Mark HibbsHelping Iran Make TRR Fuel

In January I discussed the prospects for including a serious nuclear cooperation component in a future comprehensive deal to resolve the Iran crisis, permitting Russia to provide Iran with PWR fuel-fabrication infrastructure to complete the front end of its civil nuclear fuel cycle and at the same time incentivize Moscow to firmly line up with the EU-3/US group. At the end of the day, it was evident that going this route would require a lot of Russian heavy lifting and more than a modicum of Iranian trust.

But if such a scheme is ambitious, then the gambit put forth this week by Peter Jenkins and Yousaf Butt must be categorized as minimalist. Their proposal was hatched on a Reuters blog on Thursday.

The bottom line would be that if the P5+1 aims to reduce Iran’s stockpile of UF6 enriched to 20% U-235, now being churned out at Iran’s underground Fordow site, then the powers should offer Iran to help it convert this inventory into uranium metal and process it into fuel plates for the TRR reactor. The driver behind this idea is that the IAEA has documented that Iran has continued to produce more and more 20% EUP, but also that it has dedicated a large share of that output to the production of U3O8, which is what Iran right now has to work with if it want to produce TRR fuel.

The key passage seems to be this:

But Iran appears to be having some technical problems in doing the conversion – especially in fabricating the enriched uranium fuel plates. So the world powers could encourage the IAEA to assist Iran in this conversion and subsequent fuel plate fabrication. Iran would get fuel for its reactor and, in the process, turn potential bomb fuel into a safer form.

We’ve been all over the TRR before, for five years, in fact. Are the facts on the ground right now any different than they were in 2009? Does a “TRR deal” today make any sense?

Apart from the fussy nit that what is happening in Iran is the re-conversion of the UF6 to oxide, followed by a separate process of using U3O8 product to manufacture fuel the TRR reactor, and what looks like a misunderstanding or oversight concerning the potential value of converting UF6 to uranium metal, the critical facts would appear to be these:

Iran’s Inventory of 20% U-235

After commencing with the enrichment of uranium to 20% U-235 in early 2010, Iran accumulated about 150 kilograms of EUP at this enrichment by the end of 2011, and it crossed the 200 kg threshold sometime in the middle of 2012.  The latest report from the IAEA in February says Iran had produced 280 kg of UF6 enriched to 20% U-235, of which 167  kg was still in the form of UF6. Virtually all of the rest has been introduced into the reconversion plant to produce U3O8 for fuel fabrication. When the IAEA accounted for Iran’s declared activities in February, the plant had produced U3O8 containing 50 kg of uranium, leaving about 60 kg of uranium in the process inventory. According to the IAEA data, the current rate of production of feedstock at this enrichment level in its centrifuges is about 15 kg per month. Based on Iran’s ongoing installation of centrifuges, in theory Iran could produce as much as three times that amount, as spelled out in this ISIS report.

Iran’s Requirements for the TRR

On the basis of the performance history of most material test reactors, my notes distill that at a rating of 5 MW, if the reactor were to operate for half a calendar year, the TRR would use just under 13 kg of fuel per year enriched to 19.75% U-235., assuming an average discharge burnup of the fuel of 45%. That kind of duty cycle length and fuel burnup is de rigueur for material test reactors like the TRR. The exact amount of fuel needed for a year would depend on the exact operating schedule of the reactor, fabrication losses, and the real power level (not the nominal rated level) of the reactor. If we assume fabrication losses of about 15%, Iran might need 15 kg/year to make enough fuel to operate TRR like most MTRs are operated.

This morning I found a previous post by Geoff Forden back in 2009, which suggests that my generic MTR data track well with an unreferenced IAEA report showing that average burnup of the TRR fuel has been 42%, and that the reactor has been operated on-off on a weekly basis.

The record would also suggest that the TRR in the past was operated like other MTRs worldwide. But Iran since 2009 has claimed that it wants to operate the TRR to make medical isotopes. If we assume that, then Iran would want to operate the TRR on a near-continuous basis. Operating at 80% of the time, the fuel fabrication requirement would be about 24 kg/year enriched to 20% U-235 if all the other variables were the same.

What to Conclude from the Numbers

The current inventory of 280 kg of 20% EUP which Iran has accumulated since it began enriching at this level just three years ago is enough to meet Iran’s requirements for more than twenty years if the reactor were to continue to operate as in the past, enough for a decade if Iran were to be successful in operating the 1960s-vintage machine flat out to produce isotopes round-the-clock. Beyond this, if Iran were to continue to produce 20% enriched EUP at current rates, it will produce annually seven times the amount of EUP Iran would need for TRR if it were to be operated at a high duty cycle for isotope production. If Iran increases output of 20% enriched uranium based on its deployed capabilities, it might make enough TRR fuel feedstock in a single year to operate the reactor for two to four decades.

So a gambit to build down Iran’s higher-enriched uranium inventory might fit into the picture as a CBM for the diplomats. But at Iran’s current rate of production of 20% enriched EUP, it is already piling up an inventory of higher-enriched feedstock which in relative terms would resemble the kind of surpluses which Russia has accumulated at its civilian reprocessing plant at Ozersk because Russian reactors can’t burn the plutonium fast enough.

The Butt/Jenkins piece proposes that the P5+1 help Iran process the UF6 into metal fuel instead of U3O8. They argue that the metal form would be “safer” and “more proliferation resistant” than UF6.  It would be more chemically stable and hence safer. But uranium metal can be fairly quickly converted back to UF6.  There might be a virtue in converting the UF6 to metal that the authors didn’t mention: Metallic uranium would be an intermediate stage in the production of uranium silicide that is used in fuel worldwide for MTRs like the TRR which since the 1980s were converted from HEU to LEU fuel.

Right now Iran has neither the technology nor the expertise to produce fuels on the basis of U3Si2 in powder form. There are a number of fabricators that do have this knowhow–in Argentina, Chile, China, France, Indonesia, and South Korea–and perhaps one of these, as part of a P5+1 deal with Iran, could step in and provide assistance.  On the basis of experienced accumulated by the U.S. DOE RERTR program worldwide, it would not likely be difficult for Iran to replace U3O8 with U3O2 fuel in the core of the TRR.

But if the powers negotiating with Iran were to contemplate a new TRR fuel deal with Iran and it failed to materialize, the most likely showstopper might not be whether Iran could be provided access to fuel production technology but the fact that Iran already has lots of feedstock to make the fuel on the basis of U3O8, the IAEA’s inventory data which suggest that Iran made enough fuel feedstock to operate the TRR for perhaps 3-4 years, and the expectation that Iranian scientists and technicians will on their own solve their fuel re-conversion and fabrication problems.

Back in 2009, even before Iran began enriching uranium to 20% U-235, Tehran warned that it would do that to produce medical isotopes in the TRR unless the P5+1 powers accomodated Iran and struck a deal. That didn’t happen, and now, four years later, Iran still has its TRR, and it has enough enriched uranium fuel to operate the reactor for many years. The P5+1 could repeat its offer to fabricate TRR fuel (in this case, help Iran fabricate it, as Butt and Jenkins plead), but given the fact that Iran has carried out its 2009 vows and is now making the fuel, the powers have little to offer here that would interest Iran. Iran will probably figure out how to get the temperatures and pressures right to reconvert the UF6 to U3O8 at the FPFP re-conversion plant at Esfahan, just as it figured out over the last several years how to operate the UCF plant in the same location.

A lot more than this will have to be put on the table by both sides before diplomacy shows real results.


  1. Cheryl Rofer (History)

    It seems to me that proposals like Jenkins’s and Butt’s assume that of course Iran will be glad to accept what the P5+1 offer if the proposal authors think it is a good idea. Jenkins and Butt are not the only ones who make this assumption.

    But Iran takes offense readily and loudly if they feel that there is something wrong with what looks to the rest of us like a proposal that could engender cooperation.

    So not only is there a question as to whether such proposals will be received, there is the problem of having Iran blow it up into an attack on their sovereignty or somesuch.

    Mark, you address Iran’s (lack of) motivation toward such a proposal in your last few paragraphs. If it wasn’t a game-changer in 2009, it’s less likely now.

    • Joe (History)

      Well, there are reasons for that. Iran has a long list of grievances with west going back to the days of British empire, the days of The Great Game and Anglo-Iranian oil company. With regard to nuclear issue, it is no different. Iran in mid 1970’s invested billions in Eurodif enrichment plant and still technically owns 10% of that plant which provides enriched uranium to European reactors chiefly French reactors. As per “agreements” Iran had been promised 10% of the production of that plant. To this day after almost four decades, Iran has not seen a single gram of uranium from that plant. From the perspective of Iranians, it was a fraud.

      Let’s go over another example. Iran having little uranium reserves inside the country, bought 15% shares of a uranium mine in Namibia. The mine in question used to be the world’s second largest. Iran was hoping to get uranium from the mine. Instead all they got was nothing. The mine today is under management of a western company providing uranium for western reactors. Again, Iranians have always seen this as another colonial fraud.

      I do not think, Iranians will ever be able to trust the west with such track records. Any kind of deal with Iranians if at all possible, will be small, temporary and purely transactional in nature. And this mistrust cuts across Iranian population. Even the man incharge of Iran’s nuclear program before Iran’s revolution, someone who has lived in France since the revolution and is by no means a mullah, is of the view that Iran should continue resisting the western demands to put restraints on its nuclear program and should push for a full fledged sovereign and national nuclear program. It appears, the time for tricking Iranians have come to an end.

    • Rob Goldston (History)

      Let me double down on Cheryl’s remark. Western “helpers” would be perceived by Iran as spies. And of course Iran would be right, from a practical point of view. At a minimum, the helpers would be thoroughly debriefed by the IAEA on their return. To make an offer like that proposed could be perceived as insulting both to Iran’s technical know-how and to its intelligence.

      What is Farsi for “How stupid do they think we are?”

    • Rene (History)

      It would be “have you considered me a donkey?” 🙂

    • Ricki (History)

      Citations on these statements requested, please.

  2. yousaf (History)

    thanks for the post and excellent background material.

    You say: “The P5+1 could repeat its offer to fabricate TRR fuel (in this case, help Iran fabricate it, as Butt and Jenkins plea), but given the fact that Iran has carried out its 2009 vows and is now making the fuel, the powers have little to offer.”

    It is not clear that Iran can actually make the fuel plates, at least not very well. They may boast they can but it seems they are having issues (see links in our piece). They may be able to make dummy unenriched plates but the real enriched plates?

    In any case, a offer to reduce proliferation risk cannot hurt.

    @Cheryl: no we don’t assume Iran will automatically accept the offer — but it is worth making and checking.

    • Cheryl Rofer (History)

      Making offers to check if they will be accepted might work in continuing negotiations where both sides are willing to refuse or modify the offers without public comment. That would constitute a real negotiation.

      But that’s not what is happening now. The time for negotiation is extremely limited, and Iran takes up a great deal of it with rhetoric or proposals that obviously aren’t going anywhere.

      It appears to me that the basis for claims that Iran is having problems with fabrication of TRR fuel elements is largely speculation. Iran may have had some problems at first, but it has been producing 20% enriched plates for some time now. I would link to a couple of posts by Susan Voss at Nuclear Diner, but we are having some problems with the site that we hope to have resolved early next week.

      It might have been, also, that Iran was accumulating 20% enriched UF6 as a negotiating chip. That explanation works in a great many areas.

  3. Johnboy (History)

    This may be a naïve question, but is there an export market for this stuff i.e. are there MTRs in other countries that buy their reactor fuel from overseas suppliers rather than make it themselves?

    Because if there are then isn’t one of the “lot mores” that can be put on the table an offer from the P5+1 to help Iran muscle its way into that supply chain so it can sell this stuff overseas, rather than simply demand that Iran surrender it?

    After all, Iran insists that its developing its nuclear program because it wants a nuclear INDUSTRY….. so why not help it develop a nuclear EXPORT industry.

    • Karu (History)

      Indeed there are a handful of countries that import MTR fuel for their research reactors. However, they largely rely on France’s CERCA facility, with the US, Argentina and in the near future, S. Korea, also exporting. Nevertheless this is a limited (and troubled, in the larger context of research reactor fuel production) market likely of little interest to Iran.
      A naïve question would be how much AREVA, Babcock & Wilcox, INVAP and KAERI desire an additional competitor.

  4. Amir (History)

    Diplomacy can get to somewhere if west accept Iran’s rights otherwise Iran will not concede under pressure. If this is the hope of these sanctions, then I bet west will be disappointed at the end of the day. The more they wait, more Iran advances in its nuclear persuade and more become inflexible. Sanctions will not affect a country with 80 million people and 1 trillion dollar of GDP with a young and educated population

  5. JO (History)

    Iran’s nuclear submarine program is going to require a lot of 20% enriched uranium.

    For a good exposition of sub fuel requirements google:
    “effects of variation of uranium enrichment on nuclear submarine reactor design”

    They will build a land-based test unit first. 50MWt with a 15 year refuelling interval so not enough yet for one core yet.

    • mark (History)


      The techniques applied and the fuel fabrication process for 1960s-vintage TRR-spec MTR fuel using U3O8 would not be the same as required for the high-density fuel that Iran would need for a nuclear submarine.

      If Iran really intends to use its 20%-enriched uranium for submarine fuel, an offer to help Iran make a small amount of MTR fuel would not interest Iran.

      Leaves open the question as hinted in a different context at by Butt/Jenkins that Iran’s naval propulsion fuel program is not for real but is something Iran could put on the table and then at an expedient moment withdraw.

    • JO (History)


      The push in the 80’s and 90’s to replace HEU with LEU fuel in research reactors had a parallel in nuclear propulsion. There was an international effort to encourage use of LEU. Iran would likely make use of this effort and take the LEU (~20%) route as they have done with the TRR modernisation.

      The french LEU-powered submarines use U02 heavy-clad zircalloy plate fuels (“caramel” type) rather than cermet. So the difference is only geometry and fabrication.

      To clarify a little,- the size of reactor I described earlier would be suitable for powering a 636 (Kilo-type) submarine. Good for 1200 days at rated power since its not powered up all the time. But NB. I am filling in the blanks a little on the size.

      The Iranians, when they are serious (not like that fighter plane shemozzle) do things sensibly and in steps. They have the Kilo technology well in hand. A small nuclear SSN (larger than the french nuclear SSNs) in the Indian ocean would be an extremely effective defensive asset.

      Also the project itself is absolutely a manageable size for Iran – the device is small maybe 2 meters diameter. Small circulation pumps and a third-party steam turbine. They need to come up with a fuel geometry, heat exchanger/RPV, and controls. As I said there is a legacy of international research in this area they can capitalise on.

    • Rene (History)


      Sorry, this question is not related to the present post, but since you referred to the Iranian fighter, I have a question: do you think it was a mock up, or that the design doesn’t make sense, or that essentially they won’t be able to pull it off, even with inferior electronics and engines?

      I ask this because recently one of their generals said in an interview that they’re going to have its taxi trials this year (he also said something like “westerners ridiculed us, but it’s not a sin to show an incompletely assembled aircraft.” he seemed to imply that the engine and air intakes were not installed, but that the shown aircraft was indeed the prototype, not a mock up).

    • JO (History)


      I doubt much real work has been done, its psyops. Everything is too small – 3/4 scale prototype? The fiberglass ribbing on the inside of the cockpit is stage-prop tier and the canopy is hobby acrylic. Also the aerial shots released were of an R/C plane. It could be aspirational, project for undergrads type stuff. The other twin-fin F-5 was frivolous too.

      However their rocket program is real; multiple models, logical steps, small to large, many tests, many successes, exports. The counterstrike on gulf facilities from these is their deterrence element, the air force is deprecated.

      Similar with their submarine program, which is designed to mine the gulf. Nobody knows what those mines are like, secret but I bet they’re serious about them too.

      This focus is linked to the war with Iraq. Mines, and rockets. That’s a carryover by the people in charge now of their experiences then.

  6. Rob Goldston (History)

    Two comments, which cut in opposite directions with respect to interpreting Iran’s intentions:
    1) The core currently in the TRR has been there a long time. It could be that Iran desires to replace it completely, rather than maintain its current level of burn-up. This would justify production of quite a bit more 20% enriched U.
    2) No one seems to take into account that Iran has tremendous capabilities to convert uranium oxide to UF6, in their normal fuel cycle. Yes, they couldn’t start a breakout with oxide, but they could convert oxide to finish it, it seems to me and to those with whom I have discussed this,
    Some advice: We certainly don’t want to help Iran learn how to make metal out of UF6 quickly and efficiently, for reasons that should be obvious to all, so if the IAEA is to offer help, let’s not include that.

    • mark (History)


      1.) I have discussed with safety regulators and MTR design engineers the possibility you raise of Iran increasing the power level of the TRR. This would be possible. Iran could increase the power level to, say, 10 MW, implying that its requirements for uranium fuel if operated in isotope-production mode might increase to perhaps 35-50 kg/y of 20% EUP. To do that, a lot of engineering work and physical upgrading of the entire facility would be necessary. In advanced country nuclear oversight regimes it would require a robust recalculation not only of the core of the reactor but also of virtually the entire balance of plant. On the basis of those calculations, extensive hardward modificaions would be necessary. Iran is not a signatory of the International Safety Convention. The outside world has little information about what kind of safety standards it applies for the TRR. I’m confident however that the IAEA Department of Nuclear Safety has some data points. In any case Iran upgrading the TRR to double its power level would not be a trivial undertaking and on the basis of my discussions of this subject with people who have done this work elsewhere in the world it is not likely that Iran currently has the capability to do it alone, even if we factor in what Iran has learned from Russia during the restructuring of the Bushehr project. Most of that work was carried out by two Russian nuclear power design institutes–not by Iran.

      2.)The Butt/Jenkins proposal assumes as given that Iran would have problems in reconverting oxide to hex:

      “Reconversion back to gaseous form is difficult and time-consuming and a major roadblock if Iran intends to “race to the bomb.”

      If that’s not the case, then the nonproliferation value of the proposal would appear to be compromised.

      I take your point on Iranian metal production. The Butt/Jenkins proposal as described above in the link provides no information about what the authors assume are Iran’s present capabilities for quickly and efficiently producing uranium metal from other chemical forms. If Iran would learn more about weapons-applicable metallurgy as a result of the P5+1 helping Iran in this area, then the nonproliferation benefits of the proposal would appear to be compromised.

    • Rob Goldston (History)

      1) Just to be clear, I wasn’t talking about upgrading the power level so much as just replacing the whole existing core with fresh fuel, which would presumably allow for substantially more neutron-absorbing targets, because of the more favorable neutron economy with fresh fuel.
      2) I don’t understand why it would be hard for Iran to convert oxide into hex quickly. They are making hex at large scale from natural uranium yellowcake to feed into their centrifuges already. Obviously they would not want to mix 20% enriched with natural, but they could presumably clean out a production line.

  7. mamdali (History)

    I echo the previous comments that, at the end of the day, Iran has no interest in relying on external resources (especially the ‘West’) for obvious reasons . I must add the article implicitly ‘rations’ Iranian nuclear resources to this activity and that. A definite no go.

    Also, Iran has found significant uranium deposits so the premise that Iran doesn’t have enough ore is probably not a correct one. If so, Iran may want to, for example, export plates and/or the processing technology and/or the medical isotopes. Again, orthogonal to the ‘rationing’ proposed in this article.

    Nothing short of acceptance Iranian sovereign rights to take any peaceful action with *any* resource it has, en totale, will be accepted by Iran.

  8. Cyrus (History)

    Iran has repeatedly stated that the TRR has reached the end of its life and must be replaced by a larger reactor. But apart from that, Iran has repeatedly offered to entirely cease 20% enrichment, and had agreed to the Brazil/Turkish deal too. In fact Iran offered multinational enrichment on its soil. All it asked in return is recognition of its right to enrichment. These offers were summarily rejected or simply ignored. I’ll say it again: the nuclear issue is a pretext for imposing regime-change on Iran. It is a POLITICAL dispute, and won’t be resolved until and unless the US recognizes Iran…which it can’t due to domestic political considerations. So this is all really a non-sequitur.

  9. rwendland (History)


    Are you sure you have properly allowed for the conversion losses from UF6 in the numbers above?

    For example, the 280kg in the recent IAEA report is of UF6. 67.5% of UF6 is U, so that contains 189kg of 20% enriched U. If your 15 kg/year max consumption for the TRR is of U, and Iran can convert to U3O8 with 100% efficiency, then that is “more than 12.6 years”, rather than the “more than twenty years” you give above.

    However Susan Voss in the article linked to below, says Iran’s UF6 to U3O8 conversion in 2011/2 was only 39% efficient, with 61% going to non immediately useful forms of U, presumably other compounds of U. (To be precise, 39% of the U from UF6 was converted to the U in U3O8.) So at this efficiency the 280kg UF6 would create U308 containing 73.7kg of U, or about 5 years at your 15kg U/year max consumption rate.

    As Susan Voss says, a critical factor in Iran’s 20% UF6 requirement is whether they can improve the conversion efficiency, or can quickly develop processes to convert the 61% of non immediately useful forms of U into something that can be fed back into the main process.

    As an illustrative thought experiment, if Iran cannot improve on the 39% efficiency, and they want to produce a similar batch of fuel as Argentine supplied (115.8 kg U), by my reckoning they would need about 440kg of 20% enriched UF6:

    (115.8 / 0.39) / 0.675 = 439.9

    I’m not certain about all this, but they look like plausible numbers worthy of further consideration. If Susan Voss is right about the efficiency, perhaps it would be in the west’s interest for the IAEA to give project assistance on improving this – otherwise Iran will have quite a bit of 20% U sitting around in forms that they cannot use for fuel.

    Susan Voss also says it is likely that the early Iranian fuel elements would probably have a short in-reactor life, as they do qualification tests on the fuel.

    • rwendland (History)

      … correction to detail: In my illustrative thought experiment I forgot about the likely 15% wastage in fabricating fuel elements, so it should have been:

      As an illustrative thought experiment, if Iran cannot improve on the 39% efficiency, and they want to produce a similar batch of fuel as Argentine supplied (115.8 kg U), by my reckoning they would need about 520kg of 20% enriched UF6:

      ((115.8 / 0.85) / 0.39) / 0.675 = 517.5

  10. Mohammad (History)

    In case anyone is interested, IRIB channel 1 is right now (beginning on Wednesday 20:00 GMT) airing a 30 min documentary on Iran’s production of fuel rods for the TRR, (interviewing some of the key people involved in the project, albeit anonymously – one of them, Majid Shahriari, has been assassinated before) as a part of the Sorayya TV program. I guess someone will put it in a downloadable format soon.

  11. Miles Pomper (History)

    Also, my colleague Ariane Tabatabai and I have been looking into the claimed uses Iran wants for the reactor.

    First of all, Iran didn’t use facility as a significant medical isotope production facility until its Argentinian 19.75% enriched fuel began to ran out and it needed a reason for it to produce this level of enriched LEU.

    Before 2007, Iran had been content to rely on imports of Molybdenum-99 (Mo-99), a necessary precursor in the production of technicum-99m, a key isotope used in medical diagnostic procedures.

    In addition to problems it producing fuel for the reactor, Iran relies on the technique of “neutron activation” to irradiate targets with enriched levels of Molybdenum-98 at TRR and to produce Mo-99. This neutron activation technology is significantly less efficient for producing Mo-99 than the technology most Mo-99 producers use (irradiating enriched uranium targets in a research reactor). However, it has the benefit from Iran’s point of view of not requiring a separate processing facility to purify the Mo-99, something Iran lacks. A potential carrot that the p5 +1 might consider dangling before Iran in negotiations, therefore, is help in fabricating appropriate LEU targets and fuel from Iran’s near 20% enriched uranium hexafluoride stockpile and constructing an appropriate Mo-99 processing facility.

    • Mohammad (History)

      The documentary I pointed to above, based on interviews with current or former AEOI officials (including Ali Akbar Salehi) and a few medics involved with nuclear medicine says that Iran’s import of Mo-99 was curbed by sanctions in 2007, and Iran’s nuclear medicine centers went into one or two months of crisis, before the AEOI project to produce the radioisotope in the TRR beared fruit. Before that, the TRR was only used in research and training and was only “turned on” a few hours a week.

      According to the documentary, the request for buying new 20% fuel rods for the TRR was first made by Gholamreza Aghazadeh, then head of AEOI, in a 2008 letter to the IAEA. (Iran expected to simply buy the rods, but the Western powers insisted on a swap) Then the TRR zombie fuel swap deal began to come into its short living.