Innovation

In Cadarache, France, a tokamak held plasma steady for 1,337 seconds on a February morning, and that twenty two minute run pushed fusion’s endurance record up by 25 percent

By Hugo Rojas · November 11, 2026 · 10:50 AM · 5 min read
Tokamak reactor vessel glowing during 1337 second plasma record test, tokamak held plasma

A control room near Cadarache, France went silent for twenty two minutes straight on a February morning.

Inside a doughnut shaped steel vessel, a loop of ionized gas hotter than the center of a star hung suspended by magnetic fields.

Nobody touched it.

Engineers elsewhere are tackling parallel endurance problems of their own, including grid scale storage tools that fusion power would eventually need to pair with.

It simply held.

When the superheated ring of plasma finally cooled, the clock read 1,337 seconds, a new mark for any fusion machine on Earth.

So what actually kept that gas from collapsing into the wall around it?

What actually keeps a loop of 50 million degree gas from touching the wall

A tokamak like WEST works by trapping plasma in a magnetic cage shaped like a hollow bagel, so the superheated gas never makes contact with the metal vessel walls around it. The instant plasma brushes metal, it cools, contaminates and the reaction dies, which is the entire struggle of fusion engineering.

So the real achievement was not the heat itself but the stamina of the hardware wrapped around it. Tungsten lines the divertor region, where the plasma’s heat and particles get steered away from the core chamber.

It tolerates extreme heat loads, but only with careful management to stop stray particles from drifting back into the plasma. Which means the record was really a test of a metal, not just a gas.

That tungsten lining had to shed an enormous heat load for over twenty minutes without flaking a single particle into the plasma it was guarding, a feat of materials endurance as much as physics.

Inside the machine that broke its own continent’s record

WEST sits at a national atomic energy research campus in Cadarache, in southern France, the same sprawling site that hosts the much larger international ITER project now under construction nearby. The tokamak itself is modest by comparison, a research scale device built specifically to test plasma facing components rather than to generate power.

That modesty is the point. Before any larger reactor can run for minutes, let alone hours, engineers need working proof that interior surfaces survive sustained contact with plasma at temperatures no structural material was ever meant to touch directly.

The run beat a rival machine by a clear margin. On February 12, WEST maintained blistering plasma for 1,337 seconds, 22 minutes and 17 seconds in total, edging past the previous duration record of 1,066 seconds set only weeks earlier by China’s EAST tokamak.

That gap placed two national fusion programs within weeks of each other, trading the same record back and forth like a relay baton.

How the number was measured, and why 22 minutes matters more than it sounds

The 1,337 second figure came from a single, dated experimental run rather than a projection or simulation, logged by instruments tracking the plasma’s temperature and density in real time throughout the shot. According to the lab’s own account of the run, the plasma, numbered 61299, was sustained for 1337 seconds, with a power injected by the heating system and extracted by the actively cooled tungsten components facing the plasma of 2.6 GJ. That precision matters, because it means WEST topped EAST’s mark by 4 minutes and 29 seconds, a genuine 25 percent improvement in plasma duration.

Twenty five percent does not sound dramatic until you remember that every additional second asks the tungsten divertor to shed more heat without degrading. Sustaining plasma for over twenty minutes, rather than mere seconds, is the difference between a laboratory curiosity and a working reactor cycle.

Researchers were candid that the run was a step, not a finish line. Anne-Isabelle Etienvre, the CEA’s Director of Fundamental Research, said WEST has achieved a new key technological milestone by maintaining hydrogen plasma for more than twenty minutes through the injection of 2 megawatts of heating power, with the next goal being several hour long runs at higher temperatures.

Where the record stops mattering and the next problem begins

Duration alone does not make electricity. The WEST shot ran at roughly 50 million degrees Celsius, well above the surface of the sun, yet still short of the density and heat combination a commercial reactor would need to produce more energy than it consumes.

Even so, the test answered a narrower and arguably harder question. Reaching net energy depends on several factors at once: plasma density, temperature, and how well the machine retains heat and particles.

WEST was built to isolate just the heat retention piece of that puzzle, and for 1,337 seconds it did not fail. Other fusion efforts chase different pieces of the same puzzle elsewhere, a reminder that the field is now a crowded, competitive floor rather than any single lab’s private project.

One team recently tested mechanical compression methods instead of magnetic confinement, a sign that no single approach has pulled ahead for good.

What a twenty two minute run buys the next generation of reactors

The honest caveat is that WEST is a test bed, not a power source, and the record will likely be broken again within a year by either EAST or a newer machine. Fusion’s history is a relay of records traded in months, not decades, and this one is no exception.

What the run does buy is confidence for the plasma facing components inside ITER, the much larger reactor rising at the same Cadarache campus, since its tungsten walls will need to survive far longer exposures than 1,337 seconds. The lab’s own published account of the run frames it explicitly as groundwork for that larger machine.

Engineers elsewhere are tackling parallel endurance problems of their own, including grid scale storage tools that fusion power would eventually need to pair with.

For now, the twenty two minutes in southern France stand as the longest a tokamak has ever kept its star in a bottle, measured, logged and already being chased by the next shot on someone else’s schedule.

RE+
Hugo Rojas
Hugo Rojas

Hugo is an engineer with strong technical expertise. Multilingual from an early age, his writing combines technical clarity with a strong interest in science and energy.

Hugo_writer
Hugo Rojas

Hugo is an engineer with strong technical expertise. Multilingual from an early age, his writing combines technical clarity with a strong interest in science and energy.