Built to ship in a standard 20-foot cargo container, a 1-megawatt microreactor in Idaho sustained its first chain reaction in 150 days flat, and three more followed before Independence Day
Inside a concrete test bay at the edge of the high desert, a reactor the size of a shipping container sustained a chain reaction for the first time on the first of July.
No cooling towers loomed outside.
The whole machine, fuel and all, was designed to leave the factory in a standard 20-foot box.
The federal energy secretary visited the laboratory before the test and inspected a fuel rod, an unusual level of cabinet attention for a zero-power demonstration.
It took just 150 days from formal kickoff to get there.
Stranger still: three more reactors hit the same milestone before the week was out. How did startup teams move that fast?
How a reactor goes critical, and why 150 days matters so much
Criticality is the moment a reactor crosses a threshold: neutrons released by splitting atoms sustain more splits on their own, without any external source pushing the reaction forward. A reactor that cannot reach criticality cannot produce power.
In conventional programs, the road to that moment runs through years of regulatory review, site preparation and staged fuel loading. What compressed the timeline was a deliberate structural change to how the federal government treats prototype reactors.
An executive order established a Reactor Pilot Program mandating that at least three advanced reactor concepts achieve criticality by July 4, 2026, with companies fully responsible for costs while national laboratories acted as facilitators. That program opened a separate door, distinct from the conventional multi-year licensing path, allowing small teams to move at a pace the industry had not seen before. The result was a race measured in weeks, not years.
A reactor designed to go anywhere the grid cannot reach
The Unity demonstration reactor reached initial criticality at the National Reactor Innovation Center at Idaho National Laboratory on July 1, 2026. The test was conducted at what engineers call zero power operation: enough neutrons to prove the chain reaction is self-sustaining, but deliberately held well below the thermal output it would produce in commercial service.
The Unity Nuclear Battery is a 1 MWe gas-cooled microreactor using an actively cooled helium primary loop and standard low-enriched uranium dioxide fuel. Helium does not corrode, and that fuel type is the same one commercial light water reactors have used for decades, meaning components can be checked and replaced without bespoke supply chains. The whole package is engineered for factory manufacture and shipment in a standard 20-foot container for remote, maritime and defense applications.
One megawatt is modest by grid standards, roughly enough to power 800 homes, but the design targets places where the grid does not exist at all: remote mines, forward operating bases and island communities far beyond the nearest substation.
Three reactors in a month, four before the holiday
Unity did not arrive at the milestone alone. In June, two other reactors also achieved criticality: the Mark-0, a high-temperature sodium heat pipe reactor using entirely passive liquid metal heat pipes, and the Ward 250. Each used a different approach to cooling and fuel geometry, meaning the program simultaneously tested three distinct engineering bets rather than variations on a single design.
A fourth reactor cleared the same threshold in the days that followed. “Achieving criticality in roughly 150 days is a remarkable accomplishment, and Idaho National Laboratory is proud to have provided the facilities and expertise that helped make this milestone possible,” said the laboratory’s director in a press statement.
The federal energy secretary visited the laboratory before the test and inspected a fuel rod, an unusual level of cabinet attention for a zero-power demonstration. The administration had set a goal of three microreactors reaching criticality by the nation’s 250th birthday, and four did so.
What the program does not yet prove
Zero power criticality is a rigorous test, but it is not the same as producing electricity. The next stages each carry their own engineering and regulatory weight: raising the reactor to full thermal output, connecting a generator, running for thousands of hours without unplanned stops, and demonstrating that the containerized package can be transported and reinstalled somewhere remote.
None of the four reactors has completed those steps, and the question of how a commercially deployed microreactor would obtain a full operating license for a remote or mobile site remains open. The program’s structure allowed these machines to bypass the conventional licensing path, but that exemption does not automatically extend to revenue-generating deployment.
Cost discipline will matter enormously when teams move from a single demonstration unit to factory production volumes. For a sense of the capital complexity that larger nuclear projects carry even after funding is secured, a $1.9 billion loan was recently closed to bring a single conventional reactor back online.
What four simultaneous milestones signal for nuclear’s next chapter
The last time the United States saw multiple novel reactor concepts reach criticality within a single month, the programs were government-owned and measured in billions of dollars over decades. This summer’s sequence happened inside startup teams working on compressed private timelines with private capital at risk, which is a fundamentally different industrial model.
Whether that model holds as companies scale from one demonstration unit to a factory line is the question the next two years will answer. Data centers, remote mines and military installations are all hunting for firm, carbon-free generation that does not depend on a transmission line, and the pressure is acute.
The Department of Energy confirmed Unity’s milestone as fulfillment of a precedent-setting directive to reignite nuclear innovation in the United States. The Idaho criticality tests and the broader Duane Arnold restart sit at opposite ends of the nuclear size spectrum, yet both reflect the same underlying reality: the grid needs more firm power faster than conventional construction allows. Four reactors going critical in a single month does not yet mean four new power sources are headed to a remote airstrip, but it does mean the chain reaction, in every sense, has started.
Hugo is an engineer with strong technical expertise. Multilingual from an early age, his writing combines technical clarity with a strong interest in science and energy.
