A Nonlocal Orientation Field Phase-Field Model for Misorientation- and Inclination- Dependent Grain Boundaries

We propose to incorporate grain boundary (GB) anisotropy in phase-field modeling by extending the standard partial differential equations formulation to include a nonlocal functional of an orientation field. Regardless of the number of grains in the simulation, the model uses a single orientation field and incorporates grain misorientation and inclination information obtained from sampling the orientation field at optimized locations in the vicinity of the grain boundary. The formalism enables simple and precise tuning of GB energy anisotropy while reducing an extensive fitting procedure. The functional includes an explicit GB anisotropy function to control the GB energy as a function of both misorientation and inclination. The model is validated by reproducing the linear grain growth rate, Wulff shapes with varying misorientations and anisotropic coefficients, and analytical equilibrium dihedral angles at triple junctions. Polycrystalline simulations demonstrate grain growth, coalescence, triple junction behavior, and the influence of anisotropy on grain morphology.

💡 Research Summary

The authors introduce a novel phase‑field framework that captures both misorientation‑ and inclination‑dependent grain‑boundary (GB) energetics using a single scalar orientation field, θ(x). Traditional multi‑phase‑field (MPF) approaches require a separate field for each grain, leading to poor scalability, while earlier orientation‑field models (e.g., KWC) cannot represent arbitrary GB energy functions because the local field alone does not contain enough information about the two adjoining grains. To overcome these limitations, the paper proposes a “non‑local” free‑energy functional that explicitly samples the orientation field at points displaced a distance Δ (the equilibrium GB width) on either side of the current location along the GB normal. These sampled values, denoted θ⁺ and θ⁻, serve as proxies for the bulk crystal orientations of the two grains that meet at the GB. The misorientation is then Δθ = |θ⁺ – θ⁻|, and the GB normal n̂ provides the inclination angle.

The total free‑energy density consists of two contributions weighted by a smooth function w(|∇θ|): an “inner” term that dominates inside the GB and contains the anisotropic GB energy function γ_aniso(Δθ, n̂), and an “outer” term that dominates in the transition region and contains an isotropic function γ_iso(Δθ). Both γ functions are prescribed analytically; γ_iso follows a Read‑Shockley‑type dependence on Δθ, while γ_aniso adds a cosine modulation to encode crystal symmetry (m‑fold) and a strength parameter β that controls the degree of inclination anisotropy. Additional terms include a gradient‑penalty s|∇θ|² for numerical stability and a double‑well potential ψ(θ) that forces θ toward the bulk values θ⁺ or θ⁻ in the outer region.

Applying the calculus of variations yields an Allen‑Cahn‑type evolution equation ∂θ/∂t = –M δF/δθ, where M is a (here constant) mobility. Because the non‑local sampling distance Δ is fixed, the functional derivative does not generate higher‑order non‑local terms; the contribution of the rotation of ∇θ in the definition of θ⁺/θ⁻ vanishes to first order, preserving the conventional PDE structure. The weighting function w(|∇θ|) smoothly transitions from 1 (deep inside the GB) to 0 (in the bulk), ensuring that the inner anisotropic term is active only where the GB actually exists.

The model is validated through a series of benchmark tests. First, a planar GB exhibits the classic curvature‑driven growth law R ∝ t¹ᐟ², confirming that the kinetic formulation is correct. Second, equilibrium shapes (Wulff constructions) generated for various prescribed γ(Δθ, inclination) match analytical predictions, demonstrating that the anisotropy can be tuned precisely by adjusting β and the symmetry parameters. Third, triple‑junction simulations reproduce the analytical dihedral angles dictated by Young’s law for anisotropic surface tensions, confirming that the model correctly balances forces at grain‑boundary nodes.

Large‑scale polycrystalline simulations further illustrate the practical utility of the approach. When β is set to zero, grains evolve into near‑circular shapes, as expected for isotropic GB energy. Increasing β leads to pronounced faceting aligned with the preferred crystallographic directions, and the grain‑size distribution evolves more slowly because high‑misorientation boundaries carry higher energy and thus lower mobility. Importantly, the computational cost scales linearly with the number of grid points, not with the number of grains, because only a single θ field is solved.

The authors discuss several advantages: (i) memory and CPU savings due to the single‑field formulation, (ii) straightforward incorporation of arbitrary misorientation‑inclination functions without an extensive fitting procedure, and (iii) compatibility with existing phase‑field solvers because the governing equations remain standard PDEs. Limitations are also acknowledged. The current implementation is restricted to 2‑D systems with a single rotation axis; extending to full 3‑D will require a vectorial orientation representation and a more sophisticated sampling scheme. Moreover, the fixed sampling distance Δ assumes a constant GB width; in situations where the width varies (e.g., temperature‑dependent GBs), the accuracy may degrade.

Future work outlined includes (a) dynamic adaptation of Δ based on local GB thickness, (b) generalization to three‑dimensional orientation tensors, and (c) coupling the prescribed γ functions to atomistic GB energy databases, thereby achieving truly material‑specific GB anisotropy without empirical fitting.

In summary, this paper presents a compact, efficient, and mathematically transparent phase‑field model that captures the full richness of grain‑boundary energetics—both misorientation and inclination dependence—through a non‑local orientation‑field formulation. The approach bridges the gap between atomistic GB energy calculations and mesoscopic microstructure evolution, offering a powerful tool for predictive modeling of polycrystalline materials.