The effects of liquids on intrinsic interfacial sliding have been studied in NaCl crystals. The primary liquids in the study were water, methanol, and mixtures thereof. Sliding experiments were performed using a simple geometry in which a shear and normal compressive component of force were exerted on the interface. The geometry consisted of two single crystals joined at a boundary whose normal was inclined at an angle, $\theta$, to an axis along which a compressive load, P, was applied. The specimens were found to deform in two distinct ways: (1) by sliding along the interface, and (2) by indenting into one another in a direction normal to the interface. The introduction of liquids into the interface through channel-like defects was found to increase both the rate of sliding and indentation, with the increases being much greater for liquids with high water contents. It was found that the overall rate of displacement along the axis of the specimen was effectively independent of P but increased in roughly a linear fashion with $\theta$. A model for the process is developed in which displacement is produced primarily by interfacial sliding, with the liquid acting to promote the rate by undercutting the boundary and reducing the effective area of contact. The area of contact is determined by adaptations of friction theory, which lead to the observed P and $\theta$ dependence of the displacement rate. In addition, results of other experiments are presented which describe how grain boundaries in NaCl and KCl bicrystals are penetrated by water and methanol. Water is found to penetrate at much greater rates. This is discussed in terms of the differences in wetting and solubility exhibited by the two liquids. Both the sliding and penetration experiments are important in the understanding of liquid enhanced creep.

The objective of this work was to select and process an electrically insulating thin film suitable for the photoelectrodes used in solar to chemical energy conversion where the current flow would occur by quantum tunneling. A number of aluminum oxide coatings with various processing parameters were deposited onto silicon to explore the relationships between the deposition variables and the performance of the electrodes in 1 N H$\sb2$SO$\sb4$ aqueous solution subjected to visible light. The oxide thin films deposited by ultra high vacuum (UHV) reactive evaporation technique possessed 70-90% transmittance in UV-visible light range. Their optical transmittance ($\tau$) and the open-circuit photopotentials ($V\sb{oc}$) were found to depend on the deposition variables. The film growth on single crystal Si occurred from initially oriented islands and clusters, to a coalesced amorphous film as the atomic coverage increased. The surface compositions and depth profiles showed clean and uniform films. It was found that the oxide films can improve the open-circuit photovoltage of silicon and also act as anti-reflection (AR) coatings on Si. (Abstract shortened by UMI.)

This work uses three-dimensional Finite Element Analysis (FEA) to investigate the effect of geometric arrangement of particulate reinforcement in highly filled polymer composites (such as polymer concrete) on the setting stresses that develop in these materials during cure due to resin shrinkage during polymerization. These composites were initially modeled by systems reinforced with spherical particles packed in simple cubic (SC) and face-centered cubic (FCC) arrangements within the polymer matrix. A pronounced decrease in setting stresses was observed in the FCC system, which has a greater aggregate to resin ratio and more of resin domains per unit cell. A hexagonal-close-packed arrangement of hexagonal, prism-shaped aggregate was also analyzed and found to develop higher stresses, indicating that aggregate shape has an effect on setting stresses. A second set of models investigated the effect of size gradation and geometric arrangement of spherical reinforcing particles on setting stresses. The maximum stresses occur at the particle-resin interface, underlining the importance of resin/aggregate adhesion. Reduction of setting stresses by a factor of two was observed in systems with efficient packing, achieved with proper size gradation and close-packed geometry. A microstructural model for a polymer composite system based on a fairly random arrangement (FRA) of aggregate particles was also developed. This model gives a realistic representation of actual particle reinforced polymer composites. FEA results were used to develop an empirical equation for maximum setting stresses for Particle reinforced polymer composites. A probabilistic model for the distribution of voids in polymer composites was developed by solving a non-linear constrained optimization problem. The probability distributions of voids was used with a specially developed algorithm to generate the voids distributions in specific composites. The effect of voids on setting stresses in FRA models was discussed. In polymer composites voids tend to act as stress relief. This effect is more pronounced in poorly packed systems. This study provides an understanding of setting stress distribution in polymer composites. This work provides guidelines for optimizing the amount, shape and particle size distribution of the reinforcing aggregate in polymer composites so as to minimize setting stresses, thus leading to composites with significantly enhanced strength.

The inherent high strength, thermal conductivity, and electrical conductivity make nanotubes attractive reinforcements for polymer matrix composites. However, the structure that makes them desirable also causes highly anisotropic properties and limited reactivity with other materials. This thesis isolates these problems in two separate studies aimed at improving mechanical properties with single wall nanotube (SWNT) reinforced thermoplastic polymer composites. The two studies demonstrate the effect of solid freeform fabrication (SFF) and chemical functionalization on anisotropy and limited reactivity, respectively. Both studies showed mechanical property improvements. The alignment study demonstrates a maximum increase of 93% in tensile modulus with single wall nanotubes (SWNTs). The chemical functionalization study shows a larger increase in storage modulus for functionalized SWNTs as compared to purified SVWNTs with respective increases of 9% and 44% in storage modulus. Improved interfacial properties are also observed as a decrease in mechanical damping. Maximum property increases in composites are obtained when nanotubes are aligned, requiring additional processing consideration to the anisotropic structure. Melt spinning and extrusion processing effectively align nanotubes, but the end product of these techniques, composite fibers, requires further processing to be incorporated into finished parts. Extrusion-based SFF is a novel technique for processing nanotube reinforced composites because it allows for the direct fabrication of finished parts containing aligned nanotubes. SFF processing produces parts containing preferentially oriented nanotubes with improved mechanical properties when compared to isotropic composites. Functionalization of the nanotube surface disrupts the rope structure to obtain smaller ropes and promote further interfacial bonding. The chemically inert nature of nanotubes resulting from a structure containing few defects and the formation of larger, ordered ropes of SWNTs limits the amount of interfacial bonding and load transfer that occurs between nanotubes and a polymer matrix. Improved dispersion, interfacial properties, and mechanical properties are achieved through chemical functionalization. Subsequent partial removal of the functional groups created a direct bond between the nanotubes and the polymer matrix. The alignment and functionalization studies in this thesis further the knowledge of the use of nanotubes as reinforcements in polymer composites through understanding the sensitivity of the nanotubes' anisotropic properties and the nanotube/polymer interface.

A novel technique is developed to measure quantitatively the adhesion strength of metallizations deposited on substrates such as silicon. Electrostatic adhesion testing employs electrostatic forces to generate delaminating stresses in thin metallic films. The interfacial adhesion strength is readily calculated from the electrode geometry and the applied electrostatic field at failure. Unlike other adhesion tests, this technique does not require any mechanical contact and is virtually independent of the plastic deformation of the film. Furthermore. this test provides direct strength measurements as opposed to work or energy of adhesion measurements obtained by the common peel-test. The adhesion strengths of several metallizations (Cu, Al, Al-Cu alloy, and TiN) are characterized using this electrostatic technique. The distribution of stress-at-failure data follows Weibull statistics. Field emission scanning electron microscopy reveals that films are delaminated in a micro-blister-type mode. Annealing of metallizations causes reactions and changes flaw distributions. The presence of brittle compounds near the interface may create easy fracture paths and can act as stress concentrators to initiate and propagate the fracture. These stressed areas may lead to localized adhesion failure under applied stress. It is shown that electrostatic adhesion testing is effective in providing quantitative values for the adhesion strengths and failure probabilities of thin-film metallizations.

This thesis studies metal-hydrogen systems. The interaction between hydrogen-atoms and vacancies in metals have been elucidated in Fermi-Dirac statistics. Calculations have been presented and compared for specific models in which H-atoms act both as simple interstitial species and form either decorated vacancies or substitutional defects. A model has been presented to explain the superabundant vacancy formation under high hydrogen pressures. The solutions based on these models apply to much lower temperatures and higher concentrations than the traditional ones. These results show abundant vacancies will be formed in the presence of hydrogen; the vacancy concentration is many orders of magnitude larger than those in the H-free lattice. A study of the diffusion of hydrogen in the crack tip area has been provided. The slow diffusivity of H-atoms at low temperatures and the interaction between H-atoms and vacancies in the crack tip plastic zone give an explanation of the experimental data which show a maximum crack growth rate at room temperature. This work is associated with the embrittlement of steel by hydrogen.

An investigation is conducted into the reduction of thermal conductivity in a porous ceramic through the inclusion of small amounts (1--3wt.%) of dispersed nanotube bundles and nanofibers. The thermal properties and thermal stability of single walled carbon nanotubes (SWNT) are examined at high temperature, and their impact on the micro and nanostructure of the ceramic matrix is evaluated. An isotropic dispersion of SWNT bundles between small (∼100nm) ceramic particles results in thermal conductivity reductions by altering the nature of the porosity, contributing to an extended grain boundary region, and by serving as the basis for polycrystalline templates. Vapor grown carbon fibers (VGCF)---an order of magnitude larger in diameter than SWNT bundles---are used in low concentrations within the matrix as a comparison. Partially stabilized zirconia (PSZ) is selected as the matrix due to its extensive use in the gas turbine industry in thermal barrier coatings, and a tape casting technique is used to produce samples of varying compositions for the measurement of thermal diffusivity and calculation of thermal conductivity. Significant total thermal conductivity reductions (40--50%) are achieved in the PSZ composite and the results are compared with conventional models for porosity to evaluate the nature of the contributions.

In this work, we present a new computational method based on a nonconforming domain decomposition technique for modeling of phase transitions. Phase transitions are the result of thermal or mechanical loading in ferromagnetic materials or shape memory materials. Modeling of phase transitions is important because it can help to predict or control the behavior of these materials. This thesis will focus on phase transitions characterized by two directions of magnetization in the case of ferromagnets and two variants of Martensite in the case of shape memory materials. In both types of materials, branching occurs near an internal surface which is characterized by complex microstructures. These microstructures occur at a minimum energy state. The new computational method simulates the branching behavior of these microstructures near an internal surface. We approximate the microstructures via energy minimization. We minimize the total stored energy stored in vicinity of internal surface with the minimizing function representing the microstructures. We compare the numerical results obtained by the new technique with those obtained by a more standard technique, one not incorporating nonconforming domain decomposition. Furthermore, we verify the various energy scaling laws used to predict the total stored energy near an internal surface. Among these laws, we verify the local-in-y scaling property which has been conjectured but not proven.

Colloidal processing methods were developed in order to disperse highly concentrated 1.0, 2.0, and 6.0 vol% single-walled carbon nantoube (SWNT)-Si 3N4 aqueous composite suspensions. Interparticle pair potentials were developed between individual Si3N4 particles and SWNT bundles by coating them with cationic surfactant molecules of cetyltrimethylammonium bromide (CTAB). Zeta potential, viscosity, and sedimentation measurements were conducted on SWNTs and Si3N4 particle suspensions in order to optimize the pH and amount of adsorbed CTAB. The composite suspension viscosity was pH sensitive and adjusted accordingly before consolidation into three-dimensional solid parts using a rapid prototyping fabrication method called robocasting. High-density composites were produced using spark plasma sintering and structurally intact SWNTs were directly observed in the final sintered microstructure using scanning electron microscopy and Raman spectroscopy. When processed with SWNTs the highly insulative ceramic became electrically conductive and resulted in increased grindability for the otherwise hard to machine ceramic. The high hardness, fracture toughness and density of Si 3N4 was maintained for the composite due to the detailed development of colloidal processing and sintering methods used during fabrication. In addition, the thermal conductivity of the ceramic was reduced with the incorporation of well-dispersed SWNTs. Indentation load studies on the composites revealed sub-surface chipping and deformation around the indent before radial crack development indicating a degree of damage tolerance over the monolith. Along the wake of the crack SWNTs were also observed bridging the crack therefore showing their potential to act as toughening agents in brittle ceramics.

Thin films containing ZrO2 and dispersed fullerenes have been synthesized using codeposition processing, and the role that dispersed fullerenes play in the thermal behavior of co-deposited films has been characterized. The presence of C60 within the films has been confirmed using a new application of electron impact mass spectrometry, and the structure and stability of dispersed C60 in Zr and ZrO2 have been studied using Raman spectrometry, TEM, XRD, and WDS. Strong interactions between neighboring ZrO2 grains and dispersed C60 are manifested in the decreased desorption rate of C60 from the ZrO2 surface, and effects of dispersed C60 on the ZrO2 microstructure are observed using TEM. The potential for the use of fullerenes and fullerene-like structures in thermal barrier applications has been examined using both steady state and transient thermal conductivity measurement methods to measure the effective film conductivity. Significant reductions have been observed in the co-deposited film conductivity due to the contact resistance between dispersed fullerenes and neighboring ZrO2 grains, and contributions of fullerenes and the affected grain boundary region to thermal conductivity reductions have been identified. Dispersed C60 is proposed to affect phonon scattering in ZrO2 by imposing an interfacial resistance that is strongly dependent upon fullerene particle size, shape, and distribution.