Benchmarking simulation methods for understanding warm dense matter

8 of the co-authors meeting at the recent “Physics of Nonideal Plasmas” conference in Oxford, England. From left: Hanno Kählert, Pontus Svensson, Christopher Makait, Mandy Bethkenhagen, Gianluca Gregori, Ronald Redmer, David Ceperley, and Michael Bonitz. Credit: Michael Bonitz

Understanding the properties of hydrogen under high compression, a type of Warm Dense Matter (WDM), is crucial for astrophysics and laboratory experiments like inertial confinement fusion. However, experiments in WDM are complex and affected by various factors, often limiting data accuracy.

To improve this, researchers Bonitz et al. have worked to gauge the accuracy of computer simulations that provide insights into WDM properties.

“We carefully analyzed and compared over ten simulation approaches, evaluating their strengths, weaknesses, and accuracy. Our conclusion is that the most promising strategy is to develop several methods in parallel and combine them intelligently,” explained author Michael Bonitz.

The researchers used methods including the Quantum Monte Carlo simulations, which yield nearly exact results, alongside less precise techniques like density functional theory (DFT), semiclassical MD, and the average atom model. They benchmarked these methods and integrated first-principles simulations with simpler chemical models to calculate complex quantities, such as the ionization potential depression.

“When we began this project, our goal was to provide a comprehensive overview of thermodynamic properties. As we engaged with leading experts, their interest helped significantly expand the scope of our research, which now also includes the entire phase diagram as well as dynamic and transport properties.” Bonitz noted.

Bonitz hopes this work will encourage critical assessment of simulations and a rigorous evaluation of individual methods’ accuracy and applicability. He also anticipates this will lead to the development of more powerful and predictive simulations.

By doing these benchmarks, people with simpler methods can find out which approximations could be favorable for dense hydrogen, which will be important for alternative computation methods in the future.

Source: “Towards first principles-based simulations of dense hydrogen” by Michael Bonitz, Jan Vorberger, Mandy Bethkenhagen, Maximilian P. Böhme, David M. Ceperley, Alexey Filinov, Thomas Gawne, Frank Graziani, Gianluca Gregori, Paul Hamann, Stephanie B. Hansen, Markus Holzmann S. X. Hu, Hanno Kählert, Valentin V. Karasiev, Uwe Kleinschmidt, Linda Kordts, Christopher Makait, Burkhard Militzer, Zhandos A. Moldabekov, Carlo Pierleoni, Martin Preising, Kushal Ramakrishna, Ronald Redmer, Sebastian Schwalbe, Pontus Svensson, and Tobias Dornheim, Physics of Plasmas (2024). This article can be accessed at https://doi.org/10.1063/5.0219405 .