Computational modelling of polymer network formation
Repository URI
Repository DOI
Change log
Authors
Abstract
There exists a number of methods for the computational simulation of polymerisation processes. This thesis provides an overview of these methods, and establishes a connection between them on a theoretical level, based on the First Shell Substitution Effect (FSSE) model of polymerisation. The FSSE framework was used to evaluate a coarse-grained simulational method, benchmarked against a fine-grained, atomistic model that was constructed using Reactive Molecular Dynamics (RMD).
A modified version of RMD was implemented using a non-reactive force field, and used it to estimate the physical properties of a 3D printing thermosetting-thermoplastic copolymer, in order to validate the method. When comparing to experimental results, we found a good matching for the glass transition temperature, but inaccuracies when estimating elastic properties. We believe that this is due to the limited system size we had to resort to, when using the computationally intensive fully atomistic models.
Next, we assessed whether the FSSE model can capture the most relevant properties of the forming polymer network in step-growth polymerisation. As model systems, epoxy-amine copolymers were used, with different monomer functionalities of the hardener molecules. Using numerical simulations on a graph, which is essentially a simulated random process where edges are step-wise added to it, we showed that the FSSE model is indeed capable of reproducing the results of the benchmark RMD simulations, in terms of network evolution. A core property within this model is the monomer reactivity, which is a function of both the monomer degree and the overall functional group conversion. The conversion dependent monomer reactivities provide a basis for an intuitive comparison of different methods modelling the polymerisation process.
Finally, we implemented a novel coarse-graining method, Trajectory Matching, which can be used to scale up simulations. Using the FSSE framework, we looked at whether the coarse-grained model can reproduce the dynamics of network evolution throughout the polymerisation process. We used identical epoxy-amine systems as for the FSSE study, and found good matching between the results of coarse-grained DPD, and benchmark RMD simulations. We also showed that steric effects only become significant for higher degree monomers and at high conversions, while the reactivity of low degree monomers are diffusivity controlled.
Description
Date
Advisors
Keywords
Qualification
Awarding Institution
Rights
Sponsorship
Engineering and Physical Sciences Research Council (1948669)