Sherwood 2015

Abstract Details

files Add files

Nonlinear Diamagnetic Stabilization Effects on m=2, n=1 Cylindrical Double-Tearing Modes in Hall MHD Simulations

Author: Stephen R. Abbott
Requested Type: Poster Only
Submitted: 2015-01-19 09:20:18

Co-authors: K.Germaschewski

Contact Info:
University of New Hampshire
8 College Rd
Durham, NH   03824

Abstract Text:
Reverse-shear magnetic configurations may be unstable to double-tearing mode instabilities (DTMs) when two nearby rational surfaces of the same safety factor couple and reconnect. The emergence of large magnetic islands can trigger a fast phase of the instability which collapses the annular current ring and generates strong, sheared flows. Preventing the onset of this explosive growth regime may be necessary to prevent disruptions in advanced tokamaks with non-monotonic safety factor profiles. Simulations of DTMs in the presence of sheared poloidal flows has indicated that differential rotation can have a stabilizing effect by decoupling the two tearing surfaces. The introduction of diamagnetic drifts through pressure profiles (of forms similar to ITBs) creates both this differential rotation and an addition local stabilization of the reconnection process which has been previously studied in m=1 kink-tearing modes. Our simulations of a cylindrical m=2, n=1 DTM with the Hall MHD code MRC-3d show that the location of the pressure gradient has a large impact on nonlinear stability. Differential rotation is most effective at small amplitudes and initially decoupled modes will recouple when the islands grow large. We identify that, for our equilibrium, the island on the outer q=2 surface is primarily responsible for this recoupling and that locating the peak diamagnetic drift at this surface significantly reduces the mode growth and delays disruption.
The Magnetic Reconnection Code 3D (MRC-3d) is one of a suite of simulation codes built on the libmrc scientific computing library. MRC-3d implements a fully nonlinear, 3D Hall-MHD model in arbitrary curvilinear geometry through a Python based code generator. It features both implicit and adaptive explicit time-stepping via the PETSc computational toolkit in a high-performance, parallel manner, and data analysis is simplified by use of portable hdf5 output and a Python interface to code internal discretizations.


March 16-18, 2015
The Courant Institute, New York University