Lectures

Americans hate to wait in line. In 1973, OPEC declared an oil embargo against the United States causing the price of oil to quadruple and creating the longest gas station lines in American history. This crisis solidified the American public’s commitment to energy independence and spawned the formation of the Department of Energy, the Gas Research Institute, and a comprehensive tax policy. Over the next 30 years, the U.S. invested a little over a billion dollars in unconventional gas production technology, such as directional drilling and massive hydraulic fracturing. As a result of these investments, the U.S. has become the largest gas and second largest oil producing country in the world, added $1.2 trillion to the gross domestic product (GDP), and created more than 9.3 million jobs. The stunning success of U.S. investments in unconventional oil and gas production research and development provides a more generic roadmap for what it takes for technology innovation. Often overlooked or undervalued is the element of “crisis”—a key ingredient for technology innovation that creates a sense of urgency, a well-defined mission, and secures the public’s financial commitment to research and development. While most scientists agree the crisis of our day is global climate change, the slow 0.14-0.20°C/decade creep in global temperatures has been insufficient to create the collective hysteria necessary for the public to demand action. As scientists and engineers, how do we continue to work on the most important problems facing society in the absence of public consensus and what are the most promising research directions related to thermal and fluids engineering to address the crisis of our day? A few examples of game-changing concepts and major technological advances in thermal engineering will be presented.

Frosting on the heat exchanger surface of the heat pump was unfavorable phenomena causing deterioration of heating capacity and increase of the fan power due to the increased pressure drop of the heat pump system, while the system was operating as heating mode under 0℃. Prediction and verification of transient frost characteristics and its effect on heating capacity and pressure drop of the heat exchanger were required to suggest the improved way in order to lengthen operating time of the heat pump system by the frost delay. The frosting issue required understanding of multi-phase flow as well as heat and mass transfer.  
Frost characteristics including frost thickness, frost surface temperature, frost density, and frost mass on the heat exchanger of the heat pump were numerically predicted by applying frost modelling, and then experimentally verified under various operating conditions of dynamic and geometric parameters. They were applied for evaluating local and total heat transfer rates, blockage ratio, and pressure drop of the heat exchanger under the different frosting condition.
Local frost characteristics on the heat exchanger surface were always non-uniform until the heat exchanger was almost blocked. Pressure drop of the heat exchanger were predicted by using the maximum frost thickness and curve between pressure drop and volume flow rate. Frosting on the heat exchanger surface of the heat pump using refrigerant was grown faster than that using ethylene glycol. It made heating capacity of the heat pump using refrigerant be decreased faster than that using ethylene glycol under frosting condition.

The world's renewable fresh water supply may be considered to be the difference between precipitation over land and evaporation; and this supply is essentially fixed while world population rises and the water demands of growing economies rise. As a result, water scarcity is an increasing problem throughout the world, with water shortages impacting both agricultural and urban populations and with substantial environmental damage as a result of water diversions for human use. In addition, agricultural and industrial pollution of waterways further impedes fresh water supplies.

Desalination of seawater, brackish ground water, and even wastewater, has become a major tool in meeting the world’s growing water demand, but energy intensity remains high and more efficient systems are needed. The tools of heat and mass transfer, and of thermal system systems engineering are directly applicable. This lecture discusses design analogies between desalination systems and thermal power systems, with a focus on reverse osmosis systems. Entropy generation minimization is shown to be the major aim in energy efficient design. Analogies between heat exchangers and osmotic mass exchangers are developed. The potential use of ultrapermeable membranes to reduce system size or energy consumption is considered. The impact of concentration polarization, analogous to incondensable gases in heat exchangers and dependent upon mass transfer coefficients, is shown to be controlling is system performance. Finally, the role of varying mass flow rates in membrane feed channels is examined using analogy to heat transfer and a superposition technique, showing how mass transfer coefficients can be accurately modeled in these systems.

Studies on combustion limit have been conducted from the middle of nineteenth century to prevent undesired accidents in coal mines. The first theoretical description on the mechanism of combustion limit was presented in 1940’s among other numerous experimental approaches. Nevertheless, another fifty years was required until novel microgravity experimental approach in 1990’s eventually proved the fundamental limit mechanism of deflagration waves is due to the radiative heat loss from high temperature zone including burned gas.
Apart from the limit of deflagration wave, the existence of “flame ball” was first predicted by Zel’dovich in 1940’s and it was proved through series of microgravity experiments by Ronney and collaborators in drop tower experiments in U.S. from 80’s, JAMIC (Japanese) drop shaft experiments in 90’s, and Space Shuttle experiments in late 90’s to 2000’s. However, limit mechanism and interactions of those two kinds of flame regimes have never been investigated in the same platform to date.
Our final goal is to construct comprehensive combustion limit theory which covers both the limits of conventional deflagration wave and flame ball. Before proceeding to the space experiments in the International Space Station, preliminary airplane-based microgravity experiments with low-speed counterflow flame technique have been conducted to date. This lecture presents the apparent transitions from counterflow deflagration wave to flame ball-like phenomena near the combustion limit in the counterflow field. A hypothesis on the comprehensive combustion limit theory based both on microgravity experiments and three dimensional computations with diffusive-thermal model will be introduced.

Numerical Simulation of Solid Oxide Fuel Cell Electrodes

Solid oxide fuel cell (SOFC) has been attracting large attentions because of its high efficiency and high fuel flexibility.  Since SOFC operates at high temperature, great efforts have been made by many research groups to overcome the durability issues.  It is widely recognized that the electrode microstructures have significant impacts on cell performance as well as cell durability.   For the anode, porous nickel-yttria stabilized zirconia (Ni-YSZ) is the most commonly used material in SOFCs. It is known that coarsening of Ni by sintering is one of the major degradation mechanisms during the long time operation of SOFC.  Thus, controlling the sintering process is of key importance for achieving high performance as well as long term durability.  In the present study, phase field and lattice Boltzmann methods are applied to numerically predict the degradation phenomena in the SOFC anode.   Temporal evolutions of the three dimensional microstructures are validated by the FIB-SEM reconstruction.

The heat generated by semiconductor devices and electronic components is a big problem for a variety of products and systems including smartphones, electric vehicles, servers, and satellites.  “Extreme” is a unifying theme, from nanometer features and 10+ kW chips to severe materials heterogeneity.  This presentation will summarize these challenges and our progress on research topics including electron and phonon transport in transistors and novel electronic materials, nanostructured packaging materials, and microfluidic two-phase heat sinks.  Thermal conduction at extreme nanoscale dimensions is discussed in the context of compact logic and memory devices for low-power chips.  For these devices, fundamental challenges include electron-phonon interactions at metal-semiconductor and interfaces and the impact of atomic-scale disorder.  Progress on convective boiling at extreme heat fluxes is described in the context of high power radar chips, and includes diamond microfluidic heat sinks and 3D separation and routing strategies for the liquid and vapor phases.  This presentation will also highlight two decades of collaborations with the semiconductor industry and silicon valley startups.

Modern gas turbine airfoils are subject to gas temperatures in excess of 3000⁰ F that are well above the material limits for reliable operation. The airfoils have to be therefore actively cooled to prevent engine failure.  Cooler air from the compressor is circulated through serpentine channels with turbulators (internal cooling) and is discharged as a film through inclined holes drilled on the airfoil surface (film cooling). While there is a significant body of experimental and computational literature on both internal and film cooling problem, most of the effort has been focused on improving geometric designs to enhance cooling effectiveness and in developing correlations or the predictive capability for use in turbine design. A fundamental understanding of the flow structures and their relationship with the surface heat transfer is largely missing. In this paper, attention is focused on the flow physics of internal and film cooling flows, and the current understanding of this flow physics is reviewed.  Large Eddy Simulation (LES) results are analyzed to understand the origin and development of flow structures. The spectral characteristics of both the flow and thermal fields will be studied in order to identify the frequencies associated with these structures and to examine which of these play an important role on the temperature distributions and heat transfer coefficients near the surface.

Rearch on thermal hydraulics and severe accident are important for the safety of nuclear power plants. Thermal-hydraulic safety research aims at achieving a reliable cooling of reactor core at normal and accident conditions to prevent the melting of nuclear fuels containing most radioactive materials. Severe accident research aims at managing core melt accident scenarios to minimize the release of radioactive materials to the environment. Along with the expanding nuclear power program, nuclear safety R&D has been active in Korea during the last two decades. This lecture introduces the current issues, on-going programs and major achievements in nuclear safety R&D in Korea, focusing on the activities of Korea Atomic Energy Research Institute, which is a key organization in nuclear safety research. Topics to be covered include: (a) fundamental experiment and modeling for boiling and two-phase flow in the reactor, (b) separate effect and integrate effect tests for advanced safety systems, (c) investigation of severe accident phenomena in reactor and containment, (d) development and application of safety analysis codes, and finally (e) development of advanced thermal-hydraulic systems for safety enhancement.

Computational Fluid Dynamics (CFD) faces many numerical and physical challenges when applied on IC-Engines. These start with moving boundaries and meshes, then continue with the modelling of the flow including turbulence, heat transfer, spray, combustion and pollutant formation. Despite clear challenges and complexities in every subject, CFD has to provide accurate and reliable results in a cost effective way. This leads to compromises in different fields from the choice of numerical schemes, spray and wall film models, to the reaction mechanisms and other phenomenological models. One of important tasks is the modelling of turbulence and answering if the standard Reynolds-Averaged Navier-Stokes method fulfils the latest quality criteria for CFD IC-Engines or if it is necessary to use Large-Eddy Simulations or at least some of the novel hybrid RANS/LES models. As it would be impossible to provide all these answers here, only latest developments of a few selected topics will be presented. The Partially-Averaged Navier-Stokes (PANS) approach, which represents variable resolution turbulence models, will be explained in detail and its performance will be assessed on engine flows. The advanced wall heat-transfer model that has a minimum dependency on the wall normal distance will also be presented. Furthermore, a novel combustion method based on the Flame-Tracking-Particle (FTP) approach, which implies continuous tracing of the mean flame surface and application of the laminar/turbulent flame velocity concepts, will be shown. Some aspects of the FTP method like knocking predictions will be also discussed. Additionally, the progress in the spray modelling that is based on a new model of drop parcel that allows minimization of grid dependence of spray penetration and vaporization will be reported. This will all be analyzed having in mind the impact of the mesh generation process on the modelling itself and on simulation results e.g. computational unstructured meshes, the wall cell layers and the adaptive mesh refinements. As a reference point, the present state of the art engine calculations and measurements will be used.

This paper is attempted to clarify the causes and physical insight of the thermal-lag phenomena by presenting analytical solutions. The analytical solutions presented are able to answer critical questions about the thermal-lag engine which have not been answered since it was proposed roughly thirty years ago. Authors find the criteria for onset of the thermal-lag oscillation in the engine.  Tailer’s postulation that a thermal lag can be created to a certain extent by the imperfect heat transfer is confirmed. It is also found that the thermal lag angles can be altered by artificially adjusting the geometrical and physical parameters. Finally, the influence of thermal lag on the indicated work output of the engine can be evaluated. In the present study, a nonlinear instability analysis is performed and theoretical solutions are carried out by perturbation method.  In parallel, a prototype engine is developed and tested to partially verify the present theory. Experiments are conducted by measuring the displacement of piston under various heating temperatures and design parameters. Close agreement between the experimental data and the theoretical predictions, particularly in the steady oscillation regime, has been observed.

Continuous attempts have been made to raise the turbine inlet temperature of a gas turbine engine since the performance of the engine is proportionally increasing with this temperature. Major parts in gas turbine engine such as combustor, nozzle and blade are exposed to extremely high temperature which exceeds melting temperature of its material. Therefore, cooling is essential element for designing of high performance gas turbine engine and various advanced cooling methods have been developed to protect gas turbine parts from high temperature environment. However, as cooling technology is applied, large temperature gradient is presented in hot parts and it could induce thermal stress, which is major cause of failure in gas turbine engine. Moreover, the strength of material is deteriorated in high temperature condition so that the turbine parts are more vulnerable to the thermal stress in operation. Hence, designing method for hot components plays a key role in developing gas turbine engines to guarantee its performance and service lifetime. In this lecture, the advancement of cooling technology and thermal design process for hot components is introduced. In terms of cooling technology advancement, details of cooling methods will be covered from single cooling element to combination of each cooling elements referring experimental study in the laboratory.

In engineering problems where radiation is the dominant mode of heat transfer, such as high-temperature combustion and material processing, fire and atmospheric radiation, renewable solar energy, space exploration, microwave and laser applications, accurate and complete solutions of the governing equation of radiative transfer (ERT) are desired. The integro-differential nature of the ERT makes analytical solution difficult, and thus numerical methods, such as the finite volume method (FVM) and discrete-ordinates method (DOM), are preferred.  Numerical methods have garnered increasing attention in the field of radiation heat transfer, as they provide cost-effective alternatives to costly experimentation and their efficiency and accuracy has been improved with the advance of computational technology.
Ray effect and false scattering are two well-known numerical errors associated with DOM. However, they also exist in all discretization-based numerical methods including the FVM. The prevailing notion in the field is that these two errors are “separate” from one another, although they may act in a compensatory manner. Ray effect and numerical smearing exhibit dependence on both spatial and angular discretization.  Proportionality expressions for various orders of numerical smearing errors were derived.
The false scattering error originated from spatial discretization in the dimension domain reflects numerical smearing or diffusion, similar to the artificial diffusion in CFD. It is not directly related to radiation scattering both mathematically and physically. The directional discretization in the solid-angle domain generates the real-meaning false scattering effect. Thus, there are three types of numerical errors, namely, ray effect, numerical smearing, and angular false scattering. Using the DOM, individual and combined impacts of these three numerical errors were examined for various spatial and angular discretizations and medium optical properties. 

Over the past several decades, air transportation is expanding remarkably, as the world economy becomes increasingly globalized. The growth of the aviation industry is expected to continue into the coming decades. During this period of air transport expansion, environmental concerns will also attract public attention. After scientists gave proof of global warming, gaseous emissions as well as noise emission problems have been cited with respect to aircrafts. It is worth noting that reductions in gaseous emissions are directly related with fuel consumption and thus operation costs. Environmental issues and airline operators require gas turbine manufacturers to produce environmentally friendly gas-turbine engines with lower emissions and higher specific fuel consumption (SFC) ratings. The requirements can be met when heat exchangers are incorporated into gas turbines. In this presentation, a variety of heat exchangers equipped in the heat management system of existing aero-engines is introduced, and advanced thermo-dynamic aero-engine cycles that use the heat exchanger technology for recuperation and/or intercooling are addressed. In addition, the principle and detailed methodology for the design of high efficiency and ultra-light heat exchanger are discussed.

In this keynote lecture the multiscale simulation of heat/mass transfer and fluid flow problems in thermal an environmental engineering are presented in detail. Following sections are involved.
In the first section the meaning of multiscale problems is briefly introduced. The multiscale heat/mass transfer and fluid flow problems are divided into two categories: multiscale process and multiscale system. For a multiscale system all processes at different scales are governed by the same governing equations, while for a multiscale process different governing equations should be used for different scales. Then the numerical approaches for different scale problems are classified: the first one based on the continuum assumption, i.e.,  macroscopic method, the second one is so-called meso-scale method, including direct simulation of MC (DSMC), and lattice Boltzmann method. The third approach is the molecular dynamics simulation (MDS). A typical multiscale process is the transport of mass/heat transfer and fluid in a PEMFC. In the PEMFC the fuel gas flows in polar plate channels occurs at the length scale of millimeters or centimeter, while the chemical reaction at the porous catalyst layer and the transport of proton in the memory occur at micrometer even nanometer scale level.
In the second section the numerical methods for the multiscale problems are described. For the multiscale processes two kinds of method are widely used, i.e., solving the entire process by a general method and solving separately and coupling at the interface. The key issue in the “solving separately and coupling at the interface” is the information transfer techniques at the interface. Transfer information at the interface from mirco-results to macro-results is relatively easy, which can be done by a compressor operator, usually being different statistical averaging methods, while transfer information from macro results to micro results is quite difficult, where a re-constructor operator should be developed which can transform less to much. Such re-constructors developed in the authors group are introduced
In the third part, several simulation example of multiscale problems are presented in detail. These include: (1) Flow around/through a porous media; (2) Transport processes in a mini PEMFC; (3) Scale effect on flow slip and temperature jump in microchannel; (4) Hybrid method for fluid flow around carbon nanotube ;(5) Impingement of droplet onto a liquid film ;(6) VOC (Volatile organic compounds) emission process in chamber;(7) Multiscale transport processes during shale gas extraction

In all convective heat transfer situations losses occur in the flow field (by dissipation) as well as in the temperature field (by conduction). Typically these losses are more or less quantified by the friction factor f with respect to losses in the flow field and the Nusselt number Nu for the heat transfer quality. Assessing the process of convective heat transfer as a whole then becomes problematic because two different non-dimensional quantities, f and Nu, have to be combined somehow.  From a thermodynamics point of view a fundamentally different approach is appropriate, however. All losses (here: in the flow and in the temperature field) become manifest in corresponding entropy generation rates. They can and should be determined in order to quantify losses uniquely and consistently, assess thermal engineering processes and suggest optimization strategies with respect to convective heat transfer processes. Within this approach, called second law analysis (SLA), an energy devaluation number is introduced, see [1]. It basically determines how much of the so-called entropic potential of the energy involved in a process is used within it. The entropic potential, introduced in [2] is defined with the overall entropy which by a certain amount of energy is discharged to the ambient on the way from being primary energy (pure exergy, i.e. available work) to being part of the internal energy of the ambient (pure anergy, i.e. no more available work). Examples will illustrate the advantage of this kind of analysis (SLA).
[1] H. Herwig (2011): The role of entropy generation in momentum and heat transfer, Journal of Heat Transfer, Vol. 134, 031003-1-11

The mathematical modelling and numerical simulation of thermal processing of advanced and emerging materials are complicated because of multiple transport mechanisms and complex phenomena that commonly arise. The materials are often not easily characterized and typically involve large property changes over the ranges of interest. The boundary conditions are often not properly defined and may be unknown, leading to an inverse problem. The configuration and interactions between different components are also often quite complicated. However, it is necessary to obtain accurate, realistic and dependable analytical and numerical results in order to predict, design, and optimize the thermal processes of current and future interest. The mathematical and numerical models employed must be validated and the accuracy of the results established if the simulation is to form the basis for improving existing systems and developing new ones. This paper focuses on the main challenges that are encountered in obtaining accurate numerical simulation results in practical thermal processes and systems in the area of manufacturing and materials processing. Of particular interest are concerns like verification and validation, imposition of appropriate boundary conditions, and modelling of complex, multimode transport phenomena at multiple length and time scales. The coupling of multiscale simulations is critical to an overall accurate model. Additional effects such as viscous dissipation, surface tension, buoyancy and rarefaction that could arise and complicate the modeling are outlined. Uncertainties that arise in typical processes are considered since these are important in design and optimization. Several different processes and systems are considered. The methods that may be used to meet the challenges in an accurate simulation are discussed, along with typical results for a few important processes. Future needs in this important area are also outlined.

The performance degradation of air-source heat pumps cannot be avoided when they operate at both very low and high ambient temperatures. The refrigerant injection technique has rapidly developed in recent years due to its outstanding performance at low ambient temperatures. This study measured the heating performance of air-source heat pumps in which novel vapor injection techniques of a combined flash tank and sub-cooler (FTSC) cycle and a double expansion sub-cooler (DESC) cycle were applied. Two-stage air-source heat pumps, which had a rated heating capacity of 5.5 kW using R-410A, were tested according to the injection ratio and compressor frequency. The performance of these cycles was compared with that of a flash tank (FT) and a sub-cooler (SC) cycle. In addition, the performance of an air-source heat pump using liquid injection technique was measured and compared with that of the non-injection heat pump.

Organized Structures of Single and Two Phase Turbulent Flows with Scalar Transport and Some Applications to Solar Engineering Problems

The organized structures of the single and two-phase turbulent flows have been explored, and some findings are being applied to find a solution to the solar engineering problems derived from organized convection of solid / gas two phase flows. Super parallel computing enabled to simulate spatially advancing round jets, curved channel turbulent flows, turbulent flows with obstacles and solid / gas two phase flows. The perfect set of the time-dependent data revealed the turbulent flow structures regarding energy and scalar transport. The research work is being extended to be applied to solar engineering problems. The organized motion of particle flows is studied to develop a new type of a particle receiver for solar concentration.

Enhanced Figure of Merit of a Self-assembled Micro-porous Bismuth Telluride Thin Film

Thermoelectric conversion has been an active research area due to its potential as a solid-state energy conversion technology to directly convert thermal energy to electricity.  The performance of thermoelectric materials is determined by the dimensionless figure of merit ZT, defined by thermal conductivity, electrical conductivity and Seebeck coefficient.  However it is difficult to realize the materials with both low thermal conductivity and high electrical conductivity in bulk semiconductors.  In this decade, it has been increasingly recognized that thermal conductivity in nanostructured materials is not an intrinsic parameter, unlike for bulk materials.  When the characteristic length of micro-structures is shorter than the mean free path of the carriers, the transport of the carriers becomes ballistic.  If the pore spacing of the porous materials is shorter than the mean free path of the phonons and is longer than the mean free path of electrons, the lattice thermal conductivity can be decreased while simultaneously maintaining high electrical conductivity.  It has been found that the reduction of thermal conductivity by the introduction of a microporous structure is an effective approach for increasing thermoelectric performance. Here, we made scalable nano-porous thermoelectric materials using self-assembled nano-porous materials such as an anodizing alumina, and a block copolymer (BCP).  The porous Bismuth Telluride thin films were made by vacuum deposition process on the porous materials, and their measured non-dimensional figure of merits were higher than that of bulk materials due to extremely low thermal conductivity.  We made the in-plane thermoelectric generator with porous Bismuth Telluride, and measured the output power.  The thermoelectric generator performance of porous Bismuth Telluride was 1.5 times higher than that of flat thermoelectric thin films.

From heat dissipation in electronic devices to the heating of our homes, thermal transport is ubiquitous in today’s technology. In the past few decades, a drive to reduce device dimensions, in some cases to length scales smaller than the intrinsic mean free paths of the energy carriers, while concurrently increasing efficiency, has pushed thermal management to a primary consideration in the design and implementation of technological devices.  Of particular importance in thermal management is the ability to measure, predict, and understand the conduction of energy across an interface, and furthermore, the ability to tune the conduction at the interface. Depending on the application, a high interfacial conductance may be ideal, allowing energy carriers to flow unimpeded, such as on a computer chip where large temperature gradients can cause hot spots and cause device failure.  In other cases, a high interfacial resistance may be ideal to confine heat within a particular region, such as in heat pipes. This talk will present an overview on our recent work at the University of Virginia Nanoscale Energy Transport Laboratory to combine experimental and computational methods to model, measure, and predict the thermal transport across a variety of solid-vapor, crystal-crystal, and crystal-crystal interfaces to serve as a guide for future thermal management techniques in a broad range of applications.

The powerful measurement methods, i.e., particle image velocimetry (PIV) and laser induced fluorescence (LIF), have been widely spread to micro/nanofluidics research in combination with two-wavelength fluorescence, evanescent wave illumination, and so forth. However, the fluorescence-based techniques have an inherent disadvantage of employing extrinsic fluorescent labels which can potentially alter the sample properties and flow fields due to the electrochemical and toxic influence. Non-intrusive measurement techniques for thermofluid dynamics have been proposed by sophisticated integration of Raman scattering, multi-wavelength detector for different vibration modes of molecules, total internal reflection and nonlinear optical process. Two-wavelength spontaneous Raman imaging was applied to obtain the spatiotemporal distribution of temperature in microfluidic device. Total internal reflection Raman imaging was first proposed to quantitatively visualize the concentration distribution of molecules in the vicinity of interface. Ideal non-intrusive measurements of molecule concentration in micro gas chip and micro chemical chip were achieved by coherent anti-Stokes Raman scattering (CARS) microscopy.

About 2.8% of all electricity consumed today goes for datacenters, not counting telecom communication centers, innumerable cell phone towers, etc. and the rate of annual increase is about 15% per year. Thus, a significant effort is underway to develop much more energy efficient micro-two-phase cooling systems to replace the air-cooled systems that consume about 40% of the electricity in a datacenter. Such cooling systems at the server level will normally include a multi-microchannel evaporator cold plate on each CPU, a compact condenser and a driver (pump, compressor or gravity). In this context, while the field of micro-fluidics has greatly emerged over the past several decades, the tandem topic of “hyper” micro-fluidics is now emerging for high speed two-phase flows, that is those in the approximate flow range from 100 > Re > 5000 as opposed to Re < 1-5 for micro-fluidics. Indeed, instead of being interested in directing the movement of 1 or 2 bubbles, as many as 1000 bubbles per second flow through each microchannel as it transitions from one two-phase flow pattern to the next. To develop mechanistically based models for these high speed two-phase flows in the microchannels, extensive use has been made of high speed digital videos and image processing to characterize and investigate the various flow patterns, transitions between flow patterns and important phenomena. The result is the emergence of a “smart” flow pattern map approach with embedded mechanistic based models for each flow regime, which is getting off to a good start for unifying all of these methods. To stabilize such high speed two-phase flows, one applies micro-orifices to “fluidize” these flows to try to obtain proper distribution of the phases in as many as 1200 parallel microchannels to date. For this, a new simulator for predicting the threshold of parallel channel two-phase instabilities has been developed in the LTCM lab and compared to flow visualisations. For the micro-two-phase cooling systems, new steady-state and transient operation simulators have been proposed for pump and compressor driven systems and for gravity-driven thermosyphon systems (the latter with zero energy consumption). To demonstrate this cooling approach, the LTCM lab is applying this knowledge to the cooling of: CPU’s and GPU’s for high performance servers and supercomputers, future 3D stacked computer chips with interlayer micro-cooling channels inside, IGBT’s of power electronics, high energy physics particle detectors for CERN, fiber optic laser diodes, etc. The present lecture will give an overview, primarily drawn from the speaker’s own work, including: selected videos of the flow patterns and coalescence phenomena, several experimental and image processing techniques (time-strip analysis, micro-particle shadow velocimetry, and planar laser induced fluorescence), and a few simulations of these cooling systems under dynamic operation, and thus highlight the status, challenges and future directions of this field, referred to tentatively here as “hyper-micro-thermal fluidics”.

Application of Wavelength Control Technology of Radiation to Drying Furnace

Technologies for drying a material with a flammable solvent at a low temperature are needed. A hot air dries a material in a conventional drying furnace. The hot gas could ignite the flammable solvent. A cold air is flowed in a drying furnace to prevent from igniting a flammable solvent. The cold air can not supply enough heat to evaporate a flammable solvent. The radiative heat transfer can compensate for the lack of the heat. It is reasonable to emit radiation at the absorption band of a flammable solvent in order to evaporate the flammable solvent.
It is known that a cavity structure and a meta material structure have a high emissivity at a specific wavelength. A cavity structure and a meta material structure were designed by using a cavity resonance model and an LC circuit model, respectively, in such a way that the cavity structure and the meta material structure have a high emissivity at the absorption band of a flammable solvent. It has been clarified that the cavity structure and the meta material structure have smaller reflectances than 0.2 at the absorption band of the flammable solvent by using an RCWA analysis method. Assuming Kirchhoff's law, the emissivity equals to 1 minus reflectance in the case that the transmittance is about zero. The cavity structure and the meta material structure were formed by using MEMS technologies. The cavity structure were made of a glassy material by an imprinting method to prevent from the deformation of the shape at a high temperature and the gold film was foamed on the cavity structure by an sputtering method. The meta material with Au/Al₂O₃/Au was fabricated by using an sputtering equipment, an atomic layer deposition equipment, and an electron beam lithography equipment. The reflectances of the cavity structure and the meta material structure were measured by using an FT-IR equipment with an integral sphere. It is clarified that the emissivities of the meta material structure are larger than 0.8. The emissivities of the cavity structure and the meta material at a high temperature are measured by using a blackbody furnace.
A numerical simulation using a radiosity method were carried out to clarify the heating effect of the cavity structure and the meta material structure. It was clarified that the cavity structure and the meta material structure with larger emissivity than 0.8 can dry a material with a flammable solvent at a low temperature. Experiments are performed to verify the heating effect of the cavity