PSI - Issue 33

A.F.M.V. Silva et al. / Procedia Structural Integrity 33 (2021) 138–148 Silva et al. / Structural Integrity Procedia 00 (2019) 000–000

140

3

behavior of the joints in adverse conditions. Few numerical works can be found in the literature to predict the adhesive joints’ behavior under impact loads, either by proposed analytical or numerical models. Moreover, a constant development in the impact behavior modelling is noticed, leading to results that become gradually closer to the reality. Morin et al. (2013) developed a new concept of cohesive element skilled to model the materials’ elasto-viscoplastic behavior, aiming to be used in the design for impact structures and crashworthiness joint design evaluations. The proposed element is an interface between two structural parts, consisting of an 8-node hexahedron element. The adhesive thickness ( t A ) is assumed to be insignificant when compared with the size of the assembled part, and is modelled as a common spring attached to the midpoint of the structural parts, similar to the classic cohesive element. This method requires coincident meshes between the two assembled parts, which can be modelled by shell elements, solid elements or by a combination of both. A good correlation between the numerical and experimental results was found, thus validating the proposed approach. The authors shown that the lower computational effort and the simpler mechanical testing campaign that is required to feed the model make this approach very interesting when compared with traditional cohesive elements. Valente et al. (2019) performed a CZM analysis on tensile impact loaded bonded joints evaluating the strength and the stress propagation. The study included two types of adherends, the AW6082 T651 aluminum alloy and a CFRP with unidirectional lay-up. Three adhesives with different ductility grades were used to bond the adherends in a SLJ geometry. The validation of the numerical model was accomplished by comparison with experimental data from impact tests performed in an instrumented falling weight impact tester. The authors found a good agreement between experimental and numerical load-displacement ( P -  ) curves in terms of P m . However, the experimental data presented a higher failure displacement. Regarding the influence on the joint behavior by the type of adherends’ material, it was shown that by using aluminum, the plastic deformation can be higher depending on the adhesive, thus reducing P m . Nonetheless, the higher stiffness of the CFRP adherends is prone to affect the load transfer, and especially with stiffer adhesives, leading also to a reduction of P m . As a final remark, the CZM analysis enabled to select the best design of material/adhesive conditions to provide the best results under impact. In this work, the impact strength of tubular adhesive joints with AW6082-T651 aluminum alloy adherends and the adhesive Araldite ® AV138 is studied. For this purpose, the modification of the main geometric parameters and is considered: L O and t SE . The analysis includes s y and t xy stress distributions, enabling to better compare the different geometries. P m and U prediction is accomplished using CZM. Previous CZM validation with SLJ was accomplished. 2. Experimental and numerical conditions 2.1. Adherend and adhesive materials The adherend material was different between the validation study with SLJ and this work’s tubular joint analysis. For the validation study, the adherends were made of a spring steel (DIN 55 Si7), a silico-manganese alloy, treated in the quenched and tempered condition typically. The mechanical properties of this alloy are detailed in a previous work (Silva et al. 2016): Young’s modulus ( E )=210 GPa, tensile yield stress (  y )=1078 MPa, tensile strength (  f )=1600 MPa, tensile failure strain (  f )=6%, Poisson’s ratio (  )=0.3 and density(  )=7.8. The AW6082-T651 aluminum alloy was considered for the subsequent tubular joint analysis. The main properties, according to the standard ASTM-E8M 04, and defined in reference (Campilho et al. 2011), are: E =70.07±0.83 GPa,  f =324.00±0.16 MPa,  y =261.67±7.65 MPa,  f =21.70±4.24% and  =2.7. In both analyses, an epoxy brittle adhesive was considered to bond the adherends (Araldite ® AV138). Mechanical and fracture characterization of this adhesive was carried out in a previous work (Campilho et al. 2011). The tensile mechanical properties were defined by tensile tests in bulk specimens. In addition, Thick Adherend Shear Tests were used to estimate the shear mechanical properties. The relevant fracture properties of the adhesive (tensile fracture energy or G IC and shear fracture energy or G IIC ) were obtained from Double-Cantilever Beam and End-Notched Flexure tests, respectively. Table 1 includes the aforementioned properties. In addition, since this study considers an impact event, the tensile cohesive strength ( t n 0 ), shear cohesive strength ( t s 0 ), G IC and G IIC at 1 mm/min and 100 mm/min were experimentally estimated, whereas the properties for 105000 mm/min (1.75 m/s) were computed by logarithmic extrapolation (Zgoul and Crocombe 2004) from the properties at the lower test velocity. The results of this analysis are presented in Table 2. It should be noted that E and shear modulus