- Email: [email protected]
Time effect on wetting transition of smart surface and prediction of the wetting transition for critical heat flux in pool boiling
International Journal of Heat and Mass Transfer 114 (2017) 735–742
Contents lists available at ScienceDirect
International Journal of Heat and Mass Transfer journal homepage: www.elsevier.com/locate/ijhmt
Time effect on wetting transition of smart surface and prediction of the wetting transition for critical heat flux in pool boiling Jin Man Kim a,⇑, TaeJoo Kim a, Dong In Yu b, Moo Hwan Kim c, Kiyofumi Moriyama c, Hyun Sun Park c a
Neutron Science Center, Korea Atomic Energy Research Institute, Daejeon 34057, Republic of Korea Maritime Reactor Development Center, Korea Atomic Energy Research Institute, Daejeon 34057, Republic of Korea c Division of Advanced Nuclear Engineering, Pohang University of Science and Technology, Pohang 37673, Republic of Korea b
a r t i c l e
i n f o
Article history: Received 5 April 2017 Received in revised form 22 June 2017 Accepted 26 June 2017
Keywords: Critical heat flux Wettability Time effect TiO2 Pool boiling
a b s t r a c t A smart surface that is a TiO2-coated surface (TCS) is a hydrophobic surface initially, but becomes a hydrophilic surface when heated. Therefore, such a surface can be used to enhance both boiling heat transfer (BHT) and critical heat flux (CHF) in pool boiling. In the present study, the time effect of the wetting transition of TCS was focused on. The CHF on TCS was enhanced more when the holding time of the heat flux in high-temperature regime was increased. By observing changes in contact angles on TCS through heat treatment in air, it was found that the wetting transition was affected not only by the temperature, but also by the time. Thus, a variation of the receding contact angle was correlated in the form of an exponential function. The suggested empirical correlation includes temperature and time, and it describes the transition of the receding contact angle. The correlation was also used to predict the CHF on TCS in pool boiling. As a result, CHFs on TCS could be explained using the correlation. Ó 2017 Elsevier Ltd. All rights reserved.
1. Introduction Boiling is an efficient heat transfer method because it uses latent heat. For this reason, many applications have used the boiling system. However, the critical heat flux (CHF) imposes a limitation on the practically available boiling regime. When CHF is reached, a vapor blanket forms on the heated surface, resulting in an interruption in phase changes and a deterioration in the boiling heat transfer (BHT). Then, the heating substrate melts due to the reduced heat removal capacity. In this respect, for safe and efficient heat transfer systems, it is important to improve CHF and BHT. As technology has advanced, many techniques have been suggested to improve the CHF and BHT. To be specific, the surface modification technique, due to its applicability into real systems, e.g., machining, chemical etching, coating of thin layers, and deposition of particles, is one of the most promising methods for the improvement of boiling performance. Furthermore, the effect of nanoparticles in pool boiling has been reported on substantially since the advent of nanofluids and their applications [1–9]. Bang et al. [5] investigated the effect of alumina (Al2O3) nanofluid on the pool boiling characteristics. The authors prepared
⇑ Corresponding author. E-mail address: [email protected] (J.M. Kim). http://dx.doi.org/10.1016/j.ijheatmasstransfer.2017.06.114 0017-9310/Ó 2017 Elsevier Ltd. All rights reserved.
alumina nanofluids that had different volume concentrations. In pool boiling, the nanofluids enhanced the CHF, but reduced the BHT. In addition, the higher the concentration of the nanofluid, the lower the BHT. Kim et al. [6] used alumina, zirconia (ZrO2), and silica (SiO2) nanofluids to examine the effect of nanoparticles on the pool boiling characteristics. When the authors fixed the volume concentration at 0.01%, the CHF was enhanced for all the nanofluids. It was believed that the improved wettability via nanoparticle deposition contributed to the enhancement of CHF. However, the nanofluids lowered the BHT. Kim et al. [7] utilized alumina and titania (TiO2), and Stutz et al. [8] adopted iron oxide (Fe2O3) nanofluids for the enhancement of CHF in pool boiling. In their studies [7,8], they also agreed that the improved wettability due to the nanoparticle deposition was the main cause of the enhancement of CHF. The previous studies [6–8] showed that the wettability had an important role in enhancing CHF, regardless of the nanoparticle material. In this respect, different surface modifications were suggested with a focus on the wettability. Ahn et al. [10] fabricated nano-micro structures on zircaloy-4 using an anodic oxidation. The authors controlled the anodization time of zircaloy-4 so that the static contact angle on the surface decreased from 49.3 to 0°, while the modified surface enhanced the CHF by 90%. Chu et al. [11] fabricated hierarchical surfaces by depositing nanoparticles on micropillar posts. In pool boiling, the hierarchical surface showed a CHF enhancement of 200% in comparison to that of a
736
J.M. Kim et al. / International Journal of Heat and Mass Transfer 114 (2017) 735–742
Nomenclature a A b B C hlv k q00 R SCS t T T Ts T1 TCS
coefficient constant coefficient constant experimental constant latent heat (kJ/kg) rate constant (/min) heat flux (kW/m2) constant SiO2-coated surface time (min) temperature (K) normalized temperature, T (°C)/200 °C surface temperature (K) ambient temperature (K) TiO2-coated surface
smooth surface. In their study, the higher the roughness, the higher the CHF. However, BHT decreased as the roughness increased. The researchers believed that the thermal resistance of the hierarchical surfaces was the reason for the deterioration of BHT. Even though nanoparticle deposition enhanced the CHF due to the improved wettability, some studies [5–8,11] have identified a deterioration of BHT. However, other studies [9,12] have overcome the deterioration of BHT. Kim et al. [9] used microelectromechanical systems (MEMS) to manufacture nanostructures, microstructures, and nano-micro combined structures on silicon substrates. In pool boiling, the nano-micro structured surface showed a CHF enhanced by more than 107%, which was contributed to the increased wettability. In particular, the structures improved not only the CHF, but also the BHT. Betz et al. [12] focused on wetting control without structural changes. The authors fabricated mixed wetting surfaces: one type had hydrophobic hexagon patterns on a hydrophilic surface, while another surface had hydrophilic hexagon patterns on a hydrophobic surface. The mixed wetting surface improved the BHT and CHF by 100% and 65%, respectively, in comparison to those for a plain surface. As reported in previous studies [9,12], enhancement of both BHT and CHF requires intricate technology because there is a trade-off between BHT and CHF when the wettability is modulated. In studies [13,14], nanometrically smooth hydrophobic surfaces showed better BHT than did hydrophilic surfaces in pool boiling; hydrophobic surfaces have been found to be advantageous for high BHT, but are known to have low CHF. On the other hand, hydrophilic surfaces have showed a BHT lower than that of hydrophobic surfaces, but can have a higher CHF. In this regard, Kim et al. [15] suggested a smart surface, which is a TiO2-coated surface (TCS), to enhance both BHT and CHF in pool boiling. The authors noted that the properties of hydrophobicity and hydrophilicity have different effects on the boiling performance. So, they focused on the wetting transition of TCS. In several studies [16–18], TiO2 was found to have a wetting transition that is dependent on the temperature. TiO2 initially shows a hydrophobic characteristic, but it becomes hydrophilic when heated. Kim et al. [15] conducted pool boiling experiments using TCS and observed that TCS increased not only the BHT, but also the CHF. They reported that this result was contributed to the wetting transition of TCS. In addition, it was suggested that time was an important factor in the change of wettability, and could lead to an additional enhancement of CHF.
Greek symbols / orientation of heating surface (°) h static contact angle (°) h normalized receding contact angle, hr =90 h0 contact angle constant (°) hr receding contact angle (°) qv vapor density (kg/m3) r surface tension (N/m) or Stefan-Boltzmann constant (5.67 108 W/m2 K4) Subscripts a the coefficient of a b the coefficient of b l liquid n coefficient indication v vapor lv phase change from liquid to vapor CHF CHF
The present study focused on the time effect of the wetting transition of TCS. Through heat treatment with changes in the treatment temperature and time, the variation of the receding contact angle was examined, and an empirical correlation was provided. In addition, the empirical correlation was combined with Kandlikar’s correlation [19] to predict the CHF trend on TCS. 2. Experiments 2.1. Surface fabrication A 500 lm thick silicon wafer was used as a base substrate. Since the silicon surface had a 1 nm roughness, we considered that the roughness effect could be excluded based on Hsu’s theory [20]. For the heating of the substrate, a heating element was fabricated on the back of the substrate, as shown in Fig. 1. On one side, an SiO2 thin film with a thickness of 500 nm was deposited for electrical insulation. A photoresist (GXR601, Electronic Materials) was coated on the insulated silicon wafer using a spin coater (ATIS, Midas System) at 3000 rpm. Then, the coated photoresist (PR) was exposed to ultraviolet (UV) light through a patterned mask using a mask aligner (MA6, Suss Microtec). The exposed wafer was immersed in tetramethylammonium hydroxide (TMAH) solution, and so a reversed pattern of the PR was made on the wafer. On this surface, using an electron beam evaporator (E-beam), a platinum (Pt) layer with a thickness of 120 nm was deposited. When the Pt coated wafer was immersed in acetone, only the Pt pattern remained after the residual PR layer was removed. As a result, a heating element that has an effective heating area of 10 10 mm was fabricated. On the other side of the substrate, thin films for boiling surfaces were deposited. A 500 nm-thick SiO2 thin film was deposited on the heating surface as a reference surface; a 200 nm-thick TiO2 thin film was deposited to investigate the wetting transition. A TiO2 layer was deposited using radio frequency sputtering (RF sputtering) at room temperature conditions (25 °C). 2.2. Experimental facilities For the investigation of CHF on TCS, pool boiling facilities were constructed. The main pool was made of stainless steel pipe (304 L) that could hold 12 L of deionized (DI) water as a working fluid. In
737
J.M. Kim et al. / International Journal of Heat and Mass Transfer 114 (2017) 735–742
Fig. 1. Fabrication procedures of the heating element.
pool boiling under atmospheric condition, CHF occurred at 150 °C. However, a complete wetting transition was observed near 200 °C, as was mentioned in a previous study [15]. This will be briefly reviewed in Section 3.1. In this respect, the saturated temperature of the working fluid was increased using relief valves (VRV1MH, HY-LOK), which opened at 2.0, 3.3, and 4.1 bar. When the working fluid was boiled by an immersion heater (600 W), a relief valve was opened to discharge excess steam at the set pressure. According to the set value of the relief valve, the saturated temperature and pressure were determined during boiling. The experimental pressures were 1.0, 2.0, 3.3, and 4.1 bar; the saturated temperatures were 99.6, 120.1, 136.8, and 144.9 °C, respectively, at each pressure condition. The overall system was constructed as depicted in Fig. 2. The heater was connected to the data acquisition system (34970a, Agilent) and the power supply (XG150, Sorensen) via a shunt resistor. The power supply controlled the input voltage to manipulate the heat flux, and the data acquisition system collected data for the voltage and current of the heater and shunt resistor. The shunt resistor maintained constant resistance in the thermostat and the resistance was used to calculate the circuit current. The saturated pressure was calculated using the correlation suggested by Wagner
et al. [21]. Details on the experimental ranges and uncertainties were listed in a previous report [15]. 2.3. Experimental procedures All experiments were conducted under saturated condition. Using an immersion heater, the working fluid was boiled in advance for 6 h. The first 3 h was used to degas the DI water at the 1.0 bar condition. After degassing, only one relief valve was opened and we waited another 3 h to achieve the saturated condition at the set pressure of the relief valve. The saturated condition was maintained for more than 10 min, and the heat flux was then increased step-by-step. At each desired level of heat flux, the heat flux was held for 2 min to obtain averaged data. During the experiments, CHF was defined, and the power supply was shut down when an abrupt increase of the heating element temperature was detected. 2.4. Data reduction The wall temperature denotes the surface temperature where the surface contacts the working fluid. In the present study, the heating element was on the backside of the heating substrate and exposed to air. Therefore, we estimated heat losses to air and calculated the wall temperature based on the heating element temperature. Heat can be transferred to the air by three mechanisms: convection, radiation, and conduction. The surrounding circumstances of the heating element are illustrated in Fig. 3. Air heated by the heating, because it is heated from the top, is trapped and stationary in the space below the heating element. Thus, the natural convection was negligible and there was no forced convection. The radiation heat transfer of an ideal black body can be expressed by the following equation:
q00 ¼ rðT 4s T 41 Þ
ð1Þ
where r is the Stefan-Boltzmann constant (5.67 10 W/m K4), T s is the surface temperature (K), and T 1 is the ambient temperature (K). With the assumption that SiO2 is an ideal black body, the heat loss to the air by radiation was calculated using the experimental data for SiO2. The calculated heat loss to the air by radiation was below 1.0%, during which the heat flux was higher than 200 kW/m2. For the real surface, emissivity should be multiplied to Eq. (1). Then, the heat loss will be much lower than the calculated value because the emissivity is less than one for the real surface. 8
Fig. 2. Schematic image of the overall system for pool boiling.
2
738
J.M. Kim et al. / International Journal of Heat and Mass Transfer 114 (2017) 735–742
Fig. 3. Cross-sectional view of heater assembly: rendering image (left) and schematic image (right).
Additionally, the maximum heat loss to the air by conduction was 0.43%. As a result, the heat loss to the air was ignored. For the calculation of the wall temperature, we used Fourier’s law, with a thermal conductivity of 130 W/m K for the silicon substrate. When it was assumed that the heat was transferred in one direction from the heating element to the boiling surface, the wall temperature was easily obtained. After each experiment, the wall temperature was calculated using the method and the experimental data.
3. Results and discussion 3.1. Review of the effect of temperature on the wetting transition of TCS In the previous study [15], in the heat treatment test, TCS showed a wetting transition, as shown in Fig. 4. For the heat treatment test, the SiO2-coated surface (SCS) and TCS were heated in the oven (air condition) at a specific temperature for 10 h. The temperature ranged from 100 to 200 °C; this range corresponded to the experimental temperature range of the heating surface. After the heat treatment was conducted, the test surface was taken out of the oven and cooled at a room temperature of 25 °C. The contact angles on the surfaces were measured using a droplet goniometer
(SmartDrop, FemtoFab) with 2 ll water droplets. Because the instrument has an accuracy of ±0.1°, it was believed that the contact angles were precisely measured. A surface that had been heated once in the oven was never reused for another test. After heat treatment was conducted at 100 °C, SCS and TCS showed contact angles of 68 and 83°, respectively. As the heat treatment temperature increased, a decrement of the static contact angle on TCS was observed, while the contact angle on SCS maintained its initial value (in Fig. 4(a)). After the heat treatment at 180 °C, the static contact angle on TCS was smaller than that on SCS. With an increase in the heat treatment temperature to 200 °C, TCS became hydrophilic in comparison to SCS; TCS showed a static contact angle of 33°, while SCS showed a static contact angle of 62°. The receding contact angles also showed tendencies similar to the variation of the static contact angles, as shown in Fig. 4(b). The receding contact angle on TCS changed from 53 to 14° with increasing of the heat treatment temperature, while SCS showed an averaged receding contact angle of 42°. The wetting transition can be explained by the changes in the crystal structure. There have been previous studies [16,18,22–26] on the wetting transition induced by UV illumination. Sun et al. [16] investigated the water contact angles on TiO2 thin film before and after UV illumination. The authors confirmed that the TiO2 thin film became hydrophilic when it was exposed to the UV light. They explained that the wetting transition was due to the changes in the
Fig. 4. (a) Static contact angles and (b) receding contact angles after heat treatment for 10 h [15].
J.M. Kim et al. / International Journal of Heat and Mass Transfer 114 (2017) 735–742
crystal structure. When the surface was treated under UV light, it had oxygen vacancies. If a water molecule contacts an oxygen vacancy, the molecule will be dissociatively adsorbed onto the vacancy. This induced the wetting transition. Sakai et al. [23] also investigated hydrophilic conversion by UV illumination. They fabricated TiO2 film using a dip-coating method. On the TiO2 coated substrate, they conducted UV irradiation at various UV intensities and observed a conversion to a hydrophilic surface, resulting in a contact angle smaller than 10°. They also explained that the reconstruction of the hydroxyl groups occurred at the oxygen vacancies when UV light illuminated the TiO2 film. 3.2. Time effect of the static contact angle For the clarification of the time effect for the wetting transition, we examined the time effect of the wetting transition for TCS through heat treatment tests; this was done because the authors in the previous study [15] showed that an increase of the holding time of the heat flux at high temperature regime affected the additional enhancement of CHF on TCS. TCSs with dimensions of 10 10 mm were thermally treated at 200 °C in an oven while the heat treatment time was varied from 0 to 10 h. One TCS was cooled at 25 °C after 1 h of heat treatment and we measured the static contact angle. The next TCS was also cooled at 25 °C after heat treatment for 2 h, and the static contact angle was measured. Experiments were performed using the same procedure at intervals of 1 h. Results are presented in Fig. 5 for the time control tests in the heat treatment. Initially, SCS showed a static contact angle of 57°, but TCS showed a relatively hydrophobic characteristic, with a contact angle of 83°. As the heat treatment time increased up to 10 h, TCS became hydrophilic, with a contact angle of 23°, while SCS did not show any significant change in contact angle. These results indicate that a sufficient amount of time is needed to achieve the wetting transition of TCS. The time tendency for the reduction of the contact angle was also reported on in previous studies [16,18,22–25]. Stevens et al. [24] fabricated TiO2 surfaces to examine the alteration of the wettability by UV irradiation. As the illumination time of UV light increased, the advancing contact angle was found to decrease. Borras et al. [25] also fabricated TiO2 thin films using metal-organic chemical vapor deposition (MOCVD). They controlled the UV illumination time and measured the water contact angle on the surfaces. The contact angle decreased as the illumination time increased. In addition, the contact angle became small when the UV intensity increased. Even though the previous studies [24,25] used UV illumination to induce the wetting transition, the
100
739
tendency of the wetting transition with increasing illumination time of UV light showed a form of variation similar to that found in the present study. Since Kandlikar [19] proposed a CHF model that reflected a wetting effect using the receding contact angle, the receding contact angle has been considered as an important parameter for the prediction of CHF. This means that measuring the receding contact angle on the heating surface is essential to understanding boiling phenomena and, in particular, for predicting CHF. In this respect, we investigated the variation of the receding contact angle with changes in heat treatment temperature and time. For the heat treatment, the temperature ranged from 140 to 200 °C because, in the present study, CHF occurred at a temperature higher than 140 °C; the time was adjusted from 2 h to 10 h at 2 h intervals. In detail, TCSs were heated at 140 °C in the oven. The first sample was heated for 2 h and its receding contact angle was measured on the surface after it was cooled to a room temperature of 25°. The second sample was heated for 4 h and the receding contact angle was measured using the same procedure. In total, heat treatments for 20 cases were conducted at various temperatures and times, and each case was repeated more than 3 times. The measured receding contact angles on TCS after the heat treatment are displayed as solid points in Fig. 6. As expected, decrements in the receding contact angle were observed as the temperature increased. However, even though the heat treatment occurred at 200 °C, the decrease of the receding contact angle was not significant when the heat treatment time was short. For the heat treatment at 200 °C, the receding contact angle was about 23° after heat treatment for 2 h, but it became 14° after heat treatment for 10 h. As the heat treatment time increased, the receding contact angles approached certain asymptotic values as a result of heat treatment for 10 h. The explanation for the solid lines in the graph is treated in Section 3.3. 3.3. Empirical correlation of the variation of the receding contact angle For the future application of TCS, a prediction of the wetting transition is essential. For this reason, certain attempts have been made to predict the change of the contact angle of TiO2. Stevens et al. [24] deposited 100 nm TiO2 thin film on a silicon wafer. The authors exposed the surface to UV light with changes in irradiation time and measured the advancing contact angle on the surface. When the surface was exposed to UV light, the advancing
SCS TCS
90
Contact angle (°)
80 70 60 50 40 30 20 10 0
0
2
4
6
8
10
12
Heat treatment time (h) Fig. 5. Static contact angles on SCS and TCS as heat treatment time increases at 200 °C.
Fig. 6. Receding contact angles after heat treatment with changes of temperature and time; solid points are the experimental data; lines are predictions based on empirical correlation.
740
J.M. Kim et al. / International Journal of Heat and Mass Transfer 114 (2017) 735–742
Fig. 7. Fitting of coefficients as a function of temperature.
contact angle decreased. Additionally, the advancing contact angle decreased more as the irradiation time increased. They determined that the decrease in contact angle after UV irradiation could be expressed in exponential form as follows:
h ¼ h0 expðkUV tÞ
ð2Þ
where h0 is the contact angle (°), kUV is the rate constant (/min), and t is the time (min). The proposed equation fitted the experimental data well. Borras et al. [25] also identified a decrease in the contact angle on a TiO2 coated surface after UV irradiation. The authors used the same form that was suggested by Stevens et al. [24] to express the decline of the contact angle for the TiO2 coated surface. In two previous studies [24,25], the exponential form was found to represent the change of the contact angle on TiO2 well. In the present study, we also tried to show the variation of the receding contact angle on TCS using the function in Eq. (3),
h ¼ a þ b expðktÞ
ð3Þ
where h is a normalized receding contact angle, k is a rate constant (min1), t is a time (min), and other values are coefficients; the coefficients are functions of the temperature. In Eq. (3), the normalized receding contact angle is determined by the combination of a and b when k is fixed. For the transformation of the crystal structure of TiO2, the rate constant can be expressed using Arrhenius equation [27]. Because the Arrhenius equation has been used in many solid state reactions [28,29], we also adopted the form of Arrhenius equation to depict the rate constant with the variation of the temperature. This form shows the agreement of the relationship between the rate constant and temperature in the previous studies [30,31]; the rate constant increases with increasing temperature. Based on the previous studies [27–31], we used Eq. (4), in which T is an absolute temperature (K).
k ¼ 0:0118 expð12:84=TÞ
ð4Þ
In addition, to determine a and b in Eq. (3), we postulated that these factors can be represented by Eq. (5), which is a function of the temperature:
aðor bÞ ¼ A þ B expðRT Þ
ð5Þ
where A, B, and R are constants, and T⁄ is the normalized temperature divided by 200 °C. In this correlation, we determined constants to fit our data; the results are shown in Fig. 7. The value of a decreased as the temperature increased and the correlation in Fig. 7(a) matched the data points, with an R2 of 0.9811. The value of b increased as the temperature increased and matched the fitting
Table 1 Fitting constants for the coefficients. Constants
Values
Aa Ba Ra Ab Bb Rb
0.9278 0.1311 1.8205 0.2933 0.1182 1.8997
line, with an R2 of 0.9817. The values of the constants used in Eq. (5) are listed in Table 1. As a final step, the receding contact angle was predicted by substituting Eqs. (4) and (5) for Eq. (3). Since the final form of the correlation is a function of time and temperature, the transition of the receding contact angle with changing time and temperature can be described well. In Fig. 6, the curves were plotted using the correlation in Eq. (3), with a value of R2 of 9743. The correlation matched well with the measured receding contact angles after heat treatment was conducted at various temperatures and treatment times. 3.4. Critical heat flux In the previous study [15], the CHF on TCS was enhanced more when the heat flux holding time was increased to 20 min in the high-temperature regime. The authors postulated that the time also contributed to the enhancement of CHF. In this situation, there was no appropriate model to predict the CHF on TCS because it depends on the temperature and the time. Even though Kandlikar [19] suggested a CHF prediction model, represented by Eq. (6), that includes the receding contact angle, it also cannot predict the CHF on TCS due to the wetting transition of TCS.
q00CHF ¼ Chlv q1=2 v
1=2 1 þ cos hr 2 p þ ð1 þ cos hr Þ cos / 16 p 4
ð6Þ
In Fig. 8, a comparison of CHFs on TCS is displayed: the blue circles are CHFs in the normal operation, the red triangles are CHFs in the time effect tests, and the black solid line is a CHF prediction line on TCS using the model suggested by Kandlikar [19]. For each pressure condition, the fitting constant was multiplied to fit the reference data, SCS, which was explained in the previous study [15]. The correlation overestimated the experimental data because an asymptotic receding contact angle under CHF condition was used without consideration of the time effect in the wetting transition. Furthermore, due to the time effect of the wetting transition
Critical heat flux (kW/m2)
J.M. Kim et al. / International Journal of Heat and Mass Transfer 114 (2017) 735–742
2200 2000 1800 1600 1400 1200 1000 800 600 400 200 0
741
3.5. Time constant
Kandlikar [19] Exp. (normal operation) Exp. (time effect)
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Pressure (bar) Fig. 8. Comparison of CHFs on TCS with prediction model suggested by Kandlikar [19].
We observed the wetting transition via the heat treatment tests using the oven in air condition. Also, an additional enhancement of CHF on TCS was identified when the holding time of the heat flux increased. The enhancement of the CHFs on TCS was contributed to the wetting transition to the hydrophilic surface because the roughness and thermal resistance effects were negligible [15]. However, the time rates of the wetting transition on TCS were different for air and water. In Fig. 9, the time, represented by the solid line, denotes the time duration in the case of heat treatment in air. The solid data points (holding heat fluxes for 2 min) correspond to the CHF calculated using the receding contact angle obtained from the heat treatment for 1.5 h. The hollow data points (holding a heat flux for 20 min in high-temperature regime) correspond to the CHF calculated using the receding contact angle for 4 h heat treatment. It seems that the heat treatment times of 1.5 and 4 h in air correspond to heat flux holding times of 2 min and 20 min in water. The time rate of the wetting transition seems to have been affected by the surrounding environment. Therefore, the relation between environment and the time constant should be clarified in the future. 4. Conclusions
Fig. 9. CHF prediction with consideration of the wetting transition of TCS; the solid points are the CHFs on TCS in normal tests; hollow points are the CHFS on TCS in time effect tests; solid lines are CHF predictions for changing t in Eq. (3).
for TCS, the CHFs on TCS were different for normal operation and for the time effect test, as was previously mentioned. Thus, to reflect the wetting transition effect of TCS, we substituted Eq. (3) for Eq. (6). The prediction results using Eqs. (3) and (6) are shown in Fig. 9; the CHF ratio was defined as the proportion of CHF on TCS to CHF on SCS. The solid points are the CHF ratios on TCS in the normal tests, the hollow points are the CHF ratios on TCS in the time effect tests, and the solid lines are the CHF ratios of TCS predicted using the empirical correlation with changing T and t. In the correlation, time is the heat treatment time assuming the heat treatment was performed in air, not in water. When we assumed the heat treatment time was zero, the prediction line underestimated the experimental CHFs on TCS because TCS would be hydrophobic when the heat treatment time was zero. When we input a 1.5 h heat treatment time into the correlation, the prediction line approximated the CHFs in normal operation. In this case, the maximum error was 7.99% at a 3.3 bar condition. When we assumed that the heat treatment time was 4 h, the prediction line was quite well matched to the experimental CHFs in the time effect tests, which maintained heat flux for 20 min in the high-temperature regime. In this case, the maximum error was 2.34% at a 2.0 bar condition. Therefore, the experimental CHFs on TCS can be explained using Kandlikar’s model [19], combining the empirical correlation.
The time effect of the wetting transition was examined using a smart surface (TCS). In the present study, it was confirmed that, in the air condition, the wettability of TCS was affected not only by the heat treatment temperature but also by the time. From the perspective of predicting the CHF on TCS in pool boiling, the variation of the receding contact angle was analyzed through heat treatment tests. The receding contact angle decreased with increasing heat treatment temperature and time. This was expressed as an exponential function having parameters of temperature and time. By determining the constants in the empirical correlation, the calculated receding contact angles were found to be well matched to the measured receding contact angles on TCS. Additionally, the CHFs on TCS were calculated by combining the empirical correlation and Kandlikar’s model [19]. When the heat treatment time was changed in the empirical correlation, the CHFs on TCS could be explained. Even if there was the wetting transition on TCS, the time constant was different according to the environment. In the future, the relation between the time rate and the environment should be studied. Conflict of interest The auth ors declare that they have no conflict of interests for financial. Acknowledgements This work was supported by National Research Foundation of Korea (NRF) grants funded by the Korean Government (MSIP) (2015M2A8A2074795) and the Korean Government (MSIP) (2012M2A2A6004262). The fabrication of the test sections was supported by Ulsan National Institute of Science and Technology (UNIST), and National Nanofab Center (NNFC). References [1] 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) 3374–3376, http://dx.doi.org/10.1063/1.1619206. [2] H. Kim, J. Kim, M.H. Kim, Effect of nanoparticles on CHF enhancement in pool boiling of nano-fluids, Int. J. Heat Mass Transf. 49 (2006) 5070–5074, http://dx. doi.org/10.1016/j.ijheatmasstransfer.2006.07.019.
742
J.M. Kim et al. / International Journal of Heat and Mass Transfer 114 (2017) 735–742
[3] S.J. Kim, I.C. Bang, J. Buongiorno, L.W. Hu, Effects of nanoparticle deposition on surface wettability influencing boiling heat transfer in nanofluids, Appl. Phys. Lett. 89 (2006) 153107. [4] C.-C. Hsu, P.-H. Chen, Surface wettability effects on critical heat flux of boiling heat transfer using nanoparticle coatings, Int. J. Heat Mass Transf. 55 (2012) 3713–3719. [5] I.C. Bang, S. Heung Chang, S.H. Chang, Boiling heat transfer performance and phenomena of Al2O3–water nano-fluids from a plain surface in a pool, Int. J. Heat Mass Transf. 48 (2005) 2407–2419, http://dx.doi.org/10.1016/j. ijheatmasstransfer.2004.12.047. [6] 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. 50 (2007) 4105–4116, http://dx.doi.org/10.1016/j.ijheatmasstransfer. 2007.02.002. [7] H. Kim, H.S. Ahn, M.H. Kim, On the mechanism of pool boiling critical heat flux enhancement in nanofluids, J. Heat Transfer. 132 (2010) 61501, http://dx.doi. org/10.1115/1.4000746. [8] B. Stutz, C.H.S. Morceli, M. de F. da Silva, S. Cioulachtjian, J. Bonjour, Influence of nanoparticle surface coating on pool boiling, Exp. Therm. Fluid Sci. 35 (2011) 1239–1249, http://dx.doi.org/10.1016/j.expthermflusci.2011.04.011. [9] S. Kim, H.D. Kim, H.D. Kim, H.S. Ahn, H. Jo, J. Kim, M.H. Kim, Effects of nanofluid and surfaces with nano structure on the increase of CHF, Exp. Therm. Fluid Sci. 34 (2010) 487–495, http://dx.doi.org/10.1016/j.expthermflusci. 2009.05.006. [10] H.S. Ahn, C. Lee, H. Kim, H. Jo, S. Kang, J. Kim, J. Shin, M.H. Kim, Pool boiling CHF enhancement by micro/nanoscale modification of zircaloy-4 surface, Nucl. Eng. Des. 240 (2010) 3350–3360, http://dx.doi.org/10.1016/j.nucengdes. 2010.07.006. [11] K.-H. Chu, Y. Soo Joung, R. Enright, C.R. Buie, E.N. Wang, Hierarchically structured surfaces for boiling critical heat flux enhancement, Appl. Phys. Lett. 102 (2013) 151602. [12] A.R. Betz, J. Xu, H. Qiu, D. Attinger, Do surfaces with mixed hydrophilic and hydrophobic areas enhance pool boiling?, Appl Phys. Lett. 97 (2010) 141909. [13] H. Jo, H.S. Ahn, S. Kang, M.H. Kim, A study of nucleate boiling heat transfer on hydrophilic, hydrophobic and heterogeneous wetting surfaces, Int. J. Heat Mass Transf. 54 (2011) 5643–5652, http://dx.doi.org/10.1016/j. ijheatmasstransfer.2011.06.001. [14] B. Bourdon, P. Di Marco, R. Rioboo, M. Marengo, J. De Coninck, Enhancing the onset of pool boiling by wettability modification on nanometrically smooth surfaces, Int. Commun. Heat Mass Transf. 45 (2013) 11–15, http://dx.doi.org/ 10.1016/j.icheatmasstransfer.2013.04.009. [15] J.M. Kim, D.I. Yu, H.S. Park, K. Moriyama, M.H. Kim, Smart surface in pool boiling: thermally-induced wetting transition, Int. J. Heat Mass Transf. 109 (2017) 231–241.
[16] R.-D. Sun, A. Nakajima, A. Fujishima, T. Watanabe, K. Hashimoto, Photoinduced surface wettability conversion of ZnO and TiO2 thin films, J. Phys. Chem. B 105 (2001) 1984–1990. [17] Q. Ye, P.Y. Liu, Z.F. Tang, L. Zhai, Hydrophilic properties of nano-TiO2 thin films deposited by RF magnetron sputtering, Vacuum 81 (2007) 627–631, http://dx. doi.org/10.1016/j.vacuum.2006.09.001. [18] Z.W. Zhao, B.K. Tay, Study of nanocrystal TiO2 thin films by thermal annealing, J. Electroceram. 16 (2006) 489–493. [19] S.G. Kandlikar, A theoretical model to predict pool boiling CHF incorporating effects of contact angle and orientation, J. Heat Transfer. 123 (2001) 1071, http://dx.doi.org/10.1115/1.1409265. [20] Y.Y. Hsu, On the size range of active nucleation cavities on a heating surface, J. Heat Transfer 84 (1962) 207–213. [21] W. Wagner, A. Pruss, International equations for the saturation properties of ordinary water substance. Revised according to the international temperature scale of 1990. Addendum to J. Phys. Chem. Ref. Data 16, 893 (1987), J. Phys. Chem. Ref. Data 22 (1993) 783–787. [22] T. Watanabe, Photocatalytic activity and photo-induced wettability conversion of TiO2 thin film prepared by Sol-Gel process on a Soda-Lime glass, J. Sol-Gel Sci. Technol. 19 (2000) 71–76. [23] N. Sakai, A. Fujishima, T. Watanabe, K. Hashimoto, Quantitative evaluation of the photoinduced hydrophilic conversion properties of TiO2 thin film surfaces by the reciprocal of contact angle, J. Phys. Chem. B 107 (2003) 1028–1035. [24] N. Stevens, C.I. Priest, R. Sedev, J. Ralston, Wettability of photoresponsive titanium dioxide surfaces, Langmuir 19 (2003) 3272–3275. [25] A. Borras, C. Lopez, V. Rico, F. Gracia, A.R. Gonzalez-Elipe, E. Richter, G. Battiston, R. Gerbasi, N. McSporran, G. Sauthier, A. Borra, E. Richter, G. Battiston, Effect of visible and UV illumination on the water contact angle of TiO2 thin films with incorporated nitrogen, J. Phys. Chem. C 111 (2007) 1801–1808. [26] S. Yamauchi, Y. Imai, Plasma-assisted chemical vapor deposition of TiO2 thin films for highly hydrophilic performance, Cryst. Struct. Theory Appl. 2 (2013) 1–7. [27] C.N.R. Rao, Kinetics and thermodynamics of the crystal structure transformation of spectroscopically pure anatase to rutile, Can. J. Chem. 39 (1961) 498–500. [28] A.K. Galwey, M.E. Brown, A Theoretical Justification for the Application of the Arrhenius Equation to Kinetics of Solid State Reactions (mainly ionic crystals), in: Proc. R. Soc. London A Math. Phys. Eng. Sci., The Royal Society, 1995, pp. 501–512. [29] A.K. Galwey, M.E. Brown, Application of the Arrhenius equation to solid state kinetics: can this be justified?, Thermochim Acta. 386 (2002) 91–98. [30] J. Yu, X. Zhao, Q. Zhao, Photocatalytic activity of nanometer TiO2 thin films prepared by the sol–gel method, Mater. Chem. Phys. 69 (2001) 25–29. [31] J. Yu, X. Zhao, Q. Zhao, Effect of surface structure on photocatalytic activity of TiO 2 thin films prepared by sol-gel method, Thin Solid Films. 379 (2000) 7–14.
Contents lists available at ScienceDirect
International Journal of Heat and Mass Transfer journal homepage: www.elsevier.com/locate/ijhmt
Time effect on wetting transition of smart surface and prediction of the wetting transition for critical heat flux in pool boiling Jin Man Kim a,⇑, TaeJoo Kim a, Dong In Yu b, Moo Hwan Kim c, Kiyofumi Moriyama c, Hyun Sun Park c a
Neutron Science Center, Korea Atomic Energy Research Institute, Daejeon 34057, Republic of Korea Maritime Reactor Development Center, Korea Atomic Energy Research Institute, Daejeon 34057, Republic of Korea c Division of Advanced Nuclear Engineering, Pohang University of Science and Technology, Pohang 37673, Republic of Korea b
a r t i c l e
i n f o
Article history: Received 5 April 2017 Received in revised form 22 June 2017 Accepted 26 June 2017
Keywords: Critical heat flux Wettability Time effect TiO2 Pool boiling
a b s t r a c t A smart surface that is a TiO2-coated surface (TCS) is a hydrophobic surface initially, but becomes a hydrophilic surface when heated. Therefore, such a surface can be used to enhance both boiling heat transfer (BHT) and critical heat flux (CHF) in pool boiling. In the present study, the time effect of the wetting transition of TCS was focused on. The CHF on TCS was enhanced more when the holding time of the heat flux in high-temperature regime was increased. By observing changes in contact angles on TCS through heat treatment in air, it was found that the wetting transition was affected not only by the temperature, but also by the time. Thus, a variation of the receding contact angle was correlated in the form of an exponential function. The suggested empirical correlation includes temperature and time, and it describes the transition of the receding contact angle. The correlation was also used to predict the CHF on TCS in pool boiling. As a result, CHFs on TCS could be explained using the correlation. Ó 2017 Elsevier Ltd. All rights reserved.
1. Introduction Boiling is an efficient heat transfer method because it uses latent heat. For this reason, many applications have used the boiling system. However, the critical heat flux (CHF) imposes a limitation on the practically available boiling regime. When CHF is reached, a vapor blanket forms on the heated surface, resulting in an interruption in phase changes and a deterioration in the boiling heat transfer (BHT). Then, the heating substrate melts due to the reduced heat removal capacity. In this respect, for safe and efficient heat transfer systems, it is important to improve CHF and BHT. As technology has advanced, many techniques have been suggested to improve the CHF and BHT. To be specific, the surface modification technique, due to its applicability into real systems, e.g., machining, chemical etching, coating of thin layers, and deposition of particles, is one of the most promising methods for the improvement of boiling performance. Furthermore, the effect of nanoparticles in pool boiling has been reported on substantially since the advent of nanofluids and their applications [1–9]. Bang et al. [5] investigated the effect of alumina (Al2O3) nanofluid on the pool boiling characteristics. The authors prepared
⇑ Corresponding author. E-mail address: [email protected] (J.M. Kim). http://dx.doi.org/10.1016/j.ijheatmasstransfer.2017.06.114 0017-9310/Ó 2017 Elsevier Ltd. All rights reserved.
alumina nanofluids that had different volume concentrations. In pool boiling, the nanofluids enhanced the CHF, but reduced the BHT. In addition, the higher the concentration of the nanofluid, the lower the BHT. Kim et al. [6] used alumina, zirconia (ZrO2), and silica (SiO2) nanofluids to examine the effect of nanoparticles on the pool boiling characteristics. When the authors fixed the volume concentration at 0.01%, the CHF was enhanced for all the nanofluids. It was believed that the improved wettability via nanoparticle deposition contributed to the enhancement of CHF. However, the nanofluids lowered the BHT. Kim et al. [7] utilized alumina and titania (TiO2), and Stutz et al. [8] adopted iron oxide (Fe2O3) nanofluids for the enhancement of CHF in pool boiling. In their studies [7,8], they also agreed that the improved wettability due to the nanoparticle deposition was the main cause of the enhancement of CHF. The previous studies [6–8] showed that the wettability had an important role in enhancing CHF, regardless of the nanoparticle material. In this respect, different surface modifications were suggested with a focus on the wettability. Ahn et al. [10] fabricated nano-micro structures on zircaloy-4 using an anodic oxidation. The authors controlled the anodization time of zircaloy-4 so that the static contact angle on the surface decreased from 49.3 to 0°, while the modified surface enhanced the CHF by 90%. Chu et al. [11] fabricated hierarchical surfaces by depositing nanoparticles on micropillar posts. In pool boiling, the hierarchical surface showed a CHF enhancement of 200% in comparison to that of a
736
J.M. Kim et al. / International Journal of Heat and Mass Transfer 114 (2017) 735–742
Nomenclature a A b B C hlv k q00 R SCS t T T Ts T1 TCS
coefficient constant coefficient constant experimental constant latent heat (kJ/kg) rate constant (/min) heat flux (kW/m2) constant SiO2-coated surface time (min) temperature (K) normalized temperature, T (°C)/200 °C surface temperature (K) ambient temperature (K) TiO2-coated surface
smooth surface. In their study, the higher the roughness, the higher the CHF. However, BHT decreased as the roughness increased. The researchers believed that the thermal resistance of the hierarchical surfaces was the reason for the deterioration of BHT. Even though nanoparticle deposition enhanced the CHF due to the improved wettability, some studies [5–8,11] have identified a deterioration of BHT. However, other studies [9,12] have overcome the deterioration of BHT. Kim et al. [9] used microelectromechanical systems (MEMS) to manufacture nanostructures, microstructures, and nano-micro combined structures on silicon substrates. In pool boiling, the nano-micro structured surface showed a CHF enhanced by more than 107%, which was contributed to the increased wettability. In particular, the structures improved not only the CHF, but also the BHT. Betz et al. [12] focused on wetting control without structural changes. The authors fabricated mixed wetting surfaces: one type had hydrophobic hexagon patterns on a hydrophilic surface, while another surface had hydrophilic hexagon patterns on a hydrophobic surface. The mixed wetting surface improved the BHT and CHF by 100% and 65%, respectively, in comparison to those for a plain surface. As reported in previous studies [9,12], enhancement of both BHT and CHF requires intricate technology because there is a trade-off between BHT and CHF when the wettability is modulated. In studies [13,14], nanometrically smooth hydrophobic surfaces showed better BHT than did hydrophilic surfaces in pool boiling; hydrophobic surfaces have been found to be advantageous for high BHT, but are known to have low CHF. On the other hand, hydrophilic surfaces have showed a BHT lower than that of hydrophobic surfaces, but can have a higher CHF. In this regard, Kim et al. [15] suggested a smart surface, which is a TiO2-coated surface (TCS), to enhance both BHT and CHF in pool boiling. The authors noted that the properties of hydrophobicity and hydrophilicity have different effects on the boiling performance. So, they focused on the wetting transition of TCS. In several studies [16–18], TiO2 was found to have a wetting transition that is dependent on the temperature. TiO2 initially shows a hydrophobic characteristic, but it becomes hydrophilic when heated. Kim et al. [15] conducted pool boiling experiments using TCS and observed that TCS increased not only the BHT, but also the CHF. They reported that this result was contributed to the wetting transition of TCS. In addition, it was suggested that time was an important factor in the change of wettability, and could lead to an additional enhancement of CHF.
Greek symbols / orientation of heating surface (°) h static contact angle (°) h normalized receding contact angle, hr =90 h0 contact angle constant (°) hr receding contact angle (°) qv vapor density (kg/m3) r surface tension (N/m) or Stefan-Boltzmann constant (5.67 108 W/m2 K4) Subscripts a the coefficient of a b the coefficient of b l liquid n coefficient indication v vapor lv phase change from liquid to vapor CHF CHF
The present study focused on the time effect of the wetting transition of TCS. Through heat treatment with changes in the treatment temperature and time, the variation of the receding contact angle was examined, and an empirical correlation was provided. In addition, the empirical correlation was combined with Kandlikar’s correlation [19] to predict the CHF trend on TCS. 2. Experiments 2.1. Surface fabrication A 500 lm thick silicon wafer was used as a base substrate. Since the silicon surface had a 1 nm roughness, we considered that the roughness effect could be excluded based on Hsu’s theory [20]. For the heating of the substrate, a heating element was fabricated on the back of the substrate, as shown in Fig. 1. On one side, an SiO2 thin film with a thickness of 500 nm was deposited for electrical insulation. A photoresist (GXR601, Electronic Materials) was coated on the insulated silicon wafer using a spin coater (ATIS, Midas System) at 3000 rpm. Then, the coated photoresist (PR) was exposed to ultraviolet (UV) light through a patterned mask using a mask aligner (MA6, Suss Microtec). The exposed wafer was immersed in tetramethylammonium hydroxide (TMAH) solution, and so a reversed pattern of the PR was made on the wafer. On this surface, using an electron beam evaporator (E-beam), a platinum (Pt) layer with a thickness of 120 nm was deposited. When the Pt coated wafer was immersed in acetone, only the Pt pattern remained after the residual PR layer was removed. As a result, a heating element that has an effective heating area of 10 10 mm was fabricated. On the other side of the substrate, thin films for boiling surfaces were deposited. A 500 nm-thick SiO2 thin film was deposited on the heating surface as a reference surface; a 200 nm-thick TiO2 thin film was deposited to investigate the wetting transition. A TiO2 layer was deposited using radio frequency sputtering (RF sputtering) at room temperature conditions (25 °C). 2.2. Experimental facilities For the investigation of CHF on TCS, pool boiling facilities were constructed. The main pool was made of stainless steel pipe (304 L) that could hold 12 L of deionized (DI) water as a working fluid. In
737
J.M. Kim et al. / International Journal of Heat and Mass Transfer 114 (2017) 735–742
Fig. 1. Fabrication procedures of the heating element.
pool boiling under atmospheric condition, CHF occurred at 150 °C. However, a complete wetting transition was observed near 200 °C, as was mentioned in a previous study [15]. This will be briefly reviewed in Section 3.1. In this respect, the saturated temperature of the working fluid was increased using relief valves (VRV1MH, HY-LOK), which opened at 2.0, 3.3, and 4.1 bar. When the working fluid was boiled by an immersion heater (600 W), a relief valve was opened to discharge excess steam at the set pressure. According to the set value of the relief valve, the saturated temperature and pressure were determined during boiling. The experimental pressures were 1.0, 2.0, 3.3, and 4.1 bar; the saturated temperatures were 99.6, 120.1, 136.8, and 144.9 °C, respectively, at each pressure condition. The overall system was constructed as depicted in Fig. 2. The heater was connected to the data acquisition system (34970a, Agilent) and the power supply (XG150, Sorensen) via a shunt resistor. The power supply controlled the input voltage to manipulate the heat flux, and the data acquisition system collected data for the voltage and current of the heater and shunt resistor. The shunt resistor maintained constant resistance in the thermostat and the resistance was used to calculate the circuit current. The saturated pressure was calculated using the correlation suggested by Wagner
et al. [21]. Details on the experimental ranges and uncertainties were listed in a previous report [15]. 2.3. Experimental procedures All experiments were conducted under saturated condition. Using an immersion heater, the working fluid was boiled in advance for 6 h. The first 3 h was used to degas the DI water at the 1.0 bar condition. After degassing, only one relief valve was opened and we waited another 3 h to achieve the saturated condition at the set pressure of the relief valve. The saturated condition was maintained for more than 10 min, and the heat flux was then increased step-by-step. At each desired level of heat flux, the heat flux was held for 2 min to obtain averaged data. During the experiments, CHF was defined, and the power supply was shut down when an abrupt increase of the heating element temperature was detected. 2.4. Data reduction The wall temperature denotes the surface temperature where the surface contacts the working fluid. In the present study, the heating element was on the backside of the heating substrate and exposed to air. Therefore, we estimated heat losses to air and calculated the wall temperature based on the heating element temperature. Heat can be transferred to the air by three mechanisms: convection, radiation, and conduction. The surrounding circumstances of the heating element are illustrated in Fig. 3. Air heated by the heating, because it is heated from the top, is trapped and stationary in the space below the heating element. Thus, the natural convection was negligible and there was no forced convection. The radiation heat transfer of an ideal black body can be expressed by the following equation:
q00 ¼ rðT 4s T 41 Þ
ð1Þ
where r is the Stefan-Boltzmann constant (5.67 10 W/m K4), T s is the surface temperature (K), and T 1 is the ambient temperature (K). With the assumption that SiO2 is an ideal black body, the heat loss to the air by radiation was calculated using the experimental data for SiO2. The calculated heat loss to the air by radiation was below 1.0%, during which the heat flux was higher than 200 kW/m2. For the real surface, emissivity should be multiplied to Eq. (1). Then, the heat loss will be much lower than the calculated value because the emissivity is less than one for the real surface. 8
Fig. 2. Schematic image of the overall system for pool boiling.
2
738
J.M. Kim et al. / International Journal of Heat and Mass Transfer 114 (2017) 735–742
Fig. 3. Cross-sectional view of heater assembly: rendering image (left) and schematic image (right).
Additionally, the maximum heat loss to the air by conduction was 0.43%. As a result, the heat loss to the air was ignored. For the calculation of the wall temperature, we used Fourier’s law, with a thermal conductivity of 130 W/m K for the silicon substrate. When it was assumed that the heat was transferred in one direction from the heating element to the boiling surface, the wall temperature was easily obtained. After each experiment, the wall temperature was calculated using the method and the experimental data.
3. Results and discussion 3.1. Review of the effect of temperature on the wetting transition of TCS In the previous study [15], in the heat treatment test, TCS showed a wetting transition, as shown in Fig. 4. For the heat treatment test, the SiO2-coated surface (SCS) and TCS were heated in the oven (air condition) at a specific temperature for 10 h. The temperature ranged from 100 to 200 °C; this range corresponded to the experimental temperature range of the heating surface. After the heat treatment was conducted, the test surface was taken out of the oven and cooled at a room temperature of 25 °C. The contact angles on the surfaces were measured using a droplet goniometer
(SmartDrop, FemtoFab) with 2 ll water droplets. Because the instrument has an accuracy of ±0.1°, it was believed that the contact angles were precisely measured. A surface that had been heated once in the oven was never reused for another test. After heat treatment was conducted at 100 °C, SCS and TCS showed contact angles of 68 and 83°, respectively. As the heat treatment temperature increased, a decrement of the static contact angle on TCS was observed, while the contact angle on SCS maintained its initial value (in Fig. 4(a)). After the heat treatment at 180 °C, the static contact angle on TCS was smaller than that on SCS. With an increase in the heat treatment temperature to 200 °C, TCS became hydrophilic in comparison to SCS; TCS showed a static contact angle of 33°, while SCS showed a static contact angle of 62°. The receding contact angles also showed tendencies similar to the variation of the static contact angles, as shown in Fig. 4(b). The receding contact angle on TCS changed from 53 to 14° with increasing of the heat treatment temperature, while SCS showed an averaged receding contact angle of 42°. The wetting transition can be explained by the changes in the crystal structure. There have been previous studies [16,18,22–26] on the wetting transition induced by UV illumination. Sun et al. [16] investigated the water contact angles on TiO2 thin film before and after UV illumination. The authors confirmed that the TiO2 thin film became hydrophilic when it was exposed to the UV light. They explained that the wetting transition was due to the changes in the
Fig. 4. (a) Static contact angles and (b) receding contact angles after heat treatment for 10 h [15].
J.M. Kim et al. / International Journal of Heat and Mass Transfer 114 (2017) 735–742
crystal structure. When the surface was treated under UV light, it had oxygen vacancies. If a water molecule contacts an oxygen vacancy, the molecule will be dissociatively adsorbed onto the vacancy. This induced the wetting transition. Sakai et al. [23] also investigated hydrophilic conversion by UV illumination. They fabricated TiO2 film using a dip-coating method. On the TiO2 coated substrate, they conducted UV irradiation at various UV intensities and observed a conversion to a hydrophilic surface, resulting in a contact angle smaller than 10°. They also explained that the reconstruction of the hydroxyl groups occurred at the oxygen vacancies when UV light illuminated the TiO2 film. 3.2. Time effect of the static contact angle For the clarification of the time effect for the wetting transition, we examined the time effect of the wetting transition for TCS through heat treatment tests; this was done because the authors in the previous study [15] showed that an increase of the holding time of the heat flux at high temperature regime affected the additional enhancement of CHF on TCS. TCSs with dimensions of 10 10 mm were thermally treated at 200 °C in an oven while the heat treatment time was varied from 0 to 10 h. One TCS was cooled at 25 °C after 1 h of heat treatment and we measured the static contact angle. The next TCS was also cooled at 25 °C after heat treatment for 2 h, and the static contact angle was measured. Experiments were performed using the same procedure at intervals of 1 h. Results are presented in Fig. 5 for the time control tests in the heat treatment. Initially, SCS showed a static contact angle of 57°, but TCS showed a relatively hydrophobic characteristic, with a contact angle of 83°. As the heat treatment time increased up to 10 h, TCS became hydrophilic, with a contact angle of 23°, while SCS did not show any significant change in contact angle. These results indicate that a sufficient amount of time is needed to achieve the wetting transition of TCS. The time tendency for the reduction of the contact angle was also reported on in previous studies [16,18,22–25]. Stevens et al. [24] fabricated TiO2 surfaces to examine the alteration of the wettability by UV irradiation. As the illumination time of UV light increased, the advancing contact angle was found to decrease. Borras et al. [25] also fabricated TiO2 thin films using metal-organic chemical vapor deposition (MOCVD). They controlled the UV illumination time and measured the water contact angle on the surfaces. The contact angle decreased as the illumination time increased. In addition, the contact angle became small when the UV intensity increased. Even though the previous studies [24,25] used UV illumination to induce the wetting transition, the
100
739
tendency of the wetting transition with increasing illumination time of UV light showed a form of variation similar to that found in the present study. Since Kandlikar [19] proposed a CHF model that reflected a wetting effect using the receding contact angle, the receding contact angle has been considered as an important parameter for the prediction of CHF. This means that measuring the receding contact angle on the heating surface is essential to understanding boiling phenomena and, in particular, for predicting CHF. In this respect, we investigated the variation of the receding contact angle with changes in heat treatment temperature and time. For the heat treatment, the temperature ranged from 140 to 200 °C because, in the present study, CHF occurred at a temperature higher than 140 °C; the time was adjusted from 2 h to 10 h at 2 h intervals. In detail, TCSs were heated at 140 °C in the oven. The first sample was heated for 2 h and its receding contact angle was measured on the surface after it was cooled to a room temperature of 25°. The second sample was heated for 4 h and the receding contact angle was measured using the same procedure. In total, heat treatments for 20 cases were conducted at various temperatures and times, and each case was repeated more than 3 times. The measured receding contact angles on TCS after the heat treatment are displayed as solid points in Fig. 6. As expected, decrements in the receding contact angle were observed as the temperature increased. However, even though the heat treatment occurred at 200 °C, the decrease of the receding contact angle was not significant when the heat treatment time was short. For the heat treatment at 200 °C, the receding contact angle was about 23° after heat treatment for 2 h, but it became 14° after heat treatment for 10 h. As the heat treatment time increased, the receding contact angles approached certain asymptotic values as a result of heat treatment for 10 h. The explanation for the solid lines in the graph is treated in Section 3.3. 3.3. Empirical correlation of the variation of the receding contact angle For the future application of TCS, a prediction of the wetting transition is essential. For this reason, certain attempts have been made to predict the change of the contact angle of TiO2. Stevens et al. [24] deposited 100 nm TiO2 thin film on a silicon wafer. The authors exposed the surface to UV light with changes in irradiation time and measured the advancing contact angle on the surface. When the surface was exposed to UV light, the advancing
SCS TCS
90
Contact angle (°)
80 70 60 50 40 30 20 10 0
0
2
4
6
8
10
12
Heat treatment time (h) Fig. 5. Static contact angles on SCS and TCS as heat treatment time increases at 200 °C.
Fig. 6. Receding contact angles after heat treatment with changes of temperature and time; solid points are the experimental data; lines are predictions based on empirical correlation.
740
J.M. Kim et al. / International Journal of Heat and Mass Transfer 114 (2017) 735–742
Fig. 7. Fitting of coefficients as a function of temperature.
contact angle decreased. Additionally, the advancing contact angle decreased more as the irradiation time increased. They determined that the decrease in contact angle after UV irradiation could be expressed in exponential form as follows:
h ¼ h0 expðkUV tÞ
ð2Þ
where h0 is the contact angle (°), kUV is the rate constant (/min), and t is the time (min). The proposed equation fitted the experimental data well. Borras et al. [25] also identified a decrease in the contact angle on a TiO2 coated surface after UV irradiation. The authors used the same form that was suggested by Stevens et al. [24] to express the decline of the contact angle for the TiO2 coated surface. In two previous studies [24,25], the exponential form was found to represent the change of the contact angle on TiO2 well. In the present study, we also tried to show the variation of the receding contact angle on TCS using the function in Eq. (3),
h ¼ a þ b expðktÞ
ð3Þ
where h is a normalized receding contact angle, k is a rate constant (min1), t is a time (min), and other values are coefficients; the coefficients are functions of the temperature. In Eq. (3), the normalized receding contact angle is determined by the combination of a and b when k is fixed. For the transformation of the crystal structure of TiO2, the rate constant can be expressed using Arrhenius equation [27]. Because the Arrhenius equation has been used in many solid state reactions [28,29], we also adopted the form of Arrhenius equation to depict the rate constant with the variation of the temperature. This form shows the agreement of the relationship between the rate constant and temperature in the previous studies [30,31]; the rate constant increases with increasing temperature. Based on the previous studies [27–31], we used Eq. (4), in which T is an absolute temperature (K).
k ¼ 0:0118 expð12:84=TÞ
ð4Þ
In addition, to determine a and b in Eq. (3), we postulated that these factors can be represented by Eq. (5), which is a function of the temperature:
aðor bÞ ¼ A þ B expðRT Þ
ð5Þ
where A, B, and R are constants, and T⁄ is the normalized temperature divided by 200 °C. In this correlation, we determined constants to fit our data; the results are shown in Fig. 7. The value of a decreased as the temperature increased and the correlation in Fig. 7(a) matched the data points, with an R2 of 0.9811. The value of b increased as the temperature increased and matched the fitting
Table 1 Fitting constants for the coefficients. Constants
Values
Aa Ba Ra Ab Bb Rb
0.9278 0.1311 1.8205 0.2933 0.1182 1.8997
line, with an R2 of 0.9817. The values of the constants used in Eq. (5) are listed in Table 1. As a final step, the receding contact angle was predicted by substituting Eqs. (4) and (5) for Eq. (3). Since the final form of the correlation is a function of time and temperature, the transition of the receding contact angle with changing time and temperature can be described well. In Fig. 6, the curves were plotted using the correlation in Eq. (3), with a value of R2 of 9743. The correlation matched well with the measured receding contact angles after heat treatment was conducted at various temperatures and treatment times. 3.4. Critical heat flux In the previous study [15], the CHF on TCS was enhanced more when the heat flux holding time was increased to 20 min in the high-temperature regime. The authors postulated that the time also contributed to the enhancement of CHF. In this situation, there was no appropriate model to predict the CHF on TCS because it depends on the temperature and the time. Even though Kandlikar [19] suggested a CHF prediction model, represented by Eq. (6), that includes the receding contact angle, it also cannot predict the CHF on TCS due to the wetting transition of TCS.
q00CHF ¼ Chlv q1=2 v
1=2 1 þ cos hr 2 p þ ð1 þ cos hr Þ cos / 16 p 4
ð6Þ
In Fig. 8, a comparison of CHFs on TCS is displayed: the blue circles are CHFs in the normal operation, the red triangles are CHFs in the time effect tests, and the black solid line is a CHF prediction line on TCS using the model suggested by Kandlikar [19]. For each pressure condition, the fitting constant was multiplied to fit the reference data, SCS, which was explained in the previous study [15]. The correlation overestimated the experimental data because an asymptotic receding contact angle under CHF condition was used without consideration of the time effect in the wetting transition. Furthermore, due to the time effect of the wetting transition
Critical heat flux (kW/m2)
J.M. Kim et al. / International Journal of Heat and Mass Transfer 114 (2017) 735–742
2200 2000 1800 1600 1400 1200 1000 800 600 400 200 0
741
3.5. Time constant
Kandlikar [19] Exp. (normal operation) Exp. (time effect)
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Pressure (bar) Fig. 8. Comparison of CHFs on TCS with prediction model suggested by Kandlikar [19].
We observed the wetting transition via the heat treatment tests using the oven in air condition. Also, an additional enhancement of CHF on TCS was identified when the holding time of the heat flux increased. The enhancement of the CHFs on TCS was contributed to the wetting transition to the hydrophilic surface because the roughness and thermal resistance effects were negligible [15]. However, the time rates of the wetting transition on TCS were different for air and water. In Fig. 9, the time, represented by the solid line, denotes the time duration in the case of heat treatment in air. The solid data points (holding heat fluxes for 2 min) correspond to the CHF calculated using the receding contact angle obtained from the heat treatment for 1.5 h. The hollow data points (holding a heat flux for 20 min in high-temperature regime) correspond to the CHF calculated using the receding contact angle for 4 h heat treatment. It seems that the heat treatment times of 1.5 and 4 h in air correspond to heat flux holding times of 2 min and 20 min in water. The time rate of the wetting transition seems to have been affected by the surrounding environment. Therefore, the relation between environment and the time constant should be clarified in the future. 4. Conclusions
Fig. 9. CHF prediction with consideration of the wetting transition of TCS; the solid points are the CHFs on TCS in normal tests; hollow points are the CHFS on TCS in time effect tests; solid lines are CHF predictions for changing t in Eq. (3).
for TCS, the CHFs on TCS were different for normal operation and for the time effect test, as was previously mentioned. Thus, to reflect the wetting transition effect of TCS, we substituted Eq. (3) for Eq. (6). The prediction results using Eqs. (3) and (6) are shown in Fig. 9; the CHF ratio was defined as the proportion of CHF on TCS to CHF on SCS. The solid points are the CHF ratios on TCS in the normal tests, the hollow points are the CHF ratios on TCS in the time effect tests, and the solid lines are the CHF ratios of TCS predicted using the empirical correlation with changing T and t. In the correlation, time is the heat treatment time assuming the heat treatment was performed in air, not in water. When we assumed the heat treatment time was zero, the prediction line underestimated the experimental CHFs on TCS because TCS would be hydrophobic when the heat treatment time was zero. When we input a 1.5 h heat treatment time into the correlation, the prediction line approximated the CHFs in normal operation. In this case, the maximum error was 7.99% at a 3.3 bar condition. When we assumed that the heat treatment time was 4 h, the prediction line was quite well matched to the experimental CHFs in the time effect tests, which maintained heat flux for 20 min in the high-temperature regime. In this case, the maximum error was 2.34% at a 2.0 bar condition. Therefore, the experimental CHFs on TCS can be explained using Kandlikar’s model [19], combining the empirical correlation.
The time effect of the wetting transition was examined using a smart surface (TCS). In the present study, it was confirmed that, in the air condition, the wettability of TCS was affected not only by the heat treatment temperature but also by the time. From the perspective of predicting the CHF on TCS in pool boiling, the variation of the receding contact angle was analyzed through heat treatment tests. The receding contact angle decreased with increasing heat treatment temperature and time. This was expressed as an exponential function having parameters of temperature and time. By determining the constants in the empirical correlation, the calculated receding contact angles were found to be well matched to the measured receding contact angles on TCS. Additionally, the CHFs on TCS were calculated by combining the empirical correlation and Kandlikar’s model [19]. When the heat treatment time was changed in the empirical correlation, the CHFs on TCS could be explained. Even if there was the wetting transition on TCS, the time constant was different according to the environment. In the future, the relation between the time rate and the environment should be studied. Conflict of interest The auth ors declare that they have no conflict of interests for financial. Acknowledgements This work was supported by National Research Foundation of Korea (NRF) grants funded by the Korean Government (MSIP) (2015M2A8A2074795) and the Korean Government (MSIP) (2012M2A2A6004262). The fabrication of the test sections was supported by Ulsan National Institute of Science and Technology (UNIST), and National Nanofab Center (NNFC). References [1] 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) 3374–3376, http://dx.doi.org/10.1063/1.1619206. [2] H. Kim, J. Kim, M.H. Kim, Effect of nanoparticles on CHF enhancement in pool boiling of nano-fluids, Int. J. Heat Mass Transf. 49 (2006) 5070–5074, http://dx. doi.org/10.1016/j.ijheatmasstransfer.2006.07.019.
742
J.M. Kim et al. / International Journal of Heat and Mass Transfer 114 (2017) 735–742
[3] S.J. Kim, I.C. Bang, J. Buongiorno, L.W. Hu, Effects of nanoparticle deposition on surface wettability influencing boiling heat transfer in nanofluids, Appl. Phys. Lett. 89 (2006) 153107. [4] C.-C. Hsu, P.-H. Chen, Surface wettability effects on critical heat flux of boiling heat transfer using nanoparticle coatings, Int. J. Heat Mass Transf. 55 (2012) 3713–3719. [5] I.C. Bang, S. Heung Chang, S.H. Chang, Boiling heat transfer performance and phenomena of Al2O3–water nano-fluids from a plain surface in a pool, Int. J. Heat Mass Transf. 48 (2005) 2407–2419, http://dx.doi.org/10.1016/j. ijheatmasstransfer.2004.12.047. [6] 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. 50 (2007) 4105–4116, http://dx.doi.org/10.1016/j.ijheatmasstransfer. 2007.02.002. [7] H. Kim, H.S. Ahn, M.H. Kim, On the mechanism of pool boiling critical heat flux enhancement in nanofluids, J. Heat Transfer. 132 (2010) 61501, http://dx.doi. org/10.1115/1.4000746. [8] B. Stutz, C.H.S. Morceli, M. de F. da Silva, S. Cioulachtjian, J. Bonjour, Influence of nanoparticle surface coating on pool boiling, Exp. Therm. Fluid Sci. 35 (2011) 1239–1249, http://dx.doi.org/10.1016/j.expthermflusci.2011.04.011. [9] S. Kim, H.D. Kim, H.D. Kim, H.S. Ahn, H. Jo, J. Kim, M.H. Kim, Effects of nanofluid and surfaces with nano structure on the increase of CHF, Exp. Therm. Fluid Sci. 34 (2010) 487–495, http://dx.doi.org/10.1016/j.expthermflusci. 2009.05.006. [10] H.S. Ahn, C. Lee, H. Kim, H. Jo, S. Kang, J. Kim, J. Shin, M.H. Kim, Pool boiling CHF enhancement by micro/nanoscale modification of zircaloy-4 surface, Nucl. Eng. Des. 240 (2010) 3350–3360, http://dx.doi.org/10.1016/j.nucengdes. 2010.07.006. [11] K.-H. Chu, Y. Soo Joung, R. Enright, C.R. Buie, E.N. Wang, Hierarchically structured surfaces for boiling critical heat flux enhancement, Appl. Phys. Lett. 102 (2013) 151602. [12] A.R. Betz, J. Xu, H. Qiu, D. Attinger, Do surfaces with mixed hydrophilic and hydrophobic areas enhance pool boiling?, Appl Phys. Lett. 97 (2010) 141909. [13] H. Jo, H.S. Ahn, S. Kang, M.H. Kim, A study of nucleate boiling heat transfer on hydrophilic, hydrophobic and heterogeneous wetting surfaces, Int. J. Heat Mass Transf. 54 (2011) 5643–5652, http://dx.doi.org/10.1016/j. ijheatmasstransfer.2011.06.001. [14] B. Bourdon, P. Di Marco, R. Rioboo, M. Marengo, J. De Coninck, Enhancing the onset of pool boiling by wettability modification on nanometrically smooth surfaces, Int. Commun. Heat Mass Transf. 45 (2013) 11–15, http://dx.doi.org/ 10.1016/j.icheatmasstransfer.2013.04.009. [15] J.M. Kim, D.I. Yu, H.S. Park, K. Moriyama, M.H. Kim, Smart surface in pool boiling: thermally-induced wetting transition, Int. J. Heat Mass Transf. 109 (2017) 231–241.
[16] R.-D. Sun, A. Nakajima, A. Fujishima, T. Watanabe, K. Hashimoto, Photoinduced surface wettability conversion of ZnO and TiO2 thin films, J. Phys. Chem. B 105 (2001) 1984–1990. [17] Q. Ye, P.Y. Liu, Z.F. Tang, L. Zhai, Hydrophilic properties of nano-TiO2 thin films deposited by RF magnetron sputtering, Vacuum 81 (2007) 627–631, http://dx. doi.org/10.1016/j.vacuum.2006.09.001. [18] Z.W. Zhao, B.K. Tay, Study of nanocrystal TiO2 thin films by thermal annealing, J. Electroceram. 16 (2006) 489–493. [19] S.G. Kandlikar, A theoretical model to predict pool boiling CHF incorporating effects of contact angle and orientation, J. Heat Transfer. 123 (2001) 1071, http://dx.doi.org/10.1115/1.1409265. [20] Y.Y. Hsu, On the size range of active nucleation cavities on a heating surface, J. Heat Transfer 84 (1962) 207–213. [21] W. Wagner, A. Pruss, International equations for the saturation properties of ordinary water substance. Revised according to the international temperature scale of 1990. Addendum to J. Phys. Chem. Ref. Data 16, 893 (1987), J. Phys. Chem. Ref. Data 22 (1993) 783–787. [22] T. Watanabe, Photocatalytic activity and photo-induced wettability conversion of TiO2 thin film prepared by Sol-Gel process on a Soda-Lime glass, J. Sol-Gel Sci. Technol. 19 (2000) 71–76. [23] N. Sakai, A. Fujishima, T. Watanabe, K. Hashimoto, Quantitative evaluation of the photoinduced hydrophilic conversion properties of TiO2 thin film surfaces by the reciprocal of contact angle, J. Phys. Chem. B 107 (2003) 1028–1035. [24] N. Stevens, C.I. Priest, R. Sedev, J. Ralston, Wettability of photoresponsive titanium dioxide surfaces, Langmuir 19 (2003) 3272–3275. [25] A. Borras, C. Lopez, V. Rico, F. Gracia, A.R. Gonzalez-Elipe, E. Richter, G. Battiston, R. Gerbasi, N. McSporran, G. Sauthier, A. Borra, E. Richter, G. Battiston, Effect of visible and UV illumination on the water contact angle of TiO2 thin films with incorporated nitrogen, J. Phys. Chem. C 111 (2007) 1801–1808. [26] S. Yamauchi, Y. Imai, Plasma-assisted chemical vapor deposition of TiO2 thin films for highly hydrophilic performance, Cryst. Struct. Theory Appl. 2 (2013) 1–7. [27] C.N.R. Rao, Kinetics and thermodynamics of the crystal structure transformation of spectroscopically pure anatase to rutile, Can. J. Chem. 39 (1961) 498–500. [28] A.K. Galwey, M.E. Brown, A Theoretical Justification for the Application of the Arrhenius Equation to Kinetics of Solid State Reactions (mainly ionic crystals), in: Proc. R. Soc. London A Math. Phys. Eng. Sci., The Royal Society, 1995, pp. 501–512. [29] A.K. Galwey, M.E. Brown, Application of the Arrhenius equation to solid state kinetics: can this be justified?, Thermochim Acta. 386 (2002) 91–98. [30] J. Yu, X. Zhao, Q. Zhao, Photocatalytic activity of nanometer TiO2 thin films prepared by the sol–gel method, Mater. Chem. Phys. 69 (2001) 25–29. [31] J. Yu, X. Zhao, Q. Zhao, Effect of surface structure on photocatalytic activity of TiO 2 thin films prepared by sol-gel method, Thin Solid Films. 379 (2000) 7–14.