The thermal conductivity of one dimensional nanostructures, such as nanowires, nanotubes, and polymer fibers, is of significant interests for understanding nanoscale thermal transport phenomena as well as for practical applications in nanoelectronics, energy conversion and thermal management. This has fueled a significant body of research aiming to identify causes and enable rational engineering of thermal conductivity, which has predominantly been achieved by limiting the phonon mean-free-path by the characteristic size of the nanostructure. Thermal conductivity in such small nanostructures could be further altered by way of phonon confinement as the structure approaches the phonon wavelength and reduction of the phonon velocity, signified via change in the elastic constants.

An enhanced experimental technique ~100X more sensitive than the original method to characterize the thermal properties of one-dimensional nanostructures such as semiconductor nanowires, nanotubes, and polymer fibers. The thermal conductivity and mechanical properties of semiconductor nanostructures was measured using the developed calorimetry setup and an in-situ FIB tensile testing setup. The thermal conductivities of sub-20 nm Ge and Ge-Si core-shell nanowires were measured and found to be significantly lower than bulk SiGe alloys or smooth Si NWs due to phonon confinement. Meanwhile, ~5 nm thick shell crystalline Si nanotubes were found to have thermal conductivity lower than the apparent boundary scattering limit and lower than even ~5 nm thick amorphous Si nanotubes, which is understood by elastic softening measured via tensile testing. Next, further mechanical characterization of crystalline and amorphous nanotubes is performed, and strong size-dependent viscoelastic behavior is observed in amorphous nanotubes due to a fluid-like surface, with room temperature viscosity equivalent to that of bulk glass above 1000 C. Finally, the thermal and mechanical properties of polymer nanofibers with diameters down to sub-100 nm are measured and discussed. The axial thermal conductivity of individual nanofibers is strongly correlated with the crystalline morphology while the elastic modulus stress relaxation is found to be highly size-dependent, enabling the smallest diameter fibers to break the stiffness-toughness tradeoff present in most material systems.

Flexible thermoelectric (TE) refrigerators are attractive for personalized thermal management, for example, when integrated into clothes. Flexible TE power generators can also find applications in wearable self-powered mobile electronics. These devices, however, need high thermoelectric performance and mechanical flexibility under compressive and tensile strains, a combination of device properties that are often difficult to achieve. Despite significant efforts in the field, flexible TE devices, such as the ones made from organic materials, often exhibit lower cooling power than the actual needs. In this thesis, I studied flexible TE devices that offer high coefficient of performance.

In chapter 1, The motivation of personalized thermal management was provided. Fundamental aspects of thermoelectric devices were discussed.

In chapter 2, Technical approaches of making flexible TE devices were described. A flexible TE device was demonstrated.

In chapter 3, the coefficient of performance (COP) measurement results were presented and discussed. Specifically, COP for cooling and heating applications was measured.

In chapter 4, conclusions drawn from this study were summarized. Future work was suggested.

Desiccant based dehumidification allows for independent treatment of latent and sensible properties in air conditioning systems and enables the use of energy-saving solar thermal evaporative cooling systems. These systems can reduce dependence on vaporcompression air conditioning which is a significant source of greenhouse gases. Recent research has focused on polymer-based desiccant materials which have higher adsorption capacities than conventional materials such as silica gel and zeolite. Solid desiccant materials used in dehumidification systems are typically utilized in a cycle in which they absorb water vapor from a process air stream and are then regenerated by a source of hot, dry air which evaporates and removes water contained in the material, allowing it to again absorb water from the process air. This research focuses on the development and testing of a hygroscopic salt impregnated, NIPAM hydrogel-based thermo-responsive desiccant material that can be regenerated by elevating the material to its lower critical solution temperature (LCST), enabling a significant energy saving over materials that require regeneration through water evaporation. At this LCST, the water adsorbed by the desiccant becomes insoluble in its NIPAM polymer matrix and is removed through droplet formation induced by gravity. Adsorption and regeneration testing was performed on the material to evaluate the feasibility of using this temperature-sensitive desiccant into a dehumidification system. Testing shows this composite material has a maximum SR of > 2 and analysis indicates improvement in regeneration heat requirements over existing desiccant materials.

Thermal properties (thermal conductivity and specific heat) of the disordered materials, such as amorphous silicon (a-Si), polymer, and nano-crystalline semiconductors, are of significant interests for fundamental understanding of thermal transport process and for technical applications in thermal energy management and conversion. Due to the random distribution of atoms or molecules in disordered materials, the study of thermal transport is more challenging than that in crystalline materials. Understanding of the heat carrier transport behavior can be utilized to engineer the thermal properties in disordered materials, which can be applied for better devices thermal design and improving thermal energy conversion efficiency.

We have studied the size dependent thermal conductivity of a-Si thin films and nanotubes, and observed unusually high and anisotropic thermal conductivity in the isotropic a-Si nanostructure. This manifests surprisingly broad mean free path distribution of the propagating modes (propagons), which is found to range from 10 nm to 10 μm, in the disordered and isotropic structure. Constraining the long MFP propagons by boundary scattering in thin film and nanotubes explains the appreciable size effect in a-Si. Additionally, we developed a novel platform to measure the specific heat of low-dimensional disordered materials. By measuring the frequency dependent temperature rise data along the Nylon nanofibers (NFs), we are able to extract the specific heat and thermal conductivity simultaneously. While the thermal conductivity is increased by 50% over the bulk value in the 600 nm NFs, the specific heat exhibits bulk-like behavior. Finally we engineered the thermal conductivity in nano-crystalline bismuth-antimony-telluride (BST) by embedding SiO2 or diamond nanoparticles (NPs) at temperature below 300K, which has important application in thermoelectric cooling. We have shown that the embedded NPs work as additional scattering centers for lattice vibration (or called phonons), and can efficiently scatter the long MFP phonons in BST. We have observed 23% reduction of thermal conductivity, and 15% improvement of thermoelectric figure of merit (ZT) in the 0.5 vol. % Diamond NPs mixing sample, compared to the non-NPs nano-crystalline BST.

The rapid growth of modern economy and the increasingly high quality of human life over the last century are essentially attributed to the development of energy technologies, including the conversion, transport, storage, and utilization of energy. The transfer of thermal energy plays an instrumental role in virtually all energy related technologies. As such, exploring the fundamentals of heat transfer is of significant importance for both scientific advancement and technological development. Each of the basic heat transfer modes is governed by different length scales, understanding and engineering of which thus underpin the development of heat transfer technologies. In this dissertation, quantification and engineering of characteristic lengths in two different heat transfer mechanisms, namely, phase change heat transfer (PCHT) and radiation heat transfer, are studied.

Liquid-vapor PCHT utilizes latent heat of vaporization to efficiently transport a large amount of thermal energy, which is being used in a variety of thermal energy conversion and management applications. The critical heat flux (CHF) of PCHT, either evaporation or boiling, is fundamentally limited by mass flux of the vapor departing from the liquid-vapor interface, known as the kinetic limit. This limit could be in theory greater than 1 kW cm-2 on a planar surface, but its experimental realization has remained elusive. In the first part of this dissertation, by leveraging the small length scale for heat conduction and vapor bubble departure within a thin liquid film, a new “thin film boiling” mechanism is proposed and realized by confining the boiling liquid within a microscale film on a nanoporous membrane. Superior heat transfer performance is demonstrated with a high CHF of over 1.8 kW cm-2, which is among the highest reported values for PCHT on a planar surface and is within a factor of four of the calculated theoretical kinetic limit. Furthermore, by continuously shrinking the liquid film, a universal transition from boiling to evaporation is further identified for different fluids with varying surface tension values (water, ethanol, isopropanol, and perfluorohexane), and the heat transfer characteristics of the transition points are found to be close to the kinetic limit. The limiting factors dictating the transition between the two different heat transfer modes are also elucidated.

In advanced materials for high temperature thermal transport such as thermal barrier coatings for turbine blades and heat transfer media in concentrated solar power plants, the radiative thermal transport, which is often negligible at room temperature, becomes comparable to or even more important than heat conduction. However, an effective thermal conductivity encompassing both heat transfer modes is often used at high temperatures, due to the lack of a convenient methodology to separate the contributions from conduction and radiation, especially for high temperature thermal measurement using the prevailing instrument called the laser flash analyzer (LFA). In particular, the characteristic length scale of radiation heat transfer is often elusive. In the second part of this dissertation, a transient coupled conduction/radiation model, with the realistic boundary conditions used in the LFA, is developed to study the transient heat transfer in high temperature materials. By identifying the propagation length of photons in different portions of the spectrum, heat transfer due to photons with direct transmission and diffusive transport is delineated. In the conductive thermal conductivity of semi-transparent materials obtained from the LFA, conventional models would still cause non-negligible error at high temperatures, while the coupled model developed here can accurately quantify the radiative contribution.

Nowadays, 80% of the developed world’s energy comes from fossil fuels. With the aim to peak the carbon emission by 2030 and neutralize it by 2050, sustainable energy sources are being actively explored. It is estimated that 90% of the current energy budget is in the form of thermal energy, implying the significance of efficient harnessing and management of thermal energy towards an energy-efficient world. Concentrating solar power (CSP) coupled to thermal energy storage have gained extensive attention in recent years due to their capability to store high-grade thermal energy at low cost for on-demand electricity generation. However, current R&D effort in CSP has been hindered by the limited knowledge in heat transfer of high-temperature components in CSP, such as solar-absorbing coatings, alloy containments, granular flow and molten salt, etc. One of the challenges stems from the lack of suitable in-situ measurement tool for these materials due to their high temperature and corrosive natures. To tackle this difficulty, a non-contact laser-based modulated photothermal radiometry (MPR) thermal measurement system was developed. The MPR uses the thermal emission from the specimen surface as the thermometry probe, which is suitable for high-temperature measurement under harsh environment. The system was first characterized by bulk materials and solar-absorbing coating. The MPR system was modified for the measurement of the intrinsic thermal conductivity and convection heat transfer coefficient of flowing fluid in a channel. Afterwards, the MPR system was applied for the measurement of moving particle bed for CSP. The effective thermal conductivity of moving particle with variation in the flowing velocity, temperature, particle size, and particle morphology was obtained. We disseminated the near-wall thermal resistance from the bulk thermal resistance. The discrete element modelling (DEM) indicated that the near-wall thermal resistance was induced by the diffusive motion of particles away from the wall in a moving particle bed. Low-grade thermal energy is a universal form of energy with easy access and low cost. The low-grade heat was utilized for desalination and water harvesting using hydrogel materials. We developed a new class of thermo-responsive hydrogels for forward-osmosis desalination where the hydrogel draws water molecules across the membrane from seawater due to its high ionic strength. This new class of draw agent processed the unique LCST phase transition to release water in liquid phase at 35 ℃ to avoid the latent heat penalty in conventional FO desalination. Subsequently, we applied polyelectrolyte hydrogels for continuous solar-driven evaporative desalination, where solar energy was directly converted to heat to evaporate the seawater. Due to the unique osmotic pumping mechanism within the hydrogel, the salt crystallization problem was avoided as opposed to conventional capillary wicks. Lastly, we demonstrated that the hydrogel materials could also be applied for water harvesting from moisture in the air by the adsorption-desorption process. The generic idea was to coat a thin hydrogel layer on a metallic scaffold with 100-1000-fold enhanced effective surface for fast heat and mass transfer kinetics. We achieved a daily water yield an order-of-magnitude higher than the state-of-the-art techniques. Due to the evaporative latent heat in the desorption process, we found that the water harvesting device also manifested itself a superior heat sink for power electronic thermal management. Based on the similar geometric design, we developed a moisture thermal battery (MTB) and demonstrated its capability of passive thermal management of 5G device and CPU.

With the increased demand for energy usage and the lack of sufficient renewable energy sources, this has led to a rise in CO_2 emissions. In turn, it causes global warming, therefore, to combat this issue, we must develop new sources of renewable energy. One such technology is thermophotovoltaics (TPV), a technology that converts thermal energy into electricity using photovoltaic (PV) cells [1]. A TPV system consists of a heat source that will heat an emitter to high temperatures above 700°C, by using a variety of heat sources such as fuel from natural gas, biomass, or reusing wasted heat. Once it is heated up, the emitter will act like a filter and only emit photons at a specific wavelength. These photons will then be absorbed by a low bandgap photovoltaic (PV) cell of less than 1.0 eV to generate a current through the photovoltaic effect. To prevent alteration of the material's optical properties caused by oxidation [2, 3], most TPV studies' system setups are under vacuum and with emitters that are not stable at high temperatures for an extended period. Here, we have created a TPV system that can accommodate selective emitters under air and at high temperatures for long-term stability. By using various ceramic compounds to help dissipate heat and prevent heating elements from localized heating, we also incorporated a thermal management system, which effectively dissipates unwanted heat from the PV cells, thus allowing the placement of PV cells as close as 3 mm away from the 1000°C emitter.

Hydrogen (H2) is promising alternative to replace fossil fuels, but its transport and storage has been challenging. As H2 fuel cell vehicles are gaining traction, the infrastructure for storing large amounts of liquid H2 is needed. However, liquid H2 would suffer from boil-off loss, and traditional vapor compression refrigeration systems would not be able to economically recover the lost H2 due to the low efficiencies at cryogenic temperature. Magnetocaloric (MC) refrigeration systems could possess much higher coefficient of performance (COP) at cryogenic temperature compared to the vapor compression ones. Previous work on cryogenic MC systems, however, have only focused on large scale applications which use superconducting magnets to provide a large magnetic field but are prohibitively expensive to operate for small scale applications, such as that of a H2 refilling station. In this work, we modeled the performance of a MC refrigeration cycle using 1-Tesla permanent magnets for H2 liquefaction, with the objective of cooling H2 from 80 K (using liquid nitrogen as the heat sink) to 20 K (boiling point of hydrogen). We evaluated main performance metrics including the total work input to the refrigeration system, COP, total MCM mass in the system, and total volume of the permanent magnets, etc. Our modeling results indicate that such a permanent magnet-based MC cooling system is feasible for small-scale H2 liquefaction, with projected COP values significantly higher than those of vapor compression systems. This work provides design guidelines for future experimental efforts on permanent magnet MC cooling systems for cryogenic cooling.

Calorimetry is a heat measuring process that quantifies heat generation, loss, and transport. Calorimetry holds significant values to a multitude of scientific disciplines such as nanoscale heat transfer, bolometer infrared detection, and drug discovery. This dissertation focuses on two aspects of calorimetry: 1) application of a calorimetry technique to investigate fundamentals of thermal transport at micro/nanoscale level, and 2) development of high-resolution microfluidic calorimeters for biological applications.

A modulated calorimetry technique, namely, the 3ω method, was used to study thermal transport across highly mismatched interfaces, where Au/Si multilayers (MLs), were used as the model material. By leveraging thermal resistance at interfaces of Au and Si, which have highly dissimilar acoustic impedance, we experimentally demonstrated an ultralow thermal conductivity, κ, of 0.33 ± 0.04 W m-1K-1 at room temperature with a high interfacial density of ~0.2 interface nm-1. The measured κ was the lowest amongst inorganic MLs with similar interfacial density. The 3ω method was also applied to measure thermal conductivity of the near-surface regime of tungsten damaged by ion irradiation used in plasma facing components (PFCs) in fusion reactors. The measurement showed a nearly 60\% reduction in κ in comparison to pristine tungsten, which can be detrimental to the mechanical integrity of the PFCs under heat fluxes

The second thrust of the dissertation is focused on the development of high-resolution calorimeters. A high performance resistive thermometer material, NbNx, was developed and integrated into a suspended bridge calorimeter. When used with a modulated heating current and a differential instrumentation schemes, the NbNx-based calorimeter demonstrated an exceptional high power resolution of 0.26 pW at room temperature. Furthermore, the high-resolution calorimetry was extended to microfluidic platform for biological applications. Owing to the low parasitic heat loss and a long-term temperature stability platform, the biocalorimeters developed from this study possess a significant enhancement over the state-of-the-art long-term power detection limit, which enable the detection of metabolic rate of individual living cells in the microfluidic channels. Such a biocalorimetry technique will open up new opportunities in a biomedical field such as in cell and tissue metabolism, drug discovery and screening, and antibody-antigen detection.

As electronics become more powerful and have higher energy densities, it is becoming more and more necessary to find solutions to dissipate these high heat fluxes. One solution to this problem is nanopore evaporative cooling. Based on current literature, the experimental data is far below what is expected from the theoretical calculations.

In this thesis, the experimental results produced heat fluxes much closer to the theoretical values. Experimentally, a maximum heat dissipation of 103 W was achieved on a area which corresponds to a heat flux of on the overall AAO surface or a heat flux of on the active evaporating pore area which is close to the theoretical heat flux of that can be obtained given the membrane parameters. While the results are promising, it still needs to be further studied to obtain higher heat fluxes closer to the theoretical one. In order to do so, membrane parameters of the working porous membrane need to be able to be adjusted.

For the nanopore evaporative experiments done so far, commercially purchased AAO was used. However, commercially purchased AAO is only sold in certain set parameters and the AAO produced has a disordered structure, both of which are not useful for future experimental purposes. Therefore, AAO needs to be fabricated in the lab to meet experimental needs.

The fabrication of AAO on high purity aluminum (99.997%) with 0.3 M Oxalic Acid at 40 V at was achieved. With the given fabrication parameters an average pore size of ~80 nm was achieved with a circularity of 0.91 while the commercially purchased AAO had an average pore size of ~200 nm with a circularity of 0.80.

The initial nanopore evaporative cooling experiments will be discussed along with the process of fabricating usable AAO for these experiments.