The Materials and Performance Optimization (MPO) focus area within CASL has recently developed and released a 3D modeling framework known as MAMBA (MPO Advanced Model for Boron Analysis) to predict CRUD deposition on nuclear fuel rods. CRUD, which refers to Chalk River Unidentified Deposit, is predominately a nickel-ferrite spinel corrosion product that deposits on hot fuel clad surfaces in nuclear reactors. CRUD has a lower thermal conductivity than the zirconium fuel clad or zirconium oxide surface layer, and also incorporates lithium and boron impurities. Boron, in particular, is a strong neutron absorber. CRUD deposits are responsible for local temperature increases that can further accelerate CRUD formation rates, leading to localized corrosion induced failures. Furthermore, the boron incorporation in CRUD can produce severe reactor power level oscillations known as axial offset anomaly that can necessitate significant reductions in plant power to avoid these occurrences. Thus, it is critically important to accurately predict CRUD formation, including both the thickness, composition and the effects of CRUD on temperature and reactor power level.
MAMBA solves a general time-dependent 3D heat transport equation along a single fuel rod, with localized heat sinks due to sub-cooled nucleate boiling. MAMBA also models the fluid transport within the crud layer and the chemistry/thermodynamics required to treat the coolant chemistry (Li, B, H2, Ni, Fe, B(OH)3, and several ionic species) and for accurately determining the rate of CRUD precipitation. MAMBA includes an adaptive 3D mesh which “grows” the 2D crud surface as mass deposits onto the surface elements. The erosion of the CRUD’s surface due to the turbulent flow of coolant along the surface of the CRUD is also included. The figure below shows a prediction of CRUD thickness along a zirconium fuel rod with three spacer grids with mixing vanes. The effects of turbulent flow through the mixing vanes on each spacer are clearly visible (the blue “rings”). A magnified view of the region around the first spacer grid is also shown. As the CRUD forms on the hottest regions of the fuel clad, the local crud thermal conductivity varies in both space and time due to the changes in porosity. The change in porosity results from the slow, internal deposition of nickel ferrite (NiFe2O4) within the pores of the CRUD. The local porosity can also change more quickly due to the precipitation of lithium-tetraborate (Li2B4O7), which is the primary mechanism for boron incorporation that can produce axial offset anomalies. The output of MAMBA predictions of CRUD thickness and composition are now being used within CASL as input to neutronic models of reactor power level.