Fusion is a nuclear reaction in which two light atomic nuclei merge to form a single heavier nucleus, releasing energy in the process. The energy comes from a small amount of mass that “disappears” during the reaction. The combined nucleus weighs slightly less than the two original nuclei did separately, and that missing mass converts directly into energy, following Einstein’s famous equation E=mc². This mass-to-energy conversion is the same process that powers the sun and every other star in the universe.
How Fusion Works at the Atomic Level
Atomic nuclei are positively charged, which means they naturally repel each other, the same way two magnets push apart when you hold matching poles together. To force nuclei close enough for fusion to occur, you need extreme speed and temperature. At around 100 to 200 million degrees, nuclei move fast enough to overcome that electrical repulsion. Once they get close enough, a much stronger fundamental force takes over and binds them together into a new, heavier nucleus.
Interestingly, the ideal density for a fusion plasma is surprisingly low, about a million times less dense than air. At higher densities, collisions between nuclei and electrons generate so much radiation that all the energy in the plasma gets radiated away before fusion can sustain itself. So fusion needs extraordinary heat but relatively thin fuel.
The Fuels That Make It Possible
The most commonly studied fusion reaction combines two heavy forms of hydrogen: deuterium and tritium. Deuterium is easy to find. About 1 out of every 6,500 hydrogen atoms in seawater is deuterium, giving Earth’s oceans an enormous supply. A single reaction between deuterium and tritium releases 17.6 million electron volts of energy, a huge amount for a single atomic event.
Tritium is the harder ingredient. It’s radioactive with a half-life of about 12 years, meaning it decays quickly and barely exists in nature. Small amounts form when cosmic rays interact with the atmosphere, but nowhere near enough for energy production. The practical solution is to breed tritium by exposing lithium (specifically lithium-6) to the high-energy neutrons that fusion itself produces. Future fusion power plants would essentially manufacture their own tritium fuel in a surrounding “blanket” of lithium, making the reactor self-sufficient.
Containing a Miniature Star
Since fusion plasma is hotter than the core of the sun, no physical material can hold it. Engineers have developed two main approaches to keep the fuel contained long enough for reactions to occur.
The first and most widely pursued method is magnetic confinement. Devices called tokamaks and stellarators use powerful magnetic fields to shape, heat, and suspend the plasma in midair, preventing it from ever touching the reactor walls. The international ITER project in France, the largest fusion experiment ever built, uses a tokamak design and reached a major milestone with its first plasma scheduled for December 2025.
The second approach is inertial confinement. Instead of holding plasma in place with magnets, this method fires enormous lasers at a tiny capsule of fuel from all directions simultaneously. The lasers compress the capsule so dramatically that it’s comparable to squeezing a basketball down to the size of a pea. That compression creates the extreme conditions needed for fusion in a brief, intense burst. The U.S. National Ignition Facility (NIF) uses this approach and has achieved ignition (more energy out than laser energy in) eight times as of May 2025. Its best result came in April 2025, when 2.08 megajoules of laser energy produced 8.6 megajoules of fusion energy, a gain of more than 4.
Turning Fusion Heat Into Electricity
In a deuterium-tritium reactor, about 80% of the energy from each reaction is carried away by high-speed neutrons. These neutrons slam into the blanket surrounding the reactor, transferring their kinetic energy as heat to a coolant flowing through the structure. That heated coolant then drives a turbine to generate electricity, similar to how conventional power plants work. The blanket does double duty: it captures energy and breeds tritium from lithium at the same time.
This steam-turbine cycle, while proven, loses a significant fraction of energy at each conversion step. That’s one reason researchers are also exploring alternative fusion fuels that could skip the steam cycle entirely.
Alternative Fuels and Aneutronic Fusion
Not all fusion reactions produce neutrons. Aneutronic fusion uses heavier elements like helium-3 or boron-11 instead of hydrogen isotopes. A reaction between a proton and a boron-11 nucleus, for instance, releases its energy primarily as fast-moving charged particles rather than neutrons. Because charged particles carry an electrical charge, their energy can be captured directly through electromagnetic induction, potentially converting nuclear energy to electricity without a steam cycle at all.
Aneutronic reactions also produce little to no penetrating radiation, which means less damage to reactor components and fewer safety concerns for nearby workers. The tradeoff is that these reactions are much harder to ignite. Proton-boron fusion yields about 8.7 million electron volts per reaction compared to 17.6 for deuterium-tritium, and it requires even more extreme temperatures. Researchers in China and Russia are experimenting with proton-boron, deuterium-lithium, and helium-helium fuels, but practical aneutronic reactors remain further off than deuterium-tritium designs.
How Fusion Compares to Fission on Safety
Fusion is fundamentally different from fission (the splitting of heavy atoms like uranium) when it comes to accident risk. A fission reactor sustains a chain reaction that must be actively controlled. If that control fails, the reaction can accelerate dangerously. Fusion is the opposite: it requires such precise, extreme conditions that any disruption causes the reaction to simply stop within seconds. A runaway chain reaction is physically impossible. As the International Atomic Energy Agency puts it, fusion is self-limiting: if you lose control, the machine switches itself off.
Fusion also avoids the long-lived radioactive waste that makes fission cleanup so difficult. The main radioactive concern is tritium, which has a half-life of only 12.3 years, compared to the thousands or hundreds of thousands of years for some fission byproducts. Reactor components will become somewhat radioactive over time from neutron bombardment, but this activated material is far less hazardous and shorter-lived than spent fission fuel. Managing tritium leakage remains an active area of safety research, but the overall waste profile is dramatically smaller.
Where Fusion Stands Today
Fusion has crossed several critical scientific thresholds in recent years. The NIF’s repeated ignition shots have proven that net energy gain from fusion is physically achievable, not just theoretical. ITER aims to demonstrate that magnetic confinement can sustain a burning plasma at power-plant scale. Dozens of private companies are also pursuing compact fusion designs with varying approaches, hoping to reach commercial electricity production in the 2030s or 2040s.
The remaining challenges are engineering problems, not physics mysteries. Building materials that can withstand years of neutron bombardment, breeding tritium efficiently enough for a reactor to fuel itself, and maintaining plasma stability for long continuous operation are all active areas of development. The core chemistry and physics of fusion are well understood. The question now is whether the engineering can catch up to the science at a cost that makes fusion electricity competitive.