FDBACE code for fast Doppler broadening ACE format based on parallel processing

Abstract

In this article we introduce a code called FDBACE (Fast Doppler Broadening ACE) which has been developed as a tool for MCNP users to enhance the generation of high temperature ACE data tables. In this code, we developed new broadening, thinning and unionization subroutines, implemented in FORTRAN programming language, for directly broadening the ACE data table at desired temperature. For fast and exact broadening of the ACE data tables, we used a new parallel broadening algorithm based on the kernel broadening approach and shared memory scheme (using OpenMP). The thermal scattering table for molecules and crystalline solids, and probability table for unresolved resonances are computed by interpolating data between two bracketing temperatures. This code, which reads the ACE data table file and broadens it to the desired temperature and then writes it to a new ACE data table file, can be easily converted to an internal module embedded in Monte Carlo source code for In-line Doppler broadening.Accuracy, performance and formatting of FDBACE are evaluated for validation of the proposed method by three different test cases. The results obtained from FDBACE code, shows that it has a good performance over previous codes such as NJOY, SIGACE and MAKXSF in generating high temperature ACE data tables.