China has just completed the largest superconducting magnet ever built for a fusion reactor and is preparing a machine that will attempt to generate electricity by 2030. But it has not ignited a star, it does not yet have an unlimited energy power plant, and not all the records announced measure the same thing. This is the complete story of a project advancing rapidly between real scientific discoveries, gigantic industrial ambitions, and a considerable cloud of misinformation.
No, this is not a series starring Zhang Linghe. But the resemblance is tempting: the actor has a legion of followers who turn on screens just by seeing him, and China has a legion of engineers trying to ignite a plasma with magnetic fields. One of those things is real. The other is entertainment. Let’s go with the real one.
What China has just done, and which has triggered headlines around the world, is complete the largest superconducting magnet ever built for a nuclear fusion reactor. The Institute of Plasma Physics of the Chinese Academy of Sciences finished manufacturing and testing in June 2026 two decisive magnetic systems in Hefei, Anhui province. The most spectacular is a toroidal field magnet weighing 582 tons: 21 meters long, 12 meters wide and 3.3 meters high. According to Chinese institutions, its volume is equivalent to 1.3 times that of a comparable magnet from the international ITER project and can store nearly three times more magnetic energy. Alongside it, a central solenoid made with high-temperature superconductors was tested.
Both components passed technical acceptance procedures on June 27. They were not switched on to produce fusion, they did not generate electricity, and they do not even constitute a reactor by themselves. What was achieved was to demonstrate that China can now manufacture, with technologies developed within the country, a part of the magnetic musculature that its future reactors will need. It is as if a Formula 1 team had manufactured the best engine in the world and tested it on the test bench. The car is not yet running, but the engine is already theirs.
But no matter how many magnets they put on this machine, no sun shines brighter than Zhang Linghe. So, let’s see what parameters the Chinese technical teams are considering so that their “artificial sun” is not just a front-page headline, but a true milestone in physics.
The news of the magnet multiplied during the last days of July, accompanied by headlines such as “China turns on its artificial sun,” “unlimited energy closer than ever,” or “Chinese reactor reaches temperature six times hotter than the Sun.” Each of those phrases contains a portion of truth, but together they produce a completely false image.
Let’s start with the principles of physics, which are more reliable than the plot of a historical romance series.
HOW A STAR IGNITES (AND WHY EARTH CANNOT IMITATE IT)
Everything we know is made of atoms. Fusion occurs when two light nuclei manage to get close enough to merge into a heavier one, transforming a small part of their mass into a enormous amount of energy. At the center of the Sun, the gravity of a gigantic mass compresses matter until it reaches extreme pressures and temperatures. Under those conditions, hydrogen nuclei overcome the electrical repulsion that normally keeps them apart.
To understand it, imagine two magnets facing each other with the same poles: they repel each other. The same happens with atomic nuclei. To make them collide, we must hurl them at each other with enormous speed. Temperature, at the atomic scale, is movement: the hotter the fuel, the faster its particles move.
The Sun has an advantage we do not: its own gravity. Earth cannot gather a stellar mass inside a laboratory. That is why terrestrial reactors, called tokamaks (doughnut-shaped chambers), must compensate for the lack of pressure by greatly increasing the temperature. The center of the Sun is about 15 million degrees Celsius; tokamaks must reach 100 million degrees or more. The paradox is explained: the reactor is hotter than the Sun, but much less dense. Since it cannot squeeze the fuel with stellar gravity, it needs to stir its particles at higher speeds.
And how do you place something at 100 million degrees inside a machine without it melting? The answer is that you do not let it touch the machine.
A MAGNETIC TRACK FOR MATTER THAT CANNOT BE TOUCHED
At those temperatures, the fuel ceases to behave like an ordinary gas and becomes plasma (a gas of electrically charged particles). Because the particles have charge, they can be guided by magnetic fields. The tokamak is a hollow doughnut surrounded by extremely powerful electromagnets. Magnetic fields create an invisible track that forces the plasma to circulate without touching the walls.
The toroidal magnet generates the field that runs along the circumference of the doughnut; the central solenoid acts as a transformer, helping to induce the plasma current. That is why the 582-ton magnet is crucial: confining plasma requires enormous, stable, and extraordinarily precise magnetic fields. An ordinary magnet would consume prohibitive amounts of electricity and overheat to the point of being unusable. Superconductors, by contrast, can conduct current with practically zero electrical resistance when kept at cryogenic temperatures. Here another irony appears: to control matter at 100 million degrees, parts of the machine must be kept near absolute zero. Inside the same device coexist one of the hottest places created by humanity and some of the coldest.
EAST: THE LONG-DISTANCE RUNNER THAT IS NOT A POWER PLANT
When the Chinese “artificial sun” is mentioned, it generally refers to EAST (Experimental Advanced Superconducting Tokamak), installed in Hefei. Its mission is to solve one of the great problems of fusion: it is not enough to reach a very high temperature for a fraction of a second. A power plant needs to keep the plasma confined, stable, and controllable for extended periods.
On January 20, 2025, EAST maintained a steady-state, high-confinement plasma for 1,066 seconds (17 minutes and 46 seconds), far surpassing the 403 seconds achieved by the same facility in 2023. “High confinement” means that energy escapes more slowly, like a thermos that keeps coffee hot.
But beware: EAST did not sustain a self-sustaining fusion reaction generating electricity during those 17 minutes. It maintained an experimental plasma under necessary, but not sufficient, conditions for a power plant.
BREAKING THE DENSITY CEILING
In early 2026, another EAST advance came to light. For decades, researchers have observed that plasma density in a tokamak has a ceiling, known as the Greenwald limit. Exceeding it causes instabilities and loss of confinement. But an international team, led by Chinese scientists, developed a model and used electron cyclotron resonance heating (microwaves tuned to the rhythm of the electrons) to surpass that limit, reaching average densities between 1.3 and 1.65 times the threshold. The researchers described the result as experimental access to a “density-limit-free regime.” The headline that “China has eliminated the density limit” is premature: it worked under specific start-up conditions. The next step is to verify whether the method works equally well in high-confinement, high-performance plasmas.
HL-3: HEATING BOTH HALVES AND THE GEOPOLITICAL BOARD
China has another large tokamak, HL-3, located in Chengdu. During 2025, it achieved both electrons and ions (the heavy nuclei) exceeding 100 million degrees Celsius. This is crucial because electrons heat up easily, but fusion reactions depend on the ions. Achieving dual temperature brings the experiment closer to complete energy conditions.
This is where we must look at the map. HL-3 is the only official satellite device of ITER in China. In 2025, the International Atomic Energy Agency inaugurated there its first international collaborative center dedicated specifically to fusion research and training. This dismantles the image of a secretive, isolated project: China participates in ITER, manufactures essential components for that international reactor, and maintains cooperation with institutions from dozens of countries. At the same time, it uses that knowledge to accelerate its own technological trajectory. While Europe bets on the international giant ITER (which accumulates delays and cost overruns, with a budget exceeding 20 billion euros) and the United States drives private initiatives such as SPARC, China has opted for a dual path: collaborating in the global consortium while developing in parallel a constellation of domestic devices. Unofficial estimates place the accumulated investment in China’s fusion program at several billion dollars, a figure that, while lower than ITER’s budgetary monster, grows at a steady pace.
HH70: WHEN PRIVATE CAPITAL ENTERS THE SCENE
In Shanghai, the Honghuang 70 (HH70), built by the private company Energy Singularity, is in operation. It is small, but its magnetic system uses exclusively high-temperature superconductors, allowing for more intense fields in less space. During 2026, it conducted 5,755 experiments, maintained a plasma current for 1,337 seconds, and also confirmed it had reached temperatures of 100 million degrees, validating that the private, compact route can also heat plasma under extreme conditions.
Here another misleading comparison arose: some headlines claimed HH70 had surpassed EAST’s record. This is not a valid comparison. HH70’s 1,337 seconds measure plasma current duration; EAST’s 1,066 seconds measure operation in a high-confinement regime. It is like comparing the time a car keeps its engine running with the time another travels at a controlled speed. The importance of HH70 is not that it “defeated” EAST, but that it demonstrates that China’s fusion ecosystem no longer depends exclusively on the state: a private industry capable of attracting capital and developing compact designs is flourishing.
BEST: THE POINT WHERE HISTORY CAN CHANGE (AND ITS INVISIBLE CHALLENGES)
EAST learns to maintain plasma; HL-3 explores temperature conditions; HH70 tests the private route. BEST (Burning Plasma Experimental Superconducting Tokamak) will attempt the definitive leap. Under construction in Hefei, its completion is scheduled for around 2027, and its declared objective is to work with a deuterium-tritium mixture, achieve a burning plasma, reach net fusion energy gain, and demonstrate electricity generation by 2030.
A burning plasma is one in which the heat from the reactions themselves (alpha particles) maintains the temperature, like a campfire that no longer needs matches. Then comes the Q gain: if we put in 10 units of energy and get 100, the gain is 10. But we must distinguish between the energy that goes into the plasma and all the electricity consumed by the plant (magnets, cryogenics, pumps). A machine can achieve net gain in the plasma and still consume more energy as a complete installation than it delivers.
Moreover, BEST faces two technical ghosts that rarely appear in headlines. The first is tritium. Deuterium is abundant in water, but tritium is a radioactive isotope of hydrogen, scarce in nature, and must be produced artificially. Reactor designs envisage surrounding the reactor with a blanket containing lithium; the neutrons released by fusion would react with it to generate new tritium. That self-sufficiency in tritium is a cycle that has never been demonstrated on an industrial scale. The second is neutrons. These particles cross the magnetic field and hit the reactor walls, weakening materials and making them radioactive. Fusion does not generate the same kind or quantity of long-lived waste as a fission plant, and it cannot experience an uncontrolled chain reaction like a conventional reactor. But saying it produces no radioactive waste or is completely risk-free is also false. There will be activated materials and components to be replaced by robots.
THE GEOPOLITICS OF THE “FUSION CITY”
China is not building just one reactor; it is building an industrial ecosystem. Authorities have announced the creation of a “Fusion City” in Changfeng County, designed to integrate scientific facilities, companies, industrial centers, and residential areas. That decision reveals the scale of the bet: attempting to dominate an industry before a consolidated global market even exists, and doing so with a coordination between state, academia, and private sector that few countries can match.
But caution remains indispensable. The history of this technology is full of missed deadlines, rising costs, and announcements that commercial energy was always thirty years away. ITER, the largest international fusion project, has accumulated delays that show the gap between manufacturing extraordinary components and assembling a machine capable of operating as designed. However, caution should not turn into automatic denial. It would be just as wrong to announce that China has already solved humanity’s energy problem as to ignore that it is advancing from experimentation toward the construction of industrial-scale systems, and doing so with a speed and coordination that deserve attention.
INFINITE ENERGY? NO
No. Fusion could use extremely abundant fuels, and the International Atomic Energy Agency estimates that the deuterium contained in one liter of water could, in theory, generate as much energy as the combustion of about 300 liters of oil. But it will require lithium, superconductors, turbines, and materials capable of withstanding extreme conditions. All infrastructure has physical, economic, and environmental costs. The most rigorous expression is “potentially abundant, high-density, low-carbon energy.” Nor would it solve the climate crisis on its own. If commercial plants take decades to deploy, humanity cannot wait for fusion to reduce its consumption of coal, oil, and gas now. Fusion could transform the energy matrix of the second half of the century, but it does not replace the urgency of the present energy transition.
SO, WHERE IN THE STORY IS CHINA?
China has already passed the stage of demonstrating that it can produce plasma and reach extreme temperatures. It has also demonstrated that it can sustain high-confinement plasmas for unprecedented times, overcome certain density limits, heat electrons and ions simultaneously beyond 100 million degrees, and manufacture large superconducting components with domestic technology.
It is now entering the most difficult phase: turning plasma science into energy engineering. The 582-ton magnet represents precisely that transition. It is not a power plant, but neither is it a small laboratory experiment. It is an industrial component built for machines that will have to confine burning plasmas and operate under much greater demands.
BEST will be the decisive test of this stage. If it meets its schedule, China will attempt toward the end of the decade to demonstrate net fusion gain and some form of electricity generation. Afterward will come the CFEDR, the future China Fusion Engineering Demonstration Reactor, conceived as a bridge toward a larger-scale demonstration plant and, eventually, toward commercial plants.
China has not reached the end of the artificial sun story. But neither is it in its prologue. It is in the chapter where scientific records must be turned into complete machines; where controlling plasma is no longer enough and it is necessary to extract energy, produce fuel, withstand neutrons, maintain components, and generate electricity. It is the chapter where many fusion promises have stalled for decades. The difference is that China has placed behind that challenge something more than a laboratory: a long-term national policy, a superconducting industry, heavy machinery manufacturers, universities, private companies, public investment, international cooperation, and a city — Hefei — beginning to organize its economic development around fusion.
There is not yet a Sun trapped in Hefei. There is something less poetic and perhaps more important: thousands of scientists, engineers, and workers trying to build the magnetic fields, materials, and industrial chains that one day might keep a small fusion reaction alight without the help of a star’s gravity.
The first lamp has not yet been switched on. But China is already manufacturing the switch. And Zhang Linghe continues to shine brighter than ever before on screen. The other plasma, for now, still does not reach him in his brilliant artistic career. But who knows. Science, unlike entertainment, does not need a script. It needs time, money, and a superpowerful magnetic track. The project is truly astonishing. The beauty of science in fusion and function for peace, as it always should have been.
REFERENCES
Direct sources
Chinese Academy of Sciences. EAST record of 1,066 seconds in high confinement: Chinese “Artificial Sun” Sets New Record
Xinhua. Development and testing of superconducting magnetic systems, June 28, 2026: China’s “artificial sun” project achieves new engineering milestone
Science Advances. Scientific article on the density-limit-free regime in EAST: Accessing the density-free regime with ECRH-assisted ohmic start-up on EAST
Xinhua. Explanation of the experiment that surpassed the density limit: China’s “artificial sun” experiment finds way to break fusion plasma density limit
Xinhua. BEST objectives and electricity demonstration by 2030: China accelerates nuclear fusion engineering, targeting power generation demonstration by 2030
Xinhua. Construction of the BEST tokamak in Hefei: Burning Plasma Experimental Superconducting Tokamak under construction
Government of China. HL-3, EAST, ITER and international cooperation: China contributes innovative power to global fusion development
Official Chinese information on the private HH70 tokamak: Shanghai’s “artificial sun” achieves landmark milestone
Energy Singularity. Statement on HH70 results, including 100 million degrees: official company announcements, June 2026
Investment estimates: market analysis and statements from Chinese officials on the fusion budget (unofficial figures but cited in specialized media)
Technical and reference sources
International Atomic Energy Agency. General explanation of nuclear fusion: What is Nuclear Fusion?
International Atomic Energy Agency. Burning plasma and alpha particle heating: Burning plasma
International Atomic Energy Agency. Fuels, waste, tritium and material activation: Fusion: Frequently Asked Questions
International Atomic Energy Agency. Deuterium, tritium, magnetic confinement and electricity production: Nuclear Fusion Basics
