PSI - Issue 47

L.A.R. Gomes et al. / Procedia Structural Integrity 47 (2023) 94–101 Gomes et al. / Structural Integrity Procedia 00 (2019) 000 – 000

96

3

direction. The mechanical properties are presented in Table 1 (Ribeiro et al. 2016). Since there was no occurrence of interlaminar rupture during this study, interlaminar properties are not considered as a limiting factor.

Table 1. Elastic orthotropic properties of a unidirectional lamina of Texipreg HS 160 RM (Ribeiro et al. 2016).

E x =1.09E+05 MPa E y =8819 MPa E z =8819 MPa

ν xy =0.342 ν xz =0.342 ν yz =0.380

G xy =4315 MPa G xz =4315 MPa G yz =3200 MPa

To provide results for different types of adhesives, testing was conducted with two adhesives - one brittle and one ductile. The AV138 is a brittle yet highly resistant adhesive, often used in joints with dissimilar materials, which combines average strength with high ductility, which are desirable properties for impact loadings. Bulk specimens underwent tensile testing to determine mechanical properties such as Young’s modulus ( E ) and tensile cohesive strength ( t n 0 ) as well as Thick adherend Shear Tests (TAST) for the shear modulus ( G ) and shear cohesive strength ( t s 0 ). Double-cantilever beam (DCB) and end-notched flexure (ENF) experiments were used to retrieve results related to adhesive fracture toughness in tension and shear, namely G IC and G IIC , respectively (Silva et al. 2022). Results showed that testing speeds do not have influence over E and G (Machado et al. 2016). Similarly, G IC and G IIC are not measurably altered with testing speed increases (Machado et al. 2016). Based on said evidence, these properties at 1 mm/min were considered to be valid for impact scenarios. Being affected by strain rate increases, as is the case in impact loadings, t s 0 and t n 0 must be experimentally determined before conducting numerical experiments (Zhang et al. 2015). The strength properties for both adhesives were determined at different speeds by experimental testing, with properties at 1.75 m/s being object of logarithmic extrapolation from the lower testing speed values (Zgoul and Crocombe 2004). Table 2 presents the properties used in the numerical work of this study for each adhesive.

Table 2. Mechanical and fracture properties of the adhesives as a function of the test speed.

Adhesive

Test speed [mm/min]

0 [MPa]

t s 0 [MPa]

IC [N/mm]

G IIC [N/mm]

t n

G

AV138

1

41.0 49.1 70.2 11.5 18.4 29.9

30.2 36.2 51.7 10.2 15.7 26.4

0.35 0.35 0.35 2.36 2.36 2.36

0.6 0.6 0.6

100

105000

7752

1

5.41 5.41 5.41

100

105000

2.2. Geometries Two different joint configurations were used: SLJ (Fig. 1 a) for experimental validation and DLJ (Fig. 1 b) for numerical analysis. Both configurations share the same adherend width and total length ( L T ), and the same L O variations were tested. For the SLJ, adherends of same thickness ( t P ) were used. That was not the case for the DLJ, where the outer adherends ( t Po ) had half the thickness of the inner adherend ( t Pi ). The t A value is identical between the SLJ and DLJ. Table 3 displays all configurations measurements.

a)

b)

Fig. 1. SLJ (a) and DLJ (b) geometry and dimensions.