Nuclear fusion

Atomic nuclei carry positive electric charge and repel one another. At sufficiently high energies, some nuclei can approach closely enough for the strong nuclear force to bind them together. When light nuclei fuse into a more tightly bound product, the small difference in mass is released as energy. Fusion is therefore distinct from nuclear fission, which releases energy by splitting heavy nuclei. The outcome depends on the fuel and conditions: fusion is not simply the joining of any two atoms.

Fusion supplies the energy of the Sun and other stars. In a star, immense gravitational pressure helps keep hot matter together so that nuclei collide and react. Experiments on Earth cannot reproduce stellar gravity, so researchers must create extremely energetic conditions and confine the fuel long enough for a useful number of reactions to occur. The fuel becomes plasma—an ionized state of matter made up of charged particles. For prospective power plants, the deuterium–tritium reaction is a leading candidate: it produces a helium nucleus, a neutron and energy. Deuterium is a hydrogen isotope; tritium is another, radioactive isotope that is scarce in nature and would need to be supplied or bred in a reactor system.

A practical experiment must heat the plasma, provide enough particle density for collisions and keep the fuel confined for sufficient time. Magnetic-confinement devices use strong fields to guide charged plasma; tokamaks and stellarators are two major designs. ITER is an international tokamak experiment intended to study a self-heating, or burning, plasma. In inertial-confinement fusion, a very small fuel capsule is rapidly compressed and heated, commonly by powerful laser pulses, so fusion occurs before the fuel can expand. These approaches address the same physical problem through different techniques, and experimental success in one does not by itself establish that a power plant can operate economically.

On 5 December 2022, the National Ignition Facility at Lawrence Livermore National Laboratory reported a controlled experiment in which fusion produced more energy than the laser energy delivered to its target: 3.15 megajoules of fusion yield from 2.05 megajoules of laser energy on target. This target-level gain was an important physics milestone, not a demonstration of a facility supplying net electricity: the comparison does not include the energy used to operate the entire laser facility or the systems needed to convert reactor heat into power. LLNL reported further ignition results, including an eleventh ignition experiment in June 2026. The distinction between reaction yield, energy delivered to fuel and electricity available to users is essential when interpreting fusion results.

Creating fusion reactions is only one step toward a power system. A deuterium–tritium reactor would have to capture the energy carried largely by neutrons as heat and convert that heat into electricity. It would also need a dependable tritium fuel cycle, components that withstand intense neutron exposure, effective removal of heat and particles from the plasma, and maintenance methods suitable for components exposed to intense neutron radiation. ITER describes these as continuing research and engineering challenges. The promise of fusion as a possible low-carbon energy source therefore remains conditional on solving demanding problems in materials, fuel production, plant efficiency, reliability and cost.