Copyright © 2019 by American Scientific Publishers
Journal of Nanofluids Vol. 8, pp. 1–11, 2019 (www.aspbs.com/jon)
All rights reserved. Printed in the United States of America
Thermal Enhancement Using Nanofluids on High Heat Dissipation Electronic Components Roger R. Riehl National Institute for Space Research, INPE – Space Mechanics and Control Division – DMC Av dos Astronautas 1758, São José dos Campos, 12227-010 SP, Brazil
KEYWORDS: Thermal Enhancement, Electronics Cooling, Thermal Control, Pressure Drop, Nanofluids.
1. INTRODUCTION Following today’s needs for improvement on heat transfer, new technologies and innovative solutions must be found in order to meet current requirements for both active and passive thermal control. There has been substantial growth in the heat fluxes that need to be dissipated, which require different approaches from designers with more efficient thermal management systems. With the increase of heat dissipation needs, conventional designs are not suitable due to several factors such as operation in hostile environments and high heat densities of electronics that need their temperature to be controlled. In such cases, the application of nanofluids can greatly contribute to give designers more degrees of freedom to face projects’ requirements. The need for thermal management has increased over the last decade and the prediction is that a steeper increase is yet to come for the next years. Such an increase is related to more powerful electronics used for data processing in high-tech equipments used for satellites and defense/military purposes. Investigations in this subject have been already reported, bringing the attention to new approaches for current and future applications,1–3 Email:
[email protected] Received: 1 February 2018 Accepted: 12 April 2018 J. Nanofluids 2019, Vol. 8, No. 1
2169-432X/2019/8/001/011
where technologies developed for aerospace thermal control problems have been transferred to ground systems when needed. Hybrid solutions, using single- and twophase thermal control systems together have presented to be a promising approach to solve the high heat dissipation problem in ground systems. Whilst single-phase systems rely in a pump to drive the working fluid throughout the loop, and a correct hydraulic project ensures the adequate flow velocity in their several branches, two-phase systems apply heat pipe technology as a high performance thermal control device. Both single- and two-phase systems are known to dissipate high levels of heat, but some projects require much higher capacities that lead designers and researchers to seek for new technologies, especially when considering the working fluid responsible for transporting the heat. In this case, nanofluids have been considered and applied with promising results to solve future thermal management limitations,4 5 which has contributed to bring this technology to a level of maturity that enables it to be used commercially. Several investigations related to nanofluids applications have been conducted with important contributions to many areas.6 7 Considering current and future thermal management needs, the use of nanofluids is becoming inevitable. Nanofluids present to be an important approach to enhance the heat transfer capability of heat pipes and loop heat doi:10.1166/jon.2019.1563
1
ARTICLE
The paper deals with the development of a thermal management solution for a surveillance equipment, which needs to dissipate high levels of heat loads using both active and passive thermal control devices, that has been designed, simulated, built and tested in real operating conditions. The thermal management system was designed to use both a single-phase forced circulation loop and heat pipes using copper oxide (CuO)-water nanofluid, applied to promote the thermal management up to 50 kW of heat generated by several arrays of electronic components, being dissipated to the environment by a fan cooling system. The heat pipes collect the heat from electronic components that were far from the main single-phase forced circulation loop, rejecting the heat directly in its cold plates. Tests results of the thermal management system operating in real conditions show that with an addition of up to 20% by mass of CuO nanoparticles to the base fluid in the single-phase system, enhancements of up to 12% in the heat transfer coefficients were achieved but the increase in the pressure drop was around 32%. This shows that the use of nanofluid in the heat pipes resulted in a substantial decrease in the heat source temperature. When applying nanofluids in heat pipes, the maturity of this technology has reached Technology Readiness Level (TRL) of 8 for surveillance systems.
ARTICLE
Thermal Enhancement Using Nanofluids on High Heat Dissipation Electronic Components
pipes systems, which has already been proven.8 9 Other applications are related to the use of nanofluids in regular heat exchanger devices already installed in industries.10 Evaluation of nanofluids have been performed by many researchers in order to better understand the effects of nanoparticles in the transport properties, which are important for the prediction of the pumping requirements.11 12 Applications related to PCB thermal management using nanofluids have also been reported.13 However, important issues require attention, especially when considering the verification of a nanofluid regarding its own design, since many authors have reported different results for the same combination of base fluids and nanoparticles.14 Another issue that requires attention is the operational condition at which the nanofluid will be facing, being at atmospheric or saturation conditions. Differences in the thermal capacity of nanofluids have been reported when a given nanofluid had its thermophysical properties obtained in atmospheric conditions, and its behavior was completely different when operating in saturation conditions.14 Heat pipes and thermosyphons9 15–17 have been investigated with different nanofluids as working fluids, and the gains in their overall performances have been reported. Important improvements have been achieved when applying nanofluids in micro channels heat sinks,18 which can potentially be used in more reliable thermal management projects, especially those related to surveillance systems. However, investigations related to the effects of solid nanoparticles added to a given base fluid to form the nanofluid are required, especially to verify the direct impact of nanoparticles sizes and shapes, purity and chemical compatibility with thermophysical properties like thermal conductivity, viscosity, density, surface tension and, most importantly, wettability angle.19–22 Technological issues still require full attention from researchers and developers, as well as designers, in order to overcome resistances from applying this promising technology and guarantee that systems will operate according to their requirements. Another aspect that required attention is the technology development of new solutions that result in a maturity for application in the real world, being proved to operate in levels above laboratory conditions and beyond numerical simulations. In this case, systems that present TRL (Technology Readiness Level) equal or above 7 are desirable. Many researches have shown the potentiality in applying nanofluids in laboratory controlled environments, but their operation in the real world are needed to prove their full capability in the long term, which also should consider chemical compatibility and performance variation along time. Evaluating the technological baseline that are considered for designing a reliable and effective high heat dissipation thermal management systems that need to operate in hostile environments, with potential use of nanofluid, 2
Riehl
this article presents a solution applied to promote the dissipation of 50 kW of rejected heat. Considerations were given to a hybrid thermal management system, which uses both single- and two-phase thermal control devices operating with nanofluids to enhance the heat dissipation’s capabilities, in order to prove the nanofluid application for real systems. Real operation tests were carried out with the thermal management system designed for this purpose and its thermal analysis is presented. The information presented and discussed in this work is important for the evaluation of a real product operating with nanofluids as a new and promising technology.
2. NANOFLUID’S CHARACTERISTICS: TRANSPORT PROPERTIES CONSIDERATION The great influence by adding solid nanoparticles in a base fluid is enhancing its thermal conductivity. However, not only the thermal conductivity is influenced by this addition, but all other transport properties, which must be carefully considered during the design of the nanofluid and its application in the thermal management system.14 The base fluid’s transport properties are influenced by the addition of the solid nanoparticles, which also contribute to enhance its liquid density and viscosity. Considerations must be made regarding the addition of the nanoparticles as those properties directly influence the thermal management and pumping analysis. Some models have been developed to describe the influence of the addition of nanoparticles in pure substances and the gain in the liquid thermal conductivity that might represent23 as usually the Maxwell model is applied on this case as kn =
kp + 2kl + 2kp − kl f k kp + 2kl − kp − kl f l
(1)
Since the liquid thermal conductivity is affected by the addition of a nanoparticle in the base fluid, proper consideration and evaluation of the solid nanoparticles in a liquid must be taken according to the two-phase theory.24 Thus, the nanofluid density (n ) can then be calculated as 1 f 1−f = + (2) n p l The nanofluid dynamic viscosity (n ) can then calculated as25 1 n = l (3) 1 − f 25 The modification of the transport properties indicated by Eqs. (1) to (3) should be included in any analysis to correctly address their influence on the system’s thermal performance, even though they represent a simple approach to the problem when considering the multiple models presented in the literature.14 Thus, Eqs. (1) to (3) were implemented in a computer code26 used to predict the nanofluid J. Nanofluids, 8, 1–11, 2019
Riehl
Thermal Enhancement Using Nanofluids on High Heat Dissipation Electronic Components
influence on the thermal and hydraulic performance of the surveillance thermal management system, for both singleand two-phase thermal control devices. Such code has been constantly validated with different nanofluids’ models and the results obtained with Eqs. (1) to (3) have shown to be most reliable, when applying them for an overall thermal analysis. The single-phase device was designed as a liquid cooling system, circulating a working fluid to collect the heat generated by the electronic components and reject it to the environment using a fan cooling system. In this case, the electronic components (placed as arrays) were in direct contact with the cold plates connected to the single-phase cooling system. However, some electronic components were not able to be directly connected to the liquid cooling system’s cold plates, thus heat pipes were used as two-phase systems to collect the heat and transport it to the remote heat sinks.
3. THERMAL MANAGEMENT SYSTEM DESIGN
J. Nanofluids, 8, 1–11, 2019
3.1. Thermal Control Design Firstly, the overall thermal control design was performed to define the capacities for the single-phase pumping requirement, fan cooling system and bulk heat dissipation to the environment. This first task allowed the definition of the baselines for the project, and helped in the definition of the project’s requirements especially when the surveillance system would operate at its full capacity at the highest ambient temperature and humidity conditions. Then, as a mandatory task, the detailed thermal control design was done to evaluate the thermal behavior of each electronic component and its integrated operation in the PCB. Considering that electronic components have their sizes reduced substantially as technology evolves while keeping their heat dissipation, higher heat densities are found and must be carefully considered. Such level of information is extremely important to be shared between those responsible for designing the PCB and those responsible for the thermal control system, as smaller components will present very reduced areas to dissipate the heat, which can result in concentrated heat and high levels of temperature that can damage the electronic component. For example, if an electronic component dissipates 0.5 W and has a surface area of 2.5 mm by 2.5 mm, the total heat flux that needs to be dissipated from it will be 80 kW/m2 , which is substantially high. As a methodology for a detailed thermal control design, the PCB is simulated for conjugated heat transfer (when applied) to verify the heat and temperature concentrations. The parameters involved in this analysis are directly related to the project’s requirements and will vary from one to another. Figure 1 presents a mathematical simulation analysis made for this project, considering the PCB
Fig. 1.
Schematics of a PCB and its electronic components.
3
ARTICLE
The design of the thermal management system is a complex task that involves the consideration of several variables related to the conjugate heat transfer process (conduction, convection and radiation), along with materials used, working fluid, operation orientation, temperature, humidity, etc. The approach given for any thermal design should consider both the overall heat dissipation system analysis and the detailed thermal control design. The overall heat dissipation system analysis considers the total amount of heat to be dissipated and the available areas for the heat rejection, along with other overall aspects important for solving the problem. Usually, this is a first approach and it is used to verify the overall capacity of the thermal management system and will point the direction that should be followed to conceive a reliable and efficient system. The detailed thermal control design considers each and every component of the thermal management system operating separately (as a sub-system) and then together with all other parts (as a system) including all thermal couplings to evaluate, with high degree of details, how the thermal management system will behave during its operation and identify any potential issue. Both approaches have to deal with well-known thermal solutions in order to verify issues and limitations related to the heat dissipation process, therefore using already known systems like liquid pumped loops with dry coolers and heat pipes, which already present TRL = 9. However, upon applying a nanofluid in such systems, the real capabilities of both systems must be extensively verified not only for the thermal management process, but also for reliability over time, chemical stability during long term operation, NCG generation, among other parameters. The information presented in this work are important for the overall analysis of the thermal management system designed, built and tested for this application.
Additional information related to details of its hardware and final configuration is protected by a non-disclosure agreement.
Thermal Enhancement Using Nanofluids on High Heat Dissipation Electronic Components
Table I. Example of a list of components and their levels of heat fluxes. Dimension (mm) Component
ARTICLE
R15 R16 R17 R32 L1 L5 L12 L9
Q (W)
X
Y
010 010 014 014 115 115 115 220
165 165 135 135 140 140 140 140
325 325 220 220 220 220 220 220
Heat flux (W/m2 ) 1939394 1939394 4713805 4713805 37402597 37402597 37402597 71525974
architecture based on the electronic components positioning and heat dissipation levels, using as examples the values shown in Table I. The results from the mathematical simulation presented by Figure 2 were obtained using a computer code based on a nodal network26 for the PCB, which considers the conjugated heat transfer with the project’s parameters while operating in ambient conditions. Figure 2 shows that the PCB presents concentration of temperatures that must be properly addressed in order to avoid any component to shut down due to overheating, which is always referred to the components’ manufacturer cold junction temperature levels. Such information is also important for the design of the mechanical interface that will be used to transfer the heat generated to the dedicated heat sink. In many cases, a rugged design of the mechanical interface is not enough to guarantee the adequate heat transfer from the source to the sink, therefore new solutions must be applied. Once considering each PCB as a sub-system that integrates an entire system composed of several PCBs with their respective thermal control device, interface materials, thermal couplings etc., all interconnected with the main thermal control system (single-phase liquid cooling),
Fig. 2. Thermal simulation for the PCB operating in ambient conditions.
4
Riehl
the complexity of this design is evident. Therefore, proper consideration of all details must be done which relies in the designer’s experience. For this specific project, a surveillance system design dissipating high levels of heat has been conceived to operate in hostile environments where the ambient temperatures can range from +5 to +50 C and humidity levels up to 95%. In this case, a single-phase thermal control loop has been designed to use a nanofluid, presenting a forced circulation using a pump to drive the working fluid throughout the circuit to remove heat from the electronic components through dedicated cold plates, rejecting this heat to the environment by a fan cooling system. For this thermal management system, a hybrid design has been applied where the heat generated by some PCBs (located far from the cold plates) was removed by heat pipes configured as pulsating open loops, delivering the heat to the heat sinks allocated thorough the surveillance equipment (cold plates). The heat sinks were then connected to the single-phase thermal control loop that collected all the heat and dissipated it to the environment. The schematics of such arrangement is presented by Figure 3 and the surveillance equipment where the system was installed is shown by Figure 4, whilst Figure 5 presents the liquid cooling schematics. The heat pipe architecture is presented by Figure 6, where its configuration as an open loop pulsating heat pipe (OLPHP) is shown integrated to the mechanics for a given PCB. It is important to mention that each mechanics is mechanically and thermally connected to their PCB to ensure a fast heat removal and control of the electronic components’ temperatures, thus configuring a low thermal
Fig. 3. Arrangement schematics of the thermal control system. J. Nanofluids, 8, 1–11, 2019
Riehl
Fig. 4.
Thermal Enhancement Using Nanofluids on High Heat Dissipation Electronic Components
Schematics of the surveillance equipment.
inertia thermal control approach. Figures 7 and 8 present the OLPHPs unit 1 and unit 2, respectively, used for this analysis.
Fig. 5.
Liquid cooling setup schematics.
J. Nanofluids, 8, 1–11, 2019
OLPHP integrated with the PCB’s mechanics.
for both base fluid, solid nanoparticles and the heat pipe and liquid cooling system’s tubing, as the constant heat cycling along with high temperatures will accelerate the reaction and generate NCGs. Therefore, an analysis based on the Arrhenius Model can be performed to evaluate the amount of NCG that will be generated based on the interaction between the base fluid and the solid nanoparticles and, upon performing a so-called aging process, the NCG generation can be accelerated and flushed out of the system thus avoiding the negative influence of it during the system’s lifetime.27 Several solid nanoparticles have been tested during the last years in order to verify their thermal behavior when applied to single- and two-phase systems, like liquid cooling and heat pipes, respectively. The selection range from commercial available and custom made solid nanoparticle, which present high purity and quality in guaranteeing the particles’ sizes and distribution. The nanofluid used in this work were obtained using the 2-step method14 to result in a reliable nanofluid. It is important to mention that commercial nanoparticles usually present an acceptable quality related to their purity, geometry and size’s distribution. As those parameters increase in reliability, the costs for those nanoparticles will increase significantly. Such parameters are important to be considered as several publications have shown dispersed results from many authors, related to the nanofluid thermophysical properties, even though the same nanoparticle and base fluid, in the same concentration, have been reported to be used.14 If discrepancies in results related to the same solid nanoparticle, base fluid and concentration are obtained, one should consider that other parameters are important to guarantee that the same results will be obtained, such as particle shape, its distribution among a sample, etc. Such information is considerably important to be take into account and should be the object of future investigations in order to obtain a model with a statistical approach to design a nanofluid.14 28 5
ARTICLE
3.2. Nanofluid Selection and Application As one of the most important component of the thermal management system, the nanofluid selection required careful consideration regarding the selected solid nanoparticles and the base fluid. For the sake of chemical compatibility for both single- and two-phase thermal control systems (liquid cooling and heat pipes, respectively), the most appropriated selection was the use of copper nanoparticles and water as the base fluid. Purity levels for both substances had to be as high as possible to avoid the interaction of other substances with the nanofluid, which could potentially lead to the generation of non-condensable gases (NCGs) that can negatively impact the thermal operation of the system, especially the heat pipes. The assessment to this issue is based on the chemical reaction analysis
Fig. 6.
ARTICLE
Thermal Enhancement Using Nanofluids on High Heat Dissipation Electronic Components
Riehl
Fig. 7. OLPHP unit 1 under analysis.
The augmentation on the thermal conductivity of a given working fluid will directly be dependent on the solid nanoparticle size and geometry, as well as the amount of particles with those characteristics. For better results,
the solid nanoparticles must present their size and geometry in at least 95% of the sample to ensure that the nanoparticles have a homogeneous characteristics. This is an important information to be considered, otherwise the
Fig. 8. OLPHP unit 2 under analysis.
6
J. Nanofluids, 8, 1–11, 2019
Riehl
Thermal Enhancement Using Nanofluids on High Heat Dissipation Electronic Components
nanofluid with the same size of nanoparticles and base fluid will present different results, due to the fact that it is not homogeneous. This can represent a very wide range of results and can consequently impact the thermal management system performance. Two solid nanoparticles were used for this analysis, which are identified as follows: 1. High purity CuO: purity > 98% and particle sizes (spherical) dp = 29 nm; 2. High purity Cu: purity > 99.98% and particle sizes (spherical) dp = 2 nm.
Table II. Wettability characteristics for various working fluids and materials. Base fluid Water
Acetone
Methanol
Nanoparticle Wick material Copper Nickel Polyethylene
Al2 O3 CuO NiO Al2 O3 CuO NiO Al2 O∗∗ CuO NiO 3 ∗ ∗ ∗
Good Poor Poor Good Poor Poor Good Poor Poor Good Poor Poor Poor Poor Poor
Poor Poor Poor
Good Poor Poor Good Poor Poor
Notes: ∗ Al2 O3 —Water are incompatible substances—NCGs are generated with long term operation during power cycling. ∗∗ Al2 O3 —Methanol have presented chemical incompatibilities and unstable solubility with long term operation.
J. Nanofluids, 8, 1–11, 2019
7
ARTICLE
Sample 1 was obtained using CuO-water (darker nanofluid) and Sample 2 was obtained using pure Cu and water as the base fluid. Each samples clearly present different characteristics that directly impact in their final thermal performance. Guaranteeing that the sample has a high purity and homogeneous size distribution will certainty cause a direct impact in the final performance. Both nanofluids were prepared without surfactants using the 2-step method, but to reach a homogeneous nanofluid without precipitation of the solid nanoparticles, Sample 1 needed 2 hours in an ultrasonic bath whilst Sample 2 required only 15 min. This difference was attributed to the fact that Sample 2 used the pure Cu with homogeneous particle sizes and distribution, which contributes to a high quality nanofluid. The homogeneous distribution represents a high impact in the nanofluid’s performance while operating in the thermal management system, and will cause an important impact in other transport properties as well, such as the surface tension and wettability angle. This last one is a very important variable for a high efficient heat transfer system, since a high wettability angle will ensure that the nanofluid will be wetting the heat transfer surface more efficiently for values close to 180. Table II presents a laboratory investigation performed for heat pipes application with various wick structures, housing materials and working fluids, where the observed wettability and chemical compatibility characteristics were compiled. The indication of “Good” wettability means that the working fluid was able to spread on the surface and be absorbed by the
wick material (typical of contact angles between 90 and 180 24); the indication “Poor” wettability means that the working fluid remained as a bubble without spreading on the surface and without being absorbed by the wick material (typical of contact angles lower than 90 24 ). One of the most investigated nanofluid is composed with water as base fluid and Al2 O3 solid nanoparticles. However, when operating in thermal cycles for a long period of time, this combination will generate hydrogen as a NCG that will cause a negative impact in the thermal management system in the long term. Several reports have described the NCG generation when water and Al2 O3 were used, especially in heat pipes applications, which resulted in severe recommendations to avoid this combination for thermal control devices used.3 8 Also, the use of surfactants to improve the solubility of the solid nanoparticles in the base fluid shall not be used as it will also generate NCGs during the thermal management system’s operation in the long term. The project requirements define the maximum operational temperature for the liquid cooling system, as well as the highest temperature that the electronic components can operate, due to their according cold junction limits, which was imposed to the passive system. The design of the thermal management system resulted in the following structural setup for both liquid cooling and passive systems, as they showed to be the most optimized characteristics for this application. Thus, the most important operational requirements imposed by this application are as follows • Liquid cooling system: Top = 55 C maximum Operating with de-ionized water and CuO (Sample 1) selected as the nanofluid combination at atmospheric condition Over 400 m of lines with ID = 11 mm No use of surfactants due to chemical issues • Passive system: Top = 90 C maximum—due to cold junction temperature limitation for the electronic components Connects the PCBs that are far from the cold plates Operating with de-ionized water and CuO (Sample 1) selected as the nanofluid combination at saturation condition Total of 240 pulsating heat pipes divided in 4 arrays Pulsating heat pipes characteristics: Unit 1: 3.0 mm OD; 1.6 mm ID; Leff = 190 mm Unit 2: 3.0 mm OD; 1.6 mm ID; Leff = 225 mm No use of surfactants due to chemical issues. Given the complexity of the presented thermal management system, the use of a nanofluid represents a significant gain in its thermal performance. Considering all important variables for this application based on in-house research and development, evaluating also the project’s requirements, de-ionized water has been selected as the base fluid. The performance verification was done using CuO (Sample 1) for the liquid cooling system as well as for the heat
Thermal Enhancement Using Nanofluids on High Heat Dissipation Electronic Components
Riehl
pipes. The nanoparticles concentration (f ) shall vary from 3.5% to 20% (by mass of the base fluid) for the liquid cooling system and up to 3.5% for the heat pipes, in order to verify their effect on the overall thermal performance of the system. Tests were limited to Sample 1 due to the high costs involved when using Sample 2 in both systems.
ARTICLE
4. RESULTS AND DISCUSSION The presented results were well correlated to the thermal tests applied to this equipment, and further analysis will be disclosed in future reports. For the sake of presenting the most important results obtained so far, the following data were selected among several hours of operation. However, due to the sensitive technology that used this thermal management systems solution, additional information related to the electronic arrays and their architecture cannot be disclosed. The following results are related to the liquid cooling system. Figure 9 presents the results for the liquid cooling pressure drop as a function of the operational temperature; Figure 10 presents the results for the heat transfer coefficients as a function of the operational temperature; Figure 11 presents the results for the volumetric flow rate as a function of the operational temperature; Figure 12 presents the results for the pumping power as a function of the system’s operation temperature. Figures 9 and 10 present results for the pressure drop and heat transfer coefficients observed during the operation of the liquid cooling system, respectively, on a comparison between the use of pure water and the addition of CuO nanoparticles at different concentrations by mass percentage of the base fluid in the system. The results are related to each individual electronic module (composed of 3 PCBs), which dissipate a maximum of 50 W of heat, thus, based on the module’s footprint and heat dissipation, the calculation for the heat transfer coefficient was
Fig. 9. Liquid cooling system—pressure drop as a function of operational temperature.
8
Fig. 10. Liquid cooling system—heat transfer coefficient as a function of operational temperature.
performed. The pressure drop was measured using a differential pressure transducer (Sensotec A5/882–22A5–D, accuracy of 0.25%, reading scale of 0–172.25 kPa), which measure the inlet and outlet pressure in the pump. The heat transfer coefficient was calculated using the Newton’s Law of Cooling Q (4) h= ATi − To with maximum uncertainties of ±10%. As shown by Figure 11, less volumetric flow rate is necessary to promote the heat dissipation when using the nanofluid, as the heat transfer coefficient increases. It is clear that as the nanoparticle concentration increases, the pressure drop also increases up to 32% for f = 20% as more solid nanoparticles are present in the system (Fig. 12). The pump needs to overcome the extra resistance as the transport properties are changed with the addition of the nanoparticles, which is also highly dependent on the system’s operation temperature. However, the increase on the
Fig. 11. Liquid cooling system—volumetric flow rate as a function of operational temperature. J. Nanofluids, 8, 1–11, 2019
Riehl
Thermal Enhancement Using Nanofluids on High Heat Dissipation Electronic Components
Fig. 12. Liquid cooling system—required pumping power as a function of operational temperature.
Fig. 13. Unit 1 OLPHP—maximum heat source temperature as a function of the applied heat load to the evaporator. J. Nanofluids, 8, 1–11, 2019
G=
Q Te − Tc
(5)
with maximum uncertainties of ±8.8%. The results show that higher values were obtained for the heat pipe operating with the nanofluid, which is highly dependent on the augmentation of the working fluid’s thermal conductivity by adding the solid nanoparticles. Even better performances were observed for the Unit 2 heat pipe, even though this device presented a longer effective length, which impacts in the final heat source temperature as shown in Figure 15. Figure 16 shows the
Fig. 14. Unit 1 OLPHP—thermal conductances as a function of the applied heat load to the evaporator.
9
ARTICLE
heat transfer coefficients is also clear and can represent a gain around 12% for the same f = 20% which cannot be neglected. The deviation between the results below and the simulated results using the computer code26 were less than 5%, thus showing reliability in the thermal management design conceived for this application. Regarding solid nanoparticles deposition in the single-phase cooling system’s liquid reservoir, this showed to be a minor issue to be consider only for f = 20%. However, during the long term operation of the single-phase liquid cooling system, the deviation on the results were negligible, showing that the solid nanoparticles were constantly mixed and incorporated with the base fluid. Upon individually analyzing the heat pipes used to transport the heat from the electronic components to the heat sink connected to the liquid cooling system, the direct impact in their performance caused by the use of a nanofluid is very clear. Figure 13 presents the operational
results for the Unit 1 heat pipe while transferring the heat from the electronic components (their respective PCB) and dissipating in the heat sink. The heat loads at which the heat pipes were operating were obtained from the heat dissipation of their PCB upon measuring their electronic operation. For this measurement, the software responsible for controlling the PCB was responsible for reading and acquiring this information, which was a feature promoted by its manufacturer. Figure 13 presents the results for the Unit 1 heat pipe heat source temperatures according to the heat load given by the PCB. At a higher heat load, the operation with nanofluid presented its reliability as the heat source temperature was lower than the operation with pure water. Since the electronic components used in military applications are very sensitive, it should operate far from their cold junction temperature limitation, which is related to their qualification purpose and set by the manufacturer. For this case, the highest cold junction temperature allowed was 90 C, thus while operating at a temperature below 70 C (for the heat pipe using nanofluid), the electronic components could reach their highest performance. Figure 14 shows the thermal conductances observed during the operation, which are calculated as
Thermal Enhancement Using Nanofluids on High Heat Dissipation Electronic Components
Riehl
equilibrium temperature of the heat source was substantially decreased (83 C), which shows the potentially in using the nanofluid in this application. At the highest heat load, the observed temperature of the heat source while the Unit 2 heat pipe was operating with nanofluid was 35 C lower than the operation with de-ionized water, which is was significant impact in the passive system’s operation. The beneficial impact while using the nanofluid in the heat pipe can also be observed in the results related to the thermal conductances (Figs. 14 and 16), with a increase of up to 55% when compared to the operation with de-ionized water.
5. CONCLUSIONS
ARTICLE
Fig. 15. Unit 2 OLPHP—maximum heat source temperature as a function of the applied heat load to the evaporator.
results for the thermal conductances during Unit 2 heat pipe operation. Since the working fluid needs to travel farther to reject the heat acquired in the heat source, due to a greater effective length for the Unit 2 heat pipe, higher pressure drops during the flow will result in higher equilibrium temperatures, which directly impact in the heat source temperature. Thus, the temperatures observed were higher than those verified for Unit 1 heat pipe. Since the cold junction temperature limit was 90 C for the electronic components, operating the heat pipe with de-ionized water as working fluid showed to be a real concern as the final equilibrium temperature was higher than the limit (118 C), which certainly caused the components in the PCB to switch off due to the temperature protection. Operating the heat pipes in this condition is really harmful for the PCBs as well as for the mission requirements imposed by the surveillance equipment. Upon operating with the nanofluid, the final
Fig. 16. Unit 2 OLPHP—thermal conductances as a function of the applied heat load to the evaporator.
10
In general, the main conclusions that can be derived from this investigation are: 1. Higher heat transfer coefficients can be reached with the increase of the solid nanoparticles concentration, representing an enhancement of up to 12% for f = 20% at 55 C when compared with the operation with water; 2. The pressure drop also increases as the concentration of nanoparticles increases, which could compromise the pump operation; 3. Lower volumetric flow rates are observed for higher concentration of nanoparticles, as this factor contributes to increase the working fluid’s viscosity and density; 4. Even with the use of solid nanoparticles, the required pumping power does not represent to be the major issue on this specific project, as this requirement can be easily addressed as the calculated values are rather low; 5. Pulsating heat pipes can keep the heat source temperatures below 90 C; 6. The overall analysis indicates that the application of the nanofluid with higher concentrations can be used, as the major parameter for this analysis is the heat transfer coefficient, which is reducing the size of the thermal management system applied to control the temperature of the electronics components. When considering that the thermal management system is operating at higher capacities, while keeping the working fluid’s temperature within the project’s parameters, the use of a nanofluid presents to be an important innovative approach for this project. This is directly resulting in more gains than loses for the overall thermal system analysis and should remain as the most indicated solution for this application. Another important conclusion that can be taken from this investigation it that while operating the surveillance equipment in real conditions, the thermal management system (single-phase liquid cooling and two-phase heat pipes) has achieved a maturity related to TRL = 7. Since other systems have been already equipped with heat pipes operating with nanofluids and have already being used as final products, this technology is mature enough to be related as TRL = 8. J. Nanofluids, 8, 1–11, 2019
Riehl
Thermal Enhancement Using Nanofluids on High Heat Dissipation Electronic Components
NOMENCLATURE AND ACRONYMS A dp f ID G h kn
Heat Transfer Area [m ] Particle Diameter [nm] Nanoparticle Mass Fraction Inner Diameter [m] Thermal Conductance [W/ C] Heat Transfer Coefficient [W/m2 C] Effective Thermal Conductivity of a Homogeneous Nanofluid [W/m · K] Particle Thermal Conductivity [W/m · K] Base Fluid Thermal Conductivity [W/m · K] Heat Pipe Effective Length [m] Non-Condensable Gases Outer Diameter [m] Open Loop Pulsating Heat Pipe Printed Circuit Board Applied Heat Load to the Evaporator [W] Average Evaporator Temperature [ C] Average Condenser Temperature [ C] Fan cooling outlet temperature [ C] Fan cooling inlet temperature [ C] Operation Temperature [ C] Technology Readiness Level Nanofluid Density [kg/m3 ] Nanoparticle Density [kg/m3] Base Fluid Density [kg/m3 ] Nanofluid Dynamic Viscosity [Pa · s] Base Fluid Dynamic Viscosity [Pa · s].
References and Notes 1. R. R. Riehl and L. Cachuté, Thermal control of surveillance systems using heat pipes and pulsating heat pipes, Proceedings of 49th AIAA/ASME/SAE/ASEE Joint Propulsion Conference and Exhibit 11th International Energy Conversion Engineering Conference, San Jose, USA, July (2013). 2. R. R. Riehl and L. Cachuté, Thermal management of surveillance equipments electronic components using pulsating heat pipes, Proceedings of IEEE-ITherm Conference—Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronics Systems, Orlando, USA, May (2014). 3. R. R. Riehl, Passive thermal management systems using pulsating heat pipes, Proceedings of 9th Minsk International Seminar Heat Pipes, Heat Pumps, Refrigerators, Power Sources, Minsk, Belarus, September (2015).
J. Nanofluids, 8, 1–11, 2019
4. Z. H. Liu and Y. L. Li, Int. J. Heat M. Transf. 55, 6786 (2012). 5. S. M. S. Murshed and C. A. N. de Castro, Green Solvents I: Proper Appl. Chem. 1, 397 (2012). 6. E. Ebrahimnia-Bajestan, H. Niazmand, W. Dungthondsuk, and S. Wongwises, Int. J. Heat M. Transf. 54, 4376 (2011). 7. A. Ghadimi, R. Saidur, and H. S. C. Metselaar, Int. J. Heat M. Transf. 54, 4051 (2011). 8. R. R. Riehl and N. Santos, Appl. Therm. Engineer. 42, 6 (2012). 9. K. Alizad, K. Vafai, and M. Shafahi, Int. J. Heat M. Transf. 55, 140 (2012). 10. K. Y. Leong, R. Saidur, T. M. I. Mahlia, and Y. H. Yau, Int. J. Heat M. Transf. 55, 808 (2012). 11. S. M. S. Murshed, C. A. Nieto de Castro, M. J. V. Lourenço, M. L. M. Lopes, and F. J. V. Santos, Ren. Sust. Ener. Rev. 15, 2342 (2011). 12. S. M. S. Murshed and P. Estellé, Ren. Sust. Ener. Rev. 76, 1134 (2017). 13. L. Colla, L. Fedele, S. Mancin, S. Bobbo, D. Ercole, and O. Manca, Nano-PCMs for electronics cooling applications, Proceedings of the 5th International Conference on Micro/Nanoscale Heat and Mass Transfer, Biopolis, Singapore, January (2016). 14. E. Marcelino and R. R. Riehl, A review on thermal performance of cuo-water nanofluids applied to heat pipes and their characteristics, Proceedings of the 15th IEEE Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITherm), Las Vegas, USA, May–June (2016). 15. A. B. Solomon, K. Ramachandran, L. G. Airvatham, and B. C. Pillai, Int. J. Heat M. Transf. 75, 523 (2014). 16. T. Grab, U. Gross, U. Franzke, and M. H. Buschmann, Int. J. Therm. Science 86, 352 (2014). 17. Z. H. Liu and Y. Y. Li, Int. J. Heat M. Transf. 55, 6786 (2012). 18. L. Snoussi, N. Ouerfelli, K. V. Sharma, V. Vrinceanu, A. J. Chamkha, and A. Guizani, Phys. Chem. Liq. 56, 281 (2018). 19. K. Khanafer and K. Vafai, Int. J. Heat M. Transf. 54, 4410 (2011). 20. V. Mikkola, S. Puupponen, H. Grahbohm, K. Saari, T. Ala-Nissila, and A. Seppälä, Int. J. Therm. Science 124, 187 (2018). 21. D. Cabaleiro, L. Colla, S. Barison, L. Lugo, L. Fedele, and S. Bobbo, Nanosc Resear. Lett. 12, 53 (2017). 22. J. P. Meyer, S. A. Adio, M. Sharifpur, and P. N. Nwosu, H. Transf. Engineer. 384 (2015). 23. J. Koo and C. Kleinstreuer, J. Nanopar. Resear. 6, 577 (2004). 24. V. P. Carey, Liquid–Vapor Phase Change Phenomena: An Introduction to the Thermophysics of Vaporization and Condensation Processes in Heat Transfer Equipment, 1st edn., Taylor and Francis, USA (1992). 25. Y. Xuan and W. Roetzel, Int. J. Heat M. Transf. 43, 3701 (2000). 26. R. R. Riehl, Design and operation of capillary driven two-phase loops as HP/CPL/LHP, Brazilian Patent 03/0030, August (2002). 27. R. R. Riehl, Heat Pipe Scie. Techn. Int. J. 8, 69 (2017). 28. H. D. Koca, S. Doganay, A. Turgut, I. H. Tavman, R. Saidur, and I. M. Mahbubul, Ren. Sust. Ener. Rev. 82, 1664 (2018).
11
ARTICLE
kp kl Leff NCG OD OLPHP PCB Q Te Tc To Ti Top TRL n p l n l
2
