Recent DIII-D advances in runaway electron measurement and model validation

Abstract

Novel measurements and modeling of runaway electron (RE) dynamics in DIIID have resolved experimental discrepancies and validated predictions for ITER, improving confidence that RE avoidance and mitigation can be predictively optimized without empirical tuning. Results are organized in terms of the RE life-cycle, going from 1) RE formation, where RE avoidance is desired, next to 2) the mature RE plateau, where RE dissipation is desired, and ending with 3) the RE final loss. Considering RE formation, first measurements of the RE seed current demonstrates that present hot-tail theories are not yet accurate and require improved treatment of the pellet dynamics. Novel measurements of kinetic instabilities in the MHz-range have been made in the RE formation phase, and the intensity of these modes have been correlated with previously unexplained empirical thresholds for RE generation. Controlled RE dissipation experiments in quiescent regimes have validated RE distribution function dependencies on collisional and synchrotron damping, both in terms of distribution function shape and dissipation rates. Measurements of RE bremsstrahlung and synchrotron emission are now used in tandem to resolve energy and pitch-angle effects, respectively. A resolution to long-standing anomalies in the RE dissipation rate in quiescent regimes is offered by, taking into account, kinetic instability effects, with kinetic instabilities in the 100-150 MHz range now directly observed. Recent experiments also observe kinetic instabilities in the mature post-disruption RE plateau phase if the collisional damping rate for the instabilities is reduced. Further experiments in high-Z collisional dissipation have found that the dissipation rate saturates with high-Z injection quantity, likely due to diffusion rates being slower than vertical instability rates in DIII-D. Considering the final loss, a model has been developed that enables estimation of the first-wall Joule heating that is in good agreement with experiment. Finally, controlled access to edge safety factor of 2 RE equilibria have identified novel dynamics brought about by large-scale kink instabilities. These dynamics are typified by fast (tens of micro-second) RE loss rates without RE beam regeneration. The above measurements and comparison with theory substantially improve confidence that model-based optimization of RE avoidance and mitigation can be achieved, and indicate possible new opportunities for RE avoidance or mitigation via kinetic instabilities.