Monte Carlo method with complex weights for neutron leakage-corrected calculations and anisotropic diffusion coefficient generations

A Monte Carlo algorithm that considers the neutron leakage effect specified by buckling to generate group constants for use in reactor core designs has been developed. For the neutron leakage effect, the physical flux is factored as a product of the spatial fine structure within a unit cell or assembly and a macroscopic cosine distribution, similar to traditional lattice physics calculations. In this algorithm, complex weights are explicitly treated. The implicit capture and Russian roulette methods are applied to the complex weights. The fission sources with negative weights must be cancelled with an appropriate weight cancellation technique. From the B1 method, anisotropic diffusion coefficients can be generated with this Monte Carlo algorithm. This newly developed algorithm has two different modes. One mode is the k-eigenvalue mode, in which the eigenvalue keff corresponds to a fixed buckling. For the other mode, the eigenvalue is the buckling that corresponds to a user-specified keff (e.g., keff = 1). This newly developed method has been successfully demonstrated with a production level continuous energy Monte Carlo code. Critical buckling values, group constants and anisotropic diffusion coefficients have been obtained for the unit fuel pin cells and the fuel assembly, which are similar to the results obtained with a deterministic reactor core design code.

► The B1 method is implemented into Monte Carlo calculation to include leakage effect.
► A Monte Carlo algorithm where complex weights are treated has been developed.
► Group constants and diffusion coefficients can be generated for a unit cell.
► The method is verified by comparing its predictions with a deterministic core design code.