Ensuring national security with a modernized nuclear weapons stockpile and a responsive nuclear security enterprise
LLNL’s foremost responsibility is strategic deterrence. The Laboratory is sustaining weapons systems in the nation’s nuclear arsenal as it engages in programs and partnerships that jointly modernize the stockpile and the weapons production enterprise. Effective deterrence relies not only on the nation’s deployed weapons systems, but also on the extensive science and technology (S&T) capabilities of its national security laboratories and the agility of the nuclear security enterprise (NSE) to respond to emerging national needs.
Strategic Modernization Programs
LLNL is the design agency partnering with Sandia National Laboratories (SNL), production agencies, and the U.S. Air Force to develop and certify the W80-4 warhead for the bomber-delivered Long-Range Standoff (LRSO) missile. While challenges remain, the team is making good progress in the Life Extension Program (LEP), which currently is in Phase 6.4 (production engineering) with plans to conduct the System Final Design Review in 2025. The W80-4 will reuse existing designs while modernizing some components and incorporating additional safety features. The LEP is taking advantage of new manufacturing methods that minimize costs, increase throughput, and reduce the need for environmentally sensitive materials and processes. The First Production Unit (FPU) of the W80-4 is scheduled for September 2027.
Production engineering fully engages partnerships across the NSE to improve the producibility of components. The W80-4 team completed the first full-lifetime Full System Engineering Test with development hardware and diagnostics with positive results. The Laboratory also delivered test assemblies to SNL and the U.S. Air Force for ground and flight tests. Final design reviews were conducted for multiple components. Other activities included an integrated hydrodynamic experiment conducted at Los Alamos National Laboratory (LANL) that included newly manufactured high explosives (HE) and final-design components, aging and compatibility testing, and HE acceptance tests. Where technical challenges remain, options have been developed to meet system delivery timelines.
LLNL and SNL are also the design agencies for the W87-1 Modification (Mod) Program. The laboratories are working with partners across NNSA and the U.S. Air Force to deliver a warhead to replace the aging W78. To be deployed on the Sentinel ballistic missile (under development) in the early 2030s, the W87-1 will be the first modern warhead that is 100 percent newly manufactured. Currently in Phase 6.3 (engineering development), the W87-1 Mod Program is working toward component production with seven systems and sub-systems tested this year. The W87-1 Mod team is pursuing innovative, transformational partnerships with NNSA design agencies to introduce new materials, manufacturing techniques, and processes to meet W87-1 and future stockpile systems requirements. Advances in the tools of science-based stockpile stewardship will enable confident certification of the new warhead without nuclear explosive testing.
A key accomplishment, announced at the end of FY 2024, was the manufacture at LANL of the FPU plutonium pit for the W87-1 nuclear warhead. This first fully qualified pit was “diamond stamped” after meeting all requirements, signifying its readiness for deployment to the U.S. nuclear stockpile at “war reserve” quality. Achieving FPU of the W87-1 pit is an important milestone for the United States’ nuclear weapons stockpile modernization program. The success was a collaborative team effort with LLNL responsible for the design of the W87-1 plutonium pit and leading the product realization team. Experimental capabilities at LLNL contributed to assessment and certification of the manufactured W87-1 pit. An array of capabilities at the Superblock’s Plutonium Facility were used to prepare samples and perform qualification tests on pit coupon samples. The effort spanned more than eight years to develop and mature qualification, certification, and product acceptance processes required to manufacture this FPU pit.
Enclaves and Production Partnerships
Success in the modernization programs requires an integrated effort within the NSE and strong partnerships with the U.S. Air Force and their contractors. LLNL is spearheading groundbreaking initiatives to develop innovative materials and manufacturing techniques. Researchers from the Laboratory and the Kansas City National Security Center (KCNSC) work side-by-side at LLNL’s Polymer Enclave and equivalent resources at KCNSC to meet rate production goals for key components. In addition, LLNL and the Y-12 National Security Complex have teamed up to rapidly modernize technology and production methods for crucial special materials.
Launched in 2022, LLNL’s Energetic Materials Development Enclave Campus (EMDEC) has strengthened partnerships within the NSE. An important collaborative effort of LLNL and the Pantex Plant is focused on the time-urgent need to manufacture insensitive HE components for the strategic modernization programs. EMDEC serves as a test bed for transformative approaches to production. The center integrates the extensive suite of capabilities at the Laboratory—ranging from synthesis and formulation of new energetic molecules to pressing, machining, and assembly of HE components for evaluation and dynamic testing. Focused on the goal of reducing the time to achieve high-yield-rate production, LLNL and Pantex have installed identical equipment at the two sites to support the joint effort. EMDEC features the Facility for the Advanced Manufacture of Energetics (FAME), where first-of-their-kind technologies have demonstrated safe 3D-printing of precision insensitive HE components. Plans are to expand EMDEC with pilot-scale manufacturing capabilities for scaleup and use at the production agencies.
Annual Stockpile Assessment
In FY 2024, LLNL completed Cycle 29 of the annual stockpile assessment. The process included a formal comprehensive peer review between LLNL and LANL of each other’s weapons systems. Testing and analysis activities enabled an assessment of the condition and sustainment of the B83, W80-1, and W87-0 stockpile systems. These efforts benefited from steady increases in understanding legacy explosives, the results of subcritical experiments that helped inform the effect of pit aging, and high-resolution 3D analyses that provided high-confidence understanding of anomalies in the stockpile.
El Capitan: The World’s Most Powerful Supercomputer
In November 2024, LLNL, in collaboration with NNSA, Hewlett Packard Enterprise (HPE), and Advanced Micro Devises, Inc. (AMD), officially unveiled El Capitan as the world’s most powerful supercomputer and the first exascale system dedicated to national security. The TOP500 organization verified the system’s performance at 1.742 exaFLOPs (1.742 quintillion calculations per second) using its standard benchmarking tool. With more than 11,000 compute nodes and 5.43 petabytes of total memory, El Capitan has a peak performance of 2.79 exaFLOPs. An exascale computer can in one second outperform the combined effort of a billion people calculating day and night for more than a decade.
NNSA’s first exascale supercomputer provides Tri-Lab (LANL, LLNL, and SNL) scientists and engineers a factor-of-20 leap forward in modeling weapons performance and safety in high-fidelity resolution with quantified uncertainties. Capabilities to routinely run large-scale 3D models are essential for maintaining an aging stockpile and certifying modernized weapons systems. El Capitan will also accelerate progress in inertial confinement fusion, enable discoveries in material behavior under extreme conditions, and support other critical nuclear security missions such as nonproliferation and counterterrorism. In addition, unclassified research projects at the NNSA laboratories will greatly benefit from use of a companion system, Tuolumne, built with the same architecture and components as El Capitan and one-tenth the scale (288 petaFLOP peak performance). El Capitan will transition to classified use early in 2025. As current efforts demonstrate (see National Ignition Facility and Partnerships), the combination of exascale computing, major advances in AI accelerator technology, and the Laboratory’s high-performance computing expertise will greatly benefit national security in wide-ranging applications.
A key foundation for El Capitan was DOE’s Exascale Computing Project (ECP). Launched in 2016, ECP enabled NNSA and DOE Office of Science (SC) researchers to collaboratively overcome imposing technical challenges and deploy the exascale Aurora and Frontier systems at SC laboratories. The NNSA laboratories focused on the next steps to meet their challenging needs. They formed working groups with HPE and AMD to collaborate on co-development of first-of-its-kind hardware, the required software ecosystem, and compatible scientific applications. The HPE Cray supercomputer features an innovative high-speed interconnection network and first use of new local data storage units, called Rabbits. El Capitan is powered by AMD InstinctTM MI300A accelerated processing units that deliver unmatched computational performance, energy efficiency, and reliability for the laboratories’ AI-assisted simulation codes. The completed $100-million Exascale Computing Facility Modernization project at LLNL provides the power and cooling infrastructure for El Capitan, which—in spite of its high energy-per-node efficiency—requires 35 megawatts.
Supported by NNSA’s Advanced Simulation and Computing program, Laboratory computer scientists developed the Tri-Lab Operating System Stack (TOSS), which serves as the operating system for commodity computing clusters at the three laboratories. El Capitan is running an upgraded version of TOSS. LLNL and collaborators also created many software tools that support applications development and manage scientific computing workflows, nearly a dozen of which won R&D100 awards and are widely used open-source software. Teams of code developers used early-access systems as test beds to design and test multi-physics applications for use on El Capitan. Years of preparation are about to pay off.
Nightwatch and a New Role for LLNL
In 2024, NNSA accepted Nightwatch, the first digital product developed and produced by LLNL, for integration into the nuclear stockpile. Nightwatch is a self-contained temperature logger that is designed to sit inside the shell of a nuclear weapon. It records temperature fluctuations over its multi-decade service life, which can pose risks to the longevity and reliability of weapon components. This data will provide crucial insights into how weapon materials are aging under various conditions and help ensure the safety and reliability of the stockpile.
The acceptance was celebrated with the Livermore Field Office at a ceremonial “stamping” event. Stamping marked Nightwatch’s compliance with all technical and quality requirements. NNSA’s official stamp of approval is the culmination of years of effort. The project required extensive collaboration with other NNSA entities to establish processes, develop governing documents, and ensure rigorous quality standards were met. The Laboratory is now an approved production agency for digital products—NNSA’s first new production agency since 1996. Hardware and electrical components will still be manufactured and produced by NNSA partners. In its new production role, LLNL is developing additional digital products that are expected to reach the stamping process in 2025 and 2026.
A Successful Subcritical Experiment
In May 2024, LLNL led execution of the first U.S. subcritical experiment (SCE) since 2021 at the Nevada National Security Site. The successful multi-institutional experiment—the first in a series of SCEs named “Nimble”—was staged nearly 1,000 feet underground at the Principal Underground Laboratory for Subcritical Experimentation (PULSE) facility, formerly known as the U1a Complex. To study the properties of special nuclear material under extreme conditions, the research team detonated high explosives in a containment vessel designed to prevent the release of radiological material. SCEs are designed to remain below the threshold of criticality. The years of preparation and combined efforts of engineers, scientists, and technical experts across multiple organizations were instrumental in the experiment’s success. The high-fidelity diagnostic data from the Nimble experiments will answer key weapons physics questions and enable essential advances in modeling and simulation capabilities that support certification efforts.
