Controlling Hydrodynamic Mixing in NIF Implosions

While it’s true that oil and water don’t mix—unless you add detergent—the same unfortunately isn’t true for the target capsule material and the fusion fuel in NIF inertial confinement fusion (ICF) experiments.

In fact, one of the most significant challenges to achieving ignition on NIF has been understanding and controlling the growth of mixing caused by hydrodynamic instability, a type of fluid dynamics, in the imploding fusion fuel as it’s compressed. Rayleigh-Taylor (RT) and Richtmyer–Meshkov (RM) instabilities can lower ICF capsule performance by substantially degrading the ablator’s ability to compress the fusion fuel and by mixing ablator material into the fuel.

The RT and RM instabilities occur when fluids of different densities, such as the material in the outer layers of an imploding NIF target capsule (the ablator) and the hydrogen fusion fuel inside the capsule, are accelerated together (RT) or shocked (RM) and begin to interpenetrate at their boundary. High levels of fuel-ablator mixing occurred in many of the implosions studied during the National Ignition Campaign (NIC), which failed to approach ignition conditions even though ignition-relevant implosion velocities were reached. Suspecting that the mixing might be due to capsule surface roughness, a multi-institutional Science of Ignition workshop recommended in 2012 that NIF conduct experiments with a roughened ablator and the full NIC drive pulse and compare the results against the predictions of simulations.

The result was the development of special target capsules with pre-imposed sinusoidal modulations, or “ripples,” with various wavelengths on the capsule surface for use in a series of hydrodynamic growth radiography (HGR) experiments. The experiments measured the growth of hydrodynamic instabilities as a function of the wavelength of the ripples. They tested instability growth for two ignition-relevant laser drives: the “low-foot” drive used during NIC, and the new “high-foot” pulse shape (see the February 2014 Photons & Fusion Newsletter). The HGR platform has been tested up to a convergence ratio of four (convergence ratio is the initial radius divided by the imploded radius).

The high-foot implosions have a higher adiabat (more internal energy, making for lower density at the same pressure) than their low-foot counterparts due to the higher initial laser pulse, and were predicted to be less prone to hydrodynamic instability. The experiments showed that unstable growth at the ablation front was indeed dramatically reduced in recent implosions with high-foot drives, which helps to improve the performance of layered deuterium-tritium (DT) implosions when compared to the previous low-foot experiments.

On Oct. 31, Don Cook, deputy administrator for Defense Programs at the National Nuclear Security Administration (NNSA), presented an NNSA Defense Programs Award of Excellence to the High-Foot Team. The award citation noted that the High-Foot Team used face-on HGR experiments to directly test the degree of ablation front stabilization gained from the high-foot pulse shape, and the implosions show low levels of inferred mix and excellent agreement with one- and two-dimensional implosion models without the aid of mix models.

Fabrication and metrology of the tiny capsules—about the size of a peppercorn—with the high-quality sinusoidal ripples required by the experiments, and design of the targets to allow radiography of the ripples, were significant research and development challenges for the target fabrication teams at General Atomics (GA) of San Diego and LLNL. The GA target fab team used diamond-turning techniques for producing the modulations on plastic shells, and laser ablation for high-density carbon (diamond) shells. The LLNL team developed the design of the target and a cryogenic cone-on-shell assembly.

LLNL lead engineer Jeremy Kroll said the LLNL team was able to leverage many existing target assembly techniques, “but HGR created at least three unique challenges. First, the HGR capsule modulations needed to be aligned to within plus-or-minus five-tenths of a degree of the hohlraum axis. This required development of tooling and techniques for assembly under an optical coordinate measuring machine.

“Second,” he said, “ HGR also required the inner surface of the gold cone to be coated with plastic to prevent gold from ablating into the radiography line-of-sight, which would blank the image. This required significant process and tooling development. And finally, when assembling the capsule onto the gold cone, we needed to develop tooling to hold the capsule in place during gluing without damaging the capsule modulations.”

Along with validating model predictions of single mode instability growth and testing the effects of different pulse shapes, the HGR experiments are being used to study the role of capsule surface roughness, target features such as the capsule support “tent” inside the hohlraum—which had been suspected of contributing to hydrodynamic instabilities—and the stability aspects of alternate ablators such as diamond and beryllium.

“HGR was initiated as a high energy density (HED) experiment to validate models of ICF instability growth,” said physicist Kumar Raman of LLNL’s Weapons and Complex Integration (WCI) directorate, who led the design effort during development of the HGR platform. “Before the HGR experiments, there was considerable uncertainty about the ability of our models to predict instability growth in this new physical regime. Therefore, much of the design effort was spent in making the platform robust enough to accommodate scenarios where the design models differed significantly from reality.

“HGR was part of the same WCI-HED ignition science effort that spawned the high-foot campaign and viewfactor and alternate ablator platforms,” he said, “and it involved a close collaboration between WCI and NIF&PS. The experiment was developed within the mix campaign of the ICF program and, in Fiscal Year 2014, transitioned to being an ICF experiment. It is currently being used by the ICF program to investigate a variety of issues that will impact future ignition designs.

“The experiments have provided new insights into a number of modeling aspects that impact our ability to predict instability growth,” he added, “including the way in which shock timing and viewfactor data are incorporated into the simulations. The results so far have largely validated the tuned drive approach to modeling instability growth used in state-of-the-art ICF calculations.”

In the experiments, 184 of the 192 NIF laser beams are focused on the hohlraum walls, generating soft x-rays which drive the capsule, with peak drive corresponding to radiation temperatures around 300 electron volts (eV). The remaining eight beams are focused on a 12.5-micron-thick vanadium foil outside the hohlraum to generate 5.4-keV backlighter x rays which radiograph the capsule in flight. The backlighter x rays enter the hohlraum through the cone, pass through the rippled capsule surface, and exit the hohlraum through a window toward a gated x-ray camera.

LLNL’s Luc Peterson, who took over the design work for the HGR experiments at the beginning of FY14, said that while the implosion models do an adequate job of simulating the hydrodynamic growth of known perturbations, the researchers are now extending the measurement “to look at capsules with potentially unknown ‘native roughness’ perturbations. We are also using the HGR platform to test theories on how to control growth in next-generation implosion designs,” he said. “As we push toward ignition, it will be important to keep hydrodynamic instabilities in check; the HGR experiments help us stay on the right track.”

“One remarkable feature of the results is the experimental confirmation that the instability growth is close to what the codes predict,” added LLNL researcher Steve Haan. “While these results do confirm the expected reduction in growth with the high-foot pulse, they do not explain why the low-foot shots behaved as if they had larger-than expected growth. Exploring that puzzle is a top priority for future work.”

Participating in the HGR experiments along with Raman, Peterson and Haan were LLNL colleagues Vladimir Smalyuk, Dan Casey, Omar Hurricane, Bruce Remington, Harry Robey, Dan Clark, Bruce Hammel, Nino Landen, Marty Marinak, David Munro, and Jay Salmonson; and Kyle Peterson of Sandia National Laboratories. The GA target fabrication team included Martin Hoppe, Jr., Anna Nguyen, Neal Rice, Lane Carlson, Denise Hoover and Abbas Nikroo. The LLNL team was composed of Lucian Sweitzer, Steve Andrews and Jack Reynolds, who developed the cone plastic coating process; Jean Jensen and Elias Piceno, who developed the capsule loading fixture process and the modulation alignment process; and the engineering team of Kroll, Dave Barker, Suhas Bhandarkar, Brian Yoxall and Nick Hash.

—Follow us on Twitter: @lasers_llnl