Nanoporous anodic alumina oxide layer and its sealing for the enhancement of radiative heat dissipation of aluminum alloy

Nano Energy 31 (2017) 504–513

Contents lists available at ScienceDirect

Nano Energy journal homepage: www.elsevier.com/locate/nanoen

Nanoporous anodic alumina oxide layer and its sealing for the enhancement of radiative heat dissipation of aluminum alloy ⁎

MARK

⁎

Junghoon Leea,1, Donghyun Kimb,c,1, Chang-Hwan Choia, , Wonsub Chungb, a b c

Department of Mechanical Engineering, Stevens Institute of Technology, Hoboken, NJ 07030, USA Department of Materials Science and Engineering, Pusan National University, Busan 46241, Republic of Korea Analysis and Certification Center, Korea Institute of Ceramic Engineering and Technology, Jinju-si, Gyeongsangnam-do 52851, Republic of Korea

A R T I C L E I N F O

A BS T RAC T

Keywords: Energy transfer Heat dissipation Emissivity Aluminum Anodizing

Various types of nanoporous anodic aluminum oxide layers and their sealings were studied to enhance the thermal emissivity and hence improve the heat dissipation of aluminum alloy for energy application. Dissipating heat fluxes from the anodized aluminum surfaces were measured using a modified steady-state method and investigated with respect to the various nanoporous morphologies obtained with different anodizing conditions and sealing methods. Results show that the anodized nanoporous oxide layers significantly enhance the thermal emissivity and heat dissipation of aluminum alloy, compared to bare aluminum alloy, and such enhancement is further improved with sealings. A thicker nanoporous oxide layer anodized in oxalic acid results in higher thermal emissivity and better heat dissipation than that in sulfuric acid, showing a darker color which is attributed to the more irregular and disordered pore size and pattern of the nanoporous oxide layer. The nanoporous oxide layer with cold NiF2 or black sealing shows further enhancement in thermal emissivity and heat dissipation, demonstrating the highest enhancement in emissivity up to 0.906 in case of the nanoporous oxide layer anodized in oxalic acid with black sealing, which is seven times greater than that of bare aluminum. The nanoporous oxide layer with black sealing also results in the significant improvement of the cooling efficiency of a heat exchanger system of aluminum alloy by 36.4%, suggesting great energy saving for real energy application.

1. Introduction Heat transfer by thermal radiation has been explored to overcome limitations of convection and conduction modes for managing heatrelated problems in heating and cooling in vacuum environments and buildings [1–7]. Thermal radiation is traditionally explained by the Stefan-Boltzmann equation, where the radiative heat flux is dependent on the emissivity (ε) and temperature of the surface. In order to enhance the heat transfer by thermal radiation, both the higher emissivity close to that of an ideal black body (ε=1) and the higher temperature of the radiative surface, which are controllable by surface treatment, are required [8–10]. Heat dissipation by thermal radiation is related with a passive cooling state, which does not involve a fan or fluid circulation for cooling [11–13]. Since the passive cooling does not require additional energy to operate cooling units, it has been considered to be an economical and environmental-friendly technology [14]. This system has been applied to various electric devices, which do not require

⁎

1

considerably active cooling [15,16]. In such a passive cooling state, thermal radiation of a heated surface is an effective way to transmit the heat, and thus the heat dissipation can be improved by the thermal radiation of cooling parts (e.g., heat sinks). Therefore, surface modification technologies to enhance the emissivity (ε) have a great potential in thermal management [17,18]. Ceramics [19–24] or carbon-based materials [25–27], which are known to have a high emissivity, can be coated on metallic materials having high thermal conductivity (e.g., Al and Cu alloys) to improve heat dissipation, because typical metallic materials have a low surface emissivity of ~0.1. In particular, anodizing or coatings with ceramic materials on aluminum alloys, which have been employed to various heat sinks, can be a promising technique to solve heat management problems [28–30]. Especially, an enhancement of the thermal emissivity and absorptivity of anodic aluminum oxide (AAO) with cobalt sulfide (CoS) sealing (i.e., black sealing) was reported for the application to solar absorbing films [31–33]. However, most of ceramic materials examined so far have relatively lower thermal conductivity

Corresponding authors. E-mail addresses: [email protected] (C.-H. Choi), [email protected] (W. Chung). These authors contributed equally to this work.

http://dx.doi.org/10.1016/j.nanoen.2016.12.007 Received 1 October 2016; Received in revised form 18 November 2016; Accepted 5 December 2016 Available online 07 December 2016 2211-2855/ © 2016 Elsevier Ltd. All rights reserved.

Nano Energy 31 (2017) 504–513

J. Lee et al.

than metallic materials [34,35]. Therefore, when only considering the conduction, the ceramic coating layer works as a thermal barrier, suppressing an effective heat transfer toward the coating surface with an increase in its thickness. In contrast, if the thickness of a coating layer is not enough to have a high emissivity, effective radiative heat dissipation could not be implemented [36]. Therefore, when considering the heat dissipation by thermal radiation, the optimized thickness of the coating layer is as critical as the coating material itself. Thus, it is critical to design the surface coating layer to have the optimized heat dissipation performance with the careful consideration of such complicated characteristics of the surface layer for efficient thermal management. In this study, for a new surface coating layer, we employ anodizing and sealing techniques, which are scalable for mass production and readily available for industrial applications, to explore and maximize the combined effects of the synergistic increase in thermal emissivity and conductivity to enhance the overall heat dissipation performance of aluminum alloy. A nanoporous oxide layer directly grown on an aluminum substrate by the anodizing process enhances the surface's radiative emissivity. In addition, unlike the composite coatings using carbon-based materials or polymer resin that have relatively high interfacial thermal resistance with substrate metals [7,25–27], the anodic aluminum oxide has negligible interfacial thermal resistance with the aluminum substrate [37], which is beneficial for the overall heat dissipation performance. In order to find out the optimized surface conditions for the best heat dissipation performance of the anodized aluminum oxide (AAO) surface, the effects of anodizing conditions (i.e., anodizing time and temperature) and additional sealing treatments (e.g., hydrothermal, cold NiF2, and black sealings) on the heat dissipation of aluminum alloy are explored and discussed with respect to the porous nanostructures of the AAO and thermal emissivity. For the effective measurement of such properties, we introduce a measurement method for the dissipating heat flux from a modified surface, such as anodized aluminum alloy, so that it can provide a reliable estimation for the heat dissipation performance of such a modified metallic surface. Furthermore, an enhancement of cooling performance of the treated aluminum alloy with improved dissipating heat flux is demonstrated for real applications.

Table 1 Conditions of anodizing and sealing processes. Electrolyte

1.63 M (15.0 wt%) sulfuric acid and 0.3 M (2.7 wt%) oxalic acid

Temperature Current density Process time Sealing solution

0, 15, 30 °C 50 mA/cm2 10–40 min Hydrothermal Cold NiF2 Black

98 °C, 30 min 2.5 g/l, 25 °C, 30 min 1st step: 250 g/l cobalt acetate (Co(C2H3O2)2), 45 °C, 20 min 2nd step: 20 g/l, ammonium sulfide ((NH4)2S), 25 °C, 30 min

scanning electron microscopy (FE-SEM, Hitachi S-4700, Japan). The thickness of anodic aluminum oxide was measured from the crosssectional image obtained by the FE-SEM. In order to obtain clear images showing pore structure, the surface of anodized aluminum was mildly polished with 1 µm magnesia (MgO) and diamond paste using a nap cloth. Polished specimens were cleaned with 10 vol% H3PO4 for 30 s and washed in deionized water with ultrasonication. Captured optical and SEM images were analyzed using Image-Pro software to determine the averaged values and the standard deviations of a gray level of surface appearance, pore diameters, eccentricity of pore, interpore distance, and porosity. Emissivity of the anodized aluminum surface was measured using Fourier transform infrared (FT-IR) spectroscopy (Nicolet Avatar 360 FT-IR spectroscopy, Thermo Scientific, USA) at 200 °C, where the measured range of the wavelength was 5–20 µm. Four samples fabricated in the same conditions were measured to estimate the average value and the standard deviation at the given condition. The improved heat dissipation performances were compared to those of the bare aluminum alloy by cooling hot water contained in a cylindrical aluminum tube with anodized oxide layers. 2.2. Dissipated heat flux measurement setup Fig. 1 presents a schematic diagram of the designed radiation heat flux measurement setup modified from the thermal conductivity measurement setup using a flow-meter method [38,39]. Stainless steel 304 block (thermal conductivity=16.2 W/m K at 200 °C, POSCO, Korea) cut into 50×50×120 mm was used as a meter bar and thermally insulated to minimize heat loss. The fabricated aluminum alloy sample

2. Materials and methods 2.1. Sample preparation and evaluation methods A plate of Al 1050 alloy (99.5%Al – 0.24% Fe – 0.15% Si, Alcoa Inc., USA) was used as a substrate and cut into 50×50×1.5 mm in size. The plates were cleaned ultrasonically in acetone before use. The substrates were electropolished in ethyl alcohol (99.9+%, SK Chemicals, Korea) and perchloric acid (Junsei Chemical Co., Ltd., Japan) mixture solution (3:1 in volumetric ratio) at 25 °C under constant voltage of 20 V for 2 min. Two different types of electrolytes were employed for anodizing, including 0.3 M (2.7 wt%) oxalic acid (Junsei Chemical Co., Ltd., Japan) and 1.63 M (15 wt%) sulfuric (Junsei Chemical Co., Ltd., Japan). Aeration into the electrolytes was used for agitation, and the temperature of the electrolyte was regulated in a range from 0 to 30 °C. Constant current density of 50 mA/cm2 was applied for anodizing. In order to control the thickness of anodic oxide layer, the anodizing time was also regulated in a range from 10 to 40 min. After the anodizing, hydrothermal, cold NiF2, and black sealing methods were applied respectively for comparison. Detail conditions for the anodizing and sealing processes were summarized in Table 1. Dissipated heat from the anodized aluminum surface was measured (see Section 2.2 for the details of the dissipated heat flux measurement setup). During the measurement of the dissipated heat, the temperature of ambient was maintained at 25 °C using an air conditioner. Pore structures of anodic oxide layer were observed using field emission

Fig. 1. Schematic diagram of dissipated heat measurement setup.

505

Nano Energy 31 (2017) 504–513

J. Lee et al.

(50×50 mm) was placed on the top of the meter bar with thermal interfacial material (TIM, 3.8 W/m K, Evercool TC-200, Taiwan), which reduces an air gap generating interfacial thermal resistance. Three T-type thermo-couples (T1, T2, and T3) were inserted with 10 mm spacing one another from the top surface of the meter bar, and temperature changes were recorded using a monitoring device (MV1000, Yokogawa Electric Corporation, Japan). The meter bar was heated by an electrical heater under the meter bar, and the heat source (electrical heater) was maintained at 200 °C. Steady-state temperatures, which were considered when the temperature fluctuations were within ± 0.1 °C for more than 20 min, were used to calculate the heat flux. 2.3. Data analysis The heat generated by the heat source is dissipated to ambient through the meter bar and the sample surface. Therefore, dissipated heat flux can be calculated from the heat flux in meter bar. The temperature of inserted thermo-couples and the thermal conductivity of the meter bar were used to calculate the heat flux. The heat flux (Qxy, W/m2) between two thermo-couples (x and y) can be calculated by following equation [38,39]:

Qxy = (λ ⋅ A / dx − y) · (Tx −Ty),

(1)

where Tx and Ty are the temperatures of the arbitrary two thermocouples x and y. λ, A, and dx-y are the thermal conductivity of the meter bar (16.2 W/m K for the 304 stainless steel), the cross-sectional area of the meter bar (25 cm2), and the distance between two thermo-couples (10 mm), respectively. Since three thermo-couples (T1, T2, and T3) were used, three values (Q12, Q23, and Q13) were estimated and averaged. Moreover, four samples fabricated in the same conditions were tested for each condition to compute the averaged value and the standard deviation. Detail procedure to obtain the dissipated heat flux of anodized Al 1050 alloy can be found in Supplementary information.

Fig. 2. (a) Emissivity of electropolished bare aluminum 1050 alloy. (b) Temperatures of thermo-couples in the meter bar for the electropolished bare aluminum 1050 alloy.

the anodizing temperature. In case of the anodizing at 0 °C, the surface color becomes darker brown with the increase in the anodizing time. In case of the anodizing at 15 and 30 °C, the surface color becomes darker black. The surface with a darker color would lead to higher thermal emissivity, being closer to that of a black body. Estimated pore diameter, eccentricity of pores, inter-pore distance, and porosity from SEM images (Fig. 3b) using an image analysis software are shown in Fig. 3d–g and summarized in Table S1 in Supplementary information. The anodizing temperature significantly affected the pore morphology (e.g., pore size, pore shape (i.e., eccentricity), inter-pore distance, and porosity) of AAO nanostructures, but the anodizing duration showed negligible effect. The anodizing at higher temperature in oxalic acid resulted in larger-sized pores with higher porosity of the AAO layer [37]. In addition, the AAO layer anodized at 15 and 30 °C showed more irregular size and shape of the pores (the higher values of the eccentricity of pores and the standard deviation of the pore diameter and eccentricity) as well as more disordered pattern (the higher standard deviation of the inter-pore distance) than those of the AAO layer anodized at 0 °C. The differences in the pore morphology make the surfaces appear different colors with a different refractive index [40,41]. The AAO layer with more irregular diameter and shape of pores which are also more randomly distributed would lead to enhance the scattering and absorbing of the light at the AAO layer, indicating more strongly saturated color appearance [41–43]. Fig. 3h shows the thicknesses of AAO films fabricated with the different anodizing temperatures and durations. At the given temperature, the thickness of the AAO film increases with the increase in the anodizing time, resulting in an increase of the pore depth (i.e., the aspect ratio of the cylindrical porous nanostructure). The pore diameter and the porosity do not change much with the anodizing time but remain almost the same despite the increase in the anodizing time. Since the higher aspect

3. Results and discussions 3.1. Heat dissipation of bare aluminum 1050 alloy Heat dissipation performance of electropolished bare aluminum 1050 alloy, which is used as a base substrate material and also as a control in this work, was first investigated to verify the enhancement of the heat dissipation by anodizing treatment. Fig. 2 shows an emissivity and recorded temperatures of thermocouples in the meter bar. The emissivity of mirror-like (after electropolishing) aluminum substrate is in a range of 0.050–0.200 within the scanned wave length, and an averaged emissivity is 0.132 at 200 °C. This low emissivity value indicates a relatively low thermal radiation from the surface following the Stefan-Boltzmann equation. Temperature slopes ((Tx−Ty)/dx−y) between thermocouples T1 and T2, T2 and T3, and T1 and T3 are 1.2 °C/cm, 1.3 °C/cm and 1.25 °C/cm, respectively. By substituting these slopes into Eq. (1), the averaged heat flux of 1984.5 W/m2 for the bare aluminum substrate used as a control is measured. 3.2. Heat dissipation of anodized aluminum with nanoporous oxide layer Fig. 3a and b show the optical and SEM images of the nanoporous anodic aluminum oxide (AAO) surfaces fabricated in oxalic acid with the different anodizing temperatures and durations. At the given anodizing temperature, the surface color of the AAO film becomes darker (i.e., low value of gray level obtained from optical images (Fig. 3a), see Fig. 3c and Table S1 in Supplementary information) with an increase of the anodizing time. At the given anodizing time, the surface color of the AAO film also appears darker with an increase of 506

Nano Energy 31 (2017) 504–513

J. Lee et al.

Fig. 3. Surface appearance and nanoporous structure of anodic aluminum oxide anodized in 0.3 M oxalic acid with different anodizing durations and temperatures. (a) Optical and (b) SEM images of nanoporous anodic aluminum oxides. (c) Gray levels obtained from optical images. (d) Pore diameter (Dp), (e) eccentricity, (f) inter-pore distance (Dint) and (g) porosity obtained from SEM images. (h) Thickness of anodic aluminum oxides. White and black scale bars in (a) and (b) indicate 2 cm and 200 nm, respectively. In (c)–(h), the symbols represent the averaged values for the given conditions and the error bars for the standard deviations.

same way. Fig. 4a and b show the optical and SEM images of the AAO surfaces fabricated in surfuric acid with the different anodizing temperatures and durations. Similar to the case of oxalic acid, the surface color of the AAO film becomes darker (i.e., lower value of gray level, see Fig. 4c and Table S2 in Supplementary information) with the increase in the anodizing time at the given anodizing temperature, implying higher thermal emissivity. However, unlike the case of oxalic acid, the surface color becomes darker with the lower anodizing temperature, showing the darkest black color for the AAO layer anodized at 0 °C for 40 min. The estimated pore diameter, eccentricity of pores, inter-pore distance, and porosity from SEM images (Fig. 4b) are shown in Fig. 4d–g and summarized in Table S2 in Supplementary information. Similar to the case of oxalic acid, the AAO layer with larger pores, higher porosity, and lower inter-pore distance is obtained

ratio of pores with the increase of the thickness of the AAO layer at the given pore diameter and porosity contribute to enhance the scattering of light and reducing the intensity of light reflected from the substrate/ AAO interface, the darker surface color appears with the increase in the anodizing time [41,44]. The average growth rates of the AAO films for 0, 15, and 30 °C are 1.35, 1.33, and 1.24 µm/min, respectively. Although the thickness of the AAO layer anodized at the lower temperature is slightly thicker at the given anodizing time, the surface color of the AAO surface anodized at the lower temperature does not appear darker since the pore size and the porosity of the AAO layer anodized at the lower temperature are smaller with the less randomlydistributed pore shape than those at the higher temperature. The investigation of surface color, pore morphology and thickness of the AAO layers anodized in surfuric acid was also conducted in the 507

Nano Energy 31 (2017) 504–513

J. Lee et al.

Fig. 4. Surface appearance and nanoporous structure of anodic aluminum oxide anodized in 1.63 M sulfuric acid with different anodizing durations and temperatures. (a) Optical and (b) SEM images of nanoporous anodic aluminum oxides. (c) Gray levels obtained from optical images. (d) Pore diameter (Dp), (e) eccentricity, (f) inter-pore distance (Dint) and (g) porosity obtained from SEM images. (h) Thickness of anodic aluminum oxides. White and black scale bars in (a) and (b) indicate 2 cm and 200 nm, respectively. In (c)–(h), the symbols represent the averaged values for the given conditions and the error bars for the standard deviations.

film (i.e., the surface nanostructures with higher aspect ratios), which is more pronounced with the increase in the anodizing time. The average growth rates of the AAO films for 0, 15, and 30 °C are 1.68, 1.46, and 1.22 µm/min, respectively. These results imply that if AAO layer has a more regular pore shape with a smaller size and more-ordered pattern, the structural features become less important than the effect of thickness for the saturation of color appearance [41,45]. Fig. 5a shows the average emissivity values of the AAO surfaces made at the different anodizing temperatures and durations in oxalic acid, estimated within the scanned wavelength values of 5–20 µm. The emissivity value measured at each wavelength and averaged emissivity can be found in Fig. S1a and Table S3 in Supplementary information. Considering the heat dissipation toward ambient, the temperature of the anodized surface should be higher than that of ambient so that the

with the increase in the anodizing temperature, while the anodizing duration does not show any significant effect on the pore morphology. However, the results show that the AAO layer obtained with the sulfuric acid has less irregular shape of pores with much smaller size (smaller diameter and lower eccentricity of pore with lower standard deviation) and more ordered pore pattern with a shorter inter-pore distance (lower inter-pore distance with lower standard deviation) than the AAO layer obtained with the oxalic acid. Fig. 4h shows the thicknesses of AAO films fabricated in sulfuric acid with the different anodizing temperatures and durations. Agreeing with the case of oxalic acid, the thickness of the AAO film increases with the increase in the anodizing time at the given temperature, and it is more pronounced at the lower anodizing temperature [37]. At the given anodizing time, a lower anodizing temperature results in a higher thickness of the AAO

508

Nano Energy 31 (2017) 504–513

J. Lee et al.

Fig. 5. (a), (b) Emissivity of the anodized surfaces in oxalic acid (a) and sulfuric acid (b), respectively, with different durations and temperatures. (c), (d) Measured heat flux dissipating from the anodized surfaces toward ambient, including the anodized surfaces in oxalic acid (c) and sulfuric acid (d), respectively, with different durations and temperatures. The dotted lines in (c) and (d) represent the heat flux of bare aluminum alloy (Al 1050) for comparison. In (a)–(d), the symbols represent the averaged values for the given conditions and the error bars for the standard deviations.

were 0.647, 0.631, and 0.534, respectively. Similar to the case of oxalic acid, the emissivity of aluminum alloy with anodizing was significantly increased compared to that of the electropolished bare aluminum surface (0.132). The emissivity was also slightly increased with an increase in the anodizing time, showing the maximum value at the anodizing for 40 min, such as 0.736 (69.2 µm) for 0 °C, 0.710 (60.7 µm) for 15 °C, and 0.580 (50.3 µm) for 30 °C. Similar to the case of oxalic acid, the darker surface color with the increase in the anodizing time attributes the increase in the emissivity. Agreeing with the case of the oxalic acid, this result shows that the thicker AAO film with a darker surface color (i.e., AAO films having the porous structure with higher-aspect-ratio cylindrical pores) can effectively enhance the thermal emissivity [41,44]. However, unlike the case of oxalic acid, the AAO film anodized at the lowest temperature (0 °C) shows the highest emissivity, appearing the darkest and black-like color. It should also be noted that the emissivity values of the AAO surfaces obtained with the sulfuric acids are generally lower than those with the oxalic acids, due to the more ordered pore pattern with smaller and more uniform size and shape of pores. The high amount of sulfate ions and water crystallization contained in the AAO layer fabricated in sulfuric acid is also attributed to further reduce the efficiency of a AAO layer for thermal radiation [37,48–51]. Measured heat flux toward ambient from the anodized surfaces made in oxalic and sulfuric acids with different anodizing time and temperature are shown in Fig. 5c and d, respectively. Overall, the heat flux was significantly increased by the anodizing of the aluminum surface, both with oxalic acid (Fig. 5c) and sulfuric acid (Fig. 5d). Generally, the heat flux was slightly increased with longer anodizing time, which is attributed to the increase of the thermal emissivity. It indicates that the heat dissipation from the aluminum surface can

emissivity was measured at 200 °C. The emissivity values of the AAO surfaces obtained with the oxalic acids at 0, 15, and 30 °C for 10 min were 0.746, 0.820, and 0.801, respectively. The emissivity of aluminum alloy with anodizing was significantly increased compared to that of the electropolished bare aluminum surface (0.132). The emissivity was slightly increased with an increase in the anodizing time, showing the maximum value at the anodizing for 40 min, such as 0.788 (47.8 µm) for 0 °C, 0.875 (40.6 µm) for 15 °C, and 0.845 (36.2 µm) for 30 °C. The increase in the emissivity with the increase in the anodizing time is attributed to the increase in the AAO film thickness with a darker surface color (i.e., the porous surface nanostructures with higher aspect ratios at the given anodizing temperature). In case of the AAO film anodized at 0 °C, the emissivity was the lowest despite the thicker AAO layer than those anodized at higher temperatures, appearing rather brownish than black. It is attributed to the smaller pore diameter and porosity with the less randomly-distributed pore shape, which negate the effects of the thickness of the AAO film. In contrast, the AAO layer anodized at 15 °C shows the highest emissivity, indicating that the AAO film with the more randomly-distributed pore structures with higher aspect ratios at the given pore diameter and porosity can enhance the thermal emissivity more effectively due to the significantly increased effective surface area for the radiation as well as the scattering/ absorption of light [40,46,47]. Fig. 5b shows the average emissivity values of the AAO surfaces made at the different anodizing temperatures and durations in sulfuric acid, estimated within the scanned wavelength values of 5–20 µm. The emissivity value measured at each wavelength and averaged emissivity can also be found in Fig. S1b and Table S4 in Supplementary information. After the anodizing for 10 min, the emissivity values of the AAO surfaces obtained with the sulfuric acids at 0, 15, and 30 °C 509

Nano Energy 31 (2017) 504–513

J. Lee et al.

and SEM images, the thermal emissivity, and the heat flux data of the AAO films showing the best performance in each anodizing case are shown in Fig. 6. Fig. 6a shows the apparent colors of the AAO surfaces with the sealing treatments. In case of the hydrothermal and the cold NiF2 sealings, the changes in apparent color were not significant after the sealings for the AAO films anodized in both oxalic and sulfuric acids. However, the nanoporous AAO surface with black sealing shows a darker black color (i.e., lower gray level obtained from optical images; see details in Table S5 in Supplementary information), compared to the AAO with no sealing and also other the sealings. Fig. 6b shows the SEM images of the AAO surfaces with hydrothermal, cold NiF2, and black sealing treatments. Whereas the AAO surfaces anodized in oxalic and sulfuric acids showed different morphology of pore structures, no significant differences in the surface morphology are found after the sealings since the surfaces of the nanoporous AAO layers are now covered with precipitated sealants (hydrated alumina for hydrothermal sealing; Al(OH)3, Ni(OH)2, and AlF3 for cold NiF2 sealing; cobalt sulfide (CoS) for black sealing) accompanied with a dissolution of the AAO nanostructures during the sealing reaction [31,32,48,51,55,57,58]. No significant difference in the surface morphology of the AAO layers is also found for the different sealing treatments. The surface thermal emissivity for varying wavelength and the heat flux measured for the AAO surfaces with sealings are shown in Fig. 6c and d, respectively. The averaged emissivities of the surfaces are also summarized in Table 2. In case of the hydrothermal sealing, the changes in both emissivity and heat flux were not significant after the sealing for the AAO films anodized in both oxalic and sulfuric acids. However, the cold NiF2 and the black sealing resulted in noticeable enhancement of the thermal emissivity and heat dissipation, compared to the AAO films with no sealing. The emissivity and heat flux of the AAO film with cold NiF2 sealing were 0.883 (0.9% increase compared with no sealing) and 4623.6 W/m2 (3.0% increase compared with no sealing) for oxalic acid anodizing, and 0.744 (1.1% increase compared with no sealing) and 3956.3 W/m2 (3.8% increase compared with no sealing) for sulfuric acid anodizing, respectively. More significant improvement of the thermal emissivity and heat flux were measured after black sealing, such as 0.906 (3.5% increase compared with no sealing) and 5088.0 W/m2 (13.3% increase compared with no sealing) for the AAO surface anodized in oxalic acid, and 0.749 (1.8% increase compared with no sealing) and 4096.1 W/m2 (7.5% increase compared with no sealing) for the AAO surface anodized in sulfuric acid. The result also shows that, similar to the AAO films with no sealing, the higher emissivity and heat flux were obtained even after the sealing with the AAO film anodized in oxalic acid than that in sulfuric acid. In this study, the nanoporous AAO film anodized at 15 °C in oxalic acid along with the black sealing showed the highest enhancement in heat dissipation (256.4% increase compared with the bare aluminum alloy) with the largest improvement of the thermal emissivity (686.4% increase compared with the bare aluminum alloy). As discussed earlier, the difference in the nanoporous structural morphology depending on the anodizing conditions becomes ineffective after the sealings and the surface morphology after the sealings are not also significantly different regardless of the different sealing methods. Thus, the different behaviors in the enhancement of the thermal emissivity and the heat flux for the different sealing methods are attributed to the nature of the precipitated sealants for each sealing method. In case of the hydrothermal sealing, it leads to react the pore walls of the AAO film with water to precipitate hydrated alumina (e.g., boehmite (AlOOH)), which has similar chemical component with AAO [48,50,56,59]. In contrast, during the cold NiF2 sealing, Al(OH)3, Ni(OH)2, and AlF3 are precipitated in pores [56–58]. In case of the black sealing, naturally black-colored cobalt sulfide (CoS) is formed in pores [31,32]. Such heterogeneous materials formed in the cold NiF2 sealing or the black sealing, compared to the hydrated alumina formed in the hydrothermal sealing, are attributed to improve the emissivity of the AAO film [46,47]. In particular, heterogeneous materials with black

significantly be enhanced by the nanoporous oxide layer resulted from anodizing. Specifically, in case of the AAO anodized in oxalic acid (Fig. 5c), the heat flux measured from the anodized surface for 10 min were 3904.0, 4265.6, and 4204.0 W/m2 for 0, 15, and 30 °C, respectively, which are significantly higher than the heat flux from the electropolished bare aluminum 1050 alloy (1984.5 W/m2). The heat flux was slightly increased with the anodized surfaces with a longer anodizing duration, showing the highest heat flux with the anodized surfaces for 40 min, such as 4120.4 W/m2 for 0 °C, 4490.0 W/m2 for 15 °C, and 4366.0 W/m2 for 30 °C. In case of the oxalic acid, the highest emissivity was measured with the anodized surface at 15 °C. The heat flux of the AAO surfaces anodized at 15 °C in oxalic acid was greater than those at 0 and 30 °C by 8.2% and 2.8% in average, respectively. The measured heat flux of the AAO surfaces in the different anodizing time and temperature is in good agreement with the emissivity results, verifying that the increased thermal emissivity by the surface nanostructures resulting from the anodizing is effective to enhance the heat dissipation from the surface. In case of the AAO anodized in sulfuric acid (Fig. 5d), the heat flux measured from the anodized surface for 10 min were 3600.0 W/m2 for 0 °C, 3492.4 W/m2 for 15 °C, and 3005.2 W/m2 for 30 °C, respectively. Although these values are slightly lower than those in oxalic acid for the same anodizing time, which is mainly due to the lower emissivity value of the AAO surface anodized in sulfuric acid, they are still significantly higher than the heat flux from the electropolished bare aluminum 1050 alloy, verifying again that the heat dissipation from the aluminum surface can significantly be enhanced by the nanoporous oxide layer resulting from anodizing. Similar to the case of oxalic acid, the heat flux was slightly increased with the anodized surfaces with a longer anodizing duration, showing the highest heat flux with the anodizes surfaces for 40 min, such as 3810.4 W/m2 for 0 °C, 3715.6 W/m2 for 15 °C, and 3184.6 W/m2 for 30 °C. In case of the sulfuric acid, the heat flux from the AAO surface anodized at 0 °C was highest, being higher than those at 15 and 30 °C by 2.5% and 16.4% in average, respectively. Unlike the case of oxalic acid, the higher emissivity of the AAO surface was obtained with the lower anodizing temperature in case of the sulfuric acid. Thus, this result also supports that the surface heat dissipation can effectively be improved by the enhanced surface emissivity due to the surface nanostructures formed by anodizing. Meanwhile, it should be noted that the AAO film fabricated in sulfuric acid contains a lot of hydrated aluminum oxide (Al2O3) and aluminum sulfate (Al2(SO4)3), which are reported to reduce the velocity of phonon [37,48,52]. Thus, the thermal conductivity of the AAO film made in sulfuric acid is considerably lower than that fabricated in oxalic acid. When considering the thermal conduction through the AAO film, the low thermal conductive oxide layer works as a thermal barrier, reducing the heat flow. In addition to the lower emissivity of the AAO surface anodized in sulfuric acid, the lower conductivity also contributes to the lower heat flux from the AAO surface made with sulfuric acid, compared to the heat flux from the AAO surface made with oxalic acid. 3.3. Sealing of nanoporous anodic aluminum oxide In order to improve surface properties such as corrosion and wear resistance, the pores of AAO films are usually filled with solid materials by sealing as a post-treatment of anodizing [51,53–55]. Such a sealing also affects and can improve the surface's thermal emissivity as well as heat dissipation. In this study, three different types of sealing methods including hydrothermal, cold NiF2, and black sealing were applied to the nanoporous AAO layers to investigate their combined effects on the thermal emissivity and heat dissipation [31,32,56]. Following the same trends of the AAO films with no sealing treatment, the highest emissivity and heat flux were obtained with the sealings with the AAO film anodized at 15 °C for 40 min for oxalic acid, and the AAO film anodized at 0 °C for 40 min for sulfuric acid, respectively. The optical 510

Nano Energy 31 (2017) 504–513

J. Lee et al.

Fig. 6. (a) Optical images and (b) SEM images of anodized AAO surface (for 40 min in 0.3 M oxalic acid at 15 °C and 1.63 M sulfuric acid at 0 °C) without (w/o) sealing and with hydrothermal (HT), cold NiF2 (CN) and black (BL) sealing. (c) Surface emissivity and (d) measured heat flux of anodized aluminum 1050 alloy with sealings. In (c), (d), the symbols represent the averaged values for the given conditions and the error bars for the standard deviations.

uated by water cooling test. A schematic diagram of the test setup is shown in Fig. 7a. The anodizing and sealing were made only on the outer surface of a cylindrical Al 1050 tube. The inside of the tube was filled with hot water at 92 °C, and its temperature transient to ambient in cooling was recorded for 30 min. The cooling rate of the water was estimated to verify the heat dissipation performance of the anodized aluminum with sealing, compared to that without sealing as well as the bare Al 1050. The result of the water temperature as a function of time is shown in Fig. 7b. In the case of a bare Al 1050, the temperature of water was dropped from 92 to 53.2 °C for 30 min (cooling rate of 1.29 °C/min). In the case of anodized surfaces without sealing, the water temperatures were decreased to 43.8 °C (cooling rate of 1.61 °C/ min) and 47.9 °C (cooling rate of 1.47 °C/min) for the anodized surfaces in oxalic and sulfuric acids, respectively. The lower temperatures of water in the cases of anodized Al 1050 tubes indicate improved heat dissipation, resulting from the enhanced surface emissivity by the nanoporous oxide layers. The final water temperature was lower in the case of the anodized surface in oxalic acid than that in the sulfuric acid, which is due to the higher emissivity and the heat flux shown in the anodized surface in oxalic acid than that in the sulfuric acid. In the case of the anodized surfaces with the black sealing, the decrease of the temperature was more significant, such as 39.3 °C (cooling rate of 1.76 °C/min) for the anodized surface in oxalic acid and 45.5 °C (cooling rate of 1.55 °C/min) for the anodized surface in sulfuric acid. Agreeing with the higher emissivity and heat flux shown in the anodized surfaces with black sealing than those without sealing, the final temperatures were lower with the anodized surfaces with black sealing, compared those without sealing. However, it should be noted that the final temperature for the anodized surface in sulfuric acid with

Table 2 Averaged emissivity values of anodized aluminum (15 °C oxalic acid and 0 °C sulfuric acid, respectively, both for 40 min) with the different sealing methods.

Without sealing Hydrothermal sealing Cold NiF2 sealing Black sealing

Oxalic acid

Sulfuric acid

0.875 0.876 0.883 0.906

0.736 0.740 0.744 0.749

color (i.e., CoS) are more favorable to enhance the surface emissivity since the surface absorbance of light is identical with emissivity in such black-colored materials. Therefore, the naturally black-colored material such as CoS formed in the black sealing is more effective to enhance the emissivity of the AAO film. In addition, the solid-sate sealing of the nanoporous oxide layer helps to improve the thermal conductivity of the AAO films and hence the heat dissipation [52]. As revealed in Fig. 6, the results demonstrate that the heat dissipation performance of the nanoporous AAO layer can further be improved by proper sealing processes such as cold NiF2 and black sealing, which will fill the empty nanopores with heterogeneous materials that help to increase the thermal emissivity as well as the thermal conductivity.

3.4. Cooling performance The heat dissipation performances of the anodized aluminum surfaces (15 °C for 40 min in oxalic acid and 0 °C for 40 min in sulfuric acid, respectively) without and with black sealing were further eval511

Nano Energy 31 (2017) 504–513

J. Lee et al.

(a)

Fig. 7. (a) Schematic diagram of cooling test setup and (b) the measurement result of water temperature in the aluminum 1050 alloy tube with anodizing (for 40 min in 0.3 M oxalic acid at 15 °C and 1.63 M sulfuric acid at 0 °C) and black sealing (BL).

Acknowledgements

black sealing was higher than that for the anodized surface in oxalic acid without sealing. It is because the emissivity of the anodized surface in sulfuric acid even with the black sealing did not exceed that of the anodized surface in oxalic acid without sealing. The result of the heat dissipation test by water cooling demonstrates that the enhancement of the heat dissipation of an aluminum surface by anodizing and additional sealing is effective due to the enhanced emissivity and radiative heat transfer. Aluminum alloys have excellent thermal conductivity with a light weight, and thus they are widely used in various industrial fields, especially for heat exchange systems. In addition, anodizing and sealing processes are commercially or readily available with advantages in cost efficiency and mass productivity. They can also be applied to substrates with complex shapes. Therefore, the heat dissipation and cooling performances of heat exchange systems or parts composed of aluminum substrates can significantly be enhanced by tailored anodizing and sealing processes as demonstrated in this study.

This work was supported by the Korean Ministry of Trade, Industry and Energy, Korea (MOTIE) under Grant M0000529. The anodizing work was also partly supported by the US Office of Naval Research (ONR) Award N00014-14-1-0502. Appendix A. Supplementary information Supplementary data associated with this article can be found in the online version at doi:10.1016/j.nanoen.2016.12.007. References [1] [2] [3] [4] [5] [6] [7]

4. Conclusions

[8]

The significantly improved heat dissipation of aluminum alloy surface was achieved by the surface treatment of anodizing and sealing that led to increase the radiative emissivity. The anodizing of aluminum allows the surface to have nanoporous oxide layer which can itself enhance the thermal emissivity but also to provide nanoporous reservoirs that can be sealed with solid heterogeneous materials that can further enhance the thermal emissivity. The increase of the thermal emissivity of the nanoporous anodized aluminum oxide layer got more significant with the surface with a darker black color (i.e., a thicker oxide layer with high-aspect-ratio nanopores), enabling to enhance the heat dissipation from the surface. The nanoporous oxide layer anodized in oxalic acid was more effective to improve the thermal emissivity and the heat dissipation performance than that anodized in sulfuric acid due to the more irregular pore size and shape with the more disordered pattern, which lead to make the oxide surface darker black in color. Black and cold NiF2 sealings of the anodized nanoporous oxide layer resulted in additional improvement in emissivity and heat dissipation. Especially, black sealing to the nanoporous oxide layer anodized in oxalic acid resulted in the largest enhancement in emissivity up to 0.906, which is seven times greater than that of bare aluminum. Consequently, the aluminum oxide film anodized in oxalic acid with black sealing, having the emissivity close to that of a black body, significantly improved the radiative heat dissipation and cooling capability of the aluminum alloy surface. The advanced understanding and the real demonstration of the enhanced thermal emissivity and heat dissipation performance of the aluminum alloy by anodizing and sealing are of a great significance in the design and the practical applications of aluminum substrates for effective heat management.

[9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30]

[31] [32]

512

B. Givoni, Energy Build. 17 (1991) 177–199. A.A. Dehghan, M. Behnia, J. Heat Transf. 118 (1996) 56–64. B.K. Larkin, S.W. Churchill, AIChE J. 5 (1959) 467–474. R. Petela, J. Heat Transf. 86 (1964) 187–192. X. Cheng, X. Liang, Int. J. Heat Mass Transf. 54 (2011) 269–278. F.C. Lockwood, N.G. Shah, Symp. (Int.) Combust. 18 (1981) 1405–1414. D. Kim, J. Lee, J. Kim, C.-H. Choi, W. Chung, Energy Convers. Manag. 106 (2015) 958–963. M. Brogren, G.L. Harding, R. Karmhag, C.G. Ribbing, G.A. Niklasson, L. Stenmark, Thin Solid Films 370 (2000) 268–277. D. Willits, Biosyst. Eng. 84 (2003) 315–329. A. Agudelo, C. Cortés, Energy 35 (2010) 679–691. A. Srivastava, J. Nayak, G. Tiwari, M. Sodha, Energy Build. 6 (1984) 3–13. S. Rau, S. Vierrath, J. Ohlmann, A. Fallisch, D. Lackner, F. Dimroth, T. Smolinka, Energy Technol. 2 (2014) 43–53. A.P. Raman, M.A. Anoma, L. Zhu, E. Rephaeli, S. Fan, Nature 515 (2014) 540–544. R. Belarbi, F. Allard, Renew. Energy 22 (2001) 507–524. J. Cho, K.E. Goodson, Nat. Mater. 14 (2015) 136–137. M. Chandrasekar, S. Suresh, T. Senthilkumar, Energy Convers. Manag. 71 (2013) 43–50. A. Vidal, Remote Sens. 12 (1991) 2449–2460. D. Starikov, C. Boney, R. Pillai, A. Bensaoula, G. Shafeev, A. Simakin, Infrared Phys. Technol. 45 (2004) 159–167. Y. Wang, H. Tian, X. Shen, L. Wen, J. Ouyang, Y. Zhou, D. Jia, L. Guo, Ceram. Int. 39 (2013) 2869–2875. X. Yan, G. Xu, Mater. Sci. Eng.: B 166 (2010) 152–157. R. Raman, A. Thakur, Appl. Energy 12 (1982) 205–220. W.C. Michels, S. Wilford, J. Appl. Phys. 20 (1949) 1223–1226. F. Watanabe, J. Vac. Sci. Technol. A 11 (1993) 432–436. D. Kim, D. Sung, J. Lee, Y. Kim, W. Chung, Appl. Surf. Sci. 357 (2015) 1396–1402. G. Zhang, S. Jiang, H. Zhang, W. Yao, C. Liu, RSC Adv. 6 (2016) 61686–61694. T. Eyassu, T.-J. Hsiao, K. Henderson, T. Kim, C.-T. Lin, Ind. Eng. Chem. Res. 53 (2014) 19550–19558. C.N. Suryawanshi, C.T. Lin, ACS Appl. Mater. Interfaces 1 (2009) 1334–1338. M. Zhou, T. Lin, F. Huang, Y. Zhong, Z. Wang, Y. Tang, H. Bi, D. Wan, J. Lin, Adv. Funct. Mater. 23 (2013) 2263–2269. Y.L. Hsin, C.K. Chang, M.H. Wang, H.C. Yeh, T.Y. Su, J. Chin. Chem. Soc. 60 (2013) 1309–1316. C.L. Chapman, S. Lee, B.L. Schmidt, Semiconductor thermal measurement and management symposium, SEMI-THERM X., in: Proceedings of the 10th IEEE/ CPMT, 1994, pp. 24–31 Y. Goueffon, L. Arurault, C. Mabru, C. Tonon, P. Guigue, J. Mater. Process. Technol. 209 (2009) 5145–5151. Y. Goueffon, L. Arurault, S. Fontorbes, C. Mabru, C. Tonon, P. Guigue, Mater.

Nano Energy 31 (2017) 504–513

J. Lee et al. Chem. Phys. 120 (2010) 636–642. [33] M. Zemanova, M. Chovancova, Z. Galikova, P. Krivošík, Renew. Energy 33 (2008) 2303–2310. [34] T. Nomura, K. Tabuchi, C. Zhu, N. Sheng, S. Wang, T. Akiyama, Appl. Energy 154 (2015) 678–685. [35] S. Dhibar, P. Bhattacharya, G. Hatui, S. Sahoo, C. Das, ACS Sustain. Chem. Eng. 2 (2014) 1114–1127. [36] J.R. Howell, R. Siegel, M.P. Menguc, Thermal radiation heat transfer, 5th ed., CRC press, Boca Raton, FL, 2010. [37] J. Lee, Y. Kim, U. Jung, W. Chung, Mater. Chem. Phys. 141 (2013) 680–685. [38] ASTM, D5470-06: standard test method for thermal transmission properties of thermally conductive electrical insulation materials, in: ASTM Int., America, 2006 [39] J.C. Tan, S.A. Tsipas, I.O. Golosnoy, J.A. Curran, S. Paul, T.W. Clyne, Surf. Coat. Technol. 201 (2006) 1414–1420. [40] K. Kant, S.P. Low, A. Marshal, J.G. Shapter, D. Losic, ACS Appl. Mater. Interfaces 2 (2010) 3447–3454. [41] Q. Xu, H.-Y. Sun, Y.-H. Yang, L.-H. Liu, Z.-Y. Li, Appl. Surf. Sci. 258 (2011) 1826–1830. [42] J.W. Diggle, T.C. Downie, C.W. Goulding, Chem. Rev. 69 (1969) 365–405. [43] X. Hu, Z.-Y. Ling, S.-S. Chen, X.-H. He, Chin. Phys. Lett. 25 (2008) 3284. [44] W. Lee, J.-C. Kim, U. Gösele, Adv. Funct. Mater. 20 (2010) 21–27. [45] J. Zhu, Z. Yu, G.F. Burkhard, C.-M. Hsu, S.T. Connor, Y. Xu, Q. Wang, M. McGehee, S. Fan, Y. Cui, Nano Lett. 9 (2008) 279–282. [46] I. Laurent, M. Mario, B. Abderrahim, D. Stefan, C. Yves, L. Jean, Meas. Sci. Technol. 17 (2006) 2950. [47] X.H. Wang, T. Akahane, H. Orikasa, T. Kyotani, Y.Y. Fu, Appl. Phys. Lett. 91 (2007) 011908. [48] S. Feliu, M.J. Bartolomé, J.A. González, S. Feliu, J. Electrochem. Soc. 154 (2007) C241–C248. [49] J.A. Treverton, N.C. Davies, Electrochim. Acta 25 (1980) 1571–1576. [50] B. Schnyder, R. Kötz, J. Electroanal. Chem. 339 (1992) 167–185. [51] V. López, M.J. Bartolomé, E. Escudero, E. Otero, J.A. González, J. Electrochem. Soc. 153 (2006) B75–B82. [52] J. Lee, Y. Kim, J. Kim, W. Chung, J. Nanoelectron. Optoelectron. 9 (2014) 136–140. [53] J.A. Gonzalez, V. Lopez, E. Otero, A. Bautista, R. Lizarbe, C. Barba, J.L. Baldonedo, Corros. Sci. 39 (1997) 1109–1118. [54] J. Lee, Y. Kim, H. Jang, W. Chung, Surf. Coat. Technol. 243 (2014) 34–38. [55] N. Hu, X. Dong, X. He, J.F. Browning, D.W. Schaefer, Corros. Sci. 97 (2015) 17–24. [56] J. Lee, U. Jung, W. Kim, W. Chung, Appl. Surf. Sci. 283 (2013) 941–946. [57] J.A. González, S. Feliu Jr, A. Bautista, E. Otero, S. Feliu, J. Appl. Electrochem. 29 (1999) 845–854. [58] H. Herrera‐Hernandez, J.R. Vargas‐Garcia, J.M. Hallen‐Lopez, F. Mansfeld, Mater. Corros. 58 (2007) 825–832. [59] Y. Chang, Z. Ling, Y. Li, X. Hu, Electrochim. Acta 93 (2013) 241–247.

Donghyun Kim received his Ph.D. in Materials Science and Engineering from the Pusan National University, South Korea, 2016. Then, he joined the Center of Analysis & Certification at Korea Institute of Ceramic Engineering & Technology (KICET), South Korea. His researches are focused on electrochemical surface treatment and thermal radiative heat dissipation.

Chang-Hwan Choi joined Department of Mechanical Engineering at Stevens Institute of Technology as an Assistant Professor in 2007 and has been working as an Associate Professor since 2013. He received his Ph.D. in Mechanical Engineering from University of California at Los Angeles in 2006, specializing in MEMS/ Nanotechnology. In 2010, he received the Young Investigator Program award by US Office of Naval Research. In 2015, he was awarded the Humboldt Research Fellowship for Experienced Researchers by the Alexander von Humboldt Foundation. His current research includes large-area nanopatterning, fluid/thermal physics at nanoscale interfaces, and applications of nanostructured surfaces for energy saving/harvesting.

Wonsub Chung received his Ph.D. in Materials Science and Engineering from the Kyushu University, Japan, in 1989. From 1989 to 1992, he was a tenure-track researcher at Research Institute for Science and Technology (RIST) and then he joined the Department of Materials Science and Engineering at Pusan National University, South Korea. He is leading a nano-electrochemistry lab pursuing researches aimed to develop surface treatment method for metallic materials. His research interests include the corrosion prevention of metallic materials, thermal radiative materials for heat transfer and energy saving.

Junghoon Lee received his Ph.D. in Materials Science and Engineering from the Pusan National University, South Korea in February 2014. He is currently a postdoctoral scholar in the Department of Mechanical Engineering at Stevens Institute of Technology. His research interests include electrochemical anodic oxidation and deposition for corrosion prevention and heat transfer of metallic materials. He also works on development of lubricant infused nanoporous metal oxides for anti-corrosion, anti-biofouling and heat transfer.

513