Multiscale Modeling of Polymers: Fracture Simulations, Constitutive Modeling and Crystallization Analysis
Skip to main content
eScholarship
Open Access Publications from the University of California

UC Berkeley

UC Berkeley Electronic Theses and Dissertations bannerUC Berkeley

Multiscale Modeling of Polymers: Fracture Simulations, Constitutive Modeling and Crystallization Analysis

Abstract

The dissertation investigates the complex behavior of polymers through computational mechanics,employing numerical techniques that include peridynamics (PD), finite element methods (FEM), molecular dynamics (MD), and deep learning. Our aim is to develop a deeper understanding of the relationship between the polymer microstructure and the resultant mechanical properties, in the context of fracture mechanics, large deformation problems, and additive manufacturing.

In the second chapter, we introduce a numerical framework to simulate the fracture response of elastomeric materials. The approach utilizes the bond-based peridynamics formulation, which is a nonlocal and meshfree alternative to the continuum-based methods. We introduced a novel bond potential into the framework, based on the non-Gaussian chain statisticstheory, which treats the material as a complex network of randomly jointed chains with extensible links. The resultant formulation is capable of capturing the fracture and large deformation process of elastomeric materials, as demonstrated in several numerical examples and comparisons with the experimental data.

The remaining chapters focus on the semicrystalline polymers that are used in additive manufacturing. In the third chapter, we have developed a constitutive law for crystalline polymers, based on molecular dynamics and machine learning. We have collected data from the MD simulations of Polyamide12 (PA12), which is used to train deep neural networks to capture the mechanical response of the material. We demonstrated that this novel approach can accurately provide a three-dimensional molecular-level anisotropic constitutive relation that can be used in macroscale mechanics methods such as the FEM. The final chapter consists of an analysis of the crystallization of polymers during the additive manufacturing process. By incorporating crystallization and melting models for PA12, we performed a heat transfer analysis using FEM to predict changes in crystallinity during the manufacturing process, which can help achieve the desired mechanical properties in the final product.

Main Content
For improved accessibility of PDF content, download the file to your device.
Current View