A.B. de Morais

Mode I interlaminar fracture of carbon/epoxy multidirectional laminates

An experimental study was conducted on the mode I interlaminar fracture of carbon/epoxy multidirectional specimens with starter delaminations in 0/0 and 0/θ interfaces. The specimens selected for Double Cantilever Beam (DCB) tests had [(02/90)6/02//(02/90)6/02] and [(02/90)6/02//θ/(02/90)6/02] stacking sequences, where // denotes the position of the starter delamination and θ = 22.5 to 90 degrees. The incorporation of the thick symmetric [(02/90)6/02] block of 20 plies aimed at minimising elastic couplings related problems, such as non-uniform width-wise distributions of the strain energy release rate GI, spurious modes and effects of residual stresses. This was verified in preliminary three-dimensional (3D) Finite Element (FE) analyses, which also showed the applicability of the Corrected Beam Theory data reduction scheme. The DCB specimens were tested according to the procedures defined in the ISO 15024:2001 standard for unidirectionally reinforced specimens. In the tests, intraply damage occurred in the θ-oriented ply soon after initiation from the starter crack, as it is commonly observed in multidirectional specimens. For θ = 22.5 and 45 degrees, a second crack propagated close to a neighbouring interface, while for θ = 67.5 and 90 degrees the starter crack ran inside the θ-oriented ply and proceeded in a wavy propagation trajectory. This resulted in extensive fibre bridging and pronounced R-curves with artificially high final values of the critical strain energy release rate, GIc. Therefore, only GIc values of initiation could be considered true interlaminar properties. Moreover, it was found that they were fairly independent of the ply angle θ. An additional FE analysis of the interlaminar stresses ahead of the crack tip was performed to interpret the delaminating interface effect on GIc values. The results indicated the existence of an interface independent fracture toughness, KIc, probably resin controlled, thus reinforcing the idea that the measured initiation GIc were pure interlaminar properties.

Mode-I interlaminar fracture of carbon/epoxy cross-ply composites

Abstract Mode-I double-cantilever beam (DCB) tests were performed on carbon/epoxy [0°/90°]12 specimens. The starter crack was created at mid-thickness, between the 0 and 90° mid-layers. During the tests, however, the crack also propagated along the neighbouring 0°/90° interface and within the 90° mid-layer. Nevertheless, the test results were apparently consistent with the assumptions of the corrected beam theory (CBT) that was used to obtain the interlaminar critical strain energy release rate, GIc. The measured values were higher than those of unidirectional [0°]24 specimens, especially the final propagation values. A finite-element analysis confirmed the applicability of the CBT for interlaminar propagation along the two 0°/90° interfaces. The results also indicated that the intralaminar GIc is significantly smaller than the interlaminar GIc. This will prevent pure interlaminar propagation in multi-directional specimens with high interlaminar fracture toughness.

Analysis of crack propagation in double cantilever beam tests of multidirectional laminates

Abstract This paper presents numerical analyses of crack propagation in double cantilever beam (DCB) tests of multidirectional laminates. Three-dimensional finite element models were used for application of the virtual crack closure technique and for simulation of crack growth with a progressive damage model. The analyses were concerned with three features that may affect the measurements of the mode I critical strain energy release rate G Ic : residual stresses, mode-mixity and curved delamination fronts. The results showed that the effects of those features could be minimised by a proper selection of specimen stacking sequences. The suitability of other DCB specimens proposed in the literature to avoid intralaminar damage and crack jumping was also evaluated.

Strength of adhesively bonded stainless steel joints

An experimental study was conducted on the strength of adhesively bonded stainless steel joints, prepared with two epoxy and one acrylic adhesives. The mode I critical strain energy release rate, GIc, was measured in double cantilever beam (DCB) tests. Lap-shear tests were performed on various types of single-lap and double-lap joints. Finite element analyses were used to explain joint strength results. It was found that the strengths of epoxy adhesive joints depended mostly on the level of peel stresses near the bondline edges. When expressed in terms of effective overlaps, joint strengths were generally insensitive to the presence of defects created at the overlap ends.

Mode II Interlaminar Fracture of Filament Wound Angle-ply Specimens

This paper presents a study of the interlaminar fracture of filament wound composites. Mode II end notched flexure (ENF) tests were performed on flat glass/polyester specimens. The tested specimens had asymmetric [±Θ]4 angle-ply stacking-sequences, with Θ values from 0.8° (hoop winding) to 30°. Due to the low stiffness and probable high toughness, it was not possible to propagate the crack in [±60°]4 specimens before they suffered considerable permanent deformations. A short support span had to be employed for [±30°]4 specimens in order to obtain crack propagation. Nevertheless, the results for those specimens should be viewed with caution, as some nonlinear behaviour and small permanent deformations were detected. No other unusual features, such as crack jumping to other interface, were observed. The scatter in the critical strain energy release rate values (GIIc) was higher for [±10°]4 and [±30°]4 specimens than for the quasi-unidirectional ones. GIIc values from the insert were generally lower than those from mode II pre-cracks, except for quasi-unidirectional specimens. A plot of average GIIc values against Θ showed a minimum at Θ=5°.

Beam analysis of the double cantilever beam specimen with fibre bridging

The fibre bridging phenomenon commonly seen in double cantilever beam tests is considered an artifact of the standard unidirectional specimen. Therefore, it limits the scope of the test to the measurement of the critical strain energy release rate GI c for crack initiation. This paper presents a beam model based on a linear traction–separation law that has been shown to be a good approximation to the early stages of fibre bridging. The present model enables the evaluation of accurate propagation GI c values through the fitting of experimental load–displacement curves. Furthermore, fairly good fits can be obtained without employing sophisticated optimization algorithms by considering simple relations between model parameters and curve features. Therefore, the present model is able to extract a broader set of results from a single test, thereby increasing the confidence in toughness measurements.

Analysis of the cohesive zone and crack length correction in the adhesively bonded double cantilever beam specimen

ABSTRACTThe double cantilever beam specimen has been increasingly employed to enable the development of cohesive zone models for adhesive joints. Evaluation of the traction–separation law (TSL) requires elaborate experimental techniques and usually relies on data measured until the crack initiation point. Nonetheless, current standards stipulate fracture energy measurements under steady-state crack propagation. This paper investigated the influence of the cohesive zone on the commonly used corrected beam theory data reduction scheme. Analytical solutions for the elastic–perfectly plastic, bilinear, and trapezoidal laws were developed using a beam model. The role of the elastic traction decay zone was found to be significant for high strength moderately tough adhesives. Nevertheless, the results showed that the sensitivity of the crack length correction to the cohesive zone can be exploited to obtain approximate TSLs.ABSTRACT The double cantilever beam specimen has been increasingly employed to enable the development of cohesive zone models for adhesive joints. Evaluation of the traction–separation law (TSL) requires elaborate experimental techniques and usually relies on data measured until the crack initiation point. Nonetheless, current standards stipulate fracture energy measurements under steady-state crack propagation. This paper investigated the influence of the cohesive zone on the commonly used corrected beam theory data reduction scheme. Analytical solutions for the elastic–perfectly plastic, bilinear, and trapezoidal laws were developed using a beam model. The role of the elastic traction decay zone was found to be significant for high strength moderately tough adhesives. Nevertheless, the results showed that the sensitivity of the crack length correction to the cohesive zone can be exploited to obtain approximate TSLs.

Analysis of the fracture process zone and effective crack length in the adhesively bonded end-notched flexure specimen

ABSTRACT The mode II fracture of adhesive joints is well-known to involve large fracture process zones. Their effect in fracture energy measurements can be taken into account by the effective crack length approach. Moreover, fracture process zones can be simulated by cohesive zone models, which are increasingly used for structural analysis of adhesive joints. This paper aimed at evaluating the influence of the traction-separation law on the fracture process zone and on the effective crack length in end-notched flexure tests. Novel analytical cohesive zone models were developed for the bilinear and trapezoidal traction-separation laws. The latter were shown to affect significantly the energy dissipation rate versus effective crack length curve prior to crack initiation. Therefore, this effect seems to provide a simple approach for evaluating approximate traction-separation laws. The models here developed are easy to apply and provide simple approximate expressions useful for specimen selection.

Mode II Fracture Characterization of Pinus Pinaster Wood

This paper describes experimental and numerical studies on the application of the End Notched Flexure (ENF) and End Loaded Split (ELS) tests to mode II wood fracture characterization. In this context, ENF and ELS specimens were used to determine GIIc of a clear Pinus pinaster wood in the RL system, which is the most relevant for structural design. In mode II fracture tests the crack faces are in contact, thus hindering a rigorous visualization of the crack tip. This makes classic methodologies based on crack length measurement during experimental tests inadequate, since they induce significant errors on the mode II fracture properties. To overcome this experimental problem a Compliance Based Beam Method (CBBM) is used. This new data reduction scheme does not require crack length monitoring and includes the effect of the Fracture Process Zone (FPZ) ahead of crack tip. Furthermore, the clamped cross-section rotation of the ELS specimen is also taken into account. In the present work a numerical analysis considering a cohesive damage model was performed with a cohesive damage model in order to validate the application of the CBBM to the experimental results. The results confirmed the adequacy of the CBBM and the applicability of the ENF and ELS tests for mode II wood fracture characterization.