A swimming pool-type reactor similar to the TRR in Iran whose fuel is apparently the subject of a deal between Iran and the P5+1. The eerie blue light is Cherenkov radiation given off as subatomic particles streak through the surrounding water.

( Techno-Wonk Alert: There are a lot of numbers in this post so if you are not into that, skip to the Summary and Discussion section.)

Friday’s apparent agreement to send Iranian LEU out of the country for further enrichment, so that it can be used to produce medical isotopes in the Tehran Research Reactor (TRR), will undoubtedly cause a lot of people to want to know more about the production of medical isotopes. It certainly made me want to know more. Fortunately, a great deal of information about Iran’s plans for isotope production can be gleaned from a paper in Annals of Nuclear Energy (vol. 30, pp. 883-895, 2003) but Sayareh, Ghannadi Maragheh, and Shamsaie; three researchers at Amir Kabir Technical University and the Atomic Energy Organization of Iran.

Most importantly, they are planning on producing 20 Curies (Ci) of Molybdenum 99 (99Mo) every other week. This is about half their diagnostic requirements for 99Mo, which they currently import. Currently, 95% of the world’s 99Mo supply is produced in six reactors around the world. As of 2006, the reactors in Canada (which produces 40% of the world’s needs), the Netherlands, Brussels, France, Germany, and South Africa used weapons grade uranium as the target . That means that sheets of 90+% Uranium 235 are inserted into the high neutron densities found inside these reactors for very brief periods of time. The US supplies about 25 kg of weapons grade uranium each year to Canada’s NRU reactor alone. This HEU is inserted into the reactor for a short time, just long enough to “burn up” about 5% of the uranium 235. The irradiated fuel is removed and the molybdenum is extracted from the fission products; molybdenum is produced in about six percent of the fissions. Thus, in a year, Canada’s NRU reactor burns about 1.25 kg of HEU and produces about 32 grams of 99Mo each year. (The remaining 23.75 kg of HEU exported to Canada each year is considered “waste” and is not recycled. That’s a lot of nuclear bombs waiting around in the trash can north of the border.) The entire world’s 99Mo production is 80 grams per year. Iran currently imports about 0.2 grams of 99Mo each year. That is about 0.25% of the diagnostic molybdenum produced in the world while Iran has 1% of the world’s population. By way of contrast, the United States with 4.5% of the world’s population, uses about 40% of the world’s diagnostic molybdenum.

Technetium Generators

Molybdenum 99 is not the isotope used for medical diagnostics. Instead, 99Mo is the “long half-life” warehouse for storing and transporting Technetium-99m. Technetium-99m is written as 99mTc where the “m” means “metastable;” just a fancy way of saying that it has a 6 hour half life and decays by emitting a gamma ray, just a very powerful X-ray. As soon as a quantity of 99Mo is made, it starts to decay into 99mTc, however, since the Tc’s half-life is a tenth of the molybdenum, the amount of 99mTc at essentially any time is a constant fraction of the amount of molybdenum. The solution that arrives in the mail is a solution of 99Mo (with a proportionate amount of 99mTc) dissolved in sodium hydroxide. When a dose is required, the same volume of Technetium removed in a column separator (technically a chromatograph) to get a similar amount of 99mTc isotope. This device is known as a Technetium generator or even, apparently, a “moly cow” because it is being “milked” for isotopes, though I would have thought “techni-cow” would be more appropriate. Iran’s weekly requirement of 20 Ci of 99Mo corresponds to, I think, about 4,000 diagnostic doses of 99mTc when the decay of the molybdenum is taken into account.

TRR Specific Information

As is well known by now, the TRR (or Tehran Research Reactor) uses 19.75% enriched uranium as its fuel. This is just shy of the 20% threshold for the fuel not being considered Low Enriched Uranium. When the TRR was supplied to Iran by the United States in 1967 (the date of first criticality), the reactor used weapons grade uranium but in 1992 Argentina sold Iran the present LEU supply. According to Khalafi and Gharib (Ann. Nuc. Ener., vol. 26, pp. 1601-1610), an initial load of the reactor consists of 32 kg of 19.75% uranium in the form of U3O8. The IAEA reports that a reload occurs when a maximum of 42% burn up has occurred. The reactor itself is run, when it is run at full operating power, essentially every other week and burns up 2% of its fuel each week. (I believe that refers to 2% of its initial load since they use the control rods to maintain a constant neutron density over time but I could be wrong about that. Wonk-readers?) If I am correct then this burn up rate means they replace the fuel load every year and a five year supply would amount to 5 × 32 kg = 160 kg (of uranium “metal”). Higher amounts of fuel required for five years have been reported in the press, but I believe that this is the correct amount and will be used in the rest of this post.

Of course, irradiating natural uranium for short amounts of time is a flag for potential weapons grade plutonium production. However, Iran will only be irradiating about 100 g of natural uranium every two weeks. This corresponds to a plutonium production of about 26 milligrams of plutonium each year.

Cost/Benefit Analysis

It costs Iran about a $1 million per annum to import its current needs for diagnostic 99Mo. About half of that is “wasted” in transit as the molybdenum decays, an amount that could be saved if the isotope was produced locally. If, as has been discussed in the media, Iran transfers enough LEU (currently at an average enrichment of 3.5% U235) for five years worth of fuel for the TRR (about 160 kg of 19.75% U235 metal according to my calculations) then it will need to ship about 2200 1500 kg of LEU to Russia. However, according to the IAEA, Iran only has about 1508 kg of UF6 LEU, as of 31 July 2009. So Iran will either have to get less TRR fuel or process more 3.5% LEU. The required amount nicely matches the amount of LEU Iran had as of the beginning of August.

How much does this cost? Assuming $90 per SWU for enrichment services, this further enrichment will cost $170,000, plus shipping. Not counting operating expenses, this is a considerable savings over the $5 million continuing to import diagnostic Molybdenum would cost over the five years the TRR fuel is expected to last. However, this does not count the sunk costs associated with the Iran’s original LEU enrichment. If that is included, again at the artificial rate of $90/SWU, the total opportunity cost to Iran for this fuel would be $380,000. Again, a remarkable percentage savings when compared to the $5 million it would cost to import the same amount of 99Mo.

The real benefit to Iran for completing this deal, however, will not be the savings of a few million dollars or even the savings of nearly half the imported diagnostic radioisotopes from unavoidable wastage due to decays during shipment. The real savings will be the foot up Iran gets in its health care from starting to develop its own nuclear medicine industry. The discrepancy between the use of diagnostic isotopes in Iran and the developed world can, and should, be dramatically reduced; as it should for the entire world.

Summary and Discussion

It seems ironic, considering the problems for enrichment that naturally occurring molybdenum in Iran’s uranium feed stock has already caused, that Iran is going to these lengths to produce it. However, Iran has developed plans to use naturally occurring uranium as a “target” for producing an important medical diagnostic isotope of molybdenum, an isotope whose decay product can be used to scan for cancers in bone, heart, lung, and kidney. Iran already imports a sizable quantity of this pharmacological radionuclide but producing it indigenously would not only save Iran a considerable amount of money each year, much more than it would pay for the fuel for the reactor it would use to produce it, but also allow a more efficient use of this short lived isotope by preventing the decay of nearly half of the amount bought before it even reached the patients. Perhaps the biggest incentive indigenous production of 99Mo in Iran would be the encouragement of its entire nuclear medicine infrastructure; an infrastructure that might right the imbalance of medical isotopes into this developing country relative to other nations.

Sending essentially all of its current LEU stock that it has produced over the past two and half years out of the country is a big deal for Iran. Iran has experienced a history of being denied access to the nuclear infrastructure it bought into in the West. (I’m primarily thinking of the $1 Billon it investing in Eurodiff but Iran can, and does, list other examples.) It is a tremendous leap of faith for Iran to send this material out of country and the world should appreciate it. It is also, or at least should be, a big deal for the West. Even though this uranium has been under safeguards and most analysts feel it would be unlikely for Iran to use it to build a bomb, if they decided to do so, it should still be very reassuring. After all, it was only in early September that high US officials were publicly worrying about just such a diversion of materials.

If Iran goes ahead and sends its LEU out of country, the West needs to respond with a bold diplomatic option of its own. The best such proposal would be a multinational enrichment center in Iran.

Note on units: The customary unit for medical doses is curies (or Becquerels, I suppose, for us SI enthusiasts) which is related to the number of radioactive decays per second at the time of administration. This makes a lot of sense for practitioners since they care very much about the amount of radioactivity they are exposed to and to how much they use on their patients. Thus, the “standard” diagnostic dose of 99mTc is 25 mCi and Iran imports 20 Ci of 99Mo per week. Note that since 99mTc has one tenth the lifetime, a gram of 99mTc is ten times as radioactive as a gram of 99Mo, which has a “typical” radioactivity of 1,850 Ci. I find this unit convention not very informative for my purposes so I have tried to consistently use grams produced.

Update: Unfortunately, I did not incorporate some edits I had planned on making before posting this article. As a consequence, the wrong number for the amount of 3.5% LEU inadvertently ended up on the blog. I’ve correct the numbers after “striking out” the wrong numbers so you can see where the error was.