Jeffrey LewisUpdate from Tokyo

Ok, so that earthquake joke I made the other day is no longer funny. Apparently, those quakes were just the warm-up act for the massive 8.9 behemoth that struck us today.

I was just exiting the Tokyo subway when it hit.  News reports quote people saying that the quake seemed to go on forever.  That was certainly what struck me.  I also noticed that plenty of locals were really scared. This was different. (The exception were a few intrepid senior citizens who kept shopping, completely at peace with the concept that each minute might be the last one.)

Although I am sure that the devastation in the northeast is every bit as bad as the news reports, Tokyo is calm.  Well, most of Tokyo is calm.  Then there is me.  I cannot believe that this morning I was sitting in a Japanese bath in Hokkaido. That seems like years ago.

On the 14th floor, I feel each and every sway of each and every aftershock, which come regularly.  (Though not quite as frequently as the phantom aftershocks I imagine.) Still, I am conscious of my relative comfort with a large room in a luxury hotel. There are lots of people stranded in the hotel lobby.  The highways are blocked off.  It is cold.  Still, order prevails.

Anyway, thanks to everyone who asked after me.  I am fine and am looking forward to getting home.


  1. krepon (History)

    One more reason to travel less.
    Safe journey.

  2. FSB (History)

    Glad you are OK!

    Here is a more minor earthquake from FP magazine on DNI Clapper:

    “Clapper also said at the hearing he has high confidence in his assessment that Iran has not restarted its nuclear weapons program and the intelligence community does not know if it ever will. ”

    How does that fit in with sanctions and bombing etc.? What does the IAEA think of the US DNI’s views?

    • kevin (History)



  3. ataune (History)

    Stay safe Jeffrey !

  4. earthenberm (History)

    Travel safely!

  5. hmd (History)

    Thanks for the news. The only person I knew who is over there right now is you and I worried, came here to ask for your well being news. Stay safe!

  6. Andy (History)


    Glad you’re safe and hope you have a safe an uneventful trip back.


    This thread isn’t about Iran. Could you please give it rest just this once?

  7. Magoo (History)

    Glad you are safe. We are praying for Japan and its peoples. You out did yourself in yesterdays Post.
    Your ellusion to California is most important.

  8. Gridlock (History)

    Well now there appears to be one, if not more, reactor breaches occurring not to mention primary cooling circuit venting. Story is that power failed to the coolant pumps, and the tsunami took care of the diesel backups.

    They are referring to it as a reference accident though, ie within design norms for the region.

  9. George William Herbert (History)

    Ah, oops. Looks like a hydrogen explosion at Fukushima I. Building’s gone.

    Bad day 8-(

  10. Jochen (History)

    Hmm, looks more like it could have been a rupturing high-pressurized steam vessel to me. I wouldn’t expect a core meltdown in a water-moderated reactor type, though (particularly if the moderating medium “has left the building”, as i’m suspecting). But then, on the other hand, i’m no nuclear scientist, so others might be more qualified to comment on this.


    1.) Glad to hear you’re o.k. (at least so far – i wish you good luck, particularly in the near future! Stay safe!)!

    2.) I’d say that this would perhaps not be the worst time to get away from Japan, even though adventure might have an addictive side-effect…

  11. yousaf (History)

    fyi, UCS has some posts on the subject:

    This is still speculative but at Fukushima it appears that hydrogen gas was released which subsequently exploded. The process could be as described in the papers below:

    see eg: p.17

    4. Cladding Embrittlement and Hydrogen Release

    Hydrogen may be stored in the cladding due to hydrogen uptake during a period of steam-starved oxidation6,7 and then be released during a quenching period when the cladding may crack due to thermal stresses in cladding embrittled by a combination of oxygen and hydrogen uptake.7,8 The embrittlement of the cladding decreases its ductility to the point that the stresses induced by a temperature gradient during quenching may result in cracking of the cladding.



  12. Nick (History)

    I wonder how EPR with its passive safety measures would have done. This incident will impact negatively deployment of new ones in US.

    Looks like Bushehr’s latest delayed commission due to the cooling pump is a wake up call for the region.

  13. thermopile (History)

    Careful with the doom-and-gloom, please. Yes, it was a (likely) hydrogen explosion at Unit 1, but it was external to the reactor. (That implies to me that the reactor room circulation pumps were not ignition protected, but this is a requirement for US reactors for just this reason.) “Breach” has a lot of bad connotations, and there has not been a reactor breach … and probably won’t be.

    A bad situation, yes, but the bigger concern is the millions without power and clean water in Japan.

    Anyhow, there’s good coverage here:

  14. yousaf (History)

    There is an INL study which describes the putative process: (I posted on this a while back but it may be in Jeffrey’s email queue) —

    4. Cladding Embrittlement and Hydrogen Release

    Hydrogen may be stored in the cladding due to hydrogen uptake during a period of steam-starved oxidation6,7 and then be released during a quenching period when the cladding may crack due to thermal stresses in cladding embrittled by a combination of oxygen and hydrogen uptake.7,8 The embrittlement of the cladding decreases its ductility to the point that the stresses induced by a temperature gradient during quenching may result in cracking of the cladding.

    • yousaf (History)


      and can be found via google.

  15. François Heisbourg (History)

    Is there any news about the accident-prone reprocessing facility on the coast at Tokai Mura south of Sendai?
    François Heisbourg

  16. Geoff Forden (History)

    Glad you are safe! Now get out of there as soon as you can.

    ps try not to breath!

  17. Hairs (History)

    Yousaf, I think you’d find that there is not enough hydrogen stored in embrittled zircaloy to initiate an explosion.

    Far more likely – in my opinion – is that any hydrogen involved in the explosion was formed by high temperature oxidation of metals (could be zircaloy, could be stainless steel, could be lots of things).

    Taking zircaloy as an assumption, there is significant hydrogen production once zircaloy gets above about 1000 deg C in a partial steam atmosphere. All the time there is still greater than about 50 – 60% steam around, the hydrogen concentration in the air / oxygen will be too low to permit deflagration. But if the steam concentration is lowered (e.g. by the use of containment cooling sprays intended to condense some steam, and hence relieve the containment structure’s over-pressure) then the hydrogen:oxygen ratio easily achieves ignition point.

    The auto-ignition temperature for deflagration is around 500 deg C, which is much lower than the temperature required for production of hydrogen in the first place. Consequently if a metal surface is hot enough to produce hydrogen by a steam reaction, then it is almost certainly hot enough to set off an almighty big bang as soon as the hydrogen concentration is within the flammable limits.

    • yousaf (History)

      Thank you!

      I believe the scenario you outline agrees with what apparently transpired — adding cooling water made the steam and then when the steam concentration went down, pop!

      “The spokesman also explained an explosion that destroyed a building at the Fukushima plant early this morning. He said efforts to add cooling water to the reactor core had resulted in the production of hydrogen gas. The gas built up inside the building and then exploded.”

    • Allen Thomson (History)

      A couple of sentences earlier, it also says “A spokesman for the Japanese government said the reactor’s core would be flooded with sea water and boric acid.”

      That’s pretty drastic. Sea water is going to trash things thoroughly, and boric acid is to lessen the probability of parts of the fuel, perhaps slumping in meltdown, from going critical. Apparently there are some very concerned people there.

    • Allen Thomson (History)

      From the IAEA:

      AEA update on Japan Earthquake
      Staff Report

      2110 CET [2010 GMT], 12 March 2011 Japanese authorities have informed the IAEA that the explosion at Unit 1 reactor at the Fukushima Daiichi plant occurred outside the primary containment vessel (PCV), not inside. The plant operator, Tokyo Electric Power Company (TEPCO), has confirmed that the integrity of the primary containment vessel remains intact.

      As a countermeasure to limit damage to the reactor core, TEPCO proposed that sea water mixed with boron be injected into the primary containment vessel. This measure was approved by Japan’s Nuclear and Industrial Safety Agency (NISA) and the injection procedure began at 20:20 local Japan time.

      Japan has reported that four workers at Fukushima Daiichi were injured by the explosion.

      NISA have confirmed the presence of caesium-137 and iodine-131 in the vicinity of Fukushima Daiichi Unit 1. NISA reported an initial increase in levels of radioactivity around the plant earlier today, but these levels have been observed to lessen in recent hours.

      Containment remains intact at Fukushima Daiichi Units 1, 2 and 3.

      Evacuations around both affected nuclear plants have begun. In the 20-kilometre radius around Fukushima Daiichi an estimated 170000 people have been evacuated. In the 10-kilometre radius around Fukushima Daini an estimated 30000 people have been evacuated. Full evacuation measures have not been completed.

      The Japanese authorities have classified the event at Fukushima Daiichi Unit 1 as a level 4 ‘Accident with Local Consequences’ on the International Nuclear and Radiological Event Scale (INES). The INES scale is used to promptly and consistently communicate to the public the safety significance of events associated with sources of radiation. The scale runs from 0 (deviation) to 7 (major accident).

      Japan has also confirmed the safety of all its nuclear research reactors.

      The IAEA continues to liaise with the Japanese authorities and is monitoring the situation as it evolves.

  18. George William Herbert (History)

    Useful photo at NY Times site shows the parts of the building, the gone part and the remaining part:

  19. thermopile (History)

    Borated water is commonly added to reactor coolant to control reactivity; it’s part of the normal operating procedure. , for one example.

    So, putting boric acid in the coolant is No Big Deal. Seawater, however, is pretty corrosive, obviously. The Japanese won’t be able to use that core anymore, but I’m not convinced that it trashes the reactor. Yes, it’s a BWR, but I would assume they have isolated as much of the other components as possible, and I don’t think that a few days exposure to seawater is going to completely ruin the reactor vessel. There will be some expense and some (figurative) elbow grease involved, but it should be recoverable. Technically. (I make no prediction about the political ramifications.)

    PS – Although a Friend of Wonk, I’m disappointed to see Joe Cirincione’s comments on NBC saying that this is going to be as bad as Chernobyl. That’s a terrible, fear-mongering exaggeration.

    • Allen Thomson (History)

      > Borated water is commonly added to reactor coolant to control reactivity; it’s part of the normal operating procedure.

      Yeah, but this wasn’t normal operation. As I understand it, the reactor scrammed automatically when the quake hit. I.e., the control rods (or whatever they used as an equivalent) went in and the reactor became way subcritical. So why introduce additional neutron absorber unless there was a concern that things might get significantly abnormal?

      > I’m disappointed to see Joe Cirincione’s comments on NBC saying that this is going to be as bad as Chernobyl.

      I haven’t seen his comments, but yes, Chernobyl, at least in terms of physical effects, was much, much worse than what seems to be going on in Japan.

    • Jeffrey (History)

      It is never a good day when one is trying to figure out if this is going to be a bad Three Mile Island or a less serious Chernobyl.

      When I think of Chernobyl, I think of the explosions that breached the reactor vessel — something that didn’t happen here.

      This is a really terrible Three Mile Island scenario where the core melts.

    • rwendland (History)

      Re the Fukushima v, TMI comparison, I’ve come across something not very reassuring in the NRC WASH-1400 report:

      “BWR LOCA events … result in larger [than PWR] radioactive releases … because the BWR containment volume is considerably smaller, and when the core melts, the noncondensable gasses gasses generated … will overpressurize and rupture the containment. Thus, the most likely path to containment failure is by overpressure, rather than by melt-through. This failure also negates the reactor building filter system.” (NUREG 75/014,, page 65)

      So we’ve already had one BWR most likely LOCA failure mode happen[*], compared to the TMI PWR most likely LOCA failure mode (melt-through) not happening. So you could argue this has gone beyond TMI already.

      [*] Though extensive core melt is not certain, and hydrogen-air explosion seems to be the currently speculated cause.

  20. yousaf (History)

    Good ref:

    the oxidizing reactions w/ steam are highly exothermic in addition to releasing hydrogen.


    “1.1 The hydrogen problem in Light Water Reactor safety
    Light Water Reactor (LWR) safety studies have shown [1, 2] that in the very rare event of
    1055 of coolant accidents (LOCA) or plant transients, combined with complete failure of the
    multiple safety injection systems, the reactor core will heatup and can be damaged severely.
    At temperatures above 1300 K metallic components of the core and the fuel can react with
    steam to produce hydrogen according to the following reactions:
    Zr + 2H20 -1 Zr02 + 2H2+ 576 kJ/mole Zr,
    2Cr + 3H20 -1 Cr203 + 3H2+ 2 ·186 kJ/mole Cr)
    Fe + H20 -1 FeO + H2+3 kJ/mole Fe,

    Oxidation of the Zircaloy c1adding is accompanied by a very significant energy release, whereas
    on the other hand Fe oxidation is energetically nearly neutral and the fuel oxidation by steam
    is negligible.

    In large Pressurized Water Reactors (PWRs) of the 1000 to 1500 MWe dass the mass of
    hydrogen produced by the above reactions in case of a severe accident can be in the range
    of about 500 to 1500 kg. This hydrogen will be released at the break location of the primary
    coolant system into the containment atmosphere. Mixing with air and simultaneous steam
    condensation on the cold containment walls can lead to quite reactive H2-air-steam mixtures,
    depending on details of the accident sequence.
    Ignition of such mixtures would then create significant pressure loads on the containment walls
    and endanger the containment leak tightness. t Containment failure in the early accident phase,
    during wh ich the atmosphere still contains high concentrations of volatile fission products and
    radioactive aerosols, would result in large contaminations of the environment, comparable to
    the Chernobyl accident. This can not be accepted.”

    One would hope that there would be a better material than H20 to use here in the future!

    The Japanese nuclear safety agency says that the water level is not rising even though they are pumping in more water.

  21. JO (History)

    Apparently Reactor 3 and possibly Reactor 1 at Fukushima were running MOX fuel. So there are additional fissile materials monitoring implications.

  22. Gridlock (History)

    “Assuming meltdown occurred” at 1 reactor, chief Cabinet secy says #quake
    web • 13/03/2011 03:01

    I think some parties in all this might be substantially understating the progress of this accident.

    I understand the sensitivities and obviously none of us have downwind radionuclide detection systems or whatever but doesn’t it look like there’s much more concern about this externally than internally?

    Clinton spoke about the US supplying coolant – would this have been boric acid, and would this request have set off alarm bells?

    • Dan (History)

      The BBC is reporting that reactor three at the No. 1 nuclear plant, which has lost its cooling system, has both uranium and plutonium fuel. This apparently differs from reactor one (the one whose external shell exploded).

      Can someone comment on how having plutonium as fuel changes the safety picture?

    • Ian (History)

      My guess would be not much. Pu builds up in fuel loads over time anyway. The main difference between Uranium-based fuels and Pu MOX is the slightly lower percentage of delayed neutron emission from MOX, and this obviously does not matter so long after induced fission stopped.

  23. Miles P (History)

    Greetings from your fellow earthquake survivor….

    According to the World Nuclear Association it is indeed reactor 3 at the #! plant
    To wit “Tepco started up Fukishima-Daiichi-3 BWR with MOX fuel in September 2010. ”

    The WNA backgrounder on nuclear power in Japan also includes a description of Tepco’s questionable safety record and Japan’s 2008 decision to grant lifetime extensions beyond 40 years…..

  24. Hairs (History)


    When plutonium is used as fuel e.g. in the form of MOX, the neutron spectrum becomes significantly harder than it would be if pure uranium were used. Consequently the effect of neutron absorbers such as boron, xenon and control rods becomes proportionately less i.e. there will be less absorption than if the reactor were fuelled with uranium and the neutron spectrum were softer.

    In addition to the above, the moderator temperature coefficient around MOX fuel tends to be less than it would be for uranium fuel. Consequently the self-limitation of the core becoming less reactive the more the moderator temperature increases will be reduced. This is relevant because in an accident the moderator (water in the case of Fukushima) gets hotter and this temperature rise diminishes the thermalising of the neutrons and therefore dimishes the reactivity of the core. Of course, most of the heat in a tripped reactor is simply the decay heat generated by fission products, but there will still be SOME neutrons created by on-going fission even if they are only the product of spontaneous neutron emission, so it’s good if those neutrons are not thermalised and thus have a better chance of not causing further fissions.

    Lastly the Doppler temperature coefficient is lower in a MOX fuelled core than in a pure uranium core. Just as the moderator temperature coefficient affects the spectrum of the neutrons reaching the fuel, the DTC affects the absorption cross-section of the fuel itself, and thus affects whether a given neutron will be captured and cause a fission.

    All in all, MOX fuelled reactors have less “buffer” in the way they react to temperature changes and, compared to uranium fuelled reactors, require a bit more in the way of absorbers and poisons to keep them under control. In theory this shouldn’t be a problem because a MOX fuelled core will have been designed specifically for the MOX fuel. In practice, though, when adapting an original uranium design to a MOX design some compromises undoubtedly have to be made because the basics of the core e.g. the pressure vessel dimensions can’t be re-optimised. As a consequence there might be a bit of a reduction in the safety margin when using MOX fuel in a reactor design that was originally intended for uranium.

    On top of everything else, a huge amount of experience and literature exists for uranium fuelled designs and accidents, whereas for MOX fuel there is less knowledge and experience. As a result it is easier to imagine a miscalculation or error of judgement when dealing with MOX emergencies.

    Having said all the above, I have confidence in the abilities of the Japanese nuclear engineers, and (standing on my personal soapbox) if I was in the region I’d be more worried about the toxic effects of smoke from refinery fires, water-borne diseases, and straightforward exposure and hypothermia because of loss of electricity production than I would about whether MOX makes another accident more or less likely.

  25. rwendland (History)

    Reading up about the BWR Mark I Containment building, as used in Fukushima Unit 1, is interesting. It appears that the BWR Drywell Head that closes off the concrete primary containment is a known source of leakage; in tests it unseated itself at 27 psig and at 117 psig created a leakage area of between 35 and 63 in^2.

    The Drywell Head is directly under that steel roof we saw blow off. I presume the explosion was in the primary containment under the Drywell Head – could the Drywell Head have launched off into the air leaving the top of the pressure vessel exposed to the sky? In a slow-mo video of the explosion there is a large blast wave into the sky – how could that happen with the heavy Drywell Head still in place? Doesn’t sound good.

    Maybe improvements have been retrofitted since these tests in 1984ish.

    If you want to read up, these are some useful sources: (basic info and diagrams pages 6-21) (1985 containment study BNL-NUREG-36339, page 5) (1984 containment study BNL-NUREG-35286, pages 4-5)

  26. archjr (History)

    Not that I follow ice-skating, of course.

  27. yousaf (History)

    MOX is considerably nastier if it gets out — fortunately, #3 is not topped up w/ MOX:

    “One particular concern with Unit 3 is the presence of mixed-oxide (MOX) fuel in the core. MOX is a mixture of plutonium and uranium oxides. In September 2010, 32 fuel assemblies containing MOX fuel were loaded into this reactor. This is about 6% of the core.

    I have done considerable analysis on the safety risks associated with using MOX fuel in light-water reactors. The use of MOX generally increases the consequences of severe accidents in which large amounts of radioactive gas and aerosol are released compared to the same accident in a reactor using non-MOX fuel, because MOX fuel contains greater amounts of plutonium and other actinides, such as americium and curium, which have high radio-toxicities.

    Because of this, the number of latent cancer fatalities resulting from an accident could increase by as much as a factor of five for a full core of MOX fuel compared to the same accident with no MOX. Fortunately, as noted above, the fraction of the fuel in this reactor that is MOX is small. Even so, I would estimate this could cause a roughly 10% increase in latent cancer fatalities if there were a severe accident with core melt and containment breach, which has not happened at this point and hopefully will not.”