0

Boiling

2005;():1-7. doi:10.1115/ICMM2005-75010.

We previously evidenced the influence of confinement and inlet conditions on convective boiling stability in a minichannel. The experiments were realized based on an upward n-pentane two-phase flow. Here, we present results of convective boiling in a minichannel for several minichannel orientations which can be modified from the horizontal (heating surface on the top) through the vertical (previous situation already studied of upward flow) to the horizontal (heating surface on the bottom). We present the results obtained for the same heat flux provided to the minichannel (Q W = 92 kW.m−2 ), the same range of inlet mass velocity (73 to 2300 Kg.m−2 .s−1 ) for 5 different minichannel’s orientations: −90°, −45°, 0°, 45° and +90°. The consequences on the minichannel total pressure drop, fluid-wall temperature, and two-phase flow stability are discussed.

Commentary by Dr. Valentin Fuster
2005;():9-14. doi:10.1115/ICMM2005-75026.

Flow boiling heat transfer characteristics of R134a were experimentally investigated in a horizontal stainless steel mini-tube. The inner diameter of the test tube is 1.3 mm and the tube wall thickness is 0.1 mm. Local heat transfer coefficients are obtained over a range of vapor qualities up to 0.8, mass fluxes from 310 to 860 kg/m2 s, heat fluxes from 21 to 50 kW/m2 , and saturation pressures from 6.5 to 7.5 bar. The mass flux, heat flux, saturation pressure, and vapor quality dependences of heat transfer coefficients are demonstrated. Based on an available model in recent literature potential heat transfer mechanisms are also analyzed.

Commentary by Dr. Valentin Fuster
2005;():15-22. doi:10.1115/ICMM2005-75048.

This paper addresses a non-dimensional analytical stability model aimed at predicting the occurrence of flow instabilities at micro-scale. In this context, linear stability model using homogenous flow was considered. Towards that, a linear stability model was developed using perturbation method. A characteristic equation (the response of pressure drop to a hypothetical perturbation in inlet velocity) obtained in this analysis, is shown to be a function of sub-cooling number, Zuber number, Froude number, friction number and inlet and outlet restriction coefficients. Then, a neutral dynamic stability curve is obtained using D-Partition approach. Similarly, static or excursive stability curve is also obtained from the characteristic equation. The derived analytical form for static and dynamic instability threshold is represented in the form of simplified correlations. The experimental data reported by other researchers agree well with these correlations. From the results, it is amply clear that for all practical purposes, two-phase cooling will be unstable. The question to be answered in future is, therefore, whether the oscillations that accompany can be tolerated from the application viewpoint.

Commentary by Dr. Valentin Fuster
2005;():23-29. doi:10.1115/ICMM2005-75054.

An experimental study is undertaken to investigate microcapillary pumped loop (micro-CPL) evaporator behaviours. The experimental set-up is made of a horizontal square cross section glass microchannel which is locally heated up in order to analyze vaporization. A specific procedure allowed us to investigate the capillary pumping due to vaporization: the mass flow rate according to heat flux applied is determined. A laser-induced fluorescence technique has been used to observe phases distribution in the microchannel during vaporization, especially the liquid raising in the corners. The extended meniscus zone length has been determined as a function of the heat flux using image processing.

Topics: Microchannels
Commentary by Dr. Valentin Fuster
2005;():31-39. doi:10.1115/ICMM2005-75097.

Arrays of 68 microchannels, each one 200 μm wide, 100 μm deep and 40 mm long, were fabricated by mechanical micro-machining of brass foils, and used to evaporate water in the mass flux range between 61 kg m−2 s−1 and 123kg m−2 s−1 in an electrically powered device. Three different microchannel designs were tested: straight channels with and without impactor structures at the microchannel exits, and curved channels following a sinusoidal path. Variations of the average steam quality inside the microchannel arrays (pulsation) were observed by high-speed videography. Even when the device as a whole produced overheated steam, pulsation inside the channel array near the outlet, spatial maldistribution of the two phase flow, and liquid water reaching the outlet plenum were observed in many cases. From these observations and their analysis, conclusions are drawn for the further empirical optimization of microchannel array evaporators.

Topics: Water , Microchannels
Commentary by Dr. Valentin Fuster
2005;():41-47. doi:10.1115/ICMM2005-75124.

Experimental investigations on thermofluid-dynamic phenomena in a vertical narrow rectangular microchannel with the hydraulic diameter dh = 0.27 mm were carried out. The experiments are performed under fluid-inlet subcooling conditions with de-ionised and degassed water for different mass fluxes (50–2000 kg/m2 s) and heat fluxes (2–150 kW/m2 ). Moreover, flow visualisation of the two-phase flow patterns along the channel is performed using a digital high-speed video camera. Investigations on pressure drop during single- and two-phase flow have been carried out. The present work is concentrated on two-phase heat transfer. The mean heat transfer coefficient and the local heat transfer coefficient at saturated conditions were calculated and the latter ones was compared with available correlations.

Commentary by Dr. Valentin Fuster
2005;():49-56. doi:10.1115/ICMM2005-75128.

In the present study, the characteristics of heat transfer and flow patterns are experimentally investigated on the falling film evaporation of pure refrigerant HCFC123 in a vertical rectangular minichannel consisting of offset strop fins. The refrigerant liquid is supplied to the channel through 37 holes of a distributor. The liquid flowing down vertically is heated electrically from the rear wall of the channel and evaporated. To observe the flow patterns during the evaporation process directly, a transparent vinyl chloride resin plate is placed as the front wall. The experimental parameters are as follows: the mass velocity G = 28∼70 kg/(m2 s), the heat flux q = 20∼50 kW/m2 and the pressure P ≈ 100 kPa. It is clarified that the heat transfer coefficient α depends on G and q in the region of vapor quality x ≥ 0.3 while there is little influence of G and q in the region x ≤ 0.3. From the direct observation using a high speed video camera and a digital still camera, flow patterns are classified into five typical ones: plane liquid film, wavy liquid film, liquid film accompanied with dry patch, liquid film accompanied with dripping and liquid film accompanied with mist. Then the relation between heat transfer and flow pattern is clarified. The results of heat transfer characteristics are also compared with some previous correlation equations.

Commentary by Dr. Valentin Fuster
2005;():57-64. doi:10.1115/ICMM2005-75141.

A micro-grooved flat plate evaporator is modeled and its heat transfer characteristics are investigated numerically and experimentally. A test model is developed for the vapor compression cycle evaporator, where pressure gradient drives the vapor and the liquid flow. In this study, the effect of pressure gradient is implicitly introduced through the Smith’s equation for predicting void fraction from given quality. The film thickness profile in the micro region near the contact line is obtained by solving the 4th order differential equation. Then the local heat flux is obtained by assuming that the heat conduction through the liquid is one dimensional in the wall normal direction. The shape of liquid-vapor interface is assumed to be a circular arc in the macro region, whose radius is directly linked to the void fraction. This curvature radius is used as the boundary condition for the micro region model at the micro-macro interface. Finally, the heat transfer coefficient on a micro-grooved flat plate evaporator is measured in a HFC134a experimental loop and compared with the numerical prediction. The present model assumptions are validated and assessed.

Commentary by Dr. Valentin Fuster
2005;():65-72. doi:10.1115/ICMM2005-75142.

The paper continues the discussion of experimental and numerical investigations of forced convection boiling heat transfer in vertical minichannels covered by two former editions of this conference and our previous papers. Liquid crystal thermography technique was used for measuring the two-dimensional heating surface temperature distribution and boiling front detection. Influence of selected parameters on boiling heat transfer and nucleation hysteresis was observed and discussed. The two-dimensional heat transfer model and the analytic-numerical heat polynomial method were applied to solve the inverse boundary value problem and determine the temperature distributions in the heating foil and protecting glass and the boiling heat transfer coefficient as well. This paper shows how to modify and improve the heat polynomial method if we know the measurement errors and implement them into the numerical procedure. The accuracy of temperature measurements on the heating surface with liquid crystal method was estimated and the analysis of experimental results was given. The functions sought in numerical calculations describe temperature distribution in the protecting glass and the heating foil of the minichannel. They are presented in the form of linear combination of heat polynomials. The adopted boundary conditions and temperature measurements are used to construct error functionals. The latter express the root-mean-square errors, with which computed solutions satisfy relevant boundary conditions. On the basis of functional minimalisation unknown coefficients of linear combinations are determined. The solutions obtained satisfy the differential equations in the exact manner whereas the adopted boundary conditions are met in the approximate fashion. The unknown boiling heat transfer coefficient is the function computed from the boundary condition of the third kind. In the modified method, measurement errors are weights for individual temperature measurements. The more accurate is the measurement, i.e. has a smaller error, the greater is the weight put to it in further calculations. Therefore, it is possible to heighten the accuracy with which glass and foil temperature distributions, determined experimentally, fulfil the assumed equality conditions on the contact surface. Temperature distributions in the glass and the foil, computed on the basis of the modified method, are closer to real values than those obtained with the basic one. Local heat transfer coefficients obtained for two-dimensional boiling heat transfer model with both the basic and the modified heat polynomial methods are also compared.

Commentary by Dr. Valentin Fuster
2005;():73-80. doi:10.1115/ICMM2005-75143.

Flow boiling through microchannels is characterized by nucleation of vapor bubbles on the channel walls and their rapid growth as they fill the entire channel cross-section. In parallel microchannels connected through a common header, formation of vapor bubbles often results in flow maldistribution that leads to reversed flow in certain channels. The reversed flow is detrimental to the heat transfer and leads to early CHF condition. One way of eliminating the reversed flow is to incorporate flow restrictions at the channel inlet. In the present numerical study, a nucleating vapor bubble placed near the restricted end of a microchannel is numerically simulated. The complete Navier-Stokes equations along with continuity and energy equations are solved using the SIMPLER method. The liquid-vapor interface is captured using the level set technique. The results show that with no restriction the bubble moves towards the nearest channel outlet, whereas in the presence of a restriction, the bubble moves towards the distant but unrestricted end. It is proposed that channels with increasing cross-sectional area may be used to promote unidirectional growth of the vapor plugs and prevent reversed flow.

Commentary by Dr. Valentin Fuster
2005;():81-88. doi:10.1115/ICMM2005-75144.

In fouling, the accumulation of poorly conducting materials on the surface of process equipment, results in an increased resistance to heat transfer and hence reduces heat exchanger effectiveness. Under most conditions fouling is more severe during boiling heat transfer, due to the mechanisms of bubble formation and detachment. Thus, in order to enhance heat transfer and mitigate fouling in boiling processes, a new type of vapour-liquid-solid (three-phase) circulating fluidised bed boiling system has been designed, combining circulating fluidised bed technology with boiling heat transfer. Experiments are conducted in a glass mini-channel of square cross sectional area 21.5 mm × 11mm, height 1000mm, and fitted with an electrically heated cartridge heater rod of 8mm diameter × 730mm length. The set-up uses stainless steel balls to investigate the effect of particle properties (specifically particle size) on three-phase boiling heat transfer enhancement. Experimental results show that overall, compared with two-phase flow boiling, the presence of solid particles in the three-phase boiling system augments the heat transfer coefficient. Results are presented and discussed.

Commentary by Dr. Valentin Fuster
2005;():89-96. doi:10.1115/ICMM2005-75157.

The effects of dissolved air in the dielectric liquid FC-77 on flow boiling in a microchannel heat sink containing 10 parallel channels, each 500 μm wide and 2.5 mm deep, were experimentally investigated. Experiments were conducted before and after degassing, at three flow rates in the range of 30 to 50 ml/min. The dissolved air resulted in a significant reduction in wall temperature at which bubbles were first observed in the microchannels. Analysis of the results suggests that the bubbles observed initially in the undegassed liquid were most likely air bubbles. Once the boiling process is initiated, the wall temperature continues to increase for the undegassed liquid, while it remains relatively unchanged in the case of the degassed liquid. Prior to the inception of boiling in the degassed liquid, the heat transfer coefficients with the undegassed liquid were 300–500% higher than for degassed liquid, depending on the flow rate. The heat transfer coefficients for both cases reach similar values at high heat fluxes (over 120 kW/m2 ) once the boiling process with the degassed liquid was well established. The boiling process induced a significant increase in pressure drop relative to single-phase flow; the pressure drop for undegassed liquid was measured to be higher than for degassed liquid once the boiling process became well established in both cases. Flow instabilities were induced by the boiling process, and the magnitude of the instability was quantified using the standard deviation of the measured pressure drop at a given heat flux. It was found that the magnitude of flow instability increased with increasing heat flux in both the undegassed and degassed liquids, with greater flow instability noted in the undegassed liquid.

Commentary by Dr. Valentin Fuster
2005;():97-102. doi:10.1115/ICMM2005-75163.

A series of visualized experiments were conducted to investigate the boiling nucleation and bubble dynamics restricted within parallel microchannels on a silicon chip. The cross-section of each channel was 100 μm (W) × 100 μm (H). A high-speed CCD camera (up to 8,000 fps) was employed together with a microscope to record the boiling process. Under the present experimental conditions, the incipience of boiling was captured. The rates of bubble growth were measured at various flow and heating conditions. The interaction between vapor bubbles, vapor-liquid interface, and solid wall, was analyzed.

Commentary by Dr. Valentin Fuster
2005;():103-108. doi:10.1115/ICMM2005-75180.

Evaporation heat transfer and pressure drop characteristics of carbon dioxide were investigated in a multi-channel micro tube. The aluminum tube has 3 square channels with a hydraulic diameter of 2mm, a wall thickness of 1.5mm, and a length of 5m. The tube was heated directly by electric current. Experiments were conducted at heat fluxes ranging 4–16 kW/m2 , mass fluxes from 150 to 750 kg/m2 s, evaporative temperature from 0 to 10°C, and qualities from 0 to superheated state. The heat transfer coefficient measured was in the range of 6–15kW/m2 K, and the pressure drop was 3–23kPa/m. For the qualities lower than 0.5, the heat transfer coefficient was found to increase with the quality, which is assumed to be the effect of convective boiling. For the qualities higher than 0.6, sudden drop in heat transfer coefficients was sometimes observed due to local dry-out. It was found that dry-out occurred at lower quality if mass flux was smaller. The average heat transfer coefficient was found to increase with increasing heat flux, mass flux, and evaporation temperature, of which the effect of heat flux was the greatest. At given experimental conditions the pressure drop increased almost linearly with increasing quality. The total pressure drop was found to increase with increasing heat flux, mass flux, and evaporation temperature, of which the effect of mass flux was the greatest. From the experimental results simple correlations for heat transfer coefficients and pressure drop were developed.

Commentary by Dr. Valentin Fuster
2005;():109-113. doi:10.1115/ICMM2005-75185.

This paper reports an experimental study of flow boiling in a microchannel and this work has been focused on the investigation on the effect of operating pressure. The experimental apparatus consisted mainly of peristaltic pump, preheater, microchannel test tube, and vacuum chamber for control of operating pressure. Deionized water was used as the working fluid. The test section was a round tube of 310 μm inside diameter, made of 304 stainless steel. The experiment has been performed for the conditions of heat flux of 35∼85 kW/m2 , mass flux of 200∼300 kg/m2 s (160∼250 of liquid Reynolds numbers), and the operating pressure of 10∼20 kPa. The measured flow boiling heat transfer coefficients in the microchannel were in the range of 3.0 to 27 kW/m2 K and the experimental data showed that the flow boiling heat transfer coefficients in microchannels were affected mainly by the wall heat flux and the operating pressure.

Commentary by Dr. Valentin Fuster
2005;():115-124. doi:10.1115/ICMM2005-75197.

The flow boiling process suffers from severe instabilities induced due to nucleation of vapor bubbles in a minichannel or a microchannel in a superheated liquid environment. In an effort to improve the flow boiling stability, several modifications are introduced and experiments are performed on 1054 × 197 μm microchannels with water as the working fluid. The cavity sizes and local liquid and wall conditions required at the onset of nucleation are analyzed. The effects of an inlet pressure restrictor and fabricated nucleation sites are evaluated as a means of stabilizing the flow boiling process and avoiding the backflow phenomena. The results are compared with the unrestricted flow configurations in smooth channels.

Commentary by Dr. Valentin Fuster
2005;():125-130. doi:10.1115/ICMM2005-75209.

While silicon microchannel heat sinks are promising for high heat flux integrated circuits, they have not reached their potential because microscale convective boiling is poorly understood. Previous work integrated sensors and heaters into a silicon chip to provide distributed thermometry, but did not specifically examine hotspots or thoroughly treat experimental uncertainty. This work microfabricates a single channel in a thinned silicon beam, instrumented with doped sensors and aluminum heaters, to study the wall temperature and fluidic response to flow boiling induced by non-uniform heating. Uncertainty analysis shows a need for better measurements of the fabricated channel including channel cross section and surface roughness. Refined data from this work will suggest improvements to existing boiling flow models, which may then be implemented into a design tool for optimizing boiling flow microchannel heat sinks.

Commentary by Dr. Valentin Fuster
2005;():131-137. doi:10.1115/ICMM2005-75228.

Visualization experiments of convective boiling in transparent single micro conduits with the same hydraulic diameter but different cross sections are carried out with simultaneous measurement of local heat transfer coefficients and pressure losses. Two different cross sections with the same similar hydraulic diameters are applied: A circular microtube of 210μm in diameter and a square microchannel of 214μm × 214μm cross section. ITO/Ag thin film of 100 nm is sputtered on the outer surface of the conduits for the direct joule heating. The convective boiling shows some periodic variation of different flow patterns in both square and circular conduits. These flow patterns include bubbly, plug, slug, annular and capillary flows. The capillary flow pattern is the independent liquid droplets moving in the flow direction and very rarely observed in conventional tubes. The reason of such variation of flow patterns is that confined spaces limit the bubble growth in radial direction. So the nucleation bubble grows in both upstream and downstream and makes the flow pattern varies radically. The square microchannel conduit has more simple flow pattern variation, more nucleation bubbles and larger local heat transfer coefficients at lower vapor quality. It is due to that corners of the square microchannel act as helps nucleation cavities. Corners also promotes the formation of liquid film and the contact line between liquid and wall, which can stabilize the flow field. Local heat transfer coefficients decrease with increasing local vapor qualities. Local heat transfer coefficients increase with increasing boiling number but have their maximum value when boiling number reaches critical value. Such peculiar heat transfer characteristics can also be explained by the visualization results.

Commentary by Dr. Valentin Fuster
2005;():139-143. doi:10.1115/ICMM2005-75231.

Highly subcooled flow boiling at high heat flux in a rectangular minichannel of 2 mm × 2 mm in dimension was experimentally investigated. The channel was heated only from the bottom wall of the channel and it is regarded as a model of cooling components on electric devices in mobile gears. Under the condition of high subcooling and high heat flux in the stagnant pool and conventional scale channel flow, many fine vapor bubbles are emitted from primary vapor bubbles on heated surfaces. This boiling regime is called microbubble emission boiling (MEB). If we realize MEB in a minichannel, high heat flux due to vaporization will be obtained and large pressure loss due to confinement of bubble will be avoided. This paper first reports the results of our early experimental investigations on MEB in a pool boiling system to introduce the appearance of MEB. In flow boiling experiment, we reports heat transfer characteristics and photographic observation.

Commentary by Dr. Valentin Fuster
2005;():145-151. doi:10.1115/ICMM2005-75253.

In our previous investigations the formation of liquid bump of locally heated laminar liquid film with co-current gas flow was obtained [1,2]. The evaporation of liquid was left out of account. Heat transfer to the gas phase was approximately specified by a constant Biot number [2,3]. The aim of this work is an investigation of the evaporation effect, the hydrodynamics and the heat transfer of liquid film flow in a channel 0.2–1 mm height. The 2-D model of locally heated liquid film moving under gravity and the action of co-current gas flow with low viscosity in a channel are considered. The channel can be inclined at an angle with respect to horizon. It is supposed that the height of the channel is much less than its width. Surface tension is assumed to depend on temperature. The velocity profiles for gas and liquid regions are found from problem of joint motion of isothermal non-deformable liquid film and gas flow. Using the findings the joint solution of heat transfer and diffusion problem with corresponding boundary condition is calculated. Having the temperature field in the whole of liquid and gas flow region we find a local heat transfer coefficient on the gas-liquid interface and Biot number as a function of flow parameters and spatial variables.

Commentary by Dr. Valentin Fuster

Heat Pipes

2005;():153-158. doi:10.1115/ICMM2005-75008.

Recently microchannels or minichannels are widely used for cooling the high density power electronic devices. Especially, the channels are used in small and high efficient equipments such as heat pipes and heat exchangers. The interfacial behaviors between liquid and gas phases are very important to understand the flow characteristics in micro or minichannels. One of the problems at the CPL heat pipes is the backflow of vapor through the liquid line toward the evaporator. Therefore, we applied the converging minichannels at the condenser to avoid the backflow. In this paper, a numerical analysis on the interfacial velocities and the total pressures depending on variations of the converging rate and the depth of the minichannel was performed. And it was shown that the converging minichannel can reduce the backflow effect. Eventually the proper dimension of the minichannel to reduce the vapor backflow was found. In the numerical experiment, CFD-ACE+, a commercial program, was used.

Commentary by Dr. Valentin Fuster
2005;():159-165. doi:10.1115/ICMM2005-75045.

As reduction in the size of electronics creates demand for smaller, less expensive and faster-to-produce spacecraft, the use of high heat flux electronics or advanced nuclear propulsion systems will increase the stress on the thermal subsystem. This work presents a thermal management solution to this problem using liquid-cooled microchannel heat sinks. First, a simple computer model is used to illustrate the need for an atypical cooling method when high-heat flux electronics are used. Second, a thermal/fluid model of microchannel heat sinks is developed and applied to address the satellite thermal need. The total thermal resistances and pressure drops show excellent comparison with published experimental and analytical results. Finally, the model of the microchannel heat sink is optimized to remove 25 W/cm2 over a footprint of 3.7cm2 . The mass flow rate needed was significantly lower (almost 5–10 times lower) when compared to other published results, which means that micro-pumps available on the market will be sufficient. The integration of the microchannel system with the satellite is also discussed.

Commentary by Dr. Valentin Fuster
2005;():167-174. doi:10.1115/ICMM2005-75066.

A simple design micro-heat pipe was proposed. It was composed of a 20.0 × 20.0 mm square flow circuit which had two adjacent narrow-sides (1.0 × 1.0 mm2 or 0.5 × 1.0 mm2 ) and two adjacent wide-sides (5.0 × 1.0 mm2 or 2.5 × 1.0 mm2 ). A heating spot was at the narrow side and a cooling spot was at the wide side. Working fluid was ethanol. The flow circuit was placed horizontally. Bubbles generated at the heating spot migrated toward the wide side, the bubbles coalesced there to form a large bubble, and then the large bubble moved to the cooling spot. Finally, the large bubble was condensed at the cooling spot. This cycle repeated continuously. As a result of it, heat transport from the heating spot to the cooling spot was produced in the micro heat pipe even if it was arranged horizontally. It was confirmed that this simple device works as the heat pipe. An analysis of a flow mechanism was performed by solving a simple flow equation based on the flow resistance. It was proved that one-way circulation flow could be formed in the flow circuit. Predicted flow velocities were close to measured velocities. The heat transport performance of the proposed micro heat pipe was much better than the heat conduction of a stainless steel plate.

Topics: Heat pipes
Commentary by Dr. Valentin Fuster
2005;():175-181. doi:10.1115/ICMM2005-75115.

In order to elucidate the effects of working fluid’s properties on the heat transport capacity of a micro heat pipe, 3 commonly used fluids are selected for this study: water, ammonia and methanol. From the results obtained, it shows that for operating temperatures lower than 50°C, ammonia is preferred, but if the operating temperature exceeds 50°C, water is more suitable in transferring heat. Over the temperature range of 20°C∼100°C, the behavior of the heat transport capacity is found to be dominated by a property which is the ratio of the working fluid’s surface tension and liquid viscosity. This property which has the dimension of velocity has a controlling effect on the working fluid’s rate of circulation and therefore, the heat transport capacity.

Topics: Heat , Fluids , Heat pipes
Commentary by Dr. Valentin Fuster
2005;():183-189. doi:10.1115/ICMM2005-75122.

A series of tests have been carried out with a miniature loop heat pipe (mLHP), which has been developed for consumer electronics cooling, for horizontal and four vertical orientations under different sink temperatures. The mLHP has a cylindrical evaporator of 5 mm outer diameter and 29 mm length. The steady-state operating characteristics are similar for different orientations except for the orientation where the evaporator is above the compensation chamber. At an evaporator temperature of 75 °C, an evaporator heat load up to 70 W can be reached with thermal resistance of about 0.2 °C/W. The transient behavior of the mLHP is studied in detail. In general, the mLHP can be started up with very low power input (5 W). Big temperature oscillations in the liquid line were found in many cases, however, the temperature oscillations in the evaporator are minimum. The orientations greatly influence the operating characteristics of the mLHP. At least for the horizontal orientation, the overall performance of the tested mLHP is satisfying.

Commentary by Dr. Valentin Fuster
2005;():191-197. doi:10.1115/ICMM2005-75227.

Effects of cross-section geometry of capillary on the evaporation from the meniscus have been investigated by adopting several circular and rectangular capillaries. The evaporating meniscus shape, evaporation rate and flow near the evaporating meniscus of various liquids such as water, ethanol and methanol are determined. The shapes of water and ethanol menisci in circular capillary are quite different from each other due to the difference in surface tension. But the difference in meniscus shapes is relatively small in rectangular channel. The averaged evaporation fluxes in rectangular channel are much larger than that in circular capillary. The rotating vortex motion is observed near the evaporating menisci of ethanol and methanol except for the case of methanol in 200 × 20-μm capillary. The reason for this is considered to be the existence of the corner menisci at the four corners.

Topics: Evaporation , Geometry
Commentary by Dr. Valentin Fuster

Electronics Cooling

2005;():199-207. doi:10.1115/ICMM2005-75027.

This paper presents three generically similar impingement liquid coolers that have been engineered for cooling power electronics on future aero gas turbines. The thermal and hydraulic performances of the coolers have been compared with that of a commercial, state-of-the-art pin fin liquid cooler. It is demonstrated that the impingement liquid coolers outperform thermally the baseline pin fin cooler, and with significantly lower pressure drops. The impingement liquid coolers could also be easily modified to trade reduced pressure drop against higher flowrate or reduced thermal performance. A scaling model has also been developed to predict the thermal performances of the coolers for other types of coolants and flow conditions. The model has been applied for predicting the convective thermal performances of the coolers assuming hot aircraft fuel as the coolant. Future work would include an investigation of alternative convective applications in which the cooling system could be systematically explored.

Topics: Electronics , Coolers
Commentary by Dr. Valentin Fuster
2005;():209-214. doi:10.1115/ICMM2005-75068.

Increased processing speed, miniaturisation and higher packing densities of electronic processing chips has lead to the development of high power, high heat flux electronics cards used by the telecom and server industries to process digital signals. The drive towards higher system performance has put a large demand on forced air convection cooling techniques and is heading towards the thermal limit of the technology. Thermal management is becoming the limiting factor in the development of higher power electronic devices. To meet future heat transfer demands, innovative methods of thermally managing electronic devices are required. This paper looks at the creation of a test facility and single phase, prototype microchannel heat exchanger design to cool high power silicon and optical chips.

Commentary by Dr. Valentin Fuster
2005;():215-221. doi:10.1115/ICMM2005-75078.

A computational investigation of the heat transfer for a high performance integrated chip by using an electrohydrodynamic (EHD) pump was studied. This paper presents a fully computational system bundle with electro field, fluid flow and heat transfer for a cooling device. The micro pump provides the required pumping power by using the dipole moment generated from polarizing molecules and induces the flow to cool down the heat source. The computational domain of the micro channel for length and depth are kept in 1500μm and 500μm with parallel electrodes pitch (20μm, 40μm, 80μm). The effects of different applied voltage VE ranging from 100V to 500V, using oil as the working fluid and the heat flux of the heat source fixed at 2.5W/cm2 is investigated in detail. It is found that the EHD micro pump is more effective for lower channel pitch and higher applied voltage. For VE = 500V and electrodes pitch = 20μm, this study identifies a maximum performance of 49.36kPa in the pressure head and 9.55W/cm2 in the heat transfer. In addition, the performance of flow rate, liquid velocity and averaging Nusselt number for the specific condition are 0.94 L/min-mm2 , 0.12 m/s, and 106.10. However, it also identifies the performance of the heat transfer for electrodes pitch = 40μm is about 146.0% of that for pitch = 80μm. But for pitch = 20μm, it is only 10.5% higher than that for pitch = 40μm.

Commentary by Dr. Valentin Fuster
2005;():223-229. doi:10.1115/ICMM2005-75156.

Pressure drops and friction factors related to forced flow of de-ionized water over staggered/ in-line micro pin fin bundles (100-μm long, with hydraulic diameter of 50-μm and 100-μm, and with horizontal and vertical pitch ratios 1.5 and 5) possessing different cross-sections (circular/ diamond) have been experimentally investigated over Reynolds numbers ranging between 5–128. The applicability of conventional scale correlations to evaluate micro flow tests results has been assessed. It has been shown that available large-scale correlations do not adequately predict the pressure drop obtained at the micro scale.

Commentary by Dr. Valentin Fuster
2005;():231-236. doi:10.1115/ICMM2005-75239.

A heat sink consisting of microchannels of rectangular or trapezoidal cross-section through which a polar fluid is circulated by means of an electro-osmotic pump was studied numerically. The equivalent pressure head–volume flow rate curve was determined for both geometries and the influence of the aspect ratio was investigated. The dimensionless temperature profile was determined keeping also the effect of Joule heating into account. The cross-sectional Nusselt number was calculated for the above conditions and was found to be strongly influenced by the ratio of Joule heating to convective heat flux, Mz . The dependence of the Nusselt number on the dimensionless electro-osmotic diameter (kDh ) was also investigated for the two geometries and for increasing values of Mz , and a comparison with the values obtained analytically for slug flow under the same conditions was made. The value of the Nusselt number as a function of the aspect ratio was also calculated for increasing values of Mz . The numerical data presented in this paper can be useful to optimize the thermal performance of silicon micro heat-sinks.

Commentary by Dr. Valentin Fuster
2005;():237-248. doi:10.1115/ICMM2005-75241.

BTeV is a high-energy physics experiment, which is designed to proof several aspects of the so-called Standard Model. Precise measurements will reveal if the Standard Model contains breakdowns and therefore they will hint new matters for a more fundamental theory. One of BTeV’s main goals is to precisely measure CP violation in the beauty quark system. CP violation was first observed in strange quarks in 1963 and recently in beauty quarks at BaBar and Belle. CP violation causes particles and antiparticles to behave differently. The BTeV experiment was approved by FERMILAB and was currently being developed. In fact a very recent decision from the Department of Energy (February 2005) cancelled the project due to budget restrictions. A prototype of an innovative detector, called μ-strip detector, is under construction in a team leaded by an Italian group at INFN. The detector is based on copper strips deposited onto 300μm thick high resistivity bulk silicon. A hybrid electronic linked at the two terminals of the strips is positioned on the silicon layer. The system is inserted in a carbon fiber structure and then finally located around the particle beam. Even if the details of the electronic power dissipation and the chip geometry are not yet completely defined, the major constraints of the experiment (radiation hard structure, no mechanical vibration, high signal noise ratio with extremely low electrical charges, low atomic number of the components) have led the μ-strip team to make an effort in direction of a heat-sink based on a PEEK mini-tube system.

Commentary by Dr. Valentin Fuster
2005;():249-252. doi:10.1115/ICMM2005-75250.

A high heat flux cooling system using an array of micro-jets is presented. The Micro-Jet Cooling Array (MJCA) is based on an array of small diameter micro-jets, with jet diameters as low as 300 microns. The return flow in MJCA is arranged so that the effect of neighboring jets is essentially eliminated and the high heat fluxes achievable in a single jet are reproduced over large areas. MJCA has been machined using the LIGA process and has been extensively tested. Extremely high heat fluxes have been achieved with MJCA, making is suitable for a variety of high heat flux applications.

Commentary by Dr. Valentin Fuster

MEMS, Systems and Devices

2005;():253-260. doi:10.1115/ICMM2005-75029.

Surface structure modification and wettability control are very crucial for the advancement in microfluidic systems. The wettability control of the surface is achieved by depositing a Self-Assembled Monolayer (SAM) on the surface. The technology of SAM has been advanced by coating the selective interior regions of silicon wafer. This was accomplished using a mask wafer with open slots attached to define the areas that need to be coated. Besides the capability of controlling the wettability, structure modification such as controlling hole size leads to the magnification of the influence of surface tension force. As the size of the hole gets smaller, the surface tension could become a dominant force. This paper utilizes a unique approach to improve the water and air management at the cathode of a micro Direct Methanol Fuel Cell (DMFC). Both structure modification and local surface wettability control are utilized. A two-inch silicon wafer is formed of alternate strips of hydrophobic and hydrophilic zones with arrays of holes of different sizes. Water will be guided along the hydrophilic wetting zones with large hole openings; while the air goes into the cathode from the hydrophobic dry areas with smaller holes. This process ensures that the cathode is not flooded by water while air passes easily. The excess water at the cathode could be pumped back by a micro-pump to the anode so that the size of water storage will be minimized. Therefore, most of the storage reservoir space is used for the pure methanol, which enables the achievement of a high power density of the system. In this study, a silicon-based membrane is built accordingly to observe the water and air management. Images of the CCD camera showed clearly that the water drained from the big holes without blocking the air passages.

Commentary by Dr. Valentin Fuster
2005;():261-266. doi:10.1115/ICMM2005-75179.

We report a method of producing minute micro-droplets through nanoscale molecular self-assembly from a given liquid volume without external input. By the branching of hydrophobic restrictions, the drop is forced to separate while translating on a flat surface. Symmetry in the drop front wetted perimeter conduces to equal division of the drop. We show that at least 3 divisions can be performed sequentially on a 1.5 microliter drop to give minute droplets of approximately equal volumes. A division of carrier liquid volume by 1/23 enables multiple analyses on many separate stations in one microfluidic application.

Topics: Self-assembly
Commentary by Dr. Valentin Fuster
2005;():267-272. doi:10.1115/ICMM2005-75181.

In this study, the flow characteristics have been experimentally investigated for various shapes of the diffuser/nozzles, and the results are compared with the numerical simulation. Three chambers (inlet, exit, and middle) and two identical diffuser/nozzles are fabricated on a silicon wafer. The inlet and middle chamber is connected by one diffuser/nozzle; and the middle and exit chamber is connected by another diffuser/nozzle. The experiments are performed in a pump mode in which pressure is applied to the middle chamber and a supply mode in which pressure is supplied to the inlet and exit chamber. The net flow rate is determined by the flow difference between the pump and supply mode. The important parameters considered in this study are the throat width (30–12μm) and the taper angle (3.15–25.2°). For the taper angle and the throat width, it is found that there exists an optimum at which the net flow rate is the greatest. The optimal taper angle is in the range of 10–20° for all the pressure differences; and the throat width indicates an optimal value near 90μm for the case of 35kPa pressure difference. This tendency has been verified by the numerical simulation. From the numerical simulation, it is also found that the net flow rate is influenced by the size of the middle chamber. With decreasing chamber size, the net flow rate is reduced because of the interference between two streams flowing into the middle chamber.

Commentary by Dr. Valentin Fuster
2005;():273-279. doi:10.1115/ICMM2005-75207.

Parallel-plate and transverse comb-drive types of electrostatic microactuators are commonly used MEMS-based devices. Although they have the advantages of favorable scaling, fast response, and low power consumption, these electrostatic microactuators have had a fundamental limitation in that the allowable travel range is limited to 1/3 of the total gap between comb capacitor plates. Travel beyond this allowable range results in “pull-in” instability, independent of mechanical design parameters such as stiffness and mass. This paper presents the extension of stable travel ranges through the development of an active control system that stabilizes electrostatic microactuators and allows travel almost over the entire available gap between comb capacitor plates, providing a practical approach to extending travel range of electrostatic microactuators for applications that require high fill factors. The addressed challenges include the nonlinear dynamics of microactuators and system parameters that vary among fabricated devices. A nonlinear model inversion technique was proposed to address the nonlinear dynamics, which allows the use of traditional linear controller design methodologies for obtaining a desired linear system response. An adaptive controller was developed to provide improved position tracking in the presence of device parameter variations caused by fabrication imperfections. For experimental verification, the control system was implemented on a transverse comb-drive electrostatic microactuator fabricated using deep reactive ion etching on silicon-on-insulator wafers. Experimental results demonstrate that the resulting system is capable of traveling 4.0μm over a 4.5μm full range without “pull in.” Satisfactory tracking performance was obtained over a wide frequency band.

Commentary by Dr. Valentin Fuster
2005;():281-286. doi:10.1115/ICMM2005-75208.

A micropump was developed for a fluidic system that requires fluid transport in the 100+ μL/min flow range. The constraints on the design included the ability to control the flow rate over a reasonable range of flow rates, operate at fairly high pressure loads, and non-contact of the working fluid with the pump actuation system. The design was based on a displacement style pump, actuated by a piezoelectric element, with one-way polymer membrane check valves. The valves provided essentially zero backflow based on the elastic character of the material. Results are presented for three membrane thicknesses, three valve opening diameters, over a range of operating frequencies. The flow rate versus frequency curves show a characteristic trend with three regions of operation, the first a linear region at low frequencies, the second a region of decreasing slope resulting in a maximum flow rate regime and the third a region of reduced flow rate with increasing frequency. Results also show that a suction lift could be overcome with the micropump whose value depends on the valve size. The performance is compared to an ideal case indicating that larger diameter valves with thinner membranes obtain the best performance.

Commentary by Dr. Valentin Fuster

Mixing and Micro-Mixers

2005;():287-292. doi:10.1115/ICMM2005-75053.

In this paper, we present the acceleration of mixing and chemical reaction by a split-and-recombine (SAR) mixing method quantitatively, which was performed by numerical computation using the heat fluid analysis software, Star-CD. The authors have newly defined the mixing efficiency, which is a quantitative measure of the mixing of two fluids. The calculated result of the mixing efficiency in SAR device with two different channel configurations, angled and curved channels, showed that the secondary flow is important in increasing mixing efficiency. The angled channel is more effective than the curved channel, because the secondary flow is much stronger in the angled channel. The abrupt increase in sectional area also increases mixing efficiency. The split angle at the split point in the SAR device also affects mixing efficiency, because the secondary flow becomes stronger with the split angle. The mixing efficiency was greater (about 1.3 times) with a split angle of 45 degrees than that with a split angle of 15 degrees. According to the above-mentioned results, the authors designed a three-successive-SAR device, whose mixing efficiency was approximately 7.5 times greater than that of straight channel. The present findings are different from the existing mixing increase concept of the SAR device, which produces many thin layers from two fluids.

Commentary by Dr. Valentin Fuster
2005;():293-301. doi:10.1115/ICMM2005-75076.

Thermocapillary induced convection in thin liquid films on a horizontal wall with microgrooves is studied experimentally and numerically. To this end, we carry out experiments with silicon oil on a heated copper wall with parallel groves. The flow is visualized by tracking glass spheres seeded within the liquid film. The results of velocity measurements are reported. A numerical model for a liquid film on a structured wall is proposed. The full incompressible Navier-Stokes equations and the energy equation are integrated by a finite difference algorithm, whereas the mobile gas-liquid interface is tracked by the volume-of-fluid method. The numerical model is verified by comparison with the experimental data showing good agreement. The model is used to study flow patterns and film rupture caused by thermocapillary forces. We demonstrate that at any Marangoni number, either positive or negative, thermocapillary convection characterized by rolls develops within the film. In the experiments, two rolls in each groove are observed. The numerical solutions predict that at certain conditions the rolls are doubled under the influence of the wall structure guiding the flow. It is also found that an abrupt increase in wall temperature may rupture the liquid film near the structure crest. The results of this study may be applied to the design of microfluidic mixers and heat exchangers.

Commentary by Dr. Valentin Fuster
2005;():303-308. doi:10.1115/ICMM2005-75125.

Convective mixing in microstructures gives good mixing results in a very short time. In this work a theoretical and experimental study was performed on convective micro mixing in different mixing structures and their combinations. Various mixing elements had been integrated on a silicon chip to achieve a device for a high mass flow above 15 kg/h. These test structures were fabricated and tested concerning their flow behavior and mixing characteristics. Flow measurements with pH neutralisation and indication by Bromothymol Blue confirm the numerical simulations of the flow characteristics and mixing behaviour. The integral mixing quality in the micro mixer is measured with the iodide-iodate-reaction (Villermaux-Dushman) and shows excellent values for high Re numbers. This opens the potential of microstructures for new applications in the production of chemicals.

Commentary by Dr. Valentin Fuster
2005;():309-316. doi:10.1115/ICMM2005-75165.

Electro-osmotic flow (EOF) in microchannels is restricted to low Reynolds number regimes. Since the inertia forces are extremely weak in such regimes, turbulent conditions do not readily develop. Therefore, species mixing occurs primarily by diffusion, with the result that extended mixing channels are generally required. In this paper we present an investigation to predict the optimal applied voltage for the side channel type micromixer (SCTM) which is capable of continuous sample mixing for microfluidic applications. The device uses electrokinetical focusing which is an important EOF phenomenon. In this study, according to the conservation of mass, a simple theoretical model, based on the ‘flow-rate-ratio’ model and Kirchhoff’s law, is first proposed to predict the performance of the device. Computational fluid dynamics simulations are performed to investigate the effect of this model on the mixing efficiency. The results reveal that the mixing efficiency can be enhanced by using ‘flow-rate-ratio’ model and Kirchhoff’s law to predict the optimal applied voltage.

Commentary by Dr. Valentin Fuster
2005;():317-322. doi:10.1115/ICMM2005-75187.

This paper shows the microfabrication process of a split and recombination (SAR) micromixer and the effect of the cross-sectional rotation of fluidic interfaces, which is based on three dimensional microchannels composed of two poly-dimethysiloxane (PDMS) layers. When fluids pass through the nth SAR mixing unit, the number of interfaces increases to 2n+1 −1 through SAR mixing. The cross-sectional rotation of interfaces induced by the different time of expanding at slanted walls enhances mixing efficiency. The effect of the rotation is compared by three types of the SAR micromixer characterized by the existence and the direction of the rotation. The first type, No-R type, has flat walls which do not induce the rotation. The second type, Co-R type, has slanted walls which induce counter-clockwise rotations. The third type, Count-R type, has slanted walls and the counter-clockwise rotation alternates with the clockwise rotation. Water and blue dye is used in the mixing experiment as mixing fluids. Each inlet’s flow rate range is between 1 μl /min (Re0.1179) and 100 μl /min (Re 11.79). In No-R type, there is a clear contrast in interfaces which means that there is no rotation effect in SAR mixing process. In Co-R and Count-R, the contrast is unclear because of the rotation. Mixing in the SAR micromixer is almost complete after the 7th unit, whose length is 4200μm. The performance of the micromixer is estimated by numerical analysis using CFD ACE+. The flow rate and the diffusion coefficient are set 5 μl /min (ReO.5893) and 10−10 m2 /s, respectively for each inlet. The cross-sectional view of simulational SAR mixing agrees with experimental observations. The simulation estimates that the degree of mixing is more than 90% after the 6th unit of the No-R and Co-R type SAR micromixer.

Commentary by Dr. Valentin Fuster
2005;():323-328. doi:10.1115/ICMM2005-75199.

Micro technology supported aerosol processes provide a basis for integrated control of complex processes and therefore a promising research subject. A key aspect in aerosol technology is to control particle deposition, either to avoid clogging or to achieve a well defined coating of surfaces. As a first step we conducted an experimental and theoretical study of the particle deposition in a simple static T-shaped micro mixer. For the experiments monodisperse sodium chloride particles in the particle size range between 10 nm and 700 nm were used. The aerosol was introduced into one branch of the micro reactor and mixed with a particle-free air stream. The predominant particle deposition effect within the mixer is due to impaction, which is induced by the high curvature of stream lines at the inlet and in the mixing zone. Additional CFD calculations confirm the experimental results and show ways of optimizing the inlet geometry of the mixer, which should result in a significant reduction in impaction losses.

Commentary by Dr. Valentin Fuster
2005;():329-336. doi:10.1115/ICMM2005-75236.

Mixing in microfluidic channels or chambers is typically dominated by molecular diffusion. A common method used to evaluate mixing involves the examination of a time series of instantaneous concentration maps of fluid tracers (colorimetric or fluorescent) that enable visualization of fluid layering and simultaneous diffusive mixing. A scale often used to characterize micromixer performance is the global deviation of these concentration maps. While useful, this measurement scale does not provide a sensitive metric for evaluating fluid layering in the mixing process. This paper proposes an analytical approach that examines spatial concentration gradients and a global gradient-based scale, a normalized L2 norm of the gradient map, for micromixer performance evaluation. This gradient-based scale is complementary to deviation-based scales and is especially useful for the class of micromixers that enhance mixing by stretching and folding of fluids, whether the dominant mode of mixing is diffusion or chaotic advection. The algorithm is easy for micromixer designers to implement and will reveal performance metric information that remains implicitly hidden when deviation-based scales are used. The use of gradient-based mixing performance evaluation is illustrated with baker’s transform, a series of discrete mappings similar to kneading dough. The changes in both the deviation-based and gradient-based scale created by discrete fluid stretching and folding are discussed. The results from the one-dimensional discrete mixing problem are extended to a realistic mixing problem that simulates continuous stretching and folding.

Topics: Microfluidics
Commentary by Dr. Valentin Fuster
2005;():337-342. doi:10.1115/ICMM2005-75243.

Understanding and controlling stirring in micro-systems is necessary for the design of efficient passive micro-mixer. In this study, we focus on the dispersion of passive tracers injected in flows in between two rough surfaces under weak inertia influence (small but non-zero Reynolds number). The flow is induced by a constant applied pressure gradient between two cross-sections of the channel and the velocity field is calculated thanks to an extension of the lubrication approximation taking into account the first order inertial corrections. Tracers trajectories are obtained by integrating numerically the quasi-analytic velocity field. Our purpose is to examine the flow structure for various surface patterns and various Reynolds number. We focus on a simplified aperture field which is a smooth periodic function. This study puts forward interesting behavior of streamlines and show the dispersion of passive tracers in various geometries.

Commentary by Dr. Valentin Fuster

Microfluidics and Lab-on-a-Chip Devices

2005;():343-348. doi:10.1115/ICMM2005-75005.

Two-fluid flows in microchannel are often found in biological analysis, such as during ion exchange or solvent extraction from one phase to another. In this article, a numerical scheme is presented to describe a two-fluid flow in microchannel with electroosmotic (EO) effects. In this two-fluid system, the interfacial viscous force of a high EO mobility fluid drags a low EO mobility fluid; the high EO mobility fluid is driven by electroosmosis. We particularly analyze the electric double layer (EDL) regions close to the wall and the interface in the high EO mobility fluid. As the governing equation of the electrical potential is singularly perturbed, finer meshes are adopted to capture these EDL regions. In simulation, the interface between the two fluids evolves along the flow direction as the flow develops. Level set method is used to capture the interface implicitly. A localized mass preservation scheme is used to ensure mass conservation. A finite-volume method is used to solve the coupled electric potential equation, level set equations and Navier-Stokes equation. The validity of the numerical scheme is evaluated by comparing its predictions with the results of the analytical solutions in the fully developed regions. The interface positions; pressure gradients; mass flow rates and velocity profiles of the two fluids along the channels are obtained numerically.

Commentary by Dr. Valentin Fuster
2005;():349-354. doi:10.1115/ICMM2005-75015.

A bubble-powered micropump was fabricated and tested in this study. The micropump consists of a pair of nozzle-diffuser flow controller and a pumping chamber. The two-parallel micro line heaters were fabricated to be embedded in the silicon dioxide layer above a silicon wafer which serves as a base plate for the micropump. The pumping chamber, the pair of nozzle-diffuser unit and microchannels including the liquid inlet and outlet port were fabricated by etching through another silicon wafer. A glass wafer having two holes of inlet and outlet ports of liquid serve as upper plate of the pump. The sequential photographs of bubble nucleation, growth and collapse were visualized by CCD camera. Clearly liquid flow through the nozzle during the period of bubble growth and slight back flow of liquid at the initial period collapsing can be seen. The mass flow rate was found to be dependent on the duty ratio and the operation frequency. The mass flow rate decreases as the duty ratio increases in the micropump with either circular or square pumping chamber.

Commentary by Dr. Valentin Fuster
2005;():355-360. doi:10.1115/ICMM2005-75017.

We studied experimentally the particle electrophoretic motion in converging-diverging microchannels on a poly(dimethylsiloxane) (PDMS) chip. The whole process of particle acceleration and deceleration was visualized through traditional optical microscopy, with which the accelerated particle electrophoretic separation is demonstrated. The effects of electric field, particle size, particle moving passage, and channel configuration on particle electrophoretic motion are examined individually. We find that the ratio of particle velocity in the throat to that in the straight channel is insensitive to both the particle moving passage and the length of converging/diverging channel, but increased for smaller particles moving through symmetric converging-diverging channels under lower electric fields. Moreover, we find that the particle velocity ratio in electrically driven flows is significantly lower than the cross-sectional area ratio of the straight channel to the throat. We have attributed this discrepancy to the particle-induced distortion in the electric potential distribution. The computed contour of electric field in a converging-diverging microchannel has revealed that the electric field is locally higher around the two poles of a particle than all other regions inside the channel.

Commentary by Dr. Valentin Fuster
2005;():361-368. doi:10.1115/ICMM2005-75021.

There has been tremendous interest in developing micro technologies towards the integration and automation of Biochips or Lab-on-a-Chip devices due to their wide range of applications in environmental, chemical and biomedical engineering fields. The laminar flow nature in microfluidic devices offers opportunities to microfabricate the desired structures inside microchannels and pattern culturing medium inside microchannels. However, no analysis tools are available to provide optimized configurations for control the flow for microfabrication. Therefore, the goal of this study is to develop a numerical model to study transport phenomena in a cross-linked microchannels aiming to explore an optimized configuration for the microfabrication of specific desired features inside microchannel networks through investigating the effects of controlling parameters on the multistream flow. In this study, electroosmotic flow with induced pressure-driven flow will be employed. This model consists of a set of equations describing the applied potential field, flow field and concentration field in such geometries. The effects of various operational parameters are investigated based on the simultaneous solution to this model, to explore optimized configurations for flow and mass transport control in crossing linked microchannels.

Commentary by Dr. Valentin Fuster
2005;():369-376. doi:10.1115/ICMM2005-75023.

Three-dimensional lattice Boltzmann method based simulations of a microduct have been undertaken in this paper. The objective is to understand the different physical phenomena occurring at these small scales and to investigate when the flow can be treated as two-dimensional. Towards this end, the Knudsen number and aspect ratio (depth to width ratio) are varied for a fixed pressure ratio. The pressure in the microduct is non-linear with the non-linearity in pressure reducing with an increase in Knudsen number. The pressure and velocity behaves somewhat similar to two-dimensional microchannels even when the aspect ratio is unity. The slip velocity at the impenetrable wall has two components: along and perpendicular to the flow. Our results show that the streamwise velocity near the centerline is relatively invariant along the depth for aspect ratio more than three, suggesting that the microduct can be modeled as a two-dimensional microchannel. However, the velocity component along the depth is never identically zero, implying that the flow is not truly two-dimensional. A curious change in vector direction in a plane normal to the flow direction is observed around aspect ratio of four. These first set of three-dimensional results are significant because they will help in theoretical development and flow modeling at micro scales.

Commentary by Dr. Valentin Fuster
2005;():377-384. doi:10.1115/ICMM2005-75033.

A backwards-Euler time-stepping numerical method is applied to simulate the transient response of electroosmotic flow in a curved microtube. The velocity responses of the flow fields induced by applied sinusoidal AC electric fields of different frequencies are investigated. The transient response of the system is fundamentally important since both the amplitude and the time duration of the transient response must be maintained within tolerable or prescribed limits. When a sinusoidal AC electric field is applied, the transient response of the output velocity oscillates in the time-domain. However, after a certain settling time, the output velocity attains a sustained oscillation with the same amplitude as the driving field. In this study, the transient response of the electroosmotic flow is characterized by the time taken by the velocity response to reach the first peak, the peak of the sustained oscillation, the maximum overshoot, the settling time, and the bandwidth of the sustained oscillations in the time-domain. Meanwhile, the performance of the system is identified by plotting the output velocity response and the output velocity phase-shift against the frequency of the applied signal. A finite time is required for the momentum to diffuse fully from the walls to the center of the curved microtube cross-section. As the applied frequency is increased, the maximum overshoot and the bandwidth and peak of the sustained oscillations gradually decrease since insufficient time exists for the momentum to diffuse fully to the center of the microtube. Additionally, the phase-shift between the applied electric field and the output velocity response gradually increases as the frequency of the applied signal is increased.

Commentary by Dr. Valentin Fuster
2005;():385-390. doi:10.1115/ICMM2005-75041.

Microfluidic devices can take advantage of novel physics only available at the micro-scale. As a result, new physical models for the flow behaviour within these devices are required. In recent years micron resolution particle image velocimetry (Micro-PIV) has established itself as a reliable verification tool for these models. Micro-PIV measures the velocity profile across a single 2D plane within a microfluidic device. Here, Micro-PIV data is obtained from several planes within a device to create a complete mapping of the flow field. Combination of the data from all the planes allows the flow profile on the plane perpendicular to the original images to be studied and the volumetric flowrate through the device to be measured. By fitting the measured velocities to known flow profiles, the width and depth of the device can also be measured with sub micron precision. The technique is applied to an isotropically etched microchannel, where the results are compared to an independent computational fluid dynamics solution. Good agreement is found between the two data sets. The technique is then demonstrated in a more complex flow in a mixing channel device. In order to image an entire device a large field of view may be required. This dictates the use of low magnification lenses, which in turn have a large depth of field. A theoretical model for the measured Micro-PIV velocity demonstrates that, if care is taken when measuring near to curved walls, the use of these low magnification lenses does not significantly reduce the quality of the data obtained.

Topics: Microfluidics
Commentary by Dr. Valentin Fuster
2005;():391-396. doi:10.1115/ICMM2005-75044.

This paper investigates the modeling of styrene free radical polymerization in two different types of micro-mixer for which the wall temperature is kept constant. The simulations are performed with the help of the finite elements method which allows solving simultaneously partial differential equations resulting from the hydrodynamics, thermal and mass transfer (convection, diffusion and chemical reaction). The different micro-mixers modeled are on one hand an interdigital multilamination micro-mixer with a large focusing section and on the other hand a simple T-junction with three different radii followed by a tube reactor having the same radius. The results are expressed in terms of reactor temperature, polydispersity index, number-average degree of polymerization and monomer conversion for different values of the chemical species diffusion coefficient. Despite of the heat released by the polymerization reaction, it was found that the thermal transfer in such microfluidic devices is high enough to ensure isothermal conditions. Concerning the polydispersity index, the range of diffusion coefficients over which the polydispersity index can be maintained close to the theoretical value for ideal conditions increases as the tube reactor radius decreases. The interdigital multilamination micro-mixer was found to act as a T-junction and tube reactor of 0,72 mm ID but gives up to 15% higher monomer conversion. This underlines that the use of microfluidic devices can lead to a better control of the polymerization.

Commentary by Dr. Valentin Fuster
2005;():397-402. doi:10.1115/ICMM2005-75059.

A novel technique based on direct thick film writing has been developed for the rapid prototyping of microfluidic devices. The direct writing process is based on pressure driven dispensing of precursor materials through a micro-capillary tip. The process exhibits wide latitude in both the materials that can be patterned and the substrate formats and shapes that can be accommodated. A fabrication process flow sequence with general applicability to microfluidic devices was developed and its efficacy was demonstrated by the construction of two-input mixer devices. Integration of fluidic components with electrical circuitry was also demonstrated.

Commentary by Dr. Valentin Fuster
2005;():403-410. doi:10.1115/ICMM2005-75062.

In this article we report on a planar miniaturized dielectrophoretic (DEP) microfluidic device developed for the purpose of continuous fractionation and purification of sample suspensions of microscopic particles or biological cells, employing specially shaped nonuniform (isomotive) electric fields. The device integrates three fully functional and distinct sub-units consisting of 1) sheath and sample injection ports, arranged to achieve hydrodynamic focusing of the cell stream; 2) the DEP fractionation region and 3) two sample collection ports. In the DEP fractionation region, the magnitude of the field induced DEP force acting on the particle is essentially constant and independent of the particle’s position and furthermore only dependent on the intrinsic polarization response of the particle, for identical sized particles. The operation and performance in terms of sample throughput, separation efficiency and repeatability of the device was evaluated using test microscopic sized dielectric particles and biological particles, including cancerous cell lines.

Topics: Microfluidics
Commentary by Dr. Valentin Fuster
2005;():411-418. doi:10.1115/ICMM2005-75063.

Dielectrophoresis (DEP) has been successfully applied and demonstrated to provide novel and non-invasive means for characterizing, manipulating, trapping, separating and isolating microscopic sized particles, including biological cells. In this article, we report on the design, fabrication and performance of a novel, low cost, integrated Poly(dimethylsiloxane) (PDMS)/DEP microfluidic device capable of controlled manipulation of microscopic sized cells and particles that can be simultaneously utilized both for DEP spectral analysis and cell sorting. We have prototyped microfluidic channels, with DEP microelectrodes incorporated within PDMS channels. Previously, we have evaluated the operation and performance of a prototype device using various dielectric and biological particles, including yeast cells and polystyrene latex beads. In this paper, we report initial experimental observations on malignant cancerous cells. Non-viable cells, due to positive DEP, were attracted to the planar electrodes at frequencies between 200–600 kHz and were clearly repelled from the electrodes, due to negative DEP, at frequencies above 10 MHz.

Commentary by Dr. Valentin Fuster
2005;():419-426. doi:10.1115/ICMM2005-75079.

A reduced-order model is developed allowing for a fast computation of the temperature field in multichannel microreactors. The model regards the fluid and the solid phase as interpenetrating continua and incorporates heat exchange between the two phases via a heat-transfer coefficient characteristic for the channel geometry under study. The geometry of the channel walls determines the components of the thermal conductivity tensor which govern conductive heat transfer to the envelope of the reactor. The mean-field model is solved numerically for a test case inspired from practical applications. Parallel to that, a detailed model is set up with the purpose to benchmark the results of the mean-field model. This full model incorporates the geometric details of the multichannel reactor and contains consider-ably more degrees of freedom than the mean-field model, resulting in a much larger computational effort. It is found that the temperature fields computed with the two models agree reasonably well. Thus, the mean-field model appears as an efficient tool to evaluate the thermal performance of multichannel microreactors, especially in the context of parameter studies or system optimization.

Topics: Heat transfer
Commentary by Dr. Valentin Fuster
2005;():427-432. doi:10.1115/ICMM2005-75080.

This paper presents a simple microheater design for microfluidic devices by embedding resistance wire into a PDMS chip, and the results of an experimental study of the thermal response of liquid samples in the PDMS chip with the embedded local heater. Temperature-dependent fluorescent dye was used to measure the temperature distribution within a microchannel heated by the local heater. Two heater configurations were built, tested, and compared with numerical simulation. Through comparing the performance of these two configurations, heating and cooling rates and uniformity of the temperature field were evaluated. Additionally, thermal cycling at two different temperature levels was achieved by controlling the power of the local heater.

Commentary by Dr. Valentin Fuster
2005;():433-440. doi:10.1115/ICMM2005-75102.

The Joule heating induced temperature development and its effects on the electroosmotic flow in a capillary packed with microspheres is analyzed in this paper using finite-difference based numerical method. The model incorporates the coupled momentum equation for the electroosmotic velocity, the energy equations for temperature distributions, and the mass and electric current continuity equations. The temperature-dependent physical properties of the electrolyte solution are taken into consideration. The simulation predicts that, in the presence of Joule heating, there exists a significant axial temperature gradient in the thermal entrance region. This high temperature gradient strongly enhances the local electric field at the entrance, resulting in a non-uniform distribution along the flow direction. The temperature shows a parabolic radial profile but the gradient is very small due to the small system Biot number. The non-uniform temperature distribution in turn greatly affects the EOF velocity by means of changing the local viscosity and the dielectric constant of the fluid phase, and the local electric field strength. The results by this model are found to be in a good agreement with published analytical and experimental works in the literature.

Commentary by Dr. Valentin Fuster
2005;():441-448. doi:10.1115/ICMM2005-75104.

The topic of single phase liquid flow in submicron or nanochannels is a nascent field. There have only been a couple papers that have dealt with this area directly. The most probable reason for this is that currently most research in fluid mechanics or heat transfer is being focused on micron size channels. To help facilitate researchers to focus on this undeveloped area, this paper serves as a review for some of the micro-fabrication processes that will make it possible for engineers and scientists to study this field in greater detail.

Commentary by Dr. Valentin Fuster
2005;():449-456. doi:10.1115/ICMM2005-75113.

Numerical simulation based a design tool devised by an aposteriori error estimation technique termed the ‘bound method’ is applied in this paper to examine and analyze fluidic flow and species transport phenomena of the electro-osmotic flow in a heterogeneously charged T-shaped micro-mixer. This novel technique provides fast, inexpensive and reliable bounds to the ‘output’ for the slip velocity model of the electro-osmotic flow. The bound method presented here-in is extended to use of an adaptive refinement to generate the ‘ideal’ mesh for computations, the direct equilibration to evaluate the ‘hybrid-flux’ very efficiently, and a parallel calculation to accelerate the error estimates of the output of interest.

Commentary by Dr. Valentin Fuster
2005;():457-464. doi:10.1115/ICMM2005-75116.

Creation of concentration gradients is important in the study of biological and chemical processes that are sensitive to concentration variations. This paper presents a simple method to generate a linear concentration gradient in electroosmotic flow in microchannels with converging and diverging geometries. The method is based on the enhanced diffusive mixing inside the microchannel. By varying the converging-diverging geometries, the degree of diffusive mixing can be controlled. Different concentration gradients can be obtained by varying the applied potential and the geometry. Concentration profiles with minimal axial variations can be achieved with a deviation of 7% and 3% over a channel length of 3mm and 1mm, respectively, for a 400μm wide microchannel. Although the underlying physics and mechanisms for creating concentration profiles in a converging-diverging microchannel are the same as a T-shaped micromixer, the converging-diverging microchannel can produce desired concentration profiles in a much shorter distance (shorter by a factor of 2∼3.5 compared to a T-shape mixer). A serially connected concentration gradient generator is also realized with the ability to generate two concentration gradient ranges in the same microchannel. Numerical simulations and experiments were carried out to investigate the factors contributed to the generation of the concentration gradients.

Topics: Microchannels
Commentary by Dr. Valentin Fuster
2005;():465-472. doi:10.1115/ICMM2005-75119.

Concentration gradient in a chamber appended to a microchannel is important to cell movement control and to the concentration gradient based assays on Lab-on-a-Chip devises. In this paper, the effects on the concentration field of the asymmetrical injection, the Peclet number, the mobility ratio of electrophoresis to electroosmosis, the chamber’s downstream position, and the chamber’s geometry parameters, are investigated. The most sensitive parameter is the asymmetrical injection, which can increase the concentration gradient twice as large as to that in the symmetrical injection. Furthermore, using heterogeneous surface patches is a very effective way to enhance the concentration gradient generated in the chamber. Different patches for certain chambers are investigated. Finally, experimental visualization of the concentration fields was conducted, and good agreements were found between the numerical simulation results and the experimental results of the concentration fields generated in a micro-chamber with/without a heterogeneous patch.

Topics: Microfluidics
Commentary by Dr. Valentin Fuster
2005;():473-480. doi:10.1115/ICMM2005-75132.

Tailor-designed AC electro-osmotic (AC-EO) stagnation flows are used to convect bioparticles globally from a bulk solution to localized dielectrophoretic (DEP) traps that are aligned at flow stagnation points. The multi-scale trap, with a typical trapping time of seconds for a one cc sample, is several orders of magnitude faster than conventional DEP traps and earlier AC-EO traps with disjoint electrodes. A novel serpentine wire resistor loop capable of sustaining a high field, up to 20,000 V/cm, is fabricated to produce strong AC electro-osmotic flow with two separated stagnation lines, one aligned with the field minimum and one with the field maximum. The continuous loop design allows a large applied voltage without inducing Faradaic electrode reactions. Particles are trapped within seconds at one of the traps depending on whether they suffer negative or positive DEP (n-DEP, p-DEP). The particles can also be rapidly released from their respective traps (and recaptured in the opposite traps) by varying the frequency of the applied AC field below particle-distinct cross-over frequencies. Zwitter ion addition to the buffer allows further geometric and frequency alignments of the AC-EO and DEP motions. The same device hence allows fast trapping, detection sorting and characterization of a sample with realistic conductivity, volume and bacteria count.

Topics: Wire , Convection , Design
Commentary by Dr. Valentin Fuster
2005;():481-486. doi:10.1115/ICMM2005-75137.

In this work, micro particle imaging velocimetry (micro-PIV) was performed on the fundamental components of a micro total analysis system. Specifically, high aspect ratio passive valves and mixers were designed, fabricated, and characterized. The components were built using Micralyne Protolyne technology on a glass substrate and operated at reasonably achievable pressures. The flows through the components were analyzed both qualitatively and quantitatively with the goal of developing a more complete understanding of internal device performance. Using the results of the micro-PIV developed velocity fields it was found that the high aspect ratio passive valves are able to perform at reasonably achievable pressures. However, it was determined that the high aspect ratio passive mixers offer limited performance enhancements because of the low Reynolds number flows. The results of this work contribute to the understanding of passive component operation and address some of the challenges associated with developing completely integrated micro total analysis systems that use passive devices.

Commentary by Dr. Valentin Fuster
2005;():487-493. doi:10.1115/ICMM2005-75145.

An electrokinetic micro-pump fabricated by a sol-gel process has been designed which can be used as a robust fluid-driving unit on a chip-scale analytical system. An overall monolithic silica matrix with morphology of micron-scaled through pores was formed within 100-μm inner diameter fused silica capillary. This pump utilizes electroosmotic flow to propel liquid solution with no moving parts. The Nafion® house design in the cathode chamber separates the electrolytic bubble interference from flow channels. The maximum flow rate and maximum pressure generated by the pump are 3.0 μL/min and 3.5 atm, respectively, at 6 kV. The flow rate can be controlled in the range 200 nL-3.0 μL/min by adjusting applied electric filed. As the monolith is silica-based, this pump can be used for a variety of fluids, especially for organic solvents such as acetonitrile and methanol, without swelling and shrinking problems. These results indicate that the pump can provide sufficient pressure and flow for micro-total-analysis systems (μTAS).

Topics: Bubbles , Design , Pumps
Commentary by Dr. Valentin Fuster
2005;():495-502. doi:10.1115/ICMM2005-75148.

A novel automatic electrokinetically-controlled immunoassay lab-on-a-chip was developed in this paper. The microchip was made of poly(dimethylsiloxane) (PDMS)/glass using photolithography and replica molding. The immunoassay technique using anti-Helicobacter pylori antibody was applied to detect H. pylori protein antigens. Rhodamine-labeled secondary antibody was employed for signal generation. Experiments were first conducted on a straight microchannel to prove the feasibility of an electrokinetically-driven immunoassay. The detection limit for the coating antigen was found to be 1 ng/μL. Automatic electrokinetically-controlled immunoassay experiments were further carried out on a microchannel network. Numerical simulation and experimental studies were combined for the first time to demonstrate an integrated, electrokinetically-controlled immunoassay lab-on-a-chip. The electrokinetically driven, time-dependent reagent delivery processes were simulated using finite element method (FEM). Fully automatic on-chip experiments were accomplished by sequentially changing the applied electric field. It was found that the lab-on-a-chip can realize much shorter assay time, reduced reagent consumptions and automation while the detection limit is better than the conventional colorimetric immunoassay.

Commentary by Dr. Valentin Fuster
2005;():503-509. doi:10.1115/ICMM2005-75154.

For the first time we report a series of micro air bubble manipulations (transporting, merging and eliminating of air bubbles) in two-dimensional microchannels filled with a water solution. Air bubbles (∼1.2 mm3 in volume) are driven by electrowetting-on-dielectric (EWOD) principle. By sequentially energizing an array of electrodes covered with dielectric layers, micro air bubbles can be transported along a programmed path, merged into a single larger one, and eliminated out of a bulk water solution. These three fundamental micro bubble operations demonstrated will facilitate efficient micro bubble handling essentially required in micro gas analysis systems.

Topics: Bubbles
Commentary by Dr. Valentin Fuster
2005;():511-515. doi:10.1115/ICMM2005-75162.

This paper examines the geometries of basic straight microneedle arrays, slanted channel arrays with varying angles, and arrays with diverging and converging interior cross sections for the purpose of interstitial fluid extraction and transdermal drug delivery. Flow behaviour is analyzed under biometric pressure driven conditions including frictional losses, minor losses due to the array geometry, and losses due to electrokinetic effect in microchannels. This paper also presents design and fabrication details of preliminary work that will lead to a design for microneedle arrays.

Commentary by Dr. Valentin Fuster
2005;():517-522. doi:10.1115/ICMM2005-75170.

In this paper, we studied liquid-solid slip by employing a mean-field free-energy lattice Boltzmann approach recently proposed [Zhang et al., Phy. Rev. E. 69, 032602, 2004]. With a general bounce-back no-slip boundary condition applied to the interface, liquid slip was observed because of the specific fluid-solid interaction. The slip length is clearly related to the interaction strength: the stronger the interaction, the less hydrophobic the surface and hence results in less slipping. Unlike other lattice Boltzmann models, a contact angle value between 0–180° can be generated here without using a less realistic repulsive fluid-solid interaction. We found that system size does not affect the absolute slip magnitude; however, the ratio of the slip length to system size increases quickly as the system becomes smaller, illustrating that slip becomes important in smaller-scale systems. A small negative slip length can also be produced with a strong fluid-solid attraction. These results are in qualitative agreement with those from experimental and molecular dynamics studies.

Commentary by Dr. Valentin Fuster
2005;():523-528. doi:10.1115/ICMM2005-75171.

In this paper, we have studied the droplet movements and continuous flows confined between two rough and hydrophobic surfaces. A recently proposed mean-field free-energy lattice Boltzmann model was employed. The movement of contact point over a well-patterned rough surface displays a periodic sticking-jumping-slipping behavior; while the dynamic contact angle changes accordingly from maximum to minimum values. These complex varying behaviors are totally different from those on flat surfaces and implies more carefulness is necessary in interpreting measured contact angles on rough surfaces. Two regimes were found of the droplet velocity changing with the surface roughness: first decreasing and then increasing; and qualitative analysis was given. We have also studied the continuous flow rates and two cases, with and without vapor trapped, were compared. Simulation results show the vapor trapped can indeed reduce the resistance to fluid motion from the channel surfaces, and such information could be useful for microfluidic applications.

Commentary by Dr. Valentin Fuster
2005;():529-534. doi:10.1115/ICMM2005-75174.

Electrokinetic (EK) pumping efficiency has recently been a hot topic due largely to its applications in microfluidics for liquid transport. As the Onsager reciprocity relation suggests that the exact opposite of EK pumping is EK generation, experimental study on the efficiency of the latter is seldom reported. This paper examines electrokinetic generation efficiency in the presence of finite external loads through the use of different ceramic filters. The electrical output power was measured experimentally as a function of different external loadings. When the experiments were performed continuously on subsequent days, the EK generation efficiency for all the tested filters can be made to increase and approach a maximum limit. The experimental data also indicated that, when pore size of the filter decreases, the curve for potential versus current becomes more non-linear and the maximum efficiency does not occur when the current is half the zero-potential current.

Topics: Stress
Commentary by Dr. Valentin Fuster
2005;():535-539. doi:10.1115/ICMM2005-75175.

Electrokinetic phenomena play an important role in microfluidic transport behavior. Review of literature suggests that surface energetic can also be an important factor, but rarely explored. Typically, surface energetic is taken into account by consideration as an arbitrarily selected slip boundary condition in the modified Navier-Stokes equation. In this paper, instead of selecting this arbitrary slip condition, we examine how solid-liquid energy parameters influence the transport of microfluidics in terms of streaming potential. The simultaneous effects of surface energetics and electrokinetics will be conducted by means of a mean-field free energy lattice boltzmann approach recently proposed. Rather than using the conventional Navier-Stokes equation with a slip condition, the description solid-liquid energetic is manifested by the more physical energy parameters in the mean-field description of the method. As a result, the magnitude of liquid slip can be related directly to the solid-liquid interfacial slip. These results will be employed in conjunction with the description of electrokinetic transport phenomena for streaming potential. The variation of streaming potential as a function of the energy parameters (solid-liquid interaction) is clearly demonstrated. In pressure-driven liquid microfluidics, the flow rate may be decreased due to the counter-effect between the electrokinetic and slip.

Commentary by Dr. Valentin Fuster
2005;():541-549. doi:10.1115/ICMM2005-75176.

Microfluidic systems have profoundly transformed chemical analysis, separation and detection techniques over the past decade by enabling rapid manipulation of extremely small volumes of fluid. Electrokinetic (EK) flow, i.e., flow of an electrolyte in narrow capillaries driven by the combined influence of electric field and pressure, is of significant interest in microfluidic devices. Review of literature reveals that most studies on microchannels are either for steady state solution or infinite length microchannels. In this paper, we examine the development of a transient streaming potential for pressure-driven EK flow in a finite length microchannel. A transient numerical simulation of ion transport leading to the development of a streaming potential across a finite length circular cylindrical microchannel connecting two infinite reservoirs is presented. The solution based on finite element analysis shows the transient development of ionic fluxes, currents, and the streaming potential across the channel. The influence of the entrance and exit effects on the evolution of the streaming potential is clearly depicted in this study. Our results will be employed to discuss some of the limitations of literature streaming potential analysis based on infinite length microchannels.

Topics: Microchannels
Commentary by Dr. Valentin Fuster
2005;():551-556. doi:10.1115/ICMM2005-75200.

A microfluidic chip-based solid phase extraction method for isolation of nucleic acids is demonstrated. The chip was fabricated in a cyclic polyolefin by hot embossing with a master. The solid phase was made by in-situ UV polymerization of a monolithic column impregnated with silica particles, and separation was achieved due to irreversible binding of the nucleic acids to the silica particles in the monolith. The porous monolithic column was formed within the channels of the device by photoinitiated polymerization of a mixture of methacrylate and dimethacrylate monomers, UV sensitive free radical initiator and porogenic solvent. The channel surface was photografted with a thin interlayer polymer prior to preparation of the monolith in the channel. The grafted layer covalently attached the monolith and prevented the formation of voids between the monolith and the channel surface. The solid-phase prepared by this method allowed for successful extraction and elution of nucleic acids.

Commentary by Dr. Valentin Fuster
2005;():557-562. doi:10.1115/ICMM2005-75204.

Direct measurement techniques are employed to quantify the kinematics of DNA flows in micro-contraction devices. Flow through micro-contractions subjects the fluid to large spatial gradients in velocity, thereby eliciting viscoelastic effects. Additionally, in this microfluidic flow environment, the fully extended length of the macromolecule L will approach the characteristic length scale of the channel geometry h. This is a unique flow environment that is not yet well understood. Knowledge of the fundamental physics that govern this flow regime will have a profound impact on optimization of lab-on-a-chip systems incorporating macromolecular flows. This study investigates the flow of semi-dilute λ-DNA solutions in a 2:1 micro-contraction where L/h ∼ 0.32. Video microscopy and streak images of semi-dilute DNA flows reveal large vortex regions in the corners of the contraction, which are indicative of strong elastic behavior. Velocity fields constructed using Digital Particle Image Velocimetry (DPIV) demonstrate the first use of this tool for obtaining velocity measurements of viscoelastic flows in microfluidic systems.

Commentary by Dr. Valentin Fuster
2005;():563-568. doi:10.1115/ICMM2005-75205.

Recent advancements in micro- and nanoscale fluidic manipulation have enabled the development of a new class of tunable optical structures which are collectively referred to as optofluidic devices. In this paper we will introduce our recent work directed towards the development of a spectrographic optofluidic memory. Data encoding for the memory is based on creating spectrographic codes consisting of multiple species of photoluminescent nanoparticles at discrete intensity levels which are suspended in liquids. The data cocktails are mixed, delivered and stored using a series of soft and hard-lithography microfluidic structures. Semiconductor quantum dots are ideally suited for this application due to their narrow and size tunable emission spectra and consistent excitation wavelength. Both pressure driven and electrokinetic approaches to spectral code writing have been developed and will be experimentally demonstrated here. Novel techniques for data storage and readout are also discussed and demonstrated.

Topics: Microfluidics
Commentary by Dr. Valentin Fuster
2005;():569-574. doi:10.1115/ICMM2005-75206.

Recent advances in the development of lab-on-a-chip devices have been rapid and broad ranging. In general however these devices, while containing micro- or even nano-scale components, rely heavily on macroscale infrastructure (e.g. microscopes, chip readers and power sources) to perform much of the actual product detection and subsequent analysis. As such to enable the next generation of portable lab-on-chip devices, techniques for simply and cheaply integrating on-chip analysis functionalities will be required. In this work we present our work directed towards the development of a new concept in rapid on-chip imaging which we refer to as “optofluidic microscopy (OFM)”. Here we present an overview of the imaging theory, fabrication procedure and operational details of the initial prototype. Preliminary experimental results of this on-chip optical imager are also reported. A significant advantage of the technique is that through proper spatial scaling, sub-wavelength resolution can be achieved without bulk optics.

Topics: Microscopy
Commentary by Dr. Valentin Fuster
2005;():575-578. doi:10.1115/ICMM2005-75213.

Each microstructured stainless steel foil was brazed in vacuum for stacking. Inner surface of micro channels was coated with Al2 O3 layer to support Pt catalyst by sol-gel method. The stack was designed like a cross-flow type heat exchanger to perform the combination of exothermic and endothermic reactions simultaneously. It is expected to apply to the micro reformer which produces hydrogen for micro PEMFC (Proton Exchange Membrane Fuel Cell). As the first step in our study, we measured experimentally the heat transfer rate and the spatial temperature distribution of the stack. An then, the reaction of C3 H8 -air with heat transfer to cold air flow was performed in the stack. As a consequence, quantitative and qualitative thermal characteristics of the stack for reaction were investigated.

Topics: Heat exchangers
Commentary by Dr. Valentin Fuster
2005;():579-583. doi:10.1115/ICMM2005-75219.

We present a proposed method to obtain three hematology parameters: red blood cell count, mean red blood cell volume and red blood cell distribution width on a microfabricated microchannel device. Detection will be conducted with an impedance device, in a series of three or more filter beds with channel dimensions on the order of 2–7 μm wide × 5 μm high × 5–15 μm long, and a total device volume on the order of 1 μl.

Commentary by Dr. Valentin Fuster
2005;():585-591. doi:10.1115/ICMM2005-75233.

SU-8 photopolymer is employed to fabricate interlocking structures through which high-density fluidic connections are provided between microchannel containing substrates. Each interconnect is composed of a compliant SU-8 cylinder that deforms to fit when inserted into a silicon or SU-8 hole that is 1–2 μm smaller in initial diameter. Interconnects up to 200 μm in outer radii in SU-8 films 200 μm thick were analyzed and simulated using ANSYS to determine relative strengths and weaknesses of different interconnect designs, as well as determine forces required for assembly and disassembly. Fabricated 200 μm radius interconnects were tested both mechanically and for fluid pressurization. First-order pull-out forces using weights were measured to be 35 mN for fully assembled 100 μm deep interconnects. Pressure testing on a microfluidic board resulted in a maximum pressure of 30 psi (200 kPa) with no leakage using a computer controlled pressure source and water.

Commentary by Dr. Valentin Fuster
2005;():593-597. doi:10.1115/ICMM2005-75234.

Rapid electric field switching is an established microfluidic mixing strategy for electrokinetic flows. Many such microfluidic mixers are variations on the t- or y-channel geometry. In these configurations, rapid switching of the electric field can greatly improve initial mixing over that achieved with static-field mixing. Due to a fundamental lack of symmetry, however, these strategies suffer from lingering cross-channel concentration gradients which delay complete mixing of the fluid stream. Presented here is a field switching microfluidic mixing strategy which utilizes a symmetric sequential injector and an expansion chamber to achieve rapid and effectively complete microfluidic mixing. The three-inlet symmetric injector sequentially interlaces the two dissimilar incoming solutions. Just downstream of the injector, the sequence enters an expansion chamber and increased axial diffusion results in rapid mixing. The completely mixed solution is refocused into the outlet stream. The microfluidic chips are designed such that only the minimum number of independent fluid reservoirs is required. Chips are manufactured in polydimethylsiloxane using established soft-lithography based microfabrication methods. Fluorescence microscopy is employed to analyze, quantify and demonstrate the effectiveness of this mixing strategy, and determine a preferred operating frequency range. The microfluidic chip design is based on the findings of a recent numerical modelling based work that demonstrates the sequential injection micromixing concept.

Topics: Microfluidics
Commentary by Dr. Valentin Fuster
2005;():599-603. doi:10.1115/ICMM2005-75235.

Electrokinetically-driven flow circulations resulting from heterogeneous surface patches have previously been employed to improve mixing in microchannels. Here, numerical simulations demonstrate local in-channel hydrodynamic focusing through the use of strategically-patterned surface charge. Presented first is the case of a single straight channel with an axially-localized cross-sectional surface patch (ring). The surface patch exhibits a zeta potential equal in magnitude to the native microchannel surface but opposite in sign. The unsteady species transport in the presence of the electrokinetically-induced circulations is modelled, and a mean residence time is quantified. In general, residence times indicate the potential application of these circulations to microfluidic-based memory storage. Next, an improved focusing process for pinched-injection is demonstrated that exploits non-uniform surface patches. Lastly, surface patches are applied to enhance stream focusing in the microfluidic cross geometry. It is demonstrated that with this technique three-dimensional hydrodynamic focusing can be achieved in a single planar microfluidic structure. In one case, the microfluidic fluid stream was constrained to the centre of the channel and focused to 12% of its original cross-sectional area. Extensions of this work are discussed, as are the microfabrication and surface modification processes required for lab-on-chip implementation of these numerically simulated processes.

Topics: Microfluidics
Commentary by Dr. Valentin Fuster

Nanoscale Effects and Nanoparticles

2005;():605-612. doi:10.1115/ICMM2005-75004.

The paper features the mathematical model of calculation of thermal conductivity and viscosity for nanofluids on the basis of statistical nanomechanics. Calculation of transport properties for nanofluids for real substances is possible by the classical and statistical mechanics. Classical mechanics has no insight into the microstructure of the substance. The equations obtained by means of classical thermomechanics are empirical and apply only in the region under observation. Contrary to classical mechanics, statistical mechanics calculates the thermomechanic properties of state on the basis of intermolecular and intramolecular interactions between particles in the same system of molecules. It deals with the systems composed of a very large number of particles. For the first time in scientific literature are presented the analytical results for viscosity and thermal conductivity for nanofluids on the basis of statistical nanomechanics. The analytical results obtained by statistical mechanics are compared with the experimental data and show relatively good agreement.

Commentary by Dr. Valentin Fuster
2005;():613-616. doi:10.1115/ICMM2005-75172.

In this paper, we will demonstrate a selective surface patterning method by a micro-plasma discharge. In this method, argon plasma is ignited through a hole of copper clad polyimide microstructure electrodes. As an illustration, experiments were performed in which an octadecanethiol (CH3 (CH2 )17 SH) self-assembled monolayer (SAM) on a gold film is exposed to a microdischarge and subsequently followed by immersion into the 16-mercaptohexadecanoic acid (COOH(CH2 )15 SH) solution. The octadecanethiol SAM is desorbed upon Ar plasma exposure, allowing the formation of a second SAM on the damaged region [Chai et al, App. Phys. Lett., 86, 034107 (2005)]. The patterned samples are viewed by using optical microscope and scanning electron microscopy. The advantage of this approach is that it is noncontact and eliminates the need of photolithography. The patterned samples can be employed to microfluidic self-propelled movement.

Commentary by Dr. Valentin Fuster
2005;():617-620. doi:10.1115/ICMM2005-75177.

Alignment of carbon nanotubes (CNTs) has been studied in details for the past decades. In this paper, we examine the presence of single wall carbon nanotubes on experimental microfluidic systems during streaming potential measurements. The PMMA surface modified by CNTs has been characterized by high resolution scanning electron microscope (SEM), contact angles and streaming potential measurements. Preliminary results demonstrate clearly the effect of CNTs on microfluidic experiments in terms of streaming potentials.

Commentary by Dr. Valentin Fuster
2005;():621-626. doi:10.1115/ICMM2005-75178.

In the present paper, molecular dynamics is applied to study the effect of interaction between liquid-solid on liquid transport properties, especially a stagnant layer to a slip layer, in a circular nano tube under a constant injection flow rate boundary condition. In simulations, a full 12-6 Lennard-Jones potential, truncated at 2.2σ, is used to govern the interaction between liquid-liquid and liquid-solid molecules. Six cases with different interactions (or energy scales) between liquid-solid are carried out. The non-equilibrium molecular dynamics (NEMD) simulation results show that the interaction between liquid-solid causes a wavelike density distribution in normal wall direction and the amplitude of wave is proportional to the interaction between liquid-solid; the interaction between liquid-solid dominates the motion of first liquid layer close to the solid substrate, i.e., the strong interaction between liquid-solid can result in the a stagnant layer; the weak interaction between liquid-solid produces a slip layer and the intermediate interaction between liquid-solid made a slip and sub-slip layers.

Commentary by Dr. Valentin Fuster
2005;():627-633. doi:10.1115/ICMM2005-75210.

Development of nano-devices for various applications has drawn great attention recently driven by the need of miniaturizing the devices for the integration and automation of Biochips or Lab-on-a-Chip devices. Fundamental understanding of transport phenomena in nanofluidic channels is critical for systematic design and precise control of such devices. The goal of this study is to develop a theoretical model to study electroosmotic flow in nanochannels. Instead of using the Boltzmann distribution, the conservation condition of ion number and the Nernst equation are used in this new model to find the ionic concentration field in the nanochannels. A correct boundary condition for the concentration field at the wall of the channel is developed and the symmetry condition of the potential field at the center of the nanochannel is applied to this model. The ionic concentration field, electrical potential field and flow field are obtained by numerically solving this model. Comparisons of area-average velocity between the numerical simulations and experimental results reported in literature are provided.

Commentary by Dr. Valentin Fuster
2005;():635-640. doi:10.1115/ICMM2005-75211.

The average concentration of ions in a liquid depends on the size of a channel if the charges on solid surface do not change. The relation between them is that the concentration of ions is inversely proportional to the channel size. When a channel decreases from a micro to a nano size, the concentration of ions will increase 1000 times. In this case, the ion’s distribution in liquid may not be considered as dilute if the charges on solid surface is large, and interactions among ions have to take into account. In this paper, molecular dynamics is applied to study the effect of extra-pairs of positive/negative ions on liquid transport properties in a nano syringe under a constant injection flow rate boundary condition. In simulations, the Coulomb’s law and 12-6 Lennard-Jones potential are used to govern the interaction between ion-ion, ion-liquid, ion-solid, liquid-liquid and liquid-solid molecules. Four different cases (no ions, counter-ions, and counter-ions combining with small and large extra-pairs of positive/negative ions in liquid) are carried out. The non-equilibrium molecular dynamics (NEMD) simulation results show that the concentration of extra-pairs of positive/negative ions has significant influence on liquid velocity profile and ion distributions. For liquid flow without ions, a quasi-parabolic velocity distribution was obtained. When the counter-ions and extra-pairs of positive/negative ions are considered, the flow approaches a plug flow as the number of extra-pair of ions increases. We also found that charges in liquid do not follow the Poisson-Boltzmann distribution, especially for the net charges which have a valley located at about 1.5 molecular sizes away from the solid surface.

Commentary by Dr. Valentin Fuster
2005;():641-645. doi:10.1115/ICMM2005-75237.

We are reporting electroosmotic flows in nanochannels having different surface roughness. Molecular dynamics simulation technique has been applied to understand microscopic or molecular aspects of solid-liquid interactions. The surface roughness in this study was modeled as a succession of expanding and contracting steps along the flow direction, determined by the electric field direction. Water and sodium ion density profiles for the smooth wall show strong layering of water molecules near the solid wall. For rough walls, the density profiles are very similar for all cases except where the steps are located. We observed that this disturbance grows with the amplitude of the roughness and decreases with the period of the roughness. To further investigate strong layering of the water molecules, the net electric dipole moment was computed for the smooth and rough walls. It shows ordering of water molecules near the wall. The rough wall result show water molecules are ordered more in the expanded region. The velocity profiles and the flow rate were calculated for all cases. From these, electric double layers are found to overlap. The maximum velocities and the flow rate decreased for the lower period and the higher amplitude of surface roughness.

Commentary by Dr. Valentin Fuster

Heat Exchangers: Minichannels and Microchannels

2005;():647-655. doi:10.1115/ICMM2005-75025.

Heat transfer from arrays of circular and non-circular ducts subject to finite volume and constant pressure drop constraints is examined. It is shown that the optimal duct dimension is independent of the array structure and hence represents an optimal construction element. Solutions are presented for the optimal duct dimensions and maximum heat transfer per unit volume for the parallel plate channel, rectangular channel, elliptic duct, circular duct, polygonal ducts, and triangular ducts. Approximate analytical results show that the optimal shape is the isosceles right triangle and square duct due to their ability to provide the most efficient packing in a fixed volume. Whereas a more exact analysis reveals that the parallel plate channel array is in fact the superior system. An approximate relationship is developed which is very nearly a universal solution for any duct shape in terms of the Bejan number and duct aspect ratio. Finally, validation of the relationships is provided using exact results from the open literature.

Commentary by Dr. Valentin Fuster
2005;():657-665. doi:10.1115/ICMM2005-75071.

For a couple of years, microstructure heat exchangers have been numerically simulated, designed, manufactured and tested at the Institute for Micro Process Engineering of the Forschungszentrum Karlsruhe. The microstructure heat exchangers consist of single metal foils which are connected by a diffusion bonding process to form a nearly monolithic body. The number of integrated microchannels is in the order of several hundreds to several thousands. This internal numbering-up leads to a sufficiently low pressure drop. The devices have an extremely high heat transfer to volume ratio of about 30,000 m2 · m−3 which makes it possible to transfer thermal power in the range of several kilowatt within a volume of some centimeters cubed only.

Commentary by Dr. Valentin Fuster
2005;():667-678. doi:10.1115/ICMM2005-75114.

The development of advanced microchannel heat exchangers and microfluidic devices is dependent upon the understanding of the fundamental heat transfer processes that occur in these systems. Several researchers have reported significant deviation from the classical theory used in macroscale applications, while others have reported general agreement, especially in the laminar region. This fundamental question needs to be addressed in order to generate a set of design equations to predict the heat transfer performance of microchannel flow devices. A database is generated from the available literature to critically evaluate the reported experimental data. An in-depth comparison of previous experimental data is performed to identify the discrepancies in the reported literature. It is concluded that the classical theory is applicable to microchannel and minichannel flows. The literature reporting discrepancies do not account for developing flows, fin efficiency, erros in channel geometry measurements and experimental uncertainties. It is further concluded that if all these factors are accounted for, the available data have good general agreement with macroscale theories. A similar approach is presented for pressure drop in microchannels in an accompanying conference paper, Steinke and Kandlikar (2005).

Commentary by Dr. Valentin Fuster
2005;():679-684. doi:10.1115/ICMM2005-75218.

The heat transfer and pressure drop characteristics of compact heat exchangers having mini channels and copper wire spring fins are investigated experimentally. A copper capillary tube is used as heat transfer tube (length: 250 mm, inside diameter: 1.4 mm and wall thickness: 0.2 mm). The copper wire spring fin (CSF) is made of thin copper wire whose thickness is 0.5 mm and length is 8.4 m, respectively. The heat transfer area density of heat exchanger is 1,880 m2 /m3 . Water is employed inside the circular tube to transfer heat with air for convenience. The air flow entering into the test section is a fully developed duct flow and velocity is varied from 0.5 to 2.0 m/s with 0.5 m/s intervals. Based on the experimental data, heat transfer rate, pressure drop, f–factor and j–factor are investigated.

Commentary by Dr. Valentin Fuster
2005;():685-689. doi:10.1115/ICMM2005-75249.

The performance of a micro-channel gas-liquid cross flow heat exchanger, manufactured by the LIGA technique is presented. Large heat transfer coefficients are achieved on the gas side by achieving gas-flow passage dimensions as low as 300 microns. Cross flow heat exchanger panels have been produced as large as 20 cm by 15 cm. These panels can be arranged in a variety of ways to produce heat exchangers capable of handling large thermal loads. Experimental results have shown that these heat exchangers are approximately one order of magnitude better, in terms of heat transfer per unit volume, than the commercially available tube-fin heat exchangers with characteristic cross flow channel dimensions that are typically three times larger.

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