Skip to Content

A New Approach to Fusion

A startup snags funding to start early work on a low-budget test reactor.

General Fusion, a startup in Vancouver, Canada, says it can build a prototype fusion power plant within the next decade and do it for less than a billion dollars. So far, it has raised $13.5 million from public and private investors to help kick-start its ambitious effort.

Power pistons: General Fusion’s reactor is a metal sphere with 220 pneumatic pistons designed to ram its surface simultaneously. The ramming creates an acoustic wave that travels through a lead-lithium liquid and eventually accelerates toward the center into a shock wave. The shock wave compresses a plasma target, called a spheromak, to trigger a fusion burst. The thermal energy is extracted with a heat exchanger and used to create steam for electricity generation. To produce power, the process would be repeated every second.

Unlike the $14 billion ITER project under way in France, General Fusion’s approach doesn’t rely on expensive superconducting magnets–called tokamaks–to contain the superheated plasma necessary to achieve and sustain a fusion reaction. Nor does the company require powerful lasers, such as those within the National Ignition Facility at Lawrence Livermore National Laboratory, to confine a plasma target and compress it to extreme temperatures until fusion occurs.

Instead, General Fusion says it can achieve “net gain”–that is, create a fusion reaction that gives off more energy than is needed to trigger it–using relatively low-tech, mechanical brute force and advanced digital control technologies that scientists could only dream of 30 years ago.

It may seem implausible, but some top U.S. fusion experts say General Fusion’s approach, which is a variation on what the industry calls magnetized target fusion, is scientifically sound and could actually work. It’s a long shot, they say, but well worth a try.

“I’m rooting for them,” says Ken Fowler, professor emeritus of nuclear engineering and plasma physics at the University of California, Berkeley, and a leading authority on fusion-reactor designs. He’s analyzed the approach and found no technical showstoppers. “Maybe these guys can do it. It’s really luck of the draw.”

The prototype reactor will be composed of a metal sphere about three meters in diameter containing a liquid mixture of lithium and lead. The liquid is spun to create a vortex inside the sphere that forms a vertical cavity in the middle. At this point, two donut-shaped plasma rings held together by self-generated magnetic fields, called spheromaks, are injected into the cavity from the top and bottom of the sphere and come together to create a target in the center. “Think about it as blowing smoke rings at each other,” says Doug Richardson, chief executive of General Fusion.

On the outside of the metal sphere are 220 pneumatically controlled pistons, each programmed to simultaneously ram the surface of the sphere at 100 meters a second. The force of the pistons sends an acoustic wave through the lead-lithium mixture, and that accelerates into a shock wave as it reaches the plasma, which is made of the hydrogen isotopes deuterium and tritium.

If everything works as planned, the plasma will compress instantly and the isotopes will fuse into helium, releasing a burst of energy-packed neutrons that are captured by the lead-lithium liquid. The rapid heat buildup in the liquid will be extracted through a heat exchanger, with half used to create steam that spins a turbine for power generation, and the rest used to recharge the pistons for the next “shot.”

The ultimate goal is to inject a new plasma target and fire the pistons every second, creating pulses of fusion reactions as part of a self-sustaining process. This contrasts with ITER, which aims to create a single fusion reaction that can sustain itself. “One of the big risks to the project is nobody has compressed spheromaks to fusion-relevant conditions before,” says Richardson. “There’s no reason why it won’t work, but nobody has ever proven it.”

He says it look longer than expected to raise the money for the prototype project, but the company can now start the first phase of building the test reactor, including the development of 3-D simulations and the technical verification of components. General Fusion aims to complete the reactor and demonstrate net gain within five years, assuming it can raise another $37 million.

If successful, it believes it can build a grid-capable fusion reactor rated at 100 megawatts four years later for about $500 million, beating ITER by about 20 years and at a fraction of the cost.

“I usually pass up these quirky ideas that pass my way, but this one really fascinated me,” says Fowler. He notes that there are immense challenges to overcome, but the culture of a private startup may be what it takes to tackle them with a sense of urgency. “In the big programs, especially the fusion ones, people have gotten beat up so much that they’ve become so risk averse.”

General Fusion’s basic approach isn’t entirely new. It builds on work done during the 1980s by the U.S. Naval Research Laboratory, based on a concept called Linus. The problem was that scientists couldn’t figure out a fast-enough way to compress the plasma before it lost its donut-shaped magnetic confinement, a window of opportunity measured in milliseconds. Just like smoke rings, the plasma rings maintain their shape only momentarily before dispersing.

Nuclear-research giant General Atomics later came up with the idea of rapidly compressing the plasma using a mechanical ramming process that creates acoustic waves. But the company never followed through–likely because the technology to precisely control the speed and simultaneous triggering of the compressed-air pistons simply didn’t exist two decades ago.

Richardson says that high-speed digital processing is readily available today, and General Fusion’s mission over the next two to four years is to prove it can do the job. Before building a fully functional reactor with 220 pistons on a metal sphere, the company will first verify that smaller rings of 24 pistons can be synchronized to strike an outer metal shell.

Glen Wurden, program manager of fusion energy sciences at Los Alamos National Laboratory and an expert on magnetized target fusion, says General Fusion has a challenging road ahead and many questions to answer definitively. Can they produce spheromaks with the right densities, temperature, and life span? Can they inject two spheromaks into opposite ends of the vortex cavity and make sure they collide and merge? Will the acoustic waves travel uniformly through the liquid metal?

“You can do a good amount of it through simulations, but not all of it,” says Wurden. “This is all very complex, state-of-the-art work. The problem is you’re dealing with different timescales and different effects on materials when they’re exposed to shock waves.”

Los Alamos and General Fusion are collaborating as part of a recently signed research agreement. But Richardson isn’t planning on a smooth ride. “The project has many risks,” he says, “and we expect most of it to not perform exactly as expected.” However, if the company can pull off its test reactor, it hopes to attract enough attention to easily raise the $500 million for a demonstration power plant.

Says Fowler, “Miracles do happen.”

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

OpenAI teases an amazing new generative video model called Sora

The firm is sharing Sora with a small group of safety testers but the rest of us will have to wait to learn more.

Google’s Gemini is now in everything. Here’s how you can try it out.

Gmail, Docs, and more will now come with Gemini baked in. But Europeans will have to wait before they can download the app.

This baby with a head camera helped teach an AI how kids learn language

A neural network trained on the experiences of a single young child managed to learn one of the core components of language: how to match words to the objects they represent.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.