Particle physics experiments

Understanding hotspot conditions in National Ignition Facility implosion experiments

Research at Lawrence Livermore National Laboratory (LLNL) describes a validation exercise of simple models used to understand the hotspot conditions reached during an implosion, which find good agreement against a set of simulations.

Progress toward ignition requires accurate diagnosis of current conditions and evaluation of proximity measurements for implosion experiments at LLNL’s National Ignition Facility (NIF). Hotspot conditions are not directly measured, but rather inferred, often using simple 0- and 1-dimensional (1D) models.

“When we perform firing experiments on NIF, we have a phenomenal range of diagnostics that can measure many aspects of firing and its performance,” explained LLNL physicist Alex Zylstra. “Unfortunately, some things that we really care about in burn physics aren’t directly measurable, like the pressure or the amount of energy in the hot spot.”

Plasma distribution plots

This image shows the distribution plots for the model parameters pressure, ion temperature, burn width, and radius. 1 credit

Researchers therefore rely on simple models to deduce these quantities from the data. But for the results derived from the inferences to be credible, researchers must then compare these models.

The findings are presented in Plasma physicswith LLNL physicists Zylstra, Ryan Nora, Prav Patel, and Omar Hurricane as the paper’s authors.

The research is also a much deeper validation exercise for these simple models, using more than 20,000 2D simulations varying performance as well as various things that can “go wrong” in an experiment.

“We find that simple models still do a fairly good job over a reasonable range of parameters,” Zylstra said. “We have also started using a new Markov Chain Monte Carlo algorithm to produce probabilistic distributions for inferred quantities based on the measurement uncertainties of the inputs.”

Hurricane said these simple models are important for assessing some of the burning plasma criteria in the recently released LLNL. Nature article, as well as future articles.

“These simple models have been used in the literature for quite some time,” Hurricane said. “What’s new here is that the development of ‘ensembles’ by the cognitive simulation group within the ICF program has now produced simulation ensembles large enough to support this type of study, testing our model against a known truth using the synthetic data from the simulation.”

Reference: “Model validation for inferred hot-spot conditions in National Ignition Facility experiences” by AB Zylstra, R. Nora, P. Patel and O. Hurricane, December 1, 2021, Plasma physics.
DOI: 10.1063/5.0069366