Find a Security Clearance Job!

Weapons of Mass Destruction (WMD)


A Covert Reactor?

Does Iran have a plutonium production reactor, covertly hidden in some magic mountain unseen by outsiders? Or are the overt facilities the sum of Iran's program? In this provocative analysis, Mark Gubrud concludes that while Iran could hide a reactor, it probably has not. Construction of the heavy water plant at Arak is five years ahead of reactor construction because it take five years of heavy water production to accumulate sufficient heavy water for the reactor.

A plutonium production reactor similar in size to the existing reactor at Yongbyon in North Korea or the one that Iran has declared its intent to construct at Arak (adjacent to the existing heavy water plant that would appear to be intended for providing D2O needed by such a reactor) would be expected to produce some 20-40 MW of waste heat that could not be efficiently converted into electrical power and would have to be disposed of somehow.

A conventional assumption would be that the infrared signature of this waste heat would be readily observable by satellite and therefore such a reactor could not be operated covertly. This would indeed be the case if the heat were to be dissipated by conventional means such as a single large cooling tower or by discharging hot water into a nearby river at a single point. However, any analysis of the potential for covert construction (underground or disguised as another type of facility) and operation of a plutonium-production reactor would have to consider the possibility that unconventional means could be used to hide or disguise the waste heat.

Covert construction and operation would significantly complicate and increase the cost of such a project, and would expose the facility to critical security risks if security measures that would normally be taken were forgone in order to avoid raising suspicions as to its true nature. Besides the heat release, there are additional signatures of reactor operation and plutonium separation, such as the release of particular isotopes as gases, contaminants of the reactor secondary coolant, or wastes from the chemical separation process, that could be difficult to fully control. However, the heat signature should not be impossible to conceal.

A typical large electrical generating plant may dissipate thousands of MW of waste heat, and typical oil refineries and large-scale chemical processing plants may dissipate hundreds of MW. Therefore it should not be difficult to hide an additional 20-40 MW of wate heat in the thermal budget of such a facility if the reactor were located nearby. The additional heat from the reactor could either be used in the existing process or inserted into the waste heat stream from the process, cutting back its operating level if necessary.

The heavy water plant at Arak is reportedly to use the Girdler sulphide process. Canadian plants using this process required approximately 33 TJ of steam heat at moderate temperature (130 C) per metric ton of D2O produced . The Arak plant is to have an initial capacity of 8 t/yr. Therefore the Arak plant alone could dispose of around 10 MW and could be combined with another disposal method, or perhaps a covert reactor may be smaller (and it may be difficult to do calorimetry within a factor of 2 by remote observation).

Controlling the thermal signature from the reactor itself and from steam piping carrying the heat away from the reactor to the disposal site would require a combination of insulation and active cooling with careful temperature regulation, but this would be feasible if the distance were not too great.

Other possibilities for disposing of the waste heat would include evaporation of water into ambient dry air, not in a conventional cooling tower but perhaps at many points in open country or in a single large building vented horizontally; discharge into a nearby river at many points up- and downstream (this method could also be used to disguise a somewhat elevated waste heat output); discharge into an underground river or aquifer; or injection of steam into a deep well as is done conventionally to improve recovery in old oilfields.

The principal difficulty with such scenarios would be the risk of detection either by surveillance during construction, or during operation due to an accident, or at any point due to a breach of security or detection of some signature by technical means. Refusal of a demand for on-site inspection might be taken as prima facie evidence of a clandestine facility. It may be implausible or excessively risky, as well as undermotivated, for Iran to undertake such a large and complicated project and try to keep it secret from Israeli, American and allied intelligence agencies, but that is more difficult to assess on a purely technical basis.

Iran's nearly-operational D2O production line at Arak and announced plans for an isotope prodcution reactor to begin construction this year at an adjacent site follow a logical sequence. The IR-40 reactor is expected to need 80-90 t of D2O to get started, 5 yrs. production at Arak once its capapcity is doubled, although Iran must have other stocks of D2O that can contribute. Once the reactor is online, part of its heat output can be used to support the D2O plant, whose output will then exceed the operating needs of the reactor, but could fuel further ambitions.

In conclusion, a covert reactor cannot be ruled out on the basis of the heat signature alone. However, such a reactor seems unlikely, particularly at Arak where a declared reactor is to be built and will be subject to inspection (while also giving Iran an option to openly break out of safeguards and produce Pu). Challenge inspections at Arak would likely reveal any provisions for a covert reactor to be used as a steam source, and under inspections it would be difficult to hide the reactor, even underground, since there would need to be access for moving large pieces of equipment in and out. The D2O facility alone is not subject to IAEA inspections, but the declared reactor would be.




NEWSLETTER
Join the GlobalSecurity.org mailing list