5 potential paths to a fusion energy breakthrough

Dozens of companies are racing to deliver electricity from fusion — a form of carbon-free electricity often compared to a “star in a bottle.”

While fusion power had been thought to be decades away, in May, Microsoft signed the world’s first contract to purchase fusion-generated electricity. It’s a sign that the industry, long seen as akin to science fiction, is achieving a new level of commercial maturity.

Helion, the company that signed the deal with Microsoft, is just one of the dozens of companies, backed by billions in venture capital, competing to bring the energy source to commercial use — and under heavy pressure to begin showing results.

The particular physics of fusion makes that a diverse field. 

“There’s almost a staggering a variety of technological approaches to try to make fusion work, which has been part of the excitement of the last few years,” said Dennis Whyte, director of the MIT Plasma Science and Fusion Center, told The Hill.

Here’s an overview of the main paths — and the entities pursuing them — starting with the only known example of successful fusion power: our sun.

A star in a bottle?

In a sense, virtually all electricity on Earth comes from fusion.

The light and heat that warms the Earth, powers its wind and water cycles, and grows its plant life comes from the sun — an enormous fusion reaction 94 million miles away. 

Because of the sun’s contribution to weather, all electricity that comes from wind, solar, biomass and hydropower is a product of fusion, as well.

And because fossil fuels come from the unimaginably old, highly compressed bodies of ancient plants, that makes oil, gas and coal a product of fusion, too.

Fusion energy is released when two atoms are forced together until they collapse into each other. It’s an opposite process to the atom-splitting, or fission, that powers current forms of nuclear energy.

But it takes a great deal more power to force atoms together than to split them, which is why — so far — the only place a continuous fusion reaction has been observed is in stars.

In stars like our sun, the crushing gravity produced by their enormous size can keep the fusion fires burning. (The sun is 1.3 million times the size of Earth.)

That crucial role of gravity (and size) poses a serious problem for earthlings attempting to generate fusion power.

“We say [fusion power is] building a star, but you can’t literally replicate a star on Earth, because there’s a reason stars need to be that size,” said Whyte, the MIT fusion researcher.

“That’s why there’s no small stars around,” Whyte added.

Tokamak: a magnetic donut

Without access to the sun’s powerful gravity, the quest for Earth-based fusion requires another powerful force to drive atoms together.

In most cases, that means magnets. The dominant form of fusion uses a donut-shaped reactor wound around with magnetic coils — what the Soviet scientists who invented it called a tokamak. (The word comes from the Russian acronym for “toroidal chamber with an axial magnetic field.”)

Inside that donut, which was first successfully operated in 1958, clouds of gas could be forced together to achieve temperatures sufficient to accomplish fusion, and that approach is now seen in projects from Europe’s ITER to China’s EAST.

Without access to the sun’s great size, fusion reactors also have to produce a lot more heat, which is where tokamaks have delivered. 

In 2021, the international team at EAST (the Experimental Advanced Superconducting Tokamak) ran at 120 million degrees Celsius, or 216 million degrees Fahrenheit, for an unprecedented 101 seconds — a runtime it increased by a factor of four in April.

That’s also the approach that Whyte’s team at MIT is pursuing in conjunction with Commonwealth Fusion Systems, a company that has $2 billion invested. 

Stellarator: twisting the donut

The tokamak isn’t the only piece of magnet-based Cold War-era fusion technology to receive a modern infusion of cash. 

In March, Breakthrough Energy Ventures — an investment firm owned by Microsoft founder Bill Gates — provided $29 million in initial funding to Type One Energies. (Gates was also a funder of Commonwealth fusion.) 

Type One’s approach is built around the stellarator, an attempt to build fusion power designed at Princeton in the 1950s but largely abandoned in favor of tokamaks.

If the tokamak is like a glazed donut, the stellarator — which twists around itself like a Mobius strip — is more like a cruller, a Type One spokesperson said.

In theory, the twisted shape cancels out the instabilities in the circulating plasma, giving the magnetic “cage” that contains it far more stability than it would have in a tokamak, Type One CEO Christofer Mowry told The Hill.

That makes it much easier to maintain a protracted fusion reaction and ultimately get power out.

But modeling the complex physics to produce a stellarator was impossible “until the last 20 years when supercomputing really came forward, and based on that, they were actually able to start designing these machines,” Mowry said.

In 2018, the publicly funded Wendelstein 7-X fusion in Germany achieved a record three seconds of fusion, inaugurating a new era in stellarator research that Type One plans to piggyback off.

Lasers and beams: rapid ignition

The particular physics of fusion represents an almost infinite number of methods to release power, Whyte told The Hill.

Fusion is the product of two factors: the speed of the reaction, and the density of the available fuel. 

To see this principle in action, imagine a bonfire, Whyte said. For it to burn quickly, “the wood can’t be spread over half an acre: it has to be together, so each log burns the one beside it.” 

By contrast, if it’s spread across the yard, the wood may still burn — but it will take a lot longer. “So the density can be rather modest, and energy confinement time needs to be larger,” he said.

At one extreme, the sun burns (relatively) slow and cold; at the other are pulses and lasers, in which a pocket of plasma 10 million times denser than that in a tokamak is fused over a period 10 million times shorter.

That was the approach seen in the landmark results from the Lawrence Livermore National Laboratory in December, in which a fusion reaction for the first time ever released more energy than had been put in.

Fusion engine: skipping the hard part

Somewhere between the steady power of a tokamak or stellarator and the flash-in-the-pan intensity of a laser is the hybrid approach Microsoft invested in last week: Helion’s fusion engine.

Tokamaks and stellarators can’t generate electricity directly. Instead, their fusion reactions release a flood of radioactive neutrons, which batter the metal walls of their container, releasing heat that can be tapped to boil water for a steam turbine.

That’s the form of electricity generation that has run fossil fuel- and nuclear-powered “thermal” power plants for more than a century.

Helion CEO David Kirtley has also long argued that it is unnecessary — and, he has sometimes implied, a bit unimaginative. “I thought, ‘You’re catching a star and you’re using it to boil water?’” he told Rolling Stone in 2017.

Helion’s approach seeks to cut out the middleman by an unconventional means: generating electricity directly off the fusion reaction itself.

It takes advantage of the fact that magnetic fields and electric currents are two aspects of the same phenomenon — or more specifically, that disruptions in a magnetic field, caused in this case by a series of tiny fusion explosions, creates outpourings of electricity.

Because that approach doesn’t require a stable, self-sustaining reaction, “we’ve been able to build fusion systems that are much, much smaller than any of the other approaches to fusion,” Kirtley told The Hill last week.

“And that means you can build them faster, you can learn more, and you can build power plants sooner.”

How soon? In the 2017 Rolling Stone interview, Kirtley predicted fusion electricity within 10 years — a prediction that he is now financially obligated to (more or less) fulfill. 

In early May, Helion committed to provide Microsoft with 50 megawatts of fusion power by 2028, and the company faces financial penalties if it fails.