This paper presents comparisons between experimental and numerical studies of low-velocity impact damage for thermoplastic (IM7/PEEK) and thermoset (IMS65/MTM) carbon fibre reinforced composites. The experiments were conducted at two key impact energies (8 and 30J) under identical conditions allowing a systematic comparison to be made. Three LS-dyna Finite Element Analysis (FEA) models (standard, continuum damage mechanics (CDM) and discrete) were implemented, all using cohesive interface elements for delamination. The role of Mode II fracture toughness is highlighted. The predictive capabilities of different modelling techniques are compared and discussed and the CDM model gave better correlation with experiments. Fibre failure was predicted by the numerical approaches. The thermoplastic materials did not show noticeably superior behaviour to the thermoset materials and were governed by unstable delamination damage propagation for the same impact energy.

Mode II fracture energy, GIIC, is a critical parameter for determining the propagation of delamination in composite laminates. Its value can be affected by Through-Thickness Compression (TTC) stress acting on the crack tip and here this effect has been studied using IM7/8552 carbon/epoxy laminates with cut central plies. External TTC loads were applied through bi-axial testing. Unidirectional (UD) cut-ply specimens were used to determine the TTC enhancement factor, ηG, for GIIC. A similar enhancement effect was also found in Quasi-isotropic (QI) specimens with 2 extra cut central 0° plies inserted into the layup. The TTC enhancement factor was implemented in a Finite Element Analysis (FEA) framework using cohesive interface elements, showing that the determined ηG can be successfully used to model the effect of TTC on delamination.

The change in the critical strain energy release rate as damage evolves, known as the Rcurve, is of crucial importance to the understanding of fracture behaviour. The examination of damage evolution ahead of the crack tip in order to determine accurately the crack increment is key for the determination of the R-curve. Conventional in situ methods such as optical measurements only examine the specimen surfaces. X-ray Computed Tomography (CT) offers satisfactory image quality, but conventional CT scanning requires the removal of the specimens from the test machine. If no dye penetrant is used, the specimens can be re-loaded, but some important information will be missing such as the early load drops corresponding to damage initiation. If dye penetrant is used, the specimens can no longer be re-tested. In contrast, in situ CT scanning can capture the detailed damage states ply-by-ply while the specimen is loaded and diminish the need for multiple specimens. In situ characterization of trans-laminar fracture toughness of composites using CT has not been attempted in the past. It has been proven successful in this research, and shown that a partial R-curve can be constructed with a single Extended Single-Edge-notch Tension (ESET) specimen.

No abstract