Japan Earthquake: Nuclear Plants at Fukushima Daiichi

[joewein]

The progress of the water decontamination is a bit puzzling.
TEPCO reports that 80,000 cubic meters have been decontaminated and over 30,000 desalinated, but the stored volume yet to be treated, at 107,000, is down less than 20,000 from its peak of about 125,000. At this rate, it will take several years to drain the water from the plant, despite the addition of SARRY.

It is to be expected that the total amount of contaminated water will not decrease by as much as is treated: Much of the desalinated water goes right back into the reactors (close to 400t a day), where it gets contaminated again and from which it leaks back into the basement. That's Tepco's current concept of circulation cooling, and not much will change about it until the containments can be repaired, for which there really is no viable plan yet.

A certain amount of cooling water will evaporate into the atmosphere from inside the buildings, at least until the covers are built. The steel frame around unit 1 looks complete, from what I see on the http://www.tepco.co.jp/en/nu/f1-np/camera/index-e.html" today. Now they'll have to add the plastic boards and repeat the whole process for two other units.

Meanwhile, the waste build up will accelerate as the spent SARRY columns add to the AREVA sludge and Kurion zeolites. That pile of very nasty material will need tending for many decades.
Are there any specifics on the storage space used? TEPCO does claim that this problem was adequately provided for, but no detail or photographs have been released, afaik.

When fuel melts into corium, about 2/3 of the radioactive inventory evaporates. Much of it will will end up in cooling water at some stage. There were hundreds of tons of fuel in the three cores. Expect those terabecquerels to gradually accumulate as Areva/Kurion/SARRY sludge. This is one big difference with TMI-2: There, most of the nasty stuff stayed inside until the RPV was opened for cleanup, while in Fukushima Daiichi the radioactivity gets washed into basements as long as they keep cooling it.

 

[tsutsuji]

http://mainichi.jp/photo/archive/news/2011/09/09/20110909k0000m040167000c.html Interview of Hiroaki Koide, assistant professor at the Kyoto University Research Reactor Institute. He makes a distinction between unit 1 for which Tepco reported in May after adjusting the water level gauge that "there is already no more water in the core", and units 2 and 3 for which there is no such report. If some unmelted fuel remains in units 2 and 3, a cooling failure could result in a steam explosion. If what Tepco reported about unit 1 is correct, "pouring water is nearly meaningless", and the priority is to build the ground water shielding wall. That the leaks consist only in the hole in concrete near unit 2 water intake that was repaired in April "is impossible". "The nuclear plant site is covered with concrete, so it can be thought that concrete was broken by the earthquake or the tsunami. A concrete which does not break is impossible". The concrete hole that was found is "the tip of the iceberg". (About cold shutdown) "Tepco itself is saying the RPV bottom has fallen out". When the roadmap was created in April, with the "cold shutdown" goal, it was under the assumption that "there is fuel". "Now that this assumption has collapsed, I am asking the meaning of keeping the cold shut down goal". An abbridged version is available in English at http://mdn.mainichi.jp/perspectives/news/20110909p2a00m0na016000c.html

 

[Dmytry]

Yes the /s is important as well - if that is the right unit. Add that to my list of errors :tongue:. I have done calculations of activation rates from neutron flux, and well as participated in neutron irradiation of vessels and internals. I have forwarded the summary of the study to othe colleagues and none of us have been able to figure out how they reached their numbers and conclusions.

Specifically, they counted their samples for 4 hours and came up with 1502 atoms of S-35. Is that the peak activity - 1502 decays per second (Bq) of S-35? Or is it 1502 total atoms of S-35 detected over the four hour count? With that kind of uncertainty should it be n/m^2*s or /hr or /fortnight? All we can tell for sure is that whatever they measured was higher than they have seen previously and the concluded that the source of that peak was Fukushima.

I think paper says that ~1500 atoms per cubic metre is what they got in air at San Diego... I didn't look closely right now but in general you put a lot of air through a filter, then measure activity of the filter over some hours, then calculate the concentration.

Any help deciphering that study would be appreciated. At a minimum they should get a big fat F- on the abstract or summary they wrote. Any ideas, Dmytry?

Dunno, going to look over it more closely this or probably next weekend.

For the flux I think it's not per time but total. The whole thing looks like if they are outside of their field when it comes to reactors.
The graph does not look like some global event that globally makes more sulphur-35 which subsequently decays exponentially, but more like a cloud that passes over measurement site.

 

[etudiant]

I think paper says that ~1500 atoms per cubic metre is what they got in air at San Diego... I didn't look closely right now but in general you put a lot of air through a filter, then measure activity of the filter over some hours, then calculate the concentration.

Dunno, going to look over it more closely this or probably next weekend.

For the flux I think it's not per time but total. The whole thing looks like if they are outside of their field when it comes to reactors.
The graph does not look like some global event that globally makes more sulphur-35 which subsequently decays exponentially, but more like a cloud that passes over measurement site.

Afaik, Scripps is more an oceanographic/atmospheric research institute, so their focus would be on the airborne dissemination of Fukushima emissions. The corresponding oceanographic survey was done by Woods Hole earlier this year. ( https://www.whoi.edu/page.do?pid=67796). It does seem that this survey was quite narrowly focused, as the radioactive sulfur is a minute component of the overall emissions from the site.
Perhaps the narrow focus reflects a specific work assignement, for instance completion of a PhD thesis.
Imho. a bit of broader background would have elevated this to a A paper.

 

[LabratSR]

Something about the core re-melting 10 days after the accident

http://www.tepco.co.jp/en/nu/fukushima-np/images/handouts_110909_04-e.pdf

 

[clancy688]

Wait, what? First fuel elements outside the pressure vessel and now remelting ten days after the tsunami? But I don't quite understand what the handout wants to tell us... is it disproving the theory of a remelt? Or is it actually proving one?

 

[LabratSR]

Wait, what? First fuel elements outside the pressure vessel and now remelting ten days after the tsunami? But I don't quite understand what the handout wants to tell us... is it disproving the theory of a remelt? Or is it actually proving one?

Yeah, that's what I can't figure out.

 

[luciablue]

I'd be more inclined to believe in recriticality if iodine levels exceeded cesium levels again and sampling around the site perimeter spiked, which it hasn't really since the end of March:

http://fukushima.grs.de/sites/default/files/Entwicklung_ODL.jpg"

what do you make of this:

Iodine spike August 25th
http://fukushima-diary.com/2011/09/breaking-news-fukushima-in-recriticality/

Yellowknife had a very small I-131 detect on August 31st
http://www.hc-sc.gc.ca/hc-ps/ed-ud/respond/nuclea/data-donnees-eng.php#cntbtns

 

[LabratSR]

Report to NISA in relation to the impact of Tohoku-Chihou Taiheiyo-Oki Earthquake to Nuclear Reactor Facilities at Fukushima Daiichi Nuclear Power Station (Japanese language) English Translation to come - maybe...hopefully


http://www.tepco.co.jp/cc/press/betu11_j/images/110909m.pdf

"English translations of the reports and the appendixes are now being
developed and it takes a while to complete them.
We will post the translations one by one when it is prepared.
(The documents written in Japanese below will be replaced by English
translations.)
We apologize for this inconvenience caused."

 

[joewein]

what do you make of this:

Iodine spike August 25th
http://fukushima-diary.com/2011/09/breaking-news-fukushima-in-recriticality/

Yellowknife had a very small I-131 detect on August 31st
http://www.hc-sc.gc.ca/hc-ps/ed-ud/respond/nuclea/data-donnees-eng.php#cntbtns

2300 Bq/kg in sludge looks like a good size spike and I frankly don't know what to make of it.

If it came from Fukushima it would have to have been in rainwater, which in Tokyo is closely linked to drinking water: Most of the drinking water comes from surface water in the mountains west of here. However, no spikes were detected testing the tap water. The last time any I-131 was detected was 4 months ago (early May):
http://monitoring.tokyo-eiken.go.jp/monitoring/w-past_data.html

We should be 22 half lives down the decay of I-131 by now, a factor of about 1:4,000,000 reduction of quantity since the accidents.

One theory mentioned on ex-SKF in July was that some hospitals were dumping I-131 used for diagnostic purposes into the sewage system. I'm not sure how plausible that is.
http://ex-skf.blogspot.com/2011/07/iodine-131-still-detected-in-sewage.html

 

[joewein]

http://www3.nhk.or.jp/news/genpatsu-fukushima/20110907/index.html Confirming that Daiichi units 1,2,3,4 will be decommissioned, Tepco president Toshio Nishizawa said the restarting or decommissioning of Daiichi units 5,6 and Daini four units will be "decided taking prefecture and local opinion into account".

They are really reluctant to simply say: "No, we won't restart units 5 and 6 or any of Daini."

From a practical point of view, radiation levels near the 4 damaged units are pretty high. The measuring point "South of Central Building", only a few hundred metres from 1F5 and 1F6, is recording over 300 uSv/h, or 5000 times Tokyo levels.

Even if you ignore the fact that 4 out 5 diesels were knocked out at units 5+6 and they have water in their basements, they are not going to be the greatest work environments for the engineers having to run them, if Tepco were to ever start them up again.

 

[Bodge]

Occam's razor and the circumstantial evidence strongly suggest that recriticality has occurred somewhere at Fukushima Daiitchi.

I have lost track of the number of posts I've made on this 'theory'.


I would also like to know what has and is going up the vent stack?

Are we talking about aerosolized actinides including polonium?

Are the releases ongoing - and at what levels?

Why isn't a helicopter collecting samples from the vents now?

 

[LabratSR]

Wait, what? First fuel elements outside the pressure vessel and now remelting ten days after the tsunami? But I don't quite understand what the handout wants to tell us... is it disproving the theory of a remelt? Or is it actually proving one?

According to Ex-SKF, the charts show that no re-melting occurred

http://ex-skf.blogspot.com/2011/09/tepco-dumps-565-page-report-on-early.html

 

[joewein]

The http://2.bp.blogspot.com/-nUcXAB3Vm...sv-E/s1600/fukushimareactor3coreremelt-4.JPG" showing a close match between rainfall at three different locations and spikes in radiation levels around March 21 makes a good case.

TEPCO then explains the drop of water injection to almost 0 in the chart from 3/21 to 3/25 with a switch in measuring instrument, arguing that it really, really kept injecting water all the time, it just wasn't being measured.

It leaves me wondering how they then explain the rise in temperatures at the RPV flange, from around 230 C on 3/20 to way over 500 C on 3/22 (one expects it would have been a lot hotter inside). Something doesn't quite add up.

 

[tsutsuji]

A certain amount of cooling water will evaporate into the atmosphere from inside the buildings, at least until the covers are built. The steel frame around unit 1 looks complete, from what I see on the http://www.tepco.co.jp/en/nu/f1-np/camera/index-e.html" today. Now they'll have to add the plastic boards and repeat the whole process for two other units.

According to the simulation video ( http://www.tudou.com/programs/view/Cw5FeFIJxeI/ ) the side panels are put first and the roof panels are added at the very end.
[URL]http://www.tepco.co.jp/en/news/110311/images/110910_1t.jpg[/URL]
picture taken on 9 September: large size: http://www.tepco.co.jp/en/news/110311/images/110910_1.jpg

http://www.asahi.com/national/update/0910/TKY201109100129.html The removal of debris from the top part of unit 3 is starting today. It will take until next summer to remove an estimated 3500 m³ volume. This is done using remote-controlled machines installed on the top a U-shaped workstand (8 m wide, 30 m high, total length 150 m) surrounding the building. Each debris will be checked for radiations and sorted according to the radiation level. Debris removal will begin at unit 4 in the middle decade of this month. The radiation being lower, this will be performed with manned machines, while spreading a dust fixing agent with an elevated water spreading truck. To prevent debris from falling into the pool, the pool will be covered with a floating cover.

http://www.tepco.co.jp/en/nu/fukushima-np/images/handouts_110909_05-e.pdf Tepco press release on units 3 and 4 debris removal (English)

Report to NISA in relation to the impact of Tohoku-Chihou Taiheiyo-Oki Earthquake to Nuclear Reactor Facilities at Fukushima Daiichi Nuclear Power Station (Japanese language) English Translation to come - maybe...hopefullyhttp://www.tepco.co.jp/cc/press/betu11_j/images/110909m.pdf

"English translations of the reports and the appendixes are now being
developed and it takes a while to complete them.
We will post the translations one by one when it is prepared.
(The documents written in Japanese below will be replaced by English
translations.)
We apologize for this inconvenience caused."

It is a big report (565 pages, 25MB)

http://www.nikkei.com/news/category...39797E3E2E2E2;at=DGXZZO0195165008122009000000 This is the first time Tepco is providing the government with a report gathering details on the accident. Concerning the cores, it repeats previous statements. Concerning spent fuel pools, it says "it can be thought that the fuel was not exposed above the water level". At unit 4, pool temperature rose up to 90°C and water level decreased to 1.5 m above fuel.

 

[tsutsuji]

Something about the core re-melting 10 days after the accident

http://www.tepco.co.jp/en/nu/fukushima-np/images/handouts_110909_04-e.pdf

This is probably Tepco's answer to Fumia Tanabe's presentation at a meeting of the Atomic Energy Society of Japan (probably the annual/fall meeting in Kitakyushu, 19-21 September 2011) which was announced by Asahi at http://www.asahi.com/english/TKY201108080276.html (see also the diagrams on the Japanese language page at http://www.asahi.com/national/update/0807/TKY201108070330.html )

http://www.asahi.com/national/update/0909/TKY201109090584.html It was found by the investigation performed in August by Tepco, that the measuring instruments were being temporarily replaced/switched [?] between 21 and 25 March. Rather, adding pumps, the flow rate was being increased, as could be understood from the records - deemed to be close to the real values - taken by a worker who was working at supplying the pumps with oil.

http://www.yomiuri.co.jp/science/news/20110909-OYT1T01029.htm instead of 24 tons per day as was previously thought, the actual flow rate was 1600 tons per day or more.

http://www.tepco.co.jp/en/nu/fukushima-np/images/handouts_110909_06-e.pdf Tepco press release with the new March 21-24 flow rate data.

The http://2.bp.blogspot.com/-nUcXAB3Vm...sv-E/s1600/fukushimareactor3coreremelt-4.JPG" showing a close match between rainfall at three different locations and spikes in radiation levels around March 21 makes a good case.

As I said in April at https://www.physicsforums.com/showpost.php?p=3258585&postcount=4508 , the radiation peak starts at around 04:00 AM (maximum reached at 05:40 AM), but the rain starts after 07:00 AM on 21 March in Mito, Ibaraki prefecture. So it is not possible to say that the peak is created by the rain, at least in Mito city.

Source : http://www.bousai.ne.jp/vis/tgraph.php?area_id=108&post_id=1080000037

You may compare with the graph for Yokosuka, Kanagawa prefecture, where the (smaller) peak starts clearly after the rain :

source : http://www.bousai.ne.jp/vis/tgraph.php?area_id=114&post_id=1140000006

 

[hellbet]

Nice graph tsusuji, every single word from Tepco should be double-checked and that's quite a tough job ;) Keep up the good work!

 

[NUCENG]

I think paper says that ~1500 atoms per cubic metre is what they got in air at San Diego... I didn't look closely right now but in general you put a lot of air through a filter, then measure activity of the filter over some hours, then calculate the concentration.

Dunno, going to look over it more closely this or probably next weekend.

For the flux I think it's not per time but total. The whole thing looks like if they are outside of their field when it comes to reactors.
The graph does not look like some global event that globally makes more sulphur-35 which subsequently decays exponentially, but more like a cloud that passes over measurement site.

Ok, I now have the report and understand a lot more. The study uses the HYSPLIT NOAA model to confirm that the air in San Diego on March 28 included air that was over Fukushima on March 20. I have no problem with that. I am now convinced that the S-35 peak they measured included radiation from Fukushima. The problem is that unless they were in the peak of the plume their numbers for neutron release from Fukushima may be under-stated.

Next point- the HYSPLIT model was not used to calculate the concentrations of the S-35 along the plume trajectory. That was done wiith a moving box model that I am still working through. Here I have hit a snag in nomenclature. They discuss the removal processes and dilution in terms of "lifetimes" which I believe are the "average times for a particle to be removed by the process" rather than a half life as we are familiar with for radioactive decay. Are their any oceanographers or meteorologists out there that can confirm that terminology?

The study shows that dilution lifetime is the most sensitive parameter. A 10% change in this value creates a 20% change in results. That makes results highly suspect since they do not discuss how they set that input. In addition they seem to use a constant value of 4.9 days for this lifetime for all dilution paths modeled.

More to follow

edit: By the way if my belief is correct about the average lifetime I do know that half life is the ln(2) times the mean lifetime. I am proceding under the assumption that is a correct interpretation of their method.

 

[NUCENG]

...

I went back to the ORIGEN2 source term analysis I have quoted previously. My question was whether there were enough neutrons available to generate the amount of S-35 assumed in the study.

For a typical BWR at shutdown the rate of neutron emission from spontaneous fission and fission product decay is about 1 to 2E7 n/sec/MWt. I assumed 1.5E7 n/sec/MWt. Using the design thermal ratings of 1F1 to 1F3 only for the cores (1380 MWt, 2381MWt, and 2381 MWt). This would generate 9E10 n/sec.

Conservatively assume that the start of seawater injection was on 3/13 and that there was seawater exposed to the cores until the air mass left Fukushima enroute to San Diego. That would be 3/20 per the study (7 days). That gives a total of about 5E16 neutrons to produce S-35 and get it into the atmosphere with steam. According to the study they needed 2E8 neutrons to react with Cl-35 in [n,p] reactions that produced S-35.

Thus, ignoring the question about their use of a flux (n/m^2/s) there were physically enough neutrons to do the trick.

 

[Dmytry]

I went back to the ORIGEN2 source term analysis I have quoted previously. My question was whether there were enough neutrons available to generate the amount of S-35 assumed in the study.

For a typical BWR at shutdown the rate of neutron emission from spontaneous fission and fission product decay is about 1 to 2E7 n/sec/MWt. I assumed 1.5E7 n/sec/MWt. Using the design thermal ratings of 1F1 to 1F3 only for the cores (1380 MWt, 2381MWt, and 2381 MWt). This would generate 9E10 n/sec.

Conservatively assume that the start of seawater injection was on 3/13 and that there was seawater exposed to the cores until the air mass left Fukushima enroute to San Diego. That would be 3/20 per the study (7 days). That gives a total of about 5E16 neutrons to produce S-35 and get it into the atmosphere with steam. According to the study they needed 2E8 neutrons to react with Cl-35 in [n,p] reactions that produced S-35.

Thus, ignoring the question about their use of a flux (n/m^2/s) there were physically enough neutrons to do the trick.

well that's not very meaningful tbh. The study claims specific concentration at Fukushima, without specifying volume that should have had this concentration. That just doesn't make any sense.
The measurement at San Diego would have been same if at Fukushima it was 1 cubic metre with concentration q or 1000 cubic metres with concentration q/1000 , and conversely the result would differ by factor of 1000 if it was 1 or 1000 cubic metres with concentration q.
After such long distance, you can't tell the original concentration any more, you can only tell amount. It's like you drop some solution into a cup of water and stir. From concentration in the cup you won't be able to tell original concentration in the drop, only the amount.

edit: I think I know what they may have done, they applied it as if they were estimating solar wind fluence on the pole, getting the fluence per square metre of Earth area, for the fluence that is distributed over many square kilometres. Getting the rays per square metre all right in the end. Maybe someone here worked on something like that and can shed some light?

 

[Gary7]

Conservatively assume that the start of seawater injection was on 3/13

Was looking for information on this. According to the Japanese article below it says freshwater ran out on the 12th at 2:53pm, and they began injecting seawater at 7:04pm. They injected seawater until 7:25pm when they were told (basically) to wait for permission from the Prime Minister, which eventually came at 8:20pm.

So seawater basically began to be injected into reactor #1 on the evening of the 12th (Japan time).

http://www.asahi.com/special/10005/TKY201105260339.html

 

[NUCENG]

well that's not very meaningful tbh. The study claims specific concentration at Fukushima, without specifying volume that should have had this concentration. That just doesn't make any sense.
The measurement at San Diego would have been same if at Fukushima it was 1 cubic metre with concentration q or 1000 cubic metres with concentration q/1000 , and conversely the result would differ by factor of 1000 if it was 1 or 1000 cubic metres with concentration q.
After such long distance, you can't tell the original concentration any more, you can only tell amount. It's like you drop some solution into a cup of water and stir. From concentration in the cup you won't be able to tell original concentration in the drop, only the amount.

edit: I think I know what they may have done, they applied it as if they were estimating solar wind fluence on the pole, getting the fluence per square metre of Earth area, for the fluence that is distributed over many square kilometres. Getting the rays per square metre all right in the end. Maybe someone here worked on something like that and can shed some light?

Sorry, you missed my point.

If there weren't enough neutrons as they estimated reacted with Cl-35 it would have required an assumption of recriticality or their method was wrong. It was a simple check that doesn't prove their method, but could have disproved it.

As I said I am still working out their moving box model. The study says the size of the box was not significant and that is consistent with other uses of this moving box model I have found. But I have the same uneasy feeling you do about that.

I have two new questions.

1. They have been continuously monitoring cosmogenic S-35 at the Scripps pier since February 2009. Cosmic radiation and generation of spallation isotopes is affected by solar wind, sunspot activity and even weather that can cause atmospheric turnover. Solar wind and sunspot activity have a 22 year cycle. Further, even with the peak of March 28, the March-April average of S-35 at the pier was less than the averages in 2009 and 2010. This once again raises the question of how well they have characterixzed background levels of S-35.

2. I have found an article which questions the use of the large voolume atmospheric models for evaluating plumes such as aircraft contrails or a radiation point source plume.

Overall, I still think it is more than coincidence that they observed this peak S-35 concentration. But I'm still not sure I believe their method allows them to calculate neutrons released.

 

[tsutsuji]

http://www.tepco.co.jp/nu/fukushima-np/images/handouts_110912_04-j.pdf (not translated yet) on September 13-14 the Areva and Kurion systems will be shut down to perform the following maintenance : replacement of a pump at the Kurion system ; replacement of a mixer in the Areva system ; installation of the pump to carry water from the waste facility process main building basement to the underground tanks as shown on the other diagram: http://www.tepco.co.jp/nu/fukushima-np/images/handouts_110912_05-j.pdf . The SARRY system will keep running.

Tepco video showing the water injection system (from buffer tank to each unit's RPV)

 

[Bodge]

...
I would also like to know what has and is going up the vent stack?

Are we talking about aerosolized actinides including polonium?

Are the releases ongoing - and at what levels?

Why isn't a helicopter collecting samples from the vents now?

Anyone got any ideas?

 

[jim hardy]

"Overall, I still think it is more than coincidence that they observed this peak S-35 concentration. But I'm still not sure I believe their method allows them to calculate neutrons released"

as always - not quite enough information to be sure.

Given Cl35's 40 barn cross section
and Nuceng's spontaneous neutron production rate in a dirty core
i wouldn't ascribe the sulfur to anything more than the seawater.

Didn't somebody say there were tons of sea salt in the cores?
Assume they're uniform in three core volumes with seven days worth of Nucengs neutrons 5E16 or so?, and 40 barn cross section -
- and you only need 1E8 captures?

i think it'd calculate out to show only how remarkably sensitive are today's instruments.

i might try back of envelope calc later on - but i am awkward, and on a "honey-do" list today..

 

[zapperzero]

Anyone got any ideas?

Oh I got lots of ideas. No data, though. TEPCO said at some point quite late on that they would start monitoring the steam. They haven't got around to publishing any results.

 

[Caniche]

Occam's razor and the circumstantial evidence strongly suggest that recriticality has occurred somewhere at Fukushima Daiitchi.

I have lost track of the number of posts I've made on this 'theory'.I would also like to know what has and is going up the vent stack?

Are we talking about aerosolized actinides including polonium?

Are the releases ongoing - and at what levels?

Why isn't a helicopter collecting samples from the vents now?

Who knows ,the chimney is pretty irrelevant since the RPV got melted the fuel went walkies,any connection to the turbines was probably frazzled and containment of any kind was blown to smithereens .
As for ongoing releases ,nobody is talking . Iaea updates have petered out, RADNET and RIMNET went silent months ago ,TEPCO ,well there is no meltdown oh hang about. In short,the track record of official bodies when providing information leaves much to be desired in terms of accuracy and promptness.I'm guessing their taciturn approach is not a good sign.
But you already knew that

 

[robinson]

I would also like to know what has and is going up the vent stack?

I think any concerned party would want to know that. I want to know why it is impossible to simply find out. Or if they know, why is it top secret information?

 

[joewein]

I would also like to know what has and is going up the vent stack?
(...)
Why isn't a helicopter collecting samples from the vents now?

The blowers of the standby gas treatment system haven't worked since the blackout, so they aren't going to push anything up the stacks.

Somebody more knowledgeable please correct me if I'm wrong, but the stacks should not be a likely suspect for emissions at this stage. They would have been while the RPV and containment were under high pressure from steam and hydrogen and emergency venting took place from the wet well (S/C). Since then the pressure gradually dropped and some (or all) of the primary containments are suspected of no longer being able to hold pressure. The assumption is that gas can now leak either way, which is why nitrogen is being injected to try to ensure that the containment atmosphere remains inert.

TEPCO was monitoring radiation levels near the perimeter of the plant over several months after when they started nitrogen injections at the plant, to see if N2 flushes radioactive gases from the containment into the environment but apparently didn't detected anything ineresting:
http://www.tepco.co.jp/en/nu/fukushima-np/f1/index8-e.html

A lot of the radioactivity should now be contained inside the wet wells, either from venting or from being washed there via the downcomer tubes from the dry well, if cooling water leaks out of the bottom of the RPV or steam from the RPV condenses inside the dry well.

The decay heat should keep the S/C pretty warm. The 4000 mSv/h spot inside unit 1 was near steam from the S/C and there's a lot of gamma radiation coming up through the concrete floor of the first floor (1F) of the reactor buildings, which they're shielding with heavy steel plates to protect crews performing work inside (new instruments, etc).

 

[zapperzero]

http://www.power-eng.com/news/2011/09/1499357613/team-finds-possible-cause-of-fukushima-no-4-reactor-blast.html

Radiogenic hydrogen as a cause for explosion in #4 reactor - theory is being put forward by

team of researchers from the University of Tokyo, the Japan Atomic Energy Agency and others

. Apparently they did some small-scale experiment.

 

[LabratSR]

Recently released Tepco video showing the water injection system with English dubs.

http://www.youtube.com/user/AtomicPowerReview

 

[NUCENG]

http://www.power-eng.com/news/2011/09/1499357613/team-finds-possible-cause-of-fukushima-no-4-reactor-blast.html

Radiogenic hydrogen as a cause for explosion in #4 reactor - theory is being put forward by . Apparently they did some small-scale experiment.

Aw shucks, we were there back in May on PF!

AntonL found a DOE report and suggested this back on May 7:

https://www.physicsforums.com/showpost.php?p=3287847&postcount=6068

I did some additional calcs and concluded it was a possibility and would explain initial inspection results that fuel in SFP4 was not significantly damaged:

https://www.physicsforums.com/showpost.php?p=3329698&postcount=8496

Now there is an experiment that supports the phenomenon. Still isn't enough for proof, but it is good to know.

Recently they also determined that the Unit 4 side of the SBGT and hardened vent ducting showed relatively low contamination levels. This would also be consistent with detonation of radiolytic hydrogen without releasing a lot of the fission products from the pool.

 

[Caniche]

Recently released Tepco video showing the water injection system with English dubs.

http://www.youtube.com/user/AtomicPowerReview

"the boric acid solution will be injected immediately in the event of a re-criticality"verbatim hmmm that's one debate signed and sealed then, maybe

 

[zapperzero]

Now there is an experiment that supports the phenomenon. Still isn't enough for proof, but it is good to know.

More experiments
http://web.engr.oregonstate.edu/~petersr/RadiolyticEnhancementLiterature/Water%20Decomposition%20Paper%204%20Kalinichenko.pdf

 

[tsutsuji]

Unit 2:

At 2:59 pm on September 14, we started injecting water to the reactor
through core spray system water injection piping arrangement in addition
to the water injection through the reactor feed water system piping
arrangement.
Water is currently injected at approx. 4 m3/h through reactor feed water
system piping arrangement, and at approx. 1 m3/h through core spray
system water injection piping arrangement.
http://www.tepco.co.jp/en/press/corp-com/release/11091502-e.html

http://www.tepco.co.jp/en/nu/fukushima-np/images/handouts_110914_01-e.pdf more details about unit 2 cooling plan.