Reactive Molecular Dynamic Simulations of Network Polymers:Generation, Characterization and Mechanical Properties.

Network;Percolation;Engineering;Materials Science and Engineering

Shankar, ChandrashekarLahann, Joerg ;

University of Michigan

关键词: Network;    Percolation;    Engineering;    Materials Science and Engineering;   

Others  :  https://deepblue.lib.umich.edu/bitstream/handle/2027.42/62243/cshankar_1.pdf?sequence=1&isAllowed=y

瑞士|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

The goal of this research was to gain a fundamental understanding of the properties of networks created by the ring opening metathesis polymerization (ROMP) of dicyclopentadiene (DCPD) used in self-healing materials.To this end we used molecular simulation methods to generate realistic structures of DCPD networks, characterize their structures, and determine their mechanical properties.Density functional theory (DFT) calculations, complemented by structural information derived from molecular dynamics simulations were used to reconstruct experimental Raman spectra and differential scanning calorimetry (DSC) data. We performed coarse-grained simulations comparing networks generated via the ROMP reaction process and compared them to those generated via a RANDOM process, which led to the fundamental realization that the polymer topology has a unique influence on the network properties.We carried out fully atomistic simulations of DCPD using a novel algorithm for recreating ROMP reactions of DCPD molecules. Mechanical properties derived from these atomistic networks are in excellent agreement with those obtained from coarse-grained simulations in which interactions between nodes are subject to angular constraints.This comparison provides self-consistent validation of our simulation results and helps to identify the level of detail necessary for the coarse-grained interaction model.Simulations suggest networks can classified into three stages: fluid-like, rubber-like or glass-like delineated by two thresholds in degree of reaction α: The onset of finite magnitudes for the Young’s modulus, αY, and the departure of the Poisson ration from 0.5, αP.In each stage the polymer exhibits a different predominant mechanical response to deformation. At low α < αY it flows. At αY < α < αP the response is entropic with no change in internal energy. At α > αP the response is enthalpic change in internal energy.We developed graph theory-based network characterizations to correlate between network topology and the simulated mechanical properties. 1) Eigenvector centrality 2) Graph fractal dimension 3) Fiedler partitioning and 4) Cross-link fraction (Q3+Q4). Of these quantities, the Fiedler partition is the best characteristic for the prediction of Young’s Modulus.The new computational tools developed in this research are of great fundamental and practical interest.