Penetrating Radiography of Imploding and Stagnating Beryllium Liners on the $Z$ Accelerator

The implosions of initially solid beryllium liners (tubes) have been imaged with penetrating radiography through to stagnation. These novel radiographic data reveal a high degree of azimuthal correlation in the evolving magneto-Rayleigh-Taylor structure at times just prior to (and during) stagnation, providing stringent constraints on the simulation tools used by the broader high energy density physics and inertial confinement fusion communities. To emphasize this point, comparisons to 2D and 3D radiation magnetohydrodynamics simulations are also presented. Both agreement and substantial disagreement have been found, depending on how the liner’s initial outer surface finish was modeled. The various models tested, and the physical implications of these models are discussed. These comparisons exemplify the importance of the experimental data obtained.

Figure 1

(a, b) Schematic illustrations of the two-frame monochromatic backlighter. The $Z$ Beamlet Laser (ZBL) [14] delivers two $\sim 1\mathrm{\text{−}}\mathrm{kJ}$, 527-nm beams to Mn targets, generating x rays. Quartz crystals ($22\underline{4}3$) select the $6151\pm 0.5\mathrm{\text{−}}\mathrm{eV}$ photons for imaging. (c) Half-section illustration of the load region. The $125\mathrm{\text{−}}\mu \mathrm{m}$-thick, 26-mm-diameter Be return current can is approximately uniformly transparent to the 6151-eV backlighter. The minor attenuation that it causes is corrected for by the radiograph normalization and gradient-correction processes discussed in the text. (d) Drive current and radiograph times (vertical lines) from each experiment and a reference implosion trajectory from a 1D simulation that used the ALEGRA radiation magnetohydrodynamics code [15]. Each current pulse was measured by four $\dot{B}$ probes located 6 cm from the cylindrical axis of symmetry [16].

Figure 2

(a) Radiographs from $Z$ experiments. The vertical dashed lines indicate the initial positions of the inner and outer liner surfaces (inner and outer radii of 2.89 and 3.47 mm, respectively). (b)–(e) Synthetic radiographs from radiation magnetohydrodynamics simulations using the 3D GORGON code [17] (b),(c) and the 2D LASNEX code [3] (d),(e).

Figure 3

Liner surface finish data. (a) Sample of surface height variation illustrating azimuthally correlated structure (striations) due to the single-point, diamond-turned fabrication process (the liners were not polished or further modified). (b) Power spectra for axially aligned wave vectors ($600\mathrm{\text{−}}\mu \mathrm{m}$ axial sample length). The liner surface finishes had a root-mean-square roughness of 100–250 nm.

Figure 4

Example volume density images from $Z$ experiments (a) and GORGON 3D simulations (b),(c). The density images in (a),(b) were generated by Abel inverting the corresponding radiographs of Figs. 2a, 2c. The images shown in (c) are cut-through slices taken directly from the GORGON simulation output.

Figure 5

Example volume density profiles obtained by Abel inverting the axially averaged transmission data. The uncertainties in the experiment profiles are less than $\pm 20\%$. Here the imploding liner has been compressed to about $2.5\times \mathrm{\text{solid}}$ density.

Figure 6

Inner, bubble, and spike radii (determined from the volume density reconstruction data) and reference trajectories from a 1D ALEGRA simulation. The horizontal error bars represent the $\pm 1\mathrm{\text{−}}\mathrm{ns}$ timing uncertainty of the overall system.

Figure 7

MRT characterization for an $\mathrm{AR}=6$ Be liner on $Z$. (a) MRT amplitude growth. (b) MRT wavelength growth. (c) Axially averaged liner $\rho R$ and $\Delta (\rho R)$ evolution.