Wiki · Concept
Nuclear fusion: how a fusion reactor works
Nuclear fusion is the reaction that powers the Sun: two light nuclei join into a heavier one and release even more energy per gram than fission. A fusion reactor is a machine built to sustain that reaction on Earth, and none has done so commercially yet.
In fusion, two light nuclei, typically deuterium and tritium, join to form helium and release a very energetic neutron. Per unit of mass it releases about four times more energy than fission, and its fuel comes from water and lithium.
The problem is that nuclei repel each other electrically. To make them fuse they must be heated above 100 million degrees and kept confined, with magnetic fields in a tokamak or a stellarator, or by compressing them with lasers. Getting more energy out than is put into heating the plasma is the goal that ITER, in France, and several private companies are chasing this decade.
That is why this atlas shows only fission reactors: they are the only ones delivering electricity to a grid. The 167 fusion devices in existence, from JET to ITER, are experiments. The IAEA maps them in its FusDIS system.
Types of devices. Magnetic confinement by an induced current is a tokamak; by twisted coils, with no current, a stellarator. Inertial confinement compresses a capsule with lasers instead of holding the plasma for seconds. And there is a family of alternative concepts — magnetic mirrors, field-reversed configurations, Z-pinches, spheromaks, magnetized target fusion — chasing the same goal by other paths.
Quick facts
| Easiest reaction | D + T → He-4 + n, 17.6 MeV |
|---|---|
| Temperature required | above 100 million °C |
| Devices worldwide | 167 according to FusDIS, none commercial |
| ITER | under construction at Cadarache, start of research operation planned for 2034 |