Back to search

South Dakota State University

A Multiscale Computational Framework Bridging Molecular Dyanmics, Phase-Field, and Elasto–Viscoplstic Fast Fourier Transform: Applications to Process–Structure–Property Relationships in Metallic Materials

Abstract

dc:description.abstract

Metallic materials exhibit process–structure–property relationships that arise from physical mechanisms operating across multiple length and time scales. Atomistic processes control diffusion, thermal response, and interfacial behavior; mesoscale mechanisms govern grain morphology, interface migration, and microstructure evolution; full-field crystal-plasticity calculations resolve heterogeneous mechanical fields; and continuum-scale finite-element (FE) models connect material response to component-level deformation, damage, and performance. Because no single simulation method can resolve all of these mechanisms, predictive modeling requires a framework in which information obtained at one scale can be transferred to the next. This dissertation develops a multiscale computational framework that bridges molecular dynamics (MD), phase-field (PF) modeling, and elasto-viscoplastic fast Fourier transform-based (EVPFFT) crystal plasticity simulations for metallic materials. At the atomistic scale, MD simulations of the aluminum–silicon (Al–Si) system are used to compute thermophysical, interfacial, and kinetic quantities relevant to aluminum foam systems, including self-diffusion coefficients, specific heat capacity, solid–liquid interfacial energy, and grain-boundary mobility. These quantities provide physically informed inputs for PF formulations and establish the atomistic-to-mesoscale connection within the framework. The PF method is then used to describe grain growth and recrystallization and is coupled with EVPFFT-based crystal plasticity to simulate static recrystallization (SRX) in three-dimensional polycrystalline copper. In the coupled formulation, the PF model describes nucleation, grain growth, and grain-boundary migration, while the EVPFFT solver computes the heterogeneous stress, strain, and stored deformation-energy fields generated during plastic deformation. A subcycling strategy exchanges grain identity, crystallographic orientation, and mechanical-field information between the two solvers as the simulation proceeds, allowing the mechanical response to be updated as recrystallization proceeds and the microstructure evolves. Overall, this dissertation demonstrates how atomistic, mesoscale, and full-field mechanical modeling approaches can be integrated into a unified computational framework for the design and analysis of metallic materials. The Al–Si system is used to establish the MD-informed PF parameterization, while copper is used to develop and validate the coupled PF–EVPFFT recrystallization model. The framework provides a foundation for studying microstructural evolution, recrystallization kinetics, mechanical-field redistribution, and process–structure–property relationships, with future extensions toward texture evolution, dynamic recrystallization (DRX), and thermomechanical processing.

Degree

thesis:*
Name thesis:degree_name
Doctor of Philosophy (PhD)
Level thesis:degree_level
Dissertation - Open Access
Discipline thesis:degree_discipline
Mechanical Engineering

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Jouhari, Chaimae
Contributors dc:contributor
  • Yucheng Liu

Rights

dc:rights

Identifiers

dc:identifier.*
Repository record dc:identifier
https://openprairie.sdstate.edu/etd2/2125

Chain of custody

source
Harvested from
South Dakota State University
Base URL
openprairie.sdstate.edu/do/oai/
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Jouhari, Chaimae. A Multiscale Computational Framework Bridging Molecular Dyanmics, Phase-Field, and Elasto–Viscoplstic Fast Fourier Transform: Applications to Process–Structure–Property Relationships in Metallic Materials. Dissertation - Open Access thesis, https://openprairie.sdstate.edu/etd2/2125