(no subject)
So, this is what I've gleaned regarding Japan's Fukushima Dai-ichi. Most of it is boring, but that's a good thing. When things about reactors become exciting, that's when you should start being worried.
I should say, that I'm not a nuclear engineer. I do, however, have the same qualifications it seems as a lot of people writing articles for the media, which is to say a Bachelor's degree and the ability to read Wikipedia. So I wrote this to make myself learn the material, and try to keep my head straight. There are probably factual errors. In fact, I guarantee there are factual errors. You should take the word of real nuclear engineers over me.
Furthermore, this is not in any way up to date. The best place to find updates is probably on Facebook, from the page for the International Atomic Energy Agency, or (if you trust them) from the English press release page for the Tokyo Electric Power Company (TEPCO) (both are slower then normal news, but more accurate).
Fukushima Dai-ichi One is a Boiling Water Reactor (BWR), a type of reactor that is used widely in the civilian sector because it's very safe (for comparison, see RBMK, the graphite-type reactor famous for its disastrous disaster recovery, which fell out of favor post-Chernobyl). In a boiling water reactor rods of nuclear fuel are placed into the reactor chamber and surrounded by water. The heat from the nuclear reactions are transferred to the water, which then circulates to the turbine room. In the turbine room the coolant exchanges its heat with another external cooling system while it drives the turbines. In this way the water cools the reactor, and produces power, at the same time.
(There's a diagram here which I can't make display properly)
The reactor works because of the fissile nature of heavy matter. What this means is that extremely heavy atoms (the ones you rarely see, like uranium), tend to be unstable, in the same way that a rock balancing on top of a mountain is unstable. If you kick it off it ends up speeding up and releasing a lot of energy. It does this by splitting up into smaller components, often alpha particles (two protons and two neutrons), and rapid neutrons. The neutrons, having been splintered off of one of the atoms, go spinning off and slamming into other atomic nuclei, which then themselves undergo fission, sending even more neutrons to go and knock apart other atoms. For this to happen there has to be a "critical" amount of fuel and interaction in the reactor. Without enough fuel, the reaction burns itself out (the neutrons get absorbed faster then they create new fissions).
The fuel for these reactors is formed into rods. The rods are then covered with a layer of cladding to prevent direct contact of the fuel with water, and are inserted into the reactor. All of this is then monitored very carefully by people in the control room. Unlike in sci-fi, there is no overload button for the reactor. But there are a lot of Off buttons.
Here's what I think is a description of the emergency procedures:
First, the moment the reactor picked up seismic tremors, it SCRAMed. SCRAM is an acronym of uncertain origin (Wikipedia likes the term Safety Control Rod Axe Man - from the days when the safety system was suspended above the reactor, and could be inserted by cutting the emergency rope), but it means that the reactor is "shut down". What happens is that the control rods are inserted rapidly into the reactor.
The control rods are large rods of neutron-absorbent material that are inserted between the fuel rods. This isolates the fuel rods, absorbing neutrons flying between fuel rods and preventing neutrons from one fuel rod from causing a reaction in another fuel rod. This slowly kills the reaction - as fuel decays more neutrons end up being absorbed then starting new reactions, slowly reducing the number of reactions.
SCRAMing immediately puts the reactor into what's called a "sub-critical" state, that is the reactor is now decreasing in energy output. The SCRAM system is mechanical. It's actually several times mechanical - the primary system has several mechanical backups (in the case of failure). In some reactors I think the control rods are actually positioned above the reactor - in catastrophic failures, they just fall into the reactor.
This worked in Japan - following the earthquake, every reactor SCRAMed. Energy output immediately started decreasing. And that's where everything went totally and completely wrong.
A BWR is a bit like an electric stove that you use to boil water in an aluminum pan. When they SCRAMed the reactors, they essentially pulled the plug. But the burner coils were still hot. Like an electric stove, they cool down with time, but in the case of nuclear reactors the time is measured in days instead of seconds or minutes. During the cooldown time you don't want to empty the water out of your pan - you'll end up with a melted pan. A melted reactor is even worse.
In case of a "standard" emergency SCRAM you can keep cycling water through the reactor - after all, that's how you've been keeping it cool all this time. The problem is that to cycle water properly you need to pump it into the reactor. To pump it into the reactor, you need pumps. To run pumps, you need power. Normally this isn't a problem - Fukushima alone has six reactors, you can probably usually get power from one of them. If the whole plant gets shut down, then you can connect to the outside grid (nuclear power plant cooling systems have grid priority one would assume).
Of course, if you have an earthquake over magnitude 7.0 then chances are you probably won't have a power grid either. So now you're screwed.
If power can't be restored, you can do what the Japanese did and fall back on the backups, including the Emergency Core Cooling System (ECCS). The ECCS is installed on every reactor (yes, even Chernobyl had an ECCS), and in a BWR can just consist of a system to inject more water into the reactor core, since that was what you were using for coolant anyway. The ECCS is not just one system, it's several systems, each designed to somehow cool the core. Any one of them can be effective.
A diagram of just how many bloody injection and release systems there are is included:
Unfortunately if you've reached this point, the core is probably starting to heat up. This is bad, because it means you're starting to overheat, but it's also bad because it means that the water in the core is now turning into steam, and like a kettle, it's beginning to pressurize. Pressure is bad, because it's as difficult to insert water into a high-pressure container as it is to blow into a high-pressure balloon.
Because of this, the ECCS needs to be powered (remember, it takes days to cool down a reactor). With the grid gone, you have to go to backups. Generally, an ECCS can have up to three backup sources if external and internal power is gone. There's a steam backup that can be used (I don't know if Fukushima had one, but the pipe could have been broken). If that fails, there are diesel generators standing by. Normally there are multiple diesel generators, each one large enough to run the ECCS. If both of those fail, there's a battery backup.
At Fukushima, there either was not a steam system, or the pipes broke. The diesel generators turned on, and then after an hour or so, cut out for unknown reasons probably dealing with debris in the water cooling supply, and the fact that the housings were hit by a goddamned tsunami. That left the battery backups. There were about eight to ten hours of power in the battery backups, designed to hold until emergency power arrived. That's gone now.
So, if you lose power to all your pumps, then what do you do? Well, what I would do is call 911.
What? I'm serious.
Think about it. You need to pump pressurized water into the reactors. To do that, what you want is some sort of truck or something that carries a large pump you can use to pressurize water. Who has trucks with high-pressure water pumps? In some plants, as I understand it, you can basically drive the fire trucks up to the reactor building and basically plug them into the system.
But what if that fails? What if every method to pump water into the reactor core fails?
Well then, you have a problem. The water already in the reactor starts to heat and turn to steam. As it does, the steam escapes the reactor (this is intentional, just like steam escaping a tea kettle). As the steam escapes, the water level is also dropping. This exposes the fuel rods to air, which means that there's nothing to cool them down. This starts the meltdown cycle.
Well, maybe. You see, there's no real good definition of a meltdown. The easiest is that it means that something in the reactor is starting to melt. Without any water around them, there's nothing to cool the fuel rods, and they start to heat up. The first thing that gives is the cladding around the fuel cells. Once that's gone the fuel itself is directly exposed. Once it begins to melt it begins to drip, like candle wax. Very hot, very heavy, and very radioactive candle wax. Obeying the call of gravity, the molten fuel drips to the bottom of the reactor chamber.
Since there's no set definition, there's no real way to tell how bad this is. You could have a few drops of hot wax on your table. Or your entire candle can melt and run down into the carpet.
Once you're in crisis mode, your main concern is to keep as much radiation as possible on the site of the reactor. There are three sources of radiation. The first is the actual reactor components - over the years the parts of the reactor get somewhat radioactive. This is usually no concern unless the reactor itself explodes (which so far hasn't happened - more on the explosions later). Second is the water that goes through the reactor core. The water is of course irradiated during its time in the core creating radioactive steam. Normally this steam is contained, but if it gets released it does have some low level of radioactivity. Fortunately a lot of it is fairly benign because water is made of light elements, and even when irradiated light elements don't decay in the same catastrophic way that heavier elements do.
The biggest source of radiation and concern is the actual fuel rods. They are chalk-full of heavy radioactive elements. If they get out of the plant (as was the case in Chernobyl via massive explosion), you're looking at a serious problem. There are basically two ways that this happens - first, the rods melt and burn through the reactor housing and into the ground. Second, the heat in the reactor chamber causes the pressure to increase to the point where the reactor vessel itself explodes like an overheated pressure cooker.
So it's time to move on to plan B - containment.
To understand how reactions are contained, you have to look at the reactor itself:
The Fukushima reactors are older BWRs, contained in what's called a Mark I Containment system (or sometimes a drywell). There are three (at least) layers of containment here. The reactor core itself is placed inside a reactor vessel. In Chernobyl, this reactor vessel (the red thing) was then put in a hole in the ground in the middle of a large warehouse and left there. In these older, less safe Japanese reactors, the reactor vessel is made of thick steel (15cm by some accounts) and then put inside of a separate steel containment vessel. The steel containment vessel, which is shaped like a giant metal test tube, is then put inside the concrete drywell wall, which is shaped like a giant lab flask, and sealed inside. The drywell functions to isolate the reactor vessel from the pump room. Steam from inside the drywell is pumped out and into the wetwell, the torus (the ring) around the reactor. The wetwell is filled partway with water, and the steam vents are under the surface, so steam from the inner reactor is injected into cold water, causing rapid cooling. Everything is then placed inside a containment building, an airtight (in theory) negative pressure structure built around the whole complex (in American Pressurized Water Reactors the containment building is built to withstand impacts from, say, hijacked airliners filled with fuel. I think it's much the same in Japan).
In order to get out into the open air, the fuel has to melt. Then it has to melt its way out of the reactor core. Then it has to melt its way out of the steel containment vessel. Then it has to melt its way out of the concrete drywell. Then it has to melt its way through the secondary containment building. Then you're in trouble. But there is a lot of concrete, steel, and more concrete between anything that comes out of the reactor and the open air. So except for very small leaks even a partial meltdown should be safely contained. Even if the internal pressure vessel violently ruptures, the particles in the steam have to be forcefully propelled through several layers of protection. So containment should keep things in check.
Unless you do something unexpected, like blow up the secondary containment building.
The official explanation, which I think is reasonable enough that it's probably true, runs something like this. As the reactor overheats, you build up steam. The steam is slightly radioactive and so what you do is keep it inside the secondary containment building, the building that houses the reactor. Unfortunately, if the reactor really overheats, you get a lot of steam, and you start to overcome the safety parameters of the building. This leads to the release of radioactive steam in controlled bleed offs, like we've been seeing since the whole affair started.
But as the cladding overheats on the fuel rods, it can produce, in contact with water, hydrogen gas. The hydrogen gas escapes in the same way as the steam and is trapped in the secondary containment building. However, hydrogen gas is not steam, when exposed to oxygen and a spark it explodes, which means that once you get hydrogen in the containment building sooner or later you get a big boom. This is apparently a known problem, and modern reactors come with hydrogen igniters and other safety systems (see this brochure from Westinghouse for an example list of all modern safety features in PWRs), but it was either unknown or unsolved at the time they built Fukushima Dai-ichi.
At this point you begin to sense some panic from Japan's nuclear techs. They had systems prepared for if power failed, if cooling failed, and even if all the redundant cooling failed. But nobody really asked the question "What if all the cooling systems fail, and then someone blows up the building?" There might have been plans for it, but whatever they were went by the wayside as someone finally decided to push the panic button.
Up until now the plan had been to contain and filter the radiation in the containment building, and slowly restore cooling to the reactors. With the containment building gone and the cooling equipment largely non-functional (on account of being exploded), they went into panic mode one. In a hurry you can actually reduce core breach chances, and provide emergency cooling, simply by flooding the reactor plant. In some modern plants this is done through the use of a massive on-site water reserve. Fukushima has a plan B - the plant is right on the ocean, a nearly limitless supply of cold water.
This is not an approach favored by the owners of nuclear plants, because as anyone who has owned a car in a region of the world where they salt the roads knows, salt causes havoc with mechanical parts. Normally the reactor is cooled with distilled water. Cooling it with salt water corrodes a lot of the machinery, and can do significant damage as it causes steel parts to corrode. On the other hand, meltdowns also cause significant damage, and quite frankly if the worst problems in the reactors at the end of this is saltwater corrosion, the engineers will die happy. It's not like they're going to try restarting the reactor.
Mixed in with the salt water is boronic acid, mostly to put some boron in the water. Boron absorbs neutrons, the little particles that zoom around causing collisions, essentially turning the cooling water into a control rod, and increasing the speed at which the reactor cools off. Once you start seawater injection the reactor is basically ruined until they do full maintenance, but it should cool things off in a hurry, and at least keep steel containment intact.
This is where we now seem to be on unit one and unit three. Both of them have lost their secondary containment buildings, but the steel containment shell seems to be intact. It's possible that they drywell concrete shell also remains, in which case there are still at least two safety layers between the overheating core and the outside world. Unit two is in a more perilous condition, having possibly achieved a partial meltdown when water levels dropped, resulting in the fuel rods being exposed. With the rods being exposed to air for possibly several hours, the temperature rose dramatically and its quite possible that some of the fuel started to melt. They have since exploded the containment building on that reactor too, although that was expected ever since they got hydrogen in the building.
Of the other nuclear power plants, Onagawa declared a state of emergency when radiation was detected outside the plant. Radiation levels have since returned to normal and it is speculated that the radiation was from Fukushima. No evacuations have been ordered, and no equipment failure reported. Fukushima Daini is in a much more precarious state. Equipment failure resulted in partial loss of cooling, although cooling has since been restored and off-site power is available. As of the latest update I have, no venting has been ordered indicating that not only is the reactor cooling, but it's cooling well enough to avoid steam buildup. Three of its four reactors have already been reported as entering cold shutdown.
So far the reports indicate that all three steel containment vessels are intact. It's also possible that the concrete walls that actually house the reactor are intact. In that case things are relatively contained - not to mention being cooled by seawater.
But what about radiation leaks?
As you may remember from high school chemistry, there are three types of radiation, alpha, beta, and gamma, to which I'm going to add another, neutron. You get these types of radiation when unstable, radioactive nuclei, decay. In a reactor you start with a large number of unstable radioactive nuclei, in the fuel. You also create some, the water used for cooling absorbs neutrons, turning Oxygen in Nitrogen-16 (see the note in this NYTimes article).
Before you can tell how dangerous nuclear radiation is, you have to know the following: what kind it is, what products were left behind, and how fast it decays (releases radiation). Big, heavy fissile elements, like uranium, are dangerous because not only do they decay, but they're big enough to release a lot of particles when they decay, and some of the products that they decay into are also big enough to be highly radioactive. Lighter radioactive isotopes are more likely to decay into stable, and very boring products.
How fast is also key. Decay times are measured in half-lives, which is the amount of time it takes for half of the material to undergo decay. Nitrogen-16 has a half life of about seven seconds, which means that if it's floating in the air, by the time it reaches you, it's already decayed. More dangerous radioactive elements like the cesium and the iodine that come out of the core, have half-lives measured in years, which makes them more dangerous, because it takes centuries for those deposits to become as harmless as Nitrogen-16 is after a minute. Of course, too long a half-life and it basically doesn't decay at all - things like lead have a ridiculously long half-life.
And finally, you worry about what type of radiation. Gamma radiation is composed of photons, and high-energy photons are dangerous. Not as dangerous as high-energy electrons, which have mass as well as energy, and are known as beta radiation. Alpha, which is thousands of times heavier then a beta particle, consists of two protons and two neutrons, and is absolutely devastating to your cells.
In order to protect yourself against extremely harmful alpha radiation (which is usually produced by massive isotopes, like the ones used in nuclear fuel), you need to shield yourself with something thick enough to absorb the radiation. Like, for instance, a T-shirt. Or a piece of paper. Or, actually, your skin. Alpha radiation is often unable to penetrate the dead skin cells on the surface of your skin. Alpha radiation is extremely damaging because it's heavy - it destroys everything it hits. The corollary of that is that everything it hits destroys it, including running into your dead skin cells. Beta radiation also does damage (though not quite as much) and also has mass, you have to upgrade to several layers of aluminum foil to shield yourself from beta radiation (a wall will also do the trick).
So if radiation is so easily blocked, why do we worry about it? Because if you get it in the air, the stuff gets everywhere. Take one possible source of radiation, Iodine 131, which decays within a matter of days via beta decay (into Xenon-131, which is stable), releasing a beta particle. If you just cover over a deposit of Iodine 131, you'll shield out the Beta radiation. But what if it gets into the air? Then you have Iodine particles floating everywhere. Chances are if you're in the area, you'll breath some of them. They'll get absorbed into your body, get into your blood and accumulate in your thyroid. Then, some days later, they'll decay, releasing their beta particles directly into your thyroid, irradiating you directly without any need to pass through shielding.
Unless you take iodine pills, which flush the stuff from your body pretty fast. But iodine pills don't work on Cesium 137 (another beta emitter), which can stick around for years, slowly building up in your body. And of course enough direct exposure to alpha and beta radiation will make you pretty sick (also enough gamma, which blasts right through you). Which is why Japan is repeatedly emphasizing the fact that the reactor vessels are still intact. While there may be traces of Cesium and Iodine on the outside, the vast majority of the fuel is still sealed away in a giant steel chamber, in a giant concrete chamber, and so far isn't going anywhere.
In order to get a high level nuclear disaster, that stuff needs to get in the air. In Chernobyl the reactor exploded with enough force to rip off the 2000 ton covering plate, exploded again, an then caught on fire, spreading the remains of the reactor all over the atmosphere. Without the helping hand of a few large-scale explosions, well, it's a lot less impressive. Even if you do have core breach and containment breach, a lot of the stuff stays on the ground. The rest of it tends to float around within a few miles of the reactor, settling to the ground, unless the wind blows it somewhere (in this case out to sea). The end result is devastating, but it's not Chernobyl level devastating - unless you manage to blow up your reactor in the process.
But even minimal radiation leaks can have serious side effects. Radiation exposure from the environment (even if you don't breath it in) can cause radiation sickness. Certainly, you don't want to be exposed to high levels of background radiation. The energy constantly bombarding your cells will eventually kill you...if nothing else does first.
Radiation levels at the station perimeters are reported as having risen to a level consistent with several millirem (mrem) an hour. This is clearly above background, and in our radiation training, a dose of ten mrem was considered a safety-related event. Ten mrem is estimated to have permanent effects on your health approximately consistent with smoking 1.4 cigarettes (approximately one in a million chance of causing your death). The radiation dosage at that level is approximately consistent with the radiation exposure you receive on an airplane flight. It's so low that it's difficult to measure health effects (see this note by the NRC who point out that the entire city of Denver is in a high-radiation zone and hasn't shown much in the way of symptoms except in airport planning).
Unfortunately, that doesn't mean that you would be safe. Acute exposure to even small amounts of radiation can have health effects (although nobody knows how much). The human body is chaotic, and sometimes incredibly delicate. So to keep people away from that, they've evacuated the area.
And there are spikes. The blast that destroyed unit two sent radiation spiking up to 800 mrem or so, which is very, very dangerous. No word on whether this has reduced after the first breach - if it doesn't it probably means that they have a leak somewhere, and that they're streaming radiation (although not actually that much, and hopefully it's mostly from the steam).
Of course, if the core breeches, you'll have radioactive material outside of the core. That's the long term fear. That something will leak out of the core of one of the reactors and become so pervasive, so long-term, that it will be impossible to get rid of the stuff. It will simply lurk there in the background, waiting for some unwary person to get too close. But for now the prospects of the entire reactor exploding Chernobyl style are still very remote.
So, here's the summary in a nutshell.
The reactors all SCRAMed immediately. That killed the ongoing chain reaction. Since then the intensity of nuclear decay within the reactors has been steadily decreasing. Every hour that passes reduces the intensity of any remaining radiation (even if it does increase the total heat). The explosions at the reactors were in the pumping equipment, caused by hydrogen gas buildup. This has kept the engineers from being as successful at screening out radiation from steam, hence the larger evacuation zone, but the radiation from the steam is short-lived and not likely to be long-term harmful. The second explosion was predicted before it happened, as they now know something of what to look for. Despite the explosions, the interior containment layers (the plural is important) remain intact in all the reactors. This has kept the vast majority of the fuel isolated from the environment in an airtight steel case. The biggest cause of radiation leakage is probably from the steam, which has picked up N-16, tritium and some trace particles of cesium and iodine. The N-16 should be mostly harmless given a large evacuation zone. The Cesium and Iodine contributions have yet to be measured. All three reactors are now having boron enriched seawater pumped into them, which seems to be holding as a cooling method, although there's now a leak in Unit Two. So far, shielding has held. If there was a major radiation leak of the Chernobyl level, everyone would know (Canada would know by now, and I think they're watching). The biggest threat is that the reactor will overpressurize despite constant seawater pumping, and explode. If the reactor can somehow undergo a steam explosion it can vent the internal reactor steam, including particles directly from the fuel rods, into the atmosphere. This would require it to breach the reactor core, the steel containment vessel, and then the concrete containment vessel before steps could be taken to blow off steam, or in the worst case to just concrete the reactor. This would require a rapid rise in pressure, either due to a failure of seawater pumping, a failure in equipment, and a failure in containment, probably all at once. If I had to pick a doomsday scenario this would be it - and it would still be unlikely. The other possibility is that the fuel will melt down fast enough to burn through the reactor shell. This was stopped at Three Mile Island because the fuel did not melt fast enough, but it might succeed here. In that case the fuel would probably end up exposed to air, but first it has to burn through several layers of containment - before someone can come up with some clever way to stop it. Several of the reactors might already have experienced a partial meltdown. The fuel might have melted and dripped to the bottom of the reactor vessel. So far it doesn't seem to have melted its way out, which is sort of anti-climactic (and hopefully will stay that way). To repeat, so far the containment of the fuel has held. And to repeat, the reactors SCRAMed successfully. That means that the primary reaction was killed, and with every minute that passes the energy output of the reactors gets lower, and the radiation level drops. So cross your fingers.
I'm hoping things get better. So far the past two days seems to have been a long, painful exercise in hanging on by the fingernails. Now it looks like the wetwell on Unit Two was damaged during the explosion that destroyed their containment system. Also, it appears that part of the building for Unit Four is on fire (this is less of a concern, Unit Four was shut down for maintenance at the time of the earthquake, and so there's nothing to cool in the meantime). Unit Two was already having problems - it's hot enough that they have problems pumping the water in. Meanwhile, Kan has moved the evacuation zone now to twenty kilometers, and the stay-at-home to thirty.
I'm going to post this and hope that things get better, and that we keep hanging on. I think they have a fighting chance. I think that even if they do have a breach, that the effects will be mostly minimal when compared to the effects of the rest of the tsunami. But I'm not all that confident anymore. Hopefully optimism will prove to be correct here.
I should say, that I'm not a nuclear engineer. I do, however, have the same qualifications it seems as a lot of people writing articles for the media, which is to say a Bachelor's degree and the ability to read Wikipedia. So I wrote this to make myself learn the material, and try to keep my head straight. There are probably factual errors. In fact, I guarantee there are factual errors. You should take the word of real nuclear engineers over me.
Furthermore, this is not in any way up to date. The best place to find updates is probably on Facebook, from the page for the International Atomic Energy Agency, or (if you trust them) from the English press release page for the Tokyo Electric Power Company (TEPCO) (both are slower then normal news, but more accurate).
Fukushima Dai-ichi One is a Boiling Water Reactor (BWR), a type of reactor that is used widely in the civilian sector because it's very safe (for comparison, see RBMK, the graphite-type reactor famous for its disastrous disaster recovery, which fell out of favor post-Chernobyl). In a boiling water reactor rods of nuclear fuel are placed into the reactor chamber and surrounded by water. The heat from the nuclear reactions are transferred to the water, which then circulates to the turbine room. In the turbine room the coolant exchanges its heat with another external cooling system while it drives the turbines. In this way the water cools the reactor, and produces power, at the same time.
(There's a diagram here which I can't make display properly)
The reactor works because of the fissile nature of heavy matter. What this means is that extremely heavy atoms (the ones you rarely see, like uranium), tend to be unstable, in the same way that a rock balancing on top of a mountain is unstable. If you kick it off it ends up speeding up and releasing a lot of energy. It does this by splitting up into smaller components, often alpha particles (two protons and two neutrons), and rapid neutrons. The neutrons, having been splintered off of one of the atoms, go spinning off and slamming into other atomic nuclei, which then themselves undergo fission, sending even more neutrons to go and knock apart other atoms. For this to happen there has to be a "critical" amount of fuel and interaction in the reactor. Without enough fuel, the reaction burns itself out (the neutrons get absorbed faster then they create new fissions).
The fuel for these reactors is formed into rods. The rods are then covered with a layer of cladding to prevent direct contact of the fuel with water, and are inserted into the reactor. All of this is then monitored very carefully by people in the control room. Unlike in sci-fi, there is no overload button for the reactor. But there are a lot of Off buttons.
Here's what I think is a description of the emergency procedures:
First, the moment the reactor picked up seismic tremors, it SCRAMed. SCRAM is an acronym of uncertain origin (Wikipedia likes the term Safety Control Rod Axe Man - from the days when the safety system was suspended above the reactor, and could be inserted by cutting the emergency rope), but it means that the reactor is "shut down". What happens is that the control rods are inserted rapidly into the reactor.
The control rods are large rods of neutron-absorbent material that are inserted between the fuel rods. This isolates the fuel rods, absorbing neutrons flying between fuel rods and preventing neutrons from one fuel rod from causing a reaction in another fuel rod. This slowly kills the reaction - as fuel decays more neutrons end up being absorbed then starting new reactions, slowly reducing the number of reactions.
SCRAMing immediately puts the reactor into what's called a "sub-critical" state, that is the reactor is now decreasing in energy output. The SCRAM system is mechanical. It's actually several times mechanical - the primary system has several mechanical backups (in the case of failure). In some reactors I think the control rods are actually positioned above the reactor - in catastrophic failures, they just fall into the reactor.
This worked in Japan - following the earthquake, every reactor SCRAMed. Energy output immediately started decreasing. And that's where everything went totally and completely wrong.
A BWR is a bit like an electric stove that you use to boil water in an aluminum pan. When they SCRAMed the reactors, they essentially pulled the plug. But the burner coils were still hot. Like an electric stove, they cool down with time, but in the case of nuclear reactors the time is measured in days instead of seconds or minutes. During the cooldown time you don't want to empty the water out of your pan - you'll end up with a melted pan. A melted reactor is even worse.
In case of a "standard" emergency SCRAM you can keep cycling water through the reactor - after all, that's how you've been keeping it cool all this time. The problem is that to cycle water properly you need to pump it into the reactor. To pump it into the reactor, you need pumps. To run pumps, you need power. Normally this isn't a problem - Fukushima alone has six reactors, you can probably usually get power from one of them. If the whole plant gets shut down, then you can connect to the outside grid (nuclear power plant cooling systems have grid priority one would assume).
Of course, if you have an earthquake over magnitude 7.0 then chances are you probably won't have a power grid either. So now you're screwed.
If power can't be restored, you can do what the Japanese did and fall back on the backups, including the Emergency Core Cooling System (ECCS). The ECCS is installed on every reactor (yes, even Chernobyl had an ECCS), and in a BWR can just consist of a system to inject more water into the reactor core, since that was what you were using for coolant anyway. The ECCS is not just one system, it's several systems, each designed to somehow cool the core. Any one of them can be effective.
A diagram of just how many bloody injection and release systems there are is included:
Unfortunately if you've reached this point, the core is probably starting to heat up. This is bad, because it means you're starting to overheat, but it's also bad because it means that the water in the core is now turning into steam, and like a kettle, it's beginning to pressurize. Pressure is bad, because it's as difficult to insert water into a high-pressure container as it is to blow into a high-pressure balloon.
Because of this, the ECCS needs to be powered (remember, it takes days to cool down a reactor). With the grid gone, you have to go to backups. Generally, an ECCS can have up to three backup sources if external and internal power is gone. There's a steam backup that can be used (I don't know if Fukushima had one, but the pipe could have been broken). If that fails, there are diesel generators standing by. Normally there are multiple diesel generators, each one large enough to run the ECCS. If both of those fail, there's a battery backup.
At Fukushima, there either was not a steam system, or the pipes broke. The diesel generators turned on, and then after an hour or so, cut out for unknown reasons probably dealing with debris in the water cooling supply, and the fact that the housings were hit by a goddamned tsunami. That left the battery backups. There were about eight to ten hours of power in the battery backups, designed to hold until emergency power arrived. That's gone now.
So, if you lose power to all your pumps, then what do you do? Well, what I would do is call 911.
What? I'm serious.
Think about it. You need to pump pressurized water into the reactors. To do that, what you want is some sort of truck or something that carries a large pump you can use to pressurize water. Who has trucks with high-pressure water pumps? In some plants, as I understand it, you can basically drive the fire trucks up to the reactor building and basically plug them into the system.
But what if that fails? What if every method to pump water into the reactor core fails?
Well then, you have a problem. The water already in the reactor starts to heat and turn to steam. As it does, the steam escapes the reactor (this is intentional, just like steam escaping a tea kettle). As the steam escapes, the water level is also dropping. This exposes the fuel rods to air, which means that there's nothing to cool them down. This starts the meltdown cycle.
Well, maybe. You see, there's no real good definition of a meltdown. The easiest is that it means that something in the reactor is starting to melt. Without any water around them, there's nothing to cool the fuel rods, and they start to heat up. The first thing that gives is the cladding around the fuel cells. Once that's gone the fuel itself is directly exposed. Once it begins to melt it begins to drip, like candle wax. Very hot, very heavy, and very radioactive candle wax. Obeying the call of gravity, the molten fuel drips to the bottom of the reactor chamber.
Since there's no set definition, there's no real way to tell how bad this is. You could have a few drops of hot wax on your table. Or your entire candle can melt and run down into the carpet.
Once you're in crisis mode, your main concern is to keep as much radiation as possible on the site of the reactor. There are three sources of radiation. The first is the actual reactor components - over the years the parts of the reactor get somewhat radioactive. This is usually no concern unless the reactor itself explodes (which so far hasn't happened - more on the explosions later). Second is the water that goes through the reactor core. The water is of course irradiated during its time in the core creating radioactive steam. Normally this steam is contained, but if it gets released it does have some low level of radioactivity. Fortunately a lot of it is fairly benign because water is made of light elements, and even when irradiated light elements don't decay in the same catastrophic way that heavier elements do.
The biggest source of radiation and concern is the actual fuel rods. They are chalk-full of heavy radioactive elements. If they get out of the plant (as was the case in Chernobyl via massive explosion), you're looking at a serious problem. There are basically two ways that this happens - first, the rods melt and burn through the reactor housing and into the ground. Second, the heat in the reactor chamber causes the pressure to increase to the point where the reactor vessel itself explodes like an overheated pressure cooker.
So it's time to move on to plan B - containment.
To understand how reactions are contained, you have to look at the reactor itself:
The Fukushima reactors are older BWRs, contained in what's called a Mark I Containment system (or sometimes a drywell). There are three (at least) layers of containment here. The reactor core itself is placed inside a reactor vessel. In Chernobyl, this reactor vessel (the red thing) was then put in a hole in the ground in the middle of a large warehouse and left there. In these older, less safe Japanese reactors, the reactor vessel is made of thick steel (15cm by some accounts) and then put inside of a separate steel containment vessel. The steel containment vessel, which is shaped like a giant metal test tube, is then put inside the concrete drywell wall, which is shaped like a giant lab flask, and sealed inside. The drywell functions to isolate the reactor vessel from the pump room. Steam from inside the drywell is pumped out and into the wetwell, the torus (the ring) around the reactor. The wetwell is filled partway with water, and the steam vents are under the surface, so steam from the inner reactor is injected into cold water, causing rapid cooling. Everything is then placed inside a containment building, an airtight (in theory) negative pressure structure built around the whole complex (in American Pressurized Water Reactors the containment building is built to withstand impacts from, say, hijacked airliners filled with fuel. I think it's much the same in Japan).
In order to get out into the open air, the fuel has to melt. Then it has to melt its way out of the reactor core. Then it has to melt its way out of the steel containment vessel. Then it has to melt its way out of the concrete drywell. Then it has to melt its way through the secondary containment building. Then you're in trouble. But there is a lot of concrete, steel, and more concrete between anything that comes out of the reactor and the open air. So except for very small leaks even a partial meltdown should be safely contained. Even if the internal pressure vessel violently ruptures, the particles in the steam have to be forcefully propelled through several layers of protection. So containment should keep things in check.
Unless you do something unexpected, like blow up the secondary containment building.
The official explanation, which I think is reasonable enough that it's probably true, runs something like this. As the reactor overheats, you build up steam. The steam is slightly radioactive and so what you do is keep it inside the secondary containment building, the building that houses the reactor. Unfortunately, if the reactor really overheats, you get a lot of steam, and you start to overcome the safety parameters of the building. This leads to the release of radioactive steam in controlled bleed offs, like we've been seeing since the whole affair started.
But as the cladding overheats on the fuel rods, it can produce, in contact with water, hydrogen gas. The hydrogen gas escapes in the same way as the steam and is trapped in the secondary containment building. However, hydrogen gas is not steam, when exposed to oxygen and a spark it explodes, which means that once you get hydrogen in the containment building sooner or later you get a big boom. This is apparently a known problem, and modern reactors come with hydrogen igniters and other safety systems (see this brochure from Westinghouse for an example list of all modern safety features in PWRs), but it was either unknown or unsolved at the time they built Fukushima Dai-ichi.
At this point you begin to sense some panic from Japan's nuclear techs. They had systems prepared for if power failed, if cooling failed, and even if all the redundant cooling failed. But nobody really asked the question "What if all the cooling systems fail, and then someone blows up the building?" There might have been plans for it, but whatever they were went by the wayside as someone finally decided to push the panic button.
Up until now the plan had been to contain and filter the radiation in the containment building, and slowly restore cooling to the reactors. With the containment building gone and the cooling equipment largely non-functional (on account of being exploded), they went into panic mode one. In a hurry you can actually reduce core breach chances, and provide emergency cooling, simply by flooding the reactor plant. In some modern plants this is done through the use of a massive on-site water reserve. Fukushima has a plan B - the plant is right on the ocean, a nearly limitless supply of cold water.
This is not an approach favored by the owners of nuclear plants, because as anyone who has owned a car in a region of the world where they salt the roads knows, salt causes havoc with mechanical parts. Normally the reactor is cooled with distilled water. Cooling it with salt water corrodes a lot of the machinery, and can do significant damage as it causes steel parts to corrode. On the other hand, meltdowns also cause significant damage, and quite frankly if the worst problems in the reactors at the end of this is saltwater corrosion, the engineers will die happy. It's not like they're going to try restarting the reactor.
Mixed in with the salt water is boronic acid, mostly to put some boron in the water. Boron absorbs neutrons, the little particles that zoom around causing collisions, essentially turning the cooling water into a control rod, and increasing the speed at which the reactor cools off. Once you start seawater injection the reactor is basically ruined until they do full maintenance, but it should cool things off in a hurry, and at least keep steel containment intact.
This is where we now seem to be on unit one and unit three. Both of them have lost their secondary containment buildings, but the steel containment shell seems to be intact. It's possible that they drywell concrete shell also remains, in which case there are still at least two safety layers between the overheating core and the outside world. Unit two is in a more perilous condition, having possibly achieved a partial meltdown when water levels dropped, resulting in the fuel rods being exposed. With the rods being exposed to air for possibly several hours, the temperature rose dramatically and its quite possible that some of the fuel started to melt. They have since exploded the containment building on that reactor too, although that was expected ever since they got hydrogen in the building.
Of the other nuclear power plants, Onagawa declared a state of emergency when radiation was detected outside the plant. Radiation levels have since returned to normal and it is speculated that the radiation was from Fukushima. No evacuations have been ordered, and no equipment failure reported. Fukushima Daini is in a much more precarious state. Equipment failure resulted in partial loss of cooling, although cooling has since been restored and off-site power is available. As of the latest update I have, no venting has been ordered indicating that not only is the reactor cooling, but it's cooling well enough to avoid steam buildup. Three of its four reactors have already been reported as entering cold shutdown.
So far the reports indicate that all three steel containment vessels are intact. It's also possible that the concrete walls that actually house the reactor are intact. In that case things are relatively contained - not to mention being cooled by seawater.
But what about radiation leaks?
As you may remember from high school chemistry, there are three types of radiation, alpha, beta, and gamma, to which I'm going to add another, neutron. You get these types of radiation when unstable, radioactive nuclei, decay. In a reactor you start with a large number of unstable radioactive nuclei, in the fuel. You also create some, the water used for cooling absorbs neutrons, turning Oxygen in Nitrogen-16 (see the note in this NYTimes article).
Before you can tell how dangerous nuclear radiation is, you have to know the following: what kind it is, what products were left behind, and how fast it decays (releases radiation). Big, heavy fissile elements, like uranium, are dangerous because not only do they decay, but they're big enough to release a lot of particles when they decay, and some of the products that they decay into are also big enough to be highly radioactive. Lighter radioactive isotopes are more likely to decay into stable, and very boring products.
How fast is also key. Decay times are measured in half-lives, which is the amount of time it takes for half of the material to undergo decay. Nitrogen-16 has a half life of about seven seconds, which means that if it's floating in the air, by the time it reaches you, it's already decayed. More dangerous radioactive elements like the cesium and the iodine that come out of the core, have half-lives measured in years, which makes them more dangerous, because it takes centuries for those deposits to become as harmless as Nitrogen-16 is after a minute. Of course, too long a half-life and it basically doesn't decay at all - things like lead have a ridiculously long half-life.
And finally, you worry about what type of radiation. Gamma radiation is composed of photons, and high-energy photons are dangerous. Not as dangerous as high-energy electrons, which have mass as well as energy, and are known as beta radiation. Alpha, which is thousands of times heavier then a beta particle, consists of two protons and two neutrons, and is absolutely devastating to your cells.
In order to protect yourself against extremely harmful alpha radiation (which is usually produced by massive isotopes, like the ones used in nuclear fuel), you need to shield yourself with something thick enough to absorb the radiation. Like, for instance, a T-shirt. Or a piece of paper. Or, actually, your skin. Alpha radiation is often unable to penetrate the dead skin cells on the surface of your skin. Alpha radiation is extremely damaging because it's heavy - it destroys everything it hits. The corollary of that is that everything it hits destroys it, including running into your dead skin cells. Beta radiation also does damage (though not quite as much) and also has mass, you have to upgrade to several layers of aluminum foil to shield yourself from beta radiation (a wall will also do the trick).
So if radiation is so easily blocked, why do we worry about it? Because if you get it in the air, the stuff gets everywhere. Take one possible source of radiation, Iodine 131, which decays within a matter of days via beta decay (into Xenon-131, which is stable), releasing a beta particle. If you just cover over a deposit of Iodine 131, you'll shield out the Beta radiation. But what if it gets into the air? Then you have Iodine particles floating everywhere. Chances are if you're in the area, you'll breath some of them. They'll get absorbed into your body, get into your blood and accumulate in your thyroid. Then, some days later, they'll decay, releasing their beta particles directly into your thyroid, irradiating you directly without any need to pass through shielding.
Unless you take iodine pills, which flush the stuff from your body pretty fast. But iodine pills don't work on Cesium 137 (another beta emitter), which can stick around for years, slowly building up in your body. And of course enough direct exposure to alpha and beta radiation will make you pretty sick (also enough gamma, which blasts right through you). Which is why Japan is repeatedly emphasizing the fact that the reactor vessels are still intact. While there may be traces of Cesium and Iodine on the outside, the vast majority of the fuel is still sealed away in a giant steel chamber, in a giant concrete chamber, and so far isn't going anywhere.
In order to get a high level nuclear disaster, that stuff needs to get in the air. In Chernobyl the reactor exploded with enough force to rip off the 2000 ton covering plate, exploded again, an then caught on fire, spreading the remains of the reactor all over the atmosphere. Without the helping hand of a few large-scale explosions, well, it's a lot less impressive. Even if you do have core breach and containment breach, a lot of the stuff stays on the ground. The rest of it tends to float around within a few miles of the reactor, settling to the ground, unless the wind blows it somewhere (in this case out to sea). The end result is devastating, but it's not Chernobyl level devastating - unless you manage to blow up your reactor in the process.
But even minimal radiation leaks can have serious side effects. Radiation exposure from the environment (even if you don't breath it in) can cause radiation sickness. Certainly, you don't want to be exposed to high levels of background radiation. The energy constantly bombarding your cells will eventually kill you...if nothing else does first.
Radiation levels at the station perimeters are reported as having risen to a level consistent with several millirem (mrem) an hour. This is clearly above background, and in our radiation training, a dose of ten mrem was considered a safety-related event. Ten mrem is estimated to have permanent effects on your health approximately consistent with smoking 1.4 cigarettes (approximately one in a million chance of causing your death). The radiation dosage at that level is approximately consistent with the radiation exposure you receive on an airplane flight. It's so low that it's difficult to measure health effects (see this note by the NRC who point out that the entire city of Denver is in a high-radiation zone and hasn't shown much in the way of symptoms except in airport planning).
Unfortunately, that doesn't mean that you would be safe. Acute exposure to even small amounts of radiation can have health effects (although nobody knows how much). The human body is chaotic, and sometimes incredibly delicate. So to keep people away from that, they've evacuated the area.
And there are spikes. The blast that destroyed unit two sent radiation spiking up to 800 mrem or so, which is very, very dangerous. No word on whether this has reduced after the first breach - if it doesn't it probably means that they have a leak somewhere, and that they're streaming radiation (although not actually that much, and hopefully it's mostly from the steam).
Of course, if the core breeches, you'll have radioactive material outside of the core. That's the long term fear. That something will leak out of the core of one of the reactors and become so pervasive, so long-term, that it will be impossible to get rid of the stuff. It will simply lurk there in the background, waiting for some unwary person to get too close. But for now the prospects of the entire reactor exploding Chernobyl style are still very remote.
So, here's the summary in a nutshell.
I'm hoping things get better. So far the past two days seems to have been a long, painful exercise in hanging on by the fingernails. Now it looks like the wetwell on Unit Two was damaged during the explosion that destroyed their containment system. Also, it appears that part of the building for Unit Four is on fire (this is less of a concern, Unit Four was shut down for maintenance at the time of the earthquake, and so there's nothing to cool in the meantime). Unit Two was already having problems - it's hot enough that they have problems pumping the water in. Meanwhile, Kan has moved the evacuation zone now to twenty kilometers, and the stay-at-home to thirty.
I'm going to post this and hope that things get better, and that we keep hanging on. I think they have a fighting chance. I think that even if they do have a breach, that the effects will be mostly minimal when compared to the effects of the rest of the tsunami. But I'm not all that confident anymore. Hopefully optimism will prove to be correct here.