In this thesis we study the thermal and inertia effects on atomic friction. The Friction Force Microscope (FFM) is simulated with the Tomlinson model. We introduce a method to calculate atomic friction including thermal effects. A novel scaling form is proposed and tested. We also investigate the inertial effects of the bulk and the tips. We find if the bulk is weakly damped, there is hysteresis in the velocity versus driving force curve. When the tips are weakly damped, the friction force shows large deviations. / In the first half of the thesis, we study the atomic friction between an FFM tip and a solid substrate at nonzero temperatures. Through theoretical analysis, we find the frictional force follows an universal scaling form F ∝ const---T2/31n 2/3(T/v), where T is temperature and v is scanning velocity. We test the theoretical results with numerical simulations. The FFM is simulated with the Tomlinson model. A tip is connected to a support with a spring. The support is pulled with a constant scanning velocity. We model the substrate with two different systems: one is a sinusoidal surface potential function for rigid substrate; the other is a ball-spring model for an elastic substrate. We find the friction force follows the scaling form well, with both kinds of substrates. This confirms the universality of our findings. The results also compare well to recent experimental work, permitting the quantitative extraction of microscopic parameters. / In the second half of the thesis, we study the inertial effects on atomic friction with a multiple-tip Tomlinson model. The model consists of an array of tips mounted on a common bulk support. We find if the bulk is weakly damped and driven by an external force, there is hysteresis in the velocity versus driving force curve. The relative amount of hysteresis decreases with the number of tips for incommensurate systems. We also find that if the tips are weakly damped and experience longjumps, then when the bulk is pulled with a constant velocity, the total friction force shows a large deviation. The deviation exists even for incommensurate systems and cannot be averaged out by incorporating more tips. However, the deviation decreases with increasing temperatures. These findings provide new insights on the influence of the inertia of the bulk and the tips. They also shed some light on future multiple-tip FFM experiments.

We use an atomic force microscope to perform nanolithography by local oxidation on two-dimensional electron gases. We study the oxidation process under different environmental conditions. It is found that oxide growth is hindered at high temperature and affected by ozone. These findings help us to understand the role of the water film present on the surface, and lead us to propose a mechanism that would avoid its widening effect. We develop a new lithography technique, diamond scribe scratching, and use it to make coarse mesoscopic structures. Then, we combine diamond scribe scratching and local oxidation to fabricate working mesoscopic structures, namely quantum point contacts.

We investigated the feasibility of using X-rays to study self-assembled monolayers on inorganic surfaces. Variable-temperature X-ray powder diffraction measurements of metal alkylphosphonate salts measured the contraction and expansion of planar inorganic layers as the material passed through a series of phase transitions when heated and cooled. Small-angle X-ray scattering measurements of alkanethiolate-capped gold nanoclusters allowed the determination of the average nanocluster size and the average distance between nanoclusters. These techniques have been demonstrated to contribute useful information which, when combined with results from other probes, lead to a detailed model of the materials' structures and properties.

Thermal conductivity is a powerful probe of electronic quasiparticles, especially at low temperatures. At higher temperatures it gives very useful information on the behavior of quasiparticles and phonons, however both contributions are more difficult to identify precisely. / We carried out a comparative study of the thermal conductivity in the hole and electron-doped cuprates, focussed on Bi2Sr2CaCu 2O8 (hole-doped) and Pr1.85Ce0.15CuO 4 (electron-doped). After a brief review of the literature it was clear that these families show very different physical properties, although they present similar features in their structure and phase diagram. / We detected the presence of a residual normal fluid in Bi2Sr 2CaCu2O8, in rather good agreement with the theory for d-wave superconductors, and showed its absence in Pr 1.85Ce0.15CuO4, firm indication of a nodeless gap. At higher temperatures we observed for the first time a peak below Tc in the thermal conductivity of Pr1.85Ce 0.15CuO4.

This thesis is dedicated to understanding the mechanism of the in situ synthesis of magnetic nanocomposites based on cellulosic substrates and characterizing the products of this synthesis. The preparation of uniform membranes containing large quantities of magnetite particles (Fe$ sb2$O$ sb3$) of defined site was our specific objective. For that purpose, cellulosic substrates of two different kinds: one of bacterial origin bacterial cellulose--BC), the other (Lyocell) derived from dissolving-grade wood-pulp through a dissolution/coagulation process (film-casting) were used as never-dried gel membranes. BC has also been used in the form of a suspension of open fibrillar pellets which were dried to a parchment after the in situ treatment. Characterization of the resulting magnetic materials was performed using transmission electron microscopy (TEM) in imaging and diffraction mode, X-ray diffraction (XRD), vibrating sample magnetometry (VSM) and Mossbauer spectroscopy. In BC membranes, needle-like lepidocrocite ($ gamma$-FeOOH) formed along the cellulose fibrils, using the crystalline surface as a nucleation site. Spherical magnetite particles subsequently formed around the needles. The less swollen Lyocell substrates produced needle-like feroxyhite ($ delta$-FeOOH) concentrated at the membrane surface while spherical magnetite particles formed within the membrane after several cycles of treatment. The treated BC and Lyocell membranes were both superparamagnetic at room temperature. The BC suspensions had no space constraint hence the treatment conditions were more critical for controlling the morphology of synthesized ferrites. These conditions could be defined to yield homogeneous membranes containing magnetite particles uniform in size, too large however to behave superparamagnetically at room temperature.

The goal of this thesis is to design a furnace to be used in in-situ time-resolved x-ray scattering studies of processes such as crystallization in 25 mum thick metallic ribbons. Two approaches are tested thoroughly. The first is to use forced convection to heat the sample. This is tested experimentally up to about 500 K with 1.5 mm thick samples and theoretical calculations are used to resolve these data and to predict the behaviour of the thin ribbons at higher temperatures. The results show that the ideal configuration for this heating mechanism involves heating the ends of the samples while allowing hot gas to flow over the surface. Despite its expected advantages, such as small thermal gradients within the sample, this system should have poor long term stability and require a difficult calibration in order to couple the gas heater to the heaters at the ends of the sample. (Abstract shortened by UMI.)

We study the interaction between dislocations and smooth elastic strains in thin film growth. In the first part of the thesis, a continuum model is introduced which, for the first time, includes nucleation, interaction, and dynamics of dislocations in an external stress field in two spatial dimensions. The model implicitly includes the presence of boundaries and their coupling to the elastic strains in the system. In particular, it naturally gives rise to the two well-known strain relaxation modes in thin films: the Asaro-Tiller-Grinfeld (ATG) morphological instability, leading to a grooved morphology of the film-vapor interface, and the nucleation of misfit dislocations for films thicker than the Matthews-Blakeslee critical thickness. The novelty of the model resides in the fact that both of these mechanisms are explicitly incorporated within a unified approach. Therefore, this is a generic model for dislocations in strained heterogeneous systems. / In the second part of the thesis, this model is applied to thin film growth with dislocations. It is shown that the film undergoes an ATG instability. However, the accumulation of dislocations at the film-substrate interface leads to an effectively screened stress in the film, and hence buckling occurs at longer wavelengths. In particular, it is shown that the film remains planar for sufficiently effective screening. Furthermore, the effect of dislocations depends very strongly on their equilibrium density and mobility. It is also shown that, in the late-time regime, dislocations interact strongly with the stress enhancement at the bottom of the grooves. In particular, this leads to a buildup of localized dislocations around the stress concentrations, in addition to dislocations at the film-substrate interface. This in turn leads to very complicated film morphologies, in qualitative agreement with experiments. / In the last part of the thesis, the dynamics of a single groove is studied in a strip geometry, both with and without dislocations. In the absence of dislocations, it is argued theoretically and shown numerically, that the groove attains a steady-state. A theoretical argument is given for the shape of the groove, and good agreement with numerics is found. Upon including noise in the dynamics, a morphological transition from straight to oscillatory propagation is found. (Abstract shortened by UMI.)

Previous measurements show that the thermal conductivity of $Y Ba sb2 Cu sb3 O sb{7- delta}$ in the basal plane is anisotropic with a large peak in the superconducting state. The magnitude of this anisotropy in the superconducting and normal states, and the dominant mechanism for heat conduction in the superconducting state are currently the subject of debate. We have measured the thermal conductivity of high quality $Y Ba sb2 Cu sb3 O sb{7- delta}$ for deoxygenated, twinned and detwinned samples along the a and b axes to shade light on this issue. We were able to measure the electrical and thermal conductivity using the same contacts and hence determine the Lorenz number L = $ kappa$p/T accurately. / Attributing the normal state anisotropy in the heat transport to electrons in the Cu-O chains, the Lorenz number takes on its full Sommerfeld value i.e. $L = L sb0.$ Under this assumption, the phonon conduction is about the same in the superconducting and deoxygenated samples. / Our results are discussed in connection with the two possible mechanisms for heat conduction in the superconducting state. We find that although a strong case can be made for the "electronic scenario" whereby the peak is due to rapidly increasing electron mean free path below $T sb{c},$ it is still not compelling at this stage. / In addition, it is found that the thermal conductivity along the a and b axes is isotropic at low temperatures, with a nonzero linear term in $ kappa,$ indicative of some uncondensed electrons as $T rightarrow$ 0. This low temperature isotropy contradicts previous explanations in terms of non-superconducting chains.

The amorphization of the Bi-Sr-Ca-Cu-O system by mechanical alloying starting from the individual oxides is examined. Several reactions concur to finally yield an amorphous matrix containing nanocrystalline copper oxide precipitates. Annealing the amorphous material leads, in a first step, to a crystalline solid solution, and, in a second step, to a low-Tc superconducting phase. Long term annealing allows the formation of the 85K superconducting phase.

Bismuth-based high T$ sb{c}$ superconductors were prepared from amorphous precursors made by melt-extraction. As-made amorphous fibers ranged from 0.7 $ mu$m to 100 $ mu$m in diameter and 0.2 cm to 5 cm in length. Fibers as thin as 1 $ mu$m in diameter were crystallized to form continuous filaments composed of single-grain chains. / Superconducting transitions at 105 K and 82 K were measured by SQuID magnetometry performed on annealed fibers of initial composition Bi$ sb{1.8}$Pb$ sb{0.2}$Sr$ sb2$Ca$ sb3$Cu$ sb4$O$ sb{x}$. X-ray diffractometry performed on annealed fibers revealed Bi-2212 (T$ sb{c}$ = 85 K) as the majority superconducting phase and Bi-2223 (T$ sb{c}$ = 110K) as the secondary phase. The volume fractions of superconducting phases were estimated to have lower bounds of 30% for 2212 and 5% for 2223. Differential scanning calorimetry measurements made on as-made amorphous fibers indicate a glass transition at T$ sb{g} / simeq$ 685 K, followed by a series of exothermic peaks associated with the formation of precursor crystalline phases identified as the low T$ sb{c}$ 2201 phase and a BCC Bi-Sr-Ca-Cu-O solid-solution with lattice parameter 4.25 A. Further work is required to determine the suitability of these new materials for applications such as multifilamentary conductors.