Note added 1/8/2015: The article discussed below has disappeared from the nuclearenergy.ir website. Find it here.
The semi-governmental nuclearenergy.ir website has published a fascinating report detailing what it claims would be an 18 month long breakout period for Iran to develop nuclear weapons. The report has since been retweeted and apparently endorsed by Foreign Minister Javad Zarif. The willingness of Iranian actors to engage on the sensitive issue of nuclear breakout is both significant and useful, as it implicitly acknowledges that there are concerns that Terhan’s future nuclear program could be diverted to weapons purposes, a risk that it appears willing to address head on. However, the methodogy and conclusion of the report are questionable and need serious scrutiny.
Most significantly, the nuclearenergy.ir article ignores the risk of clandestine nuclear facilities, such as centrifuge plants, conversion plants, hot cells. Given Iran’s past behavior, this is a real concern and factors into western estimates about breakout. Granted, there are inspectors in Iran, but the P5+1 countries are understandably concerned about Iran pursuing what have been dubbed “sneak out” scenarios where parallel facilities are built and then used in a rapid push for a bomb. 100% verification is unreachable, which is why other indicators are significant in gauging intentions; indicators like the past possible military dimensions of Iran’s program and the status of its missile program. Iran may not intend to breakout, but the P5+1 the international community might not be willing to take that chance.
Below is a detailed response to particular points contained in the nuclearenergy.ir report.
“Iran is also constructing a heavy water research reactor. The nascent reactor in Arak, known as the IR-40, has a capacity of 40 Mega-watt (MW) and is designed to meet Iran’s need for radioisotopes, material test and other neutron therapy and neutron studies. The United States and its allies claim that this reactor can produce weapon-grade plutonium. Although plutonium is an inherent element of any reactor of any type, Iran has declared repeatedly that it does not need the IR-40’s plutonium in any shape or form.”
It is true that every reactor can make the “element” plutonium, but not every reactor is well suited for producing weapons grade plutonium. In fact, ordinary light water reactors tend to produce reactor grade plutonium when the fuel is removed not weapons grade plutonium.  Weapons grade plutonium is much easier to use in a bomb than reactor grade plutonium, although, it must be said that they are both dangerous. Simple, back-of-the-envelope estimates and calculations that are more elaborate, predict approximately 8-10 kg of weapons grade plutonium produced per year from the IR-40 reactor. The type of reactor that Iran has chosen to build is exactly the type of reactor that is of concern for weapons grade production.
The fact that Iran has declared repeatedly that it does not need plutonium is beside the point, Iran’s significant investment in heavy water reactors is the issue. While there may be legitimate reasons for building heavy water reactors, there is a very serious precedent for the P5+1’s concern. In the 1950’s Canada built a very similar 40 MW (the same power as the IR-40) heavy water reactor in India which was used to produce weapons grade plutonium (also had similar plutonium production capacity) and this led to the first Indian nuclear test. This was despite the fact that Canada expressly stipulated in their agreement with India that should only be used for peaceful purposes.
“Enriching uranium above 90 percent U-235: although Iran now has the required technology and infrastructure for enrichment of uranium, Iran’s current centrifuge machines are not capable of directly enriching uranium from natural U-235 concentration (or even up to 5 percent) to above 90 percent.”
The wording in the article is troublesome. Iran is certainly capable to enrich uranium to 90% enrichment. Say 25 kg of 90% enriched uranium is the goal. You can think of this as a final destination on a long road. A centrifuge is like a vehicle to get to the final goal. You can decide to stop at the 5% milestone or you can continue on along the road to get to 90% final destination. The same vehicle is used either way.
The quote continues:
“With Iran’s current capabilities, the latter process would have to be step-by-step, including reconfiguration of Iran’s current centrifuge cascades which have been installed and piped for enrichment up to 5 percent. Now, how long will it take for Iran to hypothetically conduct the process of producing HEU?”
This should not take a long time. You can go to what I call Issue 4 and Issue 5 where you can see my response. I believe the change in configuration should not take longer than it took to change it back to current post-Jan 20’th mode. In a tandem centrifuge mode Witt et al. predict 20 kg feed of near-20% enrichment could produce 2.5 kg 90% in one month.
“Assuming the highest estimate of Iranian SWU capacity, the required time for producing 6,000 SWU will be 6.6 months. But a very obvious and underlying principle has been neglected here . The theoretical critical mass is not reachable even within that timeframe for any non-nuclear weapon states which does not have the expertise. There is some loss and waste of material in the process of learning. Moreover, any chemical conversion and transformation process has its inherent loss in the form of solid and liquid waste as well as in-process holdup. Thus, as a rule of thumb, there is a need for more than 6,000 SWU of HEU as raw material for diversion.”
No underlying principle has been neglected. The 25 kg Significant Quantity (SQ) already takes into account processing losses etc. Many analysts suggest that the 25 kg SQ rule is too high. It is NOT the minimum amount that you would need to have to make a nuclear bomb, and it is NOT the critical mass. You could make as much as two 10 kt bombs with that much HEU for a nuclear weapon newcomer country. See Table 1 for an independent critique by the Natural Resources Defense Council (NRDC) for the amount of uranium necessary for countries of different levels of nuclear weapon capabilities.   Obviously, the true details are classified but we know that at least one well-known weapon scientist, Theodore Taylor, testified in 2004 NRC hearings that: “Amounts of U-235 as small as 1 kg are significant quantities. He did not state that anyone can build a bomb with 1 kg of U-235, but did suggest that this is roughly the amount that good designer would need.”  The NRDC estimate for high technical capability for HEU is 2.5 kg for nuclear weapon countries with high technical capabilities. For countries that are new to building nuclear weapons considered to be “low” technical capability, only 8 kg of 90% HEU is necessary to produce a 1 kt bomb. So the 25 kg SQ rule should really not be used as a criterion, it is too high. It is also important to realize that you can always manipulate the configuration of the fissile material to decrease the critical mass. For example, this can be accomplished by adding a reflecting material like Beryllium. According to Alexander Glaser’s (Princeton University) estimate adding a 15 cm reflector to 93% enriched uranium metal sphere decreases the bare critical mass from 54 kg to just under 12 kg. 
|Weapon-Grade Plutonium (kg)||Highly-Enriched Uranium (kg)|
|Yield||Technical Capability||Technical Capability|
Table 1: Figure taken from NRDC document showing the mass of fissile material needed as a function of yield. Refer to the originating document for the details.
The prerequisite of the processes outlined as involved in HEU production is a reconfiguration of cascade piping and other auxiliary equipment. This process is a civil-mechanical process and demands heavy physical work that has its lead time.
With its current generation of centrifuge machines, Iran must reconfigure the current cascades to tandem cascades, which minimally takes 6 months for 54 cascades. The current configuration is based on single cascades, which in the best condition yields up to 5% U-235 concentration.
It would not take 6 months to reconfigure the configuration. In an important paper by William C. Witt et al., they calculate the performance of Iran’s centrifuge cascades in a breakout scenario where the configuration of the centrifuge cascades are the same as they were before the JPOA.  They assume the following conditions:
1) The configuration of the centrifuge cascades are converted back to what it was before the JPOA
2) Near-20% UF6 is fed into the cascades rather than near-5% UF6. For this to happen the uranium oxide must be converted back to UF6. The total quantity of near-20% enriched material that is currently in the form of uranium oxide is 303.2 kg. 
3) The near-20% UF6 is fed into the centrifuges just as near-5% enriched UF6 was fed into the same cascades to produce near-20% UF6. However, in this case 90% enriched UF6 is produced rather than near-20% enriched UF6.
The authors find that 20 kg near-20% UF6 can be fed into the cascades every month to produce 2.5 kg UF6 with 90% enrichment if the configuration of the centrifuges are changed back to the tandem cascade configuration. This means that the piping and feed rates can stay identical to the configuration before the JPOA and in less than 3 month Iran could produce as much as 7.5 kg near-20% HEU. However, it is important to note that as far as we know, Iran does not have a reconversion facility (U3O8 to UF6) that we know of.
Now, how long would it take for Iran to resort back to the tandem cascade configuration that it had before the JPOA went into effect? We can estimate this since the IAEA reported that the near-20% enrichment has continued until just before the JPOA went into effect.  We can then estimate the time for the piping change based on the time when production of only near-5% enrichment UF6 started. Unfortunately, the IAEA only quotes the near-5% production quantities on 2 dates after the Jan 20th, but not the start date of production. However, it is reasonable to assume that production would be linear. On the date near-5% production started the initial quantity would have been 0 kg. So we can calculate what the date was when the production quantity was zero. Or in other words, when the line of 2 points intersects the X-axis (see Figure 1 below). This date is Jan 20 for Natanz and Jan 22 for Fordow. It is not hard to see, qualitatively, that the 3 points make a line when the initial point is close to Jan 20th .
The crucial point is this, it did not take a long time for Iran to get up to production after the change. As long as the equipment is under safeguards it would be hard to cheat, but if Iran kicks out inspectors it would not be hard to reconfigure the cascades, convert the uranium oxide back to UF6 and produce weapons grade UF6.
“Therefore, Iran needs a minimum of 18 months merely to produce enough raw fissile material for a single nuclear bomb.”
If you would accept the fact that a SQ is overestimating what is needed for a bomb then it would take significantly less time than 18 months to produce enough raw material. If it takes 6 months to produce 25 kg of 90% HEU, then in two months 8 kg could be produced enough for a 1 kT bomb for a low tech nuclear weapon newcomer (see Table 1).
“The IAEA is present in Iran, conducting announced and unannounced inspections, and has daily access to Iran’s enrichment facilities. Thus, it can provide early warning to the international community at the very beginning of the reconfiguration of cascades, which is the prerequisite for any HEU production.”
This is true, but the problem is that the DPRK has shown that clandestine facilities can be constructed without the international community knowing. Now, Iran is a very different case because the IAEA can’t inspect DPRK facilities, but the P5+1’s concern is that a hidden facility may exist.
“While all of Iran’s nuclear facilities are under the supervision of the IAEA’s robust inspection system, where could Iran convert the HEU to other chemical compositions? Iran does not have the necessary facility or unit operating in its current facilities, all of which are under safeguarded supervision. Constructing a conversion unit with its criticality considerations for HEU cannot be done overnight or even in several months.”
This is true. A facility can’t be constructed overnight. However, the fear that the P5+1 has is that a clandestine facility could be built without knowledge by the international community or that it has already been constructed.
“This process itself takes an additional 6 months and cannot be carried out in parallel with the conversion outlined as the second step.”
It is true that uranium is pyrophoric (ignites in air) and this does make it more difficult to handle. But this is not an insurmountable problem for a country as advanced as Iran. Countless references can be found on the net on how to work with uranium. See for example the detailed reference: James A. Aris, Machining of Uranium and Uranium Alloys, ASM Handbook, Volume 16: Machining, 1989.  However, it can be handled in an inert atmosphere and uranium is less radioactive then plutonium so it will be easier to handle. Essentially, an inert glove box will work, there is much less concern with radioactivity with uranium then plutonium.
“Another important part of hypothetical breakout is the warhead system, which inherently involves very complicated and high technology mechanisms. Iran has no experience in this field, and thus has no experience of the complicated relevant mechanisms and technology.”
True, Iran may not have experience with nuclear weapon construction, but neither did the United States in the Manhattan project. Again, once you have the weapons grade HEU, you don’t even need to test it, a weapon can definitely be designed by a country as advanced as Iran. As the famous physicist Louis Alvarez quoted:
“With modern weapons-grade uranium, the background neutron rate is so low that terrorists, if they had such material, would have a good chance of setting off a high-yield explosion simply by dropping one half of the material onto the other half. Most people seem unaware that if separate HEU is at hand it’s a trivial job to set off a nuclear explosion . . . even a high school kid could make a bomb in short order.”
“There is no significant amount of plutonium in Iran. Thus, the claimed breakout in this manner is tied to the commissioning of the IR-40 reactor, which is planned for 2015. However, even after commissioning, the reactor must work for many months to irradiate the fuel for production of plutonium. In this regard, the first step, which is to produce the required fissile material, would take at least 2 years after the commissioning of the IR-40. ”
I am not sure it would take 2 years. However, commissioning the reactor as a violation of safeguards would bring back sanctions and would make a start of the reactor very difficult for Iran. Regardless of Iran’s stated intentions, there is once again a valid concern because of historic precedents: the first nuclear weapon of many countries (Soviet Union, UK, France, India, DPRK (probably)) have been with plutonium.  While at the moment Iran may not have ill intentions, it is hard to predict the future.
The irradiated fuel assemblies comprise different high radioactive materials, which cannot be contacted or worked with except via special facilities called “hot cells.”
Iran has no “hot cell”, which is needed for plutonium extraction and further processes. Construction of such a facility would require at least 4 to 5 years, and the commissioning and the operation would require another 1 to 2 years.
It is true that Iran would need to have a hot cell and a reprocessing facility to extract the plutonium. However, I think 5 years as an estimation seems too long. Iran’s intentions have to be questioned, because of previous declarations that Iran planned to construct a hot cell for the IR-40. See para 44 and GOV/2003/75:
“In its letter of 21 October 2003, Iran acknowledged that two hot cells had been foreseen for this project. However, according to the information provided in that letter, neither the design nor detailed information about the dimensions or the actual layout of the hot cells was available yet, since they did not know the characteristics of the manipulators and shielded windows which they could procure. On 1 November 2003, Iran confirmed that it had tentative plans to construct at the Arak site yet another building with hot cells for the production of radioisotopes. Iran has agreed to submit the relevant preliminary design information with respect to that building in due course. “
See also para 74 (Annex) in GOV/2003/75:
In its letter of 21 October 2003, Iran acknowledged that two hot cells had been foreseen for this project. However, according to the information provided in that letter, neither the design nor detailed information about the dimensions or the actual layout of the hot cells were available at the present time, since they did not know the characteristics of the manipulators and shielded windows for the hot cells which they could procure. Iran indicated in that letter that manipulators would be needed for: 4 hot cells for the production of medical radioisotopes, 2 hot cells for the production of Co-60 and Ir-192 sources, 3 hot cells for waste processing, and 10 back-up manipulators. The 21 October 2003 letter included a drawing of a building which Iran said would contain hot cells for the production of isotopes. In the meeting on 1 November 2003, upon further Agency inquiry, Iran confirmed that there were tentative plans to construct at the Arak site an additional building with hot cells for the production of radioisotopes. Iran stated that that first building was to contain hot cells for the production of “short lived” isotopes, and that it intended to construct the other building to produce “long lived” radioisotopes. Iran agreed to provide preliminary design information for the second building.
Clearly, there was interest in 2003 for a hot cell that could be used to produce “long-lived” isotopes. Many analysts have taken this these long-lived isotopes to refer to plutonium. Now, hot cells are used for peaceful purposes, but some analysts are concerned that there may be clandestine facilities which could be used to extract plutonium, in batch form, smaller quantities in smaller labs that would pose less risk.
“Conclusion: It is impossible for Iran to break out in months through the plutonium route. The required time span is in years.”
It is true that Iran would not be able to produce plutonium in short order. After construction it would take several month to commission the reactor, but once it is running, assuming there are no problems it could immediately start to produce plutonium. So I would say after 1 year of running it will have enough plutonium for a nuclear bomb, recall Table 1.
 The fact is that weapons grade plutonium becomes reactor grade plutonium depending on the length of time that the fuel is exposed in the reactor. In power reactors the exposure time between refueling ensures that spent fuel will have been converted to reactor grade plutonium, but this is not the case with heavy water reactors.
 Thomas B. Cochran and Christopher Paine, The Amount of Plutonium and Highly-Enriched Uranium Needed for Pure Fission Nuclear Weapons, NRDC doc, 1995. See: https://www.nrdc.org/nuclear/fissionw/fissionweapons.pdf
 See further discussion here on SQ and how they should be interpreted: http://lewis.armscontrolwonk.com/archive/5028/significantly-wrong-about-significant-quantities
 See: Alexander Glaser, On the Proliferation Potential of Uranium Fuel for Research Reactors at Various Enrichment Levels, Science and Global Security, 14:1–24, 2006. See: fissilematerials.org/library/sgs14glaser.pdf
 William C. Witt, Patrick Migliorini, David Albright and Houston Wood, Modeling Iran’s Tandem Cascade Configuration for Uranium Enrichment by Gas Centrifuge, INMM 54th Annual Meeting July 14-18, 2013, Palm Desert, California USA.
 “As of 20 January 2014, when Iran ceased production of UF6 enriched up to 20% U-235” See: paragraph 29 and 35 in safeguards report GOV/2014/10.