What inertial confinement is

Inertial confinement does not confine anything for long: it bets on compressing and igniting the fuel so fast that it fuses before its own inertia lets it expand and cool. A spherical capsule the size of a pinhead, with a frozen layer of deuterium and tritium, is compressed with laser beams to densities on the order of a hundred times lead's and temperatures above 100 million degrees at its center, all within nanoseconds. There are two ways to deliver that energy: in direct drive, the lasers aim straight at the capsule; in indirect drive, they aim at the inner walls of a metal cylinder called a hohlraum, which the walls convert into X-rays, and these compress the capsule from every direction more evenly. Indirect drive is less energy-efficient, but more tolerant of implosion imperfections, and it is what the world's two largest facilities use.

The National Ignition Facility (NIF), at Lawrence Livermore National Laboratory (LLNL), in California, fires 192 laser beams that deliver up to 2 megajoules of ultraviolet light onto a hohlraum the size of a pencil eraser. On 5 December 2022, a NIF shot delivered 2.05 MJ of laser energy to the capsule, and it released 3.15 MJ of fusion energy: the first time in history a controlled fusion experiment produced more energy than was delivered directly to it, what the Department of Energy and LLNL called ignition and scientific breakeven. The result was repeated and improved many times since: the highest yield LLNL has reported is 8.6 MJ from 2.08 MJ of laser input, in April 2025, a target gain of 4.1. Even so, no NIF shot came anywhere close to returning the grid electricity its lasers consumed to fire: those lasers are inefficient, and much of the energy is lost before it reaches the fuel.

France has its own equivalent: the Laser Mégajoule (LMJ), run by the CEA at the CESTA center near Bordeaux, designed, in its full configuration, to converge 176 laser beams onto deuterium-tritium targets by indirect drive — a fraction of which are already operating while the rest are commissioned progressively — delivering more than a megajoule of ultraviolet light. LMJ pursues the same goal as NIF — achieving ignition and propagating burn in the laboratory — and is, per the CEA itself, the centerpiece of its Simulation programme for maintaining the French nuclear deterrent without nuclear testing.

Neither NIF nor LMJ was built to generate electricity, and neither is aimed at doing so any time soon. Their stated mission is stockpile stewardship without nuclear testing — NIF answers to the US National Nuclear Security Administration, and LMJ to the CEA's deterrence program — studying how matter behaves under the extreme conditions of a thermonuclear explosion without detonating one. That a single shot releases more fusion energy than the lasers receive does not mean the facility is a net energy source: the electrical energy the lasers consume to produce those few megajoules of light must be subtracted, and today it is one or two orders of magnitude greater than the fusion energy released. And neither facility fires more than a few times a day, far from the several shots per second a hypothetical inertial-confinement power plant would need. Turning inertial confinement into a power source would also require a driver — a laser or some other kind — able to withstand hundreds of millions of shots per year at about 10 Hz without replacing the target chamber components, and a way to manufacture and inject a fresh, precisely shaped capsule for every pulse — engineering problems neither NIF nor LMJ was designed to solve.