Seaborg Technologies aims to put compact molten-salt reactors on barges that can be moored outside cities.
4 minute Read
A Danish startup named after an American physicist is working to make nuclear power cheap and scalable—even if it will still be illegal in its home country.
Where standard reactors use dense fuel rods kept under high pressure in heavy containment vessels, molten-salt reactors use liquid fuel and operate at close-to-normal atmospheric pressures, allowing them to be smaller and lighter. In the 1950s, those virtues led the Air Force to explore using molten-salt reactors to power strategic bombers—yes, it may sound like something out of Dr. Strangelove, but it actually happened—but then the plan was ultimately scrapped.
Instead of bombers, Seaborg (which was founded in 2015 and took its name from pioneering nuclear physicist Glenn Seaborg) aims to put compact molten-salt reactors on barges that can be moored outside cities.
“It significantly reduces the cost and complexity,” says cofounder and CEO Troels Schönfeldt of the molten-salt architecture. And it’s safer: The end state of a reactor malfunction in a molten-salt design should be that the fuel congeals into a stable solid.
And waste from this reactor isn’t as high-level radioactive as the leftovers from a standard nuclear plant. “We’ll still have nuclear waste, but that waste will need to be stored for a few hundred years,” Schönfeldt says—not the thousands of years that high-level radioactive waste requires.
Seaborg plans to build these reactors, with Samsung Heavy Industries constructing the barges at a shipyard in Korea, starting in 2026. (The reactor assembly is scheduled to begin around the same time as the power barge.) The company will float them to cities in the tropics needing clean energy, an approach Schönfeldt calls cheaper and faster than on-site construction. He estimates this mass-production-friendly design can initially generate power at $50 per megawatt-hour, slightly above solar and wind costs but well below that of gas and coal.
“We are going straight for southeast Asia and developing countries,” Schönfeldt says, citing Vietnam as a particularly compelling case. “These are fast-growing economies; they need a lot of energy.”
He also sees these target markets as especially ill-suited for solar and wind: The wind varies too much; and during monsoon season, it gets too cloudy.
The minimum size would be a 200-megawatt reactor on a 300-foot barge, although an installation can be scaled up to 800 MW. In comparison, Denmark’s largest offshore wind farm produces 400 MW; the country has no nuclear plants, having voted in 1985 to exclude nuclear from its energy plans; but conventional reactors in France, Europe’s most aggressive adopter of nuclear power, range from 880 MW to 1,500 MW.
Prakash Sharma, vice president for multi-commodity research at the Wood Mackenzie energy-research firm, rejects the idea that solar and wind aren’t reliable enough in the tropics but acknowledges that they can’t provide enough power there.
In Vietnam, for example, Wood Mackenzie estimates that while those two renewable sources provide 12% of the country’s electricity today and can rise to provide 40% by 2050, the overall power output will rise about 265 terawatt-hours today to some 950 TW hours by 20250.
“The potential for power-demand growth is significant,” Sharma says. “They would need to be considering all options, including nuclear.”
That’s also true for the rest of the world, he adds. “There is a realization now that solar and wind cannot meet those climate transition goals,” he says.
Nuclear power can in theory provide that consistent backup power without needing much land, but around the world it requires long-term storage of radioactive waste—and in the U.S., new large nuclear plants are also ruinously expensive. The only civilian project now under construction, a set of two reactors near Augusta, Georgia, is now expected to cost $30 billion.
“There’s definitely momentum toward small-scale nuclear reactors,” Sharma says. “The cost is typically much better than a conventional, large-scale nuclear reactor.”
His employer hasn’t researched Seaborg, in particular, but that startup and others exploring molten-salt designs (including the Bill Gates-backed TerraPower) face the same technical barrier that the Air Force struggled with some 70 years ago: The chemical mixtures involved can corrode reactor structures.
A 2018 Department of Energy report on molten-salt reactor designs, which lists Seaborg among 10 other organizations developing MSR designs, noted the persistence of problems with metal structures cracking or suffering “embrittlement” and suggested launching a “systematic development program” that might identify better candidate alloys by 2022 or 2023.
Acknowledging the risk of this using molten salt—”It eats everything, including your own reactor”—Schönfeldt professes himself confident in Seaborg’s ability to find the right alloy to solve that problem.
“I don’t see that as a big issue,” he says. But the company isn’t describing that alloy yet.
With a 2028 goal to start generating power—“The first commercial product will be there floating and generating electricity in 2028,” as Schönfeldt says without qualification—Seaborg doesn’t have much time to pivot from metallurgical research to bending metal.
“There are a lot of things that can break us,” he allows. “Yes, we are high risk. But at least we are realistic.”
About the author
Rob Pegoraro writes about computers, gadgets, telecom, social media, apps, and other things that beep or blink. He has met most of the founders of the Internet and once received a single-word e-mail reply from Steve Jobs.