Lattice benchmarking of deterministic, Monte Carlo and hybrid Monte Carlo reactor physics codes for the soluble-boron-free SMR cores
Since the use of deterministic transport code WIMS can significantly reduce the computational time compared to the Monte Carlo (MC) code Serpent and hybrid MC code MONK, one of the major objectives of this study is to observe whether deterministic code WIMS can provide accuracy in reactor physics calculations while comparing Serpent and MONK. Therefore, numerical benchmark calculations for a soluble-boron-free (SBF) small modular reactor (SMR) assembly have been performed using the WIMS, Serpent and MONK. Although computationally different in nature, these codes can solve the neutronic transport equations and calculate the required neutronic parameters. A comparison in neutronic parameters between the three codes has been carried out using two types of candidate fuels: 15%235U enriched homogeneously mixed all-UO2 fuel and 18%235U enriched micro-heterogeneous ThO2-UO2 duplex fuel in a 2D fuel assembly model using a 13×13 arrangement. The eigenvalue/reactivity (k∞) and 2D assembly pin power distribution at different burnup states in the assembly depletion are compared using three candidate nuclear data files: ENDF/B- VII, JEF2.2 and JEF3.1. A good agreement in k∞ values was observed among the codes for both the candidate fuels. The differences in k∞ between the codes are ∼200 pcm when cross-sections based on the same nuclear data file are used. A higher difference (up to ∼450 pcm) in the k∞ values is observed among the codes using cross-sections based on different data files. Finally, it can be concluded from this study that the good agreement in the results between the codes found provides enhanced confidence that modeling of SBF, SMR propulsion core systems with micro-heterogeneous duplex fuel can be performed reliably using deterministic neutronics code WIMS, offering the advantage of less expensive computation than that of the MC Serpent and hybrid MC MONK codes.