Session JO7: Hydrodynamic Instability I

Three-Dimensional Simulations of Flat-Foil Laser-Imprint Experiments at the National Ignition Facility

Control of shell nonuniformities imprinted by the laser and amplified by hydrodynamic instabilities in the imploding target is critical for the success of direct-drive ignition at the National Ignition Facility (NIF). To measure a level of imprint and its reduction by the NIF smoothing by spectral dispersion (SSD), we performed experiments that employed flat CH foils driven with a single NIF beam with either no SSD or the NIF indirect-drive SSD applied to the laser pulse. Face-on x-ray radiography was used to measure optical depth variations, from which the amplitudes of the foil areal-density modulations were obtained. Results of 3-D, radiation--hydrodynamic code \textit{HYDRA}\footnote{ M. M. Marinak \textit{et al.}, Phys. Plasmas \textbf{8}, 2275 (2001).} simulations of the growth of the imprint-seeded perturbations are presented and compared with the experimental data. This work was supported by the U.S. Department of Energy National Nuclear Security Administration under Award Number DE-NA0001944 and under the auspices of the Lawrence Livermore National Security, LLC, (LLNS) under Contract Number DE-AC52-07NA27344.   

P2 Asymmetry of Au's M-band Flux and its smoothing effect due to high-Z ablator dopants

X-ray drive asymmetry is one of the main seeds of low-mode implosion asymmetry that blocks further improvement of the nuclear performance of "high-foot" experiments on the National Ignition Facility [Miller et al., Nucl. Fusion 44, S228(2004)]. More particularly, the P2 asymmetry of Au's M-band flux can also severely influence the implosion performance [Li et al., Phys. Plasmas 23,072705(2016)]. Here we study the smoothing effect of mid- and/or high-Z dopants in ablator on M-band flux asymmetries, by modeling and comparing the implosion processes of a Ge-doped and a Si-doped ignition capsule driven by x-ray sources with asymmetric M-band flux. As the results, (1) mid- or high-Z dopants absorb M-band flux and re-emit isotropically, helping to smooth M-band flux arriving at the ablation front, therefore reducing the P2 asymmetries of the imploding shell and hot spot; (2) the smoothing effect of Ge-dopant is more remarkable than Si-dopant due to its higher opacity than the latter in Au's M-band; and (3) placing the doped layer at a larger radius in ablator is more efficient. Applying this effect may not be a main measure to reduce the low-mode implosion asymmetry, but might be of significance in some critical situations such as Inertial Confinement Fusion (ICF) experiments very near the performance cliffs of asymmetric x-ray drives.   

3D broadband Bubbles Dynamics for the imprinted ablative Rayleigh-Taylor Instability

We report on highly nonlinear ablative Rayleigh-Taylor growth measurements of 3D laser imprinted modulations. These experiments are part of the Discovery Science Program on NIF. Planar plastic samples were irradiated by 450 kJ of 3w laser light and the growth of 3D laser imprinted modulations is quantified through face-on radiography. The initial seed of the imprinted RTI is imposed by one beam focused in advance (-300 ps) without any optical smoothing (no CPP, no SSD). For the first time four generations of bubbles were created as larger bubbles overtake and merge with smaller bubbles because of the unprecedented long laser drive (30 ns). The experimental data, analyzed both in real and Fourier space, are compared with classical bubble-merger models [1], as well as recent theory and simulations predicting 3D bubbles reacceleration due to vorticity accumulation caused by mass ablation [2]. These experiments are of crucial importance for benchmarking 2D and 3D radiation hydrodynamics code for Inertial Confinement Fusion. [1] D. Oron et al., Phys. Plasmas 8\textbf{, }2883 (2001). [2] R. Yan et al., Phys. Plasmas 23, 022701 (2016).   

Signatures of Intermediate-Mode Asymmetries in OMEGA Implosions

On the OMEGA laser, the 60-beam port geometry creates intermediate-mode asymmetries in the illumination pattern\footnote{ S. Skupsky and K. Lee, J. Appl. Phys.~\textbf{54}, 3662 (1983).} that can potentially degrade implosion performance. Recently, some x-ray images of Ge- and Cu-doped shell implosions have exhibited structures that could be related to these mid-mode nonuniformities.~These images are processed to emphasize those structures using a feature in the detection algorithm that subtracts the uniform background from the image and removes the high-frequency noise.~The hydrodynamic code \textit{DEC3D} is being used to determine whether some of those features are produced during the deceleration or disassembly phase of the implosion.~ The goal of the work is to develop a method to identify and measure the magnitude of the mid-mode asymmetry resulting from the laser-port geometry and to assess its impact on the implosion performance. This material is based upon work supported by the Department of Energy National Nuclear Security Administration under Award Number DE-NA0001944.   

Nonlinear Excitation of the Ablative Rayleigh--Taylor Instability for All Wave Numbers

Small-scale modes of the ablative Rayleigh--Taylor instability (ARTI) are often neglected because they are linearly stable when their wavelength is shorter than a linear cutoff. Using 2-D and 3-D numerical simulations, it is shown that linearly stable modes of any wavelength can be destabilized. This instability regime requires finite amplitude initial perturbations. Compared to 2-D, linearly stable ARTI modes are more easily destabilized in 3-D and the penetrating bubbles have a higher density because of enhanced vorticity. It is shown that for conditions found in laser fusion targets, short-wavelength ARTI modes are more efficient at driving mixing of ablated material throughout the target since the nonlinear bubble density increases with the wave number and small-scale bubbles carry a larger mass flux of mixed material. This work was supported by the Office of Fusion Energy Sciences Nos. DE-FG02-04ER54789, DE-SC0014318, the Department of Energy National Nuclear Security Administration under Award No. DE-NA0001944, the Ministerio de Ciencia e Innovacion of Spain (Grant No. ENE2011-28489), and the NANL LDRD program through project number 20150568ER.   

Finite Atwood Number Effects on Deceleration-Phase Instability in Room-Temperature Direct-Drive Implosions

Performance degradation in direct-drive inertial confinement fusion implosions can be caused by several effects, one of which is Rayleigh--Taylor (RT) instability growth during the deceleration phase. In room-temperature plastic target implosions, this deceleration-phase RT growth is enhanced by the density discontinuity and finite Atwood numbers at the fuel--pusher interface. For the first time, an experimental campaign at the Omega Laser Facility systematically varied the ratio of deuterium-to-tritium (D-to-T) within the DT gas fill to change the Atwood number. The goal of the experiment was to understand the effects of Atwood number variation on observables like apparent ion temperature, yield, and variations in areal density and bulk fluid motion, which lead to broadening of neutron spectra along different lines of sight. Simulations by the hydrodynamic codes \textit{LILAC} and \textit{DRACO} were used to study growth rates for different D-to-T ratios and identify observable quantities effected by Atwood number variation. Results from simulations and the experiment are presented. This material is based upon work supported by the Department of Energy National Nuclear Security Administration under Award Number DE-NA0001944.   

Abstract Withdrawn

We use detailed continuum hydrodynamics and molecular dynamics simulations to characterize bubble and spike growth in shock-driven ejecta. Insights from the simulations are used to suggest a modified expression for the velocity associated with ejected spike structures, while a recently proposed model [1] explains the observed bubble velocities. For spikes, existing models [2] can overpredict observed spike velocities if they do not include the modification of the initial spike growth rates due to nonlinearities. Instead, we find that using the potential flow model of [2], corrected with a suitable nonlinear prefactor leads to predictions in close agreement with our simulation data. We propose a simple empirical expression for the nonlinear correction for spike velocities that is able to reproduce results from our simulations and published experimental and simulation data over a wide range of initial conditions and Mach numbers. We verify these ideas with simulations (continuum and MD) at different amplitudes, initial perturbation shapes, and shock strength. This work was supported by the Los Alamos National Laboratory. [1] K. O. Mikaelian, Phys Rev Lett 80, 508 (1998). [2] Q. Zhang, Phys Rev Lett 81, 3391 (1998).   

Experimental platform for shock-driven Rayleigh-Taylor / Richtmyer-Meshkov evolution before and after re-shock

The growth of Richmyer-Meshkov and Rayleigh-Taylor instabilities at an interface that is impulsively accelerated, for example by the passage of a shock, have been studied in many laser-driven experiments. However, investigation of instability growth subject to a second shock (``reshock'') has to date been limited to ``classical'' (non-high-energy-density) shock tubes. Here we describe the results of experiments, performed on the National Ignition Facility, to directly measure the growth \textit{vs.} time of the non-linear instability at a planar interface before and after reshock. In this work the unstable mixing region is directly imaged with side-on x-ray radiography, and we highlight the unique advantages of laser-driven experiments over classical shock tubes. These include precise control over the initial conditions of the instability, as well as tailored x-ray opacity to ensure accurate measurement of the entire region of material interpenetration.   

Shock-driven Rayleigh-Taylor / Richtmyer-Meshkov 2D multimode ripple evolution before and after re-shock

Laser-driven hydrodynamic experiments allow for the precise control over several important experimental parameters, including the timing of the laser irradiation delivered and the initial conditions of the laser-driven target. Our experimental platform at the National Ignition Facility enables the investigation of the physics of instability growth after the passage of a second shock (``reshock"). This is done by varying the laser to change the strength and timing of the secondary shock. Here we present x-ray images capturing the rapid post-reshock instability growth for a set of reshock strengths. The radiation hydrodynamics simulations used to design these experiments are also introduced.   

Computational investigation of reshock strength in hydrodynamic instability growth at the National Ignition Facility

Experiments at the National Ignition Facility (NIF) are studying Richtmyer-Meshkov and Rayleigh-Taylor hydrodynamic instabilities in multiply-shocked plasmas. Targets feature two different-density fluids with a multimode initial perturbation at the interface, which is struck by two X-ray-driven shock waves. Here we discuss computational hydrodynamics simulations investigating the effect of second-shock (``reshock'') strength on instability growth, and how these simulations are informing target design for the ongoing experimental campaign. A Reynolds-Averaged Navier Stokes (RANS) model was used to predict motion of the spike and bubble fronts and the mixing-layer width. In addition to reshock strength, the reshock ablator thickness and the total length of the target were varied; all three parameters were found to be important for target design, particularly for ameliorating undesirable reflected shocks. The RANS data are compared to theoretical models that predict multimode instability growth proportional to the shock-induced change in interface velocity, and to currently-available data from the NIF experiments.   

3D Simulations of NIF Wetted Foam Experiments to Understand the Transition from 2D to 3D Implosion Behavior

The high convergence ratio (CR) of layered Inertial Confinement Fusion capsule implosions contribute to high performance in 1D simulations yet make them more susceptible to hydrodynamic instabilities, contributing to the development of 3D flows$^1$. The wetted foam platform is an approach to hot spot ignition to achieve low-to-moderate convergence ratios in layered implosions on the NIF$^2$ unobtainable using an ice layer. Detailed high-resolution modeling of these experiments in 2D and 3D, including all known asymmetries, demonstrates that 2D hydrodynamics explain capsule performance at CR 12 but become less suitable as the CR increases. Mechanisms for this behavior and detailed comparisons of simulations to experiments on NIF will be presented. To evaluate the tradeoff between increased instability and improved 1D performance, we present a full-scale wetted foam capsule design$^3$ with 17$<$CR$<$42 and evaluate its sensitivities to asymmetries. Simulations predict that, given currently achievable levels of asymmetry, their effects negate all advantages of increased CR. $^1$B.~M.~Haines et al. Phys.~Plasmas 23:072709, 2016 $^2$R.~E.~Olson et al. Phys.~Rev.~Lett. 117(24):245001, 2016 $^3$B.~M.~Haines et al. Phys.~Plasmas, 24:0727   

Depth and Extent of Gas-Ablator Mix in Symcap Implosions at the National Ignition Facility

A longstanding question in ICF physics has been the extent to which capsule ablator material mixes into the burning fusion fuel and degrades performance. Several recent campaigns at the National Ignition Facility have examined this question through the use of separated reactants. A layer of CD plastic is placed on the inner surface of the CH shell and the shell is filled with a gas mixture of H and T. This allows for simultaneous neutron signals that inform different aspects of the physics; we get core TT neutron yield, atomic mix from the DT neutrons, and information about shell heating from the DD neutron signal. By systematically recessing the CD layer away from the gas boundary we gain an inference of the depth of the mixing layer. This presentation will cover three campaigns to look at mixing depth: An ignition-like design (“Low-foot”) at two convergence ratios, as well as a robust, nearly one-dimensional, low convergence, symmetric platform designed to minimize ablation front feed-through (HED 2-shock). We show that the 2-shock capsule has less ablator-gas mix, and compare the experimental results to mix-model simulations.   

Mix Models Applied to the Pushered Single Shell Capsules Fired on NIF1

The goal of the Pushered Single Shell (PSS) experimental campaign is to study the mix of partially ionized ablator material into the hotspot. To accomplish this goal, we used a uniformly Si doped plastic capsule based on the successful Two-Shock campaign [1]. The inner few microns of the capsule can be doped with a few percent Ge. To diagnose mix, we used the method of separated reactants [2]; deuterating the inner Ge-doped layer, CD/Ge, while using a gas fill of Tritium and Hydrogen. Mix is inferred by measuring the neutron yields from DD, DT, and TT reactions. The PSS implosion is fast (\textasciitilde 400 km/sec), hot (\textasciitilde 3KeV) and round (P2 \textasciitilde 0). This paper will present the calculations of RANS type mix models such as KL along with LES models such as multicomponent Navier Stokes on several PSS shots. The calculations will be compared to each other and to the measured data. [1] S.F. Khan et al. Physics of Plasmas \textbf{23}, 042708 (2016) [2] D.C. Wilson et al. Physics of Plasmas~\textbf{18}, 112707 (2011)   

Ultra High Mode Mix in NIF NIC Implosions

This work re-examines a sub-set of the low adiabat implosions from the National Ignition Campaign [1] in an effort to better understand potential phenomenological sources of ‘excess’ mix observed experimentally. An extensive effort has been made to match both shock-timing and backlit radiography (Con-A)[2] implosion data in an effort to reproduce the experimental conditions as accurately as possible. Notably a ~30% reduction in ablation pressure at peak drive is required to match the experimental data. The reduced ablation pressure required to match the experimental data allows the ablator to decompress, in turn causing the DT ice-ablator interface to go Rayleigh-Taylor unstable early in the implosion acceleration phase. Post-processing the runs with various mix models indicates high-mode mix from the DT ice-ablator interface may penetrate deep into the hotspot. This work offers a potential explanation of why these low-adiabat implosions exhibited significantly higher levels of mix than expected from high-fidelity multi-dimensional simulations. Through this new understanding, a possible route forward for low-adiabat implosions on NIF is suggested. References [1] Lindl et al., Physics of Plasmas 21, 020501 (2014) [2] Hicks et al., Physics of Plasma s19, 122702 (2012)