Iran's Nuclear Options After Operation Midnight Hammer
The U.S. hit sites at Fordow, Natanz, and Esfahan. Now what?
Hello from a very hot and sunny Philadelphia, dear reader. I’m here to join Aaron Stein at the Foreign Policy Research Institute this evening for a talk about my book.
I was quiet here and on social media over the weekend as the United States took the step of hitting three sets of targets across Fordow, Natanz, and Esfahan in Iran, drawing itself into the war that Israel began some two weeks ago. As I wrote with my colleague Nicole Grajewski in the Boston Globe, I don’t think it was a good idea for the United States to do this; I think diplomacy, as difficult as it may have been, presented a better option, particularly for preserving some modicum of verification and monitoring capability for what is now a very messy situation.
In any case, what is done is done, and the present situation is not particularly encouraging. President Trump, Vice President Vance, and others, including SECDEF Pete Hegseth, have all celebrated the weekend’s strikes, which were an impressive display of tactical coordination. But the broader claim being made by members of the administration that Iran’s pathway to the bomb has been “obliterated,” or “substantially” set back are just … not true, I think. The Iranians could probably assemble and detonate an initial nuclear explosive device1 in probably around a year.
Both James Acton (in Politico) and Jeffrey Lewis (on BlueSky) have broken down the facts in ways that I completely endorse. I suggest you read their contributions.
The core problem is Iran’s 400 kg stockpile of highly enriched uranium (enriched to 60 percent). This likely has been dispersed from its erstwhile storage location near Esfahan. This is the most valuable “card” in Iran’s possession and the shortest near-term input for weapons-useable material. In addition, Iran’s new covert enrichment site (the one that they might have told us about soon had this war not started), ability to manufacture components for centrifuges, and mass base of nuclear-relevant human resources and talent remain intact. The last of these is something that can’t reasonably be addressed with military force, but the other two—and, presumably, the HEU stockpile—could have been. They weren’t, and there’s no suggestion they will be at this point. (I’m aware of the messy ceasefire situation, but that could very quickly be overtaken.)
With the caveat that I have tremendous uncertainty about Iran’s political decision-making in the coming days, weeks, and months, the technical picture is pretty unambiguous, I think. The Iranians could probably reconstitute enough of a uranium enrichment capability at an unknown, potentially buried site, to quickly enrich the existing enriched UF6 (uranium hexafluoride) stockpile to ~90% levels—suitable for use in an initial nuclear explosive device in a test. Testing would be a political choice, likely accompanied by some change in status with regard to the NPT and compliance with IAEA safeguards. Matters could move quickly on this front, but I won’t comment since I’m not a close observer at this point of Iran’s domestic political churn. (Analysts who are do have me thinking that things could move in a negative direction on this front very quickly.)
If the Iranians did choose to test, they would face some other hurdles that would have to be overcome. First, their uranium metal production capabilities at Isfahan no longer exist; they would need to reconstitute something of a capability to do this. This would not be infeasible within a year and would benefit from the post-May 2019 activities they’ve undertaken. Second, there’s a strategic consideration with testing: can you test a single device and not have a broader weaponization check that you can cash in extremis? What do I mean by this? Well, outside of the existing 400 kg stockpile, Iran no longer has (as far as we know) an active conversion facility that can produce more UF6 feedstock for reconstituted enrichment capabilities. If they do test in a year, they’ll likely do so with a very short remaining bit of material that could probably result in a 7 to 8 devices. There’s the separate question of getting something that’d be a deliverable nuclear weapon (as distinct from a test device). This is trickier; what I’ve seen from the Israeli intelligence-extracted Project Amad files leads me to think that the Iranians would need on the order of an additional six months to a year to convert a working physics package into a Shahab-3 or Khorramshahr-deliverable nuclear warhead (with any reasonable level2 of reliability).
I’m not saying all of this will happen, to be clear, but this is more of a thought experiment at thinking through a scenario that is, in my opinion, not all that unlikely. A friend recently asked me my subjective probability estimate that we see an Iranian nuclear weapons test by January 20, 2029. I had to think for over a day, since I’m not even sure at this point that the regime lasts that long. But baking in all conditionalities, I landed on 0.6—so a 60 percent chance (significantly more likely than not) that we get a nuclear test for the first time in the 21st century outside of North Korea. Make of that what you will.
The problems I’ve described above are best addressed, I think, through bold, creative, and painstakingly nuanced diplomacy. I haven’t seen much under this U.S. administration that suggests this sort of diplomacy is possible. I’m not begrudging the United States and Israel their tactical ingenuity; I think the air operations we’ve seen are among the most tactically impressive in recent memory. But these are problems that cannot be bombed away without intelligence of the sort that I think is unlikely to manifest at a high enough confidence level. My understanding is that, based on the concept of operations we’ve seen for the GBU-57A/B Massive Ordnance Penetrator, the U.S. inventory will be low. And unlike centrifuges, which are highly sensitive to vibrations, power delivery interruptions, and physical perturbation of all kinds, underground weaponization activities will be less susceptible to MOP-ing up.
That’s all today. See some of you later this evening in Philly.
I am deliberately not using the term nuclear “weapon”.
My experience with the North Korean case suggests that the Iranians could probably accept a total system reliability—from missile booster ignition to airburst on target—of around 50 percent. This could be totally wrong, though. Unlike conventional missiles, they don’t need to care much about CEP for the deterrent effects. I am more certain of this.