Bulletin of the American Physical Society
63rd Annual Meeting of the APS Division of Plasma Physics
Volume 66, Number 13
Monday–Friday, November 8–12, 2021; Pittsburgh, PA
Session ZO03: HED: Computational and Analytic TechniquesOn Demand
|
Hide Abstracts |
Chair: Adam Sefkow, University of Rochester Room: Rooms 302-303 |
Friday, November 12, 2021 9:30AM - 9:42AM |
ZO03.00001: A non-LTE model for spectroscopy based on self-consistent average-atom orbitals Stephanie B Hansen High energy density experiments and simulations rely on understanding the material properties of matter at extreme conditions. These include equations of state (EOS), transport coefficients, and radiative properties such as opacities and emissivities. Radiative properties are typically calculated by specialized multi-configuration codes that are based on data for isolated atoms, and which are thus not explicitly consistent with the density-functional-theory (DFT) codes typically used to generate high-quality EOS and transport data. Here, we show that self-consistent orbitals from a DFT-based average atom code can be used to build multiconfiguration atomic structure [1] suitable for collisional-radiative modeling, thereby extending the functionality of DFT-based models to non-LTE conditions and increasing the consistency of opacity and emissivity calculations with EOS and transport tables based on DFT models. The detailed non-LTE emission and absorption spectra produced by this model natively account for plasma density effects such as continuum lowering and is compared with spectra from the widely used Spectroscopic Collisional-Radiative Atomic Model (SCRAM) [2]. |
Friday, November 12, 2021 9:42AM - 9:54AM |
ZO03.00002: A theoretical approach for transient shock strengthening in high-energy-density laser compression experiments Michael J Wadas, Griffin S Cearley, Jon H Eggert, Eric Johnsen, Marius Millot Laser-compression experiments in high-energy-density systems typically utilize shock waves passing through a series of different materials to achieve a desired state of compression. In this study, a theoretical approach for strengthening such shock waves is examined. A method based on characteristics analysis is used to semi-analytically solve the problem of a shock passing through an intermediate region of non-uniform impedance between the experimental apparatus and the sample under study that increases the strength of the shock initially transmitted into the sample. It is shown that an exponential discretization of impedance in the intermediate region is the most efficient distribution for shock strengthening, leading to as much as a 25% increase in the pressure of the sample, significantly extending the range of achievable states in laser-driven dynamic compression experiments. The results of the analysis are verified via comparison to simulations performed with the HYADES hydrodynamics code. |
Friday, November 12, 2021 9:54AM - 10:06AM |
ZO03.00003: Revealing the Atomic Motion Composing the B1–B2 Structural Transformation of MgO Under High Pressures Brenda M McLellan Magnesium oxide (MgO) is an abundant material on Earth and plays an important role as as an optical window in high-energy-density experiments. At ambient conditions MgO is stable in the B1 (NaCl) structure. When shock compressed1 a structural transformation has been observed where it morphs into the B2 (CsCl) structure. Ramp compression and x-ray diffraction experiments2 have additionally detected a possible intermediate structure along the B1–B2 phase transition. Our goal is to comprehend the atomic motion during the transformation. Calculations based on density functional theory have been performed throughout a structural coordinate space connecting the B1 and B2 structures. The physical process of the transformation is represented as a set of thermodynamically optimal intermediate structures that form a pathway across the structural coordinate space. The structural transformation based on this prediction will be presented. The predicted atomic motion composing this transformation will be shown and compared with previously unexplained experimental observations. This material is based upon work supported by the Department of Energy National Nuclear Security Administration under Award Number DE-NA0003856. |
Friday, November 12, 2021 10:06AM - 10:18AM |
ZO03.00004: Uncertainty Quantification using Two-Shock Campaign data and simulations Paul A Bradley, Brian M Haines The 2-shock campaign$^1$ was a series of NIF capsule implosions that used a 675 micron outer radius capsule with a roughly 175 micron thick 1-{\%} Si-doped ablator. These capsules were used to test several hypotheses for yield degradation, including: shock convergence mis-timing, enhanced surface roughness, implosion asymmetry, and increases to the convergence ratio (initial to final inner radius ratio). We use the xRAGE Eulerian Adaptive-Mesh-Refinement computer code to model these implosions in 1-D and 2-D. We start by comparing our results to DD, DT, and TT yields, along with the DT/TT ratio, burn weighted DT Tion value, x-ray hot spot size and burn width. We use normalized differences between the simulation and experiment for these quantities and compare these results to those derived from measurement uncertainties. These comparisons show that 2-D simulations better match the data than 1-D simulations, but there is room for improvement for some gas filled capsules and capsules where the convergence ratio is great than 15. |
Friday, November 12, 2021 10:18AM - 10:30AM |
ZO03.00005: Implementation of a Two-Dimensional Unsplit Volume of Fluid Interface-Capturing Method for Multi-Fluid Compressible Flows in the FLASH Code Adam Reyes, John W Grove, Marissa B Adams, Abigail Armstrong, Kasper Moczulski, Periklis Farmakis, Edward C Hansen, Yingchao Lu, David Michta, Don Q Lamb, Petros Tzeferacos We present an implementation of the volume-of-fluid (VOF) method to model multiple immiscible compressible fluid species within the FLASH code’s unsplit hydrodynamics solver. FLASH is a highly capable, parallel, adaptive-mesh refinement, finite-volume Eulerian hydrodynamics and magnetohydrodynamic code with extended physics capabilities. FLASH assumes a Dalton mix of the species within each computational cell, and advects the corresponding mass fractions with the flow, resulting in the mixing of species across contact discontinuities. In VOF, species are assumed to occupy distinct volumes whose interfaces may cut the computational cells and are assumed to be in mechanical equilibrium with a single velocity field shared by all species. Special care needs to be taken to allow for the compressibility of the different species and for the modeling of shocks and discontinuities in the flow, maintaining sharp interfaces between species even at contact discontinuities. Additional considerations are also necessary within a dimensionally unsplit formulation to prevent fluxing the same volume into multiple cells. We highlight the capabilities of this VOF implementation in FLASH for simple gamma-law equations of state (EOS); the formulation is readily extended to tabulated EOS for simulations of high-energy-density physics and laser-driven experiments. |
Friday, November 12, 2021 10:30AM - 10:42AM |
ZO03.00006: Multiphysics code validation and sensitivity analysis through integrated modelling of convergent shock tube experiments Adam R Fraser, Dave A Chapman, James D Pecover, Mila D Fitzgerald, Nicolas-Pierre L Niasse, Aidan C Crilly, Nicholas Hawker, Nathan Joiner, Jeremy P Chittenden A fundamental requirement to have confidence in the predictive capabilities of multi-physics simulation methods is their ability to reproduce verification and validation data. This submission presents a sensitivity study performed using the convergent shock tube experiments of Setchell et al. [1]. Depending on the Mach number of the incident shock, the experiment demonstrates deviation from ideal gas and the onset of multi-level ionisation after successive overlapping of shocks on-axis following reflection from the inner walls, significantly impacting the subsequent dynamics. Despite substantially lower temperatures than typical ICF conditions, with the transition from a high-temperature gas to a plasma the experiment provides important validation data for codes modelling coupled physical phenomena. |
Friday, November 12, 2021 10:42AM - 10:54AM |
ZO03.00007: Understanding Shock-Release Experiments Using a Numerical Simulation of VISAR Daniel H Barnak, Riccardo Betti, Varchas Gopalaswamy, Aarne Lees, Alexander Shvydky Shock release, or rarefaction, is a ubiquitous yet not well-understood phenomenon. An experiment using planar foils and the velocity interferometer system for any reflector (VISAR) diagnostic was performed to quantify the release mass and velocity. A synthetic VISAR diagnostic based upon previous work[1] was developed as another possible analysis tool to compare 1-D hydrodynamics simulations directly with real experimental data. The synthetic VISAR is discussed in detail with an emphasis on constructing useful reflectivity models for future experiments. Discrepancies between the model results and the experiments are discussed in detail with an emphasis on heat transport, equation of state, radiation transport, and 2-D effects. [1] S. Laffite et al., Phys. Plasmas 21, 082705 (2014). |
Friday, November 12, 2021 10:54AM - 11:06AM |
ZO03.00008: Self-Consistent Description of Dense Plasma Mixtures Liam G Stanton, Abdou Diaw, Luciano G Silvestri, Michael S Murillo High energy-density laboratory plasmas are often multi-component, and a self-consistent hydrodynamic description can be challenging to construct due to effects such as partial pressures. Simplifications are typically used to mitigate these challenges, such as velocity/temperature fields that do not resolve individual species and mixing rules between single-component equations of state; however, these approximations can break down in experimentally relevant regimes [1, 2]. Here, we extend the work of Diaw and Murillo [3] to construct a self-consistent model of plasma mixtures within the framework of the BBGKY hierarchy. Not only does this extension provide a more consistent description of a plasma mixture that resolves every field for each species, it also establishes a direct connection to the underlying atomic correlations. Simulation results are presented for interfacial mixing within ICF targets, and comparisons are made to more traditional hydrodynamic models. |
Friday, November 12, 2021 11:06AM - 11:18AM |
ZO03.00009: Investigating State Relocalisation in Mg and Mg Compounds Thomas D Gawne, Patrick Hollebon, Gabriel Perez-Callejo, Oliver Humphries, Justin S Wark, Sam M Vinko Continuum lowering predictions from standard plasma models have come under scrutiny in light of significant discrepancies with the measured continuum lowering at recent experiments. We attribute this to the difficulty in defining states as purely bound or purely free in average atom models. Here, we use the inverse participation ratio of valence states calculated using finite-temperature density function theory to study the localisation (“boundness”) of states in Mg and Mg compounds. We apply this with a hybrid Kohn-Sham scheme to perform calculations with electron temperatures in excess of 100 eV, investigating the effect of the density around the Mg sites on the recombination of the M-shell. It is found that the M-shell clearly relocalizes in isochorically heated Mg; however, in MgF2, while peaks associated with the Mg M-shell form in the density of states, these states do not fully localise. |
Friday, November 12, 2021 11:18AM - 11:30AM |
ZO03.00010: Mechanisms for laser-plasma interaction and their macroscopic effects within the extended MHD framework James Young, Matthew Evans, Hannah R Hasson, Imani West-Abdallah, Marissa B Adams, Pierre-Alexandre Gourdain Modeling lasers interacting with z-pinch plasma is made particularly computationally intense due to the range of scales required. Z-pinch simulations typically require μm/ns spatiotemporal scales, while laser-plasma interaction (LPI) can require sub-nm and pico/femtosecond resolutions. This problem is usually handled by hybrid codes, but for short laser interactions the fluid approximation may give a relatively accurate picture of flow dynamics on macroscopic scales due to strictly enforcing mass, momentum and energy conservation. A question that deserves attention is exactly what mechanism within the XMHD framework (such as PERSEUS) performs the coupling between electromagnetic wave and plasma. For hybrid codes, the laser energy can be deposited by ray-tracing and ponderomotive force. However, for XMHD that includes electron inertia, although the full set of Maxwell's equations are solved, the interaction occurs within Generalized Ohm's Law. The fluid approach replicates several features usually associated with PIC such as a cut-off density, hole-boring, and density fringes. We will explore these to understand how they are related to the analogous theoretical effects. |
Friday, November 12, 2021 11:30AM - 11:42AM |
ZO03.00011: A Deterministic Collisional Ionization Module for Particle-in-Cell Codes Stephen E DiIorio, Benjamin J Winjum, Frank S Tsung, Jennifer Elle, Ricardo A Fonseca, Alexander G Thomas We present updates to our collisional ionization module for particle-in-cell (PIC) codes. Our method treats and calculates ionization events deterministically as each particle's rate is calculated explicitly and deposited onto a grid. This grid of ionization rates is then used to advance ion densities, which allows us to track how much new charge is generated each timestep, so we can create newly ionized electrons accordingly. Additionally, the ionization rate grid, with little modification, keeps track of how much energy is lost per grid cell due to ionization physics. We interpolate this information back onto the particles; this allows for a continuous decrease in the energy of macro-particles as they participate in ionization events and allows for the easy calculation of the new momentum of ionized electrons. Collectively, this particle-to-grid and grid-to-particle information transfer act as a "smoothing" process, reducing noise considerably compared to other current algorithms. This module has been tested for its accuracy and integrated into the PIC code OSIRIS. In addition to this, we present several simulations highlighting the new physics that is captured when considering collisional ionization in different scenarios. |
Friday, November 12, 2021 11:42AM - 11:54AM |
ZO03.00012: Enabling Predictive Scale-Bridging Simulations through Active Learning Jeff Haack, Abdourahmane Diaw, Robert S Pavel, Irina Sagert, Brett Keenan, Daniel Livescu, Nick Lubbers, Mike McKerns, Christoph Junghans, Timothy C Germann Designing effective methods for multiscale simulation is a longstanding challenge. Our goal is to advance the state of the art for Machine Learning (ML) beyond sequential training and inference and facilitate scale bridging through novel techniques. Active Learning (AL) is a special case of semi-supervised ML in which a learning algorithm is able to interactively use the fine-scale model to obtain the desired outputs at new data points, making it ideal for concurrent scale-bridging. Our AL procedure will dynamically assess uncertainties of the ML model, query new fine scale simulations as necessary, and use the new data to incrementally improve our ML models. This capability will be demonstrated on two applications: transport in nanoporous media (e.g., for hydraulic fracturing) and inertial confinement fusion (ICF), validating against experimental data. Although the physics is quite dissimilar, both applications represent problems that suffer from inaccurate macro-scale predictions due to subscale physics that are ignored. |
Friday, November 12, 2021 11:54AM - 12:06PM |
ZO03.00013: BARS and mini-BARS for Iteratively Learned Sampling and Tiling of Phase Space in High Energy Density Plasma Kinetic Simulations Bedros B Afeyan, Sean M Finnegan, Luis Chacon We will show the efficacy of Bidirectional Adaptive Refinement Scheme, BARS and mini-BARS (a reduced functionality version), to tackle plasma kinetic simulations with much more control on speed, accuracy and interpretability than traditional PIC code (Monte Carlo) sampling. We will take the initial value probelm of nonlinear kinetic electron plasma waves as an example, and show how sparse sampling in the bulk and dense sampling in the tail of the electron velocity distribution function will guarantee both sufficient accuracy and great speed at the same time. We will indicate how these methdos can be turned into full Machine Learning algorithms adapting what is learned at one set of plasma parameters to efficiently run nearby parameter cases, such as changed perturbation amplitude and wavenumber. Strict error control and novel Python diagnostics with reconstructed velocity distribution functions will be used. All our proof of principle results will be shown with DPIC, a Los Alamos Electrostatic PIC code. |
Friday, November 12, 2021 12:06PM - 12:18PM |
ZO03.00014: A Machine Learning-Based Analysis for Efficient Predictive Modeling of Negative Hydrogen Ion Sources Tiernan Casey, Simone Venturi Negative hydrogen ion sources (NHIS) are the preferred mode of plasma heating in tokamak devices due to superior neutralization efficiency. As such, accurate predictions of the H- density are crucial for the design of such machines. This work exploits state-of-the-art statistical and data science tools developed for machine learning to reducing the computational cost of a global NHIS model while preserving its predictive capabilities. This model includes detailed pathways composed of hundreds of reactions to achieve high fidelity in reproducing the important kinetics features. Multiple approaches are deployed here with the aim of revealing which reactions are responsible for generating critical species that catalyze H- production during the temporal evolution of the chemical system. Some of these techniques are grounded on graph theory and tackle the complex network of reactions based on dominant trajectories across the chemical graph. Unsupervised machine learning techniques are then applied for reducing the dimensionality by removing the uninformative reactions and clustering the species showing similarities in their dynamics. The result is a reduced order model, and a computationally inexpensive surrogate is constructed via probabilistic neural networks for emulating its steady-state solutions. The probabilistic attribute of this surrogate allows taking into account the uncertainties on rate coefficients while predicting the quantities of interest. Finally, non-linear manifold learning is used to study the global model's dynamical structure and introduce simplifications to the reaction set based on time scale analysis. |
Follow Us |
Engage
Become an APS Member |
My APS
Renew Membership |
Information for |
About APSThe American Physical Society (APS) is a non-profit membership organization working to advance the knowledge of physics. |
© 2024 American Physical Society
| All rights reserved | Terms of Use
| Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 20740-3844
(301) 209-3200
Editorial Office
100 Motor Pkwy, Suite 110, Hauppauge, NY 11788
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700