Working at a think tank with lots of proliferation specialists offers certain advantages, especially the casual conversations.
It turns out Ferenc Dalnoki-Veress, Miles Pomper and I all share a pet-peeve: When people incorrectly convert amounts of plutonium and especially highly enriched uranium from kilograms into some arbitrary number of nuclear weapons.
There is a right way and a wrong way to do this.
Significantly Wrong on Significant Quantities
or
A cautionary tale by way of a slightly technical rant about the proper use of certain concepts pertaining to the conversion of actual units of measure into hypothetical units of destruction and the pitfalls encountered therein
by
Ferenc Dalnoki-Veress, Jeffrey Lewis and Miles Pomper
James Martin Center for Nonproliferation Studies
If you are a casual reader, you may be forgiven for thinking that the official unit of measure for plutonium and highly enriched uranium is a “bomb’s worth” – so often is the term employed by pundits, wonks and other forms of ink-stained wretches attempting to describe how much the United States, Russia, Iran or North Korea has of this or that. Of course, the proper measure is the kilogram. Even if you are an American and stubborn, you should still go metric.
Of course, kilograms are not the least bit exciting, certainly not when compared with nuclear weapons. Tell an audience that the US and Russia, though the “Megatons to Megawatts” program, have downblended more than 400 metric tons of weapon-grade HEU and you may receive some polite applause. Tell them it is enough for thousands of nuclear weapons and you have their attention.
With that attention, however, comes a responsibility to be accurate. All too often, pundits make serious mistakes when converting a real measurement like kilograms into the more evocative “bombs worth.” Few pundits make clear their assumptions on how much nuclear material is needed for a weapon in which context. Invariably, the result is to exaggerate either the danger faced or averted. The problem is especially bad when the topic is an amount of highly enriched uranium.
For example, this week’s Economist cover story on Iran paraphrases an Israeli official, Director of Military Intelligence Major-General Aviv Kochavi, explaining that “if further enriched to 90 percent,” Iran’s stockpile of 100 kg of 20percent HEU “would be enough for up to four nuclear weapons.” Similarly, a recent White House blog post on the upcoming Nuclear Security Summit, for example, notes that since the last summit in April 2010, the United States has removed more than 400 kg of highly enriched uranium removed from over 10 countries – “enough for 16 nuclear bombs.” These calculations are just plain wrong. Perhaps it doesn’t matter to the casual reader, but it matters to us. And, if you read this blog, it should matter to you, too.
The obvious place to begin our discussion is the IAEA safeguards concept of a “significant quantity” of HEU. Readers certainly know that current IAEA safeguards standards define a significant quantity as “the approximate amount of nuclear material for which the possibility of manufacturing a nuclear explosive device cannot be excluded. Significant quantities take into account unavoidable losses due to conversion and manufacturing processes and should not be confused with critical masses.” And the current IAEA glossary defines that threshold amount for HEU as 25kg of U-235 in HEU. (U235 is the stuff that goes boom.)
Now, “significant quantity” is not a perfect measure. Tom Cochran and Chris Paine have long and convincingly argued that the IAEA should lower the definition of a significant quantity, at least in some contexts. State actors using implosion designs can and do build devices with much smaller amounts of material, even accounting for material lost along the way. A “bomb’s worth” of HEU for China is much less than a “bomb’s worth” for a terrorist group willing to settle for a gun-type device. Still, for all its flaws, the Significant Quantity is the standard agreed to by the international community as the basis for safeguards. As such, it serves as a handy reference point without respect to the design expertise of a particular state or group. Moreover, a White House official will find it much easier to clear the IAEA’s concept of a “bomb’s worth” than asking Z Division for an estimate. So, for all its flaws, the SQ remains useful. Now, would it be so difficult to use the concept correctly?
Think about it. Twenty-five kg of U235 is not the same thing as 25 kg of “highly enriched” uranium – highly enriched can mean anything from 20 percent U-235 to well over 90 percent U-235. Let’s revisit the comment of the Israeli intelligence official. He says the material in Iran is sufficient for four nuclear weapons. The Economist does not detail his assumptions, but if his estimates are based on IAEA significant quantities, then he is way off base. One hundred kilograms of 20 percent HEU contains 20 kg of U-235 — less than one significant quantity. Similarly, the recent White House blog post also overlooks the fact that some of material is former Soviet HEU with enrichment levels of 30 or 40 percent. For example, Soviet-origin HEU fuel removed from the WWR-M reactor in Ukraine is at a 36 percent enrichment level. Using 36 percent as an estimate of the average level of enrichment, the material might contain as little 144 kilograms of U235. (Of course, some of the material may be of a higher enrichment.)
Moreover, in both cases, the material would need to be further enriched to make a nuclear weapon. If you grandmother had balls, the old saying goes, she’d have been your grandfather. A better measure of how many “bomb’s worth” of material is how many bombs one might make from the material at its existing level of enrichment – after all, one can theoretically build a bomb with material of lesser enrichment. This would involve large amounts of fissile material in a ginormous device, but it is possible. As Frank von Hippel and Alexander Glaser demonstrate, the critical mass needed for a bomb is roughly inversely proportional to enrichment level. So at 20 percent U235, the critical mass is about 400 kilograms of HEU assuming a 5-cm-thick beryllium neutron reflector. At 90 percent U235, the similarly reflected critical mass is 28 kg.*
As Matt Bunn likes to point out, the IAEA actually has a rule of thumb to estimate a “bomb’s worth” of material across different enrichment values: the notion of the effective kilogram. In the case of uranium with an enrichment of 0.01 (1%) or greater, the IAEA calculates the number of effective kilograms in the material as “its weight in kilograms multiplied by the square of its enrichment.” Using this method, Iran’s stockpile of 100 kg of 20 percent HEU is just 4 effective kilograms – well below the 25 kilogram “significant quantity” threshold. Similarly, the 400 kilograms of “HEU” removed since the last nuclear security summit, again assuming an average enrichment level of 36 percent, could shrink to a mere 52 effective kilograms – just enough for two significant quantities.
This is the correct way to do these calculations. It is fine if someone wants to point out why, in a particular context, significant quantities and effective kilograms don’t capture an important feature of the real world. (Your proverbial Iranian grandmother, for example, could grow balls and further enrich her stockpile of uranium.) That’s the sort of expert discussion that makes this field interesting and fun, especially if one can make grandmother jokes. But understanding how these measures relate to the real world is different from simply not understanding them at all. There is a right and wrong way to use these measures. If you don’t know the difference, stick to kilograms.
*What’s a technical gripe without a footnote? Twenty-eight kilograms of 90 percent HEU just happens to be almost exactly one Significant Quantity. (Such material would contain 25 kilograms of U235.) Although one would expect the reflected critical mass of weapons-grade highly enriched uranium to be very similar to a Significant Quantity, the fact that they are identical is just a coincidence. The IAEA Significant Quantity is not a critical mass calculation. A panel of experts recommended twenty-five kilograms to the IAEA. We know they included an allowance for material lost to processing, but the three of us have not seen further discussion about device design. We are comparing apples to, well, applesauce. But it’s still nice the numbers are copacetic.
Well that was overdue…nicely phrased, and with the usual ACW snark no less.
Nice piece, have always found it somewhat reminiscent of the standard law enforcement line of [insert mass/volume] of [insert drug] with a street value of [insert arbitrary but impressive sounding cash value].
However if you’re going to be that much of a stickler for detail, would it be unsporting to suggest that the word tonnes be used instead of the rather more clumsy ‘metric tons’. If only to avoid confusion with short tons, long tons, beef wellingtons etc…
Not unsporting at all.
There are always little choices that one fusses over when writing something like this. In this case, even the BIPM notes that “In English speaking countries this unit [tonne] is usually called ‘metric ton’.” I thought about it for a bit, but concluded that “metric ton” would cause less confusion and had the reluctant blessing of BIPM. (On a related note, the little essay on why the unit of measure is a kilogram, and not gram, is unintentionally hilarious and made my day.)
I also fussed over the IAEA’s use of “weight” rather than “mass” in its definition of an effective kilogram. In the end, I decided to leave the quote as is, rather than complaining about Newtons and whatnot.
In the end I let it go. Excessive attention to such details has been known to result in unpleasant dinner invitations.
This ordinary house brick here contains one metric gajillion bombs-worth of material – assuming I find a way to covert its mass to energy, o’course.
Similar issue with chem: “this much toxin is enough to kill the entire human population!” Sure, if you teleport a drop directly into every person’s heart, it is. My brick can also kill the entire human population – through blunt force trauma, if they all line up and my arm doesn’t get too tired.
Man, I should put this brick on eBay.
The big issue in the reporting, for me as a layman, is the huge yawning gulf of a difference between almost any quantity of 20% HEU and almost any quantity of 90%. My own interest in “lost” or poorly-stored/guarded material is in the context of non-state actors potentially being irrational dickheads. I realise the context is different for state-level arms control, but from a CT point of view there is a certain threshold beyond which I care a great deal more.
If some nut-job terrorist organisation wanted to give me their 30kg of 90% HEU in exchange for my 1000kg of 20%, I’d take that trade in a second. The 1000kg is comparatively useless in the hands of folk without access to enrichment or, let’s face it, the means to move, machine, hide, deploy or otherwise do anything useful with 1000kg of radioactive material. Even dirty-bomb wise, they’d do better going for a chemical attack with 1000kg of about a million other substances.
The 30kg of 90%, though, is something even my grandmother (whose current status vis-a-vis balls is, happily, purely speculative at this time) could use to nuke a city.
Any time there’s a report or rumour of unaccounted-for 90% (or similar) HEU, this is something I care about, even at quantities much less that 1SQ. There are plenty of people who have a very great interest in collecting enough of those sub-SQ quantities to get into the 30kg range. Sure, you don’t *want* 1000kg of 20% running around, but it’s not in the same league as even a small quantity of 90%.
Poor reporting means I end up with very little usable information without having to dig, and there’s no reason for it. Is page space really so tight they can’t give the actual level of enrichment? Three characters and 2 spaces? C’mon.
I would not describe a “huge yawning gulf” between 20 percent and 90 percent enriched uranium — 20 percent is something like 9/10ths of the way to 90 percent.
To understand why, recall that “enrichment” is a misleading term — the method is separation or the removal of unwanted material, nothing is added. The simple example that I like starts with 1000 atoms of uranium. Only seven of them will be the fissile isotope Uranium 235. The rest are useless Uranium 238. (We can ignore the U234 in this example.)
To make typical reactor fuel, Iran or any other country 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 need only to remove 27 of the remaining 35 atoms, leading a ratio of 7:8 (~90 percent).
Which takes less work? Moving 965 atoms or 27 atoms? This simplification does omit the important concept of tails and some other details at very high levels of enrichment (say 95 percent and beyond). But those details aren’t enough to alter the nice smooth curve by Drell et al (on page 59) that shows cumulative separative work units to various levels of enrichment. Notice the sharp “knee bend” around 20 percent.
There is a reason that Iran’s enrichment to 20 percent is provoking a real crisis.
Right – the “SWU remaining” to weapons-actual-usable (whether that’s 60%, 75%, 80%, 90%, 93.5%, whatever) from 20% is much lower than from 5%, and either one is much lower than the SWU to go from natural U to 5%.
While it might take large, visible factories full of centrifuges to do the enrichment to 5% or 20%, taking 200 kg of 20% and getting 20 kg of 90% (plus tails) could be done in a small lab with a few calutrons or small number of centrifuges, etc. Uranium enrichment is a massive industrial undertaking, and weaponization from natural U is nearly unavoidably visible from a distance, but weaponization from LEU is much more compact and hideable. Anyone who could credibly build enrichment devices could probably hide a program to take a ton of LEU to a bomb’s worth of weapons-usable HEU in a year.
By the way, Magpie, 30 kg of 90% will take some effort to nuke a city. The public information on gun-type bombs is that Little Boy used 62 kg of around 83% enrichment HEU, and the South African gun weapon used 55 kg of around 93% enrichment HEU. Criticality calculators show plenty of reflector configurations for 30 kg of 90% which give you supercritical systems, but for actual useful explosions the time value of how long neutrons take to get reflected back in matters, and systems relying too much on reflection “melt down” too quickly rather than exponentially exploding effectively. Alpha matters.
Imploding it is another matter entirely.
Which takes less work?
Depends on whether you have the ability to move any atoms anywhere. That’s my point: for ***non-state actors***, all that matters is the ability to use an existing substance. They can’t enrich, so it’s not work they can do, “less” or otherwise.
They can use ~90%-enriched.
They can’t really use 20%-enriched.
The difference between can and can’t *is* huge and yawning.
My point re: poor reporting is as an addition to your own (wonky?) interest. That is, there are plenty of people with specific interests in specific areas of the issue, and that can be addressed simply by reporters actually including that detail of enrichment.
Re: 30kg, I’m hand-waving a bit. That’s about the area that – again – non-state actors could have a reasonable shot at producing something viable (given access to decent personnel).
Again, they CAN’T enrich, not going to happen. They COULD produce a device of some sophistication, though, given that the hard part (enrichment) is done.
Sorry, just to further clarify, as I said in the first post, but should perhaps have made clearer:
“I realise the context is different for state-level arms control, but from a CT point of view there is a certain threshold beyond which I care a great deal more”.
…that is, quite apart from your own interest in arms control, poor reporting annoys we CT semi-wonks just as much, for slightly different reasons. We don’t care about the ability to perform enrichment. We treat that as something that happens on a higher plane, or as a natural process, like continental drift – we’re dimly aware it happens, but hey, whaddayagunna do?
We’re interested in the location, security, and quantity of existing materials that are, *in their present form*, capable of being turned into a viable device. Tracking it, protecting it, getting rid of it, working out who you’d need and what you’d need to turn it into something useful. Working out, in rough terms, how much of it might be around, and how hard it might be to get more.
Sorry to have confused.
Jeffrey: Isn’t a ratio of 7:8 “only” 47% HEU? To get to 90% you need to remove 34 of the remaining 35 atoms, down to a ratio of 7:1.
You’re right, I meant 7 of 8. I was in slow math in elementary school.
Jeffrey,
I must object to your claim that the curve in the Drell et al. report (or an equivalent calculation from standard formulas) has a “sharp knee around 20 percent” which explains why “Iran’s enrichment to 20 percent is provoking a real crisis.”
As curve in the report shows, enrichment to 5% is already most of the way there. The “bend” in the curve is not sharp around 20% but extends rather from about 5% to 25%.
You can say that, politically, enrichment to 20% is significant because the IAEA has defined it to be.
But while it brings Iran somewhat closer to a sig quant of bomb-grade HEU, they are already pretty close with a large stockpile of 5%, plus the centrifuge cascades to take it the rest of the way. The difference between 5% and 20% as the starting material is quantitative, not qualitative, and depends entirely on the capacity of the centrifuges.
As I understand it, Iran is permitted to possess 19.75% U-235 anyway. Are they known to be making more than can be justified by the needs of the TRR?
In no way does this justify manufacturing a “crisis” which consists of overt Israeli threats to launch a major war, in the context of escalating covert-ops violence by all sides — and then pretending that the blame falls solely on Iran for enriching some U to the magic number (just under) 20%.
Mark:
Stating that the knee “extends rather from about 5% to 25%” is not materially different from suggesting the bend occurs at about 20 percent.
The point is that after 20 percent, relatively little work is left to reach 90 percent.
I am sorry to say this, but I find your intervention a bit tendentious and I object strongly to your characterization of twenty percent as some sort of politically-driven and arbitrary threshold.
As one of the rare individuals in the United States who doesn’t think zero centrifuges is a necessary or even desirable outcome, I am still extraordinarily concerned about enrichment levels above about 5 percent.
Jeffrey,
Sorry if you found my comment tendentious, but of course I was saying that I found your comment about Iran being to blame for “provoking a real crisis” to be for you uncharacteristically tendentious in your interpretation of technical facts.
I would suggest that the really critically significant quantity, in terms of military and political implications, is not so much the level of enrichment, if that is still below 80% or so, or even the quantity of uranium enriched to that level, but rather the quantity of time it would take Iran to go from breakout to a bomb, or in this limited context, to a bomb’s worth. That depends not only on the available bomb technology (implosion vs. gun, reflectors), as you point out, but also on the total capacity (SWU rate) of Iran’s centrifuges.
More centrifuges (apparently only about twice as many) and 5% starting material are as good as fewer centrifuges starting with 20%. The difference can also be made up by willingness to accept a high tails assay.
Since Iran is adding centrifuges all the time (and presumably fixing the ones damaged by stuxnet), and their number remains uncapped, it is difficult to see how the enrichment of some U to 20% by itself defines “a real crisis.”
Re “Iran is permitted to possess 19.75% U-235 anyway” – as I understand, Iran (or anyone else, for that matter) _is permitted_ to possess 100% U-235 as long as it is safeguarded.
Two brief points:
1.) The SQ is a convenient unclassified conversion number. While DOE’s RDD-7 can provide a convenient conversion ratio for plutonium (Section V.A2.u), no such measure exists for HEU, that I know of.
2.) This little chart may help graphically make your point between enrichment, mass, and work (SWUs) necessary — and how getting to 20% takes you 2/3 of the way to 90%, as far as SWUs are concerned. You just need a lot of stuff to start with.
Aaack, no edit function. Meant to provide this link:
http://www.anengineerindc.com/2012/02/iranian-uranium.html
I don’t know why we persist talking about guns and Hiroshima and Nagasaki as examples of what one might do, when we know the advantages of implosion compression in requiring much smaller quantities of fissile materials and increased yield efficiency.
The “public” numbers are as follows:
1) 94%Pu239 pit mass is 2.0 kg (Orion Pulse Unit Design, Report GA-5009 vol. iii, pg. 104-106. These pulse units were Ted Taylor’s optimized implosion designs of 1960’s, yielding from sub-kiloton to about 15 kT.
2) For 93% U-235 HEU, from LA-10860MS (1967, Table 28, pp.96) the critical mass without compression can be 14 to 35 kg, depending on the reflector material.
Assuming a “run of the mill” compression factor of two, the mass goes down roughly by a factor of 4, down to about 6 kg to 9kg of HEU for an implosion design and sufficient “alpha” to produce a decent yield, particularly if one uses boosting and a modern neutron generator (5000 to 15000 n/ns injection rate.
In conclusion, the significant quantities today (2012) are roughly 2kg for 94% Pu239 or 6kg of 93% HEU.
Ara writes:
…
The Orion pulse units were an extremely odd bird in the fission device spectrum – absolute minimization of the fissile material for both cost reasons and to minimize he fallout problem. Conversely, other mass in the unit – such as extraordinarily large implosion assemblies by “modern standards” – was not detrimental at all, as reaction mass was needed for the unit to work properly and was in fact additionally supplied as inert plastic beyond the bomb’s explosive system residue.
Military utility of a fission device starts at the delivery mass one can put on a cargo transport plane, is significantly higher for what a medium bomber can carry, is higher for what a fighter can carry. For some users it may peak when it’s small enough for compact primaries in a two-stage system.
The Orion pulse units are at least one step up from ideal in that way of looking at it.
These also assume that the weaponizing party has a sophisticated engineering program, and high tolerance for failure or an ability to do a weapons proof test. The degree of compression one can reliably achieve will increase as ones design concepts and engineering and materials improve.
If one can’t do a hot nuclear yield test, one has to build in more design margins or assume a high failure probability. Doing covert tests has eluded proliferators so far.
There’s also reason to believe that the state of the art of proliferator implosion systems – and likely non-state actor implosion systems – has been overestimated in the field. The writeup and implications haven’t been published yet in an organized fashion.
Doing covert tests has eluded proliferators so far.
Other than, perhaps, the Vela event?
George,(GWH)
I am talking about SQ in the hands of backward, primitive entities that can put together a crude implosion system a la Pakistan (242 detonators copy of CHIC-4 Chinese gift)overloaded with HE….Not exactly state of the art, but far superior to the gun that everybody’s talking about..
But Iran is not Pakistan, launched a bunch of earth sattelites, and they certainly can make a reasonable (not perfect) replica of R-265…and we need to keep that in mind.
Fortunately, for the moment, they don’t have a Ted Taylor around…
In regards to converting 20% LEU to 90% HEU, what is the impact of converting ~20% LEU into reactor fuel? Would Iran’s stated intention to use its LEU to power its research reactor alter the time lines for converting it into HEU? If so, by how much? This is one aspect of Iran’s nuclear program that I have not seen discussed and it would be interesting to understand the implications of Iran’s intention to create fuel rods should it actually do so.
It’s not just a calculation error that’s the problem here – the problem is about a fundamental bad faith presentation of information intended to inflame and scaremonger. Like Cheryl Rofer pointed out, equating uranium atoms with nuclear weapons is just bad reporting, even if the math works out.
http://whirledview.typepad.com/whirledview/2009/02/whoohoo-atoms-of-fissionable-material-everywhere.html
Enriching to 20% under safeguards (or whatever number you like) is not cause to provoke a crisis — or, to be clear, the crisis is self-created and is political not substantive or legal.
All the nuclear material is safeguarded.
The IAEA says that 20% is anyway, “a fully adequate isotopic barrier to weapons usability of 235U” (quote from IAEA)
20% is also considered L. E.U. as in Low by the IAEA and ASTM.
There is a cause for concern in Iran so doing but Iran may itself be concerned that its fuel supply will be bombed and so wants excess on hand.
It’s funny to think that one can threaten Iran’s nuclear facilities and that Iran should continue with business as usual.
In any case, as Dan Joyner as argued “concerns” of the IAEA are not material: what is material is legal breaches of the CSA.
To quote:
===
“Since the IAEA is acting outside of its legal authority in this section of the [Nov] report, it does not have a legal standard to apply to its conclusions regarding possible nuclear weapons related activities not involving fissile material.
Throughout the report, the director general expresses “concern” about the information being presented, and requests “clarification” from Iran in order to address these concerns. However, since there is no treaty language in Iran’s, or any other state’s, safeguards agreement that deals with non-fissile-material activities related to nuclear weapons, there is no prohibitive or regulatory standard that the director general can point to against which to make a conclusion of compliance or non-compliance.
In short, as the ancient legal maxim states, there can be no illegality where there is no law. The IAEA is simply “concerned.”
===
So one can be concerned about Iran’s safeguarded 20% work until one gets blue in the face — or a job at State or the IAEA — but there is little one can do about it legally unless one indulges in politicizing the IAEA and UNSC.
Is HEU really used to refer to u-235 quantities as low as 20%? I would have figured that seeing as the effort required to enrich from 20% to 90% is virtually nothing compared to the effort to get from 0.7% (natural) to 20%.
Perhaps it would be simpler if 20% were referred to as low-enriched uranium and 90% – as those are the significant benchmarks – as HEU.
As an aside how many significant quantities of HEU are missing from US inventories?
I’ve heard numbers like 14,000 kg of US HEU are unaccounted for abroad?
you are incorrect — it is actually 16,000 kg of US HEU that is missing. 16 Tons of HEU…or is it tonnes?
http://motherjones.com/mojo/2011/09/usa-lost-tons-nuclear-weapon-uranium
I believe this could be a real threat to the peace and a possible cause to kick of Chapter 7 UNSC sanctions on the US?
Oh forget it!
That material is NOT missing. It’s transferred to other countries as a regular transaction and is under their safeguards obligations. It’s missing the same way one “loses” a ten dollar bill they use to pay a lunch. Sure, they don’t know where it is, but it’s someone else’s now.
Only way to call that material lost is if one are either totally unaware of how the safeguards obligations work or you are intentionally trying to be dramatic, the turth being damned.
I don’t think the GAO is being dramatic.
Evidently the GAO thinks that US-origin HEU is the responsibility of the US.
Which legal interpretation are you offering to say that it is the responsibility of Libya/Iran/other nations?
Picky, picky, picky.
When folks are trying to scare people out of their wits about the threat of imminent nuclear holocaust they need SCARY, not REALITY.
Besides, most folks have no clue about ‘nuclear’ and do not know the difference between a research reactor and a bomb.
So all the headline need read is ‘”UN atomic agency report notes rapidly expanding Iranian activity that could be used for arms,” for panic to set in. Never mind that that same UN Agency has been reporting calmly about the progress in plant expansion AND the composition of the output.
Your insistence upon reality just shows and pro-Iranian bias in line with the anti-Semitic Obama administrations efforts to aid the Iranian weapons program. ;-0
What a great post and string. As someone who tried hard to avoid major math, thanks to all who explained it so well. One crude observation that comes to mind is that grandmother’s growing balls, for whatever reason, which reason explained sooner is better than later. That is why the time factor (not to say “timeliness,” for fear of delving into theology) is so important for any effective verification regime when enrichment and reprocessing are involved. The explicators in Iran have done a good job of explaining this to the decision-makers. And that’s where the politics comes in, in a knot continuing to be tied more tightly, the better to frustrate a respected IAEA staff (including the DG), the better to jerk us all around.
Hello gang from a cold but sunny Cambridge, Ma. Thanks for citing my colleague Matthew Bunn. I thought it might be worth drawing your attention to one of the more egregious examples of miscalculation that Prof. Bunn often cites: the fanfare around the 10 tons of HEU in spent fuel from the shut down BN-350 fast neutron reactor at Aqtau in Kazakhstan. This spent fuel also contains approximately 3 tons of better-than-weapon-grade plutonium. The U.S. GTRI program, which has various initiatives to eliminate weapons-usable material around the globe, helped Kazakhstan move this material from the former reactor site on the Caspian Sea to a more secure location ain Kazakhstan, at a cost of hundreds of millions of dollars.
Statements from NNSA sometimes exaggerate the importance of this material, by failing to note that (a) most of the HEU had an initial enrichment in the 22-27% range, which was further reduced by fission of some of the U-235 in the reactor, making it quite difficult to use to make a nuclear bomb; and (b) both the HEU and the plutonium would require reprocessing to recover the material for use in weapons.
Most relevant to this discussion, however, is that NNSA statements have consistently exaggerated the number of weapons that could be made from this material, by using IAEA “significant quantity” figures incorrectly. In its press release announcing the completion of the shipment, for example, NNSA described the material as “775 weapons’ worth.” [See “NNSA Secures 775 Nuclear Weapons Worth of Weapons-Grade Nuclear Material From BN-350 Fast Reactor in Kazakhstan” DOE/NNSA press release (Washington D.C.,18 November 2010) http://nnsa.energy.gov/mediaroom/pressreleases/bn35011.18.10%5D. This figure came from dividing 10 tons of HEU by the IAEA significant quantity figure of 25 kilograms, to get 400 weapons from that material, and dividing the 3 tons of plutonium by the IAEA significant quantity figure for plutonium of 8 kilograms, arriving at 375 bombs’ worth from that source. But as you point out, the IAEA significant quantity figure is not 25 kilograms of HEU, but 25 kilograms of U-235 contained in HEU. Even when the material had not yet been irradiated, it likely had an average enrichment in the range of 25%, so a correct use of this approach would lead to an estimate of 100 bombs’ worth from this source, not 400.
Moreover, the approach itself is wrong, since more HEU is required to make a bomb at lower enrichments than at higher enrichments: here we might want to use the rule of thumb known as “effective kilograms,” which is the number of kilograms of HEU multiplied by the square of the enrichment (expressed as a decimal fraction): using that approach would cut the number of bombs’ worth by a further factor of four, to 25 bombs’ worth of HEU. Irradiation further reduced the average enrichment and probably led to much of it no longer being HEU, further reducing the number of bombs that could be made. While the IAEA significant quantity figures are higher than the amounts actually needed for a bomb, that effect is small by comparison to the large exaggeration from using the significant quantity methodology incorrectly. For the plutonium, NNSA’s calculation is correct, though getting at the plutonium would require reprocessing, which I think we can agree would be a very difficult task for non-state actors.
I hope that provides a nice example for you! As I think all of us know, GTRI does great work and deserves the support of the nonproliferation community—but it is telling that even their official statements fall prey to some of the shortfalls you describe in determining the danger posed by certain stocks of material.
All the best,
Eben Harrell
Research Associate
Managing the Atom
Harvard University