What Three Mile Island “Proved” About Meltdowns

Screencap from the entry movie at William Tucker's site, TerrestrialEnergy.org.

In an opinion piece from March 14, William Tucker writes some good stuff and some bad stuff about Fukushima in the Wall Street Journal. Tucker’s book is called Terrestrial Energy: How Nuclear Power Will Lead the Green Revolution and End America’s Energy Odyssey. In short, he glows green.

The main point Tucker makes is that the Generation III reactors that have been designed for future use in the United States are safer than the Fukushima reactors, which are Generation II. ALL existing civilian plants in operation in the U.S. are Generation II, the same generation as Fukushima. The Generation III reactors have been held up at the NRC for years, because of political opposition to nuclear power. The probable result of that is that nuclear power in the United States today is less safe, not safer, than if those Gen-III plants had been developed.

For the record, the Generation III plants reportedly eliminate the central flaw of the design of Fukushima — and every other Generation II plant — which is the lack of a fail-safe in coolant circulation. A fail safe is a system that, in the case of an event, fails into a safe state rather than a hazardous one.

In other words, the failures at Fukushima are in no way directly associated with using uranium or MOX fuel to heat water — that’s what these damn things are designed to do. The flaw is in not being able to maintain the coolant circulation in the event of a catastrophic disaster like this one. The Generation III plants eliminate that problem with a design that keeps water circulating in the case of a shutdown.

I’m with Tucker on several points, most importantly that the press needs to stop forecasting disaster, and that (once again!!) this is not Chernobyl and it’s not going to be.

But I have to say yet again — and I feel like I say it constantly — that the time for nuclear advocates to call for development of more nuclear power plants is AFTER the Fukushima events are under control. Which will be soon, I feel confident, but in the meantime STFU about building new power plants.

In any event, here’s Tucker:

Once the reactor has shut down, there remains “decay heat” from traces of other radioactive isotopes. This can take more than a week to cool down, and the rods must be continually bathed in cooling waters to keep them from overheating….If the pumps are knocked out in a Generation II reactor—as they were at Fukushima Daiichi by the tsunami—the water in the cooling system can overheat and evaporate. The resulting steam increases internal pressure that must be vented. There was a small release of radioactive steam at Three Mile Island in 1979, and there have also been a few releases at Fukushima Daiichi. These produce radiation at about the level of one dental X-ray in the immediate vicinity and quickly dissipate.

If the coolant continues to evaporate, the water level can fall below the level of the fuel rods, exposing them. This will cause a meltdown, meaning the fuel rods melt to the bottom of the steel pressure vessel.

Early speculation was that in a case like this the fuel might continue melting right through the steel and perhaps even through the concrete containment structure—the so-called China syndrome, where the fuel would melt all the way to China. But Three Mile Island proved this doesn’t happen. The melted fuel rods simply aren’t hot enough to melt steel or concrete.

…None of this amounts to “another Chernobyl.” The Chernobyl reactor had two crucial design flaws. First, it used graphite (carbon) instead of water to “moderate” the neutrons, which makes possible the nuclear reaction. The graphite caught fire in April 1986 and burned for four days. Water does not catch fire.


Right, right, and right, on the decay  heat, the Gen II cooling pumps, and the steam venting. His comment about the dental X-ray is red herring-fueled propaganda; the radiation release at Fukushima is probably not over and has not been evaluated in the calm following the storm. More importantly, information coming out of Japan has been spotty. It’s ludicrous to claim he knows how much radiation has been released, yet.

Here’s where he’s misrepresenting the facts:

Early speculation was that in a case like this the fuel might continue melting right through the steel and perhaps even through the concrete containment structure—the so-called China syndrome, where the fuel would melt all the way to China. But Three Mile Island proved this doesn’t happen. The melted fuel rods simply aren’t hot enough to melt steel or concrete.


Hell’s bells! Did this guy go to fourth grade? That’s the kind of science that gave us spontaneous generation of frogs from grain!!

Three Mile Island did not “prove” anything. Three Mile Island was a single case. Science “proves” things by repeated experimentation, not individual cases thirty years apart. Even then, in the case of very speculative situations with very small numbers of cases to draw on, science tends to “demonstrate” things, not prove them. Experimentation, modeling, evaluation of real-world results — those are all very important. All of those have been done in the case of building nuclear power plants.

But you don’t “prove” things by pointing to a related but certainly not identical situation thirty years ago! Just because we (thankfully!!) don’t have a whole pile of other real-world cases of this magnitude doesn’t mean you get to extrapolate from the only case that even smells vaguely like this!

Core meltdowns at civilian power plants are (again, thankfully) events so fantastically rare as to be utterly individual. The bottom of the containment vessel at Three Mile Island was not breached despite a significant amount of the core melting. As pro-nuclear people told us again and again and again at the beginning, Fukushima is not Three Mile Island or Chernobyl. Only now…Tucker is trying to tell us that it’s, oh, you know, still not Chernobyl, but it IS Three Mile Island.

Except it’s not Three Mile Island, and implying that it is is beyond ludicrous. Three Mile Island did not “prove” that the containment vessel could not be breached. It established only that it was not breached in that case. That’s an important data point. That’s a really important data point. But it doesn’t “prove” the floor of the containment vessel won’t breach.

Especially since the single containment vessel that contained the partially melted core at Three Mile Island was not leaking water. One of the cores at Fukushima is.

Am I saying that if there is a 100% core meltdown in all three cores at Fukushima — a highly, highly, highly unlikely scenario — the containment vessel will be breached? Does it mean the China Syndrome will happen? Will the melted corium hit groundwater and cause an explosion? Hell, no, it doesn’t mean any of that! It doesn’t mean anything!

In fact, Three Mile Island suggests that the melted core may not, in fact, breach the bottom of any of the containment vessels, even if there is a 100% core meltdown in all three cores (again, extremely unlikely). The China Syndrome appears extremely-extremely unlikely, for reasons of basic physics even when you do have molten metal that generates its own decay heat. Decay heat is not the heat from active chain reactions, which would be happening if the control rods had been blown to smithereens as at Chernobyl.

Is that heat less extreme than it was at Chernobyl? Hell yes, without question. Is it hot enough to melt steel? Nobody knows with absolute certainty, but probably not. It’s almost certainly impossible that the core would be hot enough to melt the thick steal of the containment vessel. But nobody knows with absolute certainty. It hasn’t been “proven.”

As Rod Adams, a highly (to me) abrasive pro-nuker but a reasonably rational nuclear advocate when he lets himself be, put it:

At TMI [Three Mile Island], the widely predicted and discussed “China Syndrome” did not happen, even though 20-30% of the core melted and slumped to the bottom of the pressure vessel. That melted corium froze again once it contacted the thick metal walls – the maximum measured penetration was just 5/8ths of an inch. Anyone who has ever watched as welder employs a torch to cut through a thick steel wall will understand just how much concentrated power it takes to melt several inches of steel. Avoiding the China Syndrome was not a matter of luck – the scenario is imaginary and only works in fiction. Physics and material science make it impossible.


My opinion is that Adams is carrying the existing information a bit far in coming to that conclusion that “Physics and material science make it impossible,” but at least he doesn’t use the P-word.

Adams is not a scientist; he was trained as a nuclear engineer in the U.S. Navy, so I trust his gut feeling on this way more than I trust any amount of scientific understanding from William Tucker, who is “just” a journalist — not unlike some other people who seem to have decided, unilaterally as far as I can tell, that solid zirconium alloy explodes at 2,000 degrees.

Adams, at least in this paragraph, doesn’t claim that science “proves” anything about a real-world nuclear meltdown, for which I am thankful.

What’s important here is to stick to demonstrable, documented facts — not wild extrapolations from the only case we have to consider.

That’s how one creates a cogent argument on a scientific or engineering matter. Shouldn’t that be obvious to any professional journalist?

What is this, rocket science?

Oh, one last thing — It’s nothing, really. Nothing at all! Just a…well, I guess I should mention it.

Adams does appear to be wrong on a central fact, which would be nit-picky of me to point out — maybe even a little rude — if it wasn’t kind of a big one. Here’s a quote from the Nuclear Regulatory Commission’s Backgrounder on the Three Mile Island Incident:

Because adequate cooling was not available, the nuclear fuel overheated to the point at which the zirconium cladding (the long metal tubes which hold the nuclear fuel pellets) ruptured and the fuel pellets began to melt. It was later found that about one-half of the core melted during the early stages of the accident. Although the TMI-2 plant suffered a severe core meltdown, the most dangerous kind of nuclear power accident, it did not produce the worst-case consequences that reactor experts had long feared. In a worst-case accident, the melting of nuclear fuel would lead to a breach of the walls of the containment building and release massive quantities of radiation to the environment. But this did not occur as a result of the three Mile Island accident.


Get it? About one-half the core melted at Three Mile Island, not 20%-30% as Adams has stated. Frankly, I like Adams’s writing. But the explicit misrepresentation of a number so central to his assertion calls everything he says into question.

I’m not a nuclear engineer — I’m a fucking horror writer. I write amusing pieces about zombie attacks at Dartmouth! I drop one-liners about autogiros! I debate e-book rights, not containment vessels! I got a “C” in Chemistry, for the love of God.

So it’s possible there’s some nuclear engineer ba-da-bing Adams could lay on me that would indicate the NRC’s talking out it’s ass. Hey, it wouldn’t be the first time. But, just speculating, I find it more likely that Adams carelessly misremembered the fact. He does appear to have retained the important takeaway — whatever percentage of the core melted at Three Mile Island, the containment vessel remained intact.

If that’s true, or even if it’s not, what else are any of these people being careless about?

Not sure what Adams’ response would be to that one, but I’ve got metal fatigue.

I’m going home to pet my dog and turn the lights off for a while.

Screencap from the entry movie at William Tucker’s site, TerrestrialEnergy.org.

Possibly related posts: