In the blogosphere I have noticed that one blogger claimed that the radioactive silver spread around by the Fukushima accident was due to the use of silver in control rods. While one paper I read suggested that the reactor one used boron carbide, I can not rule out that the reactors used silver control rods.
Control rods are used on almost all reactors to control the rate of reaction, the further out they are pulled from the core the faster the reaction occurs. You can think of the control rods as the accelerator pedal of the reactor.
Silver has a high cross section for neutrons and as a result would make a good control rod material, the ideal control rod would.
- Last forever
- Cost nothing to make
- Not become radioactive while in service
- Behave nicely even during a horrible accident
The first issue is an interesting one, designs for control rods vary from reactor to reactor. One common choice is to use boron; this is because one of the isotopes of boron has a very large cross section for neutrons.
The cross section for neutrons is expressed in barns; this is an old unit of measurement which dates back to the Manhattan project. The idea was that if everyone expressed the cross section areas in barns rather than square meters then if a spy saw a cross section then it would be just a meaningless number. I suspect that the term barn relates to barn door.
The choice of boron has a sting in the tale which can come back to bite you on the arse. The problem is that any isotope with a very high cross section for neutrons will not be needed in large atom numbers. Each time it catches a neutron then one less atom will be present, so this can cause a change in the effectiveness of the control rod. This idea is known as burning out the neutron poison.
While the slow weakening of a control rod’s effect is an undesirable effect, this effect can be used in a beneficial way. Some fuels which have very high fissile contents have a little boron blended in. The idea is that as the fissile atoms are used up the boron is also burnt up by the neutron bombardment. The overall idea is that the fuel keeps the same reactivity level throughout its whole life inside the reactor.
The reaction by which the boron works is
n + 10B –> 4He + 7Li
This reaction generates helium gas; one Russian design for a control rod uses a boron steel alloy. The problem with this design is that the life of the control rod is limited because the helium starts to form bubbles in the steel. These bubbles then harm the properties of the rod.
A common western design is to use boron carbide (B4C), the rods are made of a steel and have holes into which are place pellets of boron carbide. As the steel is separated from the boron we do not have the helium bubble problem, but if the rod is overheated then a reaction can occur between the stainless steel and the boron carbide. This is an exothermic reaction which forms metal borides and some carbon. For example
B4C + 4Fe –> 4FeB + C
One other disadvantage of boron carbide is that during an accident it can form methane; the methane can lead to the formation of methyl iodide during an accident.
An alternative is to use a cadmium-silver alloy; the nice thing about cadmium is that it is very selective. It has a very large cross section for slow thermal neutrons while for fast neutrons is has next to no cross section. As the thermal neutrons are more able to cause fission then the cadmium has the nice effect of selectively mopping up these neutrons thus altering the energy spectrum of the neutrons in the core.
As a result of the fact that cadmium is selective for slow neutrons, I think that a control rod based on only cadmium would be a poor choice for a fast reactor such as a sodium cooled fast breeder, for such a reactor I would be inclined to use boron as it has a simple broad graph of absorption cross section as a function of neutron energy.
Cadmium is a metal which has a series of non radioactive isotopes, so when the cadmium-113 (the isotope with the largest thermal cross section) swallows up a neutron it forms cadmium-114 which is non radioactive and has a small capture cross section for neutrons. In this way many of the cadmium atoms can swallow up a series of neutrons without forming much radioactivity. Also the radioactive isotopes of cadmium are mostly well behaved short lived isotopes.
106Cd, 1.25 %
107Cd, half life of 6.5 hours decays to 107Ag
108Cd, 0.89 %
109Cd, half life of 463 days decays to 109Ag
110Cd, 12.49 %
111Cd, 12.80 %
112Cd, 24.13 %
113Cd, 12.22 %, very long half life (7.7 x 1015 years or 7.700000000000000 years)
114Cd, 28.73 %
115Cd, half life of 53.46 hours decays to 115In
116Cd, 7.49 %
117Cd half life of 2.49 hours decays to 117In
On the other hand silver has two stable isotopes, both of which form radioisotopes when they swallow up a neutron. This means that silver control rods will make more long (half life > 1 day) radioactivity per million neutrons which they absorb than a cadmium control rod will.
Some time ago I visited a disused nuclear power plant in Sweden; it was a small heavy water plant which produced only 10 MW of electric power and heat for district heating. In the reactor containment I saw the area where the used fuel used to be stored (the fuel had been taken away long ago) but the control rods remained locked inside the storage area. They were being left there to decay while everyone is waiting to decommission the reactor building. The thing about decommissioning is that the longer you wait the lower the levels of many irksome isotopes. For example the 60Co which forms as a result of the cobalt impurities in stainless steel will become half as strong every five years, thus by waiting for 50 years this radiation source will become one thousand times weaker.
Due to the fact that the control rods are exposed to such high neutron fluxes when in use, and as they are intended to absorb neutrons they can become very active.
When an accident occurs and a core melt occurs, it is likely that silver containing control rods will melt and start to form fine silver rich particles. This is likely to be a good thing as iodine has a strong affinity for the silver; this aerosol of silver may help to trap out radioactive iodine inside the plant. On the other hand if the silver particles are of the “wrong size” then maybe they will assist the escape of the radioactive iodine. One of the key features of the Chernobyl accident was that ruthenium tetroxide (RuO4) was formed; this is a very volatile and strongly
oxidizing metal oxide. The RuO4 enabled the 103Ru and 106Ru to form a coating on steel surfaces. These steel surfaces included both parts of the plant and also fine steel particles which were then able to escape from the plant. This is likely to be the reason why ruthenium rich hot particles were observed after the Chernobyl accident.
Ruthenium is a nice metal which I have a deep love of; I picked up this liking for it when I worked for Tony Hill. He joked that he had a special attraction to [RuHCl(CO)(PPh3)3] which is a complex formed by heating ruthenium chloride in methoxyethanol with triphenyl phosphine. I think I can see why Tony likes this complex; it is a useful starting material for a series of other things. It is also possible to make an osmium version of this complex but I will save my views on osmium for another day. The formation of this rather interesting looking compound is related to the work of Vaska. Vaska is a chemist from Eastern Europe who is something of a genius, he did a lot of nice chemistry with elements such as iridium. But lets get back to control rods.
So sometimes formation of solids or particles are a good thing and sometimes a bad thing.
One alternative to boron carbide, boron steel and indium-silver-cadmium alloys is to use hafnium. This is an interesting element; while zirconium has a very low cross section for neutrons (it is close to transparent to neutrons) hafnium is a very strong absorber of neutrons. As the elements are so similar in chemistry hafnium is commonly found in zirconium minerals, the zirconium used for nuclear reactor applications is normally a special low hafnium grade.
The hafnium is used in the form of hafnium hydride in control rods; these control rods are unlikely to react violently during a reactor accident or to form troublesome gases.