What methods are used to model/predict fission reactions?

Topher925

This is a stupid question, but it has always been a question of mine that I have never been able to find a strait answer. To be more specific, is quantum electrodynamics or special relativity used to model or predict fission reactions such as those found in a nuclear explosion or inside a nuclear reactor? Do the classical laws of conservation of mass and energy still apply or only conservation of mass/energy?

Astronuc

In brief QFT/QED/QCD are not used in the prediction or modelling of fission reactions in reactors or weapons. Rather, special solutions of either neutron diffusion or transport theory are used.

The cross-sections (probabilities of a neutron reaction) are known from experiment. The cross sections are based on the attenuation measurements of neutron beams and the measurement of resulting reactions, which include scattering, absorption (with gamma emission), and absorption (with fission).

For nuclear fuel and reactors, the microscopic cross-sections are developed into macroscopic cross-sections which are used in so-called lattice codes, like PHOENIX or CASMO, or a proprietary code used by one of the reactor suppliers. The lattice codes then determined the relative power level of a cross-seciton of the fuel assembly as a function of burnup (exposure/depletion), and these results are used in core simulator to determine the fission rate and more particularly the thermal power distribution during operation of a reactor.

Most commercial fuel suppliers use a modified 2-group diffusion theory.

Nuclear weapons are quite different the commerical reactors, and they use very specialized codes the model the physics of those systems.

Topher925

Thanks for the reply Astronuc. Very informative.

vanesch

I could add that one uses also a lot Monte Carlo codes to simulate neutronic behaviour. They have the advantage of not having computer time to depend strongly on the complexity of the situation, and not needing many a priori assumptions about the behaviour of the solution. On the other hand, they have statistical errors, and they are CPU hogs.

But as Astro said, all nuclear happenings in all these codes are based upon experimentally derived cross sections. No nuclear theory is used in any of these codes. The experimental cross sections are freely available and can be found on the website of the NNDC for instance.