Earlier this morning, the International Thermonuclear Experimental Reactor (ITER) Organization announced what has long been known: The largest tokamak in the world will be delayed further, prolonging the awaited nuclear fusion machine’s operations by at least a decade.

ITER is a massive doughnut-shaped magnetic fusion device called a tokamak. Tokamaks use magnetic fields to control superheated plasmas in a way that induces nuclear fusion, a reaction by which two or more light nuclei come together to form a new nucleus, releasing a huge amount of energy in the process. Nuclear fusion is seen as a potentially viable carbon-free energy source, but there are many engineering and economic challenges to overcome to make that a reality.

The project’s previous baseline—its timeframe and the benchmarks within it—was established in 2016. The global pandemic that started in 2020 interrupted much of ITER’s ongoing operations, delaying matters further.

Watch: Florida Family Takes NASA to Court Over Home Damaged by Space Trash [embedded content] video Florida Family Takes NASA to Court Over Home Damaged by Space Trash

As reported by Scientific American, ITER’s cost is four times initial estimates, with the most recent numbers putting the project at over $22 billion. Speaking at a press conference earlier today, Pietro Barabaschi, ITER’s director general, explained the cause of the delays and the updated project baseline for the experiment.

“Since October 2020, it has been made clear, publicly and to our stakeholders, that First Plasma in 2025 was no longer achievable,” Barabaschi said. “The new baseline has been redesigned to prioritize the Start of Research Operations.”

Barabaschi said that the new baseline will mitigate operational risks and prepare the device for operations using deuterium-tritium, one type of fusion reaction. Instead of a first plasma in 2025 as a “brief, low-energy machine test,” he said, more time will be dedicated to commissioning the experiment and it will be given more external heating capacity. Full magnetic energy is pushed back three years, from 2033 to 2036. Deuterium-deuterium fusion operations will remain on schedule for roughly 2035, while the start of deuterium-tritium operations will be delayed four years, from 2035 to 2039.

Oak Ridge National Laboratory” class=”size-full wp-image-2000376386″ /> A mockup of the ITER tokamak and its integrated systems. Illustration: Oak Ridge National Laboratory

ITER is paid for by its member states: the European Union, China, India, Japan, South Korea, Russia, and the United States. Progress on ITER is being made, albeit slowly, and at greater costs than initially projected.

Earlier this week, the ITER Organization announced that the tokamak’s toroidal field coils—very large magnets that help provide the conditions necessary for the machine to hold plasma—had finally been shipped, a moment 20 years in the making. The 56-foot tall (17-meter) coils will be cooled to -452.2 degrees Fahrenheit (-269 degrees Celsius) and will be wrapped around the vessel that contains the plasma, allowing the ITER scientists to control the reactions inside.

The scale of its infrastructure is as massive as its investment; the largest cold mass magnet currently in existence is a 408-ton (370-tonne) component of CERN’s Atlas experiment, but ITER’s newly completed magnet—the combined size of the toroidal field coils—has a cold mass of 6,614 tons (6,000 tonnes).

ITER’s stated projected goals are to demonstrate the sort of systems that need to be integrated for industrial-scale fusion, to achieve a scientific benchmark called Q≥10, or 500 megawatts of fusion power out of the machine for 50 megawatts of heating power into the plasma, and to achieve Q≥5 at steady state operation of the device. These are not easy goals to achieve, but nuclear fusion experiments in laboratory settings, in tokamaks and using lasers, are helping scientists inch towards fusion reactions that produce more energy than it takes to power the reactions themselves.

Now for the obligatory caveats about the difference between progress towards fusion’s scientific viability and its actual utility in addressing global energy demands, as we reported on Monday:

A wry truism—so rehashed it’s a cliché—holds that nuclear fusion as an energy source is always 50 years away. It’s forever just beyond the technologies of today, and, like an irredeemable ex, we’re always told “this time it will be different.” ITER is intended to prove fusion power’s technological feasibility, but importantly not its economic viability. That’s another vexing issue: making fusion power not only a workable energy source, but a viable one for the power grid.

In the remarks, Barabaschi also noted that the plasma-facing material in ITER’s tokamak will now be made of tungsten, rather than beryllium, “because it is clear that tungsten is more relevant for future ‘DEMO’ machines and eventual commercial fusion devices.” Indeed, back in May the WEST tokamak sustained a plasma over three times hotter than the Sun’s core for six minutes using a tungsten casing, and the KSTAR tokamak in Korea replaced its carbon diverter with one made of tungsten.

As Gizmodo has previously reported, nuclear fusion is a worthwhile field for R&D, but it should not be relied upon as the energy source to get humans away from fossil fuels, which drive global warming. The science is coming along, but nuclear fusion was always going to be an ultra-marathon, not a sprint.

More: What to Know About the DOE’s Big Nuclear Fusion Announcement