Bubble Departure Research Articles

BubbleID: A deep learning framework for bubble interface dynamics analysis

This paper presents BubbleID, a sophisticated deep learning architecture designed to comprehensively identify both static and dynamic attributes of bubbles within sequences of boiling images. By amalgamating segmentation powered by Mask R-CNN with SORT-based tracking techniques, the framework is capable of analyzing each bubble's location, dimensions, interface shape, and velocity over its lifetime and capturing dynamic events such as bubble departure. BubbleID is trained and tested on boiling images across diverse heater surfaces and operational settings. This paper also offers a comparative analysis of bubble interface dynamics prior to and post-critical heat flux conditions.

A visualization study of a confined vapor chamber with conical microstructure applied in data centres

Abstract With the development of the informatization technology, the scale of the data centre is expanding rapidly. The energy consumptions of the electronic equipment in the data centre are rising regularly, which lead to the thermal management becoming an argent issue to be settled. To realize the sustainable development of green data centre, the passive two-phase vapor chamber (VC) has turned into the focus of electronic cooling research. Constructing nucleation induced structures on the evaporation surface is an effective method to improve the performance of the vapor chamber. To address the problem of achieving enhanced boiling, a novel conical microstructure was designed in the vapor chamber with 50 mm height. The conical structure of 1 mm axial height was fabricated on the evaporation surface by computer numerical control (CNC) machining technology. A visualization experimental system was developed to investigate the effect of the conical microstructure on the two-phase behaviours, and the boiling heat transfer characteristics under different heating conditions (Q in = 35 W, 50 W, 65 W) in a confined vapor chamber. The high-speed camera was used to capture the bubble behaviours. Experimental results found that compared with the smooth surface, the integration of the conical structure increasing the number of bubble nucleation sites and the bubble departure frequency. The bubble growth period at stable heating stage is 162 ms shorter than initial heating stage on the evaporation surface with conical structure. The thermal resistance (R vc) of vapor chamber with conical structure is improved by 5.85% compared to the smooth one at Q in = 65 W, which indicate that the conical microstructure can enhance the boiling heat transfer performance. This study aims to provide a reference for the design of thermal management system for green data centre.

Towards understanding effects of ultrasonic waves on subcooled boiling using particle image velocimetry

In this study, effects of ultrasonic waves on heat transfer, bubble behavior and temporal variation of liquid flow near the heating surface in subcooled boiling was investigated with the help of high-speed particle image velocimetry method. Experimental results show that the enhancement effect of ultrasound on heat transfer and critical heat flux considerably depends on subcooling and boiling space height. In nucleate boiling regime, the ultrasound-induced acoustic streaming alters the upward liquid flow and changes the vortex structure associated with bubble growth, departure and rising. As heat flux increases, the bubble-induced convection accelerates and competes with the acoustic streaming near the heating surface, weakening the ultrasonic effect on heat transfer. In transition boiling regime, effects of ultrasonic waves on heat transfer of microbubble emission boiling depends on the competition between acoustic streaming and oscillating flow resulting from vapor film oscillation, with enhancement occurring when acoustic streaming dominates the liquid flow near the heating surface. Based on the visualization results, a scaling law is developed to determine the potential effects of ultrasound on critical heat flux.

Bubble departure and sliding in high-pressure flow boiling of water

Bubble growth, departure and sliding in low-pressure flow boiling has received considerable attention in the past. However, most applications of boiling heat transfer rely on high-pressure flow boiling, for which very little is known, as experimental data are scarce and very difficult to obtain. In this work, we conduct an experiment using high-resolution optical techniques. By combining backlit shadowgraphy and phase-detection imaging, we track bubble shape and physical footprint with high spatial ( $6\,\mathrm {\mu }{\rm m}$ ) and temporal ( $33\,\mathrm {\mu }{\rm s}$ ) resolutions, as well as bubble size and position as bubbles nucleate and slide on top of the heated surface. We show that at pressures above 1 MPa bubbles retain a spherical shape throughout the growth and sliding process. We analytically derive non-dimensional numbers to correlate bubble velocity and liquid velocity throughout the turbulent boundary layer and predict the sliding of bubbles on the surface, solely from physical properties and the bubble growth rate. We also show that these non-dimensional solutions can be leveraged to formulate elementary criteria that predict the effect of pressure and flow rate on bubble departure diameter and growth time.

Prediction of bubble departing diameter in pool boiling of mixtures by ANN using modified ReLU

In this research, bubble departure diameter in pool boiling have been measured in aqueous amine and ethylene glycol solutions for various concentrations. The experimental data have been compared with major existing predictive correlations. It is shown that the effect and identity of the independent variables on bubble diameter proposed in the previous studies are inconsistent. The predictions of different correlations have on average a deviation of about 40% from the experimental data. This is mainly due to the complicated interactions between bubbles on the heterogeneous boiling medium, which provides a complex condition. This complexity limits any mathematical modelling of the forces acting on the developing bubbles. Particularly in liquid solutions, where mass transfer by back diffusion through micro-sub-layers adds further complexity. In this work, the classical artificial neural network, ANN, with rectified linear unit, ReLU, activating function, AF, has been modified. This modification is based on adding a numerical matrix to each layer to adjust the slope of AF for each neuron independently. The addition of this parameter, together with the adjustment of the bias matrix, makes the activation function more flexible than the classical ReLU. To find the tuning parameters, a genetic algorithm was implemented instead of the back-propagation technique. It is shown that the predictions of the trained ANN with modified ReLU AF agree within an absolute average error of 10%, which is equal to the total uncertainty of the measurements. Prediction of bubble departing diameter in boiling phenomena is a key parameter for accurate design, operation and optimisation in many industrial systems.

Roles of Wettability and Wickability on Enhanced Hydrogen Evolution Reactions.

Bubble dynamics significantly impact mass transfer and energy conversion in electrochemical gas evolution reactions. Micro-/nanostructured surfaces with extreme wettability have been employed as gas-evolving electrodes to promote bubble departure and decrease the bubble-induced overpotential. However, effects of the electrodes' wickability on the electrochemical reaction performances remain elusive. In this work, hydrogen evolution reaction (HER) performances are experimentally investigated using micropillar array electrodes with varying interpillar spacings, and effects of the electrodes' wettability, wickability as well as bubble adhesion are discussed. A deep learning-based object detection model was used to obtain bubble counts and bubble departure size distributions. We show that microstructures on the electrode have little effect on the total bubble counts and bubble size distribution characteristics at low current densities. At high current densities, however, micropillar array electrodes have much higher total bubble counts and smaller bubble departure sizes compared with the flat electrode. We also demonstrate that surface wettability is a critical factor influencing HER performances under low current densities, where bubbles exist in an isolated regime. Under high current densities, where bubbles are in an interacting regime, the wickability of the micropillar array electrodes emerges as a determining factor. This work elucidates the roles of surface wettability and wickability on enhancing electrochemical performances, providing guidelines for the optimal design of micro-/nanostructured electrodes in various gas evolution reactions.

Phase Transition Heat Transfer Enhancement of a Graphene-Coated Microporous Copper Surface Using Two-Step Electrodeposition Method

Abstract Owing to their exceptionally high thermal conductivity, there is a growing demand for graphene nanoparticles in phase transition heat transfer applications. This research delves into the exploration of various critical phenomena within the realm of surface science, specifically focusing on interactions at solid-liquid and liquid-liquid interfaces. In this work, graphene nanoparticles at varying concentrations are subject to electrochemical deposition on a microporous copper substrate to form graphene coated over microporous copper (GCOMC). The study encompasses a comprehensive analysis of surface characteristics, such as porosity, roughness, and wettability. Furthermore, the study involves the calculation of two key heat transfer metrics, the critical heat flux (CHF) and the boiling heat transfer coefficient (BHTC), through the execution of pool boiling experiments. The findings of this research underscore the remarkable superiority of GCOMC surfaces over their uncoated copper counterparts in terms of boiling performance. Particularly, the GCOMC surface showcases an impressive 87.5% enhancement in CHF and a 233% increase in BHTC compared to the bare copper surface. Furthermore, this investigation delves into a detailed quantitative analysis of bubble behavior, encompassing parameters such as bubble departure diameter, bubble departure frequency, and nucleation site density, employing high-speed camera techniques to comprehensively understand the underlying processes.

Influence of solution pH on the dynamics of oxygen bubbles on the surface of TiO2-NTAs electrodes

The rise of concentration resistance caused by bubble growth is a key factor affecting the photoelectrochemical water splitting efficiency. In this research, we employed an electrochemical workstation, a synchronized measuring system, and a CCD camera to investigate the in-situ evolution of oxygen bubbles on the surfaces of TiO2-NTAs and C/TiO2-NTAs electrodes. The connection among the critical nucleation, growth and departure behavior of a single bubble on the photoanode surface and photocurrent curves at different solution pH (1–13) has been investigated. For TiO2-NTAs, it was found that strong alkalinity significantly reduces the voltage required for critical bubble nucleation. The photocurrent rises approximately 1.5 times when the control voltage is 0.10 V (vs. Ag/AgCl) by pH increasing from 1 to 13 and while the bubble evolution cycles decline by approximately 3.2 times. The photocurrent rises by 10.4 times compared with pH = 9 and by 7.3 times compared with pH = 13 when the control voltage is 0.71 V (vs. RHE), respectively. We also found that the oxygen bubble evolution of C/TiO2-NTAs is similar to TiO2-NTAs. When pH = 3–13, the average photocurrent of C/TiO2-NTAs electrode is higher than that of TiO2-NTAs, and the average bubble evolution cycles are lower than those of TiO2-NTAs at the same pH. According to the results, carbon films can rise the absorption capacity of TiO2-NTAs and significantly improve the effectiveness of photogenerated holes and electrons departure. Meanwhile, based on the oxygen bubble force balance model on the photoanode surface, the concentration Marangoni force rises with the increase of pH, which leads to the growth of bubble departure diameter. The outcomes demonstrate that oxygen bubbles on the photoanode surface can be successfully eliminated in strongly alkaline settings.

Bubble dynamics in high-pressure flow boiling conditions

High resolution measurements of key parameters relevant to bubble departure processes in high-pressure water systems has been a major challenge over the past several decades due to the small size and fast motion of vapor bubbles, making conventional optical measurements unreliable. Addressing this difficulty is critical, as accurate bubble departure data and computationally inexpensive models are needed for future two-phase heat transfer models. To close this gap, we use high-speed imaging to track bubble size and velocity in high-pressure flow boiling conditions (10 – 40 bar). It is observed that bubbles are almost always perfectly spherical in shape and slide along the boiling surface immediately after nucleating. To predict sliding, we conduct a force balance on the bubble and derive an equation of motion in the direction of the bulk flow. Interestingly, we find that at high pressure conditions, the equation of motion can be significantly simplified, resulting in an analytical expression for the bubble motion, growth time and departure diameter, allowing us to develop simple and computationally efficient equations. The applicability of the model and departure criterion across a broader range of high-pressure conditions is discussed, and recommendations for further characterization of bubble departure processes are made.

Boiling of binary mixtures on superhydrophobic surfaces with artificial cavities

This study experimentally investigates bubble behaviour from an isolated artificial cavity on a superhydrophobic substrate, the working fluid is a binary mixture of Novec649 and Novec7000 with mole fractions between 0.05 – 99.5 as well as their pure counterparts. The results show an earlier onset of bubble departure as well as deceasing trends in bubble departure diameter in the presence of the binary mixture compared to pure cases. The bubble departure diameter is at its minimum and fastest growth rate at a mole fraction of approximately 0.50. Theoretical models from literature were tested and satisfactory agreement with different models were achieved depending on the concentration.

Direct Numerical Simulation of single bubble dynamics in nucleate pool boiling with micro-region modeling and thermal coupling to a solid wall

In this study, we conduct two-dimensional, axisymmetric simulations to investigate the growth and departure of a single bubble, with a particular focus on the conjugate heat transfer between the fluid and the heating wall. A multiscale modeling approach is employed to account for nanoscale effects near the liquid-vapor-solid triple contact line (CL). At the macro (bubble size) scale, the interface dynamics is captured with the front-tracking method by using the open-source code TRUST/TrioCFD. The macro scale algorithm is coupled with a sub-grid micro-region model. It is driven by the wall superheating at the CL as input, and predicts the macroscopic apparent contact angle and heat fluxes. To validate our modeling approach, we first conduct quantitative comparisons using the data on the bubble growth experiment RUBI and its simulations. Next, an investigation of the case of water under atmospheric pressure and gravity is performed. Notably, we observe a significant spatial variation in temperature near the nucleation site during the expansion and contraction of the bubble base. This variation greatly affects the thermal boundary layers in the fluid and solid domains and the bubble growth, demonstrating the need to resolve the conjugate heat transfer in the considered cases.

Effect of gas flow rate and nozzle diameter on bubble size and shape distributions in bubble column

The influence of gas flow rates and nozzle diameters on bubble size and aspect ratio distributions was studied using high-speed photography with an image processing algorithm. Results reveal bimodal probability density distributions (PDD) of bubble diameter under all nozzle diameters, with one peak near 1.5–2 mm and the another in the larger bubble range. Increasing gas flow rates from 0.1 to 0.2 L/min leads to a higher probability density of large bubbles, indicating prevalent bubble coalescence. As the gas flow rate rises to 1.2 L/min, the peak shifts to smaller bubble diameters, and the bubble breakage becomes dominant. For 1–2 mm bubbles, shape is less influenced by gas flow rates, while 3–9 mm bubbles exhibit aspect ratio PDD peaks at an aspect ratio (E) of 0.5 across all gas flow rates. The Iguchi and Chihara model can better predict the variation of bubble departure diameter with increasing gas flow rates.

A hybrid atomistic-continuum framework for multiscale simulations of boiling

Boiling is a multiscale physics process where the nucleation of vapour bubbles occurs due to molecular-scale interactions between the fluid and a heated wall, but it also depends on the larger-scale hydrodynamics and thermal boundary layers determined by the outer system boundary conditions. Modelling boiling from the nanometre up to the millimetre scales at which bubble departure occurs is not possible via state-of-the-art simulation methods: Molecular Dynamics (MD) simulations can capture nucleation from first principles but are limited to nanometre scales due to their computational cost, whereas computational fluid dynamics (CFD) simulations based on the continuum Navier-Stokes equations cannot capture nucleation. Here, we present a novel multiscale simulation method which merges MD and CFD descriptions into a single modelling framework, where MD resolves the near-wall region where molecular interactions are important, and a CFD solver resolves the bulk flow. We model the progressive heating of a Lennard-Jones fluid via contact with a solid wall until a vapour bubble nucleates in the MD region of the domain and grows by entering in the CFD domain. Our results show that an incompressible CFD flow model based on the Volume Of Fluid method with interphase mass transfer calculated via the Hertz-Knudsen-Schrage equation is sufficient to obtain seamless coupling of phase fraction, velocity and temperature fields, with the hybrid MD-CFD framework yielding bubble dynamics closely matching those of MD alone.

Modeling and study of pressure-dependent boiling crisis in membrane-based heat sink

A theoretical model is developed to study the pressure-dependent boiling crisis of membrane-based heat sink. It has been reported that the critical heat flux of this heat sink can be increased to 1.8 kW/cm2 by using a piece of hydrophobic membrane to promote bubble departure. In the modeling, two boiling crisis mechanisms are applied. The free energy and the forces acting on the bubble are considered in the calculations of the bubble's geometric dimensions, the vapor venting rate, and the pressure drop for driving the wicking flow. Wetting and the evaporating rate beneath the bubble are computed based on the reported coupled wicking and evaporation model. The modeling explains why there exist two regions in the profile of the critical heat flux against the pressure drop. The time for rewetting the dry spot is obtained with the modeling to establish the correlation between the critical heat flux and the pressure drop at higher heat flux. When the boiling surface has a relatively high area ratio, the modeling error is 3 % with the suggested membrane permeability and critical temperature rise of the dry spot.

Boiling heat transfer enhancement by a pair of elastic plates

Boiling heat transfer enhancement is essential for energy utilization. Traditional passive boiling enhancement methods aiming at bubble departure promotion usually depend on the modulation of surface tension force by surface modification. However, the limited driven force which benefits bubble departure cannot help the further improvement of heat transfer performance facing the increased heat dissipation amounts. In this paper, a novel boiling enhancement method using a pair of elastic plates (i.e., flexible parallel plates made of stainless steel) which promotes bubble departure by its swing effect for boiling on Pt wire in confined space is proposed and investigated. Compared to boiling with a pair of rigid plates (the thickness of 0.5 mm and the length of 12 mm), the heat transfer coefficient (HTC) is improved by 55.8% of those with elastic plates (the thickness of 0.01 mm and the length of 12 mm).Periodic expansion and recovery process of plates coupling bubble dynamics is observed and studied. Experimental results indicate that elastic plates enhance boiling heat transfer performance significantly by smaller bubble departure diameter and lower wall superheat at the same wall heat flux. A correlation for bubble departure diameter prediction in elastic channel is proposed which matches well with experimental results. It is found that the stronger pumping effect induced by higher swing rate of plates enhance HTC significantly. The mechanism where expansion process of plates promotes bubble departure while its subsequent recovery process benefits the supply of the cold liquid is verified by simulation. This study proposes a new approach for boiling heat transfer enhancement by inducing the energy exchange between bubbles and elastic plates, and provides theoretical and technical support for boiling study based on flexible materials.

Monodispersed Bubble Generation Using Hydrophobic Orifices: The Extended Tate's Law.

The use of submerged orifices for bubble generation is ubiquitous in industries with wettability known to influence the bubble departure diameter. In this study, we investigated bubble generation and departure from the orifices (0.3-2 mm) drilled on hydrophobic perfluoroalkoxy (PFA) tubes in water. By varying the gas inflow rate (33 to 200 mL/min), we found that the Sauter mean diameter closely matched those generated by hydrophilic quartz orifices. However, monodispersed bubbles were formed on the PFA tube compared to those on quartz with much wider size distributions. By examining the dynamic bubbling process, we confirmed its agreement with Tate's law, which was originally developed for quasi-steady conditions and emphasizes a balance between capillary and buoyancy forces. However, it should be noted that dynamic conditions lead to an increase in bubble volume compared to the quasi-steady condition despite following the same principle, which is explained by the continuous gas inflow when the bubble departs from the orifice at a necking stage. The above understandings enable generation of monodispersed bubbles under dynamic conditions, benefiting industries requiring precise control on bubble size, such as the bubble assisted wet etching and cleaning processes in semiconductor fabrication.

Pool boiling performance for GaldenR HT 55 subject to smooth, pin fin, and sintered pin fin heat sink

The dielectric fluid of GaldenR HT 55 (Perfluoropolyether) is regarded as a potential candidate for two-phase immersion cooling since its boiling point at 1 atm is 55 °C which is quite suitable for electronics cooling. The fluid is of special interest since the main vender of dielectric fluid will no longer produce FGAS after 2024. However, very rare nucleate boiling performance data for HT55 were reported in the literature. The objective of this study aims to provide nucleate boiling performance data for various heat sinks applicable for nucleate boiling. The test samples include a smooth reference surface (#1), 2 pin fins (the pin diameters are 1 mm (#2) and 2 mm (#3), and 3 pin fin with various sintered coatings (powder size: 150–200 μm (#4), 200–250 μm (#5), and 250–350 μm (#6)). Heat flux spans from 80 to 600 kW/m2. HTC for small pin fin (#2) is marginally higher than that of smooth surface, and large pin fin (#3) shows about 40 % enhancement compared to smooth surface at 140 kW/m2. HTCs for sintered pin fins are much higher than those of pin fin or smooth surface with approximately 90–95 % enhancement relative to the smooth surface at low heat flux. The visual observations indicate that sample #6 yields the largest bubble departure diameter and departure frequency. Yet bubble generations are especially pronounced on top of the pin fin. The CHF of the tested surfaces are following the sequence: #5 > #6 > #4 > #3 > #2 > #1. The sample #5 contains a CHF over 600 kW/m2 which is about four times higher than that of the smooth surface.

Bubble Interaction and Heat Transfer Characteristics of Microchannel Flow Boiling With Single and Multiple Cavities

Abstract Flow boiling in microchannels can effectively address the challenges of high power density heat dissipation in electronic devices. However, the intricate bubble dynamics during the two-phase flow in microchannel necessitates understanding the characteristics of complex bubble hydrodynamics. In this study, we perform 2D numerical simulations of flow boiling using the Cahn-Hilliard phase-field method for a 200-µm width microchannel with single and multiple cavities in COMSOL Multiphysics (V5.3). The numerical model successfully captures bubble dynamics, encompassing vapor embryo generation, bubble growth, departure, coalescence, sliding, and stable vapor plug formation. The heat transfer mechanism inside the microchannel is dominated by bubble nucleation and thin-film evaporation. Elevated wall superheats in a single nucleation cavity, and increased mass flux facilitates higher bubble departure frequency and heat transfer performance. Temporal pressure fluctuations are observed inside microchannels in multiple cavities due to bubble coalescence, departure, and subsequent nucleation. Increasing the nucleating cavities from 2 to 5 within the microchannel while maintaining consistent cavity spacing of 100 µm has resulted in nearly 32% enhancement in heat transfer performance. This study offers valuable findings that can help improve the thermal management of electronic devices.

Toward a complete understanding of quasi-static bubble growth and departure

The distortion by gravity of a quasi-static bubble attached on an upward facing surface in a quiescent liquid is investigated. The contact angle evolution during the growth of such a bubble is studied, and the consequences on the motion of the contact line between the solid and the interface are discussed. From the initial case of a bubble attached to the rim of a nucleation site, the contact line can move inside the cavity for a highly wetting fluid, causing premature departure. For a higher wettability, the contact can either remain attached to the rim of the cavity or spread out of the cavity, depending on the cavity size and geometry. For the latter case, the bubble growth is investigated taking into account a contact angle hysteresis. Finally, a comprehensive map detailing various geometrical characteristics of bubbles is presented, ranging over several orders of magnitude of Bond numbers and normalized volumes. The map aims at being used as a tool for investigating bubble growth in a similar situation.

Turbulent flow boiling simulation based on pseudopotential lattice boltzmann method with developed wall boundary treatments and unit conversion

The lattice Boltzmann method (LBM) is widely studied for complex flow simulations, including liquid–vapor phase change phenomena. Although previous studies extended the LBM simulation capability from single-phase to two-phase boiling simulations, most of them are restricted to no-flow (pool boiling) or low-flow-rate (laminar flow) boiling conditions owing to the stability problem in collision at low viscosity, representing high Reynolds numbers. However, a high Reynolds number flow is essential for investigating the effects of turbulence on bubble dynamics. Recently, the central moment-based collision method is used to enable high Reynolds numbers for bulk flow; however, extending the method for multiphase-flow simulation with high Reynolds numbers is limited owing to the boundary treatment near the wall. In this study, we augment the previous boundary treatment in LBM simulation with a forcing term, which increases the stability of the simulation even under two-phase conditions. With the improved boundary treatment, the flow boiling phenomena at high Reynolds numbers (up to Re = 107,000) are successfully reproduced. Moreover, in this study, unit conversion is reconsidered with the relationship between conversion factors because the poor outcomes of bubble dynamics are attributed to misinterpreted unit conversion. The new unit conversion method satisfies the necessary conditions for mapping the physical bubble behavior to lattice frame, reproducing bubble dynamics effectively following the Reynolds number, with reasonable spatial and temporal resolution. Consequently, the flow boiling simulation can accurately predict the bubble departure diameter from the real experimental results and the heat transfer trend as a function of the Reynolds number.