eprintid: 10204809 rev_number: 12 eprint_status: archive userid: 699 dir: disk0/10/20/48/09 datestamp: 2025-03-07 13:14:03 lastmod: 2025-03-07 13:14:03 status_changed: 2025-03-07 13:14:03 type: thesis metadata_visibility: show sword_depositor: 699 creators_name: Müller, Werner title: The Multiscale Simulation of Graphene Polymer Nanocomposites ispublished: unpub divisions: UCL divisions: B04 divisions: C06 divisions: F56 note: Copyright © The Author 2025. Original content in this thesis is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) Licence (https://creativecommons.org/licenses/by-nc/4.0/). Any third-party copyright material present remains the property of its respective owner(s) and is licensed under its existing terms. Access may initially be restricted at the author’s request. abstract: Through high-performance computing, we can simulate and accurately predict the properties of materials at a fraction of the time and cost of an experimental study. Moreover, we can select the appropriate resolution and, on the microscopic scale, understand the structure that may not have been practical empirically. This project is a UCL IMPACT PhD studentship co-funded by Hexcel, a company that develops and manufactures composite materials. By combining simulation techniques of different time and length scales, this thesis aims to predict the macroscopic properties of graphene-based nanocomposites commonly used in paints, coatings and adhesives in the aerospace industry. The primary tool used for simulation in this thesis is molecular dynamics. While molecular dynamics simulations can accurately predict numerous material properties, independent simulations can converge to different metastable states due to the inherent randomness associated with the initial positions and velocities, even averaging over long timescales. For the simulations to be accurate and reproducible, the study of verification, validation, and uncertainty quantification is an integral part of this research, as only through averages over ensembles of molecular simulations can robust and actionable theoretical predictions be yielded. This thesis covers four distinct scenarios relating to the modelling of nanocomposite materials. First, we present a study on the mechanical stability of a graphene polymer bilayer used in capacitive micromachined ultrasound transducers through repeated loading and unloading cycles. We then propose a new, faster method to obtain the glass transition temperature more suited to parallel supercomputers. We perform a rigorous statistical analysis comparing this new parallel method to the commonly used step-wise method. We follow this by investigating the effect of water molecules on the elastic moduli of graphene epoxy nano-composites. We conclude by developing a machine learning surrogate for a multiscale scheme that couples the finite element to molecular dynamics. date: 2025-02-28 date_type: published oa_status: green full_text_type: other thesis_class: doctoral_open thesis_award: Ph.D language: eng primo: open primo_central: open_green verified: verified_manual elements_id: 2360570 lyricists_name: Muller Roa, Werner lyricists_id: WMULL37 actors_name: Muller Roa, Werner actors_id: WMULL37 actors_role: owner full_text_status: public pages: 137 institution: UCL (University College London) department: Chemistry thesis_type: Doctoral citation: Müller, Werner; (2025) The Multiscale Simulation of Graphene Polymer Nanocomposites. Doctoral thesis (Ph.D), UCL (University College London). Green open access document_url: https://discovery.ucl.ac.uk/id/eprint/10204809/2/Thesis.pdf