Why we swoon for SWU
Or, how I learned to stop worrying about centrifuges and measure SWU
Ahead of the March deadline for an Iran Deal, international negotiators are back at the table in Vienna to hash out the fine print for an Iranian nuclear program. Undoubtedly, limiting uranium-enriching technology is at the heart of the debate.
Yet, before we further add to the noise of the conversation or rattle off another opinion on the matter, let’s discuss some critical details behind uranium enrichment! Already, you wonks and aficionados reading this article are probably familiar with the fundamental builds of centrifuges and cascades — centrifuges are cylindrical machines that are organized into rows of cascades, all while each centrifuge manages three basic flows: feed, product, and tails.
Taking a closer look at an individual centrifuge, the ‘feed’ that is initially delivered into the cascades consists of raw uranium hexaflouride (UF6). At the start of the cascade, the feed to each centrifuge will have close to ‘natural enrichment,’ meaning out of all the uranium atoms in the supplied UF6, only about 0.711% will be U-235 at this stage.
The enrichment process begins when the feed is introduced to the centrifuge, which then spins at incredible speeds (you spin me right round baby right round, like an Iranian IR-1 baby right round round round). Simply, the centripetal force during the spin phase slightly separates the U-238 and U-235 gradually. Following this, the centrifuge then filters U-235 and U-238 atoms into two separate flows. Only then does the tails flow contain a lower enrichment than the feed, because the process has separated out the U-235 atoms. Lastly, the product flow out of a centrifuge contains a higher enrichment — around 0.75% of U-235.
Some of you out there might be thinking, “that’s a long way from HEU, and even further from weapon’s grade material, yo.” Well, you are correct.
Certainly, one centrifuge would take forever and a day (yes, a scientific measure of time) to produce uranium enriched at 90%, but crafty engineers have since built the centrifuges in series, where the product of one centrifuge goes to another, and to another, and to another—getting a bit more enriched at every level to create a cascade. But obviously, centrifuges are not similar. The builds of earlier Iranian IR-1 centrifuges weren’t nearly as efficient at enriching uranium as modern IR-4s, yet were still better than sticking a bunch of uranium in a bucket on a rope and swinging it around as if you were practicing for the Olympic discus throw (we tried that here at StrategicSwag before agreeing we were better off using our strength to mix margaritas). This efficiency, or the centrifuge’s separative work required to separate any amount of uranium has its own measurement that needs to be considered.
This separative process is measured in SWU, or the separative factor (separative work units) required to achieve the enriched product. But, why is SWU so important? Perhaps because a more defining characteristic of SWU is that it allows us to measure enrichment potential per kilogram of uranium over a year within a cascade, which is much more important than nitpicking over a number of centrifuges. Essentially, the SWU figure can inform us of the efficiency or amount of “separation” the cascade is actually doing.
A simple analogy: we like to think of it in terms of pressing orange juice. If we use a particularly crummy machine and four oranges to make one glass of juice, that’s okay. Or, we can go out and buy a more efficient press that only requires two oranges to yield the same product — et voilà! You’ve got a healthy glass of ‘enriched’ orange juice, sans pulp.
In a broader scope, it may be hard to conceptualize how enriching uranium can be compared to pressing orange juice, so let’s return to a real-life example. Indeed, we’ve already established that different types of centrifuges vary in their SWU ability. According to the Institute for Science and International Security (the good ISIS), an Iranian IR-1 centrifuge provides about .75 to .9 SWU per year. So if we have a facility, say, the StrategicSwag Enrichment Plant with 10,000 IR-1 centrifuges, the output of SWU would fall between 7,500 to 9,000 SWU. Now, let’s pretend we also have the StrategicSwag Pilot Enrichment Plant (we aren’t very creative in naming our plants) that uses the Iranian IR-2m and IR-4 centrifuges, containing approximately 4–5 SWU with a mere 2,000 centrifuges. The reality is that this facility could optimally provide somewhere between 8,000 to 10,000 SWUs per year — rivaling that of our first fictional enrichment plant (which is still better than our bucket and rope). Yet, many ignore the difference during negotiations, instead focusing only on the number of centrifuges in the cascade.
Therefore, drastic reduction or limitation of centrifuge numbers doesn’t always correlate to its separative work capabilities. Even if a reduction of 10,000 out of 12,000 total centrifuges in any given facility is negotiated, this only corresponds to a reduction of about 50% in total separative work capacity if we eliminate all of the IR-1 centrifuges. With that, hopefully this more realistic example illustrates that SWU needs to be considered when brought to the negotiation table.
For the Iran case, it comes as no surprise that the Iranians seek to continue a program of about 10,000 centrifuges, enabling it to whip up nice amounts of enriched uranium, though could also be quickly ramped up for non-peaceful purposes. Previously, the West encouraged deep cuts in centrifuges, traditionally pushing for numbers below 4,000 to hedge against diversion. But, the lesson here is that it really isn’t all about the numbers, it’s about the ability of separation.
Now, you can now go enjoy a tall glass of freshly pressed orange juice with your breakfast (please don’t consume enriched uranium for breakfast) and if you happen to find yourself chatting with colleagues or office cat about Iranian centrifuges and SWUs, hopefully you can speak confidently knowing StrategicSwag has your back!
Written by: Marianne Nari Fisher (@MarianneFisher) & Cervando A. Bañuelos II (@NuclearFarmboy)
