Thesis (Ph. D.)--University of Hong Kong, 1993.

Thesis (M.S.)--Rutgers University, 2007. / "Graduate Program in Ceramic and Materials Science and Engineering." Includes bibliographical references (p. 49-50).

Thesis (Ph. D.)--University of Washington, 2005. / Vita. Includes bibliographical references (leaves 129-133).

Kyoto University (京都大学) / 0048 / 新制・課程博士 / 博士(工学) / 甲第14589号 / 工博第3057号 / 新制||工||1455(附属図書館) / 26941 / UT51-2009-D301 / 京都大学大学院工学研究科化学工学専攻 / (主査)教授 山本 量一, 教授 宮原 稔, 教授 大嶋 正裕 / 学位規則第4条第1項該当

This research describes a two-dimensional modelling effort of heat and mass transport in simplified intrusive models of sills and their feeder dykes. These simplified models resembled a complex intrusive system such as the Great Dyke of Zimbabwe. This study investigated the impact of variable geometry to transport processes in two ways. First the time evolution of heat and mass transport during cooling was investigated. Then emphasis was placed on the application of convective scavenging as a mechanism that leads to the formation of minerals of economic interest, in particular the Platinum Group Elements (PGEs). The Navier-Stokes equations employed generated regions of high shear within the magma where we expected enhanced collisions between the immiscible sulphide liquid particles and PGEs. These collisions scavenge PGEs from the primary melt, aggregate and concentrate it to form PGEs enrichment in zero shear zones. The PGEs scavenge; concentrate and 'glue' in zero shear zones in the early history of convection because of viscosity and dispersive pressure (Bagnold effect). The effect of increasing the geometry size enhances scavenging, creates bigger zero shear zones with dilute concentrate of PGEs but you get high shear near the roots of the dyke/sill where the concentration will not be dilute. The time evolution calculations show that increasing the size of the magma chamber results in stronger initial convection currents for large magma models than for small ones. However, convection takes, approximately the same time to cease for both models. The research concludes that the time evolution for convective heat transfer is dependent on the viscosity rather than on geometry size. However, conductive heat transfer to the e-folding temperature was almost six times as long for the large model (M4) than the small one (M2). Variable viscosity as a physical property was applied to models 2 and 4 only. Video animations that simulate the cooling process for these models are enclosed in a CD at the back of this thesis. These simulations provide information with regard to the emplacement history and distribution of PGEs ore bodies. This will assist the reserve estimation and the location of economic minerals.

No description available.

Elastic effects on the hydrodynamic instability of inviscid parallel shear flows are investigated through a linear stability analysis. We focus on the upper convected Maxwell model in the limit of infinite Weissenberg and Reynolds numbers. Specifically, we study the effects of elasticity on the instability of a few classes of simple parallel flows, specifically plane Poiseuille and Couette flows, the hyperbolic-tangent shear layer and the Bickley jet. The equation for stability is derived and solved numerically using the Chebyshev collocation spectral method. This algorithm is computationally efficient and accurate in reproducing the eigenvalues. We consider flows bounded by walls as well as flows bounded by free surfaces. In the inviscid, nonelastic case all the flows we study are unstable for free surfaces. In the case of wall bounded flow, there are instabilities in the shear layer and Bickley jet flows. In all cases, the effect of elasticity is to reduce and ultimately suppress the inviscid instability. The numerical solutions are compared with the analysis of the long wave limit and excellent agreement is shown between the analytical and the numerical solutions. We found flows which are long wave stable, but nevertheless unstable to wave numbers in a certain finite range. While elasticity is ultimately stabilizing, this effect is not monotone; there are instances where a small amount of elasticity actually destabilizes the flow. The linear stability in the short wave limit of shear flows bounded by two parallel free surfaces is investigated. Unlike the plane Couette flow which has no short wave instability, we show that plane Poiseuille flow has two unstable eigenmodes localized near the free surfaces which can be combined into an even and an odd eigenfunctions. The derivation of the asymptotics of these modes shows that our numerical eigenvalues are in agreement with the analytic formula and that the difference between the two eigenvalues tends to zero exponentially with the wavenumber. / Ph. D.

The phenomenon of transition in a laminar flow has been a topic of continued interest for many years. Recent experiments in shear flows have revealed a series of instabilities that lead to breakdown to turbulence. We have completed an analysis of the mechanisms which drive the primary (TS wave) and secondary instabilities in plane Poiseuille flow. This was accomplished by studying the solutions of linear primary and secondary stability theory with energy methods. We found that primary instability occurred when the viscous stresses overpowered dissipative forces near the channel walls. For the secondary instability, we saw that the TS wave catalyzes the instability and then mediates the transfer of brge amounts of energy from the mean flow into the three-dimensional disturbance, thus driving the instability. In addition, we have compiled an extensive catalog of the loc!l.l energy and vorticity field distributions which result from each instability. / Master of Science

The design of fiber-reinforced thermoplastic (FRT) parts is hindered by the determination of the various empirical parameters associated with the fiber orientation models. A method for obtaining these parameters independent of processing doesn't exist. The work presented here continues efforts to develop a rheological test that can obtain robust orientation model parameters, either by fitting directly to orientation data or by fitting to stress-growth data. First, orientation evolution in a 10 wt% long-glass-fiber-reinforced polypropylene during two homogeneous flows (startup of shear and planar extension) was compared. This comparison had not been performed in the literature previously, and revealed that fiber orientation is significantly faster during planar extension. This contradicts a long-held assumption in the field that orientation dynamics were independent of the type of flow. In other words, shear and extension were assumed to have equal influence on the orientation dynamics. A non-lubricated squeeze flow test was subsequently implemented on 30 wt% short-glass-fiber-reinforced polypropylene. An analytical solution was developed for the Newtonian case along the lateral centerline of the sample to demonstrate that the flow is indeed a superposition of shear and extension. Furthermore, an existing fiber orientation model was fit to the gap-wise orientation profile, demonstrating that NLSF can, in principle, be used to obtain fiber orientation model parameters. Finally, model parameters obtained for the same FRT by fitting to orientation data from startup of steady shear are shown to be inadequate in predicting the gap-wise orientation profile from NLSF. This work is rounded out with a comparison of the fiber orientation dynamics during startup of shear and non-lubricated squeeze flow using a long-fiber-reinforced polypropylene. Three fiber concentrations (30, 40, and 50 wt%) were used to gauge the influence of fiber concentration on the orientation dynamics. The results suggest that the initial fiber orientation state (initially perpendicular to the flow direction and in the plane parallel to the sample thickness) and the fiber concentration interact to slow down the fiber orientation dynamics during startup of shear when compared to the dynamics starting from a planar random initial state, particularly for the 40 and 50 wt% samples. However, the orientation dynamics during non-lubricated squeeze flow for the same material and initial orientation state were not influenced by fiber concentration. Existing orientation models do not account for the initial-state-dependence and concentration-dependence in a rigorous way. Instead, different fitting parameters must be used for different initial states and concentrations, which suggests that the orientation models do not accurately capture the underlying physics of fiber orientation in FRTs. / Ph. D. / In order to keep pace with government fuel economy legislation, the automotive and aerospace industries have adopted a strategy they call “lightweighting”. This refers to decreasing the overall weight of a car, truck, or plane by replacing dense materials with less-dense substitutes. For example, a steel engine bracket in a car could be replaced with a high-temperature plastic reinforced with carbon fiber. This composite material will be lighter in weight than the comparable steel component, but maintains its structural integrity. Thermoplastics reinforced with some kind of fiber, typically carbon or glass, have proven to be extremely useful in meeting the demands of lightweighting. Thermoplastics are materials that can be melted from a feedstock (typically pellets), reshaped in the melted state through use of a mold, and then cooled to a solid state, and some common commodity-grade thermoplastics include polypropylene (used for Ziploc bags) and polyamides (commonly called Nylon and used in clothing). Although these commodity applications are not known for their strength, the fiber reinforcement in the automotive applications significantly improves the structural integrity of the thermoplastics. The ability to melt and reshape thermoplastics make them incredibly useful for highthroughput processes such as injection molding. Injection molding takes the pellets and conveys them through a heated barrel using a rotating screw. The melted thermoplastic gathers at the tip of the barrel, and when a set volume is gathered, the screw is rammed forward to inject the thermoplastic into a closed mold of the desired shape. This process typically takes between 30-60 seconds per injection. This rate of production is crucial for the automotive industry, as manufacturers need to put out thousands of parts in a short period of time. The improvement to mechanical properties of the thermoplastics is strongly influenced by the orientation of the reinforcing fibers. Although design equations connecting the part’s mechanical properties to the orientation of the fibers do exist, they require knowledge of the orientation of the fibers throughout the part. Fibers in injection-molded parts have an extremely complicated orientation v state. Measuring the orientation state at each point would be too laborious, so empirical models tying the flow of the thermoplastic through the mold to the evolving orientation state of the fibers have been developed to predict the orientation state in the final part. These predictions can be used in lieu of direct measurements in the part design equations. However, the orientation models rely on empirical fitting parameters which must be obtained before injection molding simulations are performed. There is currently no standard test for obtaining these parameters, nor is there a standardized look-up table. The work presented in this dissertation continues efforts to establish such a test using simple flows in a laboratory setting, independent of injection molding. Previous work focused exclusively on using shearing flow (e.g. pressure-driven flow found in injection molding) to obtain these parameters. However, when these parameters were used in simulations of injection molding, the agreement between measured and predicted fiber orientation was mediocre. The work here demonstrates that another type of flow, namely extensional flow, must also be considered, as it has a non-negligible influence on fiber orientation. this is crucial to injection molding, as injection molding flows have elements of both shearing and extensional flow. The first major contribution from this dissertation demonstrates that extensional flow (e.g. stretching a film) has a much stronger influence than shearing flow, even at the same overall rate of deformation. The second major contribution used a combination shear/extensional flow to demonstrate that the empirical model parameters, thought to be characteristic of the composite, are actually strongly influenced by the type of flow experienced by the sample, and that no single set of model parameters can fit the full orientation state. The final major contribution extends the previous case to long-fiber reinforcement at multiple fiber concentrations which are of industrial interest. This finds the same results, that the model parameters are dependent on the type of flow experienced by the sample. The flow-dependence of the parameters is a crucial point to address in future work, as the flows found in injection molding contain both shearing and extensional flow. By further developing this flow-type dependence, future injection molding simulations should become more accurate, and this will make computer-aided injection-molded part design much more efficient.

The lift and drag forces exerting on a sphere immersed in a shear flow above a flat channel bed are evaluated by solving the steady three-dimensional Navier-Stokes equations. The numerical technique which combines the Newton iteration method and the finite element method is used to solve the non-linear Navier-Stokes equations. The technique first linearizes the non-linear terms in the partial differential equations, then solves the linearized equations by the finite element method. The Newton iteration method is used to linearize the non-linear equations. Since the iteration method requires a good initial guess, the linear solution of the partial differential equations is used for the initial guess, where the linear solution is the obtained by solving the differential equations without non-linear terms. The computer model developed can evaluate the lift coefficients of a sphere stationed at various distance from the channel bed. The computational results agree very well with the experimental measurements cited in the literature. The lift coefficient of the sphere changes with the undisturbed approaching velocity profile as well as the gap ratio which is the ratio of the distance between the sphere and the channel bed and the diameter of sphere. For fixed gap ratios, higher Reynolds number gives smaller lift coefficient than that of the lower Reynolds number. On the other hand, the lift coefficient also changes with the diameter of sphere for each fixed gap ratio. For small gap ratios, the lift coefficient increases as the diameter of sphere increases. For large gap ratios, the lift coefficient increases in the negative (downward) direction as the diameter of sphere increases. / Ph. D.