ICF13A

13th International Conference on Fracture June 16–21, 2013, Beijing, China -3- age within the machine is done with an optical far field microscope Questar QM100. The microscope is focused on the shallow notched area in the middle of the gauge length of the specimen shown in Fig. 2 (a). The microscope is continuously taking pictures of the surface, thus a measurement of the crack growth is possible. The miniature specimens are used for phase contrast tomography (PCT) and the relatively new developed diffraction contrast tomography (DCT) to map the grain microstructure in the gauge length of the specimen. The digital reconstruction of the three-dimensional volume contains the grain shape and distribution and also the crystallographic orientation data of every individual grain. By using these data, it is possible to calculate the stress distribution within the two-phase microstructure by means of the finite element method (FEM) in combination with an elastic-anisotropic, crystal-plastic material model. 3. Results The fatigue tests were carried out under uniaxial tension-compression (R=-1) at room temperature. As it was found in certain studies [5], a strong slip band formation in the austenite grains took place during the tests. The sample shown in Figure 3 was loaded with an stress amplitude of 400 MPa up to 1.6·106 cycles until the test was stopped. The crack initiation took place at a phase boundary between the grains 1 and 3. The crack initiation site has been analyzed by EBSD to calculate the corresponding slip systems and Schmid factors. For the austenite grain 1 a Schmid factor of Ms=0.46 was determined and the slip planes in grain 1 are oriented in the (111¯) direction; in comparison, the ferrite grain 3 has an corresponding Schmid factor of Ms=0.35. During the test, grain 1 shows a continuously increasing number of activated slip bands, thus it seems that the cracking of the phase boundary in this case is depending on the amount of accumulated micro strain caused by the local plastic deformation of grain 1. Figure 3: Crack initiation site of a fatigue sample loaded at Pa, 1.6·10 6 cycles. Figure 4 shows the optically measured crack length vs. the number of cycles for the left and right crack front. The crack shown in Figure 3 corresponds only to the first 50 µm of the final crack length. The crack length is measured starting from its crack initiation site to the left and right direction, to show the varying crack propagation rates being visible at the surface. It was found that the propagating crack follows within the first 1.3·106 cycles the phase boundary in the left direction until it reaches grain 5. In the other direction it took the same number of cycles to overcome the grain boundary between grain 1 and grain 2 to reach grain number 4. The crack propagation starts to accelerate after the crack crosses the phase boundaries to the grains 4 and 5. The individual crack propagation rate was calculated for grains 4 and 5. For the ferrite grain 4 the rate was 1.1·10-9 m/cycle, the rate for the austenite grain 5 was lower with 4.5·10-10 m/cycle. This effect can be correlated to the different kinds of active slip mechanisms in austenite and ferrite. One possible explanation of this discrepancy between the different crack propagation rates in austenite and ferrite could be the splitting of the energy on multiple slip systems (= lower da/dN in austenite), which is necessary for local plastic deformation [5]. γ α γ γ α

RkJQdWJsaXNoZXIy MjM0NDE=