Three small nuclear reactors are now on a path to operate at Idaho National Laboratory, with the Energy Department tracking fueled experiments by Westinghouse, Radiant, and the Department of Defense’s Project Pele for testing inside a single facility called DOME. As of early July 2026, a third advanced reactor has already reached criticality under the agency’s accelerated program, and Radiant has begun a scheduled test campaign. The convergence of three separate reactor projects on one test bed raises a pointed question: who gets to go first, and does that sequence determine which design reaches commercial licensing before the others?
Why three reactors sharing one test bed changes the race
DOME is a repurposed containment structure, originally built for the Experimental Breeder Reactor-II at Idaho National Laboratory, and it is the only facility of its kind in the world designed to host privately developed, fueled microreactor experiments. Its thermal capacity tops out at 20 MWth, according to the DOE Office of Nuclear Energy. That single-facility constraint means each developer’s test campaign must be scheduled around the others. The result is a de facto queue, and the developer that completes fueled testing first will hold months or even years of operational data that competitors lack when they approach the Nuclear Regulatory Commission for commercial licensing.
DOE has framed DOME as a centerpiece of its microreactor strategy, emphasizing that the test bed will give private companies a way to demonstrate real hardware instead of relying solely on models. In an announcement describing the initial experiments selected for the facility, the department highlighted how shared infrastructure lowers costs and shortens timelines for multiple developers at once. But that shared approach also forces those developers into a common bottleneck: only one fueled system can occupy the test bed at a time.
DOE made conditional selections for Westinghouse and Radiant to conduct the first fueled microreactor tests in DOME. The department stated the first fueled reactor experiment would start as early as spring 2026, and follow-on public updates reiterated that these two designs remained at the front of the line. Radiant has already begun its scheduled test campaign, giving it an early operational foothold. The Defense Department’s Project Pele, a separate microreactor built by BWXT Advanced Technologies for military use, is also advancing demonstration site work at the same Idaho campus. DOE projected that Pele could become one of the earliest advanced units to operate in the United States as soon as 2026.
The practical tension is straightforward. DOME cannot run three fueled experiments simultaneously. Each campaign requires installation, safety reviews, fuel loading, testing, and decommissioning steps before the next developer can move in. Whichever company accumulates real-world performance data first gains a measurable head start in proving its design to regulators and potential customers, from data centers to remote military bases. That sequencing advantage is informal, but it is real, and it may influence which design reaches long-term power purchase agreements and serial production first.
Environmental clearances, safety approvals, and reactor specifications
The regulatory path to fueled testing inside DOME cleared two major administrative gates in 2025. DOE published a Final Environmental Assessment for DOME Test Bed Operations in May 2025 and followed with a Final Finding of No Significant Impact in June 2025, both documented under DOE/EA-2268. Those determinations allowed the facility to accept fueled experiments without a full environmental impact statement, removing what could have been a multi-year delay and signaling that, from a federal perspective, microreactor testing at the site posed no significant environmental risk with appropriate controls.
Alongside the private-sector experiments, INL is advancing its own microreactor called MARVEL. DOE’s Idaho Operations Office approved the Preliminary Documented Safety Analysis for MARVEL, a step required before the reactor can reach initial criticality. MARVEL is rated at 85 to 100 kWth and roughly 20 kWe, making it far smaller than the 20 MWth ceiling DOME can accommodate. The safety approval for MARVEL is significant because it demonstrates the regulatory pathway that Westinghouse, Radiant, and Project Pele will each need to navigate for their own designs, including detailed accident analyses, emergency planning, and fuel handling procedures.
The broader DOE program hit a separate milestone in early July 2026, when the agency confirmed that a third advanced reactor achieved criticality. That announcement framed the achievement as meeting a presidential goal for accelerated reactor deployment. While the three DOME-bound microreactors are distinct projects from the reactors counted in that tally, the milestone reflects the speed at which DOE is pushing new nuclear technology from paper designs to operating hardware. In that context, DOME is less an isolated facility than a key node in a broader push to normalize advanced reactor operation across multiple sites and use cases.
Unanswered scheduling and the identity of the third developer
DOE has said it was tracking three developer experiments for potential first operation in DOME starting as soon as 2026. Westinghouse and Radiant are named explicitly. But the identity of the third microreactor developer referenced in DOE’s own tracking has not been disclosed in publicly available primary documents. That gap matters because the unnamed third entrant could be competing for the same DOME time slots, and outside observers cannot assess the full competitive picture without knowing who it is or how mature its design may be.
Public-facing DOE materials that highlight three microreactor efforts to watch beginning in 2026 confirm that multiple designs are being advanced in parallel but stop short of providing a detailed Gantt chart of when each will occupy DOME. Instead, they emphasize the diversity of applications-from remote installations to potential grid-support roles-while leaving the precise order of testing open. For investors and potential customers trying to map out which vendor will be first to market, that lack of specificity makes it difficult to translate technical progress into commercial timelines.
Equally unclear is how DOE will allocate DOME’s 20 MWth capacity across the three projects over time. No primary documentation spells out a published test sequence, priority framework, or scheduling calendar. The conditional selections for Westinghouse and Radiant confirm intent, and Radiant’s active test campaign confirms execution, but the order in which remaining campaigns will proceed has not been made public. It is also unknown whether DOE will reserve contingency windows for troubleshooting or extended testing that could push other developers further back in the queue.
For energy companies, defense contractors, and utilities watching these experiments, the next development to track is whether DOE releases a formal scheduling plan for DOME or whether the sequencing continues on an ad hoc basis driven by readiness and funding. The company that finishes its fueled test campaign first will likely be in the strongest position to move into follow-on demonstration projects and early commercial deployments, while those that wait may have to differentiate on cost, safety case, or specific customer needs rather than on simple timing. Until DOE clarifies how it will juggle three high-profile microreactors inside a single test bed, the race to be first will remain as much about access to DOME as about the underlying reactor designs themselves.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
