Role of the equilibrium and perturbative central current density in sawtoothing and non-sawtoothing discharges in KSTAR

Abstract

The experimental measurement [1,2] of the central current density before and after the sawtooth crash event puzzled the theoretical community for about two decades and has since largely been forgotten: the measurement confirmed that the q0 changes very little (i.e. from 0.75 to 0.8) whereas the theoretical community has believed that the q0 should return to ~1 from ~0.75 right after the crash to justify the m/n=1/1 kink instability believed to cause the crash. Note that there are experimental results contradicting with those two measurements. These days, many discharges are routinely operated with q0>1 and it is therefore rare to observe sawtoothing behaviors. Recent experimental and theoretical work on multiple mode structures (m/n=2/2, 3/3, etc.) in addition to the m/n=1/1 mode in the core of sawtoothing plasmas (q0<1) in KSTAR provided a new avenue to revisit this problem. A localized electron cyclotron heating/current drive (ECH/CD) inside the q~1 surface induced higher order modes right after the crash in the sawtoothing discharge and the time evolution of the higher order mode was studied using a 2D Electron Cyclotron imaging (ECEI) diagnostic system [3]. The time evolution of the higher order modes were compared with a reduced MHD simulation with an empirical source term for the radially localized current drive on a flat q-profile with q0~1 [4]. After each crash, the higher m/n mode transforms into the lower m/n mode and eventually merges into a single m/n=1/1 mode before the crash. In order to understand the observed time evolution of the multiple mode structures, the dependence of growth rates of the time evolving higher order modes for three different core equilibrium conditions are studied in detail using the M3D-C1[5] code in cylindrical geometry: (1) q0 changes from 0.8 to 0.75 (2) q0 changes from ~1 to 0.75 (3) q0 >1 as in non-sawtoothing H-mode plasmas.