are reports of research work in progress and are accordingly subject to change or modification