Thesis (Ph. D.)--University of Hong Kong, 2000. / Includes bibliographical references.

The entire thesis text is included in the research.pdf file; the official abstract appears in the short.pdf file; a non-technical public abstract appears in the public.pdf file. Title from PDF of title page (University of Missouri--Columbia, viewed on October 5, 2009). Thesis advisor: Dr. Alan G. Whittington. Includes bibliographical references.

Thesis (Ph.D)--Mechanical Engineering, Georgia Institute of Technology, 2010. / Committee Co-Chair: Haynes, Comas; Committee Co-Chair: Wepfer, William; Committee Member: Fedorov, Andrei; Committee Member: Liu, Meilin; Committee Member: Paredis, Chris; Committee Member: Teja, Amyn. Part of the SMARTech Electronic Thesis and Dissertation Collection.

Additive Manufacturing (AM), over the years, has seen a tremendous amount of research for improving the manufacturability of materials into final products. The main advantages of additive manufacturing are the minimizing of waste material as it is an additive process. As well as the ability to create custom low-volume products without the need for creation of expensive tooling or programming before manufacturing begins. Because of these advantages, however, AM is susceptible to unique challenges in the quality side of manufacturing. These challenges include minimizing and detecting defects during the build. The focus of this research looks at the capability of using Pulse Thermography (PT), a nondestructive testing method, with longer than typical pulse length on additively manufactured parts for surface and sub-surface defect detection as well as thermal property determination based on a known void depth. The first and second part of this research will look at a range of pulse lengths greater than 100ms to determine if the previously defined assumption is necessary for accurate defect detection. The significance of increasing the pulse length is to have the ability to increase the overall energy input into the part without having to increase the power. Allowing for the capability of defect detection for both shallow and deeper defects with the same overall setup. One-dimensional simulations r using Forward Time Center Space (FTCS) approximation, show that the assumption of an instantaneous pulse is relative, and defects can be accurately calculated within a range of pulse lengths. Based on the simulations, experimentation was conducted to determine the capability of calculating sub-surface defect depths with a longer pulse on a FDM printed ABS part with 100% in fill. The defect depths will range from 0.3mm to 1.8mm and the widths of the defects used for depth calculation will be 8x8mm. Results of the experiments show that even with FDM printed parts defect depths were accurately calculated up to a depth of 1.2mm. The third aspect of this research looks at the infrared reflections emitting off the surface during the longer pulse. With a longer pulse length, there is more time for the infrared camera to collect thermograms of the surface during the pulse. It was noticed during sub-surface defect detection that the infrared reflections paint a picture of the surface characteristics of the part. Characteristics that include surface imperfections not intended in the original build parameters such as under extrusions and cracks. Defects as small as 150μm with a thermal pixel resolution 75μm are detected. The third and final aspect of this research looks at the ability to use PT with a longer pulse to determine thermal properties of a binder jetted additively manufactured part as well as packing factors that may be otherwise be unknown. When a product is binder jetted a chemical binder is added to the powder layer by layer until a product is formed.

O objetivo deste trabalho foi desenvolver um software, de fácil operação e eficiente, para determinar a difusividade térmica de biomateriais. A necessidade de se conhecer a difusividade térmica de materiais como, por exemplo, esmalte e dentina, é essencial para o estabelecimento de protocolos de utilização clínica laser, para evitar danos colaterais ao paciente. O software desenvolvido, denominado CZ ThermaDiff, baseia-se no processamento de imagens térmicas adquiridas por uma câmera termográfica no infravermelho (ThermaCam SC3000, FLIR System, EUA). Foi desenvolvido em ambiente LabView, o que permitiu criar um painel de controle de fácil operação, contendo apenas duas funções básicas (Start e Stop). O software arquiva os dados em formato de tabela contendo todas as medidas de difusividade térmica, suas médias para intervalos de 10 °C e suas respectivas temperaturas, para uma amostra. Foi observado, tanto para o esmalte quanto para a dentina, que os valores de difusividade não são constantes e aumentavam em função da temperatura. Os valores encontrados foram aplicados a um modelo de transferência de calor, simulando um dente molar humano com as seguintes estruturas: esmalte, dentina e polpa. O modelo baseia-se no método de resistores térmicos, sendo que para a polpa foi utilizado o modelo de difusão de calor, considerando a circulação sanguínea. Os valores de temperatura obtidos neste modelo teórico, utilizando difusividades dependentes da temperatura foram maiores que as obtidas com um valor constante de difusividade, medido à temperatura ambiente. Este fato realça a importância da mensuração da difusividade em função da temperatura e da interface desenvolvida neste trabalho. / The aim of this work was to develop a software, easy to operate and efficient, to determine the thermal diffusivity of biomaterials as enamel and dentin. It is necessary to know the thermal diffusivity of these materials to establish laser irradiation protocols, to avoid collateral damage to the patient. The software developed named called CZ ThermaDiff, processes thermographic images from a thermographic camera (ThermaCam SC300, FLIR System, USA). The software was programmed in LabView environment, allowing easy operation from a control window with only two buttons (start and stop). Thermal diffusivity values, the mean values for intervals of 10 °C and its respective temperature, for one sample are saved in table form. For both biomaterials, thermal diffusivity increased as function of the temperature increase. The experimental thermal diffusivity data were used in a heat transfer model, for a human molar tooth with three layers: enamel, dentin and pulp. The model was based on the thermal resistors method, for the three layers and for the pulp, it was applied the heat diffusion model, taking into account the blood circulation. Using temperature dependent diffusivities, temperatures where values were higher than the temperatures of the theoretical model using a constant diffusivity value, obtained at ambient temperature. This fact emphasized the importance of both: the temperature dependent diffusivity measurement and the software developed in this work.

A graphite foam was developed at Oak Ridge National Laboratory (ORNL) by Dr. James Klett and license was granted to POCO Graphite, Inc. to manufacture and market the product as PocoFoam™. Unlike many processes currently used to manufacture carbon foams, this process yields a highly graphitic structure and overcomes many limitations, such as oxidation stabilization, that are routinely encountered in the development of carbon foam materials. The structure, thermal properties, electrical resistivity, isotropy, and density uniformity of PocoFoam™ were evaluated. These properties and characteristics of PocoFoam™ are compared with natural and synthetic graphite in order to show that, albeit similar, it is unique. Thermal diffusivity and thermal conductivity were derived from Fourier's energy equation. It was determined that PocoFoam™ has the equivalent thermal conductivity of metals routinely used as heat sinks and that thermal diffusivity is as much as four times greater than pure copper and pure aluminum. SEM and XRD results indicate that PocoFoam™ has a high degree of crystalline alignment and near theoretical d spacing that is more typical of natural flake graphite than synthetic graphite. PocoFoam™ is anisotropic, indicating an isotropy factor of 0.5, and may yield higher thermal conductivity at cryogenic temperatures than is observed in polycrystalline graphite.

The purposes of this study were to determine as a function of temperature the thermal diffusivity and/or thermal conductivity of SiC-Mullite and SiC-SiC, and to explain the observed behavior in terms of changes in temperature, microstructure, composition, and/or orientation. Materials used in the SiC-Mullite study consisted of single crystal SiC whiskers (prepared from rice hulls or by the vapor-liquid-solid process) dispersed within a polycrystalline mullite matrix. During measurement of thermal diffusivity, the samples were heated to l500°C and cooled back to room temperature. No hysteresis occurred. However, both thermal diffusivity and conductivity exhibited maximum values at room temperatures, perpendicular to the hot pressing direction, at high volume percentages of SiC whiskers, and when VLS whiskers were employed. The SiC-SiC samples consisted of a crossweave of polycrystalline SiC fibers that were coated with phenolic resin and surrounded by a chemically-vapor-deposited matrix of SiC. The two types of samples examined were prepared with different amounts of resin. The matrices of the high resin samples were found to be dominated by the presence of char. Samples were cycled to 1000, 1400, and l800°C; hysteresis occurred on some of the cycles. Thermal diffusivity was highest parallel to one set of fibers. These results allow the qualitative tailoring of the heat flow properties of these ceramic composites, for particular applications, and set forth limitations on the use of the SiC-SiC composites at high temperatures. / Master of Science

The Fluorescence Scanning Thermal Microscope (FSTM v3.0), was designed to create an inexpensive, and easily manufactured, device for measuring the diffusivity of samples with microscopic locational precision. This was accomplished by using a Blu-ray device known as a PHR-803T, referred to in this work as a PHR. The optics in the PHR are nearly identical in function to conventional devices used in thermoreflectance microscopy, making the PHR extremely useful to integrate into the FSTM design. The focus of this thesis is the application of the FSTM as a confocal microscope using 3D printed components and various low-cost devices to operate with comparable sampling accuracy to existing confocal microscopes. The electronics and optical filters were then adapted to enable the measurement of thermal waves, particularly by detecting a linear relationship between phase delay and the spacing between heating and sensing lasers, as predicted by previous work on the FSTM.

Recent global efforts have focused on developing new lightweight alloys specifically designed for high-pressure die casting (HPDC) processes, aiming to achieve the lightweight of electrified vehicles. HPDC offers a distinct advantage by allowing the production of neat-net-shape automotive components, minimizing the need for further processing. An inherent characteristic of HPDC is its rapid cooling rates, making the understanding and characterization of the thermophysical properties of these newly developed lightweight alloys under high cooling rates a must. These properties have a significant effect on controlling the HPDC process and developing suitable filling and solidification models to simulate the HPDC process intricacies for commercial production adaptation. The thermophysical properties of these alloys are shown to exhibit considerable variability with temperature, particularly under rapid solidification conditions, like in HPDC. Hence, an essential step in developing such alloys is to thoroughly investigate the variation of their thermophysical properties with temperature under high cooling rates. To fulfill such a need, an experimental setup has been developed to allow the solidification of molten metal samples under varying cooling rates using a set of impinging water jets. An inverse heat transfer algorithm has been developed to estimate the thermal conductivity and thermal diffusivity as a function of the temperature of the solidifying samples under high cooling rates. To validate the accuracy of the inverse heat transfer algorithm and the experimental methodology, a set of experiments has been carried out using pure Tin, which is a well-characterized material. Its thermal diffusivity and thermal conductivity are readily available in the literature. The estimated thermal diffusivity and thermal conductivity of Tin have been compared with the published data. The estimated thermal diffusivity and conductivity of the solid phase were in good agreement with the published values. A maximum deviation ranging from +10.1% to -3.47% was observed in the estimated thermal diffusivity. The maximum deviation in the estimated thermal conductivity was between +7.8% and -13.6%. Higher deviations have been observed in the estimated thermal diffusivity and conductivity of the liquid phase with deviations in the range of +33.71% to -4.86% and +0.76% to 26.53%, respectively. The higher deviations observed for the liquid phase might be attributed due to the natural convection that developed in the tested liquid sample. The effect of natural convection was examined using a set of numerical simulations that confirmed the existence of a convection-induced movement within the liquid phase. A sensitivity analysis was carried out to examine the impact of the accuracy of thermocouple positions and the effect of temperature sensing accuracy on the estimated thermal properties. / Thesis / Master of Applied Science (MASc) / An inverse heat transfer algorithm along with an experimental setup has been developed to estimate the temperature-dependant thermophysical properties during solidification of metallic alloys under high cooling rates. To verify the accuracy of the developed algorithm and the experimental setup the estimated thermal conductivity and diffusivity of pure Tin have been compared with data available in the literature. The results showed reasonable agreement.

The purpose of this investigation was to develop an apparatus to measure the thermal conductivity of polymer melts, and to use the apparatus to measure the thermal conductivity of selected melts as a function of melt temperature. The steady-state, coaxial cylinder method with guard heaters was used and the annular gap was 0.075 inch. The polymer was melted in a cylindrical melt chamber, then metered to the thermal conductivity measuring apparatus. Cartridge heaters provided heat input and temperature measurements were made with calibrated, differential, iron-constantan thermocouples. The thermal conductivity of polyethylene, polystyrene and nylon melts tested increased with increased temperature. The thermal conductivity of the polypropylene sample was temperature independent. Complexity of molecular structure lowered melt thermal conductivity. Radiation losses were accounted for and convection was determined to be absent. The results were shown to be within a 3 percent experimental measurement error. Meaningful confidence limits cannot be calculated because of the limited number of data points. / Master of Science