Conference Agenda

This study presents a computational fluid dynamics (CFD) methodology for simulating boiling at bubble scale without prescribed flowrate in the context of the quenching fabrication process. The approach is based on the Volume Of Fluid (VOF) method and the incorporation of phases sources allowing the water/steam phase change. This method is based on the use of an adaptive mesh allowing a very high refinement along the water/steam interface during the development of the bubble. The primary objective was to validate this methodology by comparing CFD results with experimental data in pool boiling cases. The quantity of interest is the wall temperature at different heat fluxes and for different surface topologies (roughness is modeled explicitly). Analysis shows promising agreement between CFD results and measurements regarding the wall temperature for low heat fluxes. This work demonstrates the model’s ability to create and develop multiple interacting bubbles as well as predict relevant wall temperatures for low heat fluxes.

This work represents a significant advancement towards developing a methodology to numerically assess the heat removal capability of a given surface as a function of its topology.

The strengthening of pool boiling heat transfer capacity is of great significance. As an active enhancement technology, ultrasound can effectively enhance the boiling process by influencing the growth and detachment of bubbles. With low cost and simple operation, it has a broad application potential while the mechanism still needs to be studied.

This study presents a multi-physical model which considers acoustics and fluid dynamics based on Multiphysics software. The volume force term is added to describe the nonlinear effects caused by the sound field; and the level-set method is used to track the phase interface that brings the saturated boiling bubble behavior under the influence of ultrasound. As a commonly used numerical value in the industry, the ultrasonic frequencies are set to 20,28 and 40 kHz. Numerical simulation has found that the acoustic streaming caused by ultrasound can cause the enhancement of fluid flow which generate shear forces on the bubbles. The acoustic streaming also make perturbations on the surface, which can accelerate bubble detachment and further enhancing the surface heat transfer capacity. As a result, the heat transfer efficiency has a considerable increase. Increased frequency and ultrasonic power can effectively enhance the acoustic streaming then act on the heat transfer process. Meanwhile, experimental research has been carried out to verify the results of numerical simulations. The derived conclusions could be useful for the application of ultrasonic treatment on boiling heat transfer.

Nucleate pool boiling is a useful heat transfer technique present in many engineering applications such as space and aircraft industries, thermal design of electronic components, refrigerants, and nuclear reactors. Despite its wide use, the physical mechanisms linking heat transfer to bubble dynamics and turbulence remain largely unexplored. Depending on the subcooling temperature, which is the difference between the wall temperature and the liquid saturation temperature, bubbles may either depart, merge rapidly, coalesce slowly or shrink. These complex dynamics affect heat transfer and lead to intricate vortex patterns in the flow, influencing velocity and temperature statistics.

The present work uses Direct Numerical Simulations with an in-house solver for incompressible multiphase flow to study the effects of subcooling on boiling behavior. The level set method captures the interface between liquid and vapor phases, while a computer vision algorithm was developed to track bubble properties. The findings show that subcooling temperature significantly impacts heat flux and bubble behavior, which in turn alters the flow's vorticity structures, producing ring vortices or irregular patterns. Statistical analysis is provided to better understand these complex interactions, shedding light on the relationship between bubble dynamics and heat transfer in boiling flows.

The authors are grateful for the financial support by the National Aeronautics and Space Administration (NASA) Grant number: 80NSSC21K0470 monitored by Dr. David F. Chao. We thank NASA Ames Research Center and NASA Advanced Supercomputing Division (NAS) for their generous allocation on Pleiades to perform three dimensional CFD simulations.

Predicting maximum sheath temperature accurately is important from the safety perspective (to demonstrate fuel and fuel channel integrity). The geometry and dryout mechanisms for CANDU fuel channels differ significantly from the light water fuel bundle assemblies. A new Eulerian multiphase computational fluid dynamics wall boiling model is being developed to model liquid film thickness-induced dryout in CANDU fuel channels. The operating conditions in CANDU fuel bundles are especially challenging because they span over a range of two-phase flow regimes. The proposed liquid film thickness-induced dryout model, which leverages advanced boiling closures for water at high pressures, was assessed in an earlier study using single-element CHF tests performed at Stern Laboratories, and encouraging results were obtained. This paper documents the model development for CANDU fuel bundles, its implementation in the STAR‑CCM+ software, and a qualitative assessment of the predicted dryout power using data from heated tests on the modified 37-element CANDU fuel bundle configuration. The approach adopted in the analyses is anticipated to yield advanced predictive capabilities that can be leveraged to improve traditional reactor safety analyses.

High resolution computational fluid dynamics (CFD) studies of turbulent flow boiling are challenging due to the simultaneous requirements that turbulence and bubbles are resolved adequately. These interface-capturing simulations can support the understanding of the mesoscale mechanisms of the heat removal process. Detailed analysis of such simulations has been demonstrated to uncover important physics of the heat transfer process and support macroscopic heat transfer model development. One area in which CFD can support the understanding of complex fluid mechanics is high heat flux boiling conditions. In pursuit of this goal, a benchmark problem has been developed to mimic the conditions of a previously completed high resolution experiment. The topic covered in this paper is a scoping study designed to assess the performance of the selected approach applied to this problem. The domain in this case is accurate to the experiment, with realistic Reynolds number, inlet subcooling, and heat flux. However, for simplicity and computational cost concerns, only a limited number of nucleation sites are considered. The mesh design is covered in detail to illustrate the consideration of both liquid turbulence and nucleate boiling. The bubble dynamics and associated wall temperature are compared against experimental values to assess the suitability of the approach for future multiple nucleation site studies and to which surface conditions the approach is generalizable. Based on the observed results, CHT & microlayer evaporation models are expected to introduce additional physics to improve predictions. Therefore implementation of these models is evaluated for future work.

This research aims at simulating and contributing to understand the generic physics which could occur in nuclear spent fuel pools during loss of cooling accidents. Based on the limitations inherent to the Direct Numerical Simulation approach in terms of Rayleigh number and geometry, we also intend to provide relevant reference results for RANS simulations.
This paper presents in a stepwise layout the different necessary components to simulate the natural convection of water in a cuboid domain combining vapor bubbles dynamic and evaporation through the free surface resulting into a descent of the water level.

Heat transfer due to evaporation is accounted for using the model presented by W. H. Hay et al. (2021), while the related mass transfer relies on a new remeshing procedure which attributes the descent proportionally to all cells by re-meshing the grid at each time-step. This allows to avoid any field changes at the boundaries whilst distributing the error along the height. This remeshing procedure, although apparently simple, involves a change in the temporal discretization of the governing equations. On the other hand, an Eulerian-Lagrangian approach is implemented and allows to compute the motion and growth/shrinkage of vapor bubbles while the effect of the bubbles on the fluid is accounted via momentum and energy exchanges between the two phases in a two-way coupling.

First, we detail the different models implemented. We then present a validation and verification procedure against analytical, experimental and numerical results. Finally, we present and discuss results of both models separately and combined.