PSI - Issue 13

S.A. Atroshenko et al. / Procedia Structural Integrity 13 (2018) 1373–1377 S. A. Atroshenko et al. / Structural Integrity Procedia 00 (2018) 000–000

1376

4

Fig. 5. Cracks in copper ring sample loaded with a period of 6 . 5 µ s , x1000

Fig. 6. Shear bands, x2500 (a) and twins, x1600 (b) in copper ring sample loaded with a period of 1 µ s

Fig. 7. Neck regions in copper ring samples loaded with a period of 1 µ s , x1000 (a) and loaded with a period of 6 . 5 µ s , x1600 (b).

3.2. Microstructure investigations

As expected (Atroshenko et al. (2014), Atroshenko et al. (2015)), the fraction of the viscous component on the fracture surface of a sample loaded with a shorter period of 1 µ s was much smaller ( S = 61 . 8 %) than in a sample loaded with a period of 6 . 5 µ s ( S = 98 . 1 %). I.e. the sample breaks more brittle at a higher loading rate. At the same time, a study of the microstructure of the cross-section of the tested samples revealed the presence of cracks similar to spalling in samples with a longer pulse duration (Fig. 5). While, shear bands and twins are observed (Fig. 6)under loading with a shorter period duration. Neck regions were found in all samples. But more of them are observed in the samples after short-period loading (Fig. 7). It may be concluded that more energy is spent on the formation of shear bands, twinning and neck formation than on the nucleation of cracks at the high loading rate.

4. Conclusions

The following conclusions based on the presented research results can be drawn:

• The developed technique of magnetic-impulse loading made it possible to determine the strength of copper ring samples in a wide range of microsecond-scale loading rates • The fractography of fracture surfaces confirmed a more fragile character of failure with increasing strain rate.