Pacific Fusion Company Just Broke Ground Next to a Nuclear Weapons Lab — Here’s Why
In short: Pacific Fusion, a private fusion energy company, has started building a $1 billion “net facility gain” fusion demonstration system in Albuquerque, New Mexico — right next to Sandia National Laboratories, a US nuclear-weapons lab. That’s not a coincidence: Sandia is formally partnered with Pacific Fusion to optimize the underlying pulsed-power technology for both civilian energy and nuclear-weapons research simultaneously. Independent fusion scientists say the approach is promising but caution there’s still a long way to go before it produces usable electricity.
A private company building next to a weapons lab, on purpose
Pacific Fusion is assembling components for its fusion machine in a warehouse in Los Lunas, New Mexico, and has now broken ground on a separate, larger site in Albuquerque’s Mesa del Sol district — directly behind Kirtland Air Force Base and down the road from Sandia National Laboratories. Sandia isn’t a passive neighbor. The lab, whose focus is national security research, has a formal cooperative agreement with Pacific Fusion specifically to refine the company’s pulsed-power fusion technique for both energy generation and weapons applications at once.
Keith LeChien, Pacific Fusion’s chief technology officer and co-founder, described the stakes plainly to Nature: the work is “going to be transformative for both fusion energy and defence applications.” That’s an unusually direct admission from a privately funded energy startup that its research has a second customer beyond the electrical grid.
Why fusion research and nuclear weapons research overlap at all
This isn’t a new idea specific to Pacific Fusion — it’s the same logic that has justified US government-funded fusion research for decades. Since the United States stopped conducting live nuclear weapons tests in 1992, government scientists have needed another way to understand how aging warhead materials behave under the extreme heat, pressure, and radiation a detonation produces. That’s the core mission of the US Stockpile Stewardship Program, and it’s why facilities like the National Ignition Facility in California — better known for chasing fusion energy headlines — have always been jointly funded and run under the National Nuclear Security Administration rather than purely as civilian energy research.
A machine that can briefly recreate extreme energy densities in a small, controlled, non-nuclear way is useful for both goals at once: pushing toward a fusion energy breakthrough, and generating physics data relevant to how nuclear materials behave without an actual explosion. Pacific Fusion’s arrangement with Sandia follows that same established pattern — a private company’s commercial fusion ambitions and a national lab’s weapons-stewardship mission, running on the same hardware.
What Pacific Fusion is actually building
| The approach | Pulsed power — using extremely brief, extremely powerful electrical pulses to compress a fuel capsule, rather than the laser-based approach used at facilities like the National Ignition Facility. |
| The core hardware | An impedance-matched Marx generator (IMG), a pulsed-power architecture co-invented by LeChien, developed as a direct evolution of Sandia’s own Z Machine — itself nearly half a century into pulsed-power research. |
| Scale | More than 150 truck-sized IMG modules, each a bank of capacitors, arranged in a ring around a central reaction chamber, all firing in parallel toward a single pea-sized hydrogen fuel capsule at the center. |
| The pulse itself | Nature’s reporting describes the combined electrical pulse as thousands of times more powerful than a lightning bolt, generating a magnetic field that compresses the fuel capsule until hydrogen isotopes fuse into helium and release energy. |
| The goal | “Net facility gain” by 2030 — the point where the machine produces more energy than the entire facility consumes to run it, not just more energy than hits the fuel target directly. |
Why pulsed power instead of lasers. At the National Ignition Facility, less than 1% of the energy used to fire its 192 lasers actually reaches the fuel capsule — the rest is lost converting electrical charge into light, then X-rays, then kinetic energy, with losses at every stage. LeChien told Nature that Pacific Fusion’s pulsed-power design gets roughly 12% of its initial electrical charge to the target — a more direct route with far fewer conversion losses, which is the central engineering bet behind choosing this approach over lasers.
The numbers, and why independent scientists are still cautious
That last figure matters more than it might first appear. LeChien has been candid that even a successful 100-megajoule shot against an 80-megajoule charge — a genuine net facility gain — would still only represent about one-fifth of the energy gain an actual power plant would need per pulse. Reaching that milestone would prove the physics works at facility scale; it would not yet prove the physics works at power-plant scale, and Nature’s reporting notes that operating the facility continuously enough to generate sustained, usable power involves separate, substantial engineering challenges beyond the initial gain target.
Independent researchers echo that caution. Michael Campbell, a fusion engineer at UC San Diego not affiliated with Pacific Fusion, told Nature the pulsed-power approach is promising but said plainly: “there’s still a long ways to go.” Kyle Peterson, a plasma physicist at Sandia itself, framed the broader field’s uncertainty simply: no single fusion approach among the current wave of well-funded startups has an obviously correct design yet, which is part of why so many different companies are pursuing different physical mechanisms in parallel rather than one dominant strategy emerging.
What hasn’t been demonstrated yet
Pacific Fusion has real, verifiable engineering milestones behind it. The company’s own scaled pulser prototype has delivered 440 gigawatts of peak power in an 80-nanosecond test pulse, and a separate prototype built with Lawrence Livermore National Laboratory under a cooperative research agreement, called Sirius, completed more than 3,000 test shots at 95% energy efficiency. What hasn’t happened yet is the full-scale demonstration system actually running and hitting net facility gain — that’s the entire purpose of the facility now under construction, not something already proven. Pacific Fusion’s roughly $1 billion in funding to date has come in staged tranches tied to hitting specific technical milestones rather than as a lump sum, which means continued funding itself depends on the company continuing to hit targets like this one.
A groundbreaking is a construction milestone, not a physics result. Starting construction on a facility designed to demonstrate net facility gain is not the same as having demonstrated it. The specific 2030 target is the company’s own stated goal, not an independently verified timeline, and Nature’s own sourcing — including scientists with no financial stake in Pacific Fusion’s success — frames the approach as promising but unproven at the scale that would actually matter for a power grid.
My Take — Mr Wangdoo
The technology here is interesting, but the part I keep coming back to is the governance question sitting underneath it. Pacific Fusion is a privately funded company, backed by venture capital and answerable to its investors, deliberately building infrastructure that a national-security agency will also rely on for nuclear-stockpile research. That’s not new in principle — the National Ignition Facility has run on exactly that dual mission for years — but it’s unusual to see a venture-backed startup, rather than a government lab, sit at the center of that arrangement from day one. Worth watching isn’t just whether the physics works by 2030; it’s how a privately funded company balances shareholder pressure for commercial fusion progress against a national lab’s slower, more methodical stockpile-stewardship research needs, when both are running on the same hardware.
Common questions
What is Pacific Fusion actually building?
A fusion energy demonstration facility in Albuquerque, New Mexico, using pulsed-power technology (an impedance-matched Marx generator) to compress a hydrogen fuel capsule and trigger fusion, aiming to demonstrate “net facility gain” — producing more energy than the facility consumes — by 2030.
Why is it being built next to a nuclear weapons lab?
Sandia National Laboratories, located nearby, has decades of pulsed-power research experience that Pacific Fusion’s core technology builds on, and the two organizations have a formal cooperative agreement to develop the technology for both civilian fusion energy and nuclear-weapons stockpile research simultaneously.
Is Pacific Fusion working on nuclear weapons?
Not directly. The company is building energy infrastructure that Sandia, a national-security laboratory, can also use to study how aging nuclear weapons materials behave under extreme conditions — a research approach the US has relied on since ending live nuclear testing in 1992, and one the National Ignition Facility has used for the same dual purpose for years.
Who leads and funds Pacific Fusion?
Pacific Fusion was co-founded in 2023 by Eric Lander (its CEO), Keith LeChien, Will Regan, Carrie von Muench, and Leland Ellison. Former Google CEO Eric Schmidt sits on its board of directors alongside other investors, but is not the company’s CEO. The company has raised roughly $1 billion since founding, in tranches tied to technical milestones, led by General Catalyst.
Has this technology been proven to work?
Not at the scale that would matter for a power grid. A scaled prototype has demonstrated reliable, repeated pulses at a national laboratory, but the full demonstration facility now under construction has not yet run, and independent fusion scientists describe the overall approach as promising but still requiring significant further progress.
Sources
- Nature — “Billion-dollar machine aims to set a fusion-energy record — and study nuclear weapons,” on-the-ground reporting from Pacific Fusion’s Los Lunas facility, 24 August 2026. nature.com
- Wikipedia — Pacific Fusion company history, founders, and funding. en.wikipedia.org
- Pacific Fusion — company founders’ letter and technical approach. pacificfusion.com
