0

Research Papers: Evaporation, Boiling, and Condensation

J. Heat Transfer. 2015;137(5):051501-051501-9. doi:10.1115/1.4029593.

Wall nucleation experiments have been performed in a vertical annulus test section for investigation of the bubble departure diameter and bubble departure frequency. The experimental data in forced convective subcooled boiling flow is presented as a parametric study of the effect of wall heat flux, local bulk liquid subcooling, liquid flow rate, and system pressure. The data are shown to extend the database currently available in literature to a wider range of system conditions. Along with the current database in forced convective flow, the available models for bubble departure size and frequency are reviewed and compared with the existing database. The prediction of the bubble departure frequency is shown to require accurate modeling of the bubble departure diameter which has poor agreement with the experimental database.

Commentary by Dr. Valentin Fuster
J. Heat Transfer. 2015;137(5):051502-051502-11. doi:10.1115/1.4029647.

Single-phase convective heat transfer of nanofluids has been studied extensively, and different degrees of enhancement were observed over the base fluids, whereas there is still debate on the improvement in overall thermal performance when both heat transfer and hydrodynamic characteristics are considered. Meanwhile, very few studies have been devoted to investigating two-phase heat transfer of nanofluids, and it remains inconclusive whether the same pessimistic outlook should be expected. In this work, an experimental study of forced convective flow boiling and two-phase flow was conducted for Al2O3–water nanofluids through a minichannel. General flow boiling heat transfer characteristics were measured, and the effects of nanofluids on the onset of nucleate boiling (ONB) were studied. Two-phase flow instabilities were also explored with an emphasis on the transition boundaries of onset of flow instabilities (OFI). It was found that the presence of nanoparticles delays ONB and suppresses OFI, and the extent is correlated to the nanoparticle volume concentration. These effects were attributed to the changes in available nucleation sites and surface wettability as well as thinning of thermal boundary layers in nanofluid flow. Additionally, it was observed that the pressure-drop type flow instability prevails in two-phase flow of nanofluids, but with reduced amplitude in pressure, temperature, and mass flux oscillations.

Commentary by Dr. Valentin Fuster
J. Heat Transfer. 2015;137(5):051503-051503-13. doi:10.1115/1.4029743.

We studied experimentally the behavior of a single cold water droplet that falls onto a hot horizontal flat solid surface in the film boiling region. We found that when the droplet hits the surface (for the first time), three different regimes may occur. These regimes depend on the ratio of fluid's inertia and its surface tension (the Weber number, indicated as We) and on the surface temperature. For relatively low We numbers or surface temperatures, the droplet completely bounces back from the surface and no breakup occurs. For intermediate We numbers or surface temperatures, the spreading stage is faster and the droplet undergoes spreading and partial recession before it breaks up into bouncing small secondary droplets that leap inward and successively coalesce above the surface to form a single droplet once again. For high We numbers or surface temperatures, the spreading velocity is higher, the contact area with the surface is greater, and the liquid film thickness is smaller. Thus, during the expansion of the spreading stage, the droplet breaks up into bouncing small secondary droplets that uninterruptedly leap outward and travel independently. We also present the limiting conditions differentiating between the different behaviors found. This work shows droplet film boiling behaviors that are essentially different than droplet levitation on top of a thin vapor layer, as mainly assumed in theoretical models. We also observed that when a droplet hits the surface for the second, consecutive time (and on), the droplet behaves somewhat differently due to its preheating, very low impact velocity, different shape, spin, orientation, and the surface temperature. At the second impact on the surface (and on), the droplet can continue its bounce in a unique and different manner than in the first impact or it can explode violently to small secondary droplets. Both are unique and differentiating mainly by the droplet's shape and orientation at the exact moment of impact on the surface. Additionally, a rare and unique view of droplet–droplet collision during film boiling is presented. This type of collision behaves in a different way than other droplet–droplet collisions and compared to the adiabatic case of droplet–droplet collision on a nonheated surface. The behaviors found, presented, and discussed in this study change our view of the droplet–surface and droplet–droplet interactions that occur in spray cooling in the film boiling region.

Commentary by Dr. Valentin Fuster
J. Heat Transfer. 2015;137(5):051504-051504-8. doi:10.1115/1.4029815.

Spray cooling is critical in many industrial applications to extract large heat fluxes from metal parts, such as hypervapotron in nuclear fusion reactors, heat treatment of steel plates in run-out table (ROT), electronic parts, and many more. The objective of the present study is to enhance the heat dissipation in transition and nucleate boiling regimes using an air-atomized water spray with water-based copper nanofluid as a coolant. The nanoparticle used in this study is energetic metal Cu, which has been prepared by mechanical milling (MM) process. The nanofluid has been prepared by suspending 0.1 vol. % Cu nanoparticles in water, with or without a dispersing agent (surfactant). The effect of type of dispersing agent on augmentation of boiling heat transfer has also been studied. The spray cooling experiments are conducted on a 6 mm thick stainless steel plate of initial temperature above 900 °C. The transient surface heat flux and temperatures are estimated using commercial inverse heat conduction software named intemp. The experimental results illustrated that transition and nucleate boiling heat flux as well as critical heat flux (CHF) increased significantly using nanofluid spray. A maximum ultrafast cooling (UFC) rate of 267 °C/s is achieved using surfactant-based nanofluid spray, which is 31.53% and 59.88% higher as compared to the nanofluid without any dispersant and pure water sprays, respectively. Overall, the surfactant-based copper nanofluid spray can serve as a better coolant on the ROT of steel processing industry.

Commentary by Dr. Valentin Fuster

Research Papers: Experimental Techniques

J. Heat Transfer. 2015;137(5):051601-051601-11. doi:10.1115/1.4029595.

A methodology to evaluate the specific heat of metallic thin films at constant pressure and 300 K by means of the heating profile is proposed. Changes on the electrical resistance of metallic films after the application of short electric pulses (20–500 μs) are correlated with changes of temperature of the films. Electric pulses are applied on films by an implemented electronic device. A proposed analytical thermal model predicts the correlation between the duration of the electric pulses and the thermal profiles of the film/substrate systems. The analytical thermal model and the measured thermal profiles results are useful to evaluate the specific heat of films. Following this methodology, Au and Al nanofilms evaporated on glass substrates were analyzed. Results indicate that specific heat values of Au films decrease from (229 ± 15) J/kg K to (125 ± 8) J/kg K, and for Al films from (1444 ± 89) J/kg K to (947 ± 53) J/kg K, for film thicknesses from 20 to 200 nm.

Commentary by Dr. Valentin Fuster

Research Papers: Forced Convection

J. Heat Transfer. 2015;137(5):051701-051701-10. doi:10.1115/1.4029596.

Entropy generation is squarely linked with irreversibility, and consequently with exergy destruction within a thermal system. This study concerns with the temperature distribution, and local and volumetric averaged entropy generation rates within a cylindrical system with two solid co-rotating inner and outer parts and the middle nanofluid flow part. Temperature-dependent thermal conductivities for solid materials are included within the modeling. To obtain the temperature formula within all three sections, a combined analytical–numerical solution technique is applied. An exact analytical solution is also obtained, when constant thermal conductivities for solid materials are assumed. The resultant data from the analytical–numerical solution technique is verified against the investigated exact solution. Thereafter, the velocity and temperature fields from the combined analytical–numerical solution technique are incorporated into the entropy generation formulations to obtain the local and volumetric averaged entropy generation rates. Using abovementioned procedure, the effects of thermophysical parameters such as nanoparticles volume concentration, Brinkman number, thermal conductivity parameter ratios, outer temperature boundary condition, internal heat generation rates and velocity ratios on the temperature field, and entropy generation rates are investigated.

Commentary by Dr. Valentin Fuster

Research Papers: Heat Exchangers

J. Heat Transfer. 2015;137(5):051801-051801-11. doi:10.1115/1.4029764.

Computational fluid dynamics (CFD) and heat transfer simulations are conducted for a novel heat exchanger. The heat exchanger consists of semi-circle cross-sectioned tubes that create narrow slots oriented in the streamwise direction. Numerical simulations are conducted for Reynolds numbers (Re) ranging from 700 to 30,000. Three-dimensional turbulent flows and heat transfer characteristics in the tube bank region are modeled by the k-ε Reynolds-averaged Navier–Stokes (RANS) method. The flow structure predicted by the two-dimensional and three-dimensional simulations is compared against that observed by the particle image velocimetry (PIV) for Re of 1500 and 4000. The adequate agreement between the predicted and observed flow characteristics validates the numerical method and the turbulent model employed here. The three-dimensional and the two-dimensional steady flow simulations are compared to determine the effects of the wall on the flow structure. The wall influences the spatial structure of the vortices formed in the wake of the tubes and near the exit of the slots. The heat transfer coefficient of the slotted tubes improved by more than 40% compare to the traditional nonslotted tubes.

Commentary by Dr. Valentin Fuster

Research Papers: Natural and Mixed Convection

J. Heat Transfer. 2015;137(5):052501-052501-10. doi:10.1115/1.4029594.

Natural convective flow over narrow plates induces an inward flow near the edges of the plate causing the flow to be three-dimensional near the edges of the plate. This influences the heat transfer rate near the edges of the plate and is referred to as the edge effect. The primary objective of this paper is to numerically study this edge effect and the interaction of the flows over two inclined vertically separated narrow heated plates of the same size embedded in a plane adiabatic surface. The cases where the plates and surrounding adiabatic surface are inclined at positive or negative angles to the vertical have been considered. Results were obtained by numerically solving the full three-dimensional form of governing equations using the commercial finite volume based software Fluent©. Results have only been obtained for a Prandtl number of 0.7; this being the value existing in the application which involved airflow that originally motivated this study. The results presented here cover Rayleigh numbers between 103 and 107, at all values of W considered, plate width-to-height ratios between 0.2 and 1.2, gap, at all values of W considered, to the plate height ratios of between 0 and 1.5, and, at all values of W considered, angles of inclination of between −45 deg and +45 deg. The effects of the Rayleigh number, dimensionless plate width, dimensionless gap between plates, and inclination angle on the heat transfer rate have been studied in detail. Empirical correlations defining the effect of these parameters on the heat transfer rate have been derived.

Commentary by Dr. Valentin Fuster
J. Heat Transfer. 2015;137(5):052502-052502-10. doi:10.1115/1.4029681.

A numerical study is conducted for the Joule heating effects on fluid flow and heat transfer of radiatively participating magnetohydrodynamics (MHD) under different levels of thermal radiation considering the Hall effects in a square cavity. In the cavity, the vertical walls are isothermal with constant but different temperatures, while the horizontal walls are adiabatic. The absorption, emission, and scattering of the fluid and the reflection, absorption, and emission of the walls are all taken into account. The governing equations for momentum and energy together with the boundary conditions are solved by the finite volume method (FVM), while the governing equation for radiative transfer is solved by the discrete ordinates method (DOM). Tabular and graphical results are presented in terms of streamlines, isotherms, Nusselt number, and the average temperature of the fluid. After detailed analysis, we found that the Joule heating has notable effects on fluid flow and heat transfer in the cavity and Joule heating cannot be neglected in certain range of parameters.

Commentary by Dr. Valentin Fuster
J. Heat Transfer. 2015;137(5):052503-052503-10. doi:10.1115/1.4029742.

Convection heat transfer below a horizontal, hot, and isothermal strip of infinite length and width of 2L embedded in fluids with different Prandtl number (Pr) and Nusselt number (Nu) is analyzed with the aid of integral method. A new concept is utilized to determine the boundary layer thickness at the strip's edge, which is based on matching the flow rate of the boundary layer below the strip at its edge and the flow rate of the plume, which forms after the heated fluid detaches from the strip's edge. In addition to these novelties, a numerical model is developed to verify the analytical framework, and an excellent agreement is observed between the analytical and numerical models.

Commentary by Dr. Valentin Fuster
J. Heat Transfer. 2015;137(5):052504-052504-12. doi:10.1115/1.4029763.

The effects of inclination $180deg≥φ≥0deg$ on steady-state laminar natural convection of yield-stress fluids, modeled assuming a Bingham approach, have been numerically analyzed for nominal values of Rayleigh number Ra ranging from 103 to 105 in a square enclosure of infinite span lying horizontally at $φ=0deg$, then rotated about its axis for $φ>0deg$ cases. It has been found that the mean Nusselt number $Nu¯$ increases with increasing values of Rayleigh number but $Nu¯$ values for yield-stress fluids are smaller than that obtained in the case of Newtonian fluids with the same nominal value of Rayleigh number Ra due to the weakening of convective transport. For large values of Bingham number Bn (i.e., nondimensional yield stress), the mean Nusselt number $Nu¯$ value settles to unity ($Nu¯=1.0$) as heat transfer takes place principally due to thermal conduction. The mean Nusselt number $Nu¯$ for both Newtonian and Bingham fluids decreases with increasing $φ$ until reaching a local minimum at an angle $φ*$ before rising with increasing $φ$ until $φ=90deg$. For $φ>90deg$ the mean Nusselt number $Nu¯$ decreases with increasing $φ$ before assuming $Nu¯=1.0$ at $φ=180deg$ for all values of $Ra$. The Bingham number above which $Nu¯$ becomes unity (denoted $Bnmax$) has been found to decrease with increasing $φ$ until a local minimum is obtained at an angle $φ*$ before rising with increasing $φ$ until $φ=90deg$. However, $Bnmax$ decreases monotonically with increasing $φ$ for $90deg<φ<180deg$. A correlation has been proposed in terms of $φ$, Ra, and Bn, which has been shown to satisfactorily capture $Nu¯$ obtained from simulation data for the range of Ra and $φ$ considered here.

Commentary by Dr. Valentin Fuster

Research Papers: Porous Media

J. Heat Transfer. 2015;137(5):052601-052601-5. doi:10.1115/1.4029773.

The model developed in our previous paper (Nield and Kuznetsov, 2011, “The Effect of Vertical Throughflow on Thermal Instability in a Porous Medium Layer Saturated by a Nanofluid,” Transp. Porous Media, 87(3), pp. 765–775) is now revised to accommodate a more realistic boundary condition on the nanoparticle volume fraction. The new boundary condition postulates zero nanoparticle flux through the boundaries. We established that in the new model, oscillatory instability is impossible. We also established that the critical Rayleigh number depends on three dimensionless parameters, and we derived these three parameters from the governing equations. We also briefly investigated the major trends.

Commentary by Dr. Valentin Fuster

Research Papers: Radiative Heat Transfer

J. Heat Transfer. 2015;137(5):052701-052701-9. doi:10.1115/1.4029546.

A general formulation of the spherical harmonics (PN) methods was developed recently to expand the method to high orders of PN. The set of N(N + 1)/2 three-dimensional second-order elliptic PDEs formulation and their Marshak boundary conditions for arbitrary geometries are implemented in the openfoam finite volume based cfd software. The results are verified for four cases, including a 1D slab, a 2D square enclosure, a 3D cylindrical enclosure, and an axisymmetric flame. All cases have strongly varying radiative properties, and the results are compared with exact solutions and solutions from the photon Monte Carlo method (PMC).

Commentary by Dr. Valentin Fuster

Research Papers: Two-Phase Flow and Heat Transfer

J. Heat Transfer. 2015;137(5):052901-052901-8. doi:10.1115/1.4029694.

Aluminum heat pipes have traditionally been incompatible with water and water-based fluids because they quickly react to generate noncondensable hydrogen gas (NCG). Two different inorganic aqueous solutions (IAS) are tested in a flat heat pipe (FHP). Grooved aluminum plates were used as the heat pipe wick and the tests were run with the heating section raised above the condenser. Compatibility between the working fluid and the aluminum heat pipe was established by running the device to dryout and observing thermal resistance results along the way. De-ionized (DI) water was also tested, as a baseline for comparison, to establish that it did indeed fail as expected. Operating performance of each mixture was obtained from zero heat input until dryout was reached for two angles of inclination. The data suggest that both IAS mixtures are compatible with aluminum heat pipes and exhibit performance similar to that of a copper and water heat pipe. It is demonstrated that IAS and aluminum heat pipes show potential for replacing existing copper and water devices for some applications and provide alternative options for heat pipe designers who value both the thermophysical property advantages of water and reduced weight of aluminum devices.

Commentary by Dr. Valentin Fuster

Technical Brief

J. Heat Transfer. 2015;137(5):054501-054501-5. doi:10.1115/1.4029646.

Experimental and numerical results of heat transfer by mixed convection in a ventilated cavity are presented. The results were obtained for a 1.0 m × 1.0 m × 1.0 m cavity. The inlet and outlet dimensions were of 0.08 m × 0.08 m, and the air velocity at the inlet was set to 0.1 and 0.5 m/s. The left wall receives a uniform and constant heat flux whereas the right wall was maintained at a constant temperature. Experimental and numerical results of temperature profiles and heat transfer coefficients are presented and compared. The results showed that the variation of the Rayleigh number increases about 1% the percentage differences between experimental and numerical values, and the maximum percentage differences on heat transfer coefficients are 2.0% for Ra = 2.7 × 108 and 3.0% for Ra = 4.5 × 108.

Commentary by Dr. Valentin Fuster