Artificial Sun Breaks Fusion Record: The Challenge of Sustained Fusion Plasma
Updated: Aug 21
Dr. Marcus L. Reinhardt¹, Dr. Yuna S. Chen², Prof. Rafael A. Domínguez³
¹ Department of Plasma Energy Systems, Helios Institute of Technology
² Center for Magnetic Confinement Science, East Pacific Research University
³ Laboratory for Fusion Engineering, Institute of Advanced Energy
[Disclaimer: This is a sample academic article. All author names, affiliations, and institutional details are fictional and have been created solely for illustrative and educational purposes.]
Abstract
A major milestone in long-pulse magnetic-confinement fusion research was achieved by the Experimental Advanced Superconducting Tokamak (EAST) in China when steady-state high-confinement plasma operation was sustained for 1,066 seconds on 20 January 2025, extending substantially beyond the previous 403-second EAST record. The achievement was produced through improvements in superconducting magnet operation, auxiliary heating, current drive, plasma control, impurity management, tungsten-divertor performance, and particle and heat exhaust, all of which are required if tokamak plasmas are ultimately to be maintained for reactor-relevant timescales [10.1007/s42524-025-5502-3; 10.1063/5.0146690]. Nuclear fusion is produced when light nuclei are combined into heavier nuclei and a fraction of their mass is released as energy, but terrestrial magnetic-confinement fusion requires extremely hot ionized plasmas to be created and maintained while simultaneously preventing excessive heat loads, instabilities, impurity accumulation, and loss of confinement. Long-duration high-confinement operation is therefore regarded as a critical step toward future burning-plasma and steady-state fusion systems, although duration alone should not be confused with fusion ignition, net electrical energy production, or commercial reactor operation [10.1088/0029-5515/47/6/S01]. EAST has been developed specifically as a fully superconducting tokamak platform through which steady-state plasma physics and technologies relevant to ITER and future Chinese fusion reactors can be investigated, and long-pulse high-performance H-mode operation has progressively been extended through advances in radio-frequency heating and current drive, plasma-facing components, instability control, and exhaust physics [10.1063/5.0146690]. Particular attention is given in this article to the physical meaning of the 1,066-second record, the role of H-mode confinement, non-inductive current drive, tungsten plasma-facing components, plasma stability, tritium-fuel considerations, and the relationship between EAST and ITER. The remaining barriers to practical fusion power—including sustained fusion gain, neutron-resistant materials, tritium breeding, heat extraction, divertor lifetime, reactor availability, and conversion of fusion energy into reliable electricity—are also examined. The EAST result is therefore interpreted not as the arrival of commercial “limitless energy,” but as an important demonstration that increasingly long, controlled, high-confinement tokamak plasmas can be engineered, bringing one of the central requirements of steady-state fusion reactors closer to experimental validation.
1. Introduction
Controlled nuclear fusion is pursued as a potential large-scale energy source because the fusion of light nuclei can release substantial energy without direct carbon dioxide emissions during reactor operation, but practical power production requires an ionized plasma to be heated, confined, and controlled under conditions where fusion reactions occur rapidly enough to compensate for energy losses. For the deuterium–tritium reaction, reactor concepts are generally designed around plasma temperatures of order (10^8) K, sufficiently high density, and sufficiently long energy-confinement time, parameters historically summarized through the Lawson criterion and its modern triple-product formulations [10.1088/0370-1298/70/1/303]. Tokamaks address the confinement problem by using strong toroidal and poloidal magnetic fields to restrict charged-particle motion inside a toroidal vacuum vessel, with improved confinement regimes such as H-mode having become central to the operating scenarios envisioned for ITER and future reactors [10.1103/PhysRevLett.49.1408; 10.1088/0029-5515/47/6/S01]. EAST, the Experimental Advanced Superconducting Tokamak operated by the Institute of Plasma Physics of the Chinese Academy of Sciences, was constructed as a fully superconducting facility in which long-duration plasma confinement, non-inductive current drive, plasma-facing components, heat exhaust, and steady-state control can be investigated under conditions increasingly relevant to next-step fusion devices [10.1088/1741-4326/ad4270]. A 1,056-second improved-confinement Super I-mode discharge was reported in 2023, a reproducible 403-second H-mode discharge was subsequently documented, and on 20 January 2025 a high-confinement plasma was sustained for 1,066 seconds, representing a further step in the progression toward reactor-relevant long-pulse operation [10.1126/sciadv.abq5273; 10.1088/1741-4326/ad4270; 10.1007/s42524-025-5502-3]. The scientific significance of this achievement lies not in demonstrating net fusion-energy production, but in showing that a complex superconducting tokamak and its heating, current-drive, plasma-control, exhaust, cooling, and diagnostic systems can be operated coherently over approximately eighteen minutes while a high-confinement plasma state is maintained [10.1007/s42524-025-5502-3].
2. Results and Discussion
The EAST record should be interpreted as an integration milestone in magnetic-confinement fusion rather than as an isolated duration record, because a useful steady-state reactor must simultaneously sustain plasma current, preserve adequate energy confinement, avoid damaging magnetohydrodynamic instabilities, control impurities, exhaust helium and fuel particles, protect plasma-facing components, and continuously remove heat [10.1126/sciadv.abq5273; 10.1088/1741-4326/ad4270]. Increasing pulse duration from seconds to hundreds and ultimately thousands of seconds forces physical processes with long characteristic times—such as current-profile evolution, wall saturation, impurity accumulation, thermal equilibration, and plasma–surface interactions—to become experimentally visible, making long-pulse devices particularly important for evaluating whether apparently stable short-duration regimes remain stable on reactor-relevant timescales [10.1126/sciadv.abq5273].
2.1. Fusion Conditions: Temperature Alone Is Not Enough
Extremely high temperature is required because positively charged nuclei must approach closely enough for quantum tunnelling through the Coulomb barrier to permit nuclear fusion, but temperature by itself cannot determine whether a plasma produces useful fusion power. Fusion performance depends jointly on ion temperature, particle density, and energy-confinement time, which is why the Lawson criterion was formulated to describe the minimum combination of density and confinement required for a self-sustaining thermonuclear system [10.1088/0370-1298/70/1/303]. In a deuterium–tritium plasma, the reaction (D+T\rightarrow{}^4He+n) releases approximately 17.6 MeV, with most of the energy carried by a 14.1-MeV neutron and the remainder deposited locally by the alpha particle, so a burning plasma is obtained only when alpha-particle self-heating becomes a major contributor to maintaining the plasma temperature [10.1088/0029-5515/47/6/S01]. Consequently, statements that a tokamak has reached temperatures above 100 million degrees must be separated from claims of ignition or net-energy gain, because high temperature can be sustained experimentally by external heating even when fusion power remains far below the heating power supplied to the plasma [10.1088/0029-5515/47/6/S01].
2.2. Why Tokamaks Use Magnetic Confinement
A material container cannot directly confine a plasma at fusion temperatures because contact with a solid wall would cool the plasma and damage the surface, so charged particles are instead confined primarily by magnetic fields. In a tokamak, a strong toroidal magnetic field is combined with the poloidal field associated with the plasma current to produce helically wound magnetic field lines that constrain charged-particle trajectories inside nested magnetic surfaces [10.1088/0029-5515/47/6/S01]. The resulting configuration nevertheless remains susceptible to turbulence, neoclassical transport, magnetohydrodynamic instabilities, energetic-particle effects, and edge phenomena that can degrade confinement or terminate the discharge [10.1088/0029-5515/47/6/S01]. The tokamak problem is therefore not simply one of generating a sufficiently large magnetic field, but of maintaining an optimized magnetic equilibrium while plasma pressure, current profile, turbulence, heating, fueling, and boundary conditions are continuously controlled.
2.3. H-Mode and the Importance of High Confinement
A major improvement in tokamak performance was produced by the discovery of the high-confinement mode, or H-mode, in which a transport barrier forms near the plasma edge and reduces turbulent energy and particle losses compared with low-confinement operation [10.1103/PhysRevLett.49.1408]. H-mode subsequently became the reference operating regime for ITER because improved energy confinement permits reactor-relevant temperatures and pressures to be reached with lower external heating requirements than would otherwise be necessary [10.1088/0029-5515/47/6/S01]. However, conventional H-mode frequently produces edge-localized modes, or ELMs, in which bursts of energy and particles are expelled toward plasma-facing components, potentially generating unacceptable transient loads in larger fusion devices [10.1007/s41614-023-00119-2]. EAST research has therefore included the development of small-ELM, no-ELM, radiative-divertor, and other high-confinement regimes in which adequate core performance can be combined with more manageable conditions at the plasma boundary [10.1007/s41614-023-00119-2].
2.4. From 1,056 Seconds to 403 Seconds and 1,066 Seconds
Different EAST duration records have represented different plasma regimes and should not be conflated. In 2021, a 1,056-second fully non-inductive discharge was produced in a newly identified Super I-mode regime combining an I-mode edge with an electron internal transport barrier, with density, impurity accumulation, and divertor heat flux remaining controlled throughout the discharge [10.1126/sciadv.abq5273]. A reproducible 403-second long-pulse steady-state H-mode plasma was later obtained using predominantly radio-frequency heating and current drive, demonstrating sustained H-mode confinement together with impurity, recycling, MHD, and heat-flux control [10.1088/1741-4326/ad4270]. The 2025 experiment extended high-confinement operation to 1,066 seconds, surpassing the previous EAST H-mode duration record and demonstrating a thousand-second-class high-confinement scenario in the superconducting device [10.1007/s42524-025-5502-3]. The progression is important because the scientific target has not merely been longer plasma duration but the integration of duration with progressively better confinement and more reactor-relevant operating conditions.
2.5. Why 1,066 Seconds Matters Physically
A tokamak pulse lasting approximately eighteen minutes is long compared with several important characteristic times within the plasma and surrounding engineering systems, allowing quasi-steady interactions among current drive, particle recycling, impurity transport, wall conditions, divertor exhaust, superconducting magnets, cryogenics, diagnostics, and active feedback to be tested simultaneously [10.1126/sciadv.abq5273]. The earlier 1,056-second EAST experiment demonstrated that approximately 2 GJ of externally injected energy could be handled while density and peak divertor heat load were controlled and strong core impurity accumulation was avoided [10.1126/sciadv.abq5273]. Long-duration experiments are therefore valuable because problems that remain hidden during a ten-second discharge may accumulate over hundreds of seconds and eventually terminate a reactor-scale plasma. The 1,066-second high-confinement achievement should consequently be viewed as evidence of improved integrated steady-state capability rather than as evidence that a fusion power plant has already been demonstrated [10.1007/s42524-025-5502-3].
2.6. Non-Inductive Current Drive and the Steady-State Problem
Conventional transformer-driven tokamak operation is intrinsically pulsed because the central solenoid induces plasma current through changing magnetic flux, and indefinitely increasing this flux is impossible. A genuinely steady-state tokamak must therefore obtain much of its plasma current from non-inductive mechanisms such as lower-hybrid waves, neutral-beam current drive, electron-cyclotron current drive, ion-cyclotron systems, and the self-generated bootstrap current associated with pressure gradients [10.1088/0029-5515/47/6/S01]. EAST has placed particular emphasis on lower-hybrid current drive and radio-frequency heating, and fully non-inductive high-performance plasmas have been demonstrated through optimization of externally driven current and bootstrap-current contributions [10.1088/1741-4326/ab443a]. In the 1,056-second Super I-mode discharge, the plasma current was sustained non-inductively using lower-hybrid current drive together with a substantial bootstrap contribution, demonstrating precisely the kind of continuous-current physics required by future steady-state reactor concepts [10.1126/sciadv.abq5273].
2.7. Plasma Stability and Edge-Localized Modes
Long-duration high-performance plasmas can be terminated by instabilities even when sufficient heating and current drive are available, making stability control one of the central requirements of steady-state fusion. Edge-localized modes are particularly important because repeated impulsive energy deposition onto divertor surfaces could substantially shorten component lifetime in future high-power devices [10.1007/s41614-023-00119-2]. EAST experiments have demonstrated naturally occurring and impurity-induced small- or no-ELM regimes compatible with high density, significant bootstrap current, tungsten impurity exhaust, radiative-divertor operation, and fully non-inductive conditions [10.1007/s41614-023-00119-2]. Resonant magnetic perturbations and impurity seeding have additionally been investigated for ELM suppression and detachment control in ITER-relevant scenarios [10.1088/1741-4326/ad4270]. A commercially useful fusion plasma will therefore require not only strong confinement but a form of confinement whose edge behavior remains compatible with the thermal and mechanical limits of surrounding materials.
2.8. The Divertor: Where Fusion Heat Becomes an Engineering Problem
The divertor is one of the most demanding components of a magnetic-fusion device because heat and particles escaping from the confined plasma are intentionally directed toward a relatively small surface area, where extremely high steady and transient heat fluxes can be produced. Future reactor divertors must simultaneously remove thermal power, withstand erosion, survive neutron irradiation, exhaust helium ash and impurities, and retain structural integrity during prolonged operation [10.1016/j.fusengdes.2022.113010]. EAST has progressively adopted actively cooled metallic plasma-facing components and tungsten divertor technology so that long-pulse operation can be investigated under conditions more closely related to ITER and future reactors [10.1088/1741-4326/ad4270]. The 1,056-second experiments showed that peak divertor heat flux could be actively controlled for more than a thousand seconds, but examination of tungsten/copper components has also revealed the severe materials challenges associated with repeated high-heat-flux operation [10.1126/sciadv.abq5273; 10.1088/1741-4326/adf3c7]. Thus, successful plasma confinement does not remove heat; it determines where and how that heat must eventually be engineered out of the system.
2.9. EAST and ITER Address Different Fusion Questions
EAST and ITER are complementary rather than equivalent machines. EAST is optimized as a superconducting experimental platform for long-pulse and steady-state plasma physics, whereas ITER is designed to study deuterium–tritium burning plasmas in which substantial heating is supplied by fusion-produced alpha particles [10.1088/0029-5515/47/6/S01]. ITER's central fusion-performance objective has historically been expressed as a plasma fusion gain (Q=P_\mathrm{fusion}/P_\mathrm{external}) of approximately 10 in its high-gain scenario, meaning that fusion power produced in the plasma would substantially exceed the external auxiliary heating delivered directly to that plasma [10.1088/0029-5515/47/6/S01]. EAST's 1,066-second record does not represent (Q>1), ignition, or net electrical output; instead, it provides experimental information on technologies and operating regimes required if future burning plasmas are eventually to be sustained for comparable durations [10.1007/s42524-025-5502-3]. Long-duration EAST operation is consequently relevant to ITER and successor reactors because high fusion gain is of limited practical value unless plasma performance can also be sustained and controlled reliably.
2.10. Net Fusion Energy Is More Demanding Than Plasma Gain
Even when plasma fusion gain exceeds unity, a power plant must satisfy a more demanding engineering energy balance because electricity is required for magnets, cryogenics, pumps, heating and current-drive systems, tritium processing, cooling systems, diagnostics, and other plant equipment. Thermal energy carried largely by fusion neutrons must first be absorbed in a blanket, converted into heat, transferred through a coolant system, and subsequently converted into electricity with finite thermodynamic efficiency. Thus, (Q>1) at the plasma level is not equivalent to net electrical power from the complete facility, and commercial viability requires sufficiently large fusion gain, high plant availability, durable components, and efficient heat conversion. ITER is intended principally as a burning-plasma experiment rather than as an electricity-generating power station, so additional DEMO-class systems will be required before integrated net-electric fusion technology can be assessed under power-plant conditions [10.1016/j.fusengdes.2019.01.141].
2.11. Tritium Breeding and the Closed Fuel-Cycle Challenge
Deuterium is abundant in natural water, but tritium is radioactive, occurs naturally only in very small quantities, and cannot support a large global fusion industry from existing inventories alone. A deuterium–tritium power plant must therefore use energetic fusion neutrons to generate replacement tritium through reactions with lithium inside a breeding blanket surrounding the plasma [10.1016/j.fusengdes.2019.01.141]. Such a blanket must simultaneously breed and extract tritium, absorb neutron energy, transfer heat to a power-conversion system, shield external components, withstand irradiation, and remain maintainable over the reactor lifetime [10.1088/1741-4326/ad00cb]. No full-scale power-reactor breeding blanket has yet been operated, making closed-fuel-cycle demonstration one of the largest engineering steps separating present tokamak experiments from self-sufficient fusion power plants [10.1016/j.fusengdes.2019.01.141]. Tritium self-sufficiency must therefore be demonstrated experimentally rather than assumed from the underlying nuclear reaction alone.
2.12. Neutron Damage and the Materials Frontier
In deuterium–tritium fusion, approximately four-fifths of the released energy is carried by high-energy neutrons that are unaffected by magnetic confinement and penetrate the surrounding first wall and blanket. These neutrons displace atoms, produce transmutation products, generate helium and hydrogen within structural materials, and progressively alter mechanical and thermal properties, making radiation resistance a central constraint on reactor lifetime [10.1088/0029-5515/51/11/113006]. Tungsten is widely considered for high-heat-flux plasma-facing regions because of its high melting point and favorable sputtering characteristics, while reduced-activation ferritic–martensitic steels and other advanced alloys are being developed for structural components [10.1088/0029-5515/51/11/113006]. A practical reactor must therefore demonstrate not merely that a plasma can be sustained, but that surrounding components can survive years of combined neutron irradiation, thermal cycling, mechanical stress, and plasma exposure while remaining economically replaceable.
2.13. Can Fusion Become a Practical Energy Technology?
The transition from experimental fusion physics to electricity production will depend on successful integration of several requirements that have often been demonstrated separately but not yet simultaneously in one power-producing facility: high fusion gain, long-duration or steady-state plasma operation, non-inductive current sustainment, stable confinement, controlled exhaust, neutron-resistant materials, tritium self-sufficiency, efficient heat extraction, remote maintenance, and sufficiently high plant availability [10.1016/j.fusengdes.2019.01.141; 10.1016/j.fusengdes.2022.113010]. EAST directly addresses only part of this chain, but long-duration high-confinement experiments are essential because a future reactor cannot function economically if high-performance plasmas can be maintained only for brief intervals. Fusion should therefore not be described as already providing “limitless energy”; rather, an exceptionally large fuel resource is potentially available if a highly demanding combination of plasma physics, nuclear technology, materials science, thermal engineering, and reactor economics can be made to operate reliably together.
3. Conclusion and Outlook
The 1,066-second EAST experiment represents an important advance in sustained magnetic-confinement fusion because high-confinement plasma operation was extended to a thousand-second timescale in a fully superconducting tokamak, building upon earlier 1,056-second Super I-mode and 403-second H-mode achievements [10.1126/sciadv.abq5273; 10.1088/1741-4326/ad4270; 10.1007/s42524-025-5502-3]. Its primary significance lies in the integration of plasma confinement with long-duration heating, current drive, cryogenic operation, superconducting magnets, plasma control, impurity management, particle exhaust, and thermal management rather than in net fusion-energy generation itself [10.1007/s42524-025-5502-3]. EAST also provides a platform on which high-confinement regimes, non-inductive current drive, ELM mitigation, tungsten plasma-facing components, and reactor-relevant exhaust strategies can be tested over timescales approaching those required by future steady-state devices [10.1007/s41614-023-00119-2; 10.1088/1741-4326/ad4270]. However, the transition to commercial fusion will additionally require high-gain deuterium–tritium burning plasmas, viable tritium breeding, neutron-resistant materials, durable divertors, closed fuel cycles, efficient thermal-to-electric conversion, maintainability, and reliable plant availability [10.1016/j.fusengdes.2019.01.141; 10.1016/j.fusengdes.2022.113010]. The most useful interpretation of the EAST record is therefore not that the fusion-energy problem has been solved, but that one essential reactor requirement—stable high-confinement operation for genuinely long timescales — is being progressively transformed from a theoretical requirement into an experimentally controllable engineering capability.
References
Lawson, J. D. (1957). Some criteria for a power producing thermonuclear reactor. Proceedings of the Physical Society. Section B, 70, 6–10. [10.1088/0370-1298/70/1/303]
Wagner, F., Becker, G., Behringer, K., et al. (1982). Regime of improved confinement and high beta in neutral-beam-heated divertor discharges of the ASDEX tokamak. Physical Review Letters, 49, 1408–1412. [10.1103/PhysRevLett.49.1408]
ITER Physics Expert Group Chairs and Co-Chairs, ITER Joint Central Team and Physics Integration Unit. (2007). Progress in the ITER Physics Basis. Nuclear Fusion, 47, S1–S414. [10.1088/0029-5515/47/6/S01]
Huang, J., Qian, J. P., Garofalo, A. M., et al. (2020). Improved high-performance fully non-inductive discharge by optimizing the fast-ion confinement on EAST. Nuclear Fusion, 60, 016002. [10.1088/1741-4326/ab443a]
Song, Y., Zou, X., Gong, X., et al. (2023). Realization of thousand-second improved confinement plasma with Super I-mode in Tokamak EAST. Science Advances, 9, eabq5273. [10.1126/sciadv.abq5273]
Xu, G. S., Wang, Y. F., Yang, Q. Q., et al. (2023). Recent advances in developing natural and impurity-induced small/no-ELM H-mode regimes in EAST. Reviews of Modern Plasma Physics, 7, 14. [10.1007/s41614-023-00119-2]
Gong, X., EAST Team and Collaborators. (2024). Overview of recent experimental results on the EAST Tokamak. Nuclear Fusion, 64, 112013. [10.1088/1741-4326/ad4270]
Yan, J., Kong, D., & EAST Team. (2025). China’s “Artificial Sun”: Experimental Advanced Superconducting Tokamak (EAST). Frontiers of Engineering Management, 12, 442–445. [10.1007/s42524-025-5502-3]
Federici, G., Boccaccini, L. V., Cismondi, F., et al. (2019). An overview of the EU breeding blanket design strategy as an integral part of the DEMO design effort. Fusion Engineering and Design, 141, 30–42. [10.1016/j.fusengdes.2019.01.141]
Federici, G., Bachmann, C., Barucca, L., et al. (2022). Divertor of the European DEMO: Engineering and technologies for power exhaust. Fusion Engineering and Design, 175, 113010. [10.1016/j.fusengdes.2022.113010]
Stork, D., Agostini, P., Boutard, J.-L., et al. (2011). From materials development to their test in IFMIF: An overview. Nuclear Fusion, 51, 113006. [10.1088/0029-5515/51/11/113006]
Cismondi, F., Spagnuolo, G. A., Boccaccini, L. V., et al. (2020). Progress of the conceptual design of the European DEMO breeding blanket, tritium extraction and coolant purification systems. Fusion Engineering and Design, 157, 111640. [10.1016/j.fusengdes.2020.111640]
Fischer, U., et al. (2023). Testing needs for the development and qualification of a breeding blanket for DEMO. Nuclear Fusion, 63. [10.1088/1741-4326/ad00cb]
Comments