TIPE / CPGE PCSI
Study of the behavior of epoxy resin under mechanical stress in composite material assemblies. Tensile testing, failure analysis and comparison with theoretical models.
Epoxy resins are widely used in structural applications, from aeronautics to civil engineering, because of their excellent adhesion, low shrinkage and high resistance to chemical and thermal degradation.
In composite materials, the resin acts as the matrix binding the reinforcing fibers. Its mechanical behavior under load directly determines the structural integrity of the whole assembly. Understanding where and how it fails is a prerequisite for reliable design.
How do the mechanical properties of an epoxy resin vary as a function of its curing conditions and the geometry of the composite assembly in which it is embedded?
The goal was to identify the influence of cure time, temperature and sample geometry on the tensile strength and Young's modulus of the resin.
Sample preparation
Epoxy samples cast in standardized molds (dogbone geometry, ISO 527) under controlled temperature and humidity. Several cure cycles tested: ambient temperature 24h, 40C oven 12h, 60C oven 6h.
Composite assemblies
Glass fiber reinforced samples produced with hand layup. Fiber orientation varied (0, 45, 90 degrees) to measure anisotropy of the composite as a whole.
Tensile testing
Uniaxial tensile tests performed on a universal testing machine. Load-displacement curves recorded; Young's modulus, yield stress and ultimate tensile strength extracted from each curve.
Failure analysis
Visual and macroscopic observation of fracture surfaces. Comparison with theoretical failure criteria (maximum stress, Tsai-Hill for composites).
The results confirmed several expected behaviors and revealed one unexpected trend regarding the relationship between cure temperature and ductility.
+23%
Increase in Young's modulus between ambient cure and 60C oven cure, consistent with higher crosslink density.
0
Fiber orientation showing maximum stiffness. 90 degree samples showed 60% lower modulus, confirming strong anisotropy.
Brittle
All pure resin samples showed brittle fracture with no visible plastic deformation before failure.
The composite assemblies consistently outperformed the pure resin on tensile strength but showed more complex and less predictable failure modes due to fiber-matrix interface debonding. The Tsai-Hill criterion gave reasonable predictions for 0 and 90 degree samples but underestimated failure loads at 45 degrees, suggesting interfacial shear plays a larger role than the model assumes.
The TIPE was my first real encounter with experimental methodology at this level. The gap between what the theory predicts and what the samples actually do is where the interesting questions live.
I would have liked more time to investigate the 45-degree failure anomaly. The next step would be to use a scanning electron microscope to characterize the fracture surfaces and confirm the debonding hypothesis.
The work also reinforced something practical: preparation quality matters far more than testing quality. Most of the scatter in the data came from sample imperfections introduced during casting, not from the machine or the measurement.