WOSInternational audienceThe objective of this study is to define a reliable tool for dimensioning of adhesively bonded assemblies, particularly for marine and underwater applications. This paper presents experimental and numerical results, which describe the non-linear behaviour of an adhesive in a bonded assembly for various loadings. A modified Arcan fixture, well-suited for the study of the behaviour of bonded metal–metal assemblies, was developed in order to focus on the analysis of the behaviour of the adhesive in thin films. The edge effects are minimised by using an appropriate design of the substrates so that experimental results give reliable data. To analyze the kinematics of the deformation of the adhesive joint non-contact extensometry techniques were employed. Thus, various characteristics of the non-linear behaviour of the adhesive were observed. The results obtained for mixed assemblies (steel, aluminium, composite substrates) show similar behaviour of the adhesive using the proposed procedure. Furthermore, quality controls by DSC, and studies of curing by DMA have been performed to verify the bonding procedure

This paper describes a study in which the shear behavior of a structural epoxy adhesive has been measured using the standard thick adherend shear test (TAST) specimen and a modified Arcan test. A numerical study of the TAST test taking into account the nonlinear behavior of the adhesive and the finite deformations of the adhesive joint, shows that there is a localization of plastic zones close to the adhesive-substrate interface near the free edge of the adhesive. Experimental tests carried out with steel and aluminum substrates and with various adhesives also show that failure initiates in this region. These edge effects in the TAST fixture can lead to an incorrect analysis of the behavior of the adhesive (for instance, underestimation of the shear stress in the joint at failure), particularly when an adhesive failure mode is dominating. The modified Arcan fixture provides a more homogeneous stress state. A similar improvement of the TAST fixture is proposed.

WOSInternational audienceWhile the geometry of aerospace assemblies is carefully controlled, for many industrial applications such as marine structures bond line thickness can vary significantly. In this study epoxy adhesive joints of different thicknesses between aluminium substrates have been characterized using physico-chemical analyses (differential scanning calorimetry, DSC; dynamic mechanical analysis, DMA; spectroscopy), nano-indentation and mechanical testing. Thermal analyses indicated no influence of thickness on structure. Nano-indentation revealed no evidence of an interphase at the metal/epoxy interface, nor any change in modulus for different thicknesses, though Raman spectroscopy suggested there may be slight variations in composition close to the substrates. However, mechanical testing using the modified Arcan fixture indicated a significant drop in strength and failure strain under pure tension and a smaller reduction for tension/shear and pure shear loads as thickness increased. Examination of sections through joints did not indicate any physical reason for this, but numerical analysis of the stress state revealed larger stress concentration factors for tensile loading in thick joints, which may explain the thickness effect. It is recommended that joint thickness should be kept below 0.8 mm to avoid obtaining artificially low values with the Arcan test

No abstract