Exploring the Physics of the Plasma Liner Experiment: A Multi-dimensional Study with FLASH, OSIRIS, and HELIOS

The Plasma Liner Experiment (PLX) at Los Alamos National Laboratory (LANL) is a platform that seeks to achieve fusion via the Plasma-Jet-Driven Magneto-Inertial Fusion (PJMIF) concept. The experiment utilizes a constellation of plasma guns to generate fusion-relevant conditions and consists of three main phases: (1) target formation, in which up to four plasma guns shoot magnetized hydrogen or deuterium-tritium jets to form a quasi-spherical target, (2) liner formation, in which a 36 guns fire high-atomic-number (e.g., xenon) jets to form a liner shell, and (3) target compression, in which the formed liner implodes the pre-formed target. Each phase of the PLX operates in different plasma regimes, with different physics at play, thus each phase must be simulated separately with appropriate codes. In this study we highlight 1-, 2-, and 3-D simulation results of all three phases using the FLASH, OSIRIS, and HELIOS codes. Some of the key physical processes involved include shock dynamics, kinetic effects, anisotropic thermal conduction, resistive magnetic diffusion, radiation transport, and magnetized jet dynamics. The simulations show that the PLX can form a preheated ($\sim$40 eV), magnetized (electron Hall parameter $>$1) target plasma, and a quasi-collisional liner shell, which can subsequently compress the target to fusion-relevant conditions, reaching temperatures in excess of 1 keV.

💡 Research Summary

The paper presents a comprehensive multi‑code, multi‑dimensional simulation campaign of the Plasma Liner Experiment (PLX) at Los Alamos National Laboratory, which is a test‑bed for Plasma‑Jet‑Driven Magneto‑Inertial Fusion (PJMIF). PLX proceeds through three distinct phases—target formation, liner formation, and target compression—each operating in a different plasma regime and therefore requiring tailored numerical tools. The authors employ the adaptive‑mesh‑refinement (AMR) radiation‑magnetohydrodynamics code FLASH for fluid‑scale physics, the fully relativistic particle‑in‑cell (PIC) code OSIRIS for kinetic effects, and the 1‑D Lagrangian code HELIOS for high‑fidelity material‑interface benchmarking.

In the target‑formation stage, a 2‑D axisymmetric FLASH simulation of two counter‑propagating, magnetized hydrogen jets is performed. The model solves three‑temperature resistive MHD with anisotropic thermal conduction, electron‑ion energy exchange, and six‑group radiation diffusion, using PROPACEOS EOS and opacity tables. Jets are initialized with a density of 2 × 10⁻⁸ g cm⁻³, velocity 8 × 10⁶ cm s⁻¹, and a 2.5 kG azimuthal magnetic field that peaks at the gun nozzle. J × B pinching and pressure imbalance accelerate the jets to an average speed of ≈1.25 × 10⁷ cm s⁻¹. Upon collision a quasi‑spherical target forms with a volume‑averaged density of 1.67 × 10⁻¹⁰ g cm⁻³, electron temperature ≈40 eV, and an electron Hall parameter ⟨χₑ⟩ > 5, while the ion Hall parameter remains < 1. The plasma β ≈ 100 indicates that the magnetic field is strong enough to reduce thermal conductivity without dominating the dynamics, satisfying a key PLX design goal.

For liner formation, the authors compare FLASH (fluid) and OSIRIS (kinetic) simulations of two high‑Z (xenon) jets, a proxy for the full 36‑gun array. FLASH inevitably produces a strong collisional shock at the jet interface, whereas OSIRIS, equipped with a relativistic binary‑Coulomb collision operator and a hybrid MHD‑PIC algorithm for dense regions, shows partial interpenetration and reduced shock strength. By varying initial jet density and impact angle, they map the transition from a fully collisional regime (plasma frequency ≫ collision frequency) to a semi‑collisionless regime (plasma frequency ≈ collision frequency). The collisional regime yields a dense, low‑temperature liner with minimal radiative loss, while the semi‑collisionless case suffers from kinetic mixing and lower compression efficiency. This comparative study clarifies the parameter space in which the liner behaves as a quasi‑fluid shell suitable for efficient target compression.

In the compression phase, a full 3‑D FLASH simulation of the 36‑gun xenon liner imploding onto the pre‑heated, magnetized target is presented. The liner converges at ≈5 × 10⁶ cm s⁻¹, generating a strong inward shock that compresses the target core to temperatures exceeding 1 keV and densities on the order of 10¹⁴ cm⁻³. Multi‑group radiation diffusion and tabulated opacities capture the dominant radiative cooling in the high‑Z liner, while anisotropic thermal conduction, governed by the Braginskii transport coefficients, continues to suppress heat loss from the core. Throughout compression the electron Hall parameter remains > 1, preserving magnetic insulation. The simulation demonstrates that, despite radiative losses in the liner, the high‑Z material’s large emissivity limits net energy loss, allowing the target to reach fusion‑relevant conditions.

HELIOS, a 1‑D Lagrangian code, is used as a reference to quantify numerical diffusion inherent in Eulerian FLASH runs. By preserving sharp material interfaces, HELIOS provides a baseline for assessing mixing arising from grid resolution versus physical instabilities such as Rayleigh‑Taylor.

Overall, the study validates that PLX can produce a pre‑heated (≈40 eV), magnetized (electron Hall > 1) target and a quasi‑collisional xenon liner capable of compressing the target to >1 keV and ≈10¹⁴ cm⁻³. The combined use of FLASH for macroscopic MHD and radiation, OSIRIS for kinetic jet‑merging physics, and HELIOS for interface fidelity offers a robust, cross‑validated modeling framework. The results provide concrete guidance for experimental parameter optimization—jet density, velocity, magnetic field strength, and timing—to achieve the desired fusion‑grade plasma in future PJMIF experiments.