Dans le cadre du renouvellement du parc nucléaire, six modèles de réacteurs de 4ème génération ont été proposés, dont le Réacteur à Sels Fondus. Ce réacteur a la particularité d’utiliser un combustible à base de fluorures fondus, type LiF-ThF4. Pour développer ce concept, il est important de caractériser d’un point de vue structural ces mélanges de fluorures fondus, pour remonter aux propriétés physico-chimiques du combustible et optimiser ce procédé. Les systèmes fondus MF-ZrF4 (M = Li, Na, K), choisis comme modèle des systèmes au thorium, ont été étudiés expérimentalement par Résonance Magnétique Nucléaire et Absorption des Rayons X à hautes températures, ainsi que par calcul de dynamique moléculaire, en s’intéressant plus particulièrement aux environnements locaux du fluor et du zirconium. Afin d’interpréter les données RMN recueillies dans les milieux fondus, une étude préliminaire sur des halogénures de zirconium et des fluorozirconates de terres rares et d’alcalins solides a été menée par RMN du zirconium 91Zr et des corrélations structures/ paramètres RMN ont été établies. A haute température, dans les systèmes MF-ZrF4 on montre la coexistence de différents complexes du zirconium, avec des coordinences comprises entre 6 et 8, leurs proportions évoluant en fonction de la teneur en ZrF4 du mélange, et du type de l’alcalin. En fonction de la teneur en fluorure de zirconium, nous avons mis en évidence le rôle du fluor dans le bain fondu : fluor libre à faible teneur, il intervient progressivement dans la formation des complexes pour devenir pontant à plus haute teneur. Cette approche originale et innovante des systèmes fluorés fondus, combinant RMN et EXAFS à haute température, avec les calculs de dynamique moléculaire, s’avère particulièrement efficace pour leur description structurale, permettant ainsi de remonter à des données fondamentales, telles que leur spéciation ou leur fluoroacidité. / In the frame of the renewal of the different nuclear plans, the molten salt reactor is one of the six concepts of reactors of IVth generation. This reactor has the particularity to use a liquid fuel based on LiF-ThF4 mixtures. In order to develop and to optimize this concept, it is important to characterize the structure of the melt and to describe its physical and chemical properties. Our work has been based on the study of the system MF-ZrF4 (M = Li, Na, K) selected as a model of ThF4 based systems. We have combined two spectroscopic techniques, the Nuclear Magnetic Resonance and the X-ray Absorption at high temperature, with molecular dynamics calculations. We particularly focused on the local environnements of the fluorine and the zirconium. In order to interpret the NMR data obtain in the molten state, we performed a preliminary study on zirconium halides and rare earth and alkali fluorozirconates using the 91Zr solid-state NMR at very high magnetic fields. New correlations between structural parameters and NMR data have been established. At high temperature, in MF-ZrF4 melts we have shown the coexistence of three different kind of Zr-based complexes with different proportions depending on the amount of ZrF4 and on the nature of the alkali. Depending on the ZrF4 content, three kinds of fluorine have been characterized: form free fluorines at low amount of zirconium fluorides, fluorines involved in Zr-based complexes and bridging fluorines at higher ZrF4 content. This original and innovative approach of molten fluorides mixtures, combining NMR and EXAFS at high temperature with molecular dynamics calculations, is very efficient to describe their speciation and thus their fluoro-acidity.

Computer simulation techniques for exploring the microscopic world are quickly gaining popularity as a tool to complement theoretical and experimental approaches. Molecular dynamics (MD) simulations allow the motion of an N–body soft matter system to be solved using a classical mechanics description. The scope of these simulations are however limited by the available computational power, requiring the development of multiscale methods to make better use of available resources. Dual scale models are a novel form of molecular model which simultaneously feature particles at two levels of resolution. This allows a combination of atomistic and coarse-grained (CG) force fields to be used to describe the interactions between particles. By using this approach, targeted details in a molecule can be described at high resolution while other areas are treated with fewer degrees of freedom. This approach aims to allow for simulating the key features of a system at a reduced computational cost. In this thesis, two generations of a methodology for constructing dual scale models are presented and applied to various materials including polyamide, polyethene, polystyrene and octanol. Alongside a variety of well known atomistic force fields, these models all use iterative Boltzmann inversion (IBI) force fields to describe the CG interactions. In addition the algorithms and data structures for implementing dual scale MD are detailed, and expanded to include a multiple time step (MTS) scheme for optimising its peformance. Overall the IBI and atomistic force fields were compatible with each other and able to correctly reproduce the expected structural results. The first generation methodology featured bonds directly between atoms and beads, however these did not produce the correct structures. The second generation used only atomistic resolution bonds and this improved the intramolecular structures greatly for a relatively minor cost. In both the polyamide and octanol systems studied, the models were also able to properly describe the hydrogen bonding. For the CG half of the force field, it was possible to either use preexisting force field parameters or develop new parameters in situ. The resulting dynamical behaviour of the models was unpredictable and remains an open question both for CG and dual scale models. The theoretical performance of these models is faster than the atomistic counterpart because of the reduced number of pairwise interactions that must be calculated and this scaling was seen with the proposed reference implementation. The MTS scheme was successful in improving the performance with no effects on the quality of results. In summary this work has shown that dual scale models are able to correctly reproduce the structural behaviour of atomistic models at a reduced computational cost. With further steps towards making these models more accessible, they will become an exciting new option for many types of simulation.

University of Macau / Faculty of Science and Technology / Department of Computer and Information Science

A variational principle is applied to the Schroedinger equation for theXY₂ linear molecule. Trial solutions are synthesized from the nuclear eigenstates, which are assumed to be simple harmonic oscillator eigenstates, and from the unperturbed electronic states, whose azimuthal dependence is known because of the cylindrical symmetry of the field of the nuclei. The secular equation is discussed, and an expression for the Renner splitting of the π state is obtained. / Science, Faculty of / Physics and Astronomy, Department of / Graduate

In most barrier applications, both mechanical and diffusion properties of the material are important. In this thesis the evaluation of molecular mechanisms responsible for the enhanced elastic properties of Polymer Clay Nanocomposites (PCNs) and the molecular mechanisms of Oxygen diffusion in PolyButylene Terephthalate polymer are presented. Interaction energy between PCN constituents, conformational changes of polymer, interaction energy between Oxygen molecule and polymer, rate of Oxygen and Oxygen diffusion coefficient are evaluated. Molecular simulation studies of PolyButylene Terephthalate (PBT) clay nanocomposite and Nylon6 clay nanocomposite show that a higher crystallinity polymer such as PBT would require higher attractive and repulsive interactions with organic modifier in order to make significant change in the crystallinity of PBT in the nanocomposite and in turn enhance the elastic modulus and hardness. Molecular interactions energy between Oxygen molecule and polymer, change in polymer conformation caused by thermal energy assist the Oxygen molecule to diffuse through polymer. / NSF-EPSCoR FlexEM Grant

Using molecular dynamics (MD) and boundary element method (BEM), different aspects of solidification in the aluminum silicon system are studied. The angular embedding atom model (AEAM) was implemented on LAMMPS, and the necessary potentials are developed. Firstly, a modified version of the Stillinger-Weber (SW) interatomic potential for pure Si is proposed. The advantage of this potential is that, in contrast to the original SW form, the modified version allows one to grow diamond cubic crystal structures from the melt at high temperatures. Additionally, an Al-Si binary potential of the AEAM type is able to accurately predict the experimental enthalpy of mixing. It is also able to predict an Al-Si phase diagram with a eutectic concentration for the liquid that agrees with experiment within 4 at% and a eutectic temperature that differs from experiment by only 13 K. Considering the importance of step mobility and step free energy on the solidification growth rate, chapters 3 and 4 are devoted to calculation of these concepts using MD simulations. In chapter 3 the step mobility, which is the proportionality constant between the velocity and driving force, was determined for the alloy with melt composition of Al-90%Si as a function of temperature and composition. It was found that mobility decreases fairly rapidly with the addition of Al solute. Also, from the variation with temperature, it appears the mobility is proportional to the interdiffusion coefficient in the liquid. It is observed that for the Al-60%Si alloy diffusion-controlled growth is the dominant scenario, even for a few degrees of undercooling. In chapter 4 equilibrium molecular dynamics (MD) simulations and the capillaryfluctuations method (CFM) are employed to calculate crystal-melt step free energies at three different melt compositions. Anisotropy of steps are investigated by setting up the systems with different crystal orientations of steps on the high-symmetry interface plane, (111) in this case. A complete isotropy of step free energy is observed for Al-60%Si and Al-90%Si alloying systems, while CFM failed in determining step free energy in Al-30%Si due to lack of step roughness. In chapter 5 the BEM is utilized to numerically compute the concentration profile in a fluid phase in contact with an infinite array of equally spaced surface steps. In addition, under the assumption that step motion is controlled by diffusion through the fluid phase, the growth rate is computed and the effect of step spacing, supersaturation and boundary layer width is studied. BEM calculations were also used to study the phenomenon of step bunching during crystal growth and it is found that, in the absence of elastic strain energy, a sufficiently large perturbation in the position of a step from its regular spacing will lead to a step bunching instability. / Thesis / Doctor of Philosophy (PhD)

Polymers and polymer mixtures play such important roles in our lives that it is hard to imagine life without them. Although a lot of progress has been made in the past few decades in our understanding of polymer dynamics and rheology using theoretical, computational, and experimental approaches, there are significant gaps in what is still left to be done. For example, polydispersity is the norm in industrially-produced polymers. However, a lot of the theories that have been propounded are predicated on monodisperse polymers. Most recently, molecular simulations have significantly improved our understanding of polymer dynamics by either validating theories or offering novel insights into their dynamics at the micro-structure level of detail. This thesis also uses molecular simulation to explore three different classes of polymer systems – monodisperse, polydisperse, and polymer-additive mixtures. Two of the most significant theories of polymer dynamics are the Rouse model describing the dynamics of short and unentangled chains, and the tube and reptation model, which describes the motion of long, entangled chains. The reptation model predicts different dynamical regimes that are marked by distinct time scales and further introduced the concept of an entanglement length Ne – which signifies the length scale at which topological interactions between a test chain and surrounding matrix chains become significant. These distinct time scales have been investigated by various researchers, but the reported values vary with different groups. Here, we devised a protocol for the accurate determination of these time scales. We also calculated Ne using these time scales and compared our results with those reported in the literature. Furthermore, using Rouse mode analysis – a technique that resolves the coupled motion of monomers into distinct and uncoupled modes, we showed that the method can be used to determine Ne, with the values obtained closer to that using monomer displacement at longer time scales. The computational calculation of the linear viscoelastic (LVE) properties of polymers is very expensive due to the long relaxation times associated with polymers. Using three different methods – Equilibrium molecular dynamics (EMD) via the Green-Kubo relation, non-equilibrium molecular dynamics (NEMD), and corrected Rouse Mode Analysis (cRMA), we determined the LVE properties of polymer melts for both unentangled and entangled chains and compared the uncertainty associated with each of them. Specifically, we demonstrate that the cRMA method, although applicable to only the shorter chains, yielded the lowest uncertainty. Compared to earlier reported results, we also show that, using shorter computational runs, the NEMD gives an acceptable level of accuracy in the calculation of the LVE properties. Beyond methodology development, this thesis also studied the viscoelasticity of realistic polymer systems of practical interest, including polydisperse polymers and polymer-additive mixtures. We idealized a polydisperse system using a bidisperse model wherein we combined longer chains with shorter chains at different concentration levels. By investigating the individual motion of the different chains, we showed that the dynamics of the longer chains are sped up by the shorter chains, whereas the longer chains impeded the dynamics of the shorter chains, although the nature of the dynamics of the shorter chains was not altered. We also showed that the shorter chains reduce the extent of the entanglement effect of the longer chains using an RMA approach. By using the individual LVE profiles of the long and short chains, we tested a semi-empirical mixing rule for predicting the stress relaxation modulus G(t) of the bulk mixture. We found that a simple mixing rule works well when the shorter chains are the majority while the double reptation model – that assumes a simultaneous relaxation of both the test chains and surrounding matrix chains, does a better job of the prediction when the longer chains are the majority. We further explored the effect of molecular structure of single-bead additives on the thermo-physical properties, such as the glass transition temperature and Young’s modulus on the polymer-additive mixture. By varying the dimensions of the molecular additive over a range of concentration of the molecular additives, we found that smaller-sized additives are better able to reduce the glass transition temperature Tg and increase the Young’s modulus Y of the mixture due to the improved packing efficiency. On the other hand, larger-sized particles are only marginally able to reduce the Tg. The LVE properties, specifically the zero-shear viscosity of the mixtures showed an opposite trend to the Y, where once again the smaller-sized particles better reduce the zero-shear viscosity. This essentially shows the decorrelation of traditional plasticization markers. A reduction in one property does not imply a reduction in another property and this varies with the dimensions and concentration of the additives. We finally describe our various attempts at developing a multi-bead plasticizer model. For the models we have tried, we tuned the chain lengths of the beads and their interaction with the polymer. Detailed micro-structure analysis and viscoelasticity calculations reveal that they are either incompatible with the polymer, resulting in phase separation or only marginally compatible over a very limited range. / Thesis / Doctor of Engineering (DEng)

No description available.

Single-molecule stretching experiments reveal a distinct plateau region in force-extension curves of nucleic acids such as long double-stranded deoxyribonucleic acids (DNA) and ribonucleic acids (RNA). The dissertation comprises two parts. In the first part, we propose an approach to help analyze polymer force-extension curves that exhibit a distinct plateau region. When coupled to a bead-spring dynamic model, the approach qualitatively reproduces a variety of experimental force-extension curves of long double-stranded (ds) DNA and RNA, including torsionally constrained and unconstrained DNA, and negatively supercoiled DNA. In the plateau region of the force-extension curves, our molecular dynamics simulations show that the polymer separates into a mixture of slightly and highly stretched states without forming macroscopically distinct phases. In the second part, we hypothesize that, depending on the sequence composition, multiple distinct plateau regions can be seen in force-extension curves of long dsDNA fragments under physiological solvent conditions. We explore specific long double-stranded DNA sequences where we expect the phenomenon to occur, and to characterize the distribution of states along the polymer. Our molecular dynamics simulations show that multi-plateau regions are observed in the force-extension curves of specific long double-stranded DNA fragments. The formation of mixed states of slightly and highly stretched DNA, co-existing with macroscopically distinct phases in several segments in the plateau regions, is also predicted. / Doctor of Philosophy / Single-molecule stretching experiments reveal a distinct plateau region in force-extension curves of nucleic acids such as long double-stranded DNA and RNA. In this dissertation, we propose a simple bead-spring dynamic model that qualitatively reproduces a variety of experimental force-extension curves of long double-stranded DNA and RNA that exhibit a distinct plateau region. Based on the model, we make multiple predictions. In particular, we predict that multiple distinct plateau regions can be seen in force-extension curves of long composite double-stranded DNA fragments under physiological solvent conditions. We believe our findings should be of interest to both experimentalists and theoreticians. Experimentalists might find our model useful for routine analysis of force-extension curves of double-stranded DNA and RNA. Theoreticians may appreciate our general approach.

The presence of ion-pairs in electrolyte solutions affects the activity of dissolved species as well as the solubility of minerals. The extent of ion-pairing in a system is predicted by an association constant, K[subscript A], which for sparingly soluble salts are frequently determined experimentally in binary or ternary systems. This introduces complex activity coefficient calculations that often require unavailable parameters. Barite (BaSO₄) and celestite (SrSO₄) are sparingly soluble minerals with interest in the oil and mining industry, yet the values of K[subscript A] for the ion-pairs BaSO₄(aq.) and SrSO₄(aq.) are still uncertain. Molecular dynamics simulations are used to obtain the K[subscript A] values for these two salts through potential of mean force (PMF) calculations. The molecular mechanisms involved in the association reactions are also explored, in particular the role of the association intermediates in the overall reaction as described by the Eigen mechanism. Additionally, the kinetics of water exchange around the free and paired ions is examined and the residence time of a water coordinated to the free and paired cation is calculated.