Slip localization and fatigue crack nucleation near a non-metallic inclusion in polycrystalline nickel-based superalloy

Abstract

Fatigue crack nucleation at a non-metallic agglomerate inclusion has been studied by high spatial resolution Digital Image Correlation (HR-DIC) and high angular resolution Electron Backscatter Diffraction (HR-EBSD). Spatial and temporal characterization and correlation of deformation with underlying microstructures has been performed, with distributions of plastic strain measured from HR-DIC; and residual stress and density of geometrically necessary dislocations (GND) measured from HR-EBSD. Initial residual stress and GND fields, as a consequence of differing thermal expansivities in the metallic and oxide phases, localized around the agglomerate have been quantified using HR-EBSD. The localization of the pre-existing stress and dislocation states appear to lead to early onset of plasticity upon subsequent mechanical loading. Heterogeneous distributions of plastic strain have been observed in the course of the fatigue test by HR-DIC. Crack nucleation via agglomerate/nickel interface decohesion and particle fracture has been demonstrated and this is correlated with the elevation in strain and dislocation density. The measurements of residual stress, strain, and dislocation density provide key information for the mechanisms of fatigue cracking and the development of damage nucleation criteria in these material systems.2

An HR-EBSD and computational crystal plasticity investigation of microstructural stress distributions and fatigue hotspots in polycrystalline copper

“…In recent cyclic fatigue studies [39], the point-wise evolution of stored energy has been demonstrated to stabilise and achieve steady-state evolution after a few loading cycles so that in order to identify stored energy hotspots, it is not unreasonable to investigate a series of monotonically loaded polycrystals for the purposes of establishing sufficient polycrystal analyses to be statistically representative. Experimental investigations of accumulated slip at the microstructure-sensitive level also show that the form of the distribution has already been established within the first loading cycle and that subsequent cycles tend to increase the intensity of the localisations through, for example, progressive ratcheting processes [6]. In a similar manner to Fig.…”

“…Other approaches to understand generalised slip include the Taylor method [5], which assumes all grains undergo the same shape change as the entire polycrystal, driving the need for a minimum five independent slip systems to be activated simultaneously in order to maintain geometric compatibility. However, with recent developments of strain mapping techniques such as sub-micron Digital Image Correlation (DIC) [6] and crystal plasticity finite element modelling (CPFEM) [7][8][9], localised strain heterogeneity in polycrystals is observed which cannot be determined from macroscopic considerations, and the assumption of homogeneous deformation is simply wrong. In advanced CPFEM models [7][8][9], Schmid factor is utilised in constitutive formulations to determine the resolved shear stress on a given crystal system.…”

Microstructurally-sensitive fatigue crack nucleation in Ni-based single and oligo crystals

“…2. DIC analysis was conducted using in-house developed Matlab scripts in which the rigid body rotation and translation between frames were removed to improve the accuracy of strain measurement [29][30][31]. 100×100 pixels of the subset windows with an overlap of 90% were used as local strain gauges to measure and calculate strain maps [26].…”

Toward Predictive Understanding of Fatigue Crack Nucleation in Ni-Based Superalloys

“…We have developed a simple and effective surface coating method that was first reported by Jiang et al 17 where nanosize ($250 nm) silica particles were stably and uniformly applied onto the free surface of the samples. 17,21,22 These nanosize particles enable spatial high-resolution strain measurements as these highcontrast features are imaged at high-image magnifications. SEM-based imaging of the surface displacement field using our in-house MATLAB-based DIC codes enables us to obtain high-resolution, fullfield strain maps.…”