How Nuclear Reactors Work, And How They Fail

[madokie]

http://www.popsci.com/node/52720/?cmpid=enews031711
I'm just throwing this out here for further discussion: madokie

Several of Japan's nuclear power plants are experiencing serious damage from the earthquake and ensuing tsunami. Here's what you need to know to understand the news, as it happens
By Dan Nosowitz Posted 03.14.2011 at 4:19 pm 69 Comments

Fukushima Dai-ichi Nuclear Plant, March 14, 2011 DigitalGlobe via Getty Images

Several of Japan's nuclear power plants, especially the Fukushima Naiishi plant in northeastern Japan, are experiencing serious problems in the wake of Friday's earthquake and tsunami. If you've been following the news, you've seen some pretty alarming stuff going on at this plant--terms like "explosion," "partial meltdown," "evacuation," and "radiation exposure." With details sparse from the chaotic scene, here's what you need to know to understand and make sense of the news unfolding in Japan.
What Is a Nuclear Reaction?

A nuclear reaction is at its most basic nothing more than a reaction process that occurs in an atomic nucleus. They typically take place when a nucleus of an atom gets smacked by either a subatomic particle (usually a "free neutron," a short-lived neutron not bound to an existing nucleus) or another nucleus. That reaction produces atomic and subatomic products different from either of the original two particles. To make the kind of nuclear reaction we want, a fission reaction (in which the nucleus splits apart), those two original particles have to be of a certain type: One has to be a very heavy elemental isotope, typically some form of uranium or plutonium, and the other has to be a very light "free neutron." The uranium or plutonium isotopes are referred to as "fissile," which means we can use them to induce fission by bombarding them with free neutrons.

In a fission reaction, the light particle (the free neutron) collides with the heavy particle (the uranium or plutonium isotope) which splits into two or three pieces. That fission produces a ton of energy in the form of both kinetic energy and electromagnetic radiation. Those new pieces include two new nuclei (byproducts), some photons (gamma rays), but also some more free neutrons, which is the key that makes nuclear fission a good candidate to generate energy. Those newly produced free neutrons zoom around and smack into more uranium or plutonium isotopes, which in turn produces more energy and more free neutrons, and the whole thing keeps going that way--a nuclear fission chain reaction.

Nuclear fission produces insane amounts of energy, largely in the form of heat--we're talking several million times more energy than you'd get from a similar mass of a more everyday fuel like gasoline.

[RoccoR5955]

1. Using atomic energy to boil water is like using a chain saw to cut butter! n/t

[Fledermaus]

2. They fail because it like a big jackpot at a casino. Sooner or later someone wins the jackpot.

Apparently there is a big payoff every twenty or thirty years. But when you have more players?? And people keep playing??? And if you are sitting next to a nuclear winner you win too!!!

[madokie]

3. Did you read the whole thing?

[Fledermaus]

4. No, sory

[madokie]

5. You might want too

[TheMadMonk]

9. 2 (or 5 if we count each of Fukushima's plants individually.) "Jackpots"

And so far, your "jackpots" have not, I repeat NOT, measured up to the hootin' and the hollerin', the bells, whistles and flashing lights.


The official figure for Chernobyl is about 1000 deaths IIRC. The anti's claim a little under 100,000. A good rule of thumb when trying to reconcile those sort of disparity in figures (particularly when both sides have an "agenda") is to take a logarithmic average. So in this case a good bet would be for about 10,000 deaths related to Chernobyl. Not a good figure, by any means, but those 10,000 deaths are spread out over a considerable period of time, as much as 50 years. This is roughly on a par with Bhopal.

Baring an exposure event during cleanup in the future, Fukushima's, death toll is highly unlikely to rise to even 100 over the next 50 years. This puts it on a par with any number of major industrial accidents that occur several times a year such as coal mine collapses. And even then it's comparing acute casualties occuring in a moment, minutes or at most days, to chronic casualties


Chernobyl was very much an extreme and atypical example of a worst case nuclear "accident". The proximate cause was direct human intervention. And the severity was a result of an inherrently unsafe design, exacerbated by the actions taken during the causative human intervention. It is pretty much the baseline event for: "As bad as it can possibly get."

Chernobyl did make clear the design flaws of the reactor, AND the outcome of a major failure in that particular design. And that particular design has now been pretty much totally abandonned, with legacy reactors being decommissioned as fast as their generating capacity can be matched from other sources.

Fukushima I think represents a typical extreme nuclear event in second generation kit. It could of course be or get worse, but FFS it would just about take a direct meteor strike, or similarly cataclysmic event, ON TOP OF the disasters to already hit the plant, for that to happen. Even then it remains highly unlikely that a total collapse in ALL FOUR Fukushima reactors (or their ilk) would match the severity of the Chernobly event simply due to design differences.

Now let's look at what it took to actually create this event: An earthquake 5 times more powerful than the reactors were designed to withstand. They held anyway. It actually took a tsunami bigger than the site defenses were designed to withstand to knock out the backup generators. If there was any error here, it was in assuming that the earthquake which created a tsunami big enough to overwhealm those defences would have gotten the reactor first, so no need for one last generator on a pole.

And even then, the reactors held together. If only just. And more or less for the past eight days, fairly minimally effective efforts (heroic as they have been) have managed to keep the reactors JUST holding together.

Sad thing is, if the public (and I include world sentiment in this) had been willing to accept an early deliberate release of radioactive steam (which ceases to be DANGEROUSLY radioacticive after just a few minutes) directly into the atmospere, there is a very good chance most of what has happened in the days following the earthquake could have been avoided.

Venting into the outer containment building was a stupid compromise, designed to limit public upset at the idea of venting radioactive steam from the reactor core, to catch it and hold it long enough for the worst of the radioactivity to disipate. Unfortunately this also held it long enough for the steam to begin to condense and for hydrogen to build up to explosive concentrations.

If the steam had been released directly into the atmosphere, the plant grounds and immediately surrounding area would have been irradiated (not contaminated) for as long as the steam was released and a few minutes afterwards. The release could be indefinite and the area irradiated would not grow beyond the distance the wind could carry the steam before it ceased being radioactive. There would be no ever-growing plume.

Instead by trying to hold it in, we got the hydrogen explosions. And instead of irradiation, we're getting (thankfully mild) contamination from a number of sources exposed in the explosions.

Irradiation ceases to be a problem in the moment it is gone. Contamination remains a problem until it is removed, or a sufficiently long enough time elapses for it to "go away".

Fukushima is the legacy of a wrong headed public opinion driven policy of total containment under all conceivable circumstances.

And languishing on the drawing boards are a number of third and fourth generation designs which are configured to either make any form of thermal runaway event physically impossible, or to channel thermal runaway events safely along a non-harmful pathway.

The most elegant, Liquid fuel reactors, deleberately make the "core" out of materials with a relatively low melting point (500-600 Kelivn) but a high boiling point (~1950 K). The physical layout of the main reaction chamber and its plumbing combined with the thermal properties of the molten fuel cause the reactor core to be self regulating. If it gets hot, thermal expansion moves a portion of the fuel out of the reaction chamber causing the nuclear reactions to slow down. As it cools the fuel flows back allowing reactions to increase.

Virtually all regulation of a liquid fuel reactor is done by varying the rate at which a working fluid passes through pipes immersed in the liquified reactor core. AND THAT CORE IS DESIGNED AND BUILT TO ACCOMODATE a flow rate of zero.

Liquid fuel reactor cannot melt down because they begin life molten. And because the fuel is molten, it can be easily physically separated into portions small enough to solidify into portions too small to sustain a chain reaction, and the same will happen even if it's allowed to simply pour out onto the floor.

There is litterally no manner in which a liquid fueled reactor can "blow up" in a dangerous manner. And short of detonating a nuke on site, even the total physical disruption of the reactor vessel will result in nothing more than an admittedly fucking great, but VERY LOCALISED and managable mess.

And because the fuel is in the form of a flouride, atmospheric oxygen is never an issue even if the stuff does somehow manage to get hot enough to vapourise a portion of itself.


A less elegant solution is to simply park a dirty great water (or boric acid) tank over the reactor and if the reactor ever overheats, it melts the bum out of the tank long before it itself is compromised and the whole mess in immersed. Ports in the top and bottom of the reactor vessel can then be opened to allow convection to draw the water through the core.


We can't prevent "the jackpot". However, we can make the "payout" sufficiently miserly that even an endless string of "jackpots" won't break the bank.

Furthermore with only a limited amount of aditional effort, we can further offset the outcome of occasional nuclear catastrophe enormously, not by slamming our heads up against built in limits, but by using the "limitless" energy to destroy persitant pollutants such as PCBs by reducing them to their component atoms.

With enough "overgeneration capacity" it would become possible to effectively winnow nuclear waste atom by atom. To completly separate the radioactove from the non, and to further separate waste into short, median and long lived isotopes. This would make sequestration and long term storage a much easier proposition. Beyond sequestration are at least two methods which offer the promise of completely neutralising nuclear waste.

[madokie]

10. The sad thing is

You've done all that typing and it's all bullshit warmed over now for 50 some odd years. To some of us it's really starting to smell.

No radiation is good. I've got a spot on one of my lungs that can't be watched close enough to really monitor it because of the danger from the radiation that it takes to do that monitoring. I don't think my pulmonary doctor has a dog in this fight and he's the one thats telling me this.

[jpak]

6. reccing for useful info

n/t

[Nederland]

7. Good article (nt)

[Name removed]

8. Deleted message

Message removed by moderator. Click here to review the message board rules.