China fusion breakthrough tackles heat and edge bursts in EAST reactor

A minute-long plasma state in China’s EAST reactor may have found a rare balance fusion engineers chase: it eased heat on the divertor while holding the edge steady enough to avoid damaging bursts. The team led by Guosheng Xu at the Institute of Plasma Physics in Hefei reported the result in Physical Review Letters, which lists the paper as published on March 23, 2026 after receipt on October 31, 2025 and acceptance on February 17, 2026.

That combination matters because divertors have to absorb enormous heat flux from the plasma exhaust, and edge-localized modes, or ELMs, can hammer reactor surfaces with sudden pulses of heat and particles. Push cooling too far and confinement can slip; push confinement too hard and the wall takes the punishment. The EAST result points to a regime that tries to soften both problems at once instead of forcing one tradeoff.

The researchers call the state the Detached divertor and Turbulence-dominated Pedestal, or DTP, regime. In EAST, a metal-wall tokamak in Hefei, the plasma was sustained on a minute scale with feedback-controlled seeding of a light impurity gas. The paper says the device reached partial divertor detachment, remained ELM-free in high-confinement mode, and still kept strong pedestal performance. A closed divertor geometry helped trap and pump neutral particles in the divertor region, which eased the cooling of the pedestal that usually follows aggressive exhaust handling.

What the team demonstrated is the operating state itself. What the paper infers is the mechanism: reduced ionization and stronger pumping lowered pedestal cooling, which raised the pedestal temperature gradient and drove broadband turbulence identified as temperature-gradient-driven trapped electron modes. That turbulence carried particles and heat outward in a way that kept the edge stable. The Physical Review Letters abstract says the result may be especially relevant for ITER, where pedestal conditions could more readily excite the same kind of turbulence.

For now, the real test is replication, first in other metal-wall devices and then over longer pulses and harsher heat loads. If that balance can be repeated, EAST will have done more than notch a plasma-physics milestone. It will have shown a plausible path toward a reactor edge that does not force fusion to choose between confinement and survival.