The process by which light atomic nuclei combine to form heavier nuclei, releasing vast amounts of energy — the same reaction that powers the Sun and all stars.
Nuclear fusion occurs when two light atomic nuclei overcome their mutual electrostatic (Coulomb) repulsion and merge to form a heavier nucleus. Because the mass of the product nucleus is slightly less than the sum of the reactant masses, energy is released according to Einstein’s mass–energy equivalence, E = mc². The energy appears primarily as kinetic energy of the reaction products — fast neutrons, alpha particles, or protons — and can in principle be captured as heat to drive a power cycle.[1]
At the temperatures required for fusion (tens to hundreds of millions of degrees), matter exists as plasma: a gas of free electrons and ions. In this state, ions move fast enough that a small fraction can quantum-mechanically tunnel through the Coulomb barrier even at energies below its classical peak. The probability of a fusion reaction occurring is quantified by the cross-section σ(E), which depends sharply on the relative kinetic energy of the colliding ions.[2]
In 1957, John Lawson established the minimum conditions under which a fusion plasma can produce net energy. His criterion requires that the product of fuel-ion density n and energy confinement time τE exceed a threshold that depends on temperature. For D–T fuel near the optimal temperature of ~15 keV, the requirement is approximately:
When temperature is included, the “triple product” n·T·τE becomes the standard figure of merit for confinement performance. Decades of experimental progress have steadily increased the achieved triple product toward the break-even threshold.[1]
The ratio of fusion power produced to heating power supplied is called the fusion energy gain factor, Q. A plasma with Q = 1 (“scientific break-even”) produces as much fusion energy as is injected to sustain it. A power plant requires Q ≥ 10 to be economically viable after accounting for recirculating power. The condition Q → ∞ corresponds to ignition, where the plasma sustains itself entirely through self-heating by fusion-born alpha particles.[3]
Two principal approaches have been pursued for over six decades. Magnetic confinement fusion uses strong magnetic fields to hold a low-density plasma for seconds or longer — the tokamak is its most developed device. Inertial confinement fusion compresses a tiny fuel capsule to extreme density in nanoseconds, relying on the fuel’s own inertia for confinement. Each approach faces distinct physics and engineering challenges, and both have demonstrated significant fusion energy production in laboratory experiments.[4]
As of 2025, no facility has yet demonstrated sustained net electricity from fusion. The ITER tokamak, under construction in France, aims to achieve Q ≥ 10 in deuterium–tritium plasmas. The U.S. National Ignition Facility demonstrated target energy gain greater than unity (energy out exceeding laser energy delivered to the target) in December 2022, a verified result confirmed by peer-reviewed publication. These milestones, while not yet constituting a power plant, represent measurable progress toward the goal of commercial fusion energy.[4]