Results

4.3.1 Hydrogen effects on mechanical response

Fig. 4.5 shows engineering stress-engineering strain curves in the hydrogen-charged and uncharged smooth specimens. The serrations on the curves are known to be caused by dynamic strain aging as reported in previous studies [14, 15]. The total strain and ultimate tensile strength of the uncharged specimen were 64 % and 870 MPa, respectively. A loss of approximately 50% in ductility and 20% in tensile strength is observed in the presence of hydrogen. The elastic deformation behavior and yield strength were unaffected after hydrogen charging.

Fig. 4.6 displays net section stress-elongation curves of hydrogen-charged and uncharged specimens consistence of precrack length of 1 and 1.8 mm as a definition of short precracks.

The net section stress is calculated by dividing the load during tensile test per area of initial net ligament size. This figure indicates the degradation of mechanical properties of short precracks in the presence of hydrogen which deterioration of strength and ductility in specimen with 1 mm precrack is higher than one with 1.8 mm precrack.

Fig. 4.7 shows net section stress-elongation curves of hydrogen-charged and uncharged specimens with precracks of 2.5 and 3.2 mm lengths as a definition of long precracks. Unlike specimens with short precrack, results indicate negligible changes in tensile strength and ductility of hydrogen-charged and uncharged specimens. Results suggest that macroscopic mechanical properties were not affected by the hydrogen charging in long precracked-specimens.

The effect of precrack length on tensile strength and equivalent elongation examined in hydrogen-charged and uncharged specimens is illustrated in Figs. 4.8a and 4.8.b, respectively. The dependence of tensile strength upon precrack shows an obvious decrease in the tensile strength compared the smooth specimen in both hydrogen-charged and uncharged specimens. Results indicate that by increasing the precrack length the effect of hydrogen has been reduced, consequently, the loss of tensile strength and elongation is

110

decreased, while there is no considerable effect of hydrogen degradation on strength and ductility of specimens with long precrack.

4.3.2 Fractographic analysis

Fig. 4.9 displays the fracture surfaces of hydrogen-charged and uncharged smooth specimens. Observation of fracture surface of uncharged specimen demonstrates a typical ductile fracture surface composed of numerous fine dimples in all areas as shown in overall image of fracture surface in Fig. 4.9a and also in magnified images of Figs. 4.9a1 and 4.9a2 of the parts highlighted in Fig. 4.9a. In contrast, a clear region of intergranular combined with partially quasi-cleavage features is observed on the fracture surface of the hydrogen-charged specimen as shown in overall image of charged specimen in Fig. 4.9b and magnified in a Fig. 4.9b1. The remain part of the fracture surface shows considerable dimples as shown in a magnified image in Fig.

4.9b2.

Fig. 4.10a and 4.10b showthe overview of typical features of the fracture surfaces of uncharged specimens with 1 mm and 3.2 mm precrack as an examples of short and long precrack, respectively. The rest of figures are magnified images corresponding to the highlighted locations indicated in the overviews. The fracture surfaces of both uncharged specimens are characterized by a combination of quasi-cleavage feature regarding to the cyclic loading introduced by fatigued-precrack [16] and follows ductile features with dimples (Figs. 4.10a1, a2, b1 and b2).

Fig. 4.11 demonstrates the fracture surfaces of hydrogen-charged specimens with 1, 1.8 and 3.2 mm precrack. In case of 1 and 1.8 mm precracked specimens, the fracture surface exhibits a shallow intergranular cracking zone with secondary crack along prior austenite grain boundaries after quasi-cleavage features of precrack.

Quasi-cleavage-like features are also observed in the brittle region, as shown in Figs.

4.11a1, a2, b1 and b2. This brittle region surrounded by dimples and the rest of fracture surface shows ductile feature due to the microvoid coalescence as indicated in Figs.

4.11a3 and b3. In contrast, the charged specimens with 3.2 mm precrack shows no brittle feature on the fracture surface, as shown on an overview image of fracture surface in Fig. 4.11c. The fracture surface shows typically ductile with fully dimples same as uncharged one after quasi-cleavage features of precrack region (Figs. 4.11c1

-111

c3). This is to say, there is no effect of hydrogen charging on fracture feature of long-precracked specimens.

Fig. 4.12 shows the surface of uncharged and hydrogen-charged specimens with 1 and 3.2 mm precrack lengths at the vicinity of the fracture surface. There is no observable surface damage of fractured uncharged specimens as shown in Figs. 4.12a and b for 1 and 3.2 mm precracked specimens, respectively, and the magnified images of 4.12a1 and b1. In contrast, considerable surface damages due to hydrogen charging can be observed beside the fracture surfaces of both specimens (Figs. 4.12c and d).

However, the fraction area of these surface cracks is higher in 1 mm precrack than 3.2 precracked charged specimens as shown clearly in magnified images in Figs. 4.12c1

and d1, respectively.

Fig. 4.13 displays the brittle-fracture region length for smooth and precracked specimens under electrochemical hydrogen charging measured from fracture surfaces.

This area corresponds to the maximum length of hydrogen affected zone which can be observed on the fracture surface as a brittle area. The area fractions of the intergranular fracture decreased and replaced to the ductile fracture by increasing the precrack length. While, this brittle features diminished completely in the specimens including precrack more than 2 mm (as shown by black arrows) indicated no effect of hydrogen on failure.

4.3.3 Plastic strain rate distribution

Fig. 4.14 displays the distribution of plastic strain rate particularly in the vicinity of the crack tip of specimen with 3.2 mm precrack. This parameter is in the maximum value at the crack tip and has a reduction tendency to the far from crack tip.

Fig. 4.15 shows the calculation results of the distribution of the maximum plastic strain rate for the specimens with various precrack lengths. The strain rate presented here is an average of equivalent plastic strain rate during the 500 s of experiment correspond to the 0.5 mm displacement. Since the stress intensity factor is higher in longer precrack length the plastic strain rate has an increase tendency from 1 to 3.2 mm precrack. This tendency slightly decreases up to the convergence of curves together at approximately 2 mm in front of the crack tip.

112