Prediction of damage-growth based fatigue life of polycrystalline materials using a microstructural modeling approach M. Sistaninia ⇑ , M. Niffenegger Paul Scherrer Institut, Nuclear Energy and Safety Research Department, Laboratory for Materials Behaviour, 5232 Villigen PSI, Switzerland article info Article history: Received 8 January 2014 Received in revised form 17 March 2014 Accepted 20 March 2014 Available online xxxx Keywords: Microstructural modeling Fatigue damage Inelastic hysteresis energy Finite element Austenitic stainless steel abstract A new finite element-based mesoscale model is developed to simulate the localization of deformation and the growth of microstructurally short fatigue cracks in crystalline materials by considering the aniso- tropic behavior of the individual grains. The inelastic hysteresis energy is used as a criterion to predict the fatigue crack initiation and propagation. This criterion in conjunction with continuum damage modeling provides a strong tool for studying the behavior of materials under cyclic loading at the level of the micro- structure. The model predictions are validated against an austenitic stainless steel alloy experimental data. The results show that a combined microstructural and continuum damage modeling approach is able to express the overall fatigue behavior of crystalline materials at the macroscale based on the micro- structural features. It correctly predicts the crack initiation on slip bands and at inclusions in low-cycle and high-cycle fatigue, respectively, in agreement with experimental observations reported in the literature. Ó 2014 Elsevier Ltd. All rights reserved. 1. Introduction Fatigue is the primary damage mechanisms of the components of rotating machineries and aircrafts, where they experience a suf- ficiently large number of cycles. It is also an important issue in the safe operation of nuclear power plants where the materials fre- quently experience fatigue damage due to cyclic thermal-shock loads. For this reason, fatigue behavior of metallic alloys has earned high interest in the design of these structural components. Fatigue failure results from cyclic loads that are far below the maximum strength of the material. This phenomenon occurs in three stages: crack initiation, growth of short cracks and growth of long cracks. The first two stages cover most of the fatigue life of metallic alloys especially in high-cycle regime. Because of that, studying the initi- ation and propagation of fatigue short cracks has been the aim of many studies over the past few decades [1–6]. The third stage which corresponds to the propagation and per- colation of the long cracks and overall failure of the materials is studied by fracture mechanics. In this regard, Paris law has been widely accepted as a promising method for quantifying the resid- ual life of the materials with a particular crack size. For short cracks optical devices as well as electron microscopes can be used to study their growth behavior in strain-controlled or stress-controlled fatigue tests [1–3]. Using these observations, a few practical relations have been suggested in the literature, e.g., [4] to predict the specifications of fatigue short cracks. Employing this methods, Mu et al. [3] showed that in most grains of austenitic stainless steel in low-cycle regime, only 1 or 2 slip systems are acti- vated. The activated slip system usually has the highest Schmid factor, except in some rare cases. Dislocation theory can also be used to study the initiation of the short fatigue cracks and also primary stages of their growth [5]. The main shortcoming related to this approach is its high compu- tational cost. This method can hardly be used for modeling a unit- cell with many grains. Using this approach, Tanaka and Mura [7] developed a dislocation pile-up model which gives a relation be- tween the local plastic shear strain amplitude and the number of cycles to crack initiation. They assumed that the cracks nucleate in a grain on the surface which has a high value of cyclic shear stress resolved from the applied stress on the slip plane in the slip direction. The number of cycles to crack nucleation, N c , in a slip band was defined by the following energy criterion: 2N c DU ¼ 4dx s ð1Þ where DU is the stored energy of dislocations per cycle, x s the spe- cific fracture energy, d a characteristic length. This model needs the local plastic strain range, Dc, as input in order to define DU in the slip band. However, there is no reliable method for calculating this critical parameter. In some studies Dc has been approximated as a http://dx.doi.org/10.1016/j.ijfatigue.2014.03.018 0142-1123/Ó 2014 Elsevier Ltd. All rights reserved. ⇑ Corresponding author. Tel.: +41 563104045. E-mail address: sistaninia@gmail.com (M. Sistaninia). International Journal of Fatigue xxx (2014) xxx–xxx Contents lists available at ScienceDirect International Journal of Fatigue journal homepage: www.elsevier.com/locate/ijfatigue Please cite this article in press as: Sistaninia M, Niffenegger M. Prediction of damage-growth based fatigue life of polycrystalline materials using a micro- structural modeling approach. Int J Fatigue (2014), http://dx.doi.org/10.1016/j.ijfatigue.2014.03.018