On April 30, a fusion company took a step that would have seemed like science fiction just five years ago. It applied to connect a 400-megawatt fusion power plant directly to the largest electricity grid in the United States. Commonwealth Fusion Systems told the regional grid operator PJM that it plans to supply fusion-generated electricity from its Virginia plant, the Fall Line Fusion Power Station, aiming to deliver power to the grid by the early 2030s.
For fifty years, fusion has been the subject of energy jokes, always said to be 30 years away. Now, that timeline is finally starting to change. Private fusion companies have raised about $9.8 billion so far. The U.S. Nuclear Regulatory Commission has officially separated fusion from fission in its rules, and at least three U.S. companies are actively seeking permits or building grid-scale plants. This progress does not guarantee that commercial fusion will arrive on time.
Still, by 2026, the policy, funding, and engineering questions are no longer just theoretical. Today’s decisions will shape how the next decade of clean energy develops.
Fusion vs. Fission: Two Opposite Reactions
Both fusion and fission release energy from atomic nuclei, but they do so in opposite ways.
Fission is the reaction in every commercial nuclear plant operating today, which splits a heavy atom (typically uranium-235 or plutonium-239) into lighter fragments, releasing energy and a cascade of neutrons that sustain a chain reaction.
Fusion does the inverse: it forces two light nuclei together to form a heavier one. Most fusion designs use deuterium and tritium, both of which are isotopes of hydrogen. The reaction produces helium plus a high-energy neutron, releasing energy in the process. It is the same reaction that powers the Sun.
The practical differences are important. Fission needs a certain amount of fuel and a controlled chain reaction. If cooling fails, leftover heat can cause a meltdown, as happened at Fukushima and Three Mile Island. Fusion does not require a chain reaction or a critical mass, so it does not melt down. The plasma created by fusion reactions must be kept at about 100 to 200 million degrees Celsius for the reaction to continue. If those conditions change, the reaction stops on its own.
The U.S. Nuclear Regulatory Commission (NRC) found that fusion machines do not produce the kind of residual heat that requires emergency cooling. That is why, in 2023, it decided to regulate fusion as a byproduct material rather than as a power reactor.
Environmental Impacts: Where Fusion and Fission Diverge
During normal operation, neither fusion nor fission plants release carbon dioxide or other greenhouse gases. The main environmental concerns are about waste, managing fuel cycles, and the materials used to build each type of reactor.
Fission’s Long Tail
Spent nuclear fuel from fission reactors contains isotopes that remain hazardous for very long periods. Plutonium-239 has a half-life of roughly 24,100 years; uranium-235, about 700 million years. Cesium-137 and strontium-90 — major radiological contributors in spent fuel — have half-lives near 30 years but require shielded storage for centuries. The global inventory of spent nuclear fuel exceeds 400,000 metric tons, and no country has yet opened a permanent geological repository, although Finland’s Onkalo facility is near operational status.
Fission also requires uranium mining, milling, and enrichment. These are energy-intensive steps that affect land use, water, and create waste. After a plant is built, decades of carbon-free electricity can help balance out those early impacts, but the effects are real and mostly felt near mining communities.
Fusion’s Smaller, Shorter Footprint
A fusion reactor mainly produces helium, a valuable element, as direct waste; it is a non-toxic and non-radioactive gas. The main radiation concerns relate to two other sources: tritium, the radioactive hydrogen isotope used as fuel, and the reactor’s structural materials, which become radioactive over time as they are hit by high-energy neutrons during operation.
Tritium has a half-life of about 12.3 years. This is short for nuclear materials, but still long enough that any release into the environment is a real concern. Tritium can combine with water to form tritiated water, which living things can absorb. The main way to manage this is to contain and recycle tritium within a closed fuel loop. Reactor structures, usually made of special steels and ceramics, become radioactive during use. When removed, they generally become safe to handle within 50 to 100 years, which is much shorter than the thousands of years needed for fission waste.
Fusion also avoids the risk of nuclear weapons proliferation that comes with fission. Fusion systems do not use fissile material, so there is no uranium enrichment, no plutonium production, and no chain reaction that could be used for weapons. This is one reason the NRC decided that fusion’s risks are more like those of particle accelerators and medical isotope facilities than those of traditional nuclear plants.
| At a Glance Fusion vs. Fission: Opposite Reactions |
||
|---|---|---|
| Fission | Fusion | |
| Reaction | Heavy atom splits into lighter fragments | Light atoms combine into a heavier one |
| Typical fuel | Uranium-235, plutonium-239 | Deuterium (from seawater) and tritium (bred from lithium) |
| Chain reaction? | Yes — must be actively controlled | No — reaction halts if conditions falter |
| Long-lived waste | High-level waste hazardous for tens of thousands of years | Mostly activated reactor materials, hazardous on the order of decades to about a century |
| Meltdown risk | Decay heat can damage core if cooling fails | No decay heat sufficient to require emergency cooling |
| Greenhouse gases (operation) | None directly | None directly |
| Commercial status (2026) | Mature; ~440 reactors operating worldwide | Pre-commercial; first grid connections targeted 2028–early 2030s |
| Source: Earth911 analysis of U.S. Nuclear Regulatory Commission, IAEA, and Fusion Industry Association data. | ||
The Environmental Caveats
Saying fusion is environmentally clean does not mean it has no environmental impact. There are three concerns that anyone interested in sustainability should consider:
- Tritium is scarce. Worldwide, civilian tritium stocks are only about 25 to 30 kilograms, mostly made as a byproduct of Canada’s CANDU heavy-water fission reactors. Many of these reactors are set to retire this decade. A 1-gigawatt fusion plant would use more than 50 kilograms of tritium each year. The industry plans to make tritium inside the reactor by lining the walls with lithium, but this has never been proven to work at commercial scale.
- Lithium-6 and the Minamata problem. To breed tritium effectively, reactors need lithium enriched in the rare isotope lithium-6, which represents only about 7.6 percent of natural lithium. The old industrial process for separating it, called column exchange or COLEX, uses a lot of mercury and is now banned for new use under the Minamata Convention on Mercury. Right now, only Russia and China are thought to produce enriched lithium-6. Cleaner methods are being developed, but supply chain issues remain a real challenge.
- Neutron damage and decommissioning. The 14-MeV neutrons generated by deuterium-tritium fusion damage reactor materials more than fission neutrons do. Reactor walls and components will need to be replaced from time to time, producing low- and intermediate-level radioactive waste that must be managed. Over a plant’s lifetime, fusion produces more waste by weight than fission, but the radioactivity fades much faster.
Where Commercialization Stands in 2026
Fusion is now much more than a single lab experiment. According to the Fusion Industry Association’s 2025 Global Industry Report, there are 53 private fusion companies that have raised a total of $9.77 billion. Of that, $2.64 billion came in the 12 months ending July 2025, the second-largest yearly increase since the report started. The F4E Fusion Observatory said that by September 2025, total global private fusion funding was about $15.2 billion.
Three U.S. companies are now further along than the rest:
Commonwealth Fusion Systems (Massachusetts and Virginia)
Commonwealth, which started at MIT, is building a tokamak—a doughnut-shaped magnetic chamber—called SPARC at its Devens, Massachusetts, campus. The demonstration machine is about 75 percent finished and is expected to start operating by late 2027. If SPARC achieves net energy gain, the company plans to build the 400-megawatt Fall Line Fusion Power Station in Chesterfield County, Virginia. Google and the Italian energy company Eni have already signed agreements to buy power from that plant. An application to connect to the grid filed in April 2026 is the first step in a process that will take four to six years before approval. Without the grid connection, there’s no place for the electricity generated to go.
Helion Energy
Everett, Washington-based Helion uses a different approach called a field-reversed configuration, which aims to generate electricity directly from the fusion reaction’s magnetic field and avoids using a steam turbine. It has signed the world’s first fusion power purchase agreement, promising to deliver 50 megawatts of fusion electricity to Microsoft data centers starting in 2028. Helion began construction of the Orion plant in Malaga, Washington, in July 2025 and obtained its Conditional Use Permit from Chelan County in October 2025. Its prototype, Polaris, has reached plasma temperatures of 150 million degrees Celsius. Many see the 2028 deadline as ambitious.
Inertia Enterprises
Inertia was founded in 2024 to bring the laser-driven inertial confinement method, developed at Lawrence Livermore National Laboratory’s National Ignition Facility, to market. In April 2026, it announced a $450 million funding round and one of the largest public-private research partnerships in the history of DOE national labs. The company is working with LLNL to scale up the fusion-target manufacturing techniques used in NIF’s December 2022 ignition shot, which was the first lab experiment to achieve target gain by producing 3.15 megajoules of fusion energy from 2.05 megajoules of laser energy.
ITER and the International Track
ITER, an international tokamak project involving 35 countries and being built in southern France, updated itsrelease schedule in 2024. The first plasma is now expected in the mid-2030s, with operation starting in 2035 and full deuterium-tritium fusion beginning in 2039. ITER will not produce electricity, but it is still the most ambitious test site for the physics and engineering challenges that future commercial fusion plants will face.
The Regulatory Picture: Fusion Is Not Fission
In April 2023, the U.S. Nuclear Regulatory Commission unanimously voted to regulate fusion machines under 10 CFR Part 30 — the byproduct materials framework that already governs particle accelerators, medical isotope facilities, and industrial irradiators — rather than under the regime that governs fission reactors. Congress reinforced this approach in the bipartisan ADVANCE Act of 2024.
In February 2026, the NRC released its proposed rule to formalize this framework. The rule focuses on regulating tritium handling, neutron-activation products, and waste streams, instead of emergency cooling systems, because fusion machines do not create the leftover heat that fission reactors do. This is a significant policy change that addresses fusion’s real risks directly, which can speed up permitting for serious developers but also means those developers must clearly show their safety plans.
The Skeptical Case
Fusion’s commercial supporters are confident, but not everyone agrees. Daniel Jassby, who spent 25 years as a fusion researcher at Princeton’s Plasma Physics Laboratory, wrote in the Bulletin of the Atomic Scientists that fusion plants will need a lot of support infrastructure, even when the reactor is not running. He also says they may need more workers than fission plants of similar size and could create more low- to intermediate-level waste than fission, although the waste is much less radioactive.
The Sierra Club’s 1986 policy on fusion is still in place; it raised concerns about tritium release, decommissioning costs, and whether fusion is a better investment than renewables. A more recent Sierra Club essay says things have changed enough to reconsider fusion, but questions about cost, fuel-cycle viability, and how soon fusion can be deployed are still unanswered.
Even within the industry, 83 percent of fusion companies surveyed in 2025 said securing investment remains a major challenge. They estimate they need another $77 billion to build the first commercial plants, which is about eight times the money raised so far.
What This Means for the Energy Transition
The reason to pay attention to fusion in 2026 is not that it will solve the climate crisis this decade. Solar, wind, batteries, geothermal, and existing nuclear plants are already helping, with falling costs and a 15-year head start. The real point is that the next decade’s electricity demand, driven by AI data centers, the electrification of heating and transport, and industrial decarbonization, will require a diverse mix of reliable, low-carbon sources.
If fusion works at scale, it can provide reliable electricity with low emissions over its life, create little long-lived waste, and carry a low risk of nuclear proliferation. Whether fusion makes it to the grid by 2030 depends on scientists, funding, and regulations aligning. Maybe Helion, possibly with a smaller-than-promised first delivery, will win the race. Commonwealth’s Virginia plant in the early 2030s will need its grid interconnection process to move on schedule. Other players will follow later. None of these events is a sure thing.