ancona - these are worthy questions that deserve a better answer than I can give without more info.
I'd like to know the source of some of it. As in, if that source if full of it on some topic I'm expert on, then I know to ignore some other things they say - some doesn't add up, as I'll explain.
First, a quick course on fission - yes, I know I'm in some danger of writing a disorganized book here, but we're halfway there anyway.
Fission of U or Pu gives off roughly 2.5 or 2.8 neutrons per fission, most right away, some milliseconds later (a good thing or reactors would be hard to control). These neutrons are fast (we say the spectrum goes to about 2 million electron volts, or ev - hauling ass in layman's terms).
In what seems like it's counter-intuitive, this means that they have a very small "cross section" to create more fission - slow neutrons do better (many thousands/one) and are actually attracted into the nuclei. Only extremely fast neutrons (as are given off by some fusion reactions) can prompt-fission things like U-238. So, to get to a chain reaction, you (in general) need fuel enriched in either U-235 or Pu which can fission with slow neutrons.
Here's the issue. U-238 eats medium speed neutrons like pac man - huge cross section for medium speed ones, and this breeds it into Pu, but doesn't make heat to speak of. So after a fission, you have to find a way to slow them down to "thermal" speeds (.025 eV) before letting them back into any fuel that has U 238 in it (everything but bombs, in general, it's hard/expensive to get rid of and keep only the part you want).
This is why Fermi had to break up his fuel into "chunks" and also the "moderator" which is some substance (carbon in his case, light water in most modern reactors, heavy water in some the Canadians make - more on that later). Fermi used carbon, which was insanely difficult to purify well enough that in the process of slowing neutrons down by collisions with the carbon, they didn't get "eaten" by even extremely low levels of boron found in most sources of carbon - and since we have zero control over which way neutrons come out, you have to make a reactor of a certain volume to surface area such that the ones that escape out the sides aren't so many you still have enough left to get a chain reaction. In fact, carbon of the grade Fermi required is still a state secret. You can't buy it, and you can't find out how they did it. It only takes PPB of boron to ruin it for reactors. In a beaurocratic foul up I'm sure you'll find it easy to believe - I can buy and have bought heavy water, which is better anyway.
Light water, regular H2O, is a decent moderator, but you lose some neutrons to it too - sometimes one will be captured by one of the H's and make it into a D. This is why enriched fuel is required for light water reactors. Basically, the water moderator eats too many of the neutrons to make a natural uranium reactor ever go critical no matter what you do. It's actually quite a difficult task.
Heavy water is the best moderator we know about. D's (deuterium is one proton and one neutron and what I use for fuel here - for completely other reasons) just don't capture neutrons, but weighing only about twice a neutron, are really good at slowing them down in collisions. You want light atoms for this job. For example, bouncing a bb off a bowling ball merely deflects the bb without slowing it down much - almost no energy is imparted to the bowling ball in the "momentum is conserved" equation. The Canadians use this, since heavy water is so non-neutron absorbing that you don't even need to enrich the Uranium (remember, this was something we worried a lot about in WWII with the Germans and their heavy water plant up north - and did suicide missions to put a stop to). If you've got to do fission at all, I like the Canadian way better than most. Heavy water is nowadays not so expensive to make, particularly for them, since you can process regular water to get the 7 parts in 1000 heavy out, with enough electricity - and they have more electricity from hydro than they can pipe down here to sell us. So, like moonshine, you make your product more valuable per size, and ship that instead(!). Same idea.
So, you need chunked up fuel, separated by a moderator. This allows a neutron emitted by a chunk of fuel to slow down all the way to "thermal" or about .025 ev, before encountering any more fuel - most make it right out of a fuel pellet/rod at full speed, since nothing much absorbs fast neutrons, and it's the bb and bowling ball - mass 1 vs 235 in this case (or 238 or 9). Without these things (or using pure 235 or Pu and a different set of special cases) you can't have a chain reaction, period. As luck would have it, when a reactor gets hot - atoms are moving fast, and the doppler effect takes neutrons back out of "thermal" status to a higher energy that isn't as good at making a chain reaction - they are somewhat self limiting - a good thing.
Pure U235 or Pu239 are only used in bombs, as the expense is "out of this world" to get either one pure enough, and free of spontaneous neutron emitters (odd isotopes of Pu and U and Cf do that) - for reasons I probably shouldn't mention here. Too much spontaneous neutron emission means the bomb starts to go off before it's fully pushed together, and is a fizzle - like the last one the Norks tried. That or one other easy mistake is almost certainly the reason they don't really have a good bomb yet.
OK, back to reactors. You need a large volume to surface area ratio such that at least one of those 2.5-2.8 neutrons is successfully slowed, and is captured by a
fissile atom (U238 isn't fissile in this situation but does eat some too and is bred into Pu-239 after a beta decay, which is fissle). So a couple new rods (what idiot would put them in a water pool, they're not even very radioactive, make little heat etc etc, and there is danger in slowing any neutrons around them).
If this is sounding complex, man you have no idea - when you start taking into account the changing fuel composition, which parts of the reactor are hotter than others, how much neutron eating poison has built up in the fuel, and so on - it's hairy, and practically the entire periodic table gets involved. Only now we also have unstable isotopes that don't normally exist in nature, and we are not only worried about their chemistry, but also about their nuclear cross sections for eating neutrons and other issues.
Super-hairy. Probably about 100 people on earth actually "get it".
So back to 1. Fuel rods, whether U type or MOX (mixed oxides of U and Pu, used to use up all the spurious bomb parts the world thankfully thinks it no longer needs - we've already used all we could buy from Russia ourselves) are oxides of the metals, compressed into pellets, clad in Zirconium. Do we know what type of rods they had? I know they did use some MOX, but I don't know what they used in this particular set of reactors. It makes a small difference to some of the answers, as MOX is a little more fissile with fast neutrons than regular enriched U is.
The reason for the zirconium is that zirconium is just about the only element that doesn't absorb either fast or slow neutrons, that will also handle hot water and all the rest of the conditions - which are hairy, as mentioned.
Zirconium is a bit flammable - about like a lighter flint (which is misch metal, another thing, but they really are about the same re flammability) - nice sparks when filed.
Lighter flints decompose water at room temp - zirconium doesn't.
Exposure to air does nothing at all to it, and the fuel is super-hermetically sealed as stacked pellets in each rod. It has to be super hermetic, because you don't want water leaking in - but the more important reason is you don't want radioactive xenon or radon gas leaking out - you want them to decay in there to solids. These rods are designed therefore to withstand quite a lot of internal pressure. And external pressure too - you have to pressurize the primary cooling/moderating water quite a lot to keep it from boiling. You work around NASA? These things would be the best built things you've ever seen, no kidding.
There is absolutely no reason air is a problem, and dunking them in a pool is just foolish -
I'd question the source's reliability on that one. I've never seen a lighter flint spontaneously ignite (it's different stuff, but not so as you could tell with a file or grinding wheel) - the filed off bits burn, that's it. Very hard to light on fire in bulk without getting all of it white hot first, and maybe using pure oxygen or fluorine. Yellow hot water will "burn it" but that's a rather hard case to achieve - possible, and it looks like it happened somehow there - normally a pressure relief would have (and probably did) open up first, so not much water (which at that point would be extremely thin vapor, not much of it) was around, and
it's likely that's why we can even talk about this at all - most of the rods didn't burn through or melt, in other words, which could have been quite the mess (not a bomb, just a hell of a mess). Again, with no water (moderator), no fission, when there's U-238 in the mix. So a meltdown is just that - you get a hot pool of fuel, that stays hot a long time, but that's it.
If they have a whole reactor's worth of new rods in a pool, and spaced them just so - yes, you could get a chain reaction, but "just so" doesn't happen by accident with this particular class of materials (you can have a mistake with pure Pu, easily, however, which is (one reason) why NO ONE EVER uses it for anything but weapons, and uses at most a few percent in reactors). But here's what happens - heat is generated, the water boils (it's a pool, not a pressure vessel) and that removes the moderator, the chain reaction is self regulating and stops since steam is a bad moderator - not dense enough - the bubbles don't moderate. In fact, there's a reactor design called
the "water boiler" that does just this - it's highly enriched U nitrate dissolved in water, and the boiling keeps it under control automatically. A couple universities had (and may still have) them as very hot neutron sources.
I'd question (but without a definitive source, I can't necessarily heap scorn on) that whole idea.
I'd have to ask a fission reactor operator if they ever, ever, put new fuel rods under water, and why. It's kind of a don't ask and I won't tell (or reveal I don't know) business.
I doubt they do. Before any fission byproducts build up, U and Pu mainly emit alpha rays that don't make it through the cladding. Pure Pu will get warm from it's natural alpha decay rate (fast). But that's all - warm to the touch. A mox rod might not even get warm, and certainly isn't very radioactive outside the rod - that's for Hollywood, not real life. Zirconium is pretty air-proof.
That one sets off my BS detector, is all I can say for now.
2. Hell, they can't get close enough for a good look...of course they can't locate them. This stuff is evil! Yes, I'm sure the cores are leaking - water - which may have nasty stuff in it (obviously it does) from a few rods that melted or burned through (more likely but it doesn't matter how a hole got there if there is one). This is, I'm pretty sure, the source of the radioactive "hot" water they are trying to keep in all those makeshift tanks, and trying to filter the nasty stuff out so they can recycle it both it and the water. It does not bode well for the ground water nearby.
However, in an accidental test, a US bomb research/mauf facility had a similar leak - right after WWII, of very similar (actually, nastier) stuff, it went into the ground and since WWII has made it ~300 yards. Ground water is not a river flowing fast. By the time they figure it makes it off-campus, it'll be safe again(!). Remember, no one lives near Fuk any more and won't for quite awhile, it's not a huge worry till they do.
The ocean would be both a better and worse place for that - it's big and would dilute it such that you really had to work hard (evaporate maybe a few tons of ocean water, remove all the salt, then measure to even see it). But before that dilution, near Japanese shores, it could do some real damage to the ecosystem (which would be short lived and soon buried under silt and safe again).
And the ocean is a rocket compared to ground water in re motion and mixing. Basically, there's so much ocean that by the time you're a few hundred miles off Japan, it won't matter one bit, but closer in, not so great. Hah, the only guys who really hate the Japanese and might toss a bomb at them are too close for comfort to do that!
Yes, decontaminating that water is a pure bitch (pardon my french, it's called for here). First you have particulates, that's pretty easy. But then you have soluble compounds, like say cesium hydroxide. There might be some radioactive iodine, but that's mostly already decayed (and this is an issue I have with my sodium iodide gamma ray scintillators - exposure to a lot of neutrons makes some of the I radioactive, and you see the line whether an external source is there or not, you have to wait months for it to cool down if exposed to too many fusor runs for example). But it's been many months, and I assume by now the iodine threat is a thing of the past.
So, zeolite (selective filtering), chemical reactions that precipitate things out, stuff like that, are required. Since fission is a semi-random process - there are a lot of different elements in the by-products, and more are created when the radioactive byproducts beta-decay (the usual way for them, since there are too many neutrons for those smaller atoms, sometimes one emits an electron and becomes a proton.)
What I heard, though, is that in fact they are filtering the water well enough for their needs, and what happened is someone (or some animal perhaps) knocked open a drain valve, and that's our current issue. Saw the pic of that - it's just a common ball valve with a lever handle, easy to upset. Oops.
In fact, under much less stress, I made that mistake today myself, getting water for my new system - I left the ball valve on my "go get water tank" half open and didn't realize it till the tank was half full, since I was pumping it in faster than it was leaking out...then I saw water pouring out of my tailgate and did the O s**t thing.
Not all that water is "incredibly radioactive". The raw stuff is (right after coming out of the core), and after what I understand is pass 1 of 3, it's still pretty hot - and that's what leaked if my source is correct. After 2, not so bad, after 3, ready to run back through, and safe to be around and get splashed with and so on. I believe, but do not know for certain that these three passes are of various types for the reasons I gave above - some mechanical "dirt" filtering, some zeolite type stuff, some chemical precipitation reactions and a final filtering so the nasty precipitates don't get washed out into the "pure" stuff.
At least that's what I think is true. Your first question gives me some questions I don't have answers to - it doesn't make sense to me, but it might to someone I could ask. I've never, ever heard of fresh fuel put into water for the heck of it. It's a dumb idea to bring it near any moderator till you're ready to stuff it in the core. That's nuke 101, not grad student stuff.
Not that these guys haven't pulled a few serious boners, as I started off saying. The only reason I could think of is they might have some automated process for putting things in and out of pools and like just using the machines, but it's still VERY stupid, though not for the fear-monger reasons.
The old rods make a ton of heat - from a 1gw reactor (billion watt) at our end-of-life for fuel, the rods decay of fission byproducts make 8 Mw (millions watts) heat so need artificial cooling. The Japanese go to around 12 Mw from decay before pulling them (they say), which is not smart but saves money. That's why they have the pools.
That number is not for one rod, it's for all of them.
Oh, one pertinent fact - old fuel isn't as reactive, fission wise, as new, even though it's hot as hell.
The reason is, some of the byproducts eat neutrons (they're called reactor poisons for that reason).
Almost all the reactors designed in the past 40 or so years (these were earlier, so I dunno) use a boron compound in water to control the reaction rate, using less as the fuel ages due to those poisons - you don't need to add as much artificial poison (borax, basically) to keep things in line.
Very little is controlled by control rods, from the beginning till now. They just do that final 1% or so adjustment in things, though Fermi's reactor, being first, had many that would auto-drop (for a certain geiger counter reading) since they didn't know enough to be sure they weren't making a bomb - they didn't know about non-prompt neutrons then. But yes, if you "scram" a reactor, it quits fissioning, which they claim did work, and for now, I believe that - it's so hard to make it work and so easy to make it not be critical, it just doesn't take much to shut one off.
That is, as far as fission goes. But if you've been running awhile - there's all those super radioactive byproducts that don't need fission, just normal beta decay - to make heat/energy/nasty dna damage if you get that in you. That's our 8 megawatt limit, or better stated, the reason for that.
We don't want to have those turkeys just melt when we pull them out. Remember, though, that a reactor has a lot of rods that those 8mw are distributed around.
Background info - the reason we and almost everyone uses boron to control the reactor is that putting a control rod partway into one messes up the distribution of fission - near the rod, it stops. You get non-uniform fuel usage that way, it's a real problem to keep under control if part of your reactor is a lot more chain-reactive than some other part. Trying to use zillions of tiny rods, well, that's just not practical or reliable. It's easy to re-fit (and I thought they all were) for using boron compound in the primary water loop, which is under high pressure and not boiling due to that - you just add some borax to it. That water in the primary loop becomes part heavy water due to some neutron capture, and very slightly radioactive, but not so bad. It's still the reason they use two loops, one to get the heat out of the core, highly pressurized and so non-boiling, and another loop that uses tubes full of that primary coolant as the heater of a boiler in what is otherwise a conventional steam power plant. Normally, then in the event of some sort of accident with the primary loop (like it getting exposed to unclad used fuel) then won't propagate a problem, and it's
relatively easy to fix in that case.
Couldn't get this up here, so I loaded it to a "secret" part of my site. 4 pages from Halliday saying more or less the same, but from the real deal, not just me. And in fact, this was the textbook used to train most Japanese Nuclear Physicists. I highly recommend it - too bad they didn't read it well. Actually, they probably did. They just don't put physicists in the job of controlling things like this.
http://www.coultersmithing.com/data/Icons/Halliday1.gif
http://www.coultersmithing.com/data/Icons/Halliday2.gif
I know ancona will have fun searching through the coultersmithing.com/data directory, particularly the /icons subdir (mostly my collection of funnies/satirical pix). Go for it guys, but if you find anything copyrighted, you didn't get it from me, right?