Modeling of single-walled carbon nanotubes, multi-walled nanotubes and nanotube reinforced polymer composites using both the Finite Element method and the Molecular Dynamic simulation technique is presented. Nanotubes subjected to mechanical loading have been analyzed. Elastic moduli and thermal coefficient of expansion are calculated and their variation with diameter and length is investigated. In particular, the nanotubes are modeled using 3D elastic beam finite elements with six degrees of freedom at each node. The difficulty in modeling multi walled nanotubes is the van der Waal's forces between adjacent layers which are geometrically non linear in nature. These forces are modeled using truss elements. The nanotube-polymer interface in a nano-composite is modeled on a similar basis. While performing the molecular dynamic simulations, the geometric optimization is performed initially to obtain the minimized configuration and then the desired temperature is attained by rescaling the velocities of carbon atoms in the nanotube. Results show that the Young's modulus increases with tube diameter in molecular mechanics whereas decreases in molecular dynamics since the inter-atomic potential due to chemical reactions between the atoms is taken into consideration in molecular dynamics unlike in molecular mechanics.

Bioactive glass is a type of third generation bioactive material that can bond to both soft and hard tissue with applications ranging from bone defect repair, coatings for metallic implants, to scaffolds for tissue engineering. Design of bioactive glasses for these applications rely on a detailed understanding of the structures of these glasses which are complicated and multicomponent. In this thesis, I have applied molecular dynamics (MD) simulations with interatomic potentials developed in our group to understand the effect of modifier cation substitution on the structures and properties of two series of bioactive glasses. Particularly, MD simulations are used to understand K2O to Na2O and MgO to CaO substitution on the short and medium range structures (such as cation coordination number, pair distribution function, Qn distribution, and ring size distribution) and properties (such as bulk and Young's moduli and CTE) of 55S4.1 bioactive glasses. As Na2O is incrementally substituted with K2O in 55S4.1, a decrease of the glass transition temperature (Tg) and an increase of CTE was observed, as well as a decreasing trend in the moduli. For the MgO to CaO substitution series, Mg2+ is mainly four-fold coordinated that suggests that it can play a role as a network former in this series. Results of both series showed characteristics of the phenomena of the mixed alkali effect (MAE) that has been known to show non-linear variations in trends like Tg in glasses with alkali and alkali earth ion substitution.

Les zéolithes sont des matériaux aluminosilicates nanoporeux largement utilisés dans l'industrie pour la filtration de l'eau, la séparation de gaz et le craquage des bruts pétroliers. Leurs propriétés d'adsorption et de catalyse sont reliées à la géométrie de leur réseau poreux d'une part et aux cations qui sont présents à leur surface d'autre part. Dans cette thèse, nous étudions par simulation moléculaire l'adsorption de molécules polaires, i.e. l'eau et le méthanol, dans deux zéolithes sodiques de type Faujasite. L'objectif de ce travail est d'étudier l'effet de la présence de molécules adsorbées sur la position et la dynamique des cations sodium et inversement l'effet des cations sur l'adsorption et la diffusion des molécules d'adsorbat. Nous nous sommes tout d'abord attachés à simuler par la méthode Monte Carlo le processus d'adsorption en fonction de plusieurs paramètres: i) répartition et densité des substitutions Si/Al, ii) valeur des charges électriques portées par le réseau zéolithique et iii) nature des potentiels d'interaction. Nous avons ainsi pu simuler les isothermes et les chaleurs d'adsorption d'une part et d'autre part la redistribution cristallographique des cations en fonction du nombre de molécules adsorbées, Nous nous avons ensuite porté notre intérêt à l'aide de la méthode de Dynamique Moléculaire, aux propriétés dynamiques (diffusion, transport) du système molécules adsorbées/cations/zéolithe. L'originalité de ce projet réside dans l'utilisation de ces deux techniques complémentaires qui permettent d'accéder aux propriétés thermodynamiques, structurales et dynamiques de tels systèmes. Les résultats de ces simulations sont comparés, en particulier, à ceux obtenus expérimentalement par Spectroscopie de Relaxation Diélectrique. / Zeolites are nanoporous aluminosilicate materials which attract a great deal of attention because of their use in industry for water filtration, phase separation and for cracking of crude oil. Their adsorption and catalysis properties depend on their pores geometry and on the cations which are located at their surface. In this work, we study by means of molecular simulation techniques the adsorption of polar molecules, i.e; water and methanol, in two zeolites type Faujasite. The aim of this study is to determine, on the one hand, the effect of the adsorbate on the cation distribution and dynamics at the surface of the zeolite and, on the other hand, the effect of the cation density on the adsorption and diffusion of the adsorbate molecules. We first simulate by Monte Carlo methods the adsorption of water and methanol and investigate the influence of several parameters: i) localisation and density of Si/Al substitution, ii) electrical charges at the nanopore surface and iii) expression of the interaction potential between the various atoms of the considered system. We thus calculated isotherms and heat of adsorption as well as the crystallographic re-localisation of the sodium cation as a function of the adsorbed molecule loading. . Then, we study by Molecular Dynamics the dynamics of the system adsorbate/cations/zeolite. The originality of this project lies in the use of both Molecular Dynamics and Monte Carlo techniques, which are complementary to study the thermodynamic, structural and dynamical properties of confined systems. Our simulation results are compared with those obtained by means of Dielectric Relaxation Spectroscopy experiments.

Ce travail est consacré à l’étude par simulation de dynamique moléculaire l’effet d’une perturbation contrôlée, utilisant des molécules photoisomérisables sur les effets de taille finie et de surface dans les matériaux amorphes. Le présent modèle utilise le potentiel de Lennard-Jones et l’algorithme de Verlet pour résoudre les équations du mouvement. Nous avons réalisé plusieurs boîtes de simulation de tailles différentes dans le but d’étudier l’effet de la taille de la boîte sur les propriétés du matériau. Nous avons utilisé ces boîtes pour simuler le matériau à différentes températures, le matériau était avec un chromophore dilué à l’intérieur, le chromophore s’isomérisant ou non,et avec une période courte ou longue. L’étude utilise des simulations à l’échelle atomique par dynamique moléculaire où l’énergie produite par le processus de photo-isomérisation est absorbée par un thermostat,après sa dégradation thermique à l’intérieur de la matrice hôte. Par une modification continue de la forme des photochromes trans vers cis et inversement, une photo isomérisation est introduite de manière périodique. Dans ces conditions, nous avons montré que ce modèle reproduit bien les propriétés statiques et dynamiques d’un matériau amorphe. Nous avons trouvé que des hétérogénéités dynamiques apparaissent dans un liquide surfondu pour une température qui égale 120K. Nous avons constaté que le temps de relaxation diminue en même temps que la force des mouvements coopératifs lorsque la taille du système est devient plus petite. Ces résultats suggèrent la présence de deux mécanismes physiques concurrents différents avec différentes échelles de longueur. / This work is devoted to the study by molecular dynamics simulation the effect of a controlled disturbance, using photo-isomerizable molecules on the finite size effects and surface in amorphous materials. This model uses the potential of Leonard Jones and the Verlet algorithm to solve the equations of motion.We realized several simulation boxes of different sizes in order to study the effect of the size of the box on the material properties. We used these boxes to simulate the material at different temperatures, the material was diluted with a chromophore inside, the chromophore isomerized or not, and with a short or long period. The study uses simulations at the atomic scale molecular dynamics where the energy produced by the photoisomerization is absorbed by a thermostat after thermal degradation within the host matrix. By a continuous change in the form of photochromes trans tocis and vice versa, a photo isomerization is introduced periodically. Under these conditions, we have shown that this model reproduces well the static and dynamic properties of amorphous material. We found that dynamic heterogeneities appear in a supercooled liquid to a temperature equal to 120 K. We found that the relaxation time decreases along with the strength of cooperative movements when the system size is becoming smaller. These results suggest the presence of two different physical mechanisms competitors with different length scales

Made available in DSpace on 2014-06-11T19:30:54Z (GMT). No. of bitstreams: 0 Previous issue date: 2007-08-16Bitstream added on 2014-06-13T20:01:02Z : No. of bitstreams: 1 gaspar_rt_dr_sjrp.pdf: 649983 bytes, checksum: 1fcf82fd9f75312b80b50e055388690d (MD5) / Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) / O Poli(etileno glicol) (PEG) é um polímero sintético cujas características tem despertado grande interesse em diversas áreas. Suas aplicações podem ser vistas nas mais variadas áreas, desde biotecnologia e medicina até aplicações industriais e cosméticos. Alguns aspectos físicos como a estrutura adotada por esse polímero em diferentes solventes e detalhes sobre a interação entre essas moléculas ainda necessitam de maiores esclarecimentos, o que o torna objeto de intensa investigação. Essa tese visa desenvolver um modelo para moléculas de PEG, que possa ser utilizado em experimentos de dinâmica molecular. Resultados de simulações com esse modelo foram comparados a dados experimentais presentes na literatura, de forma a verificar o comportamento do modelo em diferentes condições, avaliando assim sua adequação. Os valores de densidade, obtidos dos sistemas simulados, apresentaram erro máximo de 1,14% para concentrações de até 50% de PEG400. A densidade do sistema em função da temperatura concorda com os dados da literatura, mantendo um erro fixo de 0,35%, que está relacionado com a concentração de 50% utilizada nessa simulação. A estrutura helicoidal, apresentada pelas moléculas de PEG ao final do processo de preparação dos modelos, é perdida rapidamente em todas as diferentes condições em que o sistema foi simulado, indicando que tal estrutura é energeticamente desfavorável em água. Com o aumento da concentração de PEG, as seguintes estruturas foram encontradas: moléculas de PEG livres em solução em concentrações inferiores a 5%, aglomerados de PEG entre 5 e 50%, com uma transição gradual entre uma estrutura e outra. Os resultados obtidos para concentrações acima de 50% não são conclusivos. Seguindo o procedimento aplicado ao modelo inicial, de PEG400, foi desenvolvido... / Poly(ethylene glycol) (PEG) is a synthetic polymer whose characteristics have attracted great interest in different fields. It has been applied in very different areas, from biotechnology and medicine to industry and cosmetics. Physical aspects like the structure PEG assumes in different solvent and details on the interaction between these polymers still lack clarity, make PEG an object of intense investigation. This Thesis aims do develop a model for PEG molecules that can be used in molecular dynamic simulations. Results of simulations using this model were compared to published experimental data, in order to investigate the behavior of the model under different conditions to evaluate its validity. The density values obtained from the simulations exhibit a maximum error of 1.14% for PEG400 concentrations up to 50%. The system density as a function of temperature agrees with experimental data from the literature within an error of 0.35% for the 50% PEG in the simulation. The helicoidal structure assumed by the PEG molecules at the end of the procedure of the model preparation is quickly lost under every simulation condition, thus indicating that the helicoidal structure is not energetically favorable for PEG in water. As PEG concentration is increased, the following structures were found: free PEG molecules below ca 5%, PEG clusters from ca 5-50%, with a gradual transition from one structure to another. The results for concentrations higher than 50% are not conclusive. Following the procedure applied to the initial PEG400 model, a second model was developed, almost four times larger, and used to investigate possible molecular effects capable to induce phase thermoseparation. The transition from different system states took place on average temperatures between 423.3 K and 424.1 K at the average pressure of 8.98 Bar.

Adsorption processes are utilized in both medicine and industry. It is important to have an understanding of adsorption processes to better predict the outcomes and discern potential difficulties. The primary objective of this research is to further the understanding of the nature and extent of the adsorption process in solution, which is a function of the chemical composition of the adsorbates, adsorbents, and solvent. This was accomplished by employing experimental studies as well as thermodynamic calculations and molecular dynamic simulations. Four activated carbons were used as the model adsorbents in this study. And, barbital, phenobarbital and primidone were used to elucidate the structural features of the adsorbates that were most responsible for the interaction with activated carbons. A Two-Mechanism Langmuir-Like Equation (TMLLE) was proposed to describe the independent presence of two adsorption mechanisms: non-site-specific adsorption and site-specific adsorption. The analyses of data generated by both previous investigators and current studies, suggest that the TMLLE allows an accurate analysis of the adsorption process. Based on the parameters in the TMLLE, the Modified Crisp Model and the van’t Hoff Model were employed to determine the Gibbs free energy changes for both site-specific adsorption and non-site-specific adsorption. Comparing the Gibbs free energy changes calculated by the Modified Crisp Model and the van’t Hoff Model (site-specific adsorption case), it is concluded that 5 water molecules are displaced by a phenobarbital molecule on the surface of activated carbons. And, for non-site-specific adsorption, it is concluded that 12 water molecules are displaced by a phenobarbital molecule on the nonpolar (hydrocarbon) part of the activated carbon surface. The adsorption of phenobarbital from solution by activated carbons has been simulated by employing Molecular Dynamic (MD) Modeling. The predicted differential Gibbs free energy values for site-specific adsorption at pH 2-9 were consistent with the thermodynamic calculations. And, the present MD simulations provide a good basis for the further understanding and quantitatively assessment of the adsorption driven by hydrophobic bonding. The conclusions reached in the current studies are expected to be applicable to a wide range of similar adsorption processes.

The biodesulfurization 4S-pathway is a promising complementary enzymatic approach to remove sulfur from recalcitrant thiophenic derivatives in petroleum products that remain from conventional hydrodesulfurization method without diminishing the calorific value of oil. The final step of this pathway involves the carbon-sulfur bond cleavage from HBPS, and the production of the final products 2-hydroxybiphenyl (HBP) and sulfite, has been recognized as the rate-limiting step, partially as a result of product inhibition. However, the mechanisms and factors responsible for product inhibition in the last step have not been fully understood. In this work, we proposed a computational investigation using molecular dynamic simulations and free energy calculations on 2’-hydroxybiphenyl-2-sulfinate (HBPS) desulfinase (DszB) with different bound ligands as well as different solvent conditions to develop a fundamental understanding of the molecular-level mechanism responsible for product inhibition. Based on available crystal structures of DszB and biochemical characterization, we proposed a “gate” area close to substrate binding site of DszB is responsible for ligand egress and plays a role in product inhibition. We have conducted biphasic molecular dynamic simulations to evaluate the proposed gate area functionality. Non-bonded interaction energy analysis shows that hydrophobic residues around the gate area produce van der Waals interactions inhibiting translocation through the gate channel, and therefore, the molecules are easily trapped inside the binding site. Umbrella sampling molecular dynamics was performed to obtain the energy penalty associated with gate conformational change from open to close, which was 2.4 kcal/mol independent of solvent conditions as well as bound ligands. Free energy perturbation calculations were conducted for a group of six selected molecules bound to DszB. The selections were based on functional group representation and to calculate binding free energies that were directly comparable to experimental inhibition constants, KI. Our work provides a fundamental molecular-level analysis on product inhibition for the biodesulfurization 4S-pathway.

A new method is proposed for correctly modeling the long range interaction between a fluid and a bounding wall in atomistic simulations. This method incorporates the molecular structure of the solid substrate while allowing for a finite interaction cutoff by making a proper estimation of long range correction for the fluid-wall interaction. The method is then applied to a molecular dynamic simulation of a spreading droplet. Conparison to simulations using several other previously used methods shows that the long range correction can be significant in some circumstances. / Singapore-MIT Alliance (SMA)

In this study, we used the molecular statics, molecular dynamics, and density function theory to investigate structural, electronic, and mechanical properties of ultra-thin silica nanowires. There are two parts in this study. In the first part, we used basin-hopping method to get different diameters of silica nanowires, nemed 2MR, 2MR-2O, 3MR-3O, 4MR-4O, 5MR-5O, 4MR-3f, 4MR-4f, and 4MR-5f. The various silica nanowires were optimized by density function theory to obtain the projected density of states, Mulliken charge, and electronic density difference, and we also compared this results to £\-quartz. In the second part, the molecular dynamics simulations were performed to investigate deformation behavior of silica nanowires under axial tensile loading at 10K. The Young¡¦s modulus increases when the diameter decreases. We also used angular correlation function to study the mechanical properties and variation of structures.

For decades, epitaxy is used in nanotechnologies and semiconductor fabrications. So far, it's the only affordable method of high quality crystal growth for many semiconductor materials. Heterostructures developed from these make it possible to solve the considerably more general problem of controlling the fundamental parameters inside the semiconductor crystals and devices. Moreover, as one newly arising study and application branch of epitaxy, selective area growth (SAG) is widely used to fabricate materials of different thicknesses and composition on different regions of a single wafer. All of these new and promising fields have caught the interests and attentions of all the researchers around the world. In this work, we will study the stress and strain analysis of epitaxy in nano-scale materials, in which we seek a methodology to bridge the gap between continuum mechanical models and incorporate surface excess energy effects, which can be obtained by molecular dynamical simulations. We will make a brief description of the elastic behavior of the bulk material, covering the concepts of stress, strain, elastic energy and especially, the elastic constants. After that, we explained in details about the definitions of surface/interface excess energy and their characteristic property tensors. For both elastic constants and surface excess energy, we will use molecular dynamic simulations to calculate them out, which is mainly about curve-fitting the parabola function between the total strain energy density and the strain. After this, we analyzed the stress and strain state in nanoisland during the selective area growth of epitaxy. When the nanoisland is relaxed, the lattice structure becomes equilibrated, which means the total strain energy of system need to be minimized. Compared to other researcher's work, our model is based on continuum mechanics but also adopts the outcome from MD simulations. By combining these microscopic informations and those macroscopic observable properties, such as bulk elastic constants, we can provide a novel way of analyzing the stress and strain profile in epitaxy. The most important idea behind this approach is that, whenever we can obtain the elastic constants and surface property tensors from MD simulations, we can follow the same methodology to analyse the stress and strain in any epitaxy process. This is the power of combining atomistic simulations and continuum method, which can take considerations of both the microscopic and macroscopic factors.