Thesis (Ph.D.)--Hong Kong University of Science and Technology, 2006. / Includes bibliographical references (leaves 222-237). Also available in electronic version.

Thesis (Ph.D.)--University of California, Berkeley, 2002. / Chair: John W. Morris, Jr. Includes bibliographical references.

Th.--Sci. phys.--Besançon, 1977. N°: 119.

Thesis (Ph. D.)--Lehigh University, 2000. / Includes vita. Includes bibliographical references (leaves 103-111).

No description available.

Thesis (MTech (Mechanical Engineering))--Cape Peninsula University of Technology, 2009. / Materials in almost all components are subjected to some kind of loading that must be correctly predicted to produce reliable designs. The understanding of a material's properties significantly impacts appropriate selection for a structure. This kind of material characterization is also important in the development of improved or new materials for high strength and novel applications. There are numerous metallurgical variables (composition and process parameters) that influence the physical and mechanical properties of materials. The aim of this work has been to study the influence of microstructure on mechanical properties of steel, specifically the effect of grain sizes within solid phase mixtures. Parameters for simple models of the variation of material properties with grain size can be determined. These models can then be incorporated in the material data sets of Finite Element Analysis programs which will then allow for structural analysis with zones in a material having different grain sizes. The deformation and damage behaviour of EN 8 steel have been stUdied with emphasis on the effects of grain size on the elastic-plastic response of the material. For that purpose, EN 8 specimens with a range of microstructures (grain size and phase) were prepared by heat treatment The microstructural features were carefully characterized using two different experimental surface microscopy techniques; Light Optical Microscope and Scanning Electron Microscope. The deformation and hardness characteristics have been studied with the help of tensile and hardness tests. The mechanical properties were determined as a function of microstructure (grain size and phase). The yield stress followed the classical Hall-Petch relation. The results indicated that tensile strength and hardness increases with decrease in grain size while elongation decreases. The main philosophy behind this research has been the study of the microstructure and information from the iron-carbon phase diagram together with numerical analysis of stress-strain data, in order, to understand the influence of grain size on mechanical behaviour of EN8 steel. This combination was then used to make general conclusions on mechanical behaviour of EN 8 based on heat treatment history.

The use of optical interconnects is a promising solution to the increasing demand for high speed mass data transmission used in integrated circuits as well as device to device data transfer applications. For the purpose, low cost polymer waveguides are a popular choice for routing signal between devices due to their compatibility with printed circuit boards. In optical interconnect, coupling from an external light source to such waveguides is a critical step, thus a variety of couplers have been investigated such as grating based couplers [1,2], evanescent couplers [3], and embedded mirrors [4–6]. These couplers are inherently micro-optical components which require fast and scalable fabrication for mass production with optical quality surfaces/structures. Low NA laser direct writing has been used for fast fabrication of structures such as gratings and Fresnel lenses using a linear laser direct writing scheme, though the length scale of such structures are an order of magnitude larger than the spot size of the focused laser of the tool. Nonlinear writing techniques such as with 2-photon absorption offer increased write resolution which makes it possible to fabricate sub-wavelength structures as well as having a flexibility in feature shape. However it does not allow a high speed fabrication and in general are not scalable due to limitations of speed and area induced by the tool’s high NA optics. To overcome such limitations primarily imposed by NA, we propose a new micro-optic fabrication process which extends the capabilities of 1D, low NA, and thus fast and scalable, laser direct writing to fabricate a structure having a length scale close to the tool's spot size, for example, a mirror based and 45 degree optical coupler with optical surface quality. The newly developed process allows a high speed fabrication with a write speed of 2600 mm²/min by incorporating a mask based lithography method providing a blank structure which is critical to creating a 45 degree slope to form the coupler surface. In this method, instead of using an entire exposure in a pixelated manner, only a portion of the Gaussian profile is used, allowing a reduced surface roughness and better control of the surface shape than previously possible with this low NA beam. The surface figure of the mirror is well controlled below 0.04 waves in root-mean-square (RMS) at 1.55 μm wavelength, with mirror angle of 45±1 degrees. The coupling efficiency is evaluated using a set of polymer waveguides fabricated on the same substrate as the complete proof of concept device. Device insertion loss was measured using a custom built optical test station and a detailed loss analysis was completed to characterize the optical coupling efficiency of the mirror. Surface roughness and angle were also experimentally confirmed. This process opens up a pathway towards large volume fabrication of free-form and high aspect ratio optical components which have not yet pursued, along with well-defined optical structures on a single substrate. In this dissertation, in Chapter 1, we provide an overview of optical surface fabrication in conjunction with current state of the art on fabrication of free form surfaces in macro and microscopic length scale. The need for optical interconnects is introduced and fabrication methods of micro-optical couplers are reviewed in Chapter 2. In Chapter 3, the complete fabrication process of a mirror based coupler is presented including a custom alignment procedure. In Chapter 4, we provide the integration procedure of the optical couplers with waveguides. In Chapter 5, the alignment of two-lithographic methods is discussed. In Chapter 6, we provide the fabrication procedure used for the waveguides. In Chapter 7, the experimental evaluation and testing of the optical coupler is described. We present a custom test station used for angle verification and optical coupler efficiency measurement. In Chapter 8, a detailed loss analysis of the device is presented including suggestions for future reductions in loss. Conclusions and future work considerations are addressed in Chapter 9.

The effects of variations in alloying content on the microstructure and mechanical properties of high strength steel weld metals have been studied. Based on neural network modelling, weld metals were produced using shielded metal arc welding with nickel at 7 or 9 wt. %, manganese at 2 or 0.5 wt. % while carbon was varied between 0.03 and 0.11 wt. %. From mechanical testing, it was confirmed that a large gain in impact toughness could be achieved by reducing the manganese content. Carbon additions were found to increase strength with only a minor loss to impact toughness as predicted by the modelling. The highest yield strength (912 MPa) in combination with good impact toughness (over 60 J at -100 °C) was achieved with an alloying content of 7 wt. % nickel, 0.5 wt. % manganese and 0.11 wt. % carbon. Based on thermodynamic calculations and observed segregation behaviour it was concluded that the weld metals solidify as austenite. The microstructure was characterised using optical, transmission electron and high resolution scanning electron microscopy. At interdendritic regions mainly martensite was found. In dendrite core regions of the low carbon weld metals a mixture of upper bainite, lower bainite and a novel constituent - coalesced bainite - formed. Coalesced bainite was characterised by large bainitic ferrite grains with cementite precipitates and is believed to form when the bainite and martensite start temperatures are close to each other. Carbon additions were found to promote a more martensitic microstructure throughout the dendrites. Mechanical properties could be rationalised in terms of microstructural constituents and a constitutional diagram was constructed summarising microstructure as a function of manganese and nickel contents.

Use of load-bearing materials whose functionality arises from architectured microstructures, so called architectured materials, has been hindered by the challenge of connecting them. A solution in nature is found at the tendon enthesis, a tissue that connects tendon and bone, two vastly different natural architectured materials. The tendon enthesis provides stability and allows for mobility of a joint though effective transfer of muscle forces from tendon to bone, while exhibiting toughness across a wide range of loadings. Unfortunately, many painful and physically debilitating conditions occur at or near this interface when the enthesis architecture is compromised due to injury or degeneration. Surgical and natural repairs do not reconstitute the natural toughening mechanisms of the enthesis and often fail. Hence, understanding the architectural mechanisms by which healthy and pathologic tendon entheses achieve strength and toughness would inform the development of both biological and engineered attachments.Integrating biomechanical analyses, failure characterizations, numerical simulations, and novel imaging, this thesis presents architectural mechanisms of enthesis toughening in a mouse model. Imaging uncovered fibrous architecture within the enthesis, which controlled trade-offs between strength and toughness. Ex vivo enthesis failure modes exhibited nanoscale differences in damage, milliscale differences in fiber load-sharing, and macroscale differences in energy absorption that depended on structure, composition, and the nature of loading. The elastic and failure responses of the tendon enthesis also varied with the direction of loading. This variation was due to the fibrous nature of the tendon enthesis, with a clear role for bony anatomy and fiber recruitment in enthesis toughening behavior. In vivo, , the loss of toughening mechanisms at the enthesis due to pathologic loading was evaluated by either increased (i.e., overuse) loading via downhill treadmill running or decreased (i.e., underuse) loading via botulinum toxin A induced paralysis. These loading environments led to changes in the mineralization and architecture at the tendon enthesis. These micro-architectural adaptations compromised the trade-offs between strength and toughness; overuse loading prompted active reinforcement and stiffening of the underlying trabeculae, leading a maintenance of strength and a compromise in overall toughness, whereas underloading prompted active resorption of the underlying trabecular architecture, leading to a compromise in both strength and toughness. The mouse models of the tendon enthesis failure revealed a correlation between tendon enthesis architecture and injury prevention (i.e., toughening) mechanisms. To test this concept in a clinical setting, an injury classification system was developed for patellar tendinopathy and partial patellar tendon tears. This classification system identified the stages of tear progression and prognosis by tracking changes to patellar tendon architecture. Results revealed a relationship between patellar tendon thickness and likelihood of improvement with nonoperative treatment. Taken together, this dissertation revealed how fibrous architecture can be tailored to toughen attachments between vastly different materials. This understanding can have prognostic value: tracking changes to enthesis architecture can be used as a tool for identifying candidates for various treatment options, as we showed for the patellar tendon clinical example. Furthermore, the toughening mechanisms identified here provide guidance for enhancing enthesis surgical repair and designing enthesis tissue engineered scaffolds, as well as motivating biomimetic approaches for attachment of architectured engineering material systems.

PhD - Science / The relationships between densification and microstructure, and between microstructure and mechanical and electrical properties of liquid phase sintered silicon carbide were studied in detail using hot pressing, gas pressure sintering and ultra–high pressure sintering techniques. Silicon carbide was sintered with 10 mass-% addition of the Y2O3-Al2O3 system, with various molar ratios. Hot pressing was carried out at 1925oC under 30 MPa, in argon, for half an hour. Materials were gas pressure sintered at 1925oC, under a final gas pressure of 80 bars (8MPa), in argon, for an hour. Ultra-high pressure sintering was done at ca. 1550oC, under 5.5 GPa pressure, for 15 minutes. The hot pressed and gas pressure sintered materials were subsequently heat treated at 1925oC and 1975oC. Most of the silicon carbide materials were sintered to a density around 99% of theoretical density. The heat treatment of the hot pressed materials resulted in an increase in density not changing the porosity. The densities of the heat treated hot pressed materials corresponded to the density of the gas pressure sintered materials. This resulted from the difference in composition of grain boundary phases – yttrium silicates in the hot pressed materials and yttrium aluminates in the gas pressure sintered and heat treated materials. The average silicon carbide grain size in the materials strongly depended on the densification method. In gas pressure sintered and heat treated materials the mean grain size was up to three times higher than that in the hot pressed materials. Grain growth appeared to be higher in the highest alumina-content materials. The heat treatment at 1975 °C resulted in more pronounced anisotropic grain growth. The ratio of the silicon carbide polytypes of sintered materials and materials heat treated materials at 1925oC, did not change significantly. In the materials heat treated at 1975oC Rietveld analysis revealed a decrease in SiC-6H polytype and an increase in amount of 4H and 15R polytypes, compared to the materials heat treated at 1925oC. This can be attributed to the increase in diffusion rates of aluminium into the SiC lattice at 1975oC. Segregation patterns were observed in the high yttria content materials, with Y2O3:Al2O3 molar ratios greater than or equal to two, after gas pressure sintering and heat treatments. This was suggested to be due to he poor wetting of the silicon carbide grains by the yttria-rich grain boundary phase. On heat treatment, the Vickers hardness of hot pressed materials was found to be increased from 20 to 26 GPa and elastic modulus from 318 to 338 GPa. In addition, the log of the electrical conductivity of liquid phase sintered silicon carbide (measured at 330oC) ranged from 10-8 to 10-3 with the changes in grain boundary phases observed after the heat treatments. The grain boundary phase composition also influenced the strength of the materials, The highest strength, 657 + 50 MPa, was measured for the hot pressed material containing the YAG phase. Indentation fracture toughness was mostly influenced by the SiC grain growth during heat treatments. The most significant increase in fracture toughness, the largest being from 3.7 MPa.m1/2 up to 5.6 MPa.m1/2, was observed in the higher alumina content materials after heat treatment at 1975oC. The increase in fracture toughness was attributed to the presence of a higher amount of platelet-like SiC grains within a broader grain size distribution. These elongated grains increased fracture toughness by increasing crack path deflection and crack bridging. The electrical properties were evaluated by Impedance Spectroscopy measurements between room temperature and 330oC. The LPS SiC materials can be classified into three groups with different electrical properties. This classification could be related to the grain boundary phases present in the materials. The materials with the lowest conductivity were all hot pressed materials, containing crystalline silicates and amorphous grain boundary phases. The materials with intermediate conductivity include gas pressure sintered materials and a hot pressed material, which contained crystalline aluminates (Y3Al5O12, YAlO3 and Y4Al2O9) in their grain boundaries. The materials with the highest conductivity only contained the aluminates, YAlO3 and Y4Al2O9. A pseudopercolation model of conduction was proposed, in which electrons move along a path which goes through the thinner intergranular layers, with largest nearest neighbour contact. The temperature dependence of the log of the conductivity of hot pressed and gas pressure sintered materials showed that the conduction mechanism in these liquid-phase sintered silicon carbide materials was variable range hopping conduction of electrons between defect sites. The non-Arrhenius behaviour, together with the observed wide range of peak frequencies, led to the conclusion that the effect of silicon carbide itself was not observed in the impedance spectra. The 1/T0.25 log conductivity dependence showed that the Cole-Cole arcs are due to insulating grain boundary phases rather than semiconducting SiC. In the Cole-Cole plots of the hot pressed and heat treated hot pressed materials only the effect of one phase could be observed. In the gas pressure sintered materials and the hot pressed material containing mainly YAG phase, the effects of two phases were seen in the frequency range measured. Ultra-high pressure liquid-phase sintered silicon carbide materials showed ultra-fine SiC grains, which were highly inter-grown. Segregated grain boundary “core-rim” structures, consisting of an inner core of nonequilibrium yttria and outer rim of equilibrium yttrium silicate were observed in materials containing 4 mass-% to 15 mass-% sintering additives. The hardness of ultra-high pressure sintered 10 mass-% materials increased with alumina-content, from 20 GPa – 22 GPa, and increased with decrease in sintering additive, up to 23 GPa (for the 4 mass-% material).