China's EAST Artificial Sun: 1.65x Density and Limits

China’s EAST reached electron densities 1.3~1.65 times the Greenwald limit. This expands the possibilities for high-density operation. It does not, however, demonstrate commercial power generation or the end of U.S. technological dominance.

The experimental figures are from a Science Advances paper published in January 2026.

What is China’s artificial sun, EAST?

EAST is a superconducting tokamak research device that uses magnetic fields to confine plasma. A tokamak holds plasma in a doughnut-shaped space. “Artificial sun” is an informal name for a nuclear fusion research device.

Nuclear fusion is a reaction in which light atomic nuclei combine and release energy. Plasma is matter in which electrons and ions are separated. To use fusion for power generation, energy must be confined effectively at sufficiently high temperatures and densities.

The measurement in the EAST study is line-averaged electron density. It should not be read as saying that power output increased by the same multiple. The scope of the experiment is described in the EAST paper abstract.

Is the Greenwald limit an absolute law of nature?

The Greenwald limit is an empirical density benchmark derived from tokamak experiments. Its reference density is calculated using plasma current and cross-sectional size. It is not a law setting the same maximum particle count for every device.

At higher densities, plasma becomes harder to sustain under certain conditions. Cooling at the edge and radiation from impurities can be particularly important. The conditions near the wall must therefore be controlled along with density.

The Greenwald benchmark had been exceeded before EAST. In 2024, the U.S. DIII-D reported achieving high density and strong confinement performance at the same time. It would therefore be inaccurate to describe this result as humanity’s first time exceeding the limit. U.S. Department of Energy research explanation

What conditions did EAST change?

The researchers adjusted both the gas conditions at startup and the electron heating. The main heating method was electron cyclotron resonance heating. It uses microwaves to transfer energy to electrons in the plasma.

The Chinese Academy of Sciences describes a reduction in the physical sputtering of tungsten impurities. Sputtering occurs when particles strike a surface and knock atoms loose. Reducing it can lower radiation losses caused by impurities. Chinese Academy of Sciences research announcement

The researchers interpreted the results using the theory of plasma-wall self-organization, abbreviated PWSO. The term “self-organization” does not itself mean an automatic coating technology.

It is reasonable to describe this as an improvement in how an existing device was operated. But calling it a software-only achievement overlooks the experimental conditions. The heating equipment, gas supply, and wall conditions all played a part. Researchers’ publicly available manuscript

Example calculation for 1.65 times the density

A density 1.65 times the Greenwald benchmark is 65% higher than the benchmark. A density 1.3 times the benchmark is 30% higher. Both figures are density ratios from the study published in January 2026.

Calling 1.65 times a 165% increase would exaggerate the increase. Nor does this calculation compare the result with EAST’s previous record. The denominator is the Greenwald reference density under the conditions in question.

Comparing the density record with power generation

A density record must be assessed separately from progress toward commercial power generation. Temperature and fuel composition also affect fusion output. A power plant must account for power conversion losses and the electricity consumed by its equipment.

ITER defines Q as the ratio of fusion output to external heating power. That ratio does not fully account for the electricity consumed by the entire power plant. High Q must therefore also be distinguished from net power production. ITER’s explanation of energy gain

The EAST research announcement identified further verification under high-performance conditions as the next task. At the time of the January 2026 announcement, the researchers planned to apply the approach to high-confinement operation. The paper cannot be used to calculate generation costs or a commercialization date. Chinese Academy of Sciences press release

Common misconceptions about fusion fuel and waste

It is inaccurate to describe fusion as waste-free power generation that can run indefinitely on seawater alone. A leading approach to power generation uses deuterium and tritium as fuel. The two fuels have different supply requirements.

• Deuterium is a hydrogen isotope that can be obtained from water.
• Tritium is a radioactive isotope that occurs rarely in nature.
• Researchers are studying how to produce tritium using lithium to make the fuel supply self-sufficient.
• Structural materials exposed to neutrons can become radioactive.

The U.S. Department of Energy identifies tritium production and fuel self-sufficiency as research challenges. Systems to recover and reuse fuel are also needed. The potential abundance of raw materials and a power plant’s ability to supply its own fuel are different issues. U.S. Department of Energy explanation of fusion fuel

Fusion also requires management of activated structural materials. Activation occurs when exposure to particles makes a material radioactive. The characteristics of the waste depend on the materials used and the design. U.S. Department of Energy overview of fusion

Does China’s achievement mean the end of U.S. technological dominance?

EAST’s result alone cannot establish that U.S. technological dominance has ended. There is evidence that China’s capacity to produce research papers has grown. But publication measures do not directly measure the competitiveness of fusion power plants.

In a 2025 announcement, Springer Nature described China’s growing research contribution. The data covered 2024. Nature Index’s Share is a measure of contributions to papers allocated according to authors’ affiliations. Springer Nature’s 2025 announcement

Comparisons between countries must use the same field and period. Nature Index also updates tables for past years when its data changes. Ratios without a source and version are therefore difficult to reproduce. Nature Index tables and methodology

Individual cases of insufficient travel funding also make it difficult to assess the research capacity of the U.S. as a whole. Such claims require evidence about the institution, event, and budget. Conversely, there is no reason to downplay EAST’s achievement.

Opportunities for Korean companies in the fusion supply chain

Opportunities for Korean companies should be assessed through their actual ability to deliver and their contracts. There is a verifiable precedent for Korean participation in the fusion supply chain. In 2013, ITER announced that Korea had completed production of niobium-tin superconducting strand. ITER’s record of Korean procurement

Advances in operational control do not eliminate the need for magnets and structures. ITER’s magnets use niobium-tin and niobium-titanium. It would therefore be inaccurate to treat the entire superconducting supply chain as identical to the rare earth supply chain. ITER magnet components

These distinctions are an analysis based on the functions a power plant needs. They are not a forecast of increased revenue for any particular company. When assessing a company, check its disclosures for contract scope and delivery terms.

Comparing the industrialization of solar power and fusion

There is insufficient basis to apply solar power’s path to market expansion directly to fusion. Fusion still faces equipment challenges beyond plasma operation. The U.S. Department of Energy treats the fuel cycle as a separate technical challenge. Explanation of the fusion fuel cycle

The possibility that mass production will lower costs and the viability of commercial power generation are separate claims. The EAST paper is not a cost study estimating savings from mass production. It cannot be used as evidence that China will dominate the market.

What to check when comparing fusion records

To compare records, check the experimental conditions as well as the numbers. A density ratio alone makes it difficult to rank devices. Reading results in the following order reduces the chance of confusing different achievements.

1. Check whether the measurement is electron density or fuel density.
2. Check whether it is an average or a value at the center.
3. Distinguish startup from a period of sustained high performance.
4. Check how long the conditions were maintained and the confinement performance.
5. Distinguish a density record from energy gain and net power.

These points are a checklist for comparing research results. Each value can be checked in the paper’s methods and results. Record both the research institution’s announcement date and the period covered by the data.