13th International Conference on Fracture June 16–21, 2013, Beijing, China -4- The 1D approach considers hydrogen diffusion is exclusively developed in radial direction (r), i.e., through the wire radius placed at notch symmetry plane. However, in 2D approach the hydrogen diffusion proceeds in both radial (r) and axial direction (z). Therefore, differences in the hydrogen concentration given by the two considered approaches can be associated to a geometric factor. The FEM simulation of the hydrogen diffusion assisted by stresses was performed with linear elements for both approaches (1D and 2D). In these simulations the exposure time to harsh environment was chosen equal to the fracture time (tf) obtained in the study [21], where similar specimens –material and notched geometries– were tested in an inert environment (air). This way, data related to fracture time due to HAF (tHAF) are included in the results of the simulation performed since tHAF < tf [12]. The results of the displacement at fracture instant, uf(A) = 0.4 mm, uf(B) = 0.13 mm, uf(C) = 1.2 mm and uf(D) = 0.42 mm, obtained in [21] with an extensometer of 25 mm gage length, were used to determine the time of fracture in air (tf) according to the following relation tf = uf/u for each one of the two considered extension rates u (0.1 and 0.001 mm/min). 3. Numerical Results The hydrogen accumulation in the material is represented by the relative hydrogen concentration Cr (Cr = C/C0), which can be defined as the hydrogen concentration normalized with the equilibrium hydrogen concentration in a virgin material C0, i.e., free of stress and strains. Fig. 2 shows the distribution of the relative hydrogen concentration (C/C0) through the considered notch symmetry plane for a diffusion time lower than the fracture time in air (tf) obtained for each one of the two extension rates considered, 0.1 mm/min and 0.001 mm/min. As can be shown in Fig. 2 two different trends were obtained according to the extension rate: on one hand, for the highest extension rate of 0.1 mm/min (Fig. 2 left) slightly differences appear in hydrogen concentration obtained by 1D and 2D approaches of diffusion model. On the other hand, when the lowest extension rate of 0.001 mm/min is applied (Fig. 2 right) a clear influence of the geometric factor on hydrogen diffusion is revealed, especially, in the notched geometry type A where high differences in relative hydrogen concentration are obtained considering 1D and 2D approaches. Therefore the use of 1D approach (less realistic than 2D approach) leads to a loss of accuracy in the determination of relative hydrogen concentration, it being more accused for low extension rate tests and sharp notches. According to these results hydrogen diffuses not only through the notch symmetry plane but also towards other directions. However, most of hydrogen is preferentially accumulated in the radial notch symmetry plane direction, as could be expected since H diffuses towards the location where the highest hydrostatic stress appears (situated in the notch symmetry plane [13]). So, according to Fig. 2, the lower the extension rate the more accused the effect of the geometric factor on hydrogen diffusion, it becoming inappreciable for high extension rate test where the hydrogen diffusion is performed practically through the notch symmetry plane with a negligible loss of directionality. The influence of geometric factor on hydrogen diffusion is dependent of notch geometry parameters (radius and depth). So, the influence of this factor is lower for blunt notched samples with a high notch radius (notches C and D) than for sharply notched geometries (notches A and B) with a low notch radius (Fig. 2). According to that, a stronger influence of the geometric factor on hydrogen diffusion appears for high values of the depth (distance) from the notch tip (x).
RkJQdWJsaXNoZXIy MjM0NDE=