ICF13B

13th International Conference on Fracture June 16–21, 2013, Beijing, China -9- where during the process of cyclic loading the modified layer behaves much like a thin polymeric film placed on the water surface which is manifested both in dumping the deformation development and remarkable smoothing reliefto form at the specimen’s surface. The hindering formation of strain induced relief under tests by another scheme (cyclic tension) when the surface relief to a large extent determines the origin of the main crack brings to increasing by more than twice the time period prior to origin of fatigue crack in the case of the ion beam modification. Of particular interest in the light of the results obtained is observed difference in the character of the fatigue fractureunder cyclic tension and bending. It should be reminded that in the first case the crack originwas fixed at cyclic deformation not less than 0.75 NF (the total number of loading cycles). At the same time, under alternating bending tests the main crack is already initiated at cyclic load of 0.15-0.2 NF which in absolute terms corresponds to the same number of bending cycles of the non-treated specimens and ones with modified surfacelayer. Thus, for the case of cyclic bending the main crack in the specimen without the treatment originated very early but the rate of its growth was rather low in contrast with one under cyclic tension tests. This result, in principle, can be quite simply interpreted, because under the cyclic bending a surface layer is subjected to the periodic effects of maximum tensile and compressive stresses. Therefore, presence of the nanostructured layer on the surface hinders noticeable delay the origination (but not propagation!) of the crack. However, its subsequent growth is realized significantly slower that was interpreted a result of fewer microcracks arising along the grain boundaryin the specimen without treatment, while the nanostructured layer hinders the formation of microcracks and, consequently, the rate of the crack growth. Another factor that may have a significant impact on the character and the rate of the origination and propagation of the crack is its depth. Despite the fact that in the work direct measurements of the depth of a forming fatigue crack was not carried out but judging by the size of its opening the depth of a crack in a specimen of the initial state should be noticeably higher, which in it turn should bring to its more rapid growth. In contrast, under cyclic tension test a crack is originated sufficiently later that could be associated with roughly equal distribution of stresses through the cross section. As a result, it is the deformation relief of surface that can be a key factor in terms of origin of the main crack. A visible hindering in its formation has led not only to a multiple increasing of fatigue life-time of nanostructured specimens but also significantly slower rate of the main crack propagation. It is proved by the notable difference in the rate of its propagation which under LCF (when influence of processes of localized plastic deformation is dominant) amounted up to 3.5 times (nanostructured layer hinders the development of plastic deformation), while under HCF when crack growth is governed by dominant mechanisms of the brittle fracture, the difference in the rate of the crack growth made two times. In conclusion, one should point the role of multiple cracking processes, especially under conditions of high ductility of the substrate. In our view, the greater contribution to improving fatigue durability is associated with suppression of plastic deformation processes by nanostructured layer of higher strength causing substantial delay of the moment of the fatigue crack initiation. Noticeably smaller contribution to increasing the fatigue life-time is associated with the processes of multiple cracking. At the same time, usingmaterials of higher strength in contrast with substrate, as well as greater thickness of a coating will result in its cracking with their fast coalescence into the main one

RkJQdWJsaXNoZXIy MjM0NDE=