Nuclear power start-up Deep Fission wants to put a reactor where no reactor has gone before: a mile underground. The California-based company received for its reactor design from the US Department of Energy earlier this month, and is preparing the site for its commercial pilot outside Parsons, a small city in south-east Kansas.
The company hopes that placing a reactor deep underground will cut installation and operational costs, particularly since the bedrock can provide natural containment and pressure for the reactor. This eliminates the need for a heavy-duty concrete containment dome like those that enclose conventional nuclear reactors, which account for a of their cost and the time it takes to build them. But independent engineers say it is unclear how successful the approach will be.
Deep Fission has already drilled a 1830-metre data-collection well, and will soon begin drilling a 762-metre proof-of-concept borehole. Its next phase will involve installing the reactor canister, heat exchanger and other components into the borehole, with a nuclear demonstration targeted for 2027.
Advertisement
The company’s uses one of the most common nuclear technologies: a pressurised water reactor, which runs on standard low-enriched uranium fuel and is water-cooled. It is , but less than a metre in diameter, and is designed to produce 15 megawatts of electricity.
Drilling deep boreholes is routine in the oil and gas industry, and the heat exchanger technology that will create electricity is standard in geothermal energy production: the heat created by the reactor converts water to steam, which rises to the surface and turns a turbine.
While the components of Deep Fission’s model aren’t novel, combining them is. “The biggest technical challenge is proving that all of these individually familiar technologies will work together in this unfamiliar configuration,” says , a nuclear engineer at independent consulting firm Neutronic Designs and a specialist in advanced reactor technologies.
The company thinks its design will not only work, but improve safety and reduce costs compared with conventional nuclear power projects. Putting the reactor a mile underground means the surrounding rock will serve as a built-in containment structure, and the weight of the water column above will create a pressure of 160 atmospheres – high enough to keep water in liquid form at ultra-high temperatures and eliminating the need for a pressuriser that is a necessary component of ground surface reactors.
, Deep Fission’s chief operating officer, says the reactor design also cuts the need for “engineered active emergency core cooling systems with pumps and backup power”.
“We anticipate that could translate into real capital and operating savings,” he says.
“They’re using the environment itself to perform functions that we normally have to engineer into a nuclear power plant,” says Dewan. However, she adds, the trade-off is a lack of easy access to the reactor, which will make inspection and maintenance “both more difficult and more important”.
, a nuclear engineer at the University of Michigan, sees the potential for maintenance problems, too. Even if Deep Fission designs the Gravity reactor for minimal maintenance, he says – for example, by cooling based on natural circulation rather than pumps – it is impossible to prepare for all eventualities. “I’d be more concerned about something happening that you didn’t anticipate,” says Allen.
Brasel says the company isn’t planning to raise the reactor to the surface for routine maintenance, but will be able to do so if it is needed.
No reactor has ever operated at this depth; the closest is Norway’s , which was installed inside a cavern excavated into a mountainside about 100 metres underground, also for containment purposes. It is uncertain how Deep Fission’s borehole, casing, reactor vessel and surrounding geology will hold up over potentially decades of operation. “Just due to the fact that you’re putting a reactor underground, there will be a learning curve,” says Allen.
It is a curve the company hopes to overcome quickly. Its modular approach would ultimately group multiple Gravity reactors at one site, scaling the 15-megawatt capacity to 150 megawatts or even over a gigawatt.
Deep Fission isn’t alone on its quest to make nuclear power smaller, cheaper and easier to build. Small modular reactors are intended to replace conventional nuclear power with smaller standardised units that can be manufactured and deployed in series. Several companies have made headway in their efforts: the US Nuclear Regulatory Commission issued TerraPower a construction permit in March for its Natrium reactor in Wyoming. Kairos Power and X-Energy are also moving reactors towards construction.
A demonstrated the capacity to run stable and self-sustaining reactors by 4 July 2026 – a “criticality deadline” set by the US Department of Energy’s Reactor Pilot Program that increases the odds of the companies in question finding a path to commercial deployment.
Small modular reactor technology could challenge long-standing assumptions within the nuclear industry. For instance, for many companies working on small modular reactors, “their business bet is that the changes in technology and improvements in safety will mean you won’t need containment”, says Allen. “The question with the Deep Fission idea, where you’re going to the work of drilling a very deep borehole, is: is it necessary?”
For now, Deep Fission is betting that it is, and will continue to work towards showing that a reactor can work at the bottom of a hole. If it succeeds, it could offer a radically different way to build small nuclear plants.
The most valuable experimental results to look forward to, says Dewan, will be “demonstrating the installation and retrieval of the prototype reactor equipment, measuring thermal performance, and quantifying their ability to inspect and maintain the equipment over its operating lifetime”.
