488 IEEE TRANSACTIONS ON PLASMA SCIENCE, VOL. 30, NO. 2, APRIL 2002 Two-Dimensional MHD Simulations of a Neon Pinch on Hawk Joseph W. Schumer, Member, IEEE, David Mosher, Bryan Moosman, Bruce V. Weber, Robert J. Commisso, Senior Member, IEEE, Niansheng Qi, Member, IEEE, Jochen Schein, Member, IEEE, and Mahadevan Krishnan, Member, IEEE Abstract—Two-dimensional magnetohydrodynamic (MHD) simulations using MACH2 are benchmarked against laser shearing interferometer (LSI) images of the evolving electron-den- sity sheath during 250-ns neon gas-puff pinch implosions on the Naval Research Laboratory Hawk generator. The initial density distribution for the MHD simulations is calculated using a ballistic-flow-model fit to the measured initial gas-density distri- butions. The implosion is modeled using an applied current profile, single-temperature energy equations, and black-body-limited optically thin radiation. For consistency with the radiation model, neon ionization and equation-of-state models have been added to MACH2. Computed MHD ion-density distributions compare well with LSI images as the snowplowed plasma channel evolves during implosion. The MHD results also show features that may be helpful for understanding early and weak -shell radiation observed near the nozzle in Double-Eagle gas-puff experiments. Current-channel evolution derived from one-dimensional snow- plow calculations compare well to the MHD results for the same current history and initial density distribution, indicating that axial mass flow does not strongly impact the implosion dynamics. Index Terms—MHD simulations, neon gas puff, plasma radia- tion source, snow-plow implosions, pinch. I. INTRODUCTION I N RECENT years, a variety of multidimensional magneto- hydrodynamic (MHD) fluid simulations have been carried out to simulate the pinch implosion dynamics of high-atomic- number plasma radiation source (PRS) loads designed to pro- duce intense X-ray bursts at stagnation [9]–[17]. In most cases, MHD simulations are performed with analytic idealizations or approximations of the initial puff-gas or wire-array mass distri- butions. Perturbations of the initial mass distributions are then used to seed instabilities and more accurately match the exper- imentally observed current history, stagnation time, and X-ray signature. Although the MHD simulations have evolved with Manuscript received October 15, 2001; revised January 30, 2002. This work was supported by the Defense Threat Reduction Agency (DTRA) and the De- partment of Defense Common High performance computing Scalable Software Initiative (DoD CHSSI) (CEA-10). J. W. Schumer, D. Mosher, B. V. Weber, and R. J. Commisso are with the Plasma Physics Division, Naval Research Laboratory, Washington, DC 20375 USA (e-mail: schumer@calvin.nrl.navy.mil). B. Moosman was with the Plasma Physics Division, Naval Research Labora- tory, Washington, DC 20375 USA. He is now with Cymer, Inc., San Diego, CA 92127 USA. N. Qi, J. Schein, and M. Krishnan are with Alameda Applied Sciences Cor- poration, San Leandro, CA 94577 USA. Publisher Item Identifier S 0093-3813(02)05326-2. the use of improved physics and numerical models to more accurately match such stagnation-phase measurements, bench- marking algorithms and techniques have been restricted due to limited direct experimental diagnosis of the implosion-phase dynamics. Laser shearing interferometry (LSI) [2], [3] provides two- dimensional (2-D) images of the electron-density sheath during gas-puff pinch implosions. Coupled with the high repetition rate of the Naval Research Laboratory (NRL) Hawk generator [5], this technique has been used to document the detailed evolution of a neon gas-puff pinch at many times during a series of 250-ns implosions [2], [4]. LSI, therefore, provides a desirable dynamical benchmark for MHD codes that model such pinch implosions. In this paper, results from the 2-D MHD code MACH2 [1] are compared to these LSI measurements. The MHD simulations were initiated with the measured initial gas-density profile and driven by the measured pinch current, so that comparisons with the detailed time- and space-resolved LSI measurements provide a dynamical benchmark against which various code configurations, algorithms, and boundary conditions can be tested. With the code optimized by this process, computed MHD density distributions compare well with LSI images of plasma-channel boundaries [2], [4] during implosion. The MHD results also show features that may be helpful for understanding early and weak -shell radiation observed near the nozzle in Double-Eagle gas-puff experiments [8]. In addition, current-sheath evolution derived from one- dimensional (1-D) snowplow calculations compare well to the MHD results for the same current history and initial gas-density distribution, indicating that axial mass flow does not strongly impact the implosion dynamics for this experiment, and that simple snowplow modeling may be sufficient to address some PRS processes. A point of disagreement between measurements and the code is the ionization level in the imploding current channel: elec- tron density inferred from LSI images is about twice the electron density computed from the code results. It is argued that this dis- agreement is likely due to inadequate radiation modeling. Issues of energy conservation, treatment of boundaries, and the effects of resolution on Raleigh Taylor (RT) instabilities and stagnation are also addressed. It is believed that further code improvements based on LSI benchmarking and improved atomic physics can establish an MHD capability for assessing the radiation perfor- mance of various gas-puff nozzles and pinch drivers. 0093-3813/02$17.00 © 2002 IEEE