当前位置：文档库 › Effect of fabrication parameters on capillary performance of composite wicks

Effect of fabrication parameters on capillary performance of composite wicks

Effect of fabrication parameters on capillary performance of composite wicks for two-phase heat transfer devices

Yong Tang a ,Daxiang Deng a ,b ,?,Guanghan Huang a ,Zhenping Wan a ,Longsheng Lu a

Key Laboratory of Surface Functional Structure Manufacturing of Guangdong High Education Institutes,School of Mechanical and Automotive Engineering,South China University of Technology,Guangzhou 510640,China b

Department of Mechanical Engineering,The Johns Hopkins University,Baltimore,MD 21218,USA

a r t i c l e i n f o Article history:

Received 30August 2012

Received in revised form 23September 2012

Accepted 23September 2012

Keywords:

Two-phase heat transfer devices Heat pipes

Composite wick

Capillary performance Permeability Capillary rise

a b s t r a c t

Advanced thermal management solutions for various applications have promoted the development of composite wicks for two-phase heat transfer devices (TPHTDs).In this study,a number of composite wicks by covering a layer of sintered copper powder on micro V-grooves were developed.A plough-ing–extrusion (P–E)method,as a material-saving fabrication means,was utilized to process the micro V-grooves on copper plate as the base of composite https://www.wendangku.net/doc/a013407383.html,ing an infrared (IR)thermal imaging method,the capillary rate-of-rise tests with ethanol and acetone were carried out to characterize the cap-illary performance,which integrates both capillary pressure and permeability.The effects of fabrication parameters,including groove depth and pitch,sintering temperature and time,on the capillary perfor-mance of composite wicks were focused on and examined for the purpose of design optimization.Test results show that there is an optimal groove geometry with the groove depth of 0.85mm and pitch of 0.45mm to achieve the maximum capillary performance,and sintering processes of 950°C along with 30min should be chosen.Both working liquid test results exhibit fairly good agreement and demonstrate that the IR thermal imaging provides an accurate means to evaluate the hydraulic properties of composite wicks.

1.Introduction

Two-phase heat transfer devices (TPHTDs),such as heat pipes,vapor chambers (heat spreader),loop heat pipes and capillary pumped loops have played an increasingly crucial role in thermal management of microelectronic systems,spacecraft components,heat recovery systems and light emitting diode (LED)lighting mod-ules,etc.[1,2].Relying on evaporation and condensation processes and the circulation of working liquid,TPHTD possess a large heat transport capacity.A wick,as a key element of TPHTD,provides the capillary pressure for driving the two-phase circulation,and also serves as the ?ow path for the return of working liquid.It is usually composed of homogeneous microgrooves,sintered pow-der,or meshes.For the aim of achieving high heat transfer rates,especially in high heat ?ux situations,a wick should have large capillary pressure and high liquid permeability to facilitate liquid return.However,these two demands are apparently contrasting for a homogeneous wick [3],as the former is usually maintained by porous structure with numerous of small pores,while the latter requires a grooved type.Therefore,homogeneous wicks must balance capillary pressure and liquid permeability to achieve some compromise in performance.In order to address this dilemma,more sophisticated,or composite wicks have been developed in both scienti?c and industrial areas [4–12].Composite wicks with combination of sintered metal powders and mesh screens were proposed and tested by Canti et al.[4],and high capillary pressure head and low liquid and vapor pressure drop were registered.Hybrid wick ?at heat pipes combined rectangular groove with meshed layers were proposed for cooling LED lighting module by Hsieh et al.[6],and superior thermal performances were achieved at inclined angles.In addition,Oshman et al.[7]devel-oped a hybrid wick heat spreader with woven mesh bonded atop rectangular microgrooves and found that it was able to eliminate the gravitational effect on working liquid,thus can work at adverse gravitational and dynamic acceleration situations.Furthermore,composite wick combined rectangular grooves with deposited por-ous coating atop was developed and tested by Grote et al.[8]and Khrustalev et al.[9],and was further analyzed numerically by Khrustalev and Faghri [10].The evaporative heat transfer coef?-cient and capillary pumping ability were found to increase signi?-cantly due to the presence of porous coating.Wang and Cotton [11]theoretically analyzed the evaporative heat transfer of triangular groove covered with a thin porous layer.The results exhibited a

Corresponding author.Tel./fax:+862087114634.

E-mail address:dengdaxiang88@https://www.wendangku.net/doc/a013407383.html, (D.Deng).

three to six times increase in evaporative heat transfer compared

to the groove without a porous layer.Tang et al.[12]developed a composite wick consisted of micro-groove and sintered powder, and found that it exhibited a superior capillary pumping ability to the sintered and grooved wicks.

As enumerated above,the fabrication of composite wicks plays an important role in the thermal and hydraulic performance of TPHTDs.Thus,it necessitates an in-depth investigation on the fab-rication methods and parameters of composite wicks for the opti-mization of TPHTDs.In general,the fabrication of microgrooves employs a rolling[13],dicing[14],wire electrical discharge machining(EDM)[15],chemical etching[16]or laser microma-chining method[17].The porous powder or mesh wick are gener-ally fabricated by sintering[18–20].There are suf?cient researches concerning the effects of fabrication parameters on the perfor-mance of homogeneous wicks,such as Ma and Peterson[21],Hop-kins and Faghri[13],Suman and Hoda[22],Hanlon and Ma[23], and Williams and Harris[24].However,to the authors’best knowl-edge,except for the analytical study of the triangular groove cov-ered with a porous layer regarding the effect of groove geometry on the evaporative heat transfer performance by Wang and Cotton [11],the fabrication parameter effects on the hydraulic and ther-mal performance of composite wicks of TPHTD are scarcely re-ported.In view of this situation,it is essential to conduct such experiments to provide an in-depth insight into this for the opti-mal design of composite wicks.It is also the focus of this work.

As stated before,capillary pressure and permeability of a wick are two key parameters for TPHTDs,which determine the capillary limit and the heat transfer performance.In general,these two parameters are characterized by experiment separately,as the cap-illary pressure of a wick is usually assessed by bubble point test [25]or capillary rate-of-rise method[26],and the permeability is determined by the forced liquid?ow method[25].In recent years, capillary rate-of-rise experiments coupled with analysis method have been found to be able to determine these two properties of a porous wick by Holley and Faghri[27].Byon and Kim[28,29]also used capillary rate-of-rise experiments in conjunction with numer-ical simulations to characterize the capillary performance parame-ter(K/R eff)of copper wicks with micropost arrays and bi-porous glass wicks.In order to characterize the capillary performance of wicks using capillary rate-of-rise tests,it is crucial to locate the risen meniscus accurately and obtain the capillary rise height precisely.Nevertheless,as most of the working liquid is colorless and transparent,the risen meniscus may be vague in wicks[27] if observed by optical sight,such as the CCD camera.While using another means,i.e.,weight methods[27],the attachment of an outer meniscus may results in a large mass gain which cannot be distinguished from the?nal results[30].Hence,this issue should be addressed to employ the capillary rate-of-rise method to characterize the capillary performance of wicks.

In our previous studies[12,31],a novel infrared thermal image method was developed to accurately record the capillary rate-of-rise processes of wicks using ethanol as working liquid,and the ef-fects of copper powder size and shape on the capillary performance of composite wicks,which integrates both permeability and capil-lary pressure,were assessed.However,the in?uences of fabrication and geometrical parameters on the capillary performance of com-posite wicks are still unknown.It is necessary to perform such studies to get the comprehensive information for the design opti-mization of the composite wicks.In the present study,a mate-rial-saving microgroove fabrication method was presented,and the effects of groove depth,groove pitches,sintered temperature, sintered time and working liquid on the capillary performance of composite wicks were systematically evaluated.

2.Experimental

2.1.Fabrication of micro V-grooves and preparation of composite wick samples

The?rst step to fabrication the composite wick is to form the micro V-grooves in plates of pure copper.Unlike the aforemen-tioned fabrication means[13–17],in this study we employed a ploughing–extrusion(P–E)method to fabricate the micro V-grooves.Upon developed by our research group[32,33],the P–E method,as a special kind of drawing process,enables the fabrica-tion of high-aspect-ratio micro grooves material-saving,thus very economic.A specially designed tool,namely,ploughing–extrusion tool,was utilized in the fabrication process.By grinding using steel W18Cr4V,the P–E tool included a ploughing edge,a major extru-sion face A c,a minor extrusion face A0c,a major forming face A b,a minor forming face A0

,and a tool?ank A a,as illustrated in

Nomenclature

A p cross-section area of the sintered porous layer(m2)

g gravitational acceleration(kg/m2s)

h capillary rise height(mm)

h g depth of grooves(mm)

h g/w aspect ratio of grooves(dimensionless)

h pe ploughing–extrusion fabrication depth

h t total height of grooves(mm)

d h=d t capillary ris

e velocity(mm/s)

p pitch of groove

t sintering time(min);capillary rise time(s)

T sintering temperature(°C)

w width of grooves(mm)

K permeability(mm2)

L length of wick(mm)

m p mass of sintered powder(g)

r m meniscus radius(m)

r eff effective capillary radius(m)

W width of wick(mm)Greek symbols

D P pressure drop(kPa)

D P cap capillary pressure(Pa)

e porosity,dimensionless

d thickness of wick(mm)

q density of working liquid(kg/m3) q Cu density of solid copper(kg/m3)

h liquid–solid contact angle,rad

l dynamic viscosity of ethanol(Pa s) r surface tension of ethanol(N/m)

Subscripts

g groove

pe ploughing–extrusion

t total

cap capillary

p pore

Cu copper

eff ef?cient

Y.Tang et al./Energy Conversion and Management66(2013)66–7667

Fig.1a.A c and A0c,A b and A0

were symmetric with the ploughing edge as the center line.The O–O cross-section of the tool was a wedge structure(a–tool clearance of P–E tool,K r–edge inclina-tion angle,b–forming angle,c0–extrusion angle).As depicted in Fig.1,the fabrication mechanisms of micro V-grooves are as fol-lows:The front end of ploughing edge ploughs the metal and makes the metal?ow along the major extrusion face A c and the minor one A0c due to the plastic deformation of copper.Then A c ex-trudes the split metal and makes metal?ows along the major forming face A b,and the groove?n forms under the extruding ef-fect of A b.The minor extrusion face A0c trims the?ns and makes them higher.Meanwhile,the minor forming face A0

can trim and extrude the adjacent grooves in the last run,resulting that the ratio of depth to width is further increased.Owing to this trimming and extruding effect,the?ns of micro V-grooves may become inclined if the spacing interval of grooves is small and the P–E depth is large,and the top of the groove?ns may fall apart form the?n base,as shown in Figs.2e and3a.From the above,one can see that the micro V-groove is formed not by cutting the copper base and diminishing the material,but by extrusion due to the excellent plastic deformation characteristic of copper,thus there are seldom cutting chips during the forming processes.

In the present study,the parameters of P–E tool were c0=35°, b=25°,K r=40°.The fabrication processes were carried out on a planer.Grooves samples with different geometries were fabricated as the base of the composite wicks,in which?ve samples were of different P–E depth ranged from0.4to0.8mm with the same pitch of0.45mm,and six were of different groove pitches ranged from 0.35to0.6mm with the same P–E depth of0.7mm.Each sample has a width of10mm and length of100mm.The dimensions of these samples were listed in Table1.

With the P–E depth increasing from0.4mm to0.7mm,the groove depth increased from0.6mm to0.85mm,as can be seen in Fig.2.In general,the groove depth was0.15mm higher than the P–E depth except that the case of P–E depth of 0.4mm was0.2mm.This unique phenomenon is attributed to the extrusion effect of the tool which makes the groove?ns grew higher than the actual cutting depth.However,this was not true for the P–E depth of0.8mm,as in this case the groove depth was slightly smaller than the P–E depth.As the groove depth reaches a critical point,the tool extrudes and trims the ?ns so seriously that groove?n became inclined and the groove was much narrower than the others in the lower portion.Thus the groove depth did not increase.Similar groove shapes can be seen in the case of groove pitch of0.35mm,as shown in Fig.3a.With the groove pitch increasing from0.35mm to 0.6mm,the groove width was nearly as same as the pitches. At pitch of0.6mm,the top of groove?n was not a single peak, as the extrusion effect of the tool on the adjacent grooves in the last run was relatively weak in this case.The plastic defor-mation areas in two consecutive fabricating runs were just intersected,without overlapping.

After the forming of grooves,a layer of copper powder was sintered on top of the micro V-grooves.Loose sintering method was used and the sintering module was sketched in Fig.1c.As our previous study’s results[31]show that the irregular shaped powder with the size of75–110l m exhibits superior capillary performance to the spherical shaped ones and other powder sizes, only the irregular shaped copper powder of75–110l m was selected for the composite wicks in this study.This kind of copper powder,produced by water atomization,is of purity over99.5% and supplied by Beijing Xinrongyuan Powder Tech.Co.,China. The sintering process was carried out in a box-type furnace with a programmable temperature controller under hydrogen protec-tive atmosphere.For the purpose of optimization of sintering process,four samples sintered at different temperatures(850, 900,950,1000°C)with the sintering time of60min,and three samples sintered for different times(30,60,90min)at the same temperature of950°C were fabricated.All the samples were 10mm wide and100mm long.The thickness of the powder layer above the top of groove?n was0.25mm.All the samples are listed in Table1

68Y.Tang et al./Energy Conversion and Management66(2013)66–76

2.2.Capillary rate-of-rise experiment

The capillary rate-or-rise experiments were conducted using an IR thermal imaging method,as sketched in Fig.4.Two common working liquid for heat pipes,ethanol and acetone,were used. The wick samples were dipped vertically into the liquid reservoir, and the liquid spontaneously rose along the wick due to the capil-lary pressure of wick.Two or more samples can be?xed in the sample holder and tested simultaneously in each run for the con-venience of comparison.The liquid level of reservoir is thought to be constant during the capillary rise process due to the very small amount of the absorbing liquid in wicks.The glass cover makes sure that the test space nearly saturated with working liquid vapor,thus the evaporation of working liquid during the

Images of micro V-grooves fabricated with different groove pitches:(a)–(f)correspond to the pitches of0.35–0.6mm,respectively.The ploughing–extrusion same(0.7mm).

capillary rise processes can be ignored.A FLIR ThermaCAM SC3000IR camera,with a thermal sensitivity of 0.02K at 30°C and an accu-racy of 1%for temperatures below 150°C of the full scale,was uti-lized to capture the capillary rate-of-rise processes through the observing hole.The meniscus can be accurately identi?ed by the meniscus locating method in the IR thermal imaging software as described in Ref.[12],and the capillary rise height over time can be obtained with an uncertainty of <0.3mm.Before the experi-ments,all the test samples were dried for half an hour for the pur-pose of deaeration.This was also done after the ethanol test and before the acetone test.The tests were conducted in a close room with the ambient temperature of 24.7±0.1°C.The capillary rise time for each test run is 120s.The capillary rise height was ex-tracted from the IR thermal imaging software at an interval of 1s during the ?rst 10s since in this period the height changes dramat-ically,then every 2s during the 10–20s,subsequently every 5s during the 20–90s,and every 10s in the last 30s.3.Theory and data reduction

During the capillary rate-of-rise experiment,the work liquid climbs spontaneously along the wick.The following assumptions hold for the present study [27]:(i)one-dimensional and steady-state laminar ?ow in the wick,(ii)uniform saturation with liquid

along the wetted length,(iii)no inertial and entry effects in the li-quid reservoir;(iv)evaporation of liquid and inertial effects is ne-glected as same as other studies [28,29],as it was minimized by the closed space maintained by the closed space.

The capillary pressure of the wicks generated by the liquid–va-por interface plays the driving force,which is known as the La-place–Young equation,

D P cap ?

2r cos h r p

e1T

where r is the surface tension of liquid,r p is the pore radius,and h is contact angle formed between solid and liquid.In order to charac-terize the wicks more simply,the pore radius and the contact angle cosine is usually replaced by effective capillary radius (r eff ),thereby yielding

D P cap ?

2r r eff

e2T

During the capillary rise processes,due to momentum balance,the capillary pressure should equal the pressure loss,as presented as follows,

2r r eff ?l e K h

d h

d t

tq gh e3T

Table 1

Test sample speci?cations.Sample code

Dimensions of wicks Groove fabrication parameters Final groove geometries Sintering parameters Porosity

Length (L )?width (W )?thickness (d )(mm)

P–E depth grooves (h pe )Pitch of grooves p (mm)Depth of grooves h g (mm)Width of grooves w (mm)Aspect ratio of groove h g/w Sintering

temperature T (°C)Sintering time t (mm)D1100?10?1.50.40.450.60.46 1.30950600.589D2100?10?1.50.50.450.660.46 1.43950600.594D3100?10?1.50.60.450.750.45 1.67950600.592D4a 100?10?1.50.70.450.850.45 1.89950600.596D5100?10?1.40.80.450.780.44 1.75950600.591P1100?10?1.40.70.350.770.34 2.29950600.593P2100?10?1.50.70.400.850.40 2.13950600.599P3a 100?10?1.50.70.450.850.45 1.89950600.596P4100?10?1.50.70.500.850.49 1.73950600.588P5100?10?1.50.70.550.850.55 1.55950600.590P6100?10?1.50.70.600.840.60 1.4950600.587T1100?10?1.50.70.450.850.45 1.89850600.598T2100?10?1.50.70.450.850.45 1.89900600.597T3a 100?10?1.50.70.450.850.45 1.89950600.596T4100?10?1.50.70.450.850.45 1.891000600.597t1100?10?1.50.70.450.850.45 1.89950300.594t2a 100?10?1.50.70.450.850.45 1.89950600.596t3

100?10?1.50.7

0.45

0.85

0.45

1.89

950

0.591

The same sample.

IR camera

Computer

Sample holder Stage Gravity

Controller

Glass cover Observing hole

Y.Tang et al./Energy Conversion and Management 66(2013)66–76

in which the?rst item on the right hand side is viscous friction ex-pressed by Darcy’s law,and the second one is the hydrostatic pres-sure produced by gravity.l is the viscosity of working liquid,h is the capillary rise height,d h=d t is the capillary rise velocity,q is the liquid density,g is the gravitational acceleration,e is the poros-ity of wick structure,which is generally determined by a density method,

e?1àm p

A p L q Cu

e4T

where q Cu is the density of solid copper,A p is the cross-section area of the porous powder layer,m p is the mass of sintered powder of wicks.

If neglecting the gravity effect in the earliest stage of capillary rise process,integration of Eq.(3)with the initial condition h(t?0)=0leads to the well-known Washburn’s equation[34]

h2?

eff K

l e te5T

Rewriting Eq.(5)using the capillary performance parameter,i.e., the product of capillary pressure and permeability(D P capáK),the following can be obtained

D P capáK?l e h2

e6T

It can be seen that there is a linear correlation between the term D P capáK and h2=t by the Washburn’s equation.

If the gravity effects cannot be neglected,the correlation of cap-illary rise height versus time is a complicated transcendental equation

à2r

r eff

ln1à

q gr

eff

tq gh

?K q2g2

e l te7T

Holley and Faghri[27]provided a closed form solution for Eq.(7) and obtained the K and r eff separately.Unfortunately,the capillary performance parameter D P capáK cannot be obtained by this meth-od.A different treatment is required.

By integrating the parameter D P capáK,Eq.(3)can be rewritten as follows

D P capáK?h l e d h

d t tq gK

e8T

Therefore,the capillary performance parameter is determined by the capillary rise height h,capillary rising velocity d h=d t and perme-ability K with the known viscosity of working liquid and porosity of wick.The rise velocity,d h=d t,can be obtained from the derivative of the curve of capillary rise height versus time[35].Rearranging Eq.

(8)yields

eD P capáKTá1

àq gK?l e d h d te9T

If we set x?1=h,y?d h=d t,which represents the reciprocal of risen height and the value of capillary rise velocity,respectively, Eq.(9)can be rewritten as

y?D P capáK

l eáxà

q gK

l ee10T

From Eq.(10),a best linear?tting can be performed with the sets of x and y from the experiments,and the capillary performance parameter D P capáK can be determined from the slope of this?tting line.

In the present study,since the capillary rise time is long(120s), the gravitational effect cannot be ignored.Meanwhile,in our pre-vious study[31],it was demonstrated that the quadratic Wash-burn’s equation(Eq.(5))should not be applied to characterize the capillary rate-of-rise processes of wicks,as it resulted in a greatly underestimated value of capillary performance parameter. Therefore,Eq.(10),rather than Eq.(6),is used to determine the D P capáK for two working liquid respectively.Once the D P capáK is determined,the parameter K/r eff can be calculated using Eq.(2).

The measurement uncertainties of the cross-sectional dimen-sions of the groove,the thickness of the sintered porous layer and the length of composite wick were estimated to be3%,1.5% and0.5%,respectively.The mass measurement of sintered powder of wicks by the precise balance yields an uncertainty of0.5%.From a standard error analysis method[36],the uncertainty of porosity was calculated to be4.5%.The thermophysical properties of the two test liquid were collected from Ref.[37]based on the ambient temperature and the environmental pressure of1atm,and the uncertainty due to possible variation of the ambient temperature was assumed to be0.5%.Using the IR thermal imaging method, the measurement uncertainty of capillary rise height is well con-trolled to be within1.5%.The uncertainty of capillary rise velocity was estimated to be2.5%.The standard deviation of capillary per-formance parameters by linear?tting of Eq.(10)was calculated separately and presented using the error bars.

4.Results and discussion

4.1.Effect of groove depth

Fig.5depicts the capillary rise processes for the composite wicks fabricated at?ve different P–E depths.One can see that dur-ing the early times of capillary rise process,the difference of capil-lary rise height and velocity was small for these samples with different groove depths.However,the difference became distinc-tive after a period of rising.Sample D3and D4,with the P–E depth of0.6mm and0.7mm respectively,produced the largest capillary rise height and velocity.The capillary performance parameters were also found to be larger than the others,as shown in Fig.6. The capillary performance of composite wicks increased as the P–E depth increased from0.4mm to0.7mm,but decreased as the P–E depth reached0.8mm.This indicted that the there exists an optimum P–E depth(0.7mm)and corresponding groove depth of0.85mm for the composite wicks that have a maximum capil-lary performance.Interestingly,this is consistent with previous re-searches for the triangular groove wicks[38,39].Since other parameters such as the sintered powder and the groove pitch and width were identical,the variation of D P capáK can be attrib-uted to the effect of groove depth.Though most of the micro V-grooves were covered by porous powder,the apex of the groove provides free?ow channel for the working liquid,as can be seen in Fig.1d.As the groove depth increased,the groove became nar-rower and much acuter in the apex.This induces a decrease in the meniscus radius,thereby yielding a larger capillary pressure. However,on the other hand,the friction resistance due to dissipa-tion of energy and pressure drop in liquid?ow also increases at a narrower?ow passage,thereby yielding a decrease of permeabil-ity,inducing further decrease of the capillary?ow rate.The maxi-mum capillary performance with respect to different grooves depths occurs when the trade-off of the capillary pressure and fric-tion resistance reaches an optimum,which coincides with the sit-uation in which the groove is just enough to house the meniscus [39].When the P–E depth increased from0.4mm(D1)to0.7mm (D4),the groove depth increased from0.6mm to0.85mm.The in-crease of capillary pressure plays the major role in the capillary rise processes,thereby inducing the increasing trend of D P capáK.Nev-ertheless,for the sample D5,the large friction resistance of the grooves dominates as the grooves became very narrow,thus the D P capáK was inferior to the sample D4.

Y.Tang et al./Energy Conversion and Management66(2013)66–7671

4.2.Effect of groove pitch

The capillary rise processes of six samples with different groove pitches were depicted in Fig.7.The sample P3(D4)with groove pitch of0.45mm exhibited the largest capillary rise height and velocity,while the sample P6with groove pitch of0.6mm presented the smallest ones.The D P capáK?rstly increased with the increase of groove pitches,reached a maximum at groove pitch of0.45mm,and then decreased,as can be seen in Fig.8. The trend is similar to that of the aforementioned cases of differ-ent P–E depths.The trade-off between the capillary pressure gen-eration and the?ow frictional resistance of grooves is also believed to attribute to this.As the groove pitches increased, the groove became wider and the groove apex changed to be less acute.The capillary pressure generated in the groove tended to decrease,while the friction resistance also decreased.The maxi-mum capillary performance existed when these two combining effect reached the optimum.Before this optimum point,the cap-illary performance increased;beyond that it would tend to decrease.

If considering the depth–width aspect ratio of groove,one can see that the capillary performance of composite wicks mostly fol-lowed a monotonic order with the aspect ratio of groove,though two cases(P1,P2)did not follow this trend as the decreasing of per-meability dominates the capillary performance.As the groove pitch increases from0.45mm to0.6mm,the D P capáK decreased with the decrease of the aspect ratio.This is also true for the sam-ples D1to D5,in which the D P capáK follow an increasing trend with the increase of aspect ratio.This trend accords with the

72Y.Tang et al./Energy Conversion and Management66(2013)66–76

previous researches about the grooved wicks[13,40].As analyzed by Xu and Carey[40]and experimentally demonstrated by Hop-kins and Faghri[13],microgrooved wicks with deep and narrow grooves presented a larger capillary limit,hence enhanced the heat transport rates of heat pipes.Though in composite wicks the micro V-grooves were covered by porous powders,it seems that the as-pect ratio of groove exerts similar in?uence on the capillary perfor-mance of composite wicks.4.3.Effect of sintering temperature

Fig.9reveals the effect of the sintering temperature on the cap-illary rise processes of composite wicks.For both working liquid, the sample T3with the sintering temperature of950°C mostly exhibited the largest capillary rise height and velocity.Fig.10illus-trates a series of consecutive IR thermal images in the early stages of capillary rise processes of acetone in these four composite wicks. It can be seen that the IR thermal image camera clearly captured the advancing process of meniscus in each samples.

As shown in Fig.11,the D P capáK of sample T3(950°C)was found to be the largest,following is the sample T2(900°C),T4 (1000°C)and the minimum T1(850°C).Therefore,the capillary performance of composite wicks did not follow a monotonic order with the increase of sintering temperature.However,the differ-ences of D P capáK were not very distinctive,indicating that the ef-fect of sintering temperature on the capillary performance is not as evident as the aforementioned groove geometries.The morphology of sintered porous matrix of these samples from the observations of scanning electron microscopy(SEM)were similar to those in our previous work[12]and it is not presented here.During the sin-tering processes,due to the bonding by diffusion and agglomera-tion of powder particles,a network of large inter-agglomerate pores formed and interconnected among the clusters of particles. Also,there are numerous of small intra-agglomerate pores in the powder itself caused by incomplete joining between the lumps of crystal grains.As indicted by Leong et al.[41],the size and num-bers of these two kinds of pores changed along with the sintering

Fig.10.IR images of capillary rise processes of four composite wicks fabricated at different sintering temperatures during the?rst ten seconds.From left to right,it represents the sample T4(1000°C),T3(950°C),T2(900°C),and T1(850°C),respectively.The working liquid is acetone.

Y.Tang et al./Energy Conversion and Management66(2013)66–7673

temperature,that is,the numbers of inter-agglomerate pores re-duced but the intra-agglomerate pores increased with the increase of sintering temperature.This induced the increase of the capillary pressure,which is due to the existence of small meniscus radius caused by small intra-agglomerate pores;however,on the other hand,it also resulted in the reduction of permeability as the?ow resistance in inter-agglomerate pores increased.The overall capil-lary performance of composite wicks was in?uenced by these two competing factors.Therefore,as the sintering temperature change to be higher,the capillary performance of composite wicks in-creased?rst and reached the maximum at950°C,and then de-creased at the1000°C.

4.4.Effect of sintering time

The capillary rise processes for the composite wicks with three different sintering times were plotted in Fig.12.It can be noted that as the longer sintering time seems to be unfavorable for the capillary rise processes of composite wicks.Fig.13presents a monotonic decrease tendency of D P capáK as the sintering time in-creased from30min to90min.During the powder sintering,the impact of sintering time on the sintered matrix was somewhat analogous to that of sintering temperature.The size and numbers of the aforementioned two kinds of pores in composite wicks chan-ged under different sintering time.It was believed that with the longer sintering time,the reduction of inter-agglomerate pores plays the dominant role for the?ow passages of working liquid, thereby yielding a larger?ow resistance and further a reduced D P capáK.Therefore,in the present study,it suggested that the com-posite wicks should be fabricated at the sintering temperature of 950°C,sintering time of30min,along with the P–E depth of 0.7mm and groove pitch of0.45mm.

4.5.Evaluation of capillary performance by two working liquid

From Figs.5,7,9and12,it is evident that the capillary rise pro-cesses of acetone were much quicker than ethanol,as the viscosity of acetone is much lower than ethanol,whereas the surface tension of these two liquids were almost the same.Conversely,the D P cap-áK obtained by the acetone test was slightly smaller than the eth-anol,which is true for all the samples,as can be seen in Fig.6,8,10 and13.However,the discrepancies of D P capáK of composite wicks between the ethanol and acetone tests were all less than4.7%,indi-cating a good agreement.If utilizing the parameter K/r eff to charac-terize the capillary performance of wicks without considering the surface tension effect of working liquid,the discrepancies were slightly larger due to the slight difference of surface tension be-tween these two liquids,but all within the range of8.3%,as can be seen in Table2.Possible reason may be the difference of the contact angle formed between these two working liquid and the copper powder and solid groove of the composite wicks.Addition-ally,if considering the experimental uncertainties,the capillary performance parameters obtained by different working liquid seems to agree fairly well with each other.Therefore,using the capillary rate-of-rise experiments together with the IR thermal imaging method,it can provide an accurate means to evaluate the hydraulic properties of wicks.This,obviously,extend the appli-cation of IR thermal imaging method in the area of heat transfer,as normally it was just used to explore the thermal properties of heat transfer devices[42].

It is also noted that as the capillary rise processes lasted for a period,the capillary rise height increased and the rise velocity be-came much smaller,which showed very little effect for the slope of the linear?tting of Eq.(10).For all the test samples in the present study,eliminating the capillary rise height and velocity results of the last10s just resulted in a maximal deviation of0.29%for D P capáK.Therefore,for the present experiments with a period of two minutes,the value of capillary performance parameter is acceptable to be true,without needing the effort to conduct the capillary rate-of-rise experiments for much longer time,such as several days[43].The time scales of capillary rise in this study were comparable to the previous work by Holley and Faghri[27],

74Y.Tang et al./Energy Conversion and Management66(2013)66–76

in which the effective pore radius and permeability of porous wick were also successfully obtained.Furthermore,the present ap-proach eliminates the dif?culties to prepare the very long samples to reach the equilibrium height of capillary rise.Hence,using the approaches as described before,the capillary rate-of-rise test pro-vides an easy way to determine the capillary performance param-eter of wicks.

5.Conclusions

In the present study,the composite wicks combined with micro V-grooves and sintered powder were developed.A ploughing–extrusion method was presented to fabricate the micro V-grooves as the base of composite wicks.The capillary performance,which integrates both capillary pressure and permeability,was character-ized by capillary rate-of-rise experiments using an IR thermal im-age method.The effects of groove geometries and sintering temperature and time on the capillary performance of composite wicks were assessed for the aim of design optimization.The main conclusions can be summarized as follows:

(1)Micro V-grooves with a depth-width aspect ratio of about2

was fabricated by the ploughing–extrusion fabrication

method.This provides a material-saving economic method for the fabrication of microgrooves.

(2)Both P–E depth and groove pitches were shown to have a

nonmonotonic effect on the capillary performance of com-posite wicks.There exists an optimum geometry with the groove depth of0.85mm and groove width of0.45mm.

The apex of microgrooves is believed to play a dominant role on the capillary pressure and?ow resistance.

(3)The sintering temperature and time exhibit a less evident

role on the capillary performance of composite wicks.The optimal sintering parameters were suggested to950°C along with30min.

(4)Using acetone as working liquid,it produces a much larger

capillary rise height and velocity than that of ethanol.How-ever,the capillary performance parameters obtained by these two working liquids presents fairly good agreement.

The IR thermal imaging method provides an accurate means to characterize the hydraulic properties of composite wick.

Further researches will be focused on the thermal characteriza-tions,such as the thermal conductivity and thermal resistance in two-phase heat transfer devices.The heat transfer performance of TPHTD with integration of this kind of wicks can be also explored.

Acknowledgments

The research was?nancially supported under the grants of the National Nature Science Foundation of China-Guangdong Joint Funds,Project No.U0834002,National Nature Science Foundation of China(Nos.51275180,51075155)and the PhD Programs Foun-dation of Ministry of Education of China(No.20100172120001). The authors also would like to acknowledge the Joint-training PhD Program(No.2011615063)sponsored by the China Scholar-ship Council and the Doctorate Dissertation Innovation Funds sup-ported by South China University of Technology.The authors also wish to thank Qiaoxia Yuan,Jianbo Qing and Xinrui Ding from our research group for their help with experiment.Finally,the helpful suggestions by Professor Cila Herman in the Johns Hopkins University are gratefully appreciated.

References

[1]Chang YW,Cheng CH,Wang JC,Chen SL.Heat pipe for cooling of electronic

equipment.Energy Convers Manage2008;49:3398–404.

[2]Hsieh SS,Lee RY,Shyu JC,Chen SW.Thermal performance of?at vapor

chamber heat spreader.Energy Convers Manage2008;49:1774–84.

[3]Faghri A.Heat Pipe Science and Technology.Washington DC:Taylor&Francis;

1995.

[4]Canti G,Celata GP,Cumo M,Furrer M.Thermal hydraulic characterization of

stainless steel wicks for heat pipe applications.Rev Gen Therm1998;37:5–16.

[5]Franchi G,Huang X.Development of composite wicks for heat pipe

performance enhancement.Heat Transf Eng2008;29:873–84.

[6]Hsieh JC,Huang HJ,Shen SC.Experimental study of microrectangular groove

structure covered with multi mesh layers on performance of?at plate heat pipe for LED lighting module.Microelectron Reliab2012;52:1071–9.

[7]Oshman C,Li Q,Liew L,Yang R,Lee YC,Bright VM,et al.Thermal performance

of a?at polymer heat pipe heat spreader under high acceleration.J Micromech Microeng2012;22:045018.

[8]Grote MG,Stark JA,Tefft EC.Enhanced evaporative surface for two-phase

mounting plates.In:Proceedings of the16th intersociety conference on environmental systems,San Diego,CA;1986.p.617–26.

[9]Khrustalev DK,Grakovich LP,Denisevich SV,Yanitski B,Sheleg VK.Heat pipe

with a combined capillary-porous structure.J Eng Phys Thermophys 1987;52:587–92.

[10]Khrustalev D,Faghri A.Enhanced?at miniature axially grooved heat pipe.J

Heat Transf1996;118:261–4.

[11]Wang JL,Catton I.Enhanced evaporation heat transfer in triangular grooves

covered with a thin?ne porous layer.Appl Therm Eng2001;21:1721–37. [12]Tang Y,Deng D,Lu L,Pan M,Wang Q.Experimental investigation on capillary

force of composite wick structure by IR thermal imaging camera.Exp Therm Fluid Sci2010;34:190–6.

Table2

Comparison of the capillary performance parameter determined by ethanol and acetone test.

Wick sample K/r eff(l m),

ethanol test

K/r eff(l m),

acetone test

Relative deviation of

K/r eff(%)

D10.7610.7017.8

D20.7740.7118.1

D30.8180.767 6.3

D4a0.8260.778 5.7

D50.7870.7317.1

P10.7980.750 6.0

P20.8050.761 5.5

P3a0.8260.778 5.7

P40.7830.7267.3

P50.7730.7098.3

P60.7640.7038.0

T10.7910.740 6.5

T20.8090.759 6.1

T3a0.8260.778 5.7

T40.8020.748 6.7

t10.8440.797 5.6

t2a0.8260.778 5.7

t30.8040.754 6.3

a The same sample.

Y.Tang et al./Energy Conversion and Management66(2013)66–7675

[13]Hopkins R,Faghri A,Khrustalev D.Flat miniature heat pipes with micro

capillary grooves.J Heat Transf1999;121:103–9.

[14]Benson DA,Mitchell RT,Tuck MR,Palmer DW,Peterson GP.Ultrahigh-capacity

micromachined heat spreaders.Micro Thermophys Eng1998;2:21–30. [15]Cao Y,Gao M,Beam JE,Donovan B.Experiments and analyses of?at miniature

heat pipes.J Thermophys Heat Transf1997;11:158–64.

[16]Kang SW,Huang D.Fabrication of star grooves and rhombus grooves micro

heat pipe.J Micromech Microeng2002;12:525–31.

[17]Lim HT,Kim SH,Im HD,Oh KH,Jeong SH.Fabrication and evaluation of a

copper?at micro heat pipe working under adverse-gravity orientation.J Micromech Microeng2008;18:105013.

[18]Liang TS,Hung YM.Experimental investigation on the thermal performance

and optimization of heat sink with U-shape heat pipes.Energy Convers Manage2010;51:2109–16.

[19]Elnaggar MHA,Abdullah MZ,Mujeebu MA.Characterization of working?uid

in vertically mounted?nned U-shape twin heat pipe for electronic cooling.

Energy Convers Manage2012;62:31–9.

[20]Li C,Peterson GP,Wang Y.Evaporation boiling in thin capillary wicks(I)–wick

thickness effects.J Heat Transf2006;128:1312–9.

[21]Ma HB,Peterson GP.Experimental investigation of the maximum heat

transport in triangular grooves.J Heat Transf1996;118:740–6.

[22]Suman B,Hoda N.Effect of variations in thermophysical properties and design

parameters on the performance of a V-shaped micro grooved heat pipe.Int J Heat Mass Transf2005;48:2090–101.

[23]Hanlon MA,Ma HB.Evaporation heat transfer in sintered porous media.J Heat

Transf2003;125:644–52.

[24]Williams RR,Harris DK.Cross-plane and in-plane porous properties

measurements of thin metal felts applications in heat pipes.Exp Therm Fluid Sci2003;27:227–35.

[25]Adkins DR,Dykhuizen RC.Procedures for measuring the properties of heat-

pipe wick materials.In:Proceedings of the28th intersociety energy conversion engineering conference,vol.2.SAE.p.911–7.

[26]Freggens RA.Experimental determination of wick properties for heat pipe

applications.In:Proceedings of the4th Intersociety energy conference engineering conference,Washington DC,22–26September;1969.

p.888–97.

[27]Holley B,Faghri A.Permeability and effective pore radius measurements for

heat pipe and fuel cell applications.Appl Therm Eng2006;26:448–62.[28]Byon C,Kim SJ.The effect of meniscus on the permeability of micro-post

arrays.J Micromech Microeng2011;21:115011.

[29]Byon C,Kim SJ.Capillary performance of bi-porous sintered metal wicks.Int J

Heat Mass Transf2012;55:4096–103.

[30]Fries N,Odic K,Conrath M,Dreyer M.The effect of evaporation on the wicking

of liquids into a metallic weave.J Colloid Interface Sci2008;321:118–29. [31]Deng D,Tang Y,Huang G,Lu L,Yuan D.Characterization of capillary

performance of composite wicks for two-phase heat transfer devices.Int J Heat Mass Transf2013;56:283–93.

[32]Wang X,Tang Y,Chen P.Investigation into performance of a heat pipe with

micro grooves fabricated by extrusion–ploughing process.Energy Convers Manage2009;50:1384–8.

[33]Tang Y,Chen P,Wang X.Experimental investigation into the performance of

heat pipe with micro grooves fabricated by extrusion–ploughing process.

Energy Convers Manage2010;51:1849–54.

[34]Washburn EW.The dynamics of capillary?ow.Phys Rev1921;17:273–83.

[35]Lago M,Araujo M.Capillary rise in porous media.J Colloid Interface Sci

2001;234:35–43.

[36]Taylor JR.An introduction to error analysis.2nd ed.University Science Books;

1997.

[37]Linstrom P,Mallard W,editors.NIST Chemistry WebBook.NIST Standard

Reference Database Number69.Gaithersburg(MD):National Institute of Standards and Technology;2009.

[38]Ma HB,Peterson GP.The minimum meniscus radius and capillary heat

transport limit in micro heat pipes.J Heat Transf1998;120:227–33.

[39]Ghajar M,Darabi J,Crews Jr N.A hybrid CFD-mathematical model for

simulation of a MEMS loop heat pipe for electronics cooling applications.J Micromech Microeng2005;15:313–21.

[40]Xu X,Carey VP.Film evaporation from a micro-grooved surface an

approximate heat transfer model and its comparison with experimental data.AIAA J Thermophys Heat Transf1990;4:512–20.

[41]Leong KC,Liu CY,Lu GQ.Characterization of sintered copper wicks used in heat

pipes.J Porous Mater1997;4:303–8.

[42]Vollmer M,M?llmann KP.Infrared thermal imaging:fundamentals,research

and applications.Weinheim(German):Wiley-VCH;2010.

[43]Siebold A,Walliserb A,Nardina M,Oppliger M,Schultz J.Capillary rise for

thermodynamic characterization of solid particle surface.J Colloid Interface Sci 1997;186:60–70.

76Y.Tang et al./Energy Conversion and Management66(2013)66–76