Disruption Mitigation in Tokamaks with Massive Material Injection

Licentiate thesis, 2023

The sudden loss of confinement of the energy content of fusion plasmas in off-normal events, called disruptions, is among the most severe threats to the future of fusion energy based on the tokamak design. An efficient disruption mitigation system will therefore be of utmost importance for future large, high-current devices such as ITER. The potentially greatest threat to be mitigated is posed by currents carried by highly energetic electrons, called runaway electrons, which may cause severe damage upon wall impact. The disruption mitigation system must also ensure a sufficiently homogeneous deposition of the thermal energy on the plasma-facing components, and avoid excessive forces on the machine due to currents flowing in the surrounding structures. The currently envisaged mitigation method is to initiate a massive material injection, e.g. in the form of a pressurized gas or a shattered cryogenic pellet, when an emerging disruption is detected, and so attempt to better control the plasma cooling and energy dissipation.

In this thesis, we develop modeling tools for the various physical phenomena present during a tokamak disruption mitigated by a massive material injection. This includes extending the numerical tools GO and DREAM with the capability to handle more advanced geometry, effects of partial ionization in the cooling plasma on the generation of runaway electrons, and the material assimilation in the plasma following a shattered pellet injection. These tools are then used to perform integrated numerical simulations, assessing the mitigation performance for a wide range of injection scenarios in reactor-scale tokamak devices. Finally, we also develop an analytical model for the radial transport of the relatively cold and dense material recently ablated from a shattered pellet upon exposure to the hot plasma.

Our results indicate that the severity of a disruption in a reactor-scale device can be significantly reduced by a carefully chosen injection scheme and composition of the injected material. In particular, a two-stage shattered pellet injection might efficiently reduce the localised heat loads and the runaway generation due to the hot-tail mechanism, by allowing for an intermediate equilibration of the superthermal electron population between the injections.
However, the strong runaway avalanche associated with a high plasma current was found to be able to amplify even a very small runaway seed, such as those produced by tritium decay and Compton scattering during nuclear operation, to several mega-amperes. The reason is that the intense cooling from the injected material leads to a high induced electric field and a substantial recombination, resulting in an enhanced avalanche multiplication. Our calculations also indicate that this mitigation scheme might be further complicated by a relatively large outward drift of the recently ablated pellet material.