Improvements in Neutronics/Thermal-Hydraulics Coupling in Two-Phase Flow Systems Using Stochastic-Mixture Transport Models

In this NEER project, researchers from Oregon State University have investigated the limitations of the treatment of two-phase coolants as a homogeneous mixture in neutron transport calculations. Improved methods of calculating the neutron distribution in binary stochastic mixtures have been developed over the past 10-15 years and are readily available in the transport literature. These methods are computationally more expensive than the homogeneous (or atomic mix) models, but can give much more accurate estimates of ensemble average fluxes and reaction rates provided statistical descriptions of the distributions of the two materials are know. A thorough review of the two-phase flow literature has been completed and the relevant mixture distributions have been identified. Using these distributions, we have performed Monte Carlo criticality calculations of fuel assemblies to assess the accuracy of the atomic mix approximation when compared to a resolved treatment of the two-phase coolant. To understand the benefit of more advanced binary stochastic mixture models, we have also compared Levermore-Pomraning BSM transport calculations to atomic mix and benchmark calculations in mixtures of a material in a near void. These comparisons yield valuable information about both the ensemble average scalar flux and the variance in the scalar flux.