13th International Conference on Fracture June 16–21, 2013, Beijing, China -7- Figure 9. Geometry and dimensions of DCB specimens made in X52 steel. Effective T stress has been determined by computed T stress distribution by Finite Element method along ligament and applying the Stress Difference Method, SDM [4]. Effective T stress distribution is presented in Figure 10. T is positive and indicates a low of constraint. Tef,c at failure is equal to (+151.657 MPa). Observation of hydrogen effects by SEM on crack paths is shown in Figure 11 on CT and DCB specimens. Presences of hydrogen also induce another feature on crack growth that will contribute to the acceleration effect. A comparison is made with CT specimen which develops higher constraint. For CT in air and hydrogen crack grows perpendicular to the principal opening stress in mode I. For DCB under hydrogen, there is facility to extend crack in perpendicular direction and this could cause microscopic deviations in the crack path (Figure 11). These deviations can lead into crack branching or kinked crack paths [12]. 0 2 4 6 8 1012141618202224 -800 -700 -600 -500 -400 -300 -200 -100 0 100 200 300 400 Xef=0.628 mm Tef=151.657 MPa DCB Specimen a/w=0.5 Tef=151.657 MPa Xef=0.628 mm Relative gradient of T-stress, χ -80 80 60 40 20 0 -20 -60 -40 T-stress, (MPa) Distance from notch-tip, r(mm) Figure 10. Evolution of the T-stress along of ligament for DCB specimen. Figure 11. Observation by SEM of the cracks formation in pipeline during its service in different environments: air and hydrogen for CT and DCB specimens. Crack propagation in air shows perlite debonding and ferrite matrix cracking as in Figure 12. In air, crack propagates by mode I in ferrite phase. Somes cracks are arrested in the perlitic phase, and continue as second crack in ferrite. Extension of second crack, is probably prolongation of the first crack in third dimension. The perlite phase play the role of crack arrestator.
RkJQdWJsaXNoZXIy MjM0NDE=