Big Ideas Lab - NIF Diagnostics

Episode Date: August 18, 2026

Inside the National Ignition Facility, scientists create bursts of fusion hotter than the center of the sun that vanish in 100 trillionths of a second. The only way to trust what happened is to measure it - and Lawrence Livermore has spent decades building the instruments that can see the unseeable, the same instruments that proved the world's first fusion ignition was real. This is the story of NIF's diagnostics: how Lawrence Livermore learned to photograph a dying star, prove a nuclear stockpile still works without ever testing it, and turn an event too fast, too hot, and too dense to touch into knowledge. Guests featured (in order of appearance): Dave Schlossberg - Experimental Physicist, National Ignition Facility, LLNL Cara Vennari - Experimental Physicist, LLNL -- Big Ideas Lab is a Mission.org original series. Executive Produced by Levi Hanusch. Sound Design, Music Edit and Mix by Matthew Powell. Script by Caroline Kidd Story Editing by Levi Hanusch. Audio Engineering and Editing by Matthew Powell. Narrated by Matthew Powell. Brought to you in partnership with Lawrence Livermore National Laboratory. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

December 5th, 2022, 127 a.m. The house is dark and quiet. Dave Schlossberg is asleep at home. Hours earlier, he'd told the scientist running the overnight experiment to call if there's anything to report. And then, the phone rings. Yeah. Not like two in the morning, Alex calls me up and says, hey, Dave, I think something injured to happen. Dave jumps off the bed and grabs his computer,

pulling up signals from instruments surrounding the target chamber at Lawrence Livermore's National Ignition Facility. And as the numbers resolve, he sees something extraordinary, something they'd chased for 50 years. We started looking at the data and said, you know, I think this is actually ignition. This is more energy out than we used to start the reaction. But the shot is already over.

The target is gone. The chamber is dark again and no one saw fusion ignition happen. Only the signals remain. At NIF, Lawrence Livermore's diagnostics turn an invisible instant into evidence that can shape the next experiment. And strengthened confidence in the nuclear stockpile. Welcome to the Big Ideas Lab, your exploration inside Lawrence Livermore National Laboratory. Hear untold stories meet boundary-pushing pioneers and get unparalleled access inside the gates. From national security challenges to computing revolutions,

discover the innovations that are shaping tomorrow today. Inside Lawrence Livermore's National Ignition Facility, 192 laser beams race toward a target smaller than a peppercorn and arrive almost simultaneously. Then in a fraction of a second, it's gone. No one can look inside while the lasers fire. The event is too fast, too hot, and too dense for anyone to observe directly. But understanding exactly what happens is essential. So NIF relies on diagnostics.

Dave Schlossberg is an experimental physicist at NIF, where he leads the nuclear diagnostics team. People are very familiar with diagnostics in the medical field. For example, if you want to know what your temperature is, you use a thermometer. If you want to look inside your body and see if you've a broken bone, you use an x-ray machine. Those diagnostics, those instruments, look inside a very complex system to understand what's going on. And while the principle is familiar, the environment isn't. We can't just stick a thermometer in our experiment because it's 200 million degrees and hotter than the center of the sun. And we can't use an x-ray machine because it's 50 times denser than lead.

Whatever happened inside has to be measured from the light, radiation, and particles that escape. We have to develop really clever, innovative techniques that let us measure the system of interest. The longest NIF experiment is 50 billionths of a second. That's two million times faster than you can blink your eye. Two million times. Blink once right now. Go ahead. That blink took you about a tenth of a second. In that sliver of time, this first thing.

fusion event, the one we've been describing, could happen and disappear two million times over. And in that sliver of time, the diagnostics have to catch everything. We measure temperatures, densities, shape, pressure, fusion yields, structural composition, x-ray emission, particles, electromagnetic pulses coming off the experiment. If you name it, we're trying to measure it or we already have measured it. Setting up a fusion ignition experiment takes at least, eight full hours. The whole thing is over in a hundred trillions of a second.

I often sit back and reflect that when I go to work, it's an amazing opportunity to really understand a process that is like something out of science fiction. And the stakes of those measurements reach far beyond a single experiment. The United States has a stockpile of nuclear weapons that since the early 1990s we can't do any testing on. But we need them to be guaranteed to work. It's like taking care of a car you want to last for 40 years. Except you can't ever turn it on to make sure it runs.

But when you do turn it on, you need it to work every time, all the time, no questions asked. That's one of the problems NIF helps solve. Without underground nuclear testing, scientists need another way to study how materials and systems may behave as the stockpile ages. At NIF, they can recreate pieces of those extreme environments in the laboratory, and use diagnostics to measure what happens inside them. But the stockpile isn't standing still either. The National Nuclear Security Administration is rapidly modernizing key systems, introducing new designs and materials into weapons that must still perform exactly as intended.

Before any of those changes reach production, designers need proof that they will work. NIF gives them a place to build that proof, recreating the pressures and temperatures a new design will face, and using diagnostics to measure whether it behaves the way the models predict. In developing new diagnostics and even using the ones on the NIF, they come from a very long and strong family tree. The diagnostics we've developed and used have a history from actually underground testing that the U.S. did before the 90s. Back then, they were looking at an incredibly bright, very intense source that you can't physically go and measure. So you need to develop diagnostics that are stood off and can tell you about what's going on in there.

And so what we've done over the last several decades is refine and improve those measurements and customize them on the National Commission facility. That evolution continues today, with each generation of diagnostics designed to capture more information faster and under more extreme conditions. One of the newest is called Fiddle. Kara Vennari is an experimental physicist at NIF. She came to Lawrence Livermore with a background in geology, studying how matter behaves deep inside planets,

places humans cannot directly observe. At NIF, FIDL allows her to watch similar transformations unfold in the laboratory, frame by frame. Fiddle is a time-resolved X-ray diffraction diagnostic, so it takes snapshot images that are two billionth of a second long, and takes four of them per NIF shot. So we're able to measure the density and understand the structure of a material on very short time scales over the course of a single experiment. These four snapshots create something like a stop motion view of a material changing under pressure.

If you compare graphite and diamond, those are both pure carbon materials, but one is extremely hard and one is extremely soft. Those two materials have the same chemical structure, but their physical structure is completely different. and that changes the properties of them really drastically. So understanding as we apply pressure, that will change a material from one phase to another, like graphite to diamond. Under enough pressure, those atoms can rearrange, transforming one material into something with entirely different properties. We understand roughly what pressure and temperature materials should change from one phase to

another, but we don't really understand how the kinetics affect that phase transition, how long it will actually take all the atoms to rearrange and generate a new structure. That transition is the missing piece. Under extreme conditions, the way a material transforms can be just as important as the state it ultimately reaches. We're using that as a test to develop the diagnostic for future applications that can be for discovery science or stockpiled modernization. To produce that atomic transformation in the laboratory, scientists need an extraordinary amount of pressure, measured in gigapascals or GPA.

At ambient pressure, like on the surface of the Earth, we're at zero GPA, one atmosphere. The experiments that I'm working on are compressing lead. We understand what structures it can exist at, at high pressures and high temperatures. We are basically compressing that sample from zero to 110 GPA over about 15 nanoseconds. 110 gigapascals is roughly equivalent to the weight of 500,000 elephants pressing down on a single city manhole cover.

And the instrument trying to capture what happens sits just centimeters from the target. The closest part of the target to the closest part of fiddle is 2.5 centimeters away. That proximity gives fiddle the view it needs. But it also places its sensitive electronics directly in the path of everything the experiment throws off. kind of like needing to shield your eyes from the headlights of an oncoming car or the sun on the horizon. We have this problem where regenerate a plasma and a plasma is basically a gas with electrons pulled off of it. Those hot electrons fly around and I've called them the poltergeist of NIF because they hit everything. And then unfortunately when they hit stuff, they make it start glowing in x-rays.

Basically when an electron hits a material, it excites the electrons inside of it, lets up more light. Unfortunately, that process happens in our experiments, and those x-rays that get generated increase the background crazy amounts on our sensors. So there's a lot of work to understand how to control the secondary x-rays that are generated during the experiment. So the team redesigned part of Fiddle targets shielding, hoping they solved the problem. We were like, oh, we're going to be so smart. We'll let's put plastic on it and then we won't get any background. This plastic shielding created a barrier intended to stop the x-rays created by hot electrons before they could reach the instrument's sensitive detectors.

They installed the new component on the target, positioned fiddle close to the target, and prepared for another experiment. Then the warning lights came on. The target chamber closed, the lasers fired. And it was, I would say a catastrophe. At NIF, building something new often means learning first from the version that doesn't work. The team had hoped the plastic shielding would block the unwanted background. Instead, the failed shot revealed something different. It was as if we had nothing on there, and we realized that the problem actually was that we were generating maybe lower energy x-rays, but the transparency of that material to those

x-rays was much lower than the metal was. So they were making x-rays and then had a complete line of sight to our sensors. The change didn't remove the unwanted signal. It gave it a clearer path in. They needed a different type of shielding. You can beat yourself up for being wrong and not making the correct prediction, or you can really view it as an opportunity to learn and say, hey, here's something that we don't quite understand and we can really dive in and try to figure out what's going on.

In experimental science, an unexpected result isn't the end. of the work. It's new information. And the build, test, learn, and redesign cycle is how NIF's diagnostics continue to evolve. Diagnostics are both the scoreboard and the map. They're the scoreboard for Ignition to say how much fusion came out, how symmetric was the implosion, what was the performance like, and they're also the map to tell us how do we move forward next? What's the impact of the changes that we made and what should we change next? At NNIF, If that feedback not only shapes the next shot, it can shape the next diagnostic. We're constantly refining and evaluating and upgrading and even conceptualizing new diagnostics.

It's a really nice feedback loop of here's what we measured, here's what we think we understand about it, but we still don't understand this process. So let's build something else to actually tell us what we need to know. And if it's really difficult to measure, it makes it even more fun. Fiddle itself grew out of that progression. There was a predecessor diagnostic called G3D, and that was a testing diagnostic to see if the sensors that we wanted to use in Fiddle would survive the extreme conditions that are inside of NIF. And then once that proved possible with these H-C-Moss sensors, then we built FIDL, which has more of the sensors and can cover more of the diffraction angles that are required for these kinds of experiments. Now the team is developing the next generation of that diagnostic.

We will be building a new diagnostic called Falcon in the next few years that will have eight sensors and we'll have a different type of sensor that we'll be able to get instead of just four snapshots, six snapshots during an experiment. More snapshots create a richer record of the experiment. The next challenge is turning that record into a measurement scientist can understand. There's a couple things we do with the data that gets collected. The first thing is we take raw signals that we collect like voltage and time and light. And we convert them into something that's physically meaningful. And the way we do that is we apply calibrations and corrections

and we compare the data with backgrounds. It's not just you get a voltage and you know, aha, that's what the temperature is. There's a whole slew of supplemental and background work that needs to happen in order to understand what that signal is telling you. That means removing the background noise and examining the x-ray diffraction peaks left behind by the material. For any one shot, we get many different diagnostics, and we want to understand every aspect of it.

We subtract the rod shot, which is any background that could be contributed by just the sensor being turned on. And then we analyze the peaks using Labai fits, which is a standard X-ray diffraction technique. So we know what structure the material should be in. We can generate a synthetic X-ray diffraction pattern, and then we can pull out the density that we measure during the experiments. Each diagnostic preserves a different piece of the same event. Scientists also compare those pieces, looking for independent measurements that tell a consistent story. Another thing we do with the data that we collect is we compare it with different diagnostics. For example, if one system tells us the fusion yield of some value, and then another diagnostic that's independent confirms that value,

then it gives us more confidence in that measurement. On the NIF, we do have some diagnostics that measure the same quantity, and that's by design so that we understand, yes, this is a value that we can have high confidence in. Scientists also compare those measurements with what computer models predicted before the shot. How the pressure should rise, how the material should respond, and what the diagnostics should see. We run the experiment and collect data, and that data either confirms the model's predictions, or says, hey, nature doesn't quite behave like you think it does. You need to update your model on it and change your understanding of things. And so one thing we do with the data is we compare it to these models and use it to refine our understanding.

That understanding can then travel beyond the experiment itself, informing new designs and even bringing knowledge to related fields, like the quest to make fusion energy practical. There's synergy between NIF Diagnostics and Research and the needs of future fusion energy plants. In terms of fusion energy, there's a laboratory institute called Lyft, Livermore, Institute for Fusion Technology, and that is geared towards taking some of the understanding that we've developed

and saying, how do we leverage that to move towards fusion energy? It's a great organization that interfaces between public-private partnerships, as well as trying to push forwards the entire field. The shot may be over in an instant, but scientists and engineers can return to its data for hours, days, and years, reconstructing what happened, testing new explanations, and designing better experiments. The shot on NIF itself is gone almost instantly, But the data remains for hours, days, years after that.

And so what we can do is we can have hundreds of scientists and engineers reconstruct what happened over time, test hypotheses, and build better experiments for the next round of operations. So the real product of diagnostics is not just the numbers that get produced, but the understanding of what's going on in these physical processes. At NIF, the most important part of the experiment is also the part no one can see directly. The target compresses. materials transform, fusion reactions unfold, and within a fraction of a second, the event is gone.

But the diagnostics do more than explain a single experiment. I think what is so unique and fulfilling about the work that the National International Facility and Lawrence Livermore National Laboratory do is that it's built in with purpose. The work I do is really relevant for stockpile modernization, it's really relevant for national security, it's really relevant for fusion and fusion energy. The instruments make hidden physics visible. What scientists do with that view is what turns an experiment into something the country can use. Thank you for tuning in to Big Ideas Lab.

If you loved what you heard, please let us know by leaving a rating and review. And if you haven't already, don't forget to hit the follow or subscribe button in your podcast app to keep up with our latest episode. Thanks for listening. listening.