13th International Conference on Fracture June 16–21, 2013, Beijing, China -7- known about the damage on a micro scale. The conditions at the crack tip are probably controlled by the environment and the diffusion of oxygen towards and into the grain boundaries. This process is in turn controlled by the conditions at the grain boundary where slip bands, local stress, voids, oxide formation and local chemistry will affect the diffusion of embrittleing elements like oxygen. Depending on the local condition at the grain boundaries micro cracks will form and propagate till they are interfering and sometimes arrested by microstructure like δ phase or blunted at different locations in different directions creating a damaged volume where the final crack path will be formed when the ligaments between the micro cracks will break after plastic deformation or embrittlement. This crack model can explain the apparent non continuous crack growth and all the other phenomena like oxidized voids and plastic deformation together with embrittlement. Crack closure caused by local plasticity and observed mixed mode character along the crack path may also play a role both due to the unloading cycle between the cycle but also due to plasticity and oxide formation[10]. The environmental crack propagation during dwell at 700°C in Allvac718plus is observed as close to purely intergranular with limited fractographic information of microscopic direction of growth while the fatigue crack growth during the same conditions show clear crack arrest markings during intergranular growth. The fact that the there are virtually no crack arrest markings present during dwell does unfortunately not answer the question if the crack growth is continuous or not. Analysis of the crack tip with [15] advanced methods (SIMS) has shown that the penetration of oxygen is very limited in the crack tip region in front of the crack tip so the triggering of the crack growth event is probably caused by oxidation and growth of nano sized voids in the crack tip region. The voids can either be pure creep nano voids or γ’’ or γ´particles from the matrix identified in this study. The origin of the voids found in this study could be from γ’’ and/or γ´ growing under the influence of local stress and dislocation activity [10] and subsequently close to the crack tip being opened up and oxidized to form the observed pores. The oxidation of γ’’ is probable in this environment since the δ-plates exposed to oxygen, with the similar composition as γ’’ or even sometimes could be transformed to δ by dislocation particle [11] shearing, have been observed to be severely oxidized. Since the analysis with SIMS has shown that diffusion of oxygen in front of the crack tip is nonexistent a growth mechanism based on oxygen diffusion must be extremely local in character. This supports a mechanism where nanometric voids are created by a combined action of creep, transformation/oxidation of γ’’ or γ´ into pores where the pore walls are thin enough to allow transportation of oxygen before they break and gradually open up the crack. In a review by Woodford [12] the cavities are supposed to be created by oxidation of carbon into gas phase where the voids are created from gas bubbles. The bubbles are not only acting as creep cavities but also to reduce the grain boundary sliding that normally takes place to relive stress concentrations in the grain boundary at higher temperatures. This is not contradicted by the observations in this work and could also explain the lack of expected oxide in front of the crack tip and also the fact that no oxygen at all was found by SIMS in ref. [16]. This also supports the observations in this work that there is a large amount of plastic deformation involved together with intergranular fracture in IN718 and no signs of embrittlement since the oxygen is not primarily causing embrittlement according to this model. Unfortunately evidence of a possible carbon source for the creation of gas bubbles to support the model proposed in Woodford has not been observed in this study but on the other hand not specifically been a subject of study. Generally the growth of a crack during hold time constant loading is for IN718 a result of a number of growth mechanisms where microstructure together with plastic deformation and environment preferably oxygen are acting. For Allvac718Plus the dwell time effect at high temperature is more clear with pure intergranular growth mode but the growth mechanism during intergranular growth is very similar to IN718 with signs of oxidized pores in the [11]crack tip region.
RkJQdWJsaXNoZXIy MjM0NDE=