Date of Award
Doctor of Philosophy (PhD)
Chemical and Biomolecular Engineering
Dr. Christopher L. Kitchens
Dr. David A. Bruce
Dr. Amod A. Ogale
Dr. Srikanth Pilla
An ongoing challenge in advanced materials design is the development of accurate multiscale models that consider uncertainty while establishing a link between knowledge or information about constituent materials to overall composite properties. Successful models can accurately predict composite properties, reducing the high financial and labor costs associated with experimental determination and accelerating material innovation. Whereas early pioneers in micromechanics developed simplistic theoretical models to map these relationships, modern advances in computer technology have enabled detailed simulators capable of accurately predicting complex and multiscale phenomena.
This work advances domain knowledge via two means: firstly, through the development of high-fidelity, physics-based finite element (FE) models of composite microstructures that incorporate uncertainty in predictions, and secondly, through the development of a novel inverse analysis framework that enables the discovery of unknown or obscure constituent properties using literature data and Gaussian process (GP) surrogate models trained on FE model predictions. This work presents a generalizable approach to modeling a diverse array of composite subtypes, from a simple particulate system to a complex commercial composite.
The inverse analysis framework was demonstrated for a thermoplastic composite reinforced by spherical fillers with unknown interphase properties. The framework leverages computer model simulations with easily obtainable macroscale elastic property measurements to infer interphase properties that are otherwise challenging to measure. The interphase modulus and thickness were determined for six different thermoplastic composites; four were reinforced by micron-scale particles and two with nano-scale particles.
An alginate fiber embedded with a helically symmetric arrangement of cellulose nanocrystals (CNCs) was investigated using multiscale FE analysis to quantify microstructural uncertainty and the subsequent effect on macroscopic behavior. The macroscale uniaxial tensile simulation revealed that the microstructure induces internal stresses sufficient to rotate or twist the fiber about its axis. The reduction in axial elastic modulus for increases in CNC spiral angle was quantified in a sensitivity analysis using a GP surrogate modeling approach.
A predictive model using GP regression was employed to investigate the link between input features and the mechanical properties of fiberglass-reinforced magnesium oxychloride (MOC) cement boards produced from a commercial process. The model evaluated the effect of formulation, crystalline phase compositions, and process control parameters on various mechanical performance metrics.
Arp, Joshua, "Multiscale Modeling and Gaussian Process Regression for Applications in Composite Materials" (2023). All Dissertations. 3369.
Author ORCID Identifier