|Institution:||Delft University of Technology|
|Keywords:||glass fibre-reinforced composites|
|Full text PDF:||http://resolver.tudelft.nl/uuid:c8de1961-2950-49c7-84f5-34eb65b3f8f1|
Fibre-reinforced polymer composites (FRPs) have been used in structural applications for years mainly due to their outstanding specific mechanical properties. In this thesis, the characterisation of the strain field upon transverse tensile loading of glass fibre-reinforced polymers (GFRPs) was investigated by applying digital image correlation (DIC) to scanning electron microscopy (SEM) images acquired during in situ mechanical testing. Full arrays of strain values were successfully obtained for regions of interest (ROI) atin the microscopicmeter length scale. With an eye towards the effect of microstructure, matrix and fibre/matrix interface properties on the overall mechanical behaviour of the material, GFRPs with two different thermoset polymer matrices were tested under 3-point-bending and compared. A systematic procedure for the generation of speckle patterns based on the deposition of Al2O3 and Fe3O4 nanoparticles has been developed, granting the DIC analysis a high resolution even at high magnifications 5000x. An optimum subset size was determined by using the Gray Level Co-Occurrence Matrix. Cracking at some fibre/matrix interfaces could be detected visually and by means of high strain concentrations displayed in the DIC strain contour plots. Good agreement of displacement and strain fields between DIC analyses and finite element analyses (FEA), that were performed as a validation method, was achieved after taking the cracks into account in the FE-model. A novel double matrix concept was proposed as a means to reduce the stress concentrations that result during transverse tensile loading. FEA were used as validation of the results provided by the DIC analyses. The strain at rupture achieved with GFRP samples with this new concept upon transverse tensile loading is 3 times higher as the state-of-the-art at the expense of 98% in Young’s modulus.