Search

Copyright © 2021 The Author(s). The pool boiling process is one of the most effective heat transfer modes capable of transferring large amounts of heat with small temperature difference between the heated surface and the fluid. In addition, fundamental knowledge of pool boiling processes is the starting point of flow boiling research and applications. It is therefore no surprise that it has been, and still is, the subject of extensive research globally for quite some time and a critical analysis is now required in order to move forward with enhanced surface designs. The current on-going research focuses on the understanding of boiling fundamentals including bubble generation, growth and bubble dynamics. In this context, fluid-surface interaction is critical. In the first part of this two-part paper we present the factors and parameters affecting the above, starting with the criteria for gas/vapour entrapment, nucleation site stability and the superheat required for heterogeneous nucleation. The models predicting the incipience superheat are critically described, classified into phase instability and superheated boundary-layer based models. This first part includes bubble growth and departure models, elucidating the effect of surface topology and wettability that can inform and facilitate the design of enhanced surfaces that are presented in Part II [10]. Three fluids of industrial interest, i.e. FC-72, HFE7100 and water were used through the discussion, as examples, to represent low and high surface tension fluids and help the understanding of surface-fluid interactions and relation to possible heat transfer enhancements. Engineering and Physical Sciences Research Council, UK (Grant Reference: EP/S019502/1).

Recent advancements in the electronics industry led to smaller and more powerful systems. Therefore, efficient heat dissipation for cooling of microelectronics systems, integrated circuit chips, power semiconductor devices such as IGBTs and laser diodes is required. In such systems, heat fluxes in the order of MW/m2 need to be removed from small spaces, while maintaining the temperature below a certain design limit (Karayiannis and Mahmoud in Appl Therm Eng 115:1372–1397, 2017 [1]). Flow boiling in microchannels is one of the most promising methods for achieving these high cooling demands because it can dissipate large heat fluxes over a small surface area by utilizing the latent heat of the coolant (Wang in Appl Therm Eng 110:369–381, 2017 [2]). Literature review indicates that there are still disagreements on the prevailing flow patterns, heat transfer rates, and pressure drop trends; see Mahmoud and Karayiannis (Encyclopedia of Two-Phase Heat Transfer and Flow IV, 2018, pp 233–301 [3]). The main objective of the present work was to investigate the flow boiling flow patterns, heat transfer coefficient, and pressure drop characteristics in a square metallic microchannel at different inlet subcooling, mass flux and heat flux conditions, using water as the working fluid.

Fanno theory provides an analytical model for one-dimensional confined viscous compressible flows. The model holds under the assumptions of adiabatic flow and constant cross-section channel. From theory, the differential of every flow-related quantity is expressed as a function of Mach number and friction factor. One-dimensional flow numerical models can be derived by discretizing Fanno equations. However, theory does not assess how to evaluate friction, while the model works properly only if friction is estimated correctly. Compressibility and turbulence act by deforming the velocity profile making it flatter. Assuming the friction factor function of the Reynolds number alone, in line with incompressible flow theory, is thus not correct. Better correlations should include the Mach number to address compressibility effects. Here, the impact of turbulence and compressibility on the velocity profiles in a micro-channel is analysed by means of CFD simulations. Friction factor correlations are deduced for turbulent micro-flows. The impact of the velocity profile on other quantities, such as dynamic pressure and bulk temperature, needed for the numerical model operation, is also evaluated. Additional correlations for these quantities overcome the instrinsic limits of the one-dimensional model, necessarily unaware of local velocity profiles, in a quasi-2D fashion significantly improving its predicting capabilities.

The main objective of this research was the design and test of a multi-microchannels heat sink for electronics cooling applications, which operates at system pressure near atmospheric and low mass flow rates. HFE-7100, a dielectric and eco-friendly coolant, was chosen as the working fluid. Twenty five rectangular microchannels, of width 0.7 mm and height 0.35 mm giving a hydraulic diameter of 0.46 mm, were fabricated from oxygen-free copper with a base area of 500 mm2. The channels in-between wall was 0.1 mm thick. Five mass fluxes ranging from 50 to 250 kg/m2 s were tested at fixed inlet sub-cooling near 5 K. The effect of heat flux, mass flux and vapour quality on the local heat transfer coefficient was investigated. Four flow patterns namely; bubbly, slug, churn and annular flow, were visualized using a high-speed camera mounted on a microscope. In this study, the maximum flow boiling heat transfer coefficient was 12.71 kW/m2 K at mass flux of 250 kg/m2 s. A comprehensive comparison with experimental results was conducted including flow pattern maps, heat transfer and pressure drop correlations. Some of the correlations, proposed for conventional channels and microchannels, showed good agreement with the present results.

Engineering & Physical Sciences Research Council (Grant EP/N011112/1 ).

The effect of channel aspect ratio on flow boiling characteristics (flow patterns, heat transfer and pressure drop) of HFE-7100 in copper multi-microchannel heat sinks was investigated experimentally. Three heat sinks with base area 500 mm2, channel hydraulic diameter 0.46 mm and channel aspect ratio (ratio of channel width to channel height) of 0.5, 1.0 and 2.0 were tested. The average surface roughness of the channel bottom surface was nearly the same in the three heat sinks and the measured values were 0.271, 0.286 and 0.304 µm. The local heat transfer rates were determined simultaneously with flow visualisation at mass flux ranging from 50 to 250 kg/m2s, wall heat flux from 9.6 to 191.6 kW/m2, system pressure of 1 bar and low inlet sub-cooling of 5 K. The results showed that, when the channel aspect ratio increased, the heat transfer coefficient increased, while the flow boiling pressure drop decreased. However, the heat transfer rate calculated using the heat sink base area was higher in the heat sink with the smallest channel aspect ratio, indicating an enhancement due to the largest surface area.

From Poiseuille theory, it is known that incompressible laminar fully-developed flow of a Newtonian fluid in a constant cross-section channel is characterised by steady parabolic velocity profiles after a fully-developed flow condition is attained. In turbulent fully-developed flow the velocity profiles are non-parabolic and become more flat for higher Reynolds numbers. When the incompressible hypothesis does not hold, as in the case of high velocity ideal gas flow, the velocity profile becomes flatter, as if more turbulent, due to the superposition of compressibility and turbulence effects, if applicable. This is typical in micro-channel flows, where pressure gradients are high and the gas is rapidly accelerating, eventually up to the sound velocity. As the flow accelerates the effects of compressibility grow stronger and the velocity profile keeps changing shape. The radial velocity component does not zero as in fully-developed flow but reverses after the entrance effects have damped out and grows with the Mach number. A net mass transfer toward the walls is thus generated making the velocity profile more flat. This affects the friction factor which is no longer constant, being proportional to the normal-to-wall velocity gradient, and needs to be evaluated. In the present work, the compressible friction factor is numerically investigated and correlations are proposed based on the velocity profile shape evolution as a function of the Mach number. This, together with other considerations on the velocity profile shape change, is shown to enhance the predictive capability of the Fanno theory for compressible flows.From Poiseuille theory, it is known that incompressible laminar fully-developed flow of a Newtonian fluid in a constant cross-section channel is characterised by steady parabolic velocity profiles after a fully-developed flow condition is attained. In turbulent fully-developed flow the velocity profiles are non-parabolic and become more flat for higher Reynolds numbers. When the incompressible hypothesis does not hold, as in the case of high velocity ideal gas flow, the velocity profile becomes flatter, as if more turbulent, due to the superposition of compressibility and turbulence effects, if applicable. This is typical in micro-channel flows, where pressure gradients are high and the gas is rapidly accelerating, eventually up to the sound velocity. As the flow accelerates the effects of compressibility grow stronger and the velocity profile keeps changing shape. The radial velocity component does not zero as in fully-developed flow but reverses after the entrance effects have damped out and grows with ...

Numerical simulations were performed using Fluent 14.5 to investigate single phase flow and conjugate heat transfer in copper rectangular microchannels. Two different configurations were simulated: (1) single channel with hydraulic diameter of 0.561 mm and (2) multichannel configuration consisting of inlet and outlet manifolds and 25 channels with hydraulic diameter of 0.409 mm. In the single channel configuration, four numerical models were investigated namely, 2D thin-wall, 3D thin-wall (heated from the bottom), 3D thin-wall (three side heated) and 3D full conjugate models. In the multichannel configuration, only 3D full conjugate model was used. The simulation results of the single channel configuration were validated using experimental data of water as a test fluid while the results of the multichannel configuration were validated using experimental data of R134a refrigerant. In the multichannel configuration, flow distribution among the channels was also investigated. The 3D thin-wall model simulation was conducted at thermal boundary conditions similar to those assumed in the experimental data reduction (uniform heat flux) and showed excellent agreement with the experimental data. However, the results of the 3D full conjugate model demonstrated that there is a significant conjugate effect and the heat flux is not uniformly distributed along the channel resulting in significant deviation compared to the experimental data (more than 50%). Also, the results demonstrated that there is a significant difference between the 3D thin-wall and full conjugate models. The simulation of the multichannel configuration with an inlet manifold having gradual decrease in cross sectional area achieved very reasonable uniform flow distribution among the channels which will provide uniform heat transfer rates across the base of the microchannels.

The experimental results of two-phase flow boiling pressure drop of R134a in vertical microdiameter stainless steel tubes are presented in this paper. The tests were conducted using four tubes: one tube with an inner diameter of 0.52 mm and 100 mm heated length and three tubes with an inner diameter of 1.1 mm and different heated lengths (150, 300, and 450 mm). Other experimental conditions include mass flux range of 200–500 kg/m2-s, system pressure range of 6–10 bar, inlet subcooling value of about 5 K, and heat flux range of 1–140 kW/m2. The results indicated that the total measured two-phase pressure drop increases with increasing mass flux, heat flux (exit quality) and decreasing system pressure and tube inner diameter. The test section heated length was found to have a significant effect on the measured pressure drop per metre length. The total measured two-phase pressure drop results were also compared with eighteen macro- and microscale models and correlations.

Accurate predictions of two-phase pressure drop in small to micro-diameter passages are necessary for the design of compact and ultra-compact heat exchangers, which find wide application in process and refrigeration industries and in the cooling of electronics. A semimechanistic model of boiling two-phase pressure drop in the confined bubble regime is formulated, following the three-zone approach for heat transfer. The total pressure drop is calculated by time-averaging the pressure drops for single-phase liquid, elongated bubble with a thin liquid film, and single-phase vapor. The model results were compared with experimental data collected for a wide range of tube diameters (4.26, 2.88, 2.02, 1.1, and 0.52 mm) for R134a at pressures of 6–12 bar. In this model's present form, its predictions are close to those of the homogeneous flow model but it provides a platform for further development.

An experimental study of low-level turbulence natural convection in an air filled vertical square cavity was conducted. The cavity was 0.75 m high × 0.75 m wide × 1.5 m deep giving 2D flow. The hot and cold walls of the cavity were isothermal at 50 and 10 °C respectively giving a Rayleigh number of 1.58 × 109. The local velocity and temperature were simultaneously measured at different locations in the cavity and both mean and fluctuation quantities are presented, i.e. ū, u′rms, v, v′rms, T , T′rms, u ′ v ′ , u ′ T ′ and v ′ T ′ . The local and average Nusselt numbers, the wall shear stress as well as the turbulent kinetic energy and the dissipation rate of the temperature variance are also presented. The experiments were conducted with very high accuracy and as such the results can form experimental benchmark data and will be useful for validation of computational fluid dynamics codes.

An experimental study of low level turbulence natural convection in an air filled vertical square cavity was conducted. The dimensions of cavity were 0.75 m × 0.75 m × 1.5 m giving two-dimensional flow. The hot and cold walls of the cavity were isothermal at 50 and 10°C, respectively, giving a Rayleigh number of 1.58 × 109. The temperature and velocity distribution was systematically measured at different locations in the cavity, and was nearly anti-symmetrical. An experimentally obtained contour plot of the thermal field and a vector plot of the air flow in the cavity are reported for low turbulence natural convection in such cavities for the first time. The wall shear stress and the local and average Nusselt numbers are also presented. The Nusselt number compares well with previous results; the agreement on the velocity and temperature profiles at mid-height near the vertical walls is fair. Differences were found at mid-width and in the rate of velocity and temperature changes near the walls. The experiments were conducted with high accuracy. Therefore, the results can form experimental benchmark data and will be useful for CFD code validation.

Simple expressions have been developed for important variables of the flow in an open-flow solar regenerator (with applications in absorption cooling systems) through a first- and second-law analysis of the system. In particular, expressions for the regenerator figure of merit, the exergetic or rational efficiency and the entropy production rate have been given and applied to some available data with reasonably good results.

A three-dimensional (3-D) code which can predict the complete performance of high efficiency boilers is described in this paper. The model is in two parts. The combustion chamber is modelled in the first part and the heat exchanger in the second. The thermal efficiency is calculated from the predicted results. The computational model was based on the two-dimensional (2-D) TEAM code which has been extended into 3-D and extensively modified by the present authors. In the code, the gaseous combustion was modelled by Simple Chemical Reaction System concept; the Radiation process was handled by the ‘Discrete Transfer’ method. The modelling results of the combustion chamber were used as input for the calculation of the consequent finned tube heat exchanger. The heat exchanger calculation included conduction, convection and radiation. Predictions were performed for a low Reynolds number boiler and compared with existing data. Good agreement was obtained.

Experimental results for etectrohydrodynamic (EHD) effects on nucleate boiling at passively enhanced surfaces, such as Thermoexcel-HE and Gewa-T, are reported. The experiments were carried out using a single-tube test rig by applying a 0- to 30- kV dc potential to a dielectric liquid (Refrigerant-114). The EHD effect on the enhanced surfaces is compared with results of earlier experiments on smooth and Thermoexcel-C tubes and discussed. The results indicate that the boiling hysteresis at these passively enhanced surfaces can be eliminated and the nucleate boiling can be further enhanced and controlled by applying an electric field. An explanation of bubble initiation at low superheat by an electric field is offered in terms of the effect of the field on bubble critical radius.

The electrohydrodynamic (EHD) enhancement mechanism is first outlined in this paper. A comprehensive review of past work on EHD single and two-phase heat transfer, as well as past work in the related area of EHD-induced flow, is then presented. Correlation attempts are also reviewed. Recent experimental results for EHD boiling and condensation in single- and multi-tube heat exchangers are discussed. A description of the possible practical EHD electrode systems for applications in power production cycles and regrigeration is also presented. The research work needed to clarify outstanding questions in EHD and encourage its use in practical systems is discussed.

Copyright @ 2011 Elsevier Two-phase flow boiling in microchannels is one of the most promising cooling technologies for coping with high heat fluxes produced by the next generation of central processor units (CPUs). If flow boiling is to be used as a thermal management method for high heat flux electronics it is necessary to understand the behaviour of a non-uniform heat distribution, which is typically the case observed in a real operating CPU. The work presented is an experimental study of two-phase boiling in a multi-channel silicon heat sink with non-uniform heating, using water as the cooling liquid. Thin nickel film sensors, integrated on the back side of the heat sinks were used in order to gain insight related to temperature fluctuations caused by two-phase flow instabilities under non-uniform heating. The effect of various hotspot locations on the temperature profile and pressure drop has been investigated. It was observed that boiling inside microchannels with axially non-uniform heating leads to high temperature non-uniformity in the transverse direction. This research was supported by the UK Engineering and Physical Sciences Research Council through grant EP/D500109/1.

Convective heat transfer was investigated numerically for rectangular enclosures both undivided and divided in two zones by a vertical partition, and having opposite isothermal walls at different temperatures. The aspect ratio was varied from 0.1 to 16 and the Rayleigh number from 3.5 ∗ 10 3 to ∗ 10 7 (non-partitioned enclosures) and from 1.0 ∗ 10 5 to 1.6 ∗ 10 8 (partitioned enclosures). The thickness and conductivity of the partition were varied. The end wall thermal boundary conditions were adiabatic or LTP (Linear Temperature Profile). The continuity, momentum and energy equations for a 2-D laminar steady flow were solved under the Boussinesq approximation by using a finite-difference method and the SIMPLEC pressure-velocity coupling algorithm. Grid-independent results indicate that the reduction in the Nusselt number caused by a thin central partition can be predicted within a few per cent (in the range investigated) by assuming the partition to be isothermal, i.e. infinitely conducting. The finite conductivity of the partition causes a temperature distribution along its length, resulting in an increase in Nu which depends on Rayleigh number, aspect ratio and end wall thermal boundary conditions.

Bubble nucleation and growth to confinement during flow boiling in microchannels lead to high heat transfer coefficients. They may also create pressure fluctuations that change the superheat driving evaporation and cause flow reversals that promote transient dry-out and uneven distribution of flow between parallel channels. The work described in this paper is part of a programme to develop models for these processes that will aid the design of evaporative cooling systems for devices operating at high heat fluxes. Video observations of water boiling in a single copper channel of rectangular cross-section, 0.38 × 1.6 mm and a heated length 40 mm, were performed. The top side of the channel was a glass window. Results are presented for a heat flux, averaged over the area of the three metal sides, of 210 and 173 W/m2 K for incompressible and compressible inlet flow conditions. The inlet pressure was about 1.12 bar and the mass flux was 747.5 kg/m2 s for both conditions examined. The results demonstrated the strong influence of compressibility on the mode of bubble detachment and growth and therefore on flow patterns, pressure fluctuations and heat transfer rates. The fluid mechanics of boiling in this size channel were also successfully investigated by 3-D numerical simulation for bubbles growing at a defined rate with a fixed inlet flow rate using the 3-D CFD code FLUENT 6 (no upstream compressibility). The study examined the fluid mechanics of bubble motion with heat transfer, but the mass transfer across the bubble-liquid interface was not simulated in the present work. A small vapour bubble was injected at the wall to ensure the bubble generation is under a quasi nucleation condition. Its growth was driven by an internal source of vapour, at a rate derived by analysis of the experimental measurements of growth. The simulation reproduced well the observed motion and shape of the bubble. The simulation was then extended to model bubbles generated and growing randomly in a 2-D channel.Copyright © 2009 by ASME

Copyright @ 2009 ASME A simple 1-D model with low requirements for computing time is required to investigate parametric influences on the potentially adverse effects of pressure fluctuations driven by confined vapour bubble growth in microchannel evaporative cooling systems operating at high heat fluxes. A model is developed in this paper for the particular conditions of a channel of rectangular cross-section with high aspect ratio with a constant inlet flow rate (zero upstream compressibility). (The model will later be extended to the conditions of finite upstream compressibility that lead to transient flow reversal). Some parametric trends predicted by the model are presented. The simplifying assumptions in the model are examined in the light of a 3-D simulation by a commercial CFD code, described in an accompanying paper by the same authors. The predictions of pressure changes are in reasonable agreement. It is suggested that the 1-D model will be a useful design tool. This work is supported by the UK Engineering and Physical Sciences Research Council (EPSRC) under grants EP/D500095/1 and EP/D500125/1.

The effect of symmetric partitions protruding centrally from the end walls of a rectangular vertical enclosure on heat transfer rates is investigated numerically. The enclosure has opposite isothermal walls at different temperatures. The Rayleigh number is varied from 10 4 to 10 7 and the aspect ratio from 0.5 to 10. The thickness of the partitions is fixed and equal to one tenth of the width of the enclosure. Their non-dimensional length ( L / H ) is varied from 0 (non-partitioned enclosure) to 0.5 (two separate enclosures). The effect of different thermal boundary conditions at the end walls and at the partitions is included in the investigation.

Flow boiling heat transfer experimental results, obtained in two stainless steel tubes of internal diameter 4.26 mm and 2.01 mm using R134a as the working fluid, indicate that the local heat transfer coefficient increases with heat flux and is independent of vapour quality when this is less than about 40% to 50% for the 4.26 mm tube and 20% to 30% for the 2.01 mm tube, conventionally interpreted as nucleate boiling. Above these quality values, the separate graphs merge into a single line for heat transfer coefficient decreasing with increasing vapour quality. The data in the apparently-nucleate boiling condition are compared with a recent state-of-the-art three-zone evaporation model for the confined bubble flow regime without a nucleate boiling contribution. The model predicts the experimental data reasonably well but does not predict correctly the trends for changing pressure and diameter. Some suggestions are made for improving the model. The comparisons made in this paper support the statements by the developers of the model and others that the application of conventional macro flow boiling correlations to micro tube flow boiling heat transfer may not necessarily have a sound physical basis.

Triangular enclosures are typical configurations found in attic spaces of residential as well as industrial pitched-roof buildings. Previous studies on airflow within such spaces have focused on the laminar range. A numerical study has been carried out for time-dependent, two-dimensional low turbulence natural convection of air contained in the attic of a pitched roof of triangular cross-section. Two sets of boundary conditions have been considered: enclosure heated from the inclined walls and from the horizontal base wall, for aspect ratios (AR) of 0.33 ≤ AR ≤ 1.73 for Rayleigh number (Ra), 109 ≤ Ra ≤ 1011. In this paper, the flow field and Reynolds stress distribution are presented for the isothermal and isoflux cases.

An interferometric technique was used to study coupled convective heat transfer for an inclined solar collector. The collector was simulated by an inclined, closed rectangular cavity which was heated isothermally at the bottom plate and cooled by natural convection at the top plate. Both plates had dimensions 457.2 mm wide and 424.2 mm long. Infinite interferograms, for temperature field visualization, and finite interferograms for analysis were obtained. The parameters studied were Rayleigh number, aspect ratio, and angle of inclination. The Rayleigh number was varied from subcritical to 6 × 105 and the aspect ratio was changed from 6.68 to 33.4. The angle of inclination was 15° and 60° to the horizontal. The temperature on the cold plate was found to increase with distance from the lower boundary. Temperature reversal occurs in the cavity and first appears in the corner regions. The temperature gradients were obtained and thus the local heat transfer coefficients were calculated indicating a strong dependence from side boundaries. In comparison with reports where both surfaces were isothermal a significant difference in the dependence of average Nusselt number on Rayleigh number was found. At high Rayleigh numbers the average Nusselt number is less dependent on Rayleigh number for the cavity with a non-isothermal top surface than for the isothermal top surface.