The purpose of this paper is to describe a few of the suggested advanced concepts for fast breeder reactors and to compare these with the standard approach as to their potential advantage. I have attempted to estimate the economic effect of full technical success with each of the proposed concepts. The proposed concepts include: (1) single sodium system, (2) steam-cooled core concept, (3) direct cycle reactor using potassium as reactor coolant and working fluid, (4) molten plutonium-fuel alloy circulated and cooled by a jet of sodium, (5) settled-bed core, (6) molten salt concept, and (7) paste-fuel system.

Many measurement of neutron spectra have been made in water assemblies. In these system a large part of the spectrum is given by an essentially Maxwellian distribution at the moderator temperature which is insensitive to the scattering model. Hence, the test of the energy exchange scattering kernel is in the difference of the spectrum and the fundamental or Maxwellian component. The figure shows the spectrum for the Nelkin model. Once the fundamental has been subtracted neither theoretical model seems to fit the data extremely well. Below kT the Nelkin model is a closer fit than the gas model but it underestimates the deviation from the Maxwellian whereas in the joining region it overestimates the distortion.