Startups Gamble on Fusion Energy
In 1989, a pair of chemists boasted at having achieved fusion — that is, harnessing the same type of energy that’s produced by the sun — in room-temperature water. The infamous “cold fusion” announcement generated excitement around the world. But no one was able to replicate the result, and the claims were quickly rejected and ridiculed.
Since then, scientists — and, more recently, investors — have stressed that a different version of fusion is not only possible, but the future of energy. But can the fledgling industry defy skepticism of its hype?
For more than 70 years, researchers have been tinkering with various fusion technologies, from doughnut-shaped reactors to tightly focused laser beams to devices with twisted magnetic coils — all aiming to heat up a hydrogen fuel until the atoms fuse together, releasing huge amounts of energy. And in 2022, scientists at one key fusion lab, located at the Lawrence Livermore National Laboratory in California, reached a fusion-energy milestone when they were able to produce a self-sustaining reaction with a net-positive amount of energy — a threshold called ignition.
The fusion industry is small, but growing rapidly: In recent decades, nearly 50 companies have sprung up, with many founded within the past 10 years.
The accomplishment, along with previous experiments briefly generating a substantial amount of energy, were a shot in the arm for a nascent industry, bolstering the legitimacy and fundraising efforts of numerous startups, including Xcimer Energy in Denver. “That was the big inflection point,” said Conner Galloway, the CEO and chief science officer of the company, who previously studied at MIT and Los Alamos National Laboratory. Xcimer, like the Livermore lab, uses a laser-powered fusion technique. “We were convinced, myself and my co-founder, that if you build a high enough energy laser, this will work.”
Leaders of other fusion startups agree with that assessment. “For us as a company, and as a privately funded industry, this was a big data point that shows that, yeah, fusion is ready for this transition from science project, where there are successes, to building power plants,” said Brian Berzin, CEO of the New Jersey-based startup Thea Energy. Berzin co-founded the company, which is backed by the U.S. Department of Energy, with David Gates, a former researcher at the Princeton Plasma Physics Laboratory, and Matthew Miller, a tech entrepreneur.
The fusion industry is small, but growing rapidly: In recent decades, nearly 50 companies have sprung up, with many founded within the past 10 years, according to the Fusion Industry Association. The majority of them are located in the United States, in addition to a few in Western Europe, China, Japan, Canada, and elsewhere, and they’ve often popped up near national labs and major research universities. Investors have poured billions of dollars into startups exploring different approaches to fusion.
The payoff, advocates say, would be enormous: cheap, abundant, carbon-free energy, with a fraction of the radioactive waste or risk of nuclear power. At present, there’s no shortage of steep energy demands, such as the expanding artificial intelligence and data centers. Those have helped fuel a renaissance for traditional nuclear power plants, including Pennsylvania’s notorious Three Mile Island , which suffered a partial meltdown in 1979 — the worst nuclear accident in the country’s history. If fusion comes to fruition within the next decade, its technologies and reactors will presumably be similarly scrutinized for potential environmental and safety concerns, including mining for metals, risks of radiation exposure, and so on.
It remains to be seen whether these fusion startups, some now with significant funding, will be able to build reactors in the near future capable of generating hundreds of megawatts. Jackson Williams, a plasma physicist at Lawrence Livermore, wrote in an email to Undark that though he doesn’t have the details to evaluate whether the timeline and power output is plausible, “the most I could say is that no company or entity has built a fusion reactor, nor any of the grid-scale components, that would reach those power levels.”
Older nuclear techniques, whether in warheads or power stations, involve fission, where atoms, the smallest unit of any element, are broken apart to release their substantial stores of energy. In nuclear fission, a tiny particle called a neutron typically collides with the nucleus at the center of a uranium-235 atom. The process then releases more neutrons, so if there’s a critical mass of fuel, there will be a chain reaction. For atomic bombs, the chain reactions are deliberate and cause an explosion; in power plants, they must be controlled to avoid a meltdown, releasing dangerous radioactivity.
Unlike solar and wind energy, nuclear fission plants can generate continuous energy, barring disruptions by natural disasters or war. But nuclear energy comes with environmental and health risks. Uranium has to be mined, for instance, including on sites like Navajo tribal lands, which contaminates the surrounding land and water. Operating nuclear plants requires massive quantities of water to help cool the hot reactors where the fission takes place, which is why the plants are usually built next to lakes, rivers, and seas. And the normal operation of such plants leaves behind toxic nuclear waste, which has to be safely transported and stored securely for millennia.
Rather than splitting atoms, fusion smashes them together at incredible temperatures and pressures. It’s the same process that occurs in the sun’s 27-million-degree Fahrenheit interior, where hydrogen becomes so hot that it turns into an electrically charged form of matter called plasma, and the hydrogen nuclei collide with each other at tremendous speeds, combining to become helium nuclei and spewing out energized neutrons in the process. (Unlike the first nuclear weapons that were based entirely on fission, today’s nukes are hydrogen bombs that combine fusion and fission to release even more energy.)
Earthbound fusion reactors differ significantly from stellar cores, of course. Reactors typically use a small amount of heavy hydrogen fuel, like deuterium and tritium, which needs to be continually fed into the system to keep the reactions running. Rather than the crushing gravitational pressures of the sun, terrestrial fusion uses other techniques to drive the reactions. And then the reactors absorb the excess energy as heat, which produces steam that powers electricity-generating turbines. Within a decade if not sooner, Xcimer and other startups say they could have megawatt-generating reactors — enough to power data centers — and eventually gigawatts of energy to supply to cities.
However, that’s orders of magnitude beyond anything any of them have achieved so far, and fusion experts are cautiously optimistic though many challenges remain. Ellie Tubman, an experimental physicist working on laser-driven fusion at the University of California, Berkeley, expects more startups to achieve ignition and eventually start generating more energy. Tubman, who collaborates with fusion companies and may work with Xcimer in the future, said some of these startups could face technological hurdles, including developing reactor materials to withstand streams of high-energy neutrons.
These commercial outfits have been exploring many types of fusion. One approach focuses strong lasers on small amounts of hydrogen fuel to trigger fusion. Other techniques use powerful magnets to confine and control the superheated plasma. This includes doughnut-shaped reactors called tokamaks, one of which is being developed at Commonwealth Fusion Systems in Virginia, and some that are twisted, called stellarators, which are in development at Type One Energy and Thea Energy in Tennessee and New Jersey, respectively. Some companies, like General Fusion in Richmond, British Columbia, have hybrid approaches combining elements of both magnetic and laser-based systems, which, they say, will help them to achieve better energy efficiency and at lower costs.
Compared to nuclear fission, at almost every step, fusion appears to be a cleaner form of energy with fewer risks and, eventually, fewer costs — particularly when it comes to pollution and water consumption.
Compared to nuclear fission, at almost every step, fusion appears to be a cleaner form of energy with fewer risks and, eventually, fewer costs — particularly when it comes to pollution and water consumption. Rather than uranium, most of today’s fusion reactors require tritium. Global tritium supplies are so limited that it goes for about $30,000 per gram, sourced from 19 commercial nuclear reactors in Canada. An international fusion research project in France plans to consume much of that supply, so Xcimer and most other companies are banking on breeding tritium in their fusion reactors, where high-energy neutrons from the fusion reactions collide with a layer of lithium, releasing tritium atoms. As long as their breeding process produces at least 10 percent more tritium than they started with, they’ll be self-sufficient, according to a recent study.
In that tritium-generating process, the reactor walls are covered in a layer of lithium, whose atoms can recombine into helium and tritium. For this process to work, mining lithium becomes crucial. Most of the world’s supply is mined in just a couple places, and that mining does have environmental impacts. Still, the amount of lithium needed in a reactor for generating tritium and coolant materials is relatively small, though it could add up to the equivalent amount used in 20,000 electric cars, said Galloway, the CEO at Xcimer. “So we do not view Lithium as a critical supply chain risk,” he wrote in an email.
As for safety, while fusion does require hot temperatures of some 100 million degrees Celsius, there’s no chain reaction and therefore no risk of meltdown. “Fusion is very hard to get started and very easy to stop; fission is easier to get started, and harder to stop,” said Megan Wilson, General Fusion’s chief strategy officer. For all types of fusion, a reactor has no more than a gram of fuel at any given moment, just enough to keep running for the next instant of time, and a reactor can’t keep going on its own. “You need to keep continually putting fuel in, so there’s no runaway possible. If anything happens — earthquakes, something falls into the plasma, somebody overfed it, somebody did whatever — it will basically collapse, the process will stop,” said Thomas Sunn Pedersen, chief technology officer of Type One Energy.
Tritium is mildly radioactive, but it has a half-life of about 12 years, meaning it takes that long for half of it to decay. Uranium-235, on the other hand, takes hundreds of millions of years to stop posing a threat. And tritium isn’t as dangerous for people who are exposed to it because its radiation doesn’t penetrate the skin. General Fusion’s Wilson likens fusion’s radiation profile to that of a hospital that uses medical isotopes or has a cancer treatment ward. There’s no long-lived radioactive waste and fusion has minimal risks of proliferation or weaponization, said Berzin, who argues that fusion is so much safer than fission that the two energy sources have no significant similarities.
Fusion researchers have made plenty of progress, yet experts suggest that the industry is at least a decade away from powering an electricity grid. Even at Lawrence Livermore, the research team is advancing incrementally. One of their first successful fusion experiments only produced just a tiny yield, or net energy — about 3.1 megajoules, which is about 0.1 percent of a single U.S. home’s monthly energy use. Over the next few years, the researchers are working on growing that amount. According to Williams, the Lawrence Livermore fusion physicist, that involves “trying to get a little bit more energy out of the laser, and trying to use that laser energy as efficiently as possible, minimizing any losses we have in the system, trying to design the target in a way that most efficiently uses that energy.”
Fusion startups have been amassing funds for their ambitious plans, and thanks to growing investment over the past couple years, at least nine companies have already raised more than $100 million each. A 2021 report from the National Academies of Sciences, Engineering, and Medicine recommended that the Department of Energy and private sector should aim to produce net electricity in a fusion pilot plant in the 2035 to 2040 timeframe.
As an intermediate step, some startups already have smaller initial projects in the works. General Fusion’s Lawson Machine 26, for example, is a device that the company said will demonstrate its approach at 50-percent scale by 2026. Wilson said they intend to achieve scientific breakeven, or the point at which the devices are able to make the same amount of energy put into the plasma.
Berzin, the CEO of Thea, the New Jersey startup, said their prototype stellarator system will produce tritium, with a subsequent system providing on-the-grid energy by the 2030s. Xcimer’s Galloway said their prototype will be online in 2026, and he said their target is to have their fusion pilot plant up and running in the mid-2030s. Last year, Type One Energy announced plans for their first stellarator, Infinity One, at the Tennessee Valley Authority’s Bull Run Fossil Plant. In a press release on Tuesday, the company announced that it entered into a cooperative agreement with the TVA to “jointly develop plans for a potential TVA fusion power plant project” in the region called Infinity Two, which could come online as early as the mid-2030s.
But researchers have made fusion promises before. In the 1980s, at around the same time as the cold fusion fiasco, scientists proposed building a tokamak reactor — which employs hot fusion, as do all the techniques under development today — in southern France called ITER. Four decades later, that flagship project remains far from completion. Backed by a collaboration between the U.S. and several European and Asian countries, the ITER project has been beset with numerous construction delays and technical issues related to the thermal shields. The team is now targeting 2034, followed by years of research.
Nevertheless, no one at any private company has yet claimed to achieve scientific breakeven, the milestone that Livermore reached in 2022. A couple startups could reach it within a few years, General Fusion’s chief strategy officer said, but even then, the accomplishment would still be multiple steps away, both technical and financial, from erecting a full-sized power plant that could generate electricity commercially.
One major question in the field is whether fusion will make much of a difference in terms of climate change mitigation, particularly if it doesn’t come to fruition anytime soon.
Still, many in the industry remain hopeful. Berzin believes that there will be demand for commercial fusion plants of different sizes and for different sites and purposes, while Pedersen believes fusion could become a powerful climate solution, generating “24/7 electricity production with no CO2 emissions,” he said.
“Climate change is happening,” he said, “and the sooner we get a solution that can be adopted across the board, the better for humanity.”
Fusion researchers have made plenty of progress, yet experts suggest that the industry is at least a decade away from powering an electricity grid.
Perhaps the future will bring competition between fission and fusion, especially if the intermittency of solar and wind energy makes them perceived as insufficient.
Both kinds of energy could provide small reactors (each with a capacity of up to 300 megawatts) as well, Pedersen said, with some applications beyond powering data centers, like energy for desalination and carbon removal facilities. Or if fusion turns out to be as safe and environmentally friendly as its proponents claim, it could eventually prevail over its nuclear rival.
The rising industry has grand goals, assuming plans for new pilots and reactors are realized. For Xcimer’s Galloway, the development of fusion marks a critical transition for an advanced civilization. “It’s the final step for energy production,” he said. “You start off with animal power, burning wood, and then you get coal, wind, and water. Then you discover nuclear physics and fission and fusion. Past fusion, there’s nothing else.”
Gallaway omits to mention solar pv and batteries in his list of power sources over the ages. They make and store electricity. They will get cheaper and so ubiquitous that fusion, which only makes steam, will certainly never be viable.