A Comparative Study of Probabilistic and Deterministic Methods for Multi-Group Diffusion Constants Generation

Th e Monte Carlo (MC) method is a powerful and flexible approach for multi-group diffusion constants (MGDCs) generation due to its capabilities in a complex geometry modeling, high accuracy in neutronic calculations and utilizing continuous energy neutron cross sections data libraries. Th is paper describes a comparison study of a new method in the MCNP5 MC code to generate MGDCs such as scattering cross section and diffusion coefficient which are not directly obtained from MCNP code output. In this study, the results of the proposed method in the MCNP5 MC code to generate diffusion coefficient and group cross sections such as slowing-down cross section are compared with those of the WIMSD5 deterministic lattice code for two case studies including a typical slab and a PWR pin-cell benchmark problems. Th e generated MGDCs via applied method shows a good comparison to the reference deterministic values.