Impact of nanocrystalline grain boundaries on first-shock dynamics in ICF implosions

ORAL

Abstract

Repeated fusion ignition shots at NIF employ a strong 12 Mbar first shock to completely melt the high-density carbon (HDC) ablator, which is thought to be a potential seed of hydrodynamic instabilities and DT fuel/ablator mixing. Achieving a new 10+ MJ yield milestone would be significantly aided by the reduction of the fuel adiabat and robust control of instability seeds and resultant fuel/ablator mix dynamics at the nanoscale. Current ICF design moves towards nanocrystalline diamond (NCD) ablators, which possess a significant fraction of low-density (~2.2 g/cm³) graphitic sp² carbon atoms in NCD grain boundaries. However, the potential impact of grain boundary structure on the first shock and release into DT ice is currently unknown.

In this talk, we discuss results of billion-atom machine learning molecular dynamics (MD) simulations of shocked NCD ablator targets at experimental micrometer and nanosecond time and length scales on exascale DOE supercomputer Frontier, specifically focusing on investigating inhomogeneities due to incomplete melting behind the first shock front as well as the persistence of solid fragments upon the NCD release. Our goal is to explore whether the strength of the first shock can be safely reduced without a significant effect on NCD/DT mix.

Presenters

  • Ivan Oleynik

    University of South Florida

Authors

  • Ivan Oleynik

    University of South Florida

  • Joseph M Gonzalez

    University of South Florida

  • Juergen Biener

    Lawrence Livermore National Laboratory

  • Daniel S Clark

    Lawrence Livermore National Laboratory

  • Marius Millot

    LLNL, Lawrence Livermore National Laboratory

  • Vladimir A Smalyuk

    Lawrence Livermore National Laboratory, LLNL

  • Philip A Sterne

    Lawrence Livermore National Laboratory, Lawrence Livermore Natl Lab

  • Seth Davidovits

    Lawrence Livermore National Laboratory, Lawrence Livermore Natl Lab

  • Jon Henry Eggert

    LLNL