Nuclear fusion is the process of combining light atomic nuclei to release energy. Controlling fusion is straightforward so long as the goal is not to produce electricity; it is possible to build a fusion reactor on a bench top. The challenge lies in generating net power.
In 1964 Philo Farnsworth, better known as the inventor of electronic television, worked out the design for a device known as the Farnsworth fusor. It uses two concentric, spherical metal grids inside a vacuum chamber as electrodes. Deuterium (a heavy isotope of hydrogen with a neutron as well as a proton in its nucleus) is injected, ionised and accelerated inward. If the voltage is high enough, nuclei travelling through the centre can fuse on collision, producing either a helium-3 nucleus and a spare neutron or a tritium nucleus and a proton—both reactions releasing energy as heat. The design is simple enough that hobbyists can build fusors, but Farnsworth could not coax them into generating net power.
Avalanche Energy, an American startup based in Tukwila, a suburb of Seattle, has revisited the fusor concept. Its device, the Orbitron, replaces Farnsworth's spherical geometry with a cylindrical vacuum chamber and a rod-shaped cathode. Ionised nuclei spiral around the cathode rather than colliding with it, while a magnetic field forces free electrons to circle similarly, balancing charges and allowing a far higher density of nuclei to be sustained for longer. Trials with deuterium have successfully generated neutrons. Commercial reactors would use a deuterium-tritium mix, capable of a hundred times as many collisions, though tritium is radioactive, rare in nature and must be manufactured in a nuclear reactor.
Individual Orbitron units would produce around 15kW; a battery of them generating 1MW or more would fit in a shipping container. Potential uses include powering robot ships and submarines, isolated Arctic bases, military applications and spacecraft.
Global private investment in fusion rose from around $7bn in 2024 to $10bn in 2025, according to the International Atomic Energy Agency. Commonwealth Fusion Systems, an American startup, expects to produce its first plasma by 2027. Europe's commercial fusion activity lags; its bets are mostly tied to ITER, a 35-country project in southern France that will not produce plasma before 2033 and is not expected to be fully operational until 2039.
See also fusion energy for the tokamak approach and commercial fusion startups.
I don't deserve this award, but I have arthritis and I don't deserve that either.