Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Fukushima - Why did Unit 2 release so much more radioactivity than Units 1 and 3?

  1. Jun 15, 2011 #1
    Fukushima - Unit 2, what happened different to Units 1 and 3?

    There are recent discussions about this topic going on, so I think it deserves its own thread.

    According to attachement IV-2, Unit 2 is responsible for more than 90% of the overall emissions. So there's one big question: What was different at Unit 2, compared to Units 1 and 3?

    We know for sure that there are three total meltdowns in Units 1-3. Part of the Corium probably penetrated the RPV and is now at the bottom of the primary containment. As for Unit 1, there are probably holes in the secondary containment, but I think that those are mainly ruptured valves. So the primary containment is keeping most of the fission products in check. They would need to go through a labyrinth of pipes and valves to see fresh air.
    Unit 3 is a little more tricky. I'd say it's the same situation as in Unit 1, but the big bang which happened is still a mystery.

    So let's look at Unit 2 now. Unit 2s RPV lost pressure around one hour before midnight on March 14th, so that's probably the time when the corium penetrated the vessel. It's interesting to see that simultaneously the Drywell pressure (= primary containment) increased.
    Then, at 6:00 am, another interesting thing happens - the explosion near the torus, followed by a fast depressurization of both the torus and the RPV.

    Now I found the following pdf which deals with containment failures for Mark-I containments:


    It says "Drywell shell melt-through would result in blowdown to torus room or second floor of reactor building". So perhaps that's what happened at Unit 2, but not at Unit 1 and 3. Molten corium attacked the primary containment walls which resulted in a blowdown to either the torus room or the second floor. Or only one of the two. And the following depressurization, maybe coupled with a hydrogen explosion, could have further damaged the building. So that the primary containment would be directly connected to the outside, not through pipes and valves.
    If the blowdown occured in the eastern corner of the building, the shockwave may have deflagrated in the turbine building without even coming near the refueling deck. Which would explain why the outer structure of Unit 2 is still okay.

    What do you think?
    Last edited: Jun 15, 2011
  2. jcsd
  3. Jun 15, 2011 #2
    Thanks for the link to the pdf. Shows the importance of ensuring all efforts are made to cool the core and core debris while it is still in the RPV!

    The explosions at U1, 2 and 3 all seem too closely associated with the actions taken to vent the PC. U3 PC was vented sometime early in the morning of 3/15. The explosion was noted at 0614 and torus pressure dropped to 0. Although drywell pressure was almost twice design pressure when the explosion occurred, later in the morning RPV pressure and drywell pressures are not equal to torus pressure. One would think that after RPV breach by core debris and drywell liner melt-thru, pressures in these volumes would be quite low. The reliability of any indications after the explosion is certainly suspect, especially after they found the RPV water level instruments to be reading high when they should have been downscale.

    I can't help but think there is some correlation to PC venting. The design of the HPV system at these units shows the drywell and torus vent piping going to the rupture disk then the stack. But, upstream of the rupture disk is a branch to the SBGT System. This branch remains isolated as long as the AOV in the line going to SBGT System is or stays closed. (See vent drawings in Japanese gov't report to IAEA.) Downstream of the AOV is ductwork that would have easily failed at the pressures the PCs were vented. This would result in lots of hydrogen and possible CO from core-concrete interaction in the drywell to escape directly to the RB. The crane way from the ground floor of the RB to the refuel floor is a perfect chimney to get hydrogen to all floors of the RB.

    The location of the U2 blast is different from U1 and U3 and was reported to have been in or near the torus area room. I suppose your explanation is possible. Also, if liner melt-thru occurred, the corium causing the melt-thru is not likely to stop at the liner breach location but would flow and be ejected through the failure point along with the combustible gases. So, do we have RB wall melt-thru by core debris?
  4. Jun 15, 2011 #3


    User Avatar

    Staff: Mentor

    Why does this need a separate thread? Not saying it won't be allowed, but there are too many threads being started on the same subject.
  5. Jun 15, 2011 #4
    It's a pain in the *** finding something in the main thread. Moreover it's very hard to actually follow a specific topic there.

    For example there are three people discussing the explosion of Unit 3. Then suddenly somebody detects steam on the live webcam and posts a nervous response. Five people reply to that, sharing their own conspiracy theories. And perhaps two other new people join the thread in the meantime and ask for a summary, which will be provided by yet another member.
    Now anybody following the inital discussion will be doomed. How is he supposed to follow the issues regarding the Unit 3 explosion?

    It's already stated that Fukushima is the most complex nuclear accident ever happened. Unraveling its technical dimensions in a single thread is foredoomed to failure.

    So I really don't think that your shutting down of the Unit 3 explosion thread is helpful in any way. You're basically preventing what we're doing here for over three months - fact based speculation. It's hard borderline but still not against the rules - overly speculative. That you shut down the thread won't prevent those discussions, but rather create further confusion in the main thread.
  6. Jun 15, 2011 #5
    Two questions re Unit #2 (although it's not really related with the quite specific thread title.)

    1. Is it established that there was an actual explosion inside containment that caused it to fail or is the leaking containment cause an unknown at this stage?

    2. Please correct me if I have misunderstood that in a working Mark I system if the pressure in the torus drops to a given level below that of the containment building one or more vacumn breakers will activate to equalize that pressure difference -

    edit : i.e. drawing air (oxygen) from the containment building into the torus?

    EDIT: Oh, so is this the right thread or......??

    May i suggest these sub threads be nested under the main Fukushima thread if thats possible, that will at least keep the main forum index cleaner. Just a thought :)
    Last edited: Jun 16, 2011
  7. Jun 15, 2011 #6
    There was the sound of a heavy explosion near the torus heard on March 15th, 6:00am, followed by a sudden deprussurization of the Wetwell.
    Something did make a bang inside the containment. They don't know why what. Or at least they won't tell.

    My wish would be getting an own sub-forum, but it's still six months to Christmas...
  8. Jun 16, 2011 #7
    Here is what the report to IAEA says (from IV-64):

    It may be a very long time before we get anything better than this explanation.

    Im stilla bit confused about the exact timing of the release to the environment, because cetain graphs that I posted in the main thread just the other day, seem to suggest that a lot of stuff got into the environment in the hours leading up to the explosion, rather than after the explosion. Thats what their reactor analysis seems to show. But other analysis, based on radiation readings at certain locations on site, seems to have the highest magnitude releases happening after the explosion.

    First graph shows stuff going into environment after valve opened, before explosion.


    This graph of estimated total release amounts and the time periods they cover, has highest magnitude releases happening for just a few hours during the 15th, at a time after the explosion:

  9. Jun 16, 2011 #8
    I'm going to respond to a few things that were said in main thread recently, but as they involve release issues with reactor 2 I'll do it here.

    I need to work harder on translating this document, which explains how release estimates were calculated:

    http://www.nsc.go.jp/anzen/shidai/genan2011/genan031/siryo4-2.pdf [Broken]

    From what I can tell they did use actual sampling of conditions outside the plant, they were not just relying on other analysis of what may have happened at the reactors.

    However there certainly seems to be a reason to have some doubts about the very highest magnitude release estimates from unit 2. From the table on page 4 of the document, we can see that for the time period of the highest estimated releases, 9am to 3pm on the 15th, they had to rely on air sampling only, they did not have any dust sampling data. I believe this is due to the weather conditions of the 15th, I think dust sampling method doesnt work when its raining/snowing.

    I dont think it is at all safe to make this claim in such a concrete way. As I have pointed out in recent days, there is an issue here, especially with reactor 3 where wind around the time of explosion there makes it hard to imagine the scale of any release being accurately determined by on-land monitoring.

    But all this leaves us with is uncertainty, it does not prove that other reactors had the same order of release as reactor 2, only that the possibility exists for them to have underestimated release from other reactors.
    Last edited by a moderator: May 5, 2017
  10. Jun 16, 2011 #9
    The guy I was responding to had made a claim (that I see as inaccurate) in a very concrete way:

    I'm OK with Clancy stating his opinion.
  11. Jun 16, 2011 #10
    Hm, okay. Let's see, "TEPCO reports that Unit 2 is probably responsible for nearly all of the contamination."

    I think that's better than my previous claim. Of course I'd be interested in how TEPCO decided that Unit 2 released that much radioactivity.

    Did they actually measure what was coming out of Units 1, 2 and 3? (I can't really imagine how that would be possible without dozens of sensors in and on the reactor buildings). Did they make that statement based on the course of events? (Big radioactivity spike after the explosion sound in Unit 2 -> Unit 2 is entirely at fault)
    Or do they perhaps know more than us?

    But you're probably right. Before we can discuss why Unit 2 popped out so much radioactivity, we need to evaluate first if and/or why the report about that is justified.
  12. Jun 16, 2011 #11
    To accept premise of this thread requires one to accept a fallacious assumption.

    The thread should be re-titled.

    *Clancy beat me to it.*
    This post is redundant.
  13. Jun 16, 2011 #12
    Yeah, I tried that this morning. Unfortunately, it doesn't work. Perhaps only mods can change thread titles.

    I think "Unit 2 - why did the accident take a different course compared to Units 1 and 3" would be fitting description. But as I said, I can't change the title, sorry.
  14. Jun 16, 2011 #13
    There is nothing wrong with the premise of the thread. People took note of the estimations of total release and the resulting INES rating 7, so they should also take note of the fact that these numbers came from analysis where unit 2 was considered mostly to blame.

    There exists the possibility that they are wrong, and have either overestimated reactor 2 releases or underestimated reactor 3 or 1 releases, or reactor 4 fuel pool release. But this is just a possibility, not a fact that you have proven.

    There are in general not that many certainties at this point, such is the nature of the beast. So for sure the thread title is too certain, but that does not mean the entire premise of the thread is defunct.

    Given the on-site readings and data from further afield, I certainly have no reason to downplay the release from reactor 2. Its a big shame they dont have dust sampling for important 9am-3pm period on that date, but even without this it seems more than plausible that reactor 2 was greatly responsible for the contamination which has caused evacuations to the north west. I dont know how we may ever learn whether reactor 3 spat out more contamination than they estimate, but I will be looking again at reactor 1 and seeing if it is possible to learn how much of the north-west contamination came from reactor 1 rather than 2.

    Even if there remains a large dispute about reactor 3's contribution to the environment relative to reactor 2's, we can still talk about reactor 2 being responsible for the bulk of things, we just have to rephrase things so that we are talking only about land contamination rather than total release.
  15. Jun 16, 2011 #14
    Unit two was left without minders, so one SRV cycled mindlessly and the steam it released heated up the water in a very small portion of the torus. When burps of superheated steam get into hot water, you get a nasty pressure wave. Basically, that SRV water-hammered the torus wall until it gave out.
  16. Jun 16, 2011 #15
    Do you have any sources for that SRV-cycle statement? You're saying one SRV alone (while others were shut?) released superheated steam from the RPV into the torus which finally failed the walls?
    But the RPV lost pressure shortly before midnight, while the torus explosion sound was heard seven hours later. Why should the SRV open if the RPV was already depressurized?
  17. Jun 16, 2011 #16
    Umm... yes. Apparently, without operator intervention, only one of the many SRVs opens (the one with the lowest-threshold pressure gage, I suppose). Here's NUCENG reading from the paper traces provided by TEPCO:


    Also, this.
    Breaches in containment at #1 and #2 before the respective explosions:
    http://search.japantimes.co.jp/cgi-bin/nn20110526a1.html [Broken]

    The utility also hypothesized that a breach roughly 10 cm wide occurred at the No. 2 reactor's containment vessel 21 hours after the quake due to elevated temperatures, among other factors.

    They're working off a model here, they don't know where the breach was. Hydrogen would accumulate outside the torus etc etc. I'm not so sure about the timeline.

    EDIT: found it again. In one of my own posts no less.
    Last edited by a moderator: May 5, 2017
  18. Jun 16, 2011 #17


    User Avatar

    Staff: Mentor

    I can change thread title, but each post has its own tile - and these will be left. I am not sure if it makes sense.
  19. Jun 16, 2011 #18
    That TEPCO data & post you link to are about the first hour or so after the earthquake hit. For a guide to what may have happened on the 14th and 15th, I use some other TEPCO & NISAs analysis, mostly via the english version of report to IAEA.


    Table on page 23 of Attachment IV-1 tells us:

    9.20pm on 14th they open 2 SRV's and reactor pressure decreases.
    11pm it is presumed that the 1 SRV was closed because reactor pressure increased.

    The 2nd graph on page 35 shows these reactor pressure trends.

    Graphs on page 36 show the D/W and S/C pressures, both measured and those predicted by the model. Note that S/C pressure readings start to go down well before the explosion, at a time when D/W pressure readings are going up.

    Graph on page 32 of the document shows us the reason they have also assumed a D/W leak really early on, approx 21 hours after earthquake. This version of graph does not factor in this assumption, and as a result the model says that D/W and S/C pressures would have gone shooting up to very high levels days earlier. The other graphs I already mention do have the assumption of 21 hour D/W damage in them, and as a result the model D/W and S/C pressure values track the ones they actually measured pretty well.

    Note that I have chosen analysis case 2 because thats the one that reflects what they learnt about bad water level readings at reactor 1, implying less water at the other reactors. This leads to conclusions such as RPV damage having happened. Case 1 was the more optimistic version of events where fuel was only partially melted and RPV damage didnt happen, so I've paid much less attention to it.

    Also note that this document is TEPCOs version of analysis, and their versions of the FP existence ratio graphs are not like the ones I posted recently, theirs have only around 1% of stuff getting outside containment.
    Last edited by a moderator: May 5, 2017
  20. Jun 16, 2011 #19
    Assuming both analyses represent reality, it would seem to imply that (some) radioactivity was released from unit 2 before the explosion in the unit (due to venting), and quite a lot more radioactivity was released after the explosion in unit2 -- however not from unit 2 but from one or more of the other units.

    Perhaps I do not fully understand the graph, but it bothers me that it seems to show that all the compartments had the fraction 0 of fission products, until the safety release was opened at about 78h. Also (caveat: judged visually) in the following period there are times when the sum of increases in compartmental fractions seem to be not matched to the sum of decreases.

    There's certainly a peak in the release rate at that time, but the graph also shows the release rate to remain high for days thereafter. in particular as regards I-131. If we think in accumulated releases, the release during the peak would seem dwarfed by the sum of releases followingly.
  21. Jun 17, 2011 #20
    Here's my proposed sequence:

    - quake, tsunami, blah blah
    - SRV starts cycling
    - a small D/W leak appears, steam goes who knows where, in any case some coolant loss happens.
    - S/C wall goes pop, pressure drops (you can see it on your graph), water level drops too
    - operators prop open two SRVs. Meltdown is now in progress.
    - huge pressure spike in the D/W
    - some of the new, hydrogen-laden steam makes it out of the S/C out of the water and through the crack. boom.
    - water starts gushing out of the S/C
    - a short time later, D/W pressure drops, as it is now venting to the outside, through the broken torus.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook