PSI - Issue 13

Jean-gabriel Sezgin et al. / Procedia Structural Integrity 13 (2018) 1615–1619 Jean-Gabriel Sezgin/ Structural Integrity Procedia 00 (2018) 000 – 000

1617

3

Then, aligned penny-shaped cracks included in an elastoplastic medium have been considered in the FE analysis. A crack spacing of 15 µm and a total crack length ranging from 3 µm to 30 µm have also been into consideration. The criterion for void growth was herein defined as the critical J Ic value. However, its actual value is unknown as it depends on the configuration the voids (shape, fraction, distribution…) . This critical stress has to be calculated numerically. Thus, the zero-pressure case for each spacing has been taken as a reference in order to define the failure criterion, J Ic . Figure 1-b) shows quantitatively the influence of the internal pressure on the critical stress. The critical stress of the H-charged specimen was normalized by that of the non-charged one. If fast diffusion related to dislocation pipe diffusion caused a higher pressure (133 MPa), the maximal loss in critical stress was 15%. However, this loss could be judged to be insufficient to cause void growth, since the corresponding loss in the critical strain was 35% and the experimental loss in the critical strain was nearly 50% (Matsuo et al. 2014).

Figure 1 – Results from simulation: a) internal pressure build-up in a 1-nm thick void, b) relationship between normalized critical stress and internal pressure

3. Experimental validation of the modelling procedure 3.1. Experimental protocol

In order to validate experimentally the simulation results stated in the previous section, some SSRT tests have been carried out with smooth, round-bar specimens with a diameter of 4 mm at room temperature under a strain rate of 5 × 10 − 5 s − 1 . To take the influence of hydrogen into account, some of the specimens have been exposed to 100-MPa hydrogen gas at 270 ℃ for 200 h before the SSRT test. The SSRT tests have been conducted under the following conditions: (I) non-charged specimen tested in vacuum (10 Pa), (II) H-charged specimen tested in vacuum (10 Pa), (III) H-charged specimen tested in 115-MPa nitrogen gas, and (IV) non-charged specimen tested in 115-MPa nitrogen gas. Cases (III) and (IV) consist in the application of an external pressure. This external pressure targets to cancel the internal pressure of around 130 MPa, which is the most conservative case. By comparing SSRT properties of Cases (I) to (IV), the contribution of the internal pressure to the SSRT properties can be clarified experimentally. 3.2. Effect of internal pressure on the tensile properties Figure 2 shows the results of the SSRT tests for all the aforementioned conditions. In parallel, Table 1 presents the values of RA and RRA for the different testing conditions. All the specimens failed after reaching their tensile strength; however, the H-charged specimens showed a slight degradation of RRA, regardless of application of external pressure. The RRA values were 0.84 for (II), 0.88 for (III) and 1.01 for (IV), respectively. These experiment results suggest that there was no notable contribution of the internal pressure to the tensile ductility of the H-charged specimen.

Table 1. Results of the SSRT tests for all the conditions

Case

Condition

RA (%)

RRA

I

Non-charged in vacuum (10 Pa) H-charged in vacuum (10 Pa) H-charged in 115 MPa N 2 Non-charged in 115 MPa N 2

92 77 81 93

--

II

0.84 0.88 1.01

III IV