Flow boiling enhancement of structured microchannels with micro pin fins

Flow boiling enhancement of structured microchannels with micro pin fins

International Journal of Heat and Mass Transfer 105 (2017) 338–349 Contents lists available at ScienceDirect International Journal of Heat and Mass ...

5MB Sizes 11 Downloads 52 Views

International Journal of Heat and Mass Transfer 105 (2017) 338–349

Contents lists available at ScienceDirect

International Journal of Heat and Mass Transfer journal homepage: www.elsevier.com/locate/ijhmt

Flow boiling enhancement of structured microchannels with micro pin fins Daxiang Deng a,⇑, Wei Wan a, Yu Qin a, Jingrui Zhang b, Xuyang Chu a a b

Department of Mechanical & Electrical Engineering, School of Aerospace Engineering, Xiamen University, Xiamen 361005, China Department of Instrumental & Electrical Engineering, School of Aerospace Engineering, Xiamen University, Xiamen 361005, China

a r t i c l e

i n f o

Article history: Received 10 July 2016 Received in revised form 21 September 2016 Accepted 27 September 2016

Keywords: Microchannel heat sinks Flow boiling Structured microchannels Micro pin fins Heat transfer enhancement

a b s t r a c t Structured microchannels with micro pin fins (SM-MPF) were developed for the utilizations in advanced microchannel heat sinks to cool high heat-flux devices. These micro cone pin fins were fabricated on the bottom surface of rectangular microchannels by a laser micromilling method. Flow boiling performance of these structured microchannels with micro pin fins was experimentally explored together with comprehensive comparisons with conventional rectangular microchannels with smooth bottom walls surfaces. Flow boiling tests with two coolants, deionized water and ethanol, were performed at inlet subcoolings of 40 °C and 10 °C, mass fluxes of 200–300 kg/m2s and effective heat fluxes up to 1067 kW/m2. Experimental results demonstrated that the structured microchannels presented significant flow boiling heat transfer enhancement, i.e., an enhancement of 10–104% in water tests, and 90–175% in ethanol cases in general, compared to the conventional rectangular microchannels. Moreover, the critical heat fluxes were promoted notably, and the severe two-phase flow instabilities were also mitigated for the structured microchannels in the low inlet temperature cases. The structured microchannels provided lots of stable nucleation sites by producing lots of tiny reentrant cavities, and introduced significant wicking effects to maintain the liquid rewetting and hinder the local dry-out, which contributed to the above notable flow boiling enhancement. Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction Microchannel heat sinks incorporating two-phase flow boiling have recently emerged as a highly efficient cooling technique to address the current and future high heat flux dissipation issues in many cutting-edge areas, such as microelectronic devices, high power laser diode arrays, chemical reactors and so on [1,2]. They offer several distinct benefits over other cooling methods due to their high heat transfer capabilities via latent heat exchange, small overall heat sink size, small rate of coolant flow and uniform wall temperatures distributions [3]. Since the concept of microchannel heat sinks has been proposed by Tuckerman and Pease [4], a large amount of research efforts have been undertaken to better understand the flow boiling phenomena in microchannels [5]. Most of researches have been focused on conventional parallel microchannels, such as rectangular [6,7], triangular [8] or trapezoidal [9] ones. Recent advancement in the manufacturing techniques, however, allows microchannels to be fabricated into more advanced

⇑ Corresponding author. E-mail address: [email protected] (D. Deng). http://dx.doi.org/10.1016/j.ijheatmasstransfer.2016.09.086 0017-9310/Ó 2016 Elsevier Ltd. All rights reserved.

structures, which makes them possible to enhance heat transfer, promote critical heat fluxes (CHF), reduce pressure drop or mitigate two-phase flow instabilities [10]. Among the reported enhancement methods, one way is to modify the flow passages. Cross-linked microchannels [11], tapered microchannels [12], diverging microchannels [13], expanding microchannels [14], and X-shaped reentrant microchannels [15] have been developed and fabricated. Comparative experimental and numerical studies have demonstrated their superiorities over conventional parallel microchannels. Another direction is to produce structured surfaces on the microchannel bottom wall or sidewall. During the two-phase boiling flow in microchannels, the boiling surface played a vital role on the bubble behaviors, i.e., bubble nucleation, detachment and movement, which is critical for the flow boiling heat transfer, pressure drop and flow instabilities. Therefore, various methods have been developed to alter the boiling surface, such as fabricated artificial micro cavities [16,17], addition of porous layers [18–20] and nano coatings [21,22], have been proposed to promote the flow boiling performance. Kandlikar et al. [16] fabricated micro cavities on the bottom of rectangular copper microchannels via a laser drilling method to produce artificial nucleation sites. This means,

D. Deng et al. / International Journal of Heat and Mass Transfer 105 (2017) 338–349

339

Nomenclature Ach At Dh G hfg htp kCu km ks L Li lCu lhs

m _ m N ONB Pin qeff q q00 eff

total heat transfer area of microchannels, m2 footprint area of the copper block top platform, m2 hydraulic diameter, mm mass flux, kg/m2s latent heat of vaporization, kJ/kg local two-phase heat transfer coefficient, kW/m2K thermal conductivity of copper block, W/mK thermal conductivity of microchannels, W/mK thermal conductivity of solder, W/mK length of heat sink, mm distance from the inlet to thermocouple location in the stream-wise direction, m distance between the thermocouple and the top surface of copper block, m distance between backside surface of microchannels sample and the bottom wall surface of each microchannel, m fin parameter mass flow rate, kg/s number of reentrant microchannels onset of nucleation boiling inlet pressure, kPa effective heat power, W total power input, W effective heat flux based on platform area, kW/m2

together with the installation of inlet restrictor, was found to mitigate the severe two-phase flow instabilities significantly. Utilizing deep reactive ion etching (DRIE) methods, Kuo and Peles [17] processed reentrant micro cavities on the sidewall surface of rectangular silicon microchannels. They experimentally demonstrated that such microchannels presented significant reduction in the wall superheat of the onset of nucleation boiling (ONB), and mitigated the rapid bubble growth instabilities compared to plain-wall microchannels. Ammerman and You [18] assessed that a small rectangular channel with sprayed painting porous coatings reduced the wall superheat for ONB, and promoted the heat transfer performance and critical heat fluxes (CHF) compared to the uncoated ones. Subsequent similar approaches, such as rectangular channels covered with wire mesh by Wang and Peterson [19], and sintered porous copper powder layer by Sun et al. [20], also demonstrated the merits of porous coating for the promotion of bubble nucleation. Besides, the nanocoatings, such as the electrochemical deposited carbon nanotubes on copper channel surfaces by Khanikar et al. [21], and Si nanowires on the three sides of rectangular silicon microchannels by Yang et al. [22], also contribute to the enhancement in boiling heat transfer and CHF. Besides of the above means, micro pin fins is another good choice to promote the flow boiling. As a matter of fact, pin fins have been successfully used in compact heat exchanger passages due to the pronounced merits in enlarging the heat transfer area, increasing the nucleation sites and flow intersection. The above merits of pin fins have been also explored in micro-scale domain. Krishnamurthy and Peles [23] fabricated micro circular silicon pin fins of a diameter of 100 lm, spacing of 150 lm and height of 250 lm in a rectangular channel using MEMS procedures. In their latter studies [24], a single row of inline micro circular pin fins was entrenched in a microchannel. The flow boiling results of coolant HFE7000 demonstrated heat transfer enhancements compared to the plain microchannel. The pressure drop characteristics were not reported but it is expected to be very large due to pin fin obstructions. Also using MEMS methods, Kosar and Peles [25] fab-

Ttci Tin Tw,tci

DTsct,tci Tsat, tci W Wfin x

thermocouple reading (i = 1–5), °C inlet fluid temperature, °C channel bottom wall temperature at thermocouple location, °C wall superheat, °C local saturation temperature of thermocouple location, °C width of heat sink, mm width of fin between two rectangular microchannels, mm thermodynamic quality

Greek symbols g fin efficiency q density of fluid, kg/m3 u heat transfer ratio Subscripts Cu copper hs heat sink fin fin tci thermocouple location in inlet sat saturation

ricated micro pin fin heat sinks with arrays of hydrofoil pin-fins with chord thickness of 100 lm and fin height of 243 lm. Qu and Siu-Ho [26] studied the saturated flow boiling heat transfer of water in an array of staggered square micro pin fins with a cross-section of 200  200 lm2 and a height of 670 lm. They found that the heat transfer coefficient was enhanced by inlet subcooling in the low quality region, but kept fairly constant in the high quality region. Ma et al. [27] reached considerable flow boiling heat transfer enhancement in a suqre micro-pin-finned chip surface compared to a smooth one in FC-72. Chang et al. [28] explored the saturated and subcooling flow boiling heat transfer and associated bubble characteristics of FC-72 over a micro pin fin surface flush mounted on the bottom of a horizontal rectangular channel. Law and Lee [29] conducted a comparative study on the flow boiling performance of copper oblique-finned microchannels and straight parallel microchannels. Significant augmentation in heat transfer and delay in the onset of critical heat flux have been reached for the oblique-finned microchannels. Unfortunately, it accompanied with higher pressure drop penalty. Recently, Woodcock et al. [30] developed a 2.4 mm wide silicon structured microchannel with an array of 150 lm piranha pin fin entrenched inside. Single-phase and two-phase heat transfer performances and pressure drop characteristics were evaluated in both open flow and extraction flow configurations. The extraction flow configuration performed better in heat transfer, but induced larger pressure drop than the open one. From the above literature review, one can note that structured surfaces and micro pin fins were two promising means for flow boiling enhancement. To combine both merits of them, we in this study proposed a novel type of structured microchannels with micro pin fins on the bottom surface. Laser micromilling method was utilized to process micro cone pin fins on the bottom surface of rectangular microchannels. Flow boiling performance of these structured microchannels with micro cone pin fins were comprehensively explored and compared with rectangular microchannels with smooth bottom walls.

340

D. Deng et al. / International Journal of Heat and Mass Transfer 105 (2017) 338–349

2. Fabrication of structured microchannels Both rectangular microchannels with and without micro pin fins on the bottom surfaces were fabricated in pure oxygen-free cooper (99.9% Cu) plates. The rectangular microchannels were firstly processed by the micro wire electrical discharge machining (EDM) method. The obtained rectangular microchannels with smooth bottom surface were shown in Fig. 1. The obtained microchannels samples measured to be 45 mm long, 20 mm wide and 2 mm thick. They consisted of 14 parallel rectangular microchannels, and their geometric parameters are given in Table 1. Then one of the rectangular microchannels samples were further installed in a prototype pulsed fiber laser machine (IPG, No: YLP-1-100-30-30-HC-RG, Russia) to process the micro pin fins on the bottom surfaces by a laser micromilling method. The cross machining route and loop multiple-pass reciprocating scanning strategy are selected to fabricate the micro pin fins, and the line spacing is set to be 5 lm. The material removal is achieved by layer-by-layer milling process around the pin fins. The laser was set to produce 100 ns pulses with a 1064 nm fundamental wavelength (k) at a repetition rate of 20 kHz. The micro pin fins were fabricated at the laser output power of 27 W, scanning speed of 500 mm/s and scanning times of 10. Details of the laser micromilling process of micro pin fins can be available in our previous study [31]. Finally, the micro cone pin fins were formed on the bottom surface, as shown in Fig. 2. It was of five stagger arrays of cone pin fins, which provide numerous tiny pores for bubble formations. After all the fabrication procedures, both microchannels samples were thoroughly cleaned with kerosene oil in an ultrasonic bath for approximately one hour, and then ultrasonically cleaned with deionized water for half an hour to remove any possible contaminants.

3. Experiments 3.1. Experimental setup Fig. 3 shows the schematic diagram and photograph of the flow loop. The test liquid is stored in stainless steel reservoir, in which there is a coil heater to boil the liquid vigorously for degassing. The liquid is pumped through a 7 lm filter by a magnetically coupled gear pump (Micropump, GA-T23-DB-380B). The flow rate of the liquid is monitored by a micro turbine flowmeter (Omega, FTB-311D). The liquid is subsequently pre-heated before being fed into the test section, which is implemented by flowing through a copper coil immersed inside a constant temperature water bath. Tubes from the exit of the coil to the inlet of the test section are

(a)

insulated. The designed inlet subcoolings are thus maintained by installing a thermocouple before the microchannel sample in the test section. After that, the liquid enters the test section and heat exchange occurs. The hot liquid is cooled to the ambient temperature by a coiled condenser before it returns back to the reservoir to finish the loop. The test section is heated by a copper block which incorporates ten cartridge heaters inside, supplying a maximum total power of 1000 W. The cartridge heaters are connected to a digital power meter and powered by a variac. The test section, as illustrated in Fig. 4, are similar to those utilized in our previous studies [32–34], and is described briefly for completeness. The heating copper block with cartridge heaters inside consists of an upper rectangular section of 20  45 mm, which is identical to the projection area of the microchannel sample. The heating block is embedded in a polyetheretherketone (PEEK) flow housing, and the interface of the copper block and flow housing is filled by RTV silicon rubber to prevent any leakage of liquid. The bottom part of copper block is hosted by a PEEK insulation block. The test microchannels sample is soldered on the top of the copper block using a thin layer of solder (Pb-Sn-Ag-Sb, thermal conductivity (ks) of 50 W/mK given by the supplier) to minimize the contact thermal resistance. As the height difference between the top surface of copper block and the flow housing is designed to be 0.5 mm to form a gap to host the solder, the solder layer is ensured to be 0.1 mm thick after the soldering process is completed. A Pyrex 7740 glass cover plate is assembled above the microchannels sample, above which there is a top plate to assemble the glass cover. Good contact between the top surfaces of microchannels sample and the glass cover is maintained by sealing with an O-ring when all the top plate, glass cover, flow housing and insulating block are clamped together by screws. Any bypass flow is prevented for the microchannels. Details of the above constructions were shown in the previous study [32] with an exploded view. The flow passages are constructed by the connections of the flow housing and two inlet and outlet plenums. Horizontal inlet and outlet manifold arrangements are utilized. None deep inlet/outlet reservoirs are adopted to reduce the compressive upstream volume. Two-phase flow instabilities are expected to be reduced by the above approaches together with the installation of stiff stainless steel tube connecting the test section and the upstream flow loop [9]. Rubber plates are set before the entrance and after the exits of microchannels to prevent liquid leakage. The inlet and outlet pressure are measured by two pressure transducer (WH-PTX7517, Shenzhen Oriental Vanward Instrument Co.) with a response time of 3 ms. The inlet and outlet temperatures are monitored by two type-K thermocouples, which are set at about 5 mm before and after the microchannels sample. The stream-wise wall temperature distributions are measured using

(b)

Fig. 1. SEM images of the rectangular microchannels with smooth bottom surface (a) top view; (b) cross-section view.

341

D. Deng et al. / International Journal of Heat and Mass Transfer 105 (2017) 338–349 Table 1 Specification of parameters of two microchannels. Sample

Material

Sample dimension, W  H  L(mm)

Width of each microchannel, Wch (lm)

Depth of each microchannel, Hch (lm)

Hydraulic diameter, Dh (lm)

Spacing of microchannel, s, (mm)

Number of microchannels, N

Rectangular microchannels (RM) Structured microchannels with micro pin fins (SMMPF)

Oxygen-free pure copper Oxygen-free pure copper

20  2  45

640

1000

776

1.2

14

20  2  45

640

1000

750

1.2

14

(a)

(b) Fig. 2. SEM images of the structured microchannels with micro cone pin fins on the bottom surface (a) top view; (b) cross-section view.

five type-K shielded thermocouples with a diameter of 1 mm. All the thermocouples are of a response time of 50 ms. These shielded thermocouples are set at a distance of 2 mm below the top surface of copper block, and are distributed at the interval of 10 mm as detailed in Fig. 4. All the temperatures, pressures and flow rates are collected by an Agilent 34970A data acquisition system. The flow behaviors during the two-phase boiling process are visualized by a microscope (XTL-850P Shanghai Guangmi Instrument Co.,Ltd.) together with a high speed camera (Fastec HiSpec DVR 2F).

increased in a small increment of 15–30 W. When the steady state is reached, all temperature and pressure data are recorded at 1 s interval for 2 min. Heat transfer coefficients and pressure drop are then determined using the averaged values from the measured data for 2 min. The experiments stops when the critical heat flux is reached or the maximum value of wall temperatures reached as high as 160 °C to avoid the solder melting.

3.2. Experimental procedure

In order to calculate the heat transfer coefficients of microchannel heat sinks, the effective heat absorption (qeff) should be determined firstly. During the two-phase boiling regions, a common method of the heat transfer ratio (u) [8,9,26,35] based on the percentages of the sensible heat gain by fluid against the total heat power in the single-phase convection regions were utilized to determine the qeff. Prior to boiling at each test cases, a series of single phase heat transfer tests are conducted to obtain the effective heat power applied to the heated surface during the two-phase flow boiling. The sensible heat gain by single-phase fluid is determined as follows,

Flow boiling tests are conducted for both microchannels samples in two liquid, i.e., deionized water and ethanol (A.R., 99.5%). Experiments are done at different inlet subccolings (DTsub = 10, 40 °C) and mass flow rate (G = 200–300 kg/m2s). Prior to each test, the test liquid is fully degassed via vigorous boiling for about half an hour. The deaeration method is believed to be effective as no bubbles are observed in the microchannels before the ONB [6,26].The vent value at the top of the reservoir is periodically open to allow release the dissolved gas and to maintain the test pressure to near atmospheric. In each test, the supplied flow rate and inlet temperature are maintained to be constant. The heat power is

3.3. Data reduction

_ p ðT out  T in Þ qeff ¼ mc

ð1Þ

342

D. Deng et al. / International Journal of Heat and Mass Transfer 105 (2017) 338–349

(a)

(b) Fig. 3. The flow boiling loop:(a) schematic diagram; (b) photograph.

where cp denotes the specific heat of fluid, T in and T out denote the _ denotes the inlet and outlet liquid temperature, respectively. m _ ¼ V_ q, in which V_ is mass flow rate, which can be obtained by m

the volumetric flow rate measure by the flow meter, q is the liquid density. Then the heat transfer ratio (u) can be obtained by

_ p ðT out  T in Þ=ðVIÞ u ¼ mc

ð2Þ

where q = V  I, V and I are the input voltage and current readings from the digital power meter, respectively. The single-phase tests showed that u ranged from 0.8 to 0.9 depending on the certain test conditions. The mean values of heat transfer ratio (u) were then utilized to calculate the effective heat-

ing power in the flow boiling experiments, i.e., qeff ¼ uq. Such approaches have been adopted by many previous works [8,9,26,35]. Then the effective heat flux (q00 eff) can be obtained by

q00eff ¼ qeff =At

ð3Þ

where At is the top platform area of the copper block, At ¼ W  L, in which W and L are the width and length of microchannels sample. As direct wall temperature measurements at the bottom surface of the microchannels are not available, the wall temperatures of microchannels (Tw, tci, i = 1–5) is calculated by the extrapolations of the temperature readings at the streamwise direction (Ttci, i = 1–5) using one-dimensional heat conduction assumption.

D. Deng et al. / International Journal of Heat and Mass Transfer 105 (2017) 338–349

343

Fig. 4. Schematic of the cross-section of the test section.

 T w;tci ¼ T tci  qeff

lCu lhs ls þ þ kCu At km At ks At

 ð4Þ

where T tci is the local thermocouple reading (i = 1–5). The wall temperature of microchannels (T w;tci ) is deduced from a thermal resistance network detailed in the previous work[32], where kCu, ks, km are the thermal conductivities of copper, solder and the microchannel base, respectively. lCu, ts and lhs represents the distance from the thermocouple location to the top heating surface of copper block, thickness of solder, and the distance between the backside surface of microchannels sample and the bottom wall surface of each microchannel, respectively. The local heat transfer coefficient can be then obtained by

htp;tci ¼

qeff DT sat;tci Ach

ð5Þ

where DT sat;tci is the local wall superheat of the thermocouple location (Ztci, i = 1–5), which can be defined as DT sat;tci ¼ T w;tci  T sat;tci , in which T sat;tci is the liquid saturation temperature. Ach is the total heat transfer area of microchannels, which is given by the fin analysis method as follows,

perature measurements by the type-K thermocouples yield an uncertainty of ±0.3 °C. The measurement errors for the flow meter and pressure transducer are 1% and 0.1% of full scale, respectively. The supply power measured by the digital power meter yields an uncertainty of 1%. The wall temperature uncertainty comes from the thermocouple errors and from the correction from the temperature drop from the copper block as in Eq. (5). Using the standard error analysis method [36], the maximum uncertainties in the vapor quality and two-phase heat transfer coefficient for rectangular microchannels can be estimated to be 9.3% and 9.2%, respectively. For the structured microchannels with micro pin fins, the maximum uncertainties in the vapor quality and two-phase heat transfer coefficient can be estimated to be 10.1% and 10.3%, respectively. 4. Results and discussion 4.1. Single-phase heat transfer validation

where Wfin is the width of the fin between two adjacent rectangular microchannels. Eqs. (5)–(8) are iteratively solved to obtain htp. The vapor quality can be obtained as follows,

In order to access the capabilities of the experimental setup and data reduction methods, single-phase convective flow tests were conducted for the conventional rectangular microchannels using water. Two inlet subcooling cases, i.e., DTsub = 67 °C and 40 °C, are performed at certain heat fluxes and different flow rates. The measured results of single-phase heat transfer are then compared to the common correlations of Shan and London [37]. Fig. 5 illustrates the comparison results of average Nusselt numbers against Reynolds number. The convective flow in the experiments is related to the hydraulically developed but thermally developing flow under uniform heat flux boundary conditions. The increasing trend of measured Nuave with increasing Re results accords with the prediction of Shan and London [37] correlation. The deviations between the measurement results and the theoretical correlations are all within 9.5%, indicating that fairly good agreement is achieved. It suggests that the measurement system and data reduction method are adequate for high heat transfer coefficients measurement in boiling flow.



4.2. Boiling curves

Ach ¼ NLðW ch þ 2gHch Þ

ð6Þ

where Wch and Hch are the width and depth of each microchannel, respectively. N is the total number of microchannels in a sample, g is the fin efficiency, which can be calculated as follows assuming the adiabatic fin tip condition,



tanhðmHch Þ mHch

ð7Þ

in which m is the fin parameter as given by

sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi htp;tci  2ðL þ W fin Þ m¼ km LW fin

  1 qeff Li   C p ðT sat;tci  T in Þ _ L hfg m

ð8Þ

ð9Þ

where hfg is the latent heat of vaporization, Li is the distance from the inlet to the thermocouple location. The dimensional errors of the microchannels and the solder layer thickness are estimated to be ±0.01 mm. The individual tem-

Fig. 6 shows the boiling curves of both rectangular microchannels and structured microchannels with micro pin fins in two coolants tests. These plots show the variations of the wall superheat at the most downstream thermocouple location are versus the

344

D. Deng et al. / International Journal of Heat and Mass Transfer 105 (2017) 338–349 14

RM, water 2

12

ƸTsub=67ć, q''eff =267kW/m Correlation of Shah&London [37]

ƸTsub=40ć, q''eff =200kW/m2 Correlation of Shah&London [37]

10

Nuave

8

6

4

2

0 100

200

300

400

500

600

700

800

900

1000 1100 1200

Re Fig. 5. Comparison of experimental data of rectangular microchannels with theoretical correlation of Shah and London [37] in single phase heat transfer.

single-phase heat transfer was obtained for the SM-MPF. The promotion of convective heat transfer was more pronounced in theDTsub = 40 °C case of ethanol tests, which may be due to that the micro pin fins inside the microchannels induced intense flow mixing and disruption of boundary layer in the ethanol liquids with much smaller density and surface tension than water. After an initial single-phase region, all the boiling curves shift abruptly with a small increase in the supplied heat fluxes, which indicates that the onset of nucleation boiling (ONB) occurred. It can be noted that both microchannels initiated the ONB at more or less the same wall superheat. It seems that the structured microchannels with micro pin fins did not play a significant role on the occurrence of ONB. Nevertheless, beyond the point of ONB, the slopes of boiling curves of SM-MPF are much steeper than the smooth rectangular microchannels. It suggests that the SM-MPF sample presented a pronounced boiling heat transfer enhancement over the latter one, which is true for all the tests cases with different mass fluxes and inlet subcooling in two coolants experiments. With further increase in the heat fluxes, the boiling curves of SM-MPF ascended more or less steadily to high heat fluxes, i.e., over 1000 kW/m2 in water tests and 650–900 kW/m2 in ethanol tests. On the contrary, the rectangular microchannels showed an abrupt flattening of boiling curves at 430–480 kW/m2 in water tests and 310–390 kW/m2 in ethanol tests, indicative of the incipience of critical heat flux (CHF). The structured microchannels with micro pin fins delayed the dryout condition and promoted the critical heat flux significantly, which is especially promising for the high heat fluxes dissipations applications. 4.3. Flow boiling heat transfer enhancement

Fig. 6. Comparison of boiling curves of both microchannels at (a) water tests; (b) ethanol tests.

effective heat fluxes for both samples. The initial data was linked to the single-phase convective flow. It can be noted that the SM-MPF sample generally sustained slight larger heat fluxes at the same wall temperatures, indicating that a slight enhancement of

Figs. 7and 8 shows the local two-phase heat transfer coefficients of both microchannels in water and ethanol tests, respectively. The htp of the most downstream thermocouple location near the outlet of microchannel are plotted against the effective heat fluxes and vapor qualities, which relates to the greatest amount of saturated boiling and the largest vapor quality. At this initial stage of boiling process after the ONB, large values of htp can be noted, as the latent heat of coolants was released fast. Nevertheless, with a small increase in the effective heat fluxes and associated vapor qualities, the boiling heat transfer performance of both microchannels generally presented a rapid decrease, which is true for both coolants tests. The heat transfer coefficients were closely dependent on the heat fluxes at small heat fluxes regions, which is an indication of a nucleate boiling dominated mechanism. The high-speed visualization images of both samples in Fig. 9, taken near the outlet section of the microchannels, confirm the above observation. The early decrease in htp of both microchannels with and without micro pin fins accorded with previous water boiling studies of conventional parallel microchannels [7–9], and also accorded with the water results of micro pin fins heat sinks in Qu and Siu-Ho [26], and FC-72 results of oblique pin fins in Law et al. [38]. The above sharp drop in htp at the early stage of boiling is believed to be linked to the suppression of bubble nucleation due to the confinement effects of microchannels, as well as the coalescence of adjacent bubbles and formation of elongated bubbles. This is argued to be the unique phenomena of microchannels by Steinke and Kandlikar [7]. At this stage, it can be noted that the structured microchannels with micro pin fins showed enhanced heat transfer performance, which is especially notable in the large subcooling cases (DTsub = 40 °C) of water tests and all the ethanol tests. For example, the SM-MPF sample presented a 26%–61% enhancement in the water tests of DTsub = 40 °C and G = 200kg/m2s, and the enhancement reached as high as 54– 142% in ethanol tests compared to the smooth microchannels at the small heat fluxes region. As shown in Fig. 2, the addition of

345

D. Deng et al. / International Journal of Heat and Mass Transfer 105 (2017) 338–349

35

35

Water 2 SM-MPF, ƸTsub=40ć,G=200kg/m s

30

2

SM-MPF, ƸTsub=10ć,G=200kg/m s

30

Water 2 SM-MPF, ƸTsub=40ć,G=200kg/m s

25

SM-MPF, ƸTsub=10ć,G=300kg/m s

2

SM-MPF, ƸTsub=10ć,G=200kg/m s

2

SM-MPF,ƸTsub=10ć,G=300kg/m s

2

20

20

15

15

10

Water 2 RM, ƸTsub=40ć,G=200kg/m s

5

RM, ƸTsub=10ć,G=200kg/m s

Water 2 RM, ƸTsub=40ć,G=200kg/m s

10

2 2

RM, ƸTsub=10ć,G=300kg/m s 0

htp kW/m2ć

htp kW/m2ć

25

0

100

200

300

400

500

600

2

700

800

900

1000 1100

q''eff kW/m

(a)

2

RM, ƸTsub=10ć,G=200kg/m s

5

2

RM, ƸTsub=10ć,G=300kg/m s 0 -0.05

0.00

0.05

0.10

xtc5

0.15

0.20

0.25

(b)

Fig. 7. Two-phase heat transfer coefficient of water tests for two microchannels as a function of: (a) effective heat flux; (a) vapor quality.

(a)

(b)

Fig. 8. Two-phase heat transfer coefficient of ethanol tests for two microchannels a function of: (a)effective heat flux; (a) vapor quality.

micro cone pin fins on the bottom surface of microchannels introduced lots of tiny reentrant cavities, which supplied numerous nucleation sites for the bubble formation. Moreover, these reentrant cavities were surrounded by the micro cone pin fins, which hindered the subcooled liquid to condense and swamp the nucleation embryos inside the reentrant cavities. This facilitated to maintain lots of active and stable nucleation sites during the nucleate boiling process. Therefore, the enhancement of boiling heat transfer can be noted for the structured microchannels with micro pin fins at this stage. With further increase in the effective heat fluxes and vapor qualities, a transition region can be noted for the heat transfer curves of structured microchannels with micro pin fins at moderate heat fluxes, during which the htp of SM-MPF tended to flatted out. Flow visualization results (Fig. 10) indicated that an intermittent flow dominated inside the middle and downstream portion of the microchannels. As aforementioned, after the rapid bubble nucleation and growth from the microchannel surface, these bubbles were confined to agglomerate with the upstream and down-

stream ones, forming long elongated bubbles flow. These elongated bubbles were pushed downstream by the incoming flow, coalesced with more bubbles in the downstream and induced long slugs, as shown in Fig. 10. Different heat transfer tendency from the previous continuous decline was noted. After this transition region, the htp of SM-MPF began to increase slightly with increasing vapor qualities, indicative of a convective boiling process[39]. At this region, the heat transfer performance of SM-MPF sample almost showed an insignificant dependence on the heat fluxes. Annular flow with vapor core was found to dominate inside the structured microchannels, as shown in Fig. 11, and a thin liquid film layer formed between the vapor core and microchannel wall. The heat fluxes were mainly transferred via the thin film evaporation. The increase in the vapor qualities resulted in the reduction in the liquid film thickness, which increased the rate of evaporation and promote the heat transfer process [40]. On the other side, for the smooth microchannels, a general decreasing trend can be still noted with the increase in heat fluxes and vapor qualities at moderate to high heat fluxes regions, differing from the structured

346

D. Deng et al. / International Journal of Heat and Mass Transfer 105 (2017) 338–349

Bubbles

Thin film

Bubbles

(a)

Vapor core

Bubbles Fig. 11. Annular flow near the outlet location of structured microchannels with micro pin fins at ethanol tests, DTsub = 10 °C, G = 200 kg/m2s, q00 eff = 160 kW/m2.

(b) Fig. 9. Bubbly flow in two microchannels: (a) SM-MPF, at water tests, DTsub = 40 °C, G = 200 kg/m2s, q00 eff = 201 kW/m2; (b) RM, at water tests, DTsub = 40 °C, G = 200 kg/m2s, q00 eff = 202 kW/m2.

Flow Elongated bubble t0

insufficient replenishment of liquid finally induced the critical heat fluxes (CHF), as illustrated in Fig. 12. It should be noted that the heat transfer curves of rectangular microchannels tended to flatten out in the moderate heat flux regions before undergoing the final decrease in the high inlet temperature test cases of ethanol tests. The flow visualization results in our previous study [15] indicated that an alternating flow combing with single-phase liquid flow, bubbly flow, slug flow, annular flow and wavy annular flow dominated inside the rectangular microchannels in this regions. Both nucleate boiling and forced convective boiling may contribute to the total heat transfer process, resulting in no remarkable change in the heat transfer performance. The above different flow behaviors of both microchannels induced the significant difference in the heat transfer performance at moderate to high heat fluxes region. It can be noted that the

t0+48ms

Flow t0+96ms Bubbles

t1

t0+208ms

t0+224ms

t1+80ms

t0+240ms Vapor slug t0+296ms

t1+120ms

Fig. 10. Intermittent flow in structured microchannels with micro pin fins at water tests, DTsub = 40 °C, G = 200 kg/m2s, q00 eff = 306 kW/m2.

microchannels significantly. Such trend of smooth microchannels accorded with many previous studies of rectangular microchannels [6,41,42]. Wang and Sefiane [41] attributed the above continuous deterioration in htp to be the occurrence of local dryout inside the microchannels at moderate to high heat fluxes. With the continuous increase in the vapor qualities in the rectangular microchannels, the microchannels were covered with vapor flow. The local dryout firstly occurred partially in the smooth rectangular microchannels, and the liquid were still able to replenish quickly to rewet the microchannel surface. Nevertheless, when the heat fluxes and vapor qualities increased further, the partial dryout extended to longer regions in the microchannels, and the

Local dryout

t1+160ms Local dryout

t1+200ms

Fig. 12. Local dryout in rectangular microchannels at ethanol tests, DTsub = 40 °C, G = 200 kg/m2s, q00 eff = 391 kW/m2.

D. Deng et al. / International Journal of Heat and Mass Transfer 105 (2017) 338–349

structured microchannels with micro pin fins presented significant enhancement in boiling heat transfer coefficient. In water tests, a 10–105% enhancement can be reached for SM-MPF in general. The ethanol tests produced even larger enhancement for SMMPF, i.e., it introduced the heat transfer coefficients 1.9–2.75 times that of smooth microchannels. Unlike the smooth wall surface of SM sample, there were numerous tiny pores inside the micro cone

347

pin fins of the structured microchannels, which is able to supply excellent capillary forces [43]. This introduced significant wicking effects for the liquid intake on the bottom surface. As such, the SM-MPF sample was able to maintain the thin liquid film even to very high heat fluxes. This hindered the occupation of vapor on the wall surface and delayed the occurrence of dry-out considerably. Thin film evaporations can be sustained to high heat fluxes,

Fig. 13. Variation of local thermocouple readings (Ttc5), inlet temperature and pressure fluctuation of both SM-MPF and RM samples at: (a) low heat flux; (b) moderate heat flux; (c) high heat flux.

348

D. Deng et al. / International Journal of Heat and Mass Transfer 105 (2017) 338–349

and high heat transfer rates can be maintained for SM-MPF without notable deteriorations. Therefore, considerable heat transfer enhancement can be reached for the structured microchannels with micro pin fins. 4.4. Two-phase Flow instabilities In order to compare the two-phase flow instabilities of both structured microchannels with micro pin fins and conventional rectangular microchannels, the alleviation methods for twophase instabilities as described in Section 3.1 were utilized instead of the elimination methods, such as the installation of inlet restrictors in the upstream of test section [44]. The occurrence of twophase flow instabilities is directly related to the fluctuation of local thermocouple readings, inlet temperature and pressure, which is able to provide information for the wall temperature variations and flow oscillations. The amplitude of the temperature and inlet pressure fluctuations were presented using a standard deviations (r) method, which is given as follows,

rðTÞ ¼

vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi uX 2 u n u ðT n  T n Þ t n¼1

n

ð10Þ

and

rðPin Þ ¼

vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi uX 2 u n  u Pin;n  Pin;n t n¼1

n

ð11Þ

where n is number of data collected at a certain heat flux under a given mass flux condition. In the high inlet temperature cases, both SM-MPF and RM samples presented fairly stable boiling process with negligible fluctuations in the temperature and inlet pressure, which accorded with the previous study [15]. Nevertheless, it was found that twophase flow instabilities occurred for both microchannels samples in the low inlet temperature cases, which is especially notable for the conventional rectangular microchannels, as shown in Fig. 13. It can be noted that the SM-MPF samples showed much smaller fluctuation amplitude of the local wall temperatures than the rectangular microchannels, i.e., the r of local thermocouple readings can reduced by 0.6–2.2 times for the SM-MPF sample. As aforementioned, the structured microchannels produced much more stable bubble nucleation sites, and dissipated heat fluxes uniformly by continuous bubble formations and departure. Besides, they also facilitated to supply efficient wall surface rewetting due to its excellent capillary force. They hindered the occurrence of local dry-out and thus reduce the local wall temperature spikes. Therefore, significant improvement of the wall temperature stability can be noted during the flow boiling process for the structured microchannels with micro pin fins. This enhancement is especially favorable for the robust and safe operations of microchannel heat sinks for high heat fluxes dissipations, as the thermal stress can be reduced considerably. From Fig. 13, it can be observed that the structured microchannels also mitigated the severe inlet temperature and pressure oscillations. At small to moderate heat fluxes, the rectangular microchannels suffered severe flow instabilities, i.e., the amplitude of Pin and Tin can reach as high as 11 kPa and 28 °C. On the contrary, there are no notable fluctuations in the Tin and Pin of the structured microchannels with micro pin fins. The standard deviations (r) of inlet temperature and pressure is reduced as high as 2.8 and 7.7 times for the SM-MPF compared to the conventional rectangular microchannels. This may be linked to that the micro cone pin fins in the structured microchannels provided much more active nucle-

ation sites. It produced more continuous and stable bubble behaviors compared to the smooth rectangular microchannels, and resulted in much smaller flow oscillations. At high heat fluxes, despite that the SM-MPF sample began to presented notable fluctuations in the Pin and Tin, the amplitude of the inlet pressure variations were still much smaller than the rectangular microchannels. In general, the addition of micro pin fins on the bottom surface help to alleviate the detrimental two-phase flow instabilities. This helps to delay the occurrence of premature dryout which lead to CHF limitations [44], and also reduce the mechanical vibration of microchannel heat sinks. 5. Conclusions Structured microchannels with micro pin fins on the bottom surface were proposed and fabricated. The flow boiling performance of these structured microchannels was assessed together with the comprehensive comparisons with the conventional rectangular microchannels with smooth bottom walls. The tests were performed utilizing two coolants, deionized water and ethanol, at different inlet subcoolings and flow rates for a wide range of heat fluxes. The structured microchannels with micro pin fins were demonstrated to introduce significant flow boiling heat transfer enhancement compared to the conventional counterpart with smooth walls in all the test conditions, i.e., they generally presented an enhancement of 10% to 105% in water tests, and 90% to 175% in ethanol ones. The additions of micro cone pin fins on the microchannels wall surface supplied numerous stable nucleation sites due to the formation of lots of tiny reentrant cavities. Moreover, they provided excellent capillary forces and introduced significant wicking effects for the liquid rewetting, which hindered the local dryout and maintain high heat transfer coefficients even at high heat fluxes and vapor qualities. Critical heat fluxes were notably promoted for the SM-MPF sample compared to the rectangular microchannels. Furthermore, the two-phase flow instabilities were also found to be mitigated for the SM-MPF sample. The fluctuations in the local wall temperatures, inlet temperature and pressure were reduced considerably compared to the rectangular microchannels at small to moderate heat fluxes in the low inlet temperature cases. From the above, the structured microchannels with micro pin fins seem to provide a promising alternative in microchannel heat sinks for high heat flux dissipation applications. Acknowledgement The research was financially supported under the Grants of the National Nature Science Foundation of China (No. 51405407), the Natural Science Foundation of Fujian Province (No. 2015J05112), the Fundamental Research Funds for the Central Universities, Xiamen University (No. 20720150094), the Fundamental Research Funds for the Xiamen Universities (No. 20720152002), and the Science and Technology Planning Project for Industry-UniversityResearch Cooperation in Huizhou City (Grant No. 2014B050013002). Furthermore, the financial support of Collaborative Innovation Center of High-End Equipment Manufacturing in FuJian is also acknowledged. References [1] B. Agostini, M. Fabbri, J.E. Park, L. Wojtan, J.R. Thome, B. Michel, State of the art of high heat flux cooling technologies, Heat Transfer Eng. 28 (2007) 258–281. [2] I. Mudawar, Recent advances in high-flux, two-phase thermal management, J. Therm. Sci. Eng. App. ASME J. Thermal Sci. Eng. Appl. 5 (2013) 021012. [3] T. Harirchian, S.V. Garimella, Microchannel size effects on local flow boiling heat transfer to a dielectric fluid, Int. J. Heat Mass Transfer 51 (2008) 3724– 3735. [4] D.B. Tuckerman, R.F.W. Pease, High-performance heat sinking for VlSI, IEEE Electron Device Lett. ELD-2 (5) (1981) 126–129.

D. Deng et al. / International Journal of Heat and Mass Transfer 105 (2017) 338–349 [5] Z. Wu, B. Sundén, On further enhancement of single-phase and flow boiling heat transfer in micro minichannels, Renewable Sustainable Energy Rev. 40 (2014) 11–27. [6] W. Qu, I. Mudawar, Flow boiling heat transfer in two-phase micro-channel heat sinks—I. Experimental investigation and assessment of correlation methods, Int. J. Heat Mass Transfer 46 (2003) 2755–2771. [7] M.E. Steinke, S.G. Kandlikar, An experimental investigation of flow boiling characteristics of water in parallel microchannels, ASME J. Heat Transfer 126 (2004) 518–526. [8] G. Hetsroni, A. Mosyak, E. Pogrebnyak, Z. Segal, Periodic boiling in parallel micro-channels at low vapor quality, Int. J. Multiphase Flow 32 (2006) 1141– 1159. [9] G. Wang, P. Cheng, A.E. Bergles, Effects of inlet/outlet configurations on flow boiling instability in parallel microchannels, Int. J. Heat Mass Transfer 51 (2008) 2267–2281. [10] S.G. Kandlikar, Mechanistic considerations for enhancing flow boiling heat transfer in microchannels, ASME J. Heat Transfer 138 (2016) 021504. [11] A. Megahed, Experimental investigation of flow boiling characteristics in a cross-linked microchannel heat sink, Int. J. Multiphase Flow 37 (2011) 380– 393. [12] S.G. Kandlikar, T. Widger, A. Kalani, V. Mejia, Enhanced flow boiling over open microchannels with uniform and tapered gap manifolds, ASME J. Heat Transfer 135 (2013) 061401. [13] C.T. Lu, C. Pan, Stabilization of flow boiling in microchannel heat sinks with a diverging cross-section design, J. Micromech. Microeng. 18 (2008) 075035. [14] K. Balasubramanian, P.S. Lee, L.W. Jin, S.K. Chou, C.J. Teo, S. Gao, Experimental investigations of flow boiling heat transfer and pressure drop in straight and expanding microchannels – A comparative study, Int. J. Therm. Sci. 50 (2011) 2413–2421. [15] D. Deng, W. Wan, Y. Tang, Z. Wan, D. Liang, Experimental investigations on flow boiling performance of reentrant and rectangular microchannels – A comparative study, Int. J. Heat Mass Transfer 82 (2015) 435–446. [16] S.G. Kandlikar, W.K. Kuan, D.A. Willistein, J. Borrelli, Stabilization of flow boiling in microchannels using pressure drop elements and fabricated nucleation sites, ASME J. Heat Transfer 128 (2006) 389–396. [17] C.J. Kuo, Y. Peles, Local measurement of flow boiling in structured surface microchannels, Int. J. Heat Mass Transfer 50 (2007) 4513–4526. [18] C.N. Ammerman, S.M. You, Enhancing small-channel convective boiling performance using a microporous surface coating, ASME J. Heat Transfer 123 (2001) 976–983. [19] H. Wang, R.B. Peterson, Enhanced boiling heat transfer in parallel microchannels with diffusion brazed wire mesh, IEEE Trans. Compon. Packag. Technol. 33 (2010) 784–793. [20] Y. Sun, L. Zhang, H. Xu, X. Zhong, Subcooled flow boiling heat transfer from microporous surfaces in a small channel, Int. J. Therm. Sci. 50 (2011) 881–889. [21] V. Khanikar, I. Mudawar, T. Fisher, Effects of carbon nanotube coating on flow boiling in a micro-channel, Int. J. Heat Mass Transfer 52 (2009) 3805–3817. [22] F. Yang, X. Dai, Y. Peles, P. Cheng, J. Khan, C. Li, Flow boiling phenomena in a single annular flow regime in microchannels: (I) characterization of flow boiling heat transfer Int, J. Heat Mass Transfer 68 (2014) 703–715. [23] S. Krishnamurthy, Y. Peles, Flow boiling of water in a circular staggered micropin fin heat sink Int, J. Heat Mass Transfer 51 (2008) 1349–1364. [24] S. Krishnamurthy, Y. Peles, Flow boiling heat transfer on micro pin fins entrenched in a microchannel ASME, J. Heat Transfer 132 (2010) 041007.

349

[25] A. Kosar, Y. Peles, Boiling heat transfer in a hydrofoil-based micro pin fin heat sink, Int. J. Heat Mass Transfer 50 (2007) 1018–1034. [26] W. Qu, A. Siu-Ho, Experimental study of saturated flow boiling heat transfer in an array of staggered micro-pin-fins Int, J. Heat Mass Transfer 52 (2009) 1853– 1863. [27] A. Ma, J. Wei, M. Yuan, J. Fang, Enhanced flow boiling heat transfer of FC-72 on micro-pin-finned surfaces Int, J. Heat Mass Transfer 52 (2009) 2925–2931. [28] W.R. Chang, C.A. Chen, J.H. Ke, T.F. Lin, Subcooled flow boiling heat transfer and associated bubble characteristics of FC-72 on a heated micro-pin-finned silicon chip Int, J. Heat Mass Transfer 53 (2010) 5605–5621. [29] M. Law, P.S. Lee, A comparative study of experimental flow boiling heat transfer and pressure characteristics in straight- and oblique-finned microchannels Int, J. Heat Mass Transfer 85 (2015) 797–810. [30] C. Woodcock, X. Yu, J. Plawsky, Y. Peles, Piranha Pin Fin (PPF) — Advanced flow boiling microstructures with low surface tension dielectric fluids, Int. J. Heat Mass Transfer 90 (2015) 591–604. [31] D. Deng, W. Wan, Q. Huang, X. Huang, W. Zhou, Investigations on laser micromilling of circular micro pin fins for heat sink cooling systems, Int. J. Adv. Manuf. Tech. (2016), http://dx.doi.org/10.1007/s00170-016-8468-9. [32] D. Deng, Y. Tang, D. Liang, H. He, S. Yang, Flow boiling characteristics in porous heat sink with reentrant microchannels, Int. J. Heat Mass Transfer 70 (2014) 463–477. [33] D. Deng, R. Chen, Y. Tang, L. Lu, T. Zeng, W. Wan, A comparative study of flow boiling performance in reentrant copper microchannels and reentrant porous microchannels, Int. J. Multiphase Flow 72 (2015) 275–287. [34] D. Deng, W. Wan, H. Shao, Y. Tang, J. Feng, J. Zeng, Effects of operation parameters on flow boiling characteristics of heat sink cooling systems with reentrant porous microchannels, Energy Conver. Manage. 96 (2015) 340–351. [35] J. Xu, G. Liu, W. Zhang, Q. Li, B. Wang, Seed bubbles stabilize flow and heat transfer in parallel microchannels, Int. J. Multiphase Flow 35 (2009) 773–790. [36] J.R. Taylor, An Introduction to Error Analysis, second ed., University Science Books, 1997. [37] R.K. Shah, A.L. London, Laminar Flow Forced Convection in Ducts, Academic Press, New York, 1978. [38] M. Law, P.S. Lee, K. Balasubramanian, Experimental investigation of flow boiling heat transfer in novel oblique-finned microchannels, Int. J. Heat Mass Transfer 76 (2014) 419–431. [39] S. Szczukiewicz, N. Borhani, J.R. Thome, Fine-resolution two-phase flow heat transfer coefficient measurements of refrigerants in multi-microchannel evaporators, Int. J. Heat Mass Transfer 67 (2013) 913–929. [40] K. Balasubramanian, M. Jagirdar, P.S. Lee, C.J. Teo, S.K. Chou, Experimental investigation of flow boiling heat transfer and instabilities in straight microchannels, Int. J. Heat Mass Transfer 66 (2013) 655–671. [41] Y. Wang, K. Sefiane, S. Harmand, Flow boiling in high-aspect ratio mini- and micro-channels with FC-72 and ethanol experimental results and heat transfer correlation assessments, Exp. Therm. Fluid Sci. 36 (2012) 93–106. [42] E. Sobierska, R. Kulenovic, R. Mertz, M. Groll, Experimental results of flow boiling of water in a vertical microchannel, Exp. Therm. Fluid Sci. 31 (2006) 111–119. [43] Y. Zhu, D.S. Antao, K.H. Chu, S. Chen, T.J. Hendricks, T. Zhang, E.N. Wang, Surface structure enhanced microchannel flow boiling, ASME J. Heat Transfer 138 (2016) 091501. [44] W. Qu, I. Mudawar, Transport phenomena in two-phase micro-channel heat sinks, ASME J. Heat Transfer 126 (2004) 213–224.