Transverse Mechanical Properties of Unidirectionally Reinforced Hybrid Fiber Composites
Open Access
- Author:
- Ripepi, Maximilian James
- Graduate Program:
- Engineering Mechanics
- Degree:
- Master of Science
- Document Type:
- Master Thesis
- Date of Defense:
- None
- Committee Members:
- Charles E Bakis, Thesis Advisor/Co-Advisor
Kevin L Koudela, Thesis Advisor/Co-Advisor - Keywords:
- Composites
Unidirectional
Transverse
Modulus
Strength
Hybrid
Commingled
Carbon Fiber
Glass Fiber
Epoxy - Abstract:
- Fiber reinforced polymer composites have much versatility in structural design on account of their wide range of elastic and strength properties as functions of direction. Different kinds of fibers such as carbon and glass can be selected to meet mechanical property requirements as well as cost objectives. When multiple types of fibers are incorporated into a composite, the result is called a hybrid composite. Much experimental characterization and theoretical modeling on the mechanical properties of hybrid fiber composites in the fiber direction can be found in the literature. Theoretical models for the mechanical properties of hybrid composites transverse to the fiber direction can be found in the literature, but no experimental data has been published. The objective of the current investigation is therefore to manufacture unidirectional carbon and E-glass hybrid fiber composites by a filament winding process, characterize the elastic modulus and strength of beam specimens tested transversely to the fibers, and assess the capability of available analytical models and the finite element method to capture the trends in the elastic modulus as a function of the proportion of the comingled carbon and E-glass fiber in the composite. The compositions tested included 25% carbon and 75% glass, 50% carbon and 50% glass, 75% carbon and 25% glass. Transverse flexural strength and modulus were both found to increase monotonically with an increasing glass-to-carbon ratio. The series spring model, a modified version of the series spring model, a modified version of the Halpin Tsai model, and the finite element method were used to predict the modulus data. The modified series spring model and modified Halpin Tsai model showed good correlation with the modulus data after the appropriate adjustment of their curve fitting parameters.