Thermal performance comparison of oscillating heat

International Journal of Thermal Sciences 50 (2011) 1954e1962

Contents lists available at ScienceDirect

International Journal of Thermal Sciences journal homepage: www.elsevier.com/locate/ijts

Thermal performance comparison of oscillating heat pipes with SiO2/water and Al2O3/water nanofluids Jian Qu, Huiying Wu* Key Laboratory for Power Machinery and Engineering of Ministry of Education, School of Mechanical Engineering, Shanghai Jiaotong University, Shanghai 200240, PR China

a r t i c l e i n f o

a b s t r a c t

Article history: Received 6 January 2010 Received in revised form 7 April 2011 Accepted 13 April 2011 Available online 11 May 2011

Thermal performances of two same oscillating heat pipes (OHPs) charged with SiO2/water and Al2O3/ water nanofluids, respectively, were investigated experimentally. Both the average evaporator wall temperature and the overall thermal resistance of the OHPs at different nanoparticle mass concentrations (0e0.6 wt% for silica nanofluids and 0e1.2 wt% for alumina nanofluids) and at the volume filling ratio of 50% were tested and compared. Experimental results showed that different nanofluids caused different thermal performances of OHPs. Within the experimental range, using the alumina nanofluid instead of pure water enhanced the heat transfer of the OHP (reductions in the evaporator wall temperature and thermal resistance of the OHP of about 5.6 C (or 8.7%) and 0.057 C/W (or 25.7%), respectively, were obtained), while using the silica nanofluid instead of pure water deteriorated the thermal performance of the OHPs (with the evaporator wall temperature and the thermal resistance of the OHP being increased by 3.5 C (or 5.5%) and 0.075 C/W (or 23.7%), respectively). A preliminary analysis was conducted for the different effects induced by the addition of different nanoparticles to pure water, and it was found that the change of surface condition at the evaporator and condenser due to different nanoparticle deposition behaviors was the main reason for the thermal performance improvement or deterioration of the OHPs charged with different nanofluids. Ó 2011 Elsevier Masson SAS. All rights reserved.

Keywords: Oscillating heat pipe Nanofluid Thermal performance Nanoparticle deposition Improvement/deterioration in heat transfer

1. Introduction With the great demand for dissipating increasingly higher heat fluxes from devices including, but not limited to, electronics, computer and laser, many efforts have been devoted to develop heat transfer enhancement technologies. Among them, nanofluids, engineered by dispersing metallic/nonmetallic nanometer-sized particles in conventional fluids such as water, have attracted much attention for the past decade due to their excellent thermal properties [1,2]. Experimental investigations on the convective heat transfer [3,4] and pool boiling [5] of nanofluids have demonstrated their great heat transport capability in the open thermal systems. Inspired by this, some studies on the use of nanofluids in the closed thermal systems such as heat pipes have also been performed recently [6e14]. In 2004, Tsai et al. [6] introduced water-based gold particle nanofluids into a circular meshed heat pipe. They examined the effect of structural characteristics of nanoparticles on heat pipe thermal performance and found that the thermal resistance of the

* Corresponding author. Tel.: þ86 21 34205299; fax: þ86 21 34206337. E-mail address: [email protected] (H. Wu). 1290-0729/$ e see front matter Ó 2011 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.ijthermalsci.2011.04.004

heat pipe with nanofluids was lower than that with distilled water. Subsequently, Ma et al. [7,8] conducted an experimental investigation to study the nanofluid effect on the heat transport capability in an oscillating heat pipe. Their results demonstrated that the thermal performance of the OHP was significantly improved when charged with water-based diamond nanofluid, with the thermal resistance decreased to 0.03 C/W at a power input of 336 W. A similar experiment was performed by Lin et al. [9] in an OHP charged with aqueous silver nanofluid. It was found that the average temperature difference between the evaporator and condenser, and the thermal resistance of the OHP charged with nanofluids could be decreased by 7.79 C and 0.092 C/W, respectively. Kang et al. [10] investigated the thermal performance of a micro-grooved circular heat pipe charged with the silver nanofluid, and significant reductions in the thermal resistance and the evaporator temperature of the heat pipe were detected when using the silver nanofluid instead of distilled water. In addition, the thermal efficiency of a heat pipe could also be improved after using nanofluids. Naphon et al. [11] used TiO2/R11 as working fluid in a copper tube heat pipe. They found that the heat pipe charged with the TiO2/R11 nanofluid of 0.1% nanoparticle concentration operated with the efficiency 1.4 times higher than the heat pipe charged with pure refrigerant (R11). Noie et al. [12] also

J. Qu, H. Wu / International Journal of Thermal Sciences 50 (2011) 1954e1962

Greek

Nomenclature Ae cp Db f h I k Na Nc Q R T Tc Te U

1955

2

heat transfer area at the evaporator (m ) specific heat (J kg1 K1) bubble release diameter (m) bubble release frequency (Hz) heat transfer coefficient (W m2 C1) input current (A) thermal conductivity (W C1 m1) active nucleation site density (m2) micro-cavity density (m2) heat load (W) thermal resistance ( C W1) temperature ( C) average condenser temperature ( C) average evaporator temperature ( C) input voltage (V)

found a thermal efficiency enhancement of about 16% for a twophase closed thermosyphon when using the aqueous Al2O3 nanofluid at 3% volume fraction instead of pure water. Inconsistent results, however, were also found for the nanofluid-charged heat pipes. A recent study by Khandekar et al. [13] showed that a closed two-phase thermosyphon having the water-based nanofluids (with Al2O3, CuO and Laponite clay nanoparticles of 1% mass fraction, respectively) as working fluids had inferior thermal performance than the same thermosyphon having pure water as working fluid. Similar thermal performance deterioration was also reported by Xue et al. [14] when they used an aqueous solution containing carbon nanotubes with 1% volume fraction instead of pure water in a closed thermosyphon. As reviewed above, although most heat pipes charged with nanofluids had higher thermal performances, there are still discrepancies. To give an insight into the heat transfer improvement/deterioration mechanism caused by different nanofluids, an experimental investigation has been performed in this paper to compare and analyze the thermal performance of the oscillating heat pipe having same geometric parameters but charged with different aqueous oxide (SiO2 and Al2O3) nanofluids. Here, the oscillating heat pipe is chosen due to its unique wickless feature [15e17] and great potential for electronic cooling applications [18e20]. Compared with the water-charged OHP, both positive and negative effects on the thermal performance are found for the OHP charged with aqueous oxide nanofluids. To explain the experimental results, both the variation of thermal properties owing to the addition of nanoparticles and the change of surface condition due to the nanoparticle deposition are analyzed. It is found that the change of surface condition at the evaporator and condenser due to different nanoparticle deposition behaviors is mainly responsible for the enhanced/degraded heat transfer in the aqueous oxide nanofluids-charged OHP. The results presented help us understand and design more efficient nanofluid-charged heat pipes. 2. Description of experiment 2.1. Preparation of nanofluids In this study, two kinds of spherical oxide nanoparticles, SiO2 (Degussa Co., Germany) and Al2O3 (Nanophase Techonlogies Co., USA) with the mean diameters of 30 and 56 nm respectively, were

q m r

contact angle ( ) dynamic viscosity (kg m1 s1) density (kg m3)

Subscript c con e evp i lev n sat w wall

condenser section condensation evaporator section evaporation/boiling serial number of wall thermocouple liquid and vapor two-phase flow nanofluid saturation water or wall heat pipe inner wall

chosen as source materials, and pure water was used as the base fluid. The water-based nanofluids were synthesized by the twostep method [21]. No surfactant/dispersant additives were added during the synthesis process because they may affect the thermophysical properties and even deteriorate the boiling heat transfer of the dispersed fluid due to the agglutination phenomenon on the heated surface [22,23]. Instead, the electrostatic stabilization method was adopted here to avoid the agglomeration of nanoparticles [24]. For this purpose, the pH value of water was firstly adjusted to a certain value (with pH ¼ 9.7 and 4.9 for the silica and alumina nanofluids, respectively) which was far away from the corresponding iso-electric point (IEP) of silica (with pH w 3) or alumina (with pH w 9), and then nanoparticles were added into water. The dispersion solution was subsequently vibrated for about 4 h in an ultrasonic bath (SCQ-2210, Shengyan Ultrasonic Equipment Co.) to form the homogeneous suspension. The SiO2/water nanofluids (with the SiO2 particle mass concentration of 0.1e0.6%) and Al2O3/water nanofluids (with the Al2O3 particle mass concentration of 0.1e1.2%) prepared by the above method could be stably suspended for several days. Fig. 1(a) and (b) show the transmission electronic microscope (TEM) images of the dispersed silica and alumina nanoparticles in water, respectively. It is clear that the alumina nanoparticles were better dispersed. 2.2. Experimental setup Two sets of same oscillating heat pipes (OHPs) fabricated by bending stainless steel capillary tubes with an inner diameter of 2 mm and an outer diameter of 3 mm were prepared for the SiO2/ water and Al2O3/water nanofluids experiments, respectively. Each set of the OHP has 6 turns with the total length of 3.0 m. As illustrated in Fig. 2, the experimental setup was mainly composed of an OHP assembly, a multi-channel data acquisition system, a DC power supply (GPR-3060D, GW) and a water cold bath (DC-0506, BILANG). The OHP, consisting of evaporation, adiabatic and condensation sections with 50, 105 and 70 mm in length respectively, was vertically placed. The evaporation section (i.e., evaporator) was electrically heated, and the input voltage U and current I were measured by the digital multimeter (E2377A, HEWLETT). The condensation section (i.e., condenser) was cooled by cooling water pumped from the cold bath. Both the evaporation and adiabatic sections were well thermally insulated to minimize the heat loss from these two sections to the ambience.

1956


As shown in Fig. 2, sixteen OMEGA T-type thermocouples with a diameter of 0.1 mm and an accuracy of 0.1 C were mounted on the outer surface of the OHP, with eight (T1eT8) being placed at the evaporator and the other eight (T9eT16) being placed at the condenser. All wall temperature measurements were recorded by the data acquisition system (34970A, Agilent). Before the experiment, the OHPs were evacuated by a vacuum pump and then charged with the prepared nanofluids. For the convenience of comparison, filling ratios for both SiO2/water and Al2O3/water nanofluids-charged OHPs were set at 50% by volume. During the experiment, the cooling water with constant inlet temperature and flow rate was continuously supplied to the condenser and the power input to the evaporator was stepwise increased. Under each power input, temperature data at the evaporator and condenser were recorded after the system had reached a steady state. 2.3. Data reduction and uncertainty analysis The overall thermal resistance and the average wall temperature of the evaporator are two important parameters to describe the thermal performance of an OHP. The overall thermal resistance of an OHP is defined by:

R ¼

Te Tc Q

(1)

where Te and Tc are the average wall temperatures of the evaporator and condenser, respectively, which are determined by following equations:

Fig. 1. TEM images of (a) SiO2 nanoparticles and (b) Al2O3 nanoparticles suspended in pure water.

Te ¼

8 1X T 8 i¼1 i

(2)

Tc ¼

16 1X T 8 i¼9 i

(3)

Fig. 2. Experimental setup.


where Ti (i ¼ 1, 2, ., 16) are wall temperatures measured by thermocouples allocated on the OHP as shown in Fig. 2. Q in Eq. (1) is the heat load of an OHP which is calculated by follows:

Q ¼ UI

(4)

where U and I are the input voltage and current measured by the digital multimeter (E2377A, HEWLETT) with an accuracy of 0.3%. Strictly speaking, the heating power input at the evaporator obtained by Eq. (4) is a little larger than the net heat absorbed by cooling water at the condenser due to the heat loss. According to the analysis of heat balance between the evaporator and the condenser, the heat loss from the evaporation and adiabatic sections of the OHPs to the ambience in this study was within 4.7% of the heating power input. Thus, the uncertainty of Eq. (4) due to the heat loss is 4.7%. Table 1 lists the maximum uncertainties of main parameters in this study. The uncertainties of the direct measurement parameters such as Ti, U, and I were synthesized by the system uncertainty es from the precision of instruments (thermocouple, multimeter) and the random uncertainty er from the repeatability of data as follows:

e ¼

qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi e2s þ e2r

(5)

1957

those of pure water. Thus, the addition of silica nanoparticles to base water deteriorated the thermal performance of the OHP. 3.2. Thermal performance of Al2O3/water nanofluids-chareged OHP Fig. 4(a) and (b) show the average evaporator wall temperature and overall thermal resistance of the OHP charged with Al2O3/ water nanofluids at different concentrations (0e1.2 wt%) as a function of the power input, respectively. It is seen that (1) both the evaporator wall temperature and the overall thermal resistance of the Al2O3/water nanofluids-charged OHP decreased as compared with the pure water-charged OHP, i.e., the addition of alumina nanoparticles enhanced the heat transfer of the OHP; (2) the average evaporator wall temperature and thermal resistance decreased initially with increasing the particle concentration from 0 wt% to 0.9 wt%, but they increased as the concentration changed further from 0.9 wt% to 1.2 wt%. Thus, there was an optimal particle concentration, which was about 0.9 wt% for the Al2O3/water nanofluids-charged OHP in the present experiment. At the optimal concentration of 0.9 wt% and the power input of 82 W, reductions in the evaporator wall temperature and overall thermal resistance of the Al2O3/water nanofluids-charged OHP of about 5.6 C (or 8.7%) and 0.057 C/W (or 25.7%), respectively, were obtained as

The uncertainties of the indirect measurement parameters such as Te, Tc, Q and R were obtained with the corresponding definitions as aforementioned according to error propagation principle. Note that the uncertainty induced by the heat loss (4.7%) was also included in the uncertainty estimation of Q and R.

3. Experimental results and discussions It is clear from Eq. (1) that the overall thermal resistance R is a measure of the wall temperature uniformity along an OHP for a certain heat load. Usually, a lower thermal resistance and a lower average evaporator wall temperature mean a higher thermal performance of the OHP. In the following, we will discuss the thermal performances of the OHPs charged with different nanofluids.

3.1. Thermal performance of SiO2/water nanofluids-charged OHP The average evaporator wall temperature and overall thermal resistance of the OHP charged with SiO2/water nanofluids at different mass concentrations (0e0.6 wt%) as a function of the power input are shown in Fig. 3(a) and (b), respectively. Contrary to what had been expected, both the evaporator wall temperature and the overall thermal resistance increased after using SiO2/water nanofluids instead of pure water (i.e., nanofluid with 0 wt% concentration). Moreover, this increase became more obvious with increasing concentrations of SiO2 nanoparticles from 0.1 wt% to 0.6 wt%. At the concentration of 0.6 wt%, the evaporator wall temperature and overall thermal resistance were maximally increased by 3.5 C (or 5.5%) and 0.075 C/W (or 23.7%), respectively, as compared with Table 1 Maximum uncertainties of main parameters. Parameters

Maximum uncertainties (%)

Te Tc U I Q R

0.2 0.4 0.5 0.5 5.7 6.1

Fig. 3. Thermal performance of the OHP charged with SiO2/water nanofluids at different mass concentrations: (a) average evaporator wall temperature; (b) overall thermal resistance.

1958


Fig. 5. Thermal resistance model of an OHP.

thermal resistances due to the two-phase oscillation flow along the heat pipe, Rlev, the evaporation/boiling at the evaporator, Revp, as well as the condensation at the condenser, Rcon, in Eq. (6) are the major factors affecting the thermal performance of the nanofluidscharged OHP. In the following, we will focus attention on the effects of nanofluids on Rlev, Revp, and Rcon.

Fig. 4. Thermal performance of the OHP charged with Al2O3/water nanofluids at different mass concentrations: (a) average evaporator wall temperature; (b) overall thermal resistance.

compared with the water-charged OHP. Thus, the addition of alumina nanoparticles to base water improved the thermal performance of the OHP. 3.3. Explanation for the different thermal performances of OHPs charged with different nanofluids As indicated above, the addition of different nanoparticles to base water caused different thermal performances of the OHPs. To account for this phenomena, the thermal resistance model of the OHP is presented in Fig. 5. It is seen that the overall thermal resistance of an OHP between the heat source and the heat sink consists of the conductive thermal resistance in the pipe wall Rwall, the thermal resistance of the two-phase oscillation flow between the evaporator and the condenser Rlev, and the thermal resistances due to evaporation/boiling and condensation at the evaporator and condenser (Revp and Rcon), respectively, i.e.,

R ¼ 2Rwall þ Rlev þ Revp þ Rcon

(6)

Generally, the conductive thermal resistance of a metallic OHP, Rwall, is small and independent of the working fluid used. Thus, the

3.3.1. Thermal resistance Rlev At present, the study on the nanofluid two-phase oscillation flow in an OHP is rather limited. However considering that phase change in the OHP occurred mainly at the evaporator and condenser, it is likely that the convection heat transfer mechanism dominated the two-phase oscillation flow in the OHP. Thus the main thermophysical parameters concerning the convection heat transfer, such as the fluid thermal conductivity, heat capacity and dynamic viscosity, will influence the thermal resistance Rlev due to the twophase oscillation flow. The nanoparticle suspension is suggested to influence the heat transfer of two-phase oscillation flow due to the following reasons: (1) the presence of nanoparticles in water increases the thermal conductivity (k) of the working fluid. The nanoparticles can absorb liquid molecules, and cause the formation of molecular nanolayer on the surface of nanoparticles [25,26], which has a thermal conductivity higher than that of the bulk liquid and can intensify the heat transfer in the interior of the fluid; (2) the addition of nanoparticles to water changes the heat capacity (rCp) of the working fluid. For the nanofluids having a higher heat capacity than the base water, more heat will be transferred by the fluid if the flow rate remains unchanged based on the convection heat transfer theory; (3) the dynamic viscosity (m) of the fluid is also varied due to the addition of nanoparticles. For a nanofluid having a higher viscosity than base water, the increased viscosity means lower flow rate and thus less heat can be transferred by the fluid in the OHP; (4) in addition to the above thermophysical parameters, nanoparticles’ migration motion and its induced microconvection in the aqueous suspension were also assumed to enhance the convective heat transfer of the working fluid [27,28]. However, the variation of Rlev due to the above factors is small because of the relatively small mass concentration of nanoparticles used in the present experiment. Table 2 gives the ratios of the thermophysical parameters of the SiO2/water and Al2O3/water nanofluids to those of pure water based on Ref [29e31] at 60 C. It is seen that the variations in the thermal conductivity (k), heat capacity (rcp), dynamic viscosity (m) of nanofluids as compared with pure water are less than 1.1%, 0.12%, 0.85%, respectively, which are too small to induce detectable variations in Rlev. Moreover, according to the recent study by Eapen et al. [32], microconvection in nanofluid induced by nanoparticles’ migration and Brownian motion will not lead to an obvious increase of the convection heat transfer coefficient of the working fluid due to the relatively small concentration of nanoparticles used in this study. Thus, the change of Rlev due to the presence of nanoparticles is negligibly small, and it neither supports the apparent decrease of the thermal resistance in the Al2O3/water nanofluids-charged OHP (see Fig. 4(b)), nor explains the obvious increase of the overall


1959

Table 2 Ratios of thermophysical parameters of nanofluids to those of pure water calculated at 60 C. Nanofluids

Particle mass concentration (wt%)

Thermal conductivity ratio (kn/kw) [29]

Heat capacity ratio [(rcp)n/(rcp)w] [30]

Dynamic viscosity ratio (mn/mw) [31]

(krcp)n/(krcp)w

SiO2/water Al2O3/water

0.1e0.6 0.1e1.2

1.000e1.002 1.001e1.011

0.9998e0.9988 1.00001e1.00006

1.0011e1.0067 1.0007e1.0085

0.9998e1.0008 1.001e1.010

thermal resistance for the SiO2/water nanofluids-charged OHP (see Fig. 3(b)). Therefore, the change of Rlev is not the predominant factor for the improvement or deterioration in the thermal performance of the nanofluids-charged OHP. 3.3.2. Thermal resistances Revp and Rcon Thermal resistances Revp and Rcon in Eq. (6) due to evaporation/ boiling and condensation at the evaporator and condenser, respectively, are greatly influenced by the inner surface condition of the OHP. In order to analyze the effects of nanoparticle deposition on the surface condition, representative wall samples from the evaporator and condenser after the nanofluid experiment, as well as a clean sample boiled in pure water were examined by a scanning electronic microscope (SEM). Fig. 6 shows the SEM images of the surface conditions of these samples. It is seen that: (1) nanoparticle depositions occurred in both SiO2/water and Al2O3/water nanofluids-charged OHPs; (2) as compared with the Al2O3/water

nanofluids-charged OHP, more serious depositions with larger particle agglomerates appeared at the evaporator and condenser in the SiO2/water nanofluids-charged OHP; (3) the deposition of Al2O3 nanoparticles at the condenser (shown in Fig. 6(e)) was negligible as compared with that at the evaporator (shown in Fig. 6(d)). In the following, we will focus our attention on the surface condition effects on Revp and Rcon. The thermal resistance Revp due to evaporation/boiling at the evaporator is defined as

Revp ¼

1 hAe

(7)

where h is the evaporation/boiling heat transfer coefficient, and Ae is the heat transfer area at the evaporator. Considering that the nucleate boiling prevails at the evaporator of the OHP, the thermal resistance due to boiling at the evaporator can be expressed by [33]:

Fig. 6. SEM images of clean surface boiled in pure water (a), silica nanoparticles-deposited surfaces at the evaporator (b) and condenser (c), and alumina nanoparticles-deposited surfaces at the evaporator (d) and condenser (e).

1960


Fig. 7. Static contact angles of 3-mL sessile droplet on different surfaces: (a) pure water droplet on clean surface boiled in pure water, q ¼ 65 ; (b) SiO2/water nanofluid droplet on silica nanoparticles-deposited surface at the evaporator, q ¼ 29 ; and (c) Al2O3/ water nanofluid droplet on alumina nanoparticles-deposited surface at the evaporator, q ¼ 54 .

Revp ¼

1 1 pffiffiffipffiffiffiffiffiffiffiffiffiffiffiffiffi ¼ 2 hAe 2Na Db f pkrcp Ae

(8)

where Na, Db, f, k, r, cp are the surface active nucleation site density, bubble release diameter, bubble release frequency, liquid thermal conductivity, density and specific heat, respectively. Considering that the variation of product (krcp) due to the addition of nanoparticles is less than 1.1% (see Table 2), and the increase of surface heat transfer area at the evaporator (Ae) due to the deposition of nanoparticles is less than 2.1% (according to the surface morphology analysis of Fig. 6(a), (b) and (d)), Eq. (8) can be further simplified as:

Revp f

1 Na D2b

pffiffiffi f

(9)

Generally, the bubble generation is related to the surface roughness and wettability. Wang and Dhir [34] correlated their experimental data to give the expression of surface active nucleation site density as follows,

Na fNc ð1 cosqÞðTw Tsat Þ6

(10)

where Nc is the micro-cavity density on the surface, q is the surface contact angle, and (TweTsat) is the wall superheat. It is clear from Eq. (10) that decreasing contact angle q (or increasing surface wettability) tends to decrease the active nucleation site density and deteriorate the boiling heat transfer at the evaporator. To characterize the surface wettability, the apparent static contact angles of 3 mL sessile droplets of pure water, SiO2/ water nanofluid and Al2O3/water nanofluid on the clean, silica and alumina nanoparticle-deposited surfaces shown by Fig. 6(a), (b) and (d), respectively, were measured by an optical contact angle measuring device (OCA20, Dataphysics, uncertainty 0.1 ). Fig. 7 shows that the surface contact angle at the evaporator decreased from 65 for the clean surface to 29 and 54 for the silica and alumina nanoparticle-deposited surfaces, respectively. The fact that the surface contact angle at the evaporator decreased (i.e., surface

wettability increased) after the deposition of nanoparticles partly accounts for the deterioration of thermal performance of the SiO2 nanofluids-charged OHP in Section 3.1, but it does not explain the improvement of thermal performance of the Al2O3 nanofluidscharged OHP in Section 3.2. Thus the factor relating to the surface roughness should be also taken into account. Fig. 8(a)e(c) give the 2D and 3D atomic force microscope (AFM) images of the surface morphology of the clean, silica and alumina nanoparticle-deposited surfaces at the evaporator, respectively. It is clear that the nanoparticle deposition also changed the surface roughness condition. For the silica nanoparticle agglomerates shown in Fig. 8(b), their average size was about 2e3 mm, which was on the same magnitude of the clean surface cavities in Fig. 8(a) (the arithmetical mean roughness Ra and root-mean-square roughness Rq of the clean surface in Fig. 8(a) measured by an optical profiler (Wyko NT8000, Veeco) were 2.39 mm and 3.07 mm, respectively). As a result, the silica particle deposition on the evaporator surface led to the decrease of micro-cavity density (i.e., Nc in Eq. (10) decreased) by cramming the particle agglomerates into the surface cavities and thus deteriorated the nucleate boiling heat transfer based on Eqs. (9) and (10). Meanwhile, a porous layer constructed by the silica particles was formed and created an extra thermal resistance between the liquid and the inner surface of the evaporator. Thus, the thermal resistance Revp and the evaporator wall temperature Te of the SiO2/water nanofluids-charged OHP were increased due to the combination effects of the decreased nucleation sites, decreased contact angle and thicken porous layer. The deposition of silica particles not only changed the thermal resistance Revp, but also altered the thermal resistance Rcon. Fig. 6(c) shows the silica nanoparticle deposition also occurred at the condenser. Since the deposition of silica nanoparticles at the condenser increased the surface wettability which had a negative effect on the condensation heat transfer [35], the thermal resistance Rcon at the condenser was also increased after the deposition of silica nanoparticles. Thus, the total increase of Revp and Rcon caused the increase of overall thermal resistance R according to Eq. (6), and finally caused the thermal performance deterioration of the SiO2/water nanofluids-charged OHP as shown in Fig. 3. Moreover, this deterioration in thermal performance was more obvious at larger particle concentrations than that at lower concentrations. Contrary to the negative effect induced by the deposition of silica nanoparticle agglomerates, the deposition of alumina nanoparticle agglomerates had positive effect on the thermal performance of the OHP. Fig. 8(c) shows that the size of alumina nanoparticle agglomerates deposited on the boiling surface was several ten to hundred nano-meters (measured by the AFM), which were one to two orders of magnitude smaller than the cavities of clean surface. As a result, when the smaller alumina nanoparticles deposited on the clean surface, they created more new active nucleation sites by splitting a single nucleation site into multiple ones (i.e., Na is increased) [36], and thus enhanced the boiling heat transfer. In addition, the irregular nanopores formed between the deposited alumina nanoparticles which created the nanoroughness within the micrometer-roughened surface would affect the bubble release frequency and diameter. Within the cavities through the nanopore network, the nanobubbles may be continuously generated [37,38] and feed the nucleation and growth of larger bubbles at the microscale cavities [39]. This enables stable nucleation with increasing bubble release frequency f at lower wall superheat and decreases the evaporator temperature as shown in Fig. 4(a). Although the bubbles with smaller size Db may be created on the nano-microroughened hierarchical surface which plays a negative role in reducing thermal resistance as seen from Eq. (9), the dramatic increase of the active nucleation site density Na and the release frequency f will overwhelmingly intensify the boiling


1961

Fig. 8. 2D and 3D AFM images of surface morphology for (a) clean surface boiled in pure water, (b) silica nanoparticles-deposited surface at the evaporator, and (c) alumina nanoparticles-deposited surface at the evaporator.

heat transfer and decrease the thermal resistance Revp at the evaporator. As to the thermal resistance Rcon, since the deposition of the alumina nanoparticles at the condenser was negligible as seen from Fig. 6(e), the thermal resistance Rcon due to condensation is nearly the same for the Al2O3 nanofluid and pure water-charged OHPs. Thus, the decrease of Revp is fundamentally responsible for the thermal performance improvement of the Al2O3 nanofluidscharged OHP. Note that the alumina nanoparticle agglomerates will be intensified with the increase in the particle concentration. If the alumina

particle agglomerates grow to the sizes which are on the same magnitude of the clean surface cavities, the nucleation sites will decrease from a peak, and then degrade the boiling performance. Consequently, there existed an optimal concentration of the alumina nanoparticles for the maximal thermal performance as demonstrated in Fig. 4. According to the above discussions, Table 3 is presented to summarize the main reasons for the improvement or deterioration in the thermal performance of OHPs charged with different nanofluids. It is seen that for the SiO2/water nanofluids-charged OHP,

Table 3 Variations of thermal resistances of nanofluids-charged OHPs due to different factors. Different factors

SiO2/water nanofluids-charged OHP

Al2O3/water nanofluids-charged OHP

Addition of nanoparticles to water Addition of nanoparticles to water, fluid oscillation Deposition of nanoparticles at the evaporator Deposition of nanoparticles at the condenser Total effect

(Rwall) e (Rlev) w (Revp) [ (Rcon) [ (R ¼ 2Rwall þ Rlev þ Revp þ Rcon) [

(Rwall) e (Rlev) w (Revp) Y (Rcon) e (R ¼ 2Rwall þ Rlev þ Revp þ Rcon) Y

Note that symbols “Y, [, e, w” in Table 3 represent “increase, decrease, no variation and minor variation, respectively”.

1962


both increases of Revp and Rcon due to the deposition of silica nanoparticles resulted in the increase of overall thermal resistance and the deterioration in the thermal performance. As for the Al2O3/ water nanofluids-charged OHP, the deposition of alumina nanoparticles induced the decrease of Revp, which is the predominant factor for the decrease of overall thermal resistance and the improvement in the thermal performance of the Al2O3/water nanofluids-charged OHP. 4. Conclusions An experimental investigation has been carried out to compare and analyze the different thermal performances of the oscillating heat pipes, which were charged with SiO2/water and Al2O3/water nanofluids, respectively. The following conclusions are obtained from the present work: (1) Different nanofluids caused different thermal performances of the OHPs. The thermal performance of the OHP was improved when charged with Al2O3/water nanofluids instead of pure water, while it was deteriorated when charged with SiO2/water nanofluids. (2) For the alumina nanofluids-charged OHP, there existed an optimal concentration of 0.9 wt%, at which reductions in the overall thermal resistance and the evaporator wall temperature of about 0.057 C/W (or 25.7%) and 5.6 C (or 8.7%), respectively, were obtained as compared with pure water. For the silica nanofluids-charged OHP, the overall thermal resistance and the evaporator wall temperature increased with the increase in the mass concentration of silica nanoparticles. (3) The change of surface condition at the evaporator and condenser due to different nanoparticle deposition behaviors accounts for the different thermal performances of OHPs. For the alumina nanofluids-charged OHP, the deposition of nanoparticles (mostly occurred at the evaporator) increased the surface nucleation sites and thus enhanced the heat transfer of the OHP. While for the silica nanofluids-charged OHP, the deposition of nanoparticles (occurred at both evaporator and condenser) decreased the surface nucleation sites and contact angle, and thus deteriorated the heat transfer of the OHP. Acknowledgement This work was supported by the National Natural Science Foundation of China through Grant No. 50925624, and the Shanghai Municipal Education Commission through Grant No. 08GG05. References [1] S.K. Das, S.U.S. Choi, H. Patel, Heat transfer in nanofluids- a review, Heat Transfer Eng. 27 (2006) 3e19. [2] X. Wang, A.S. Mujumdar, Heat transfer characteristics of nanofluids: a review, Int. J. Therm. Sci. 46 (2007) 1e19. [3] Y. Xuan, Q. Li, Investigation on convective heat transfer and flow features of nanofluids, J. Heat Transfer 125 (2003) 151e155. [4] S. Zeinali Heris, M. Nasr Esfahany, S.Gh. Etemad, Experimental investigation of convective heat transfer of Al2O3/water nanofluid in circular tube, Int. J. Heat Fluid Flow 28 (2007) 203e210. [5] L. Cheng, E.P. Filho, J.R. Thome, Nanofluid two-phase flow and thermal physics: a new research frontier of nanotechnology and its challenges, J. Nanosci. Nanotech. 8 (2008) 3315e3332. [6] C.Y. Tsai, H.T. Chien, B. Chan, P.H. Chen, P.P. Ding, T.Y. Luh, Effect of structural character of gold nano-particles in nanofluid on heat pipe thermal performance, Mat. Lett. 58 (2004) 1461e1465. [7] H.B. Ma, C. Wilson, B. Borgmeyer, K. Park, Q. Yu, S.U.S. Choi, M. Tirumala, Effect of nanofluid on the heat transport capability in an oscillating heat pipe, Appl. Phys. Lett. 88 (2006) 143116.

[8] H.B. Ma, C. Wilson, Q. Yu, K. Park, S.U.S. Choi, M. Tirumala, An experimental investigation of heat transport capability in a nano-fluid oscillating heat pipe, J. Heat Trans. 128 (2006) 1213e1216. [9] Y.H. Lin, S.W. Kang, H.L. Chen, Effect of silver nano-fluid on pulsating heat pipe thermal performance, Appl. Therm. Eng. 28 (2008) 1312e1317. [10] S.W. Kang, W.C. Wei, S.H. Tsai, S.Y. Yang, Experimental investigation of silver nano-fluid on heat pipe thermal performance, Appl. Therm. Eng. 26 (2006) 2377e2382. [11] P. Naphon, D. Thongkum, P. Assadamongkol, Heat pipe efficiency enhancement with refrigerant-nanoparticles mixtures, Energ. Convers. Manage. 20 (2009) 772e776. [12] S.H. Noie, S. Zeinali Heris, M. Kahani, S.M. Nowee, Heat transfer enhancement using Al2O3/water nanofluid in a two-phase closed thermosyphon, Int. J. Heat Fluid Flow 30 (2009) 700e705. [13] S. Khandekar, Y.M. Joshi, B. Mehta, Thermal performance of closed two-phase thermosyphon using nanofluids, Int. J. Therm. Sci. 47 (2008) 659e667. [14] H.S. Xue, J.R. Fan, Y.C. Hu, R.H. Hong, K.F. Cen, The interface effect of carbon nanotube suspension on the thermal performance of a two-phase closed thermosyphon, J. Appl. Phys. 100 (2006) 104909. [15] S. Khandekar, M. Groll, An insight into thermo-hydrodynamic coupling in closed loop pulsating heat pipes, Int. J. Therm. Sci. 43 (2004) 13e20. [16] J.L. Xu, Y.X. Li, T.N. Wong, High speed flow visualization of a closed loop pulsating heat pipe, Int. J. Heat Mass Transfer 48 (2005) 3338e3351. [17] J. Qu, H. Wu, P. Cheng, X. Wang, Non-linear analysis of temperature oscillations in a closed-loop pulsating heat pipe, Int. J. Heat Mass Transfer 52 (2009) 3481e3489. [18] J. Qu, H. Wu, P. Cheng, Experimental study on thermal performance of a silicon-based micro pulsating heat pipe, Proceedings of MNHMT 2009, Shanghai, China, 2009, 18525. [19] H. Yang, S. Khandekar, M. Groll, Performance characteristics of pulsating heat pipes as integral thermal spreaders, Int. J. Therm. Sci. 48 (2009) 815e824. [20] H.B. Ma, B. Borgmeyer, P. Cheng, Y. Zhang, Heat transport capability in an oscillating heat pipe, J. Heat Transfer 130 (2008) 181501. [21] C.H. Li, G.P. Peterson, The effect of particle size on the effective thermal conductivity of Al2O3-water nanofluids, J. Appl. Phys. 101 (2007) 044312. [22] M. Chopkar, A.K. Das, I. Manna, P.K. Das, Pool boiling heat transfer characteristics of ZrO2-water nanofluids from a flat surface in a pool, Heat Mass Transfer 44 (2008) 999e1004. [23] Z. Liu, L. Liao, Sorption and agglutination phenomena of nanofluids on a plain heating surface during pool boiling, Int. J. Heat Mass Transfer 51 (2008) 2593e2602. [24] X.F. Li, D.S. Zhu, X.J. Wang, N. Wang, J.W. Gao, H. Li, Thermal conductivity enhancement dependent pH and chemical surfactant for Cu-H2O nanofluids, Thermochim. Acta 469 (2008) 98e103. [25] C.J. Yu, A.G. Richter, A. Datta, M.K. Durbin, P. Dutta, Molecular layering in a liquid on a solid substrate: an X-ray reflectivity study, Physica B 283 (2000) 27e31. [26] P. Keblinski, S.R. Phillpot, S.U.S. Choi, J.A. Eastman, Mechanisms of heat flow in suspensions of nano-sized particles (nanofluids), Int. J. Heat Mass Transfer 45 (2002) 855e863. [27] D. Wen, Y. Ding, Effect of particle migration on heat transfer in suspensions of nanoparticles flowing through minichannels, Microfluidic. Nanofluidic. 1 (2005) 183e189. [28] H.E. Patel, T. Sundararajan, T. Pradeep, A. Dasgupta, N. Dasgupta, S.K. Das, A micro-convection model for thermal conductivity of nanofluids, PRAMANAJ. Phys. 65 (2005) 863e869. [29] W. Yu, S.U.S. Choi, The role of interfacial layers in the enhanced thermal conductivity of nanofluids: a renovated Maxwell model, J. Nanopart. Res. 5 (2003) 167e171. [30] J. Buongiorno, Convective transport in nanofluids, J. Heat Transfer 128 (2006) 240e249. [31] H.C. Brinkman, The viscosity of concentrated suspensions and solution, J. Chem. Phys. 20 (1952) 571e581. [32] J. Eapen, W.C. Williams, J. Buongiorno, L. Hu, S. Yip, Mean-field versus microconvection effects in nanofluid thermal conduction, Phys. Rev. Lett. 99 (2007) 095901. [33] B.B. Mikic, W.M. Rohsenow, A new correlation of pool-boiling data including the effect of heating surface characteristics, J. Heat Transfer 91 (1969) 245e250. [34] C.H. Wang, V.K. Dhir, Effect of surface wettability on active nucleation site density during pool boiling of water on a vertical surface, J. Heat Transfer 115 (1993) 659e669. [35] S. Pulipaka, S.Y. Son, A study of effect of wettabllity on heterogeneous condensation-varied substrates, Proceedings of IMECE 2007, Seattle, Washington, 2007, 42451. [36] G.P. Narayan, K.B. Anoop, S.K. Das, Mechanism of enhancement/deterioration of boiling heat transfer using stable nanoparticle suspensions over vertical tubes, J. Appl. Phys. 102 (2007) 074317. [37] X.H. Zhang, X.D. Zhang, S.T. Lou, Z.X. Zhang, J.L. Sun, J. Hu, Degassing and temperature effects on the formation of nanobubbles at the mica/water interface, Langmuir 20 (2004) 3813e3815. [38] R.E. Cavicchi, Bubble nucleation and growth anomaly for a hydrophilic microheater attributed to metastable nanobubbles, Phys. Rev. Lett. 98 (2007) 124501. [39] C. Li, Z. Wang, P. Wang, Y. Peles, N. Koratkar, G.P. Peterson, Nanostructured copper interfaces for enhanced boiling, Small 4 (2008) 1084e1088.