Laboratory for Laser Energetics targets inertial confinement fusion

The Laboratory for Laser Energetics (LLE; Rochester, NY) is home to two extremely powerful lasers—OMEGA and OMEGA EP—and researchers are using them to explore laser-driven inertial confinement fusion (ICF; see video).

ICF involves compressing a small amount of fuel consisting of hydrogen isotopes, deuterium (D), and tritium (T), and heating it to temperatures greater than the center of stars.

“When these conditions are reached, the fuel undergoes fusion—releasing enormous energy that can be used for research relevant to the National Nuclear Security Administration’s (NNSA) Stockpile Stewardship Program (SSP) and to drive carbon-free power plants,” explains Valeri Goncharov, distinguished scientist and director of the Theory Division at LLE.

LLE was established at the University of Rochester in 1970 and is the largest U.S. Department of Energy university-based research program in the nation, supported by the National Nuclear Security Administration as part of its Stockpile Stewardship Program (SSP).

“As a center for exploring the interaction of intense radiation with matter, LLE is a unique national resource for research and education in science and technology,” says Goncharov. “Our current research includes exploring fusion for the SSP program and as a future source of energy, developing new laser and materials technologies, and pursuing a better understanding of high-energy-density phenomena.”

Laser-driven inertial confinement fusion

ICF experiments are a key focus area at LLE, and OMEGA (see Figs. 1 and 2) conducts approximately 10 experiments per day—compared to 1 or 2 per day on the National Ignition Facility (NIF) at Lawrence Livermore National Lab.

The total amount of energy delivered by the laser for a single experiment isn’t enormous: 30 kJ on OMEGA or 2 MJ on NIF. “With 30 kJ of energy, you can only heat 3 ounces of water from room temperature up to boiling point,” says Goncharov. “But in ICF, space and time are shrunk so much that even this modest amount of energy leads to creating exotic conditions of matter.”

An ICF implosion lasts only a few nanoseconds and, for a successful ignition experiment, plasma needs to be kept at ignition conditions for a tenth of a nanosecond. To help visualize it: light travels only a foot during a nanosecond.

Laser energy is also delivered to a tiny space, and the targets on OMEGA are less than a millimeter in size (see Fig. 3). “Laser energy is delivered to a very small volume during an extremely short time and puts the matter into a very high-energy-density regime (similar to high mass density, if mass is squeezed into a very tiny volume),” Goncharov says. “Under HED conditions, we create pressures of above 1 million atmospheric pressure (atm). For comparison, pressure inside the Earth is 3 million atm and diamonds are formed under 0.5 million atm. Material under HED conditions is relevant to planetary science, the conditions achieved inside the stars.”

FIGURE 3. An image inside the OMEGA target chamber during a target shot.

During an ICF implosion, when the laser interacts with target material, pressures of 100s million atm are created by the same mechanism as the flow of burnt fuel that generates thrust in a rocket: a thin layer of target material, heated by laser, ablates off and its blow-off flow creates a push that accelerates the rest of the target inward. As the target implodes, spherical convergence amplifies target pressure to 100s billion atm and creates fusion conditions—at a temperature of several hundred million degrees and several hundred billion atm of pressure.

The quest for ignition

On December 5, 2022, NIF achieved ignition for the first time in a laboratory—by generating more fusion energy (3.15 MJ) than laser energy (2.05 MJ) used during an experiment to compress DT fuel.

“This milestone demonstrates our understanding of how matter behaves at HED conditions and, although not complete, has advanced to the point where we can design such complex experiments and achieve fusion conditions that require high precision and accuracy in modeling, laser energy delivery, target fabrication, and data acquisition and analysis,” says Goncharov.

From Goncharov’s point of view, “there’s always a question of whether our knowledge of ICF implosions is complete enough to use lasers to create 100 million atm of pressure to compress DT fuel— reducing its volume by a factor of 10,000—and not introduce any excessive deviations from spherical symmetry (implosion needs to be highly spherically symmetric for effective fuel compression) during the process of compression, and keep the fuel cold enough so its density is maximized,” he says.

It’s no secret NIF was struggling to achieve ignition, and many people who work within the field of HED/ICF had begun to question whether lab ignition is possible at all.

“We were desperately searching for the answers, exploring how physics—not included within the current modeling—could explain the reasons for ignition failure,” says Goncharov. “Based on the data, one thing was clear: energy coupled to the fuel in earlier implosions wasn’t enough to create a ‘fusion spark’ at the target center to ignite the rest of the fuel. Hard work and the ingenuity of LLNL scientists—with many contributions from the entire HED community—led to design modifications that led to higher fuel energy and, ultimately, to ignition.”

LLE and NIF use similar targets—a layer of frozen DT overcoated with a layer of plastic—but differ in the way laser energy is delivered to the targets.

“In NIF implosions, the target is placed inside a high-Z (gold) enclosure called a hohlraum,” explains Goncharov. “A laser is pointed on the inside walls, which get heated and emit x-rays. It’s like an oven where you bake a casserole by heat emitted from the oven’s walls. In the same way, a target surface is heated inside a hohlraum by x-rays.”

When the surface material blows off, it creates a rocket effect that drives the rest of the target inward. “Since laser energy gets to the target indirectly, it’s called ‘indirect drive.’ At LLE, we use the direct drive approach, which involves using a laser to heat up the target surface directly,” Goncharov says. “Instead of an oven, it’s akin to setting a pot on a gas stove. The laser heats up the target surface and the rest of the drive process proceeds the same way as indirect drive.”

Indirect drive vs. direct drive

There are pros and cons to both indirect drive and direct drive. Indirect drive is less efficient (less laser energy gets to the fuel), but target compression is more symmetric. LLE has developed several techniques to overcome nonuniform heating by the laser beams—earlier disadvantages of direct drive are largely resolved.

“Direct drive still has a major flaw: when a laser travels through blow-off plasma to reach the target, a high level of electron and ion waves is excited,” says Goncharov. “These waves scatter away laser light before reaching the target region where it can get efficiently absorbed. This was an a-ha! moment on OMEGA, because we were struggling to understand why ablation pressure generated in direct-drive implosions by the rocket effect was below expectations based on model predictions.”

To understand the issue, the researchers measure the color (frequency) of laser light scattered from the target. “Laser light is nearly monochromatic—it only has one wavelength, as opposed to everyday white light that has many different wavelengths visible in rainbows,” says Goncharov. “But when light, as any other wave, reflects from a moving surface, its wavelength shifts due to the Doppler effect. It’s akin to hearing different pitches from the sound of a siren based on whether the emergency vehicle is moving toward or away from you. Laser wavelength also gets slightly longer when it reflects from an imploding target surface—moving away from the source.”

By measuring the wavelength of reflected laser light, they saw a smaller-than-expected shift from a moving target. “The only explanation is that the laser light doesn’t penetrate deeply enough and reflects earlier prior to reaching the target surface,” Goncharov says. “It’s reflected from ion waves that were launched into the plasma by the laser itself. Since two intersecting beams are needed to create a reflective wave pattern, this phenomenon is a cross-beam energy transfer (CBET). It’s responsible for 30 to 40% of light reflection from the target and a significant reduction in rocket effect and ablation pressure.”

This is both good and bad. Good, because the researchers finally figured out why their implosions weren’t efficient. And bad, because it was the nature of the direct drive—the direct interaction of laser light with a target plasma—causing this drive deficiency. What made this discovery even worse is the level of coupling losses scales with the target size.

If a bigger laser facility like NIF performs direct-drive implosions, CBET losses will be larger there and prevent efficient laser energy coupling to DT fuel—and it makes ignition a very difficult task for the direct-drive approach.

“A solution seemed trivial: break a big ion wave into a bunch of smaller waves,” Goncharov says. “As with waves on a water surface, they’ll have a regular structure if the source is regular in time—like if you throw stones into the water at regular intervals.”

When a monochromatic—single wavelength—laser light travels through the plasma, certain regions in the plasma (resonance region) can have regular-pattern ion waves that get excited. “What happens if I use a laser with two slightly different wavelengths? Two different waves with slightly different frequencies will be created—each with smaller amplitude and different locations,” points out Goncharov.

And what happens if you repeat the process of introducing more and more laser wavelengths and break the laser beams into several smaller beams with different frequencies? “I will totally destroy the refractive ion wave pattern and make the wave amplitude so small and spread out in space that these waves won’t be able to scatter the laser light anymore,” Goncharov says.

This is the main idea of fighting CBET, and another recent a-ha! moment that got Goncharov and colleagues very excited about the future of direct drive.

If they can demonstrate CBET is defeated by using broadband lasers (laser light that consists of several frequencies separated by a few present), “the advantage of the laser drive approach will be enormous: nearly 100% of laser light can be coupled to the target and ablation pressure can be increased to several hundred million atm,” Goncharov explains. “With such a highly efficient laser, direct drive is predicted to achieve fusion yields a factor of 50 to 100 more than currently possible with indirect drive on NIF using the same laser energy.”

Computing plays huge role in ICF experiments

It's easy to imagine matter under HED conditions behaves quite differently than our everyday experiences suggest. To model ICF experiments, researchers rely on a different type of computational tools or numerical codes.

“One set of codes model material properties such as compressibility (by how much material volume changes by applying pressures of million atm), heat conductivity (if one point has higher temperature than the other, how fast temperature will equilibrate), viscosity, etc.,” says Goncharov.

LLE also has codes to model the interaction of laser light with plasma. “It’s a complicated process because under HED conditions and temperatures of millions of degrees, electrons are separated from ions in atoms—forming a new state of matter: plasma,” Goncharov says. “Both electrons and ions are charged particles, so they respond to electromagnetic fields like the way Earth’s magnetic field deflects charged particles that stream at us from the universe.”

Laser light is an electromagnetic wave, so the amplitude of electric and magnetic field carried by this wave oscillate in time. Electrons respond to such field oscillation in two ways: First, they absorb some of light energy. Second, they oscillate and create plasma waves. Ions can oscillate and form waves as well.

Laser light’s absorption by electrons is beneficial for ICF because it represents the main mechanisms by which laser light couples its energy into plasma.

“The second process, forming electron and ion waves, is damaging for ICF because these waves can scatter light away from the target and accelerate plasma electrons to very large energies,” says Goncharov. “These energetic electrons—a.k.a. hot electrons—can travel all the way to the main target and preheat deuterium-tritium fuel, which makes the fuel very difficult to compress and reach ignition conditions.”

To model these complex details of laser-plasma interaction, LLE has special codes that include relevant plasma physics. “These physics-specific codes for material properties and laser-plasma interaction are too complex and computationally demanding to simulate the entire implosion process,” Goncharov says. “We simply don’t have enough computational resources in the world to simulate full-scale target evolution using codes that describe physics at the micro-scale (atomic) level. Even the world’s most advanced exascale computers can’t do it right now or will be able to within the foreseeable future.”

Instead, LLE researchers approximate plasma as electron and ion fluids because their temperatures and other material properties are different. Then they use hydrodynamic codes to solve fluid continuity equations that originate from Leonhard Euler’s work in the 1750s.

“Since these equations have been around hundreds of years, people have developed numerous techniques to solve them using big computers—although, after all these years of hard work, we still can’t claim we have a perfect algorithm to solve every hydrodynamic problem. Each problem requires its own code and technique,” says Goncharov.

Why do these old equations still work for such extreme conditions? Because they’re built upon the basic principles of matter: mass, momentum, and energy must be conserved.

How does modern knowledge about complex plasma physics phenomena and exotic materials properties mesh with century-old equations? “One problem with hydrodynamic equations is they don’t represent a complete set; there are more unknowns than the number of equations,” explains Goncharov. “Back in Euler’s day, it wasn’t a big problem for everyday conditions—pressures of 1 atm, room to boiling water temperatures. People could calculate material densities by simply weighing the mass and measuring its volume. Nearly every material could be treated as incompressible (density doesn’t depend on pressure).”

But as people began to build airplanes and high-pressure presses, they realized material density isn’t a constant anymore. It could change at super-high-flow velocities reached by airplane or if applying pressures much higher than atmospheric. This relationship between material density, temperature, and pressure is the equation of state (EOS), which can’t be derived using a hydrodynamic equation. It comes from the knowledge of material properties at the atomic level.

Fast-forward to today and into the world of high-energy-density physics, “using complex models of material properties and experiments on OMEGA and NIF helps us come up with EOS for these exotic conditions,” says Goncharov. “We use modern models in our hydro codes to accurately describe HED phenomena like ICF implosions. Not only EOS, but other material properties like heat conduction, viscosity, radiative properties, and transport require modern modeling tools based on plasma physics and atomic physics.”

Using these modern modeling tools requires modern supercomputers like Conesus, which was built by Dell Technologies and named after one of the Finger Lakes in upstate New York. The 43,008-core machine, acquired recently by LLE, was ranked 311 on the June 2023 Top 500 list at 2.59 petaflops. Conesus also ranked 77th on the June 2023 Green 500 list of most energy-efficient machines.

A broadband laser beam prototype

What’s next? Building a broadband laser beam prototype at LLE to be used on OMEGA to demonstrate CBET is mitigated (see Fig. 4). “We call this laser beam ‘FLUX,’ which is short for Fourth-Generation Laser for Ultra-Broadband Experiments,” says Goncharov.

FIGURE 4. A diode-pumped crystal large-aperture ring amplifier system, based on an architecture previously developed to pump optical parametric chirped-pulse amplification stages in the MTW and OMEGA EP lasers, delivers 1.5-J pulses at 5 Hz to pump the first two broadband amplifiers on the Fourth-generation Laser for Ultra-broadband eXperiments (FLUX).

These experiments will be conducted during 2024 and, if successful, “we’ll put together a proposal for a next-generation broadband implosion facility,” Goncharov adds.

Although much smaller than NIF, the facility’s main goal is to demonstrate that all direct-drive deficiencies due to laser-plasma interactions can be eliminated via a spherical implosion and that laser technology is ready for an at-scale implosion facility to ultimately deliver high fusion yields (>100 MJ) for the Stockpile’s needs and a future inertial fusion energy power plant.

FURTHER READING

1. H. Abu-Shawareb et al., Phys. Rev. Lett., 132, 065102 (Feb. 5, 2024); https://doi.org/10.1103/PhysRevLett.132.065102.

2. R. S. Craxton et al., Phys. Plasmas, 22, 110501 (2015); https://doi.org/10.1063/1.4934714.