Nuclear reactors, then. A good idea with pretty much the worst PR in the history of the planet, since when people hear the word “nuclear” they think of this or this or this or this, with little idea of what nuclear power generation is actually about. In light of the recent screw-up at Fukushima the politics surrounding nuclear power are fairly muddied and the future for it is uncertain, which is a bit of a shame because nuclear reactors are – if done correctly – actually a fairly neat way of fulfilling your country’s energy requirements.

Everyone should have learned how electricity is generated back in secondary school/high school/whatever their national equivalent is. There is a fuel material which generates heat, which vaporises water, which produces steam, which drives a turbine, which creates electricity by dint of providing the mechanical work for a electromagnetic dynamo/generator of some kind. There’s several subtypes of power plant but this is the setup everyone will be familiar with and all power plants work by roughly the same principle anyway: the conversion of mechanical work, whether generated by wind or tides or steam, into electrical energy. The only major difference is in what you use to do the mechanical work. In the case of steam-driven power plants, it’s the fuel you use to generate the heat. Fossil fuel power plants burn their fuels directly and create steam that way; however, nuclear power generation works very differently.

A nuclear fuel rod starts its life as raw uranium ore mined from the ground. The raw ore is overwhelmingly (99.3%) made up of an isotope of uranium called uranium-238. The remaining 0.7% is uranium-235. The difference between these two isotopes is that uranium-235 is fissile – that is, it’s possible to initiate a fissile chain reaction in u-235 by bombarding it with neutrons of any energy – whereas the 238 isotope is fertile. Fertile material is not immediately fissile, but if you hit it with high-energy (or “fast”) neutrons then it too will start to undergo radioactive fission and may eventually become something immediately fissile (in the case of u-238, plutonium-239) a couple of steps down the decay chain.

Now, why is the distinction between fertile and fissile material important? It’s all to do with the chain reaction. Say you’ve got your u-235 nucleus, and you hit it with a neutron. This splits the atom into two fission fragments (child atoms with a total mass of just less than the parent u-235 atom, but whose exact atomic type is impossible to predict), an average of 2.4 free neutrons and some now-unneeded binding energy, which manifests itself as heat. Obviously you can’t have 2.4 neutrons, so what we’re saying here is that the majority of u-235 fission reactions will produce either two or three free neutrons. These two or three neutrons spin off into the atomic ether, and if one of them encounters another u-235 nucleus then the fission reaction will happen all over again, releasing another 2.4 neutrons along with some thermal energy and so on.

This sort of fissile reaction is potentially very bad news if you have a lot of fissile material jammed together into what’s called a critical mass. The neutron multiplication factor k is basically how many child reactions your byproduct neutrons cause in the next step in the chain. For example, if two neutrons span out of a u-235 fission reaction, and each of them hit another u-235 nucleus, then the neutron multiplication factor would be 2 since the first reaction begat two more reactions. Having a k of 1 means your reaction is just barely self-sustaining – each reaction leads to one more reaction, and then one more reaction, and so on – and any mass of fissile material with a k of at least 1 is referred to as a critical mass. Having a k of more than 1 can be catastrophic; this is a supercritical mass and can lead to a runaway chain reaction that increases exponentially with every step unless it either self-corrects somehow or else is moderated by an outside source.

The human race has come up with several ingenious methods of converting subcritical masses of fissile material into supercritical masses very very quickly since it’s a key part of nuclear weapon design, and I may do a post on that later. For now, though, we’re focusing on the chain reaction as it pertains to power generation, and this is where the fertile vs. fissile thing comes into play. A fission reaction can be induced in a fissile material by hitting it with any neutron; the reaction produces more neutrons, which produces more reactions, and so on. Getting energy out of fissile material is quite an easy thing to do – indeed, the hard part is getting it to stop. However, only 0.7% of mined uranium is the fissile u-235 isotope. The rest is the fertile u-238 isotope which can only be induced to split by bombarding it with fast neutrons. This presents a problem if we want to do anything useful with it because after you ram a fast neutron into a u-238 nucleus it loses most of its energy due to inelastic scattering, and so any child neutrons will be slow neutrons that can’t fission any subsequent u-238 nucleii they encounter.

So the problems we face if we want to use uranium as a prospective fuel for generating heat are twofold:

1)      Using u-235 would generate plenty of heat, but it makes up a very small percentage of the overall amount of uranium mined and it would be very easy for the chain reaction to go out of control, necessitating some way of moderating the reaction.

2)      There’s far more u-238 around that we can use, but the neutrons it emits after a fission reaction generally don’t have enough energy to induce another one, stopping the chain reaction in its tracks.

U-235 appears to be our best bet for heat generation on the basis that it actually, you know, generates heat. Unfortunately the tiny quantity of uranium that is naturally occurring u-235 isn’t enough to sustain a chain reaction for any significant length of time anyway – this, incidentally, is why there aren’t any natural nuclear reactors scattered around the planet apart from the one at Oklo. In order to get enough fissile material to make a critical mass in the first place we need to increase the percentage of the uranium which is made up of the u-235 isotope, and this is called enrichment. Enrichment is achieved by a process called isotope separation, which can be summed up as “dumping some of the u-238”. The quantity of u-238 removed from the uranium (and hence the increased percentage of u-235) dictates what the enriched fuel is good for; nuclear fuel rods are about 3% u-235 which is enough to sustain the reaction. Weapons-grade uranium has a much higher percentage of u-235 (80%+) but is made using exactly the same process, which is why the West gets so uppity every time Iran says it’s totally just making some harmless nuclear fuel, honest. The removed u-238 is what’s known as depleted uranium; not much good for power generation except in very specific circumstances, but its incredibly high density makes it very good for tank armour and cannon shells — if you’re willing to overlook the fact that firing a DU slug into something hard is going to shear bits off, and then when people (including your own soldiers) inhale this radioactive dust everyone gets cancer.


Anyway, we now have a quantity of fissile material that can sustain a chain reaction. The next step is controlling this chain reaction so that it doesn’t go out of control, and this is done by surrounding the fuel rods in a medium that functions as a neutron moderator, slowing down and blocking neutrons released from the fission process, consequently reducing the percentage of neutrons that initiate further fission reactions. With u-235 this is something of a double-edged sword; fast, high energy neutrons released from fission reactions are less likely to set off a chain reaction, but low-energy neutrons that have been slowed by a moderator that do make it through to a u-235 nucleus are almost certain to. Paradoxically, then, surrounding the u-235 with a neutron moderator like water makes the reaction more efficient, not less. To properly control the fission process you need something that can outright block or absorb neutrons very very efficiently, and which can be fine-tuned to block some, none or all of the neutrons emitted by the chain reaction.

This is where the control rods come in. Control rods are made up of a material that’s very good at absorbing the type of neutron which induces a reaction in the fissile material you’re trying to control; in the case of u-235 we want to block pretty much all neutrons so a mix of materials which have different neutron absorption cross sections are used in order to cover the widest range of neutron energies possible. These control rods are lowered into and raised from the nuclear pile in order to moderate the nuclear reaction as required. The rods are lowered and raised by an electromagnet, so if there’s a serious incident where the plant staff lose control all they have to do is cut the power to the electromagnet and the rods will automatically fall all the way into the reactor in a failsafe position, shutting it down. This emergency shutdown procedure is known as a SCRAM, which comes from the possibly apocryphal story of the control rods for the very first nuclear reactor – Chicago Pile One, famously built on top of a university squash court – being held up by a rope. If there was an emergency, a man would quickly cut the rope with an axe, hence the acronym SCRAM standing for Safety Control Rod Axe Man.

This is how light water reactors function; water is used as the moderator and control rods are used to control the reaction, and the energy liberated from this chain reaction inside the fissile u-235 heats water which provides the mechanical work to drive the turbine. There are other reactors which use different moderators but all u-235 reactors work by the same principle of moderators and blockers. What about all that u-238 going to waste, though? Isn’t there some way we could use that to generate energy? As a matter of fact there is, and the next sciencey post I make will deal with fast breeder reactors, which is where the real potential for nuclear power lies. Also more safety features. I like safety features.

Tagged , , , , , ,


  1. innokenti says:

    You’ve referred top the safety issues in this post and on some previous occasions – are you planning to do a post going into detail on how the dangerous stuff arises and how we prevent it. And why it has gone wrong when it has in the past?

    More importantly, are there any lower profile dangers associated with other methods of energy production (including renewable ones) that have caused problems?

  2. Joe says:

    And so it begins… Thank you hentzau!

    This is the first I have heard about Oklo, fascinating! I did not think such a thing was possible. What would this site have been like while the reactor was cooking? Would the “reactor material” have been in a constant meltdown liquid state? I also wonder if there is any evidence of this from organisms that lived around this area at the time.

    • hentzau says:

      It was an on-off thing; groundwater acted as a moderator which enabled a chain reaction to start, this boiled away the groundwater and stopped the reaction, then the groundwater came back etc. etc. So it was self-controlling, in a way, and never got powerful enough to meltdown.

  3. Adam Benton says:

    What about thorium? I’ve heard people arguing its both not viable and the best thing since sliced bread.

    • hentzau says:

      This whole sorry story was started by a question about thorium reactors. In order to explain what I think about thorium reactors I have to explain the physics behind nuclear power plants, how you do nuclear power plants badly and how you do nuclear power plants correctly in order to discuss the merits (or not) of the thorium designs.

      • Adam Benton says:

        Well then I await further posts with great anticipation, since I keep hearing professionals espouse one position, think its settled, and promptly hear someone else disagree with it.

Leave a Reply