The annual Consumer Electronics Show in Las Vegas is an exhibition of new gadgets and technologies that are close to hitting the market. So it’s not where you’d expect to hear about fusion power. After all, one of the oldest jokes in the energy sector is that fusion is a few decades away — and always will be.
And yet when the show opened last week, one of the speakers was Bob Mumgaard, CEO of Commonwealth Fusion Systems, a company spun out from MIT.
Mumgaard told his audience that new AI simulation tools are speeding the development of the company’s prototype fusion reactor, west of Boston, which is scheduled to be demonstrated next year. Another big step is in sight, too: CFS plans to start running a commercial-scale fusion power plant in Virginia in the early 2030s. Google has already agreed to buy half of the electricity generated there.
His message, in other words, was that fusion is finally moving from science fiction to reality. “We can see a new age of abundance,’’ Mumgaard told the crowd.
Fusion would have advantages where other sources of electricity have shortcomings. It wouldn’t emit greenhouse gases. Unlike fission-based nuclear power, which involves splitting atoms, fusing atoms together — the same process that powers the stars — would not generate long-lived radioactive waste. There would be no chance of a runaway chain reaction or a meltdown. And unlike wind and solar, fusion could operate continuously.
If you squint, you may see the contours of the abundant future Mumgaard is talking about. If thousands of fusion plants sprout around the world to augment today’s existing clean electricity sources, we could bend the emissions curve downward even as developing economies ramp up their energy usage.
The elegance of that vision offers a sharp contrast to our current reality. We are on track to overshoot the internationally agreed-upon limit of 1.5 degrees Celsius of global warming above preindustrial levels. Meanwhile, much of the public is indifferent to the problem. Clean energy has powerful
Is it really possible that humankind is nonetheless on the verge of innovating its way out of this mess?
“I’m generally pretty optimistic about all the climate tech stuff overall,’’ Mumgaard told me before his CES talk. “In fusion, I’m
He’s been pursuing the fusion dream since he graduated from college in 2008 and came to MIT for graduate work in plasma physics. That department would soon demonstrate why it had a reputation for innovation. When the federal government said in 2012 that it would cut funding for the experimental fusion reactor on campus, the MIT researchers took that as an opportunity to develop new fusion techniques — which eventually formed the basis of CFS’s plans. Since Mumgaard and his cofounders spun the company out in 2018, CFS has raised almost $3 billion from a wide range of investors, including Google and Bill Gates.
CFS is one of several startups vying to bring a fusion power plant online in the next decade. (One of them just got an investment from President Trump’s media company, but it’s hard to tell if that will meaningfully aid its efforts.) Meanwhile, countries around the world have been renewing or expanding their pursuit of the technology — notably China, which appears to be pulling ahead of the United States in the race to build a working plant.
All this hopeful activity is even more remarkable when you consider that no one has yet generated a fusion reaction that comes anywhere close to being a reliable source of power, let alone an economically viable one.
Fusion works in our sun and all the other stars because they are so massive. Their gravity alone jams hydrogen atoms together. That constantly releases astronomical amounts of energy — which is how we’re all here now.
To replicate this in a power plant will require only modest amounts of fuel — generally made up of two hydrogen isotopes, including one that’s abundant in seawater. But you have to get the material ridiculously hot (CFS intends to crank it to 100 million degrees Celsius) and put it under intense pressure. If you sustain the pressure for long enough, the energy produced by the reaction will greatly exceed the amount that goes into making it happen. That energy can be converted to electricity by capturing it in a system that heats water and spins steam turbines.
This is beautiful in theory. Fusion developers know how the reactions can be made to do what they need to do. Test results have been so promising that Mumgaard doesn’t consider CFS’s prototype reactor to be a science experiment. For him, it’s just a construction project.
“There’s not a lot of decisions or R&D results that are waiting to inform it or anything. It’s baked. It’s being built,’’ he told me. “It’s a put-it-together project. It’s a big set of Legos, but we know how they’re supposed to fit.’’
Even if Mumgaard is right, though, that the pieces of working fusion reactors are beginning to fall into place, we still could be decades away from having substantial amounts of fusion power on the electric grid.
Stephen O. Dean, who has been involved with fusion science for more than 60 years and is president emeritus of Fusion Power Associates, a nonprofit research and education firm, says Mumgaard and his colleagues at CFS are “an outstanding team of people doing a great job.’’ Even so, Dean says their timelines are glossing over a phenomenon first described by Hyman Rickover, a US naval admiral who led the development of nuclear power in the 1950s: When a reactor is still a design on paper, “there’s never a problem. It always works,’’ Dean says. “The real reactors always have problems.’’
Many of the challenges that will beset real fusion plants won’t be apparent until one is actually running. For example, how long will key materials, such as the vessel surrounding the miniature star, last before needing to be replaced?
Obviously, no one can be sure what it will take to solve still-unknown issues. In the history of fusion research, it’s hard to think of a schedule or budget that didn’t get blown. “It takes five years to do things that you thought could be done in one,’’ Dean says.
So even if some fusion plants are up and running in the 2030s, Dean doubts that this first generation will produce power cheaply enough to appeal to electric utilities in the United States. Government support could change that equation. It also could speed the development of subsequent versions of the technology that figure to be less expensive. US government funding for fusion research has been flat for the past two decades, though. The technology may have an easier time getting going in China, which is willing to subsidize state-owned companies and audacious public works projects over long stretches.
By now you may be concluding that fusion is mostly hype because even if these plants arrive, they will be too late to keep us from crashing past 1.5 degrees of global warming. But it will never be too late to stop warming from getting worse. And it’s even possible to bring temperatures back down. With fantastically abundant clean power, carbon removal and other remediation efforts hold out far greater promise.
Mumgaard is well aware that countless things can go wrong. Nevertheless, he sees a flip side to the fact that much remains unknown about how fusion can scale up: It might turn out to be
Mumgaard uses the word “overshoot’’ differently than climate pessimists do: He’s thinking that fusion may start a revolution like the ones unleashed by semiconductors, agricultural breakthroughs, and other inventions that harnessed science in fundamentally new ways.
“The history of technology has been an overshoot,’’ he says. “We make stuff that is significantly better than we thought we were going to make.’’