Weird but fun: Effects of Neutron Transmutation of Silicon to Raise Thermoelectric Figures of Merit.

[NNadir]

Wandering around in a desultory fashion in the literature, I came across this paper: Gaidar, G.P. The Effect of Thermal Treatment on the Thermoelectric Figure of Merit of Silicon Doped Using Nuclear Transmutation. Surf. Engin. Appl.Electrochem. 57, 425–430

Recently my son and I were discussing the properties of silicon in a neutron flux in connection with corrosion resistant refractories, not thermoelectric devices, and drifting around the literature I came across the paper linked above. It's something I never would have thought of doing.

Thermoelectric devices convert heat directly into electricity; these devices are mostly known from space mission RTGs (Radioisotope Thermoelectric Generators) used on Space missions to Mars and the outer planets. The Voyager missions now beyond our solar system are still powered by devices of this type launched in the 1970s. As they have no moving parts, they are very reliable.

The downside is their low thermal efficiency and current research is devoted to raising the "Figure of Merit," a dimensionless number reflecting efficiency, which generally varies with temperature.

Precise nanostructure of thermoelectric devices is important in adjusting two conflicting properties, thermal conductivity and electrical conductivity in opposing direction of magnitude, not an easy trick, but doable.

The paper puts it this way:

Hence, a good thermoelectric material must have simultaneously high electric conductivity, high thermo-emf and low thermal conductivity.

In this case the author was studying phosphorous doped silicon as a thermoelectric material and evaluated two ways of making it, one by traditional melt processes in which the phosphorous dissolves in the molten Si forming an impurity, the other involved placing the silicon into a nuclear reactor and irradiating it with neutrons. Under these conditions, one of the isotopes of silicon, Si-30 is transmuted into phosphorous's only stable isotope, P-31.

The rationale for the latter approach was to more precisely locate the phosphorous atoms.

From the text:

During the irradiation of the silicon monocrystals by the flux of slow neutrons, a certain amount of silicon atoms turns into phosphorus atoms, which are uniformly distributed throughout the bulk of a crystal. However, because of a strong interaction of neutrons with Si atoms, the atoms of phosphorus are generally located in the interstitial spaces. In order that all (or at least the majority) of the phosphorus atoms (obtained as a result of the neutron radiation) appeared to be in the lattice nodes and could exhibit electric activity, a special technological annealing is needed (at 800°С for 2 h).

Both processes required annealing, and it was determined the cooling rate was an important factor in performance.

The figure of merit was better for the neutron irradiated material, not enormously so, but better.

Neutron induced transmutation is often discussed with respect to certain components of used nuclear fuel for various reasons, mostly in connection of shortening their half-lives, and in fact, it is a continuous process in nuclear fuels in use. Neutrons are also widely used as analytical tools, most typically in diffraction experiments; my son participated in some of this type of work during an Oak Ridge internship.

I can't recall if I've ever seen an application quite like this however. Maybe I have, but if so, I've forgotten it.

I wouldn't expect it to go commercial, but it's pretty cool, and in my conversation with my son about silicon in neutron fluxes, it's worth discussion.

Don't try this at home.

I wish you a happy New Year.