In the foreground of the nuclear talks between Iran and Western powers that got under way in Geneva this month were centrifuges, yellowcake and enriched uranium — all elements of what Iran calls a peaceful nuclear-energy program and what the West worries is a route to a nuclear weapon. But Iran has also charted a second route, one that could produce fuel for a possible bomb not from highly enriched uranium but out of plutonium, a product of the heavy-water reactor nearing completion in the hills outside the city of Arak, 300 km (190 miles) southwest of Tehran. Heavy water is water with an extra neutron, useful in moderating a nuclear reaction.
Because it is not yet up and running, the Arak heavy-water reactor has remained in the background of the nuclear controversy. But it looms larger every day. The reason: once Arak goes online, the option of destroying Iran’s nuclear program with air strikes becomes moot. The reactor is essentially invulnerable to military attack, because bombing one risks a catastrophic release of radioactivity. In the words of Israel’s last chief of military intelligence, Amos Yadlin, who piloted one of the F-16A’s that cratered Iraq’s Osirak heavy-water reactor in 1981 before it was due to become operational: “Whoever considers attacking an active reactor is willing to invite another Chernobyl, and no one wants to do that.”
That reality is the reason why some experts are drawing attention to a peculiar notice filed by Iran’s nuclear agency to the International Atomic Energy Agency (IAEA) in May. Iran told the U.N. agency that, as it readies the Arak plant for operation, it intends to do a practice run: instead of inserting real fuel rods filled with uranium into the reactor’s core, where nuclear fission occurs, they would insert inert “dummy” fuel rods. And instead of pumping heavy water into the reactor to moderate the nuclear reaction and absorb the thermal energy being released, Iran said it plans to use “light water,” just ordinary H2O.
The plan mystifies experts, who take particular issue with testing the system using light water. The facility would be contaminated by ordinary H2O, which if mixed with heavy water would render the latter unusable, because in order to work heavy water must be 99.75% pure.
“Anything above that is hard to achieve and testing the system with light water would leave a residual atmosphere of H2O that would degrade the heavy water when it is added,” writes one U.S. specialist of heavy-water reactors, who has worked with the Institute for Science and International Security (ISIS), a Washington, D.C.–based think tank, and who shared his assessment on condition he not be identified further. In other words, rather than save time, using ordinary water would delay the project for the weeks required to clean the system thoroughly enough to assure no trace of H2O remained; it wouldn’t take much to dilute the heavy water below 99.75%.
Iran’s stated intentions are unlikely enough that an Israeli nuclear specialist suggests that they might be a ruse. Ephraim Asculai, a scientist retired from the Israel Atomic Energy Commission, warns that Iran may have no intention of carrying out a dry run at all. It may be a cover story, he posits, for a plan to rush the installation of live fuel rods and heavy water instead — essentially getting the Arak facility “hot” before the outside world expects, at which point it becomes invulnerable to military attack. There might then be no way to stop Iran’s nuclear program short of invasion.
“At that point, they are in the ‘zone of immunity’ as it’s called,” says Asculai, who has also worked at ISIS; he is currently a senior research fellow at the Institute for National Security Studies, a think tank at Tel Aviv University.
Although Iran already has enough enriched uranium to fuel more than one bomb — should it make the decision to convert the enriched uranium to military use — that’s not all the world must worry about. “The Arak reactor is increasingly relevant and, yes, it’s been a sideshow,” says Mark Fitzpatrick, a former State Department nonproliferation chief now at the London-based International Institute for Strategic Studies, or IISS, where Asculai laid out the “ruse” possibility during a September talk. “The angle of Iran pulling a fast one isn’t something that can be dismissed,” says Fitzpatrick. “I think it’s unlikely,” he adds — because so rash an act would run counter to Iran’s patient behavior to date and it would take “a couple of years” to generate enough plutonium for a bomb. “But it’s something that should be factored into whatever is tabled in Geneva.”
U.S. officials say Arak is indeed on their radar. “We have very serious concerns about them having a plutonium capability, another pathway for fissile material for nuclear weapons,” a senior American official told reporters before the first round of talks since the election of Iranian President Hassan Rouhani; a second round is set for early November. “It is a subject,” the senior official said, “of enormous concern.”
That concern is shared by other Western nuclear experts who worry that Iran might try to sneak the plant online. Olli Heinonen, a former IAEA deputy director who is now at Harvard University’s Belfer Center for Science and International Affairs, says Iran might rush to declare the Arak plant operational before a watching world thinks it has produced enough fuel rods necessary to power the plant, which outside experts estimate at 100 to 150. Manufacture of the rods has apparently lagged, but Heinonen says Iran might install a far smaller number of fuel rods, perhaps as few as the 10 currently known to be in hand, and with that announce that Arak had become operational, and therefore too dangerous to bomb.
“It might be an exaggeration, but they could try to assert that the reactor is now loaded,” says ISIS president David Albright. Albright says that, technically, an Iranian claim that the reactor was “operating” with just 10 or 20 fuel rods would amount to a bluff, because more fuel would have to be in place to make the plant invulnerable to bombing. “Even if the fuel is partly loaded, the reactor could still be destroyed, and the radiological consequences of that would be very slight, if any,” he says. All that, of course, assumes that outsiders had a high level of confidence in how many rods are installed.
Iran claims the Arak plant is intended to produce isotopes for a variety of medical uses. But the reactor is far larger than required for that purpose, and, if fully operational, would generate enough plutonium to fuel two nuclear weapons annually. Extracting the plutonium would require another step, including the addition of a reprocessing facility Iran has not yet built, “but it wouldn’t be beyond them to get it on the black market, or more likely, from North Korea,” Fitzpatrick says.
Albright suggests another possibility, one altogether more heartening for the West: that the Iranians’ plan to test the plant with light water is sincere. “It’s not very smart,” he says, “which maybe implies something about Iranian capabilities and worries about the reactor.” The plant is years behind schedule, and the timetable slipped again after Iran’s May statement to the IAEA. The plan then was for a dry run in the final three months of 2013 and for the reactor to come online early in 2014. The timeline has been pushed back, but no one knows how far. IAEA inspectors complain that Iran has held back design information and limited their access to parts of the site.
“Before the delay was known, I estimated they could produce a plutonium device sometime toward the end of 2016, if everything went well for them,” says Asculai. “So there is still time. And meanwhile,” he says, “the enriched uranium route is really there.”