When Kitty Litter Caused a Nuclear Catastrophe(practical.engineering) |
When Kitty Litter Caused a Nuclear Catastrophe(practical.engineering) |
This accident was traced to a manager transcribing "inorganic absorbent" as "an organic absorbent". A more serious example of the need to have competent people with domain knowledge in the room and empowered when documents are written.
You don't need to know a lot of chemistry to realize that mixing organics with nitric acid is a bad idea. Why did none of the technicians doing the work say "hold on, this doesn't seem right"?
Right? We don't store nuclear waste where I work ... BUT one time we needed to buy a bunch of ethernet cables, basically the same thing. We wrote down our requirements, came up with some options. The engineers evaluated the options before purchasing and checked what we received before installing it. There wasn't even a formal process, it's just ... how you do your job?
Obviously organizational dysfunction is a real thing, particularly at LANL, so I can definitely imagine how this sort of thing can fall through the cracks for various processes. But I feel like but requirements verification should be a rigorously enforced formal procedure before storing nuclear waste in perpetuity.
It was an expensive mistake, but thankfully, no one died.
But still, I'm a bit alarmed that a trained nuclear technician would simply follow these instructions and mix organic material with acid without having any second thoughts about it...
A literal, or literary, bit-flip.
Internship started as this thing: https://youtu.be/hq03MsP1MPI?si=lVpDMLqRN4nfwMiA really great experience.
There are a lot of brilliant people there both in terms of science and project management. However, the best person I knew got driven out. But I think also a lot of nepotism and a security clearance culture that filters out really interesting people and leaves behind the dangerously milquetoast.
Why they were useless is interesting in itself. It turned out that controlling robots, when all you have is a bad TV camera, is hard. And robots also tend to get stuck on things.
As a result, the "Joker" robot that was helping to clear the roof got its tracks wedged on a firefighters water pipe.
>In July 2012, LANS issued Solution Package (SP) Report-72, Salt Waste (SP #72) (Revision 1) to address the processing steps for nitrate salt drums. This document concluded that the glovebox procedure must be revised or replaced to ensure that the final waste mixture meets or exceeds 1.2:1 kitty litter/zeolite:nitrate salt as specified by May 8, 2012, LANL-CO white paper.
>In response to SP #72, LANS prepared a major revision to the glovebox operations procedure. Section 10.6 was added to provide instructions for nitrate salt drum processing. Paragraph 10.6[3] stated “ensure an organic absorbent (Kitty Litter/Zeolite® absorbent) is added to the waste material at a minimum of 1.5 absorbent to 1 part waste ratio.” The Board concluded that specifying the use of “organic” absorbent and the omission of the word “clay” in the WCRRF glovebox procedure was not consistent with the direction provided in the white paper.
https://wipp.energy.gov/Special/AIB_WIPP%20Rad_Event%20Repor...
I remember at the time there was also some concern that the swap had taken place due to a green initiative to use renewable sources rather than something that was mined. there were no sources to back that up except the fact that the organic litter is a little over double the cost of clay litter.
It was not sufficient to just write "inorganic". Given the seriousness of possible consequences, some redundancy should have been added. E.g. "inorganic mineral-based kitty litter can be used; organic kitty litter is not acceptable". A few more words would have prevented an actual nuclear incident.
In May 2012, Los Alamos published a white paper titled “Amount of Zeolite Required to Meet the Constraints Established by the EMRTC Report RF 10-13: Application of LANL Evaporator Nitrate Salts.” In other words, “How much kitty litter should be added to radioactive waste?” The answer was about 1.2 to 1, inorganic zeolite clay to nitrate salt waste, by volume.
That guidance was then translated into the actual procedures that technicians would use to repackage the waste in gloveboxes at Los Alamos. But something got lost in translation. As far as investigators could determine, here’s what happened: In a meeting in May 2012, the manager responsible for glovebox operations took personal notes about this switch in materials. Those notes were sent in an email and eventually incorporated into the written procedures:
“Ensure an organic absorbent is added to the waste material at a minimum of 1.5 absorbent to 1 part waste ratio.”
“It’s almost unbelievable that we entrust ourselves - squishy, sometimes hapless bags of water, meat, and bones - to navigate protocols of such profound complexity needed to safely take advantage of radioactive materials.”
Maybe it’s just me but I feel like that all the time, not specifically about radioactive stuff, but about other highly complex and regulated environments where a simple mistake can have catastrophic consequences. As an example, just look at how the aerospace industry operates, there are so many talented scientists and engineers working in every aspect and yet many incidents take place every year due to trivial human errors, from pilots misunderstanding something to technicians not tightening a bolt enough.
Of course it’s not like we can trust anything (anyone?) else other than other humans to do this stuff, but it blows my mind how easily we forget about it.
Robot designers and toddler-minders will know some of these.
We're so used to so many failsafes that the "bare minimum to operate" for many things is frighteningly simple (very early electric wiring, for example).
If there weren't so many of us, it would be a lot harder to accept the losses.
As an owner of a particularly opinionated Orange Cat, I can relate.
Orange is overrepresented in house cats and underrepresented in feral cats. The interpretation of this is that what selects for a kitten at a shelter is an evolutionary disadvantage when in the wild.
I have owned a yellow girl (dilute orange) who was too clever by half, an orange boy who was very sweet but dumb as a post, and an orange boy with 3 brain cells most days, except he loans them out occasionally and without notice.
Well, I knew what I meant.
I tried to commit documentation to our source repository; I was told to upload it to the documentation site instead.
The documentation site that we've used for one year, subsequent to the previous documentation site the we used for 3 years, subsequent to the previous documentation site that we used for 5 years.
Following the trend, by the end of this year we will migrate documentation every 5 minutes, degrading every time.
At least source code can be migrated with fidelity. (The main branch anyway)
Who me? Bitter and scarred not at all!
We have a data team where I work that has a huge Confluence Wiki that helps them maintained shared reality about procedures and standards. I never log into it, if I need to know something from their point of view I go next door and ask one of them.
I'm a film photographer, and I had been taking my color film to a lab to be processed and scanned. A couple of months ago the lab let me know that the turnaround time for scanning would be a couple of weeks instead of a few days. Some two months later, I still had not gotten my film back. I went to the lab and spoke with the owner, and he said that LANL was sending him so much film to develop and scan that he couldn't get to his other customers, and he expected that the volume would increase. He was nice enough to give me my undeveloped film and and refund the prepaid bill.
I did not ask, as I didn't want to piss of the owner, but I have many questions. Why is LANL sending so much film to a lab for developing? Why can't LANL set up their own film developing and scanning lab, it's not nuclear engineering, it just requires some equipment and a little expertise. Why now? Why are the even using film these days? Why did the lab's owner feel it necessary to prioritize LANL's business over others, rather than putting it in the queue to wait its turn?
Are you asking why a government agency can't just magic up random money and employees during a time the parent government is saying privatization and getting rid of government employees is the way to go?
Also, LANL is an exception to general gutting of federal agencies. It's been given a record budget this FY. https://sourcenm.com/briefs/lanl-expecting-boon-from-congres... That budget was set in the Biden administration. The current administration wants to resume nuclear weapons testing. https://www.npr.org/2025/10/30/nx-s1-5590818/trump-nuclear-t...
Speaking of privatization, LANL is privately managed and operated by Triad National Security, LLC.
They have the disadvantage that they don't control odor well, and when composted smell even worse. So don't try to compost them at home.
Only problem I've had with it was when my dog decided that it smelled good enough to eat. Of course, it pulled all the moisture out of the dog's mouth and left it with a stuck-on clump that it couldn't get rid of.
"Hot" nuclear waste means theres still a lot of energy left to capture. Why are we disposing of materials that still have a lot of energy to capture? Seems like the closer we get to lead, in terms of capture, the better.
It would be like taking my half-full of gas car to the pump, dumping the existing gas, and getting fresh. Its a waste of energy, waste of resources, and generates worse nuclear waste.
This type of thought rings more and more true to me as I grow older.
after reading a summary, this title is very click-baity
The title reads like "a cat got into a nuclear plant, got contaminated in a and spread radioactivity with cat litter stuck to its paws"
It's kind of like why old and broken polyvinyl chloride pipes go to landfills instead of being burned as fuel in power plants. Even though PVC is flammable, the cost of burning PVC and capturing its carcinogenic combustion byproducts is a lot greater than burying waste PVC and burning fossil hydrocarbons.
In the far future, uranium mining costs might rise enough that it makes economic sense to reprocess old spent nuclear fuel. In the early days of the atomic age people thought that reprocessing and breeder reactors would be necessary because uranium was believed to be very rare on Earth. Vigorous exploration programs and new mining techniques proved this belief to be false by the end of the 1960s, and the situation hasn't changed since then. It's safer and cheaper to mine fresh fuel and just store the old fuel without any sort of reprocessing.
See e.g.
Bunn, Matthew G., Steve Fetter, John P. Holdren, and Bob van der Zwaan. 2003. The Economics of Reprocessing vs. Direct Disposal of Spent Nuclear Fuel (PDF):
https://dash.harvard.edu/server/api/core/bitstreams/7312037d...
Some of the fission products that are produced in reactors are actively harmful for sustaining a chain reaction (neutron poisons), Xenon-135 being a prime example. Xenon-135 only has a half-life of about 9 hours (which means it’s pretty spicy) but it also has a massive neutron capture cross-section. If it doesn’t capture a neutron, it emits a beta particle (electron/positron), which doesn’t contribute to sustaining the reaction; if it does capture a neutron it becomes Xenon-136 which is pretty stable. In both cases, it’s sitting in the fuel but either useless (yay) or actively hurting the neutron economy (boo).
At some point in the future it might be economically advantageous to reprocess “used once” nuclear waste and use it in a second cycle but for now it’s way cheaper to get more fresh uranium and process that into fuel instead.
And in this case, the reaction had nothing to do with nuclear energy. It was a regular chemical reaction, that got into thermal runaway.
Nuclear reactors do not work off radioactive decay. U-235, for example has a half life of 704 million years. Radioisotope thermal electric generators [0] by contrast do run off radioactive decay, an isotopes used for that application have short half-lives, such as Pu-238 with 87.7 years.
Commercial nuclear reactors use unenriched or minimally enriched fuel. This means that, within a fairly short period of time, the percentage of fissile material in the fuel drops to the point where continuing to use it is no longer economical. At that point the fuel is a mixture of extremely hot fission products, transuranics, unreacted fuel, and non-fissile (but fertile) isotopes such as U-238.
It's not practical to use the decay energy from the fission products for power. What would make much more sense would be to remove the fission products and recycle the fuel that remains into new fuel (for a reactor that's designed to use it). This would be a much more efficient use of mined nuclear fuel (allowing nuclear power to be used for thousands of years), it would vastly reduce the volume of nuclear waste, and it would mean nuclear waste would only be hazardous for decades to centuries.
The US was on the path to this with the Integral Fast Reactor and Pyroprocessing [1] developed by the Argonne National Laboratory. This was killed [2] in 1994 by the Clinton administration. Not for any technical reason, but because it was a "threat to nuclear non-proliferation". How that makes sense when, to the best of my knowledge the process developed by Argonne couldn't be used to produce weapons-grade material, and even if it could the US already had nuclear weapons so it wouldn't be proliferating it to a non-nuclear country, I don't know. But, apparently, since some other forms of nuclear waste reprocessing can be used to generate weapons-grade material (by extracting Pu-239), it was a bad symbol so it had to go.
[0]: https://en.wikipedia.org/wiki/Radioisotope_thermoelectric_ge...
[1]: https://en.wikipedia.org/w/index.php?title=Integral_fast_rea...
[2]: https://en.wikipedia.org/w/index.php?title=Integral_fast_rea...
My first take is that I'm not surprised from a fermi problem standpoint that you can destroy two computers made from small parts smashed by radiation with a similar dose. But maybe that intuition is wrong because your brain could survive losing a few neurons but a microchip could be 0% functional after losing one transistor. My rule of thumb is about right for conventional chips but you can certainly get rad-hard chips that hold up better:
https://en.wikipedia.org/wiki/Radiation_hardening
Space is a big market for that sort of thing.
Mostly it's about penetration.
Any radiation that can get through your skin can do damage. Once that happens, the question is then how much flux is there doing damage.
It would not be unusual for the person being told to write the process document to be brought into a room with a notebook, be shown written or electronic materials in the room, take notes in a provided notebook, have that notebook be handed over after the meeting for a (non-technical) security review, then receive the notebook pages some days/weeks later and have those notes be what is used to develop the document. Security culture is good for security but bad for error-free processes involving people.
When combined with a good tasting milk, I like them quite alot
Which is to say, lies.
Totally orthogonal, but you just reminded me of a pet peeve I have.
This word is correct, but I can't stand it.
I wish we spelled this "specked", even though that has a homonym.
Like trafficked, panicked, frolicking, etc.
"specced" makes my brain wince.
Those aren't truncations of longer words though?
I'd write it as spec'ed but that's German grammar with ' as a truncation mark (signifying the omitted "ifi").
https://en.wikipedia.org/wiki/Tokaimura_nuclear_accidents
was an example of where that "empowerment" went wrong. It is usual for workers in Japanese factories to make continuous improvements in process for quality and cost and it is usually a good thing... but criticality accidents involve invisible dangers and "following procedures strictly" in that kind of work saves lives.
Notably Japan has been the world leader in nuclear accidents since the 1980s and some of that is that they kept working on things like fast reactors after many other countries quit and others that are cultural. For instance at American BWR reactors it is routine to test the isolation condenser whenever the reactor is shut down so everybody knew what it sounded like (LOUD!) when it worked but when somebody at Fukushima was asked if it was working they saw a little steam coming out the ports but had never seen it work before and didn't know what to expect.
I also want to put things in perspective: far, far more people are dying from fuckups with fossil fuels, but like "Florida man" (Florida has a law that crime reports must be published) we actually report and collect accidents involved in Nuclear production, so you can see every mistake. But you don't see mass protests because natural gas infrastructure failed in Texas and building pipes burst and people froze to death, including a young boy.
Accidents in the fossil fuel industry are far more localized. Sure, you can blow up your own plant and kill a bunch of people, but it's not too hard to clean up the mess afterwards. Even something as horrific as the Deepwater Horizon disaster won't have much of a residual impact 10 years down the line.
> A seemingly-trivial change can lead to continent-sized damages and permanent condemnation of city-sized areas of land.
Chernobyl wasn't a "seemingly-trivial change" -- it was several successive groups all choosing to do the worse-possible thing, and it still killed and harmed fewer people than the https://en.wikipedia.org/wiki/Bhopal_disaster