Among the several reasons I was pleasantly surprised by the 2 April Lausanne accord was that Iran agreed not to use anything other than first-generation centrifuges to enrich uranium for ten years. The IR-1 is the Model T of the centrifuge industry – clunky, slow and prone to break-down. Given its limited efficiency, keeping Iran’s programme restricted for now to the IR-1 is a significant confidence-building measure.
Not long ago Iranians were promoting an alternative way of limiting the enrichment output which would have involved replacing the IR-1s with a smaller number of advanced models. The idea was that it wasn’t the number of centrifuges that mattered but rather the amount of enriched uranium produced. But allowing advanced models would have given Iran the capability for more rapid future expansion, so the United States and its negotiating partners pressed to limit both the enrichment output and the way it is produced.
As a result, the 1,000 second-generation models (designated IR-2m) that are now installed at Natanz will be removed and packed away. Although they won’t be dismantled, the decade they spend in storage will see inevitable deterioration. Just as North Korea’s two plutonium-production reactors that were under construction rusted away due to the freeze imposed under the 1994 Agreed Framework, a freezing of Iran’s IR-2m centrifuges should also result in effective disablement.
Iran’s willingness to pack away the IR-2m’s was probably influenced by poor performance indicators. Iran purchased Pakistani versions of the IR-2m from the A.Q. Khan network back in 1995, but in 20 years has never managed to get this model to work well. There is a reason that the 1,000 IR-2ms that were installed at Natanz in the first half of 2013 were never used to enrich uranium.
Iran has thus moved on to experiment with more advanced models: the IR-4, IR-5, IR-6 and IR-8. Several of these machines are installed at the Natanz pilot enrichment plant, where they have been operating off and on singly and in small cascade configurations. Under the terms of the November 2013 Joint Plan of Action, Iran was allowed to operate what it had at the time, but not to introduce more or better models nor to do more than it had been doing before the freeze was imposed. When the IAEA revealed in November 2014 that Iran had intermittently fed uranium gas into an IR-5 that had not previously enriched uranium, the United States complained and Iran stopped it.
The rules in existence at the time were not clear as to whether or not the IR-5 feeding was a violation. When the interim deal was extended later that month, the restrictions on research and development (R&D) were tightened in order to remove ambiguities. Under the extension, Iran agreed not to test the IR-2m in large cascades, not to feed gas into the IR-5, not to test the IR-6 on a cascade level with gas and not to complete installation of the partially installed IR-8 at the pilot plant. These provisions were designed to ‘limit research and development on advanced centrifuges that move the machines to the next level of development'.
It would be reasonable for the comprehensive agreement to incorporate similar limits on R&D. The Lausanne accord said Iran will engage in limited R&D with its advanced centrifuges, ‘according to a schedule and parameters which have been agreed to by the P5+1’. Parties have been tight-lipped about those details. Iran also agreed not to conduct uranium enrichment R&D at the Fordow facility for 15 years.
On National Nuclear Technology Day in April 2014, Supreme Leader Ayatollah Ali Khamenei said ‘scientific progress in the nuclear field should in no way be halted or slowed down’. For the parties negotiating with Iran, however, there would be no point to a nuclear agreement that did not slow the advancement. Iranian pride – which is as well developed as that of any country – will not allow it to stop nuclear progress altogether. Nevertheless, like most of his supposed red lines, Khamenei’s edict on R&D is pinkish.
Mark Fitzpatrick is Director of the Non-proliferation and Disarmament Programme.