Monday, May 16, 2011

Fukushima and Epistemology

TEPCO is releasing more and more information about Fukushima. And it's not all getting into the news. I'm on a listserv that sees a bit more than the media makes available. I'm not going to go into the details in this post, though. Rather, I'd like to consider how we know what we think we know about Fukushima and how much it matters.

Last week we had an, er, meltdown in some quarters over the news that the core of reactor #1 seems to be more damaged than previously thought. The increase was from 70% damaged to 100% damaged, which hardly seems to me worth some of the breast-beating over MELTDOWN!

That word has been used since the beginning of the crisis. Early on, most of the users seemed to have "The China Syndrome" in mind: a molten blob eats its way through the reactor vessel, concrete foundation, and keeps going under its own fission power to emerge in China, or wherever is the antipode of the reactor's original location.

I've kept to words like "damaged fuel," which can include melted fuel. It would take extremely high temperatures, higher than have been reported for the Fukushima reactors, to produce a molten blob. What seems to have happened, according to the latest reports, is that the fuel element cladding warped, split, and perhaps melted, which allowed the fuel pellets, each about as big as the tip of your thumb, to pour out of the fuel elements and settle to the bottom of the reactor vessel.

Large industrial machinery is set up to keep people away. It's hot or pressurized or radioactive or has moving parts that easily crush flesh and bone. So that humans can keep track of what it's doing, sensors for quantities like temperature and pressure are attached to meters or computer control systems, which the humans can use to modify what the machinery is doing. This is true of any power plant - gas, coal, or water-powered - and most manufacturing. When something goes wrong, the sensors may fail. This is why nuclear reactors have multiple levels of containment, at Fukushima, the inner two a thick steel pressure vessel and an outer concrete structure, to contain the radioactive materials if the worst happens.

If the sensors fail, there may be indirect ways of measuring what is happening. Temperature measurements of the outside of the concrete structure provide limits on what the interior temperatures may be. Rising pressure indicates steam production from something hot inside.

Then there is the question of who knows what. At the plant, the operators have pieces of information that need to be coordinated. This is usually done through the control panel, but an emergency means that there will be problems with the control panel. Someone will have to make decisions on what measurements are needed and what actions to take. Many of those actions are dictated by emergency plans; some may have to go outside them. In the executive offices, people are coordinating with government officials and deciding what to say to the public. Everyone needs information, but the further you get from the reactor, fewer details are needed.

I'm far, far down the line of people who need information. I'm not anywhere where Fukushima's radiation will affect me. I can't do anything to help or hinder. Maybe send some money to help the evacuees. But all I need to know to do that is that there are evacuees. As a blogger, of course, I'd like to have a direct line to TEPCO and have all my questions answered instantly.

So why the clamor every time something new appears? Was TEPCO lying in earlier estimates of reactor damage? Or have they learned something new? Why is it a big deal to go from 70% damage to 100% damage? And why the focus on "meltdown" and "re-criticality"?

I'm going to suggest that, in the absence of a detailed understanding of how things go wrong in nuclear reactors, there's a very attractive narrative line that's been furnished by "The China Syndrome." Any number of easy rebuttals have been provided; I think the movie doesn't even say that the molten mass (see how familiar those phrases become) will emerge in China. Gravity would, of course, stop it at the center of the earth, if something else, like the pressure vessel didn't stop it first.

For this narrative line, we need a molten mass, presumably from the meltdown, and we need a continuing state of nuclear criticality. So it's very important that there be a meltdown and that criticality be re-established. I suspect that the commentators who are so focused on "meltdown" and "re-criticality" don't even realize that a movie's narrative line is driving their reasoning, but I don't see another reason that these terms have taken on such importance.

A meltdown could seriously breach the reactor containment, and if it did so, the radioactive reactor contents could be distributed widely, as at Chernobyl. The question that one would focus on, however, if this were the concern, would be whether there is a reactor breach. The fact that the containment vessels hold some pressure indicates that they haven't been breached Chernobyl-style. We'd also be seeing a much greater variety of fission products and actinides if there were a big breach, particularly in conjunction with a meltdown. (Added later: A blog post isn't a scientific paper, but I think I should add that I do think there's a breach in the coolant system. More below.) With all that water around, we'd be seeing steam explosions. But the media and other focus is on meltdown.

The furor over "re-criticality" has subsided somewhat, but it's not gone. The heat from continuing fission is what would drive the molten mass through the steel and concrete containments. So "re-criticality" is essential to that narrative line. I've written before why it's unlikely. That post might be updated a little. I'm recalling from my time in a reactor design group that new reactor designs had to have a calculation in which all the fuel was on the bottom of the vessel, and it had to be less than a critical mass. I think they still have to do calculations like that in order to get reactor designs certified.

TEPCO has concealed information and lied in the past. For both the corporation and the Japanese government, there is an issue of losing face, which may lead to concealment. For TEPCO in particular, there are likely to be heavy financial penalties. The government is considering putting TEPCO into receivership and taking its money to compensate the evacuees. So the last few days' reports of possible earthquake damage before the tsunami would take some of the responsibility off TEPCO employees and perhaps argue against fault on TEPCO's part. There are many possible motives for the release of various sorts of information.

TEPCO is making more and more information available. I saw the curves of pressures in the reactors after the earthquake this morning. Not simple to interpret, so I won't try just now.

Whether it's the state of the cores and the fuel elements in the cooling ponds or TEPCO's culpability and information flow, we're not going to have definitive evidence for months or even years. So, as I've said before, I'm trying to get past the hooting and hollering every time another number is released. Or every time the MSM finally figures out something to say about it. By and large, the worst of the crisis is over. It will take some time before all the reactors are in cold shutdown, and surprises are always possible. But last week's release of information wasn't among them.

Added later: There is probably a breach in the primary coolant systems at reactor #1, and probably at others. The coolant system consists of piping and pumps, somewhat like the hot water system in your house. The weak points are where pipes are connected to pumps and to the containment vessels: fittings and seals. The earthquake could have broken these connections, or too much heat could have. But the fact that the only fission products we're seeing are the soluble ones argues that they're coming out with the water.

1 comment:

Physics Today said...

Financial Times seems to have the best coverage.