Effects of Bondline and Substrate Thicknesses on the Mode II Fracture Toughness of Composite/Adhesive Joints

Abstract

Glass fiber reinforced plastics (GFRPs) are commonly used in wind turbine blades (WTB). The joining of several GRFP components in a WTB is usually obtained using structural adhesives. The GFRP/adhesive joint usually experiences mixed-mode loading (opening and shear) conditions, with mode II being a major phase. The experimental evaluation of the joint failure under shear is usually conducted using end-notched flexure (ENF) test, which provides the value of the fracture energy required for the onset of interlaminar crack propagation by shear forces. There are standard methods such as ASTM D7905 that have provided the procedure for the ENF test. This standard test method has been particularly developed for the evaluation of crack growth from a precrack within a fiber reinforced composite, not composite/adhesive joints. However, this method has also been used for Mode II fracture energy evaluations in composite/adhesive joints. To better understand the applicability of the standard ENF method for composite/adhesive joints of WTBs, this work focuses on the dependency of the measured mode II critical strain energy release rate (GIIc) on the specimen geometry, in particular, substrate thickness and adhesive layer thickness. The substrate effect on the fracture behavior was assessed using three different substrate thicknesses of about 3, 5, and 6.3 mm, and the bondline thickness was examined from 0.25 to 3.55 mm.Overall, it was found that both substrate thickness and the bondline thicknesssignificantly affect the fracture energy and failure mode. The results showed that withan increase in the substrate thickness, GIIc was increased, and the fracture surfaceexamination suggested a more cohesive failure behavior. Moreover, with an increase inthe bondline thickness, GIIc first was increased up to a maximum value of 6.90 kJ/m2 ata thickness of 1.05 mm beyond which it started to decline. These findings suggest thatthe measured fracture energy for composite/adhesive joints using ENF standard testmethod is strongly geometry dependent and thus the geometry should be considered inthe design for failure under crack growth.