Failure Analysis of Unidirectional Ceramic Matrix Composite Lamina and Cross-Ply Laminate under Fiber Direction Uniaxial Tensile Load: Cohesive Zone Modeling and Brittle Fracture Mechanics Approach

摘要

The current research work presents the computational micromechanical analysis of the room temperature tensile failure behavior of unidirectional (UD) and cross-ply (0/90) ceramic matrix composites (CMCs). For computational micromechanical analysis, three-dimensional (3D) representative volume element (RVE) and multi-fiber multilayer RVE (M-2 RVE) models are generated that are representative of the lamina and the laminate under investigation. The RVE and M-2 RVE models are generated by replicating the fiber distribution, and the placement of the fibers observed in a microscopic image of an actual CMC laminate. The generated RVE models consist of the discrete representation of individual constituent phases of the CMC such as fibers, interphase, matrix, and the fiber-interphase interface region. Under the applied external tensile load, the fiber-interphase interface interactions are modeled using the cohesive elements that follow the bilinear traction separation law. The matrix, fiber, and interphase materials failure behavior is captured using a brittle cracking model. In order to validate the proposed numerical methodology, the predicted average stress-strain curve at the UD laminate level is compared to the experimental stress-strain curve reported in the literature. In addition, the observed different phases in the predicted stress-strain curve are validated with the literature data. Using the proposed numerical methodology, a detailed local stress-strain and damage analysis leads to an observation that the so-called ductile stress-strain behavior (kink in the stress-strain curve) of a CMC UD laminate under uniaxial fiber direction tensile loads is mainly caused by the matrix damage initiation. Apart from the SiC material properties such as strength and fracture energy, it is also observed that the RVE size influences the average strength and failure strain predictions using computational micromechanics.