NIMROD DIII-D Dual SPI Injector Simulations*

A reliable and efficient Disruption Mitigation System is essential to the safe operation of ITER and future burning

tokamaks. Past experiments on single injector systems (Massive Gas Injection or Shattered Pellet Injection) were

dominated by strong n=1 MHD activity, limiting the radiation and resulting in toroidally peaked hot spots.  Recent

DIII-D SPI experiments [J.L. Herfindal et al., ``Injection of multiple shattered pellets for disruption

mitigation in DIII-D'' (NF2019), {f 59} 106034]

 have focused on dual-injector scenarios to improve the efficiency and reduce the toroidal peaking.  

NIMROD simulations of DIII-D dual injector SPI show an asymmetry in the thermal quench efficiency; total radiation

increases as the time delay between two injectors is decreased from dt = +0.4ms ($Delta W_{rad}/Delta

W_{thermal}simeq$45\%) efficiency, to dt = -0.4ms ($simeq$70\%) (positive delay indicates SPI at $phi$=15$^circ$

leads, negative indicates SPI at $phi$=215$^circ$ leads). This unexpected asymmetry is attributed to the helicity of

the tokamak plasma and its preferred direction of evolution of the quenching plasma column.  For the high radiation

thermal quench (dt = -0.4ms), analysis shows a persisting (m,n) = (2,1) structure throughout the quench. The low

radiation case (dt = +0.4ms) shows a late break-up of the (2,1) and is overcome by a (1,1) that results in the final

thermal collapse.  These NIMROD simulation suggest a strong correlation between reduced MHD activity and an efficient

thermal quench.  We will present these NIMROD results and validation comparisons to experiments and suggest ways to

exploit them for further improvements of the SPI DMS.