0

IN THIS VOLUME


2003;():1-6. doi:10.1115/ICMM2003-1000.

Microchannels and Minichannels are found in many biological systems providing very high heat and mass transfer rates in organs such as the brain, lung, liver and kidney. Many high flux cooling applications are effectively utilizing their high heat transfer capabilities of these channels. A brief overview of the historical perspective and some of the issues that need to be addressed with microchannels and Minichannels are presented in this paper.

Topics: Microchannels
Commentary by Dr. Valentin Fuster
2003;():7-14. doi:10.1115/ICMM2003-1002.

An overview will be given about investigations on heat and mass transfer in narrow channels and narrow cavities, from work carried out in the last years up to the current status of research of some relevant scientific groups in Europe. The major topics of this report are evaporation heat transfer and the flow boiling pressure drop in narrow channels; microscale heat and mass transfer phenomena in pool boiling from enhanced evaporator tubes with sub-surface channels are also addressed. In the last years a challenging topic has been the enhancement of the efficiency of heat exchangers by employing micro-structured heat transfer surfaces. The need for smaller heat exchangers with higher heat transfer rates and/or smaller thermal approaches is caused by the ongoing miniaturisation of mechanical and electronic components, leading to higher heat fluxes which can damage or even destroy the components. On the other hand, enhanced heat transfer in big equipment, e.g. heat exchangers for the petrochemical and chemical industries, can lead to significant materials and energy savings and thus reduce environmental pollution. Therefore the European Union, European industries and national organisations have supported various projects to develop and to investigate a new generation of heat transfer surfaces, to better understand the related heat transfer phenomena and to model the heat transfer from these micro heat exchanger elements. There is a very extensive research in this scientific field, comprising both flow boiling and pool boiling. The present paper deals with heat transfer in narrow channels and/or cavities and with the flow boiling pressure drop occurring during heat and mass transfer in narrow channels. Investigations of major European institutions, carried out in the past and at the moment will be presented as a contribution to the overview on the current state-of-the-art in Europe, without claim of completeness. Some recent results on microscale pool boiling and flow boiling obtained in our institute will also be presented (Shuai et al., 2002; Kulenovic et al., 2002; Chen et al., 2002a, b).

Topics: Heat transfer
Commentary by Dr. Valentin Fuster
2003;():15-23. doi:10.1115/ICMM2003-1003.

This paper presents an overview of the research activities performed in France, and presents more precisely the work carried out by the “microthermique” group in Grenoble (“Micro thermal exchanges”), which is composed of 5 teams. Examples of fundamentals and applied research projects on micro-channel heat exchangers and systems are presented.

Commentary by Dr. Valentin Fuster
2003;():25-31. doi:10.1115/ICMM2003-1004.

The purpose of the present paper is to present research and development within the area of mini- and micro channels in Sweden. A review is made of the historical development of highly compact heat exchangers within the country, starting with plate heat exchangers. The main focus is on the research performed at the Royal Institute of Technology, where mini-channel research has been going on since more than ten years. Single-phase flow as well as two-phase flow is treated, both in single channels and in full-size heat exchangers with multiple parallel channels.

Topics: Microchannels
Commentary by Dr. Valentin Fuster
2003;():33-46. doi:10.1115/ICMM2003-1005.

The forces due to surface tension, inertia, and momentum change during evaporation in microchannel govern the two-phase flow patterns and the heat transfer characteristics during flow boiling. These forces are analyzed in this paper, and two new non-dimensional groups, K1 and K2 , relevant to flow boiling phenomenon are derived. These groups are able to represent some of the key flow boiling characteristics, including the CHF. The small hydraulic dimensions of microchannel flow passages present a large frictional pressure drop in single-phase and two-phase flows. In order to keep the pressure drop within limits, the channel lengths are generally shorter and the mass fluxes are generally lower than those with conventional channels (Dh >3 mm). The resulting lower mass fluxes, coupled with small Dh , lead to Reynolds numbers in the range 100–1000. Such low Reynolds numbers are rarely employed for flow boiling in conventional channels. In these low Reynolds number flows, nucleate boiling systematically emerges as the dominant mode of heat transfer. Aided by strong evaporation rates, the bubbles nucleating on the wall grow quickly and fill the entire channel. The contact line between the bubble base and the channel wall surface now becomes the entire perimeter at both ends of the vapor slug. Evaporation occurs at the moving contact line of the expanding vapor slug as well as over the channel wall covered with a thin liquid film surrounding the vapor core. The usual nucleate boiling heat transfer mechanisms, including liquid film evaporation and transient heat conduction in the liquid adjacent to the contact line region, play an important role. The liquid film under the large vapor slug evaporates completely at downstream locations thus presenting a dryout condition periodically with the passage of each large vapor slug. The flow boiling correlation by Kandlikar [1, 2] with (i) the nucleate boiling dominant region equation, and (ii) the laminar flow equation for single-phase all-liquid flow heat transfer coefficient hLO was successful in correlating the available R-134a data for parallel microchannels of 190 μm hydraulic diameter.

Commentary by Dr. Valentin Fuster
2003;():47-58. doi:10.1115/ICMM2003-1006.

This paper reviews recent Korean studies of flow characteristics, flow boiling, and flow condensation in micro- and mini-channels. The characteristics of local heat transfer and pressure drops were experimentally investigated using condensing R134a two-phase flow, in a single round tube, with an inner diameter of 0.691 mm. New experimental techniques were developed to measure the condensation heat transfer coefficient. Tests were performed for a mass flux of 100 to 600 kg/m2 s, a heat flux of 5 to 20 kW/m2 , and a saturation temperature of 40°C. The experimental local condensation heat transfer coefficients and two-phase frictional pressure gradients are shown. Comparisons of experimental data with existing models reveal that the correlations failed to predict the present data. This study contains the unique sub-millimeter-diameter, single round tube, condensation data reported in the literature.

Commentary by Dr. Valentin Fuster
2003;():59-67. doi:10.1115/ICMM2003-1007.

Experiments were performed with clear water and with surfactant flowing in parallel triangular micro-channels. The study is based on systematic measurements of temperature and flow pattern by infrared radiometry and high-speed digital video imaging. Different flow patterns were observed simultaneously in various micro-channels at a fixed value of water or surfactant flow rates. Depending on flow and heat flux, pressure and temperature instabilities in the heated micro-channels were studied. This work develops a practical modeling approach for two-phase micro-channel heat sinks and considers also effect of surfactant on convective boiling in micro-channels.

Commentary by Dr. Valentin Fuster
2003;():69-75. doi:10.1115/ICMM2003-1008.

High performance heat pipes are widely used for thermal control of electronic devices. Concerning heat transport limitations typical wick or capillary structures show advantages in some aspects and disadvantages in others. An advanced capillary structure was developed with high thermal effectiveness, low axial pressure drop, high capillary pressure, and a high boiling limit. It combines open minichannels with open microchannels that are manufactured perpendicular on top of the minichannels. The heat transfer coefficient in the evaporator zone which is a characteristic value for the thermal effectiveness was up to 3.3 times higher compared to a similar structure without microchannels. A model that combines micro- and macroscopic phenomena was developed. It predicts the heat transfer coefficient with quite good accuracy as long as the microchannels are not smaller than about 300μm.

Topics: Heat pipes
Commentary by Dr. Valentin Fuster
2003;():77-92. doi:10.1115/ICMM2003-1009.

The physical mechanisms for the size effects on the flow and heat transfer have been divided into two classifications: (a) variations of the predominant factors influence the relative importance of various phenomena in the flow and heat transfer as the characteristic length decrease, even if the continuum assumption is still valid; (b) the continuum assumption breaks down as the characteristic length of the flow becomes comparable to the mean free path of the molecules. The departure of most flow and thermal phenomena in the MEMS from conventional ones is due to the variation of predominant factors in the flow and heat transfer problems, rather than that Navier-Stokes equation and Fourier heat conduction equation etc are no longer valid. Due to the larger surface to volume ratio for microchannels and microdevices, factors related to surface effects have more impact to microscale flow and heat transfer. For example, surface friction induced flow compressibility makes the fluid velocity profiles flatter and leads to higher friction factors and Nusselt numbers; surface roughness is likely responsible for the increased friction factor, the early transition from laminar to turbulent flow and Nusselt number; and other effects, such as the axial heat conduction in the channel wall, the channel surface geometry, surface electrostatic charges, and measurement errors, could lead to different flow and heat transfer behaviors from that at conventional scales. The condensation/evaporation across the liquid-vapor interface and the liquid-vapor nucleation are processes at nanometer scale. In these cases the macroscopic approach is hard to reveal the details at nanometer scales, while the molecular dynamic simulation is a powerful tool to describe such microscopic processes, and has been applied to investigate the density and temperature profiles across the liquid-vapor interface. The condensation coefficient on the liquid-vapor interface under thermodynamic equilibrium condition has been well predicted based on the characteristic time method, which can distinguish the condensed particles from the reflected particles. Molecular dynamics simulations show that liquid-vapor nucleation undergoes three stages, namely, cavity growth, cavity coalescence and bubble formation.

Commentary by Dr. Valentin Fuster
2003;():93-102. doi:10.1115/ICMM2003-1010.

Carbon dioxide (CO2 / R-744) is receiving renewed interest as a refrigerant, in many cases based on systems with microchannel heat exchangers that have high pressure capability, efficient heat transfer, and compact design. A good understanding of two-phase flow of evaporating CO2 in microchannels is needed to analyze and predict heat transfer. A special test rig was built in order to observe two-phase flow patterns, using a horizontal quartz glass tube with ID 0.98 mm, externally coated by a transparent resistive film. Heat flux was obtained by applying DC power to the film, and flow patterns were recorded at 4000 or 8000 frames per second by a digital video camera. Flow patterns were recorded for temperatures 20°C and 0°C, and for mass flux ranging from 100 to 580 kgm−2 s−1 . The observations showed a dominance of intermittent (slug) flow at low x, and wavy annular flow with entrainment of droplets at higher x. At high mass flux, the annular/entrained flow pattern could be described as dispersed. The aggravated dryout problem reported from heat transfer experiments at high mass flux could be explained by increased entrainment. Stratified flow was not observed in the tests with heat load. Bubble formation and growth could be observed in the liquid film, and the presence of bubbles gave differences in flow pattern compared to adiabatic flow. The flow pattern observations did not fit generalized maps or transition lines showed in the literature.

Commentary by Dr. Valentin Fuster
2003;():103-108. doi:10.1115/ICMM2003-1011.

Adhesive interactions between white blood cells and the interior surface of the blood vessels they contact is important in inflammation and in the progression of heart disease. Parallel-plate microchannels have been useful in characterizing the strength of these interactions, in conditions that are much simplified over the complex environment these cells experience in the body. Recent computational and experimental work by several laboratories have attempted to bridge this gap between behavior observed in flow chamber experiments, and cell-surface interactions observed in the microvessels of anesthetized animals.

Commentary by Dr. Valentin Fuster
2003;():109-114. doi:10.1115/ICMM2003-1012.

Intermolecular forces in liquids — van der Waals, electrostatic, steric — range over a distance of about 1–100 nm. Therefore, when liquids are confined in regions less than 100 nm, their behavior can be different and manipulated in unique ways. Such situations occur in nanoscale channels as well as in dense collection of biomolecules such as DNA and proteins, which range in size of 1–10 nm. Understanding fundamental transport properties of liquids under nanoscale confinement can have significant impact in biotechnology.

Commentary by Dr. Valentin Fuster
2003;():115-127. doi:10.1115/ICMM2003-1013.

Microfluidics is a rapidly developing area of research with great potential for a wide range of applications in many fields. One area of microfluidics is gas-liquid two-phase flow in microchannels, which is important for the development of microreactors, lab-on-a-chip systems, micro heat exchangers and micro-heat pipes, among others, that are highly relevant to industry. Recently, much interest has also been shown toward studying the two-phase flow in micro fuel cells. This keynote paper presents a state-of-the-art review of past and present research on adiabatic two-phase flow in minichannels and microchannels, which are considered to have channel diameters between 250 μm–6 mm, and less than 250 μm, respectively. From this review, certain differences between minichannels and microchannels are identified. These notable differences are also explained, based on some of our recent experiments on two-phase flow in microchannels. Our experiments have been performed using several microchannels to determine the effects of the microchannel diameter and shape on the adiabatic two-phase flow of nitrogen gas and de-ionized water. The effect of channel geometry was examined by characterizing the two-phase flow in a circular and square microchannel of similar hydraulic diameter. A video camera was used to capture images of the gas-liquid interfacial structure. From the video recordings, it became clear that the channel size strongly influences the two-phase flow patterns occurring in the circular microchannel. The flow pattern was predominantly intermittent, exhibiting alternating sequences of liquid and gas slugs. Only slug flow was observed in the microchannel for all flow conditions tested. There were no instances of bubbly flow, churn flow, slug-annular or annular flow, as reported for minichannels. Instead, four new sub-classes of slug flow were defined to better describe the interfacial structure in the time average sense: slug-ring flow, ring-slug flow, semi-annular flow and multiple flow. The time-averaged void fraction was estimated from the recorded images of the two-phase flow structure. It was found that as the channel diameter decreased, the void fraction data deviated more from those obtained for minichannels. A new void fraction correlation was developed for both the circular and square microchannels, which differs significantly from those developed for minichannels. In both microchannels, the two-phase pressure drop was best predicted by treating the two phases as being non-homogeneous and having a large velocity difference. This result was consistent with the occurrence of slug flow and significant departure of the average void fraction from those in minichannels. A possible explanation for the strong deviation of void fraction data in microchannels from the correlations applicable to minichannels is offered based on a phenomenological examination of the flow structure. Regarding the effect of microchannel geometry, the experimental results showed little difference in the void fraction and pressure drop data. However, the two-phase flow regime maps were not the same between the circular and square microchannels. The transition boundaries of the sub-categories of slug flow were noticeably shifted. The region of ring-slug flow in the circular microchannel disappeared in the square microchannel, which can be attributed to the suppression of the liquid-ring film due to the accumulation of liquid in the corners of the square microchannel.

Commentary by Dr. Valentin Fuster
2003;():129-140. doi:10.1115/ICMM2003-1014.

The present review article focuses on the research field of heat transfer of single-phase laminar-flow and two-phase self-exciting oscillating-flow in microchannels. First, to make prominent the special features of Micro Thermal Systems (MTSs), the definition of the term “Nano Thermal Systems” (NTSs) is discussed from the viewpoint of local equilibrium. Next, to show the special features of flow and heat transfer in microchannels, some thermal functions appearing in microchannels are introduced. Further, focusing on flow and heat transfer characteristics of single-phase laminar liquid-flow in microchannels, researches in the literature and recent results at IIS (Institute of Industrial Science, the University of Tokyo) are introduced, and it is shown that the results obtained for tubes larger than 0.1mm in inner diameter are in good agreement with the conventional analyses. Finally, Japanese researches and recent results at IIS on micro SEMOS heat pipes (mSEMOSs) are introduced and it is shown that a mSEMOS of 0.5mm in inner diameter can transport a significant amount of heat.

Commentary by Dr. Valentin Fuster
2003;():141-147. doi:10.1115/ICMM2003-1016.

Critical heat flux limit is an important consideration in the design of any flow boiling unit. Before the use of microchannels under flow boiling conditions becomes widely accepted in critical applications, such as electronics cooling and laser lenses, it is essential to develop CHF data for microchannels. The experiments required to obtain this information pose unique challenges as the channel dimensions become smaller. The issues of parallel channel instability, experimental control, experimental uncertainty, and conjugate effects need to be carefully addressed. These issues are addressed in the present paper, and guidelines helpful in the design of CHF experiments are outlined.

Commentary by Dr. Valentin Fuster
2003;():149-157. doi:10.1115/ICMM2003-1017.

Recent work on transport phenomena in microsystems, carried out at the Hong Kong University of Science and Technology (HKUST) during the past few years, is summarized in this paper. This includes convective and boiling heat transfer in silicon microchannels with application to electronic cooling; periodic generation of micro bubbles as a thermal actuator for enhanced mixing in biosensors; and fractal tree-like microchannel nets for electronic cooling and for the fuel distribution in the current collector of a micro fuel cell.

Commentary by Dr. Valentin Fuster
2003;():159-169. doi:10.1115/ICMM2003-1018.

Microchannel heat sinks are widely regarded as being amongst the most effective heat removal techniques from space-constrained electronic devices. However, the fluid flow and heat transfer in microchannels is not fully understood. The pumping requirements for flow through microchannels are also very high and none of the micropumps in the literature are truly suitable for this application. A wide-ranging research program on microchannel heat sinks and micropumps is underway in the Electronics Cooling Laboratory at Purdue University. This article provides an overview of the research being conducted to understand fluid flow and heat transfer in microchannels and to identify pumping requirements and suitable mechanisms for pumping in microchannels.

Commentary by Dr. Valentin Fuster
2003;():171-179. doi:10.1115/ICMM2003-1019.

The objective of the present paper is to provide a general overview of the research carried out so far in single-phase heat transfer and flow in capillary (micro) pipes. Laminar flow and laminar-to-turbulent flow transition are analyzed in detail in order to clarify the discrepancies among the results obtained by different researchers. Experiments performed in the ENEA laboratory indicate that in laminar flow regime the friction factor is in good agreement with the Hagen-Poiseuille theory for Reynolds number below 600–800. For higher values of Reynolds number, experimental data depart from the Hagen-Poiseuille law to the side of higher f values. The transition from laminar-to-turbulent flow occurs for Reynolds number in the range 1800–2500. Heat transfer experiments show that heat transfer correlations in laminar and turbulent regimes, developed for conventional (macro) tubes, are not properly adequate for heat transfer rate prediction in microtubes.

Commentary by Dr. Valentin Fuster
2003;():181-192. doi:10.1115/ICMM2003-1020.

This paper presents an overview of the use of flow visualization in micro- and mini-channel geometries for the development of pressure drop and heat transfer models during condensation of refrigerants. Condensation flow mechanisms for round, square and rectangular tubes with hydraulic diameters in the range 1–5 mm for 0 < x < 1 and 150 kg/m2 -s and 750 kg/m2 -s were recorded using unique experimental techniques that permit flow visualization during the condensation process. The effect of channel shape and miniaturization on the flow regime transitions was documented. The flow mechanisms were categorized into four different flow regimes: intermittent flow, wavy flow, annular flow, and dispersed flow. These flow regimes were further subdivided into several flow patterns within each regime. It was observed that the intermittent and annular flow regimes become larger as the tube hydraulic diameter is decreased, at the expense of the wavy flow regime. These maps and transition lines can be used to predict the flow regime or pattern that will be established for a given mass flux, quality and tube geometry. These observed flow mechanisms, together with pressure drop measurements, are being used to develop experimentally validated models for pressure drop during condensation in each of these flow regimes for a variety of circular and noncircular channels with 0.4 < Dh < 5 mm. These flow regime-based models yield substantially better pressure drop predictions than the traditionally used correlations that are primarily based on air-water flows for large diameter tubes. Condensation heat transfer coefficients were also measured using a unique thermal amplification technique that simultaneously allows for accurate measurement of the low heat transfer rates over small increments of refrigerant quality and high heat transfer coefficients characteristic of microchannels. Models for these measured heat transfer coefficients are being developed using the documented flow mechanisms and the corresponding pressure drop models as the basis.

Commentary by Dr. Valentin Fuster
2003;():193-205. doi:10.1115/ICMM2003-1021.

In the present paper, the local characteristics of pressure drop and heat transfer are investigated experimentally for the condensation of pure refrigerant R134a in four kinds of multi-port extruded aluminum tubes of about 1 mm in hydraulic diameter. Two tubes are composed of plane rectangular channels, while remaining two tubes are composed of rectangular channels with straight micro-fins. The experimental data of frictional pressure drop (FPD) and heat transfer coefficient (HTC) in plane tubes are compared with previous correlations, most of which are proposed for the condensation of pure refrigerant in a relatively large diameter tube. It is confirmed that parameters such as tube diameter, surface tension, free convection in FPD and HTC correlations should be taken into account more precisely. Considering the effects of surface tension and kinematic viscosity, new correlation of FPD is developed based on the Mishima-Hibiki correlation. New correlation of HTC is also developed modifying the effect of diameter in the correlation of Haraguchi et al. Both new correlations are compared with experimental data for tubes with micro-fins. Satisfactory agreement between experimental and predicted results is obtained. This means that the micro-fin effect is taken into account by using hydraulic diameter and the heat transfer enhancement effect of micro-fins is mainly due to the enlargement of heat transfer area.

Commentary by Dr. Valentin Fuster
2003;():207-214. doi:10.1115/ICMM2003-1022.

Heat exchangers with microchannels (hydraulic diameter less than 1mm) are attracting significant attention lately. Performance of these heat exchangers, evaporators mostly, are greatly affected by imperfect distribution of a two-phase fluid to each channel. This paper presents overview of the issue, shows basic option to evenly distribute single and two-phase fluid, and methods to determine maldistribution. Special attention is given to describing two-phase regimes in an adiabatic horizontal header with R134a as a fluid, describing a two-phase map with area of good distribution.

Commentary by Dr. Valentin Fuster
2003;():215-222. doi:10.1115/ICMM2003-1023.

Microchannel devices with channel widths in the range from one micron to several hundred microns have become increasingly important structures for heat transfer applications. This paper will examine several classes of fabrication technologies that are employed in the field. Established technologies, such as bulk and surface micromachining, high aspect ratio machining, and conventional machining have been perfected over the last several decades and are the principal methods of creating microchannel structures. Synthesizing technologies, such as wafer bonding and micromolding techniques, allow these primary structures to be assembled together into working devices or enable high volume, low cost manufacturing using microstructured masters. Established and synthesizing technologies have reached a high level of performance and are generally the subject of refinement efforts rather than innovative investigations. Active research into new materials, processes, and fabrication strategies are found in the class of emerging technologies. This paper will briefly review the dimensional spectrum of microchannels and survey the established and synthesizing technologies. Then an exploration of emerging technologies will be made. The topic of rapid prototyping will be given particular emphasis.

Commentary by Dr. Valentin Fuster
2003;():223-229. doi:10.1115/ICMM2003-1024.

The developing flow field in a parallel plate microchannel, induced by wall motion, has been modeled numerically. The flow is driven in this scenario not by an applied pressure gradient, but by the movement of the walls in the axial direction at a constant speed. This type of flow simulates the physical driving mechanism that exists in electro-osmotically generated flow with large channel diameter-to-Debye length ratios. The results are general, however, for any microscale flow induced by wall motion and resulting viscous pumping. The dynamics of the developing flow field were explored for channel length-to-hydraulic diameter ratios (aspect ratio) of 5, 10, and 20 at ten Reynolds numbers, Re (based on the wall velocity), below Re < 2000. The results show that far from the inlet the maximum fluid velocity occurs at the walls, as is expected, and the minimum velocity occurs at the channel center. Near the channel inlet, however, the centerline velocity is not a minimum but reaches a local maximum due to a resulting pressure imbalance generated by the wall motion. The ratio of the centerline velocity to wall velocity depends on the axial distance from the channel inlet, the Reynolds number and the channel aspect ratio. As the aspect ratio increases, the centerline velocity tends to approach the wall velocity far downstream from the inlet. Increases in the Reynolds number have the opposite effect on the centerline velocity. The hydrodynamic developing region, defined by that section of the channel where the wall shear stress is changing, also depends on the channel aspect ratio and Re. In general it is found that the developing region is significantly shorter than for pressure-driven flow at the same Re.

Commentary by Dr. Valentin Fuster
2003;():231-239. doi:10.1115/ICMM2003-1025.

There is a need for increased understanding of the momentum transport phenomena in micro-fluidic geometries to aid in the design and optimization of such devices. Micro-molecular tagging velocimetry (μMTV) has been used to characterize the hydrodynamic developing flow in a microtube with an inner diameter of 180 μm. μMTV is a non-intrusive laser-based technique for obtaining detailed measurements of velocity profiles in flows dominated by a single velocity component. μMTV measurements are made by directing an ultra-violet laser beam into a flow containing phosphorescent tracer molecules. The laser beam excites a line of phosphorescence in the flow. Subsequently, two digital images, separated by a short time delay, of the line are captured by a CCD camera. The displacement of the tracer molecules between the images can be determined from the two images and the velocity of the flow is thus calculated. Velocity profile data at ten axial locations within the hydrodynamic developing region of a 180 μm diameter tube were acquired using the μMTV approach. The uncertainty for these measurements ranged from 1.5% to 5.5% of the center line velocity. Data were taken at Reynolds numbers, Re, of 60, 140, 290, and 340. It was observed that a vena-contracta existed in the first few tube diameters for all Re. The velocity profiles obtained very close to the tube entrance exhibited a uniform velocity core flow surrounded by regions of relatively stagnant fluid in the near wall regions. The profiles evolved in the downstream direction until the classical parabolic distribution was observed. The total hydrodynamic entry length agrees well with values published in the literature for macroscale flows, obtained from numerical simulation.

Topics: Flow (Dynamics)
Commentary by Dr. Valentin Fuster
2003;():241-248. doi:10.1115/ICMM2003-1026.

This paper presents the results of an experimental investigation of two-phase pressure loss of R134a in microchannel headers using various end-cut techniques. Novel experimental techniques and test sections were developed to enable the accurate determination of the minor losses without obfuscating the problem with a lengthwise pressure gradient. This technique represents a departure from approaches used by other investigators that have extrapolated minor losses from air-water experiments and the combined effects of expansion, contraction, deceleration, and lengthwise pressure gradients. Pressure losses were recorded over the entire range of qualities from 100% vapor to 100% liquid. In addition, the tests were conducted for five different refrigerant mass fluxes between 185 kg/m2 -s and 785 kg/m2 -s using two differnt end-cut techniques. More than 790 data points were recorded to obtain a comprehensive understanding of the effects of mass flux and quality on minor pressure losses. High accuracy instrumentation such as coriolis mass flowmeters, RTDs, pressure transducers, and real-time data analyses were used to ensure accuracy in the results. The results show that many of the commonly used correlations for estimating two-phase pressure losses significantly underpredict the pressure losses found in compact microchannel tube headers. Furthermore, the results show that the end-cut technique can substantially affect the pressure losses in microchannel headers. A new model for estimating the pressure loss in microchannel headers is presented and a comparison of the end-cut techniques on the minor losses is reported.

Commentary by Dr. Valentin Fuster
2003;():249-255. doi:10.1115/ICMM2003-1027.

In this study a newly designed microchannel is proposed. This design comprises periodically arranged simple blocks. In this configuration, the stirring is greatly enhanced at a certain parameter set. To characterize the flow field and the stirring effect both the numerical and experimental methods were employed. To obtain the velocity field, three-dimensional numerical computation to the Navier Stokes equations are performed by using a commercial code, FLUENT 6.0. The fluid-flow solutions are then cast into studying the characteristics of stirring with the aid of Lyapunov exponent. In this study the Lyapunov exponents are computed manually because the commercial code does not provide the corresponding option. In the experiment, flow visualization for the stirring effect is performed by using pure glycerin in one tank and glycerin mixed with a fluorescent dye in the other. The numerical results show that the particles’ trajectories in the microchannel heavily depend on the block arrangement. It was shown that the stirring is significantly enhanced at larger block-height and it reaches maximum when the height is 0.8 times the channel width. We also studied the effect of the block stagger angle, and it turns out that the stirring performance is the best at the block angle 45°.

Topics: Microchannels
Commentary by Dr. Valentin Fuster
2003;():257-267. doi:10.1115/ICMM2003-1028.

The current work focused on determining the friction factors in five rectangular channels with hydraulic diameters varying from 69.5 to 304.7 μm and with aspect ratios changing from 0.09 to 0.24. The single-channel test sections were carefully designed such that the friction factors could be determined accurately from the experimental data. R134a liquid and vapor were used as the testing fluids. During the experiments, the Reynolds numbers were varied between 112 and 9180. The measured friction factors were compared with the conventional correlations. The results support such a general agreement in the literature that the flow friction in microchannels may be different from the conventional results. However, a more important finding is that when the channel surface roughness was low, even for the smallest channel tested, both the laminar friction factor and the critical Reynolds number approach the conventional values. In the turbulent region, the surface roughness has great effect on the flow friction even for the smoothest channel tested (Ra/Dh = 0.14%).

Commentary by Dr. Valentin Fuster
2003;():269-273. doi:10.1115/ICMM2003-1029.

The presence of slip in small-scale gaseous flows leads to shear work and dissipation at the boundary. This effect has been neglected in recent studies investigating the effect of viscous dissipation on convective heat transfer in small scale channels. In this paper we illustrate the effect of shear work at solid boundaries in small-scale gaseous flows through the solution of the constant-wall-heat-flux problem in the slip-flow regime. We show that dissipation at the boundary scales with the Brinkman number similarly to viscous heat dissipation inside the channel, and increases with increasing Knudsen number. As a result, it is incorrect to neglect this effect when viscous heat generation needs to be considered. An analytical expression for the fully developed slip-flow Nusselt number under constant-wall-heat-flux conditions in the presence of viscous heat dissipation is presented. This expression is verified by direct Monte Carlo solutions of the Boltzmann equation. An expression for the skin friction coefficient under fully developed flow conditions for arbitrary Knudsen numbers is presented. Simple approximate expressions for the skin friction coefficient in the ranges 0 ≤ Kn ⪝ 0 4 . and 0.4 ⪝ Kn ⪝ 3 are also presented. These expressions are in agreement with direct Monte Carlo solutions of the Boltzmann equation.

Commentary by Dr. Valentin Fuster
2003;():275-280. doi:10.1115/ICMM2003-1030.

When the characteristic scale of a channel decreases, wave propagation is increasingly dominated by viscous effects. This was first realized by Lamb who predicted that when the channel size is small compared to the diffusion length based on the oscillation frequency, the gas inertia becomes negligible (the fluid motion is effectively quasi-steady) and the flow is isothermal. This parameter regime is referred to as the narrow channel regime. We take advantage of this observation to derive analytical results for wave propagation in small scale channels for arbitrary Knudsen numbers, since due to their small transverse dimensions micro and nanochannels satisfy the narrow channel requirement except at very high frequencies. In the slip-flow regime in particular where the equations of motion can be integrated analytically, we show that thermal effects are always negligible and that the long wavelength approximation is always satisfied for narrow channels. We also discuss how this theory can be extended beyond the narrow channel approximation, that is, to include the effects of inertia and heat conduction to first order. Our results are verified by direct Monte Carlo simulations of the Boltzmann equation.

Commentary by Dr. Valentin Fuster
2003;():281-289. doi:10.1115/ICMM2003-1031.

A numerical study is made on the fully-developed bifurcation structure and stability of the forced convection in a curved microchannel of square cross-section. Two symmetric and four asymmetric solution branches are found. Thus a rich solution structure is found with up to eleven solutions over certain ranges of governing parameters. This multiplicity is at least partially responsible for the large differences in the reported friction factors and heat transfer coefficients in the literature. Dynamic responses of the multiple solutions to finite random disturbances are examined by the direct transient computation. It is found that possible physically realizable fully-developed flows evolve, as the Dean number (or Reynolds number) increases, from a stable steady 2-cell state at lower Dean number to a temporal periodic oscillation state, another stable steady 2-cell state, a temporal intermittent oscillation, and a chaotic temporal oscillation.

Commentary by Dr. Valentin Fuster
2003;():291-297. doi:10.1115/ICMM2003-1032.

Forced convective heat transfer coefficients and friction factor for flow of water and FC-72 in microchannels with a rectangular cross section were measured. An integrated microsystem consisting of five microchannels on one side and a localized heater and seven polysilicon temperature sensors along the selected channels on the other side was fabricated by using a double side polished silicon wafer. For the microchannels tested, the friction factor constant C = f ReDh obtained are values between 35.7 and 81.9, which are close to the theoretical value of 57.0. The measured Nusselt number in the laminar regime tested could be correlated by a correlation, Nu = A ReDh 1.37 Pr1/3 where A is the value between 0.000 454 and 0.000 646.

Commentary by Dr. Valentin Fuster
2003;():299-306. doi:10.1115/ICMM2003-1033.

An experimental and theoretical study of low Reynolds number compressible gas flow in a micro channel is presented. Nitrogen gas was used. The channel was microfabricated on silicon wafers and were 50 μm deep, 200 μm wide and 24000 μm long. The Knudsen number ranged from 0.001 to 0.02. Pressure drop were measured at different mass flow rates in terms of Re and found in good agreement with those predicted by analytical solutions in which a 2-D continuous flow model with first slip boundary conditions are employed and solved by perturbation methods.

Commentary by Dr. Valentin Fuster
2003;():307-310. doi:10.1115/ICMM2003-1034.

The improved rates of heat transfer in microchannel gas flows are promising for the design and development of microfluidic systems. This research focuses on the flow characteristics of air in rectangular micro/minichannels at moderate velocities (∼100 m/sec). The 50.8 mm long channels vary from approximately 266 μm to 1090 μm in hydraulic diameter, and the aspect ratio ranges from 0.1 to 0.75. The value of Re ranged from 250 to 4300, with the intention of studying the transition to turbulence. The friction factor is found to be higher than predicted values for Re < 1400 and lower when Re > 1400 suggesting earlier transition to turbulence.

Commentary by Dr. Valentin Fuster
2003;():311-317. doi:10.1115/ICMM2003-1035.

Two-dimensional compressible momentum and energy equations are solved to obtain the heat transfer characteristics of gaseous flows in parallel-plate micro-channels. The numerical methodology is based on the Arbitrary-Lagrangian-Eulerian (ALE) method. The computations were performed for channels with adiabatic walls to obtain the adiabatic wall temperature. The channel height ranges from 10 to 100 μm and the channel length is fixed at 30 mm. The stagnation pressure varies from 1.1×105 to 4×106 Pa. The outlet pressure is fixed at the atmosphere. The computations were also performed for channels with isothermal walls. The aspect ratio of the channel height and length is 100 or 200. The bulk temperature is compared with that of the incompressible flow in the conventional sized parallel plate channel.

Commentary by Dr. Valentin Fuster
2003;():319-326. doi:10.1115/ICMM2003-1037.

This paper reports the results of an experimental investigation of fluid flow and single-phase heat transfer of water in stainless steel capillary tubes. Three tube diameters are tested: 172 μm, 290 μm and 520 μm, while the Reynolds number varying from 200 up to 6000. Fluid flow experimental results indicate that in laminar flow regime the friction factor is in good agreement with the Hagen-Poiseuille theory for Reynolds number below 800–1000. For higher values of Reynolds number, experimental data depart from the Hagen-Poiseuille law to the side of higher f values. The transition from laminar to turbulent regime occurs for Reynolds number in the range 1800–3000. This transition is found in good agreement with the well known flow transition for rough commercial tubes. Heat transfer experiments show that heat transfer correlations in laminar and turbulent regimes, developed for conventional size tubes, are not adequate for calculation of heat transfer coefficient in microtubes. In laminar flow the experimental values of heat transfer coefficient are generally higher than those calculated with the classical correlation, while in turbulent flow regime experimental data do not deviate significantly from classical heat transfer correlations. Deviation from classical heat transfer correlations increase as the channel diameter decrease.

Commentary by Dr. Valentin Fuster
2003;():327-333. doi:10.1115/ICMM2003-1038.

Single-phase convective heat transfer in microtubes was numerically studied with consideration on the heat conduction in the tube wall. It indicates that the Nusselt numbers of the fully developed laminar convective heat transfer in microtubes with convective boundary condition outside the tube vary from 3.66 to 4.36, which represent the conventional results for isothermal and constant heat flux boundaries respectively. The Nusselt number depends on the parameters of thermal conductivity ratio (k*), diameter ratio (D*), and Biot number. One-dimensional thermal resistance model could underestimate the Nusselt number if the axial heat conduction in the wall can not be ignored. Discrepancies between the experimental results for the Nusselt number based on 1-D model and the standard values might be misunderstood as being caused by novel phenomena at microscales.

Commentary by Dr. Valentin Fuster
2003;():335-342. doi:10.1115/ICMM2003-1039.

The flow characteristics of a network of parallel microchannels (hydraulic diameter: 110 μm) were investigated both experimentally and numerically in the present work. The cross-section of the micro-channels was triangular for further application to micro heat pipes. Measurements of the pressure drop across the microchannels network showed a dramatic increase of the pressure losses and a departure from the law of fully developed flow as soon as the Reynolds number of the flow exceeded about 10. Numerical computations of the flow were performed by using the classical laws of hydrodynamics in order to explain this surprizing result. They showed a good agreement with the experimental results, which suggests that there are no size effects at the length scale used in the experiments. Moreover, the mechanisms responsible of the large pressure drop in the high-range of Reynolds number are identified by the numerical analysis. They correspond to extra head losses due to separation in several parts of the test cell.

Commentary by Dr. Valentin Fuster
2003;():343-350. doi:10.1115/ICMM2003-1040.

A diffuse electric double layer (EDL) in microchannel flow created by the charged surface in contact with an electrolyte solution is characterized by the so-called Debye-Hückel screening length, which depends on the ionic strength of the solution. Usually, the electric double layer thickness, which is from several nanometers to a few hundreds nanometers, is small in comparison with the microchannel height of a few tens microns. Traditional computational fluid dynamics (CFD) methods for macroscopic hydrodynamic equations have difficulties in such complex fluid dynamics problems involving microscale surface interactions. In this paper, we employ a two-dimensional generalized lattice Boltzmann model in the presence of external forces on a rectangular grid with an arbitrary aspect ratio and nonuniform mesh grids. A modified Poisson-Boltzmann equation is applied to examine the adsorption of ions from solution to a charged surface and obtain the electrostatic potential and ion distribution. An example with electroviscous flow in microchannel is used to validate the prediction ability of the model proposed here. Excellent agreement with experimental results was found.

Commentary by Dr. Valentin Fuster
2003;():351-358. doi:10.1115/ICMM2003-1041.

Rectangular microchannel is the most popular shape to be widely used in Microelectromechanical devices. However, analytical solutions of flow in this shape of microchannels are seldom reported because of its two dimensional problem. Furthermore, microflows in microchannels are more difficult to describe. In this paper, we addressed analytically the problem of laminar electrokinetic slip flow in microchannels with rectangular cross-section subjected to a time-dependent pressure gradient and a time-dependent electric field. The analytical solution has been determined based on the Debye-Hückel approximation and its solution is useful to precisely control microflow with slip in rectangular cross-section microchannels.

Commentary by Dr. Valentin Fuster
2003;():359-364. doi:10.1115/ICMM2003-1042.

An analysis is presented for the problem of micropolar flows in the annulus between two steadily rotating concentric cylinders. The local effects arising from microstructure and intrinsic motion of the fluid element that will affect the flow motion are considered. Especially, the effects of non-zero values of micro-gyration vector on the wall boundary conditions are investigated, using micropolar fluid theory. Numerical results of velocity distribution of micropolar fluids are compared with the corresponding flow problems for a Newtonian fluid. Also, the results of the surface friction coefficient and the couple stress coefficient at the inner and outer surfaces are prepared with various values of fluid properties and flow conditions.

Commentary by Dr. Valentin Fuster
2003;():365-371. doi:10.1115/ICMM2003-1043.

We have investigated the flow and mass transport within an electroosmotically-pumped incompressible liquid through a meander microchannel system. We employ two-dimensional, time-dependent Finite Element simulations in conjunction with a matched asymptotic treatment of the electrical double layers. The electroosmotic pumping is realized for two idealized and two realistic electrical fields, while a pressure-driven flow is used for comparison. We focus on the aspects of the electroosmotic transport. We find for most of the electroosmotically-driven cases rather complex flow fields, involving recirculation regions. These recirculation regions in all cases increase dispersion. (i) The least dispersion is associated with a plug-type velocity profile, which is obtained for an idealized purely wall-tangential orientation of the electrical field. (ii, iii) We find that both, the idealized horizontal electrical field and the real electrical field between two vertical plates give considerably higher dispersion than the pressure-driven flow. Vertical plate electrodes, therefore, do not allow for a electrical field, which minimizes dispersion. (iv) The arrangement of two point electrodes at the in and out sections likewise proves to be no optimal means to reduce dispersion beyond the pressure-driven flow. Thus, meander geometries of channels, in general, cause severe problems if electroosmotic pumping needs to be achieved in combination with minimized dispersion.

Commentary by Dr. Valentin Fuster
2003;():373-380. doi:10.1115/ICMM2003-1044.

The present work’s originality lies in the evidence of a non negligible effect of the fluid ions’ and co-ions’ interaction with the wall surface in a microtube. This study is based on the EDL theory (Electrical Double Layer) which is developed here for a circular geometry. High electrical surface potentials are taken into account for the present study; they induce the nonlinearity of the problem’s main equation (Poisson-Boltzmann equation). The electrical field is determined, then the velocity profile and finally the Poiseuille number. We show that even with the EDL effect taken into account, the Poiseuille number does not depend on the mean velocity. Our model agrees with the experimental results for high surface potentials (> 25 mV). This is found by comparing with experiments previously carried out with microtubes ranging from 530 to 50 μm.

Commentary by Dr. Valentin Fuster
2003;():381-388. doi:10.1115/ICMM2003-1045.

The effect of the electric double layer (EDL) on the linear stability of Poiseuille planar channel flow is reported. It is shown that the EDL destabilises the linear modes, and that the critical Reynolds number decreases significantly when the thickness of the double layer becomes comparable with the height of the channel. The planar macro scale Poiseuille flow is metastable, and the inflexional EDL instability may further decrease the macro-transitional Reynolds number. There is a good correspondence between the estimated transitional Reynolds numbers and some experiments, showing that early transition is plausible in microchannels under some conditions.

Commentary by Dr. Valentin Fuster
2003;():389-396. doi:10.1115/ICMM2003-1046.

Isothermal gas flows in two-dimensional microchannels are investigated with the lattice Boltzmann method. The slip velocity on the solid boundaries can be obtained reasonably when bounce–back reflection is combined with specular reflection in a certain proportion. Behaviors in the microchannel flow including velocity distribution, nonlinear pressure drop, and average friction factor are examined. The pressure distribution, the average friction factors and the mass flow rates are compared with those predicted by Arkilic’s model and experimental data and the agreement is reasonably good. Furthermore, the effects of bounce-back proportion rb on the slip velocity are investigated and its value is chosen to be 0.7 to best match the data from Arkilic’s model and available experimental data.

Commentary by Dr. Valentin Fuster
2003;():397-404. doi:10.1115/ICMM2003-1047.

A Lattice Boltzmann method is employed to investigate the flow characteristics and the heat transfer phenomenon between two parallel plates separated by a micro-gap. A nine-velocity model and an internal energy distribution model are used to obtain the mass, momentum and temperature distributions. It is shown that for small Knudsen numbers (Kn), the current results are in good agreement with those obtained from the traditional Navier-Stokes equation with non-slip boundary conditions. As the value of Kn is increased, it is found that the non-slip condition may no longer be valid at the wall boundary and that the flow behavior changes to one of slip-flow. In slip flow regime, the present results is still in good agreement with slip-flow solution by Navier Stokes equations. The non-linear nature of the pressure and friction distribution for micro-channel flow is gieven. Finally, the current investigation presents a prediction of the temperature distribution for micro-channel flow under the imposed conditions of an isothermal boundary.

Commentary by Dr. Valentin Fuster
2003;():405-410. doi:10.1115/ICMM2003-1048.

Gas flows in micro-electro-mechanical systems (MEMS) owing to the small size of the systems possess a relatively large Knuden number and usually belong to the slip and transitional flow regimes. This paper employs three schemes, namely the direct simulation Monte Carlo (DSMC) method, information preservation (IP) method, and the lattice Boltzmann method (LBM), to simulation micro-channel flows at three Knudsen numbers (Kn) of 0.0194, 0.194 and 0.388. The present LBM results are in agreement with those given by Nie et al. (2002), whereas they significantly differ from the DSMC (and IP) results as Kn increases. This suggests that the present version of LBM is not feasible to simulate the micro-channel flows in transition regime.

Commentary by Dr. Valentin Fuster
2003;():411-417. doi:10.1115/ICMM2003-1049.

This paper investigates the Poiseuille flows for rectangular, regular hexagonal, and semicircular cross sections in transition regime using particle approaches, namely the direct simulation Monte Carlo (DSMC) method and the information preservation (IP) method. The DSMC and IP results compare well with each other, while the IP method is much more computationally efficient than the DSMC method. The mass flow rates given by IP and DSMC are in agreement with the BGK solutions of Hasegawa and Sone. For rectangular cross sections in the wide range of the width-to-height ratio, the simulation results of the mass flow rates and the velocity profiles along the midperpendicular line have been given for both methods to estimate the lateral wall influence. For the physical quantities, such as the mass flow rate, which are influenced by the whole field, the lateral wall influence must be considered even for width-to-height ratio as large as 10. And for the physical parameters, such as the maximum velocity and the velocity profile along the midperpendicular line, the lateral wall influence can be negligible if the width-to-height ratio is bigger than 5.

Commentary by Dr. Valentin Fuster
2003;():419-423. doi:10.1115/ICMM2003-1050.

This study investigates the rough interface influence on the thermal resistance across double-layered thin films using Non-Equilibrium Molecular Dynamics (NEMD) with Lennard-Jones potential. Layer A is solid argon with face-centered structure. Layer B is obtained by changing atomic mass only. When each contacting atomic plane from the two layers has the same kind of atoms a flat interface is formed. When all the atoms in the nearest atomic plane to the interface in layer A are replaced with atom B and all the atoms in the nearest atomic plane to the interface in layer B are substituted by atoms A “rough interface” is obtained. Under flat and interface conditions the temperature profile, vibration amplitude and structure factor are studied to exhibit the roughness effect. It is found that the rough interface can effectively reduce the normal thermal resistance caused by the mismatch of the two dissimilar materials. This result suggests a possibility to control the thermal conductivity of the double-layered structure by engineering its interface condition.

Commentary by Dr. Valentin Fuster
2003;():425-431. doi:10.1115/ICMM2003-1051.

The miniaturization of clinical analysis device by microfabrication technology has great impact on medical and biological fields. In the present work, a two dimensional model has been put forward to numerically analyzing the thermal cycling of the micro flow-through PCR chip, which is different from the conventional PCR instruments and micro chamber PCR chip. In the micro flow-through PCR chip, sample and reagent continuously flow through a microchannel in the chip with three different temperature regions (denaturation, annealing and extension) to realize the nucleic acid amplification. Two parameters, U and D, are adapted to describe the temperature uniformity and deviation from the target temperature of the sample and reagent. Effects of the microchip’s geometrical structure, materials, designed temperatures of the three temperature regions, flow rate of the samples and reagents and the thermal boundary conditions around the microchip on the thermal cycling of micro flow-through PCR chip were numerically studied. Based on the simulation results, the silicon-glass-bonding micro PCR chip is recommended, and the optimally designed temperatures of the heaters in the three regions are given. The applicability of silicon-glass-bonding micro flow-through PCR chip with denaturation region heated alone is examined.

Commentary by Dr. Valentin Fuster
2003;():433-440. doi:10.1115/ICMM2003-1052.

A precise analytical model for gaseous flows in microchannels is of great interest for various applications, as for example when these microchannels are parts of a complex fluidic microsystem. However, a decrease in the channel hydraulic diameter leads to an increase in the rarefaction effects. If the Knudsen number becomes higher than about 0.1, it is generally admitted that the Navier-Stokes equation, even with first-order slip flow boundary conditions, are no longer valid. In order to keep an analytical model for higher Knudsen numbers, a resolution of the Navier-Stokes equation with second-order boundary conditions has been proposed in rectangular microchannels. An experimental setup has been designed for the measurement of gaseous microflows under controlled temperature and pressure conditions. Data relative to nitrogen and helium flows through rectangular microchannels are presented and analyzed. The microchannels have been etched by DRIE in silicon and closed with Pyrex by anodic bounding. Their depths range from 4.5 to 0.5 μm, with aspect ratios from 1 to 9%. It is shown that for aspect ratios higher than 1%, a plane flow model is no longer accurate, and that the proposed rectangular model should be used. The different sources of uncertainty that could occur during the experiments are discussed. A method is proposed to eliminate the principal one, that is the uncertainty when measuring the dimensions of the microchannel cross-section. Theoretical and experimental mass flow rates are compared, and it is shown that in rectangular microchannels, the second-order model is valid up to about 0.25, whereas the first-order model is no longer accurate for Knudsen numbers higher than 0.05. The best fit has been found for a tangential momentum accommodation coefficient σ = 0.93 , both with helium and nitrogen. Perspectives of this work are also presented.

Commentary by Dr. Valentin Fuster
2003;():441-448. doi:10.1115/ICMM2003-1053.

This paper provides concise specifications where idealized Poiseuille flow is applicable in representing one-dimensional flow through wide, thin, rough microchannels subjected to prescribed pressures at the channel ends. Starting with the general (compressible) form of the Navier-Stokes equations, new expressions which discuss the effect of body forces on flow through thin channels are first presented, leading to upper and lower bounds on channel reference velocity where idealized Poiseuille flow dominates. These results are combined with previously published studies related to the predicability of flow through stochastically rough surfaces. An arbitrarily chosen microchannel model based loosely on a previously published experimental test setup is used as a sample application.

Commentary by Dr. Valentin Fuster
2003;():449-456. doi:10.1115/ICMM2003-1054.

Structured (in particular, micro- and minigrooved) wall surfaces may improve numerous industrial processes, including falling film evaporation, thin film evaporation in lean premixed prevaporized combustion technology (LPP), and spray and jet cooling. The advantages of such surfaces include the promotion of ultra-thin film evaporation at the apparent contact lines and the prevention of dry patches on hot surfaces. However, the behavior of thin film flow on structured surfaces has not yet been comprehensively studied. We derive a model describing the heat transfer in liquid film flowing down inclined micro- or minigrooved walls. The derived model accounts for peculiarities of the evaporation process in the vicinity of the liquid-vapor-solid contact line (“micro region”) and their effect on the overall heat transfer rate. It is shown that the effect of the micro region is to increase the overall heat transfer rate at the constant fluid flow rate. A long-wave stability analysis has been performed to quantify the effect of the capillary structure on the film stability properties. Sinusoidal and triangular longitudinal groove shapes have been considered. Two cases have been studied: (i) the film completely covers the wall structure; (ii) the film partly covers the wall structure. It is shown that the longitudinal grooves completely covered by the liquid have a stabilizing effect on the falling film flow. The performed analysis is a step towards modeling the wavy motion of the liquid film on grooved surfaces.

Commentary by Dr. Valentin Fuster
2003;():457-464. doi:10.1115/ICMM2003-1055.

A two-dimensional model of a steady laminar flow of liquid film and co-current gas flow in a plane channel is considered. It is supposed that the height of a channel is much less than its width. There is a local heat source on the bottom wall of the channel. An analytical solution for the temperature distribution problem in locally heated liquid film is obtained, when the velocity profile is linear. An analytical solution of the linearized equation for thermocapillary film surface deformation is found. A liquid bump caused by the thermocapillary effect in the region where thermal boundary layer reaches the film surface is obtained. Damped oscillations of the free surface may exist before the bump. This is obtained according to the solution of the problem in an inclined channel. It depends on the forces balance in the film. The defining criterion is found for this effect. The oscillations of free surface do not exist for horizontally located channel.

Commentary by Dr. Valentin Fuster
2003;():465-472. doi:10.1115/ICMM2003-1056.

An experimental investigation has been carried out on two-phase flow characteristics in a 10 μm circular gap. The inlet of the gap is formed by two circular surfaces with radius 10 mm. The pressure drop is measured between two points placed — respectively — at the inlet and at the outlet of the gap. The upper and lower plates of the test section are transparent. Two-phase flow patterns were determined by video recording. The single flows of liquid FC-72 and nitrogen gas, as well as two-phase flow are investigated. In two-phase flow experiments the gap is filled by liquid. Liquid is maintained continuously by surface tension in the gap and in the meniscus formed between two inlet circular surfaces. The pressure difference includes two components, surface tension component in the meniscus and viscous one. Instability appears at the entrance of the gas into the gap. At small flowrate the gas flows in the gap has the form of chains of bubbles. Part of investigation was done in microgravity during a Parabolic Flights. In order to better understand experimental results, some numerical simulations have been done both in two and three dimensions for the real geometrical configuration for one-phase flows only.

Commentary by Dr. Valentin Fuster
2003;():473-477. doi:10.1115/ICMM2003-1057.

An understanding the physics behind the flow transition when capillary forces are important is critical to the development of many technologies; both for terrestrial- and space-based applications. The flow transition of greatest concern is that between slug, or plug, flow and core-annular flow. The transition of plug flow to core-annular flow in a low-Bond number configuration is experimentally observed in a novel capillary configuration. The effect of variations in mass flow rate of each phase (gas and liquid) on the overall system dynamics is discussed for both flow regimes.

Commentary by Dr. Valentin Fuster
2003;():479-486. doi:10.1115/ICMM2003-1058.

Adiabatic experiments were conducted to measure pressure drop for single-phase liquid and gas-liquid two-phase flows through a circular microchannel with an internal diameter of 100 μm. In order to study the effects of liquid properties on the pressure drop, aqueous solutions of ethanol with different mass concentrations (4.8, 9.5, 49 and 100 wt%) in distilled water and distilled water were used as the working liquid, while nitrogen gas was used for the gas phase. The surface tension of the working liquid ranged from 0.023 N/m (100 wt% ethanol) to 0.072 N/m (water), and viscosity from 0.9 mPa·s (water) to 3.4 mPa·s (49 wt% ethanol aqueous solution). For the single-phase flow experiments, the friction factor data were obtained for each working liquid used, over a Reynolds number range of 2 < Re < 800. For the two-phase flow experiments, pressure drop data were collected over 0.2 < jG < 7 m/s for the superficial gas velocity and 0.1 < jL < 1 m/s for the superficial liquid velocity. For single-phase flows, friction factor data were shown to be in reasonable agreement with conventional theory. Furthermore, early transition from laminar to turbulent flow was not observed over the present experimental flow conditions. For two-phase flows, Lockhart & Martinelli’s correlation was found to be capable of predicting the present pressure drop data irrespective of the working liquid tested, if an appropriate constant needed in the correlation is adopted.

Commentary by Dr. Valentin Fuster
2003;():487-491. doi:10.1115/ICMM2003-1059.

A new set of experimental data of two-phase air-water flow patterns in a square micro-channel is presented. The channel has a cross-section of 1×1 mm2 and a length of 300 mm. The ranges of the gas and liquid superficial velocities are 0.1–10 m/s and 0.2–7 m/s, respectively. Bubble, bubbleslug transitional, slug, and frothy patterns are observed. The present data are compared with other experimental data reported in the literature, and a good agreement is obtained. It is also compared the present data with those obtained from reduced gravity experiments, in which the Bond number has the same order of magnitude. Some problems associated with the micro-scale modeling of microgravity two-phase flow are also discussed.

Commentary by Dr. Valentin Fuster
2003;():493-498. doi:10.1115/ICMM2003-1060.

This study provides a qualitatively visual observation of the two-phase flow patterns for air-water mixtures inside a 3 mm smooth tube with the presence of vertical return bend. The curvature ratio (2R/D) is 3.2 whereas the total mass flux is from 70 to 800 kg/m2 s. The flow can be either entering from the upper of the tube or from the lower tube. However, it is found that there is no great difference between those flow entering at the upper tube and that of the lower tube if the inlet mass flux and vapor quality is the same. For a mass flux of 70 kg/m2 s at a vapor quality to 0.009, as the flow is approaching the return bend, one can observe a fluctuating phenomenon at the tail of the long slug that leads to a liquid ripple around the periphery. When the air slug is trying to penetrate the preceding liquid in the return bend, the shape at the front of the air slug was sharpened. A further increase of the vapor quality to 0.05, the flow after the return bend was temporarily turned from stratified flow into the annular flow. At a higher mass flux of 300 kg/m2 s, unlike those flow pattern at 70 kg/m2 s, the increase of the vapor shear interacts with the centrifugal force and the accumulated liquid within the return bend forces the Taylor bubble to be completely disordered. There is no separating and re-merging phenomenon of the air slug for the slug flow pattern across the return bend even for a very low vapor quality of 0.001. This is quite different from those with larger diameter tube (Chen et al. 2002, Wang et al. 2003b, 2003c).

Topics: Two-phase flow
Commentary by Dr. Valentin Fuster
2003;():499-506. doi:10.1115/ICMM2003-1061.

In microchannel flow, gas-liquid interface behavior will be important for developing a wide range of microfluidic applications, especially in micro reactors. In this paper, we discuss some topics related to capillary action and two-phase fluid behavior in a microchannel. One of the topics is interface motion in the flow driven only by capillary action. We examined circular and rectangular microchannel with diameter of 50 μm and 100 μm × 67 μm, respectively. For the circular channel, experiments well agreed with the previous theory in the case of ethyl alcohol as the test liquid. The effects of inner surface condition are found to be critical for interface motion on a microscopic scale. We have extended our theory to a rectangular microchannel. We obtained the same formula of relation between non-dimensional time and interface position as that of the circular channel. We compared predictions with experimental results of a PDMS microchannel. They agreed qualitatively, but not quantitatively. The difference was considered to be caused by contact angle estimation.

Commentary by Dr. Valentin Fuster
2003;():507-511. doi:10.1115/ICMM2003-1062.

Two-phase pressure drops of CO2 are investigated in mini tubes with inner diameters of 2.0 and 0.98 mm and in microchannels with hydraulic diameters from 1.08 to 1.54 mm. For the mini tubes, the tests were conducted with a variation of mass flux from 500 to 3570 kg/m2 s, heat flux from 7 to 48 kW/m2 , while maintaining saturation temperatures at 0°C, 5°C and 10°C. For the microchannels, mass flux was varied from 100 to 400 kg/m2 s, and heat flux was altered from 5 to 20 kW/m2 . A direct heating method was used to provide heat into the refrigerants. The pressure drop of CO2 in mini tubes shows very similar trends with that in large diameter tubes. Although the microchannel has a small hydraulic diameter, two-phase effects on frictional pressure drop are significant. The Chisholm parameter of the Lockhart and Martinelli correlation is modified by considering diameter effects on the two-phase frictional multiplier.

Commentary by Dr. Valentin Fuster
2003;():513-518. doi:10.1115/ICMM2003-1063.

In this paper an experimental study was investigated for two-phase distribution in compact heat exchanger header. A test section was consisted of the horizontal header (circular tube: φ 5 mm × 80 mm) and 10 upward circular channels (φ 1.5 mm × 850 mm) using acrylic tube. Three different types of tube insertion depth were tested for the mass flux and inlet quality ranges of 50–200 kg/m2 s and 0.1–0.3, respectively. Air and water were used as the test fluids. The distribution of vapor and liquid is obtained by measurement of the total mass flow rate and the calculation of the quality. Two-phase flow pattern was observed, and pressure drop of each channel was measured. By adjusting the insertion depth of each channel a uniform liquid flow distribution through the each channel was able to solve the mal-distribution problem.

Commentary by Dr. Valentin Fuster
2003;():519-526. doi:10.1115/ICMM2003-1064.

In a single capillary, the frictional two-phase pressure drop in Taylor flow has been measured using various liquids, and a correlation to predict the friction factor has been developed. A carefully designed inlet section for the capillary allowed the independent variation of gas bubble and liquid slug length. Gas and liquid superficial velocities were varied in the range 0.04–0.3 m/s. If the slug length was lower than 10 times the capillary diameter, the frictional pressure drop in the liquid slug increased drastically from the single phase limit (f = 16/Re). The slug length dependence is caused by a larger contribution to the pressure drop of the end effects near the bubble caps. Increased pressure drop at the ends of the slug is caused by two separate effects: (1) near the bubbles the circulation inside the liquid slug induces extra friction, and (2) the difference in curvature of the gas-liquid interface at the front and at the rear of the bubble gives rise to extra pressure drop. The use of different liquids allowed the independent variation of the Reynolds number Re and the Capillary number Ca, and an expression for the frictional pressure drop as a function of Re, Ca and the slug length was developed. The results of this work allow the determination of slug length from pressure drop measurements in closed equipment where the slug length cannot otherwise be measured easily. The applicability of the pressure drop model to estimate mass transfer is demonstrated by combined pressure drop and gasliquid measurements in a monolith, which is essentially an array of capillary channels.

Commentary by Dr. Valentin Fuster
2003;():527-533. doi:10.1115/ICMM2003-1065.

The present study investigated mass flowrate distribution and phase separation of R-22 in multimicrochannel tubes under adiabatic condition. The test section consisted of inlet and outlet headers with the inner diameter of 19.4 mm and 15 parallel multi-microchannel tubes. Each microchannel tube had 8 rectangular ports with hydraulic diameter of 1.32 mm. The key experimental parameters were the orientation of the header (horizontal and vertical), flow direction of refrigerant into the inlet header (in-line, parallel and cross flow) and inlet quality (0.1, 0.2 and 0.3). The effect of inlet quality on the mass flowrate distribution and phase separation in the microchannel tubes was negligible. The effect of the orientation of the header on the flow mass flowrate distribution and phase separation was the largest among the test parameters. Horizontal header showed better mass flowrate distribution and phase separation characteristics than vertical header. Both parallel and cross flow conditions showed better mass flowrate distribution and phase separation than in-line flow condition.

Commentary by Dr. Valentin Fuster
2003;():535-542. doi:10.1115/ICMM2003-1066.

Air-liquid two-phase flow in a horizontal flat capillary rectangular channel has been studied to clarify the effects of concentration of surfactant solution on the flow phenomena, such as flow patterns, pressure drop, void fraction and so on. The concentrations of surfactant solution were 0, 10, 50 and 100 ppm and the surface tension of each solution was reduced to about 34mN/m from that of pure water of about 72mN/m. The dimension of the channel used was 10.0 mm × 1.0 mm. The drag reduction by mixing the surfactant was examined in both the single phase flow and the two-phase flow. The experimental data of two-phase frictional pressure drop and holdup were compared with the respective correlations which were previously proposed by the other researchers and the present authors. Finally, we proposed new correlations of two-phase frictional pressure drop and holdup in which the effect of surface tension and the aspect ratio of cross section of channel were taken into account.

Commentary by Dr. Valentin Fuster
2003;():543-550. doi:10.1115/ICMM2003-1067.

To elucidate details of the fascinating nonlinear phenomena of gas-liquid interfaces in micro- and minichannels, precise spatiotemporal knowledge of the interface in gas-liquid two-phase flows is essential. This paper presents a new method for measuring the interface of a liquid film in microchannels using a laser focus displacement meter (LFD). The purpose of the study was to clarify the effectiveness of the new method for obtaining detailed information concerning interface displacement, especially in the case of a thin liquid film, in micro- and mini-channels. In the test, water and nitrogen gas were used as working fluids. To prevent the tube wall signal from disturbing that of the gas-liquid interface, a fluorocarbon tube with water box was used; the refraction index of this device was the same as that for water. With this method, accurate measurements of the liquid film interface were achieved in real time, with a sensitivity of 0.2 μm and 1 kHz. The error caused by refraction of the laser beam passing through the acrylic water box and fluorocarbon tube was estimated theoretically and experimentally. The formulated theoretical equation can derive the real interface displacement using measured displacement in a fluorocarbon tube of 25 μm to 2.0 mm I.D. A preliminary test using fluorocarbon tubes of 1 and 2 mm I.D. showed that the corrected interface displacement calculated by the equation agreed with real displacement within a 1% margin of error. We made simultaneous measurements of the interface in fluorocarbon tubes of 0.5 and 1 mm I.D. using the LFD and a high-speed video camera with a microscope. These showed that the LFD could measure the interface of a liquid film with high spatial and temporal resolution during annular, slug, and piston flow regimes. The data also clarified the existence of a thin liquid film less than 1 μm in thickness in slug and annular flow regimes.

Commentary by Dr. Valentin Fuster
2003;():551-558. doi:10.1115/ICMM2003-1068.

To investigate the analogy between heat transfer and fluid friction of two-phase flow in minichannel, heat transfer coefficient, pressure drop and void fraction were simultaneously measured for vertical upward heated air-water two-phase flow in a 2.01 mm I.D. stainless steel tube. Void fraction was slightly different from Smith’s correlation and drift flux model, but the latter gave fairly good prediction when α>0.8. Frictional pressure losses were greater than the experimental results of Mishima-Hibiki for a 2.05 mm I.D. Pyrex glass tube and the correlation of Chisholm-Laird. Heat transfer coefficients agreed with our previous experimental results for a 8.03 mm I.D. tube. A theoretical calculation was performed for an annular liquid film flow model using experimental values of wall shear stress and liquid holdup, and satisfactory results were obtained except when the liquid flow rate was very low.

Commentary by Dr. Valentin Fuster
2003;():559-566. doi:10.1115/ICMM2003-1069.

The design, fabrication, and evaluation of forced convection boiling two-phase flow, micro scale heat exchanger are described. The micro heat exchanger consist of a heater, and 16 mm long multiple parallel triangular micro channels, with hydraulic diameter ranging from 50 μm to 200 μm. The system allowed simultaneously visualizing the flow regime, to measure the resistor temperature distribution, the pressure drop, and input power levels. Experiments were conducted using water with mass flow rate of 1–10 g /min and heat fluxes ranging from 10–60 W/cm2 in order to better understand the flow mechanism associated with micro scale forced convection boiling two-phase flow. The pressure drop, temperature, fluctuation and flow regimes map were obtained. The results are consistent with newly reported studies and show that flow fluctuation at that scale is exaggerated with respect to conventional size channels. A unique flow regime was observed and was named “rapid bubble growth”.

Commentary by Dr. Valentin Fuster
2003;():567-579. doi:10.1115/ICMM2003-1070.

The use of microchannels for advanced heat transfer applications has quickly become commonplace. They are found in automotive applications, fuel cells, and even electronics cooling. However, there are fundamental issues still unresolved with heat transfer and fluid mechanics and the application of microchannels. Researchers have reported microchannel data using very different hydraulic diameters, sometimes as much as 2 orders of magnitude. An experimental investigation of the heat transfer, pressure drop, and flow boiling in microchannels is performed. A new channel size classification has been developed based upon the manufacturing techniques as well as the underlying fluid mechanics and heat transfer theory. Six parallel channels with a hydraulic diameter of 207 micrometers is manufactured and tested. Flow boiling patterns have been observed in the channels. Observations suggest that the conventional flow boiling patterns also occur in microchannels. This suggests that there is no difference in the theory used for conventional channels. Therefore, a microchannel can be model in the conventional manor. Heat fluxes of up to 930 kW/m2 have been maintained in the microchannel. The local heat transfer coefficient and quality has been measured. The largest heat transfer coefficient achieved is 192 kW/m2 K. In addition, the highest quality achieved is 1.0. Dry-out was also observed during experimentation.

Commentary by Dr. Valentin Fuster
2003;():581-588. doi:10.1115/ICMM2003-1071.

In this paper, attempts were made to experimentally investigate the boiling incipience in a narrow rectangular vertical channel of 1 mm depth with an external 40 mm wide wall heated uniformly and others assumed quasiadiabatic. The “boiling front” location was determined from the temperature distribution of the heated wall obtained from liquid crystal thermography. Boiling incipience occurs when considerable rise in wall temperature above the saturation temperature takes place. Thus, boiling incipience is accompanied by “nucleation hysteresis”. The impact of various factors on the boiling incipience in microchannels, such as: pressure, the inlet liquid subcooling and flow velocity were investigated.

Commentary by Dr. Valentin Fuster
2003;():589-594. doi:10.1115/ICMM2003-1072.

We develop a mathematical model of a long vapor bubble in a micro-channel with given temperature distributions on the walls. We assume that the shape of the bubble is dominated by capillary forces everywhere except near the walls of the channel and use a lubrication-type analysis to find the local vapor-liquid interface shapes and mass fluxes near the walls. Both two- and three-dimensional steady-state solutions are found such that evaporation near the heated bottom is balanced by condensation in colder areas of the vapor-liquid interface. The total length in this steady regime is found from the integral mass balance and investigated as a function of heating conditions. Steady-state conditions can no longer be satisfied when the intensity of heating is above a certain level. In this regime the bubble is expanding. We investigate such expansion in the framework of a two-dimensional model in the limit of small capillary number.

Topics: Vapors , Bubbles , Modeling
Commentary by Dr. Valentin Fuster
2003;():595-602. doi:10.1115/ICMM2003-1074.

An experimental effort was undertaken to investigate the flow boiling of R-134a in minichannels with transverse ribs. The test channel consisted of two parallel plates; the lower plate was heated while the upper plate was adiabatic and fabricated from a clear material to facilitate the visualization and photography of the boiling process. A number of transverse ribs were deployed along the lower plate to form a meandering flow passage inside the minichannel. The ribs blocked 3/4 of the flow cross-section in the transverse direction and the hydraulic diameter of the channel was 1.93 mm. In these experiments, the range of test parameters were: mass flux 100 to 225 kg/m2 s, saturation temperature 15 to 30°C, heat flux 14 kW/m2 , and average vapor quality 0.12 to 0.86. With these parameters, the Convective and Boiling numbers ranged from 0.04 to 0.90 and 7.49 × 10−4 to 8.98 × 10−4 , respectively. Of a total of 60 data points recorded, 78% of the data fell in the convective boiling region, while the remaining 22% were in the nucleate boiling region. In an attempt to model and correlate the data, it was found that the Kandlikar’s correlation proposed for augmented tubes was a suitable model for the present minichannel. Using the least-squares procedure, this two-equation model was fit to a total of 60 experimental data, resulting in a correlation that predicted 58 data points to within ±10%.

Commentary by Dr. Valentin Fuster
2003;():603-608. doi:10.1115/ICMM2003-1075.

As microchannels are applied in flow boiling applications, it is becoming apparent that the Reynolds number based on all liquid flow could approach values below 100. The earlier work by Kandlikar and Steinke (2002, 2003) provided modifications to the Kandlikar correlation (1990, 1991) by extending the range of the correlation to all-liquid Reynolds numbers in the range 1000–3000. The present work utilizes the newly available data on flow boiling in microchannels that cover the all-liquid flow Reynolds number between 50–500. A new correlation is developed in this range that is able to predict the flow boiling heat transfer coefficient and its trends with quality, heat flux and mass flux accurately within less than 15 percent mean deviation. It is noted that the correlation simply accounts for the change of the flow boiling mechanism without incorporating any additional empirical constants. The heat transfer mechanism during flow boiling at such low Reynolds numbers is altered considerably indicating strong presence of nucleate boiling mode of heat transfer.

Commentary by Dr. Valentin Fuster
2003;():609-614. doi:10.1115/ICMM2003-1076.

With boiling flows in very smooth heated channels, in general, the bulk liquid is superheated before the bubble formation takes place at the wall surface. Thus the eruptive boiling occurs once the superheat requirement for nucleation is satisfied, and this can be a source of flow instability. In the present work, a preliminary consideration has been given to predict the limiting heat flux for stable boiling in microchannels. The minimum heat-flux value to avoid this eruptive boiling is inversely proportional to the square of the channel diameter, and becomes enormously large as the channel diameter decreases. The stable heat-flux limit also depends on the cavity size in the channel wall. A map showing stable nucleation criteria was given and discussions were made on the simplifying assumptions made for the analysis.

Commentary by Dr. Valentin Fuster
2003;():615-621. doi:10.1115/ICMM2003-1077.

The present work investigates experimentally the two-phase flow instability of convective boiling heat transfer in a system of two parallel microchannels at a given low mass flux and several different heating surface temperatures. The test section employed in this work is fabricated by silicon bulk micro machining and anodic bounding processes. Both channels are triangular with the channel-width of the topside of 100 ìm and hydraulic diameter of 51.7 ìm. The topside width of the central island between two channels is 20 ìm. The forced flow of de-ionized water is provided by a syringe pump. The two-phase flow visualization under boiling conditions is made possible by a high-speed digital video camera. The temperatures of inlet and outlet reservoirs, the inlet pressure as well as the pressure drop during the experiments are recorded and analyzed with flow visualization. The results of the study demonstrate clearly that two-phase flow instabilities with significant oscillations in two-phase flow properties can be developed in the system of double microchannels at high heating surface temperatures due to channel-to-channel interactions. Flow visualization confirms the presence of flow reversal during large amplitude oscillations. Such two-phase flow instabilities involving flow reversal should be of great concern for the design of a micro evaporator.

Commentary by Dr. Valentin Fuster
2003;():623-628. doi:10.1115/ICMM2003-1078.

The flow boiling heat transfer coefficient of R-22 in small hydraulic diameter tubes has been experimentally studied. Both brass and aluminum round tubes of 1.66 mm inside diameter are used for the test section. The ranges of the major experimental parameters are 300∼600 kg/m2 s of refrigerant mass flux, 10∼20 kW/m2 of the wall heat flux, 0.0∼0.9 of the inlet vapor quality. The experimental result showed that the flow boiling heat transfer coefficient in this small tubes are in the range of 2∼4 kW/m2 K and it varies only by heat flux, independent of mass flux and vapor quality. It is also observed that the heat transfer coefficients in the aluminum tube are up to 50% higher than in the brass tube.

Commentary by Dr. Valentin Fuster
2003;():629-633. doi:10.1115/ICMM2003-1079.

A simultaneous visualization and measurement study has been carried out to investigate flow boiling of water in the 8 parallel silicon microchannels heated from below. It is found that there are two large-amplitude/long-period oscillating boiling modes exist in microchannels depending on the amounts of heat flux and mass flux. When the outlet water temperature is at saturation temperature and the wall temperatures are superheated, while the inlet water temperature is still subcooled, a Liquid/Two-phase Alternating Flow (LTAF) mode appears in the microchannels. This LTAF mode disappears when the inlet temperatures reaches the saturation temperature. As the heat flux is further increased such that the outlet water is superheated while the inlet water temperature is oscillating between subcooled and saturation temperature, a Liquid/Two-phase/Vapor Alternating Flow (LTVAF) mode begins. During these two unstable boiling modes, there are large-amplitude and long-period oscillations of water and wall temperatures with respect to time. Bubbly flow as well as some peculiar two-phase flow pattern are observed during the two-phase flow periods of the two unstable modes in the microchannels.

Commentary by Dr. Valentin Fuster
2003;():635-642. doi:10.1115/ICMM2003-1080.

Two-phase pressure drop is experimentally examined in a flow boiling condition in a tube of diameter 1.45 mm using water in ranges of pressure from 10 to 100 kPa, mass flux from 18 to 152 kg/m2 s, heat flux from 13 to 646 kW/m2 , and exit quality from 0.02 to 0.77. Also, pressure drop in an adiabatic air-water two-phase flow is measured at atmospheric pressure using the same test section and mass flux ranges of liquid and gas as those in the flow boiling. Decreasing system pressure the pressure drop significantly increases at a given mass flux. Influence of vapor phase on the pressure drop is found to be large both in the adiabatic and the diabatic conditions. The frictional pressure drop correlation for the adiabatic two-phase flow is developed and applied to predict pressure drop in the flow boiling. But it cannot give satisfactory predictions. The Chisholm correlation calculating a two-phase pressure drop multiplier is modified to account the influence of vapor phase in a capillary tube and the modified correlation can predict the pressure drop in the flow boiling within an error of 20%.

Commentary by Dr. Valentin Fuster
2003;():643-650. doi:10.1115/ICMM2003-1081.

Fluid flow and heat transfer in micro-channel has received attention in developing electronic element cooling systems and in designing thermal elements of micro machines and refrigerators. In the present experiment, pressurized water in the chamber flushed out of a vertical capillary tube of 0.5mm in inner diameter and 1150mm in length. The outer surface of the tube was insulated. The chamber excess pressure was changed from 0.1 MPa to 0.6 MPa and also the inlet water temperature was widely varied. At low excess pressures, the flow is liquid single-phase flow but at high excess pressures, phase change (flushing) takes place inside and/or near the exit end of the tube and the flow becomes two-phase flow of oscillatory or intermittent. Mass flow rate as well as tube wall temperature (distribution and fluctuation) were measured, and the aspect of flushing at the tube exit end was observed using high-speed video movie. It is found that the flow regime can be classified into three (single-phase, two-phase and transient flow regimes) depending on the chamber excess pressure.

Commentary by Dr. Valentin Fuster
2003;():651-657. doi:10.1115/ICMM2003-1082.

The work is an experimental investigation of the evaporation process from a liquid meniscus formed in capillary tubes of various sizes (ranging from 200 to 900 μm). The results have been compared with those of a previous analytical work and show how the strong convection in the liquid phase is responsible for the discrepancy found. In the analytical prediction the evaporation process is sustained only by diffusion and in this model the meniscus position vs. time is a linear function of the tube size; instead, our experimental results show how this parameter is inversely correlated with the pore size. The surface roughness of the tubes was characterized and particular care has been devoted to the capillaries’ cleaning procedure from which wetting properties are strongly dependent. Marangoni convection prevails at tube sizes less than one millimeter in diameter as in the present case, while at larger sizes a coupling between Marangoni and Rayleigh convection is expected. The Marangoni roll of thoroidal shape in the liquid phase has been visualized and characterized using seeding particles. As pointed out clearly in the present study, Marangoni convection enhances the heat-mass transfer from a pore.

Topics: Convection
Commentary by Dr. Valentin Fuster
2003;():659-666. doi:10.1115/ICMM2003-1083.

Experimental investigations on heat transfer in tubular micro- or minichannel arrangements more often report on two-phase flow instabilities, pulsations or oscillations, which result in a remarkable influence on heat transfer efficiency. In order to explain the piston-like oscillations of the steam-plugs and water-slugs (-columns), the authors studied the somehow similar process which occurs in the worldwide known toy steam boat. Experiments have been performed which used a demonstration plant made of glass. By controlled electrical heating, high-speed video, pressure and local temperature measurements, the paths of energy have been disclosed. The results are as surprising as the effect of making gold from sand with respect to an equivalent axial heat-conductivity of the water-filled glass tube. Initiated by these results, an abstracting model is presented that analytically quantifies this new regenerating (oscillating and conducting) heat transfer mode e.g. concerning the combination of a heat recharging tube wall and an oscillating water column in a field of diminishing temperatures between the temperature of the boiler surface and the subcooled bulk water. By introducing these heat transfer details, the steam boat can give an answer, not only on frequency and amplitude of the oscillations, but on the steady state conditions for — or time-dependency of — the location of zero-crossing as well. Experimental results and model calculations are in good agreement and need no fitting factors. This is the base to discuss that process along with its physical parameters and compare it to the above mentioned observations in flow-boilers or pulsating heat pipes etc. which use microchannels or minichannels.

Commentary by Dr. Valentin Fuster
2003;():667-673. doi:10.1115/ICMM2003-1084.

Flow boiling in small-sized channels attracted extensive investigations in the past two decades due to special requirements for transfer of high heat fluxes from narrow spaces in various industrial applications. Experiments on various aspects of flow boiling in narrow channels were carried out and theoretical attempts were undertaken. But these investigations showed large differences, e.g. up till now the knowledge on the development of flow patterns in small non-circular flow passages is very limited. This paper deals with investigations on flow boiling of water in two rectangular channels with dimensions (width×depth) 2.0×4.0 mm2 and 0.5×2.0 mm2 (corresponding hydraulic diameters are 2.67 mm and 0.8 mm). The pressure at the test section exit is atmospheric. For steady-state experimental conditions the effects of heat flux, mass flux and inlet subcooling on the boiling heat transfer coefficient and the pressure drop are investigated. Flow patterns and the transition of flow patterns along the channel axis are visualized and documented with a video-camera. Bubbly flow, slug flow and annular flow are distinguished in both channels. Preliminary flow pattern maps are generated.

Commentary by Dr. Valentin Fuster
2003;():675-682. doi:10.1115/ICMM2003-1085.

The present paper deals with two-phase flow pressure drop modeling. This is based on solving the mass, momentum and energy balance equations in steady state conditions. In the two-phase zone, the liquid-vapor is assumed to be a homogeneous fluid. The calculated pressure drop variation is found to be similar to the experimental one when the two–phase flow is steady. In the unsteady state conditions characterized by high amplitude of the pressure fluctuations, the computed pressure drop is found to be different from the experimental mean pressure drop. This difference is all the higher as the pressure fluctuation is high.

Commentary by Dr. Valentin Fuster
2003;():683-689. doi:10.1115/ICMM2003-1087.

When boiling or condensation occurs inside very small and non-circular channels, capillary forces influence two-phase flow patterns, which in turn determine heat transfer coefficients and pressure drop. A better understanding of the underlying phenomena would be beneficial from the perspective of optimizing the design of compact evaporators and condensers. The thrust of this study was to understand the nature of up-flow boiling and condensation heat transfer in channels with a small gap. It consisted of two parts. The first part included observation of two-phase flow patterns with refrigerant R21 in a test section containing plain fins. The shape of the channels formed between fins was close to rectangular. The test section was placed in a closed refrigerant loop, and it was fabricated with a transparent wall to allow observation of the flow. An electrically heated coil was used to introduce liquid and vapor at the needed quality into the test section. Regimes of slug, froth, annular and cell flow patterns were recognized and the areas of flow pattern were determined. The second part included up-flow boiling and condensation heat transfer measurement with refrigerant R21 in a set of vertical mini-channels consisting of plain fins. An aluminum fin pad was bonded to two dividing aluminum sheets by dip brazing. Heat was supplied to the test section from a thermoelectric module, which utilized the Peltier effect. A thick copper plate was placed between the dividing sheet on each side of the fin passage and the respective Peltier module to establish a uniform wall temperature. Heat transfer coefficient measurements were done under forced flow conditions. Data are obtained for mass flow rates of 30 and 50 kg/m2 s under both boiling and condensation modes with wall superheats ranging from 1 to 5K. The dependence of heat transfer coefficient from wall superheat was not observed both for boiling and condensing modes. It shows the primary role of evaporation from thin films in a confined space when the mass flux is small. At low vapor quality the boiling heat transfer coefficients are considerably higher than that for condensation. A high heat flux in ultra thin liquid film area near the channel corner or in the vicinity of liquid-vapor-solid contact line (after the film rupture) supports the high total heat transfer coefficient in evaporation mode. In contrast with evaporation mode, at upflow condensation mode the heat transfer coefficient is strongly dependent on vapor quality. At plug flow regime the vapor velocity determines the condensing heat transfer.

Topics: Condensation , Boiling
Commentary by Dr. Valentin Fuster
2003;():691-698. doi:10.1115/ICMM2003-1088.

Very little experimental information is available in the open literature about condensation inside minichannels. Most of the experimental work has been carried out by using the Wilson plot technique. This method is simple to implement because it does not require the direct measurement of the tube wall temperature. However it becomes inaccurate when a small thermal resistance is present on the test side as compared to the opposite (cooling) side, which is actually the case with a multichannel tube at high values of the internal heat transfer coefficient. In fact, in a multi-port tube internal webs work as fins, and their efficiency is close to unity; thus the internal heat transfer area is higher than the external one. In this paper a new technique to measure the heat transfer coefficient during condensation inside a multi-port extruded minichannel tube is presented. Some R134a preliminary data is also reported.

Commentary by Dr. Valentin Fuster
2003;():699-705. doi:10.1115/ICMM2003-1089.

Heat exchange has been successfully integrated with microchannel phase separation concepts to produce devices capable of simultaneous partial condensation and phase separation in reduced gravity. An air-cooled microchannel condenser has been tested on NASA’s KC-135 reduced gravity aircraft. The condenser was fed a mixture of air and water vapor at 70–95°C, which was cooled to below 40°C thereby generating water condensate. The condensate was successfully collected and removed as a separate stream over a range of operating conditions, thereby achieving simultaneous condensation and phase separation. Ambient air was used to cool in cross-flow with inches of water pressure drop. The microchannel device is presented along with an explanation of the principles of operation. Phase separation effectiveness and heat exchanger performance are reported for reduced gravity testing. Heat fluxes, effectiveness, and overall heat transfer coefficients are reported.

Commentary by Dr. Valentin Fuster
2003;():707-712. doi:10.1115/ICMM2003-1091.

An experimental study of complete convective condensation inside narrow channels is presented in this paper. Two-phase flows patterns and their transition (annular, annular-wavy, slug and bubbly flow) are visualized for the two tube diameters under study. A significant difference is observed for the two sizes of tube. Experimental results of the bubble radius decrease are then determined and compared to a model of bubble collapse in a subcooled and infinite liquid.

Topics: Condensation
Commentary by Dr. Valentin Fuster
2003;():713-720. doi:10.1115/ICMM2003-1092.

Analytical and numerical tools for modeling microchannel heat sinks are presented, including the most important project parameters. An analytical 1D model, suitable for both design and performance calculations, is first developed under lumped capacitance assumption and resorting to standard correlations for Nu under H1 boundary conditions. 2D model extensions are then provided separately for poorly and highly conductive substrates: in the former case axial conduction is neglected; in the latter case the 2D model is modified to include axial conduction by exploiting the results from 1D model. This innovative approach provides a significant reduction in computational costs with respect to a conventional 3D model. A 3D model is then developed to evaluate the effects of thermal entrance. Finally, some optimization calculations are performed.

Commentary by Dr. Valentin Fuster
2003;():721-730. doi:10.1115/ICMM2003-1093.

The paper proposes a new design of a scalable, heat sink containing 3-D micro/nano network, utilizing liquid mixed with nano phase change materials (NPCM) and having a high surface-to-volume ratio geometry. The conceptual design is capable of reaching 105 W/cm3 using encapsulated nano-size phase change materials, which would result in an order of magnitude higher cooling capacity than typical microchannel heat sink of the same volume and same pumping power. It is also scalable to submicron range, resulting even higher cooling capacity. An analysis for a working model (10 × 10 × 1 mm) is presented utilizing energy conservation principle and uniform temperature and uniform heat flux boundary conditions. The average phase change heat transfer coefficient is obtained using the numerical model results. A process of micro electrochemical deposition to fabricate the target model is illustrated, and the issues associated with system-level applications are discussed.

Topics: Design , Heat sinks
Commentary by Dr. Valentin Fuster
2003;():731-738. doi:10.1115/ICMM2003-1095.

The pressure drop and heat transfer characteristics of heat sinks with circular micro-channels are investigated using the continuum model consisting of the conventional Navier-Stokes equations and the conventional energy equation. Developing flow (both hydrodynamically and thermally) is assumed in the fluid region and three-dimensional conjugate heat transfer is assumed in the solid region. Thermal results based on this approach are shown to be in good agreement with existing experimental data. Numerical results obtained for heat sinks with various geometries indicate that the heat absorbed by the fluid per unit length is high at the inlet and decreases steadily in the flow direction. As a result, the bulk temperature increases at a higher rate near the inlet. The geometry of the heat sink is shown to have a significant effect on the overall thermal resistance.

Commentary by Dr. Valentin Fuster
2003;():739-746. doi:10.1115/ICMM2003-1096.

A micro heat pipe will operate effectively by achieving its maximum possible heat transport capacity only if it is to operate at a specific temperature, i.e., design temperature. In reality, micro heat pipe’s may be required to operate at temperatures different from the design temperature. In this study, the heat transport capacity of an equilateral triangle micro heat pipe is investigated. The micro heat pipe is filled optimally with working fluid for a specific design temperature and operated at different operating temperatures. For this purpose, water, pentane and acetone was selected as the working fluids. From the numerical results obtained, it shows that the optimal charge level of the micro heat pipe is dependent on the operating temperature. Furthermore, the results also shows that if the micro heat pipe is to be operated at temperatures other than its design temperature, its heat transport capacity is limited by the occurrence of flooding at the condenser section or dryout at the evaporator section, depending on the operating temperature and type of working fluid. It is observed that when the micro heat pipe is operated at a higher temperature than its design temperature, the heat transport capacity increases but limited by the onset of dryout at the evaporator section. However, the heat transport capacity decreases if it is to be operated at lower temperatures than its design temperature due to the occurrence of flooding at condenser end. From the results obtained, we can conclude that the performance of a micro heat pipe is decreased if it is to be operated at temperatures other than its design temperature.

Topics: Heat , Heat pipes
Commentary by Dr. Valentin Fuster
2003;():747-752. doi:10.1115/ICMM2003-1097.

The heat dissipated by convection from the fins of the heat pipe condenser section is strongly limited by the thermal barrier of the oxide layer formed on their aluminum surface. A lot of work has been done to enhance the heat transfer coefficient of this heat pipe section by changing the fins roughness. The present experimental study demonstrates the enhancement in heat transfer coefficient by applying a more conductive coating on the condenser fins surface. A comparison between a conventional technique consisting of applying a rougher surface and this new technique is performed. Results clearly show the performance of the heat pipe exhibits a better enhancement in the case of a more conductive coating than a rougher one. The orientation of the heat pipe is also investigated to demonstrate the effect of gravity on the enhancement so observed. Hydrodynamics inside the heat pipe is considered to explain the findings.

Commentary by Dr. Valentin Fuster
2003;():753-757. doi:10.1115/ICMM2003-1098.

Two different experimental set-ups are suggested which both have the advantageous features that channels of continuously varying size between mini- and microchannels are realized, that these channels are formed between macroparts without any need for microfabrication and last but not least that the channel walls are freely accessible and can be manipulated with high precision by macro surface manipulation techniques. In the limit of vanishing channel height (h → 0) the flow approaches plane channel flow. For small but finite values of h deviations from this limiting flow case are discussed in detail. For one of the experimental set-ups construction details are shown; for the other one (which is still in the conception phase) heat transfer measurements within the microchannels are discussed.

Commentary by Dr. Valentin Fuster
2003;():759-763. doi:10.1115/ICMM2003-1099.

The behavior of micro-scale flow is significant for the performance of Micro-Electro-Mechanical-Systems (MEMS) devices. Some experiments about liquid flow through microtubes with diameters about 3∼20μm are presented here. The liquids used in our experiments include some simple liquids with small molecules, such as non-ion water and several kinds of organic liquids (CCL4 , C6 H5 C2 H5 and Isopropanol etc.). The flow rate and the normalized friction coefficients were measured in micro-flow experimental apparatus. The results show that when the driven pressure varies from 0 to 1Mpa, the flow behaviors in 20μm microtube for both polar and non-polar liquids are in agreement with Hagen-Poiseuille law of the classical theory. It means that N-S equation based on continuous medium still acts well in this case. For higher pressure drop from 1 to 30Mpa, in the microtubes with diameter of 3∼10μm, the normalized friction coefficients of organic liquids can’t keep constant with pressure increases. However the non-ion water reveals different trends.

Topics: Flow (Dynamics)
Commentary by Dr. Valentin Fuster
2003;():765-772. doi:10.1115/ICMM2003-1100.

Experimental investigations of the flow and the associated heat transfer were conducted in two-dimensional microchannels in order to test possible size effects on the laws of hydrodynamics and heat transfer and to infer optimal conditions of use from the measurements. The test section was designed to modify easily the channel height e between 1 mm and 0.1 mm. Measurements of the overall friction factor and local Nusselt numbers show that the classical laws of hydrodynamics and heat transfer are verified for e > 0.4 mm. For lower values of e, a significant decrease of the Nusselt number is observed, whereas the Poiseuille number continues to have the conventional value of laminar developed flow. The transition to turbulence is not affected by the channel size. For fixed pressure drop across the channel, a maximum of heat transfer rate density is found for a particular value of e. The corresponding dimensionless optimal spacing and heat transfer rate density are in very good agreement with the predictions of Bejan and Sciubba (1992). This paper is the first time that the optimal spacing between parallel plates is determined experimentally.

Commentary by Dr. Valentin Fuster
2003;():773-780. doi:10.1115/ICMM2003-1101.

Heatric has been involved in the commercial design and manufacturing of “micro/milli” scale heat exchanger core matrices called Printed Circuit Heat Exchangers (PCHEs) since 1985. These core matrices are formed by diffusion bonding together plates into which fluid flow microchannels have (usually) been formed by photo-chemical machining. Complex fluid circuitry is readily implemented with this technique. Diffusion bonding is a ‘solid-state joining’ process creating a bond of parent metal strength and ductility. The complete microchannel heat exchangers are highly compact, typically comprising about one-fifth the size and weight of conventional heat exchangers for the same thermal duty and pressure drops. PCHEs can be constructed out of a range of materials, including austenitic stainless steels suitable for design temperatures up to 800°C, and nickel alloys such as Incoloy 800HT suitable for design temperatures more than 900°C. Single units ranging from a few grams up to 100 tonnes have been manufactured. Currently there are thousands of tons of such microchannel matrix in hundreds of services — many of them arduous duties on offshore oil and gas platforms where the size and weight advantages of microchannel heat exchangers are of obvious benefit. Whilst these matrices are predominantly involved in thermally simple two-fluid heat exchange, albeit at pressures up to 500 bar, PCHEs have also been used for many multi-stream counter-flow heat exchangers. However the field of applications is very varied, including specialised chemicals processing, and PCHEs are even to be found orbiting the Earth in the International Space Station! Due to the inherent flexibility of the etching process, the basic construction may readily be applied to both a wider range, and more complex integration of process unit operations. Chemical reaction, rectification, stripping, mixing, and absorption, as well as boiling and condensation, can be incorporated into compact integrated process modules. Crucially, the resulting degree of compactness has led printed circuit technology to be the enabling technology in certain duties. Techniques for chemical coating onto the surfaces of channels continue to evolve, with applicability both to protective coatings and catalytically active coatings. We will describe a selection of innovative printed circuit technology examples. Alongside the more esoteric, Heatric is actively extending printed circuit technology to adapt to new market opportunities such as nuclear power generation plant and fuel cell systems. These applications perhaps represent two extremes of the both size and process integration, and thus aptly serve to demonstrate the range of industrial use of microchannel devices.

Topics: Microchannels
Commentary by Dr. Valentin Fuster
2003;():781-786. doi:10.1115/ICMM2003-1102.

In this study, an analysis of the performance of micro nozzle/diffusers is performed and fabrication of the micro nozzle/diffuser is conducted and tested. It is found that the pressure loss coefficient for the nozzle/diffuser decreases with the Reynolds number. At a given Reynolds number, the pressure loss coefficient for nozzle is higher than that of the diffuser due to considerable difference in the momentum change. For the effect of nozzle/diffuser length on the pressure loss coefficient, it is found that the influence is rather small. At a fixed volumetric flowrate, a “minimum” phenomenon of the pressure loss coefficient vs. nozzle/diffuser depth is encountered. This is related to the interactions of velocity change and friction factor. Good agreements of the measured data with the predicted results are found in this study except at a diffuser having an opening angle of 20° . It is likely that the departure of this case to the prediction is due to the separation phenomenon in a larger angle of the diffuser.

Commentary by Dr. Valentin Fuster
2003;():787-794. doi:10.1115/ICMM2003-1103.

Microchannel-based master molds or final devices are typically produced using a series of resist deposition, exposure, development and etching steps. These steps can then be repeated to create multi-layer fluidic structures. Traditional fabrication of these devices requires the use of a physical mask for the photolithographic exposure process. In the research and development environment, where designs are constantly undergoing changes, or in rapid-time-to-device applications, this can be a costly and time-consuming practice. We have employed a novel, micron-scale resolution maskless photoimaging/patterning tool that permits the creation of small, arbitrary features. This microdevice printer is useful for constructing fluidic channels, devices, structures and packages utilizing any photoimageable or photoreactive material that can be applied towards fabrication of integrated microfluidic-based systems. The fabrication technology can provide features down to 20 microns simultaneously over a 2×2 cm2 field of view. Additionally, manual stitching techniques can yield unlimited field-of-view for large area fluidic patterns with high-resolution elements. The instrument relies on the use of microoptics and spatial light modulation to create the required 2D aerial image for photoimprinting. The instrument creates mask-free designs on planar and curved surfaces and has been applied to a variety of materials, including metals, ceramics, organic polymers and semiconductors. We have demonstrated the utility of the instrument for creating mechanical, optical, fluidic and electronic components and combinations that would form the basis of integrated microfluidic systems, microanalytical systems and micrototal analysis systems (uTAS). We have also created fluidic channels having structures integrated within the channel geometry. The technology has widespread applications in the MEMS, bioMEMS, microcooling technologies and sensor markets. A further extension of the technology is the application of the direct printer to rapid prototyping of microchannels and minichannels for fuel cells, microrefrigerators, heat exchangers, and biomedical devices.

Topics: Microchannels
Commentary by Dr. Valentin Fuster
2003;():795-800. doi:10.1115/ICMM2003-1104.

A novel approach in creating circuit electrodes with features as fine as 100 μm is demonstrated using a single 38 μm diameter orifice, piezoelectrically driven print head to deposit metallic nanoparticle suspensions. The suspensions consist of gold particles of ∼20 nm diameter suspended in toluene solvent. The amount of gold nanoparticles present in the suspension is 30% wt. Inductor and capacitor electrode patterns are deposited onto a glass substrate and thermally processed at 300°C for 15 minutes to drive off the solvent and allow the nanoparticles to sinter, thereby yielding a conductive path with a resistivity of O(10−7 ) Ω m.

Commentary by Dr. Valentin Fuster
2003;():801-808. doi:10.1115/ICMM2003-1105.

The need for efficient metering control of liquids in small devices has led to a boom in the advent of different micro-fluidic actuation mechanisms. Here we present a brief study on a thermal-pneumatic actuation mechanism for an on-demand delivery of minute amounts of liquids. A closely coupled, iterative design-fabrication strategy is used for optimization of a system in which no freely moving membranes are included. Special consideration was given to the heating device, minimizing the energy consumed. The fabrication method and performance of two types of fabricated resistors are compared herein. The first, a conventional Nickel-Chromium resistor using, lift-off micro-fabrication techniques, was initially tested. The second, a Copper cladded liquid crystal polymer in conjunction with a novel mask-less patterning system was used to produce nonconventional heating micro-ohmic heaters. The heating efficiency was proven to be superior using the latter approach. Various micro-fabricated fluidic devices have been designed as case studies and have been fabricated and integrated using a variety of materials to illustrate the functionality of the approach. The combination of design and fabrication steps, the simplicity of the resistive device, and the materials selected combined, yield a direct path to making fluidic transport devices for micro-analytical and power systems.

Commentary by Dr. Valentin Fuster
2003;():809-815. doi:10.1115/ICMM2003-1106.

In order to understand heat transfer processes at the microscale, detailed temperature measurements are required. This paper begins with a review of the current state-of-the art in fluid temperature measurement at the microscale. At present, fluid temperature profiles are not measured, with verification of predicted heat transfer performance being based on global measurements. The paper describes a potential full-field technique based on micro-interferometry. The accuracy of extracting temperature data from small phase difference intensity maps is discussed, with particular reference to the high levels of signal to noise as would be found in a micro-scale flow. Benchmark optical experiments quantifying the effect of noise on phase evaluation are described and the paper concludes with an outline of the achievable resolution for a given channel length and fluid.

Commentary by Dr. Valentin Fuster
2003;():817-822. doi:10.1115/ICMM2003-1107.

Micro-resolution Particle Image Velocimetry (Micro-PIV) was used to measure the flow in a micro-branch (Micro-Bypass). In this paper, effects of particle lump at the tip of a Micro-branch and difficulties of Micro-PIV measurements for microfluidics with branch passage were described. Micro-bypass was composed of a straight channel (200μm width×80μm height) and two branches which has 100μm width×80μm height. One of branches was straight and the other was curved. Experiments were performed at three regions along streamwise direction (enterance, middle and exit of branch) and five planes along vertical direction (0,± 10,±20μm) for the range of Re = 0.24, 1.2, 2.4. Numerical simulation was done to compare with the measurements and understand the effects of particle lump at the tip of branch.

Commentary by Dr. Valentin Fuster
2003;():823-827. doi:10.1115/ICMM2003-1108.

Most microfluidic chips consist of several microchannels inside. In order to design microfluidic chips efficiently, it is important to predict the flow passage and to understand the flow characteristics on the chip. In this study, the flow structure inside microchannels has been investigated using a micro-PIV system. We focused on the flow resistance with respect to the inlet configuration of microchannels. The microchannels made of poly-dimethyl-siloxane (PDMS) material were fabricated by a micro-molding technique using SU-8 (photoresist) master. The width (w) and depth of the microchannels were fixed as 100 μm and 58 μm, respectively. Six different inlet configurations with curvature radii in the ranges from r = 0.2w to 1.5w were tested in this study. As a result, with increasing the curvature radius of the inlet corner, the streamwise mean velocity develops slowly in the entrance region, but the fully developed velocity at further downstream is increased. When the curvature radius is larger than r = 0.6w, the reduction rate of flow resistance is not so significant. For the microchannels with r = 0.6w, 0.8w and 1.0w the downstream mean velocity at channel center has nearly the same value of about 276 mm/sec, 10.5% larger than that of r = 0.2w. The simple rounding of microchannel inlet corner reduces flow resistance effectively by smoothing the incoming flow. The length of entrance region is much smaller than that of macro-scale channel.

Commentary by Dr. Valentin Fuster
2003;():829-836. doi:10.1115/ICMM2003-1109.

Micro-fluid mixing is an important aspect of many of the various micro-fluidic systems used in biochemical production, biomedical industries, microenergy systems and some electronic devices. Active or highly effective passive mixing techniques are often required. In this study, two pulsed injectors are used to actively enhance mixing in a high aspect ratio microchannel (125 μm deep and 1 mm wide). The main channel has two adjacent flowing streams with 100% dye and 0% dye concentrations, respectively. Two injectors (125 μm deep and 250 μm wide) are located on opposite sides of the channel and off-set in the downstream direction. A dye solution is used to map local mixing throughout the channel by measuring concentration variations as a function of both space and time. Images of the concentration variations within the channel are used to quantify mixing. It is shown that there is a high degree of repeatability of concentration distribution as a function of phase of the pulsing cycle. The flow rate ratio between the injectors and main channel is found to be the most influential parameter on overall mixing, and evidence of an optimal flow rate ratio and frequency is presented. A nondimensional correlation is presented that could be used to predict the level of mixing for the conditions studied.

Commentary by Dr. Valentin Fuster
2003;():837-843. doi:10.1115/ICMM2003-1110.

Non-uniform, AC electric fields created by coplanar electrodes patterned on a substrate are used to move and manipulate aqueous liquid masses, and to dispense very small droplets. This liquid dielectrophoretic microactuation scheme has potential applications for microfluidic systems in the laboratory on a chip. Simple, co-planar electrode strips are used to divide microliter-sized, sessile water droplets into large numbers of droplets down to ∼40 picoliters. The dispensing system uses the electrodes to draw a long finger or rivulet of liquid from the parent microliter droplet. When the voltage is removed, the rivulet breaks up into numbers of droplets as a result of the familiar capillary instability. We propose and provide data that supports a very simple power law dependence of the finger length upon time: Z(t)∝t, which governs the time required to fill a structure. A capillary instability, very similar to the case of the cylindrical jet, leads to droplet formation when voltage is removed. The hydrodynamic instability features a critical wavelength, below which instability is not possible, and a most unstable wavelength, which controls the volume and spacing of the droplets formed.

Commentary by Dr. Valentin Fuster
2003;():845-850. doi:10.1115/ICMM2003-1112.

Multiwall carbon nanotubes show potential for use in various micro- and nanofluidic devices, since they resemble cylindrical channels used in the macroscopic world. However, in situ experimental studies of fluid behavior in nanotubes or nanochannels have been rare. In this work, transmission electron microscopy experiments are performed on closed-end multiwall carbon nanotubes filled with an aqueous multiphase fluid. The nanotubes form an experimental apparatus that is a few orders of magnitude smaller than the smallest channels used in other fluidic experiments. These nanotubes are synthesized hydrothermally, using Ni as a catalyst, and they contain segregated aqueous liquid and gas inclusions with clearly defined interfaces. Using electron irradiation, the multiphase fluid inside individual nanotubes is excited thermally, by expanding and contracting the electron beam. The excellent wettability of the graphitic inner tube walls by the aqueous fluid and the mobility of this liquid in the nanotubes are observed in real time with nanometer-scale resolution. Interface dynamic phenomena are visualized, as driven by thermocapillary forces as well as by evaporation and condensation. The hydrothermal nanotubes examined herein offer a promising platform for studying the behavior of multicomponent, multiphase fluids in nanosize channels at high-pressure conditions. The phenomena documented in this study demonstrate the potential of implementing such tubes in future nanofluidic devices.

Commentary by Dr. Valentin Fuster
2003;():851-857. doi:10.1115/ICMM2003-1113.

Microchannel blockage phenomena by hard, spherical particles have been investigated experimentally and theoretically. The study was performed over a range of particle-to-channel diameter ratios of 0.14 < R < 0.65. Two mechanisms have been investigated: orthokinetic flocculation and hydrokinetic arching. Arching appears to be the main mechanism for large, hard particles. In the absence of Brownian motion and inter-particle repulsive forces, other than simple Hertzian contact force, the blockage phenomenon is described by three non-dimensional parameters, N, R and β. The mean total number of particles in the channel having length L is N. Ratio of a diameter of particle (dp ) and a diameter of channel (D) is R. Blockage efficiency factor β is determined experimentally. The data shows that a critical value Nc exists as a function of R. N > Nc implies high likelihood of blockage; if N < Nc , blockages were never observed. The critical number decreases dramatically with increasing R. Blockages can occur at surprisingly low values of the volume concentration (φ). The experimental results matches well with the theory for the combinations of straight glass capillary, 76 < D < 156μm, 100mm-length, and spherical polymer particle, 22 < dp < 48μm.

Topics: Microchannels
Commentary by Dr. Valentin Fuster
2003;():859-867. doi:10.1115/ICMM2003-1114.

A two-dimensional model was developed to predict concentration profiles resulting from passive, or diffusive, mixing of laminated layers formed in a fractal-like merging flow network. Both uniform and parabolic velocity profiles were considered in the model. Concentration profiles were experimentally acquired near the top surface of the flow network using laser induced fluorescence. The degree of mixing was assessed from concentration profiles at the end of each channel. Although the degree of mixing from the two-dimensional model well predicts the trend of the experimental degree of mixing, the numerical model under predicts the experimental values by approximately 25 percent. This may be due in part to the presence of top and bottom walls in the experimental device. These walls tend to slow the flow in this region, thereby increasing the residence time and improving the mixing. These top and bottom walls are neglected in the two-dimensional model. For the existing flow network, the degree of mixing is provided as a function of Peclet number. The degree of mixing is further investigated by varying the number of branching levels, the width of the initial flow channels, and the total flow length for a fixed Peclet number. A nondimensional parameter is established that serves as a design tool for predicting an optimum number of branching levels for fixed values of the total flow length, initial branch width and channel depth.

Commentary by Dr. Valentin Fuster
2003;():869-877. doi:10.1115/ICMM2003-1115.

Flow through fractal-like branching flow networks is investigated using a three-dimensional computational fluid dynamics approach. Results are used to assess the validity of, and provide insight for improving, assumptions imposed in a one-dimensional model previously developed. Assumptions in the one-dimensional model include (1) reinitiating boundary layers following each bifurcation, (2) negligible minor losses at the bifurcations, and (3) constant thermophysical fluid properties. It is concluded that the temperature dependence of fluid properties, boundary layer development, and minor losses following a bifurcation are not negligible in analyses of branching flow networks.

Commentary by Dr. Valentin Fuster
2003;():879-886. doi:10.1115/ICMM2003-1117.

Significant developments in detailed quantitative microscale fluid flow visualization have been made in recent years. However, spatially resolved surface temperature measurements in microchannels have received far less attention. This paper introduces quantitative and qualitative surface temperature measurement in microscale flows using infrared (IR) thermography. This technique is applicable for channel dimensions of up to 10 μm. In order to determine temperatures accurately, careful attention toward proper choice of windows, elimination of specular reflections from channel walls, and estimation of local emissivity changes is warranted. These concerns are addressed in the paper, and preliminary results of detailed temperature distribution in a single-phase mini-channel flow are presented.

Commentary by Dr. Valentin Fuster
2003;():887-894. doi:10.1115/ICMM2003-1118.

A micro-PIV system is presented in detail, pointing out important aspects of micro-PIV system design cruicial for its operation. The micro-PIV system is then applied on a sinusoidal microchannel, and the fluid motion inside the device is presented and discussed. The wall shear stress at the waist of the channel is measured to be up to 60% higher than the wall shear stress in a conventional parallel-plate flow. The results suggest that altering of channel geometry may contribute to better design of cross-flow microfiltration units, in terms of reduced clogging by shear-control of bacterial motion. Furthermore, the flow is shown to exhibit a strong Reynolds number dependence, characterised by the onset of periodic distortion imposed on the flow by the sinusoidal walls occuring between Re = 2 and Re = 10.

Topics: Microfiltration
Commentary by Dr. Valentin Fuster
2003;():895-902. doi:10.1115/ICMM2003-1119.

Pumping of liquids and gases in micro fluidic systems has been the focus of much attention in recent times. Miniaturisation of traditional rotating pumps such as axial and radial flow designs, has been limited by the fabrication techniques employed. As a result of these limitations, the geometry of the majority of rotating micro pumps has been two-dimensional. This paper addresses issues of scaling in micro axial flow fans. The anticipated primary application will be in cooling compact electronic systems, but the results are applicable to a much wider range of pumping applications. Using novel fabrication techniques a series of geometrically similar three dimensional fans were fabricated, ranging in size from the macro to the micro scale. Experimental techniques are described which will be used for the characterisation of these fans. A scaling analysis is used to show how reduced fan scale causes increased local loss as fan dimensions are reduced to the micro scale. Numerical simulations of flow in the channels between the fan blades were performed to investigate the validity of the scaling theory, the results of which give confidence in the scaling analysis. The fundamental finding of this work is that a reduction in scale is accompanied by a reduction in efficiency and thus fan performance.

Topics: Fans
Commentary by Dr. Valentin Fuster
2003;():903-910. doi:10.1115/ICMM2003-1120.

In this paper, measurements are presented of the velocity profile in a mini-channel at different locations. The channel is rectangular in cross-section, approximately 1.2mm wide, 1.4mm deep and 29mm long. A micro-PIV system was used to obtain the velocity profiles at the inlet, mid-length and exit of the channel. The raw image maps were processed using three different commercial PIV software packages, and compared to an exact analytical solution. The mini-channel system was also simulated using a commercial CFD code as a further check on the dataset, and the experimental rig itself. It was found that the different processing procedures had little influence on the micro-PIV data, and good agreement was found with theory, numerical prediction and experiment. This establishes confidence in micro-PIV as a measurement tool in micro-systems.

Commentary by Dr. Valentin Fuster
2003;():911-918. doi:10.1115/ICMM2003-1121.

Micro mixers are an integral part of several micro fluidic devices like micro reactors or analytical equipment. Due to the small dimensions, laminar flow is expected a priori in those devices while the mass transfer is supposed to be dominated by diffusion. A detailed numerical CFD-study by CFDRC-ACE+ of simple static mixers shows a significant deviation from strictly laminar flow in a wide range of Reynolds numbers Re, channel dimensions, and types of cross sections (square, rectangular, trapezoidal). With increasing flow velocity and Re number the flow starts to form vortexes at the entrance of the mixing channel. The vortexes are symmetrical to the symmetry planes of the mixing channel, both for the rectangular and the trapezoidal cross sections investigated here. With further increasing velocity the flow tends to instabilities, which causes a breakup of the flow symmetry. These instabilities are generally found in T-shape mixers with symmetrical flow conditions, but not always in Y-shape mixers or with asymmetrical flow conditions. Within the laminar flow regime diffusive mass transfer is dominant. In this case the mixing quality at constant channel length becomes worse with increasing velocity. This effect can almost be equalized by the onset of the vortex regime, which enhances the mass transfer by convective transport. This paper shows the mixing quality at a certain length for different geometrical parameters and flow conditions.

Commentary by Dr. Valentin Fuster
2003;():919-925. doi:10.1115/ICMM2003-1122.

This paper presents preliminary experimental results of measuring velocity fields of a transparent liquid flow in a closed circuit, through a 100 μm, 200 μm and 500 μm depth flat cell with and without included 200 μm × 200 μm heat exchanger microchannel elements. A microscopic particle image velocimetry system has been developed which consists of a 5 ns pulsed Nd:YAG laser, an epi-fluorescent microscope and a cooled interline transfer CCD camera to record high-resolution 0.7 μm diameter fluorescent particle image fields. The development of the optical micro-flow measuring instrument for the present application took into consideration the fact that the test object is aimed to be exposed to pressures up to 0.6 MPa leading to flow velocities up to 15 m/s, as well as temperatures up to 100°C. Particle fouling phenomenon proved to be the main difficulty in performing velocity field measurements in microchannels and techniques to avoid or limit it are proposed.

Commentary by Dr. Valentin Fuster
2003;():927-931. doi:10.1115/ICMM2003-1123.

Gas gaps are common structures in many sensors and MEMS. It is usually regarded that heat conduction plays a dominant role when the gap size reduces and thermal radiation between surfaces is negligible. This work compares the heat dissipated by heat conduction with that by thermal radiation under different temperatures and on various scale levels. It is found that the heat flux by thermal radiation can exceed that by heat conduction. Furthermore, a regime map is plotted to recognize the relative importance of heat conduction and thermal radiation at different gap size and temperature. The impact of the thermal accommodation efficient for gas is also discussed.

Commentary by Dr. Valentin Fuster
2003;():933-943. doi:10.1115/ICMM2003-1125.

Heat and mass transfer processes become highly efficient as the channel hydraulic diameter is reduced in size. Biological systems, such as human body, rely on the extremely efficient transport processes occurring at microscale in the functioning of its vital organs. In this paper, the transfer processes in lungs and kidneys will be reviewed. Although the flow in the microchannels present in these organs is laminar, it yields very high mass transfer coefficients due to the coupling of small channel diameters. Furthermore, the molecular transport mechanisms occurring across the membranes at nanoscales through diffusion controlled processes also become extremely important. Understanding these transport processes will enable us to develop more efficient artificial organs and processes that closely mimic the performance of the natural systems. These ideas can be extended to other microscale system designs in different technologies, such as IC cooling and MEMS micro fuel cells.

Commentary by Dr. Valentin Fuster
2003;():945-949. doi:10.1115/ICMM2003-1126.

A team from the structural biology group located at the Marshall Space Flight Center in Huntsville Alabama is developing a platform suitable for cross-disciplinary microchannel research. The original objective of this engineering development effort was to deliver a multi-user flight-certified facility for iterative investigations of protein crystal growth; that is, Iterative Biological Crystallization. However, the unique capabilities of this facility are not limited to the low-gravity structural biology research community. Microchannel-based research in a number of other areas may be greatly accelerated through use of this facility. In particular, the potential for gas-liquid flow investigations and cellular biological research utilizing the exceptional pressure control and simplified coupling to macroscale diagnostics inherent with the facility will be discussed. Also noted will be the opportunities for research-specific modifications to the microchannel configuration, control and diagnostics.

Topics: Microchannels
Commentary by Dr. Valentin Fuster
2003;():951-959. doi:10.1115/ICMM2003-1001.

The volumetric heat dissipated by computer equipment at each level of the package from the chip to the chassis is having a tremendous impact on the thermal management of computer equipment. Because of the consumer’s insatiable desire for increased performance the competitive pressures are driving the computer manufacturer to pack as much processor/memory performance within the smallest volume possible. The consumer views high performance in a compact package as a benefit. These market pressures seem to be in direct conflict with the desire to continue to provide air cooling solutions for the foreseeable future. Because of these trends in power and package design other cooling technologies beside air are now becoming viable techniques but each must be weighed with many other factors that influence the cooling technology selected. These factors will discussed along with two specific IBM server packages and their associated cooling technology employed. Finally a microprocessor liquid cooled minichannel heat sink will be described and its performance presented as applied to a current microprocessor (IBM Power4) chip.

Commentary by Dr. Valentin Fuster
2003;():961-966. doi:10.1115/ICMM2003-1073.

Critical heat flux (CHF) is a very important design factor of boiling channel, then, so many investigations have been conducted so far. In the case of small diameter channel, the main interest is related with the heat removal of high heat flux component. Therefore, CHF of that system should be predicted by DNB condition. On the other hand, CHF under low heat flux condition in small channel can be considered as the relation with two kinds of restrictions. In this investigation, the confirmation of the relationship of two restrictions in CHF was principal purpose. The CHF of this system was basically decided by the dryout condition, but it deviated from the dryout under certain conditions. In those conditions, the critical flow condition achieved in lower heat flux compared with that of the dryout. Owing to this restriction of the flow rate by critical flow condition, pseudo CHF condition occurs. Experimental results have expressed these relationships between CHF and critical flow condition well.

Commentary by Dr. Valentin Fuster
2003;():967-974. doi:10.1115/ICMM2003-1094.

One of the inter-connected factors that can lead to failures in the flow plates of PEM fuel cells is the pressure differences that exist between adjacent flow channels. These pressure differences lead to stresses in the channel supports, i.e., the ribs, which can be important in the presence of stresses arising due to other factors such as temperature gradients in the flow plates. In order to investigate the magnitudes of the pressure differences across the supports and the places where the maximum pressure differences occur, the flow and pressure variations in various forms of serpentine channels, these channels having a rectangular cross-sectional shape, have been numerically calculated. The presence of the diffusion layer has been ignored and the flow has been calculated using a commercial finite-element software package using the governing equations written in dimensionless form. Solutions have been obtained for various values of the Reynolds number for each of the flow geometries considered for two channel height-to width ratios (one and three). Except for the flow in the vicinity of the bends in channels, the pressure has been found, as is to be expected, not vary significantly across the channel cross-section. The difference between the center point dimensionless pressure in a given channel with those at the same longitudinal position in the adjacent channels has been determined. The dependence of the highest dimensionless pressure difference between channel on the input parameters has been examined.

Commentary by Dr. Valentin Fuster
2003;():975-986. doi:10.1115/ICMM2003-1124.

Vascular endothelial cells are known to respond to fluid shear stress. To gain insights into the mechanism of flow response by these cells, various types of in vitro devices in which endothelial cells can be cultured under flowing culture medium have been designed. Using such a device, one can apply known levels of (usually laminar) fluid shear stress to cultured endothelial cells. We have made two types of devices: a viscometer-based cone-and-plate flow apparatus and a parallel plate chamber. The cone-and-plate apparatus is used to do biochemical analyses of flow effects on cells while the parallel plate chamber is used to observe dynamic behavior of endothelial cells under flow. We were able to maintain confluent endothelial cell cultures under flow for over a week in the parallel plate flow apparatus. Using this chamber and high resolution time-lapse video microscopy, we studied morphological changes of endothelial cells exposed to different levels of fluid shear stress. We found that endothelial cells in a confluent monolayer exhibited three types of fluid shear stress level-dependent morphological and motile responses within a narrow fluid shear stress range between 0.1–10 dyn/cm2 . Endothelial cells cultured under no flow exhibited variable shapes and no preferred orientation of their long cell axes and showed a jiggling motion. When exposed to fluid shear stress levels of below 0.5 dyn/cm2 , endothelial cell morphology and motility were not affected. However, when fluid shear stress levels were increased to 2–4 dyn/cm2 , they became polygonal and showed increased random-walk activity. Fluid shear stress over 6 dyn/cm2 caused endothelial cells to initially become polygonal and increase their random-walk activity, but they soon became elongated and aligned in the direction of flow. As the cells elongated and aligned, they migrated in the direction of flow. The average velocity of this directed cell migration was less than that of cells moving randomly under the same flow condition at earlier times. These observations indicate that endothelial cells are able to detect and respond to a surprisingly small change in fluid shear stress. It is possible that endothelial cell physiology in vivo is also regulated by small changes in fluid shear stress and that a fluid shear stress change of a few dynes per cm2 within a certain region of an artery could trigger atherogenesis in that particular location.

Commentary by Dr. Valentin Fuster

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In