Nuclear Dopers - How is Prompt Criticality prevented?

Factual Questions
1

In a reactor, prompt criticality is the condition where the Neutrons from the fissioning fuel drive the reactor from critical to supercritical, making the reactor very hard to control. A slight mis-adjustment of a control rod can cause the reactor to go from essentially zero power to many hundreds of megawatts in only a few milliseconds, which is generally considered a bad thing. The SL-1 reactor suffered this fate, and killed three people.

Somehow, a reactor can be designed to be impossible (or at least difficult) to become prompt critical. In these reactors, the Neutrons that drive the reaction into criticality come from daughter products of the original fissions, and hence are delayed by seconds or minutes, allowing the power to build up slow enough to be easily controlled.

My question is: how is this accomplished? Is it the fuel composition? The geometry of the reactor core? The moderator? Or something else?

2

I think you’re thinking of reactors with, for example, heavy water as the moderator. This eliminates the requirement for enriched fuel.

Read about CANDU reactors on Wiki.

3

Its due to delayed neutrons that are naturally occurring. They make up a small percentage of the neutron population but they can take up to 6 seconds to appear so when you average them in with prompt neutrons they make average neutron life manageable. This is off the top of my head since I’m in ICC and it’s been 20 years since Power School but thats pretty close.

4

Off the top of my head from nuclear power school (only 18 years ago for me)…

Basically, prompt criticality in a pressurized water reactor is avoided by use of the moderator. The pressurized water reactors that I am familiar with use enriched uranium fuel and ordinary water for the moderator.

Indeed, the whole point in having a controlled nuclear reaction is to avoid prompt criticality. This accomplished by the use of Uranium-235, which does not readily fission using fast neutrons (the neutrons produced in a fission event). Uranium-235 has a relatively low nuclear cross-section for fast neutrons. (The units for nuclear cross-section were amusingly named barns, as in the chance of hitting the broad side of a barn.) After neutrons collide with similarly sized subatomic particles (such as the hydrogen in water), they slow down, producing so-called thermal neutrons.

Uranium-235 has a much higher nuclear cross section for thermal neutrons. In other words, a thermal neutron is far more likely to cause a fission than a fast neutron for U-235.

What you want to do is to set up the reactor so that criticality can just exactly be reached primarily with thermal neutrons. This helps create what is referred to a negative temperature coefficient of reactivity. As the moderator (water) heats up, it becomes less dense (though it is still very hot, it is kept liquid by the high pressure). This results in fewer thermal neutrons being produced, so the reaction rate decreases, and the temperature decreases. This results in a feedback loop that controls the reaction.

If an accident were to occur, and the moderator all leaked or flashed to steam, the reaction therefore automatically shuts down, because no more thermal neutrons are being produced.

An out-of-control reaction results if you were to increase the reactivity of the system (say by continuously withdrawing the rods past criticality) until there are enough fast neutrons present to maintain the reaction. This is referred to as prompt critical, and the reaction will go out of control. This is desired in an atomic bomb, but not a power reactor. Because of the geometry of a power reactor, though, it is not possible to get it to actually blow up like a bomb, but it will result in a big mess.

I may have missed a couple of details, but I believe that I have the basics correct here.

5

I am a nuclear engineer with over 30 years’ experience and can tell you that prompt criticality is not possible in light water reactors (the kind used in the US and most of the rest of the world). When a fission occurs, neutrons are emitted along with 2 or 3 fission fragments (parts of the nucleus of the original atoms nucleus). Delayed neutrons are emitted when the fission fragments undergo a radioactive decay and emit a neutron in the process. There are over 30 fission products that emit delayed neutrons, with half-lives that vary from a few seconds to several hours. However, these 30 groups of delayed are usually combined to form 6 delayed neutron groups. When a reactor is taken critical, 99.9935% of the neutrons are prompt neutrons. The delayed neutron fraction is 0.0065 (at least for U235 based reactors). The idea is that you take the reactor almost critical on prompt neutrons with the delayed neutrons taking the reactor from almost critical to just critical. It is the delayed neutrons that allow the reactor to be controlled. The Chernobyl reactor actually went super prompt critical; that is, it was super critical with only prompt neutrons. This was possible because it was cooled by light water but moderatored by graphite. In this type of a reactor, the water acts as a neutron poison. Therefore, when power reached the point where all of the water flashed into steam, there was a large insertion of reactivity and the reactor was super prompt critical for a short time until it blew itself apart. It is estimated that the reactor reached about 100,000 MW before it blew apart.

Mark Laris

6

Prompt and delayed neutrons should not be confused with fast and slow neutrons. Fast or high energy neutrons resulting from fission events are slowed down (lose energy) by colliding with the hydrogen atoms in light water. As stated in a previous response, U235 has a high fission cross section for slow (or thermal) neutrons but a low fission cross section for fast neutrons. Prompt criticality, as used above, refers to criticality occurring by the thermalization (or slowing down) of the prompt neutrons (which are high energy “fast” neutrons) and thermalization of the delayed neutrons. Technically, it should be called prompt moderated criticality. Prompt criticality with unmoderated fast neutrons would be a bomb and is not possible in any reactor of any design.

7

Well, it’s been three years, but i’m still here, and always interested in good answers to my questions.
Thanks mjlaris, and welcome to the SDMB!

8

Light water reactors can go prompt critical. I was taught this for naval reactors, they are highly enriched. I cant remember the details but if you added a certain amount of reactivity then some how it short circuited the equation. Hmm damnit I spent so much time learning this stuff and now I cant remember it.

9

I wrote this and I have no idea what ICC is.

10

I know that this is an old question but prompt criticality is controlled by controlling the number of neutrons you have or the power level in the core. It’s actually difficult to get a reactor to go prompt critical. As a point of reference, Chernobyl went prompt critical, actually prompt super critical.

11

I’m on a road trip to Yellowstone, and with any luck I will take a side trip to Arco, ID to see the first electricity-generating nuclear reactor (EBR-1).

12

We stopped at EBR-1today.
It’s small and sparse, but really cool, if you like technology.
If anyone is ever up near Idaho Falls, I recommend a side-trip.