St-Paul-lez-Durance, France

Construction of the ITER fusion reactor in France is beset by financial and technical problems. Credit: Altivue

The world's largest fusion experiment is finally beginning to take shape. Workers at a vast site in southern France have dug the 17-metre-deep pit that will house the ITER reactor, and will soon install 500 pillars of steel-reinforced concrete that should protect the machine during an earthquake. But even as they toil, a quake halfway around the world has struck a blow to the project.

The 11 March earthquake and tsunami that hit Japan, one of seven partners in ITER, severely damaged key facilities for testing the reactor's components. Unless repairs can be made or work reassigned quickly, the damage could cause a delay of "perhaps several years", according to Osamu Motojima, ITER's director. Motojima says that he and his team are looking at ways to reduce the impact. "At present my target is less than one year's delay," he says.

ITER's first experiments have already been pushed back from 2016 to 2019, and the project has suffered serious cost overruns since its partners agreed to go ahead in 2006. Any extra delays are likely to increase political pressure to find cost savings and speed up work.

The giant reactor is designed to prove that useful energy can be extracted from the fusion of hydrogen isotopes. By trapping the hydrogen using powerful superconducting magnets and heating it to the point of fusion at 150 million °C with specially designed heating systems, ITER is supposed to produce ten times more energy than it consumes. The Naka Fusion Institute of the Japan Atomic Energy Agency, which is about 100 kilometres north-east of Tokyo, was a key facility for testing and developing the magnets and heating systems. But it was hit hard by the quake. "The buildings are damaged and we can't go inside," says Hiroshi Kataoka, director of the fusion division at Japan's Ministry of Education, Culture, Sports, Science and Technology, which oversees Japan's contributions to the ITER project. "We are not sure, but it may have some impact on our ability to deliver products on time."

The magnet test facilities are particularly important for the project. Late last year, a sample of superconducting cable to be used in a central magnet failed testing at a facility in Switzerland (see Nature 471, 150; 2011). Follow-up tests at the Naka institute were expected to help determine the cause of the failure and aid any redesign of the cable, which is being manufactured in Japan. The cable factories themselves seem to be undamaged, according to Richard Hawryluk, one of ITER's deputy directors-general.

Motojima says he is reluctant to shift work away from Japan unless absolutely necessary, but if the facilities cannot be repaired within six months, he says that he will seek arrangements with other partners in the project. A new schedule and any adjustments to the reactor's work plan must be ready by December, he says.

Meanwhile, the project's financial woes continue. Since 2006, ITER's construction costs have roughly tripled to around €15 billion (US$21 billion). The European Union, which is paying for some 45% of the project, has yet to find the additional €1.3 billion it needs to meet its near-term commitment in 2012 and 2013. Anne Jensen, a member of the European parliament from Denmark who sits on the parliament's budget committee, says that she and other members are generally supportive of the project, as long as it does not draw money away from other areas of research. "We think ITER is an interesting project, but it should not be at the expense of the development of wind energy or smart grids," she says. But she adds that she is optimistic that ITER's funding can be found by the 2012 deadline.

The other partners — Russia, South Korea, India, China and the United States, which along with Japan are each contributing around 9% to the construction — are also struggling to come up with additional funding. In the United States, for example, battles in Congress over discretionary spending have meant that ITER got its 2011 allocation of $80 million only last month.

Given the variety of financial and technical difficulties facing ITER, a further delay of "a couple of years" is probably inevitable, says Stephen Dean, president of advocacy organization Fusion Power Associates in Gaithersburg, Maryland. But, he adds, "that doesn't mean that they can't get it done".