Centrifuge disclosure illustrates value of Additional Protocol

Posted by Andreas Persbo (andreas.persbo) on May 04 2011
VERTIC Blog >> Arms Control and Disarmament

Mikael Shirazi, London

The recently-disclosed existence an Iranian manufacturing facility involved in the production of centrifuge components for uranium enrichment serves as a useful illustration of the verification problems associated with the Islamic Republic’s nuclear programme. Whilst not a breach of its duties under the Non-Proliferation Treaty (NPT), the revelation does not build confidence in the exclusively peaceful nature of Iranian nuclear program and illustrates the value of the IAEA’s Additional Protocol in allaying proliferation concerns.

At a 7 April press conference in Washington, DC, the National Council of Resistance of Iran (NCRI) exposed the role of the TABA plant in producing centrifuge parts for Iran’s controversial uranium enrichment programme. According to the NCRI, TABA manufactures ‘casing, magnets, molecular pumps, composite tubes, bellows, and centrifuge bases’ primarily for the current generation of machines—but also for emerging next-generation centrifuges.

The NCRI further claimed that the plant had been operating for almost five years and had already produced parts for over 100,000 centrifuge machines—though a number of analysts appear sceptical of this high figure. Iranian Foreign Minister Ali Akbar Salehi responded soon afterwards by confirming the existence of the site and its involvement in centrifuge production, but he denied that any attempts had been made to deliberately disguise its use. ‘It is in no way secret’, he said. ‘There are plenty of factories in the country that manufacture equipment needed by the Bushehr power plant and the Atomic Energy Organization of Iran’.

Ali Asghar Soltanieh, Tehran's envoy to the IAEA, also refuted any allegations of concealment, pointing out that no provisions in the NPT’s Comprehensive Safeguards Agreement (CSA) requires such manufacturing installations to be monitored and that ‘the NPT merely requires inspection of centrifuge machines.’ Despite these protestations, doubts concerning Iran’s commitment to openness on these issues continue. According to the satellite images provided by the NCRI (as the only evidence for their claims), TABA is located in a nondescript industrial park roughly 80 miles from Tehran and has few distinguishing features. The facility’s generic name—a Farsi abbreviation of ‘Towlid Abzar Boreshi Iran’, meaning ‘Iran Cutting Tools Company’—also gives little away. Such a lack of distinguishing features raises concerns about the ease with which these facilities may be set up covertly, subsequently diverting sensitive equipment to other secret nuclear sites in pursuit of a clandestine weapons program. Which begs the question: how much about Iran’s centrifuge production capabilities is actually known?

Centrifuge manufacturing in Iran
Most information that is publicly available on the uranium enrichment programme was gathered between November 2004 and February 2006, when Iran agreed to implement the IAEA Additional Protocol on a voluntary basis, granting far more intrusive access than was provided by its CSA (to which it has since reverted). This information, in turn, was dated to the period before late 2003, at which point the nuclear programme was suspended for three years after the revelation of its existence internationally. (TABA was reportedly established after February 2006—that is, after Iran turned its back on its Additional Protocol.)

According to a report by proliferation expert David Albright, based on the IAEA investigations, the two central organisations involved were Iran’s Atomic Energy Organization (AEOI), which was primarily in charge of development and testing, and the Defence Industries Organization (DIO), which was primarily in charge of manufacturing and assembly.

The AEOI conducted most of its testing operations before 2002 through a company named Kalaye Electric, originally a clock factory which it acquired in the 1990s. It is believed since 2002 to be producing component parts. Much like TABA, a generic name (which translates as ‘Electric Goods Company’) and nondescript facilities in an industrial park obscured its purpose. Kalaye was the primary tester and developer of the current ‘P-1’ generation of centrifuges, and is now also believed to be working on more advanced designs. Farayand Technique, a subsidiary of Kalaye, conducted quality control activities for centrifuge parts such as rotors, and made and assembled parts of the centrifuge’s bottom bearing. Another subsidiary, Pars Trash (originally an automobile part factory), manufactured the thick aluminium tubes which encase centrifuge machines.

Since 2002, centrifuge development, testing and assembly operations have been directed from the Pilot Fuel Enrichment Plant (PFEP) at Natanz, which also houses Iran’s primary uranium enrichment site. The DIO conducted its manufacturing operations mostly via 7th of Tir Industries, a well-guarded complex in Esfahan which was also involved in missile production. Here, around 20 important rotating centrifuge parts were produced, including the bellows critical to the Iranian P-1 design. Another DIO contractor, Khorasan Metallurgy Industries and its subsidiary Kaveh Cutting Tools Complex, made simpler stationary P-1 components such as its scoops and molecular pumps. The DIO-affiliated Sanam Electronic Industry Group was another facility involved in centrifuge production in an unspecified capacity. An unnamed father-and-son workshop in Tehran also made the P-1 motor, a relatively simple part comparable to that of a vacuum cleaner. In addition to Albright’s list, two other institutions have been mentioned in this regard.

In August 2006, after Iran ended its voluntary implementation of the Additional Protocol, the NCRI claimed another company had been set up identified by the acronym TSA—or ‘Iran Centrifuge Technology Company’ in English. TSA had apparently absorbed the staff and facilities of Farayand Technique and Pars Trash, and is currently on the British WMD End-Use Control Export List. Another company, Abzar Boresh Kaveh Co., was sanctioned in Annex III of UN Security Council Resolution 1803 (2008) for being ‘involved in the production of centrifuge components’, though in what capacity is not elaborated by the resolution. The Annex likewise included Khorasan Metallurgy and Sanam Electronic. A prior resolution, number 1737 (2006), also included the AEOI, DIO, Kalaye Electric, Farayand Technique, Pars Trash and 7th of Tir in its sanctions regime.

Deteriorating understanding
We can now add to this list the TABA facility as a known production centre—but it is unlikely that this discovery alone will be particularly helpful in understanding Iran’s centrifuge production capacity. The information sketched out above shows how, prior to 2003, this capacity consisted of a network of very small-to-medium sized facilities spread across Tehran and other cities. Such a system allows for the diversification and dispersal of facilities, making detection (and, perhaps not incidentally, destruction) more difficult. It may be that such a network still exists today, with TABA as only one cog in the machine. Furthermore, the above list was out of date even as Albright compiled it in 2008. The interrelationship between these entities is now unclear (as is whether they still function at all), and the possible existence of TSA and other companies exemplifies the difficulty of gathering reliable information on an industry where relatively small workshops in unremarkable buildings can play an important part in the manufacturing process. Names can be changed, and equipment can be transported and set up relatively quickly.

Though it is not publicly known what further information has been gathered by foreign intelligence agencies, it would seem that International Atomic Energy Agency’s understanding of Iran’s centrifuge production complex is relatively poor, and deteriorating as time passes from Iran’s termination of its adherence to the Additional Protocol. This lack of knowledge offers the Iranian authorities the possibility—should they so choose—of secretly sending centrifuges to a hidden enrichment installation to produce weapons grade fissile material. It is a possibility that unsettles many governments, and recent announcements of the development of more efficient centrifuge designs will do nothing whatsoever to assuage their concerns.

The IR-2m and IR-4 centrifuges, which some reports suggest are able to enrich uranium at up to five times the rate of the P-1, appear to have moved to an advanced testing phase at the PFEP. If they enter into mass production, this could mean fewer centrifuges would be needed at a smaller secret site to produce fissile material, or they could shorten the time in which this can be accomplished.

Establishing a sound understanding of the Iranian centrifuge production would help to build international confidence in the ostensibly peaceful nature of its nuclear program, by ensuring that all equipment produced is used appropriately. Iran has correctly indicated that under the current verification regime—a Comprehensive Safeguards Agreement signed with the IAEA in 1974—there is no strict obligation to declare centrifuge production facilities. The 1974 agreement largely focuses on nuclear material accountancy, and guarantees IAEA access only to those facilities through which such material passes. This includes reactors, conversion plants, fabrication plants, reprocessing plants, isotope separation plants, separate storage installations, or any location where significant amounts of nuclear material is customarily used—but there are no requirements regarding centrifuge production facilities.

By contrast, the information detailed above was almost entirely gathered via the provisions in the Additional Protocol, which Iran has signed but not ratified and to which it temporarily adhered between November 2004 and February 2006. The Model Additional Protocol specifies in Article 2.a.(iv) that the participating state must provide the IAEA with a description of the scale of operations involved in centrifuge production. According to Annex I, this means the manufacture of centrifuge rotor tubes or the assembly of gas centrifuges. This is further detailed in Annex II, which describes the various components that function with centrifuges: for example, rotor assemblies, rotor tubes, bellows, baffles, top and bottom caps, magnetic suspension bearings, molecular pumps, motor stators, centrifuge housings, and scoops, among others. The Additional Protocol therefore constitutes a much more robust mechanism with regards to those verification difficulties currently experienced in Iran, and its restoration would be the most reliable way to shed light on the issue.

Last changed: May 04 2011 at 6:57 PM