Interaction between runaway electrons and internal kink in a post-disruption plasma

Direct interactions between the n = 1 (n is the toroidal mode number) resistive internal kink instability and relativistic runaway electrons (REs) in a post thermal quench toroidal plasma are numerically investigated. A recently developed hybrid model, where the runaway current is treated as a fluid component, is adopted to study how REs affect the internal kink stability and the associated eigenmode structure. The RE-modified internal kink perturbation is in turn superposed with the equilibrium field, in order to study the effects of 3D fields on the drift orbits of REs and consequently on the RE confinement and loss. Results are compared with that assuming the single fluid model for the runaway beam. It is found that REs destabilize the resistive internal kink mode, leading to a better recovery (compared to the standard single fluid model) of the mode growth rate scaling of γ ∼ S−1/3 at lower values of Lundquist number S. Furthermore, REs significantly modify the internal kink eigenfunction inside or near the q = 1 surface. In particular, a new narrow layer forms within the resistive layer, where a strong peaking of the perturbed parallel current develops. The RE contribution is also found to substantially enhance the coupling among poloidal harmonics, in particular between m = 1 and its nearest sidebands. All these changes to the perturbation structure consequently affect the RE drift orbits in the presence of the internal kink instability. Compared to the fluid model, 3D perturbations computed with the more consistent hybrid model lead to less loss of relativistic electrons in the RE beam, when the perturbation level reaches that of the equilibrium field. At lower (but still large) perturbation amplitude (∼10% of the equilibrium field), the internal kink mode has little effect on the RE loss. REs launched within the q = 1 surface always stay confined within the plasma, independent of the field perturbation amplitude (up to the level of the equilibrium field), the beam model adopted (fluid versus hybrid), or the particle traveling direction. On the other hand, the individual drift orbit trajectory is qualitatively sensitive to these factors.