Arnie Gunderson: Lessons from Fukushima

September 20, 2011

Gundersen expresses concerns that the nuclear industry and the Nuclear Regulatory Commission are not addressing major safety issues that have become evident since Fukushima. These issues include serious design flaws in the BWR Mark 1 containment, fundamental flaws in the Boiling Water Reactor vessel design, and problems with detonation shockwaves. The NRC and the nuclear industry are using a flawed cost benefit computer code that underestimates the value of human life and minimize property damages after an accident, which has the effect of justifying continued operation of reactors without safety modifications.


9 Responses to “Arnie Gunderson: Lessons from Fukushima”

  1. otter17 Says:

    Wow, I thought NRC would be tightening their guard more following Fukushima.

  2. […] Arnie Gunderson: Lessons from Fukushima « Climate Denial Crock of the Week […]

  3. This video is pretty good; Mr. Gunderson does an excellent job of explaining the weaknesses of the BWR Mark 1 design. I myself have always held a chary opinion of BWRs; they’re the oldest basic design, dating from the 1950s and I believe them to be obsolete. I’d like to see all of them scrapped and replaced with modern designs that are much safer. PWRs are intrinsically safer (as Mr. Gunderson observes).

    However, his second point, regarding detonation shockwaves is not so well-founded. He made his own calculation to determine that the shock waves from the detonation at Fukushima was supersonic at 1,000 mph. There’s no way that such a calculation can be reliable given the paucity of data available to us. Moreover, there are lots of complicating factors regarding the physical conditions inside the pressure vessel. In particular, the distribution and concentration of hydrogen gas inside the vessel is crucial to such calculations. The models for this calculation are quite messy and I don’t think that Mr. Gunderson is on safe ground claiming that his calculations are superior to the calculations done by other researchers.

    Lastly, I think he pulls a bit of a fast one regarding the SAMA model. He’s right that the critical number is the dollar value assigned to a human life. That number has been the subject of intense debate and has changed quite a bit over the years. In the 1970s, it was implicitly about $50,000; I have lost track of the many different versions of that number nowadays. Mr. Gunderson doesn’t actually say what that number is, but I’ll tell you: it’s $3 million; elsewhere I read that his organization seems to think that the number should be $9 million. Is this the right number? I don’t know; it’s an entirely subjective issue. However, I’d like to point out to everybody concerned that consistency requires that ANY number we settle upon should be uniformly applied throughout society. For example, we kill some 40,000 people per year on our roads and highways; if we value each of those deaths at $9 million, that adds up to $360 billion in losses — meaning that we should be willing to spend $36 billion more if that would reduce total casualties by 10%. Is that a good number? I don’t know.

    But does this not also mean that the US government should pay survivor’s benefits of $9 million for each member of the military who dies? That an airline should pay $9 million for each person who dies in an airplane accident? That any person found responsible for the death of another is automatically liable for $9 million? There are a lot of complicated considerations here. I don’t know what the right number is, but we as a society need to make sure that we set it clearly, plainly, and consistently.

  4. Scottish nuclear fuel leak ‘will never be completely cleaned up’
    The Scottish Environment Protection Agency has abandoned its aim to remove all traces of contamination from the north coast seabed

  5. My understanding is that the Dounreay facility is a weapons facility, not a civilian power facility. Am I incorrect?

  6. exasperatedradscientist Says:

    Ah, Gundersen. He’s got some nerve commenting on the safety of hydrogen detonation VS deflagration, when he’s the guy who’s claimed, and I swear I’m not making this up, that a hydrogen/oxygen mixture cannot detonate. Yes, really (hear him say it again at 6:32). And then he used that to argue that the explosions at Fukushima weren’t hydrogen explosions but prompt criticalities.

    Take anything the guy says with a pinch – nay, a dump truck – of salt.

  7. Whoa! He claimed that one of the Fukushima explosions was a prompt criticality?!?!?! Gosh, and he sounds so reasonable in the video. I watched the video you linked to and I was astonished that he claimed that a hydrogen/oxygen mixture can’t detonate. There was just enough weasel-wording that I’m willing to give him some benefit of the doubt, but the suggestion of a prompt criticality is just over the top. In that video, he declared that a comparison of xenon isotopic ratios would resolve the question, yet in the later video presented here, he does not say anything about xenon isotopic ratios; my guess is that they simply never found any xenon at all. A prompt criticality would have been far dirtier than anything we saw coming out of that area. Moreover, is he claiming that both explosions were prompt criticalities, or that one was hydrogen and the other was a prompt criticality? Either way, it doesn’t make much sense.

    I am certainly disappointed. I like to think that scientists, at least, would maintain enough intellectual integrity to permit useful discussion of complicated issues like this, but this… this is just too much.

  8. “Don’t just sit there angry. write a rebuttal.”

    Sure thing. I presume that you refer to Mr. Gunderson’s suggestion of prompt criticality, a phenomenon that occurs when enough fissile material is brought into close enough proximity to trigger a full-scale nuclear chain reaction, expanding in scale exponentially until the heat generated by the reaction pushes the components apart and brings the reactions to an end. This all happens in a fraction of a second.

    The problem here is that any kind of exponentially criticality is very difficult to achieve on the small scale. That’s why making bombs is so difficult; getting the pieces together rapidly enough and tightly enough to initiate such a reaction is immensely difficult. That’s why you need weapons-grade uranium, at 90% enrichment, to build a bomb, whereas reactors use fuel with 3% enrichment, thirty times lower than what is needed to make a bomb. It’s conceivable that you could pack tons of high explosives in a spherical configuration around a reactor core and, if you had the special switches needed to trigger them properly, compress the core enough to get supercriticality, but what we’re talking about here is a much weaker explosion that was most certainly not the kind of implosion needed to accomplish this.

    Lots of other considerations argue against any prompt criticality event. Hydrogen bubbles accumulate in the top of a vessel; any hydrogen explosion would therefore send a shock wave downwards towards the fuel rods. But the fuel rods are aligned vertically, so the shock wave wouldn’t push them together, it would just push them down. Moreover, there probably wouldn’t be enough oxygen inside the reactor vessel to sustain a big hydrogen detonation; the most common scenario has the hydrogen collecting at the top of the containment structure and there mixing with enough atmospheric oxygen to build a big enough mass for an explosion the size of what we saw at Fukushima. If the explosion initiated inside the containment structure rather than the reactor vessel, then the shock wave reaching the fuel rods would have been scattered so much that it would have little compressional effect.

    Finally, there’s the fact that there were two similar explosions at Fukushima. Mr. Gunderson can take only two positions:

    1. One was a hydrogen explosion and the other was a prompt criticality. This is a rather odd position to take, but he does seem to differentiate the explosions by noting that one had a red flash while the other didn’t. This argument is patently flimsy; for all we know, the second explosion had a flash that we didn’t see because it emerged on the far side of the reactor building. Even more important, though, hydrogen combustion occurs at a temperature far higher than is necessary to generate the kind of chemical reactions that generate visible light. Indeed, the reddish color of the flash (which really isn’t well-defined in the poor video we have) is more plausibly explainable by sodium emissions, the standard color for all flames.

    2. They were both prompt criticality explosions. The idea that we’d get two prompt criticality explosions and no hydrogen explosions is wildly implausible. We know that hydrogen bubbles build when fuel cladding gets too hot; as far back as 1979 at TMI we knew that hydrogen bubbles were a very dangerous possibility. We have never had any reason to believe that a LWR can produce anything remotely like supercriticality. Mr. Gunderson in this scenario would be claiming that two implausible events took place instead of two expected events.

    Finally, if there had been a prompt criticality accident of the magnitude we witnessed at Fukushima, there would have been a huge spike of radiation that would have pinned every measuring instrument for miles. A spike that big would have been impossible to miss, yet we have absolutely no evidence of any such spike.

    Mr. Gunderson’s speculation that the explosion at Fukushima was caused by a prompt criticality event is therefore absurd.

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: