A Review on Two Phase Flow in Micro channel Heat Exchangers

International Journal of Applied Research & Studies ISSN 2278 – 9480mixture is flowing. In general, their work pointed to a fact thatcorrelations for void fraction and pressure drop predictions arebased in a separated flow model and do not predict theexperimental results in the range of conditions investigated.Weilin Qu, Issam Mudawar (2003) [10] exploredhydrodynamic instability and pressure drop in a water-cooledtwo-phase micro-channel heat sink containing 21 parallel231×713 lm micro-channels which were- severe pressure droposcillation and mild parallel which can trigger pre-maturecritical heat flux, can be eliminated simply by throttling theflow upstream of the heat sink. Different methods wereassessed for predicting two-phase pressure drop for suitabilityto micro-channel heat-sink design. Generalized two-phasepressure drop correlations were examined. which include 10correlations developed for both macro- and mini/microchannels.They proposed a new correlation incorporating theeffects of both channel size and coolant mass velocity whichshows better accuracy than prior correlations. They gave atheoretical annular two-phase flow model which, aside fromexcellent predictive capability, possesses the unique attributesof providing a detailed description of the various channelinstability. Their work showed that the severe pressure droposcillation, transport processes occurring in the micro-channel,as well as fundamental appeal and broader application rangethan correlation. Weilin Qu, Issam Mudawar (2003) [11]developed an annular flow model to predict the saturated flowboiling heat transfer coefficient. Two-phase micro-channelflow unique features, such as laminar liquid and vapour flow,smooth interface, and strong droplet entrainment anddeposition effects, were identified and incorporated into theirmodel. Their model correctly captured the unique overall trendof decreasing heat transfer coefficient with increasing vapourquality in the low vapour quality region of micro-channels.Fairly good agreement was achieved between the modelpredictions and heat transfer coefficient data over broad rangesof flow rate and heat flux. Hassan, P. Phutthavong, and M.Abdelgawad(2004) [12] provided an overview of the researchperformed thus far in the field of micro channel heat sinks.There are many other research fields concerning microchannel heat sinks that have yet to be explored. This sectionwill summarize the possible future directions of research thatmay be followed in order to obtain greater understanding ofthese micro devices. Effect of coolant types should beinvestigated more thoroughly. Liquids seem to providesuperior cooling properties when compared to gases, sincethey offer lower thermal resistance. Water has been thecoolant of choice for most experiments because it is readilyavailable and cheap and has a high specific heat capacity. Yet,still the effect of the type of the liquid used is not studied.CFD packages could be used to simulate various types ofcoolants and help predict which one provides better coolingcapabilities at affordable costs. Experiments may beperformed later to validate the simulations‟ results. E. W.Jassim, T. A. Newell, and J. C. Chato (2006) [13] investigatedwith a purpose to develop models for two-phase heat transfer,void fraction, and pressure drop, three key design parameters,in single, smooth, horizontal tubes using a commonprobabilistic two phase flow regime basis. Probabilistic twophaseflow maps were experimentally developed for R134a at25 ºC, 35ºC, and 50 ºC, R410A at 25 ºC, mass fluxes from 100to 600 kg/m2-s, qualities from 0 to 1 in 8.00 mm, 5.43 mm,3.90 mm, and 1.74 mm I.D. horizontal, smooth, adiabatictubes in order to extend probabilistic two-phase flow mapmodelling to single tubes. An automated flow visualizationtechnique, utilizing image recognition software and a newoptical method, was developed to classify the flow regimespresent in approximately one million captured images. Theprobabilistic two-phase flow maps developed were representedas continuous functions and generalized based on physicalparameters. Condensation heat transfer, void fraction, andpressure drop models were developed for single tubes utilizingthe generalized flow regime map developed. The condensationheat transfer model was compared experimentally obtainedcondensation data of R134a at 25 ºC in 8.915 mm diametersmooth copper tube with mass fluxes ranging from 100 to 300kg/m2-s and a full quality range. Renqiang Xiong (2007) [14]performed some experimental and numerical investigations.He evaluated the pressure drops of liquid flow in straight andserpentine micro-channels with hydraulic diameters of 0.209mm, 0.412 mm, and 0.622 mm. Three types of microchannels:straight short, straight long, and long serpentine,were fabricated. Adiabatic nitrogen-water flow patterns andvoid fractions in straight micro-channels were experimentallyinvestigated. Gas and liquid superficial velocities were variedfrom 0.06-72.3 m/s and 0.02-7.13 m/s, respectively. Theinstability of flow patterns was observed. Four groups of flowpatterns including bubbly-slug flow, slug-ring flow, dispersedchurnflow and annular flow were observed in micro-channelsof 0.412 mm and, 0.622 mm while in the micro-channel of0.209 mm, the bubbly-slug flow became the slug-flow and thedispersed-churn flow disappeared. The flow regime mapsshowed that the transition lines shifted to a higher gassuperficial velocity due to a dominant surface tension effect asthe channel size was reduced. The void fractions holded a nonlinearrelationship with the homogeneous void fraction asopposed to the relatively linear trend for the mini-channels. Hedeveloped a new correlation to predict the non-linearrelationship that fits most of the current experimental datawithin ±15%. Renqiang Xiong and J. N. Chung (2007) [15]studied adiabatic gas-liquid flow patterns and void fractions inmicro channels experimentally. Nitrogen and water were usedin rectangular micro channels with hydraulic diameters of0.209 mm, 0.412 mm and 0.622 mm, respectively. Superficialvelocities of gas and liquid were varied from 0.06–72.3 m/s0.02–7.13 m/ s, respectively. Their work focused mainly onthe effects of microscale channel sizes on the flow regime mapand void fraction. Flow pattern instability was observed. Flowpatterns including four groups namely bubbly slug flow, slugringflow, dispersed-churn flow, and annular flow wereobserved in micro channels of 0.412 mm and, 0.622 mm. Themicro channel of 0.209 mm, showed the change from thebubbly slug flow to the slug flow and the dispersed-churn flowdisappeared. The obtained flow regime maps reflected that thetransition lines shifted to higher gas superficial velocity due toiJARS/ Vol. II/ Issue 2/Feb, 2013/319 5http://www.ijars.in

International Journal of Applied Research & Studies ISSN 2278 – 9480a dominant surface tension effect with the channel sizereduction. The regime maps given by other authors forminichannels were found not to be applicable for microchannelsG.P. Celata (2008) [16] presented a critical evaluationof existing works in the field of heat transfer of single (liquid)and two-phase flow (boiling) in micro channels. There ismore and more the general agreement on the validity ofmacroscale knowledge going down to microscale as far asliquid single phase flow heat transfer is concerned. The scalereduction gives relevance to some effects such as viscousdissipation (or viscous heating, due to the general largepressure drop associated with micro channel flow), thermalentrance (due to the generally short geometry of microchannels) and axial length (associated with the general largethickness of the micro channels), which becomes veryimportant in microscale. Their neglectfulness is bounded togive poor analysis of heat transfer data misleading the possiblecorrect conclusion of the experimental outcome. Once theseso-called scaling effects are properly accounted for, microscaleheat transfer in liquid single-phase flow looks well predictableusing current knowledge of macroscale, such as Gnielinski(1978) correlation for turbulent flow and Nu = 4.36 equationfor laminar flow. Occurrence of scaling effects can also beused to identify the threshold between micro- and macroscale,which depends not only on the geometry but also on fluidphysical properties. More accurate experiments are stillneeded, though the knowledge looks now more assessed.Boiling heat transfer in microscale is associated with bubbleconfinement in the reduced scale, which is definitely expectedto alter the flow dynamics inside the small size channel.Because of that, divergence between macroscale knowledge(often based on empirical bases or experimental evidences) isclearly awaited and understandable. Also conclusions drawnfrom the experimental data in tubes, estimated from thefindings of boiling heat transfer in macroscale (nucleate versusforced convective boiling), might have led to erroneousconclusions about the type of boiling regime in microscale.Recent theoretical works, supported by first visualizationworks, have supposed a different boiling mechanisms formicro channels, based on elongated bubble existence andevaporation of thin liquid layer between the elongated bubbleand the channel wall. This paper points out to need for furtherwork to possibly validate the existing theory or produce newinsights for a better mechanistic modelling of boiling heattransfer. M. M. Awad And Y. S. Muzychka (2010) [17] havepresented three different methods for two-phase flowmodelling in micro channels and minichannels . They areeffective property models for homogeneous two-phase flows,an asymptotic modelling approach for separated two-phaseflow, and bounds on two-phase frictional pressure gradient. Inthe first method, new definitions for two-phase viscosity wereproposed using a one-dimensional transport analogy betweenthermal conductivity of porous media and viscosity in twophase flow. The new definitions given by them can be used tocompute the two-phase frictional pressure gradient using thehomogeneous modelling approach. In the second method, asimple semitheoretical method for calculating two-phasefrictional pressure gradient using asymptotic analysis ispresented. Two-phase frictional pressure gradient is expressedin terms of the asymptotic single-phase frictional pressuregradients for liquid and gas flowing alone. In the final method,simple rules are developed for obtaining rational bounds fortwo-phase frictional pressure gradient in minichannels andmicro channels. In all cases, the proposed modellingapproaches are validated using the published experimentaldata.B. Numerical Study Of Two Phase Flow In MicrochannelHeat Sinks. Literature shows that the micro channels and micro channelsheat sinks were studied extensively, but it was found that thereis limited research related to the performance study of microchannel heat exchangers using CFD models. Fundamentalunderstanding of the characteristics of the heat transfer andfluid flow in micro channel are necessary for effective designof micro channel heat exchanger. The designs and relations ofmacro scale fluid flow and heat transfer were employed inearlier times. Numerical simulations give the strength toinvestigate small details that are impossible to observe inexperiments.Bogdan Alexandru Nichita (2010) [18] used FLUENT in thestudy to model adiabatic and diabatic, time dependent two-phaseflows. This work involved using a fifth order WENO (WeightedEssentially Non Oscillatory) scheme to discretize the spacederivatives (otherwise oscillations of the interface occurred),and a first order Euler method for the time integration. Inanother part of their study, a 3D dynamic contact angle modelbased on volume fraction, interface reconstruction, andexperimentally available advancing and receding static contactangles was also developed and implemented into FLUENT viaUDFs. Several validations for the developed CLSVOF methodand dynamic contact angle model were presented in this thesis,which included a static bubble, a bubble rising in a stagnantliquid for Morton numbers ranging from 102 to 10−11, dropletdeformation due to a vortex flow field, droplets spreading over awall under the gravity effect and droplets sliding over a wall dueto gravity. The validations demonstrated the high accuracy andthe stability of their methods for modelling these phenomena. Aheat and mass transfer model was also implemented into thecommercial CFD code FLUENT for simulation of boiling (andcondesation) heat transfer. Several simulations were presentedwith water and R134a as working fluids. Several 2D and 3D testwere performed for air/water on a coated silicon wafer surfacewith different gravity vectors, which proved the accuracy of ourmodel when compared to both numerically and experimentallydata available in the literature.Aakash Kumar Pandey (2011) [19] predicted the experimentalwork done by Lee and Mudawar (2007) by CFD results. Thehydrodynamics and thermal behaviour of a rectangular microchannel were studied here. The variation in wall temperature,pressure drop in the channel and the friction factors werecalculated using ANSYS Fluent . The effect of Re on thebehaviour the channel was also studied. The relation betweenheat transfer coefficient and thermal conductivity of the fluid i.e.h ∞ k was proved his study. It was found that the entrance lengthiJARS/ Vol. II/ Issue 2/Feb, 2013/319 6http://www.ijars.in

International Journal of Applied Research & Studies ISSN 2278 – 9480for the fully developed velocities depended on Reynoldsnumber. The temperature rise between outlet and inlet dependson the Reynolds number, Re and Peclet number, Pe.Ashutosh Kumar, Meryll Levet, Claude Souprayen, AnilKumar, Liam Worth And Robert Pearce (2011) [20] workedkeeping in mind the safety of a Tokamak-type fusion reactorsuch as ITER (International Thermonuclear Experimentalreactor), occurrence of a defect or a crack is very likely even ifits structural integrity is assured.Even a small water leak maycause at worst a plasma disruption. A quick detection of possibleleaks is therefore crucial. Their paper addresses the issue ofcomputing the expected mass flow rate out of a microleak with aCFD 3D solution of Navier-Stokes equations with phase changein a homogeneous equilibrium model (HEM). This model wasfirst compared to quantitative estimates based on analytical 1Dmodelling of an isothermal flow inside a microcrevice ofrectangular cross-section. In such conditions, the CFD 3Dresults were matching with the 1D analytical solution forpressures, densities, velocities and flow rates. Further analysesin the non-isothermal flow regime (with heat from liquid andfrom the steel wall surrounding structures) were also discussedin their final section. The results showed that for isothermalflow condition, there was fair agreement of the results for CFDthat of the analytical model Sharipov et al. In all the nonisothermalflow, the mass flow rate was higher than theisothermal flow. In the non-isothermal flow cases liquid vapourinterface was observed. The trend of pressure profile, profile andvapour mass fraction profile for the flow with 1D heatconduction through wall model falled in between that of theisothermal wall case and the adiabatic wall case.Table 1: Current state of understanding of transportphenomena in microchannel heat sinksPhase change can greatly enhance the performance of amicro-channel heat sink by providing higher convectiveheat transfer coefficients, better axial temperatureuniformity, and reduced coolant flow rate requirements..Unfortunately, the number of investigations into fluid flowand heat transfer characteristics of two phase microchannelheat sinks is very limited. Table 1 summarises currentstatus of understanding of transport phenomenon inmicrochannel heat sinks. [5]IV.CONCLUSION AND FUTURE DIRECTIONSThe present manuscript presents a critical evaluationof existing works in the field of heat transfer of two-phaseflow in microchannel heat sinks. The literature surveyprovides with many works in this field, the accuracy of whichis increasing with date, recent works being the most accurate.The study of flow and heat transfer in micro channels is veryimportant for the technology of today and the near future, as isevident from the diversified application areas of micro channelheat exchangers. The trend of miniaturization in all fields isthe demand of time and field of heat exchange is no different.Literature shows that the micro channels and micro channelsheat sinks were studied extensively, but it was found that thereis limited research related to the performance study of microchannel heat exchangers using CFD models. Fundamentalunderstanding of the characteristics of the heat transfer andfluid flow in micro channel are necessary for effective designof micro channel heat exchanger. It was found from literaturesurvey that results found so far are not in strong agreementwith each other, many variations were observed.Effect of coolant types should be investigated morethoroughly. Liquids seem to provide superior coolingproperties when compared to gases, since they offer lowerthermal resistance. Water has been the coolant of choice formost experiments because it is readily available and cheap andhas a high specific heat capacity. Yet, still the effect of thetype of the liquid used is not studied. CFD packages could beused to simulate various types of coolants and help predictwhich one provides better cooling capabilities at affordablecosts. Experiments may be performed later to validate thesimulations‟ results.The field of two phase flow is still in its youth andneeds much more attention. It can be concluded that in all fortwo phase flow behaviour in micro channel heat exchanger useof more sophisticated computational packages should be usedwhich are specifically tailored to its needs so that strongvalidations can be found for existing experimental results.Recent theoretical works, supported by first visualizationworks, have supposed a different flow patterns and voidfraction and pressure drop relations. Much work is stillrequired to possibly validate this theory or produce newinsights for a better mechanistic modelling of two phase heattransfer.iJARS/ Vol. II/ Issue 2/Feb, 2013/319 7http://www.ijars.in

International Journal of Applied Research & Studies ISSN 2278 – 9480Boiling instabilities in two-phase microchannel heatsinks should be studied thoroughly. Instabilities were observedand were attributed to large pressure drops. Lateralinstabilities also exist in wide microchannels. The fluid wasobserved to revert back to liquid upon boiling, and to beginboiling once again in cycles that last up to a few minutes.Hestroni et al. [6] noticed erratic pressure and temperaturevariations in two-phase microchannel heat sinks. Theseinstabilities will hinder the implementation of microchannelheat sinks in commercial applications as long as they existunexplained.Different microchannel geometries should be testedto see if one shape in particular yields a better coolingperformance. Experimental data for different geometriesshould be tested for two-phase flow in order to generateaccurate flow regime maps that may be able to predict theflow pattern. Transitional flow regimes should also be clearlyand universally defined in experiments in order to avoidconfusion when attempting to model the flow regime maps.This experimental data will also be useful in generatingmodels that may predict the heat transfer coefficient associatedwith two-phase flow in microchannels.Fluid properties are affected by variations intemperature; however, most of the studies cited did notconsider this fact. Studies accounting for the temperaturedependentfluid properties will certainly give more accurateand realistic results than constant fluid properties studies, thuspredicting the real performance of microchannel heat sinks.Spatially varying heat fluxes should be investigatedfurther in the future. There was but one study that assumedspatially varying heat fluxes, which concluded that the besttwo-phase microchannel heat sinks have 80% of the total heatflux applied to the latter half of the channel. Future workmight provide other nonuniformity patterns that may enhancemicrochannel heat sink performance.Area worthy of new investigations have beenidentified and recommended, such as identification of microto-macroscaletransition, critical heat flux, visualization, flowpattern transition, flow instability, two-phase pressure drop,ACKNOWLEDGMENTI would like to express my most sincere gratitude to myadvisor and guide Prof. Asst. Prof. Vivek C. Joshi(Mechanical Engineering Department, PIET, Limda) for hisconstant guidance, support, unending encouragement, andpersonal help.Finally, my thanks to my family, mother, husband and myloving daughter for always giving me the encouragement andunconditional support during this work; I will always be deeplyindebted. `REFERENCES[1] Roger S. Stanlely Timothy A. Ameel Randall F. Barron 1997, „TwophaseFlow in microchannel‟ slouisiana Tech University P.O. Box 7923T.S.Ruston, LA 71272[2] K.A. Triplett, S.M. Ghiaasiaan , S.I. Abdel-Khalik, D.L. Sadowski 1999,„Gas±liquid two-phase flow in microchannels Part I: two-phase flowpatterns‟, International Journal of Multiphase Flow 25 (1999) 377±394[3] Weilin Qu, Issam Mudawar 2002, „Experimental and numerical study ofpressure drop and heat transfer in a single-phase micro-channel heatsink‟, International Journal of Heat and Mass Transfer 45 (2002) 2549–2565[4] Weilin Qu, Issam Mudawar 2002, „Analysis of three-dimensional heattransfer in micro-channel heat sinks‟, international Journal of Heat andMass Transfer 45 (2002) 3973–3985[5] Weilin Qu and Issam Mudawar 2002,‟transport phenomena in two-phasemicro-channel heat sinksasme‟, International Mechanical EngineeringCongress & Exposition November 17.22, 2002[6] G. Hetsroni , A. Mosyak, Z. Segal, E. Pogrebnyak 2002,‟Two-phaseflow patterns in parallel micro-channels‟, International Journal ofMultiphase Flow 29 (2003) 341–360[7] Kawahara a,b, P.M.-Y. Chung a, M. Kawaji 2002,‟Investigation of twophaseflow pattern, void fraction and pressure drop in a microchannel‟,International Journal of Multiphase Flow 28 (2002) 1411–1435[8] Akimi Serizawa, Ziping Feng, Zensaku Kawara 2002,‟Two-phase flowin microchannels‟, Experimental Thermal and Fluid Science 26 (2002)703–714[9] V. G. Niño, P. S. Hrnjak and T. A. Newell 2002, „Characterization ofTwo-PhaseFlow in microchannel‟, air Conditioning and RefrigerationCenter University of Illinois Mechanical & Industrial Engineering Dept.1206 West Green Street Urbana, IL 61801 Prepared as part of ACRCProject #107Void Fraction and Pressure Drop in Microchannels (217)333-3115[10] Weilin Qu, Issam Mudawa 2003,‟Measurement and prediction ofpressure drop in two-phase micro-channel heat sinks‟, internationalJournal of Heat and Mass Transfer 46 (2003) 2737–2753[11] Weilin Qu, Issam Mudawar 2003, „Flow boiling heat transfer in twophasemicro-channel heat sinks––II. Annular two-phase flow model‟,International Journal of Heat and Mass Transfer 46 (2003) 2773–2784[12] Hassan, P. Phutthavong, and M. Abdelgawad 2004, „microchannel heatsinks: an overview of the state-of-the-art‟, Microscale ThermophysicalEngineering, 8:183–205, 2004[13] .E. W. Jassim, T. A. Newell, and J. Chato 2006, „Probabilistic FlowRegime Modeling of Two-Phase flow‟, University of illinoisdepartmentof Mechanical Science & Engineering 1206 West Green Street Preparedas part of ACRC Project #184 Urbana, IL 61801 (217) 333-3115.[14] Renqiang xiong 2007, „single- and two-phase pressure-driven flowtransport dynamics in micro-channels‟, a dissertation presented to thegraduate school of the university of florida in partial fulfillment of therequirements for the degree of doctor of philosophy university of florida[15] Renqiang Xiong and N.Chung 2007, „An experimental study of the sizeeffect on adiabatic gas-liquid two-phase flow patterns and void fractionin microchannels‟, Phys. Fluids 19, 033301 (2007)[16] G.P. Celata 2008, „single- and two-phase flow heat transfer inmicropipes’ ,5th European Thermal-Sciences Conference, TheNetherlands, 2008[17] M. M. Awad and y. S. Muzychka 2010, „two-phase flow modeling inmicrochannels and minichannels‟, heat transfer engineering,31(13):1023–1033, 2010[18] .Bogdan Alexandru NICHITA 2010, „An Improved CFD Tool toSimulate Adiabatic and diabatictwo-Phase Flows‟, thèse no 4776 (2010)école polytechnique fédérale de lausanne présentée le 7 septembre 2010à la faculté sciences et techniques de l'ingénieur laboratoire de transfertde chaleur et de masse programme doctoral en energie[19] Aakash kumar pandey 2011, „A Computational Fluid Dynamics Study ofFluid Flow and heat Transfer in a Micro channel‟, national institute oftechnology, rourkela (orissa)[20] Ashutosh Kumar, Meryll Levet, Claude Souprayen, Anil Kumar, LiamWorth, Robert Pearce 2011, „cfd analysis of two-phase flow throughmicrochannel using homogeneous equilibrium model‟, Proceedings ofthe 4th International Conference on Heat Transfer and Fluid Flow inMicroscale HTFFM-IV 4-9 September, 2011iJARS/ Vol. II/ Issue 2/Feb, 2013/319 8http://www.ijars.in

International Journal of Applied Research & Studies ISSN 2278 – 9480[21] Engineering Thermofluids, Fluid Mechanics And Heat Transfer,M.Moussoud Chapter 5 pp 601-672[22] Heat Transfer And Fluid Flow In Minichannels And Microchannels, D.S. Kandlikar, S. Garimella, Li. S. Colin, M. R. King, Chapter 5 ,pp 175--226iJARS/ Vol. II/ Issue 2/Feb, 2013/319 9http://www.ijars.in