Blue Light Not So Special

Every now and then I get one wrong. When I do I like to correct my errors. Thanks to Charlie Martin who made me look up some different opinions.


In my recent post Blue Light Special I said that the blue lights seen at the reactor accidents in Japan were evidence of a criticality accident. But not so fast. Maybe, despite the reports there were no blue lights.

In fact, nobody has seen any flashing lights. It is speculation on what may or may not be happening:
A partial meltdown of fuel in the No. 1 reactor building may be causing the isolated reactions, Denis Flory, nuclear safety director for the U.N.'s International Atomic Energy Agency, told a news conference in Vienna.

Nuclear experts call these reactions "localized criticality." They consist of a burst of heat, radiation and sometimes an "ethereal blue flash," according to the U.S. Energy Department's Los Alamos National Laboratory website. Twenty-one workers worldwide have been killed by "criticality accidents" since 1945, the site says.

An expert says there is a chance of such a reaction. Sometimes, that kind of reaction will include a blue flash of light inside the reactor. So, following Fox logic....OH MY GOD - BLUE FLASHES OF LIGHT OVER NUCLEAR PLANT!!
So no blue flashes.

What do we have some evidence of?

Chlorine 38 - a short half life isotope made by irradiating sea water with neutrons
Iodine 134 - a short lived decay product
Neutrons - to be detected at the approximately 1 mile distance from the plant reported there needs to be a high level source of neutrons - like a critical core or something.
Tellurium 129 - a short lived decay product found recently in reactor #1
Much higher levels of Iodine 131 in core #1 than in cores #2 and #3. The levels should be comparable if all the reactors shut down with the earthquake.

You can learn more about the last two items at Chain Reactions.

So no blue light, but still a lot of evidence of an inadvertent criticality accident.

Cross Posted at Power and Control

posted by Simon on 04.05.11 at 02:22 PM


Here's an explanation from the US Nuclear Energy Institute's site debunking the re-criticality panic:

Joseph Somsel   ·  April 5, 2011 6:21 PM


It is the ratio of I-131 in #1 vs #2 and #3 that is important in this case. About 10 to 1 IIRC. i.e. about 3 half lives or 24 days.


Te (any isotope) is not listed in this U235 fission yield table

If you have any info on how it gets made I'd like it.

Also it should be possible to tell how much of each isomer is made by taking a sample. Dividing it in two. Running one through the GCMS "instantly" and running the second sample one or two (short) half lives later. You would probably need a stop watch to get exact results but a go/no go a few days after shut down would be easy. i.e. the fact that they don't give the results for each isomer is suspect. As are most of the reports given out.

This is like an intel project. You have to try to get the correct information out of a mass of conflicting data.

I assume the data is being fudged. Or totally fabricated.

So the deal is - it is not hard to tell which isomers are in the sample. Actual quantities to a few percent is more difficult. But it should be easy to tell if there is any of the short lived isotope at all assuming they are made in roughly equal proportions. (i.e. not more than a 5 to 1 ratio).

So I like your theory about Te. That doesn't answer the question of why they don't know what they have. Or if they do know why it wasn't reported.

There may be other things that could confound a measurement. Like the equilibrium ratio changing from numerous restarts.

Given the current state of the site I do not get why recriticality is so feared. At this stage it is just another problem, not the BIG problem.

M. Simon   ·  April 5, 2011 8:45 PM

"Gundecking the logs" is a term used in the Navy.


BTW this is an interesting explanation. But a GCMS would sort it out. I used to work for a GC/LC companies and running radioactives through the column is not unusual. That was back in the 60s.

M. Simon   ·  April 5, 2011 9:04 PM

So we can't rule out recriticality although it still seems unlikely. I think we can agree that is not at the top of TEPCO's current priorities as an issue since they've dumped a bunch of boron and other poisons into the cores.

Joseph Somsel   ·  April 6, 2011 2:21 PM


I agree about the importance. OTOH if the junk is critical with fast neutrons Boron is not going to be very effective. With natural Boron the cross section for thermal neutrons is 4,000 barns. For fission neutrons (2 MeV) the cross section is under 10 barns.

M. Simon   ·  April 6, 2011 5:49 PM


This is not a hard question to resolve with a GCMS. So why wasn't it done?

Or suppose they were just using an energy spectrometer (measure the energy and amount of emitted gammas and/or betas) to determine what is there. All they needed to do was to split the sample and delay for about an hour or three the second measurement.

You have to ask yourself why the measurement wasn't done. Or if it was done why it wasn't announced.

M. Simon   ·  April 6, 2011 5:55 PM

Post a comment

April 2011
Sun Mon Tue Wed Thu Fri Sat
          1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30


Search the Site


Classics To Go

Classical Values PDA Link


Recent Entries


Site Credits