TY - BOOK

T1 - Multiscale Modelling and Simulation of Flow Behavior of Polymer/Layered Silicate Nanocomposites Under Shear Flow

AU - Gooneie, Ali

N1 - no embargo

PY - 2017

Y1 - 2017

N2 - Polymer nanocomposites (PNCs) display distinguished characteristics which originate from the interplay of phenomena at different length and time scales. Further development of these materials critically relies on a comprehensive understanding of the fundamentals of their hierarchical structure and behaviors. As such, their inherent multiscale nature is only reflected properly by a multiscale analysis which accounts for all important mechanisms. The fundamental concepts of multiscale simulations of polymeric materials along with relevant research outlooks are thoroughly addressed in this thesis. It is explained that in spite of all efforts, a framework for dynamic bridging of microstructure evolutions to macroscopic models had been hindered so far. In this PhD research, a theoretical framework was developed based on well-credited mesoscopic dissipative particle dynamics (DPD) models in order to propose a solution to this problem. First, the dynamic conformation change of linear polymer chains in response to startup of a steady shear flow was investigated. Second, the orientation patterns of anisometric layered silicate particles were studied under various shear flows. The influence of the interactions between layered silicates and polymer chains on the orientation process was carefully explored. Finally, the results of these works were incorporated to develop DPD models in order to build an upscaling method for the mesoscopic orientation patterns to the macroscopic flows. This upscaling method was tested successfully against the available standard orientation models from the literature. In this strategy, the trajectories of the orientation process of weakly-interacting layered silicates are parametrized as a function of the applied shear strain instead of the time, based on the experiments which propose strain-dependent rather than time-dependent structural evolutions in such non-Brownian materials. Benefitting from the notion that the orientation kinetics is simply the rate of change with respect to strain rather than time, the applied strain was selected to pass the orientation parameters to an upper scale through a simple combination of affine and nonaffine deformations. This combination was pictured in its simplest form to be a random mixing of DPD unit cells (simulating nonaffine deformations) in a larger cell which distributes an affine deformation over the unit cells. It was noted that this strategy could be used to perform multiscale simulations of orientation process provided that the unit cells represent a precise description of the interactions between the components. A comparison of this methodology with the strain reduction factor model showed the success of the multiscale simulation of the evolution of orientation parameters against the applied shear strain. It was also shown that the method fails to capture the microstructure evolutions if the unit cells do not provide an accurate representation of the material. The remaining research challenges which must be overcome in order to improve, extend, and generalize the developed multiscale algorithm were addressed before closing the discussion.

AB - Polymer nanocomposites (PNCs) display distinguished characteristics which originate from the interplay of phenomena at different length and time scales. Further development of these materials critically relies on a comprehensive understanding of the fundamentals of their hierarchical structure and behaviors. As such, their inherent multiscale nature is only reflected properly by a multiscale analysis which accounts for all important mechanisms. The fundamental concepts of multiscale simulations of polymeric materials along with relevant research outlooks are thoroughly addressed in this thesis. It is explained that in spite of all efforts, a framework for dynamic bridging of microstructure evolutions to macroscopic models had been hindered so far. In this PhD research, a theoretical framework was developed based on well-credited mesoscopic dissipative particle dynamics (DPD) models in order to propose a solution to this problem. First, the dynamic conformation change of linear polymer chains in response to startup of a steady shear flow was investigated. Second, the orientation patterns of anisometric layered silicate particles were studied under various shear flows. The influence of the interactions between layered silicates and polymer chains on the orientation process was carefully explored. Finally, the results of these works were incorporated to develop DPD models in order to build an upscaling method for the mesoscopic orientation patterns to the macroscopic flows. This upscaling method was tested successfully against the available standard orientation models from the literature. In this strategy, the trajectories of the orientation process of weakly-interacting layered silicates are parametrized as a function of the applied shear strain instead of the time, based on the experiments which propose strain-dependent rather than time-dependent structural evolutions in such non-Brownian materials. Benefitting from the notion that the orientation kinetics is simply the rate of change with respect to strain rather than time, the applied strain was selected to pass the orientation parameters to an upper scale through a simple combination of affine and nonaffine deformations. This combination was pictured in its simplest form to be a random mixing of DPD unit cells (simulating nonaffine deformations) in a larger cell which distributes an affine deformation over the unit cells. It was noted that this strategy could be used to perform multiscale simulations of orientation process provided that the unit cells represent a precise description of the interactions between the components. A comparison of this methodology with the strain reduction factor model showed the success of the multiscale simulation of the evolution of orientation parameters against the applied shear strain. It was also shown that the method fails to capture the microstructure evolutions if the unit cells do not provide an accurate representation of the material. The remaining research challenges which must be overcome in order to improve, extend, and generalize the developed multiscale algorithm were addressed before closing the discussion.

KW - computer simulations

KW - computational methods

KW - dissipative particle dynamics

KW - multiscale modelling

KW - hierarchical structures

KW - polymer nanocomposite

KW - layered silicate

KW - microstructure evolution

KW - morphology

KW - orientation

KW - flow field

KW - Computersimulationen

KW - Rechenverfahren

KW - dissipative Partikeldynamik

KW - Multiskalen Modellierung

KW - hierarchische Strukturen

KW - Polymer-Nanokomposit

KW - Schichtsilikat

KW - Wachstum der Mikrostrukturen

KW - Morphologie

KW - Orientierung

KW - Strömungsfeld

M3 - Doctoral Thesis

ER -