A steel-and-superconductor magnet taller than a five-story building has finally come together in one place. In late May 2026, the last shipment of the central solenoid support structure rolled through the gates of the ITER fusion reactor complex in Saint-Paul-lez-Durance, southern France, completing a procurement campaign that spanned more than 15 years and crossed an ocean. With every piece of the 59-foot-tall, roughly 3,000-ton magnet system now on-site, assembly crews can begin the painstaking work of stacking six superconducting modules into the core of the world’s largest tokamak.
“This is the culmination of more than a decade of dedication by the U.S. fusion community,” Oak Ridge National Laboratory, which manages the American contribution to ITER, said in an official announcement confirming the milestone. The modules themselves were built by General Atomics at its facility in San Diego, California, while ORNL coordinated design, materials qualification, and logistics across a network of national laboratories and industrial suppliers.
Why this magnet matters
The central solenoid is, in functional terms, the engine that starts the fusion reaction. When its coils pulse with current, they generate a magnetic field of roughly 13 tesla, about 280,000 times stronger than Earth’s magnetic field. That pulse induces an electrical current in the plasma, a superheated cloud of hydrogen isotopes confined inside the tokamak’s doughnut-shaped vacuum vessel. Without that induced current, the plasma cannot reach the temperatures (on the order of 150 million degrees Celsius) or densities needed for atomic nuclei to fuse and release energy.
The support structure that just arrived is the steel skeleton that holds the six stacked modules in precise alignment. “Precise” here means tolerances measured in fractions of a millimeter, even as electromagnetic forces slam through the assembly every time the solenoid fires. Engineers at ORNL and General Atomics spent years qualifying specialized welding procedures and transport methods before any hardware left the factory floor, because a misalignment at this scale could compromise the magnetic field geometry and, with it, the entire plasma confinement strategy.
A bottleneck, removed
The support structure had become one of the most visible physical bottlenecks in ITER’s assembly sequence. Until every piece was on-site, crews could not safely begin lowering and stacking the massive coil modules inside the reactor pit, a concrete well that also houses the vacuum vessel and the 18 D-shaped toroidal field magnets that wrap around it. Each solenoid module weighs hundreds of tons and must be craned into position with surgical care.
According to ORNL’s US ITER program page, the United States has now completed its entire central solenoid procurement package, covering everything from the superconducting coils to the structural cradle. That closes out one of the single largest hardware commitments any ITER member nation has made.
Removing the bottleneck gives the on-site integration team more scheduling flexibility, but it does not, by itself, set a firm date for first plasma. ITER is a seven-party international collaboration (the European Union, China, India, Japan, South Korea, Russia, and the United States), and the overall timeline depends on deliveries and integration work from all of them.
The long road to first plasma
ITER’s schedule has been revised repeatedly. Early targets placed first plasma in the mid-2020s. Successive reviews, driven by manufacturing delays, the COVID-19 pandemic, and the sheer complexity of assembling components built on four continents, pushed that date further out. The ITER Council, the project’s governing board, has discussed a revised baseline that places first plasma around 2035, though no final date has been formally locked in as of June 2026. Readers should treat any specific year circulating in media coverage as provisional until the Council adopts an updated schedule.
Cost has grown alongside the timeline. The project’s price tag, originally pegged at around $5 billion when the ITER Agreement was signed in 2006, has ballooned to an estimated $22 billion or more by some independent assessments, though exact figures depend on how each member nation accounts for in-kind contributions. Critics argue that private fusion ventures, such as Commonwealth Fusion Systems and TAE Technologies, could reach net energy gain faster and cheaper. Supporters counter that ITER’s scale (it aims to produce 500 megawatts of fusion power from 50 megawatts of heating input) is necessary to prove that fusion can work as a power source, not just a laboratory demonstration.
What comes next inside the pit
With the structural hardware in France, the focus shifts to integration. Stacking six superconducting modules inside a crowded reactor pit, threading cryogenic cooling lines, wiring thousands of sensor and power connections, and then cooling the entire assembly to around 4 kelvin (minus 269 degrees Celsius) so the niobium-tin superconductor can carry current without resistance is a sequence that has never been attempted at this scale.
Quality assurance will be critical. ORNL’s role formally ends at delivery; the ITER Organization’s assembly team in France takes custody from there. The gap between “delivered” and “installed and tested” can stretch for months or years on a project this complex. Detailed supplier contracts, rework records, and acceptance test data for the solenoid components have not been made public, so independent observers cannot yet confirm that every piece meets final specifications.
ITER publishes periodic progress reports, and future editions will be the first reliable indicator of whether this delivery translates into measurable schedule acceleration. Until those documents appear, the most accurate reading of the milestone is straightforward: a necessary condition for first plasma has been met, but not a sufficient one.
A precise fact in a long experiment
For anyone tracking fusion energy as a scientific frontier, a policy question, or an investment thesis, the arrival of the final central solenoid shipment lands in a specific place on the ledger. At the engineering level, it is a tangible achievement backed by verifiable procurement data and more than 15 years of American industrial effort. At the project-management level, its impact on ITER’s schedule will only become clear once the organization updates its official baseline and reports on installation progress. At the energy-policy level, it contributes to a story of incremental progress toward fusion power, without resolving open debates about cost, competitiveness, or how quickly fusion might complement other low-carbon energy sources.
What happens next depends on how effectively an international team, working in a pit in Provence, can turn a decade of manufacturing into a functioning machine. The pieces are finally all in one place. Assembling them is the next test.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
