X

Fusion Power Would Revolutionize Energy, if We Could Just Get It Working

Scientists and startups are edging closer to realizing a decades-old dream.

Stephen Shankland Former Principal Writer
Stephen Shankland worked at CNET from 1998 to 2024 and wrote about processors, digital photography, AI, quantum computing, computer science, materials science, supercomputers, drones, browsers, 3D printing, USB, and new computing technology in general. He has a soft spot in his heart for standards groups and I/O interfaces. His first big scoop was about radioactive cat poop.
Expertise Processors, semiconductors, web browsers, quantum computing, supercomputers, AI, 3D printing, drones, computer science, physics, programming, materials science, USB, UWB, Android, digital photography, science. Credentials
  • Shankland covered the tech industry for more than 25 years and was a science writer for five years before that. He has deep expertise in microprocessors, digital photography, computer hardware and software, internet standards, web technology, and more.
Stephen Shankland
11 min read
An interior photo shows the Joint European Torus fusion reactor's doughnut-shaped tokamak design

The Joint European Torus is a fusion reactor with the doughnut-shaped tokamak design.

JET

By this time next decade, you could be reading this story on a phone charged by fusion power, the same energy source that fires the sun.

There's no guarantee that future will arrive on time or at all. Fusion reactions have proved notoriously difficult for humans to sustain, despite some companies trying since the 1990s. But a recent surge in technology improvements and investment has delivered steady progress -- and optimism the environmentally friendly technology is within reach.

If successful, it could supply humanity with the kind of abundant, clean power that attendees of the United Nations' COP27 climate change conference dream about. Although fusion isn't likely to reach large scale soon enough to fix the world's climate problems, it could still prove essential to satisfying humanity's appetite for power in future decades.

"This looks like something that we're starting to get within striking distance," said Andrew Sowder, a senior technical executive at the Electric Power Research Institute, an energy research and development nonprofit. "What we have seen is something going from just a strictly national lab, government-sponsored effort to the private sector, driven by deadlines and with real innovation."

Fusion energy occurs when lighter atoms, most often hydrogen, combine into heavier ones like helium. It's hard to marshal the forces to smash the atoms together, but the reaction releases a tremendous amount of energy: A pickup truck's worth of fusion fuel has the same energy as 10 million barrels of oil.

The appeal is clear. Unlike burning coal, oil or natural gas, fusion doesn't release carbon dioxide, the greenhouse gas that's the main driver of climate change. Fusion also is free of the long-lived radioactive waste or catastrophic meltdown risks that are problems for today's nuclear power plants.

Zooey Liao/CNET

Adding new electrical energy sources is critical with energy demand increasing in coming decades thanks to changes like the shift to electric vehicles and dramatically expanded use of air conditioners. The United Nations' International Energy Agency forecasts that by 2050, electrical energy use will increase between 75% and 150% over today's levels, with the higher level required to reach more aggressive reductions in carbon dioxide emissions. By 2030, the agency said, the planet will need as much new power as is now used in the US and Europe.

Our energy demands and the world's climate problems have triggered new business activity. Entrepreneurs have launched 21 new fusion energy startups since 2017, according to the Fusion Industry Association, with nearly 60% of the total $4.8 billion raised by fusion startups occurring in the last year.

The Biden administration, a fusion proponent, said in November that fusion energy is one of five key approaches to halve carbon emissions by 2030 and reach net zero emissions by 2050. The Energy Department has earmarked $50 million to help private sector fusion projects.

The benefit of fusion energy is clear. The hard part will be actually making it work.

What exactly is fusion?

The idea of fusion has been around for decades, with scientists a century ago starting to understand how it powered the sun, first fusing atoms in 1933 in a lab, and figuring out how fusion can power thermonuclear weapons in the 1950s.

Fusion is the process of jamming atoms from lighter elements together to form a different, heavier element. It takes a lot of energy to combine the lighter elements. But the act of combination pays you back with energy much greater than what's put in. Harnessing that energy is the basis of a fusion reactor.

Matter on Earth is typically found as a solid, liquid or gas, but the temperatures required for fusion are high enough that matter transforms into a fourth state called a plasma. Atoms in a plasma are so energetic that the nucleus at the heart of each loses its electrons. Because particles in plasma are therefore electrically charged, they can be manipulated with electric and magnetic fields.

Fusion is the opposite of fission, the reaction that powers nuclear power reactors today. There, the nucleus of a heavy element like uranium splits into lighter elements and releases energy. Another difference: When fusion reactors have problems, they just fizzle into idleness, with no comparable risk to the meltdowns that afflicted Chernobyl and Fukushima fission plants.

Most fusion startups are working on combining two forms of hydrogen called deuterium and tritium, a blend known as D-T fuel for short. Hydrogen is the lightest element, and most hydrogen atoms have just a single proton in their nucleus. Deuterium has a neutron tucked in there too, and tritium has two neutrons.

When the deuterium and tritium combine, they form a helium nucleus and a lone neutron. The neutron rockets away, propelled by the extra energy of the fusion reaction. The energy output is tens of millions times greater than in combustion of the molecules in fossil fuels.

In most fusion reactors, a "blanket" of molten metal or salt surrounding the fusion chamber captures the kinetic energy of the fast-moving neutrons, heating up in the process. That heat can then boil water to drive a conventional steam turbine to generate power.

It's hard to start a fusion reaction because the positively charged nuclei repel. In the sun, there's enough pressure from gravity to squeeze the nuclei together anyway. It fuses hydrogen at the astounding rate of 620 million metric tons per second, generating enough heat to feel toasty when you bathe in the sun's rays 93 million miles away.

In earthbound fusion reactors, other means are necessary to squeeze the plasma until it reaches temperatures of 100 million degrees Celsius (180 million degrees Fahrenheit). That's where today's engineers are hard at work. Various combinations of magnetic fields, electric currents and lasers confine hydrogen in a high-temperature plasma form in devices with exotic names like tokamaks, stellarators and z-pinch machines.

Some developments now making their way into the fusion business include superconducting materials that carry huge electrical currents at relatively high temperatures, artificial intelligence technology to oversee reactions and 3D printing to make unusually shaped reactor components.

Tokamaks: Fusion in a high-tech hollow doughnut 

The leading design is a tokamak, which uses powerful magnetic fields to trigger fusion inside a doughnut-shaped chamber -- a torus, to use the technical term. The name tokamak comes from Soviet scientists who invented the design in the 1950s and 1960s.

Arguably, the leading startup pursuing tokamak fusion is Commonwealth Fusion Systems, a spinoff from the Massachusetts Institute of Technology. It had a big 2021, first demonstrating new electromagnet technology that generated the world's strongest magnetic field in September, then raising an eye-popping $1.8 billion in funding in November. 

"The reason we think that tokamaks are great is that there's been so much scientific validation of the concept," said Brandon Sorbom, Commonwealth's co-founder and chief science officer. "We wanted to have the least amount of science risk as possible."

The company is spending its money on a massive new tokamak called Sparc and has begun building a facility to house it at a new site in Devens, Massachusetts, about 25 miles west of Boston. The goal is to start operating Sparc in 2025 then take it beyond a critical threshold where the reactor produces more energy than it consumes as soon as possible.

A crane helps construct a large concrete and steel building that will house Commonwealth Fusion Systems' Sparc reactor

Commonwealth Fusion Systems has begun building its Sparc fusion reactor at a site in Devens, Massachusetts.

Commonwealth Fusion Systems

That net energy threshold, called breakeven, is only a first step. According to plans that Commonwealth detailed in a septet of peer-reviewed scientific papers about Sparc, the reactor could produce energy output 10 times its input. That means the fusion gain factor, denoted by the letter Q in scientific circles, is 10. Commonwealth hopes to cross that threshold within two or so years of Sparc's first operation.

In the world of fusion reactors, you'll see a lot of talk about pushing Q as high as possible. The current fusion energy record, set in 2021 at the Joint European Torus in the UK, produced about 11 megawatts of power but only at a Q factor of 0.33.

After Sparc, Commonwealth plans to build another test reactor called Arc that's designed to actually generate power and attach to the electrical grid. It's planned for the early 2030s and should generate about 400 megawatts of power, enough to power tens of thousands of homes. For comparison, a typical nuclear power plant produces about 2.5 times that much, a gigawatt.

Next come the real machines, each generating about the same power as Arc and embodying Commonwealth's very high hopes.

"To make a dent in climate change, the plan is to build on the order of 10,000 of these by 2050," Sorbom said.

Superconducting magnets shrink tokamak designs

The key breakthrough for Commonwealth was high-temperature superconductors, materials that can carry huge electrical currents if cooled enough. High-temperature superconductors, a class of materials developed in the 1980s and beyond, don't need to be chilled as much as the conventional superconductors discovered decades earlier.

Superconductors enable very high magnetic fields designed to confine the plasma fuel and trigger fusion. And crucially, it makes that possible with a much smaller fusion reactor than, for example, the multinational ITER fusion reactor that's been under construction in France for years, with years more work to go.

One problem with D-T fusion is obtaining the materials. Deuterium is found naturally and can be filtered from ordinary water, but because tritium is radioactive, it decays away over a few years. Commonwealth and many other fusion reactor companies deal with this by including some lithium, another lightweight element, in the energy-capturing blanket surrounding the reactor. Incoming neutrons smash into the lithium and transmute some of it into tritium, a process called breeding.

One tokamak challenge is that engineers need to get the blanket hot while also keeping the hulking D-shaped superconducting magnets cold. Sparc will have a powerful chilling system and will operate in shorter pulses to control magnet heating.

Another tokamak difficulty is that those high-energy neutrons cause problems, smashing some atoms apart and glomming onto others. That makes for structural problems and some radioactive materials, particularly on the "first wall" of material nearest the plasma.

To deal with that, tokamaks are designed with some replaceable parts. But rebuilding the interior of a tokamak means significant downtime.

Z pinches and other fusion alternatives

Several other approaches besides tokamaks are in development, including Type One Energy's "stellarators" that look like contorted tokamaks, First Light Fusion's projectile fusion and EX-Fusion's laser zapper system.

That variety is one reason that EPRI's Sowder, a self-described "skeptical optimist," is relatively bullish on fusion. Fusion researchers are taking "multiple shots on goal," he said. "All it takes is one or two of those to succeed."

Zap Energy, based just north of Seattle and expanding to 100 employees this year, hopes to score with an approach called the Z pinch that uses a tubular fusion chamber small enough to fit into a pickup truck bed. Zap's fusion technique uses a pulse of electrical current that creates the plasma and squeezes it from a diffuse column into a thin line that's stable and dense enough to fuse. No superconducting magnets are needed.

Zap's newest reactor, Fuze-Q, is designed to reach that important Q equals 1 threshold where the reactor produces net energy. That could happen as soon as 2023 if research goes well, Chief Executive Benj Conway said, though cautioning "fusion is a graveyard littered with overpromises." After that, Zap plans to double electrical current and increase energy output tenfold.

"It's not inconceivable that we can have a first ever clean power plant at the end of this decade," Conway said.

General Fusion, based in Vancouver, Canada, uses steam-powered pistons to compress plasma that's magnetically confined within a whirling cylinder of molten metal blanket. It plans to begin construction of a demonstration reactor next year and turn it on in 2026.

A diagram shows how dozens of pistons compress a glowing plasma in a cylindrical chamber to initiate fusion

General Fusion uses pistons to compress a plasma until a fusion reaction begins.

General Fusion

"We're looking at our first commercial power plant in the early 2030s," said Mike Donaldson, a General Fusion engineering vice president.

TAE Technologies, founded in 1998 and thus one of the granddaddies of fusion commercialization, has raised more than $1 billion in investments so far, most recently from Chevron (which also invested in Zap) and Google (which also invested in Commonwealth).

Its field reversed configuration design spins plasma in a loop like a tokamak, but does away with much of the toroidal machinery. To keep the plasma loop spinning, TAE gooses it with beams of hydrogen atoms. TAE's sixth-generation "Copernicus" machine is designed to produce net energy in the next two and a half years or so, said CEO Michl Binderbauer.

Next comes "Da Vinci," a prototype due in the early 2030s designed to pump power onto the electrical grid. This machine also should reach a whopping 1 billion degrees, a temperature that would let the company move from D-T fuel to a combination of hydrogen and relatively heavy boron, called p-B11. That's harder, but doesn't produce any neutrons or resulting radioactivity complications, Binderbauer said.

Fusion must compete with wind, solar

Fusion startups are racing not just against each other but also the proven clean energy options of solar and wind power.

One of the most powerful reasons wind and solar have done well is the industrial phenomenon called Wright's Law, also known as the learning curve or experience curve. It's a World War II-era observation that manufacturing improvements bring down costs as capacity doubles. It's the reason why solar power costs dropped 90% from 2009 to 2021 and wind costs dropped 72% over the same period, according to annual tracking from advisory firm Lazard.

Wright's Law applies chiefly to items produced in high volume, so it's not yet clear how well complicated fusion plants will benefit.

Fusion fans are aware of the issue, though. TAE explicitly chose a design that could be mostly assembled in a central factory and shipped out in a 40-foot container. "We believe we can actually get this down in price," Binderbauer said.

A large lab shows TAE Technologies' cylindrical "Norman" fusion reactor

TAE Technologies' "Norman" fusion reactor is named after company co-founder Norman Rostoker.

TAE Technologies

In any event, fusion could complement renewable resources. Solar and wind power are intermittent, and today battery storage to store up power for peak demand moments also is expensive at large scale. In contrast, fusion could be a good "baseload" power source, supplying energy around the clock the way coal and nuclear plants do today. 

Fusion arriving late to the climate party

Although fusion energy has been a dream for decades, today's climate problems have given it new importance. You shouldn't count on cheap, abundant, clean fusion power as our savior, though.

Even if the first commercial fusion plants arrive in the 2030s, it'll likely be years before they supply a significant percentage of power. The later they arrive, the less role they can play in dropping the planet's net carbon emissions to zero by 2050.

Even some fusion fans see it that way.

"In the best case, let's say in the US, you could get to maybe 10% of the energy mix within 20 years. You've really got to forget 2050 targets," Zap's Conway said. "This idea that fusion is going to solve climate change as we know it...is just not true." 

Even so, fusion could prove important to civilization. New power sources could make it easier to wean energy-hungry industrial processes like aluminum smelting, fertilizer production and water desalinization from fossil fuel power plants. It could even potentially power direct air capture plants that remove carbon dioxide from the atmosphere.

"If humans are going to become an advanced civilization," Conway said, "fusion is one of those gateways that we have to pass through."