On the origin of the plasma current spike during a tokamak disruption and its relation with magnetic stochasticity
E. Nardon, K. Särkimäki, F.J. Artola, S. Sadouni, the JOREK Team, Jet Contributors
Abstract
Abstract A JOREK 3D non-linear MHD simulation of a disruption triggered by an argon massive gas injection in JET, which quantitatively reproduces the plasma current ( <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <mml:msub> <mml:mi>I</mml:mi> <mml:mrow> <mml:mi mathvariant="normal">p</mml:mi> </mml:mrow> </mml:msub> </mml:math> ) spike (Nardon et al 2021 Plasma Phys. Control. Fusion 63 115006), is analyzed in order to investigate the origin of the <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <mml:msub> <mml:mi>I</mml:mi> <mml:mrow> <mml:mi mathvariant="normal">p</mml:mi> </mml:mrow> </mml:msub> </mml:math> spike and its relation with magnetic stochasticity. The <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <mml:msub> <mml:mi>I</mml:mi> <mml:mrow> <mml:mi mathvariant="normal">p</mml:mi> </mml:mrow> </mml:msub> </mml:math> spike is associated to a current density ( j φ ) profile relaxation which appears to result from Shear Alfvén Wave (SAW) propagation along stochastic field lines, as proposed by Boozer (2019 Plasma Phys. Control. Fusion 61 024002; 2020 Phys. Plasmas 27 102305), possibly complemented by a macroscopic E×B flow structure. Using axisymmetric JOREK simulations involving a mean field Ohm’s law, we verify that the level of hyper-resistivity associated to SAWs is consistent with the prediction made in (Boozer 2019 Plasma Phys. Control. Fusion 61 024002; Boozer 2020 Phys. Plasmas 27 102305), which connects the <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <mml:msub> <mml:mi>I</mml:mi> <mml:mrow> <mml:mi mathvariant="normal">p</mml:mi> </mml:mrow> </mml:msub> </mml:math> spike with the level of stochasticity. The relaxation comprises two main phases, the first one corresponding to a fast (0.1 ms) and almost complete j φ flattening in the q < 2 region, while the second one is longer (0.5 ms) and corresponds to a more gradual, global and incomplete j φ flattening. During the first phase, strong E×B flows develop that play a key role in mixing impurities into the core.