Since my last posting in January, the political situation in Israel has clarified: 2nd place finisher Likud and party leader Benjamin Netanyahu are now in charge. There have been some other changes as well. True to Obama’s campaign statements indicating he intended to limit/reduce/eliminate US ballistic missile defense programs, his administration is planning to eliminate US involvement in the joint US-Israeli Arrow 3 program, freeze U.S. land-based ABM deployment, and halt development of the Airborne Laser system, effectively killing it. In and of themselves, these decisions don’t affect the probable Israeli attack timetable because the ABM systems at issue would likely be deployed sometime after Iran crossed the nuclear threshold. I mention the Obama administration’s policy on ballistic missile defense here only because I may want to address it in a future post. The whole issue becomes somewhat less urgent if Israel defangs Iran. When we might expect the defanging to occur is the topic I’m going to discuss in the balance of this article.
In some of my earliest posts, I stated that Iran conceivably could have its first nuke (a one megaton thermonuke) in late 2007. As we know, that did not happen chiefly because the schedule for the expansion of the Natanz facility has “slipped to the right”. Recall that in Is there a “threshold”?, written 16 May 2007, I gave 6,000 to 8,000 as the largest number of operating centrifuges that would be tolerated before preemption became a necessity. I further stated that I would choose the lower number if it were my decision. I concluded by saying, “If the U.S. doesn’t attack by that point, it’s not going to attack. That’s when Israel will step up to the plate if it hasn’t already done so.”
We now have good information from a variety of sources that Iran will have 6,000 centrifuges installed and running by this summer and, most importantly, that Iran had, at the end of last year, produced enough U-235 at a 3.8% enrichment level to make at least one bomb. This latter fact is significant for reasons the popular media and the Obama administration have neglected to discuss.
To understand, let’s make use of this handy-dandy Separative Work Calculator from Urenco, the large uranium enrichment company based in Europe.
Using a neutron reflector (like beryllium) along with implosion, a much smaller amount of U-235 is needed for a critical mass than with a gun design and no neutron reflector. This is less than 20 kg, at high enrichment levels. At least this amount is known to be present in the uranium hexafluoride (UF6) “product” that Iran has stockpiled. Note that 3.8% U-235 cannot be fashioned into a bomb - at least 20% U-235 is needed. For practical reasons the enrichment level for “weapons grade” uranium is normally in the 90% range.
Iran has been enriching uranium for a couple of years to get to this point, adding centrifuge capacity as it has gone along. One might be inclined to believe that there is still a large margin of safety based upon the assumption that, since it took this long to get to 3.8%, it must require a significant amount of additional time/effort to get to 90%+. It doesn’t work that way – the relationship is logarithmic, not linear. To illustrate, let’s go to the calculator.
First a caveat: I’m not a weaponeer so I don’t know the exact amounts of U-235 that would be needed for a 1 megaton thermonuke (see here, which includes my argument why Iran will go directly to a one megaton device). I do know that, in addition to the U-235 for the trigger, there would also be U-235 needed to initiate the fusion reaction in the lithium-6 deuteride. I’m going to use 30 kg of 95% enriched U-235 as a good ballpark value for the total required.
The units in the calculator are metric tonnes (1,000 kg = 1 metric tonne), so we enter the values accordingly. For “Feed Assay” enter 0.711 (the percentage of U-235 in natural uranium); for “Product Assay” enter 95; for tails enter 0.3; for product enter 0.03 (30 kg = 0.03 metric tonnes). Hit the “Calculate” button. You get 6.14424 tSW. We could also express this as 6144.24 kg SW. The “Feed” is 6.91241 tonnes of natural uranium (the box immediately below is the amount of UF6 gas). Now, clear out the entries. For “Feed Assay” enter 0.711 again; for “Product Assay” enter 3.8; for tails enter 0.3. This time, enter the 6.91241 we got above in the “Feed” box for tU. Hit the “Calculate” button. We get 3.97689 tSW. (Incidentally, if you want to check this with another calculator, you can plug the appropriate values into the FAS Separative Work Calculator – the results will be the same to 4 decimal places).
What does all of this mean? To get from natural uranium to 95% enriched, bomb grade uranium (known as high enriched uranium – HEU) takes 6.14424 tSW. To take the same amount of natural uranium to 3.8% enrichment (known as low enriched uranium – LEU) requires 3.97689 tSW. Most of the work (around 64%) expended is to get from 0.711% to 3.8%. With 6,000 or so centrifuges running at max efficiency, it should take only a few weeks to go from 3.8% to 95%.
As I write this, Iran’s “public” nuclear program is subject to IAEA monitoring. So long as the centrifuge operation and LEU UF6 stockpile are compliant with IAEA safeguards, there is no “immediate” threat, or so the argument against preemption goes. Militating against this is the speed with which the first bomb can be produced – a few weeks if the Natanz facility is converted to bomb making. A few weeks would still offer time to issue an ultimatum and then attack. However, as Iranian enrichment capacity increases, along with the stock of LEU, the Israeli reaction time decreases. This is the basic calculus behind my original threshold argument. Israel cannot put itself into a position where it has little time and no margin for error when it attacks. Added to this problem is that if Iran has a small clandestine facility somewhere, it could begin to divert the LEU and initiate bomb production. It would be suicidal folly for the Israeli government to depend on the competence of the IAEA to catch Iranian diversion. No doubt Israel relies heavily on human intelligence (humint) on the Iranian program gathered by the Mossad – the best intel agency in the world. Still, the difficulty of monitoring what the Iranians are doing, even throwing in spy satellite sweeps, increases exponentially as the Iranian program expands.
It is the inexorable mathematics of Separative Work and the enrichment process that is driving Israeli planning now. Doubtless, the Israeli attack plan is fully developed and could be set in motion in 24 hours or less. The Israelis must immediately respond to any LEU diversion they detect, irrespective of what the IAEA or international community may say. As stated above, because of the reduced amount of Separative Work required to produce a bomb from the LEU, small, widely dispersed cascades that have thus far gone undetected could readily complete the process.
The lower limit of the threshold will be reached in at most a couple of months. Based upon this, my prediction is that Israel will attack some time this summer at the latest. Israel could potentially delay a short period beyond that, with an emphasis on the word “short”. The larger the stockpile grows, even if the number of centrifuges added at Natanz is halted, the bigger the problem for Israel.
The optimal time to attack in order to maximize the Israeli advantages is 48 hours either side of New Moon at 0200 local. The months of June through September are most likely with the odds of an Israeli preempt increasing through the summer.
Though I expect a night attack, I wouldn’t rule out a daylight strike. In fact, if I were the head of the IDF, I’d strongly advise it. One of the most important resources available to the Iranians is the experience their engineers and technicians have gained over the years the nuclear program has been in operation. It is, therefore, an unfortunate necessity to kill as many of these people as possible. A daylight attack will find the maximum number of engineers and technicians at work. In addition to the enrichment facility at Natanz and the heavy water plant at Arak (see Scenario 1), cruise missiles, possibly augmented by SpecOps, should hit research centers where scientists would be located. Like the Allied raid against the Peenemunde V-2 facility in 1943 which sought to kill von Braun and his team of rocket scientists, the Iranian nuclear scientists are prime targets.
The Israelis have shown remarkable restraint throughout their history, so I doubt they will choose my approach. But, be assured, attack they will.