An efficient cooling system without any external energy consumption that consists of a plate, on which an array of copper nanorods with an average diameter ~100 nm and length ~500 nm is integrated to a planar copper thin film coated silicon wafer surface, a heater, an aluminum base, and a pool was developed. Heat is efficiently transferred from the nanostructure coated base plate to the liquid in the pool through mechanisms of boiling heat transfer. Phase change took place near the nanostructured plate, where the bubbles started to emerge due to wall superheat. Bubble formation and bubble motion inside the pool created an effective heat transfer from the plate surface to the pool. The recorded surface temperature at boiling inception was 102.1°C without the nanostructured plate and it was successfully decreased to 100°C with the use of nanostructured plate. In this study, it is shown that a nanostructured surface approach can have the potential to be an effective method of device cooling for small and excessive heat generating micro-system applications such as micro-electro-mechanical-systems or micro-processors.
With the miniaturization of micro-processors and micro-chips, an increasing trend in their power density is inevitable. As a result, there is an urgent need for micro heat sinks with low thermal resistance. Besides electronics cooling, micro heat sink technology also finds applications in micro-reactors, micro-propulsion, biotechnology, fuel cells and air conditioning.
Heat and fluid flow (both single-phase flow and flow boiling) in microscale has been rigorously studied to achieve the goal of higher heat removal capabilities. Recently, nanostructured surfaces have been utilized to achieve high heat transfer performance due to enhanced heat transfer area and positive effect on heat transfer coefficients with diminishing length scale . Moreover, nanostructures also provide additional active nucleate sites so that they could promote nucleate heat transfer in boiling .
The applications of nanostructured surfaces in boiling mainly focus on pool boiling. Recent results of pool boiling on nanofluids [1, 3-8] and nanostructured surfaces [2, 9-11] have shown significant heat transfer enhancement compared to plain surface and unseeded liquids, respectively. The investigators working on pool boiling with nanofluids detected nanoparticle coating on their heater surface, which modified the surface characteristics [1-7]. They could visualize the increase in surface roughness with nanoparticle surface coating and the decrease in contact angle (thus the increase in wettability), both of which contributed to enhance critical heat flux (CHF). By this way, researchers were able to obtain high CHF values using pure water on nanoparticle coated surfaces. Significant increases in heat transfer coefficients and the CHF, and dramatic reductions in boiling inception temperatures have been reported by independent research groups dealing with nanostructured surfaces and nanofluids in pool boiling [1-11]. However these studies generally lack a controlled method of nanostructured coating that limits the fundamental understanding of heat-transfer mechanisms in nanoscales as well as applications of such approaches in cooling systems. In this letter, we present a unique method of nanostructured coating for micro-cooling system, with capability of producing nano-features of various shapes, dimensions, and material types. In our studies, preliminary tests on a copper nanorod array coated pool boiler were conducted and boiling curves obtained were compared to the ones from a conventional planar copper thin film surface configuration. The potential use of such a compact nanostructured pool boiler having no pumping and moving components in microscale cooling applications was exploited (up to about 10W/cm2) and promising results were obtained.
Glancing angle deposition (GLAD) technique [12-15] is a physical self-assembly growth technique that provides a novel capability of growing 3D nanostructure arrays with interesting material properties such as high electrical/thermal conductivity and also reduced oxidation compared to the polycrystalline films. It is a simple and single-step process that offers a cost and time efficient method to fabricate nanostructured arrays of various materials in the periodic table as well as compounds, alloys, and oxides. The GLAD technique uses the "shadowing effect," which is a "physical self-assembly" process through which obliquely incident atoms preferentially deposit on higher surface points of a rotating substrate (Figure 1) leading to an isolated columnar morphology. Due to the statistical fluctuations in the growth and effect of initial substrate surface roughness or pattern, some surface sites grow faster in the vertical direction. Due to their higher height, they capture most of the obliquely incident particles, while the shorter surface points get shadowed and cannot grow anymore. Through the control of deposition parameters of GLAD such as angle of oblique incidence flux, substrate rotation speed, and substrate rotation, it is possible to obtain a wide variety of nanostructured arrays with different shapes (rods, springs, zigzags, etc.) and sizes (from tens to hundreds of nanometer). In addition, vertical nanorod arrays produced by GLAD have been observed to be single crystal [16-18] that increases their resistance to oxidation due to lack of grain boundaries, and therefore making them superior conductors of heat and electricity. Previously, Li et al.  demonstrated that tilted copper nanorod arrays produced by an oblique angle deposition technique (which is similar to the GLAD method but without substrate rotation) can significantly boost bubble formation and enhance boiling heat transfer. However, tilted nanorods are more prone to oxidation due to their polycrystalline property that can result in poorer stability and robustness. In addition, tilted nanorods produced by oblique angle deposition without substrate rotation cannot be easily produced with controlled diameters and separations, making systemic investigations towards fundamental understanding of heat transfer from nanostructured surfaces and their implementations in cooling applications more difficult. Therefore, we believe that vertically aligned nanorods (with substrate rotation, i.e. by GLAD) have a potential to further improve the nucleate boiling and boiling heat transfer compared to tilted nanorods (without substrate rotation).
The schematic of the custom-made GLAD experimental setup in the present study is shown in Figure 1. For the fabrication of vertically aligned Cu nanorods array, the DC magnetron sputter GLAD technique is employed. Cu nanorods were deposited on Cu thin film surface, which is coated on Si wafer (100) substrates (1 x 1 cm2) using a 99.9% pure Cu cathode (diameter about 7.6 cm). The substrates were mounted on the sample holder located at a distance of about 12 cm from the cathode. During the growth, the substrate was tilted so that the angle Î¸ between the surface normal of the target and the surface normal of the substrate is 85°. The substrate was attached to a stepper motor and rotated at a speed of 1 rpm for growing vertical nanorods. The depositions were performed under a base pressure of 5 x 10-7 Torr which was achieved by utilizing a turbo-molecular pump backed by a mechanical pump. During Cu deposition experiments, the power was 200 W with an ultrapure Ar working gas pressure of 2.5 mTorr and the maximum temperature of the substrate during growth was below ~85 °C. The deposition time of GLAD deposited Cu nanorods was 60 min. For comparison, planar Cu thin film samples (which will be also referred to as "plain surface" in the following text) were also prepared by normal incidence deposition (Î¸ = 0o) with a substrate rotation of 1 rpm. The film thickness of the vertical columns was measured utilizing quartz crystal microbalance (Inficon- Q-pod QCM monitor, crystal: 6 MHz gold coated standard quartz) measurements and cross-sectional SEM image analysis to be ~8.6 nm/min. The scanning electron microscopy (SEM) unit (FESEM-6330F, JEOL Ltd, Tokyo, Japan) was used to study the morphology of the deposited nanorods. The top and side SEM images of Cu nanorods are shown in Figure 2 in which an isolated columnar morphology can be seen. However, surface of the conventional Cu film deposited at normal incidence was smooth as indicated by the SEM images (not shown here). At early stages of GLAD growth, the number density of the nanorods was larger, and they have diameters as small as about 5-10 nm. As they grow longer and some of the rods stop growing, due to the shadowing effect, their diameter grows up to about 100 nm. The height of an individual rod is about 500 nm and the average gap among the nanorods also changes with their length from 5-10 nm up to 50-100 nm at later stages. As can be seen from Figure 2a, the tops of vertical nanorods have pyramidal shapes with four facets, which indicate that an individual rod has a single crystal structure. This observation was confirmed by previous studies [16-18] which reported that individual metallic nanorods fabricated by GLAD are typically single crystal. Single crystal rods do not have any interior grain boundaries and have faceted sharp tips. This property will allow reduced surface oxidation which can greatly increase the thermal conductivity, robustness, and resistance to oxidation-degradation of our nanorods in the present study.
The experimental setup for the heat transfer characterization is illustrated in Figure 3. Aluminum base has air gaps on four sides to enhance heat transfer with minimum loss from the heater placed beneath the aluminum block. A container made of Plexiglas is closely fitted on top of the aluminum block to create the desired pool for the pool boiling experiments on the nanostructured plate. The heat generated by the film miniature heater is delivered to the nanostructured plate of size 1.7cmx1.5cm through the base. It provides constant heat flux to the system with constant voltage applied from the electrodes of the film heater. The heat flux values are calculated with the division of the wattage readings from the power supply by the tabulated heater active surface area. Heat losses are obtained from commercial software simulation and were minor compared to electrical power since the system is compact and isolated during experiments. Water is filled to the pool and all the results are recorded when water level is 5ml above the nanostructured plate. Thermocouples are placed near the nanostructured plate at different places for the accurate measurement of the surface temperature and an almost uniform temperature profile was observed. Since the nanostructured plate thickness is on the order of hundreds of microns, no significant temperature variation across the plate has been observed.
After the experimental setup is prepared as explained, the surface temperature readings are recorded as a function of the input voltage and passing current through the heaters by the readings from the power supply. The effective areas of the heaters are tabulated within the manufacturer's guide and their values are extracted from there. These values are used to calculate the constant heat flux input to the system. At certain values of the constant heat flux, steady state surface temperature values are recorded by the thermocouples until boiling started (referred to as single phase) and during boiling (referred to as two phase). The experiment is conducted first without the nanostructured plate to clearly account for the positive effects of the nanostructured plate.
Results and Discussion
Experimental results are shown in Figure 4a, 4b and 4c. The effect of the nanostructured plate is clearly observed from the difference in the superimposed graphs. The nanostructured plate increases heat removal rate from the system. It also decreases the boiling inception temperature by 2°C. The nanorods on the surface of the plate act effectively in the enhancement of boiling heat transfer. The data presented in Figure 4a shows the superimposed two-phase data from the experiments with and without the nanostructured plate during boiling. These results show that in the boiling region the rise in the surface temperature is suppressed with the application of the nanostructured plate. The reason could be explained by the increase in heat transfer area and the number of active nucleate sites so that more bubbles would emerge during boiling from the surface and promote nucleate boiling. This facilitates enhanced heat removal from the surface of the plate and leads to stabilization of the surface temperature (Figure 4d). In addition, recent studies [2,19] have shown a significant reduction in the macroscopic water contact angle of some metallic nanorods (such as Pt and Cu), implying the increased wettability due to the enhanced roughness caused by the nanorod structure which, in turn, contributes to enhance critical heat flux (CHF).
Heat removal in the single-phase region is also promoted with the introduction of the nanostructured plate. The single-phase linear slopes are evaluated and 13% decrease in the slope is observed with the nanostructured plate. Thus, even in the single-phase the effect of the nanostructured plate is significant due to heat transfer area enhancement (Figure 4b).
During boiling, heat transfer coefficients are deduced from surface temperatures and displayed along with heat flux in Figure 4c. The results indicate that the heat transfer coefficient behavior has improved with the nanostructured surface relative to the plain surface configuration.
The results gathered from the experiments support the advantageous effects of nanorod integrated thin plates on heat transfer magnification and nucleate boiling promotion. Even for a small area of 1.7cmx1.5cm, the nanorod integrated plate acts efficiently. Using these results, possible further models and experiments, nanorod integrated plates could be used in various cooling applications of small electronic devices, microreactors, micropropulsion, biotechnology, fuel cells and air conditioning.
The authors would like to thank the UALR Nanotechnology Center and Dr. Fumiya Watanabe for his valuable support and discussions during SEM measurements.