Y-12 Ships Too Much Uranium-235

In July, Oak Ridge’s Y-12 National Security Complex shipped too much highly enriched uranium. Frank Munger has the story. It was ten times too much. That breached the rules for shipping “special nuclear material” – material that could be used to make nuclear weapons.

Consolidated Nuclear Security, the contractor who runs Y-12, has been fined $33,620 by the Department of Transportation for shipping improperly labeled material and failing to train employees properly. A full kilogram of 93.2% uranium-235  – 1000 grams – was shipped to Mirion Technologies Corp. in Horseheads, NY, which develops nuclear instrumentation. The amount that was intended and packaged and labeled for was 100 grams. Mirion says that they immediately notified Y-12, who sent people to get the excess back to Tennessee properly.

It’s hard for me to see how a mistake like this could have been made with the precautions I’m aware of. My experience is at Los Alamos, so maybe they do things differently at Y-12.

There are many controls on fissile material, the term I prefer to the US government’s official “special nuclear material.” Its transfer requires bureaucratic signoffs, usually by two people on each side of the transfer. Safety is also an issue; fissile material is hazardous for both chemical and radiological reasons, as well as its potential for nuclear criticality. The last would not be a nuclear explosion, but rather a heavy burst of neutrons and other radiation that would kill or sicken people close to it.

Where fissile material is handled, both engineering and administrative controls exist. Engineering controls include such things as making containers large enough that if they are jumbled together no criticality will occur. Administrative controls include labels on shelves saying how much fissile material may be stored on them. Both are important, but my preference goes to engineering controls that prevent bad results from people doing stupid stuff.320 3013 can

For example, the containers that my team designed are much larger in volume than the plutonium they contain. If several somehow come together, that space prevents a criticality from occurring.

I haven’t done the calculations, but 100 grams of uranium-235 seems like the most a person would want in one container to be criticality safe. We also don’t know the form of the uranium in the Y-12 shipment. Uranium oxide is the safest way to ship; uranium metal is fairly stable, but it can catch fire under some circumstances. (Update: I missed that the second paragraph of Munger’s story says it was metal.)

Y-12 is the US center for processing highly enriched uranium. Criticality safety is an essential part of their job. They ship small amounts of uranium regularly to manufacturers like Mirion for standards and instrumentation. One would think that the processes for shipping would be highly regularized, including many signoffs and checks. Fissile material is kept in vaults that require two people to unlock and sign for materials.

My first thoughts on the incident were that the mistake must have been in tens of grams or less; I envisioned the uranium in the vault in containers designed to prevent criticality accidents, which would mean 100 grams or less per container. So a shipment of 1000 grams would require ten of those containers; hard to make that kind of mistake. If an amount were measured out of a 100-gram container mistakenly, the error would be limited to tens of grams.

I am having a hard time envisioning Y-12’s storage and shipping practices. Here is how CNS Vice President Darrell Graddy described the shipping mistake, via Frank Munger.

Personnel mistakenly placed more material into the containers than was intended[.] Once the discrepancy between the amounts listed on the container labels was identified, personnel changed the labels without verifying the amounts (inside the packages).

I infer from this that the process was to remove the material from a storage container or containers and place it in shipping containers. The people doing this should have double- and triple-checked how much material was to be shipped. They should have known the difference between 100 and 1000 grams – this is easy to see! They should have been aware of possible criticality issues and questioned the 1000 gram amount. And, OMG, they shouldn’t have changed the labels!

I have many questions: Did a single storage container contain more than 1000 grams? If so, what were the criticality safety precautions? What is the design of the shipping containers and how does it limit the amount? Was the two-person rule for handling fissile material followed? What is the procedure for verifying agreement between amount to be shipped and amount actually packaged? What does the inside of the vault look like? When was the last time it was audited?

I hope there will be a GAO investigation that will ask these questions and more.

Update (Octiber 26, 2015): Mirion finally got the correct order, and NNSA says it will address the error in CNS’s annual evaluation. No word about a GAO investigation.

Photo: Y-12 nuclear complex (DOE photo).