Idaho National Laboratory is preparing to build and operate its first new nuclear reactors in roughly 50 years, ending a construction drought at the Department of Energy site that stretches back to the 1970s. The centerpiece of this effort is MARVEL, a compact microreactor that produces about 85 kilowatts of thermal energy converted to approximately 20 kilowatts of electricity, enough to power around 10 homes. While that output sounds modest, the real significance lies in what MARVEL represents: a federal testing platform designed to let private companies prove their reactor designs work before deploying them commercially.
MARVEL’s Path From Blueprint to First Criticality
The MARVEL team recently completed the 90% final design for the reactor, and physical fabrication is already underway. The unit will be sited at INL’s Transient Reactor Test Facility, known as TREAT, which has hosted nuclear experiments for decades but has not seen a new reactor built on its grounds since the era of disco and Watergate.
In parallel with design work, engineers have begun fabrication activities on key reactor components, including the primary vessel and heat exchangers that will move energy from the core to external systems. By advancing design and manufacturing together, the project team is trying to avoid the long pauses that have plagued earlier nuclear demonstration efforts.
The Department of Energy’s Office of Nuclear Energy has published a detailed schedule for what comes next. Assembly completion is expected in 2026, with installation at the TREAT facility pit slated for late that same year. Initial criticality, the point at which a sustained nuclear chain reaction begins, is anticipated in 2027. Full-power operations are projected for 2028, with the reactor running through 2030. That timeline, if it holds, would give the United States a functioning government-owned microreactor test bed within roughly three years.
For readers outside the nuclear industry, the practical takeaway is straightforward. MARVEL is not meant to feed electricity into the grid. It is meant to prove that very small reactors can work reliably and safely, so that private developers can take similar designs and deploy them at remote military bases, mining operations, disaster relief zones, and communities that lack access to centralized power infrastructure. As INL officials have emphasized, the project is about demonstrating how microreactors can provide flexible electricity and heat in places where traditional power plants are impractical.
A White House Push to End Five Decades of Inaction
The federal government has acknowledged, in unusually blunt terms, that INL’s reactor-building capabilities atrophied after new construction concluded in the 1970s. A May 2025 executive order on reforming nuclear testing at the Department of Energy directed the agency to accelerate reactor demonstration pathways and reduce bureaucratic friction that had kept advanced designs stuck in regulatory limbo.
That executive order matters because it reframes reactor testing as a national priority rather than a niche research activity. The order specifically targets the gap between laboratory-scale experiments and commercial deployment, a gap that has allowed countries like China and Russia to move ahead on small modular and microreactor designs while American developers waited for test slots and approvals. It instructs DOE to streamline internal reviews, coordinate more closely with nuclear safety regulators, and make better use of national laboratory infrastructure such as INL.
Whether the order’s directives translate into lasting structural change or amount to a policy statement with limited follow-through will depend on sustained funding and agency execution over the next several budget cycles. For now, MARVEL and associated projects at INL function as early tests of the administration’s promise to modernize how the United States evaluates nuclear technologies.
DOME: The Larger Test Bed Taking Shape
MARVEL is not the only reactor infrastructure rising at INL. The Demonstration of Microreactor Experiments facility, called DOME, is a separate DOE test bed designed to host fueled reactor experiments at a larger scale, supporting thermal output up to approximately 20 megawatts. The facility features roughly 80 feet of diameter floor space and is currently midway through construction, with the first fueled experiment potentially beginning in 2026.
The DOE has already awarded $5 million to two companies, Radiant and Westinghouse, for second-phase design work under the DOME DEEP program. Westinghouse has completed its front-end engineering study for the first eVinci microreactor experiment, moving that project past the conceptual stage and into structured, DOE-reviewed deliverables. These are not paper exercises. The phased pathway, from initial feasibility studies through detailed experiment design, is intended to ensure that when a company brings a reactor to DOME, the facility and the experiment are ready for each other.
DOME is designed with multiple shielding configurations, test bays, and support systems so that different reactor concepts can be swapped in and out over time. That flexibility is central to DOE’s strategy: instead of building a custom facility for each new design, DOME is meant to function as a reusable platform that can host a succession of microreactors as the technology matures.
INL has also announced initial selections for the first MARVEL experiments, signaling that private-sector participants are already lining up to use the platform once it reaches operational status. The selection of outside users this early in the process suggests INL is treating MARVEL less as a government science project and more as shared infrastructure for an emerging industry.
Those initial experiments will focus on how microreactors integrate with advanced energy systems, including hydrogen production and thermal storage, as well as how they can support isolated microgrids. By testing these applications on a small scale at INL, developers hope to de-risk future deployments at military installations, Arctic communities, and industrial sites that require continuous, low-carbon power.
Fuel Innovation Running in Parallel
Building reactors is only half the challenge. The fuel that goes into them has to be manufactured, tested, and delivered on schedule. INL has made progress on both fronts, particularly for high-assay, low-enriched uranium fuels suited to advanced reactors.
The lab received its first delivery of TRISO fuel for Project Pele, a Defense Department mobile reactor program, with a next batch expected to follow in 2026. TRISO particles encase uranium fuel kernels in multiple ceramic layers, creating a robust barrier that can retain fission products even under extreme conditions. Project Pele is intended to show that a compact, factory-built reactor can be transported to remote locations and operated safely for military missions.
TRISO fuel is also relevant to civilian microreactors, many of which rely on similar particle-based designs to achieve high temperatures and passive safety. By qualifying TRISO for defense applications, INL and its partners are helping establish a broader supply chain that commercial developers can tap into as their own projects move forward.
Separately, INL produced the first batch of fuel for what would be the world’s first fast-spectrum molten-salt reactor, with the team anticipating delivery of additional batches by March 2026. Fast-spectrum molten-salt concepts dissolve nuclear material in a circulating liquid salt, allowing the reactor to operate at high temperatures and potentially use fuel more efficiently than conventional light-water reactors. Demonstrating that such fuel can be fabricated, handled, and shipped safely is a prerequisite for any future test reactor using this technology.
These fuel advances are tightly coupled to MARVEL and DOME. Microreactors and advanced test beds cannot operate without reliable fuel supplies, and fuel vendors are unlikely to invest in new production lines without clear demand from demonstration projects. By moving fuel development and reactor construction forward together, INL and DOE are trying to break that chicken-and-egg cycle.
What Success Would Look Like
If MARVEL, DOME, and associated fuel programs stay on track, the United States could enter the next decade with a functioning ecosystem for testing and refining microreactor designs. Companies would have access to government-owned facilities where they can run fueled experiments, gather performance data, and validate safety features under real operating conditions. INL would regain its role as a central hub for nuclear innovation rather than a museum of past achievements.
Success would not be measured solely in megawatts. It would also be measured in shortened development timelines, reduced regulatory uncertainty, and a pipeline of designs ready for deployment in places that currently rely on diesel generators or have no reliable power at all. For communities facing harsh climates or limited infrastructure, a proven microreactor could mean stable electricity for clinics, schools, and water treatment plants.
None of this is guaranteed. Construction schedules can slip, budgets can tighten, and political priorities can shift. But after five decades without a new reactor at INL, the combination of a dedicated microreactor test bed, a larger experimental facility, and parallel fuel innovation represents a concrete attempt to rebuild the nation’s nuclear testing capability. How well that attempt succeeds will shape not only the future of microreactors, but also the broader question of whether advanced nuclear energy can move from PowerPoint slides to real-world deployment at the pace its advocates envision.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
