PSI - Issue 28

Sergey Smirnov et al. / Procedia Structural Integrity 28 (2020) 234–238 Author na e / Structural Integrity P o edi 00 (2019) 000– 0

236

3

9.4

9.6

10.7

12.1

19.2

19.5

19.8

22.8

Angle θ , deg.

0.043

0.055

0.043

0.060

0.060

0.068

0.058 11.07

0.027 12.69

Interlayer thickness h , mm Maximum failure force Р , kN

3.43

5.80

8.21

7.22

9.14

7.09

Mean tangential stress on the glue-metal boundary, σ  , MPa Mean stress normal to the glue-metal boundary, σ n , MPa

4.87

8.22

11.61

10.16

12.42

9.61

14.99

16.83

−0.81

−1.39

−2.19

−2.18

−4.32

−3.40

−5.40

−7.07

3. Experiment and results The loading diagrams are linear up to the failure load and there were no yield plateau and load decrease, this being indicative of the elastic deformation behavior of the glue layer. In all the diagrams there are small steps and pulsations, which must be due to the formation of microdamage in the interlayer. The specimen fracture occurred along the glue metal boundary, and it was associated with maximum loads recorded by the testing machine. No catastrophic failure, characteristic of brittle rupture from normal stresses, was noted: there was retarded separation of one specimen half from the other. The failure was adhesive in all the cases, as witnessed by the absence of glue on the symmetric areas of the surfaces of the specimen halves (figure 2).

Fig. 2. The surfaces of the specimen halves after testing, with symmetric areas of adhesive failure

The mean tangential and normal stresses acting on the glue-metal boundaries are defined as σ  = P cos θ and σ n = P sin θ (table 1). The load corresponding to the moments of failure increases with the angle θ . According to the earlier results obtained in Smirnov et. al. (2014), it is here considered that resistance to adhesive failure depends on the stress triaxiality factor

0   k n 