Abstract
The role of pre-existing mobile and immobile dislocation densities on the evolution of geometrical necessary dislocation densities (GNDs) during cyclic fatigue in shear is studied using a continuum dislocation-based model incorporated in a crystal plasticity finite element scheme. Clusters with different immobile dislocation densities are implemented in a homogeneous medium containing a certain mobile dislocation density. It is found that whether GND walls are formed around the initial immobile cluster (or not) strongly depends on the absolute values of initial mobile dislocation density and on the ratio between mobile and immobile densities. The results are discussed in terms of the apparent GND densities experimentally obtained using Laue microdiffraction.
Introduction
When metallic materials are subjected to cyclic loading, or fatigue, their internal microstructure undergoes significant reorganization long before macroscopic failure occurs. This process is driven by the movement and interaction of dislocations—line defects in the crystal lattice. A critical aspect of fatigue damage is the formation of Geometrically Necessary Dislocations (GNDs), which arise to accommodate gradients in plastic strain. These GNDs often organize into stable patterns like "veins" and "channels," which serve as precursors to crack initiation.
Despite extensive study, experimental observations often reveal heterogeneous damage patterns in seemingly identical regions of a sample. This research addresses how the "starting state" of a metal—specifically pre-existing clusters of immobile dislocations introduced during manufacturing—influences the subsequent evolution of GND patterns. By modeling these immobile clusters as "seeds" for damage, the study aims to explain why certain regions stabilize damage while others homogenize under the same loading conditions.
Methodology
The researchers utilized a Crystal Plasticity Finite Element Method (CPFEM) implemented within the DAMASK (Düsseldorf Advanced Material Simulation Kit) framework. Unlike standard models that rely on simplified hardening laws, this approach tracks eight distinct dislocation state variables, including edge and screw segments, to physically represent multiplication, annihilation, and mobility.
The simulation volume consisted of a 10 × 10 × 1 μm³ block of copper. At its center, a 1 × 1 × 1 μm³ "cluster" was embedded, characterized by a higher density of immobile dislocations. The surrounding medium contained a baseline density of mobile dislocations. The model subjected this volume to cyclic shear to observe how the interaction between the cluster and the surrounding mobile dislocations dictated GND wall formation.
Results and Discussion
The study identifies that the stability of fatigue-induced damage is primarily governed by the ratio ($R$) of mobile to immobile dislocations rather than total defect count alone.
The Role of the Dislocation Ratio
Two primary behaviors were observed based on the initial $R$ value:
- High Mobile Ratio ($R=0.33$): In scenarios where mobile dislocations are relatively abundant, GND walls form initially but eventually "dissociate" or break apart by cycle 50. The material effectively homogenizes, erasing the local memory of the initial defect.
- Low Mobile Ratio ($R=0.05$): When the cluster's immobile density significantly outweighs the surrounding mobile density, the GND walls become stable. These walls act as physical barriers, creating a local threshold stress ($τ_{th}$) that prevents dislocations from returning to their original positions upon load reversal, thus preserving the damage pattern.
Experimental Validation
The simulation results were benchmarked against Laue microdiffraction data, which maps misorientation and GND density in real-world copper samples.
As shown in the experimental maps, the three regions correspond to the simulation's regimes:
- Region 1 (Homogenization): Matches the high-ratio simulation where initial misorientations disappear over time.
- Region 2 (Reorganization): Shows a transition where patterns shift but remain present.
- Region 3 (Stable Traces): Matches the low-ratio simulation where high-density GND structures remain stable and even intensify throughout the 120 cycles.
Conclusion
This work demonstrates that the "memory" of a metal's failure path is encoded in its initial dislocation distribution. The ratio of mobile to immobile dislocations determines whether a pre-existing defect will serve as a site for stable damage accumulation or be smoothed out by cyclic loading. While the current model uses simplified cubic geometries and focuses on early-stage fatigue (up to 120 cycles), it provides a robust framework for understanding the initiation of fatigue damage. Future work will explore more complex, fractal-like defect geometries and the interactions between multiple clusters in bulk engineering alloys.
Summary updated