Why only delayed neutrons matter ?

[Leb]

Hi all,

I've been seeing this again and again, that delayed neutrons are the ones that are allowing thermal nuclear reactors to work etc. etc.

What I do not understand is how are prompt neutrons controlled and how delayed ones are allowed to "do their thing" no problem ? Wiki says: "However, thanks to the delayed neutrons, it is possible to leave the reactor in a subcritical state as far as only prompt neutrons are concerned: the delayed neutrons come a moment later, just in time to sustain the chain reaction when it is going to die out."

Why would it die out if you have 2-3 neutrons per fission..? The time scales are so small that I cannot imagine control rods working that fast...

 

[mfb]

Why would it die out if you have 2-3 neutrons per fission..?

You lose some neutrons in other nuclear reactions.

With fast neutrons only, you would have either a vanishing fission rate (on average less than 1 induced fission per fission) or a nuclear explosion (on average more than 1 -> gives exponential increase in microseconds). Delayed neutrons make this timescale much longer (of the order of minutes) - long enough to use active control systems.

 

[Leb]

Thanks for the answer, but it is still not clear to me, how these prompt neutrons are controlled in the reactor ? I mean, maybe I am missing something - the prompt neutrons come from fission of your fuel and delayed come from fission from your FP. But how do you control the prompt neutrinos, to prevent all the problems (i.e. bomb) they cause. I just don't see how are they not causing uncontrollable fission in one of the cladding fuels rods...

So to clarify:

Can someone explain what happens to prompt neutrinos in a reactor (and how are we able to differentiate them from delayed ones in terms of neutron control) ?

 

[the_wolfman]

Lets imagine you discovered a banking loop hole where you invest a dollar and immediately get $1.01 back. With this loophole you can quickly make a fortune. You just continuously reinvest the money you get back. And how quickly you make money is only limited by how quickly you can invest. If you're quick enough you can go from a few bucks to a million bucks in a few minutes. And your back can't do anything about it. This is basically how a reactor would work without delayed neutrons.

Now imagine a similar scenario where you invest $1.00 and you still get $1.01 back but you only get $0.99 back immediately, and you get the other $0.02 back a day later. Again you can still make a huge profit, but now it takes time. Initially you can invest all you money, and you'll get $0.99 back for each $1. You'll probably be able to do a few more initial investments but each time you have a little less money. Eventually you'll run out of money and you have to wait a day for the remaining $0.02. The next day you can repeat the process investing all you money as often as possible, but again eventually you'll run out of money and you'll have to wait until the next day before you invest again. Each day you start with more money, and you are making a profit. But now you fortune grows slowly over days instead of moments. In the meantime the bank has time to notice what your doing and fix the loophole.

This second scenario is analogous to how delayed neutrons work in a reactor and why they give us control. Its that delay between when the initial neutron induces fission and when the delayed neutron is born that is important. Once born, all neutrons (prompt or delayed) with the same energy obey the same laws of physics.

 

[Leb]

Thanks for the analogy, I think I get it why the delayed ones do the work (because every text booork website tells you the same, but I am yet to find an answer about what happens to prompt neutrons).

So IGNORE delayed neutrons (I think the question title is a bit misleading now...)
My question is, what happens to prompt neutrons ? Where do they go ? How are prompt neutrons are stopped, while delayed neutrons are allowed to work ? (i.e. let's say I do not care how delayed neutrons make the reactor work). All I want to know is why prompt neutrons "die off" if we get 2-3 prompt neutrons per fission ? That is why using various control systems still make this work (when I am imagining the number of U-235 atoms per pellet of fuel in a single fuel rod, it's hard to imagine that there are not enough of them (U-235) to make an uncontrolled chain reaction (if I understand correctly, the control rods are placed between fuel rod assemblies, so how are they to stop neutrons from one rod interacting with the fuel from other rod, if there are more neutrons than there are U-235 in one rod...

 

[Nugatory]

Thanks for the analogy, I think I get it why the delayed ones do the work (because every text booork website tells you the same, but I am yet to find an answer about what happens to prompt neutrons).

Not every prompt neutron triggers another fission event. Some of them escape the mass of uranium completely, others will be captured by something other than another U-235 nucleus. Thus, I can get more than one prompt neutron out of each fission event and still not have a runaway self-sustaining chain reaction - it will depend on the density of the fissile material, the way it is physically arranged, and how enriched it is.

If I'm building a bomb, I arrange things so that every fission event triggers more than one new fission event from prompt neutrons (for example, by compressing a sphere of highly enriched fissile material into a smaller sphere). The slow neutrons are irrelevant, as the bomb blows itself apart long before they matter.

If I'm building a power reactor, I start by arranging things so that on average the prompt neutrons from a given fission event produce less than one new fission event, and the others escape or are absorbed by something other than another fissile nucleus. Now the additional fission events produced by the delayed neutrons are relevant; I can use the control rods to control how many of these I get, and I continuously adjust the control rods so that on average I get one new fission event for every fission event. Pull them too far out, that ratio goes up, the reactor heats up more, and I know to push them back in again.

Of course we're always overshooting or undershooting the magic ratio of one new fission event for each fission event. When we undershoot (within reason) we don't extinguish the reaction, because we have plenty of delayed neutrons from a few seconds ago when it was hotter, so we back out the rods and let these neutrons kick the rate back up. Conversely, when we overshoot by pulling the rods out too far, the number of delayed neutrons, set by operating conditions a few seconds back, doesn't immediately increase so we have time to push the rods back in and slow things down.

 

[Leb]

Ah, it's more clear now ! I don't really like the oversimplifications made. It makes sense why one has reflectors then.

So basically, it is crucial how you make the pellets, how many of them are in a rod and how many rods per assembly. I think when I get too looking properly at cross sections it might make more sense.

It's very interesting how you need to engineer them in such a way, that, let's say, the first neutron absorbed by the first U-235 atom, does not cause an uncontrolled chain reaction!

 

[mfb]

Just to extend that point:

My question is, what happens to prompt neutrons ? Where do they go ? How are prompt neutrons are stopped, while delayed neutrons are allowed to work ?

The mechanisms Nugatory explained are the same for both neutron types. You just lose some fractions of the neutrons (on purpose).

So basically, it is crucial how you make the pellets, how many of them are in a rod and how many rods per assembly. I think when I get too looking properly at cross sections it might make more sense.

It's very interesting how you need to engineer them in such a way, that, let's say, the first neutron absorbed by the first U-235 atom, does not cause an uncontrolled chain reaction!

Yes.