Path to Fusion Power Pits Giant Lasers Against Powerful Magnets

The day when humans can harness the same energy that lights the stars could come sooner than you think — getting there would unleash plentiful electricity without emitting greenhouse gases.

By Will Wade, Jonathan Tirone and David R BakerGraphics by Dave Merrill

Share this article

Humanity is on the cusp of something phenomenal: harnessing the same power source that lights the stars for nearly limitless, carbon-free energy. Scientists recently proved that dream — nuclear fusion for energy generation — is possible. Now, going from a lab experiment to building a commercial plant will be a race that pits giant lasers against powerful magnets.

After decades of experiments, two competing designs for fusion plants have emerged. One calls for high-intensity lasers to trigger a series of reactions that slam atoms together many times per second. The other would use super-strong magnets to contain a cloud of plasma that burns hotter than the sun. While lasers were used in the recent breakthrough, many experts are skeptical of the commercial prospects. The better bet, they say, is magnets.

In December, at Lawrence Livermore National Laboratory, a tiny fuel capsule, containing two forms of hydrogen, was blasted with lasers

The resulting fusion reaction generated more energy than was input by the lasers on the target

The stakes couldn't be higher. If researchers can make fusion work at scale, it would open the door to power plants that supply cheap, plentiful electricity day and night without emitting greenhouse gases and with no danger of nuclear meltdown. The idea of recreating the extreme conditions of the stars in a power plant might sound like something out of science fiction, and yet the most-optimistic experts say we’re only about a decade away from that threshold. Other scientists peg it at 20 or 30 years from now.

“Fusion has always been the apex predator of energy technologies,” said Bob Mumgaard, chief executive officer of Commonwealth Fusion Systems. “It’s a very hard problem with a big payoff.”

The race is already drawing bets from some of the world's richest people. Jeff Bezos, Bill Gates and Peter Thiel are just three of the billionaires investing in startups. Investors and governments have poured more than $4.8 billion into companies pursuing fusion, led by Commonwealth Fusion, a startup spun out of the Massachusetts Institute of Technology that’s landed $2 billion. TAE Technologies has received more than $1.1 billion. The Fusion Industry Association is tracking 33 startups. Fifteen are focused on the magnetic approach, and eight are working on the laser design. The rest are pursuing a variety of other technologies.

The road will be long and complicated. Both the laser and magnet approaches face major technical challenges, scientific puzzles and cost hurdles. But getting it right would mean a tremendous advance for the world. Humanity’s long-term climate challenges would be much more manageable, and the achievement could launch a new era for energy and science.

How Fusion Works While today’s nuclear power plants employ fission, splitting atoms apart, fusion captures energy produced when atoms merge together. Fusion is already used to give modern nuclear weapons their devastating power, but the goal is taming it for civilian energy demand.

That’s no simple task. It involves operating at extremely high temperatures, containing the reaction, capturing the energy and doing it all while generating more electricity than the process consumes.

The BreakthroughNot long after midnight on Dec. 5, scientists at Lawrence Livermore National Laboratory in California focused the world's most powerful laser on a peppercorn-sized diamond pellet filled with hydrogen isotopes. It shot 192 beams in three carefully modulated pulses.

The beams delivered 2.05 megajoules of energy, triggering a reaction that fused the hydrogen into helium and released 3.15 megajoules in the process — the difference, a little more than a megajoule or roughly the equivalent energy released by hand grenade. It was an accomplishment that scientists had been pursuing for decades. The milestone, known as net energy gain, proved that humans could unlock the power of the stars. But creating a commercial plant would mean needing to generate 1,000 times that amount of energy every second, said Steven Cowley, director of the Princeton Plasma Physics Laboratory.

“Constructing a system that robust is challenging,” Cowley said.

LasersThe laser method used at the Lawrence Livermore lab is called inertial confinement. The reaction generated was incredibly brief, about as much time as it takes for light to travel an inch. To produce energy around the clock, a fusion system would need to repeat this over and over again, as many as 10 times a second.

That’s not possible with the systems available today, said Dylan Spaulding, a senior scientist at Union of Concerned Scientists, who has done research at the California lab. The laser at the National Ignition Facility in Livermore is so powerful that it can only be fired once every few days because it generates high-intensity heat that can damage the equipment.

“You tend to break a lot of things when you push the system to its limits,” he said.

When will the first fusion plant deliver electricity to the grid?

Still, that laser was built using technology that dates to the 1980s, and there have been numerous advances since then. Spaulding is optimistic that engineers can build one strong enough, and durable enough, to operate at the level needed for a fusion system.

A bigger challenge is the fuel pellets. The NIF test used a diamond ball filled with deuterium and tritium, two hydrogen isotopes. Lab officials said it takes about seven months to produce the components, and then about two weeks to assemble them. They’ve declined to put a price tag on the work, but outside experts have estimated they may cost a few thousand dollars to as much as $20,000 each. That’s way too expensive for a plant that would likely need to blast almost about 1 million pellets a day.

MagnetsThe competing approach uses magnets to contain a cloud of super-heated plasma that produces fusion reactions. The main advantage of this method would be that once that process was achieved the plasma could, in theory, be held in a steady state producing energy for decades.

In experimental tokamaks, such as ITER, deuterium and tritium plasma is trapped and held in place primarily by two magnetic fields

The longer that this plasma remains stable, the more fusion that takes place

“The magnetic approach lends itself to much larger scale, which is what you need for a commercial power plant,” said Adam Stein, director of nuclear energy innovation at The Breakthrough Institute.

The magnet confinement process would need to be so powerful it could control plasma that’s burning as hot as the sun. Nobody has yet managed to do this at the extreme temperatures needed to produce positive energy, or for long periods of time, but researchers are making progress. Much of the advance has come around a so-called tokamak design that dates back to the Soviet Union. In it, lasers and mighty electromagnets are arrayed around a super-cooled doughnut-shaped container to hold the heated plasma in place.

Fusion method companies are pursuing

Commonwealth Fusion Systems may have addressed one of the key challenges, with a magnet it says is the strongest in the world. It expects to complete in 2026 a demonstration system using a tokamak that will be able to contain — for as long as 30 seconds at a time — heated plasma that will produce net energy, according to CEO Mumgaard. A commercial version could be ready in the early 2030s, he predicts.

The tokamak design is also at the heart of the International Thermonuclear Experimental Reactor, or ITER program, whose construction in southern France is scheduled to cost more than $23 billion. Considered the largest research project in history, it’s widely thought of as the world's best shot at demonstrating that large-sale fusion power is possible. Its 35-nation funders include China, the European Union, India, Japan, Russia, South Korea and the US, with all the countries getting access to the intellectual property ITER creates.

The project has been plagued by unexpected challenges. Just as researchers started sorting out logistics disrupted by the pandemic, Russia’s invasion of Ukraine complicated the supply of critical components. In May, the project’s long-time chief Bernard Bigot died. Then in December, cracks were discovered in key components.

The streak of bad news means that ITER's first-fusion demonstration won't happen in 2025 as planned. Its new Director-General, Pietro Barabaschi, is putting together a fresh timeline and budget expected to be presented by the end of the the year.

Road AheadNumerous challenges remain.

The industry is still evaluating different fuels and has yet to settle on which will offer the easiest path to a power plant. Commonwealth, along with several other companies, is using the same hydrogen isotopes that were used in the Livermore test, deuterium and tritium. Others are testing fuels based on boron.

Commonwealth Fusion Systems, which is building a compact tokamak power plant, raised over $1.8 billion in 2021

Deciding on a fuel source will also help determine what materials are needed for the reactor walls and other components. Every design will need to withstand high heat, but using deuterium and tritium also means companies should expect their machines to become radioactive. So while fusion doesn’t produce spent fuel waste like in fission, the machine itself, when it is eventually decommissioned, could produce tons of waste.

And no system has yet been built that can capture the energy from a fusion reaction and convert it to electricity.

Ultimately, getting all the technology right will bring enormous rewards for the planet.

“We have a real climate opportunity,” said Jane Hotchkiss, co-founder and president of Energy for the Common Good, a non-profit outside of Boston laying the groundwork for broad social acceptance of fusion power.

People should be closely following today’s startups, “knowing that even if just seven of them hit their milestones over the next 10 years, it will be a very impressive feat,” she said, adding that while the timeline might feel slow, “all these incremental steps are important.”

Editor: Millie Munshi