http://nuketesting.enviroweb.org/hew/Nwfaq/Nfaq6.html
6.2.4 Other Fissile Elements
My critical mass estimates are for bare spheres at the densest STP phase. The calculations were performed using diffusion theory and a one-group representation of neutronic properties. The one-group parameters are fission spectrum averages calculated from the authoritative ENDF-6 evaluated nuclear data base. Where available I also list estimates from other sources which are usually based on more sophisticated numerical computations than what I have used.
The one-group calculation method consistently underestimates the true critical mass - primarily because it does not take into account the effects of inelastic scattering in softening the neutron spectrum. The one-group calculated critical mass estimates are thus lower bounds on the true value. Comparison between the one-group calculations and the actual values for the highly fissile isotopes for which good experimental data is available (U-233, U-235, Pu-239, and Pu-241) shows a consistent underestimate of 70-75% of the true value. For less fissile isotopes, where critical mass estimates have been offered by others (these are mostly calculated estimates also, but with more sophisticated models), the underestimates are more severe (at worst 22-29% of the 'true' value for Pu-242). This too is to be expected because the effects of inelastic scattering is relatively greater in less fissile materials. On the other hand, the estimates for extremely fissile transuranics like californium isotopes should be quite good.
................
I have seen no actual published estimates for californium isotope critical masses, so I have done one-group calculations for the two californium isotopes that are most plausible for weapons use (Cf-249 and Cf-251) as well as the often discussed Cf-252 (see the beginning of this section for more on this method). I used theoretical estimates for the density of californium based on its position in the periodic table, which are fairly reliable, since experimental data is unavailable. The estimation model used tends to underestimate critical mass sizes but should be quite accurate for highly fissile materials like thse isotopes. The calculated bare sphere critical masses are:
Cf-249 5.90 kilograms
Cf-251 1.94 kilograms
Cf-252 2.73 kilograms (это случайно не опечатка, четные изотопы не делятся тепловыми нейтронами)
By using a thick beryllium reflector the critical masses can be reduced to 40% or so of their bare value so that the most fissile of these isotopes, Cf-251, would have a reflected critical mass of 780 grams. This is arguably "in the gram range" since it is less than one kilogram, but it is obviously much, much closer to one kilogram than to one gram. Implosive compression can reduce this further. Just as a powerful and heavy implosion systems can produce low yield nuclear explosions from as little as 1 kg of Pu-239 (yield up to 100 tons, with the implosion system weighing on the order of 1000 kg), a relatively large implosion system could produce a low yield explosion from as little as 200 grams of Cf-251. The yield would be proportionately smaller of course, around 20 tons, and the total mass on the order of 200 kg. The U.S. has tested devices with 20 ton yields using the vastly cheaper plutonium (though much more of it) which only weighed 13 kg.
Experience with small nuclear devices indicates that it is impossible to make a nuclear device with a total mass less than the bare sphere critical mass of the fissile material used. Beryllium reflectors and high explosives can reduce the fissile mass required as indicated, but at the expense of adding more weight than is saved. Thus a nuclear device smaller than 2 kilograms or so using Cf-251 is almost certainly impossible.
The isotope Cf-249 results from the beta decay of Bk-249 while the heavier isotopes are produced by intense neutron irradiation of this product. Logically one would expect that Cf-252 would be more expensive then than Cf-249. However, in Sept. 1998 the ORNL Isotopes Division charged $180.25 per microgram of californium-249, and $56.00 per microgram of californium-252, indicating that factors other than ease of prodcution were setting the price. If Cf-251 could be manufactured in quantity for the 1998 sale price of Cf-252, then a critical mass would cost $11 billion.
То есть, используя Калифорний можно получить ядерный боеприпас массой единицы килограм. Может быть используя элементы 114, 116 и 118 можно получить и меньшие массы, не знаю. А если взять 9г анти-свинца, охлажденного до температуры перехода в сверхпроводящее состояние, удерживаемого магнитным полем и слегка подогреть, то это просто мечта террориста!
К тому же не надо забывать, что калифорний слегка так нагревается собственной радиацией.