Conference Agenda

Natrium® reactor vessel design utilizes the multi-dimensional computational fluid dynamics (CFD) to investigate the various flow phenomena and heat transfer mechanisms to predict the temperature distribution and flow velocity of the sodium. CFD is one of the many tools utilized during the design phase to inform various engineering teams including but not limited to transient and safety analysis, structural design and analysis etc. As part of the validation and investigation of the prediction capability of the CFD, multiple legacy data is being investigated. MONJU reactor trip benchmark by IAEA-CRP is one of them as it is investigated within the present paper. It investigates specifically thermal stratification behavior in the upper plenum of sodium cooled reactor. Previous studies investigated uncertainties on the flow hole geometry and turbulence modeling. Present paper investigates a more recent second-generation URANS closure (STRUCT−ε) model. The approach aims at advancing the robustness of hybrid turbulence models by relying on the efficiency of an extensively validated anisotropic k−ε method, while locally including the optimum resolution of complex unsteady flow structures.

In Sodium-cooled Fast Reactors (SFRs), the sensitivity of neutron cross-sections is essential for understanding the complex relationship between core neutronics, thermal hydraulics, and reactor safety. SFRs utilize fast neutrons and liquid sodium as a coolant, which introduces specific challenges in heat transfer and neutron interaction. Sensitivity analysis of neutron cross-sections in these systems quantifies the effects of uncertainties in nuclear data on parameters like reactivity, neutron flux distribution, and power peaking factors. These parameters significantly impact the core’s thermal hydraulic behavior.

The interplay between neutronics and thermal hydraulics in SFRs is crucial due to the fast neutron spectrum and sodium’s high thermal conductivity. Variations in neutron flux and cross-section values influence localized heat generation, while changes in coolant temperature and flow affect cross-sections through feedback mechanisms. Proper modeling of these interactions ensures effective heat removal from the core, preventing excessive fuel temperatures and avoiding material degradation or fuel failure, especially during transients and accident scenarios.

In safety analysis, sensitivity calculations are vital for predicting the reactor’s behavior under normal and off-normal conditions, including critical events like Loss of Flow (LOF) or sodium boiling accidents. These analyses assess how cross-section uncertainties affect thermal hydraulic margins, guiding the development of design strategies to ensure safe reactor shutdown and decay heat removal. Sensitivity analysis thus plays a key role in optimizing SFR performance and safety by offering insights into how nuclear data uncertainties impact overall system behavior, leading to more robust safety measures.

Liquid Metal Fast Reactors (LMFR) are a promising technology for expanding nuclear energy to reduce carbon emissions from the energy sector. The ultimate goal of the project is to provide validation data for the Clad Damage Propagation (CDAP) module of SAS4S/SASSYS-1, a reactor safety code system. The Experiment on Pin Failure for LMFRs (ExPL) project at Oregon State University (OSU) aims to provide data on the heat transfer impingement due to fission gas ejection from an initial cladding failure event. The experiments in liquid sodium will consist of a 19-pin assembly with electrically heated surrogate pins in a liquid sodium flow loop. The test section will be instrumented with High-Definition Fiber Optic Temperature Sensors, placed inside capillaries in place of a solid wire wrap consistent with the simulated fuel assembly geometry. The temperature within the capillaries, determined by the heat transfer through the capillary, will necessarily be different from the heater rod surface. This paper details experiments in water and air that investigate the presence and magnitude of both spatial and temporal discrepancies and potential modes for mitigating observed error. These alternative, low-risk fluids, were utilized to establish a baseline understanding of this application of fiber optic sensors and to inform future experiments in liquid sodium.

The primary cooling loop in sodium-cooled fast nuclear reactors is achieved using a centrifugal pump immersed in liquid sodium. Under certain conditions, fluid vortices can be generated and develop into bubbles of the cover gas present on the sodium coolant free surface. This phenomenon, known as Gas Entrainment (GE), may have an impact on the reactor vessel design and on the core reactivity. The GE is difficult to predict and parameters influencing its occurrence are still poorly known. Simulations using Computational Fluid Dynamics (CFD) could help to better understand such phenomenon and identify the parameters that govern its occurrence. In this work, free surface flow simulations based on the Large-Eddy Simulations (LES) for flow hydrodynamics prediction combined with the Front-tracking method for interface modeling, were performed. It figured out that predictions of the interface dynamics is greatly influenced by the the element mesh size, on which depends the accuracy of the flow hydrodynamics prediction. Finer meshes allowed to better capture the instantaneous small eddies and local velocities, which enhanced the generation of the vortices. The coarse simulation predicted less intense pressure variations near the vortices, and smoother pressure distribution throughout the domain. On the contrary, the fine simulation exhibited more distinct clusters of high positive and negative vorticity, associated with more distinct low-pressure cores. This enhanced the development of large vortices moving along the free surface.

The Liquid Metal Cooled Fast Reactors (LMFRs) are one of the technologies being considered by the Generation IV International Forum (GIF). The unique geometric design of the rod bundle channels causes an uneven circumferential temperature distribution on fuel rod surfaces, further intensified by liquid metal's high thermal conductivity. This temperature variation may induce thermal fatigue damage to the cladding, threatening reactor safety. It is identified that the subchannel heat transfer characteristics in liquid metal reactors are predominantly influenced by the Peclet number (Pe) and the pitch-to-diameter ratio (P/D). Usually, Reynolds-Averaged Navier-Stokes (RANS) computational fluid dynamics (CFD) with turbulent models are employed to study heat transfer in liquid metal rod bundles, which fail to capture the anisotropic thermal transfer in liquid metal rod bundles, ignoring circumferential temperature differences. Conversely, Large Eddy Simulation (LES) offers detailed insights into flow and heat transfer phenomena. Accordingly, this study conducts a numerical investigation on hexagonally arranged fuel bundles using LES to explore the circumferential temperature distribution under varying Pe and P/D conditions. The LES results show that Pe and P/D can affect circumferential temperature non-uniformity. Moreover, considering the experimental costs, smaller hexagonally arranged fuel bundles with a non-prototypical cold wall are selected. Due to the cold wall effect, the central fuel rods exhibit smaller circumferential temperature differences compared to the outer rods. These findings highlight the critical impact of Pe and P/D on the circumferential temperature distribution of fuel rods, providing valuable theoretical guidance for the optimized design of liquid metal-cooled fast reactors.