For decades, a rule of thumb called the Greenwald limit told fusion physicists how much fuel they could pack into a tokamak before the plasma collapsed. Push the density too high, and the superheated gas would radiate away its energy, crash into the reactor walls, or simply disintegrate. The limit was not derived from first principles; it was stitched together from hundreds of failed and successful discharges across machines worldwide. But it held so consistently that reactor designers treated it as a hard ceiling when planning future devices, including ITER, the massive international fusion project under construction in southern France.
Now a team at China’s Experimental Advanced Superconducting Tokamak, known as EAST, has punched through that ceiling. In results published in Science Advances in early 2025, researchers at the Institute of Plasma Physics in Hefei reported stable plasma discharges at line-averaged electron densities between 1.3 and 1.65 times the Greenwald limit. The plasma did not collapse. It did not trigger the violent disruptions that have ended countless high-density experiments on other machines. It held.
How they did it
The key was changing how the plasma was born. In a conventional tokamak startup, operators ramp up the plasma current while gradually fueling the chamber with hydrogen isotopes. The EAST team instead used electron cyclotron resonance heating (ECRH) to assist the ohmic startup phase, injecting targeted microwave energy that shaped the plasma’s behavior from its earliest moments. By optimizing the prefill neutral gas density and the timing of the heating pulse, they created conditions where the plasma-wall interaction stabilized rather than spiraled toward collapse.
The theoretical explanation draws on a model called plasma-wall self-organization, or PWSO, which maps how startup conditions cascade through a chain of physical dependencies: wall recycling, radiative losses, divertor behavior, and ultimately the density ceiling itself. That model was developed and tested on the J-TEXT tokamak and described in a 2023 preprint on arXiv. The EAST team’s Science Advances paper applies the PWSO framework to explain why their discharges survived in what the authors call a “density-free regime,” an operating space where the conventional Greenwald scaling no longer dictates the upper bound.
The term “density-free regime” is the authors’ own coinage, not yet standard vocabulary in the fusion community. But the experimental data backing it are specific and peer-reviewed: detailed plasma profiles, stability measurements, and discharge-by-discharge comparisons published in the open-access version of the paper.
EAST is not alone
China’s result is striking, but it is not the only evidence that the Greenwald limit can be exceeded. At the DIII-D tokamak in San Diego, operated by General Atomics for the U.S. Department of Energy, researchers have sustained plasmas above the Greenwald threshold using a completely different technique: negative triangularity shaping. By sculpting the plasma cross-section into an inverted-D shape rather than the conventional D, the DIII-D team achieved Greenwald fractions approaching 2, roughly double the supposed ceiling.
The two approaches attack the same problem from opposite directions. EAST’s method optimizes the plasma’s birth and its interaction with surrounding surfaces. DIII-D’s method reshapes the plasma’s geometry using magnetic fields. That both techniques independently breach the same empirical barrier suggests the Greenwald limit is not a wall built into the physics of magnetized plasma. It is a conditional boundary, one that depends on choices engineers and physicists make about how to create and confine their fuel.
No peer-reviewed study has yet compared the two approaches head-to-head or tested whether combining ECRH-assisted startup with negative triangularity shaping could push densities higher still. That experiment would be a logical next step, though it would require a machine flexible enough to accommodate both techniques simultaneously.
What this means for ignition
Fusion ignition, the point at which a burning plasma sustains itself without external power, depends on three variables multiplied together: density, temperature, and confinement time. Physicists call this combination the Lawson triple product. Raising any one of the three brings the product closer to the ignition threshold, but all three must be high enough at the same time.
The EAST result moves the density lever. If tokamaks can reliably operate well above the Greenwald limit, reactor designers gain a degree of freedom they did not have before. Higher density means more fuel atoms colliding per second, which means more fusion reactions per unit volume. That could allow future reactors to reach ignition conditions in smaller, less expensive machines, or to produce more power from devices already on the drawing board.
But the Science Advances paper does not claim to have reached ignition-relevant conditions overall. The temperatures and confinement times in the EAST discharges were not simultaneously at the levels needed for a self-sustaining burn. Institutional press materials describing the Greenwald limit as “unbreakable” have amplified the result’s perceived significance beyond what the paper itself argues. The researchers demonstrated that a specific barrier can be moved; they did not demonstrate that moving it is sufficient to reach ignition.
The gaps that remain
Several questions stand between this laboratory result and a working power plant.
Duration is the most immediate. The published data describe individual plasma discharges, not sustained high-density campaigns lasting minutes or hours. A commercial fusion reactor would need to hold above-Greenwald densities continuously during operation. Whether the density-free regime remains stable over long pulses, or whether slow-building instabilities eventually reassert the old limit, is unknown.
Scale is the deepest uncertainty. EAST is a mid-sized tokamak with a major radius of about 1.9 meters. ITER’s major radius is 6.2 meters, and plasma behavior can shift dramatically as machines grow. Turbulence patterns change. Wall interactions evolve. Energy confinement scaling laws that work at one size can break at another. No publicly available analysis has mapped the density-free regime onto ITER’s projected operating parameters. Until above-Greenwald operation is demonstrated on a reactor-scale device, the practical relevance of the EAST finding for commercial energy remains a projection, not a proof.
The PWSO model itself also needs further validation. The arXiv preprint from J-TEXT provides a coherent theoretical framework, but it has not yet completed formal peer review. Its predictions are plausible and consistent with the EAST data, but the model’s status is closer to “promising hypothesis” than “established physics.” Future peer-reviewed work testing PWSO predictions across multiple tokamaks would strengthen the case considerably.
Where fusion density research goes from here
The convergence of the EAST and DIII-D results is reshaping how the fusion community thinks about density limits. For years, the Greenwald scaling was treated as a design constraint baked into every tokamak blueprint. If it turns out to be a movable boundary, reactor concepts that were previously dismissed as too ambitious may deserve a second look.
The next decisive tests will likely come from two directions: longer-duration above-Greenwald discharges on existing machines, and attempts to replicate the result on larger devices as they come online. ITER, if its construction and commissioning stay on track, would be the ultimate proving ground, though its first deuterium-tritium experiments are still years away.
Private fusion companies, several of which are designing compact tokamaks that would benefit enormously from higher operating densities, are also watching closely. A reliable path past the Greenwald limit could change the economics of their reactor designs, potentially shrinking the size and cost of machines needed to reach net energy gain.
For now, the EAST result is best understood as a genuine and carefully documented advance that opens a door physicists were not sure could be opened. Walking through it, and building a power plant on the other side, will take years of additional work. But the door is open, and more than one team knows how to push on it.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
