This thesis reports a novel route to molecular self-assembly based on simple acid-base hydrolytic chemistry involving the reactions of aminosilanes with organic species containing acidic protons. From detailed solution chemistry of trimethyl-, trimethoxy-, and triphenyl-chloro/amino-silanes, two general processes have been developed for the self-assembly of long alkyl chain terminated alcohols, thiols, and carboxylic acids on inorganic oxide surfaces i.e., glass, quartz and single crystal silicon. The first method involves three successive reactions with the surface moieties where the initially formed silylchloride layer is converted to a silylamine, which is then reacted further to give the organic monolayers. The second method involves a single reaction of the surface with surfactant species formed first by reacting trimethoxysilylchloride with the appropriate chromophore. The chemistry and comparative advantages of these two methods are discussed. A discussion on the complete characterization of newly formed thin films by employing surface techniques, such as wettability, FTIR-ATR spectroscopy, ellipsometry, and X-ray reflectivity, is provided. These results indicate that thin films formed using simple acid-base hydrolytic chemistry are comparable to those obtained from more established techniques, such as trichlorosilanes on glass, thiols on gold, and carboxylic acids on silver or alumina. The results presented in this thesis demonstrate that the acid-base hydrolytic chemistry is a viable and widely applicable method to molecular self-assembly and can be considered as a unifying approach to literature methods.

The morphology and properties of polymer blends are closely related to processing conditions. The minor phase dimensions usually range from 0.1 to 10 $ mu$m. A skin/core morphology, with a minor phase undergoing variations in orientation and aspect ratio from the surface to the core of the material is observed in processes such as the injection molding of blends. The use of optical inspection is a potential tool to control the stability and the quality of the product on-line. In this thesis, polymer composites made from different glass fibers and glass microspheres embedded in a matrix of PS are used to simulate two-phase polymer blend morphology with a skin/core configuration. Two apparatuses have been used to characterize the morphology of the specimen by light scattering in reflection. The first had an incident light beam at 45 degrees from a normal to the surface of the sample. The second apparatus projected the laser beam incident perpendicularly on the surface of the sample and used polarized light. Distinctive reflected light scattering patterns were obtained in both cases and analyzed. In the first case, ratios of the axes of the ellipsoidal iso-intensity curves were plotted as a function of the position from the center of the light beam. The anisotropy of the iso-intensity lines is also plotted as a function of the position of the center of the beam for the second case. In both situations, information is obtained from the change of the anisotropy ratio as function of the distance from the center of the beam for depth-varying morphology samples. The skin, and the core content as well as the skin thickness effect can be isolated from the processed images. The second technique gave more information and patterns which were easier to process. / A Monte Carlo program was written to simulate the propagation of photons in the samples containing a spherical morphology. The simulation is a potential tool for the analysis or two-phase systems but requires further development to be a useful tool.

The shear behaviour of unidirectional AS4/3501-6 graphite/epoxy material is fully characterized for both the in-plane and the out-of-plane shear. A comprehensive study of testing methods for in-plane shear is performed and the three-rail shear test is chosen. A two-dimensional finite element analysis is performed in order to analyze and improve the performance of the three-rail shear test specimen. Through finite element analysis, the optimal specimen configuration is determined by inserting slots at the positions where there are stress concentrations. The results are verified by three-rail shear tests performed for both the standard un-notched and the new notched specimens. The notched specimens show great improvement in both static strength and fatigue life. Using a modified three-rail testing fixture and the modified notched samples, a characterization of graphite/epoxy material under in-plane shear static and fatigue loading was performed. The in-plane shear strength and stiffness were determined. The shear stiffness demonstrated significant non-linearity which was modeled and characterized. The in-plane shear fatigue life, residual strength and residual stiffness were determined and are presented in normalized forms which can be used to predict the fatigue behaviour of the material. Next the out-of-plane shear test methods are examined. Existing shear test methods are designed primarily for measuring in-plane shear strength in the x-y plane (S$ rm sb{xy}).$ An improved double-notched shear test method is developed to measure out-of-plane shear strength, S$ rm sb{yz},$ under both static and fatigue loading conditions. Out-of-plane shear strength can be measured by careful choice of specimen configuration and testing method, verified by finite element analysis. The test method is then used to determine the out-of-plane shear strength, the fatigue life and the residual strength. The behaviour of AS4/3501-6 graphite/epoxy composite under static and fatigue shear

The influence of high-strength concrete on the seismic response of reinforced concrete exterior beam-column-slab subassemblages was investigated by the testing of two full-scale specimens. The specimens, one constructed with normal-strength concrete and the other constructed with high-strength concrete, were designed as ductile moment-resisting frames such that the flexural strengths of the beams were similar. The increase in concrete strength allowed for some reduction of member sizes. / The use of high-strength concrete for seismic design is restricted by the Canadian Standard, CSA A23.3-94, to members with a concrete compressive strength, f$ sbsp{ rm c}{ prime}$ less than or equal to 55 MPa. The applicability of the current design provisions to higher strength concrete exterior beam-column joints was investigated. Both specimens were instrumented to allow for detailed strain, load and deflection measurements to be monitored at critical locations during the testing process. The strain readings and crack patterns allowed a study of the flow of forces into the joint region. The measured strains on the slab bars permitted a determination of the amount of slab reinforcement which was contributing to the negative flexural capacity of the beam. This is of particular importance because the capacity of the beams will affect the hierarchy of yielding within a structure and may in turn give an undesirable sequence of yielding. The "effective width" specified in the current codes of practice is also examined.

The purpose of this study was to characterize metallurgical and tribological events occurring at the articulating surfaces of all metal implants in order to gain understanding of the wear characteristics of Co-Cr-Mo alloys. The surfaces of fifteen implant heads (or balls), made of either cast, low carbon wrought or high carbon wrought Co-Cr-Mo material, were examined using scanning electron and atomic force microscopes. In the fast part of the study, six of the implants were examined prior to simulator testing, three after 3 million cycles of testing at 3 times body weight, and six after 6 million cycles of testing, also at 3 times body weight. In the second part of the study, the six implants that were examined prior to testing in the first part, were progressively wear tested at 5 times body weight and were periodically examined until 2 million cycles of testing. (Abstract shortened by UMI.)

Polymer dispersed liquid crystals (PDLC's) are novel composite materials consisting of micron-size liquid crystalline droplets dispersed uniformly in a solid polymer matrix. PDLC's are formed by spinodal decomposition induced by thermal quenching or polymerization. These materials have excellent magneto-optical properties, and have great potential in applications that require efficient light scattering. Present commercial applications include switchable windows for privacy control and large-scale billboards. The optical properties depend on the droplet size, shape and positional order, which are determined during the formation stage, and reorientation dynamics of the liquid crystalline molecules confined within the droplets which occurs during product use. In this thesis, new complex mathematical models that describe the formation and performance of PDLC's are successfully developed, implemented, solved and validated. The nonequilibrium thermodynamic formation model takes into account initial thermal fluctuations computed using Monte Carlo simulations and realistic arbitrary boundary conditions. The performance model is based on classical nematic liquid crystalline magneto-viscoelastic theories, and incorporates transient viscoelastic boundary conditions. The simulations are able to reproduce successfully all the experimentally observed significant dynamical and morphological features of film formation as well as all the dynamical stages observed during the use of these thin optical films. In addition, the sensitivity of the phase separating morphology to processing conditions and material parameters is elucidated. Furthermore, a new scaling method is introduced to describe the phase separation phenomena during the early and intermediate stages of spinodal decomposition induced by thermal quenching. The droplet size selection mechanism for the polymerization-induced phase separation method of forming PDLC films is identified and explained for the first time. Last

An experimental and theoretical study of the fundamentals of the vapour phase synthesis of ultrafine aluminum nitride (AIN) particles using thermal plasma was carried out. The study used the concept of a transferred-arc reactor which produces AlN ultrafine powders in two stages: evaporation of aluminum (Al) metal by the transferred-arc in non nitriding conditions; and the reaction between Al vapour and ammonia (NH3) in a separate tubular reactor. A new version of this reactor concept was built in which the transferred-arc chamber and tubular reactor were vertically aligned. This reactor design allowed the study of both radial and axial mixing of ammonia with the plasma chamber off-gas. Ultrafine powders with a specific surface area (SSA) in the range of 38--270 m2/g were produced in two plasma chamber off-gas temperature levels (1800 and 2000 K), with different quenching intensities, and two different plasma gas compositions (pure Ar and Ar/H2 mixture). The dependence of the particle size and composition on the reactor operating conditions was investigated. Depending on the plasma gas composition, two different trends were observed for the SSA as a function of quenching intensity, with the radial injection of NH3. A two-dimensional numerical model was developed for the nucleation and growth of ultrafine particles in this system, using the method of moments. A new equation for the nucleation rate for the AlN system was developed. This equation considers the effect of reaction on the surface of clusters of the new phase. This new modelling approach could explain the trends observed experimentally. The importance of the mechanisms for the gas-to-condensed phase transition in the AlN system were examined. The sinterability of the powder produced was examined. Sintering to full density was achieved at 1550°C. Because of the high oxygen content of the powder, a second phase identified as aluminum oxynitride (ALON) was observed to form.

A novel thermal plasma process for fumed silica production was investigated. The novelty was the result of the method used for silica decomposition which consisted of transferring an arc (Ar or Ar/H2) to a silica anode. The primary objective was to examine the effect of current (150--250 A), plasma gas flow rate (10--20 lpm Ar) and plasma gas composition (0--2.8% H2) on the rate of silica decomposition. The decomposition rate (0.09--1.8 g/min) was determined to be a heat transfer limited process which occurred below the arc root where the surface had attained its boiling point. The decomposition rate was not affected by plasma gas flow rate since convective heat transfer was reduced by the counter flow of decomposition products (SiO(g) and O2) from the anode surface. Increasing current increased the decomposition rate because the heat input to the anode due to electron flow and arc radiation was increased. Adding H2 to the plasma gas increased the decomposition rate because of an increase in radiative heat transfer to the anode, a reduction in the theoretical energy requirement for decomposition and a consumption of O2 which lowered the boiling point. A mathematical model was developed to simulate the behavior of the silica anode. The predicted decomposition rates were of the same order of magnitude as those observed experimentally. A secondary objective was to examine the effect of oxidation zone parameters including SiO(g) concentration (0.09--3.24%) and temperature (1414--1801 K) of the reactor exit gas and steam flow rate (46--71 g/min) on powder properties. The oxidation zone was where SiO(g) was quenched with steam to generate fumed silica. The specific surface area of the powders (104--257 m2/g) increased with increasing steam flow rate due to the higher quench rates. The high H2O (g) concentration in the oxidation zone produced surface hydroxyl densities which approached saturation (7.85 Si-OH/nm2). The powders had typical characteristics associated with

A Mo, a Nb and a Ti steel were tested in torsion to study the characteristics of dynamic and postdynamic recrystallization. To characterize dynamic recrystallization, continuous torsion tests were carried out between 850 and 1050$ sp circ$C at strain rates of 0.02, 0.2 and 2s$ sp{-1}.$ Quenches were performed to investigate the grain refinement produced by dynamic recrystallization. Interrupted torsion tests were performed between 850 and 1050$ sp circ$C and at strain rates between 0.02 and 2s$ sp{-1}$ to study the characteristics of postdynamic recrystallization. Quenches were performed after increasing holding times to follow the evolution of the postdynamic microstructure. The evolution of the grain size distribution as a function of holding time shows that the growth of dynamically recrystallized grains is the first change that takes place. Then metadynamically recrystallized grains appear and contribute to the softening of the material. The rate of metadynamic recrystallization increases with strain rate and temperature and is observed to be independent of strain, in contrast to the observations for static recrystallization. The dependence of the metadynamically recrystallized grain size on the Zener-Hollomon parameter was established and is shown to differ from static recrystallization dependence. Simple torsion simulations were carried out using constant interpass times to study the conditions under which dynamic, metadynamic or static recrystallization takes place. Dynamic recrystallization controlled rolling (DRCR) is shown to require such short interpass times that they are not attainable in hot strip mills. A new concept, metadynamic recrystallization controlled rolling (MDRCR), is introduced to describe the case where there is 20 to 80% softening by metadynamic recrystallization. The occurrence of dynamic and metadynamic recrystallization causes the load to increase less rapidly than in the case of pure strain accumulation.

Experiments involving single pass laboratory rolling and isothermal salt bath annealing were carried out; three steels were studied: a titanium stabilized interstitial free grade and two low carbon grades, one of which contained a particularly low level of manganese ($ sim$0.009wt.%). The two low carbon grades were produced such that any complication from AlN precipitation was avoided. X-ray, neutron diffraction, optical metallography and mechanical testing measurements were carried out on the samples before and after annealing. The main aim of this work was to further the understanding of the metallurgy of recrystallization after ferrite rolling at temperatures between room temperature and 700$ sp circ$C. / Deformation textures, recrystallization kinetics, final grain sizes and recrystallization textures were quantified for all the samples and experimental conditions. A major conclusion based on these data is that the influence of rolling temperature is far greater in the low carbon samples than in the IF grade. Indeed, the IF results alter only marginally with increasing temperature. In the low carbon grades, however, the rolling texture sharpens, recrystallization slows, the final grain size coarsens, and the recrystallization texture changes when the rolling temperature is increased. / This distinct difference between the two steel types is explained in terms of their contrasting deformation behaviors. Solute carbon and nitrogen in the low carbon grades interact with dislocations causing high stored energy levels after low temperature rolling (due to dynamic strain aging) and high strain rate sensitivities during high temperature rolling (due to the solute drag of dislocations in the transition region between DSA and DRC). Nucleation during subsequent recrystallization is strongly influenced by both the stored energy and the strain rate sensitivity. The latter affects the occurrence of the flow localisations that enhance nucleation.