Nuclear fusion

Nuclear fusion wasn't discovered in a laboratory—it was first proposed as the power source of stars. In 1920, Arthur Eddington published "The Internal Constitution of the Stars," speculating that stars fused hydrogen into helium, liberating enormous energy via E = mc². He calculated stellar cores would need 40 million Kelvin—a figure so extreme it sparked fierce debate.

Throughout the 1920s, mathematical foundations fell into place: George Gamow introduced quantum tunneling in 1928; Robert Atkinson and Fritz Houtermans showed fusion could occur at lower energies than believed. In 1934, Ernest Rutherford's team achieved the first experimental fusion, slamming deuterium atoms together to produce helium and "an enormous effect."

The breakthrough came from Hans Bethe in 1938-1939. He described the two nuclear reactions stars use: the proton-proton chain and the carbon-nitrogen-oxygen cycle. This earned him the 1967 Nobel Prize and established fusion as the fundamental power source of the universe.

The central challenge is recreating stellar conditions without stellar gravity. Fusion requires 100 million degrees Celsius—ten times hotter than the Sun's core. Two approaches emerged: magnetic confinement (MCF) using powerful magnetic fields to trap plasma in toroidal chambers like tokamaks, and inertial confinement (ICF) compressing tiny fuel pellets with lasers.

Despite a century of progress, fusion remains impractical. The National Ignition Facility's December 2022 breakthrough produced 3.15 MJ from 2.05 MJ of laser energy—but the lasers required 300 MJ from the grid. Materials must withstand plasma temperatures while being bombarded by high-energy neutrons. Tritium is scarce with only 4 kg produced annually. ITER, proposed in 1985 with $5B budget, has now consumed $25B and won't achieve first plasma until 2033.

NIF has achieved ignition seven times by February 2025, with April 2025 yielding 8.6 MJ from 2.08 MJ laser energy. Most experts frame fusion as "when, not if"—but that "when" remains 2050 at earliest.