Modeling Polystyrene With and Without Confinement: Exploring the Limits of Iterative Boltzmann Inversion

Polymers are present in every aspect of our lives due to their versatile applications as plastics, rubber, adhesives and gels. Understanding the structure-property relationship of these kind of molecules is crucial because of the fact that it determines the extent of our ability to engineer materials on molecular level and thus create new materials possessing different properties. Molecular dynamics simulations, being complementary to experiments, serve as a great tool to study structural and dynamic properties of polymers. Modeling polymers, however, is a challenging task due to the enormous variety of length and time scales involved in the description of their structural and dynamic properties. Atomistic simulations are needed to describe the polymer systems on a local scale, but can only handle time scales of up to a few hundred nanoseconds. The use of mesoscale models, which link the length and time scales in higher resolution to those in coarser-level representations, is essential, as a single modeling approach cannot capture all relevant properties.;The objective of this dissertation is twofold: we have investigated the effect of solvation, concentration and confinement level on the structure and dynamics of polystyrene (PS) through detailed atomistic molecular dynamics simulations and explored the limits of a structure based coarse-graining technique, Iterative Boltzmann Inversion (IBI), on mesoscale modeling of PS in different environments. Four types of systems are studied: melt, dilute solution, concentrated solution, and confined concentrated solution at different confinement levels. IBI has been developed for polymer melts and solutions, but many important applications of polymers such as protective coatings, adhesives, lubricants, etc., involve close contact with solid surfaces. Modeling polymer systems confined between surfaces by IBI is more challenging compared to modeling bulk systems, as the system is locally in different state points due to the density fluctuations near the surfaces in the former case. We address these challenges and technical details of the coarse-grained (CG) force field development process for PS in different environments. We show that all models reproduce the local structure of PS in their corresponding environments very well, while the global structure is slightly overestimated. The efficiencies of the CG models are captured by end-to-end vector autocorrelation functions, mean square displacements and speed-up factors for each system through the comparison of their corresponding chain diffusion coefficients and characteristic relaxation times of the end monomers. Results show that speed-up increases as the concentration of PS and confinement level increases. Differences in the local and global structure of PS in different environments are addressed by means of bonded/non-bonded conformational distributions, end-to-end distances, radii of gyration, and persistence lengths.