Several people have asked questions, both in comments and in direct email so I decided to write a follow-up post to address some of these things.
Nicholas: Very good questions.
Could the HPSI systems have failed due to the lower portions of the reactor buildings being inundated with seawater? I assume that these units are sealed and should not be affected by immersion.
Likely not. The pumps are located in the basement of the reactor building which is a sealed structure. The reactor building is not a pressure vessel but it sealed and maintained at a slight negative air pressure by the ventilation system so that any leaks will be contained and processed by the stack effluent treatment system.
The turbine pump isn’t sealed per se but the only exposed rotating part is the coupling between the turbine shaft and the pump shaft. Being in water would not affect that. The only thing that might be affected by immersion in water (which I don’t think happened because of the tsunami) would be contamination of the hydraulic fluid via the reservoir vent. I can’t quite recall details of the reservoir vent but I don’t believe that it would allow that to happen. I believe that there is a float valve on the vent that would close in the presence of water. Don’t take that last statement to the bank because I simply can’t remember clearly.
Should future reactors have separate battery backup for the exhaust fans, sufficient to run them for a week or more, so that even if the offsite and emergency backup generators are lost, hydrogen explosions will be mitigated?
Another excellent question. In the US at least, many lessons were learned from the TMI incident and these were detailed in RegGuide 1.97. (should be available online) One of the major issues was hydrogen control. Some hydrogen is produced during normal operation by the breakdown of water by radiation (radiolysis). This is normally handled by the off-gas processing system.
During an cladding failure event we now know that a large amount of hydrogen is produced. Therefore the following items were required of all US nuclear plants.
Hydrogen recombiner. This is a device containing a platinum catalyst that safely recombines the hydrogen with oxygen to make water again. Its location varies but in most plants it is located right outside the containment vessel and connected to the vessel by piping.
Hydrogen Ignition system. This consists of a series of glow-plug-like devices mounted at the high point of the containment vessel. Their purpose is to ignite the releasing hydrogen in real time so that a controlled burn can happen instead of an explosion. I have read but have not confirmed that the Japanese plants have this system. Of course, without any source of power, they’re useless.
I can envision that one of the lessons learned from this event might be the requirement for passive platinum catalytic igniter that don’t require electricity. This is purely my speculation.
If the TMI reactor area is too hot to enter 30 years later does that mean these three plants will need to be sealed off for a very long time too? Was the TMI core ever removed or is it still in place?
A little mis-understanding there. At TMI, only the containment building basement is so contaminated. At a couple of years after the event, the refueling deck where most of the work was done was cool enough for workers to put in an 8 hour day. I’ve been in there and observed only low radiation levels.
The TMI core was removed and transported to a DOE facility for research purposes. The rubble bed (intact fuel pellets) was removed via an underwater vacuum system. The solidified melted fuel was drilled with diamond drills and broken apart and the chunks removed, again to a DOE facility.
P Tufts: Thanks for the paper references. I read them but I think that they over-state the problem. A modern fuel pit contains many cores’ worth of fuel, most of it many years old. After a couple of years of water cooling the fission products have decayed enough that the fuel can be moved to dry storage casks. They had started that process at Fukushima.
So the only serious concern is the new spent fuel, perhaps as much as one core’s worth, depending on how much they change out during a refueling. This is normally a third of a core. Unit 4 was in a somewhat unique situation in that the entire core had been removed from the reactor so that reactor inspections could take place. However, this fuel was over a year old and so wasn’t screaming hot.
There just isn’t enough information available yet to know what is going on in the #4 fuel pit. The smoke may be cladding combustion but more likely it’s the high density boron-loaded fuel racks that hold the fuel getting hot enough to smoke. We’ll just have to wait and see on that issue.
NY: Sent me a note in email observing that the 9.0 shake was at the epicenter of the quake about 130km away and that the shake at the plant was lower. He sent this reference.
The problem is that we don’t yet know how hard the plant itself was shaken. There is a seismic monitoring system installed in the plant so the record should be known some time in the future. I do stand by my statement that the shake exceeded the plant’s design basis as evidenced by the spent fuel pool concrete wall collapse.
Some Other Thoughts
In my opinion, the GE Mk I plants in the US are in much better shape than the Japanese ones. For one thing, the RegGuide 1.97 lessons learned were required modifications to all plants. For a second thing, the US has run a much tighter nuclear ship than the Japanese. The Japanese have a record of loose regulatory controls and operating procedures. Note that I am NOT saying that this contributed to the current situation, though I do have to wonder about that spent fuel pit wall collapse.
One important standard that US plants comply with is that they have to be able to withstand a 500 year flood. This results in the diesel generators either being located on a high hill as with Sequoyah or the air intake and exhausts being extended far above the predicted flood level. The generators themselves are located in waterproof vaults.
I think that the Japanese utility is doing an excellent job of keeping us informed about what is going on. That’s another TMI Lesson Learned – that your media relations department has to be top notch
As I type this, they should have finished making a temporary power connection to Unit 2. This is very good. Once they get the pumps running again they can cool Unit 2 and then through piping interconnects, cool the other units. And fill the spent fuel pits.
It looks to me like the worst of the event is over. The radiation levels outside the plant are decreasing as the noble gas decays and blows away. My hat’s off to the job those guys have done. Now starts the recovery and cleanup.
I’ll post more if and/or when more interesting things develop.