Parallelization of Neutron Transport Approximation via Monte Carlo

Abstract

This paper discusses the parallelization of a Monte Carlo-based neutron random walk FORTRAN code in D2O (heavy water). In nuclear environments, proper selection of material is crucial is minimizing the radiation dose that an individual receives upon exposure (this material is placed between the radioactive source and the individual). Thus, the knowledge of how radioactive particles move through such materials is of great importance since one can estimate how many of such particles can penetrate a material. Neutrons will be the radioactive particle that will be analyzed since they represent one of the most important types of radiation in nuclear environments. The motivation to parallelize a Monte Carlo-based neutron random walk code lies in the fact that actual neutron sources vary in strength, ranging from the order 1E2 to 1E14 neutrons. As such, when simulating this many particles in a serial code, the computation time will be extensive. Thus, the focus of this paper is to document the parallelization of the abovementioned code.