Pith. sign in

REVIEW 3 major objections 5 minor 2 references

Elastic energy reduction drives abnormal grain growth at room temperature

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

T0 review · glm-5.2

2026-07-07 18:51 UTC pith:RTOBPFDO

load-bearing objection Solid paper with a defensible central claim; the mobility concern is real but well-mitigated by convergent evidence the 3 major comments →

arxiv 2607.05298 v1 pith:RTOBPFDO submitted 2026-07-06 cond-mat.mtrl-sci physics.app-phphysics.comp-ph

Phase-field modeling of elastically driven abnormal grain growth

Yazhuo Liu , Yin Zhang , Kunqing Ding , Yichen Yang , Alejandro Barrios , Xavier Maeder , Olivier Pierron , Xing Liu
show 1 more author
Ting Zhu
This is my paper
classification cond-mat.mtrl-sci physics.app-phphysics.comp-ph
keywords abnormal grain growthelastic anisotropyphase-field modelcyclic loadinggrain boundary migrationelastic energy-momentum tensorpolycrystalline thin filmsnickel
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper argues that the preferential growth of specific crystallographic grain orientations in polycrystalline nickel thin films under cyclic loading at room temperature can be explained by a purely elastic thermodynamic driving force. When a polycrystal is loaded within its macroscopic elastic regime, grains whose crystal orientation makes them elastically softer along the loading direction have lower strain energy density. The paper shows, through phase-field simulations and micromechanical analysis, that the system can lower its total potential energy by allowing these softer grains to expand at the expense of stiffer neighbors. For FCC nickel, the softest orientation is [100], matching experiments where [100]-aligned grains grow abnormally. The framework generalizes: in BCC chromium, where elastic anisotropy differs, the preferred orientation switches to [110], while in nearly isotropic BCC tungsten, no orientation-selective growth occurs. The paper separates the thermodynamic driving force (elastic energy reduction, which is orientation-selective and survives cycle-averaging under symmetric loading because it depends quadratically on stress) from the kinetic role of cyclic loading (which enhances grain boundary mobility through local microplasticity but does not itself impose crystallographic selectivity). The local direction of grain boundary migration is governed by the jump in the elastic energy-momentum tensor across the boundary, which combines differences in strain energy density and stress work, not by strain energy density alone.

Core claim

The central discovery is that crystallographic elastic anisotropy provides a sufficient thermodynamic driving force for orientation-selective abnormal grain growth in the macroscopic elastic regime at room temperature. Grains with the lowest effective elastic stiffness along the loading direction grow preferentially because their expansion lowers the total elastic energy of the polycrystal. The preferred orientation is material-specific: [100] for FCC Ni (Zener anisotropy ratio 2.51), [110] for BCC Cr (ratio 0.70), and no selectivity for nearly isotropic BCC W (ratio 1.01). The paper also shows that under symmetric cyclic loading, the plastic-work contribution to the driving force averagesto

What carries the argument

Phase-field model with orientation-dependent elastic energy as driving force; Allen-Cahn evolution of order parameters; Eshelby inclusion solution combined with self-consistent polycrystal homogenization to compute orientation-dependent effective stiffness E_hkl; elastic energy-momentum tensor P_ij = W_el * delta_ij - sigma_ik * u_k,j whose jump across a grain boundary gives the local configurational force Lambda_el driving migration; cycle-averaging argument showing plastic work cancels over symmetric cycles while elastic contribution persists

Load-bearing premise

The model assumes that grain boundary mobility is constant and independent of crystallographic orientation. If mobility instead varies systematically with orientation — as suggested by grain boundary complexion literature — then the observed selectivity could be partially or fully caused by mobility differences rather than by elastic energy reduction.

What would settle it

If grain boundary mobility were shown to be systematically higher for [100]-aligned grains in Ni (or [110]-aligned grains in Cr) independent of elastic stiffness, the orientation selectivity could be a kinetic effect rather than a thermodynamic one, undermining the central claim that elastic energy reduction is the driving force.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • Materials with high elastic anisotropy ratios are more susceptible to elastically driven AGG under cyclic loading, providing a design criterion for selecting or engineering polycrystalline materials for fatigue-critical applications.
  • The framework predicts that the preferred AGG orientation in any cubic polycrystal can be identified from its single-crystal elastic constants and Zener anisotropy ratio, offering a rapid screening tool without full simulation.
  • Ultrafine-grained materials may be inherently unstable under cyclic elastic loading if their elastic anisotropy is large, because the thermodynamic driving force for orientation-selective coarsening scales with the square of applied stress and the compliance contrast between orientations.
  • The separation of thermodynamic driving force (elastic, orientation-selective) from kinetic activation (cyclic, mobility-enhancing) suggests that suppressing either factor — reducing anisotropy or preventing cyclic mobility enhancement — could inhibit AGG.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • If grain boundary mobility were orientation-dependent in a way correlated with elastic stiffness, the observed selectivity could be partially or fully reattributed to kinetic rather than thermodynamic effects. The paper's claim hinges on mobility being approximately isotropic, which remains unverified experimentally.
  • The framework could be extended to non-cubic crystals (hexagonal, orthorhombic) where the relationship between orientation and elastic stiffness is more complex, potentially yielding multiple preferred orientations or continuous texture gradients rather than a single preferred family.
  • If surface energy anisotropy were comparable in magnitude to elastic energy reduction in very thin films, the two mechanisms could compete or cooperate, producing orientation selections that neither mechanism alone would predict.
  • The cycle-averaging argument assumes symmetric loading; under asymmetric cycles or mean-stress conditions, the plastic work term would not cancel, potentially introducing an additional orientation-selective driving force that could either reinforce or oppose the elastic contribution.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. This manuscript investigates abnormal grain growth (AGG) in ultrafine-grained Ni thin films under small-strain cyclic loading at room temperature. The central claim is that elastic energy reduction, driven by crystallographic elastic anisotropy, provides a thermodynamically plausible driving force for orientation-selective AGG. The authors combine phase-field simulations with micromechanical analysis using the Eshelby energy-momentum tensor. The phase-field model incorporates orientation-dependent elastic energy as the driving force for GB migration. Simulations confirm that grains with the lowest effective elastic stiffness along the loading direction (e.g., [100] in FCC Ni) grow preferentially, and the framework predicts different preferred orientations for metals with different anisotropy (e.g., [110] in BCC Cr, no selectivity in nearly isotropic BCC W). The cycle-averaged driving force analysis shows that the plastic work contribution averages to zero over a symmetric cycle, while cyclic loading enhances GB mobility through defect activity.

Significance. The manuscript addresses a physically interesting and experimentally motivated phenomenon: AGG under macroscopic elastic cyclic loading at room temperature. The combination of phase-field modeling with Eshelby energy-momentum tensor analysis to interpret local GB migration is methodologically sound and provides physical insight. The falsifiable predictions for BCC W (no selectivity, A_Z ~ 1) and BCC Cr ([110] selectivity, A_Z < 1) are a notable strength, as they provide testable distinctions from the FCC Ni case. The provision of reproducible FEniCS code and a MATLAB script for computing effective elastic moduli is commendable. The cycle-averaged driving force analysis (Section 5.4, Eqs. 15-17) provides a clean thermodynamic argument for why elastic anisotropy, rather than plastic work, governs orientation selectivity.

major comments (3)
  1. Section 5.4.1, Eqs. (15)-(17): The derivation that the cycle-averaged plastic work contribution vanishes (Eq. 17) assumes that the plastic shear strain increment associated with GB migration, Δγ_p, is the same in both half-cycles (i.e., the same magnitude and sign convention for GB migration direction). This is a load-bearing assumption because the entire argument that cyclic loading provides only a kinetic (mobility) effect rather than a thermodynamic bias rests on this cancellation. If Δγ_p differs between the tensile and compressive half-cycles (e.g., due to directional asymmetry in dislocation-GB interactions or Bauschinger-type effects), the cancellation is incomplete and a residual orientation-dependent driving force could remain. The manuscript should explicitly state this assumption and discuss its plausibility, or at minimum acknowledge it as a limitation.
  2. Section 4, Eq. (8) and surrounding text: The constant, orientation-independent GB mobility M is identified as the weakest load-bearing assumption. The paper provides a three-part defense (random in-plane orientations, [100] having lowest resolved shear stress in FCC, and frequency independence across ~4 orders of magnitude). These arguments are reasonable but not fully conclusive. The residual risk is that cyclic-loading-induced defect processes at GBs (dislocation absorption, vacancy accumulation) could have rates that depend on GB character (misorientation + normal) in ways that correlate with in-plane grain orientation, even in a random polycrystal. The manuscript acknowledges this qualitatively but does not bound the potential magnitude of such an effect. A brief quantitative sensitivity analysis or a more explicit statement of the conditions under which the assumption would fail (e.
  3. Section 5.5, Fig. 9: The W and Cr simulations use the same GB energy and mobility as Ni, stated as a deliberate choice to isolate elastic anisotropy. However, the absolute values of single-crystal elastic constants differ substantially across these metals (Table 1), which affects the magnitude of the elastic driving force relative to the GB energy driving force. Since the GB energy is held fixed at 1.0 J/m² for all three, the ratio of elastic to capillary driving forces is not directly comparable across materials. The manuscript should clarify whether the predicted absence of AGG in W and presence in Cr would be robust to physically realistic variations in GB energy and mobility for these metals, or whether the result is specific to the parameter choice.
minor comments (5)
  1. Section 2: The text states the frequency reduction is 'by several orders of magnitude' and cites Supplementary S1.2, but S1.2 states the reduction is approximately 42x (8 kHz to 0.5 Hz). 'Several orders of magnitude' is imprecise; 42x is less than two orders. Recommend revising for consistency.
  2. Fig. 7: The contour plots of strain energy density would benefit from a color bar or scale to allow readers to assess the relative magnitudes of W_el across grains.
  3. Section 3, Eq. (1): The expression for potential energy Π under stress-controlled loading is standard, but the transition between stress-controlled and strain-controlled loading conditions (the latter used in simulations) could be stated more explicitly to avoid confusion about which thermodynamic potential governs in the phase-field simulations.
  4. Table 2: The relationship between model parameters (μ, β, L) and physical parameters (γ_GB, l_GB, M) is given in the text, but it would aid readability to include the physical parameter values in the table as well.
  5. Appendix A, Eq. (A5): The notation L_{ijkl}^{(0)} is used for the stiffness tensor in the local crystallographic frame, while C_{ijkl} is used elsewhere in the main text. Recommend unifying notation or explicitly stating the equivalence.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the careful and constructive review. The referee correctly identifies three areas where the manuscript's assumptions and parameter choices deserve more explicit discussion. We agree with all three points and will revise the manuscript accordingly: (1) explicitly stating the assumption of equal Δγ_p across half-cycles and acknowledging it as a limitation, (2) adding a more quantitative discussion of conditions under which the constant-mobility assumption would fail, and (3) clarifying the robustness of the W and Cr predictions to realistic variations in GB energy and mobility.

read point-by-point responses

  1. Referee: Section 5.4.1, Eqs. (15)-(17): The derivation that the cycle-averaged plastic work contribution vanishes (Eq. 17) assumes that the plastic shear strain increment associated with GB migration, Δγ_p, is the same in both half-cycles. This is a load-bearing assumption. If Δγ_p differs between tensile and compressive half-cycles, the cancellation is incomplete and a residual orientation-dependent driving force could remain. The manuscript should explicitly state this assumption and discuss its plausibility, or at minimum acknowledge it as a limitation.

    Authors: The referee is correct that the cancellation in Eq. (17) rests on the assumption that Δγ_p is the same in both half-cycles, and that this assumption is not currently stated explicitly enough in the manuscript. We will revise Section 5.4.1 to state this assumption clearly and to acknowledge it as a limitation. We agree that if Δγ_p differs between the tensile and compressive half-cycles—due to directional asymmetry in dislocation-GB interactions, Bauschinger-type effects, or other sources of loading-direction asymmetry—a residual orientation-dependent term could survive cycle-averaging. We note that the plausibility of the assumption is supported by the experimental observation that the same [100] selectivity is observed across widely different frequencies and amplitudes, which would be difficult to reconcile with a mechanism that depends on subtle asymmetries in half-cycle plasticity. Nevertheless, we agree that this does not constitute a proof, and the manuscript should acknowledge the assumption and its potential failure modes. We will add a paragraph to this effect. revision: yes

  2. Referee: Section 4, Eq. (8): The constant, orientation-independent GB mobility M is the weakest load-bearing assumption. The three-part defense is reasonable but not fully conclusive. Cyclic-loading-induced defect processes at GBs could have rates that depend on GB character in ways that correlate with in-plane grain orientation, even in a random polycrystal. The manuscript acknowledges this qualitatively but does not bound the potential magnitude. A brief quantitative sensitivity analysis or a more explicit statement of the conditions under which the assumption would fail is requested.

    Authors: We agree that the constant-mobility assumption is the most vulnerable point of the model and that the current defense, while reasonable, does not quantitatively bound the potential effect of orientation-dependent GB character on mobility. A full quantitative sensitivity analysis would require coupling the phase-field model to a sub-model of GB defect kinetics that is not currently available, so we cannot fully answer this request. However, we can and will strengthen the manuscript in two ways. First, we will add an explicit statement of the conditions under which the assumption would fail: namely, if the correlation between GB character (misorientation and normal) and in-plane grain orientation is strong enough that the resulting mobility variation exceeds the elastic driving force contrast between orientations. Second, we will provide a rough order-of-magnitude estimate of the elastic driving force per unit GB area (on the order of σ₀²ΔE·l_GB, which for the Ni parameters gives ~0.01–0.1 J/m²) and note that mobility variations of comparable relative magnitude would be needed to override the orientation selectivity. This does not fully bound the effect, but it makes the assumption more transparent and falsifiable. We acknowledge this as a partial revision. revision: partial

  3. Referee: Section 5.5, Fig. 9: The W and Cr simulations use the same GB energy and mobility as Ni. The absolute values of single-crystal elastic constants differ substantially, affecting the ratio of elastic to capillary driving forces. The manuscript should clarify whether the predicted absence of AGG in W and presence in Cr would be robust to physically realistic variations in GB energy and mobility.

    Authors: The referee raises a valid point. Because the GB energy is held fixed at 1.0 J/m² for all three materials while the elastic constants differ substantially, the ratio of elastic to capillary driving forces is not directly comparable across materials. We will address this in two ways. First, we will add a quantitative comparison: for W, the orientation-dependent variation in effective elastic stiffness Ẽ(θ₃) is negligibly small (Fig. 10(b)), so the elastic driving force for orientation-selective growth is near zero regardless of the GB energy value. The absence of AGG in W is therefore robust to realistic GB energy variations, because there is essentially no orientation-dependent elastic driving force to begin with. For Cr, the situation is different: the elastic anisotropy is strong (A_Z = 0.70), and the elastic driving force is comparable to or larger than that in Ni. Since AGG is observed in Ni where the elastic and capillary driving forces are of the same order (Fig. 6(b)), and Cr has comparable elastic anisotropy, the prediction of AGG in Cr should be robust to physically realistic GB energy variations (0.8–1.4 J/m² range). We will add this discussion to Section 5.5 and note that a full parametric study with material-specific GB energies and mobilities would strengthen the predictions but is beyond the scope of the current study. revision: yes

Circularity Check

0 steps flagged

No significant circularity found; the derivation chain is self-contained against external benchmarks.

full rationale

The paper's central claim—that elastic energy reduction drives orientation-selective AGG—is derived from standard elasticity theory (Eshelby inclusion solution, self-consistent homogenization) and a conventional phase-field free energy functional (Allen-Cahn evolution with elastic strain energy). The orientation selection for Ni ([100] preferred) follows directly from Eq. (3), which gives E_hkl as a function of single-crystal elastic constants and orientation index Gamma. The prediction of different preferred orientations for W (no AGG, near-isotropic) and Cr ([110] preferred) is an independent extrapolation using the same equations with different elastic constants (Table 1), not a fitted parameter renamed as a prediction. The GB mobility M is fitted from post-mortem experimental observations (Section 4: 'estimated from post-mortem experimental observations and is treated as an effective mobility'), but the paper explicitly states this is a kinetic parameter that 'does not alter the thermodynamic energy landscape and thus does not affect the predicted orientation selection trend.' The mobility is not used to determine which orientations grow—it only sets the timescale. The cycle-averaged driving force analysis (Eq. 17) shows the plastic work term averages to zero over a symmetric cycle, leaving only the elastic contribution, which is a mathematical result, not a fitted input. The self-citation to Zhang et al. (2020) for the diffraction elastic constant formula (Eq. 3, Appendix A) is a standard micromechanics result derived from Eshelby's inclusion theory and self-consistent homogenization—externally verifiable and not contingent on the present paper's conclusions. The citation to Chen et al. (2019) for the resolved shear stress argument is independent experimental/computational evidence. No step in the derivation chain reduces to its own inputs by construction.

Axiom & Free-Parameter Ledger

4 free parameters · 4 axioms · 0 invented entities

The model uses standard phase-field and micromechanics frameworks. No new physical entities are introduced. The free parameters (GB energy, mobility, GB thickness) are standard in phase-field models and are calibrated to experimental data or set to physically reasonable values. The key assumptions (constant mobility, constant GB energy) are simplifications that could affect the quantitative accuracy but are stated explicitly.

free parameters (4)
  • GB energy (gamma_GB) = 1.0 J/m^2
    Set within the typical range for Ni (0.8-1.4 J/m^2); treated as constant and orientation-independent to simplify the model.
  • GB mobility (M) = 5e-3 um/(s*MPa)
    Estimated from post-mortem experimental observations; treated as an effective mobility. The paper states this accelerates simulated kinetics but does not alter the thermodynamic energy landscape.
  • Diffuse GB thickness (l_GB) = 50 nm
    Set to discretize the grain boundary in the phase-field model.
  • Applied strain (epsilon_x) = 0.25%
    Chosen to match the experimental strain amplitude of 0.26%.
axioms (4)
  • domain assumption GB mobility is constant and orientation-independent
    Section 4, Eq. 8: 'L is a kinetic coefficient related to GB mobility, assumed constant and orientation-independent.' This is a load-bearing simplification; if mobility is strongly orientation-dependent, it could confound the elastic driving force mechanism.
  • domain assumption Cyclic loading can be represented by equivalent static loading for thermodynamic driving force
    Section 5.4.1, Eq. 17: Assumes M and plastic shear strain are constant over the loading cycle, so the plastic work term averages to zero over a symmetric cycle.
  • domain assumption GB energy is constant and orientation-independent
    Section 4: 'we adopt a constant value to simplify the model.' This removes GB energy anisotropy as a competing mechanism for AGG.
  • standard math Eshelby inclusion model for a spherical/circular grain in a homogeneous matrix
    Section 3 and Appendix A: Uses standard self-consistent micromechanics to derive effective elastic moduli of grain families.

pith-pipeline@v1.1.0-glm · 22975 in / 2401 out tokens · 80557 ms · 2026-07-07T18:51:37.935196+00:00 · methodology

0 comments
read the original abstract

Grain-refined metals typically exhibit high strength, yet their engineering applications are often constrained by grain coarsening under thermo-mechanical loading. Recent experiments have revealed abnormal grain growth (AGG) in ultrafine-grained Ni thin films subjected to cyclic loading at room temperature. Unlike conventional AGG, which generally requires significant plastic deformation or high temperatures, this phenomenon occurs within the regime of macroscopic elastic deformation. This AGG is characterized by the preferential growth of grains with an in-plane <100> orientation aligned with the loading direction. Here, we investigate the underlying physical mechanisms by combining phase-field simulations with micromechanical analysis. The results indicate that elastic energy reduction provides a thermodynamically plausible driving force for this orientation-selective grain growth. Phase-field simulations reveal the evolution kinetics of AGG and confirm that local grain geometry and stress states play critical roles in determining the grain growth pathway. By applying this framework to systems with varying elastic anisotropy, we establish a general approach for investigating elastically driven AGG in polycrystalline materials.

Figures

Figures reproduced from arXiv: 2607.05298 by Alejandro Barrios, Kunqing Ding, Olivier Pierron, Ting Zhu, Xavier Maeder, Xing Liu, Yazhuo Liu, Yichen Yang, Yin Zhang.

Figure 3
Figure 3. Figure 3: Its initial geometry at [PITH_FULL_IMAGE:figures/full_fig_p013_3.png] view at source ↗
Figure 8
Figure 8. Figure 8: Schematic illustration of GB migration under a symmetric cycle of applied shear stress (𝜏). (a) Positive GB velocity (𝑣) corresponding to GB migration from grain 2 to grain 1. (b) Applied shear stress (𝜏) vs. time (𝑡) within a loading cycle (𝑇) with zero mean stress. The mechanical driving force analyzed above accounts only for the elastic contribution in the phase-field model. Under cyclic loading, microp… view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

2 extracted references · 2 canonical work pages

  1. [1]

    Science 341, 1500-1502

    Abnormal Grain Growth Induced by Cyclic Heat Treatment. Science 341, 1500-1502. Peng, W., Gao, J., Lu, T., Sun, B., Zhang, X., Zhang, L., Tu, S., 2023. Insights into abnormal grain growth in copper thin films for reduced electrical resistivity: A quantitative multi -order- parameter phase-field study under finite element framework. Acta Materialia 260, 11...

    work page 2023

  2. [2]

    Nature 579, 67-72

    High-pressure strengthening in ultrafine-grained metals. Nature 579, 67-72. 31 Supplementary Material Phase-field modeling of elastically driven abnormal grain growth Yazhuo Liua, Yin Zhangb, Kunqing Dinga, Yichen Yanga, Alejandro Barriosc, Xavier Maederd, Olivier Pierrona, Xing Liue,*, Ting Zhua,* aWoodruff School of Mechanical Engineering, Georgia Insti...

    work page 2022