Through a mathematical and computational model of the physical behavior of shape memory alloy wires, this study shows that localized heating and cooling of such materials provides an effective means of damping vibrational energy. The thermally induced pseudo-elastic behavior of a shape memory wire is modeled using a continuum thermodynamic description based on an improved Landau-Devonshire potential. Our construction of the potential function allows the model to account for particular alloys as well as the general solid-state phase transformation, improving over traditional potentials that idealize many of the material properties or focus only on individual processes. The material's thermodynamic response is modeled using a nonlinear conservation of momentum and a nonlinear heat equation. The heat equation introduces an inhomogeneous version of the Fourier heat flux, thereby describing the discontinuous heat flow associated with shape memory materials more thoroughly than standard, continuous heat dissipation mechanisms do. This continuum thermodynamic model is then solved computationally to determine the resulting state of the wire in time. Continuous time Galerkin methods and affine finite elements treat the temporal and spatial discretizations of the model, respectively. Traditional methods for solution of the resulting finite-dimensional, nonlinear, nonconvex system of equations must introduce a significant artificial dissipation to achieve existence of solutions. The solution of the discrete system here uses a novel combination of the damped Newton method and a homotopy method for minimizing the artificial dissipation. This combination, inspired by the well-known Method of Vanishing Viscosity for the solution of scalar hyperbolic conservation laws, reduces the artificial dissipation errors introduced by traditional approaches for such nonlinear, nonconvex thermomechanical models. Computational tests show that the proposed model successfully describes the relevant physical processes inherent in shape memory alloy behavior. Further computational experiments then confirm that up to 80% of an initial shock of vibrational energy can be eliminated at the onset of a thermally-induced phase transformation.

Polypropylene composites with vapor-grown carbon nanofibers (VGCF's) as reinforcement were prepared. The fibers used have an rage diameter of 200 nm with interesting thermal, electrical and mechanical properties which make them very promising for engineering applications. Fiber purification and activation of functional groups were conducted, where amorphous carbon particles were successfully removed, achieving, high purity fibers. Sample preparation was performed using conventional plastic processing technologies. Interactions between the fibers and the matrix were analyzed by physical, mechanical and electrical properties of the composite. Thermal physical analysis on the samples showed that the presence of the fibers influenced the morphology and crystallinity of the matrix. The decomposition temperature, as well as the crystallization rate increased with increasing fiber content. The electrical resistivity of the prepared composites decreased 12 orders of magnitude providing a potential composite for ESD applications. The addition of VGCF's showed an increase in stiffness of 350. Melt viscosity values were also increased by the VGCF reinforcement. Dispersion, porosity, and bonding aspects were also analyzed.

Paramagnetic gadolinium-based carbon nanostructures are introduced as a new paradigm in high-performance magnetic resonance imaging (MRI) contrast agent (CA) design. Two Gd C60-based nanomaterials, Gd C60 [C(COOH)2]10 and Gd C60(OH)x are shown to have MRI efficacies (relaxivities) 5 to 20 times larger than any current Gd3+-based CA in clinical use. The first detailed and systematic physicochemical characterization was performed on these materials using the same experimental techniques usually applied to traditional Gd 3+-based CAs. Water-proton relaxivities were measured for the first time on these materials, as a function of magnetic field (5 x 10-4--9.4 T) to elucidate the different interaction mechanisms and dynamic processes influencing the relaxation behavior. These studies attribute the observed enhanced relaxivities completely to the "outer sphere" proton relaxation mechanism. These "outer sphere" relaxation effects are the largest reported for any Gd3+-based agent without inner-sphere water molecules. The proton relaxivities displayed a remarkable pH-dependency, increasing dramatically with decreasing pH (pH: 3--12). The increase in relaxivity resulted mainly from aggregation and subsequent three-order-of-magnitude increase in tauR, the rotational correlation time. Water-soluble fullerene materials (such as the neuroprotective fullerene drug, C3) readily cross cell membranes, suggesting an application for these gadofullerenes as the first intracellular, as well as pH-responsive MRI CAs. Studies performed at 60 MHz in the presence of phosphate-buffered saline (PBS, mice serum pH: 7.4) to mimic physiological conditions demonstrated that the aggregates can be disrupted by addition of salts, leading to a decrease in relaxivity. Biological fluids present a high salt concentration and should strongly modify the behavior of any fullerenes/metallofullerene-based drug in vivo. Gd C60[C(COOH)2]10 also showed enhanced relaxivity (23% increase) in the presence of the blood protein, human serum albumin (HSA). This result suggests a strong non-covalent interaction between Gd C60[C(COOH)2]10 and HSA leading to slower rotation and a subsequent increase in relaxivity. This also suggests Gd C 60[C(COOH)2]10 as a promising candidate for non-invasive MR angiographic applications to image the "blood pool." Finally, the various important factors or parameters discussed in this work provide valuable insight that can, in general, be used not only for the development of other carbon nanostructure-based MRI contrast agents, but also for any fullerene-based biomedical application.

Nanoindentation techniques have been widely used to measure thin film mechanical properties. One of the most commonly used methods of analysis of nanoindentation load and displacement data was developed by Oliver and Pharr. The objective of this dissertation is to examine some of the limitations of this method and develop improvements so that a more accurate hardness and elastic modulus measurements can be made. Detailed experimental studies of bulk monolithic materials and soft films on hard substrates were performed to evaluate the validity of the Oliver and Pharr experimental technique and analysis procedures. Three different indenters were used. They were the Berkovich and Vickers pyramids, and a cone with a 70.3$\sp\circ$ included angle. It is shown that there are inherent limitations in the Oliver and Pharr indenter shape-function calibration method which means that it cannot be applied to the blunt Vickers and conical indenters used in this work. A new procedure was developed which avoids these problems. The pile-up behavior of monolithic and thin film materials was extensively investigated. Experimental results for monolithic materials show that materials with low elastic modulus to hardness ratios (E/H) such as ceramics are less likely to pile-up. On the other hand, monolithic materials which have high E/H ratios and low strain hardening coefficients or soft films on hard substrates are more likely to pile-up. The pile-up generated during the indentation process in these materials can create as much as 50% more contact area between the indenter and the specimen. The effects of pile-up on the hardness and elastic modulus measurements for monolithic and thin film materials were examined. It is shown that when pile-up occurs, Oliver and Pharr method overestimates both the hardness and the elastic modulus. Only if these extra contact area generated by the pile-up is included are the correct hardness and elastic modulus values obtained. The amount of pile-up is also found to depend on the indenter geometry. The Vickers indenter generates more pile-up at the indentation corners than the Berkovich indenter for both monolithic materials and soft films on hard substrates. The absolute amount of pile-up in monolithic materials for Vickers indentations is also more than the Berkovich.

The main goal of this study is to acquire a deeper understanding of the friction and wear evolution during contact sliding on amorphous carbon overcoats. Therefore, hydrogenated and nitrogenated amorphous carbon films were produced using magnetron sputtering. The film roughness, density and thickness vary with dopant concentration in the sputtering atmosphere. A friction tester was developed and built to evaluate tribological properties of a-C films during continuous contact sliding tests. The durability of the a-C:H film with the higher density and roughness is the greatest and it exhibits the lowest friction force. The friction forces for a-C:N films are significantly higher than for a-C:H. Wear in the a-C films is evaluated using atomic and lateral force microscopy (AFM and LFM). An optimal Fourier (Wiener) filter is developed and used to quantify the wear induced anisotropy in surfaces. LFM reveals the wear tracks more readily than AFM, i.e. while topographical changes are modest, significant surface alterations may be present. A film prepared by magnetron sputtering in a 0.5% H$\sb2$ in Ar atmosphere is subjected to a detailed friction and wear study. The friction increases monotonically until debris is formed in the interface. The topographical changes before debris formation are limited to gradual smoothing preferentially in the wear direction at the top of the surface. The sliding contact has significant effect on the lateral force images. Even after only 7,500 head the surface in the wear track is partitioned into two sections: an unaltered section with unchanged lateral force, and an altered section where the local coefficient of friction has a different value. The magnitude of the local coefficient of friction does not change with wear time. FE SEM with low energy incident electrons monitors changes in surface conductivity in the wear track indicating a chemical alteration of the surface. LFM corroborates that the macroscopic friction force is proportional to the real area of contact. We develop a wear model that assumes proportionality between the real area of contact and the macroscopic friction force. The model describes the friction force in a-C:H films until debris is formed in the interface.

The nickel catalyzed reaction of the diamond (110) surface in hydrogen has been studied at temperatures below 950$\sp\circ$C. A $\sim$300 A nickel film was vaporized onto a polished diamond chip with a (110) surface orientation. The samples were scratched and annealed at various temperatures, times and hydrogen pressures. The diamond surfaces were subsequently analyzed using an atomic force microprobe (AFM), Raman spectroscopy and electron dispersive spectroscopy (EDS). Surface reconstruction was observed at temperatures $>$700$\sp\circ$C during one hour annealings in H$\sb2.$ Carbon etching was observed under these conditions for times $\geq$two hours or longer. The metal-diamond interface was found to be stable in the presence of H$\sb2$ at pressures as low as 0.1 Torr. The activation energy of surface reconstruction was calculated to be approximately 70 kcalmole$\sp{{-}1}.$ Our findings suggest the possibility of growing diamond CVD films using diatomic hydrogen and thin metal films.

The three single crystal independent elastic constants c(,11), c(,12), and c(,44) were measured over the temperature range 4.2 K to 300 K for tantalum, tungsten, and five intermediate alloys. Anomalies were found for the shear constant c(,44) as functions of both temperature and composition. These appear to be interrelated. A similar, but less pronounced effect was found for the shear constant C' as a function of temperature at about the same composition as for c(,44). These anomalies are centered at approximately 50 percent tungsten. An evaluation by a model of Peter et. al. indicates that these anomalies are due to topology changes in the Fermi surface. The interatomic force constants for nearest and second nearest neighbors were calculated by two experimentally independent methods. Only the central forces between nearest and second nearest neighbors were considered for the first method. The elastic constants measured by this investigation were used for this calculation. The second method considered interactions from the third to the seventh nearest neighbors and used phonon dispersion data from Oakridge National Laboratories on the same crystals. The nearest neighbor force constants differed between the two methods. The Debye Temperatures were calculated using the Kubic Harmonics method. The dependence upon c(,44) for the Debye Temperature calculations is reflected upon the results. No calorimetric data was available for comparison. The Ta-W system was found to be isotropic to within about 5% from 50% tungsten to pure tungsten. The system also shows a remarkable adherence to the Cauchy relation if the long-range forces are considered.

The structure of vapor-deposited fiber boron is studied using x-ray diffraction and computer modeling. The diffraction pattern displays broad peaks which have been interpreted in previous studies to arise from a microcrystalline material having an average crystallite size of about 30 (ANGSTROM). This study was initiated by quantitatively investigating the microcrystalline models suggested by workers. The Debye equation was used to compute the intensity scattered by randomly oriented crystallites. Crystallites with diameters up to 26 (ANGSTROM) were considered. An additional computer program was written to calculate the different interatomic distances and their multiplicities, and to compute their contributions to the various peaks. This method of analysis was utilized to drive the atoms in the microcrystal towards positions which gave a more favorable fit of the computed intensity profile to the experimentally obtained profile. Chapter I describes boron, the production of boron fiber and its properties and includes a survey of previous work. Chapter II addresses the experimental techniques used to obtain the diffraction pattern. In Chapter III, the methods utilized the model and compute the interference patterns are described. Chapter IV summarizes and discusses the results which indicate that a complex atomic arrangement is present, involving distorted nearest-neighbor distances, partial icosahedra and individual atoms. Some of the icosahedra display orientations present in alpha-rhombohedral boron, while others display those observed in tetragonal boron. The average size of the region of local atomic ordering is approximately 20 (ANGSTROM).

A model WC-Ni cemented carbide was designed to investigate the response of this type of composite to compressive cyclic loading. The material's monotonic stress-strain behavior was studied in the compressive range, revealing considerable micro-strain plasticity and rapid strain-hardening characteristic of this composite family. Specimens were mechanically conditioned at 20 Hz for one million cycles using a zero-compression-zero sinusoidal program in stress control. Peak stress was varied in increments from zero to 1900 MPa, where failure was observed. Bulk hardness, residual stress state, and fracture toughness were monitored as a function of cyclic stress. Only modest shifts in bulk behavior were observed. On a local scale, fatigue effects on dislocation density and structure were found in the binder phase by transmission electron microscopy. As cyclic stress levels were raised, favorably oriented grain-boundary-ledge dislocation sources were activated, leading to precipitate-free bands of intense slip. These bands broadened at the higher stress levels, saturating entire binder phase regions and triggering a change in dislocation arrangements. At the highest stress levels discontinuous cell structures were observed in some binder regions. It is concluded that fatigue damage in this type of composite accrues in the binder in a characteristic manner dictated by short binder-carbide interaction distances and local stress conditions. The principal roles of fatigue in the fracture behavior of this alloy are concluded to be: a source of incipient defects and a means of stable crack extension.

The linear coefficient of thermal expansion of Nb-Mo alloys was measured and its correlation with the anomalous behavior of the elastic constant C(,44) is discussed. The coincidence of the anomaly near the critical composition of 40% Mo is made evident. The Gruneisen parameters, both at low-temperature and high-temperature, are calculated and compared to the values of the Gruneisen parameters, obtained from the pressure derivative of elastic constants. The floating-zone technique, using a combination of R.F. induction and resistance heating, was successfully used to make large single crystals of Chronium-rich Vanadium alloys. The elastic constants C(,11) and C(,44) of Chromium alloys containing 0.67% and 1.5% Vanadium were determined by the pulse-superposition method. C(,12) was evaluated from smooth curves drawn as a function of temperature for C(,11) and C'. The addition of Vanadium to Chromium lowers both Neel and spin-flip temperatures drastically. The sharp minimum in the elastic constant C(,11) in Chromium is diminished by the addition of Vanadium. The effect of magnetic transition on the shear constants is extremely small. The high attenuation observed in Chromium at T(,N) virtually vanishes as a result of alloying with Vanadium.