Jianyang Zhou ,Xiaoping Luo *,Cong Deng Mingyu Xie Lin Zhang Di Wu Feng Guo

1 School of Mechanical and Automobile Engineering,South China University of Technology,Guangzhou 510640,China

2 School of Mechanical and Marine Engineering,Qinzhou University,Qinzhou 535011,China

1.Introduction

In recent decades,electronic equipment trends for integration with the development of science and technology.More heat is produced in the unit area and the conventional cooling technology has been unable to dissipate large amounts of heat from small surface areas.As a consequence,cooling systems with liquid have been increasingly developed[1].

Since Choifirstly proposed nano fluid which was defined as suspensions of nanoparticles into base fluid[2],nano fluid has attracted many researchers around the world as a significant alternative to enhance the heat transfer performance[3-9].Due to large specific surface areas of nanoparticles,nano fluid possess superior the heat transfer properties such as high thermal conductivity and long-term stability.

[10-14]show that thermal conductivity and heat transfer performance could be enhanced by suspending nanoparticles in a base fluid.These studies mainly concentrated on the influence of thermophysical properties and nanoparticle concentrations of nano fluid on the heat transfer.For example,Peng et al.[15]studied the influence of particle parameters and surfactant on aggregation behavior of nanoparticles in refrigerant and found that the primary size and surfactant of particle had an influence on the steady-state hydrodynamic diameter.It indicated that the particle parameter of suspended nanoparticles in nano fluid had significant impact on heat transfer.Omer's study found that the thermal conductivity and specific heat capacity of nano fluid increased with the augmentation of fluid temperatures[16].Consequently,thermo physical properties have an impact on the heat transfer performance.

There are a few literatures that studied the influence of nanoparticle concentrations on heat transfer characteristics[17-20].Mahbubul et al.studied the heat transfer performance of nano fluid from 1 to 5 vol.%in a horizontal smooth tube and found that thermal conductivity and flow boiling heat transfer coefficients increased with the increase of nanoparticle concentrations[18,19].And Mahbubul et al.also investigated that particle concentration and temperature had an influence on conductivity and viscosity of nano fluid and found that ther malconductivity and viscosity increased with the augmentation of particle concentrations and temperatures[20].The above researches show that the heat transfer coefficients increase with nanoparticle concentrations increasing.

However,other studies show different results.Henderson et al.investigated that the heat transfer coefficients decreased with the increasing of nanoparticle concentrations[21].Yang et al.studied the characteristic of nano fluid heat transfer in the evaporator with CuO/H2O and experimentalresults showed that the heattransfer coefficients nonlinearly increased with nanoparticle concentrations and existed with an optimal mass concentration corresponding to the best heat transfer enhancement[22].Therefore,the influence of nanoparticle concentrations of nano fluid on heat transfer performance has not been uniformly understood.The mechanism about the effect of nanoparticle concentrations on heat transfer needs to be further studied.

Micro- fins patterning of the closed heat exchanger produced by conventional manufacturing(e.g.milling and wire-electrode cutting et al.)for electronics cooling may be impracticable due to technological constraints.In recentyears,additive manufacturing technologies represent an interesting alternative.Complex-shaped components can be created without wasteful casting or drilling,which makes additive manufacturing an economical way to fabricate single items[23].In the present study,the micro heat exchanger is made by direct metal laser sintering.In the micro heat exchanger,0.05 to 0.4 wt.%Al2O3/R141b+Span-80 nanorefrigerant is used to investigate the flow boiling heat transfer characteristics of nano fluid under the conditions of heat flux from 8.5 to 37.3 kW·m-2,mass flow rate from 184.3 to 432.2 kg·m-2·s-1,system saturated pressure 176 kPa and inlet temperature 25°C,respectively.The objective of this study is to focus on whether there is a linear relation between the heat transfer coefficients of Al2O3/R141b+Span-80 nanorefrigerant and nanoparticle concentrations or not in a certain concentration range.If the heat transfer coefficients nonlinearly increase with nanoparticle concentrations,the optimal mass concentration corresponding to the best heat transfer enhancement needs to be found out.

2.Micro Heat Exchanger by Direct Metal Laser Sintering(DMLS)

In the present study,micro heat exchanger is made by direct metal laser sintering.The main advantage of this technology is that metal complex-shaped components are easily produced in one step.A threed imensional CAD-model of an object is designed by a computer and then the CAD-model is converted to an STL file.This file defines an optimal building route of the whole object.Then these files are transferred to the computer of the DMLS machine,which has the necessary information to build up each layer.The manufacturing process of micro heat exchanger by direct metal laser sintering is shown in Fig.1.The essential operation in the DMLS process is the laser beam scanning over the surface of a thin powder layer previously deposited on a substrate[23].The forming process goes along the scanning direction of the laser beam.Each cross-section(layer)of the part is sequentially filled with elongated lines(vectors)of molten powder.

Fig.1.Manufacture principle of micro heat exchanger by direct metal laser sintering.

The micro heat exchanger was prepared by DMLS with an EOSINT M270.The specimen material property is aluminum alloy.The heat exchanger has a length of 240 mm,width of 40 mm and height of 8 mm and contains 18 channels of cross-section with 1 mm×2 mm.The hydraulic diameter of channel is 1.33 mm.Therefore,the experiment channel belongs to the category of mini-channel[24].

After micro heat exchanger manufacturing,the 3D morphological of channel wall surface is characterized by laser scanning.The 3D morphological characterizations of lateral and bottom wall surface(a small area 1 mm×1 mm)are shown as Fig.2(a)and(b).The channel wall surface roughness is measured by a BMT Surface Pro file and Roughness Laser Measuring Instrument.The measurement instrument accuracy is 0.01 μm.The arithmetic mean deviation Rais introduced to make quantitative analysis for the channel wall surface roughness.

where xLis the length range of laser scanning on the channel wall surface and f(x)is the generic rough surface contour lines offset in the xL-f(x)two-dimensional coordinate system.After being measured,the roughness range of lateral and bottom wall surface is 7.13 to 9.33 μm.

Fig.2.Surface morphological characterization of micro heat exchanger's channel by 3D optical scanner.(a)3D morphological characterization of lateral wall surface(b)3D morphological characterization of bottom wall surface.

3.Experimental Setup and Calculation Method

3.1.Preparation and properties of nanorefrigerant

In the present experiment,the nanoparticle type of nanorefrigerant is spherical Al2O3.The SEM picture and diameter distribution of Al2O3nanoparticles are shown in Fig.3.The average diameter of Al2O3is 96 nm,which is tested by HORIBA Nanoparticle Analyzer SZ-100.

Fig.3.SEM picture and diameter distribution of Al2O3 nanoparticles.

The preparation processes are shown in Fig.4.Al2O3nano particles are dispersed into pure refrigerant R141b.The suspension is oscillated in an ultrasonic oscillator bath for 3 h.Meanwhile,an amount of Span-80 dispersant is added into the mixture to get stable nanorefrigerant as the dispersion of nanoparticles will affect the heat transfer performance.Peng et al.found that the dispersant Span-80 can enhance the heattrans fer peformance of nanorefrigerant.In addition,the increasing amount of heat transfer coefficients of nanorefrigerant with 0.05 wt.%Span-80 between the mass fraction from 0.1%to 0.5%and from 0.5%to 1.0%are almostequal[25].Therefore,in the present paper the influence of the same dispersant with 0.05 wt.%Span-80 on heat transfer of different mass concentrations(0.05%,0.1%,0.2%,0.3%,0.4%)Al2O3/R131b+Span-80 nanorefrigerant is considered as the same when discussing the effects of nanoparticle mass concentrations on heat transfer of Al2O3/R131b nanorefrigerant.

Fig.4.Al2O3/R141b+Span-80 nano fluid preparation processes.

Fig.5 shows different mass fraction concentrations(0.05%,0.1%,0.2%,0.3%,0.4%)of Al2O3/R141b+Span-80 nanorefrigerant.After Al2O3/R141b+Span-80 nanorefrigerant prepared,the zeta potential and the size of the particle after dispersion tests are carried out to identify the stability of different mass fraction(0.05%,0.1%,0.2%,0.3%,0.4%)Al2O3/R141b+Span-80 nanorefrigerant by the methods from references[26].The test results are shown in Fig.6.Fig.6(a)shows the nanoparticle diameter distribution of 0.1 wt.%Al2O3/R141b+Span-80 nanorefrigerant.The averaged particle diameter after dispersion of different mass fraction(0.05%,0.1%,0.2%,0.3%,0.4%)Al2O3/R141b+Span-80 nanorefrigerant are 99 nm,112 nm,103 nm,97 nm,108 nm,respectively.The size of the particle after dispersion is slightly higher than the size of nanoparticle.The slight increase in size may be due to a small amount of particle aggregation in refrigerant.Fig.6(b)shows that the zeta potential value of different mass fraction(0.05%,0.1%,0.2%,0.3%,0.4%)Al2O3/R141b+Span-80 nano fluid ranging from 31.8 mV to 35.1 mV and the zeta potential is measured again two days later.The zeta potential values ranged from 30.2 mV to 31.2 mV,which indicates that the relative stable Al2O3/R141b nano fluid is obtained.

The physical properties(density,thermal conductivity,specific heat,viscosity)of different mass fraction(0.05%,0.1%,0.2%,0.3%,0.4%)Al2O3/R141b+Span nanorefrigerant are directly measured respectively using GB-DMR Tuning Fork Type Densimeter,DM3615 Thermal Conductivity Meter,Viscomenter(Brook field,Model:LVDV-П+Pro)and Liquid Specific Heat Capacity Meter(Model:DL21-BD-I-301)at temperatures 49°C.Table 1 indicates Al2O3/R141b+Span-80 physical parameters at saturated temperature 49 °C. φ,w, ρ,cpand μ are volume fraction,mass fraction,density,specific heat capacity and viscosity,respectively.

Fig.5.Different mass fraction concentrations of Al2O3/R141b+Span-80.

Fig.6.Stability of Al2O3/R141b+Span-80 nanorefriger ant(a)Diameter distribution of 0.1 wt.%Al2O3/R141b+Span-80 nanorefriger ant(b)Zeta potential values of different mass fraction of Al2O3/R141b+Span-80 nanorefrigerant.

Table 1 Al2O3/R141b+Span-80 physical parameters at saturated temperature 49°C

3.2.Experimental apparatus

Fig.7(a)shows a schematic of the experimental apparatus which includes circulation control system,test module and data acquisition system.Nanorefrigerant is circulated in the system with the help of a magnetically driven pump.A particulate filter is used to filter clutter.The refrigerant is heated to the desired temperature by a temperature controller.Then the fluid medium enters into test module.Finally,the two-phase mixture leaving the test module is transported to a liquidto-liquid heat exchanger,where it is cooled down to near ambient temperature.

Fig.7(b)shows the test module construction and instrumentation and Fig.7(c)shows the cross and longitudinal section of the minichannel test module.The test module contains a top plate,quartz glass plate,micro heat exchanger and heating plate.Micro heat exchanger by direct metal laser sintering contains 18 mini-channels.Each mini-channel has a cross-section of 1 mm×2 mm(Wch×Hch).There is a fin between the channel and has a fin width of 1 mm.PT100 thermal resistances are used to measure the temperature,which are installed on the aluminum block at four axial locations of 70,115,145,and 205 mm along the channel length.Two groups of PT100 thermal resistances are mounted at each axial location,spaced 30 mm(δ=30 mm)normal to the flow direction.The distance between the channel bottom wall and the upper thermal resistance is 6 mm(δ1=6 mm).The cross-sectional dimensions of mini-channel test module are shown in Table 2.

Pressure is measured at the outlet and the inlet plenums using HC3160-HVG4 pressure transducers,which has an accuracy of 0.5%according to instrument label.The accuracy of the temperature sensor and transmitter is 0.3%.All temperatures,pressure,and flow rate are recorded using ADAM-6017 data acquisition system.Table 3 shows the experimental data acquisition device parameters.The uncertainty refers to the fractional uncertainty in the paper,which is defined as δx/|xbest|,where xbestis the best estimate for x,and δxis the error in the measurement.If various quantities x,…,y are measured with small uncertainty δx,…,δy,and the measured values are used to calculate some quantity R,according to the uncertainty transfer principle,the fractional uncertainty of R can be obtained.And the fractional uncertainty of G,qe,h,xeand Tware 0.5%,0.3%,0.52%,0.64%,and 0.42%,respectively.

3.3.Experimental procedures

Firstly,the test equipment system pressure should be checked to keep constant without falling before the experiment.Nitrogen is poured into the system under the stable pressure of 0.25 MPa for 3 h.And the system leakage is examined by observing the pressure gauge.Then the experiment system is pumped into the vacuum.Secondly,an amount of Al2O3/R141b+Span-80 nanorefrigerant is poured into the test equipment system and the nanorefrigerant is run for a period of time to clean the pipe.Thirdly,different mass concentrations of Al2O3/R141b+Span-80 nanorefrigerant are used to investigate the influence of nanoparticle concentrations on flow boiling heat transfer in the mini-channel under the conditions of heat flux from 8.5 to 37.3 kW·m-2,mass flow rate from 183.1 to 457.8 kg·m-2·s-1,saturated pressure 176 kPa and inlet temperature 25°C,respectively.In addition,micro heat exchanger channels need to be cleaned after each experiment to avoid the influence of nanoparticle deposition on the next experiment.In addition,the test equipment system also must be run using pure refrigerant after each experiment to clean the system pipe.Furthermore,the repeatability experiments are carried out to ensure the reliability of the experimental data and the number of runs of experiments for each concentration are three times.

Before the experiment,the thermal equilibrium experiment is carried out to examine the system test accuracy under different heat flux.The heat losses to the surroundings are calculated by the difference between the input heat amount Q produced by the electric heating plate and the output heat amount Qlabsorbed by working fluid without phase change.The heat transfer of the aluminum base can be approximately regarded as one-dimensional heat conduction.And the heat flux calculation can be expressed as Eq.(3).

Fig.7.(a)Schematic of the experimental apparatus.(b)Test module construction and instrumentation.(c)Cross and longitudinal section of mini-channel test module.

Table 2 Cross-sectional dimensions of channel test module

Table 3 Data acquisition device parameters

where Tw,dnand Tw,upare the temperatures of four lower and upper axial thermocouple locations,δ is the distance between lower and upper measurement points,and λais the thermal conductivity of aluminum block.

The heat loss deviation is defined as follows by Eqs.(4)to(6).

where G is the mass flux rate,and Tinand Toutare the inlet and outlet temperatures of the working fluid,respectively.The results show that the maximum heat loss deviation ε is within 4%when the experimental heat flux range is between 8.5 and 37.3 kW·m-2.It indicates that the experimental system meets the precision requirements.

3.4.Calculation model and method

Fig.7(c)shows the cross section and longitudinal section of the test module, flow regions along the mini-channel,temperature measurement locations(T1-T8)and geometrical parameters.The length of subcooled region is Lsub,which is determined by the axial location where vapor quality xe=0[1].

where cp,M,Tsat,Tinand qeare specific heat capacity,mass flux per mini channel,saturation temperature of the nanorefriger ant in the channel,the fluid inlet temperature and heat flux,η is the fin efficiency for a rectangular channel,which are given by Eqs.(9)and(10).

The average heat transfer coefficient in the saturated region is calculated by applying an energy balance in the single mini-channel.And the test module is covered by a low thermal conductivity of asbestos so that heat losses to the surroundings can be neglected.

Fig.8.Variation of heat transfer coefficient with heat flux.

where Tfis the fluid temperature,and Twis the bottom mini-channel wall temperature,which is calculated using the assumption of onedimensional heat diffusion.

The axial variation of the fluid temperature in the saturated region along the micro-channel is the saturated temperature of nanorefrigerant at 176 kPa.

The axial variation of thermodynamic equilibrium vapor quality along the mini-channel can be given by calculation formula(15).

where z is the distance from the entrance,and hfgis the latent heat of vaporization.

4.Experimental Results and Discussions

4.1.Experimental results

Figs.8,9 show the variations of heat transfer coefficients of different concentrations nanorefriger ant with mass flux rate and heat flux under the conditions of system saturated pressure 176 kPa and fluid inlet temperature 25°C in micro heat exchanger,respectively.Fig.8 displays the influence of heat flux on flow boiling heat transfer coefficients of different mass concentrations Al2O3/R141b+Span-80 nanorefrigerant for mass flux rate 183.5 kg·m-2·s-1.It is observed that the heat transfer coefficients of different mass concentrations Al2O3/R141b+Span-80 nanorefrigerant proportionally increase with heat flux and the heat transfer performance of nanorefrigerant is enhanced compared with pure refrigerant R141b under the same conditions.However,the heat transfer coefficients of Al2O3/R141b+Span-80 nanorefrigerant nonlinearly increase with the mass fraction of Al2O3/R141b+Span-80 nanorefrigerant,in which the optimal nanoparticles concentration is 0.1 wt.%.

Fig.9.Variation of heat transfer coefficient with mass flux rate.

Fig.9 shows the variation of heat transfer coefficients of0 wt.%(pure refrigerant R141b),0.05 wt.%,0.1 wt.%,0.2 wt.%,0.3 wt.%,0.4 wt.%nanorefriger ant with the mass flux rate under the conditions of system average pressure 176 kPa,heat flux 32 kW·m-2and inlet temperature 25°C,respectively.It is noticed that the heat transfer coefficients of refrigerant increase with heat flux.And the heat transfer coefficients of nanorefrigerant are higher than R141b refrigerant.The average heat transfer coefficients increased by 63.4%after mixing nanoparticles in the pure refrigerant under the same heat flux.The heat transfer coefficients of Al2O3/R141b+Span-80 nanorefrigerant increased by around 39.9%to 55.0%compared with the heat transfer coefficients of pure refrigerant R141b.Fig.10 shows the variation of heat transfer coefficients of 0 wt.%(pure refrigerant R141b),0.05 wt.%,0.1 wt.%,0.2 wt.%,0.3 wt.%,0.4 wt.%nanorefrigerant with the thermodynamic equilibrium vapor quality.The results show that heat transfer coefficients slightly change with vapor quality in the range from 0.04%to 0.16%,which indicates that nucleated boiling plays an important role in nucleated region.

4.2.Analysis on the influence of nanoparticles concentration on heat transfer characteristics

Fig.10.Variation of heat transfer coefficient with vapor quality.

Fig.11.Variation of heat transfer coefficients with mass fraction Al2O3/R141b+Span-80 nanorefrigerant.

Fig.11 reveals the variation of the average heat transfer coefficients of nanorefrigerant with the mass fraction of nanoparticles in pure refrigerant R141b under the conditions of the mass flux rate from 183.1 to 457.8 kg·m-2·s-1and heat flux from 13.6 to 37.3 kW·m-2,in which the heat transfer coefficients of nanorefrigerant are always better than the pure refrigerant.The results indicate that adding Al2O3nanoparticles in R141b pure refrigerant is beneficial to strengthen heat transfer performance.The average heat transfer coefficients of 0.05 wt.%,0.1 wt.%,0.2 wt.%,0.3 wt.%and 0.4 wt.%Al2O3/R141b+Span-80 nanorefrigerant respectively increase by 55.0%,72.0%,53.0%,42.3%and 39.9%compared with pure refrigerant R141b and the heat transfer coefficient increments of different mass concentration nanorefrigerant are shown in Table 4.The heat transfer coefficients increase with increasing nanoparticle mass concentration and reaches its maximum at the mass concentration of 0.1%and then it decreases slightly.The optimal mass concentration corresponding to the best heat transfer enhancement is 0.1 wt.%.The enhanced heat trans fer coefficient of 0.1 wt.%nanorefrigerant in comparison with other mass fraction nanorefrigerant is shown in Table 5.

The experiment results show that the heat transfer performance is enhanced after adding nanoparticles in pure refrigerant(≤0.5 wt.%inour experiment)and the heat transfer coefficients increase with the increasing of mass concentrations in low concentration range(≤0.1 wt.%in our experiment).The main reasons for this effect are listed below.First,the suspended nanoparticles increase the effective thermal conductivity of the fluid and obtain higher thermal conductivity compared to the base fluid.Second,the mixing fluctuation and turbulence of the fluid are intensified as particle migration.For a fully developed-steady state nanofluid,the study indicated that mass balance for the particles satis fies Eq.(16)[27].

Table 4 Comparison of nanorefrigerant heat transfer coefficient with pure refrigerant R141

Table 5 Comparison of 0.1 wt.%nanorefrigerant heat transfer coefficient with other mass fraction nanorefrigerant

where r denotes the radial coordinate and J is the total particle flux in r direction.The total flux of particle migration consists of three terms:the particle fluxes occurred as a result of viscosity gradient Jμ,non-uniform shear rate Jcand Brownian motion Jb.Phillips et al.[28]introduced the following Eqs.(17),(18)and(19)to evaluate particle fluxes due to viscosity gradient,shear rate and Brownian diffusion.The particle fluxes from above three terms cause particles to migrate in fluid especially in the process of flow boiling heat transfer.This migration motion enhances heat transfer between nanoparticles and fluid.Therefore,the heat transfer performance of nano fluid is enhanced.

where kμand kcare constants,kBis the Boltzmann's constant,γ.is the shear rate,T denotes the temperature,and Dnpindicates the nanoparticle diameter.

[22,29,30]also get similar investigation results,which nanoparticles can enhance the heat transfer performance as the interactions among particles enhance the turbulence intensity of the fluid and reduce the boundary layer thickness.Those investigations only explain that the heat transfer coefficients increase with the increasing of mass concentrations in low concentration ranging from 0 to 0.4 wt.%but could not explain that the heat transfer coefficients nonlinearly increase with nanoparticle concentrations in this experiment.There exists an optimal nanoparticle concentration corresponding to the best heat transfer enhancement.Itmay be caused by the influence of nanoparticle deposition on the channel surface wetta bility during the flow boiling experiment.Ahmed et al.[31]experimentally investigated the effect of particle deposition on pool boiling of nano fluid and found that nanoparticles deposition have a significant effect on heat transfer.

4.3.Influence of nanoparticles deposition on heat transfer coefficient

Fig.12.SEM image of the channel surface before flow boiling experiment.

The experiment results find that the heat transfer coefficients increase with the increasing of nanoparticle concentrations in the range from 0 to 0.1 wt.%.However,the heat transfer coefficients decrease with the increasing of nanoparticle concentrations in the range from 0.1 to 0.4 wt.%as the influence of nanoparticles deposited on the channel surface wettability.Fig.12 shows the Scanning Electron Microscopy(SEM)image of the channel surface before the flow boiling experiment.Fig.13 is the SEM image of the channel surface after different nanoparticle concentration(0.05 wt.%,0.1 wt.%,0.2 wt.%,0.3 wt.%,0.4 wt.%)of Al2O3/R141b+Span-80 experiment.Fig.13(a),(b)shows that there are few nanoparticle deposition on the channel wall in low concentrations ranging from 0 to 0.1 wt.%.But with nanoparticle concentrations increasing,nanoparticles deposition on the channel wall become more and more obvious in mass concentrations ranging from 0.2 wt.%to 0.4 wt.%(see Fig.13(c),(d),(e)for details).It plays an important role in the heat transfer performance of nanorefrigerant in mini-channel as nanoparticle deposition has an in fluence on the channel surface wettability.

Fig.13.SEM image of the channel surface after different nanoparticle concentration of Al2O3/R141b+Span-80 experiment:(a)0.05 wt.%,(b)0.1 wt.%,(c)0.2 wt.%,(d)0.3 wt.%,(e)0.4 wt.%.

Fig.14.Surface contact angle of mini-channel wall after different nanoparticle concentrations of Al2O3/R141b+Span-80 experiment((a)original experimental channel wall(b)after 0.05 wt.%experimental channel wall,(c)after 0.1 wt.%experimental channel wall,(d)after 0.2 wt.%experimental channel wall,(e)after 0.3 wt.%experimental channel wall,(f)after 0.4 wt.%experimental channel wall).

The surface contact angles of the corresponding channel wall were measured using the sessile drop technique with SCA20-Software in a cleanroom at the ambient temperature.Pictures of water droplets on the channel wall after various concentrations flow boiling experiment are shown in Fig.14.The amount of water for each contact angle measurement is 1 μL.Since R141b is volatile at room temperature and atmosphere pressure and there is a certain corrosive action for test equipment.Therefore,water droplet is used to measure the channel surface contact angles instead of using R141b.Surface wettability for different channels depends on surface conditions.The surface contact angle of the same experiment channel for the mass fraction of 0.05%,0.1%,0.2%,0.3%,0.4%nanorefrigerant boiling experiment also present similar change trend if the test liquid is pure refrigerant R141b according to Wenzel's wett ability equation:cosθ =Rfcosθ0,where Rfis the channel roughness,and θ0is the ideal smooth surface contact angle[32].Reference[33]also has indicated the similar investigation results.Furthermore,the real contact angle values after boiling experiment between channel surface and R141b have been calculated according to the basic wettability equation:cosθ=(σsv-σsl)/σlv,where σsv,σsl,σlvare the surface tension of solid and vapor interface,the surface tension of solid and liquid interface,and the surface tension of liquid and vapor interface,respectively.Table 6 shows the corresponding contact angles ofthe channelusing water and R141b.Fig.15 shows thatthe surface contact angle variation with channels after different mass fractionexperiments is similar to the trend of pure R141b,which indicates that water droplet is feasible for contact angle measurement.

Table 6 Surface contact angle of mini-channel wall after different nanoparticle concentrations of Al2O3/R141b+Span-80 experiment

Fig.15.Channel surface angle after different mass fraction Al2O3/R141b+Span-80 flow boiling experiment.

In Fig.14,the original channel wall contact angle is larger than those of channel contact angle after flow boiling experiments.The reason for this phenomenon is that a hydrophilic nanoparticle deposition layer is attached on the channel wall compared with the original channel wall.It is important to note that the channel surface wettability obviously increased during 0.2 wt.%to 0.4 wt.%Al2O3/R141b+Span-80 flow boiling experiment(see Fig.14(d),(e),(f)and Table 6 for details).This is the main reason for the heattransfer performance decline in concentrations ranging from 0.2 wt.%to 0.4 wt.%.The channel surface needs more energy to produce a bubble with the wettability increasing.The formula(20)indicates that the relationship between free enthalpy generates a bubble and surface contact angle according to uniform model[34].

where r is the bubble radius,and ?v,and ?lare vapor phase free enthalpy and liquid phase free enthalpy per unit mass,respectively.σ is the surface tension of the liquid and vapor interface.Here,the bubble radius r is considered as a constant when qualitatively analyzing the influence of surface wettability on the heat transfer performance according to the reference[34]investigation analysis.

The calculated result shows that required free enthalpy for the channel after 0.1 wt.%Al2O3/R141b+Span-80 experiment is relatively increased only 0.04%comparing with the channel after 0.05 wt.%Al2O3/R141b+Span-80 nanorefrigerant experiment.The influence of nanoparticle deposition on the channel surface wettability is small during the flow boiling experiment in low concentrations ranging from 0 to 0.1 wt.%.Therefore,the effect of the surface deposition of the nanoparticles on the heat transfer performance of nanorefrigerant can be neglected.The heat transfer performance of nanorefrigerant increases with nanoparticle concentrations ranging from 0 to 0.1 wt.%.However,nanoparticle deposition on the channel wall becomes more and more obvious in concentrations ranging from 0.2 wt.%to 0.4 wt.%(see Fig.13(c),(d),(e)for details).It makes the channel surface contact angle significantly decrease after 0.2 wt.%Al2O3/R141b+Span-80 nanorefrigerant experiment.Required free enthalpy for the channel after 0.2 wt.%,0.3 wt.%,0.4 wt.%Al2O3/R141b+Span-80 nanorefrigerant experiment are relatively increased by 22.8%,22.3%and 23.3%comparing with the channel after 0.1 wt.%Al2O3/R141b+Span-80 experiment,respectively.It illustrates that the channel needs more free enthalpies producing a bubble comparing with the channel after 0.1 wt.%Al2O3/R141b+Span-80 experiment in concentrations ranging from 0.2 wt.%to 0.4 wt.%.Hence,the heat transfer performance of nanorefrigerant decreases with nanoparticle concentrations in the range from 0.2 wt.%to 0.4 wt.%.Phan et al.found that it is a nonlinear relationship between the heat transfer performance and surface wettability(contact angle).The heat transfer coefficients increase with the increasing of contact angle for low wettability surface(contact angle 20°-110°).The experimental channel contact angles after flow boiling experiment are in the range from 28.8°to 83.7°[35].The heat transfer characteristics keep consistent with Phan's investigation result in low wettability surface.Phan's investigation result can support our experiment result.

Fig.16 reveals boiling curves of Al2O3/R141b+Span-80 nanorefrigerant.It means that the relationship between the heat flux and the wall temperature superheat Δ T which is defined as Δ T=Tw-Tsat.From Fig.15 can find that the slope of qe-ΔT curve has a radical change with heat flux.It indicates that the in flection point is the onset nucleate boiling(ONB),which illustrates that the system has entered into the bubble boiling heat transfer process.But it can be noted that the boiling incipience of Al2O3/R141b+Span-80 nanorefrigerant happened at a slight wall superheat about 3.8-5.1°C much lower than the wall temperature superheat of pure R141b refrigerant 5.5°C.And the range of bubble onset boiling point for different nanoparticle mass fractions Al2O3/R141b+Span-80 nanorefrigerant almost keep consistent with Deng concluding initial bubble boiling point range about 3.2-7.2°C[36].It demonstrates that nanorefrigerant can make the wall temperature decrease easier.Thus,it enhances the heat transfer.However,the wall temperature superheat is nonlinear with the mass concentration of nanorefrigerant.The ONB temperature of 0.1 wt.%Al2O3/R141b+Span-80 nanorefrigerant is the lowest,which the wall superheat is about 3.8°C.Meanwhile,the corresponding vapor quality also presents the nonlinear relation with the increase of nanoparticles concentration,which is shown in Fig.17.The cause of the above phenomenon is the influence of nanoparticle deposition on the channel surface wettability.

Fig.16.Flow boiling curve of nanorefrigerant.

4.4.Variance analysis on influence of nanoparticle concentrations on heat transfer

Fig.17.Axial variations of the vapor quality of saturated region in the minichannel.

Table 7 Variance analysis results about the in fluence of two-factors A and B on heat transfer coefficient

Mass flux rate,nanoparticle concentrations and heat flux are defined as impact factors A,B and D,respectively.Table 7 indicates the twofactor variance analysis results of the influence of mass flux rate and concentrations on heat transfer by the homogeneity test of variances according to mathematical statistics with the data in Table 8 from Fig.9.It obeys F-distribution of the freedom degrees of(dfi,dfe).The value of Fiis calculated by Eqs.(21)-(31)and it should be thought that the factor has a significant effect on heat transfer if Fi＞Fa(dfi,dfe)=Fcritfor a given significance level of a=0.05,otherwise there is no significant effect.Tables 7,9,10 evince that mass flux rate,concentrations and heat lfux have a significant effect on the heat transfer coefficient as the value of FA,FBand FDare alllarger than Fcritunder the significance levelof0.05.In general,concentrations have a significant effect on heat transfer of Al2O3/R141b+Span-80 nanorefrigerant.

Table 8 Variation of heat transfer coefficients with different mass flow rate and the mass fraction of nanorefrigerant

Table 9 Variance analysis results about the in fluence of two-factors B and D on heat transfer coefficient

Table 10 Variance analysis results aboutthe in fluence ofimpactfactor B on heattransfer coefficient

where r,s are the levelnumbers of impact factors,r=1,2…5,s=1,2…6.MS is the mean square which is defined by the following calculation formulas.

where df,SS are the degree of freedom and the sum of squares of deviation frommean,respectively,which can be defined by the following calculation formulas.

4.5.Comparison of the experimental results with the existing correlations

Currently,heat transfer correlations from Bo and Re numbers are used to assess the effect of flow boiling heat transfer.Several correlations of heat transfer flow boiling are listed as in Table 11.

Table 11 Correlations of flow boiling heat transfer

Where Bo,Re,We and Bd are the dimensionless parameters,which can be defined by the following calculation formulas.

It is concluded that the MAE(mean absolute error)between the experimental results and the existing correlations are 21.36%,23.75%and 12.48%,respectively.And it finds that the MAE between the experimental results and the Lararek's correlation[39]is 12.48%which illustrates that this formula could effectively predict Al2O3/R141b+Span-80 nanorefrigerant flow boiling heat transfer in our experiment.There are 78.4%predicted values located in the range of±20%relative error compared with the experimental results.Fig.18 shows the comparison between the experimental results and the predicted results from Lararek's correlation.

Fig.18.Comparison of the experimental results with the predicted results of the Lararek's correlation.

The above heat transfer models'working medium are not nanorefrigerant,which did not consider the influence of nanoparticle on the heat transfer performance of nanorefrigerant during the flow boiling experiment.Therefore,it's unable to effectively predict the heat transfer characteristics.Taking nanoparticle concentrations(w)into consideration,a new correlation Eq.(37)has been proposed based on Lararek's heat transfer correlation by fitting the experimental results for different nanoparticle concentrations.

Fig.19.Comparison of the experimental results with the predicted results for Eq.(37).

Eq.(37)can be applied to nanorefrigerant flow boiling heat transfer in mini-channel at the mass concentration from 0 to 0.4 wt.%.It is concluded that the MAE between the experimental results and the new correlation is 9.87%.And Fig.19 shows that there are 94.4%predicted values located in the range of±20%for the relative error compared with the experimental results,which could illustrate that the new correlation could effectively predict nanoparticle concentrations of Al2O3/R141b+Span-80 on the in fluence of flow boiling heat transfer in micro heat exchanger.

5.Conclusions

This study has investigated the influence of nanoparticle concentrations of Al2O3/R141b+Span-80 nanorefrigerant on flow boiling heat transfer in micro heat exchanger by direct metal laser sintering through experiments.Key conclusions can be summarized as follows.

(1)In the heat transfer experiment,it is observed that the heat transfer performance can be enhanced after adding Al2O3nanoparticles into pure refrigerant R141b.The average heat transfer coefficients of 0.05,0.1 wt.%,0.2 wt.%,0.3 wt.%and 0.4 wt.%Al2O3/R141b+Span-80 nanorefrigerant increase by 55.0%,72.0%,53.0%,42.3%and 39.9%compared with pure R141b refrigerant,respectively.

(2)The mass fraction of nanoparticles has a significant effect on the heat transfer coefficient by the homogeneity test of variances.The heat transfer coefficient nonlinearly increases with increasing nanoparticle mass concentration and reaches its maximum at the mass concentration of 0.1%and then it decreases slightly.There exists an optimal mass concentration corresponding to the best heat transfer enhancement.

(3)Static contact angle test reveals that nanoparticle deposition on the channel surface makes the wettability significantly increase during the flow boiling experiment in the mass concentration range from 0.2 wt.%to 0.4 wt.%.The channel surface needs more energy to produce a bubble.Therefore,the heattransfer coefficients decrease with nanoparticle concentrations increasing when the mass fraction is more than 0.1%.

(4)A new correlation Eq.(37)is proposed by fitting the experimental data based on Lararek's correlation considering the influence of nanoparticle mass concentrations on the heat transfer performance of nanorefrigerant.And it is concluded that the MAEofthe correlations is 9.87%compared with the experimental results,which could illustrate that the new correlation can effectively predict flow boiling heat transfer of Al2O3/R141b+Span-80 nanorefrigerant in micro heat exchanger.

Nomenclature

A Impact factor about the influence of heat flux on heattransfer

B Impact factor about the influence of nanoparticle concentrations on heat transfer

Bo Boiling number

cpSpecific heat capacity

D Impact factor about the in fluence of mass flow rate on heat transfer

DhHydraulic diameter

HchDepth of the channel

h Heat transfer coefficient

hfgLatent heat of vaporization

L Region length

M Mass flow rate per a channel

m Fin parameter

N Number of channels

qeEffective heat flux based on width of unit cell containing single channel

Re Reynolds number

r Bubble radius

T Fluid temperature

ΔT Superheat

WchWidth of channel

We Weber number

WwA half width of the fin

w Mass fraction

xeVapor quality

z Distance from the entrance

δ Distance between down and up temperature test point

δ1Distance between channel bottom and up temperature test point

η Fin efficiency

θ Contact angle

λaThermal conductivity of aluminum block

μ Dynamic viscosity

ρ Density

σ Surface tension of the liquid and vapor interface

Φ Enthalpy generating a bubble

?gGas phase enthalpy

?lLiquid phase enthalpy

φ Volume fraction

Subscripts

f Liquid,bulk fluid

g Gas

in Inlet

l Liquid

nf Nano fluid

np Nanoparticles

out Outlet

r Pure refrigerant

sat Saturation

sub Subcooling

[1]H.Lee,I.Park,I.Mudawar,M.M.Hasah,Micro-channel evaporator for space applications—1.Experimental pressure drop and heat transfer results for different orientations in earth gravity,Int.J.Heat Mass Transf.77(2014)1213-1230.

[2]S.U.S.Choi,J.A.Eastman,Enhancing thermal conductivity of fluids with nanoparticles,ASME International Mechanical Engineering Congress&Exposition,11,1995,pp.99-105.

[3]S.S.Bi,K.Guo,Z.G.Liu,J.T.Wu,Performance of a domestic refrigerator using TiO2-R600a nanorefrigerant as working fluid,Energy Convers.Manag.52(1)(2011)733-737.

[4]T.Perarasu,M.Arivazhagan,P.Sivashanmugam,Experimental and CFD heat transfer studies of Al2O3-water nano fluid in a coiled agitated vessel equipped with propeller,Chin.J.Chem.Eng.21(11)(2013)1232-1243.

[5]S.M.Lu,M.Xing,Y.Sun,X.J.Dong,Experimental and theoretical studies of CO2absorption enhancement by nano-Al2O3and carbon nanotube particles,Chin.J.Chem.Eng.21(9)(2013)983-990.

[6]S.Parvin,M.A.Alim,N.F.Hossain,Prandtl number effecton cooling performance of a heated cylinder in an enclosure filled with nano fluid,Int.Commun.Heat Mass Transfer 39(8)(2012)1220-1225.

[7]S.W.Lee,K.M.Kim,I.C.Bang,Study on flow boiling critical heat flux enhancement of graphene oxide/water nano fluid,Int.J.Heat Mass Transf.65(2013)348-356.

[8]S.Vafaei,D.S.Wen,Critical heat flux of nano fluids inside a single microchannel:Experiments and correlations,Chem.Eng.Res.Des.92(11)(2014)2339-2351.

[9]R.Kamatchi,S.Venkata chalapathy,Parametric study of pool boiling heat transfer with nano fluids for the enhancement of critical heat flux:A review,Int.J.Therm.Sci.87(2015)228-240.

[10]Alawi,N.A.C.Sidik,M.H.Beriache,Applications of nanorefrigerant and nanolubricants in refrigeration,air-conditioning and heat pump systems:A review,Int.Commun.Heat Mass Transfer 68(2015)91-97.

[11]W.T.Jiang,G.L.Ding,H.Peng,Y.F.Gao,K.J.Wang,Experimental and model research on nanorefrigerant thermal conductivity,HVAC&R Res.15(2009)651-669.

[12]D.S.Wen,In fluence of nanoparticles on boiling heat transfer,Appl.Therm.Eng.41(2012)2-9.

[13]R.Saidura,S.N.Kazia,M.S.Hossaina,M.M.Rahmanb,H.A.Mohammed,A review on the performance of nanoparticles suspended with refrigerants and lubricating oils in refrigeration systems,Renew.Sust.Energ.Rev.15(1)(2011)310-323.

[14]M.S.Patil,S.C.Kim,J.H.Seo,M.Y.Lee,Review of the thermo-physical properties and performance characteristics of a refrigeration system using refrigerant-based nano fluids,Energies 9(1)(2015)22-39.

[15]H.Peng,L.N.Lin,G.L.Ding,In fluences of primary particle parameters and surfactant on aggregation behavior of nanoparticles in nanorefrigerant,Energy 89(2015)410-420.

[16]Alawi,N.A.C.Sidik,In fluence of particle concentration and temperature on the thermophysical properties of CuO/R134a nanorefrigerant,Int.Commun.Heat Mass Transfer 58(2014)79-84.

[17]M.H.U.Bhuiyan,R.Saidur,M.A.Amalina,R.M.Mosta fizur,A.Islam,Effect ofnanoparticles concentration and their sizes on surface tension of nano fluids,Procedia Eng.105(2015)431-437.

[18]I.M.Mahbubul,S.A.Fadhilah,R.Saidur,K.Y.Leong,M.A.Amalina,Thermophysical properties and heat transfer performance of Al2O3/R134a nanorefrigerants,Int.J.Heat Mass Transf.57(1)(2013)100-108.

[19]I.M.Mahbubul,A.Saadah,R.Saidur,M.A.Khairul,A.Kamyar,Thermal performance analysis of Al2O3/R134a nanorefrigerant,Int.J.Heat Mass Transf.85(2015)1034-1040.

[20]I.M.Mahbubul,R.Saidur,M.A.Amalina,In fluence of particle concentration and temperature on thermal conductivity and viscosity of Al2O3/R141b nanorefrigerant,Int.Commun.Heat Mass Transfer 43(2013)100-104.

[21]K.Henderson,Y.G.Park,L.P.Liu,A.M.Jacobi,Flow-boiling heat transfer of R134a based nano fluids in a horizontal tube,Int.J.Heat Mass Transf.53(5-6)(2010)944-951.

[22]X.F.Yang,Z.H.Liu,Flow boiling heat transfer in the evaporator of a loop the rmosyphon operating with CuO based aqueous nano fluid,Int.J.Heat Mass Transf.55(2012)7375-7384.

[23]L.Ventola,F.Robotti,M.Dialameh,F.Calignano,D.Manfredi,E.Chiavazzo,P.Asinari,Rough surfaces with enhanced heat transfer for electronics cooling by direct metal laser sintering,Int.J.Heat Mass Transf.75(2014)58-74.

[24]S.Kandlikar,W.Grande,Evolution of microchannel flow passages—thermohydraulic performance and fabrication technology,Heat Transfer Eng.24(2003)3-17.

[25]H.Peng,G.L.Ding,H.T.Hu,Effect of surfactant additives on nucleate pool boiling heat transfer of refrigerant-based nano fluid,Exp.Thermal Fluid Sci.35(2011)960-970.

[26]R.Kamatchi,S.Venkatachalapathy,B.Abhinaya Srinivas,Synthesis,stability,transport properties,and surface wettability of reduced graphene oxide/water nano fluids,Int.J.Therm.Sci.97(2015)17-25.

[27]Y.Ding,D.Wen,Particle migration in a flow of nanoparticle suspensions,Powder Technol.149(2005)84-92.

[28]R.J.Phillips,R.C.Armstrong,R.A.Brown,A.L.Graham,J.R.Abbott,A constitutive equation for concentrated suspensions that accounts for shear-induced particle migration,Phys.Fluids A 4(1992)30-40.

[29]J.A.Eastman,U.S.Choi,S.P.Li,L.J.Thompson,S.Lee,Enhanced thermal conductivity through the development of nano fluids,Department of Energy,Cambridge Univ Press,1996 2-6.

[30]Prajapati,N.Rohatgi,Flow boiling heat transfer enhancement by using ZnO-water nano fluids,Sci.Technol.Nucl.Installations(2014)1-7.

[31]O.Ahmed,M.S.Hamed,Experimental investigation of the effect of particle deposition on pool boiling of nano fluids,Int.J.Heat Mass Transf.55(2012)3423-3436.

[32]R.N.Wenzel,Resistance of solid surfaces to wetting by water,Ind.Eng.Chem.28(1936)988-994.

[33]X.P.Cao,Y.M.Jiang,Frictional property of wetting contact line and Wenzel's behavior of solid surface tension,Acta Phys.Sin.54(05)(2005)2202-2205.

[34]S.J.Kim,I.C.Bang,J.Buongiorno,L.W.Hu,Effects of nanoparticle deposition on surface wettability in fluencing boiling heat transfer in nano fluids,Appl.Phys.Lett.89(15)(2006)1531071-15310714.

[35]H.T.Phan,N.Caney,Surface wetta bility control by nanocoating:The effects on pool boiling heat transfer and nucleation mechanism,Int.J.Heat Mass Transf.52(2009)5459-5471.

[36]D.X.Deng,W.Wan,H.R.Shao,Y.Tang,J.Y.Feng,J.Zeng,Effects of operation parameters on flow boiling characteristics of heat sink cooling systems with reentrant porous microchannels,Energy Convers.Manag.96(2015)340-351.

[37]R.Yun,J.H.Heo,Y.C.Kim,Evaporative heat transfer and pressure drop of R410A in microchannels,Int.J.Refrig.29(1)(2006)92-100.

[38]W.Li,Z.Wu,A general correlation for evaporative heat transfer in micro/minichannels,Int.J.Heat Mass Transf.53(9-10)(2010)1778-1787.

[39]G.M.Lazarek,S.H.Black,Evaporative heat transfer,pressure drop and critical heat flux in a small vertical tube with R-113,Int.J.Heat Mass Transf.25(1982)945-960.