Validation of Subchannel Code CTF for Sodium Fast Reactor Modeling

Liquid Metal Cooled Fast Reactors (LMFRs) use liquid metal as the primary coolant of the reactor core. First demonstrated in the 1950s, they were never fully deployed compared with light water reactor technologies. However, the early 2000s saw a resurgence of interest, particularly in Sodium Fast Reactors (SFRs) and Lead Fast Reactors (LFRs) as Generation IV designs, because of their potential to significantly reduce the amount and toxicity of nuclear waste in a closed fuel cycle. This investigation is part of Tractebel Engineering’s effort to evaluate new tools for SFR modeling. It evaluates the extended CTF subchannel code for thermal-hydraulic modeling of SFRs using new capabilities for low-Prandtl-number fluids, sodium heat transfer, hexagonal fuel assemblies, and wire-wrapped geometries. Verification and validation are performed against two complementary datasets within the Organisation for Economic Co-operation and Development Nuclear Energy Agency Liquid Metal Cooled Fast Reactor Thermal/Hydraulic (LMFR T/H) and Sodium Fast Reactor–Uncertainty Analysis in Modeling (SFR-UAM) benchmarks: isothermal pressure drop measurements from the Texas A&M University (TAMU) 61-pin wire-wrapped bundle, and various tests from the SEFOR facility (SFR full core). CTF-SFR reproduces the TAMU frictional pressure losses across the full Reynolds number range, confirming the suitability of the implemented wire-wrap correlations. In SEFOR single-assembly cases, the code matches 61 heat balance points and shows good agreement with reference results for average and peak sodium temperatures. Additional code-to-code comparisons with the TrioMC code highlight discrepancies in axial temperature profiles, revealing the impact of axial conduction, currently not modeled in CTF. Sensitivity studies highlight that simplified mixing and radial conduction models can overpredict corner channel peak temperatures in bare rod bundles. Overall, the results demonstrate that CTF-SFR provides reliable predictions for wire-wrapped SFR assemblies while identifying modeling assumptions requiring refinement for nonstandard geometries.