Cutting-edge science keeps the nation’s nuclear deterrent safe, secure, and effective in an era without nuclear testing.
Depending on who you talk to, the Cold War was less a conflict between the United States and the Soviet Union than it was between the United States’ own nuclear superpowers: Lawrence Livermore and Los Alamos national laboratories.
Today the two facilities, which for seven decades have designed the United States’ nuclear weapons, collaborate more than they compete. But for much of the Cold War, when the laboratories routinely vied for contracts to design the bombs and warheads that compose the nation’s nuclear arsenal, their relationship tended toward rivalry. As one Livermore weapons scientist joked to anthropologist Hugh Gusterson in the late 1980s, “The Soviets are the competition, but Los Alamos is the enemy.”
Unlike the conflict between the United States and the U.S.S.R., though, the arms race between Los Alamos and Livermore was by design—a way of putting competition to use for the benefit of the nation’s security.
“Our lab was explicitly founded to provide competition in the design of nuclear weapons,” says Mark Herrmann, program director for weapons physics and design at Livermore. “I think we’ve seen, over many decades, that the combination of collaboration and competition has really provided benefits—new ideas, better ideas—that push us forward.”
Today, in the era of stockpile stewardship—which aims to ensure the safety, security, and effectiveness of the nation’s nuclear deterrent, without resorting to nuclear testing—ties between Los Alamos and Livermore have never been closer. “Our partnership, in an era without underground nuclear testing, is even more important,” Herrmann says. “It’s even more important for us to be working together and providing peer review, because we can’t go out and test out ideas to the extent that we could.”
Nuclear rivals
Los Alamos was almost a decade old when, in 1952, Livermore was founded (as the University of California Radiation Laboratory at Livermore) by Manhattan Project alumni Ernest Lawrence and Edward Teller. A World War II-era naval air base in rural Livermore, California, 45 miles east of San Francisco, was selected as the site of the new laboratory, partly for its remoteness. The town, separated from the Bay Area by a line of golden hills, wouldn’t be serviced by a highway until the 1970s, which made the new laboratory better suited for classified work than Lawrence’s laboratory in metropolitan Berkeley, California.
From the outset, Los Alamos scientists resented their new colleagues. Researchers at Los Alamos were piqued when, in 1952, the media incorrectly attributed the world’s first successful test of a hydrogen bomb to Livermore, rather than Los Alamos. Due in part to this perceived slight, Los Alamos’ scientists were gleeful when, a year later, Livermore’s first two nuclear weapons tests “fizzled,” or failed.
Despite these early setbacks, Livermore soon distinguished itself with its efforts to design thermonuclear weapons that were small in size but high in yield. In 1956, Teller shocked attendees of a naval conference by declaring that within five years, Livermore could design a 1-megaton-yield thermonuclear weapon that would fit atop a submarine-launched missile. Such a warhead would have to be less than one-tenth the size of any such weapon hitherto developed.
Teller’s audacity proved justified. In 1960, the Navy deployed the first Polaris missiles topped with the Livermore-designed W47—a small-yet-powerful warhead that signaled a turning point in nuclear weapon design.
Acrimony between the laboratories persisted, with scientists at times publicly questioning each other’s work. But in commenting on a dispute between the laboratories in the 1970s, Teller took tensions between the two in stride: “The laboratories need each other to keep each other honest,” he said.
“Livermore and Los Alamos each have history, and culture, and perspective, and they’ve each brought things to our nuclear enterprise that are absolutely essential to how it operates today,” Herrmann elaborates. “I see the advances that have been made in our nuclear weapons capability, and how good ideas from both labs have ended up in our nuclear weapons systems. The labs’ relationship has made all of our systems stronger.”
The challenge of stockpile stewardship
Today the formerly rural town of Livermore has been absorbed into Bay Area sprawl. Once a modest air base, Lawrence Livermore National Laboratory now boasts an array of glassy buildings spread across a 1.3-square-mile campus whose walkways are shaded by oak trees and other native flora.
In the post-Cold War era, Livermore is responsible for three of the seven weapons systems that make up the nation’s enduring nuclear stockpile (Los Alamos is responsible for the other four). But while Livermore and Los Alamos each maintain responsibility for different weapons, in the past three decades the laboratories have come more and more to share their capabilities.
“In the Cold War era, we were both developing systems and expertise on all the fronts we needed,” Herrmann says. “Now, we both need to be experts in different areas, but we’re also focusing on areas and investments and capabilities that we can use in a national way to develop the experimental data, and the computational data, that we need to have confidence in our systems.”
Much of the laboratories’ increased collaboration can be traced to the cessation of underground nuclear testing in the United States. The United States hasn’t tested a nuclear weapon since 1992, when President George H.W. Bush signed a moratorium on testing into law. In 1996, President Bill Clinton signed (but Congress did not ratify) the Comprehensive Nuclear-Test-Ban Treaty, which prohibits any kind of nuclear testing. That same year, the science-based Stockpile Stewardship Program was inaugurated by the Department of Energy to ensure that the nation’s nuclear weapons continue to perform as expected, even without testing.
Bruce Goodwin, who previously served as associate director for defense and nuclear technologies at Livermore, has likened the challenge of stockpile stewardship to inheriting an aging automobile. Imagine that you’ve been given the keys to a car that hasn’t been started in 40 years. Many of the car’s parts will be corroded or broken, but replacements might no longer be available, nor the materials to fabricate new parts.
Now imagine that you’ve been challenged to ensure, with 100 percent certainty, that the car will start every time someone turns its key—and that you must achieve this feat without ever starting the engine yourself.
This, in essence, is the dilemma posed by stockpile stewardship. Of course, nuclear weapons are no aging Ford: they rely for their effects on materials like plutonium, the most complicated element on Earth, and hydrogen, which in a nuclear detonation undergoes a series of reactions that commonly take place inside stars—although in a weapon these reactions occur more than 20 orders of magnitude faster than they do in, say, the sun.
The complexities of ensuring the safety, security, and effectiveness of nuclear weapons in the post-testing era are such that no one facility can shoulder the burden of stockpile stewardship alone. Gone are the days when Los Alamos or Livermore might have worked in isolation to complete weapon initiatives like the W87-1 modification program, which is underway now at Livermore. The W87-1’s design is based on previous stockpile designs, including the W87-0, which was developed in the 1980s.
“There really is no one facility that can deliver this design to the Department of Defense,” says Alicia Williams, project engineer for the W87-1. “The design agencies design it, but they have to work with the production agencies that modify the design and make sure it’s manufacturable. The design agencies also have lots of test facilities that we need to leverage.”
As a part of the Stockpile Stewardship Program, the National Nuclear Security Administration (NNSA) administers modification programs that refurbish or build new weapons based on previously tested components. Other modification programs, such as the W76-2 program at Los Alamos, have used a mix of old and new components. The W87-1 modification program differs in that it is the nation’s first 100 percent new-manufacture nuclear weapon system since the end of the Cold War, meaning that every part of the warhead will be fabricated specifically for use in the system.
Once the W87-1 goes into production in 2030, the W87-1 will replace the W78, which has served atop the land-based Minuteman III since the mid-1970s. (The Minuteman III, which entered service in 1970, will be retired and replaced with the LGM-35A Sentinel intercontinental ballistic missile.)
To complete the W87-1 modification program, Livermore is collaborating with facilities across the country, including Sandia National Laboratories—located both in Albuquerque, New Mexico, and in Livermore, California, just across the street from Lawrence Livermore National Laboratory—the Kansas City National Security Campus, and the Pantex Plant in Amarillo, Texas.
Collaboration with Los Alamos also will be crucial. In addition to making use of facilities like Los Alamos’ Dual-Axis Radiographic Hydrodynamic Test (DARHT) facility, Livermore will rely on Los Alamos for both detonators and plutonium pits—two essential components of all nuclear weapons. Without these production capabilities, new manufacture of the W87-1 wouldn’t be possible.
“As an enterprise, we do have to work as a team,” Williams says. “When I played high school soccer, my coach would always say, ‘You’ve got to communicate with each other. You’ve got to talk to each other and make sure you understand what the other players are doing, and that you’re ultimately working toward your goals together.’ It’s really no different doing this work.”
Peer review
Today Los Alamos and Livermore conduct annual peer reviews that allow the laboratories to offer each other input on the design of weapons, such as the W87-1.
In the nuclear testing era, peer reviews between Livermore and Los Alamos were less frequent and less comprehensive, in part because data from tests often furnished each laboratory with the means to validate its designs. After nuclear testing was halted in 1992, Livermore and Los Alamos began to conduct the peer reviews that now inform each laboratory’s annual assessment of its weapons. Every year, teams of scientists from each laboratory consider a series of weapon design problems faced by researchers at the other laboratory, offering input on one another’s challenges. As the nation’s only two nuclear weapon design laboratories, Los Alamos and Livermore furnish each other with perspectives that neither can find anywhere else.
Herrmann says that this makes for a singular relationship between the two laboratories. He notes that although Livermore also works closely with partners like Sandia—which designs and fabricates nonnuclear components for bombs and warheads—the fact that Livermore specializes in nuclear explosives means that Livermore and Sandia don’t have the kind of peer review relationship that Livermore and Los Alamos do. The peer review process is a way of ensuring that both laboratories work together, if also in productive competition, for the national good.
“Between Los Alamos and Livermore, it’s that collaboration-competition relationship that’s special,” Herrmann says.
Few resources better reflect Livermore and Los Alamos’ collaboration-competition relationship than Livermore’s Sierra supercomputer.
Throughout the Manhattan Project—which led to the creation of the world’s first nuclear weapons—and the Cold War, computing was used in tandem with testing to affirm that nuclear weapons would perform as expected. The end of nuclear testing has made computing even more central to ensuring the efficacy of the nation’s stockpile. Today, as the United States’ weapons age and evolve beyond the devices that once were tested, the computer codes used to simulate nuclear explosions have evolved to become more predictive, allowing researchers to leverage legacy data acquired from testing alongside newer experimental data.
Both Los Alamos and Livermore have developed high-performance computing capabilities that support their nuclear weapons research. With a peak capacity of 41.5 petaflops (a petaflop is one quadrillion floating point operations per second), Los Alamos’ Trinity supercomputer was the seventh fastest computer in the world when it went into service in 2017. But Livermore’s Sierra system is even faster. When Sierra was commissioned in 2018, the computer became the third fastest in the world, boasting a peak capacity of more than 125 petaflops. These speeds make possible routine three-dimensional weapons simulations, which previously were used only to double-check two-dimensional simulations.
Sierra is housed in a 48,000-square-foot facility that is also home to Livermore’s other supercomputers, on a floor where the whine of cooling fans is so loud that visitors are encouraged to wear earplugs. Encrypted high-speed connections allow Los Alamos and Sandia to access the machine, too. In fact, use of Sierra is divided equally among the three laboratories. But while the laboratories share use of Sierra, the codes that each laboratory uses in its weapons simulations are developed separately.
“We make a very explicit decision to keep those codes separate from one another,” Herrmann says. “We have different computational tools that take different approaches to different scenarios, and each has its strengths and weaknesses.”
Scott Futral, who is the development environment group leader in Livermore’s computing division, agrees. “There’s no such thing as a perfect code,” he says. “Independence in methodology allows the laboratories to compare their results.”
The laboratories’ shared computing endeavor will receive a further boost in 2023, when the El Capitan supercomputer is expected to enter service at Livermore. This new system will be an exascale computer at least 10 times as powerful as Sierra. El Capitan is expected to be capable of reaching 2 exaflops, or 2 quintillion calculations, per second, making the computer the second in the world to achieve exascale speeds (after Oak Ridge National Laboratory’s Frontier supercomputer).
Trust, but validate
Although high-performance computing has reduced scientists’ reliance on legacy data from nuclear tests, researchers still depend on experimental data to validate computer codes’ accuracy. Today Livermore and Los Alamos collaborate to develop the enhanced capabilities for subcritical tests that make this validation possible.
Since 1995, both laboratories have used the U1a Complex at the Nevada National Security Site (NNSS) for subcritical tests, in which high explosives compress plutonium without bringing it to the point of a self-sustaining chain reaction. Los Alamos is leading the development of Scorpius, a particle accelerator that will be used for subcritical testing at U1a. However, Livermore is developing the pulsed power that will drive the accelerator. (Sandia and NNSS are contributing to Scorpius’ development as well. Scorpius is expected to become operational by 2030.)
The National Ignition Facility (NIF), which is housed at Livermore, also helps validate the computer codes developed by Livermore and Los Alamos. NIF is the world’s largest and most energetic laser. Since opening in 2009, the facility has allowed researchers to study in novel ways the materials used in nuclear weapons. By focusing 192 laser beams onto a target the size of a pea, NIF can induce temperatures of more than 180 million degrees Fahrenheit and pressures greater than 100 billion Earth atmospheres—making a target at NIF briefly the hottest place in the solar system.
No other facilities in the world can generate such effects. NIF allows researchers from Livermore and Los Alamos to conduct experiments on elements like plutonium and hydrogen in ways that previously would have been possible only with underground testing, but to do so with fewer hazards and greater repeatability. The heat and pressures induced at NIF mean that different kinds of tests can be conducted there than are possible at facilities like the Joint Actinide Shock Physics Experimental Research Facility (JASPER), which Livermore operates at NNSS.
NIF also allows researchers to conduct experiments in materials properties, radiation transport, hydrodynamics, and weapons survivability, making it a versatile resource for scientists who work in the Stockpile Stewardship Program.
“NIF fills a specific niche that other facilities don’t,” says Heather Whitley, associate program director for high energy density science at Livermore. “There is some overlap in terms of things that we can do at Z”—Sandia’s Z Pulsed Power Facility, which uses magnetic fields to produce extreme conditions—“but even the Z machine doesn’t operate in the same regime as NIF. Both NIF and Z are supplying data to inform design decisions for the W87-1.”
NIF operates 24 hours a day, with technicians conducting some 400 “shots” there every year. At the head of NIF’s NASA-style control room, screens show images of the inside of the facility’s aluminum target chamber, which when viewed from the outside resembles the kind of futuristic machinery one might find aboard Star Trek’s USS Enterprise. (As a matter of fact, in 2012 NIF served as a set for the film Star Trek: Into Darkness.)
Experiments conducted at NIF have addressed stockpile challenges such as the energy balance problem, which dogged researchers for decades. In the 1960s, data collected during nuclear tests suggested that some of the energy scientists expected to be produced by explosive devices was “missing”—a violation of the law of conservation of energy. Beginning in the early 2000s, Livermore physicist Omar Hurricane led a team of researchers who conducted a series of experiments, including at NIF, that helped solve the decades-old mystery. The energy balance problem’s resolution will support initiatives like the W87-1 modification program.
As NIF’s name implies, one of the facility’s goals has been to achieve fusion ignition, in which more energy is produced by a reaction than is put in. This goal was achieved on December 5, 2022, when a NIF shot delivering 2.05 megajoules of energy resulted in 3.15 megajoules of fusion energy output.
The achievement of fusion at Livermore expands the range of experiments that can be performed at NIF in support of stockpile stewardship. Moreover, researchers around the world are investigating the prospect of harnessing fusion energy as an energy source. Many barriers remain in this endeavor, meaning that technologies like fusion-powered electrical plants are years in the future. But Livermore’s fusion achievement nevertheless amounts to a major breakthrough.
“The pursuit of fusion ignition in the laboratory is one of the most significant scientific challenges ever tackled by humanity, and achieving it is a triumph of science, engineering, and most of all, people,” said Livermore Director Kim Budil in a news release. “Crossing this threshold is the vision that has driven 60 years of dedicated pursuit—a continual process of learning, building, expanding knowledge and capability, and then finding ways to overcome the new challenges that emerged. These are the problems that the U.S. national laboratories were created to solve.” ★