Pique the Geek 20110320: How Nuclear Reactors Work Part the First

With the news about the horrible failure of the nuclear reactors in Japan, it occurred to me that many people do not really understand how nuclear reactors work.  This is the first part of a short series designed to demystify how nuclear reactors work.

All methods for generation of usable amounts of electricity require some sort of energy source.  In photovoltaic units, the electromagnetic energy in solar (or other) photons is the energy source.  In geothermal plants, the interior heat energy from the earth is used, whilst in wind plants the kinetic energy of moving air is used.  Hydroelectric plants use the kinetic energy of moving water.

Fossil fuel fired plants use the potential energy contained in coal, oil, or gas by converting it to heat by combustion.  Finally, nuclear electricity uses the potential energy of a very few heavy elements’ nuclei that is released as heat in the reactor.

Let us first do away with some misconceptions.  Except for photovoltaic arrays and nuclear batteries, ALL current electricity is generated by spinning the rotor of a generator.  This rotor has wire windings that spin within a strong magnetic field provided by the stator of the generator, and thus the kinetic energy of the rotor is converted into electrical energy.  It really does not matter what energy source is used to make the rotor rotate, as long is there is enough to fulfill the requirements of the load to be serviced.

Photovoltaic cells operate differently, directly converting radiant energy into electricity, whilst nuclear batteries are also different.  In a photovoltaic cell, the energy from the photons impinging on it cause a charge separation, driving electrons through a load.  In a nuclear battery, the heat from a highly radioactive isotope is directed at thermocouples, and those thermocouples experience a charge separation because of heating a junction of two dissimilar metals.  That is also how digital thermometers work.  Unlike the other means of producing power, neither of these devices have any moving parts.  By the way, both of these produce direct current whilst most other devices produce alternating current.  It is alternating current that comes into our homes through the mains.

To make the rotor turn, some sort of energy is required, and the higher the load, the more energy is needed.  Except for wind and hydroelectric generators, heat is used to impart energy into a working fluid, typically but not always water, which is thus vaporized.  This high energy steam is then directed towards a turbine which is designed to spin efficiently, and that turbine is connected to the rotor, and the plant then produces power.  In wind and hydroelectric plants, the movement of air or water does the same thing, but when you get right down to it, they are still heat devices, because it the heat from the sun that causes the rains to fall and the winds to blow.

Another misconception is that all solar energy is photovoltaic.  This is just wrong.  There are many installations that use mirrors to focus sunlight onto tubes filled with a fluid that is heated, and that hot fluid is used to boil water and spin a turbine.  Those solar installations are just like all other turbine plants except for the heat source.

In fossil plants, and I said above, chemical fuel is burnt with air to produce heat, and that heat is then used to boil the water.  In a nuclear plant, the energy of fission of certain elements is used to boil the water.  Except for the heat source, they are just like all other turbine plants.  Some people still think that somehow nuclear energy is directly transformed into electrical energy, and that is just not correct.

For the purposes of this discussion we shall confine our scope to nuclear fission, the splitting of very heavy atoms to lighter ones.  Nuclear fusion, the merging of very light atoms into heavier ones, is the process that stars use, but we have not been able to control that reaction well enough yet for it to be a viable power source.  Except for some highly advanced and highly experimental research facilities, the best that humans have done with fusion is to develop the thermonuclear device, commonly called the “hydrogen bomb“.

Both fission and fusion “work” because, except for a single nuclide, either in isotope of iron or nickel (this is still controversial, but it look like the better and newer data favor nickel), have more mass than they should.  When very heavy nuclei are split, they become more stable in the sense that some of the excess mass is lost.  The same concept applies when very light nuclei are combined.  This excess mass is lost, and lost according to Einstein’s famous mass energy equivalency formula:

E = mc²

The amount of energy produced when mass is converted to it is staggering.

It is pretty much correct to say that most nuclei, especially very heavy ones, can be induced to undergo fission under the right conditions.  There are some exceptions and caveats, but given enough energy and the right method, most heavy nuclei can be fissioned and energy is released.  That does not mean that you always get more energy back than you insert, and this is just a very broad generalization.

Even for a given element, the nuclear properties can be very, very different.  I shall use the example of uranium in the next paragraph, but first we need to define some concepts.  The first one is the fact that, in the cases important to power plants and bombs, neutrons are involved.  Remember back to high school:  the neutron is one of the basic building blocks of nuclei.  The reason that neutrons are important is that, since they carry no electrical charge, they are not repelled electrostatically by the highly positively charged nucleus.

The second concept is that free neutrons are always in motion.  Some move slowly, at about the same velocity as regular matter at a given temperature.  These are called thermal neutrons, meaning that their velocity is relatively slow.  Most neutrons that are produced in nuclear reactions move at much higher velocities, and are called fast neutrons.  Atomic nuclei have very different behaviors when exposed to the two different speeds of neutrons.

The third concept is that when neutrons interact with nuclei, they either scatter away, just imparting some kinetic energy (think of hitting a cue ball into a rack of pool balls and the entire rack, without any relative motion with respect to each other, just is pushed back a little, with the cue ball being deflected), or is absorbed (think of the cue ball merging with the rack of 15 balls to make a 16 ball set).  It is the latter that is important.

In this case, the new nucleus becomes fundamentally unstable, highly energetic, and has to dispel the energy.  In the case of fissile materials, that nucleus splits up into (usually) two lighter ones, each roughly half the mass of the original one.  This is nuclear fission in a nutshell.  Imagine the rack of pool balls, plus the cue ball, splitting into two tightly bound groups, in the case of pool balls, eight balls each, slowly heading towards the rear pockets.  That is a pretty good visual representation of nuclear fission.

But in nuclear physics, it gets a bit more complicated.  Instead of a rack of 15 numbered balls, the rack for uranium-235 contains 235 balls, 92 of which are protons, with that positive charge.  The rest are neutrons, with no charge.  It turns out that the larger the number of “pool balls”, the more neutrons are required to stabilize the nucleus.  In this case, it takes 143 neutrons to keep that heavy nucleus from flying apart immediately.  This is the direct result of the strong nuclear force, which overcomes, at least in part, electrostatic repulsion from the protons trying to get away from each other as far as possible.  This is getting much too Geeky!

Well, some nuclei undergo spontaneous fission, in that they just naturally tend to split into two, lighter nuclei, releasing energy in the process.  This is not really a significant factor in nuclear power production, but it does occur in nature.

The next important concept that needs to be explained is that of the nuclear chain reaction.  This is extremely important, and if you do not understand it from the text here, ask more in the comments.  Let us take a given fissile nucleus, say U-235.  When bombarded by a thermal neutron, unless it is just scattered, it is absorbed to form an extremely unstable excited state of the nucleus, U-236.  This nucleus immediately ( on a time scale of attoseconds) fissions into two lighter ones, liberating much of the binding energy as heat.  But, like they say on TeeVee, there is MORE!

When that nucleus is fissioned, more neutrons are liberated than were required for the initial fission.  Those neutrons are available to make other U-235 nuclei to undergo fission, and those release even more.  Carried to the limit, it is a nuclear (very clumsily called an atomic) bomb.  But we do not want to carry it to that limit in a power plant.

Now, and I know that I am getting Geeky, there is the concept of criticality.  This is also extremely important.  Please stay with me here, because if you do not understand this concept, you might as well read another blog.

Remember those neutrons?  Well, most of them just escape out of the slug of U-235 and so do not interact with many other nuclei, unless you fool around with the geometry of the mass.  The exact amounts remain classified, but these facts are well known.

Take about 15 kilograms of highly enriched U/-235 (HEU)and put it into one, or even more, long tubes.  Nothing.  However, if you take the same mass of HEU and make a sphere out of it, a nuclear detonation occurs.  In the case of the long tubes, more neutrons escape from the material because of the large surface area, so not enough are available for it to go critical.  In the case of the sphere, more neutrons are retained to impinge on other nuclei.

Power plants do not use HEU, but rather, depending on the design, either natural uranium (about 0.72% U-235, most of the rest being the nonfissile U-238), or low enriched uranium, up to about 5% U-235.  The plants in Japan used the latter material, as do current US models.  But more than fuel is required.

The basic components required are the fuel, a moderator, and control rods for current designs of reactors.

.  The moderator is required to slow down fast neutrons to the speed of thermal ones, because thermal neutrons are much more efficient than fast neutrons for initiating fission.

Several materials are suitable as moderators.  The trick is to use a material that only slows down neutrons by multiple elastic collisions, but does not absorb them.  The very first nuclear reactor, built during World War II, used graphite as a moderator.  The Chernobyl reactor that failed 25 years ago was a graphite reactor, and although graphite is not usually thought of as combustible, it will burn at the temperatures that were reached there.  This was a major factor in the huge magnitude of the accident there, because the huge graphite fire dispersed the radioactive material widely, and also lifted it high into the atmosphere.

Another useful moderator is plain water.  Current US designs, and the ones in question in Japan, are water moderated reactors.  This has several advantages, although is far from perfect.  First of all, some working fluid is necessary to transfer the heat from the reactor core to boilers, and water is ideal for that.  Second, water will not burn, so can not contribute to the energy of a reactor failure.  Third, if a reactor loses water, (a very bad thing), the actual rate of reaction decreases because of the lack of moderation of neutrons.  However, the heat building up anyway pretty much wipes out that advantage.

Control rods are used to stop the fission process by absorbing neutrons as they are emitted by the fuel rods, essentially turning off the reactor.  There are several materials that absorb neutrons strongly enough to work, and silver is one of them.  Indium is another, and so is cadmium, which has an extremely high cross section for thermal neutrons, absorbing them without undergoing fission.  Boron is also a pretty fair control material, but not as good as cadmium.

To oversimplify heavily, to produce power in a nuclear plant, fuel rods are loaded with pellets of low enriched uranium oxide (the free metal is not used, because it burns, and also because it melts at 1132 degrees C, whilst the oxide does not burn and melts at the much higher 2865 degrees).  An array of those rods are later loaded into the core of the reactor.  Typically, the fuel rod tube is made of a zirconium allow because it is high melting and resistant to corrosion.

But first, control rods are loaded into the reactor.  These rods, in the current US and Japanese designs, are typically an alloy of 80% silver, 15% indium, and 5% cadmium.  The mixture of the three metals is better than any one alone, since each element has a relatively narrow window of neutron energies that it absorbs the best, so a wider range is absorbed with the mix.  The rods are encapsulated in either stainless steel or a zirconium alloy.  Since the control rods do not get as hot at the fuel rods, stainless steel is sufficient.

The control rods are movable, being attached to a steel beam that can be lifted, thus lifting all of the control rods at once.  Now the water is turned on and the fuel rods put in place and the reactor sealed.  All that has to be done to “turn on” the reactor is to begin lifting the control rods, thus allowing the thermal neutrons to initiate fission in the U-235.  The fuel rods begin to get hot, heating the water, which in turn is circulated under high pressure (to keep it from boiling) to a heat exchanger, which is used to heat other water.  Thus, the water in the reactor core is circulated continuously through the heat exchanger and core, and unless there is an upset, never is released to the environment.  The water in the heat exchanger that the core water is boiled and the steam used to spin the turbine, and thus the generator.

As demand for electricity increases, the control rods can be lifted higher and higher, thus increasing the heat, and so the power, output of the plant.  The total output depends pretty much solely on the size of the reactor core, the maximum output being reached when the control rods are no longer within the fuel rod zone.  As demand decreases, the control rods can be lowered as needed to reduce the output.

If the plant needs to be shut down, the control rods are completely lowered into to the core, effectively stopping the original fission chain reaction.  However, it is not quite that simple, and here is what happened in Japan:

When the earthquake hit, the control rods automatically dropped all the way into the reactor, effectively shutting down the fission chain reaction, and plants are designed for this to happen.  As a matter of fact, typically the control rod array is not mechanically held in place, but rather is lifted by electromagnets so that is is not possible for a mechanical lifting device to jam.  In other words, when an off normal situation arises, the magnets deenergize and the control rods drop into place by gravity.

One would say, then, great, the reactor is shut down.  This might be true with brand new fuel rods, but not with ones that have been in service for any length of time.  Let us think about just what fission is:  the splitting of one large, heavy nucleus into two lighter ones.  Those lighter ones are, in very many cases, intensely radioactive, and because radioactive decay releases energy, even with the control rods in place, those fuel rods are still emitting heat, and lots of it (but not as much as when the control rods are raised).

So far, so good.  Since the plant was no longer producing electricity, the Diesel generators automatically activated to keep the circulation pump running to remove the heat generated by the fission products from the core to the heat exchanger.  Then the tsunami hit, flooding the air intakes for the generators.  This is what really went wrong.  The earthquake itself seems to have done little actual damage the reactors, but when the generators were flooded, things began to go wrong very rapidly.

There was battery backup for the pumps, but that was designed for temporary outages of the Diesel generators. In all, there were only about six hours of battery capacity, then the pumps stopped.  That sealed the fate of the site.  The fission products were still decaying, producing enough heat to boil the water out of the reactor cores.  The cooling pond at Reactor 4 also began to boil, since no water was replacing the water that was evaporating.

Note that Reactor 4 was not in service, as it was being refueled as I recall the news accounts.  The spent fuel rod pond became extremely hot, and the zirconium alloy tubes began to distort and finally leak their contents.  The same thing was going on in several of the reactors.  Remember, the control rods were preventing the actual chain reaction, but the heat of radioactive decay of the fission products was more than enough to compromise the tubes.

Much of the last few paragraphs is speculation on my part, because the containment buildings are much too highly radioactive for entry, but the observations are consistent with my hypotheses.  Only time will tell what actually has happened and how extensive the damage is.  I strongly suspect that this site will finally be sealed in one way or another, made into an exclusion zone, and essentially abandoned.  We shall see.

Next time we will examine the nature of some of the fission products.  In many cases these are much worse than the original nuclear fuel that went into the new fuel rods.

Well, you have done it again!  You have wasted many more einsteins of perfectly good photons reading this hot material.  And even though Ann Coulter renounces this

AND learns the difference betwixt words “maximum” and “minimum” (note that she used “minimum” when she meant “maximum” multiple times in this short piece) when she reads me say it,  I always learn much more than I could possibly hope to teach by writing this series, so keep those comments, questions, corrections (I suspect that there will be a few this time since I tried to pack so much information into a single post), and other feedback coming.  Remember, no scientific or technical issue is off topic here.  I shall stay around tonight for Comment Time as long as comments warrant, and shall return tomorrow for Review Time to field any late ones.

Warmest regards,

Doc

Crossposted at Dailykos.com, Docudharma.com, and Fireflydreaming.com.