Fast spectrum reactors and supercriticality.

I’ve been reading this little essay on the history and progress of India’s nuclear energy programs recently. I encourage you to have a look, if you’re interested.

It’s a bit critical of India’s interest in Breeder reactor technology, suggesting that fast-neutron-spectrum “breeder” reactor technology has been tried, and rejected, throughout most of the nations of the world with significant experience in nuclear engineering.

There’s one particularly interesting claim made here:

“If the operating system failed to insert control rods fast enough, the increased reactivity would, in turn, heat up the sodium further; this chain could ultimately cause a fuel meltdown into a supercritical configuration and a small nuclear explosion. “

Helen Caldicott’s Nuclear Madness also says essentially the same thing:

“Once out of control, a fission reaction in a breeder could cause not only a meltdown but also a fully fledged nuclear explosion.”

A sufficiently enormously supercritical configuration would indeed have the potential to be the makings of a nuclear explosion – that’s how a nuclear fission bomb works, by compressing fissile material together into a massively supercritical mass.

Technically, a supercritical mass is any mass of fissile material with an overall neutron multiplication ratio, k > 1. k < 1 represents an ordinary, sub-critical mass, and k = 1 represents a, just marginally, critical mass.

k > 1 in the core of a fission reactor represents an unrestrained, massive power excursion – which usually results in the explosion of the reactor, if this condition is sustained for any significant length of time.

Such an event results in a significant release of thermal energy – in the presence of a water coolant, the resulting rapid production of steam pressure contributes significantly to the explosive force of such an event.

A significant amount of direct neutron and gamma radiation is released from such a critical assembly.

Such power excursions were responsible for the destruction of the reactor in the Chernobyl reactor disaster and the destruction of the experimental SL-1 reactor in the US in 1961.However, in all practical cases, it is not easily conceivable that the neutron multiplication factor would be great enough, and that the fissile material would stay together long enough, without destroying itself, and without reducing neutron multiplication via thermal expansion or doppler broadening, for a fully fledged nuclear weapon-style to occur. This seems to me to be quite implausible.

Any nuclear engineers out there care to share their thoughts?