Category Archives: Syria

Where Did The U Metal Come From?

A “short while ago”:, we all learned some new stuff from Mark Hibbs about the uranium oxide particles found in Syria. Long story short, the IAEA has reason to believe they were uranium metal “converted” into oxide by… “some event”: that divided the uranium into fine particles, causing it to scatter and oxidize.

Now let’s go further and assume that this uranium was, prior to its untimely fine division, fuel for a Yongbyon-type Magnox reactor. This raises an obvious question: Where did it come from?

Well, I think we can rule out “Lancashire”: Instinctively, most of us would say, it came from the Yongbyon fuel fabrication complex in North Korea. It might also have come from a duplicate facility in Syria, although this assumes that quite a lot of stuff could have been built, supplied, and operated undetected.

Ask Dr. Science

The natural place to start is with “SIGINT”: In February 2008, according to Dr. Siegfried “Sig” Hecker, North Korea had under a quarter of a load of fuel rods — apparently for the infamous 5 MWe reactor that has produced all or essentially all of North Korea’s plutonium — and a full load of uncladded fuel rods for the unfinished 50 MWe reactor.

Is that everything that was supposed to be there? As it turns out, Hibbs has already looked into this question. He asked Hecker and wrote up his answer in the Dec. 18, 2008 issue of _Nucleonics Week_.

None of the safeguarded fresh uranium fuel produced by North Korea at its Yongbyon nuclear research center for two of its own reactors was diverted to an alleged clandestine reactor project in Syria, a US expert said December 15.

Siegfried Hecker, a director emeritus of the Los Alamos National Laboratory, said in an interview that the fuel, which had been declared to the IAEA, was still at the center when he visited it in February.

OK. So what about the period after the collapse of the Agreed Framework and before the return of inspectors? Is it possible that the complex was put back into operation, making more fuel?

Let’s go back to Hecker’s “Feb. 2008 trip report”:

bq. The front end of fuel fabrication (Building 1) had been operating making uranium dioxide (UO2) from uranium ore concentrate right up to the time the facility was shut down on July 15, 2007. The back end was operational with seven conversion furnaces, two casting furnaces, and eight machining lathes. However, the middle part, the fluorination facility, had deteriorated so badly during the freeze (1994 to 2003) that the building has been abandoned (as we were shown in August 2007). However, the DPRK had recently completed alternate fluorination equipment (using dry rather than wet techniques) in one of the ancillary buildings. However, this was a makeshift operation that has limited throughput potential. It was not put into full operation by the time of the shutdown on July 15.

It sounds as if A) the complex had only limited capacity during the dark period between 2002 and 2007, but B) the North Koreans were operating some (perhaps all) of the parts that worked, and C) they made efforts to reconstitute what wasn’t working (fluorination) on an _ad hoc_ basis. This element was not “put into full operation,” but that doesn’t mean that it produced nothing, either.

How much fresh fuel could these arrangements have created? Enough for the “test assembly” that one of Hibbs’s sources suggests Syria might have had at the reactor? It’s not really clear, but seems possible. Either way, it raises the question of where the Syrians expected to get a full load of fresh fuel in 2007, when Yongbyon was once again under safeguards. And no, I’m not even going to start speculating about reprocessing facilities.

Anybody Around Here Read Korean?

One way to know what happened at Yongbyon during the dark years would be to scrutinize the operating records of the fuel fabrication complex. There are forensic tests to establish the genuineness of such documents. But according to (you guessed it!) Mark Hibbs in the Jan. 12, 2009 issue of _NuclearFuel,_ these records were not among those provided by North Korea to the United States.

All together now: _Hmmmm._

So, it’s an open question. Any enterprising journalists out there want to take a stab at this one?

Just for fun: Hibbs quotes a “senior UN official” in the Feb. 23, 2009 issue of _NuclearFuel_ as saying that the particles “looked like UO2,” not other oxides. Also, a “Western safeguards official” said that — along with uranium oxide and graphite — there was lots of aluminum and magnesium in the IAEA samples from al-Kibar. That’s what Magnox fuel cladding is made of. But it might have come from the soil, too, right?

One thing is clear, at least: the Syrians won’t be letting the IAEA back in anytime soon.

After all that, you deserve a “musical bonus”: Knock yourself out.

Hibbs on Syria U Traces

So it turns out that we’re just living in James Acton’s shadow over here.

Mark Hibbs has details on the “uranium oxide traces”: that the IAEA found in Syria. It’s in the the April 6 issue of “NuclearFuel”:

First, it appears that “James’s view”: of the 80+ particles of anthropogenic U found in Syria is shared within the IAEA:

bq. Early this year, a senior UN safeguards official said the IAEA believed that the particles appeared to be NATU that had oxidized (NF, 23 Feb., 5). According to the official, the finding was consistent with the hypothesis the uranium was metallic fuel material for a DPRK-type production reactor.

Second, Hibbs goes on to relate that there’s not a great deal of research on the behavior of U particles under various conditions, but it’s generally accepted that if they are divided finely enough, they’ll oxidize all the way through.

Reading between the lines, it seems that the number of particles found in the swipe samples — now described by Hibbs as numbering close to 100 — is simply too many to be accounted for by cross-contamination. So much for “my pet theory”: As long as we’re not talking about something off-the-wall, like HWR fuel — this was said to be a “Magnox”: reactor, remember? — then in the absence of additional information, it’s only reasonable to believe that there was a lot of uranium metal onsite, which was blasted into itty-bitty bits. (That’s a technical term.)

Hibbs also offers a great deal of insight into the state of attribution technology for particles of this type. The U.S. has traces of U from North Korea; is it sharing the “fingerprints” with the IAEA for purposes of comparison? We don’t know. But that Acton fellow has “written about the North Korea traces”:

Lastly, it seems that the Syrian government has refused requests to return to the scene of the swipes to take more samples. Once stung, twice shy. This once again raises the question of when the IAEA will be willing to use its power of “special inspections”: Any day now, folks!

If the subject interests you, Pierre Goldschmidt makes special inspections Topic A in a recent paper on “strengthening the nonproliferation regime”: Some guy named James Acton also “spoke about this recently”:

I think I’ll go find something else to blog about now.

Syria: Ups and Downs of Special Inspections

The verification provisions of many arms control or nonproliferation treaties involve some version of the “special inspection.” Under whatever name, it comes down to the ability to inspect an undeclared site. There are invariably limits on these provisions, since the parties have legitimate interests in keeping unrelated secrets and in not having government or commercial activities disrupted excessively by short-notice visits from outside.

But probably the biggest checks on special inspections are unwritten. One is the concern that, once the provision is used by a state party, it could invite retaliatory uses that might prove embarrassing, intrusive, disruptive, or just humiliating. Many governments are sensitive about their national sovereignty and don’t wish to set any precedents.

Another concern is that use of the special inspection provision could prompt a refusal, leading to a withdrawal from the treaty, perhaps leading to an unraveling of the regime. As a result, these provisions are used only very sparingly, if ever. (Has there _ever_ been a challenge inspection, as it is called, under the Chemical Weapons Convention?)

So it’s encouraging to see former IAEA official Pierre Goldschmidt — along with noted experts Mark Fitzpatrick and James Acton — calling on the IAEA to conduct a special inspection in Syria under the terms of its nuclear safeguards agreement.

An (extremely) abbreviated version of their statement appeared as a “letter in the IHT”: The whole thing is “here”:

This topic has already been “kicked”: “around”: at “ACW”:

On The Other Hand

Gaming it out a bit, though, it’s doubtful their advice will be taken. “The Syrians seem disinclined to come clean”:, and the international community may lack the stomach to undertake a protracted confrontation, with referral to the Security Council, sanctions, and so forth. Without at least one of these conditions in place, pushing for special inspections could do more harm than good.

As others have argued, proceeding regardless could set a good precedent by making this type of activity more routine. But we can assume that the Syrians have learned a thing or two from the past experiences of Iraq, Iran, Egypt, South Korea, North Korea, and (of course) Syria itself. Special inspections at well-scrubbed sites that turned up nothing could further deplete the will to pursue the matter. Following this line of thought, precedent-setting special inspections might be better done elsewhere, where less is at stake.

For these reasons, cautious officials at the IAEA could easily conclude that the risk of a setback to the NPT regime wasn’t worth it. I’m not endorsing this point of view, assuming anyone actually holds it, but it seems understandable. Of course, there also seems to be something rather less calculated in play — sheer pique that Israel bombed al-Kibar rather than tipping off the IAEA, and that the U.S. stayed mum, too.

Two Strikes

If the IAEA does continue to duck calls for special inspections, it would be a shame, because it would mean losing the best remaining chance to investigate the “three mystery sites”: That’s strike two.

What was strike one? Despite the efficacy of swipe sampling, Syria has managed to deny the international community any truly thorough opportunity to investigate al-Kibar. The simple acts of clearing the site, laying a concrete foundation, erecting a new structure, and declaring it be a military facility have sealed its contents away from the eyes of outsiders. “What do you suppose is under that slab”:

One of the reasons I was skeptical of early reports that the mystery building was a reactor was how little time it took to make it all vanish. Large amounts of rubble and the major structures of a reactor couldn’t be whisked away undetected quite so quickly, right? But as it happened, “the main structures were in a deep basement”: to begin with. And there is no reason to assume that they were completely removed.

Actually getting at this stuff would give fresh meaning to phrases like “destructive assay”: and “nuclear archaeology”: But when would the Syrian authorities allow such a thing? Under most circumstances, the basis of the nonproliferation regime is consent. So in the absence of an extraordinary exercise of coercive power — the sword of Damocles suspended over Iraq during the UNSCOM era, let’s say — no inspection power is quite special enough.

At the mystery sites, though, no such entombment has taken place. Or so I’d hope.

Let’s sum up.

It’s fascinating out here in wonk-land to observe the unfolding of this particular nuclear-forensic drama. But the bottom line is, as long as the big powers are not of one mind on the importance of nuclear nonproliferation, the international bureaucrats probably aren’t going to stick their necks out too, too far.

And that is bad news.

“Musical bonus”:

Update From The Pencil Factory

“Reuters reports”: that the IAEA’s Olli Heinonen has given a briefing to diplomats about al-Kibar’s Pencil Box — you know, the place in Syria where they found “traces of uranium and graphite”:

The briefing contained the following statement: “Eighty particles of uranium oxide is significant.”

So does “uranium oxide” mean that we are looking at “cross-contamination”: from U3O8 (i.e., yellowcake), UO3, or UO2 at another facility, as I’ve suggested, and not from “Magnox fuel”: located onsite, as James has suggested?

Unfortunately, it doesn’t give us much ability to distinguish between these two possibilities. As this “handy cheat sheet”: from Argonne National Laboratory explains, uranium metal (such as that found in Magnox fuel) “is subject to surface oxidation. It tarnishes in air, with the oxide film preventing further oxidation of massive metal at room temperature.”

Microscopic particles would presumably oxidize all the way through. In other words, whatever they might have been part of before the bombs hit the building, they’re now infinitesimal specks of rust.

At the same meeting in Vienna, the “NY Times reports”:, a Syrian diplomat stated that the replacement building is missile-related:

“He made a reference to a missile, one missile,” said a European diplomat, who spoke on the condition of anonymity under usual diplomatic protocol.

Thanks to my impeccable clandestine sources, I can now bring to you an exclusive ground-truth photograph recently taken near the site.



Not only did the IAEA find “anthropogenic natural uranium particles”: at the site of Syria’s apparent graphite-moderated reactor, “they found particles of graphite”:, according to Reuters and other outlets.

But perhaps it was a secret military pencil factory.

OK, OK, the jury’s still out. But anyone who said that it couldn’t be a graphite-moderated reactor because no graphite was found at the site will have to cross that one off their list.

More Than You Wanted To Know About Magnox

Paul “raises a really important point”: about North Korea’s uranium conversion capabilities. It’s a timely subject, too.

The fuel fabrication complex at Yongbyon is reported to involve a series of process lines for uranium conversion. Uranium ore concentrate (i.e., yellowcake) is converted to UO3, which is converted to UO2, which is converted to green salt (UF4), which is then converted to metal to produce Magnox fuel rods. (“Have a look”:

The metallic fuel is natural (unenriched) uranium, which is cast into cylindrical shapes, machined smooth, and placed inside “cans” made of a magnesium alloy. “Here’s how the last conversion step is done in the UK”:

Uranium tetrafluoride (UF4) is converted to uranium metal for Magnox fuel by mixing it with magnesium metal. When heated in a furnace to 600oC, the UF4 and magnesium react together. Uranium melts and flows into a catchpot at the bottom of the furnace and a layer of fluoride slag forms on the top. After cooling, the billet of uranium is separated from the slag, remelted, and cast into rods.

And that’s why Yongbyon doesn’t have a UF6 process line (that anyone knows about). Now you know.

Now, as far as anyone knows, the only place this process has taken place in recent years, besides Yongbyon, is the “Springfields”: facility in Preston, Lancashire, England.

This is a timely subject because the IAEA’s latest report on Syria, “GOV/2009/6”:, mainly concerns the uranium traces found at the suspect site in the wadi at al-Kibar (AKA Dair Alzour), which appears for all the world to have been a Yongbyon-style Magnox reactor. It says that

analysis of the environmental samples taken from the Dair Alzour site revealed a significant number of anthropogenic natural uranium particles (i.e. produced as a result of chemical processing)…

Now, perhaps these particles weren’t traces of Magnox fuel or one of the related compounds mentioned above. But if they were from Magnox fuel, there are only three possible sources I can think of:

Now consider the following excerpt from the IAEA’s Syria report of November 2008, “GOV/2008/60”:

14. Satellite imagery and other information available to the Agency concerning installations at the three other locations in Syria referred to above suggest that those locations may be of relevance to the activities at the Dair Alzour site. As indicated above, the Agency requested access to the three locations on 2 May 2008. Analysis of satellite imagery taken of these locations indicates that landscaping activities and the removal of large containers took place shortly after the Agency’s request for access. While these activities may be unrelated to the Dair Alzour site, it would be helpful if Syria were to provide an explanation for these activities and to permit the Agency to visit the three locations.

Unfortunately, these three locations are mentioned only glancingly in GOV/2009/6. It doesn’t sound as if access has been granted, or will be anytime soon. And as for those “large containers,” what was in them and where they went is anybody’s guess.

Congratulations. You made it to the “musical bonus”:

Syria Nuclear Evidence Destruction

I agree that the United States may be right to be concerned about “possible Syrian efforts”: to sanitize sites that may (or may have been) nuclear-related. But this whole thing would be easier to sort out if the Israelis hadn’t blown up the reactor-box thing in the first place.

Blog Influence

“”: (a sandbox in which I used to play) and its readers have had their influence publicly acknowledged a couple of times recently.

You prolly already know about the _NYT_ “article”: which quoted Jeffrey and some readers’ comments on the subject of the “photos”: that Iran released of its centrifuges.

But Mark Fitzpatrick of IISS recently paid that blog another compliment. He said during the institute’s June 19 Washington roll out of its “most recent dossier,”: _Nuclear Programmes in the Middle East: In the shadow of Iran_
that “comments”: on _ACW_ about the Syrian reactorbox that Israel blew up helped analysts who worked on the dossier to assess that the box was actually a reactor.

Anyway, credit where it’s due and all that.

P.S. Mark could have been talking about the post I linked to above, or it could be, say, “this one”: or “this one.”: I am just guessing.

CIA on Syria Reactor Production Capacity

The reactor featured last week in that video had a production capacity of 1-2 SQs of Pu per year, “according to”:;_ylt=AtAZJrysQL72UZFanfm2PwyCscEA CIA director Hayden. This is apparently based on the assessment that the reactor was the same size as the Nork reactor at Yongbyon.

The reactor was of a “similar size and technology” to North Korea’s Yongbyon reactor, Hayden said, disputing speculation it was smaller than the Korean facility.

“We would estimate that the production rate there would be about the same as Yongbyon, which is about enough plutonium for one or two weapons per year,” he said.

Too bad there isn’t some international agency with the capacity to provide an independent assessment of this sort of thing…