Heat Generation Density Research Articles

Development of Printed Circuit Boards with High Thermal Conductivity and Low Thermal Expansion by Applying Cu-Mo Composite Materials

In recent years, electronic devices are strongly required to be smaller, lighter, and more powerful. As a result, mounted components have become smaller, more powerful, and denser, and the heat generation density on printed circuit boards (PCBs) has increased. Since rising component temperatures can lead to component failure, efficient heat removal from the generated heat has become an urgent issue. Since general PCBs have low thermal conductivity, when high heat dissipation is required, methods to increase the thermal conductivity of PCBs have been employed by increasing the copper (Cu) content in the PCB or by placing an aluminum (Al) alloy in the inner layer. However, while Cu and Al have high thermal conductivity, they also have high coefficients of thermal expansion (CTE) and young’s modulus, and the CTE of a PCB with increased thermal conductivity using these materials is high. If the thermal expansion difference between the PCB and mounted components is large, reliability against heat cycles cannot be obtained. Therefore, we have focused on copper-molybdenum (CuMo) composite materials with high thermal conductivity, high elastic modulus, and low thermal expansion. We have been utilizing Cu-Mo for the development of our PCBs and fabricated multilayered CuMo composite PCBs; therefore, this report will focus on the results regarding thermal characterization of prototype PCBs.

Heat transfer characteristics of solid-liquid interface on nanostructure surface under external electric field

With the size of high-performance electronic device decreasing (down to nanoscale), and the accompanying heat dissipation becomes a big problem due to its extremely high heat generation density. To tackle the ever-demanding heat dissipation requirement, intensive work has been done to develop techniques for chip-level cooling. Among the techniques reported in the literature, liquid cooling appears to be a good candidate for cooling high-performance electronic devices. However, when the device size is reduced to the sub-micro or nanometer level, the thermal resistance on the solid-liquid interface cannot be ignored in the heat transfer process. Usually, the interfacial thermal transport can be enhanced by using nanostructures on the solid surface because of the confinement effect of the fluid molecules filling up the nano-grooves and the increase of the solid-liquid interfacial contact area. However, in the case of weak interfacial couplings, the fluid molecules cannot enter into the nano-grooves and the interfacial thermal transport is suppressed. In the present work, the heat transfer system between two parallel metal plates filled with deionized water is investigated by molecular dynamics simulation. Electronic charges are applied to the upper plate and lower plate to create a uniform electric field that is perpendicular to the surface, and three types of nanostructures with varying size are arranged on the lower plate. It is found that the wetting state at the solid-liquid interface can change from Cassie state into Wenzel state with strength of the electric field increasing. Owing to the transition from the dewetting state to wetting state (from Wenzel to Cassie wetting state), the Kapitza length can be degraded and the solid-liquid interfacial heat transfer can be enhanced. The mechanism of the enhancing hart transfer is discussed based on the calculation of the number density distribution of the water molecules between the two plates. When the charge is further increased, electrofreezing appears, and a solid hydrogen bonding network is formed in the system, resulting in the thermal conductivity increasing to 1.2 W/(m·K) while the thermal conductivity remains almost constant when the electric charge continues to increase.

Nanobrick Wall Multilayer Thin Films with High Dielectric Breakdown Strength.

Current thermally conductive and electrically insulating insulation systems are struggling to meet the needs of modern electronics due to increasing heat generation and power densities. Little research has focused on creating insulation systems that excel at both dissipating heat and withstanding high voltages (i.e., have both high thermal conductivity and a high breakdown strength). Herein, a polyelectrolyte-based multilayer nanocomposite is demonstrated to be a thermally conductive high-voltage insulation. Through inclusion of both boehmite and vermiculite clay, the breakdown strength of the nanocomposite was increased by ≈115%. It was also found that this unique nanocomposite has an increase in its breakdown strength, modulus, and hydrophobicity when exposed to elevated temperatures. This readily scalable insulation exhibits a remarkable combination of breakdown strength (250 kV/mm) and thermal conductivity (0.16 W m-1 K-1) for a polyelectrolyte-based nanocomposite. This dual clay insulation is a step toward meeting the needs of the next generation of high-performance insulation systems.

Experimental Study on Microchannel with Addition of Microinserts Aiming Heat Transfer Performance Improvement

Microchannel technology rapidly established itself as a practicable solution to the problem of the removal of extremely concentrated heat generation in present-day cooling fields. By implementing a better design structure, altering the working fluids and flow conditions, using various materials for fabrication, etc., it is possible to increase the heat transfer performance of microchannels. Two parameters that affect how well a microchannel transfers heat were only recently coupled, and the complicated coupling of the parameter that affects how well a microchannel sink transfers heat is still not well understood. The newest industrial developments, such as micro-electro-mechanical systems, high performance computing systems, high heat density generating future devices, such as 5G/6G devices, fuel cell power plants, etc., all present thermal challenges that require the use of microchannel technology. In this paper, single-phase flow in microchannels of various sizes, with or without microinserts, is described in terms of its thermal-fluid flow properties, including fluid flow characteristics and heat transfer characteristics considering the compound effects of variations of channel size and addition of microinserts. The trials were carried out using distilled water that had thermo-physical characteristics that varied with temperature. A microchannel with microinserts was developed for managing the high heat generation density equipment. The fluid flow and heat transfer characteristics are explored and analyzed for Reynolds numbers ranges from 125 to 4992, for 1 mm channel size, and from 250 to 9985, for 2 mm channel size. The cooling performance criteria are pressure drop characteristics, heat transfer characteristics, and overall performance, whereas the testing parameters were chosen for the variations in channel size and the addition of microinserts. The influence of inserting microinserts on microchannels is discussed. Results suggest that by inserting microinserts, the performance of the heat transfer of microchannels is significantly improved and, also, fluid flow resistance is increased. The criteria of the thermal performance factor are employed to assess the overall performance of the microchannel. Significant intensification of heat transfer is observed with indication that the addition of microinserts to microchannels and reduction in channel sizes exhibited improved overall performance.

An electrochemical-thermal coupling model for heat generation analysis of prismatic lithium battery

In this study, an electrochemical thermal coupling model of prismatic lithium battery was established to simulate the actual discharge process. The heat generation distribution and variation in porous electrodes and current collectors were analyzed and the relationship among the total heat generation, discharge rate and discharge time was proposed. The simulation result indicated that the heat generation distribution in the porous electrodes was non-uniform. The moving distance of the peak heat generation density became shorter with the increase of discharge rate. The heat generation density increased with the discharge time. In the porous area, the negative electrode was dominant, and the proportion of polarization heat was the highest. The heat generation of battery was non-uniformity distributed on the current collectors, which increased with the discharge rate and decreased with the progress of discharge. The heat generation expression proposed in this paper could accurately predict the relationship between the total heat generation and the discharge time under each discharge rate, which could provide theoretical support for the establishment of a battery thermal management system.

Model order reduction of three-dimensional transient heat transfer analysis to one-dimensional by using Krylov-subspace concerning cooling effect of air

In semiconductors packages, heat generation density is increasing due to miniaturization. Since heat generation is caused by voltage input at the packages, electronic components are usually cooled by air. In recent years, electrification is widely spread among many engineering products like motor vehicles, and the amount of semiconductors packages installed in those products as well. To make the electronic components more densely installed, co-simulation of heat transfer analysis and electrical circuits is needed at designing products. In this paper, by using the Krylov-subspace model order reduction method, the accuracy of one-dimensional heat transfer analysis model concerning cooling effect of air is evaluated. By using the method, co-simulated with electrical circuits is possible.

Effect of supply velocity and heat generation density on cooling and ventilation effectiveness in room with impinging jet ventilation system

This study aims to understand the effect of supply momentum and heat generation density on the cooling and ventilation effectiveness in a room with an impinging jet ventilation system (IJV). A parametric study was conducted by CFD analysis, and the number of supply terminals and occupants were varied as parameters. CFD validation was conducted before the parametric study by comparing the experimental and analytical results. RNG k-ε and SST k-ω model showed almost the same accuracy for simulating, and SST k-ω model was chosen to be used for the parametric study. It is shown that the larger number of terminals and/or occupants leads to the air distribution to be displacement flow, whereas the smaller number leads to that of mixed condition. The ventilation effectiveness and cooling effectiveness within the room could be kept higher by locating the supply terminal at the centre of the walls than at the corner of the room. When the supply velocity was smaller than 0.833 m/s, the draught rate (DR) at ankle level in the central cross-section could keep lower than: 20% in the region farther than 0.5 m horizontally away from the terminals, and 15% in the region farther than 1.0 m away from the terminals. Finally, the cooling and ventilation effectiveness is expressed as the function of Archimedes number (the balance between supply momentum and buoyancy) in the specific studied cases. The asymptotic values for both large and small Archimedes numbers are obtained for each index expressing cooling and ventilation effectiveness.

Temperature leveling of electronic chips by solid-solid phase change materials compared to solid-liquid phase change materials

The complexity and the miniaturization of portable electronic devices has resulted in an increase in the heat generation density, which is a major issue. Passive heat sinks using phase change materials (PCMs) have gained widespread attention as thermal management devices to improve the reliability of the portable devices. Paraffin is a solid–liquid PCM, which exhibits relatively large volumetric latent heat and is typically used in passive heat sinks. However, it faces certain limitations such as low thermal conductivity and liquid leakage due to melting. Vanadium dioxide (VO2), which is a solid-solid PCM, has a high thermal conductivity and large volumetric latent heat. Therefore, VO2 can potentially be used to overcome the issues of the conventional PCMs such as paraffin. This study aims to determine the feasibility of a passive heat sink using VO2 by evaluating the temperature leveling performance when VO2 is attached to a heat source. The same evaluation is performed for the paraffin for comparison. From the experimental results, it is observed that VO2 exhibits a higher temperature leveling performance than paraffin. Furthermore, the effect of each thermophysical property of the PCM on the temperature leveling performance is evaluated through numerical simulations. The simulation results highlight the advantages of using VO2 as the PCM for a passive heat sink.

Local Heat Transfer Control using Liquid Dielectrophoresis at Graphene/Water Interfaces

Graphene-based materials are considered for the solution of the thermal management problem of current and next generation micro/nano-electronics with high heat generation densities. However, the hydrophobic nature of few-layer graphene makes passing heat to a fluid very challenging. We introduced an active and local manipulation of heat transfer between graphene and water using an applied, non-uniform electric field. When water undergoes electric field induced orientation polarization and liquid dielectrophoresis, a substantial increase in heat transfer develops due to a decrease in interfacial thermal resistance and increase in thermal conductivity. By using two locally embedded pin and plate electrodes of different sizes, we demonstrated a two-dimensional heat transfer control between two parallel few-layer graphene slabs. We obtained local heat transfer increase up to nine times at pin electrode region with an ultra-low Kapitza resistance through the studied non-uniform electric field strength range creating highly-ordered compressed water in the experimentally measured density limits. With this technique, heat can be (i) distributed from a smaller location to a larger section and/or (ii) collected to a smaller section from a larger region. Current results are important for hot spot cooling and/or heat focusing applications.

Development of High Heat Dissipation and Low Thermal Expansion Printed Circuit Boards

Along with higher performance and smaller size of electronic devices, the use of high-power components and the higher density of component mounting are being promoted. When the heat generation density increases and the temperature of the mounted component rises, the component malfunctions and becomes unstable. Therefore, it is necessary to efficiently dissipate the heat generated in the component. Further, in order to secure the mounting reliability of the component, it is necessary to enhance the heat dissipation from the component and at the same time reduce the difference in thermal expansion between the component and the substrate. Therefore, we are developing a new printed circuit boards that has both high heat dissipation and low thermal expansion.

Improved melting of latent heat storage via porous medium and uniform Joule heat generation

To enhance the rate of heat transfer in phase change materials (PCM), high conductivity porous materials have been widely used recently as a promising method. This study introduces a novel approach for improving melting of PCM by incorporating uniform Joule heat generation with the porous structure compared to central heat generation. Different cases based on the heater-in foam configuration under the same heat generation rate are numerically verified and compared with the case of using the central heating element, which the heat transfer in the domain enhances by the porous medium. The effects of pore density and rate of heat generation are explored using the thermal non-equilibrium model to better deal with the interstitial heat transfer between the internal heat-generated-in-foam and the PCM. For the case with the central heating element, the effects of heater dimensions as well as the rate of heat generation are also investigated. The results show that the uniform heat generation from the porous structure can substantially reduce the melting time. Applying 100 kW/m3 for the rate of heat generation reduces the melting time by 21% compared with the best case of the localised heater. Meanwhile, applying higher pore-density foam does not bring any significant effect due to the uniform distribution of the heat generation. The results also show a small effect of localized heater size on the melting time with the same rate of heat generation density from the porous structure. However, for an identical volumetric heat source power of the localised heater, the rate of heat generation per volume is more effective compared with the heating element size due to the presence of the porous medium.

Modal control of thermoplasmonic behavior of nanostructures based on excitation of perfect absorption resonances

Controlling the localized heat generation density and temperature profile of nanostructures exploiting perfect absorption of individual resonance modes is reported. The methodology is applied to spherically symmetric nanostructures using the T-matrix method. It is demonstrated that perfect modal splitting of the absorption power at desired wavelengths and individual excitation of the modes provide the ability to localize the generated heat at desired locations, and control the resulting temperature profile in multilayer core–shell structures. By knowing the thermal behavior of individual modes, it is shown that excitation of the perfect absorption modes at desired temperatures can result in compensation for the temperature-rise drop, induced in high-temperature thermoplasmonics due to thermal shift of the resonance frequencies. Much higher temperature rises can be achieved through properly designed thermal mode-coupling schemes. The proposed methodology is very promising for the control of the thermoplasmonic behavior of nanostructures, and the design of much more thermally efficient structures, taking into account the thermally dependent parameters.

Enhancing opto-thermal performances of the reflective phosphor-converted laser diode by stacking a sapphire substrate for double-sided phosphor cooling

Phosphor-converted laser diode (pc-LD) has been a very competitive candidate in high-luminance white illumination. For the reflective pc-LD, the phosphor layer is usually bonded onto a reflective substrate, which plays the role of light reflection and heat dissipation. However, the heat generation of phosphor cannot be efficiently dissipated by only cooling the reflective surface, in which case most of heat may be confined in the phosphor layer with low thermal conductivity. In this work, a double-sided phosphor cooling scheme was presented to improve the opto-thermal performances of the reflective pc-LD by stacking a highly transparent and heat-conducting sapphire substrate onto the phosphor layer. By using an opto-thermal phosphor model, it was found that the maximum heat generation density and hotspot were both confined at the incident surface. After stacking a sapphire substrate onto the incident surface, the maximum heat was extracted and conducted from the sapphire substrate to the reflective substrate, leading to a significant decrease of the maximum phosphor temperature from 221 °C to 86 °C under excitation power of 0.8 W. In this case, a higher excitation power was allowed, contributing to a higher attainable output power and luminance. The double-sided phosphor cooling scheme was also experimentally achieved. It was found that the presented scheme exhibited higher optical power, luminance and color stability than the traditional one under higher excitation power.

Two-dimensional axisymmetric opto-thermal phosphor modeling based on fluorescent radiative transfer equation

Remote phosphor plate excited by the light-emitting diode (LED) or laser diode (LD) has been widely applied in solid-state lighting. The light propagation properties within the phosphor have been analytically characterized based on the one-dimensional assumption. However, this may be not feasible in some cases, e.g., the excitation spot diameter is comparable to the plate thickness or the incident irradiance is not uniform along the horizontal direction. In this work, we extend the one-dimensional model to a two-dimensional axisymmetric opto-thermal model based on the fluorescent radiative transfer equations (FRTEs) for the cylindrical phosphor plate excited by a collimated beam with a small spot diameter and Gaussian irradiance distribution. The radiances of the blue and yellow lights are iteratively calculated by solving FRTEs in the cylindrical coordinate system using the discontinuous spectral element method. By inputting the obtained heat generation density from FRTEs, we calculate the three-dimensional temperature distribution by solving the heat diffusion equation (HDE). In addition, the opto-thermal interaction between FRTEs and HDE is achieved by introducing the temperature-dependent phosphor quantum efficiency. Using the model, we evaluate the optical, thermal and opto-thermal interacted performances of the phosphor. Finally, the model is validated by comparing the measured output optical power and surface temperature distribution with the calculated results.

Study on Thermal Network Model for Metal Plate Shunt Resistors

In recent years, miniaturization of electronic devices and high-density mounting of electronic components have advanced, and the heat generation density on the printed circuit board (PCB) has increased. Under such circumstances, it is necessary to perform thermal design using the thermal analysis actively to control temperature of the parts properly. Therefore, high accuracy is required for the thermal analysis, and an analysis model that is capable of accurate temperature estimation is needed. In this study, we investigated the thermal network model of metal plate shunt resistors for current detection, and proposed a model with low analysis cost and high analysis accuracy.

Thermal Management in a Half-MEA Microscale Fuel Cell Architecture

With the advancement of portable technology, suitable power sources must be developed. In comparison to standard battery technology, microscale fuel cells demonstrate higher energy densities and the potential for nearly instantaneous recharging [1]. However, fuel cells are commonly less efficient than batteries, and thusly, produce more waste heat. Waste heat is generated in methanol polymer electrolyte membrane fuel cells via the oxygen reduction reaction, the methanol oxidation reaction, and ohmic heating in the membrane. The oxygen reduction reaction accounts for the vast majority of the total heat generated [2]. Consequently, heat generation and removal in conventional fuel cell architectures has been carefully investigated in order to achieve effective thermal management [3]. Here we present a novel microscale fuel cell design that utilizes a half-membrane electrode assembly (MEA). In this design, a single fuel/electrolyte stream provides an additional pathway for heat removal that is not present in traditional fuel cell architectures. There are three streams in this fuel cell design: methanol fuel, oxygen, and nitrogen. Previously a simplified model neglecting the heat transfer through the anode side and nitrogen stream has been investigated [4]. The simplified model only has one heat generation source in the cathode catalyst. The ANSYS Fluent model presented here investigates heat removal of inlet fuel temperatures of 22°C, 40°C, 50°C, 60°C, and 70°C. This new model simulates the complete fuel cell including the anode side and nitrogen stream. There are three heat generation sources in the new model: oxygen reduction in the cathode catalyst, methanol oxidation in the anode catalyst, and ohmic heating in the membrane. Continuity, momentum, and energy equations were solved simultaneously by ANSYS Fluent to obtain the results. Heat generation densities are determined experimentally for all inlet fuel temperatures. The simulations predict thermal profiles throughout this microscale fuel cell design. The exit temperature of the fuel stream, oxygen stream, and nitrogen stream were obtained to determine the rate of heat removal. Simulation results show that the fuel stream dominates heat removal at room temperature. As inlet fuel temperature increases, the majority of heat removal occurs via convection with the ambient air by the exposed current collector surfaces. The top and bottom current collector removes nearly the same amount of heat. The model also shows that heat transfer through the oxygen channel and nitrogen channel is minimal over the range of inlet fuel temperatures. Ultimately, these simulations can be used to determine design points for best performance and durability in a single-channel microscale fuel cell. [1] M.A. Goulet, E. Kjeang, Journal of Power Sources, 260 (2014) 186-196. [2] A.Z. Weber, J. Newman, Journal of the Electrochemical Society, 153 (2006) A2205-A2214. [3] H. Ju, H. Meng, C-Y. Wang, International Journal of Heat and Mass Transfer, 48 (2005) 1303-1315. [4] L. Sun, A.S. Hollinger, ASME 2017 15th International Conference on Fuel Cell Science, Engineering and Technology, V001T03A001-V001T03A001. Figure 1

Heat Transfer Characteristics during Boiling of Immiscible Liquids Flowing in Narrow Rectangular Heated Channels

The use of immiscible liquids for cooling of surfaces with high heat generation density is proposed based on the experimental verification of its superior cooling characteristics in fundamental systems of pool boiling and flow boiling in a tube. For the purpose of practical applications, however, heat transfer characteristics due to flow boiling in narrow rectangular channels with different small gap sizes need to be investigated. The immiscible liquids employed here are FC72 and water, and the gap size is varied as 2, 1 and 0.5mm between parallel rectangular plates of 30mm×175mm, where one plate is heated. To evaluate the effect of gap size, the heat transfer characteristics are compared at the same inlet velocity. The generation of large flattened bubbles in a narrow gap results in two opposite trends of the heat transfer enhancement due to thin liquid film evaporation and of the deterioration due to the extension of dry patch in the liquid film. The situation is the same as that observed for pure liquids. The latter negative effect is emphasized for extremely small gap sizes if the flow rate ratio of more-volatile liquid to the total is not reduced. The addition of small flow rate of less-volatile liquid can increase the critical heat flux of pure more-volatile liquid, while the surface temperature increases at the same time and assume the values between those for more-volatile and less-volatile liquids. By the selection of small flow rate ratio of more-volatile liquid, the surface temperature of pure less-volatile liquid can be decreased without reducing high critical heat flux inherent in the less-volatile liquid employed. The trend of heat transfer characteristics for flow boiling of immiscible mixtures in narrow channels is more sensitive to the composition compared to the flow boiling in a round tube.

Analytical asymptotic solutions for heat transfer during transient compression filling of spherical and cylindrical pressure vessels in the limit of small Reynolds number

Taking the limit of small Reynolds number for a vessel being filled with compressed gas, the energy equation was found to asymptote to the unsteady heat conduction equation with heat generation and variable density. This equation was solved analytically for cylindrical and spherical geometry. Assuming the density changes linearly with time, a solution is obtained which is identical in form to the constant density solution if the Fourier number is defined using the log-mean density rather than the instantaneous density. At steady state conditions, the Nusselt number based on the diameter for cylinders with aspect ratios larger than 1 rapidly approaches an integer solution of NuD = 8. For cylinders with aspect ratios less than 1, the Nusselt number based on the cylinder length (height) characteristic dimension rapidly approaches NuL = 6. It is shown experimentally and numerically that during compression filling, the heat transfer asymptotically approaches this analytical solution at low Reynolds numbers.

유연기판 위에 제작된 Silver Nanowire 필름의 기계 및 전기적 신뢰성 연구

본 논문에서는 AgNW 필름의 기계적 전기적 신뢰성을 연구하였다. 특히 전류가 흐르는 상태에서 굽힘 변형이 발생하였을 때의 AgNW 필름의 내구성 및 신뢰성을 연구하였다. AgNW 필름의 전압 및 전류 시험을 수행하여, AgNW에서 발생하는 발열과 전류 밀도의 변화를 관찰함으로써 AgNW의 전기적 내구성을 평가하였다. AgNW는 곡률반경 2 mm까지 굽힐 수 있었으며, 200,000회의 굽힘 반복시험에도 높은 신뢰성 및 유연성을 보여주었다. 또한 over-coating 막은 AgNW 필름의 내구성을 향상시키는 효과가 있음을 확인하였다. Over-coating이 없는 AgNW 필름의 경우, AgNW 표면에서 국부적인 발열을 보인 반면에 over-coating이 된 AgNW 필름의 경우 균일하게 발열되어 over-coating막이 AgNW 필름의 내구성을 향상시킴을 알 수 있었다. 전류를 인가한 상태에서의 굽힘 시험을 수행한 결과 굽힘 반복시험에서는 전류밀도가 지속적으로 감소하여 시험 후, 52.4%의 전류밀도 감소를 보였다. 전류가 인가된 상태에서 AgNW의 굽힘 변형이 지속되면 AgNW들의 인장, 굽힘 및 sliding 등의 기계적인 변형에 의하여 AgNW network 구조의 변형이 발생하거나, 혹은 개별 AgNW의 접촉 접합부들이 떨어지면서 접촉저항이 증가하여 주울 열에 의하여 파괴가 발생한다. 또한 over-coating막의 적용은 AgNW 필름의 전기적 신뢰성을 향상시킬 수 있음을 알 수 있었다. In this paper, we investigated the mechanical and electrical reliability of silver nanowire (AgNW) films. In particular, the durability and reliability of AgNW films were studied when the AgNW film was subjected to the bending deformation under current flow. The electrical durability of AgNW was evaluated by observing changes in heat generation and current density occurring in AgNW through voltage and current tests. The AgNW film showed a constant resistance change up to a bending radius of 2 mm and 200,000 cycles in the bending fatigue tests. The over-coating layer has an effect of improving the durability of the AgNW film. In the case of AgNW with the over-coating layer, heat was uniformly dissipated on the surface of AgNW film, whereas in the case of AgNW film without the over-coating layer, heat was generated locally. In the bending test under the current flow, the current density of the AgNW film was continuously decreased up to 52.4%. During bending, the AgNW was deformed due to mechanical deformation such as tensile, bending and sliding of the AgNW, consequently contact resistance of the AgNW was increased, leading to a electrical breakdown of AgNW by Joule heating. It was found that the application of the over-coating layer can improve the electrical and mechanical reliability of the AgNW film.

Heat in Computers: Applied Heat Transfer in Information Technology

Since the advent of modern electronics technology, heat transfer science and engineering has served in the development of computer technology. The computer as an object of heat transfer research has a unique aspect; it undergoes morphological transitions and diversifications in step with the progress of microelectronics technology. Evolution of computer's hardware manifests itself in increasing packing density of electronic circuits, modularization of circuit assemblies, and increasing hierarchical levels of system internal structures. These features are produced by the confluence of various factors; the primary factors are the pursuit of ever higher processing performance, less spatial occupancy, and higher energy utilization efficiency. The cost constraint on manufacturing also plays a crucial role in the evolution of computer's hardware. Besides, the drive to make computers ubiquitous parts of our society generates diverse computational devices. Concomitant developments in heat generation density and heat transfer paths pose fresh challenges to thermal management. In an introductory part of the paper, I recollect our experiences in the mainframe computers of the 1980s, where the system's morphological transition allowed the adoption of water cooling. Then, generic interpretations of the hardware evolution are attempted, which include recapturing the past experience. Projection of the evolutionary trend points to shrinking space for coolant flow, the process commonly in progress in all classes of computers today. The demand for compact packaging will rise to an extreme level in supercomputers, and present the need to refocus our research on microchannel cooling. Increasing complexity of coolant flow paths in small equipment poses a challenge to a user of computational fluid dynamics (CFD) simulation code. In highly integrated circuits the paths of electric current and heat become coupled, and coupled paths make the electrical/thermal codesign an extremely challenging task. These issues are illustrated using the examples of a consumer product, a printed circuit board (PCB), and a many-core processor chip.