[Version française] There are several proof of the use of sodium in the Chernobyl RBMK design in a fast reactor design. I wish first to give the explanation : because a fast reactor is a breeder allowing for a good production of Pu239 which the Soviet Union desperately needed to have as many nukes (and in fact more) than the USA. And a sodium-cooled fast reactor allows to produce a lot of sodium-24 (sodium activated by neutron capture) which is a very good neutron source (in combination with e.g. heavy water), much more powerful than alpha + beryllium neutron sources (so it improves considerably the yield of nuclear devices). Sodium-24 nevertheless has a short half life (14 hours) and must so be produced close to the potential war zone or close to the ICBM base. You’ll note that the only recognized sodium reactor in Russia today is close to a warhead plant (Sverdlovsk-45) and that Russia has many mobile ICBMs available. The main US prototypes are in Idaho, close to ICBM bases, and the three French projects are in the south-eastern part of the country (Superphénix in Grenoble, Astrid and Phénix in the Gard département, Rhapsodie in Cadarache), so closer to the Middle East (a point especially meaningful for Astrid, the new project under development, even “downsized to 300 MW”).

The very first element is that sodium is a very easy explanation for the positive void coefficient of the RBMK design ; whereas it is very weird to have a positive void coefficient with a liquid water moderator because the point is that the increase in the fission rate in the reactor gets the water to boil, which relieves the fissile material of some pressure, reducing effective mass and thus slowing down fission. This is very rarely noted. Another very easy element is that it can be found that RBMK reactors used spent VVER fuel (VVER is the common light water reactor type). The only kind of reactor that uses spent fuel is a fast neutron burner type.

“The Decay of Communism: Managing Spent Nuclear Fuel in the Soviet Union, 1937‐1991” de Per Högselius (published article https://onlinelibrary.wiley.com/doi/abs/10.2202/1944-4079.1040 ) we read that “A “tandem” system [for spent fuel management] was partly realized in this context, in which spent fuel from VVER nuclear power plants, submarine reactors, and research installations was reprocessed and used for the production of RBMK (Chernobyl‐type) fuel. […] . RBMK spent fuel was not reprocessed.”

I’d also like to point that some Soviet nuclear submarine (attack-type, the “Alfa” class) had fast neutron reactors, cooled with lead-bismuth (because it is smaller for the same power output and likely because of the desire to produce quickly neutron sources through activation) and that by listening to the documentary on the Kursk submarine by Jean-Michel Carré (for French public TV Fr3, Thalassa) we learn that “the Kursk had a reactor of the same brand than Chernobyl”. The Kursk actually was an attack-type submarine with torpedoes and cruise missiles only, suggesting indeed again use of fast neutrons in Chernobyl. Dyatlov, the Soviet engineer in charge of the experiment in Chernobyl that led to the blast, used to work on the setting up of nuclear reactors in submarines, in Komsomolosk-na-Amure, until 1973, before being sent to Chernobyl. It can be expected that an expert in fast neutron reactors for submarines was sent to work in another fast neutron reactor.

A great proof for sodium I point is simply the findings of sodium mineralized onto the uranium lava of the exploded plant by researchers. Such mineralization involves a violent compression (because mineralization usually happens mostly deep underground), linked clearly to the contribution of sodium to the blast (link to article).

One can also observe the distribution of fission products : there is a difference between the fission products of slow neutron fission and the fission products of fast neutrons (not “high energy”, this is not the Sun or a thermonuclear bomb).

By looking into a publication regarding fission products in the sky of Helsinki just after the blast, I find a maximum of 400 000 microBq/m3 of Cd115, to be compared with the Ru106 (630 000 microBq/m3 for the same sample the same day). 400 000 uBq de Cd115 = 21,19 picograms, 630 000 uBq de Ru106 = 5,1 nanograms, so Cd115 here amounts to 0,41% of the Ru106. By looking at fast neutron fission data for U238, we find 1,5% (0.046/3) page 6 of this other publication. It was quite hard to find data for Cd115 under fast neutron fission, yet the trend is identical with U238, U235, Pu239 (it’s mostly a translation but ratios do not change much). But to account for the difference between 0.41% and 1.5%, one must consider a very simple element which is the very quick radioactive decay of Cd115 (2,3 days, vs. more than one year for Ru106) and the fact that the sample studied was taken on the evening of April 28, so almost 3 days after the explosion. Which means that our ratio is relatively consistent with fast neutron fission.

Lastly I’d point to the paper by Filippov, Urutskoev, Lochak, Rukhadze (Condensed Matter Nuclear Science, Ed. J. P. Biberian, World Scientific Publishing Co., Singapore, 2006. p. 838–853)  which notes that the explosion was very quick, whereas a thermal neutron reactor nuclear blast would have needed about 20 seconds to happen… with a fast neutron reactor it is logical that the explosion is much faster. Everybody knows this when it comes to fast neutrons reactors, the “reaction time” is very short.

The positive void coefficient is used to unload plutonium from the breeder without shutting down and thus avoiding detection by satellites watching infrared signals from reactors. It means however that when fission accelerates, there is no “tamper” like bubbling water vaporizing to reduce effective mass, on the contrary sodium, when warmer, increases the pressure onto the fuel (which is the property used to reduce the power of the reactor and unload peacefully the Pu239 without shutting down, because the reactor was brought to higher temps earlier, so that the sodium is still warmer and still pressures the core once the reactor has been violently brought to a very low regime thanks to the famous pressure bars). And sodium burns very violently, even explodes, when on contact with air or water. So the nuclear reaction in the reactor is uncontrolled, the pressure of the sodium also increases, the supercriticality is explosive, and at the moment the reactor is volatilized the sodium gets outside and you simultaneously have the nuclear explosion and the sodium explosion. Hence the VERY hot temperatures suggested by the “blue flash” (up to 15 000°C). Since you have a nuclear chain reaction you must have the delayed neutrons, hence the second explosion a bit more than 2 seconds after the combined nuclear + sodium blast. So here you have the reason why “so much of the fuel was impossible to locate” and the reactor itself totally destroyed, as noted by Filippov et al in their 2006 paper. And sodium explains also why the fire was so difficult to stop.

The programme by the Direction Générale des Armements on “dual use technologies” for “new weapon technologies” called “ASTRID” alike the new programme for a sodium cooled fast reactor (launched more or less at the same time) is alike a smoke signal to the industry that the point of the new weapons is to use sodium-24 as a neutron source… inflammable and very powerful… I explain very well everywhere in this website that all conventional weapons need neutron sources too because all anti armour weapons are nuclear since WW2 and even a bit earlier, since the discovery of fission in 1934 (1936 and the Spanish Civil war, for Italian and German shells, of course). The best way to learn about it is to read my book, free to download.