Inside TerraPower’s cavernous 65,000 square-foot laboratory north of Seattle, there’s a suite of stations for experimenting with new materials and cutting-edge methods for generating nuclear power.
One area is dominated by a two-story, silver tower where engineers test the flow of liquid over mock reactor fuel rods. An enclosed space nearby features a network of insulated pipes that bear a vague resemblance to an oversized hamster Habitrail, but in fact are used to study heat transfer between melted salts and metal. Inside an oxygen-free sealed glove box, molten sodium spins in a glass beaker.
“We are definitely in a mode this year of just watch us.”
– Chris Levesque, TerraPower CEO and president
The engineers here are fine-tuning the technology needed for building small modular nuclear reactors — a design that aims to be cheaper and faster to construct than traditional reactors and that could help save the planet from the ruinous impacts of climate change.
TerraPower is among a slate of U.S. companies developing the downsized reactors, but other lead contenders have recently stumbled in their pursuits. And now the Bill Gates-backed venture is in a strong position to be the first to deploy the next generation technology in America.
This month, the company expects to submit to the Nuclear Regulatory Commission an application for the certification of its Natrium reactor, which it developed with GE Hitachi Nuclear Energy. This spring, it’s scheduled to break ground on a demonstration power plant, located near a soon-to-be-retired coal power plant in Kemmerer, Wyo. Looking further ahead, the company is exploring a potential deal with the United Arab Emirates that could lead to the purchase of hundreds of reactors.
“We’re just doing it,” said Chris Levesque, TerraPower’s CEO and president. “We’ve got the [private] investment. We have the government investment. We bought the land. Yes, we’ll need permission from the NRC. But we are definitely in a mode this year of just watch us.”
The company has raised about $1 billion, including investments from Gates as well as South Korea-based SK Inc. and SK Innovation, according to PitchBook. It has additionally been awarded roughly $2 billion from the U.S. Department of Energy.
The need for nuclear is urgent. The world is on course to get dangerously hot and nuclear power is expected to play an essential role in tempering climate change. It can provide a carbon-free energy source 24/7, complementing the intermittent power coming from wind and solar. And demand keeps rising as the use of artificial intelligence and large language models expands, requiring more power-guzzling computing.
At the United Nations climate conference in December, more than 20 countries including the U.S. pledged to triple nuclear energy capacity worldwide by 2050. Next week, TerraPower representatives will attend an inaugural international nuclear energy summit in Brussels, an event that builds on the global commitments.
TerraPower’s goal is to start splitting atoms in Wyoming by 2030 — a target that has been decades in the making.
Rethinking nuclear power
In the mid 2000s, Microsoft co-founder Gates was talking with Nathan Myhrvold, the company’s former chief strategist and technology officer. Myhrvold had by that point moved on to create Intellectual Ventures, which supports the development of boundary-pushing innovation. Gates was no longer CEO, and was spending more time on global health initiatives.
The two were considering the world’s growing hunger for energy and how to meet it without further destroying the planet.
“[T]here seemed to be a huge opportunity to rethink nuclear power,” Myhrvold reflected in a 2018 Scientific American essay. “Most of the reactors operating around the world — including the ones at Fukushima and almost all of the 100 or so plants operating in the U.S. — were built from designs drafted during the slide-rule era.”
So in 2006 in Bellevue, Wash., with support from Gates and others, Intellectual Ventures launched TerraPower.
In the years since, the company added the Everett lab north of Seattle and assembled a team of 560 employees, including top nuclear physicists and engineers. It recruited expertise from the U.S. Navy’s long-standing nuclear submarine program — including Levesque, who worked on the subs for 14 years — as well as from traditional nuclear plant operators. Gates serves as chairman of TerraPower’s board and Myhrvold as vice chairman, the latter collaborating recently with company employees to brainstorm designs for the new plant’s instrumentation and controls.
“I didn’t know what innovation was until I came to work for Bill Gates and Nathan Myhrvold,” Levesque said. “When we were in the Navy or working for today’s [traditional] nuclear industry, we were paid not to innovate. We were punished if we innovated. It’s like, you’re out of control if you do something new.”
That is not the mantra for TerraPower.
The company’s Natrium reactor could become the first sodium-cooled, power-generating reactor in the U.S., building on technology used in an experimental breeder reactor in Idaho that operated for nearly 30 years before shutting down. All of the existing commercial power plants in the U.S. today are cooled with water — an approach developed decades ago.
The company is applying the newest technologies to the task, taking the modeling it used to design the facility and evolving it into a “digital twin” that will use AI to learn directly from the physical plant once it’s operating.
“The first reactor is going to be great,” Levesque said, “but our digital twin is going to tell us how we can improve maintenance, how we can optimize the plant further, tell us what design changes we should be pursuing.”
TerraPower’s advantages
Despite significant progress across the field, it has been a tough journey for the cohort of companies striving to modernize America’s nuclear power sector. They’ve had to rely on a workforce that has undergone years of attrition and a diminished supply chain. They’ve battled economic headwinds driving up the price of materials and labor.
Those challenges triggered some troubling setbacks for the sector this past fall.
“The backing of a Bill Gates is absolutely a major positive for TerraPower.”
– David Brown, director of energy transition service at the analytics firm Wood Mackenzie
Rising costs forced NuScale Power to unexpectedly nix a long-planned project in Idaho, despite having won certification from the NRC — the commission’s first approval for a small modular reactor. X-Energy scuttled its $2 billion plan to go public and announced layoffs at its advanced reactor company. Oklo, which is backed by OpenAI’s Sam Altman, had tentatively won a military bid for its microreactors, but the offer was rescinded in November.
The companies are still forging ahead, but have been forced to readjust their plans.
TerraPower has meanwhile carried on, seemingly benefiting from Gates’ deep pockets and his patience with the pursuit of a first-of-a-kind project that carries daunting financial unknowns.
“The backing of a Bill Gates is absolutely a major positive for TerraPower in terms of the ability to wait and to ride out questions around costs,” said David Brown, director of energy transition service at the analytics firm Wood Mackenzie.
The Natrium project has another advantage thanks to a unique design feature that could give it an edge in the marketplace. The company’s nuclear reactor will produce heat that generates steam that spins turbines producing electricity — just as most power plants do. But the Natrium plant will also take some of the reactor’s heat and transfer it to a vessel of molten salt, creating a thermal battery.
The heat storage technology is what the Habitrail-like setup at the laboratory is set up to explore. The researchers are testing strategies for moving the liquid sodium metal and molten salts and transferring the heat between them.
“That’s something that’s never, that we know of, been done before. So we’re validating our assumptions for our heat exchanger,” said Monica Tarqueno, TerraPower’s director of lab operations.
The feature would allow a plant to store energy from the reactor when electricity is less in demand and prices are lower. When demand and prices rise, that stored heat can be tapped to produce more power.
“That’s significant,” Brown said, “because the power market is volatile in terms of prices.”
TerraPower’s leaders also emphasize the design and technology’s safety features — which are an inherent concern in nuclear power, even with the strong track record in the U.S.
The reactor on its own should produce 345 megawatts of power. With the battery, it could deliver up to 500 megawatts for more than five hours. At maximum output, the system should generate enough energy for 400,000 homes.
Potential for growth
While some key factors are aligning for TerraPower, there are plenty of challenges ahead.
TerraPower needs to clear essential regulatory hurdles with the NRC, and approval isn’t assured. Two years ago, the commission denied Oklo’s application to build and run a microreactor.
And costs are always a concern, whether or not there’s a billion in your corner. Levesque declined to provide an updated price tag for the Kemmerer project. In September 2022, Chief Financial Officer Marcia Burkey put the figure at $4 billion.
TerraPower leaders remain concerned about the sector’s supply chain, which needs to keep ramping up. That includes American efforts to restart and scale production of a specific type of uranium fuel the U.S. stopped making more than 70 years ago — moving instead on Russian sources.
That supply was cut off when Russia invaded Ukraine, ultimately forcing TerraPower to push back the planned 2028 start date of the Wyoming demonstration project by two years.
There is progress on America’s fuel production capabilities. In November, an Ohio company manufactured the first 20 kilograms of the uranium fuel, and the U.K. government recently announced that it will also invest in domestic fuel production.
And there are tailwinds that could bolster TerraPower and other companies’ efforts.
U.S. politicians from both sides of the aisle have supported funding and legislation to aid in nuclear power deployment. The House this month overwhelmingly approved a bill aiming to accelerate the approval of new projects, and the Senate has a similar proposal in the works.
Recent research from Wood Mackenzie reveals that the market for small modular reactors is growing internationally. The pipeline of small nuclear projects worldwide reached 22 gigawatts early this year, an increase of 65% since 2021. While customers remain cautious, projects are being considered and pursued in the U.S., Poland, Canada, the United Kingdom and South Korea.
“There’s significant interest in SMRs both within the United States and outside,” said Brown, of Wood Mackenzie. “But buyers are still investigating the potential for the technology and how it fits into their portfolios.”
TerraPower in December announced a memorandum of understanding with the UAE’s Emirates Nuclear Energy Corporation (ENEC), to explore the possibility of selling reactors to the nation — possibly hundreds of reactors, according to Gates. Mohamed Al Hammadi, head of the ENEC, is a TerraPower board member.
“Many of these small companies see the export market really as their entree to quantity production, which is one of the keys to inexpensive power plants,” said Roger Blomquist, a principal nuclear engineer.
Blomquist has worked for more than four decades at Argonne National Laboratory — the lab that operated the experimental reactor on which TerraPower is building its technology.
After so many years in a sector whose public support has cycled through ebbs and flows, he’s loathe to proclaim TerraPower’s progress or renewed U.S. interest in commercial nuclear power a “renaissance” or “rebirth.”
“I’m old enough not to be opening the champagne yet,” Blomquist said. “But I think that the stars are much better aligned.”
