This thesis represents a small contribution to our understanding of metastable patterns in various stochastic models from physics and biology. By a \emph{metastable pattern}, we mean a pattern that appears to persist in a regular fashion on some timescale, but disappears or undergoes an irregular change on a longer timescale. Metastable patterns frequently result from stochastic perturbations of patterns that are stable without perturbation. In this thesis, we study stochastic perturbations of stable spatiotemporal patterns in several classes of PDE and integral equations. In particular, we address two major questions: \begin{enumerate}[Q1.] \item When perturbed by noise, for how long does a pattern that is stable without noise persist? \item How does the stochastic perturbation affect the average behaviour of a pattern on the timescale where it appears to persist? \end{enumerate} To address these questions, we pursue two lines of inquiry: the first based on the theory of \emph{quasi-ergodic measures}, and the second based on \emph{phase decomposition techniques}. In our first line of inquiry we present novel, rigorous connections between metastability of general infinite dimensional stochastic evolution systems and the spectral properties of their sub-Markov generators using the theory of quasi-ergodic measures. To do so, we develop a novel $L^p$-approach to the study of quasi-ergodic measures. We are then able to draw conclusions about the metastability of travelling waves and other patterns in a class of stochastic reaction-diffusion equations. For instance, we obtain a rigorous definition of the \emph{quasi-asymptotic speed}~of a travelling wave in a stochastic PDE. We moreover find that stochastic perturbations of amplitude $\sigma>0$ cause the quasi-asymptotic speed of certain travelling waves to deviate from the deterministic wave speed by a constant that is approximately proportional to $\sigma^2$. In our second line of inquiry, the dynamics of our (infinite dimensional) stochastic evolution system are projected onto a finite dimensional manifold that captures some property of a metastable pattern. While most previous studies using phase reduction techniques have used the \emph{variational phase}, we take an approach based on the \emph{isochronal phase}, inspired by classical work on finite dimensional oscillatory systems. When the pattern in question is a travelling wave, the isochronal phase captures the position of the wave at a given point in time. By exploiting the regularity properties of the isochronal phase, we are able to prove several novel results about the metastable behaviour of the reduced dynamics in the small noise regime in a very large class of stochastic evolution systems. These results allow us to moreover compute the noise-induced changes in the speed of stochastically perturbed travelling waves and other patterns. The results we obtain using this approach are numerically precise, and may be applied to a very general class of stochastic evolution systems.

In this compilation licentiate thesis, two studies, concerned with genre and genre-based writing, are synthesized and discussed. Genres are often used as a pedagogical device for teaching the writing of educationally valued texts. The focus is on what is here termed genre metastability, a conceptual tool for understanding how these genres can appear both as fixed and as flexible, that is, offering both constraints and creativity for writing. The findings, based on a case study into genre-based history instruction, show how these, ostensibly opposing aspects, coexist in teaching and that they can be productively put to use by teachers as a means of differentiating instruction. Further, the thesis explores how the fixed/flexible dynamic can be used as a tool for understanding how similar types of history texts can offer diametrically opposed historical understanding of content. These findings have implications for teachers, teacher educators, and (history) textbooks in terms of a deepened understanding of how, or rather when, genres constrain, and when they foster creativity.

Thermoelectric materials are playing a larger role in the global effort to develop diverse, efficient, and sustainable energy technologies: primarily through power-generating thermoelectric modules. The principal components of thermoelectric modules are solid-state thermoelectric materials – typically heavily doped semiconductors – that convert heat directly into electricity. However, this conversion efficiency is too low to supplant traditional energy technologies – severely limiting the distribution of clean and sustainable thermoelectric energy technologies. Efforts to enhance thermoelectric efficiency, which have been underway for decades, have been slow to realize appreciable gains in thermoelectric efficiency. However, a key advance in improving efficiency – the New Paradigm in thermoelectric material research – has been the development of thermoelectric nanocomposites. Thermoelectric nanocomposites show improved efficiency; however, they are often synthesized from highly toxic elements via energetically intense and costly synthesis procedures. Therefore, this research focuses on the discovery and development of a novel procedure for synthesizing thermoelectric nanocomposites – attrition enhanced nanocomposite synthesis – from open cage-like skutterudite-based materials. With further optimization, high-performance power-generating thermoelectric materials can be produced via this technique. Therefore, attrition-enhanced nanocomposite synthesis may play a small, though instrumental, role in achieving sustainable electrical power. / Graduation date: 2012 / Access restricted to the OSU Community at author's request from Jan. 6, 2012 - Jan. 6, 2013

O dioxido de estanho na estrutura rutila (SnO2) é um semicondutor de gap largo e faz parte da classe dos óxidos condutores transparentes (TCO). Possui gap direto de 3,6 eV e condutividade do tipo n, mesmo quando não dopado intencionalmente. Estudos teóricos e experimentais atribuem este comportamento à presença de defeitos intrínsecos. Por outro lado, impurezas de hidrogênio, em sítios intersticiais ou substituindo o átomo de oxigênio, poderiam ser responsáveis pelo caráter n do SnO2. Neste trabalho apresentamos nossos resultados de estrutura eletrônica, a partir de cálculos de primeiros princípios, para o dióxido de estanho puro, assim como levando em conta a presença de defeitos intrínsecos - VO, VSn, Sni, Oi, OSn, SnO, SnO+OSn, Sni+VO - e para vários centros de impureza de hidrogênio - Hi, HO, HBC, Hi-Hi, Hi-HO, Hi-HBC, HBC-HBC, onde V significa vacância e BC a impureza localizada em um sítio entre ligação. Os resultados para a impureza de hidrogênio são confrontados com os dos defeitos intrínsicos. Nossas análises mostram, tanto para o caso das impurezas de H isoladas quanto para os pares complexos H-H, que estes centros apresentam caráter doador. Em todas as configurações, as energias de formação são suficientemente baixas, comparadas com as dos defeitos intrínsecos, mostrando competitividade e sugerindo que a impureza de hidrogênio poderia ser responsável pela característica de condutividade n do cristal SnO2. Apresentamos também resultados de propriedades eletrônicas e magnéticas para impurezas de metal de transição MT (MT = V, Cr, Fe, Mn, Co e Ni) em SnO2 em uma configuração estrutural de baixa concentração. Estes sistemas são denominados semicondutores magnéticos diluídos (DMS - diluted magnetic semiconductor), isto é, ligas diluídas do tipo Sn1-xMTxO2 e Sn1-xMTxO2-y(VO)y. Consideramos neste estudo as concentrações x = 0,04 e y = 0,02, correspondendo a valores experimentalmente possíveis de se obter. Este estudo aponta para a existência de estados magnéticos metaestáveis para estes sistemas e mostra como a vacância de oxigênio afeta este comportamento. Para todos os casos, o estado eletrônico fundamental encontrado apresenta configuraçãoo de alto spin (HS - high-spin) e o fenômeno de spin-crossover para o estado de baixo spin (LS - low-spin) é possível de ocorrer. A metaestabilidade obtida para estes sistemas DMS é estudada em conecção com as relaxações estruturais em torno da impureza, na ausência e na presença da vacância de oxigênio. Por fim, alternando respectivamente camadas magnéticas e não magnéticas de r-CrO2 e r-SnO2, foram estudados sistemas em uma configuração de super-rede (SL - superlattice), do tipo (CrO2)n(SnO2)n, com n = 1; 2; ...; 10 sendo o número de monocamadas. Para todos os valores de n foi observado comportamento meio-metal (half-metal) para os sistemas. O estado fundamental é ferromagnético (FM), com momento mangético igual a 2 mu_B por cromo ndependentemente do número de monocamadas. E como o óxido r-CrO2 é instável na temperatura ambiente, porém pode ser estabilizado, quando crescido sobre o r-SnO2, sugerimos que as super-redes (CrO2)n(SnO2)n podem ser aplicadas na tecnologia de spintrônica provendo eficiente polarização de spin de seus portadores. Os cálculos de estutura eletrônica foram realizados levando em conta a polarização de spin, usando o método PAW (Projector-Augmented-Wave) implementado no pacote computacional VASP (Vienna Ab-initio Simulation Package), dentro da teoria DFT (density functional theory) utilizando a aproximação local da densidade com correção GGA-PBE (generalized gradient corrections) e aproximação GGA-PBE+U, onde U é a correção on site de Hubbard. / Rutile tin dioxide (r-SnO2) is a wide-band-gap semiconductor and is part of a class of promising transparent conducting oxides (TCO). It has a direct band gap of 3.6 eV and shows n-type conductivity, even when not intentionally doped, which is usually attributed to intrinsic defects. On the other hand, it has been proposed theoretically that in spite of unintentionally doping, hydrogen impurities at interstitial or O substitutional sites could be responsible for the n-type conductivity in SnO2. In this work we report the results of ab initio electronic structure calculatations for pure tin dioxide as well as for some intrinsic defects VO, VSn, Sni, Oi, OSn, SnO, SnO+OSn and Sni+VO and for several H-related defect centers Hi, HO, HBC, Hi-Hi, Hi-HO, Hi-HBC, HBC-HBC, where V means vacancy and BC bond-centered sites. Our H-related results centers are confronted with those obtained for the intrinsic defects. Our findings show, for example, that hydrogen impurity has a donor character for all studied centers and that not only the isolated H, but also some of its complexes, show competitive low formation energies, suggesting that various H-related centers could be responsible for the n-type conductivity observed in the unintentionally doped SnO2. We discuss some results of the electronic and magnetic properties of TM-doped tin dioxide (TM = V, Cr, Fe, Mn, Co and Ni) in a diluted magnetic oxide configurations, i.e., Sn1-xTMxO2 and Sn1-xTMxO2-y(VO)y diluted alloys. As a prototype we will consider x = 0.04 and y = 0.02, which corresponds to a TM content just within the experimental window. Our aim is to analyze the presence of magnetic metastable states in these systems and how oxygen vacancies affect this metastability. For all cases, the ground state corresponds to the expected high spin (HS) configuration and a spin-crossover to the low-spin state is possible. The obtained magnetic metasbility in TM-doped SnO2 is discussed in connection with the structural relaxations around the impurity in absence and in presence of O vacancies. Finally, alternated magnetic and non-magnetic layers of rutile-CrO2 and rutile-SnO2 respectively, in a (CrO2)n(SnO2)n superlattice (SL) configuration, with n being the number of monolayers which are considered equal to 1, 2, ..., 10 are studied. A half-metallic behavior is observed for the (CrO2)n(SnO2)n SLs for all values of n. The ground state is found to be ferromagnetic (FM) with a magnetic moment of 2 µB per chromium atom, and this result does not depend on the number of monolayers n. As the FM rutile-CrO2 is unstable at ambient temperature, and known to be stabilized when on top of SnO2, we suggest that (CrO2)n(SnO2)n SLs may be applied to spintronic technologies since they provide efficient spin-polarized carriers. Spin-polarized electronic structure calculations were performed using the Projector-Augmented-Wave (PAW) method as implemented in the Vienna Ab-initio Simulation Package (VASP), within the spin density functional theory in the local density approximation with generalized gradient corrections (GGA-PBE) and GGA- PBE+U, where U is the Hubbard correction.

The development of functional flexible electronics is essential to enable applications such as conformal medical imagers, wearable health monitoring systems, and flexible light-weight displays. Intensive research on thin-film transistors (TFTs) is being conducted with the goal of producing high-performance devices for improved backplane electronics. However, there are many challenges regarding the performance of devices fabricated at low temperatures that are compatible with flexible plastic substrates. Prior work has reported on the change in TFT characteristics due to mechanical strain, with especially extensive data on the effect of strain on field-effect mobility. This thesis investigates the effect of gate-bias stress and elastic strain on the long-term stability of flexible low-temperature hydrogenated amorphous silicon (a-Si:H) TFTs, as the topic has yet to be explored systematically. An emphasis was placed on bias-stress measurements over time in order to obtain information on the physical mechanisms of instability. Drain current was measured over various intervals of time to track the degradation of devices due to metastability, and results were then compared across devices of various sizes under tensile, compressive, and zero strain. Transfer characteristics of the TFTs were also measured under the different conditions, to allow for extraction of parameters that would provide insight into the instability mechanisms. In addition to parameter extraction, the degradation and recovery of TFT output current was quantitatively compared for various bias-stress times across the different levels of strain. Finally, the instability mechanisms are modelled with a Markov system to further examine the effect of strain on long-term TFT operation. From the analysis of results, it was found that shallow charge trapping in the dielectric is the main mechanism of instability for short bias stress times, and did not seem to be greatly affected by strain. For longer bias stress times of over 10000 seconds, defect creation in the a-Si:H becomes a more significant contributor to instability. Both tension and compression increased defect creation compared to TFTs with zero applied strain. Compression appeared to cause the greatest increase in the rate of defect formation, likely by weakening Si-Si bonds in the a-Si:H. Tension appeared to cause a less significant increase, possibly due to a strengthening of some proportion of the Si-Si bonds caused by the slight elongation of bond length or because the applied tension relieves intrinsic compressive stress in a-Si:H film. A longer conduction path and greater dielectric area appears to increase the bias-stress and strain-related effects. Therefore reducing device size should increase the reliability of flexible TFTs.

The development of functional flexible electronics is essential to enable applications such as conformal medical imagers, wearable health monitoring systems, and flexible light-weight displays. Intensive research on thin-film transistors (TFTs) is being conducted with the goal of producing high-performance devices for improved backplane electronics. However, there are many challenges regarding the performance of devices fabricated at low temperatures that are compatible with flexible plastic substrates. Prior work has reported on the change in TFT characteristics due to mechanical strain, with especially extensive data on the effect of strain on field-effect mobility. This thesis investigates the effect of gate-bias stress and elastic strain on the long-term stability of flexible low-temperature hydrogenated amorphous silicon (a-Si:H) TFTs, as the topic has yet to be explored systematically. An emphasis was placed on bias-stress measurements over time in order to obtain information on the physical mechanisms of instability. Drain current was measured over various intervals of time to track the degradation of devices due to metastability, and results were then compared across devices of various sizes under tensile, compressive, and zero strain. Transfer characteristics of the TFTs were also measured under the different conditions, to allow for extraction of parameters that would provide insight into the instability mechanisms. In addition to parameter extraction, the degradation and recovery of TFT output current was quantitatively compared for various bias-stress times across the different levels of strain. Finally, the instability mechanisms are modelled with a Markov system to further examine the effect of strain on long-term TFT operation. From the analysis of results, it was found that shallow charge trapping in the dielectric is the main mechanism of instability for short bias stress times, and did not seem to be greatly affected by strain. For longer bias stress times of over 10000 seconds, defect creation in the a-Si:H becomes a more significant contributor to instability. Both tension and compression increased defect creation compared to TFTs with zero applied strain. Compression appeared to cause the greatest increase in the rate of defect formation, likely by weakening Si-Si bonds in the a-Si:H. Tension appeared to cause a less significant increase, possibly due to a strengthening of some proportion of the Si-Si bonds caused by the slight elongation of bond length or because the applied tension relieves intrinsic compressive stress in a-Si:H film. A longer conduction path and greater dielectric area appears to increase the bias-stress and strain-related effects. Therefore reducing device size should increase the reliability of flexible TFTs.

xvi, 117 p. ; ill. (some col.) A print copy of this thesis is available through the UO Libraries. Search the library catalog for the location and call number. / High efficiency thin film solar cells have the potential for being a world energy solution because of their cost-effectiveness. Looking to the future of solar energy, there is the opportunity and challenge for thin film solar cells. The main theme of this research is to develop a detailed understanding of electronically active defect states and their role in limiting device performance in copper indium gallium diselenide (CIGS) solar cells. Metastability in the CIGS is a good tool to manipulate electronic defect density and thus identify its effect on the device performance. Especially, this approach keeps many device parameters constant, including the chemical composition, grain size, and interface layers. Understanding metastability is likely to lead to the improvement of CIGS solar cells. We observed systematic changes in CIGS device properties as a result of the metastable changes, such as increases in sub-bandgap defect densities and decreases in hole carrier mobilities. Metastable changes were characterized using high frequency admittance spectroscopy, drive-level capacitance profiling (DLCP), and current-voltage measurements. We found two distinctive capacitance steps in the high frequency admittance spectra that correspond to (1) the thermal activation of hole carriers into/out of acceptor defect and (2) a temperature-independent dielectric relaxation freeze-out process and an equivalent circuit analysis was employed to deduce the dielectric relaxation time. Finally, hole carrier mobility was deduced once hole carrier density was determined by DLCP method. We found that metastable defect creation in CIGS films can be made either by light-soaking or with forward bias current injection. The deep acceptor density and the hole carrier density were observed to increase in a 1:1 ratio, which seems to be consistent with the theoretical model of V Cu -V Se defect complex suggested by Lany and Zunger. Metastable defect creation kinetics follows a sub-linear power law in time and intensity. Numerical simulation using SCAPS-1D strongly supports a compensated donor- acceptor conversion model for the experimentally observed metastable changes in CIGS. This detailed numerical modeling yielded qualitative and quantitative agreement even for a specially fabricated bifacial CIGS solar cell. Finally, the influence of reduced hole carrier mobility and its role in limiting device performance was investigated. / Adviser: J. David Cohen

O dioxido de estanho na estrutura rutila (SnO2) é um semicondutor de gap largo e faz parte da classe dos óxidos condutores transparentes (TCO). Possui gap direto de 3,6 eV e condutividade do tipo n, mesmo quando não dopado intencionalmente. Estudos teóricos e experimentais atribuem este comportamento à presença de defeitos intrínsecos. Por outro lado, impurezas de hidrogênio, em sítios intersticiais ou substituindo o átomo de oxigênio, poderiam ser responsáveis pelo caráter n do SnO2. Neste trabalho apresentamos nossos resultados de estrutura eletrônica, a partir de cálculos de primeiros princípios, para o dióxido de estanho puro, assim como levando em conta a presença de defeitos intrínsecos - VO, VSn, Sni, Oi, OSn, SnO, SnO+OSn, Sni+VO - e para vários centros de impureza de hidrogênio - Hi, HO, HBC, Hi-Hi, Hi-HO, Hi-HBC, HBC-HBC, onde V significa vacância e BC a impureza localizada em um sítio entre ligação. Os resultados para a impureza de hidrogênio são confrontados com os dos defeitos intrínsicos. Nossas análises mostram, tanto para o caso das impurezas de H isoladas quanto para os pares complexos H-H, que estes centros apresentam caráter doador. Em todas as configurações, as energias de formação são suficientemente baixas, comparadas com as dos defeitos intrínsecos, mostrando competitividade e sugerindo que a impureza de hidrogênio poderia ser responsável pela característica de condutividade n do cristal SnO2. Apresentamos também resultados de propriedades eletrônicas e magnéticas para impurezas de metal de transição MT (MT = V, Cr, Fe, Mn, Co e Ni) em SnO2 em uma configuração estrutural de baixa concentração. Estes sistemas são denominados semicondutores magnéticos diluídos (DMS - diluted magnetic semiconductor), isto é, ligas diluídas do tipo Sn1-xMTxO2 e Sn1-xMTxO2-y(VO)y. Consideramos neste estudo as concentrações x = 0,04 e y = 0,02, correspondendo a valores experimentalmente possíveis de se obter. Este estudo aponta para a existência de estados magnéticos metaestáveis para estes sistemas e mostra como a vacância de oxigênio afeta este comportamento. Para todos os casos, o estado eletrônico fundamental encontrado apresenta configuraçãoo de alto spin (HS - high-spin) e o fenômeno de spin-crossover para o estado de baixo spin (LS - low-spin) é possível de ocorrer. A metaestabilidade obtida para estes sistemas DMS é estudada em conecção com as relaxações estruturais em torno da impureza, na ausência e na presença da vacância de oxigênio. Por fim, alternando respectivamente camadas magnéticas e não magnéticas de r-CrO2 e r-SnO2, foram estudados sistemas em uma configuração de super-rede (SL - superlattice), do tipo (CrO2)n(SnO2)n, com n = 1; 2; ...; 10 sendo o número de monocamadas. Para todos os valores de n foi observado comportamento meio-metal (half-metal) para os sistemas. O estado fundamental é ferromagnético (FM), com momento mangético igual a 2 mu_B por cromo ndependentemente do número de monocamadas. E como o óxido r-CrO2 é instável na temperatura ambiente, porém pode ser estabilizado, quando crescido sobre o r-SnO2, sugerimos que as super-redes (CrO2)n(SnO2)n podem ser aplicadas na tecnologia de spintrônica provendo eficiente polarização de spin de seus portadores. Os cálculos de estutura eletrônica foram realizados levando em conta a polarização de spin, usando o método PAW (Projector-Augmented-Wave) implementado no pacote computacional VASP (Vienna Ab-initio Simulation Package), dentro da teoria DFT (density functional theory) utilizando a aproximação local da densidade com correção GGA-PBE (generalized gradient corrections) e aproximação GGA-PBE+U, onde U é a correção on site de Hubbard. / Rutile tin dioxide (r-SnO2) is a wide-band-gap semiconductor and is part of a class of promising transparent conducting oxides (TCO). It has a direct band gap of 3.6 eV and shows n-type conductivity, even when not intentionally doped, which is usually attributed to intrinsic defects. On the other hand, it has been proposed theoretically that in spite of unintentionally doping, hydrogen impurities at interstitial or O substitutional sites could be responsible for the n-type conductivity in SnO2. In this work we report the results of ab initio electronic structure calculatations for pure tin dioxide as well as for some intrinsic defects VO, VSn, Sni, Oi, OSn, SnO, SnO+OSn and Sni+VO and for several H-related defect centers Hi, HO, HBC, Hi-Hi, Hi-HO, Hi-HBC, HBC-HBC, where V means vacancy and BC bond-centered sites. Our H-related results centers are confronted with those obtained for the intrinsic defects. Our findings show, for example, that hydrogen impurity has a donor character for all studied centers and that not only the isolated H, but also some of its complexes, show competitive low formation energies, suggesting that various H-related centers could be responsible for the n-type conductivity observed in the unintentionally doped SnO2. We discuss some results of the electronic and magnetic properties of TM-doped tin dioxide (TM = V, Cr, Fe, Mn, Co and Ni) in a diluted magnetic oxide configurations, i.e., Sn1-xTMxO2 and Sn1-xTMxO2-y(VO)y diluted alloys. As a prototype we will consider x = 0.04 and y = 0.02, which corresponds to a TM content just within the experimental window. Our aim is to analyze the presence of magnetic metastable states in these systems and how oxygen vacancies affect this metastability. For all cases, the ground state corresponds to the expected high spin (HS) configuration and a spin-crossover to the low-spin state is possible. The obtained magnetic metasbility in TM-doped SnO2 is discussed in connection with the structural relaxations around the impurity in absence and in presence of O vacancies. Finally, alternated magnetic and non-magnetic layers of rutile-CrO2 and rutile-SnO2 respectively, in a (CrO2)n(SnO2)n superlattice (SL) configuration, with n being the number of monolayers which are considered equal to 1, 2, ..., 10 are studied. A half-metallic behavior is observed for the (CrO2)n(SnO2)n SLs for all values of n. The ground state is found to be ferromagnetic (FM) with a magnetic moment of 2 µB per chromium atom, and this result does not depend on the number of monolayers n. As the FM rutile-CrO2 is unstable at ambient temperature, and known to be stabilized when on top of SnO2, we suggest that (CrO2)n(SnO2)n SLs may be applied to spintronic technologies since they provide efficient spin-polarized carriers. Spin-polarized electronic structure calculations were performed using the Projector-Augmented-Wave (PAW) method as implemented in the Vienna Ab-initio Simulation Package (VASP), within the spin density functional theory in the local density approximation with generalized gradient corrections (GGA-PBE) and GGA- PBE+U, where U is the Hubbard correction.

Dans cette thèse, nous nous sommes intéressés à la cavitation de bulles de vapeur dans un liquide confiné sous tension (c’est-à-dire sous pression négative). Ce travail s’est développé en étudiant deux aspects différents mais complémentaires : la simulation numérique et l’expérience biomimétique. L’étude numérique utilise la simulation par dynamique moléculaire d’un liquide confiné dans une cellule solide. Cette méthode nous a permis d’étudier précisément l’effet de l’interaction entre le solide et le liquide (angle de contact), mais aussi de la géométrie sur la nucléation de bulles de vapeur. Nous avons également étudié l’interaction entre deux cellules voisines, et ainsi par comparaison avec un modèle, nous avons mis en évidence une corrélation entre deux évènements de cavitation ainsi que les paramètres importants dans ce phénomène. L’étude expérimentale, quant à elle, a été réalisée sur un dispositif en hydrogel de polymère inspiré de systèmes naturels. Cette méthode nous permet d’étudier un système ayant des caractéristiques mécaniques proches des observations naturelles sur les sporanges de fougères tout en pouvant contrôler sa géométrie. Nous avons alors mis en évidence l’effet de l’épaisseur des parois entre cellules permettant d’observer des cavitations isolées ou groupées de plusieurs centaines de bulles. La taille des cellules permet de mesurer des vitesses de propagation allant jusqu’à plus de 800 m/s. A l’aide d’un modèle acoustique nous avons mis en évidence les paramètres importants dans cette propagation. / In this work, we have been interested in the cavitation process of vapor bubbles in a confined and stretched liquid. We have followed two complimentary points of view: numerical simulation and biomimetical experiments. For the numerical study we have used molecular dynamics simulations of a liquid confined in a solid cell. This method allows us to study precisely the effect of the interaction between the solid and the liquid (contact angle), and also the geometrical properties on the nucleation of vapor bubbles. We have also studied the interaction between two neighboring cells, and by comparing with a model, we have shown a correlation between two cavitation events and the important parameters taking place in this phenomenon. For the experimental study, we have used polymer hydrogel devices inspired from natural systems (ferns sporangia). This method allows us to study a system having almost the same mechanical properties as the natural one, and showing the possibility to control its geometry. We have shown that the wall thickness between the cells can control the propagation properties from isolated cavitation to grouped propagation (up to several hundreds of bubbles). The cell size controls the propagation velocity, up to values of 800 m/s. We have shown by comparing with an acoustical model the important parameters that control this phenomenon.

Les travaux de cette thèse portent sur l’étude de la métastabilité du modèle de Blume-Capel. Il s’agit d’un modèle introduit en 1966 dans lequel évoluent au cours du temps des spins à trois états +1, -1, 0, représentant respectivement une particulechargée positivement, négativement, et l’absence de particule, sur un réseau. La thèse est structurée en deux parties. La première partie contient un travail en collaboration avec C. Landim qui est paru dans la revue Journal of Statistical Physics. L’article traite du comportement métastable du modèle de Blume-Capel lorsque la température tend vers 0, dans le cas où la taille du domaine dans lequel vit le processus est fixée durant l’évolution. La seconde partie est consacrée à l’extension des résultats du premier papier au cas où la taille de la boite croît exponentiellement vite vers +1 lorsque la température décroît vers 0. Pour ce modèle, sur une très grande échelle de temps, trois états métastables subsistent, à savoir les états où le tore est respectivement remplis par des spins négatifs, positifs, ou "nuls". Il est démontré qu’avec probabilité 1, partant de la configuration n’ayant que des spins négatifs, le processus visite la configuration n’ayant que des spins "nuls" avant de visiter la configuration n’ayant que des spins positifs.Les résultats de la thèse consistent notamment à caractériser les configurations critiques et à fournir des estimations précises des temps d’atteinte des états stables. / This thesis is about the study of the metastability of the Blume-Capel model. This model, introduced in 1966, is a nearest-neighbor spin system where the single spin variable takes three possible values +1, -1, 0. One can interpret it as a system ofparticles with spins. The value 0 of the spin corresponds to the absence of particle, whereas the values ± correspond to the presence of a particle with the respective spin. The thesis is divided in two parts. The first part is an article published in Journal of Statistical Physics with C. Landim. We prove the metastable behavior of the Blume-Capel model when the temperature decreases to 0 on a fixed size torus.The second part is dedicated to the generalization of these results to the case of a torus which size increases to +1 as the temperature decreases to 0. For this model, three metastable states -1, 0,+1 remain on a very large time scale, where -1, 0,+1 stand for the configuration where the torus is respectively filled with -1’s, 0’s and +1’s. We prove that starting from -1, the process visits 0 before reaching +1 with very high probability. We also caracterize the critical configurations and provide sharp estimates of the transition times.