Research

Thermo-Mechanics and Failure Analysis of Epoxy Resins

Epoxy, a thermosetting polymer, is widely used in coatings, adhesives, electronics packaging, and high-performance composite structural applications. Cured or cross-linked epoxies are usually made from the chemical reactions of two components – epon and amine curing agents. A wide range of thermo-mechanical properties for the composite matrix can be achieved by controlling the processing conditions as well as through the incorporation of various curing agents and tougheners into the Epon formulation. Using molecular dynamics (MD) simulations and model validated experiments, we have been developing new epoxy formulations for composite applications by identifying the processing-structure-properties relationship of epoxy matrix.

Multi-Scale Modeling of Fiber-Matrix Interphase

In the glass fiber/matrix composite, interphase is a distinct region between fiber and matrix that develops during processing through diffusion and curing reaction between the matrix and the fiber sizing. Interphase could have different morphology depending on processing and thickness could vary from few nanometers to tens of nanometers. Since load is transferred through interphase, it controls composites strength, toughness, and damage modes. In this project, we have been establishing a molecular dynamics (MD) based “Materials-by-Design” framework to design new composite interphases to improve composite performance for ballistic applications. We are developing strain rate dependent mixed-mode cohesive traction law surfaces from MD simulations. MD based traction law would be used in finite element analysis to bridge the length scale in composite micro-mechanical modeling.

Predicting Interphase Morphology from Epoxy-Amine Inter-Diffusion and Cure Kinetics

Epoxy-amine inter-diffusion and curing reaction are important mechanisms governing interphase formation in glass fiber/epoxy matrix composite materials. Inter-diffusion and curing reaction between the epoxy film former of sizing and amine harder of matrix should govern the concentration gradients of constituents, atomic structure of the interphase with different degree of cross-link density, thickness and properties of the interphase. In this project, we use a combined molecular dynamics (MD) simulation and experimental approach to predict the composite interphase structure. We develop a new all-atom MD methodology to determine the inter-diffusion between an epoxy film former (Epon828) and Jeffamine® curing agents using the general AMBER force field (GAFF). Cure kinetics and relationship between diffusivity and degree of cure are established from DSC experiment and the Cohen-Turnbull free volume theory. Comparing the gelation time with the diffusion equilibrium time, we conclude that interphase development will be diffusion-controlled for composite material where the sizing (film former) thickness is in the order of hundreds of nanometers, because diffusion equilibrium with stoichiometric ratio will be reached before curing starts. On the other hand, for thicker sizing layer, curing will starts before diffusion equilibrium and diffusion will be slowed down or stopped due to gelation giving concentration gradient in the interphase. This study will also be useful for other applications such as protective coatings and paintings where the thickness spans from micrometer to millimeter. Currently, we are doing ATR-FTIR experiment to determine the epoxy-amine concentration gradient throughout the interphase thickness.

Ballistic Kevlar and UHMWPE Fibers under Multi-Axial Loadings

Ballistic impact induces multiaxial loading on Kevlar® and polyethylene fibers used in protective armor systems. The influence of multiaxial loading on fiber failure is not well understood. Experiments show reduction in the tensile strength of these fibers after axial and transverse compression. In this study, we use molecular dynamics (MD) simulations to explain and develop a fundamental understanding of this experimental observation since the property reduction mechanism evolves from the atomistic level. An all-atom MD method is used where bonded and non-bonded atomic interactions are described through state-of-the-art reactive force field. Monotonic tension simulations in three principal directions of the models are conducted to determine the anisotropic elastic and strength properties. Then the models are subjected to multi-axial loads – axial compression followed by axial tension and transverse compression followed by axial tension. MD simulation results indicate that pre-compression distorts the crystal structure inducing preloading of the covalent bonds resulting in lower tensile properties. Currently, we are working on developing a constitutive model for the concurrent transverse compression and axial tensile loading considering morphological change in the crystal structure.

Inter-Atomic Reactive Force Field Parameterization using Machine Learning Technique

Reactive atomistic-scale molecular dynamics (MD) simulations are very useful to have in-depth understanding of interactions among S-glass fiber, sizing and matrix. Though there is reactive force field for epoxy, there is no such force field to describe the inter-atomic interaction in the S-glass system. In this study, using a novel machine learning approach comprised of a Genetic-Algorithm (GA) assisted Artificial Neural Network (ANN), we are parametrizing ReaxFF interatomic potential parameters for Al/Si/O/Mg interactions for description of S-glass and other Magnesium Aluminosilicate (MAS) glass compositions. Our training set includes the Density Functional Theory (DFT) data of equation of state for various crystals of Mg/Al/Si/O such as Pyrope, Cordierite, Sapphirine, Forsterite, Enstatite, and Spinel. DFT based Nudged Elastic Band (DFT-NEB) calculation results for Mg migration inside Mullite crystal are also calculated and incorporated into the training set for better description of Mg ion migration. Combining S-glass and epoxy force field parameters, we develop force field set for S-glass/epoxy system. This machine learning technique will be used to develop reactive force field for other materials.

Carbon Nanostructures & Nanocomposites

We have been working on multiscale and multiphysics modeling and characterization to design and develop new multifunctional carbon nanostructures & nanocomposites. Modeling approach involves understanding materials’ structure–property relationships as well as their deformation, energy absorption and failure mechanisms at different length scales and loading conditions.