ICF10C

different stresses and inspection times. It is observed that the crack contours are irregular, showing the tendency to be longer in the rolling direction than in the transverse direction of the plate (an average of 1.3 times). The meallographic observations showed that most of the cracks were initiated at elongated manganese sulfide inclusions (Figure 3); therefore, the longer crack extension in the rolling direction is attributed mainly to the elongated shape of the non-metallic inclusions. Figure 4 shows that the initial number of cracks increased with the applied external stress. In the absence of an external stress, the atomic hydrogen generated by the corrosion reaction between the sour gas and steel, diffuses towards elongated MnS inclusions, where it recombines to form molecular hydrogen, reducing the bond strength between inclusion and matrix, which produces the HIC. We suppose that effect of external stresses on the hydrogen induced cracks formation is due to the weakening of the interface between the matrix and the non-metallic inclusions by the superposed stress fields of the molecular hydrogen pressure and the applied external stress in the plate. This will promote the formation of cracks. Figure 2 shows that as charging time increases, linking between cracks takes place to form more complex shapes. This crack linking occurred preferentially in the longitudinal direction, due to the fact that the crack growth was greater in that direction. This preferential growing was more evident at the lowest external stress. Figure 2 shows also the profiles of the cracks in the thickness of the test plates. It is seen that the maximum stepwise cracking was of 10 % of the thickness. During the HIC tests, the crack linking occurred preferentially in the longitudinal direction, due to the fact that crack growth was greater in that direction. This also contributes to make the cracks longer in the longitudinal direction, with a final shape close to a rectangle with round corners. Iino(5) found that in the presence of an external stress (sufficiently high compared to zero and sufficiently low compared to the yield strength of the material) cracks linking developed by stacked arrays in a plane approximately perpendicular to the external stress axis, while in absence of an external stress, the cracks link in a stepwise manner. In the present work, it was found that the external stress did not significantly effect the manner of the formation and extension of the cracks at least in the interval of external stresses used in this work, i.e. in the presence of an external stress the cracks still link in a stepwise manner (Figure 5). From the crack contours at different inspection times, the crack length and crack growth rates were calculated. Figures 6 and 7 show the average crack growth rate and crack length in the longitudinal direction of the test plates as a function of the exposure time. Figure 6 shows that the crack growth rate decreases with the exposure time and that there is not a significant effect of the external stress on the crack growth rate; however, figure 7 shows that the presence of an external stress increases the final average crack length. This is associated with the effect of the external stress on the crack initiation. and not with the effect of the external stress on the crack growth rate. As the number of initial cracks increases, the interconnection of individual cracks is more frequent and the final crack size is larger.

RkJQdWJsaXNoZXIy MjM0NDE=