Topological defects in two-dimensional orientation-field models for grain growth

Two-dimensional orientation-field models are commonly used to simulate grain growth in polycrystalline materials. These models employ a continuous scalar field to represent crystallographic orientation. The unit circle serves as the order parameter space, which is not simply connected. This topological property leads to challenges in modeling grain boundary behavior. Singularities may form at trijunctions, and multiple grain boundary solutions exist that cannot be continuously transformed into one another. These solutions can coexist along a single grain boundary, requiring a singular point defect between them. Such defects are not well understood in classical grain boundary theory. They also introduce numerical challenges like lattice pinning in simulations. This gap motivated the development of new formulations that address these topological limitations.

The aim of this work is to address the topological limitations of two-dimensional orientation-field models for grain growth. These models face challenges due to the non-simply connected nature of the unit circle as an order parameter space. The study seeks to develop alternative formulations that eliminate the topological stability of singular defects. The motivation arises from the inability of classical models to handle multiple grain boundary solutions and the resulting numerical difficulties. By introducing new formulations, the researchers aim to improve the accuracy and robustness of grain growth simulations. The study focuses on modifying the order parameter space to be simply connected. This change is expected to remove the topological constraints that lead to singularities. The proposed solutions aim to provide a more reliable framework for modeling grain boundary dynamics.

The researchers propose two alternative formulations to address the topological issues in orientation-field models. The first approach uses a three-component unit vector field as the order parameter. This vector field allows for a more flexible representation of orientation. The second formulation employs a two-component vector field combined with an additional potential. Both methods introduce an extra degree of freedom to the system. This additional freedom changes the order parameter space from non-simply connected to simply connected. The first formulation relies on a three-dimensional vector space, while the second uses a two-dimensional vector with a constraint. The additional potential in the second method ensures the order parameter remains within the desired range. These modifications are designed to eliminate the topological stability of singular defects.

The first formulation, based on a three-component unit vector field, successfully removes the topological stability of singular defects. The second formulation, using a two-component vector field with an additional potential, also achieves the same result. Both approaches transform the order parameter space into a simply connected domain. This change eliminates the need for singular point defects between different grain boundary solutions. The models no longer exhibit the topological constraints that previously led to singularities. Numerical simulations using these formulations show improved behavior, with fewer lattice pinning issues. The grain boundary solutions can now continuously transform into one another. The results demonstrate that the proposed formulations effectively address the limitations of the standard orientation-field models.

The authors conclude that the proposed formulations successfully eliminate the topological stability of singular defects in orientation-field models. The three-component unit vector field and the two-component vector field with an additional potential both achieve a simply connected order parameter space. This change removes the need for singular point defects between grain boundary solutions. The models no longer exhibit the topological constraints that previously led to numerical difficulties. The results suggest that these formulations provide a more reliable framework for simulating grain growth. The improved behavior of the models indicates that the additional degrees of freedom are essential for addressing the topological issues. The authors propose that these formulations offer a promising alternative to standard orientation-field models. The study highlights the importance of considering topological properties in phase-field modeling.