Jeffrey LewisIran to Enrich 20 Percent LEU

Iran’s announcements that it is going to produce 20 percent Highly Enriched Uranium is bad news — not least because the Iranians said they would do so without respect to the presumably doomed TRR fuel swap. (Though this thing has risen from the dead so many times, maybe I should start calling it the Zombie Fuel Swap.)

I made some comments to Le Figaro, but they are in French. (Well, they’ve been translated. I can barely say Châteauneuf-du-Pape.)

Glenn Kessler has a news analysis on the implications of the deal, in which David Albright explains why Iran could go from 20 percent HEU to a bomb in relatively short order:

Meanwhile, enriching uranium under the guise of medical needs will get Tehran much closer to possessing weapons-grade material. Iran insists it has no interest in nuclear weapons. But Albright said 70 percent of the work toward reaching weapons-grade uranium took place when Iran enriched uranium gas to 3.5 percent. Enriching it further to the 19.75 percent needed for the reactor is an additional “15 to 20 percent of the way there.”

Once the uranium is enriched above 20 percent, it is considered highly enriched uranium. The uranium would need to be enriched further, to 60 percent and then to 90 percent, before it could be used for a weapon. “The last two steps are not that big a deal,” Albright said. They could be accomplished, he said, at a relatively small facility within months.

It must seem odd for casual readers to see 20 percent and 90 percent U235 lumped together as highly enriched uranium or to be be told that Iran will find it much easier to go from 20 to 90, than from 5 to 20. That’s not how everyday math works, where 5 and 20 are closer to “ten” and 90 rounds to “one hundred.”

For many readers (especially of this blog) the answer is obvious. But for those to whom it is not obvious, Francesco Calogero found a nice way to illustrate the same point to students at a previous ISODARCO meeting. The essential concept is understand enrichment as a process of removing undesirable isotopes (or more specifically, isolating the desirable ones).

So, imagine 1000 atoms of uranium. Seven of them will be the fissile isotope Uranium 235. The rest are useless Uranium 238. (If you are the sort of person who just said, “Hey! What about Uranium 234?” or other nitpicks this post is not aimed at you.)

To make typical reactor fuel, Iran or any other country would removes 860 of the non-U235 isotopes, leaving a U235:U238 ratio of 7:140 (~5 percent).

To make fuel for the TRR, Iran removes another 105 non-U235 atoms from the 140, leaving a ratio of 7:35 (20 percent).

To make a bomb, Iran needs only to remove 27 of the remaining 35 atoms, leading a ratio of 7:8 (~90 percent).

This is simplified illustration, of course, since some of the U235 ends up in the depleted stream as “tails” — but you get the idea.

You can see why separative work is measured as mass — the interesting question is the amount of material separated — and why the lower levels of enrichment actually require more work.

As a result, as Kessler notes, Iran’s announcement “marks a new and potentially dangerous turn” in the situation. If Iran enriches a significant amount of U235 to 20 percent — and that’s a stated goal right now, not yet an actual achievement — then Iran would be able to “top off” the enrichment a small, clandestine facility like the one revealed near Qom. I did a few calculations, but I don’t need to wade into the middle the FASISIS steel cage death match that Josh has so ably chronicled. I will just encourage people to do their own back-of-the-envelop calculations. I think the answer is pretty obvious.


  1. MarkoB

    Man, this sounds like the North Korean strategy of tit-for-tat. Obama makes the move for more sanctions, so Iran says it’s going to 20%.

    If sanctions is a non-proliferation strategy then it seems not to be working.

    At any rate, the enrichment will occur under the watch of the IAEA. Will be interesting to see what effect, if any, the old Molybdenum bug-bear will have.

  2. PC

    On the first point, continuing their tradition over the last few months, different Iranian officials are saying different things about how this effects the fuel swap:

    Ali Akbar Salehi, a vice president as well as the head of the country’s nuclear program, said the further enrichment would be unnecessary if the West found a way to provide Iran with the needed fuel. “Whenever they provide the fuel, we will halt production of 20 percent,” he told the state TV late Monday, referring to the enrichment of the country’s stockpile above its current level of 3.5 percent.

    Not that it matters, since Iran’s ultimatum is a non-starter. But it is interesting they still cannot seem to get their act together on this internally. Unfortunately, the default action is escalation.

  3. Yale Simkin (History)

    The enrichment example echos one of mine in ACW from back in 2007

    Let me give a crude explanation of the counter-intuitive process.

    Imagine a bowl with 1,000 ping-pong balls in it. 993 of the balls are green. 7 of the balls are red. The balls are at “0.7% Red Enrichment.”

    Now imagine reaching in the bowl and pulling out unwanted green balls. You are doing “separative work”. You will be leaving the red balls in the bowl.

    Remove 840 green balls, a long and tedious job. Now you have 153 green balls and 7 red balls.

    You are now at “4.4% Red Enrichment”

    Last step. This time remove only 152 green balls.

    This leaves 7 red balls and 1 green ball or an “88% Red Enrichment.”

    So note: It took EIGHT-FIVE percent of the work to go from 0.7% to 4.4%!

    That is why “Peaceful” enrichment is a fraud, and selling someone LEU for their “civilian” power plant is nutzy if they have even a miniscule and crude enriching technology – centrifuges, calutrons, whatever.

  4. Cliff

    If this is such bad news, why didn’t we accept Iran’s offer for a straight swap via Turkey?

    The US administration’s insistence that any deal require Iran to give up everything up front in exchange only for our promises to deliver the enriched uranium at some distant date obviously makes this “outreach” destined to fail. Consequently this policy is identical to Bush’s, just with a different PR spin.

    And yes we all know that weapons grade material is only “months” away. Thats why you strengthen the monitoring regime by trading acceptance of Iranian enrichment for implementation of the additional protocol. Until that has been seriously tried, there is no justification for any further steps.

  5. Paul
  6. nick (History)

    All very rough estimates, but TRR needs 20Kg per year to replace the spent fuel rods. At 3 Kg per month with the above ground PFEP cascade, in 7 months there will be enough, Uranium mass to oxidize for the fuel fabrication. If by then, fuel exchange happens, everything will remain under the IAEA seal, and could be shipped as part of the deal, otherwise it will go to FMP for assembly. FMP might be the the weakest link in Iran’s fuel cycle. This whole issue could have a more positive outcome if sanctions are avoided and a partial exchange without delay is negotiated.

  7. Azr@el (History)

    How far along is the IRI ALVIS program? Could laser isotope separation function as the topping off stage of enrichment? Thus no need to reconfigure the centrifuge assemblies and no trace contamination. Run the centrifuges for the IAEA peanut gallery to do the hard lifting up to 3% and then run a few mobile AVLIS cells to get everything up to 90% should the geostrategic need ever present itself.

    Regardless, 20% U-235 should still be able to go kaboom but that would be a nuclear device not a deliverable nuclear weapon.

  8. JO

    On the other hand, if the Iranians do as they say they are and eventually manage to fabricate domestic fuel for the TRR, Iranian enriched uranium supplies will be greatly diminished. An illustration of the problems of blunt tools like sanctions.

  9. archjr (History)

    Great idea to simplify things, although complification is what I love about this blog.

    In the interest of simplifying, I recommend Robert Wright:

    Now that the TRR swap sideshow/charade, acclaimed with such enthusiasm, is dead, or irrelevant, it’s time to refocus on the exact same choices presented the moment the Iranian enrichment and plutonium programs first entered the public consciousness, now SEVEN-AND-A-HALF years ago. Do we:

    1) Acquiesce, and plan for an inevitable Iranian bomb;

    2) Accept that Iran has every legal right to enrich uranium on its soil, and demand AP and other IAEA safeguards approaches to build confidence; or

    3) Bomb?

    Answer: reread this, the last documentable evidence of solidarity in the international community:

  10. Scott Monje (History)

    Am I correct in understanding that Iran has no capacity to make the particular fuel plates used in the TRR even if it does enrich its uranium to 20 percent? This is something currently made only in France and Argentina, right?

  11. Tim (History)

    The Iranians have been in no hurry to convert the existing enriched stock from gas to fuel elements. It is naive to assume that gas enriched to 20% will be handled any differently, at least not until stocks have been replenished. Expecting any significant depletion of the Iranian stockpile of enrichment uranium is to ignore history.

  12. Ben

    Furthering what Cliff said, it is unfortunate that there wasn’t more explanation of the Turkey option or consideration of a greatly strengthened inspections regime. Iran already has the LEU and is going to keep it if no exchange takes place, so what is so dangerous about a multi-staged swap? By refusing to do any swap with Iran, Iran may claim that it is acting to treat cancer patients and that its hands are tied.

    In general, what are we going to do if any of the other states with indigenous uranium enrichment (i.e. Egypt, Argentina, etc) suddenly become regimes we don’t like? Sanction and bomb them all?

  13. Andy (History)

    The Iranian press reported that Salehi stated a single 164 centrifuge cascade has already been set up at the pilot plant for the 20% enrichment work.

  14. Mam Dali

    Ben, I have the same question. The Iranians have been very clear that they will swap in stages. That was rejected. I don’t understand why unless the goal was to hold these amounts hostage (i.e. add a year to ‘negotiations’). The current policy is back-firing and anyone can see that. I’m not sure why this policy then persists? It doesn’t make any sense.

    Mam Dali

  15. pedram

    KIA if you know what I mean

  16. nick (History)

    Unfortunately, main stream US media outlets have not explained the staged U-235 swap proposal proposed by Mottaki of Iran. It is a grave mistake not to follow through with this option and continue the threat of sanctions at this juncture.

  17. Paul Stokes (History)

    It is implicit in your discussion that Iran is using its need for the 20% enriched uranium as a ruse, and is really planning to continue enriching to 90%.

    It would be much more productive to assume that the 20% enriched uranium is needed for their medical isotopes reactor, and to negotiate with them on something like the IAEA Additional Protocol to provide adequate inspection to assure that the 20% enriched uranium was not being diverted to a weapons program.

  18. spaceman africa

    Khamenei is quoted as saying on Monday of this week that: “The Iranian nation, with its unity and God’s grace, will punch the arrogance (Western powers) on the 22nd of Bahman (February 11) in a way that will leave them stunned.”

    I predict they announce a date and possibly conditions for their withdrawl from the NPT. I do not think this is a scenario that has been discussed here.

  19. bts (History)

    Don’t listen to everything David Albright says. As U235 goes up it becomes increasingly more difficult to find the unwanted isotopes, otherwise you can reach 100% purity with little effort.

    Even based on Albright’s numbers there is not much difference. He says “20% LEU” is 15% to 20% closer to bomb. That’s what most people would assume anyway.

    The alarming part is that he says “3.5% LEU” is 70% close to bomb, and “20% LEU” is 85% to 90% close to bomb. If this is accurate it means Iran was already 70% close to bomb, so 20% LEU is not a big leap forward.

    By the way, the international community has ordered Iran to shut down the research reactor completely:

    The international community says Iran must not make medical isotopes. Iran must not make fuel. Iran must not operate the US-made reactor because after 50 years of operation it turns out to be a serious proliferation risk. So there was never any intention to give 20% LEU to Iran, that’s why the international community refused to discuss details of the swap.

  20. George William Herbert (History)

    It is my impression that the staged swap was ruled out by higher ups within Iran, more recently than it was made.

  21. anon

    “act mad enough to be dangerous; then be conciliatory in exchange for cash.”

  22. Arnold Evans (History)

    Mam Dali:

    I don’t understand why unless the goal was to hold these amounts hostage (i.e. add a year to ‘negotiations’).

    It looks to me like you understand why.

    More specifically, the US thinks of the fuel as leverage to force Iran to suspend enrichment and actually giving Iran the fuel would cause it to lose that leverage.

  23. Andrew

    The U.S. needs to imagine what kind of Iran it wants to see and the best way to get it there.


    Option 1: An Iran which gets bombed and gives up on its nuclear program does not seem likely as Iran would simply bury its program underground, literally, even further, and be given ane excuse to cease cooperation with international bodies.

    Option 2: An Iran which ships all of its LEU out of the country after sanctions and a ‘green’ revolution does not seem likely given the effect of past sanctions and the support which opposition have stated for Iran’s nuclear program.

    Option 3: An Iran with indigenous uranium enrichment (like nations such as Egpyt, Argentina, Japan, etc) would have the support of all the factions in Iran and likely be where Iran is headed, while the international community could call for increased inspection rights given past concerns. ‘Red lines’ that Iran could not cross could be identified and developed. Such a position would give Western states a leg to stand on if Iran could be shown to be doing something demonstrably wrong. I believe Richard N. Haass, President of the Council on Foreign Relations, and Jim Walsh, a research associate in the Massachusetts Institute of Technology’s Security Studies Program, have advocated for similar approaches.

    Option 4: Another option would be to ‘acquiesce’ to Iranian development, and develop a ‘containment’ strategy. This type of option offers its own advantages and disadvantages.


    A longer term perspective of what exactly we are trying to accomplish needs to be stated, and how the Iranians are likely to react to us needs to be adopted in to any formulation. I also think it would be very useful to start thinking of Iran as ‘Country X’, as universalism would help us avoid double-standards and prepare for the next crisis instead of reacting to the current one.

    Are we convinced that the current path will get Iran to renounce its nuclear program, if that is our goal? Are we convinced that such a model will prove to be effective in detering others in to the future?

    Getting Iran to remove LEU in batches and increase transparency seems much more effective than daring Iran to increase its enrichment.

  24. tomj (History)

    The example is totally incorrect: much more work goes into higher enrichment than into the low level enrichment.

    A better analogy is getting a stain out of clothing. (The stain being the U-235) It is easy to remove the bulk of the stain, but as the number of stain molecules goes down, they are harder to remove from the remaining bulk material.

    Going from 7:9 to 7:8 will be very difficult.

    The separation potential is so small that they must employ a cascade of spinning machines.

    In addition, the operational problems will compound themselves as you are forced to deal with increasingly radioactive material, I would guess this issue would be equivalent to dealing with the scale up of explosives production to commercial scale from pilot-plant production. New issues will emerge that can’t be easily predicted, so you can’t move quickly.

  25. Jeffrey Lewis (History)

    No, the example is not “totally incorrect.” It is, however, a simplification. (Sorry that I didn’t see your example, Yale.)

    That isotopic separation is evidently not like a stain (at least not over the ranges we are discussing and for our purposes). This should be evident from the separative work requirements, which drop as the material passes from one stage to another.

    That is a counter-intuitive phenomenon for most folks, that the first pass through the cascade results in 5 percent, the second pass in 20 percent. (I am rounding.)

    Agazedah explained this to some very confused Iranians:

    Now, if we use this 3.5 percent product as fuel and inject it into this cycle the output will be 20 percent. If we inject this 20 percent into this cycle, the output will be 60 percent. If we inject this 60 percent, the output will be 90 percent. This is the stage where it can be used for two purposes. Please bear in mind that according to the Safeguards Agreement, any country which signs a safeguards agreement is making a commitment and a pledge to the IAEA that it will not enrich uranium to a grade higher than 3.5 percent. Based on its regulations and the agreement it has with that specific country, the IAEA carries out the necessary supervision so that this 3.5 percent doesnʹt change.

    (I am actually suprised it takes a fourth pass, but perhaps I am missing something about tails manipulation. Little help!)

    It is true, as you say, that there are operational problems at higher levels of enrichment, but that is not what the post was about — it was about the conceptual basis for separative work units.

    The policy implication is that this was not a showstopper for Pakistan and, although there may be some bumps along the way, it should not counted on to be a showstopper for the Iranians for relatively small amount of HEU.

  26. George William Herbert (History)

    It’s true that the separative effort expended to go up in percentage increases as the enrichment increases. But the key is understanding where, in the total picture, most of the total SWU effort are expended.

    Most of the total SWU effort expended is taking the bulk of the U-238 out of the feed. By the time you reach reactor grade LEU, you’ve done that.

    It takes more centrifuge-hours per kilogram to keep going up and up and up, but for a long time the kilograms processed drops very quickly as you extract the bulk of the U-238.

    There’s a reason that weapons tend to use 80% (Little Boy), 90%, 93-94% rather than 98% or 99% enriched HEU – that reason being as stated, the effort to keep going up the enrichment ladder eventually becomes overwhelming again, and the weapons design benefits fall rapidly past 90%.

  27. bts (History)

    The example is correct for start but the stuff in tails makes a big difference. In 860 comparisons you can accidentally remove 2 U-235 isotopes, so the number U-235 is reduced from 7 to 5. Even more work will be needed to enrich to 5%.

    Lets assume in every 3 comparison between U-235 and U-238 there is 2 success + 1 failure. There are 1000 particles and 1000 comparisons are made. I figure in each pass, 1/3 of U-235 + 2/3 of U-238 ends up in tails.

    The original stash is reduced by about 2/3 in each pass, so less work is need for next pass. When it’s at 50% mark the stash gets reduced by 50% in each pass. When it’s 99% LEU the stash is reduced by 1/3 in each pass, U-235 is reduced by the same amount so it gets harder to reach purity.

    I don’t really know what I am talking about here, don’t pay too much attention if that makes no sense at all.

  28. Jeffrey Lewis (History)

    These are wonderful comments, but they illustrate why I included the disclaimer:

    “If you are the sort of person who just said, “Hey! What about Uranium 234?” or other nitpicks this post is not aimed at you.”

  29. bts (History)

    Knit picking is important! Look at Ahmadinejad, he talks about making 20 enrichment plants and sending man to space, he doesn’t pay attention to small detail like “where to get the money for it”.

    I am sure uranium can be enriched to whatever level but it’s still a pain. There is no major break through here, it’s mostly Iranian and US officials who say this is a big deal.

  30. bts (History)

    I spoke too soon. Iran has already produced 20% enriched uranium. And it only took 2 days. LOL

    “I want to announce with a loud voice here that the first package of 20 percent fuel was produced and provided to the scientists,” Ahmadinejad said.

  31. Lars (History)

    This was a great post. Thanks for clarifying this.
    So, I was wondering if someone could clarify something else for me, since I follow Iranian politics fairly closely but don’t know too much about this proliferation stuff:
    Plutonium is produced when you put reactor fuel into your reactor. And plutonium is what the real nuclear powers use, since you can’t trigger a fusion hydrogen bomb with a weapon, but you can with a weapon made with plutonium (this at any rate is what I recall from reading The Secret That Exploded by Howard Morland many years ago). But if that’s the case then it seems like you’d want to forego making highly enriched uranium for a bomb and instead just produce low-enriched uranium as reactor fuel, since then you could make plutonium and then the kind of nuke that people are REALLY afraid of. But if that’s the case, then why is everyone in a lather over highly enriched uranium when the real threat comes from the reactor fuel that could be used to ultimately produce plutonium?
    What am I missing or not understanding here?

  32. Azr@el (History)


    There are three orders of fusion reactions employed in modern thermonuclear weapon.

    a) Neutron triggering; a neutron pulse caused by compact deuterium or tritium accelerators to set off the fission cascade at the beginning of the interval of high compression of the fissile core(Uranium233/235 or Plutonium239).

    b) Boosting; a small Lithium/Deuterium/Tritium burn initiated by a partial burn of the core to flood the compressed fission assembly(Uranium233/235 or Plutonium239) and ensure maximum burn of the fission fuel before dis-assembly of the core.

    c) Staged fusion; using the immense power of fission device(Uranium233/235 or Plutonium239) to drive the compression of a boosted device thus reducing the need for fissile content of the secondary and allowing for an increase in the mass of the fusion layer.

    In all three cases the fusion reaction is fairly blind to the species of fissile fuel. All three will work equally well for the most part. Thus if the IRI was aiming for a thermonuclear staged device then there are no hurdles to them using enriched uranium 80%> initiated by a compact neutron source, boosted with deuterium/tritium/lithium and driving a secondary with a enriched uranium spark plug. Reactor bred Plutonium239 is not a requisite to swing in the big leagues.

  33. George William Herbert (History)

    Lars –

    You seem to be somewhat confused. Plenty of hydrogen bombs are triggered by all – uranium fission bomb “primaries” – the US B53 9 megaton bomb being a well known example.

    I don’t think Howard Morland said what you think he said; if so, he was wrong on that point.

    Plutonium needs much less material to reach critical mass and explode – 10.5 kilograms vs about 50 kilograms for bare, unreflected spheres of normal density metal. So you can make much smaller bombs out of plutonium than uranium. But by “much smaller”, we mean a factor of 3-4 more fissile material, and about 3-4 in volume at the low end of really small bombs – which translates into diameter being 44-58% larger.

    Without going into gory design details and getting the posting classified by accident, the smallest all plutonium US bombs were around 10 inches in diameter. So a minimum sized HEU bomb would be (very very roughly) 14 to 16 inches in diameter. Plenty deliverable with a cruise missile or ICBM, plenty usable as the ignitor in compact hydrogen bombs.

  34. John Schilling (History)

    First, Howard Morland is almost certainly wrong – you can build a perfectly good fusion hydrogen bomb with nothing but uranium and lithium deuteride. Perhaps not so nice and compact as a W-80, but probably something that will fit on a Scud derivative and scorch a city. Well, people who really understand this stuff can probably do that. Amateurs and entry-level pros like the Iranians will probably steer clear of fusion bombs for a generation or so even if they do have plutonium.

    Second, anybody who isn’t REALLY afraid of atomic bombs, the quaint old-fashioned fission kind, needs a serious reality check. You can make a fission bomb of up to half a megaton or so using nothing but uranium. And if all you’ve got is a pathetic, tiny, insignificant little Hiroshima-style gun assembly bomb, well, setting one of those off over old Jerusalem or downtown Manhattan or wherever would probably kill a few hundred thousand people and horribly wound as many more.

    People are really afraid of this. People are also really afraid of the reaction of the victim’s surviving relatives and for that matter uninvolved innocent bystanders.

    Uranium enrichment using gas centrifuges, seems to be the easiest way to make nuclear weapons from scratch. If such weapons are used, very few people are going to be saying, “good thing it was only an atomic bomb; we sure dodged the bullet on that one”. Even the threat of their use, is really scary. Which, of course, is the point.

  35. Anon.

    Lars:“What am I missing or not understanding here?”

    That uranium bombs certainly can be used to trigger thermonuclear devices. Or antimatter bombs for that matter, the secondary’s tamper doesn’t know or care what the radiation heating it came from.