Ninety frozen hydrogen pellets, injected into a 30-million-degree plasma: the pellet injection system from Oak Ridge National Laboratory proved decisive. On May 22, 2025, it enabled a 43-second plasma discharge in which Greifswald's Wendelstein 7-X stellarator surpassed tokamak reactors for the first time in power-plant-relevant long-duration operation. The formal analysis, now published in the journal Nuclear Fusion, shows how close the reactor is to meeting power-plant requirements.
What Is This: Stellarator Versus Tokamak
Fusion requires plasma, superheated ionized gas at temperatures exceeding 30 million degrees, confined magnetically long enough that fusion reactions release more energy than is supplied. Since the 1950s, two fundamentally different designs have competed to achieve this.
The tokamak, whose most famous examples are JET in the UK and JT-60SA in Japan, uses a donut-shaped chamber and an induced electrical current in the plasma that generates part of the confining magnetic field. This design achieves high peak values but has a systemic disadvantage: the induced plasma current must be restarted regularly, forcing tokamaks into pulsed operation. A commercial power plant would need to run continuously.
The stellarator, of which Wendelstein 7-X is the world's largest example, forgoes plasma current entirely. The entire magnetic field structure is generated by intricately twisted external coils. This design theoretically enables steadier operation but remained practically weaker for decades in the central metric of fusion research: the so-called triple product.
The triple product combines plasma density, temperature, and energy confinement time into a single metric. The higher the value, the closer to conditions where a fusion power plant would produce more energy than it consumes. Tokamaks dominated the triple product, but only for short discharge times. That remained unchanged for decades. Until May 22, 2025.
The Record: 43 Seconds and 1.8 Gigajoules
On the final day of experimental phase OP2.3, Wendelstein 7-X sustained a triple product over a 43-second plasma discharge that for the first time exceeded all known tokamak values at comparable discharge lengths. The plasma reached 30 million degrees Celsius, with ten gyrotron generators supplying ten megawatts of heating power.
The breakthrough was enabled by a novel pellet injection system from Oak Ridge National Laboratory (ORNL), a US Department of Energy laboratory. The device fires cylindrical pellets of frozen hydrogen with three-millimeter diameters at speeds of 300 to 800 meters per second into the plasma. During the 43-second record discharge, the system injected 90 pellets at short intervals. This solved the most difficult technical problem of stellarators: continuous fuel supply over extended operating periods. Earlier injectors managed at most ten to fifteen pellets per experiment.
In parallel, the experiment increased cumulative energy output to 1.8 gigajoules over 360 seconds of plasma operation, compared to the previous record of 1.3 gigajoules from February 2023. Thomas Klinger, project leader at IPP Greifswald, called it a "tremendous success of the international team," impressively demonstrating Wendelstein 7-X's potential. Raising the triple product to tokamak levels was "an important milestone on the path to a power-plant-capable stellarator."
Startups, Politics, and the Still-Open Hurdle
Greifswald's results came at a politically favorable moment. On February 26, 2026, Bavaria's State Ministry, the Max Planck Institute for Plasma Physics, energy company RWE, and fusion startup Proxima Fusion signed a memorandum of understanding for construction of a demonstration reactor called "Alpha" and a possible pilot plant "Stellaris" in Bavaria. Proxima Fusion plans to complete Alpha by 2031 and feed power into the grid in the 2030s. The concept builds directly on operational experience from Wendelstein 7-X. Germany's coalition government agreement states the world's first commercial fusion reactor should be built in Germany.
Whether the timeline is realistic remains unclear. Wendelstein 7-X has not yet reached a critical power-plant threshold: the ratio of plasma pressure to magnetic pressure was three percent in the record experiment over the entire plasma volume. A commercial power plant requires four to five percent. The gap has shrunk but remains.
This also explains why the international ITER project in southern France, a tokamak with approximately 22 billion euros in total budget and thus the largest single fusion experiment globally, remains a central project of worldwide fusion research. ITER is scheduled to begin plasma operation in the mid-2030s, with successor system DEMO to feed power into the grid in the 2040s. Wendelstein 7-X's breakthrough shifts the balance in fusion research but does not displace the decades-long infrastructure of the tokamak approach.
OP2.4 Starts in August: The Goal is Four Percent
In August 2026, Wendelstein 7-X resumes operation after a one-year maintenance period. The time was used to expand and improve heating systems, magnetic coils, and approximately 50 plasma diagnostic systems. The stated scientific goal of experimental phase OP2.4 is to drive the ratio of plasma pressure to magnetic pressure to the critical four percent, the threshold beyond which conditions become power-plant-relevant.
If achieved, Wendelstein 7-X will have reached a second historic milestone and substantially strengthened the scientific foundation for Project Alpha in Bavaria. The Max Planck Institute is simultaneously planning a successor research reactor an order of magnitude more powerful, whose design optimizations will build directly on Wendelstein 7-X findings. A decision on location and funding remains pending. Wendelstein 7-X has demonstrated that the path to a fusion power plant does not necessarily run through a tokamak. The next generation must still provide proof of power-plant capability.