A Review of Criticality Accidents

December 10, 2008 11:34 PM

A Review of Criticality Accidents (3.7 MB pdf)Do you like reading comp.risks, or CVR transcripts from famous plane crashes? Then you may enjoy this technical analysis of 60 accidents where improper handling of fissile materials led to unexpected critical mass.

All the famous names are here - Oak Ridge, Los Alamos, the Argonne National Lab - as well as the famous players: the Demon Core that claimed two lives in separate accidents, and later became the heart of the Bikini atoll ABLE nuclear test; Harry Dahglian; Louis Slotin; and even a cameo by Richard Feynman, who comments on the Dragon experiment in a footnote. Watch for the telltale signs of prompt critical - the blue glow, the flash of light, the wave of heat - and look up Roentgen Equivalent Man, and tuballoy in the comprehensive glossary at the end.

I can't be the only one who likes this stuff, can I? (Via AskMe.)

posted by ikkyu2 (36 comments total)

This is fun to compare with the OSHA - Fatal Facts database. The details are far more arcane, but the root causes so depressingly similar.
posted by contraption at 12:08 AM on December 11, 2008 [1 favorite has favorites]

A laboratory miscalculation also led to the Castle Bravo weapon test at Bikini Atoll yielding 15 megatons instead of the expected 4-6 megatons.
posted by XMLicious at 12:28 AM on December 11, 2008

Previously, sort of.
posted by Blazecock Pileon at 12:41 AM on December 11, 2008

Oh, you are certainly not the only one who likes this stuff. It's riveting. I think it's the calm, neutral tone of the narration juxtaposed with the horrifyingly dramatic and suspenseful action it describes.

What's particularly striking about these accidents is how a few yards of distance between you and the event can save your life. From the last accident (JCO Fuel Fabrication Plant, 30 September 1999), on pages 53-56 of the report:

All three operators were placed under special medical care. The operator standing on the floor holding the funnel at the time of the accident died 82 days later. The operator pouring the solution into the funnel died 210 days after the accident. The least exposed operator [a few meters away at a desk] left the hospital almost three months after the accident.

He was spittin' distance away and lived.

Oh, and "unfavorable geometry vessel" is one hell of an understatement.
posted by sdodd at 12:56 AM on December 11, 2008

Inherently fascinating, so much thanks for a fix! Especially interesting to read about radiation; watching (in pictures--I would love to hunt down video some day) and reading about what it does to the human body is just amazing. Though, of course, particularly unpleasant for the person in the process of dying/suffering.
posted by librarylis at 1:17 AM on December 11, 2008

Oh fabulous! Is it Nuclear day around here? Is MeFi blue the same as prompt-criticality blue? Inquiring minds want to know!
posted by Skorgu at 3:02 AM on December 11, 2008

I was interested to read the reports from the Idaho Chemical Processing Plant since I used to work there in CPP-601, site of several criticalities. Even thirty years after the fact each accident was still discussed by the operators. The corporate culture in the eighties was vicious, very adversarial between labor and management. The weekly rotating schedule didn't improve anything either.
posted by RussHy at 4:34 AM on December 11, 2008

Is MeFi blue the same as prompt-criticality blue?

Here's a couple of pictures. They used to tell me that if you saw a blue flash from an inadvertant criticality it was from Cherenkov radiation originating in your own eyeballs.
posted by RussHy at 4:46 AM on December 11, 2008 [1 favorite has favorites]

They seem to have left out the incident at the Black Mesa facility where a theoretical physicist inadvertently caused a resonance cascade in the Anomalous Materials Lab. The cleanup on that one was atrocious if memory serves.
posted by Spatch at 5:58 AM on December 11, 2008 [6 favorites has favorites]

This accident led to the decision to set up an in–plant critical experiment measurement capability to better determine critical parameters for vessels in routine use. The next criticality accident at Mayak on 2 January 1958 involved this critical experiment set up.

*smacks forehead*
posted by sixswitch at 6:16 AM on December 11, 2008 [2 favorites has favorites]

After each experiment was completed, written procedures called for the solution to be drained through a line to favorable geometry 6 liter bottles. This process was to be repeated until the entire experiment vessel (400 liters) had been drained. After filling some of these 6 liter bottles, the experimenters judged the remaining solution volume to be highly subcritical. It was then decided to circumvent the routine, tedious draining process and manually pour the remaining solution from the vessel.

I feel like I'm watching a horror film and the heroine just ran up the stairs. "No! No! It's not subcritical at all!"

To accomplish this, the neutron source and its guide tube were removed and then the vessel was unbolted from its stand. Then three of the experimenters manually lifted the vessel and began to move it (in order to directly pour the contents into containers).

Then they all died from radiation poisoning. No, really. Their bodies reflected enough neutrons for the solution to go critical.
posted by sixswitch at 6:22 AM on December 11, 2008 [1 favorite has favorites]

Do you like reading comp.risks, or CVR transcripts from famous plane crashes?

Yes, and yes. And I've already read this as well, and can attest that it is a fascinating read.
posted by killdevil at 7:16 AM on December 11, 2008

RussHy: The blue glow observed in criticality accidents is not due to Cherenkov radiation. It's a result of ionization of the air.
posted by jedicus at 8:34 AM on December 11, 2008

jedicus, that Wikipedia article goes on to say: "The only situation where Cherenkov light may contribute a significant amount of light to the blue flash is where the criticality occurs underwater or fully in solution (such as uranyl nitrate in a reprocessing plant) and this would only be visible if the container were open or transparent." Several of the accidents described in the report meet those criteria.
posted by sdodd at 8:55 AM on December 11, 2008

Oh, and "unfavorable geometry vessel" is one hell of an understatement.

And also one hell of a band name, which I totally call.
posted by The Bellman at 9:30 AM on December 11, 2008

jedicus
Another myth shattered! Thanks for the correction.
posted by RussHy at 10:26 AM on December 11, 2008

This report will make a good companion with reading Henry Petroski's To Engineer is Human.
posted by Monochrome at 10:59 AM on December 11, 2008

I think the accident B-2 on page 47 is the one dramatized in the movie Fat Man and Little Boy. John Cusack was a real mess after that one, with a face only Laura Dern could love.

That arc in the movie transforms what sdodd notes above - when the victim's experience is visualized, the horror of the accident's impact easily transcends the cold scientific discussions of what happened.
posted by buzzv at 11:42 AM on December 11, 2008

May God receive you, great-souled scientist!
While you were with us, even strangers knew
The breadth and lofty stature of your mind
Twas only in the crucible of death
We saw at last your noble heart revealed.
posted by buzzv at 11:47 AM on December 11, 2008

From wikipedia (emph. mine):

Many criticality accidents have been observed emitting a blue flash of light and the material heats up substantially. This blue flash or "blue glow" is often incorrectly attributed to Cherenkov radiation, most likely due to the very similar color of the light emitted by both of these phenomena. This is merely a coincidence.
posted by 0xFCAF at 12:18 PM on December 11, 2008

Oh, and "unfavorable geometry vessel" is one hell of an understatement.

And also one hell of a band name, which I totally call.

Agreed, though it sounds waaay more like a GBV song title.
posted by joe lisboa at 12:28 PM on December 11, 2008 [1 favorite has favorites]

Thanks for this awesome post, ikkyu2. I haven't had a chance to read through it all yet but I am intrigued by the topic.

I had to read things like this for a research job I had in the early 90s. There was a journal, monthly I think, called Serious Accidents, which was a house organ in the 50s and 60s for government people and contractors in the nuclear weapons and related industries. Lots of the kind of stuff sdodd describes, with measured narration of unbelievably horrific events.

One "inadvertent criticality," as they called them, resulted from a guy carrying two rods too close together, to the point where they touched and, well, you know the rest. I think this might be one of 1950s accidents at Idaho discussed in the report at the first link. The guy did not survive it, needless to say, and his body was blown into the ceiling. Extrication of it proved to be very difficult.

I was in Albuquerque a few years ago and made a point of visiting the fantastic National Atomic Museum. One of the volunteers working that day, a guy in probably his 70s or 80s, had been in the industry in the 50s and we got to talking. I mentioned the Idaho accident (something I have never been able to get out of my head) and he said he was one of the guys sent in to investigate it. He was pretty excited that someone so young (i.e. not born yet in the 50s) knew about it. We had a great conversation. He had a lot of interesting things to say. He was also using an oxygen tank. Occupational hazard?

Another story in Serious Accidents was about a guy who needed to make some kind of adjustment to the top of a tank or machine and just reached up on his tippy-toes to do it. Something horrible happened, I don't remember what (sorry about that), but the article concluded with a solemn reminder always to use a step-stool.
posted by isogloss at 2:47 PM on December 11, 2008

When I read "superprompt criticality" I just had too Google it & found this reaction to the litany of accidents:

I'm never going to sleep. I might not even leave my house ever again. What if I encounter unfavorable geometry? There might be unfavorable geometry right outside my window. That stuff is scarier than bigfoot. Bigfoot is just going to throw rocks at your cabin and maybe force you to marry its sister-bigfoot. Unfavorable geometry is going to fuck up your town worse than El Niño.

Black Sabbath wrote albums about Unfavorable Geometry but they were never released because they were too scary. One time Unfavorable Geometry was invited to a Halloween party and won scariest costume even though he wasn't wearing one. Chuck Norris once got into a fight with Unfavorable Geometry. He punched Unfavorable Geometry so hard that he created a superprompt criticality resulting in serious exposure of 12,590 rads. Both arms and his beard were amputated. He died in Moscow 36 hours later.

F'ing gold.

The author also reacts to a story (via Neatorama & Cliff Pickover) about a guy who was through-headed with a particle beam," The dividing line of his life goes down the middle of his face: the right side has aged, while the left froze 19 years ago."

Uh, what? How does that work? So half of this dude's head is immortal now, but burned up from the quantum shrapnel (I don't know what this means)? Maybe we should take away the deadly burning part of the shrapnel and just harness its immortality-granting properties? Can we do that? Do we have to use favorable geometry?

posted by morganw at 3:27 PM on December 11, 2008

Man, this is fascinating and terrifying. Can you imagine: you're working with this perfect sphere of plutonium, and you're carefully setting up an experiment, and everone in the room is all professional in their white scientist coats, but your hand holding the reflective brick slips, just for a second, not even a second, and you grab the brick as fast as you can, but everyone in the room instantly freezes and looks at you, silent and horrified. And you know that, no matter what you do, and even though you feel and look fine, you have a few days to live. A walking ghost. Jesus.
posted by bepe at 5:12 PM on December 11, 2008

I don't understand what the deal was with the 2nd "demon core" fatality. He was lowering a reflective sphere onto the plutonium as a demonstration. Of what? How to make the motherfucker go critical? What went "wrong"? As far as I can tell from the report, the incident looked like a suicide mission.
posted by Jimbob at 5:38 PM on December 11, 2008

Jimbob, he was lowering the beryllium hemisphere reflector onto the plutonium sphere. He set one edge down and was holding the other edge up with a screwdriver, apparently to explain how it was going to be lowered down slowly during the real experiment, when it slipped and fell into place and started reflecting enough neutrons to make it prompt critical. What went wrong was when his screwdriver slipped; he inadvertently completed the experiment while he was standing right next to it, which was not what he'd intended to do.

The Manhattan Project was known for a lot of things, including some great science and engineering produced in a very short time, with limited resources, under intense time pressure. Modern nuclear safety standards were unheard of, and in fact I think it's safe to say that the real risks being faced were not understood very well - there had never been radiation sources like these devices before, so their potential effects on humans were not very well understood.
posted by ikkyu2 at 5:49 PM on December 11, 2008

This is great. I have lost several hours just reading up on the physics and the medicine involved. I am relearning all the physics I knew when I was in college. Amazing.
posted by dirty lies at 6:19 PM on December 11, 2008

Black Mesa

Coincidence? I don't think so.
posted by scope the lobe at 6:36 PM on December 11, 2008

It's amazing how little was known about this stuff in the early years of development. Imagine being Harry Daghlian (the first demon core victim), handling chunks of these strange metals trying to manipulate their radioactivity. Sure, theoretically you know that at a certain point it becomes a runaway chain reaction and bad things happen, but the details are sketchy. Turns out all it takes is a few milliseconds with things arranged just right and bam. Like kids playing with matches.
posted by waxboy at 6:37 PM on December 11, 2008

Does this include the accident where the facillity was eerily quiet when the rescue came? And everyone was dead except they couldn't find this one dude? And then they found him stuck onto the ceiling, impaled by a control rod that had blown out of the core? Ah yes, #8.
posted by exogenous at 6:39 PM on December 11, 2008

isogloss: The guy did not survive it, needless to say, and his body was blown into the ceiling. Extrication of it proved to be very difficult.

Another story in Serious Accidents was about a guy who needed to make some kind of adjustment to the top of a tank or machine and just reached up on his tippy-toes to do it. Something horrible happened, I don't remember what

These both sound like descriptions of the same incident: a steam explosion caused by an inadequate control scheme over an unstable reactor design. (Said control scheme being some dude pulling a control rod by hand. He was the one pinned to the ceiling after the explosion, and the bodies of the three operators were highly contaminated).

But that kind of accident is all show and bang. Nothing like the prompt criticality accidents discussed in the first link - those give you time to realize you've killed yourself. Guh.
posted by Popular Ethics at 8:46 PM on December 11, 2008

I have a question.

In the SL-1 accident, the second body was recovered 9 days after the event, the third body 24 days. The guy who lived for a few hours was so contaminated that his body was emitting about 500 R/hr. The third body belonged to the man that had been standing on the reactor istelf.

Would the bodies be preserved because of the radiation? I can imagine some mechanical decay, but all the microorganisms would die, right? The corpses were buried under lead and concrete. How would the bodies look now?
posted by dirty lies at 12:48 AM on December 12, 2008 [2 favorites has favorites]

After I found the list of criticality accidents on Wikipedia, "blue flash" became a mini-meme in our household. We'd joke about seeing a blue flash when we put too many cute animals in one place, for example. I'm not sure if we decided whether The Blue Flash would be a good name for a superhero, or supervillain.

I see we completely missed the meme potential of "unfavorable geometry", though.
posted by Kalthare at 11:17 AM on December 12, 2008

I just stumbled across the BBC miniseries Edge of Darkness on youtube, which is one of the things I always think of when I hear the word criticality.
posted by BrotherCaine at 2:14 AM on December 15, 2008

Your mother possesses such unfavorable geometry, her ass went prompt critical and fried your brain.
posted by dirty lies at 12:47 PM on December 15, 2008 [2 favorites has favorites]

via reddit today: The Radium Girls saga holds an important place in the history of both the field of health physics and the labor rights movement. The U.S. Radium Corporation hired some 70 women to perform various tasks including the handling of radium, while the owners and their scientists — familiar with the effects of radium — carefully avoided any exposure to it themselves; chemists at the plant used lead screens, masks and tongs.

Not strictly a criticality item, but in the same vein.
posted by popechunk at 6:30 AM on December 16, 2008 [1 favorite has favorites]