Enhanced boiling heat transfer on composite porous surface

Enhanced boiling heat transfer on composite porous surface

International Journal of Heat and Mass Transfer 80 (2015) 107–114 Contents lists available at ScienceDirect International Journal of Heat and Mass T...

2MB Sizes 7 Downloads 55 Views

International Journal of Heat and Mass Transfer 80 (2015) 107–114

Contents lists available at ScienceDirect

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

Enhanced boiling heat transfer on composite porous surface Pengfei Xu a, Qiang Li a,⇑, Yimin Xuan a,b a b

The School of Energy and Power Engineering, Nanjing University of Science and Technology, Nanjing, Jiangsu 210094, China College of Energy and Power, Nanjing University of Aeronautics and Astronautics, Nanjing, Jiangsu 210016, China

a r t i c l e

i n f o

Article history: Received 19 March 2014 Received in revised form 28 July 2014 Accepted 22 August 2014

Keywords: Pool boiling Heat transfer enhancement Composite porous surface Surface wettability Visualization

a b s t r a c t Composite porous surfaces were prepared to investigate the saturated pool boiling of de-ionized water. Scanning electron microscope (SEM) micrographs showed that the porous layers contains three types of structures, including macro pores above 200 lm diameter, micro pores around 2 lm diameter and dendritic structure around 400 nm diameter. Results showed that, the thickness of the coating layers was 33 lm, 84 lm, 156 lm and the contact angle of de-ionized water against the porous surface reduced with the increase of the coating thickness (35°, 8° and 0°). The highest porosity of the coating layers was 94.4%. The experimental results of pool boiling heat transfer indicated that the critical heat flux (CHF) increased with surface wettability and coating thickness, while the effect of the surface wettability on nucleate boiling heat transfer was complicated. The highest CHF of the porous surface was 239 W/cm2, which is 101% higher than that of plain surface. High speed photography was used to observe the bubble behaviors in order to investigate the mechanism of enhanced nucleate boiling. Visualization data indicated that the coalescent bubble on porous surface grew more quickly and the bubble size was larger than that on plain surface under high heat flux range. Consequently, the departure of bubble brought more liquid replenishment, and thus to obtain the continuous rise of heat transfer coefficient and higher CHF value. Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction It is urgent to find effective high-heat-flux removal techniques due to that the energy consumption is a momentous embarrassment to limit the development of electronic industry due to the advent of high-density electronic components. Phase change cooling technique is one of the most powerful cooling solutions for high-heat-flux removal due to its high heat transfer rates. However, broader applications of two-phase change heat transfer are restrained by the critical heat flux (CHF), which determines the maximum power density that could be handled by a boiling heat transfer device. Among the affecting factors of the CHF, surface morphology and wettability have been identified as the key factors. Controlling the surface characteristics would in turn provide new opportunity in CHF enhancement. Previous studies [1–4] have identified surface morphology as one of the key factors affecting the CHF. Significant enhancement in the CHF has been reported for the heat transfer surface coated with porous layers. For example, Kubo et al. [1] manufactured four kinds of treated surfaces with combination of two cavity mouth diameter (about 1.6 lm and 3.1 lm) and two number of densities ⇑ Corresponding author. Tel.: +86 25 8431 5837. E-mail address: [email protected] (Q. Li). http://dx.doi.org/10.1016/j.ijheatmasstransfer.2014.08.048 0017-9310/Ó 2014 Elsevier Ltd. All rights reserved.

of the micro-reentrant cavities (81 1/cm2 and 9600 1/cm2) by use of microelectronic fabrication techniques. Their results demonstrated that the surface with large mouth diameter and larger number density of micro-reentrant obtained higher CHF. Kim et al. [2] studied the nucleate pool boiling heat transfer enhancement of microporous surfaces coated with 390 lm diameter platinum. Their results showed the microporous coating appears to delay CHF by increasing the convection heat transfer contribution with a subsequent decrease in latent heat (vapor generation) and/ or increasing the hydrodynamic stability of the vapor leaving the surface from increased bubble vapor inertia. Li et al. [3] reported pool boiling heat transfer of de-ionized in sintered copper mesh layer with 56 lm wire diameter. Their results demonstrated that the boiling heat transfer coefficient was only related to the exposed surface area and was not affected by the layer thickness, while the CHF was strongly dependent on the layer thickness and increases proportionally with increase in the wick thickness. Mori and Okuyama [4] enhanced CHF by the attachment of a honeycombstructured porous plate with different thickness (1.2 mm, 5 mm, 10 mm) on a heated surface. As the thickness of the honeycomb porous plate on the heated surface decreased, the CHF increases to 250 W/cm2. It should be noted that these studies for the surface morphology affecting CHF were mainly focused on the largescale porous structure surfaces. Although the large-scale porous

108

P. Xu et al. / International Journal of Heat and Mass Transfer 80 (2015) 107–114

Nomenclature D h I L1 L2 q00 T u U

diameter of the copper sample (m) heat transfer coefficient (W/(m2 K)) current (A) distance between surface and the first thermocouple (m) distance between two thermocouples (m) heat flux (W/m2) temperature (K) uncertainty voltage (V)

structure surface can effectively increase CHF due to its higher active nucleate site and extended surface area, the wettability of these large-scale porous structure surfaces is not met to the need of further improvement. The surface wettability is another key factor affecting CHF Surface with high wettability has been identified to enhance phase change heat transfer [5]. The wettability of a surface can be tuned by changing its surface morphology. Recently, the nano-scale surface is boomed in changing the surface morphology. For instance, Zhang and Kim [6] proposed a three dimensional, interconnected alumina nano porous surface (ANPS) using an anodic oxidation process. Electrochemical impedance spectroscopy was utilized to studied the quantitative evaluation of liquid uptake. Their results demonstrated that the CHF augment trend is well matched with the amount of liquid absorbed into the porous media. Kim et al. [7] prepared artificially nano-scale surfaces with arrays of ZnO nanorods. The contact angle of the nanorods surface against the water was less than 5°. Experiment results revealed that the nanorods surface could increase CHF significantly. Lu et al. [8] used Si nanowire coated-surface with a zero contact angle obtained by wafer-scale electroless etching to enhanced boiling heat transfer. They presented that the Si nanowire array coated surface has a higher CHF than the plain Si surface due to its better liquid spreading. Chen et al. [9] reported a twofold enhancement in CHF on Cu and Si nanowire arrays presumably because of a large capillary force provided by the nanowire arrays. In this work, we prepared the composite structure surface combining the high nucleation site density of the large-scale structure surfaces and the high wettability of the nano-scale structure surfaces to enhance phase change heat transfer. Pool boiling heat transfer experiments with de-ionized water at atmospheric pressure were performed both on the composite structure surfaces and the plain surface to contrast the heat transfer performance. The measured parameters were bubble departure diameter and nucleation density through high-speed photographic. The ultimate goal was to use the data to get a deeper insight to the heat transfer results. 2. Preparation and characterization of the composite porous surface The composite porous copper surfaces were prepared by an electrochemical method [10]. This type of porous surface has been used extensively in different applications such as catalysis, molecular sieves, fuel cells, sorption and separation. Li et al. [11] fabricated a well-ordered 3D macro-porous metallic surface layer with nanostructured porosity by the electrochemical method, and pool boiling heat transfer experiment with R134a indicated that the heat transfer coefficient is enhanced over 17 times at 1 W/cm2 compared to a plain reference surface. The overpotential [12], electrolytic solutions [13], additive [14], deposition time [15]

Greek symbols h contact angle Subscripts w wall f fluid loss heat loss

and electrolytic process [16] have been researched to improve the preparation technology. In the preparation process, copper rod sample with 12 mm diameter was placed as cathode in the electroplating solution which was consisted of 0.4 M CuSO4 and 1.5 M H2SO4. Phosphor copper plate with 40 mm length and 40 mm width was placed 20 mm higher than the copper sample serving as anode. Two simultaneous reduction reactions for the Cu2+ and H+ ions in the electrolyte solution happens at the cathode s as soon as the electric current was supplied to the electrochemical deposition cell:

Cu2þ þ 2e !Cu

and 2Hþ þ 2e !H2

Other than most of traditional deposition cases, the hydrogen formed at a much higher rate in our cases. The rapidly generated bubbles partly occupied volume of the copper deposit sediment, leaving a porous surface with multi-scale structure. A series of porous surfaces were prepared under different deposition time. Table 1 listed the characteristics of these porous surfaces. The mass and thickness of the porous layer almost linearly increased with the increment of the deposition time. Static contact angle of 5 lL deionized water at the room temperature (about 25 °C) were measured by the Contact Angle Meter JC2000D2 (Shanghai Zhongchen Digital Technology Apparatus Co., Ltd.) on the porous surfaces and plain surface. The plain copper surface was sanded by 1600 mesh paper and the arithmetic mean surface roughness was 1 lm. Experimental results manifested that the contact angle of the plain surface, P1 surface, P2 surface was 81°, 35° and 8°, respectively. The images of contact angle were shown in Fig 1. The contact angel of the P3 surface was considered as 0° because the de-ionized water spread out immediately once it contacted the porous surface [17,18]. Therefore, the P3 surface was of super-hydrophilic characteristic and the contact angle could not represent its wettability completely. The liquid spreading ability is another momentous factor that represents surface wettability for the super-hydrophilic surface [17]. Shirazy et al. [19] develop a new method to compare the wettability of super-hydrophilic surface by measuring time-ofspreading of a droplet. SEM micrographs indicate that three types of structure were included in the porous layer: macro pores above 0.2 mm diameter, micro pores around 2 lm diameter and dendritic structure around 400 nm diameter, as shown in Fig. 2. Both the macro pores above 0.2 mm and micro pores can be active nucleation site at appropriate surface superheat in our pool boiling experiment according Table 1 Characteristics of the porous surfaces. Number

Deposition time/s

Mass/ mg

Thickness/ lm

Porosity (%)

Contact angle (°)

P1 P2 P3

10 20 40

2.8 5.3 8.77

33.2 84.2 156

91.6 93.7 94.4

35 8 0

P. Xu et al. / International Journal of Heat and Mass Transfer 80 (2015) 107–114

109

Fig. 1. Contact angle image of the surface against the de-ionized water (a) the plain surface (b) the P1 surface (c) the P2 surface.

Fig. 2. SEM micrographs of the composite porous surface (a) macro porous (b) micro porous (c) dendritic structure (d) side view of the porous surface.

Hsu’s analysis [20] Side view micrograph confirmed that the pores were cross-linked along the surface. Sintering was introduced to strengthen the structure. The porous surfaces were heated to 500 °C and kept for one hour under a reduction atmosphere. Through these processes, plenty of sintering necks were formed without changing the massive structure. 3. Experimental 3.1. Pool boiling experimental system The experimental setup is shown in Fig. 3. It consisted of five parts such as the heating module, the power supply module, the boiling vessel, the data collection module and the visualization module. The heating module consisted of five cartridges heaters, copper block and copper rod sample. Five cartridge heaters embedded in the copper block were used to provide heat to the surface. Copper rod sample was fixed to copper block by threaded connection. Teflon was selected to fix the copper rod sample due to its excellent processability and low heat conductivity coefficient. The dimensional coordination ensures the copper foam exposed in the pool liquid. The Teflon should be modified with sodium first

to ensure the tightness and then. Then the interstice between copper rod sample and teflon was filled up with flexible epoxy (3M DP190). The boiling vessel was separated into two chambers by the teflon. The chamber upon the teflon was charged with working fluid while the chamber below that was stuffed with thermal insulating material to minimize the heat losses. The data collecting module included three K-type thermocouples, a data acquisition unit (HP 34901) and a computer. The arrangement of thermocouples is shown in Fig. 3(b). Two thermocouples with 0.5 mm diameter were located at the center of the copper rod sample with 15 mm intervals, from which the wall temperature could be determined by the linear temperature distribution. To decrease the error of measurement, the first thermocouple was placed 3 mm below the heat transfer surface. The visualization module consisted of a high-speed camera (IDT Y4), a LED Lamp and two lasers, as shown in Fig. 4. The maximum frame rate of the camera is 3000 frames per second at full resolution (1024  1024 pixels). With a reduction in the field of view, higher frame rates can be achieved. The maximum frame rate used in our study was 12,000 fps. A zoom lens (Navitar Zoom6000) attached to the camera through an adapter (Navitar 2.0x) was used to magnify the area of interest. The lens was capable of a variable

110

P. Xu et al. / International Journal of Heat and Mass Transfer 80 (2015) 107–114

were opened in all heat flux text. The LED lamp was on the opposite side of the high-speed camera to serve as black light. The laser were placed on either side of the high- speed camera and aimed at the heat transfer surface. The angle between the incident light and the surface was about 20°. To enlarge the lighting area, an expander was used to increase the diameter of laser beam to 7 mm. Before the experiment, de-ionized water was charged in the chamber. The distance between the liquid level and surface was set as 150 mm. Auxiliary heaters were turned on to boil the liquid for one hour to remove the non-condensable gas. During the experiment, the temperature of the working fluid was kept as saturation temperature and the temperature fluctuation of the working fluid was less than 0.2 °C. The heat transfer was considered to reach a steady state if the variation of the wall temperature was smaller than 0.1 °C in ten minutes. After gathering temperature data, lighting source was turned on and high-speed photographs were captured. The same steps were repeated at each heat flux.

3.2. Data reduction and experimental uncertainty The heat flux, wall temperature and heat transfer coefficient could be calculated with the collected data. The heat flux can be expressed as

q00 ¼ 1

UI

pD2 4

Fig. 3. Pool boiling experimental system (a) experimental setup (b) heating module and the arrangement of the thermocouples.

ð1Þ

where U is the voltage applied on the cartridge heaters, I is the current running through the cartridge heaters and D is the diameter of the copper rod. The heat conduction in the copper rod can be consider as onedimension steady-state heat conduction after carefully preparation of thermal isolation. According to Fourier’s law, the heat flux can be expressed as:

q00 ¼ k

Tw  T1 T1  T2 ¼ k L1 L2

ð2Þ

where L1 is the distance between the top of the copper rod and first thermocouple, L2 is the distance between the two thermocouples. T w is the wall temperature. T 1 and T 2 are the temperature measured by the first and second thermocouples respectively. Thus, the wall temperature can be expressed as the following form:

Tw ¼

Fig. 4. Diagram of visualization module.

magnification of 0.7–4.5 times the original size of the object. As the sensor format is 1 inch, the field of view could be changed from 22.8  22.8 mm to 3.5  3.5 mm. The angle of the camera and the heat transfer surface was set as about 30°. Through this observation angle, the density of nucleating site, the bubble departure diameter and bubble departure frequency can be figured out. To reduce the heating effect, a LED lamp with power of 100 W and two lasers with power of 1 W were selected as lighting source. Verification experiment was performed to examine the heating effect. Experimental results indicated that the temperature rise of the heat transfer surface was less than 0.1 °C after all lighting source

  L1 L1 þ 1 T1  T2 L2 L2

ð3Þ

It should be mention that the wall temperature represents the temperature of the bottom of the porous coating for porous surface in this work. Heat transfer coefficient is given by the following formulation



q00 Tw  Tf

ð4Þ

where Tf is the temperature of fluid. The uncertainties of the measurement parameters are analyzed by the error propagation method. The uncertainty of direct measurement and indirect measurement were listed in Tables 2 and 3, respectively. It is obvious that the maximum relative uncertainty of heat flux is 6.2% under the condition of considering the heat loss. The relative uncertainty of heat transfer coefficient was less than 8%.

111

P. Xu et al. / International Journal of Heat and Mass Transfer 80 (2015) 107–114 Table 2 The uncertainty of direct measurement. Parameters

Uncertainty

The smallest measuring data

The maximum relative uncertainty

D L1 L2 U I Tf T1 T2

0.02 mm 0.02 mm 0.02 mm 0.052V 0.0041A 0.2 °C 0.2 °C

12 mm 3 mm 15 mm 3.0V 0.29A – –

0.17% 0.67% 0.13% 1.72% 1.44% – –

4. Result and discussion 4.1. Pool boiling heat transfer performance Pool boiling experiments were conducted to investigate potential enhancements in saturated nucleate boiling of de-ionized water at atmosphere on composite porous surfaces and plain surface. The results are shown in Figs. 5 and 6. The heat transfer coefficient curves curve can be divided into three regions, based on the heat transfer performance, as shown in Fig. 6. In the low heat flux region (<10 W/cm2), the composite porous surfaces enhanced the heat transfer compared with the plain surface. When the heat flux increased between 10 W/cm2 and 40 W/cm2, the heat transfer coefficient of plain surface gained more rapid increase comparing with that of the porous surfaces. In this region, the plain surface showed a little advantage against the porous surfaces. As the heat flux in excess of 40 W/cm2, the heat transfer coefficient of the plain surface showed little growth or even some decrease. On the other hand, the heat transfer coefficient of the porous surfaces kept the trend of increasing until near CHF. The heat transfer performances of the three composite porous surfaces differ in this region. The surface with contact angle of 8° showed higher heat transfer coefficient until it reached to CHF. The highest heat transfer coefficient gained on the composite porous surface with contact angle of 0°. The value was 5.9 W/(cm2 K), 120% higher than that of plain surface. Therefore, it is difficult to find a surface with certain wettability that shows best heat transfer performance during the whole range of heat flux in our experiment. A clear trend between the coating thickness and the CHF augment was observed, as shown in Fig. 7. The CHF increased with the increase of the coating thickness disproportionately. The highest CHF was obtained on the P3 sample with thickness of 153 lm (239 W/cm2), 101% enhanced compared with that of plain surface. The increased thickness of porous surface affected the CHF by the follower three factor: extended heat transfer area, added micro and nono structure, degraded thermal resistance and bubble departure resistance. The extended area was considered as an important factor in traditional heat transfer enhancement method.

Fig. 5. Saturated pool boiling curves for de-ionized water on composite porous surface and plain surface.

Fig. 6. Variation of heat transfer coefficient with heat flux.

However, the porous layers in our experiment were consist of numerous loose dendritic structure around 400 nm diameter. The heat conduction between the dendritic structure was poor. Therefore, the extended area was not believed to be the primary factor affecting CHF. As the coating thickness increased, more micro pores and nano structures were formed in the coating layer, which result in more nucleation sites and higher capillary suction. Both of the two factor benefit the CHF augment. The negative effect of thermal

Table 3 The uncertainty of indirect measurement. Parameters q00

Calculation expression uq00 q00

"

2

¼ ðuUU Þ þ 2 

u 2 I

I

2

þ ð2 uDD Þ þ



uq00

2 #1=2

6 uT w ¼ 4  2

h uh h

L1 L2

 2 31=2 uq00 2 uT w 6 ð q00 Þ þ T w T f 7 ¼4 5  u T 2 f þ T w T f

The maximum relative uncertainty 6.2%

0.23 °C



loss

q00loss

31=2 2  2 þ 1 u2T 1 þ LL12 u2T 2 þ 7 5 2  2 T 2 T 1 1 u2L1 þ LL12 T 2LT u2L2 L2 2

Tw

Uncertainty –

8%

112

P. Xu et al. / International Journal of Heat and Mass Transfer 80 (2015) 107–114

Fig. 7. Variation of CHF with coating thickness.

resistance and bubble departure resistance was insignificant due to the highly porosity (around 94%) and numerous macro pores compared with previous works [3,4]. Fig. 8 shows the CHF experiment result of the porous surfaces, as well as the Kandlikar’s predictions [5], which include the effect of surface wettability. The present data followed the same trend as Kandlikar’s prediction, however, the values were higher than the prediction. Note that the Kandlikar’s model did not consider the effect of heater size on CHF, which was regarded as an nonnegligible factor for small heater transfer surface [21]. Bar-Cohen and McNeil [22] proposed that the CHF was relatively constant for large surfaces and increased for decreasing heater size if it past a certain transfer point. The transfer point is 50 mm for present study, which is much longer than heater size (diameter of 12 mm) in this work. Simply, a correction factor could be applied to the Kandlikar’s model to consider the size effect. It can be seen that the experiment results can qualitatively match the theory when the correction factor was set as 1.4 for our experiment. A rapidly increase was found near the contact angle of 0°, which could not be explained by Kandlikar’s model. The extra enhancement may owed to the liquid spreading ability, which is supported by Ahn et al. [17]. When a thin absorption layer existed beneath a growing bubble during boiling, better liquid spreading promotes better hot-spot rewetting, thus, delaying the CHF [23].

4.2. Visualization research

Fig. 8. Variation of CHF with Contact angle of water and Kandlikar’s prediction.

High speed photography demonstrated that there were three successive regions, based on the bubble forms: isolated bubbles nucleate boiling, fully developed nucleate boiling and bubble coalescence nucleate boiling. In the isolated bubbles nucleate boiling, isolated bubbles formed on the surface, and went through a complete bubble cycle: growing up, departure and waiting. As the heat flux increased, both the active nucleate site and the bubble departure diameter increased rapidly, it came into fully developed nucleate boiling region. In this region, bubble behaviors were affected by the former departure bubbles and nearby bubbles. Violent relative movement of vapor against liquid occurred, heat transfer coefficient gained a quick increase. In the bubble coalescence nucleate boiling, all of the small bubbles coalesced into

Fig. 9. Isolated bubbles nucleate boiling (at heat flux 6.6 W/cm2) (a) plain surface (b) composite porous surface (with the contact angle of 0°, the same as following figures).

P. Xu et al. / International Journal of Heat and Mass Transfer 80 (2015) 107–114

vapor clot once they formed. Single bubble behavior could not be detected, while the coalescent bubble departure frequency and coalescent bubble shape could be analyzed. Heat transfer performance can search evidence from the bubble behavior. It was interesting to notice that the regions dividing based on the bubble behavior fit well with that based on heat transfer performance. Isolated bubbles nucleate boiling images of plain surface and the composite porous surface (with the contact angle of 0°, the same as following figures) are shown in Fig. 9. The number of active nucleation site and bubble departure frequency can be calculated from the images. The moment a particular bubble came into growing was defined as 0 ms. At the time of 10 ms, there were three bubbles formed on the plain surface, otherwise, the number of bubbles formed on the composite porous surface was in excess of ten. In particular, it took 14 ms for a bubble to depart from the

113

plain surface and 4 ms for a new bubble to form. By contrast, it only needs about 4 ms for a bubble to depart from the plain surface and nearly no waiting time. Due to higher density of the active site for bubble nucleation and higher bubble departure frequency, the porous surface enhanced heat transfer in this region. As the heat flux exceeded 10 W/cm2, fully developed nucleate boiling occurred on the surface (Fig. 10). For plain surface, more nucleation sites were activated with the increasing heat flux. The bubble waiting time decreased obviously while the bubble departure diameter increased rapidly. In addition, the bubbles were dispersive and the interaction between the bubbles only emerged obviously when the bubbles grew to large diameter. The departure of each bubble could cause violent agitation of the liquid, resulting in more liquid replenishment and intense convection. However, the dense bubbles formed a vapor blanket on the porous surface. As shown in Fig. 10(b), the boundary of the vapor blanket was

Fig. 10. Fully developed nucleate boiling (at heat flux 33 W/cm2) (a) plain surface (b) composite porous surface.

Fig. 11. Bubble coalescence nucleate boiling (at heat flux 118 W/cm2) (a) plain surface (b) porous surface.

114

P. Xu et al. / International Journal of Heat and Mass Transfer 80 (2015) 107–114

nearly unchanged, indicating that the motions of liquid near the surface were inhibited. Therefore, the boiling heat transfer coefficient of the composite porous surface deteriorated in this region. The same phenomenon was found in other literature [24]. Zhang et al. [25] pointed that the vapor blanket on the porous surface could be avoid through surface wettability control method. When the heat flux was increased up to 40 W/cm2, bubbles coalesced into vapor clot. Nearly no difference was found in coalesced bubble departure frequency between the plain surface and the porous surface. Nevertheless, there were some differences in coalesced bubble shape. As shown in Fig. 11, it took 60 ms for coalescent bubble on plain surface to grow up to 24.7 mm diameter. Meanwhile, it only took 20 ms for the porous surface to grow up to 22 mm. The coalesced bubble on porous surface grew more quickly and had larger size, which brought more supplementary liquid after bubble departure, resulting in a rise of heat transfer coefficient and higher CHF value. Visualization data indicated that bubbles behaviors on plain surface and composite porous surface differed in each region, explaining their discrepancies in boiling heat transfer characteristics properly. The composite surfaces enhanced boiling heat transfer in low heat flux range and high heat flux range. More work is required to improve the heat transfer performance in the heat flux range of 10–60 W/cm2. 5. Conclusions In this work, we investigated the nucleate pool boiling from composite porous surfaces using de-ionized water as working fluid. Through heat transfer measurements and visualization research, the enhanced boiling performance and mechanism were studied. The following conclusions can be drawn. Composite porous surface with macro pores above 0.2 mm diameter, micro pores around 2 lm diameter and dendritic structure around 400 nm diameter are prepared by an electrochemical method. The composite porous layers are of high porosity and super-hydrophilic characteristic. Contact angle of de-ionized water against the porous surface decreased with the increase of the thickness (35°, 8° and 0°). The experiment indicates that the micro pores and dendritic structure increases nucleation site density and capillary force, meanwhile, the macro pores contribute to the growing and departing of the bubbles. The critical heat flux (CHF) increased with increasing surface wettability and coating thickness, while the effect of the surface wettability on nucleate boiling heat transfer was complicated. The highest critical heat flux of the porous surfaces reaches up to 239 W/cm2, 101% higher than that of plain surface. The highest heat transfer coefficient of the porous surface is in excess of 5.9 W/(cm2 K), 120% higher than that of plain surface. Due to higher density of the active site for bubble nucleation and higher bubble departure frequency, the composite porous surfaces enhance heat transfer in low heat flux. During high heat flux range, the coalescent bubble on the composite porous surfaces grows more quickly and is of larger size, which brings more liquid replenishment after bubble departure, resulting in continuous rise of heat transfer coefficient and higher CHF value. Conflict of interest None declared.

Acknowledgment This work is sponsored by the National Science Foundation of China (Grant no. 51225602). References [1] H. Kubo, H. Takamatsu, H. Honda, Effects of size and number density of microreentrant cavities on boiling heat transfer from a silicon chip immersed in degassed and gas-dissolved FC-72, J. Enhanc. Heat Transfer 6 (2) (1999) 151– 160. [2] J.H. Kim, K.N. Rainey, S.M. You, J.Y. Pak, Mechanism of nucleate boiling heat transfer enhancement from microporous surfaces in saturated FC-72, ASME J. Heat Transfer 124 (3) (2002) 500–506. [3] C. Li, G.P. Peterson, Y. Wang, Evaporation/boiling in thin capillary wicks (l) – wick thickness effects, ASME J. Heat Transfer 128 (12) (2006) 1312–1319. [4] S. Mori, K. Okuyama, Enhancement of the critical heat flux in saturated pool boiling using honeycomb porous media, Int. J. Multiphase Flow 35 (10) (2009) 946–951. [5] S.G. Kandlikar, A theoretical model to predict pool boiling CHF incorporating effects of contact angle and orientation, ASME J. Heat Transfer 123 (6) (2001) 1071–1079. [6] B. June Zhang, K.J. Kim, Effect of liquid uptake on critical heat flux utilizing a three dimensional, interconnected alumina nano porous surfaces, Appl. Phys. Lett. 101 (2012) 054104. [7] S. Kim, H.D. Kim, H. Kim, H.S. Ahn, H. Jo, J. Kim, M.H. Kim, Effects of nano-fluid and surfaces with nano structure on the increase of CHF, Exp. Therm. Fluid Sci. 34 (4) (2010) 487–495. [8] M. Lu, R. Chen, V. Srinivasan, V.P. Carey, A. Majumdar, Critical heat flux of pool boiling on Si nanowire array-coated surfaces, Int. J. Heat Mass Transfer 54 (25– 26) (2011) 5359–5367. [9] R. Chen, M. Lu, V. Srinivasan, Z. Wang, H.H. Cho, A. Majumdar, Nanowires for enhanced boiling heat transfer, Nano Lett. 9 (2) (2009) 548–553. [10] H. Shin, J. Dong, M. Liu, Nanoporous structures prepared by an electrochemical deposition process, Adv. Mater. 19 (15) (2003) 1611–1614. [11] S. Li, R. Furberg, M.S. Toprak, B. Palm, M. Muhammed, Nature-inspired boiling enhancement by novel nanostructured macroporous surfaces, Adv. Funct. Mater 18 (2008) 2210–2215. [12] N.D. Nikolic´, K.I. Popov, L.J. Pavlovic´, M.G. Pavlovic´, The effect of hydrogen codeposition on the morphology of copper electrodeposits. I – The concept of effective overpotential, J. Electroanal. Chem. 588 (2006) 88–98. [13] N.D. Nikolic´, G. Brankovic´, M.G. Pavlovic´, K.I. Popov, The effect of hydrogen codeposition on the morphology of copper electrodeposits. II – Correlation between the properties of electrolytic solutions and the quantity of evolved hydrogen, J. Electroanal. Chem. 621 (2008) 13–21. [14] H. Shin, M. Liu, Copper foam structures with highly porous nanostructured walls, Chem. Mater. 16 (2004) 5460–5464. [15] N.D. Nikolic´, K.I. Popov, Lj.J. Pavlovic´, M.G. Pavlovic´, Phenomenology of a formation of a honeycomb like structure during copper electrodeposition, J. Solid State Electrochem. 11 (2007) 667–675. [16] N.D. Nikolic´, B. Goran, K.I. Popov, Optimization of electrolytic process of formation of open and porous copper electrodes by the pulsating current (PC) regime, Mater. Chem. Phys. 125 (2011) 587–594. [17] H.S. Ahn, H.J. Jo, S.H. Kang, M.H. Kim, Effect of liquid spreading due to nano/ microstructures on the critical heat flux during pool boiling, Appl. Phys. Lett. 98 (7) (2011) 71908. [18] X. Feng, L. Feng, M. Jin, J. Zhai, L. Jiang, D. Zhu, Reversible super-hydrophobicity to super-hydrophilicity transition of aligned ZnO nanorod films, J. Am. Chem. Soc. 126 (1) (2004) 62–63. [19] M.R.S. Shirazy, S. Blais, L.G. Fréchette, Mechanism of wettability transition in copper metal foams: from superhydrophilic to hydrophobic, Appl. Surf. Sci. 258 (17) (2012) 6416–6424. [20] Y.Y. Hsu, On the size range of active nucleation cavities on a heating surface, J. Heat Transfer 84 (3) (1962) 207–213. [21] S.G. Kandlikar, Handbook of Phase Change: Boiling and Condensation, CRC Press, 1999. [22] A. Bar-Cohen, A. McNeil, Parametric effects on pool boiling critical heat flux in dielectric liquids, Proceedings of the Engineering Foundation Conference on Pool and External Flow Boiling, ASME, Santa Barbara, CA, 1992. [23] S.J. Kim, I.C. Bang, J. Buongiorno, L.W. Hu, Surface wettability change during pool boiling of nanofluids and its effect on critical heat flux, Int. J. Heat Mass Transfer 50 (19–20) (2007) 4105–4116. [24] C. Young Lee, M.M. Hossain Bhuiya, K.J. Kim, Pool boiling heat transfer with nano-porous surface, Int. J. Heat Mass Transfer 53 (19–20) (2010) 4274–4279. [25] B.J. Zhang, K.J. Kim, H. Yoon, Enhanced heat transfer performance of alumina sponge-like nano-porous structures through surface wettability control in nucleate pool boiling, Int. J. Heat Mass Transfer 55 (25–26) (2012) 7487–7498.