In a 2005 article in Science [45], Park et al. measured in vacuum the friction between a coated atomic-force-microscope tip and the clean two-fold surface of an AlNiCo quasicrystal. Because the two-fold surface is periodic in one direction and aperiodic (with a quasiperiodicity related to the Fibonacci sequence) in the perpendicular direction, frictional anisotropy is not unexpected; however, the magnitude of that anisotropy in the Park experiment, a factor of eight, is unprecedented. By eliminating chemistry as a variable, the experiment also demonstrated that the low friction of quasicrystals must be tied in some way to their quasiperiodicity. Through various models, we investigate generic geometric mechanisms that might give rise to this anisotropy.

This thesis addresses the sampling problem in a high-dimensional space, i.e., the computation of averages with respect to a defined probability density that is a function of many variables. Such sampling problems arise in many application areas, including molecular dynamics, multiscale models, and Bayesian sampling techniques used in emerging machine learning applications. Of particular interest are thermostat techniques, in the setting of a stochastic-dynamical system, that preserve the canonical Gibbs ensemble defined by an exponentiated energy function. In this thesis we explore theory, algorithms, and numerous applications in this setting. We begin by comparing numerical methods for particle-based models. The class of methods considered includes dissipative particle dynamics (DPD) as well as a newly proposed stochastic pairwise Nosé-Hoover-Langevin (PNHL) method. Splitting methods are developed and studied in terms of their thermodynamic accuracy, two-point correlation functions, and convergence. When computational efficiency is measured by the ratio of thermodynamic accuracy to CPU time, we report significant advantages in simulation for the PNHL method compared to popular alternative schemes in the low-friction regime, without degradation of convergence rate. We propose a pairwise adaptive Langevin (PAdL) thermostat that fully captures the dynamics of DPD and thus can be directly applied in the setting of momentum-conserving simulation. These methods are potentially valuable for nonequilibrium simulation of physical systems. We again report substantial improvements in both equilibrium and nonequilibrium simulations compared to popular schemes in the literature. We also discuss the proper treatment of the Lees-Edwards boundary conditions, an essential part of modelling shear flow. We also study numerical methods for sampling probability measures in high dimension where the underlying model is only approximately identified with a gradient system. These methods are important in multiscale modelling and in the design of new machine learning algorithms for inference and parameterization for large datasets, challenges which are increasingly important in "big data" applications. In addition to providing a more comprehensive discussion of the foundations of these methods, we propose a new numerical method for the adaptive Langevin/stochastic gradient Nosé-Hoover thermostat that achieves a dramatic improvement in numerical efficiency over the most popular stochastic gradient methods reported in the literature. We demonstrate that the newly established method inherits a superconvergence property (fourth order convergence to the invariant measure for configurational quantities) recently demonstrated in the setting of Langevin dynamics. Furthermore, we propose a covariance-controlled adaptive Langevin (CCAdL) thermostat that can effectively dissipate parameter-dependent noise while maintaining a desired target distribution. The proposed method achieves a substantial speedup over popular alternative schemes for large-scale machine learning applications.

Du fait de l'augmentation de la demande de produits pour les applications grand public, industrielles et automobiles, des mémoires embarquées fiables et à faible coût de fabrication sont de plus en plus demandées. Dans ce contexte, les mémoires split-gate à piégeage discret sont proposées pour des microcontrôleurs. Elles combinent l'avantage d'une couche de stockage discrète et de la con guration split-gate. Durant ce travail de recherche, des mémoires split-gate à couche de piégeage discret ayant des longueurs de grille de 20nm sont présentées pour la première fois. Celles-ci on été réalisées avec des nanocristaux de silicium (Si-nc), du nitrure de silicium (SiN) ou un hybride Si-nc/SiN avec diélectrique de control de type SiO2 ou AlO et sont comparées en termes de performances lors des procédures d'eff acement et de rétention. Ensuite, la miniaturisation des mémoires split-gate à piégeage de charge est étudié, en particulier au travers de l'impact de la réduction de la longueur de grille sur la fenêtre de mémorisation, la rétention et la consommation. Le rôle des défauts dans le diélectrique de contrôle (alumine) utilisé dans les mémoires de type TANOS a été étudié. Des travaux ont été menés pour déterminer l'origine des pièges dans ce matériau, par le biais de la simulation atomistique ainsi que d'analyses physico-chimiques précises. Nous avons montré que la concentration de pièges dans AlO pouvait être réduite par ajustement des conditions de procédé de fabrication, débouchant ainsi sur l'amélioration de la rétention dans les mémoires à piégeage de charge. Ce résultat est convenable pour les applications de type embarqué / Due to the increasing demand for consumer, industrial and automotive products, highly reliable, and low integration cost embedded memories are more and more required. In this context, split-gate charge trap memories were proposed for microcontroller products, combining the advantage of a discrete storage layer and of the split-gate con guration. In this thesis, split-gate charge trap memories with electrical gate length down to 20nm are presented for the 1st time. Silicon nanocristals (Si-nc), or silicon nitride (SiN) and hybrid Si-nc/SiN based split-gate memories, with SiO2 or AlO control dielectrics, are compared in terms of program erase and retention. Then, the scalability of split-gate charge trap memories is studied, investigating the impact of gate length reduction on the memory window, retention and consumption. We thus studied the role of defects on alumina control dielectric employed in TANOS-like memory. We used atomistic simulation, consolidated by a detailed alumina physico-chemical material analysis, to investigate the origin of traps in alumina. We showed that the trap concentration in AlO can be decreased by adjusting the process conditions leading to improved retention behaviour in charge trap memory, suitable for embedded applications.

Orientador: Maurice de Koning / Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Física Gleb Wataghin / Made available in DSpace on 2018-08-18T18:05:04Z (GMT). No. of bitstreams: 1 Barrozo_AlexandreHernandes_M.pdf: 4359405 bytes, checksum: c11339c45a35270724d3ee11084432af (MD5) Previous issue date: 2011 / Resumo: Fizemos estudos de uma nova técnica proposta para amostrar o ensemble canônico [H. Dammak et al., Phys. Rev. Lett. 103,190601 (2009)], capaz de amostrar flutuações quânticas através de Dinâmica Molecular(DM) clássica. Esta técnica é baseada nos termostatos estocásticos clássicos, que se baseiam nas equações de Langevin, com a diferença de usar um ruído com correlação temporal, ao invés de ruído branco. Aplicamos esta técnica para o cálculo de grandezas termodinâmicas de sistemas cuja solução analítica (ou numérica) é conhecida. Mais especificamente, focamos o estudo no cálculo de energias livres através de integração termodinâmica fora do equilíbrio. Neste trabalho também desenvolvemos um método que nos permite alterar a temperatura do termostato ao longo da simulação. Este método é baseado no reescalonamento do ruído aleatório que constitui este termostato quântico. Mostramos aqui que, apesar de termos excelentes resultados para o caso do oscilador harmônico, esta técnica apresenta falhas no que se diz respeito a potenciais anarmônicos, amostrando distribuições fundamentalmente incorretas / Abstract: We study a new method devised to sample the canonical ensemble [H. Dammak et al., Phys. Rev. Lett. 103,190601 (2009)], which includes quantum uctuations by means of classical Molecular Dynamics. This method is based on classical stochastic thermostats, except that it uses colored noise, instead of a white one. We apply this method to compute thermodynamical properties of systems whose analytical (or numerical) solution for the Schr odinger equation is known. Specifically, we focus on free-energy calculations using non-equilibrium thermodynamics integration. In this work, we also develop a technique that allows one to change the thermostat's temperature during the simulation. This technique is based on rescaling the random noise sequence that constitutes the thermostat. We show that, although we have excelent results for the harmonic oscillator case, problems arise while studying anharmonic potentials, in which the method sample distributions that are fundamentally incorrect / Mestrado / Física Estatistica e Termodinamica / Mestre em Física

No description available.

The workshop is composed of two main parts: the first part devoted to atomistic Monte Carlo simulations and the second part devoted to the Phase Field modelling. In each part a lecture will be accompanied by exercise activities.

L’adaptation des surfaces pour des fonctions prédéterminées par le choix des matériaux métalliques ou des couches minces ayant des propriétés mécaniques avancées peut potentiellement permettre de réaliser des nouvelles applications à petites échelles. Concevoir de telles applications utilisant des nouveaux matériaux nécessite en premier lieu la connaissance des propriétés mécaniques des matériaux ciblés à l’échelle microscopique et nanoscopique. Une méthode souvent appliquée pour caractériser les matériaux à petites échelles est la nanoindentation, qui peut être vue comme une mesure de dureté à l’échelle nanoscopique. Ce travail présente une contribution relative à l'interprétation des résultats de la nanoindentation, qui fait intervenir un grand nombre de phénomènes physiques couplés à l'aide de simulations numériques. A cette fin une approche interdisciplinaire, adaptée aux phénomènes apparaissant à petites échelles, et située à l’intersection entre la physique, la mécanique et la science des matériaux a été utilisée. Des modèles numériques de la nanoindentation ont été conçus à l'échelle atomique (modèle discret) et à l'échelle des milieux continus (méthode des éléments finis), pour étudier le comportement du nickel pur. Ce matériau a été choisi pour ses propriétés mécaniques avancées, sa résistance à l'usure et sa bio-compatibilité, qui peuvent permettre des applications futures intéressantes à l'échelle nanoscopique, particulièrement dans le domaine biomédical. Des méthodes avancées de mécanique du solide ont été utilisées pour prendre en compte les grandes déformations locales du matériau (par la formulation corotationelle), et pour décrire les conditions de contact qui évoluent au cours de l'analyse dans le modèle à l'échelle des milieux continus (traitement des conditions de contact unilatérales et tangentielles par une forme de Lagrangien augmenté). L’application des modèles numériques a permis de contribuer à l’identification des phénomènes qui gouvernent la nanoindentation du nickel pur. Le comportement viscoplastique du nickel pur pendant nanoindentation a été identifié dans une étude expérimentale-numérique couplée, et l'effet cumulatif de la rugosité et du frottement sur la dispersion des résultats de la nanoindentation a été montré par une étude numérique (dont les résultats sont en accord avec des tendances expérimentales). Par ailleurs, l’utilisation de l’outil numérique pour une autre application à petites échelles, la manipulation des objets par contact, a contribué à la compréhension de la variation de l’adhésion électrostatique pendant micromanipulation. La déformation plastique des aspérités de surface sur le bras de manipulateur (en nickel pur) a été identifiée comme une source potentielle d’augmentation importante de l'adhésion pendant la micromanipulation, qui peut potentiellement causer des problèmes de relâche et de précision de positionnement, observés expérimentalement. Les résultats présentés dans cette thèse montrent que des simulations numériques basées sur la physique du problème traité peuvent expliquer des tendances expérimentales et contribuer à la compréhension et l'interprétation d'essais couramment utilisé pour la caractérisation aux petites échelles. Le travail réalisé dans cette thèse s’inscrit dans un projet de recherche appelé "mini-micro-nano" (mµn), financé par la Communauté Française de Belgique dans le cadre de "l'Action de Recherche Concertée", convention 04/09-310.

Fundação de Apoio a Pesquisa e à Inovação Tecnológica do Estado de Sergipe - FAPITEC/SE / In this work, we studied the h-RMnO3 (R = Er, Tm, Yb, Lu, In and Sc) manganites which are materials with important multiferroic properties. The classical atomistic simulation was used in order to obtain a set of interatomic potentials that describe the structural properties of manganite family in ferroelectric phase (FE). The structural properties such as the lattice parameters, the unit cell volume and the interatomic distances were investigated and compared with results of the literature. The influence of chemical pressure on the structural properties of these compounds was analyzed. It was observed that the lattice parameters (a and c) and the volume of the unit cell increase as the ionic radius of the R3+ ion increases. All h-RMnO3 manganites were studied under hydrostatic pressure up to 40 GPa. It was observed that the lattice parameters (a and c) and the unit cell volume decrease with an increase in the hydrostatic pressure. The paraelectric phase (PE) was studied using the interatomic potential developed in the ferroelectric phase. Thus, the structural properties in this phase were obtained and compared with the literature. Effects of doping of trivalent ions on the h-LuMnO3 manganite were studied in three crystallographic sites (Mn, Lu1 and Lu2). Solution energy was calculated for all dopants (Al, Cr, Ga, Fe, Sc, Yb, Er, Y, Ho, Gd, Eu, Sm, Nd e La) and the results showed that the Mn site has the lowest solution energy, therefore, the Mn site is energetically more favorable. The metals (Al, Cr, Ga, Fe) dopants proved to be more favorable to incorporate at the Mn site than rare earths ones (Sc, Yb, Er, Y, Ho, Gd, Eu, Sm, Nd and La), but at the Lu site the rare earths are energetically more favorable to incorporate. The calculated interatomic distances at the Mn site decreases when doping with metal (Al, Cr, Ga and Fe). However, these distances increase when doping is done with rare earth (Sc, Yb, Er, Y, Ho, Gd, Eu, Sm, Nd and La). In addition, important ferroelectric distortion properties were evaluated due the change in the tilting angle of the MnO5 bipyramid. The tilting angle increases in the pure system if the doping is done with metals (Al, Cr, Ga, Fe) for all sites. On the other hand, the tilting angle decreases with rare earth ions incorporation at the Mn and Lu1 sites and increases in the rare earths doping at the Lu2 site. / Estuda-se neste trabalho as manganitas h-RMnO3 (R = Er, Tm, Yb, Lu, In e Sc), que são materiais que apresentam importantes propriedades multiferroicas. A simulação atomística clássica foi usada com o objetivo de obter um conjunto de potenciais interatômicos que descrevam as propriedades estruturais da família de manganitas na fase ferroelétrica (FE). As propriedades estruturais como os parâmetros de rede, o volume da célula unitária e as distâncias interatômicas foram investigadas e comparadas com a literatura. Em vista disso, a influência da pressão química nessas propriedades estruturais foi analisada. Observou-se que os parâmetros de rede (a e c) e o volume da célula unitária aumentam à medida que o raio iônico do íon R3+ aumenta. Todas as manganitas h-RMnO3 foram submetidas a uma pressão hidrostática de até 40 GPa. Verificou-se que houve uma redução nos parâmetros de rede (a e c) e no volume da célula unitária à medida que a pressão hidrostática aumenta. A fase paraelétrica (PE) foi estudada utilizando os potenciais interatômicos desenvolvidos na fase ferroelétrica. Com isso, as propriedades estruturais nessa fase foram obtidas e comparadas com a literatura. O efeito da dopagem com íons trivalentes foi estudado em três sítios cristalográficos (Mn, Lu1 e Lu2) da manganitas h-LuMnO3. A energia de solução foi calculada para todos os dopantes (Al, Cr, Ga, Fe, Sc, Yb, Er, Y, Ho, Gd, Eu, Sm, Nd e La) e os resultados mostraram que o sítio do Mn possui a menor energia de solução, portanto, esse é o sítio energeticamente mais favorável. Diante disso, os metais (Al, Cr, Ga, Fe) se mostraram energeticamente mais favoráveis que os terras raras (Sc, Yb, Er, Y, Ho, Gd, Eu, Sm, Nd e La) no sítio do Mn, porém nos sítios do Lu1 e Lu2 os terras raras são mais favoráveis de serem incorporados. As distâncias interatômicas calculadas no sítio do Mn diminuem após a dopagem com os metais (Al, Cr, Ga e Fe), enquanto que essas distâncias aumentam quando a dopagem é realizada com os terras raras (Sc, Yb, Er, Y, Ho, Gd, Eu, Sm, Nd e La). Além disso, os efeito da dopagem na distorção ferroelétrica foram avaliadas com a mudança no ângulo de tilting da bipirâmide MnO5. Um aumento do ângulo de tilting em relação ao sistema puro é obtido quando a dopagem é realizada com metais (Al, Cr, Ga, Fe) para todos os sítios estudados. Por outro lado, o ângulo de tilting decresce quando ocorre a incorporação de íons terras raras nos sítios do Mn e Lu1, mas aumenta quando a dopagem dos terras raras é realizada no sítio do Lu2. / São Cristóvão, SE

Le présent manuscrit présente le développement méthodologique du calcul de la tension de surface d’interfaces solide-liquide via des simulations de dynamique moléculaire. Après une courte présentation des avancées dans le domaine du calcul de la tension de surface pour les interfaces fluide-fluide et solide-fluide, les principales méthodes de calcul de la tension de surface d’un point de vue théorique sont montrées et généralisées pour le cas des interfaces solide-liquide, puis mises en oeuvre dans le cas de simulations de dynamique moléculaire. Un système école, constitué d’une feuille de graphène pour la phase solide et d’un bain de méthane pour la phase liquide, est ensuite étudié pour observer l’influence des artefacts de simulation sur le calcul de la tension de surface, montrant en particulier des effets de taille bien plus importants que pour le cas des interfaces liquide-liquide. Un autre système constitué d’une tranche de cuivre pour la phase solide, et d’un bain de méthane pour la phase liquide, a permis d’étudier l’effet inédit aux systèmes solide-liquide appelé anisotropie, montrant en particulier l’importance du caractère tensoriel de la tension de surface pour ce type de système. L’influence des paramètres du potentiel croisé entre les atomes de cuivre et de méthane est ensuite étudié. Finalement, deux systèmes applicatifs sont abordés, d’une part le système graphène-eau permettant d’étudier les effets de l’interaction électrostatique, et d’autre part un système constitué d’un solide explosif, le 1,3,5-triamino-2,4,6-trinitrobenzène (TATB) en contact avec un bain polymère pour la phase liquide, représentatif d’un cas réel d’intérêt. / This manuscript presents the methodological development of surface tension calculation of solid-liquidinterfaces via molecular dynamics simulations. After a short presentation of the advances in the field ofsurface tension calculation for fluid-fluid and solid-fluid interfaces, the main methods of surface tensioncalculation from a theoretical point of view are shown and generalized for solid-liquid interfaces, thenimplemented in the case of molecular dynamics simulations. A school system, consisting of a graphenesheet for the solid phase and a methane bath for the liquid phase, is then studied to observe the influenceof simulation artifacts on the surface tension calculation, showing in particular much larger size effectsthan in the case of liquid-liquid interfaces. Another system consisting of a copper slice for the solid phaseand a methane bath for the liquid phase made it possible to study the novel effect of solid-liquid systemscalled anisotropy, showing in particular the importance of the tensor character of the surface tension forthis type of system. The influence of the parameters of the cross potential between copper and methaneatoms is then studied. Finally, two application systems are discussed, on the one hand the graphene-watersystem for studying the effects of electrostatic interaction, and on the other hand a system consisting ofan explosive solid, 1,3,5-triamino-2,4,6-trinitrobenzene (TATB) in contact with a polymer bath for theliquid phase, representing a real case of interest.

The present work was initiated to investigate the stability, structural and thermodynamic properties of transition metal carbides, nitrides and carbo-nitrides by atomistic simulations and experimentations. The interatomic pair potentials of Cr-Cr, Mn-Mn, Fe-Fe, C-C, Cr-C, Mn-C, Fe-C, Cr-Fe, Cr-N and Mn-N were inverted by the lattice inversion method and ab initio cohesive energies, and then employed to investigate the properties of Cr-, Mn- and Fe-carbides by atomistic simulations in this work. For the binary M7C3 carbide, the structural properties of M7C3 (M = Cr, Mn, Fe) were investigated by atomistic simulations. The results show that the stable structure for these compounds is hexagonal structure with P63mc space group. The cohesive energy of M7C3 calculated in this work indicates that the stability of carbides decreases with the increasing in metal atomic number. Further, the vibrational entropy of Cr7C3 was calculated at different temperatures and compared with the entropy obtained by experimentations. The comparison demonstrates that the main contribution to the entropy is made by the vibrational entropy. For the binary τ-carbides, the structural properties of Cr23C6 and Mn23C6, as well as the vibrational entropy of Cr23C6 were computed. Further, the site preference of ternary element Fe among 4a, 8c, 32f and 48h symmetry sites in Cr23-xFexC6 was studied. It has been seen that Fe atoms would firstly occupy 4a sites and then 8c sites. The lattice constant and stability of Cr23-xFexC6 were also computed with different Fe content. In order to understand the relative stability of the transition metal carbides and nitrides, the standard formation Gibbs energies of carbides and nitrides for Cr, Mn and Fe were compared. The order of carbon and nitrogen affinities for Cr, Mn and Fe was further clarified by the comparison of the interatomic pair potentials among Cr-C, Mn-C, Fe-C, Cr-N and Mn-N. It was found that Cr-N interaction was very strong in comparison with other binary interactions above and consequently, nitrogen addition would lead to a strong decrease in the thermodynamic activity of chromium in Cr-containing alloys. This was confirmed by the investigations of thermodynamic activities of Cr in the Fe-Cr-N and Fe-Cr-C-N alloys. The activities were measured in the temperature range 973-1173 K by solid-state galvanic cell method involving CaF2 solid electrolyte under the purified N2 gas. In addition, the analysis of nitrogen content and phase relationships in the Fe-Cr-N and Fe-Cr-C-N alloys equilibrated at 1173 K were carried out by inert-gas fusion thermal conductivity method, X-ray diffraction and scanning electron microscopy technique. The experimental results show that the solubility of nitrogen in the alloys decreases with the decreasing chromium content, as well as the increasing temperature. The addition of nitrogen to the alloys was found to have a strong negative impact on the Cr activity in Fe-Cr-N and Fe-Cr-C-N systems. / QC 20100929