- Email: [email protected]
Pool boiling heat transfer to dilute copper oxide aqueous nanofluids
International Journal of Thermal Sciences 90 (2015) 224e237
Contents lists available at ScienceDirect
International Journal of Thermal Sciences journal homepage: www.elsevier.com/locate/ijts
Pool boiling heat transfer to dilute copper oxide aqueous nanofluids M.M. Sarafraz*, F. Hormozi 1 School of Chemical, Petroleum and Gas Engineering, Semnan University, Semnan, Iran
a r t i c l e i n f o
a b s t r a c t
Article history: Received 18 March 2014 Received in revised form 6 December 2014 Accepted 7 December 2014 Available online
A set of experiments have been performed to quantify the pool boiling heat transfer coefficient of dilute copper oxide water-based nanofluids at mass concentrations of 0.1e0.4%. To stabilize the two-step nanofluids, pH control, stirring and sonication were utilized. For investigating the influence of surfactant as a surface active agent additive on the pool boiling heat transfer coefficient of nanofluids, SDS, SDBS and Triton X-100 were used. Influence of some operating parameters such as applied heat flux, mass concentration of nanofluids and other parameters such as roughness of surface, boiling contact angle and deposition on the pool boiling heat transfer coefficient of nanofluids were experimentally investigated. Results demonstrated a significant deterioration of heat transfer coefficient of nanofluids comparing with the base fluid in the absence of surfactants, however, in the presence of surfactant, higher pool boiling heat transfer coefficient was reported. According to results, roughness of surface is strongly controlled by nanofluid concentration due to deposition of nanofluids on the heating section. Rectilinear changes of deposition with time in term of fouling resistance were seen at both regions with natural convection and nucleate boiling dominant heat transfer mechanisms. © 2014 Elsevier Masson SAS. All rights reserved.
Keywords: Pool boiling Surfactant Nanofluid Bubble formation Roughness Scale formation
1. Introduction Boiling heat transfer is one of the major interests of heat transfer experts, not because of the complexity of boiling phenomena, but also its utilizations in heating/cooling systems and industrial processes and its applications in refrigeration, power generation, heat exchangers, cooling of high-power electronics components and PWR nuclear reactors. Removing a large quantities of heat in the least possible compact size has always been a key point for designing the heat exchanging media. Thus, enhancing the heat transfer coefficient and thermal efficiency of above-mentioned systems is indispensable and vital. Pool boiling involves all the interactive, but complicated, and dynamic processes such as hydrodynamics, heat and mass transfer (particularly in multicomponent mixtures) and sub-phenomena such as: nucleation, bubble coalescence, and collapse of bubbles. Many efforts have been made to experimentally investigate on the pool boiling heat transfer of different fluids including: pure liquids, multicomponent mixtures, refrigerants and none-Newtonian fluids.
* Corresponding author. Tel.: þ98 9166317313; fax: þ98 6324231683. E-mail addresses: [email protected] (M.M. Sarafraz), [email protected] (F. Hormozi). 1 Tel.: þ98 9123930495. http://dx.doi.org/10.1016/j.ijthermalsci.2014.12.014 1290-0729/© 2014 Elsevier Masson SAS. All rights reserved.
However, a simple statistical search among the published documents reveals this fact that nanofluid-related documents are less in number when comparing to other pool boiling subjects. As can be seen in Fig. 1, more researches should be conducted in context of understanding the heat transfer characteristics of nanofluids. A brief literature review on the pool boiling heat transfer of nanofluids has been represented in the following section. 2. Literature review For the first time in 1995, Choi introduced the concept of nanofluid in his investigations [1]. He provided results of a theoretical study of suspended copper nanoparticles in a base fluid and demonstrated significant improvement in thermal properties of the test colloid fluid. Further experimental investigations have reported that suspensions containing nanoparticles have substantially higher thermal conductivities than traditional heat transfer fluids [2e4]. Recently, Xuan summarized the main factors belonging to nanofluids that enhance heat transfer as follows [5]: (a) The nanoparticles can increase the surface area; (b) nanoparticles have higher interaction and collisions among the particles and fluid in comparison with other solid particles (e.g. microparticles); and (c) increased mixing fluctuation and turbulence of the fluid. Owing to these attributes, it is expected that the heat transfer performance of water, the most widely used coolant, can
M.M. Sarafraz, F. Hormozi / International Journal of Thermal Sciences 90 (2015) 224e237
225
Fig. 1. Proportion of conducted researches on boiling in terms of different subjects; investigations on nanofluid after several decades are still less in number; (Direct keywords in www.sciencedirect.com).
be improved [6]. Although many researchers such as Das [7] showed that the boiling heat transfer performance deteriorates when nanofluid is used as working fluid, but there are still controversial reports that nanofluids can enhance the boiling performance [8]. Lui et al. [9] conducted an experimental investigation on boiling of CuO/water nanofluids around the grooved copper block and reported the enhancement of pool boiling heat transfer coefficient close to 50%. Shi et al. [10] and Tu et al. [11] performed experiments on Al2O3 nanofluids and collected set of experimental data that when compared to base fluid, demonstrated enhancement up to 64%. Wen et al. [12] represented the significant enhancement of heat transfer coefficient of TiO2/water and Al2O3/ water nanofluids on the stainless steel cylinder with diameter of 150 mm. Witharana [13], Troung [14], Ahn et al. [15] and Sarafraz et al. [16,17] demonstrated the enhancement of pool boiling heat transfer coefficient using nanofluids on heaters with different geometrical properties. More information can be found on cited references. In contrast, many investigators demonstrated the reduction of pool boiling heat transfer coefficient in their results. Coursey and Kim [18], Bang and Chang [19] and Sajith [20] carried some experiments out to investigate the effect of Al2O3 water based nanofluids on pool boiling heat transfer coefficient. They also surveyed the influence of scale formation and particle deposition around the heating section. The outstanding point of their works was to show the deterioration of heat transfer coefficient. Kim et al. [21], conducted a research based on the boiling heat transfer of Al2O3, CuO ZrO2 and SiO2 water based nanofluid on a wire with diameter about 0.38 mm and showed the significant deterioration of heat transfer coefficient due to the scale formation. Chopkar et al. [22], Kim et al. [23], Narayan et al. [24], You et al. [25] and Vassallo et al. [26] performed studies on boiling of nanofluid and small change in heat transfer coefficient was reported. As can be seen in the above-mentioned, controversial reports on boiling of nanofluids may be found in literature. Briefly speaking, Nanofluid pool boiling literature is in conflict over whether nanoparticles can enhance or degrade boiling heat transfer. In the present work, CuO/water nanofluids are prepared using two-step method and stabilized using pH control, stirring, and sonication processes. By employing these methods, stability of nanofluids can be improved up to 1080hr. Pool boiling heat transfer coefficients of nanofluids, in the absence/presence of SDS, SDBS
surfactants at the same mass concentration (0.1%) are experimentally measured. Influence of surfactants on the bubble-surface contact angle and bubble formation rate is visually investigated and briefly discussed. Likewise, influence of particle deposition on the heater roughness is experimentally investigated and interesting results regarding to scale formation and roughness in term of fouling resistance is represented. 3. Experimental 3.1. Experimental apparatus and procedure This study focuses on the nucleate pool boiling of nanofluids on the surface of a horizontal cylindrical heater. A schematic of test facility has been shown in Fig. 2a. As can be seen in Fig. 2a, experimental setup consists of three main sections: measurement instruments, boiling test section and photographic system. The stainless steel vessel is equipped with the boiling test section and condenser. The condenser (which is constantly cooled by water/ethylene glycol cooling cycle) condenses the vapor produced by the applied heat and the condensed liquid is returned back to the bottom of the vessel for reevaporation. A pressure sensor is mounted on top of the vessel to control the pressure of test vessel equals to the atmospheric pressure throughout the experiment. The whole system is heavily insulated for more controllability and reduction of heat loss. The temperature of the liquid inside the test vessel is monitored and controlled at predetermined set point by a thermal regulator which involves eight of PT-100U thermometers and appropriate vertical auxiliary heater. These RTDs are used to measure the bulk liquid temperature during experiments mounted at different locations inside the test vessel (bottom, up, left, right, center and four around the heating section). Heating section can be considered as the most important section of the test apparatus. A horizontal stainless steel cylinder with outer diameter of 21 mm and length of 350 mm that only the first 105 mm of its length is internally heated using a bolt heater (manufactured by Cetal Co.). It is also equipped with eight K-type calibrated thermocouples, embedded along the circumference of the rod, very close to the heating surface. The arithmetic average of these eight thermocouples can be considered as the surface temperature. A PC-based
226
M.M. Sarafraz, F. Hormozi / International Journal of Thermal Sciences 90 (2015) 224e237
Fig. 2. a. A scheme of test apparatus; b. Geometrical details of cylindrical test section.
data acquisition system was used to record the measuring parameters. The power of the heating system is provided using a DC regulator. Note that, test vessel is equipped with two heaters such that the first one maintains the temperature of bulk fluid to any desired temperature and the second one is the bolt heater mounted inside the stainless steel cylinder to provide the applied heat flux for pool boiling. Casio EX-F1 was also used to record the bubble formation and nucleation phenomena. Taken videos/ photos then are processed using image and video editing software's. After analyzing conclusions on bubble-related phenomena are made. Fig. 2b shows the geometrical details of heating section by cross section view. Noticeably, because of the ultra-high speed (UHS) recording, Arri high speed ballasts lighting system (Luminys
Lab-Light 30k) was employed to minimize the effect of flicker on taken images. Prior to commencing the experiments, test section, test vessel and pipes were cleaned by methanol/water solution and dried to remove any scale from previous experiments. Once the system was cleaned, the test fluid was introduced to the test vessel and deaeration process was performed. Following this, the auxiliary heater was switched on and the temperature of the system increased. When the fluid had reached the desired temperature, then, the power was supplied to the test heater and kept at a pre-determined value. The data acquisition system was switched on and temperatures, pressure and heat fluxes were recorded. Noticeably, the experiments have been entirely performed at saturation temperature
M.M. Sarafraz, F. Hormozi / International Journal of Thermal Sciences 90 (2015) 224e237
under atmospheric pressure. Experiments were repeated three times to ensure the reproducibility of the results. 3.2. Uncertainty and data reduction To estimate the pool boiling experimental heat transfer coefficient, applied heat flux should be calculated by the product of electrical voltage, current and cosine of the difference between input electrical voltage and current using the following correlation (cosine of difference between input and electrical voltage is almost equal to zero):
q¼
V I p$D$L
(1)
where, I is the current (A), V is the voltage (V), D is tube diameter (m) and L is tube length (m). The temperature drop between the location of the wall thermocouples and the heat transfer surface was deducted from the measured temperature difference according to:
s q Ts Tb ¼ ðTth Tb Þ $ k A
(2)
In this equation, s is the distance between the thermocouple locations and the heat transfer surface and k is the thermal conductivity of the heater material. The value of s/k was determined for each thermocouple by calibration of the test heater using pure water. Since, ratio of diameter of cylinder to its length is insignificant, cylindrical effect can be ignored. The surface temperature was obtained by the arithmetic average of values measured by the eight thermocouples. Bulk temperature was also estimated using RTDs. The heat transfer coefficient h is calculated from:
a¼
q A
ðTs Tb Þave:
(3)
To compensate the temperature drop, AlaviFazel [27] proposed that instead of using Eq. (2), following ODE equation can be solved:
1 d dTs kr ¼0 r dr dr
(5)
and:
Da ¼ a
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2 2 Dq DT þ q T
is utilized for preparing the nanofluid. Briefly, these process includes: I. Weigh the desired mass of CuO with digital electronic balancer (A&D EK Series Portable Balances, EK-1200i). II. The weighted CuO nanoparticle were added into the weighted DI-water, while all were agitated in a clean flask. The magnetic motorized stirrer (Hanna instruments Co.) was employed for agitating the nanofluid. III. UP400S ultrasonic Hielscher GmbH (350 W/24 kHz) was used to disperse the nanoparticles into the water uniformly and crack the agglomerations and clogs. Thermal conductivity, specific heat capacity and density of test nanoparticles were 32.9 (W/m K), 550.5 (kJ/kg K), 6320 (kg/m3) respectively. Noticeably, nanofluids were prepared at percentage weight fractions of 0.1e0.4. Also, quality tests were performed to ensure about size, morphology and purity of nanoparticles and their mean size. Results have been represented in Fig. 3(aec). As can be seen in Fig. 3(a), XRD pattern depicts the single-phase CuO with a monoclinic structure which implies that there is no impurity other than CuO nanoparticles and no significant peaks of impurities are found in XRD pattern [29]. Also, by employing the Scherrer's equation [30], nanoparticle size was also verified. According to the Scherrer's formula:
t¼
0:9l B cos q
(7)
l is the wavelength of X-ray target (for copper nanoparticle, it is 1.54 Å) and value of 2q and B was determined by the XRD spectra. Result of Scherrer's formula (45e50 nm) confirmed the results obtained by the particle-count experiment (shown in Fig. 3b) and XRD results. As can be seen, the dominant size of particles is close to 50 nm. Fig. 3c depicts the SEM image of CuO nanoparticles. As can be seen, particles are spherical and have the same morphology and mean size.
(4)
In Eq. (4), k is the temperature dependent thermal conductivity of the heater, which was approximated to a linear function of temperature. However, in this work, by means of calibration, Eq. (2) was used. Also, a detailed uncertainty analysis performed in accordance with Kline and McClintock [28] estimated an overall uncertainty of the pool boiling heat transfer coefficient.
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2 Dq DV 2 DI 2 DA ¼ þ þ q Vmax: Imax: A
227
3.4. Stabilization of nanofluid There are several stabilization techniques for nanofluids which come as follows: 3.4.1. The electrostatic stabilization In which nanofluids can be stabilized by controlling the chemical characteristics of solution such as pH and concentration of existing ions. In this method, Zeta potential can be a gold parameter in determining the isoelectric point (IEP) and assessing the optimum pH, in which the zeta potential of solution should be far from (smaller or higher than) the isoelectric point. Noticeably, in IEP, zeta
(6)
The maximum calibration error of the thermocouple was 0.2 K. The maximum deviation related to the measuring of location of the thermocouples was about 0.5%. The measurement error of the power meter was 0.5%. Maximum uncertainties of the heat flux were also about 9%. The uncertainty of heat transfer coefficient according to Eq. (6) equals to 10.1%. Table 1 represents the uncertainty values of instruments used in this work. 3.3. Preparation of nanofluid In the present work, CuO nanoparticles (50 nm, PlasmaChem GmbH, Germany) were uniformly dispersed into the DI-water as the base fluid for making a stable nanofluid. The two-step method
Table 1 Uncertainties of the measurement instruments. Parameter
Instrument
Uncertainty
Cross section area (m) Surface temperature (K) Voltage (V)
Precise engineering calipers
0.001 m
Calibrated K-type thermocouple
±0.2 K
Keithley digital multi-meter
Current (A)
Keithley digital multi-meter
Bulk temperature (K) Contact angle (degree)
PT-100U thermo-meter (thermo-resistance) Visual observation, Sigma scan®, Adobe® Photoshop with filter
±1% of reading ±0.1% of reading ±0.1 K ±3
228
M.M. Sarafraz, F. Hormozi / International Journal of Thermal Sciences 90 (2015) 224e237
Fig. 3. a XRD pattern of CuO nanoparticles; b Results of nanoparticle size count; c SEM image of CuO nanoparticles.
potential is minimum (approximately zero) and the least stability of nanofluids can be seen in isoelectric point. 3.4.2. The steric stabilization In which the surface properties and potential is the key parameter in stabilizing of nanofluids. This parameter is controlled by adding surfactants or dispersants. With adding the surfactants/ dispersants, coalescence (cohesive behavior of particles) significantly decreases which leads the particles to be uniformly dispersed in the medium. However, it is hardly believed that surfactants can change the thermo-physical properties of solutions and reduces the reliability of experimental data. 3.4.3. The depletion stabilization In which a new free polymer is added into the medium of dispersion which changes the intermolecular forces between particles toward the lower sedimentation. Since, particles with lower size have higher surface energies; therefore, these types of nanofluids have higher potential to form the cluster and local agglomeration and aggregations. 3.4.4. Mechanical/ultrasonic agitations Which leads the agglomerations and aggregations to be cracked in smaller pieces; this technique strongly depends on the power
and frequency of sonication. Time is the other important parameter in this method. There are still other techniques such as time-experiment methods including the settling bed and time-experiment methods. Such techniques can be employed in parallel of the other methods. In most of these techniques, height of sedimentation and scales is measured. In this research, in order to stabilize the test nanofluid, pH control is employed such that by means of HCl þ NaOH buffer solution, pH of nanofluids is controlled. Peyghambarzadeh et al. [31] conducted some experiments on the stabilization of CuO nanofluids as working fluid inside the car radiator and concluded that at pH value of 10.1 the best stability of nanofluid may be obtained. However, he ignored the influence of agglomeration of nanofluids and did not consider the influence of sonication on the stability of nanofluids. Dey et al. [32] used the ultrasonic for more than 60hr for ZnO ethylene glycol nanofluids and prepared the faily stable nanofluids. In the present work, using the DLS analyzer (Malvern Zen-3600), zeta potential of CuO nanoparticles in water has experimentally been measured. Results showed that at isoelectric point of pH ¼ 8.33, zeta potential is 0 mv at maximum concentration of nanofluid (which is so-called isoelectric point or IEP) and therefore, for stabilizing the nanofluids, pH of nanofluids must be smaller or higher than 8.33 as far as possible. Experiments showed
M.M. Sarafraz, F. Hormozi / International Journal of Thermal Sciences 90 (2015) 224e237
229
possible cluster can be formed inside the nanofluid and the minimum of scale would form on the heating section. To measure the pH, the pen-type pH-meter AZ-8685A (AZ instruments Co) was used. Fig. 4 represents the pH values of nanofluid before the pHlevel control experiments.
3.5. Experimental measurement of thermo-physical properties of nanofluids
Fig. 4. pH values of nanofluids at different concentrations before PH level control.
us that at pH ¼ 10.2, the zeta potential is 5.43 and consequently, the best stable nanofluid (for about 1080hr) is achievable. Because, zeta potential at this pH has significant difference when compared to isoelectric point in which zeta potential is 0. Therefore, the least
Based on the literature, pool boiling heat transfer coefficient can be estimated by the predicting correlations which strongly depend on the thermo-physical properties of test fluid. Physical properties can also be obtained by the related correlations which sometimes can be tolerated up to 30%. On the other hand, these correlations have not been re-developed for thermo-physical properties of nanofluids. Therefore, one way is to find the thermo-physical properties of nanofluids experimentally. Thus, viscosity of nanofluids was measured using DV-II þ PRO digital viscometer (manufactured by Brookfield Co., Accuracy: ±1.0% of reading/ Repeatability: ±0.2%). Density was also measured using DMA 4500 ME (manufactured by Anton Paar Co., Accuracy: ±1.0% of reading/Repeatability: 0.00001 g/cm3). Thermal conductivity was also measured using DTC300 (manufactured by TA Instrument, Accuracy: ±3%e8% depending on thermal resistance/Repeatability:
Fig. 5. a. Density of deionized water and nanofluids at different temperatures; b. Viscosity of nanofluids and deionized water at different temperatures; c. Thermal conductivity of nanofluids at different temperatures; d. Representation of Newtonian behavior of nanofluids at different shear rate and shear stress.
230
M.M. Sarafraz, F. Hormozi / International Journal of Thermal Sciences 90 (2015) 224e237
Fig. 6. Parameters measured by the profile meter (more details and info can be found on Elcometer official site and user manual).
±1 to ±2% depending on thermal resistance/Standard: ASTM E1530). Results have been represented in Fig. 5(aed). The Newtonian behavior of nanofluid was investigated and represented in Fig. 5d. 4. Results and discussions Before any discussions, the experimental results must be validated. To ensure the accuracy of obtained data, Rohsenow [33] correlation was employed for pure water. Because: a) Rohsenow correlation has been proposed based on the pure water and has been tested for this DI-water in previous studies. b) The physical properties of DI-water are well known with high accuracy. c) Pool boiling heat transfer coefficient of distilled water had been investigated by several investigators over a wide range of heat fluxes and system pressures. This equation has been represented as follows:
2 sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi30:33 Cpl $DTe q s 4 5 ¼ Csf mhfg g rl rg hfg prl
(8)
In the correlation, Csf and n are 0.01 and 1 respectively [34]. Gorenflo [35] also has proposed a predicting correlation verified by the experimental data based on pure water and water-based nanofluids which includes the parameter related to surface roughness. Therefore, experimental setup is validated under comparisons of experimental data and those of calculated by these two well-known correlations. Likewise, to obtain the surface roughness of the cylinder, the profile meter (Elcometer-7061) was employed to measure the roughness of surface. Fig. 6 shows the parameters that can be measured by profile meter. Roughness meter results indicate that a uniform pattern of roughness can be seen along with the length of heating section. The mean roughness of surface was 0.34 mm which is quite higher than average particle size (50 nm). Results of validation for pure water demonstrated the good agreement between experimental data and well-known correlations which can be seen in Fig. 7. An overview on the experimental data related to the pool boiling heat transfer coefficient shows that boiling heat transfer coefficients in the nanofluids are significantly lower than that in pure water. There are several parameters influencing on the
Fig. 7. Comparisons of experimental data and results obtained by Rohsenow and Gorenflo correlations.
M.M. Sarafraz, F. Hormozi / International Journal of Thermal Sciences 90 (2015) 224e237
231
Fig. 8. Effect of heat flux and concentration on the pool boiling heat transfer coefficient of Cuo-water nanofluids at different concentrations.
deterioration of the pool boiling heat transfer coefficient of nanofluids, which are discussed in the forthcoming sections. 4.1. Influence of heat flux and mass concentration of nanofluid According to the experimental data, heat flux has a strong influence on the pool boiling heat transfer coefficient. It is more convenient to show the effect of heat flux in terms of heat flux versus heat transfer coefficient. Fig. 8 demonstrates the experimental data related to pool boiling heat transfer coefficient of CuO/water nanofluids at different mass concentrations of nanoparticles at various heat fluxes. Briefly speaking, the higher concentration of nanoparticles, the higher reduction of pool boiling heat transfer coefficient is registered all over the
experiments. Also by increasing the heat flux, heat transfer coefficient significantly increases. Deterioration of heat transfer coefficient can be an effect of scale formation and sedimentation of nanoparticles around the surface which increases the fouling resistance parameter and thermal resistance of surface that leads to the heat transfer to be reduced. Such depositions can reduce the number of active nucleation sites as well as creation of an insulation porous layer on the surface. It also decreases the surface roughness of heating surface area. Consequently, thermal resistance of heating section is locally increases and finally, heat transfer coefficient deteriorates. Although there are studies that have been carried out using nano-coated surfaces [36,37], but the factors responsible for the nanoparticle deposition are still unknown.
Fig. 9. Bubble formation during pool boiling of nanofluids at initial time without scale formation; note that scale formation is a strong function of time.
232
M.M. Sarafraz, F. Hormozi / International Journal of Thermal Sciences 90 (2015) 224e237
bubbles, bubbles with lower diameter size are seen. However, average number of the bubbles formed on the surface significantly increases. This fact can also be seen in Fig. 9(bef). 4.2. Influence of surfactants
Fig. 10. Boiling contact angle concept according to the Young theory.
When initial heat flux is applied, the natural convection is the dominant heat transfer mechanism around the heating section and no bubble is formed around the cylinder. By increasing the heat flux up to 15 kW/m2, the nucleate boiling becomes the dominant heat transfer mechanism and first bubbles may appear around the cylinder. These bubbles due to the lower local agitation and slower bubble interactions have enough resident time to grow and depart from the surface. Therefore, bulky and bigger bubbles is seen at lower heat fluxes (Fig. 9a). At higher heat fluxes, due to increase the number of nucleation sites, numerous bubbles are formed around the surface and due to the local agitation and collapsing of the
Small quantity of surfactant can increase pool/flow boiling heat transfer coefficient. The rate of enhancement has also been proved that to be dependent on the additive concentrations, its type and chemistry, wall heat flux, and the heater geometry. The concentrations are usually low enough that, the addition of surfactant to water causes no significant changes in saturation temperature and most other physical properties, except for viscosity and surface tension. Small concentration of surfactant additives can also reduce the solution's surface tension considerably, and its level of reduction depends on the amount and type of surfactant added into the solution. The activation of nucleate sites, bubble growth and bubble dynamics influence the boiling heat transfer coefficient. One can refer to the literature [38e41] to find information related to the effect of surfactants on boiling heat transfer and their effects on boiling performance and sub-phenomena [38]. In the present work, 0.1% by weight of Sodium dodecyl Sulfate (SDS), Sodium dodecyl benzene sulfonate (SDBS) and Triton X-100 were added and their influence on the pool boiling heat transfer coefficient of nanofluids as well as bubble formation was experimentally investigated. SDS is widely used in detergents for laundry with many cleaning applications and foaming industries. SDS is highly effective surfactant and can be used in any task requiring the removal of oily stains and painting spots or residues. Triton X-
Fig. 11. Bubble formation in pool boiling of CuO nanofluids; reduction of visual boiling contact angle by increasing the concentration of nanofluids at the same concentration of surfactants, HF ¼ 45 kW m2.
M.M. Sarafraz, F. Hormozi / International Journal of Thermal Sciences 90 (2015) 224e237
100 is a commonly used detergent in laboratories. As surfactant may change the properties of nanofluids and their stability, the lowest possible mass concentration of surfactant was added into the nanofluids. According to the visual observations, with adding the surfactants, visible nucleation site locations dramatically increased. Also, deposition of nanoparticles increases when concentration of nanofluids and surfactant concentration increases too. Bubble-surface contact angle is another key parameter in boiling phenomena which was under influence by surfactants. According to the available studies, most researchers measured the surface tensions of aqueous additive solutions, while did not consider the effect of the contact angles on the boiling phenomena. In the present work, contact angle of nanofluids are visually measured using UHS photographic system, however, Young correlation is the most accurate equation for estimating the contact angle and wettability of a surface however, we were unable to provide the facility to measure the exact values for surface tension of nanofluids. As shown in Fig. 10, in Young equation, contact angle, q, is the angle between the liquid and solid interface which has visually been measured by analyzing the images. As mentioned, contact angle is one of the most important parameters in understanding boiling phenomena since it characterizes the wettability of the solid surface by the liquid [42]. Young [43] proposed an equation (Eq. (9)) which is still used as a fundamental understanding concept for phenomena involving the surface wettability and contact angle (q).
q ¼ arccos
ssv ssl slv
233
(9)
where, ssv, ssl and slv are the surface tension at interface of solid/ vapour, solid/liquid and liquid/vapour phases respectively. Fig. 10 represents the Young correlation related to the contact angle for a single bubble. The contact angle is defined as an angle between solideliquid and liquidevapor interfaces. Noticeably, in the present work, results of contact angle are based on the visual observations and Young equation was not employed. As can be seen in Fig. 11, visual contact angle of bubbles formed on the surface decreased with increasing the concentration of nanofluids. In fact, with decreasing the boiling contact angle, wettability of surface increases. To interpret this phenomenon by the macroscopic view, it can be said that when concentration of nanofluids increases, rate of sedimentation of particles on the surface significantly increases and coating is formed which leads the wettability of surface to be increased, while roughness of surface to be decreased. However, this coatings and particle may not easily be seen by naked eye. Note that, the relative errors of the measurement of bubble radius and bubble coordinates are estimated to be no more than 3 for a typical bubble radius of 0.5 mm, by assuming a constant error of 0.5 pixels during the discretization in imaging process in Sigma scan®. Also, as can be seen in Fig. 12, average bubble diameters are lower in comparison with the case without the surfactants,
Fig. 12. Bubble formation in pool boiling of CuO/water nanofluids in absence/presence of surfactants.
234
M.M. Sarafraz, F. Hormozi / International Journal of Thermal Sciences 90 (2015) 224e237
although, bubbles some times are coalesced and form the bigger bubbles. Coalescence seems to be weakened when surfactants are added. To demonstrate the influence of surfactant on the heat transfer coefficient of CuO/water, enhancement ratio is defined as:
h¼
aPoolBoilingþSDS aPoolBoiling
(10)
Fig. 13 represents the pool boiling heat transfer coefficient of CuO/water under influence of SDS surfactant. Higher pool boiling heat transfer coefficient of CuO/water nanofluid is reported (up to 70.48% for 0.4% of SDS) for those nanofluid mixed with SDS wt. % ¼ 0.1 comparing to other surfactants. For other surfactants, similar behavior can be seen. The role of surfactant on heat transfer enhancement is out of goals of this work.
Also, time is another major parameter, which should be taken into the account in sedimentation and fouling experiments, but is out of goals of this work and currently is undergone. Noticeably, surface with decreased roughness (smoother surface) gives weaker bubble formation, because the smoother surface has usually less cavities than the rough surface, subsequently, less bubbles are formed during boiling and bubble transport becomes weaker. On the other hand, heating surface is covered by the generated bubbles and due to the vapor captured inside the bubbles, heating section is partially isolated. Hence, boiling heat transfer performance of surface significantly reduces. All in all, there are two factors having major roles in boiling deterioration of nanofluids: 1) less cavities due to deposition of nanoparticles 2) negative influence of formed deposition on the surface which increases the thermal resistance of surface and leads to the deterioration of heat transfer.
4.3. Influence of surface roughness
4.4. Deposition (fouling) thermal resistance
Undoubtedly, heating surface condition is one of the most important factors in pool boiling of nanofluids [44]. It is one of the most important factors in increasing/decreasing heat transfer rate since it is directly related to the active nucleation sites density on the heating surface [45]. There are many studies under several conditions of specially treated surface, coated surface, or polished surface and different methods that have been employed to change the heating surface conditions which come as follows: a) Roughening surface itself. b) Coating the surface using special metal, porous material or using electrophoretic methods. c) Adding specially manufactured irregularities and cavities or patterned shapes on the surface. d) Recently, using the nanofluids and sedimentation of particles on the surface. Nanoparticles, due to the instability and agitations and their unbalanced interactions between molecular forces tend to deposit on the surface. In some systems, gravity is also a significant parameter helping to increase the rate of deposition of the particles. To measure the roughness surface, profile-meter is used before and after the boiling experiments. According to results registered by the profile meter, roughness of surface decreased from 0.34 mm for clean surface to 0.29 mm, 22 mm, 0.19 mm and 0.13 mm for wt. % ¼ 1 to 0.4 respectively, when concentration of nanofluids increased.
Pool boiling heat transfer mainly consists of two distinguishable heat transfer regions. The first region where natural convection is dominant mechanism and nucleate region where nucleate boiling heat transfer is dominant heat transfer mechanism. In the present work, a set of experiments were conducted to measure the fouling resistance of CuO nanoparticles around the stainless steel cylinder. It is customary to present fouling data in terms of fouling resistance (Rf) which can be calculated on the basis of heat transfer [46]:
Fig. 13. Enhancement ratio for pool boiling heat transfer coefficient by adding SDS surfactant.
Rf ¼
1 1 aðtÞ aðt ¼ 0Þ
(11)
Fig. 14(a,b) show the changes of pool boiling heat transfer coefficient and fouling resistance over the time. As can be seen, the pool boiling heat transfer coefficient decreases with time due to the deposition of CuO nanoparticles on the heat transfer surface and also decreasing in number of nucleation sites on the heating surface; nanoparticles cover the heating surface and slightly change the thermal resistance of surface. On the other hand, according to previous studies, micro-cavities of surface reduces, while wettability of surface increases and some other nucleation sites are covered by fluid. On the other hand, bubbles cover the heating section too which lead to the heat transfer rate to be considerably decreased. Experimental results reveal a rectilinear changes in increase of fouling resistance with time in natural convection heat transfer region and nucleate boiling which both of these changes are results of nanoparticle sedimentation fouling. Walker et al. [45] reported an asymptotic behavior of fouling resistance in boiling flow which may be attributed to additional particulate fouling. Peyghambarzadeh et al. [46] represented that the fouling resistance in the initial stages of the test period decreases even hit the negative value at some points. He expressed that this phenomenon may be due to the fact that during the nucleation stage, the nuclei forming on the heat transfer surface increases the surface roughness. This leads to an increase in the rate of heat transfer in comparison with clean surface heat transfer rate, thereby portraying the fouling resistance as negative. However, in the present work and our earlier works [47e49], no negative value of fouling resistance is reported. Results can be seen in Fig. 14(aeb). Moreover, heating surface was constantly monitored using DNT® digital microscope imaging system. Fig. 15 shows the scale formation and sedimentation layer on the heating section during natural convection of CuO/water nanofluids. Similarly, Fig. 16 represents the nanoparticle deposition on the surface during nucleate boiling heat transfer. A rough comparison between Figs. 15 and 16 also demonstrates that thickness and rate of fouling is a strong function of time and fouling resistance for nucleate boiling is much
M.M. Sarafraz, F. Hormozi / International Journal of Thermal Sciences 90 (2015) 224e237
235
Fig. 14. a. Fouling resistance in natural convection region; b. Fouling resistance in nucleate boiling region.
Fig. 15. Scale formation and sedimentation layer during natural convection; HF ¼ 20 kW/m2, wt.% ¼ 0.4.
higher than that of visually recorded for natural convection area which is due to the higher heat fluxes and due to the formation of polarized-concentrated micro-boiling layer including the nanoparticles around the heating section which is currently undergone by our researches. 5. Conclusions Experimental study on the pool boiling heat transfer coefficient of dilute CuO/water nanofluids was conducted and thermal properties of nanofluids was also experimentally measured and following conclusions have been made: Using the pH control, stirring and sonication, nanofluids were able to have longer stability up to 1080 h at pH of 10.2 at wt. % ¼ 0.4 of CuO nanoparticles.
At any concentration of nanofluids, augmentation of the pool boiling heat transfer coefficient is reported, when heat flux increases. Deterioration of the pool boiling heat transfer coefficient in pool boiling of nanofluids is reported during the experiments. This may be due to the particle deposition which reduces the number of nucleation sites. It also formed a thick layer of nanoparticles on the surface which intensified the thermal resistance of surface. In case of using the surfactants, due to the reduction of surface tension, bubble formation was significantly intensified and consequently, bubble transport become the other phenomenon which improved the heat transfer from the heating surface. Small amount of SDS, SDBS and Triton X-100 surfactants (0.1% by weight) were added into the nanofluids. This causes the heat transfer coefficient to be increased. Also, boiling contact angle of
Fig. 16. Scale formation and sedimentation layer during Nucleate boiling; HF ¼ 65 kW/m2, wt.% ¼ 0.4.
236
M.M. Sarafraz, F. Hormozi / International Journal of Thermal Sciences 90 (2015) 224e237
bubbles with surface significantly decreased when surfactant was added and when nanofluid concentration increased. Also, wettability of surface significantly increased which is in accordance with previous studies. Surface roughness of heating section decreased, when concentration of nanofluid increases. This may be due to the deposition of nanoparticles around the heating section which covered the nucleation sites and filled the micro-cavities of surface. Rectilinear changes for natural convection and nucleate boiling regions are reported with different line slopes. Also, thickness and rate of fouling is a strong function of time and fouling resistance for nucleate boiling is much higher than that of visually recorded for natural convection area Acknowledgment Authors of the presented work, dedicates this work to Imam Mahdi and tend to appreciate Semnan State University for providing the facilities for conducting this research. Nomenclatures A C db g k P q Ra s T
area, m2 heat capacity, J kg1 C1 bubble departing diameter, m gravitational acceleration, m2 s1 thermal conductivity, W m1 C1 pressure, Pa heat, W roughness, m distance, m temperature, k
Subscripts b bulk bf base fluid l liquid nf nanofluid s saturated or surface th thermocouples Greek symbols a heat transfer coefficient, W m2 C1 D Difference q contact angle h enhancement ratio r density, kg m3 s surface tension, dyne/cm or somewhere N m1 m viscosity References [1] J. Barber, D. Brutin, L. Tadrist, A review on boiling heat transfer enhancement with nanofluids, Nanoscale Res. Lett. 6 (2011) 280. [2] S.U.S. Choi, Enhancing thermal conductivity of fluids with nanoparticles, in: Proceedings of the ASME International Mechanical Engineering Congress and Exposition, ASME, San Francisco, USA, 1995. [3] J.K. Lee, Y.J. Hwang, Y.C. Ahn, H.S. Shin, C.G. Lee, G.T. Kim, H.S. Park, Investigation on characteristics of thermal conductivity enhancement of nanofluids, Curr. Appl. Phys. 6 (2006) 1068e1071. [4] J.A. Eastman, S. Choi, S. Li, W. Yu, J. Thompson, Anomalously increased effective thermal conductivities of ethylene glycol-based nanofluids containing copper nanoparticles, Appl. Phys. Lett. 78 (2001) 718e720. [5] Y. Xuan, W. Roetzel, Conceptions for heat correlation of nanofluids, Int. J. Heat Mass Transf. 43 (2000) 3701e3707. [6] I.C. Bang, S.H. Chang, Boiling heat transfer performance and phenomena of Al2O3ewater nanofluids from a plain surface in a pool, Int. J. Heat Mass Trans. 48 (2005) 2407e2419.
[7] S.K. Das, N. Putra, W. Roetzel, Pool boiling characteristics of nano-fluids, Int. J. Heat Mass Transf. 46 (2003) 851e862. [8] R.A. Taylor, P.E. Phelan, Pool boiling of nanofluids: comprehensive review of existing data and limited new data, Int. J. Heat Mass Trans. 52 (2009) 5339e5347. [9] H. Xie, J. Wang, T. Xi, Y. Liu, F. Ai, Q. Wu, Thermal conductivity enhancement of suspensions containing nanosized alumina particles, J. Appl. Phys. 91 (7) (2002) 4568e4572. [10] M.H. Shi, M.Q. Shuai, Z.Q. Chen, Q. Li, Y.M. Xuan, Study on pool boiling heat transfer of nano-particle suspensions on plate surface, J. Enhanc. Heat Transf. 14 (3) (2007) 223e231. [11] J.P. Tu, N. Dinh, T. Theofanous, An experimental study of nanofluid boiling heat transfer, in: Proceedings of 6th International Symposium on Heat Transfer, Beijing, China, 2004. [12] D. Wen, Y. Ding, Experimental investigation into the pool boiling heat transfer of aqueous based c-alumina nanofluids, J. Nanoparticle Res. 7 (2) (2005) 265e274. [13] S. Witharana, Boiling of Refrigerants on Enhanced Surfaces and Boiling of Nanofluids, Ph.D. thesis, The Royal Institute of Technology, 2003. [14] B.H. Truong, Determination of Pool Boiling Critical Heat Flux Enhancement in Nanofluids, Undergraduate Thesis, MIT, May 2007. [15] H.S. Ahn, V. Sathyamurthi, D. Banerjee, Pool boiling experiments on a nanostructured surface, IEEE Compon. Packag. Tech. 32 (2009) 156e165. [16] M.M. Sarafraz, S.M. Peyghambarzadeh, S.A. Alavi Fazel, N. Vaeli, Nucleate pool boiling heat transfer of binary nano mixtures under atmospheric pressure around a smooth horizontal cylinder, Rperiodica Polytech. Chem. Eng. 56 (2) (2012) 1e7. [17] M.M. Sarafraz, S.M. Pyghambarzadeh, Nucleate pool boiling heat transfer to Al2O3-water and TiO2-water nanofluids on horizontal smooth tubes with dissimilar homogeneous materials, Chem. Biochem. Eng. Q. 26 (3) (2012) 199e206. [18] J.S. Coursey, J. Kim, Nanofluid boiling: the effect of surface wettability, Int. J. Heat Fluid Flow. 29 (2008) 1577e1585. [19] I.C. Bang, S.H. Chang, Boiling heat transfer performance and phenomena of Al2O3ewater nano-fluids from a plain surface in a pool, Int, J. Heat Mass Transf. 48 (12) (2005) 2420e2428. [20] V. Sajith, R.M. Madhusoodanan, C.B. Sobhan, An experimental investigation of the boiling performance of water-based nanofluids, in: Proceedings of the Micro/Nanoscale Heat Transfer International Conference, Jan. 2008. [21] 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 Transf. 12 (2007) 4105e4116. [22] M. Chopkar, A.K. Das, I. Manna, P.K. Das, Pool boiling heat transfer characteristics of ZrO2ewater nanofluids from a flat surface in a pool, Heat Mass Transf. (2007). [23] J.H. Kim, K.H. Kim, S.M. You, Pool boiling heat transfer in saturated nanofluids, in: Proceedings of IMECE 2004, Anaheim, CA, 13e19 November 2004, pp. 13e19. [24] 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). [25] S.M. You, J.H. Kim, K.H. Kim, Effect of nanoparticles on critical heat flux of water in pool boiling heat transfer, Appl. Phys. Lett. 83 (2003) 3374e3376. [26] P. Vassallo, R. Kumar, S.D. Amico, Pool boiling heat transfer experiments in silicaewater nano-fluids, Int. J. Heat Mass Transf. 47 (2004) 407e411. [27] S.A. Alavi Fazel, M. Mahboobpour, Pool boiling heat transfer in monoethyleneglycol aqueous solutions, Exp. Therm. Fluid Sci. 48 (2013) 177e183. [28] S.J. Kline, F.A. McClintock, Describing uncertainties in single-sample experiments, Mech. Eng. 75 (1) (1953) 3e12. [29] M.M. Sarafraz, F. Hormozi, Scale Formation, Subcooled Flow, Boiling heat transfer of CuO-water nanofluid inside the vertical annulus, Exp. Therm. Fluid Sci. 52 (2014) 205e214. [30] B. D. Cullity, S. R. Stock, Text Book of Elements of X-ray Diffraction. third ed., Prentice-Hall Publiser, pp. 170 (Chapter 5). [31] S.M. Peyghambarzadeh, S.H. Hashemabadi, M. Narakia, Y. Vermahmoudi, Experimental study of overall heat transfer coefficient in the application of dilute nanofluids in the car radiator, Appl. Therm. Eng. 52 (2013) 8e16. [32] M. Kole, T.K. Dey, Investigations on the pool boiling heat transfer and critical heat flux of ZnO-ethylene glycol nanofluids, Appl. Therm. Eng. 37 (2012) 112e119. [33] W.M. Rohsenow, A method of correlating heat transfer data for surface boiling of liquids, Trans. ASME 74 (1952) 969. [34] I.L. Pioro, Experimental evaluation of constants for the Rohsenow pool boiling correlation, Int. J. Heat Mass Trans. 31 (1992) 2003e2013. [35] D. Gorenflo, Pool boiling, Dusseldorf, Germany, in: VDI Heat Atlas, 1993 (chapter Ha). [36] H. Kim, M. Kim, Experimental study of the characteristics and mechanism of pool boiling CHF enhancement using nanofluids, Heat Mass Transf. 45 (7) (2007) 991e998. [37] J.H. Kim, K.H. Kim, S.M. You, Pool boiling heat transfer in saturated nanofluids, in: ASME Conference Proceedings, Anaheim, CA, 2004, pp. 621e628. [38] R.I. Elghanam, M.M.E.L. Fawal, R. Abdel Aziz, M.H. Skr, A. Hamza Khalifa, Ain Shams Eng. J. 2 (2011) 195e209. [39] V. Wasekar, R. Manglik, A review of enhanced heat transfer in nucleate pool boiling of aqueous surfactant and polymeric solutions, J. Enhanc. Heat Trans. 6 (1999) 135e150.
M.M. Sarafraz, F. Hormozi / International Journal of Thermal Sciences 90 (2015) 224e237 [40] D. Wen, B. Wang, Effects of surface wettability on nucleate pool boiling heat transfer for surfactant solutions, Int. J. Heat Mass Trans. 45 (2002) 1739e1747. [41] W. Frost, C. Kippenhan, Bubble growth and heat transfer mechanisms in forced convection boiling of water containing a surface active agent, Int. J. Heat Mass Trans. 10 (1967) 931e949. [42] L. Cheng, D. Mewes, A. Luke, Boiling phenomena with surfactants and polymeric additives: a state-of-the-art review, Int. J. Heat Mass Trans. 50 (2007) 2744e2771. [43] A.W. Adamson, Physical Chemistry of Surfaces, Interscience Publication, New York, 1967. [44] M.G. Kang, Effect of surface roughness on pool boiling heat transfer, Int. J. Heat Mass Trans. 43 (2000) 4073e4085. [45] P. Walker, R. Sheikholeslami, Assessment of the effect of velocity and residence time in CaSO4 precipitating flow reaction, Chem. Eng. Sci. 58 (2003) 3807e3816.
237
[46] S.M. Peyghambarzadeh, A. Vatani, M. Jamialahmadi, Application of asymptotic model for the prediction of fouling rate of calcium sulfate under subcooled flow boiling, Appl. Therm. Eng. 39 (2012) 105e113. [47] V. Nikkhah, M.M. Sarafraz, F. Hormozi, S.M. Peyghambarzadeh, Particulate fouling of CuOewater nanofluid at isothermal diffusive condition inside the conventional heat exchanger-experimental and modeling, Exp. Therm. Fluid Sci. 60 (2015) 83e95. [48] M.M. Sarafraz, F. Hormozi, Nucleate pool boiling heat transfer characteristics of dilute Al2O3eethyleneglycol nanofluids, Int. Commun. Heat Mass Trans. 58 (2014) 96e104. [49] M.M. Sarafraz, F. Hormozi, Convective boiling and particulate fouling of stabilized CuO-ethylene glycol nanofluids inside the annular heat exchanger, Int. Commun. Heat Mass Trans. 53 (2014) 116e123.
Contents lists available at ScienceDirect
International Journal of Thermal Sciences journal homepage: www.elsevier.com/locate/ijts
Pool boiling heat transfer to dilute copper oxide aqueous nanofluids M.M. Sarafraz*, F. Hormozi 1 School of Chemical, Petroleum and Gas Engineering, Semnan University, Semnan, Iran
a r t i c l e i n f o
a b s t r a c t
Article history: Received 18 March 2014 Received in revised form 6 December 2014 Accepted 7 December 2014 Available online
A set of experiments have been performed to quantify the pool boiling heat transfer coefficient of dilute copper oxide water-based nanofluids at mass concentrations of 0.1e0.4%. To stabilize the two-step nanofluids, pH control, stirring and sonication were utilized. For investigating the influence of surfactant as a surface active agent additive on the pool boiling heat transfer coefficient of nanofluids, SDS, SDBS and Triton X-100 were used. Influence of some operating parameters such as applied heat flux, mass concentration of nanofluids and other parameters such as roughness of surface, boiling contact angle and deposition on the pool boiling heat transfer coefficient of nanofluids were experimentally investigated. Results demonstrated a significant deterioration of heat transfer coefficient of nanofluids comparing with the base fluid in the absence of surfactants, however, in the presence of surfactant, higher pool boiling heat transfer coefficient was reported. According to results, roughness of surface is strongly controlled by nanofluid concentration due to deposition of nanofluids on the heating section. Rectilinear changes of deposition with time in term of fouling resistance were seen at both regions with natural convection and nucleate boiling dominant heat transfer mechanisms. © 2014 Elsevier Masson SAS. All rights reserved.
Keywords: Pool boiling Surfactant Nanofluid Bubble formation Roughness Scale formation
1. Introduction Boiling heat transfer is one of the major interests of heat transfer experts, not because of the complexity of boiling phenomena, but also its utilizations in heating/cooling systems and industrial processes and its applications in refrigeration, power generation, heat exchangers, cooling of high-power electronics components and PWR nuclear reactors. Removing a large quantities of heat in the least possible compact size has always been a key point for designing the heat exchanging media. Thus, enhancing the heat transfer coefficient and thermal efficiency of above-mentioned systems is indispensable and vital. Pool boiling involves all the interactive, but complicated, and dynamic processes such as hydrodynamics, heat and mass transfer (particularly in multicomponent mixtures) and sub-phenomena such as: nucleation, bubble coalescence, and collapse of bubbles. Many efforts have been made to experimentally investigate on the pool boiling heat transfer of different fluids including: pure liquids, multicomponent mixtures, refrigerants and none-Newtonian fluids.
* Corresponding author. Tel.: þ98 9166317313; fax: þ98 6324231683. E-mail addresses: [email protected] (M.M. Sarafraz), [email protected] (F. Hormozi). 1 Tel.: þ98 9123930495. http://dx.doi.org/10.1016/j.ijthermalsci.2014.12.014 1290-0729/© 2014 Elsevier Masson SAS. All rights reserved.
However, a simple statistical search among the published documents reveals this fact that nanofluid-related documents are less in number when comparing to other pool boiling subjects. As can be seen in Fig. 1, more researches should be conducted in context of understanding the heat transfer characteristics of nanofluids. A brief literature review on the pool boiling heat transfer of nanofluids has been represented in the following section. 2. Literature review For the first time in 1995, Choi introduced the concept of nanofluid in his investigations [1]. He provided results of a theoretical study of suspended copper nanoparticles in a base fluid and demonstrated significant improvement in thermal properties of the test colloid fluid. Further experimental investigations have reported that suspensions containing nanoparticles have substantially higher thermal conductivities than traditional heat transfer fluids [2e4]. Recently, Xuan summarized the main factors belonging to nanofluids that enhance heat transfer as follows [5]: (a) The nanoparticles can increase the surface area; (b) nanoparticles have higher interaction and collisions among the particles and fluid in comparison with other solid particles (e.g. microparticles); and (c) increased mixing fluctuation and turbulence of the fluid. Owing to these attributes, it is expected that the heat transfer performance of water, the most widely used coolant, can
M.M. Sarafraz, F. Hormozi / International Journal of Thermal Sciences 90 (2015) 224e237
225
Fig. 1. Proportion of conducted researches on boiling in terms of different subjects; investigations on nanofluid after several decades are still less in number; (Direct keywords in www.sciencedirect.com).
be improved [6]. Although many researchers such as Das [7] showed that the boiling heat transfer performance deteriorates when nanofluid is used as working fluid, but there are still controversial reports that nanofluids can enhance the boiling performance [8]. Lui et al. [9] conducted an experimental investigation on boiling of CuO/water nanofluids around the grooved copper block and reported the enhancement of pool boiling heat transfer coefficient close to 50%. Shi et al. [10] and Tu et al. [11] performed experiments on Al2O3 nanofluids and collected set of experimental data that when compared to base fluid, demonstrated enhancement up to 64%. Wen et al. [12] represented the significant enhancement of heat transfer coefficient of TiO2/water and Al2O3/ water nanofluids on the stainless steel cylinder with diameter of 150 mm. Witharana [13], Troung [14], Ahn et al. [15] and Sarafraz et al. [16,17] demonstrated the enhancement of pool boiling heat transfer coefficient using nanofluids on heaters with different geometrical properties. More information can be found on cited references. In contrast, many investigators demonstrated the reduction of pool boiling heat transfer coefficient in their results. Coursey and Kim [18], Bang and Chang [19] and Sajith [20] carried some experiments out to investigate the effect of Al2O3 water based nanofluids on pool boiling heat transfer coefficient. They also surveyed the influence of scale formation and particle deposition around the heating section. The outstanding point of their works was to show the deterioration of heat transfer coefficient. Kim et al. [21], conducted a research based on the boiling heat transfer of Al2O3, CuO ZrO2 and SiO2 water based nanofluid on a wire with diameter about 0.38 mm and showed the significant deterioration of heat transfer coefficient due to the scale formation. Chopkar et al. [22], Kim et al. [23], Narayan et al. [24], You et al. [25] and Vassallo et al. [26] performed studies on boiling of nanofluid and small change in heat transfer coefficient was reported. As can be seen in the above-mentioned, controversial reports on boiling of nanofluids may be found in literature. Briefly speaking, Nanofluid pool boiling literature is in conflict over whether nanoparticles can enhance or degrade boiling heat transfer. In the present work, CuO/water nanofluids are prepared using two-step method and stabilized using pH control, stirring, and sonication processes. By employing these methods, stability of nanofluids can be improved up to 1080hr. Pool boiling heat transfer coefficients of nanofluids, in the absence/presence of SDS, SDBS
surfactants at the same mass concentration (0.1%) are experimentally measured. Influence of surfactants on the bubble-surface contact angle and bubble formation rate is visually investigated and briefly discussed. Likewise, influence of particle deposition on the heater roughness is experimentally investigated and interesting results regarding to scale formation and roughness in term of fouling resistance is represented. 3. Experimental 3.1. Experimental apparatus and procedure This study focuses on the nucleate pool boiling of nanofluids on the surface of a horizontal cylindrical heater. A schematic of test facility has been shown in Fig. 2a. As can be seen in Fig. 2a, experimental setup consists of three main sections: measurement instruments, boiling test section and photographic system. The stainless steel vessel is equipped with the boiling test section and condenser. The condenser (which is constantly cooled by water/ethylene glycol cooling cycle) condenses the vapor produced by the applied heat and the condensed liquid is returned back to the bottom of the vessel for reevaporation. A pressure sensor is mounted on top of the vessel to control the pressure of test vessel equals to the atmospheric pressure throughout the experiment. The whole system is heavily insulated for more controllability and reduction of heat loss. The temperature of the liquid inside the test vessel is monitored and controlled at predetermined set point by a thermal regulator which involves eight of PT-100U thermometers and appropriate vertical auxiliary heater. These RTDs are used to measure the bulk liquid temperature during experiments mounted at different locations inside the test vessel (bottom, up, left, right, center and four around the heating section). Heating section can be considered as the most important section of the test apparatus. A horizontal stainless steel cylinder with outer diameter of 21 mm and length of 350 mm that only the first 105 mm of its length is internally heated using a bolt heater (manufactured by Cetal Co.). It is also equipped with eight K-type calibrated thermocouples, embedded along the circumference of the rod, very close to the heating surface. The arithmetic average of these eight thermocouples can be considered as the surface temperature. A PC-based
226
M.M. Sarafraz, F. Hormozi / International Journal of Thermal Sciences 90 (2015) 224e237
Fig. 2. a. A scheme of test apparatus; b. Geometrical details of cylindrical test section.
data acquisition system was used to record the measuring parameters. The power of the heating system is provided using a DC regulator. Note that, test vessel is equipped with two heaters such that the first one maintains the temperature of bulk fluid to any desired temperature and the second one is the bolt heater mounted inside the stainless steel cylinder to provide the applied heat flux for pool boiling. Casio EX-F1 was also used to record the bubble formation and nucleation phenomena. Taken videos/ photos then are processed using image and video editing software's. After analyzing conclusions on bubble-related phenomena are made. Fig. 2b shows the geometrical details of heating section by cross section view. Noticeably, because of the ultra-high speed (UHS) recording, Arri high speed ballasts lighting system (Luminys
Lab-Light 30k) was employed to minimize the effect of flicker on taken images. Prior to commencing the experiments, test section, test vessel and pipes were cleaned by methanol/water solution and dried to remove any scale from previous experiments. Once the system was cleaned, the test fluid was introduced to the test vessel and deaeration process was performed. Following this, the auxiliary heater was switched on and the temperature of the system increased. When the fluid had reached the desired temperature, then, the power was supplied to the test heater and kept at a pre-determined value. The data acquisition system was switched on and temperatures, pressure and heat fluxes were recorded. Noticeably, the experiments have been entirely performed at saturation temperature
M.M. Sarafraz, F. Hormozi / International Journal of Thermal Sciences 90 (2015) 224e237
under atmospheric pressure. Experiments were repeated three times to ensure the reproducibility of the results. 3.2. Uncertainty and data reduction To estimate the pool boiling experimental heat transfer coefficient, applied heat flux should be calculated by the product of electrical voltage, current and cosine of the difference between input electrical voltage and current using the following correlation (cosine of difference between input and electrical voltage is almost equal to zero):
q¼
V I p$D$L
(1)
where, I is the current (A), V is the voltage (V), D is tube diameter (m) and L is tube length (m). The temperature drop between the location of the wall thermocouples and the heat transfer surface was deducted from the measured temperature difference according to:
s q Ts Tb ¼ ðTth Tb Þ $ k A
(2)
In this equation, s is the distance between the thermocouple locations and the heat transfer surface and k is the thermal conductivity of the heater material. The value of s/k was determined for each thermocouple by calibration of the test heater using pure water. Since, ratio of diameter of cylinder to its length is insignificant, cylindrical effect can be ignored. The surface temperature was obtained by the arithmetic average of values measured by the eight thermocouples. Bulk temperature was also estimated using RTDs. The heat transfer coefficient h is calculated from:
a¼
q A
ðTs Tb Þave:
(3)
To compensate the temperature drop, AlaviFazel [27] proposed that instead of using Eq. (2), following ODE equation can be solved:
1 d dTs kr ¼0 r dr dr
(5)
and:
Da ¼ a
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2 2 Dq DT þ q T
is utilized for preparing the nanofluid. Briefly, these process includes: I. Weigh the desired mass of CuO with digital electronic balancer (A&D EK Series Portable Balances, EK-1200i). II. The weighted CuO nanoparticle were added into the weighted DI-water, while all were agitated in a clean flask. The magnetic motorized stirrer (Hanna instruments Co.) was employed for agitating the nanofluid. III. UP400S ultrasonic Hielscher GmbH (350 W/24 kHz) was used to disperse the nanoparticles into the water uniformly and crack the agglomerations and clogs. Thermal conductivity, specific heat capacity and density of test nanoparticles were 32.9 (W/m K), 550.5 (kJ/kg K), 6320 (kg/m3) respectively. Noticeably, nanofluids were prepared at percentage weight fractions of 0.1e0.4. Also, quality tests were performed to ensure about size, morphology and purity of nanoparticles and their mean size. Results have been represented in Fig. 3(aec). As can be seen in Fig. 3(a), XRD pattern depicts the single-phase CuO with a monoclinic structure which implies that there is no impurity other than CuO nanoparticles and no significant peaks of impurities are found in XRD pattern [29]. Also, by employing the Scherrer's equation [30], nanoparticle size was also verified. According to the Scherrer's formula:
t¼
0:9l B cos q
(7)
l is the wavelength of X-ray target (for copper nanoparticle, it is 1.54 Å) and value of 2q and B was determined by the XRD spectra. Result of Scherrer's formula (45e50 nm) confirmed the results obtained by the particle-count experiment (shown in Fig. 3b) and XRD results. As can be seen, the dominant size of particles is close to 50 nm. Fig. 3c depicts the SEM image of CuO nanoparticles. As can be seen, particles are spherical and have the same morphology and mean size.
(4)
In Eq. (4), k is the temperature dependent thermal conductivity of the heater, which was approximated to a linear function of temperature. However, in this work, by means of calibration, Eq. (2) was used. Also, a detailed uncertainty analysis performed in accordance with Kline and McClintock [28] estimated an overall uncertainty of the pool boiling heat transfer coefficient.
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2 Dq DV 2 DI 2 DA ¼ þ þ q Vmax: Imax: A
227
3.4. Stabilization of nanofluid There are several stabilization techniques for nanofluids which come as follows: 3.4.1. The electrostatic stabilization In which nanofluids can be stabilized by controlling the chemical characteristics of solution such as pH and concentration of existing ions. In this method, Zeta potential can be a gold parameter in determining the isoelectric point (IEP) and assessing the optimum pH, in which the zeta potential of solution should be far from (smaller or higher than) the isoelectric point. Noticeably, in IEP, zeta
(6)
The maximum calibration error of the thermocouple was 0.2 K. The maximum deviation related to the measuring of location of the thermocouples was about 0.5%. The measurement error of the power meter was 0.5%. Maximum uncertainties of the heat flux were also about 9%. The uncertainty of heat transfer coefficient according to Eq. (6) equals to 10.1%. Table 1 represents the uncertainty values of instruments used in this work. 3.3. Preparation of nanofluid In the present work, CuO nanoparticles (50 nm, PlasmaChem GmbH, Germany) were uniformly dispersed into the DI-water as the base fluid for making a stable nanofluid. The two-step method
Table 1 Uncertainties of the measurement instruments. Parameter
Instrument
Uncertainty
Cross section area (m) Surface temperature (K) Voltage (V)
Precise engineering calipers
0.001 m
Calibrated K-type thermocouple
±0.2 K
Keithley digital multi-meter
Current (A)
Keithley digital multi-meter
Bulk temperature (K) Contact angle (degree)
PT-100U thermo-meter (thermo-resistance) Visual observation, Sigma scan®, Adobe® Photoshop with filter
±1% of reading ±0.1% of reading ±0.1 K ±3
228
M.M. Sarafraz, F. Hormozi / International Journal of Thermal Sciences 90 (2015) 224e237
Fig. 3. a XRD pattern of CuO nanoparticles; b Results of nanoparticle size count; c SEM image of CuO nanoparticles.
potential is minimum (approximately zero) and the least stability of nanofluids can be seen in isoelectric point. 3.4.2. The steric stabilization In which the surface properties and potential is the key parameter in stabilizing of nanofluids. This parameter is controlled by adding surfactants or dispersants. With adding the surfactants/ dispersants, coalescence (cohesive behavior of particles) significantly decreases which leads the particles to be uniformly dispersed in the medium. However, it is hardly believed that surfactants can change the thermo-physical properties of solutions and reduces the reliability of experimental data. 3.4.3. The depletion stabilization In which a new free polymer is added into the medium of dispersion which changes the intermolecular forces between particles toward the lower sedimentation. Since, particles with lower size have higher surface energies; therefore, these types of nanofluids have higher potential to form the cluster and local agglomeration and aggregations. 3.4.4. Mechanical/ultrasonic agitations Which leads the agglomerations and aggregations to be cracked in smaller pieces; this technique strongly depends on the power
and frequency of sonication. Time is the other important parameter in this method. There are still other techniques such as time-experiment methods including the settling bed and time-experiment methods. Such techniques can be employed in parallel of the other methods. In most of these techniques, height of sedimentation and scales is measured. In this research, in order to stabilize the test nanofluid, pH control is employed such that by means of HCl þ NaOH buffer solution, pH of nanofluids is controlled. Peyghambarzadeh et al. [31] conducted some experiments on the stabilization of CuO nanofluids as working fluid inside the car radiator and concluded that at pH value of 10.1 the best stability of nanofluid may be obtained. However, he ignored the influence of agglomeration of nanofluids and did not consider the influence of sonication on the stability of nanofluids. Dey et al. [32] used the ultrasonic for more than 60hr for ZnO ethylene glycol nanofluids and prepared the faily stable nanofluids. In the present work, using the DLS analyzer (Malvern Zen-3600), zeta potential of CuO nanoparticles in water has experimentally been measured. Results showed that at isoelectric point of pH ¼ 8.33, zeta potential is 0 mv at maximum concentration of nanofluid (which is so-called isoelectric point or IEP) and therefore, for stabilizing the nanofluids, pH of nanofluids must be smaller or higher than 8.33 as far as possible. Experiments showed
M.M. Sarafraz, F. Hormozi / International Journal of Thermal Sciences 90 (2015) 224e237
229
possible cluster can be formed inside the nanofluid and the minimum of scale would form on the heating section. To measure the pH, the pen-type pH-meter AZ-8685A (AZ instruments Co) was used. Fig. 4 represents the pH values of nanofluid before the pHlevel control experiments.
3.5. Experimental measurement of thermo-physical properties of nanofluids
Fig. 4. pH values of nanofluids at different concentrations before PH level control.
us that at pH ¼ 10.2, the zeta potential is 5.43 and consequently, the best stable nanofluid (for about 1080hr) is achievable. Because, zeta potential at this pH has significant difference when compared to isoelectric point in which zeta potential is 0. Therefore, the least
Based on the literature, pool boiling heat transfer coefficient can be estimated by the predicting correlations which strongly depend on the thermo-physical properties of test fluid. Physical properties can also be obtained by the related correlations which sometimes can be tolerated up to 30%. On the other hand, these correlations have not been re-developed for thermo-physical properties of nanofluids. Therefore, one way is to find the thermo-physical properties of nanofluids experimentally. Thus, viscosity of nanofluids was measured using DV-II þ PRO digital viscometer (manufactured by Brookfield Co., Accuracy: ±1.0% of reading/ Repeatability: ±0.2%). Density was also measured using DMA 4500 ME (manufactured by Anton Paar Co., Accuracy: ±1.0% of reading/Repeatability: 0.00001 g/cm3). Thermal conductivity was also measured using DTC300 (manufactured by TA Instrument, Accuracy: ±3%e8% depending on thermal resistance/Repeatability:
Fig. 5. a. Density of deionized water and nanofluids at different temperatures; b. Viscosity of nanofluids and deionized water at different temperatures; c. Thermal conductivity of nanofluids at different temperatures; d. Representation of Newtonian behavior of nanofluids at different shear rate and shear stress.
230
M.M. Sarafraz, F. Hormozi / International Journal of Thermal Sciences 90 (2015) 224e237
Fig. 6. Parameters measured by the profile meter (more details and info can be found on Elcometer official site and user manual).
±1 to ±2% depending on thermal resistance/Standard: ASTM E1530). Results have been represented in Fig. 5(aed). The Newtonian behavior of nanofluid was investigated and represented in Fig. 5d. 4. Results and discussions Before any discussions, the experimental results must be validated. To ensure the accuracy of obtained data, Rohsenow [33] correlation was employed for pure water. Because: a) Rohsenow correlation has been proposed based on the pure water and has been tested for this DI-water in previous studies. b) The physical properties of DI-water are well known with high accuracy. c) Pool boiling heat transfer coefficient of distilled water had been investigated by several investigators over a wide range of heat fluxes and system pressures. This equation has been represented as follows:
2 sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi30:33 Cpl $DTe q s 4 5 ¼ Csf mhfg g rl rg hfg prl
(8)
In the correlation, Csf and n are 0.01 and 1 respectively [34]. Gorenflo [35] also has proposed a predicting correlation verified by the experimental data based on pure water and water-based nanofluids which includes the parameter related to surface roughness. Therefore, experimental setup is validated under comparisons of experimental data and those of calculated by these two well-known correlations. Likewise, to obtain the surface roughness of the cylinder, the profile meter (Elcometer-7061) was employed to measure the roughness of surface. Fig. 6 shows the parameters that can be measured by profile meter. Roughness meter results indicate that a uniform pattern of roughness can be seen along with the length of heating section. The mean roughness of surface was 0.34 mm which is quite higher than average particle size (50 nm). Results of validation for pure water demonstrated the good agreement between experimental data and well-known correlations which can be seen in Fig. 7. An overview on the experimental data related to the pool boiling heat transfer coefficient shows that boiling heat transfer coefficients in the nanofluids are significantly lower than that in pure water. There are several parameters influencing on the
Fig. 7. Comparisons of experimental data and results obtained by Rohsenow and Gorenflo correlations.
M.M. Sarafraz, F. Hormozi / International Journal of Thermal Sciences 90 (2015) 224e237
231
Fig. 8. Effect of heat flux and concentration on the pool boiling heat transfer coefficient of Cuo-water nanofluids at different concentrations.
deterioration of the pool boiling heat transfer coefficient of nanofluids, which are discussed in the forthcoming sections. 4.1. Influence of heat flux and mass concentration of nanofluid According to the experimental data, heat flux has a strong influence on the pool boiling heat transfer coefficient. It is more convenient to show the effect of heat flux in terms of heat flux versus heat transfer coefficient. Fig. 8 demonstrates the experimental data related to pool boiling heat transfer coefficient of CuO/water nanofluids at different mass concentrations of nanoparticles at various heat fluxes. Briefly speaking, the higher concentration of nanoparticles, the higher reduction of pool boiling heat transfer coefficient is registered all over the
experiments. Also by increasing the heat flux, heat transfer coefficient significantly increases. Deterioration of heat transfer coefficient can be an effect of scale formation and sedimentation of nanoparticles around the surface which increases the fouling resistance parameter and thermal resistance of surface that leads to the heat transfer to be reduced. Such depositions can reduce the number of active nucleation sites as well as creation of an insulation porous layer on the surface. It also decreases the surface roughness of heating surface area. Consequently, thermal resistance of heating section is locally increases and finally, heat transfer coefficient deteriorates. Although there are studies that have been carried out using nano-coated surfaces [36,37], but the factors responsible for the nanoparticle deposition are still unknown.
Fig. 9. Bubble formation during pool boiling of nanofluids at initial time without scale formation; note that scale formation is a strong function of time.
232
M.M. Sarafraz, F. Hormozi / International Journal of Thermal Sciences 90 (2015) 224e237
bubbles, bubbles with lower diameter size are seen. However, average number of the bubbles formed on the surface significantly increases. This fact can also be seen in Fig. 9(bef). 4.2. Influence of surfactants
Fig. 10. Boiling contact angle concept according to the Young theory.
When initial heat flux is applied, the natural convection is the dominant heat transfer mechanism around the heating section and no bubble is formed around the cylinder. By increasing the heat flux up to 15 kW/m2, the nucleate boiling becomes the dominant heat transfer mechanism and first bubbles may appear around the cylinder. These bubbles due to the lower local agitation and slower bubble interactions have enough resident time to grow and depart from the surface. Therefore, bulky and bigger bubbles is seen at lower heat fluxes (Fig. 9a). At higher heat fluxes, due to increase the number of nucleation sites, numerous bubbles are formed around the surface and due to the local agitation and collapsing of the
Small quantity of surfactant can increase pool/flow boiling heat transfer coefficient. The rate of enhancement has also been proved that to be dependent on the additive concentrations, its type and chemistry, wall heat flux, and the heater geometry. The concentrations are usually low enough that, the addition of surfactant to water causes no significant changes in saturation temperature and most other physical properties, except for viscosity and surface tension. Small concentration of surfactant additives can also reduce the solution's surface tension considerably, and its level of reduction depends on the amount and type of surfactant added into the solution. The activation of nucleate sites, bubble growth and bubble dynamics influence the boiling heat transfer coefficient. One can refer to the literature [38e41] to find information related to the effect of surfactants on boiling heat transfer and their effects on boiling performance and sub-phenomena [38]. In the present work, 0.1% by weight of Sodium dodecyl Sulfate (SDS), Sodium dodecyl benzene sulfonate (SDBS) and Triton X-100 were added and their influence on the pool boiling heat transfer coefficient of nanofluids as well as bubble formation was experimentally investigated. SDS is widely used in detergents for laundry with many cleaning applications and foaming industries. SDS is highly effective surfactant and can be used in any task requiring the removal of oily stains and painting spots or residues. Triton X-
Fig. 11. Bubble formation in pool boiling of CuO nanofluids; reduction of visual boiling contact angle by increasing the concentration of nanofluids at the same concentration of surfactants, HF ¼ 45 kW m2.
M.M. Sarafraz, F. Hormozi / International Journal of Thermal Sciences 90 (2015) 224e237
100 is a commonly used detergent in laboratories. As surfactant may change the properties of nanofluids and their stability, the lowest possible mass concentration of surfactant was added into the nanofluids. According to the visual observations, with adding the surfactants, visible nucleation site locations dramatically increased. Also, deposition of nanoparticles increases when concentration of nanofluids and surfactant concentration increases too. Bubble-surface contact angle is another key parameter in boiling phenomena which was under influence by surfactants. According to the available studies, most researchers measured the surface tensions of aqueous additive solutions, while did not consider the effect of the contact angles on the boiling phenomena. In the present work, contact angle of nanofluids are visually measured using UHS photographic system, however, Young correlation is the most accurate equation for estimating the contact angle and wettability of a surface however, we were unable to provide the facility to measure the exact values for surface tension of nanofluids. As shown in Fig. 10, in Young equation, contact angle, q, is the angle between the liquid and solid interface which has visually been measured by analyzing the images. As mentioned, contact angle is one of the most important parameters in understanding boiling phenomena since it characterizes the wettability of the solid surface by the liquid [42]. Young [43] proposed an equation (Eq. (9)) which is still used as a fundamental understanding concept for phenomena involving the surface wettability and contact angle (q).
q ¼ arccos
ssv ssl slv
233
(9)
where, ssv, ssl and slv are the surface tension at interface of solid/ vapour, solid/liquid and liquid/vapour phases respectively. Fig. 10 represents the Young correlation related to the contact angle for a single bubble. The contact angle is defined as an angle between solideliquid and liquidevapor interfaces. Noticeably, in the present work, results of contact angle are based on the visual observations and Young equation was not employed. As can be seen in Fig. 11, visual contact angle of bubbles formed on the surface decreased with increasing the concentration of nanofluids. In fact, with decreasing the boiling contact angle, wettability of surface increases. To interpret this phenomenon by the macroscopic view, it can be said that when concentration of nanofluids increases, rate of sedimentation of particles on the surface significantly increases and coating is formed which leads the wettability of surface to be increased, while roughness of surface to be decreased. However, this coatings and particle may not easily be seen by naked eye. Note that, the relative errors of the measurement of bubble radius and bubble coordinates are estimated to be no more than 3 for a typical bubble radius of 0.5 mm, by assuming a constant error of 0.5 pixels during the discretization in imaging process in Sigma scan®. Also, as can be seen in Fig. 12, average bubble diameters are lower in comparison with the case without the surfactants,
Fig. 12. Bubble formation in pool boiling of CuO/water nanofluids in absence/presence of surfactants.
234
M.M. Sarafraz, F. Hormozi / International Journal of Thermal Sciences 90 (2015) 224e237
although, bubbles some times are coalesced and form the bigger bubbles. Coalescence seems to be weakened when surfactants are added. To demonstrate the influence of surfactant on the heat transfer coefficient of CuO/water, enhancement ratio is defined as:
h¼
aPoolBoilingþSDS aPoolBoiling
(10)
Fig. 13 represents the pool boiling heat transfer coefficient of CuO/water under influence of SDS surfactant. Higher pool boiling heat transfer coefficient of CuO/water nanofluid is reported (up to 70.48% for 0.4% of SDS) for those nanofluid mixed with SDS wt. % ¼ 0.1 comparing to other surfactants. For other surfactants, similar behavior can be seen. The role of surfactant on heat transfer enhancement is out of goals of this work.
Also, time is another major parameter, which should be taken into the account in sedimentation and fouling experiments, but is out of goals of this work and currently is undergone. Noticeably, surface with decreased roughness (smoother surface) gives weaker bubble formation, because the smoother surface has usually less cavities than the rough surface, subsequently, less bubbles are formed during boiling and bubble transport becomes weaker. On the other hand, heating surface is covered by the generated bubbles and due to the vapor captured inside the bubbles, heating section is partially isolated. Hence, boiling heat transfer performance of surface significantly reduces. All in all, there are two factors having major roles in boiling deterioration of nanofluids: 1) less cavities due to deposition of nanoparticles 2) negative influence of formed deposition on the surface which increases the thermal resistance of surface and leads to the deterioration of heat transfer.
4.3. Influence of surface roughness
4.4. Deposition (fouling) thermal resistance
Undoubtedly, heating surface condition is one of the most important factors in pool boiling of nanofluids [44]. It is one of the most important factors in increasing/decreasing heat transfer rate since it is directly related to the active nucleation sites density on the heating surface [45]. There are many studies under several conditions of specially treated surface, coated surface, or polished surface and different methods that have been employed to change the heating surface conditions which come as follows: a) Roughening surface itself. b) Coating the surface using special metal, porous material or using electrophoretic methods. c) Adding specially manufactured irregularities and cavities or patterned shapes on the surface. d) Recently, using the nanofluids and sedimentation of particles on the surface. Nanoparticles, due to the instability and agitations and their unbalanced interactions between molecular forces tend to deposit on the surface. In some systems, gravity is also a significant parameter helping to increase the rate of deposition of the particles. To measure the roughness surface, profile-meter is used before and after the boiling experiments. According to results registered by the profile meter, roughness of surface decreased from 0.34 mm for clean surface to 0.29 mm, 22 mm, 0.19 mm and 0.13 mm for wt. % ¼ 1 to 0.4 respectively, when concentration of nanofluids increased.
Pool boiling heat transfer mainly consists of two distinguishable heat transfer regions. The first region where natural convection is dominant mechanism and nucleate region where nucleate boiling heat transfer is dominant heat transfer mechanism. In the present work, a set of experiments were conducted to measure the fouling resistance of CuO nanoparticles around the stainless steel cylinder. It is customary to present fouling data in terms of fouling resistance (Rf) which can be calculated on the basis of heat transfer [46]:
Fig. 13. Enhancement ratio for pool boiling heat transfer coefficient by adding SDS surfactant.
Rf ¼
1 1 aðtÞ aðt ¼ 0Þ
(11)
Fig. 14(a,b) show the changes of pool boiling heat transfer coefficient and fouling resistance over the time. As can be seen, the pool boiling heat transfer coefficient decreases with time due to the deposition of CuO nanoparticles on the heat transfer surface and also decreasing in number of nucleation sites on the heating surface; nanoparticles cover the heating surface and slightly change the thermal resistance of surface. On the other hand, according to previous studies, micro-cavities of surface reduces, while wettability of surface increases and some other nucleation sites are covered by fluid. On the other hand, bubbles cover the heating section too which lead to the heat transfer rate to be considerably decreased. Experimental results reveal a rectilinear changes in increase of fouling resistance with time in natural convection heat transfer region and nucleate boiling which both of these changes are results of nanoparticle sedimentation fouling. Walker et al. [45] reported an asymptotic behavior of fouling resistance in boiling flow which may be attributed to additional particulate fouling. Peyghambarzadeh et al. [46] represented that the fouling resistance in the initial stages of the test period decreases even hit the negative value at some points. He expressed that this phenomenon may be due to the fact that during the nucleation stage, the nuclei forming on the heat transfer surface increases the surface roughness. This leads to an increase in the rate of heat transfer in comparison with clean surface heat transfer rate, thereby portraying the fouling resistance as negative. However, in the present work and our earlier works [47e49], no negative value of fouling resistance is reported. Results can be seen in Fig. 14(aeb). Moreover, heating surface was constantly monitored using DNT® digital microscope imaging system. Fig. 15 shows the scale formation and sedimentation layer on the heating section during natural convection of CuO/water nanofluids. Similarly, Fig. 16 represents the nanoparticle deposition on the surface during nucleate boiling heat transfer. A rough comparison between Figs. 15 and 16 also demonstrates that thickness and rate of fouling is a strong function of time and fouling resistance for nucleate boiling is much
M.M. Sarafraz, F. Hormozi / International Journal of Thermal Sciences 90 (2015) 224e237
235
Fig. 14. a. Fouling resistance in natural convection region; b. Fouling resistance in nucleate boiling region.
Fig. 15. Scale formation and sedimentation layer during natural convection; HF ¼ 20 kW/m2, wt.% ¼ 0.4.
higher than that of visually recorded for natural convection area which is due to the higher heat fluxes and due to the formation of polarized-concentrated micro-boiling layer including the nanoparticles around the heating section which is currently undergone by our researches. 5. Conclusions Experimental study on the pool boiling heat transfer coefficient of dilute CuO/water nanofluids was conducted and thermal properties of nanofluids was also experimentally measured and following conclusions have been made: Using the pH control, stirring and sonication, nanofluids were able to have longer stability up to 1080 h at pH of 10.2 at wt. % ¼ 0.4 of CuO nanoparticles.
At any concentration of nanofluids, augmentation of the pool boiling heat transfer coefficient is reported, when heat flux increases. Deterioration of the pool boiling heat transfer coefficient in pool boiling of nanofluids is reported during the experiments. This may be due to the particle deposition which reduces the number of nucleation sites. It also formed a thick layer of nanoparticles on the surface which intensified the thermal resistance of surface. In case of using the surfactants, due to the reduction of surface tension, bubble formation was significantly intensified and consequently, bubble transport become the other phenomenon which improved the heat transfer from the heating surface. Small amount of SDS, SDBS and Triton X-100 surfactants (0.1% by weight) were added into the nanofluids. This causes the heat transfer coefficient to be increased. Also, boiling contact angle of
Fig. 16. Scale formation and sedimentation layer during Nucleate boiling; HF ¼ 65 kW/m2, wt.% ¼ 0.4.
236
M.M. Sarafraz, F. Hormozi / International Journal of Thermal Sciences 90 (2015) 224e237
bubbles with surface significantly decreased when surfactant was added and when nanofluid concentration increased. Also, wettability of surface significantly increased which is in accordance with previous studies. Surface roughness of heating section decreased, when concentration of nanofluid increases. This may be due to the deposition of nanoparticles around the heating section which covered the nucleation sites and filled the micro-cavities of surface. Rectilinear changes for natural convection and nucleate boiling regions are reported with different line slopes. Also, thickness and rate of fouling is a strong function of time and fouling resistance for nucleate boiling is much higher than that of visually recorded for natural convection area Acknowledgment Authors of the presented work, dedicates this work to Imam Mahdi and tend to appreciate Semnan State University for providing the facilities for conducting this research. Nomenclatures A C db g k P q Ra s T
area, m2 heat capacity, J kg1 C1 bubble departing diameter, m gravitational acceleration, m2 s1 thermal conductivity, W m1 C1 pressure, Pa heat, W roughness, m distance, m temperature, k
Subscripts b bulk bf base fluid l liquid nf nanofluid s saturated or surface th thermocouples Greek symbols a heat transfer coefficient, W m2 C1 D Difference q contact angle h enhancement ratio r density, kg m3 s surface tension, dyne/cm or somewhere N m1 m viscosity References [1] J. Barber, D. Brutin, L. Tadrist, A review on boiling heat transfer enhancement with nanofluids, Nanoscale Res. Lett. 6 (2011) 280. [2] S.U.S. Choi, Enhancing thermal conductivity of fluids with nanoparticles, in: Proceedings of the ASME International Mechanical Engineering Congress and Exposition, ASME, San Francisco, USA, 1995. [3] J.K. Lee, Y.J. Hwang, Y.C. Ahn, H.S. Shin, C.G. Lee, G.T. Kim, H.S. Park, Investigation on characteristics of thermal conductivity enhancement of nanofluids, Curr. Appl. Phys. 6 (2006) 1068e1071. [4] J.A. Eastman, S. Choi, S. Li, W. Yu, J. Thompson, Anomalously increased effective thermal conductivities of ethylene glycol-based nanofluids containing copper nanoparticles, Appl. Phys. Lett. 78 (2001) 718e720. [5] Y. Xuan, W. Roetzel, Conceptions for heat correlation of nanofluids, Int. J. Heat Mass Transf. 43 (2000) 3701e3707. [6] I.C. Bang, S.H. Chang, Boiling heat transfer performance and phenomena of Al2O3ewater nanofluids from a plain surface in a pool, Int. J. Heat Mass Trans. 48 (2005) 2407e2419.
[7] S.K. Das, N. Putra, W. Roetzel, Pool boiling characteristics of nano-fluids, Int. J. Heat Mass Transf. 46 (2003) 851e862. [8] R.A. Taylor, P.E. Phelan, Pool boiling of nanofluids: comprehensive review of existing data and limited new data, Int. J. Heat Mass Trans. 52 (2009) 5339e5347. [9] H. Xie, J. Wang, T. Xi, Y. Liu, F. Ai, Q. Wu, Thermal conductivity enhancement of suspensions containing nanosized alumina particles, J. Appl. Phys. 91 (7) (2002) 4568e4572. [10] M.H. Shi, M.Q. Shuai, Z.Q. Chen, Q. Li, Y.M. Xuan, Study on pool boiling heat transfer of nano-particle suspensions on plate surface, J. Enhanc. Heat Transf. 14 (3) (2007) 223e231. [11] J.P. Tu, N. Dinh, T. Theofanous, An experimental study of nanofluid boiling heat transfer, in: Proceedings of 6th International Symposium on Heat Transfer, Beijing, China, 2004. [12] D. Wen, Y. Ding, Experimental investigation into the pool boiling heat transfer of aqueous based c-alumina nanofluids, J. Nanoparticle Res. 7 (2) (2005) 265e274. [13] S. Witharana, Boiling of Refrigerants on Enhanced Surfaces and Boiling of Nanofluids, Ph.D. thesis, The Royal Institute of Technology, 2003. [14] B.H. Truong, Determination of Pool Boiling Critical Heat Flux Enhancement in Nanofluids, Undergraduate Thesis, MIT, May 2007. [15] H.S. Ahn, V. Sathyamurthi, D. Banerjee, Pool boiling experiments on a nanostructured surface, IEEE Compon. Packag. Tech. 32 (2009) 156e165. [16] M.M. Sarafraz, S.M. Peyghambarzadeh, S.A. Alavi Fazel, N. Vaeli, Nucleate pool boiling heat transfer of binary nano mixtures under atmospheric pressure around a smooth horizontal cylinder, Rperiodica Polytech. Chem. Eng. 56 (2) (2012) 1e7. [17] M.M. Sarafraz, S.M. Pyghambarzadeh, Nucleate pool boiling heat transfer to Al2O3-water and TiO2-water nanofluids on horizontal smooth tubes with dissimilar homogeneous materials, Chem. Biochem. Eng. Q. 26 (3) (2012) 199e206. [18] J.S. Coursey, J. Kim, Nanofluid boiling: the effect of surface wettability, Int. J. Heat Fluid Flow. 29 (2008) 1577e1585. [19] I.C. Bang, S.H. Chang, Boiling heat transfer performance and phenomena of Al2O3ewater nano-fluids from a plain surface in a pool, Int, J. Heat Mass Transf. 48 (12) (2005) 2420e2428. [20] V. Sajith, R.M. Madhusoodanan, C.B. Sobhan, An experimental investigation of the boiling performance of water-based nanofluids, in: Proceedings of the Micro/Nanoscale Heat Transfer International Conference, Jan. 2008. [21] 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 Transf. 12 (2007) 4105e4116. [22] M. Chopkar, A.K. Das, I. Manna, P.K. Das, Pool boiling heat transfer characteristics of ZrO2ewater nanofluids from a flat surface in a pool, Heat Mass Transf. (2007). [23] J.H. Kim, K.H. Kim, S.M. You, Pool boiling heat transfer in saturated nanofluids, in: Proceedings of IMECE 2004, Anaheim, CA, 13e19 November 2004, pp. 13e19. [24] 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). [25] S.M. You, J.H. Kim, K.H. Kim, Effect of nanoparticles on critical heat flux of water in pool boiling heat transfer, Appl. Phys. Lett. 83 (2003) 3374e3376. [26] P. Vassallo, R. Kumar, S.D. Amico, Pool boiling heat transfer experiments in silicaewater nano-fluids, Int. J. Heat Mass Transf. 47 (2004) 407e411. [27] S.A. Alavi Fazel, M. Mahboobpour, Pool boiling heat transfer in monoethyleneglycol aqueous solutions, Exp. Therm. Fluid Sci. 48 (2013) 177e183. [28] S.J. Kline, F.A. McClintock, Describing uncertainties in single-sample experiments, Mech. Eng. 75 (1) (1953) 3e12. [29] M.M. Sarafraz, F. Hormozi, Scale Formation, Subcooled Flow, Boiling heat transfer of CuO-water nanofluid inside the vertical annulus, Exp. Therm. Fluid Sci. 52 (2014) 205e214. [30] B. D. Cullity, S. R. Stock, Text Book of Elements of X-ray Diffraction. third ed., Prentice-Hall Publiser, pp. 170 (Chapter 5). [31] S.M. Peyghambarzadeh, S.H. Hashemabadi, M. Narakia, Y. Vermahmoudi, Experimental study of overall heat transfer coefficient in the application of dilute nanofluids in the car radiator, Appl. Therm. Eng. 52 (2013) 8e16. [32] M. Kole, T.K. Dey, Investigations on the pool boiling heat transfer and critical heat flux of ZnO-ethylene glycol nanofluids, Appl. Therm. Eng. 37 (2012) 112e119. [33] W.M. Rohsenow, A method of correlating heat transfer data for surface boiling of liquids, Trans. ASME 74 (1952) 969. [34] I.L. Pioro, Experimental evaluation of constants for the Rohsenow pool boiling correlation, Int. J. Heat Mass Trans. 31 (1992) 2003e2013. [35] D. Gorenflo, Pool boiling, Dusseldorf, Germany, in: VDI Heat Atlas, 1993 (chapter Ha). [36] H. Kim, M. Kim, Experimental study of the characteristics and mechanism of pool boiling CHF enhancement using nanofluids, Heat Mass Transf. 45 (7) (2007) 991e998. [37] J.H. Kim, K.H. Kim, S.M. You, Pool boiling heat transfer in saturated nanofluids, in: ASME Conference Proceedings, Anaheim, CA, 2004, pp. 621e628. [38] R.I. Elghanam, M.M.E.L. Fawal, R. Abdel Aziz, M.H. Skr, A. Hamza Khalifa, Ain Shams Eng. J. 2 (2011) 195e209. [39] V. Wasekar, R. Manglik, A review of enhanced heat transfer in nucleate pool boiling of aqueous surfactant and polymeric solutions, J. Enhanc. Heat Trans. 6 (1999) 135e150.
M.M. Sarafraz, F. Hormozi / International Journal of Thermal Sciences 90 (2015) 224e237 [40] D. Wen, B. Wang, Effects of surface wettability on nucleate pool boiling heat transfer for surfactant solutions, Int. J. Heat Mass Trans. 45 (2002) 1739e1747. [41] W. Frost, C. Kippenhan, Bubble growth and heat transfer mechanisms in forced convection boiling of water containing a surface active agent, Int. J. Heat Mass Trans. 10 (1967) 931e949. [42] L. Cheng, D. Mewes, A. Luke, Boiling phenomena with surfactants and polymeric additives: a state-of-the-art review, Int. J. Heat Mass Trans. 50 (2007) 2744e2771. [43] A.W. Adamson, Physical Chemistry of Surfaces, Interscience Publication, New York, 1967. [44] M.G. Kang, Effect of surface roughness on pool boiling heat transfer, Int. J. Heat Mass Trans. 43 (2000) 4073e4085. [45] P. Walker, R. Sheikholeslami, Assessment of the effect of velocity and residence time in CaSO4 precipitating flow reaction, Chem. Eng. Sci. 58 (2003) 3807e3816.
237
[46] S.M. Peyghambarzadeh, A. Vatani, M. Jamialahmadi, Application of asymptotic model for the prediction of fouling rate of calcium sulfate under subcooled flow boiling, Appl. Therm. Eng. 39 (2012) 105e113. [47] V. Nikkhah, M.M. Sarafraz, F. Hormozi, S.M. Peyghambarzadeh, Particulate fouling of CuOewater nanofluid at isothermal diffusive condition inside the conventional heat exchanger-experimental and modeling, Exp. Therm. Fluid Sci. 60 (2015) 83e95. [48] M.M. Sarafraz, F. Hormozi, Nucleate pool boiling heat transfer characteristics of dilute Al2O3eethyleneglycol nanofluids, Int. Commun. Heat Mass Trans. 58 (2014) 96e104. [49] M.M. Sarafraz, F. Hormozi, Convective boiling and particulate fouling of stabilized CuO-ethylene glycol nanofluids inside the annular heat exchanger, Int. Commun. Heat Mass Trans. 53 (2014) 116e123.