This Is AuburnElectronic Theses and Dissertations

Stress Mapping of Textile Composite Materials and Its Application in Interfacial Shear Behavior

Date

2005-12-15

Author

Shady, Ebraheem

Type of Degree

Dissertation

Department

Textile Engineering

Abstract

Mapping of the stress distribution in composite materials, both at the fiber/matrix interface and at the composite constituents themselves, is important to understand the material mechanical response. Stress mapping can help predict composite behavior under certain stresses especially failure or delamination. In this work, two analytical models were proposed to map the stress distribution at fiber, matrix and fiber/matrix interface. The first model dealt with the fiber in the longitudinal direction considering axisymmetric conditions. The second model addressed the fiber stress distribution in the transverse direction. Both models were verified using finite element models. As an application for the stress mapping models, interfacial shear behavior was chosen for its importance in modeling and design of composite materials. Two fabric structures were used to manufacture five different panels for each fabric. The number of fabric layers for each plate ranged from 5 to 9 layers systematically altering the volume fraction and nesting characteristics of each plate. Four-point flexural tests were used to obtain a pure bending state between load noses. The maximum tensile stress and crack initiation stress at the bottom layer were experimentally evaluated. Experimental data was processed using the Graphical Integrated Numerical Analysis software (pcGINA) to obtain the maximum stress in the target laminate and this value was used as the input for the two analytical models. The value for the maximum interfacial shear stress which is responsible for crack initiation in the laminate was calculated using the models and results were compared to pull-out fiber test values obtained from literature. Good agreement was observed between the model results and the literature data.