This work presents an experimental study of phonon transport in individual suspended single-wall carbon nanotubes (SWCNTs). Through the use of a micro fabricated device consisting of two adjacent suspended membranes, each with a platinum resistance heater and thermometer, the thermal conductance of several individual SWCNTs has been directly measured over the temperature range of 100 to 490 K. The effects of Umklapp phonon-phonon scattering remain weak and the thermal conductance remains roughly proportional to the calculated ballistic conductance throughout the temperature range. The macroscopic thermal conductance increases with temperature throughout the temperature range indicating static scattering processes or contact thermal resistance dominate transport in this regime. These results are an order of magnitude lower than the predicted ballistic thermal conductance calculated for a defect-free (18,0) nanotube. The results contrast with thermal conductance measurements reported using a high-bias DC self heating method. The discrepancy is discussed in terms of the differences in the contact thermal resistance, defects, and measurement methods.

In analogy to the asymmetric transport of electricity in a familiar electrical diode, a thermal rectifier transports heat more favorably in one direction than in the reverse direction. One approach to thermal rectification is asymmetric scattering of phonons and/or electrons, similar to suggestions in the literature for a sawtooth nanowire [1] or 2-dimensional electron gas with triangular scatterers [2]. To model the asymmetric heat transport in such nanostructures, we have used phonon ray-tracing, focusing on characteristic lengths that are small compared to the mean free path of phonons in bulk. To calculate the heat transfer we use a transmission-based (Landauer-Buttiker) method. The system geometry is described by a four-dimensional transfer function that depends on the position and angle of phonon emission and absorption from each of two contacts. At small temperature gradients, the phonon distribution function is very close to the usual isotropic equilibrium (Bose-Einstein) distribution, and there is no thermal rectification. In contrast, at large temperature gradients, the anisotropy in the phonon distribution function becomes significant, and the resulting heat flux vs. temperature curve (analogous to I-V curve of a diode) reveals large thermal rectification.

We report in-plane and through-plane conductivity measurements of dense YSZ films varying in thickness from 20 to 200 nm. In-plane measurements were performed on YSZ films grown on silicon wafers coated with SiO 2 or Si 3 N 4 . Micro-fabricated strips with Pt electrodes in various geometries were used to obtain conductivity as a function of temperature from 200 – 600 °C in a custom-designed micro-probe station. These films have activation energies, which vary from 0.77 to 1.09 eV. Their absolute conductivity is lower compared with other reports. Through-plane and fuel cell measurements were performed by depositing YSZ on a nitrided silicon wafer, then etching through the wafer and depositing porous platinum electrodes on both sides [6,7]. We discuss the electrochemical conduction studies in detail along with fuel cell performance and correlation with electrode microstructure.

Modeling the thermoelectric properties of nanocomposites is difficult due to the complex grain boundary scattering processes which scatter both electrons and phonons. In this work we describe a code we developed which numerically calculates the electrical and thermal properties of bulk and nanocomposite thermoelectric materials using the Boltzmann equation under the relaxation time approximation. The code is capable of calculating all the relevant thermoelectric properties over a wide range of temperatures, doping concentrations, and compositions, allowing for a full characterization of the material. We model nanocomposites by incorporating a grain boundary scattering rate based on a simple model we developed and models in the literature. The code and grain boundary scattering models are validated on bulk data and data from nano-SiGe, and are then applied to other candidate thermoelectric materials to see if they would be good candidates for nanocomposites. The analysis shows that GaAs might be promising as a nanocomposite thermoelectric material.

Palladium and its alloys have long been used as hydrogen separation membranes due to their extremely high permeability and selectivity to hydrogen over all other gases [1]. The hydrogen permeation process begins with selective chemisorption of the gas onto the metal surface. As the adsorption process is the point in the permeation sequence where the majority of gases become excluded, it follows that a cleverly designed device could be created to take advantage of the so-called ‘fast’ diffusion paths of surface and grain-boundary diffusion to further enhance permeability without sacrificing selectivity. The contribution of grain-boundary diffusion to the overall permeation rate is dependent on the relative volume in the membrane occupied by grain-boundaries versus bulk material. Typically, grain boundaries only make up a miniscule fraction of the overall volume and therefore only contribute an appreciable amount to the overall diffusion process at temperatures low enough to make the bulk diffusion process nearly stagnant. However, in the case of a nanostructured membrane this paradigm is no longer valid. The fabrication methods associated with extremely thin membrane deposition typically lead to highly non-equilibrium microstructure with an average grain size on the order of tens of nanometers [2]. In order to exploit the potential advantages of grain boundary diffusion the nano-scale grains must persist throughout operation. To avoid the tendency for the grain structure to relax to a more equiaxed, coarse-grained morphology the self-diffusion of metal atoms in the film must be minimized by operating the membranes at a temperature much lower than the membrane melting temperature. Figure 1 shows the microstructural changes in a thin, sputtered, Pd/Ag alloy film before and after annealing. The initial fine-grained structure on the bottom surface of the membrane is due to a combination of low substrate temperature during deposition and the Ti adhesion layer onto which the Pd/Ag layer was deposited. After annealing at 400 C the grains have coarsened and the top and bottom structure are identical.

The classical molecular dynamics simulation was conducted in order to clarify the effects of the surface structural clearances in nanometer scale on thermal resistance at a liquid-solid interface as well as static and dynamic behaviours of fluid molecules in the vicinity of the surface. A liquid molecular region confined between the solid walls, of which the interparticle potential was Lennard-Jones type, was employed as a calculation system. The thermal resistance between the liquid molecular region and the solid walls with nanostructures was calculated by the heat flux and the temperature jump obtained in the molecular dynamics simulations. With changing the surface structural clearances from 0 to 2.81 nm the thermal resistance between the liquid molecular region and the solid walls with nanostructures once decreased and became the minimum value when the structural clearances were about 0.7 nm. Surface area in molecular scale and fluid density at the interface were dependent on the surface structural clearances and the thermal resistance index calculated by the relative surface area in molecular scale and the relative fluid density at the interface could predict thermal resistance change depending on the nanostructural clearances. Surface nanostructural clearances affected the fluid molecular motions along the heat transfer direction only when the molecular velocity was averaged over a specific characteristic time. Surface nanostructural clearances affected the diffusion behaviours of fluid molecules in the vicinity of the surface too.

In this study molecular dynamic (MD) simulations are performed to study the interfacial thermal resistance between a nanofin and a working fluid. A (5, 5) carbon nanotube (CNT) of diameter 6.78 Å and various lengths are immersed in different fluids in these analyses. For this simulation the total numbers of the fluid molecules, and the breath and the height of the cell are kept constant. In these simulations, the nanotube is placed at the centre of the cell and the matrix molecules surround the nanotube. Periodic boundary conditions are applied in all the directions. So the system under consideration is array of long nanotubes aligned in the horizontal direction. Simulation procedure consists of first minimizing the system. During the minimization the system is allowed to relax. During the simulations, nanotube and water molecules are allowed to move but the cell size remains constant. After minimization, NVT process is performed for 10ps to scale the velocities so that the average temperature of the cell is 300K. After the ensemble is equilibrated to the base temperature of 300K, the temperature of the nanotube is raised to 750K, by scaling the velocities of the carbon atoms. In the next step the system is allowed to relax under constant energy. This is done by performing the NVE equilibration for 10ps. The difference in the temperature of the carbon nanotube and the matrix is then calculated and plotted against the equilibration time. For the all three matrix, the temperature decreases exponentially with time as predicted by various researchers in the literature. From the graphs the interfacial resistance for water, ethyl alcohol and 1-Hexene is found to be 7.76×10-8, 6.76×10-8 and 35.1×10-8 W/m2K. The value of interfacial resistance for water is consistent with results in the literature.

For the last decade, many researchers have been working to develop self-healing materials, and have obtained good results in the field of polymers. The research described in this paper applies the concept of self healing to fabricate self healing metal matrix composites, with the aid of models developed by the authors for the manufacturing processes and self-healing behavior. The development of self-healing metals is a novel idea that has not been explored in great detail yet. The concept of self-healing described in this paper consists of controlling a low temperature melting alloy imbedded within a higher temperature alloy to create a self healing composite (SHC). When the SHC is damaged or cracked, heat may be applied to the affected area whereupon the low melting alloy will melt and flow into the crack healing the damage and sealing the crack. This study consists of theoretical analysis, metallurgical design, experimental synthesis and Computational Fluid Dynamics of a self-healing aluminum alloy designed by the authors. The aluminum alloy matrix is reinforced with microtubes of alumina (Al 2 O 3 ) that are filled with a low melting point solder alloy. The objective of the study reported here was to find the influence and efficiency of the solder in healing an aluminum matrix. To check this effect a crack was created in the metal surface, piercing the microtube(s) filled with solder, and then the SHC was heated above the melting point of the solder alloy.

We present a simple theory of diffusive phonon heat transport in silicon thin films using the Two-Temperature Model (TTM). In silicon thin films, boundary scattering reduces the lifetime and hence, the mean free path of acoustic phonons. As acoustic phonons are responsible for heat transport in silicon, the latter effect leads to a reduction in the lattice thermal conductivity. However, optical phonons are unaffected by boundary scattering. As the boundary scattering rate exceeds the inverse lifetime of acoustic phonons and the energy relaxation rate between optical and acoustic phonons, it results in an energy transfer bottleneck. The reduced lattice thermal conductivity from boundary scattering and the energy transfer bottleneck are taken into account in the TTM. We apply the TTM to find the steady temperature distribution in a 2D model of a silicon-on-insulator (SOI) device. The numerical results are in good agreement with those obtained from the more sophisticated full dispersion model of the Boltzmann Transport Equation (BTE). We apply the TTM to calculate the steady state and transient temperature distributions in a simplified 1D model of a SOI device.

Thermoelectric effects can be used for direct conversion of heat into electricity using a solid-state device. We describe novel metal/semiconductor nanostructured materials where the heat and charge transport are modified at the atomic level. Hot electron filtering using heterostructure barriers is used to break the trade off between high Seebeck coefficient and high electrical conductivity. Embedded ErAs nanoparticles are used to reduce the lattice thermal conductivity without significant effect on electrical conductivity. The thermal conductivity, electrical conductivity, and Seebeck coefficient of ErAs:InGaAlAs alloys are characterized. The measured ZT is ∼1 at high temperatures 650 K. Integrated circuit fabrication techniques are used to transfer the n- and p-type thin films onto AlN plates and power generation modules are made with hundreds of thin film elements. An output power density >1W/cm2 is demonstrated at an external temperature difference of 120K across the module. Finally, the prospect of other metal semiconductor multilayer structures based on TiN/GaN for high temperature operation will be briefly reviewed.

Ab-initio molecular dynamics (MD) which employs density functional theory (DFT) is used to study thermal energy transport phenomena in nano-scale structures. Thermal equilibration in multiple thin layer structures with thicknesses less than 1 nanometer per layer is simulated. Different types of layer combinations are investigated. Periodic boundary conditions in all directions are used in all cases. Two neighboring layers are first set to different temperatures using Nose-Hoover thermostats, and then the process of energy equilibration is simulated with a “free run” (without any thermostat controlling the temperatures). The temperature evolutions in the two neighboring layers are computed. The atomic vibration power spectra are calculated and used to explain the phenomena observed in the simulation.

Metal and complex hydrides offer very promising prospects for hydrogen storage that reach the DOE targets for 2015. However, slow sorption kinetics and high release temperature must be addressed to make automotive applications feasible. Reducing the enthalpy of formation by destabilizing the hydride reduces the heat released during the hydrogenation phase and conversely allows desorption at a lower temperature. High-energy ball milling has been shown to decrease the release temperature, increase reaction kinetics and lower the enthalpy of formation in certain cases. Increased surface and grain boundary energy could play a role in reducing the enthalpy of formation, but the predicted magnitude is too small to account for experimental observations. As the particle and grain sizes are reduced considerably under high-energy treatments, structural defects and deformations are introduced. These regions can be characterized by an excess volume due to deformations in the lattice structure, and have a significant effect on the material properties of the hydride. We propose a thermodynamic model that characterizes the excess energy present in the deformed regions to explain the change in physical properties of metal hydrides. An experimental investigation using the TEM to study the effect of lattice deformations and other nanostructures on the desorption process is underway.

One of the main challenges in designing oil & gas downhole wireline logging tools for high temperature and high pressure (HTHP) environments is to put together the most efficient thermal packaging to enhance the tool’s temperature survival time. In general, not all electronic components and sensors can withstand severe downhole temperature (max 500 degrees Fahrenheit). For those heat sensitive components, their electrical response and performance either decay or in some cases they completely fail when their temperature exceeds 300° F. In oil & gas wireline logging applications, the heat sensitive components can be thermally protected inside a Dewar vacuum flask that includes one or two thermal isolators and heat sinks. Cooler electronic components results in longer logging times that lead to a much higher performance and profitability. This paper first discusses the development of a one-dimensional analytical model to determine the transient temperature of heat sensitive sensors and electronic components in wireline logging tools. Second, it introduces a new and improved thermal packaging scheme based on a newly developed and commercially available nanoporous material. This material has a very low thermal conductivity and is used as a thermal shield between the outside environment and the electronics inside the flask. The new packaging scheme also includes a new design for the heat sink which is made of several solid disks separated by this nanoporous material. Results from this new design have shown roughly a 30% improvement compared with the conventional design. Results from both analytical and laboratory tests are discussed in this paper.

Increased research into the chemistry, physics and material science of hydrogen cycling compounds has led to the rapid growth of solid-phase hydrogen-storage options. The operating conditions of these new options span a wide range: system temperature can be as low as 70K or over 600K, system pressure varies from less than 100kPa to 35MPa, and heat loads can be moderate or can be measured in megawatts. While the intense focus placed on storage materials has been appropriate, there is also a need for research in engineering, specifically in containment, heat transfer, and controls. The DOE’s recently proposed engineering center of expertise underscores the growing understanding that engineering research will play a role in the success of advanced hydrogen storage systems. Engineering a hydrogen system will minimally require containment of the storage media and control of the hydrogenation and dehydrogenation processes, but an elegant system design will compensate for the storage media’s weaker aspects and capitalize on its strengths. To achieve such a complete solution, the storage tank must be designed to work with the media, the vehicle packaging, the power-plant, and the power-plant’s control system. In some cases there are synergies available that increase the efficiency of both subsystems simultaneously. In addition, system designers will need to make the hard choices needed to convert a technically feasible concept into a commercially successful product. Materials cost, assembly cost, and end of life costs will all shape the final design of a viable hydrogen storage system. Once again there is a critical role for engineering research, in this case into lower cost and higher performance engineering materials. Each form of hydrogen storage has its own, unique, challenges and opportunities for the system designer. These differing requirements stem directly from the properties of the storage media. Aside from physical containment of compressed or liquefied hydrogen, most storage media can be assigned to one of four major categories, chemical storage, metal hydrides, complex hydrides, or physisorption. Specific needs of each technology are discussed below. Physisorption systems currently operate at 77K with very fast kinetics and good gravimetric capacity; and as such, special engineering challenges center on controlling heat transfer. Excellent MLVSI is available, its cost is high and it is not readily applied to complex shape in a mass manufacture setting. Additionally, while the heat of adsorption on most physisorbents is a relatively modest 6–10kJ/mol H2, this heat must be moved up a 200K gradient. Physisorpion systems are also challenged on density. Consequently, methods for reducing the cost of producing and assembling compact, high-quality insulation, tank design to minimize heat transfer while maintaining manufacturability, improved methods of heat transfer to and from the storage media, and controls to optimize filling are areas of profitable research. It may be noted that the first two areas would also contribute to improvement of liquid hydrogen tanks. Metal hydrides are currently nearest application in the form of high pressure metal hydride tanks because of their reduced volume relative to compressed gas tanks of the same capacity and pressure. These systems typically use simple pressure controls, and have enthalpies of roughly 20kJ/mol H2 and plateau pressures of at most a few MPa. During filling, temperatures must be high enough to ensure fast kinetics, but kept low enough that the thermodynamically set plateau pressure is well below the filling pressure. To accomplish this balance the heat transfer system must handle on the order of 300kW during the 5 minute fill of a 10kg tank. These systems are also challenged on mass and the cost of the media. High value areas for research include: heat transfer inside a 35MPa rated pressure vessel, light and strong tank construction materials with reduced cost, and metals or other materials that do not embrittle in the presence of high pressure hydrogen when operated below ∼400K. The latter two topics would also have a beneficial impact on compressed gas hydrogen storage systems, the current “system to beat”. Complex hydrides frequently have high hydrogen capacity but also an enthalpy of adsorption >30kJ/mol H2, a hydrogen release temperature >370K, and in many cases multiple steps of adsorption/desorption with slow kinetics in at least one of the steps. Most complex hydrides are thermal insulators in the hydrided form. From an engineering perspective, improved methods and designs for cost effective heat transfer to the storage media in a 5 to 10MPa vessel is of significant interest, as are materials that resist embrittlement at pressures below 10MPa and temperatures below 500K. Chemical hydrides produce heat when releasing hydrogen; in some systems this can be managed with air cooling of the reactor, but in other systems that may not be possible. In general, chemical hydrides must be removed from the vehicle and regenerated off-board. They are challenged on durability and recycling energy. Engineering research of interest in these systems centers around maintaining the spent fuel in a state suitable for rapid removal while minimizing system mass, and on developing highly efficient recycling plant designs that make the most of heat from exothermic steps. While the designs of each category of storage tank will differ with the material properties, two common engineering research thrusts stand out, heat transfer and structural materials. In addition, control strategies are important to all advanced storage systems, though they will vary significantly from system to system. Chemical systems need controls primarily to match hydrogen supply to power-plant demand, including shut down. High pressure metal hydride systems will need control during filling to maintain an appropriately low plateau pressure. Complex hydrides will need control for optimal filling and release of hydrogen from materials with multi-step reactions. Even the relatively simple compressed-gas tanks require control strategies during refill. Heat transfer systems will modulate performance and directly impact cost. While issues such as thermal conductivity may not be as great as anticipated, the heat transfer system still impacts gravimetric efficiency, volumetric efficiency and cost. These are three key factors to commercial viability, so any research that improves performance or reduces cost is important. Recent work in the DOE FreedomCAR program indicates that some 14% of the system mass may be attributed to heat transfer in complex hydride systems. If this system is made to withstand 100 bar at 450K the material cost will be a meaningful portion of the total tank cost. Improvements to the basic shell and tube structures that can reduce the total mass of heat transfer equipment while maintaining good global and local temperature control are needed. Reducing the mass and cost of the materials of construction would also benefit all systems. Much has been made of the need to reduce the cost of carbon fiber in compressed tanks and new processes are being investigated. Further progress is likely to benefit any composite tank, not just compressed gas tanks. In a like fashion, all tanks have metal parts. Today those parts are made from expensive alloys, such as A286. If other structural materials could be proven suitable for tank construction there would be a direct cost benefit to all tank systems. Finally there is a need to match the system to the storage material and the power-plant. Recent work has shown there are strong effects of material properties on system performance, not only because of the material, but also because the material properties drive the tank design to be more or less efficient. Filling of a hydride tank provides an excellent example. A five minute or less fill time is desirable. Hydrogen will be supplied as a gas, perhaps at a fixed pressure and temperature. The kinetics of the hydride will dictate how fast hydrogen can be absorbed, and the thermodynamics will determine if hydrogen can be absorbed at all; both properties are temperature dependent. The temperature will depend on how fast heat is generated by absorption and how fast heat can be added or removed by the system. If the design system and material properties are not both well suited to this filling scenario the actual amount of hydrogen stored could be significantly less than the capacity of the system. Controls may play an important role as well, by altering the coolant temperature and flow, and the gas temperature and pressure, a better fill is likely. Similar strategies have already been demonstrated for compressed gas systems. Matching system capabilities to power-plant needs is also important. Supplying the demanded fuel in transients and start up are obvious requirements that both the tank system and material must be design to meet. But there are opportunities too. If the power-plant heat can be used to release hydrogen, then the efficiency of vehicle increases greatly. This efficiency comes not only from preventing hydrogen losses from supplying heat to the media, but also from the power-plant cooling that occurs. To reap this benefit, it will be important to have elegant control strategies that avoid unwanted feedback between the power-plant and the fuel system. Hydrogen fueled vehicles are making tremendous strides, as can be seen by the number and increasing market readiness of vehicles in technology validation programs. Research that improves the effectiveness and reduces the costs of heat transfer systems, tank construction materials, and control systems will play a key role in preparing advanced hydrogen storage systems to be a part of this transportation revolution.

Alternating 10 nm thermoelectric films of N-type Si/SiGe and P-type Si/SiGe and B 4 C/B 9 C have been fabricated on various substrates, electrically joined and thermoelectric properties measured from 40°K up to 700°K. These nanoscale thermoelectric films demonstrate excellent thermoelectric power factors significantly higher than current bulk thermoelectric materials. The implications of the measured thermoelectric Seebeck coefficient data and electrical resistivity data for alternating 10 nm films that are grown to thicknesses of one to 10 microns means efficiencies of 15% at 200°C temperature differences and efficiencies of 30% at 400°C temperature differences. Utilizing Seebeck and resistivity data obtained by Hi-Z and UCSD, along with published bulk thermal conductivity data, which is conservative, unique thermoelectric module and generator concept designs for both power generation and cooling are presented over wide temperature and power ranges.

Nanocatalytic particles of Gold (Au), Platinum (Pt), and Palladium (Pd) are highly reactive at room-temperature and can be used to generate heat in micro-scale devices for portable power generation. No pre-heating is required for light-off and high steady-state operating temperatures can be sustained with high density alcohol-air premixtures. Preliminary experiments conducted in our lab and those reported by Hu and co-workers at Oak Ridge National Lab have measured peak operating temperatures ∼ 300–500 degrees Celsius using near-stoichiometric methanol/air and ethanol/air premixtures at ambient initial temperature and atmospheric pressure. The effect of particle size, morphology, mass loading, and flow residence time are reported for different mixture stoichiometries. Temperature measurements and gas species analyses are also tabulated. Interestingly, smaller particles were observed to be less reactive than larger particles for the same mass loadings for select conditions. Materials characterization of the particles has also been conducted to characterize the specific surface area of the catalyst and evaluate the importance of particle sintering, morphology changes, and particle distribution.

We have identified TiN/GaN and ZrN/ScN as two possible pure rocksalt structured metal/semiconductor combinations for fabrication of solid-state thermionic energy converters for high operational temperatures. The selection of the materials was constrained by issues that are critical to the integration of heterogeneous materials such as crystallographic compatibility and thermodynamic stability of the metal/semiconductor combinations at high operating temperatures. The first nitride superlattice system consists of TiN as the metal layer and GaN, in its metastable rocksalt phase, as the semiconductor layer, grown on rocksalt MgO substrates. The metastable rocksalt GaN (rs-GaN) phase is stabilized by pseudomorphic epitaxy on a metallic rocksalt TiN underlayer, and its existence has been verified using high-resolution x-ray diffraction and transmission electron microscopy. The critical thickness for the rocksalt-to-wurtzite phase transition has been empirically determined to be between 1 and 2 nm, although much thicker rocksalt GaN films, up to approximately 6 nm, can be maintained for several superlattice periods. The second pure rocksalt-structured superlattice system analyzed consists of alternating layers of metallic ZrN and semiconducting ScN. These epitaxial superlattices were grown on rocksalt MgO substrates using dc magnetron sputtering in a nitrogen-argon ambient.