China’s EAST Tokamak Pushes Past Greenwald Density Limit in ‘Density-Free’ Fusion Regime

In a cavernous hall outside Hefei, a superconducting ring of magnets has quietly pushed past a constraint that has shaped fusion reactor design for nearly four decades.

Physicists operating China’s Experimental Advanced Superconducting Tokamak (EAST) report they have steered its superheated hydrogen plasma into what they call a “density-free” regime, sustaining densities up to 65% higher than a long‑standing operational limit without triggering the violent collapses that usually follow.

The experiments, described Jan. 1 in Science Advances, do not produce net electricity and apply only to a startup phase of the discharge. But they directly challenge the Greenwald density limit, an empirical ceiling that has guided the size and operating conditions of major fusion projects, including the multinational ITER reactor under construction in France.

“We found a practical and scalable pathway for extending density limits in tokamaks and next‑generation burning plasma fusion devices,” said Ping Zhu, a professor at Huazhong University of Science and Technology and a co-lead author of the study, in a statement released by the Chinese Academy of Sciences.

The EAST team achieved line‑averaged electron densities between 1.3 and 1.65 times the Greenwald limit, compared with the machine’s usual operating range of about 0.8 to 1.0, according to the paper. Crucially, the plasma remained stable rather than ending in the rapid, disruptive instabilities that typically mark a breach of the limit.

A limit that wasn’t supposed to budge

First proposed in 1988 by American physicist Martin Greenwald, the density limit bearing his name is a simple relation connecting a tokamak’s maximum achievable plasma density to its plasma current and cross‑sectional area. Beyond that threshold, experiments across multiple machines showed, denser plasmas tended to cool at the edge, draw in impurities from the wall and abruptly shut down.

Although Greenwald’s formula was based on data fits rather than fundamental theory, reactor designers adopted it as a practical ceiling. ITER, designed to demonstrate large‑scale fusion energy production, and planned demonstration power plants known as DEMO devices have assumed they should operate at or below roughly the Greenwald density to avoid catastrophic disruptions.

“The density limit has long been treated as a hard boundary in design studies,” said an independent European fusion researcher who was not involved in the EAST work but has published on density limits. “If you can reliably move that boundary, you change the design space for future reactors.”

How EAST stepped around the ceiling

EAST’s breakthrough lies less in raw numbers—other experiments have exceeded the Greenwald value under specialized conditions—than in how the Chinese team reached and sustained the higher densities in a reactor‑relevant machine.

The device, operated by the Institute of Plasma Physics under the Chinese Academy of Sciences in Hefei, is the world’s first fully superconducting tokamak and is used heavily as a test bed for ITER‑like technologies.

In the new campaign, led by first author Jiaxing Liu and co‑led by Zhu and associate professor Ning Yan of the Hefei Institutes of Physical Science, researchers modified the earliest moments of the plasma’s life. They increased the neutral gas pressure in the vacuum vessel before ignition and applied targeted electron cyclotron resonance heating (ECRH)—high‑frequency microwaves that heat electrons—during the ohmic startup phase, when current in the plasma first ramps up.

According to the paper and institutional summaries, this combination kept the plasma edge and divertor region—where the hot gas interacts most directly with the metal walls—relatively cool at the outset. That, in turn, reduced sputtering of impurities from the wall and limited radiative energy losses that often lead to edge cooling and collapse as density rises.

“These experimental achievements provide new physical insights into breaking through the long‑standing density limit in tokamak operation in pursuit of fusion ignition,” the Chinese Academy of Sciences said in a statement.

The team interprets the result as experimental access to a “density‑free regime” predicted by a framework known as plasma–wall self‑organization (PWSO). In that picture, a fusion device can operate in two broad basins: one in which traditional density limits emerge from runaway impurity radiation and edge cooling, and another in which the plasma and wall settle into a stable pattern of interaction that allows density to increase without immediately triggering those instabilities.

In their Science Advances paper, the authors argue that EAST’s new operating point falls squarely in this latter basin.

Not the first breach, but a different kind

The Greenwald limit has been challenged before. In 2024, researchers at the Madison Symmetric Torus in Wisconsin reported tokamak plasmas with densities about 10 times the Greenwald value, aided by a thick conducting wall and fast‑acting power supplies. Those discharges, however, ran at low magnetic field and temperature, far from the conditions needed for a fusion power plant, and the team cautioned that their approach was unlikely to translate directly to devices such as ITER.

Other experiments on the DIII‑D tokamak in San Diego and earlier campaigns on EAST itself have operated modestly above the Greenwald limit while maintaining good energy confinement, using scenarios with high so‑called poloidal beta to suppress turbulence. And in 2022, a group at the Swiss Federal Institute of Technology in Lausanne derived a new density limit scaling from first principles, suggesting ITER might safely run at nearly double the originally assumed fuel density under high heating power.

Taken together, these efforts have gradually eroded the notion of the Greenwald limit as a fundamental wall. The EAST result adds a new piece by tying a sustained breach of the limit in a metal‑walled, ITER‑like tokamak to a specific, theory‑driven operational regime.

What it could mean for ITER and beyond

If density‑free operation proves robust and extendable beyond startup, the implications for future fusion reactors could be substantial.

Fusion power output scales strongly with plasma density at a given temperature and volume. Being able to run even 50% denser than previously assumed could allow designers to squeeze more power from the same size machine or, conversely, to design smaller and potentially cheaper reactors for a given power level.

Zhu and his co‑authors argue that the ECRH‑assisted startup they demonstrated is “scalable” to next‑generation burning plasma devices. But they and outside experts also caution that several steps remain.

So far, EAST’s density‑free regime has been accessed during the early ohmic phase and carried into an initial flat‑top, not during the high‑confinement, high‑power operating mode that a commercial plant would require. Yan said in the academy’s statement that the team plans “to apply the new method during high‑confinement operation on EAST in the near future in an attempt to access the density‑free regime under high‑performance plasma conditions.”

It is also unclear how easily the same knobs can be turned on ITER, where the startup scenario, heating systems and engineering constraints differ. Even in a density‑free regime, other limits—such as materials damage from neutrons, heat loads on the divertor and the need to breed enough tritium fuel in the reactor’s blanket—will continue to bound performance.

Physicists stress that “density‑free” does not mean density can increase without bound. It signals that the specific empirical Greenwald scaling is no longer the dominant constraint in that regime.

China’s broader fusion push

The result bolsters China’s growing profile in magnetic confinement fusion. EAST, often referred to in state media as the country’s “artificial sun,” has previously set records for plasma temperature and confinement time. China is a key partner in ITER but is also pursuing its own roadmap toward a demonstration power plant, with conceptual designs such as the China Fusion Engineering Test Reactor.

For climate and energy policy, the near‑term impact of the new experiments is limited. Even under optimistic scenarios, large‑scale fusion power plants are not expected to contribute to electricity grids until the 2030s or 2040s. The main effect of results like EAST’s is to reduce technical risk and widen the design envelope for those future machines.

Still, for a field that has long been hemmed in by operational “don’ts,” redefining one of its best‑known limits marks a notable shift. Instead of building ever larger devices to stay safely below an assumed density ceiling, fusion researchers are increasingly probing how to shape the plasma—from its very first microseconds—so that the old ceiling simply does not apply. EAST’s latest shots suggest that, at least in one machine and one regime, that strategy is beginning to work.