Physics Today David Kramer | 16 Apr 2018
The leader of the international collaboration challenges a DOE undersecretary’s claim that the price tag will be tens of billions of dollars higher than official estimates.
The US Department of Energy has nearly tripled its cost estimate for ITER, the fusion test reactor in France that’s being constructed by a seven-party international collaboration, to $65 billion. ITER headquarters is pushing back, sticking by its figure of $22 billion. Though DOE has maintained in the past that the US contribution could balloon, this marks the first time the agency has publicly challenged the ITER Organization’s overall cost assessment.
Paul Dabbar, DOE undersecretary for science, provided the estimate to the Senate Appropriations subcommittee on energy and water development on 11 April. The $65 billion covers construction alone, he said; annual operating costs once experimental operations begin in 2025 aren’t included. Yet Dabbar seemed to confuse matters by telling senators that ITER’s cost estimates are “reasonable.”
ITER director general Bernard Bigot said there’s no reason to deviate from the $22 billion construction cost estimate he provided to a House of Representatives hearing in 2016. DOE’s “figures are not endorsed by the ITER Organization,” he said in a statement. “The cost has not gone up; we continue to adhere to the cost projections defined in 2016, when the new ITER baseline was agreed ad referendum by the ITER Council,” the project’s governing board.
According to a DOE spokesperson, the agency’s estimate is based on extrapolating from $6.4 billion, the high end of the anticipated US contribution as determined by a 2013 review committee and confirmed in a January 2017 report. The spokesperson adds that ITER doesn’t provide an official estimate of construction costs because the participating countries have different methods of pricing out their in-kind contributions—mostly in the form of fabricated reactor components—and those estimates are not reported to the ITER Organization.
Bigot noted that the day after Dabbar’s testimony, the European Union Council of Ministers endorsed ITER’s nearly two-year-old baseline estimate, which covers construction from 2007 to full completion in 2035. Including a 10% contingency to account for overruns, ITER’s cost to EU members is €11.7 billion ($14.5 billion). As host, the EU is paying 46% of ITER’s cost, five times the share of each of the other six partners: China, India, Japan, Russia, South Korea, and the US.
Based on the EU’s share of the total cost and industrial procurement rates, total construction costs could be extrapolated to €25.6 billion, including a contingency. Bigot said the actual price tag is likely to be lower than that because procurement costs in India, China, South Korea, and Russia, which are collectively picking up 36% of the cost, are 15–20% lower, on average, than those in Europe.
In any case, cost estimates for the project are fraught because most of the partners’ contributions are in-kind, and accounting practices, including the use and size of the contingency, vary widely. For example, DOE adds a contingency of nearly 50%, but South Korea, China, and Japan do not include any contingency.
Dabbar’s testimony came in front of a subcommittee whose chair, Lamar Alexander (R-TN), and ranking member, Dianne Feinstein (D-CA), have tried to zero out the ITER budget in three separate appropriations bills. Each time, funding was restored in conference with the House. The fiscal year 2018 omnibus appropriations act signed into law in March provides $122 million for ITER. Through FY 2017, the US has contributed approximately $1.1 billion to the project, $975 million in-kind and $145 million in cash. The US is currently in arrears for $65 million in cash.
The Trump administration is reviewing ongoing US participation in ITER as part of a broader examination of DOE’s range of nuclear energy activities. Dabbar indicated that review won’t be completed prior to the Appropriations subcommittee’s markup of the FY 2019 budget in the coming weeks. In the meantime, the agency has requested $75 million for the project next year, which includes continued funding for the construction of ITER’s central solenoid at General Atomics in San Diego.
Although initial plasma experiments with deuterium are scheduled to begin at ITER in 2025, construction will continue through 2035, the planned date for the first ignition experiments using tritium.