Effects of bubble coalescence on pool boiling heat transfer and critical heat flux – A parametric study based on artificial cavity geometry and surface wettability

https://doi.org/10.1016/j.ijheatmasstransfer.2019.118952Get rights and content

Highlights

  • Pool boiling experiments on surfaces with artificial cavities were performed.

  • The effect of surface morphology on bubble coalescence was investigated.

  • Circular cavities with various diameters and pitch sizes were fabricated on silicon surfaces.

  • Different bubble coalescence types were observed on tested samples.

  • There exists a critical pitch size/diameter ratio for the horizontal bubble coalescence.

Abstract

Controlling the onset of boiling is highly desirable for enhancing boiling heat transfer. In this study, a systematic set of pool boiling experiments on surfaces with artificial cavities were performed to investigate the effect of surface morphology on bubble coalescence and resultant boiling heat transfer performance. Circular cavities with various diameters and pitch sizes were fabricated on silicon surfaces. The effect of surface wettability on the performance of the structured surfaces were examined with the use of 50 nm thick Teflon film. Using a high speed camera to examine the bubble dynamics, the results reveal that there exists a critical hole pitch size/hole diameter ratio (P/D = 10), below which horizontal bubble coalescence occurs at the lower wall superheats. Furthermore, the visual results indicated that surface wettability alters the critical heat flux (CHF) mechanism. In contrast to the hydrophobic surfaces, hydrodynamic instability is the main reason for CHF occurrence on the hydrophilic surfaces. The results indicate that although increasing the hole diameter enhances the CHF for all the fabricated samples, the effect of pitch size depends on the surface wettability with the CHF increasing with pitch size on the hydrophobic surface and decreasing with pitch size on the hydrophilic surface. A maximum heat transfer coefficient enhancement of 100% was achieved on the hydrophilic structured surface relative to the hydrophobic structured surface. The maximum CHF increase was 100% on the hydrophilic surface and 48% on the hydrophobic surface.

Introduction

Boiling is considered as one of the most effective heat transfer mechanisms due to the utilization of latent heat of vaporization and incepts from surface impurities (i.e. cavities), which act as potential nucleation sites for bubble formation. As with many other phase transitions, a critical nucleus size exists. Below this size, which depends on the amount of area the vapor-medium interface possesses, the vapor phase becomes unstable. At the low wall superheats, the isolated bubbles start to grow and depart from the superheated surface. During this stage, the frequency of bubble departure is one of the governing parameters, which significantly influences boiling heat transfer (BHT). As the wall temperature increases, the generated bubbles from the neighboring nucleation sites start to coalescence and form larger bubbles. The driving mechanism of the bubble coalescence is simply the tendency of the system to minimize the liquid-vapor interface. As the number and size of the generated bubbles increase, liquid rewetting and surface replenishment become difficult. At the high wall temperatures, where the coalescence of bubbles is frequent, generated bubbles form large vapor blankets on the superheated surface so that vapor and liquid interfacial velocity and wavelength dictate the upper limit of boiling heat transfer and occurrence of critical heat flux (CHF).

Given that boiling heat transfer and CHF depend on the nucleation and growth kinetics of the vapor phase, they are significantly affected by surface characteristics [1], [2], working fluid properties including the thermal conductivity and the latent heat of vaporization [3], [4], as well as liquid-solid interfacial properties such as wettability. There exist a limited number of design parameters to control boiling process. The modification of the surface, on which boiling occurs, is considered as one of the promising methods to enhance the efficiency of heat transfer systems [5], [6], [7], [8], [9], [10], [11]. Although the enhancing effects of surface modifications on the onset of boiling heat transfer and critical heat flux have been extensively considered, some results such as effect on nucleation site density are still far from being intuitive, implying a lack of systematic understanding on interfacial phenomena on structured surfaces. In one of such attempts, Dong et al. [12] investigated nano-cavities and hydrophilic micro-cavities with depths and heights of 5, 10, 20, and 50 µm. The authors reported that the number of nucleation sites per area increased with decreasing micro-cavity diameter at the low wall heat fluxes, while the nano-cavities only slightly increased the active nucleation sites compared to the smooth silicon surface. As the critical bubble size is likely beyond the nano-cavity size, it is understandable that the gain in reducing the surface energy of the bubble by attaching to such a nanocavity is surpassed by the remaining vapor/liquid interface energy, leaving no reason for the bubble to stabilize from an energy perspective.

Early studies demonstrated that there is a direct correlation between the elapsed time of bubble nucleation and the distance between the active sites [13]. Sultan and Judd [13] performed nucleate pool boiling heat transfer experiments to understand the relationship between the active site density and time elapse of bubble growth. The experiments were conducted at different values of heat flux and subcooling with different active nucleation site densities. It was reported that for the low-level heat flux test, which corresponded to the heat flux of 92.21 kW/m2 and three different subcooling values (0, 6.5, and 12 °C), the separating distance of active sites and time elapse between subsequent bubble growths was linearly correlated, which means an increase in the active site distance would lead to an increase in time elapse between subsequent bubble growths. Later studies revealed the effect of surface topology on boiling at different levels of heat flux. In this regard, Teodori et al. [14] performed a series of experiments and developed a modified Rohsenow correlation [15] to find heat transfer coefficient for pool boiling of different types of fluids on micro-structured surfaces. As a result of their study, they proposed a correlation capable of estimating the heat transfer coefficient for different combinations of fluid-surfaces within an error of ±20%. They added a geometric parameter, which takes the surface characteristics into account. In another study, Teodori and his co-workers [16] made particle image velocimetry (PIV) measurements with image post-processing to characterize the mechanism of pool pooling on micro-patterned surfaces having square shape cavities. Accordingly, large bubbles cover and isolate the boiling surface, thereby resulting in a reduction in the heat transfer coefficient. By analyzing the obtained results from cavities having edge to edge distances ranging from 200 µm to 2000 µm, they proposed an optimum distance for balancing the negative effect of the horizontally departed bubbles coalescence.

As Teodori remarked [14], [16], an increase in the number of microcavities on the surface provides more active nucleation sites. The liquid-solid contact area is thus increased due to surface roughness. However, the interaction mechanism of liquid-solid contact needs to be well considered since the departed bubble coalescence might cause horizontal and vertical vapor blanket formation because of large bubbles. This problem was recently highlighted by Moita et al. [17], who analyzed pool boiling heat transfer on micro-structured surfaces. The analysis consisted of two main parts: (i) the effect of liquid properties and (ii) the effect of micropatterned surfaces (cavities and pillars). Consequently, the results of the effect of liquid properties showed that the coalescence was influenced by the microstructured surface for the liquids having large surface tension and latent heat of evaporation. They indicated that the coalescence of the bubbles was not completely dictated with the micropatterned surface for the liquids with small surface tension and latent heat of evaporation. The analysis for the second step demonstrated that the use of microstructured surfaces with micropillars rather than the smooth surfaces caused an increase in heat transfer coefficient by 10 times for water and by 8 times for dielectric fluid [17].

Yu and coworkers [18], [19] performed boiling experiments on rectangular fin arrays and artificial cavities using FC-72 as the working fluid. According to their results, vapor/bubble coalescence was the main reason for the stronger impact of cavity density on heat transfer coefficients at higher heat fluxes than those observed at lower heat fluxes. Furthermore, the authors reported that critical heat flux was dependent on the cavity density (number of cavities per unit area), and the CHF enhancement was almost proportional to the area enhancement of the cavity surface. Zhang and Shoji [20] reported that the hydrodynamic interaction between bubbles, thermal interaction between nucleation sites, and bubble coalescence were the co-existing mechanisms influencing pool boiling heat transfer. Coulibaly et al. [21], [22] investigated the effect of bubble coalescence on bubble motion and interfacial heat transfer between bubbles during coalescence. By considering inertia and heat flux controlled bubble growth phases, they developed a mechanistic model for heat transfer rate and stated that heat transfer during bubble coalescence strongly depended on wall superheat.

One of the main mechanisms for boiling crisis is the irreversible formation of dry spots and their density expansion on the superheated surface [23]. At higher heat fluxes, dry spots can be observed beneath some of the generated bubbles, which might lead to the critical heat flux condition [24]. As the number of nucleation sites increases, and the average distance between the bubbles thus decreases, bubble coalescence is likely to occur on the heating surface and plays a vital role on the formation of these dry spots. Furthermore, rising vapor columns might trigger the interfacial instability, which further leads to liquid blockage, reducing the cross sectional flux on the surface, eventually triggering the boiling crisis. Lu et al. [25] proposed a modified hydrodynamic model for boiling CHF by considering the effect of nucleation site density. A similar task was also undertaken by Gong et al. [26], who performed pool boiling experiments to investigate the effect of liquid film on CHF. They observed that in addition to surface rewetting resulting from receding liquid dam of the ruptured bubbles, bubble departure induced liquid flows also rewet the dry spots at high heat fluxes. The authors concluded that irreversible dry spots appeared and expanded laterally near the CHF condition.

The boiling heat transfer mechanisms can be modeled using simulation methods. Lattice Boltzmann (LB) model is one of the most preferred ones for the phase change problems. For example, Hu and Liu [27] proposed an improved hybrid lattice Boltzmann model for simulating the boiling heat transfer. They added a discrete scheme of an infinite volume to the hybrid thermal lattice Boltzmann model to solve the energy non-conversion problem. Moreover, without any addition of heating fluctuation, the bubble nucleation, growth, departure, and coalescence with a bubble departure frequency, bubble nucleation site activation, bubble behavior during the horizontal movement at different regimes were successfully captured by nucleation site treatment of the wall. Recently, Zhou et al. [28] integrated Pseudo – potential model and Peng- Robinson equations of state with the 3D two-particle distribution functions into the Lattice Boltzmann method to investigate periodic bubble nucleation, growth and departure. Additionally, micro-pillar structured surfaces with different geometrical dimensions were introduced to simulations to examine the effect of micro-pillar size on the bubble departure and frequency. Accordingly, among seven different micro-pillar heights and six different micro-pillar distances, the small height of micro-pillars led to a larger heat flux and shorter departure period of a bubble. However, the bubble departure period and heat flux were not strictly affected by the distance of two pillars, while the distance significantly influenced the existence of active nucleation sites. For the micro-pillar distances bigger than 10 lattices and micro-pillar heights smaller than 8 lattices, the activated nucleation sites could be generated in their simulations. Similarly, Ma et al. [29] performed pool boiling heat transfer simulations by using liquid-vapor phase-change lattice Boltzmann model to understand the effect of micro-pillar geometry with different wettability patterns on boiling and bubble dynamics. Four different micro-pillar heat sinks types were considered. The best boiling heat transfer performance was obtained from a bio-inspired heat sink, which has hydrophobic pillars at the top with a hydrophilic base, which serves for the separation of vapor and liquid paths at low superheats, faster bubble departure frequency on the hydrophobic surface and the restriction of the bubble expansion at high superheats by triple-phase lines of the corners of micro-pillars. Moreover, it was emphasized on the importance of the orientation of hydrophilic and hydrophobic area location of the heat sink in boiling curves. Significant effects of micro-pillar width, height and distance on transition from fully developed nucleate boiling to transient film boiling were observed. In transient film boiling, the effects of wettability on maximum heat flux and the Leidenfrost temperature were more dominant than in nucleate boiling. Later, Ma and Cheng [30] did further pool boiling simulations to shed light to dry spot dynamics and wet area fraction on the micro-structures, which were micro-pillars and micro-cavities, for the constant bottom wall temperature condition by using the 3D lattice Boltzmann phase-change model. Cycle-average dry spot diameter was affected by the degree of superheating and geometry of micro-pillars and cavities.

Although available studies show the importance of active nucleation sites and surface wettability on boiling heat transfer, there is still lack of a parametric study on the combined effect of surface wettability and nucleation site interactions on boiling heat transfer and CHF. For this, the microelectromechanical systems (MEMS) technology was employed both for the fabrication of artificial cavities and modification of surface wettability in this study. The effects of hole diameter, pitch distance, surface wettability, and pitch distance/hole diameter ratio on bubble dynamics and corresponding heat transfer rate were examined during pool boiling experiments. Boiling heat transfer, critical heat flux and bubble dynamics characteristics were observed by using a high-speed camera and parametric results, and the effects of surface wettability on nucleation site interactions and critical heat flux were discussed in detail. A critical hole pitch distance/diameter was obtained for enhanced bubble coalescence on hydrophilic and hydrophobic surfaces. This ratio could serve as a valuable design guideline in the design and development of new generation thermal systems.

Section snippets

Sample preparation and structural characterization

The process flow of the fabricated samples is shown in Fig. 1. The sample preparation procedures can be summarized as follows: A 500 µm thick silicon wafer was used as the substrate of the test specimens. The MEMS based fabrication methods were adopted to prepare micro-cavities. Several drops of a positive photoresist (PR) (GXR-601, AZ) were deposited on the top side of the Si wafer, and the wafer was rotated at 2000 rpm for 30 s in a spin coater. Then, thin layer of PR was formed and was baked

Experimental setup

The experimental facility includes a glass pool, an aluminum block heater, 5 cartridge heaters, 6 thermocouples, a power supply, a reflux condenser, a Teflon insulation block, thermometers, and a high speed camera (1000 fps). Fig. 4 shows the schematic of the experimental setup. The high speed camera was placed horizontally, while the light source was located in front of the camera on the other side of the glass pool. De-ionized water was used as the working fluid in pool boiling experiments.

Results and discussion

Pool boiling experiments were performed on 14 different surfaces with artificial cavities (holes) to investigate the effects of wettability, hole diameter and pitch size on boiling heat transfer and CHF. The 32 µm deep circular cavities with the diameters of 50, 100, and 200 µm, the pitch sizes of 500, 1000, and 2000 µm were utilized. The 50 nm thick Teflon thin films were coated on the samples to investigate the effect of surface wettability. Using a high speed camera, visualization was

Conclusions

The effects of hole pitch size/diameter, as a critical parameter for bubble coalescence, on boiling heat transfer (BHT) and critical heat flux (CHF) of 14 different structured surfaces with different wettability were investigated. A parametric study was performed to study bubble departure diameter, bubble departure frequency, bubble coalescence, and critical heat flux mechanism on the fabricated samples. Circular cavities with depth of 32 µm, diameters of 50, 100 and 200 µm, and pitch sizes of

Declaration of Competing Interest

The authors declared that there is no conflict of interest.

Acknowledgements

This work was supported by TUBITAK (The Scientific and Technological Research Council of Turkey) Support Program for Scientific and Technological Research Projects grant number 216M416, SUNUM (Sabanci University Nanotechnology Research and Applications Center) and Sabanci University FENS (Faculty of Engineering and Natural Sciences). The support of Mr. Ilker Sevgen from FENS in improving the experimental setup is highly appreciated.

References (49)

  • L. Zhang et al.

    Nucleation site interaction in pool boiling on the artificial surface

    Int. J. Heat Mass Transf.

    (2003)
  • A. Coulibaly et al.

    Effect of bubble coalescence on the wall heat transfer during subcooled pool boiling

    Int. J. Therm. Sci.

    (2014)
  • A. Coulibaly et al.

    Bubble coalescence at constant wall temperatures during subcooled nucleate pool boiling

    Exp. Therm Fluid Sci.

    (2013)
  • T. Theofanous et al.

    The boiling crisis phenomenon: Part II: dryout dynamics and burnout

    Exp. Therm Fluid Sci.

    (2002)
  • T. Theofanous et al.

    The boiling crisis phenomenon: Part I: nucleation and nucleate boiling heat transfer

    Exp. Therm Fluid Sci.

    (2002)
  • M.-C. Lu et al.

    A modified hydrodynamic model for pool boiling CHF considering the effects of heater size and nucleation site density

    Int. J. Therm. Sci.

    (2015)
  • S. Gong et al.

    An experimental investigation on bubble dynamics and boiling crisis in liquid films

    Int. J. Heat Mass Transf.

    (2014)
  • A. Hu et al.

    2D Simulation of boiling heat transfer on the wall with an improved hybrid lattice Boltzmann model

    Appl. Therm. Eng.

    (2019)
  • P. Zhou et al.

    Lattice Boltzmann simulation of nucleate boiling in micro-pillar structured surface

    Int. J. Heat Mass Transf.

    (2019)
  • X. Ma et al.

    Simulations of saturated boiling heat transfer on bio-inspired two-phase heat sinks by a phase-change lattice Boltzmann method

    Int. J. Heat Mass Transf.

    (2018)
  • X. Ma et al.

    Dry spot dynamics and wet area fractions in pool boiling on micro-pillar and micro-cavity hydrophilic heaters: A 3D lattice Boltzmann phase-change study

    Int. J. Heat Mass Transf.

    (2019)
  • T. Alam et al.

    Experimental investigation of local flow boiling heat transfer and pressure drop characteristics in microgap channel

    Int. J. Multiph. Flow

    (2012)
  • M.R. Özdemir et al.

    Experimental studies on ferrofluid pool boiling in the presence of external magnetic force

    Appl. Therm. Eng.

    (2018)
  • B. Mikic et al.

    On bubble growth rates

    Int. J. Heat Mass Transf.

    (1970)
  • Cited by (45)

    • On coalescence of bubble in branching microchannel with phase change

      2024, International Journal of Heat and Mass Transfer
    • A parametric study on pool boiling heat transfer and critical heat flux on structured surfaces with artificial cavities

      2023, Applied Thermal Engineering
      Citation Excerpt :

      Thermal management of such electronic systems requires efficient heat dissipation to ensure their safe and reliable operational conditions. Boiling is a highly effective phase change heat transfer mechanism dissipating a significant amount of heat through sensible and latent heat [1–9]. The boiling heat transfer (BHT) performance is characterized by heat transfer coefficient (HTC) and critical heat flux (CHF).

    View all citing articles on Scopus
    View full text