Eli5: What is nuclear criticality?



I’ve been binge watching YouTube about nuclear criticality events, and I get the basic principle that you have two radioactive items and they come in contact long enough to create an event.

But I am not quite getting it. So for example, I watched a video about the demon core, but I am not sure how the shims were preventing criticality.

Thank you.

In: Physics

Some atoms are so big that little parts accidently shoot off sometimes. Sometimes that happens randomly and sometimes it happens because something crashed into it. If you put a bunch of these big fragile atoms close together sometimes something will shoot off one atom and smash into another atom and sometimes that crash will make something shoot off the second atom.

If you put a little bit of this stuff together you will get little chain reactions, one atom shoots off and hits another then it shoots off, and it happens a few times then dies away. There wasn’t enough atoms to keep it going.

Critical is when you have a big enough ball of the stuff that the little chain reactions don’t die out. There is enough stuff flying around that instead of going for a while and dying out the chain reaction gets bigger and bigger.

The demon core was exactly enough stuff so if it was together would start that chain reaction and not stop until it exploded in a nuclear bomb. if you put even a tiny space between the two halfs it would be just barely not enough and the chain reactions would happen and stop real quick over and over forever, the tiny gap was enough to never quite make it

Radio active materials naturally emit alpha particles. Alpha particles hitting radio active materials causes it to release more alpha particles. In small enough quantities most particles go harmless into the surrounding environment.

If you put enough radio active material together so that most particles are hitting other material, the rate of release will go exponential, and now you have lethal amount going into the surrounding environment.

When the neutron reflector halves were closed, the neutrons had nowhere to go but back into the core itself, hitting more atoms, splitting off more neutrons, etc etc, making the core go critical. I’m no scientist or anything, but that’s what I got from what happened

You need a neutron to split atoms. After splitting atoms you get more neutrons out.

If the number of neutrons that come out split the SAME number of atoms as last time, your reactor is critical, and is in a steady state condition.

If the number of neutrons that come out then go on to split LESS atoms, then your reactor is “subcritical” and the power level goes DOWN.

If the number of neutrons that come out then go on to split MORE atoms, then your reactor is “supercritical” and the power level goes UP.

If we do things that make it less likely for a neutron to split an atom then we cause the reactor power to drop. Examples are, adding neutron absorbers (poisons), removing reflectors, or making the geometry harder for the neutrons to get through.

If we do things like ADD a reflector, that means less neutrons leak out and the power level will go UP.

In the demon core, they more or less dropped a reflector on the core and it exceeded the threshold for criticality, causing a prompt power rise.

Imagine a giant pile of firecrackers which can randomly pop, so there’s at least a few pooping every second no matter what.

Imagine there’s a chance of a firecracker which pops makes another pop due to the small burst of heat.

If you surround the pile with something that reflects the shock waves and heat from popping back into the pile, you could make it so a single firecracker starts an increasing rate of popping, so the whole pile pops in a second. That’s like supercritical.

If you somehow cause a bunch of firecrackers in the pile to pop at the same time, it might blow the pile (supercritical). Or it might die back down to the normal rate (subcritical). Or, if you get it just right, you keep a constant high rate of popping (critical).

Obviously, staying right at critical is super hard, especially since for reactors the time scale for these changes is millionths of a second. Reactors wouldn’t be controllable if it weren’t for a special case — some firecrackers have a delayed pop, by up to a few minutes, i.e. they’re hit by a nearby cracker and you know they’re gonna pop, but a few minutes from now.

So you keep the pile in a condition where the “prompt” condition is subcritical, but the “delayed” condition can be sub, super, or critical. You listen to the rate of pops and very slowly change things to raise or lower the rate of popping. But be careful! If you want to raise the rate and you get impatient, you may change conditions so.much that it goes supercritical and the whole pile pops before you can react.

Criticality is the point where 10 year old you and your 10 year old friend lit a string of firecrackers in the hayloft of farmer Mcgregor’s barn, and you couldn’t run down the ladder fast enough to fetch a bucket of water from the well, and the whole barn went up in one giant fire.

Criticality is a tricky concept to visualize, which is why I created this [little simulator](http://blog.nuclearsecrecy.com/misc/criticality/), which is mostly for thinking about bombs (not reactors) but is still useful, I think.

The way to think about criticality is very, very basic. It’s nothing magical. There are atomic nuclei that can be made to split (fission) when they encounter neutrons. Not all nuclei can do this, and not all neutrons split nuclei equally well (the amount of energy the neutrons have changes the probability that they will split a nuclei — paradoxically, having _less_ energy is better splitting many nuclei).

Neutrons usually only go a few centimeters before they disintegrate. When an nucleus splits, it releases a few neutrons (on average about 2.5) in random directions.

With that in mind, “criticality” means you’ve set up your atoms (and anything else around them) in such a way that on average, the neutrons that are being released from those splitting atoms will go on to split at least the same number of atoms that had split in the first place. So if one atom splits and releases 2 neutrons, you’d need 1 of those to split another atom for it to be “critical.”

“Supercritical” means you are splitting _more_ atoms than were splitting in the first place. So that one atom splits, releases 2 neutrons, and each of those neutrons split two more atoms. That leads to an exponential growth.

What affects criticality? A lot of things. It’s really just, “what factors will make it more likely for those neutrons to split more atoms?” They include:

* How many atoms of the splittable sort are in the material. If there is only one atom to begin with, it’s not going to split any more! But more practically, those neutrons are going to be traveling through space, and if they aren’t traveling through other atoms of the splittable sort, then they won’t likely encounter them. This is why a raw amount of mass is usually associated with this.

* How many atoms are mixed in with them that won’t split, but will absorb neutrons? This is an issue in bombs (it is why you need to enrich uranium, because the most common kind of uranium won’t split easily), and reactors (where it is used for control rods among other things). If your neutrons are being “eaten” by other atoms, that will change the criticality requirements.

* Are the neutrons being slowed down, to increase their likelihood of splitting? In a reactor there are substances like water or hydrogen or carbon (the “moderator”) against which the neutrons can be made to bounce to lose some of their energy, which increases the chance that they will split any splittable atoms they find.

* How close together are the atoms? This matters for bombs more than anything else, but one way to increase the chances of neutrons finding more atoms to split is to physically squish the atoms closer together, which is how an implosion bomb works (it uses explosives to increase the density of the fuel before splitting its atoms).

* What geometry is the fuel material? A solid sphere is the most favorable geometry for criticality, because any given atom in the center is likely to have its neutrons run into more material as it moves outward. Whereas a long tube (cylinder) of material is not very favorable to criticality, because the neutrons would have to be traveling along the length of the tube in order to run into more fuel. Careful choice of geometry makes a big difference in whether systems are likely to be safe or go critical.

There are some other factors as well, like temperature and chemical structure and other things of that nature, but you get the picture. It is really just about whether those neutrons are going to find other atoms to split, and anything that affects that will affect criticality.

I find that when we speak of this only as a “critical mass,” it obscures that we are talking about a “system” with a lot of properties, and can make it seem rather magical. But it’s very physical, and somewhat straightforward if you understand what is going on “under the hood.”

First its important to understand what criticality isn’t and is. Something being critical does not mean that it will instantly explode like a nuke, but it does mean the temperature will start to ramp up and will ramp faster the further into the critical region the mass is

A mass of fissile material(atoms that can be split like U-235 or Pu-239) is critical when every time an atom randomly splits it causes one other to split. If it causes more than 1 other atom to split then its super critical, if it causes less on average then its sub critical. You can change the chances of a chain reaction by reflecting the neutrons back in so they get another shot, or adding more material so its likely to hit something on the way out, or by squeezing the material so its denser so its harder to miss.

If you have a 9.9 kg sphere of Plutonium-239 it is sub-critical so it’ll sit there being kind of warm and fairly radioactive but it’ll stay the same temperature because its subcritical.

If you have a 10 kg sphere of Pu-239 then you have a critical mass of it and the sphere will get really hot and be quite radioactive. If you do anything to squeeze it or put a beryllium shell around it to reflect the neutrons back in then it’ll become supercritical and the temperature and radioactivity will both start to rise. This is what happened with the Demon Core, they dropped the Beryllium shell closed around it so their relatively small radioactive core was now supercritical and the reaction rate started to climb causing a lot of radioactivity

Now you’ll notice that it didn’t become a bomb. That’s because there are levels of super critical. If you get it supercritical so each event causes 1.001 events then the rate of radioactivity will climb slowly, but if each event causes 3 then it will climb very quickly.

A critical mass of U-235 has 15 splits per second. If you make it 10% above the critical level then after one step you have 16 splits per second, and in 10 steps you’ll measure 39 per second. If you squeeze it down in a nuke and reflect all the neutrons back in so that every neutron causes a split then the first one to split causes 3 which cause 9 which cause 27 and 10 steps later you’re at 59,000 and 50 steps later you’re at 7×10^23.

While something being “super critical” means its going to get warmer just being super critical isn’t enough for a nuke. The core needs to be *very* super critical