Analysis of Some Computer Codes Used for Evaluation of Spent Nuclear Fuel Radiation Parameters

Abstract

The radiation parameters such as radionuclide content and activities, fluxes and energy spectrum of gamma and neutrons of spent nuclear fuel are essential when planning further spent fuel management options – interim wet or dry storage or disposal into a geological repository. Radiation parameters determine the design of a storage or disposal facility, what materials, structures and thicknesses of structures should be used to provide adequate biological shielding.
 Experimental measurements of spent fuel radiation parameters are rather complicated and expensive, therefore numerical methods are widely used. Various computer codes (APOLLO, BOXER, CASMO, FISPACT, ORIGEN-S, WIMS, etc.) have been developed to simulate the irradiation processes of nuclear fuel and to obtain resulting radiation parameters. Irrespective of the used computer code, the input data firstly must be entered into that code. When simulating nuclear fuel irradiation and burn-up in the reactor core, the geometrical parameters of the fuel assembly, materials’ data (chemical compositions, densities), the operating parameters of the reactor (power, operation time, coolant parameters, etc.) shall be entered into the program as initial data. Fairly often approximations of the input data are performed, for example, fuel rods in a fuel assembly are homogenized and geometrically described as a solid cylinder, the reactor operation time is assumed as continuous and at constant power. The particularity of the input data and accepted assumptions depend on what initial information is available and on the capabilities of the computer code. The modelled spent fuel radiation parameters depend not only on the input data and assumptions, but also on the cross-section databases that are used in computer codes. Computer codes TRITON, ORIGEN-S and FISPACT have been used to model the concentration of actinides and fission products in the spent fuel from the RBMK-1000 reactor. The obtained results are compared and possible reasons for the differences in the modelling results are discussed.