Skip to Main Content

2019 Research Highlight: Molecular Modeling of Glass Fiber

Molecular Modeling of Glass Fiber

CMEDE Researchers
Dr. Sanjib C. Chowdhury
University of Delaware
Mr. Ethan M. Wise
University Delaware (URAP Intern)
Mr. Raja Ganesh
University of Delaware
Professor John W. Gillespie, Jr.
University of Delaware
Dr. Timothy W. Sirk
CCDC Army Research Laboratory

Our model composite consists of three primary constituents: S-Glass fiber, epoxy matrix and the interphase that governs load transfer between fiber and matrix and the mechanical properties of the composite. A systematic Materials by Design approach has been developed to study failure mechanisms and bridge length scales from the atomic to the continuum length scale of a single layer of a woven composite.  At the lowest length scale, classical molecular dynamics (MD) simulations have been conducted to understand high strain rate failure mechanisms of the individual constituents as well as their interactions. MD simulations of glass fiber has identified the strain rate dependent and progressive damage mechanisms within the fiber, the sensitivity of tensile strength to surface defects and the prediction of fiber surface reactivity needed to design optimal interphase properties.

During fiber spinning, sizing and composite processing, nanometer size surface cracks develop during these handling operations that reduce tensile strength significantly below theoretical limits.  In unidirectional composites, fiber fails at these defect sites resulting in a dynamic release of stored strain energy that subjects the interphase and matrix to very high strain rates and high levels of inelastic deformation that propagate at high speed along the broken fiber.  Nearest neighbor fibers are also subjected to a tensile wave with a dynamic stress concentration that can trigger additional fiber breaks at locations with critical defects.  This process continues resulting in a localization of damage that is sufficient to cause catastrophic failure of the composite.  Our Materials by Design approach focuses on optimizing the rate dependent interphase and resin properties as a function of the defect size and spatial distribution of defects within the glass fiber to improve composite properties. Therefore, to design high performance composites, it is imperative to have a fundamental understanding of the fracture mechanisms of the fiber at the smaller length scale and how the fiber defects effect composite failure mechanisms. In this summary, we have highlighted some of our key studies on glass fiber.

Recently, we have carried out reactive MD simulations using state-of-the-art reactive force field ReaxFF to study the strain rate dependent progressive failure of the fibers as well as the effects of surface crack on the mechanical properties of glass fiber. We have developed an atomistic J-integral method to determine the fracture energy release rate (Figure 1). The J-integral method captures all sources of energy dissipation and agrees well with the Griffith fracture mechanics for crack length larger than 1 nm.  As part of this study, MD was used to partition energy absorption mechanisms from both the creation of new crack surface area and progressive damage of bonds that develop within a very narrow cohesive zone adjacent to the crack surface.  Nanometer size surface crack significantly affects the fiber strength without influencing the fiber modulus.  Strength decreases with increase in the crack length and MD predicts a 35 nm surface crack is sufficient to reduce fiber tensile strength to experimental levels measured in commercial glass fiber (average strengths of 3.5 GPa).  The developed MD framework will enable us to study crack healing mechanism of glass fibers to improve strength and the prediction of surface reactivity for interphase design.