A Particle-Based Multi-Scale Simulation Procedure within the Material Point Method Framework

Recent studies of nano energetic composites and nanoelectromechanical systems (NEMS) have underscored the need for an effective multiscale procedure for simulating the responses of discrete nano and sub-micron structures and assemblies to various extreme loading conditions. In this dissertation, a particle-based multi-scale simulation procedure is proposed with a concurrent link between the Dissipative Particle Dynamics (DPD) method and the Material Point Method (MPM), and a hierarchical bridge from Molecular Dynamics (MD) to DPD. Particularly, a simple interfacial treatment is also introduced for concurrent DPD/MPM simulations based on the features of the DPD force expression and the MPM constitutive model. First, to gain a fundamental understanding of deformation mechanisms at the atomic-scale and provide baseline results for the multiscale simulation, atomistic modelling and simulation of discrete metallic nanostructures under various loading conditions are performed with MD and analyzed using Common Neighbor Analysis (CNA). The loading conditions simulated include basic mechanical loading (tension/torsion/bending), impact loading (transverse and longitudinal impact), and hydrodynamic loading. The detailed deformation mechanisms and mechanical response of these nanostructures are investigated, and the formation of some special nanostructures (such as single and multiple fivefold twins, and icosahedral structures) are observed. Then, the proposed multi-scale procedure is illustrated using similar simulations of the dynamic and impact responses of discrete metallic nano structures, as well as rod/particle assemblies in a hydrostatic fluid system. It is shown that the DPD forces can be effectively coarse-grained using the MPM background grid, and that the concurrent link between the MPM and DPD enables near-seamless integration of constitutive modeling at the continuum level with force-based modeling at the mesoparticle level. Finally, it is shown that the multi-scale simulation procedure proposed in this work requires much less computational time than the MD simulations to simulate a similar problem.

Name thesis:degree_name

Ph. D.

Level thesis:degree_level

Doctoral

Discipline thesis:degree_discipline

Civil and environmental engineering (MU)

Grantor dc:publisher

University of Missouri--Columbia

Year dc:date.issued

2014

Author dc:creator

• Jiang, Shan

Advisor dc:contributor.advisor

• Chen, Zhen, 1958-

Statement dc:rights

• OpenAccess.

Language dc:language.iso

eng, English

Handle dc:identifier.uri

https://hdl.handle.net/10355/45879

OAI identifier oai:identifier

oai:mospace.umsystem.edu:10355/45879