Four Advanced US Reactors Achieve Criticality, Marking First New Non-Light-Water Designs in 40 Years

The prevailing narrative around the American nuclear renaissance assumes it will be driven by resurrecting the past—restarting dormant gigawatt-scale plants like Three Mile Island or Palisades to feed the insatiable appetite of hyperscale data centers. But the most significant breakthrough in nuclear technology in a generation did not happen at a massive concrete cooling tower.

It happened quietly in the high deserts of Idaho and Utah. Over a single 30-day window this summer, the United States saw four brand-new, advanced non-light-water reactors achieve zero-power criticality, proving that the atomic age can be miniaturized, accelerated, and fundamentally redesigned.[2][5]

Between June 4 and July 4, 2026, four separate startups—Antares Nuclear, Valar Atomics, Deployable Energy, and Aalo Atomics—successfully initiated controlled nuclear chain reactions in their respective microreactors. This cluster of milestones marks the first time in more than 40 years that a privately developed non-light-water reactor has reached criticality in the United States.

Driven by a May 2025 executive order that mandated the Department of Energy to establish a pilot pathway and bring at least three advanced designs online by the nation's 250th anniversary, the industry beat the deadline and overdelivered.[1][2][4]

The speed of deployment represents a structural break from traditional nuclear timelines. Antares Nuclear’s Mark-0 reached criticality on June 4 at the Idaho National Laboratory, becoming the 53rd reactor built at the site since 1951. Valar Atomics followed on June 18 at its own San Rafael Energy Lab in Utah, proving these systems can be sited outside federal reservations.

Deployable Energy’s high-temperature gas-cooled Unity reactor hit the mark on June 30, and Aalo Atomics completed the sweep on July 4 with its Aalo-X test core. In an industry accustomed to decade-long construction schedules, these developers went from authorization to criticality in a matter of months.[2][4][5]

These machines bear little resemblance to the pressurized water reactors that currently provide nearly 20 percent of United States electricity. They are microreactors, designed to produce between 1 and 20 megawatts of power, and are small enough to be transported by standard shipping containers or flatbed trucks.

More importantly, they completely abandon water as a primary coolant. Antares utilizes a sodium heat-pipe architecture, while Deployable Energy relies on high-temperature gas. This fundamental engineering shift eliminates the need for the massive, active pumping systems and proximity to large bodies of water that strictly constrain traditional nuclear siting, allowing them to operate in arid environments.[4][5]

Instead of standard uranium fuel rods, several of these advanced designs rely on TRISO (tri-structural isotropic) particle fuel containing high-assay low-enriched uranium, commonly known as HALEU. TRISO encapsulates the uranium in microscopic layers of carbon and ceramics, effectively creating an indestructible containment vessel around each individual fuel particle.

This passive safety feature ensures the fuel cannot melt down even under extreme temperatures or complete loss of coolant, fundamentally altering the risk profile. By eliminating the possibility of a catastrophic meltdown, developers can safely deploy these reactors much closer to population centers, industrial facilities, or critical infrastructure without requiring massive evacuation zones.[5]

It is crucial to define what these recent criticality tests actually prove. All four events were zero-power demonstrations. The reactors sustained a controlled fission chain reaction, validating the core physics, the fuel performance, and the regulatory authorization process. However, they were run at extremely low heat levels and were not connected to steam turbines. They did not produce commercial electricity. The Department of Energy views this pilot program as a necessary research and development exercise to compress the timeline for subsequent commercial licensing by the Nuclear Regulatory Commission.[2][5]

The immediate market pull for these systems is coming from off-grid and highly resilient applications where traditional grid connections are either impossible or too slow. The Department of Defense has closely monitored the pilot program, viewing microreactors as a strategic solution for powering remote military installations without relying on vulnerable diesel fuel supply chains, while developers have even pitched the technology for future lunar surface habitats.

Simultaneously, the data center industry—currently constrained by multi-year grid interconnection delays—is aggressively evaluating microreactors as a way to co-locate dedicated, zero-carbon baseload power directly adjacent to new hyperscale computing facilities.[3][6]

Ultimately, the success of the Department of Energy’s Reactor Pilot Program shifts the primary bottleneck of the nuclear industry from bespoke construction to standardized manufacturing. If a reactor can be built on an assembly line, shipped on a standard truck, and brought to criticality in under a year, the economics of atomic energy change from civil engineering mega-projects to scalable product lines. The next critical hurdle is translating these zero-power physics demonstrations into sustained, grid-connected electrons, a phase developers anticipate reaching by 2027 as they move toward commercial licensing.[4]