Mesoscale Domain Evolution Mechanism during Alternating Current (AC) Poling of Relaxor Ferroelectrics

Bo Wang; Long-Qing Chen; Yuan-Jie Sun

arxiv: 2605.20011 · v1 · pith:H7RODK5Mnew · submitted 2026-05-19 · ❄️ cond-mat.mtrl-sci

Mesoscale Domain Evolution Mechanism during Alternating Current (AC) Poling of Relaxor Ferroelectrics

Yuan-Jie Sun , Bo Wang , Long-Qing Chen This is my paper

Pith reviewed 2026-05-20 03:57 UTC · model grok-4.3

classification ❄️ cond-mat.mtrl-sci

keywords domain wallsAC polingrelaxor ferroelectricsphase-field simulationpolarization reversalelastic interactionsPMN-PT

0 comments

The pith

The spacing ratio between 71° and 109° domain walls decides which 71° walls get irreversibly eliminated during AC poling.

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

The paper establishes that in rhombohedral relaxor-ferroelectric crystals, the relative spacing between 71° and 109° domain walls controls whether the 71° walls survive or disappear during polarization reversal under an alternating electric field. Phase-field simulations of a quasi-two-dimensional laminated structure show that closely spaced 71° walls move out of sync because of long-range elastic interactions, breaking the expected symmetry between equivalent domain variants and causing irreversible elimination of some walls. More widely spaced 71° walls remain intact, and all 109° walls stay unchanged. A reader would care because this reveals a concrete mesoscale rule for how domain structures evolve under cyclic fields in materials used for sensors and actuators.

Core claim

In quasi-two-dimensional laminated geometries of rhombohedral Pb(Mg1/3Nb2/3)O3–PbTiO3 single crystals, the domain-wall behavior during polarization reversal under AC poling depends on the spacing ratio between the 71° and 109° domain walls. Closely spaced 71° domain walls undergo irreversible elimination due to unsynchronized motion arising from long-range elastic interactions, whereas more widely separated walls are preserved, and the 109° domain walls remain intact. A threshold ratio for elimination is identified that depends on mechanical boundary conditions.

What carries the argument

The spacing ratio between 71° and 109° domain walls, which controls the strength of long-range elastic coupling and produces collective unsynchronized wall motion during switching.

If this is right

Domain engineering can use controlled initial wall densities to achieve selective elimination of 71° walls while preserving 109° walls.
Adjusting mechanical boundary conditions changes the critical spacing ratio that triggers irreversible wall removal.
Collective elastic-driven wall motion provides a mesoscale route to break symmetry between equivalent variants during cyclic poling.
Phase-field models that include long-range strain coupling can predict which walls survive repeated polarization reversal.

Where Pith is reading between the lines

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

The same spacing-dependent elimination could appear in other ferroelectric systems that contain multiple domain-wall angles under cyclic electric fields.
Preparing crystals with deliberately varied initial domain-wall densities and measuring survival rates after AC poling would test the predicted threshold.
The mechanism may help explain differences in piezoelectric response between samples poled under different mechanical constraints.

Load-bearing premise

The quasi-two-dimensional laminated geometry together with the chosen mechanical boundary conditions accurately captures the long-range strain coupling that drives unsynchronized motion between neighboring 71° walls.

What would settle it

In-situ imaging of domain wall trajectories during AC poling that shows synchronized motion of 71° walls even when their spacing ratio falls below the simulated threshold would falsify the dependence on spacing.

Figures

Figures reproduced from arXiv: 2605.20011 by Bo Wang, Long-Qing Chen, Yuan-Jie Sun.

**Figure 1.** Figure 1: Schematic illustration of domain-configuration evolution under an antiparallel electric-field pulse. (a) Initial [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗

**Figure 2.** Figure 2: Phase-field simulation of poling dynamics and associated energy evolution in PMN–PT single crystals. (a) Do [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗

**Figure 3.** Figure 3: Distinct modes of domain-wall motion during antiparallel poling in PMN–PT. (a) Snapshot of the polarization [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗

**Figure 4.** Figure 4: Evolution of w/h ratio and domain-wall motion asymmetry. (a) Final w/h as a function of the initial w/h for lamellar heights h = 32 and 64 nm under mechanical clamping ⟨ϵ⟩ = 0 and stress-free condition ⟨σ⟩ = 0. w/h = 0 indicates complete elimination of the 71◦ domain walls. (b) Dependence of the asymmetry factor (left axis) on w/h, plotted together with the resulting w/h. The inset schematically defines th… view at source ↗

read the original abstract

Ferroelectric domain variants that are energetically equivalent are expected to remain preserved during polarization reversal under a symmetry-preserving electric field. However, recent experiments on relaxor-ferroelectric crystals have revealed irreversible elimination of inclined domain walls during AC poling, while the underlying mesoscale mechanism remains unclear. Here, we investigate the domain-wall motion during AC poling of rhombohedral Pb(Mg$_{1/3}$Nb$_{2/3}$)O$_3$--PbTiO$_3$ single crystals containing both 71$^\circ$ and 109$^\circ$ domain walls within a quasi-two-dimensional laminated geometry using phase-field simulations. The simulations reveal that the domain-wall behavior during polarization reversal depends on the spacing ratio between the 71$^\circ$ and 109$^\circ$ domain walls. Closely spaced 71$^\circ$ domain walls undergo irreversible elimination, whereas more widely separated walls are preserved, while the 109$^\circ$ domain walls remain intact. A threshold ratio for domain-wall elimination is identified and found to depend on the mechanical boundary conditions. By tracking the domain-wall trajectories during the switching process, we attribute this behavior to unsynchronized motion of neighboring 71$^\circ$ domain walls arising from long-range elastic interactions when the walls become strongly coupled. This collective motion breaks the symmetry between energetically equivalent domain variants and leads to irreversible domain-wall elimination during polarization reversal. These findings provide mechanistic insight into collective domain-wall evolution during polarization reversal and suggest that proximity-driven symmetry breaking may provide a mesoscale mechanism for domain engineering in ferroelectric materials with high domain-wall densities.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The simulations show that 71° walls in this PMN-PT setup get irreversibly eliminated only when their spacing to 109° walls falls below a threshold set by elastic coupling and boundary conditions.

read the letter

The main point is that phase-field runs of AC poling in rhombohedral PMN-PT tie irreversible 71° wall loss to a spacing ratio with the 109° walls. When the 71° walls sit close enough, their motions fall out of sync under long-range strain, so one variant wins and the wall disappears; wider spacing lets both survive. The 109° walls stay put either way. They extract a threshold that shifts with clamped versus stress-free boundaries and back it with wall trajectory plots that show the desynchronization directly. That link between ratio, collective elastic motion, and symmetry breaking is the concrete new piece. Prior work on these materials had noted wall elimination but had not isolated this spacing dependence in forward simulations of the standard TDGL equations with elastic terms. The modeling is straightforward and reproducible in principle, with no fitted parameters driving the threshold itself. The quasi-2D laminated geometry is a reasonable starting point for these layered domain patterns, yet it leaves open whether full 3D curvature and out-of-plane strain relief would move or erase the reported ratio. They also skip a systematic check on gradient energy coefficients, though that is a minor gap rather than a load-bearing flaw. The central claim holds inside the model they actually ran. This work is aimed at people who model or engineer dense domain structures in relaxor crystals. Anyone trying to predict or control wall stability under AC fields will want to see whether the spacing rule survives experiment. I would send it to peer review; the mechanism is specific enough to be worth testing and the computational evidence is clean enough to justify referee time.

Referee Report

1 major / 2 minor

Summary. The paper uses phase-field simulations based on the time-dependent Ginzburg-Landau equations with long-range elastic interactions to study AC poling in a quasi-two-dimensional laminated model of rhombohedral PMN-PT. It claims that 71° domain walls undergo irreversible elimination during polarization reversal when their spacing ratio to intact 109° walls falls below a threshold (dependent on mechanical boundary conditions), while more widely spaced 71° walls are preserved; this is attributed to unsynchronized collective motion arising from elastic coupling when walls are strongly interacting.

Significance. If the reported spacing-ratio dependence and trajectory-based attribution hold, the work supplies a concrete mesoscale mechanism for the irreversible domain-wall elimination observed in AC-poling experiments on relaxor ferroelectrics. A clear strength is the forward simulation approach: the threshold ratio and elimination outcome emerge directly from the standard equations without fitted parameters, self-referential thresholds, or circular definitions, and the collective-motion explanation is tied to explicit wall-trajectory tracking.

major comments (1)

Model setup and boundary-condition statements: the central claim that a well-defined spacing-ratio threshold governs irreversible 71° wall elimination rests on the quasi-two-dimensional laminated geometry together with the two specific mechanical boundary conditions (clamped or stress-free). In three dimensions, out-of-plane strain components and additional curvature degrees of freedom can redistribute elastic energy and modify the effective range of long-range coupling between neighboring 71° walls, which could shift or remove the reported threshold. A short 3D test case or scaling argument quantifying this effect is needed to establish whether the mechanism is robust beyond the quasi-2D approximation.

minor comments (2)

An explicit sensitivity analysis with respect to the gradient energy coefficients (the only free parameters listed) should be added to confirm that the identified threshold ratio remains stable under reasonable variations.
The precise operational definition of the spacing ratio (e.g., how distances are measured between wall mid-planes) should be stated clearly in the results section to allow direct reproduction.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the positive assessment of our work and the recommendation for minor revision. We address the major comment point by point below.

read point-by-point responses

Referee: Model setup and boundary-condition statements: the central claim that a well-defined spacing-ratio threshold governs irreversible 71° wall elimination rests on the quasi-two-dimensional laminated geometry together with the two specific mechanical boundary conditions (clamped or stress-free). In three dimensions, out-of-plane strain components and additional curvature degrees of freedom can redistribute elastic energy and modify the effective range of long-range coupling between neighboring 71° walls, which could shift or remove the reported threshold. A short 3D test case or scaling argument quantifying this effect is needed to establish whether the mechanism is robust beyond the quasi-2D approximation.

Authors: We thank the referee for this insightful comment on the model dimensionality. Our quasi-two-dimensional laminated geometry is deliberately chosen to represent the experimentally observed laminated domain structures in rhombohedral PMN-PT, in which domains extend uniformly along the out-of-plane direction; this setup allows us to isolate the in-plane spacing-ratio dependence while incorporating long-range elastic interactions via the Khachaturyan approach. We acknowledge that a full three-dimensional treatment would introduce out-of-plane strain components and additional curvature modes that could, in principle, alter the absolute elastic energy. However, because the elastic coupling between neighboring 71° walls is dominated by the in-plane compatibility constraints in the laminated configuration, the relative synchronization (or lack thereof) that produces the spacing-ratio threshold should remain qualitatively unchanged; out-of-plane contributions would shift the overall energy scale similarly for all walls without modifying the critical ratio at which collective unsynchronized motion sets in. To address the referee’s concern, we will add a concise paragraph in the revised manuscript that (i) justifies the quasi-2D approximation on the basis of experimental domain morphology and (ii) provides the scaling argument outlined above. We believe this clarification is sufficient to support the reported mechanism within the scope of a minor revision; performing new three-dimensional simulations lies beyond the present study. revision: partial

Circularity Check

0 steps flagged

Phase-field simulation of domain evolution is self-contained forward modeling

full rationale

The paper performs forward numerical integration of the standard time-dependent Ginzburg-Landau equations with long-range elastic interactions inside a quasi-2D laminated geometry. The reported threshold spacing ratio and irreversible 71° wall elimination emerge directly as simulation outcomes under the stated clamped or stress-free boundary conditions; no parameters are fitted to the target result, no self-citation supplies a uniqueness theorem or ansatz, and the central claim is not redefined in terms of itself. The derivation chain therefore remains independent of its own outputs.

Axiom & Free-Parameter Ledger

1 free parameters · 2 axioms · 0 invented entities

The model rests on the standard phase-field formulation for ferroelectrics; no new entities are postulated and the only free parameters are the usual gradient energy coefficients and mobility values taken from prior literature rather than fitted to the present AC-poling results.

free parameters (1)

gradient energy coefficients
Standard values from earlier PMN-PT phase-field studies are adopted to set the domain-wall energies; they are not re-fitted to the AC-poling trajectories.

axioms (2)

standard math Time-dependent Ginzburg-Landau equation governs polarization evolution under applied AC field and long-range elastic interactions
Invoked in the model description as the governing equation for the simulations.
domain assumption Quasi-two-dimensional laminated geometry with periodic or clamped mechanical boundaries accurately captures mesoscale strain coupling
Stated in the simulation setup section as the geometry chosen to represent the experimental crystals.

pith-pipeline@v0.9.0 · 5835 in / 1670 out tokens · 32736 ms · 2026-05-20T03:57:48.462232+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

IndisputableMonolith/Foundation/AlexanderDuality.lean alexander_duality_circle_linking unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

domain-wall behavior during polarization reversal depends on the spacing ratio between the 71° and 109° domain walls... threshold ratio for domain-wall elimination is identified and found to depend on the mechanical boundary conditions

What do these tags mean?

matches: The paper's claim is directly supported by a theorem in the formal canon.
supports: The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends: The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses: The paper appears to rely on the theorem as machinery.
contradicts: The paper's claim conflicts with a theorem or certificate in the canon.
unclear: Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

25 extracted references · 25 canonical work pages · 1 internal anchor

[1]

N. M. Brichta, L. Tegg, L. W. Giles, J. E. Daniels, J. M. Cairney,Journal of the American Ceramic Society2025,108, 5 e20346

work page
[2]

Ushakov, Q

AD. Ushakov, Q. Hu, X. Liu, Z. Xu, X. Wei, V. Y. Shur,Applied Physics Letters2021,118, 23

work page
[3]

F. Yang, L. Chen,arXiv preprint arXiv:2412.043082024

work page internal anchor Pith review Pith/arXiv arXiv
[4]

F. Yang, X. Li, D. Talbayev, L. Chen,Physical review letters2025,135, 5 056901

work page
[5]

H. Wan, C. Luo, C. Liu, W.-Y. Chang, Y. Yamashita, X. Jiang,Acta Materialia2021,208116759

work page
[6]

Xiong, Z

J. Xiong, Z. Wang, X. Yang, R. Su, W. Zhang, X. Long, Y. Liu, C. He,Scripta Materialia2022,215 114694

work page
[7]

M. J. Krogstad, P. M. Gehring, S. Rosenkranz, R. Osborn, F. Ye, Y. Liu, J. P. C. Ruff, W. Chen, J. M. Wozniak, H. Luo, O. Chmaissem, Z.-G. Ye, D. Phelan,Nature Materials2018,17718

work page
[8]

Eremenko, V

M. Eremenko, V. Krayzman, A. Bosak, H. Y. Playford, K. W. Chapman, J. C. Woicik, B. Ravel, I. Levin,Nature Communications2019,102728

work page
[9]

Zheng, T

H. Zheng, T. Zhou, D. Sheyfer, J. Kim, J. Kim, T. D. Frazer, Z. Cai, M. V. Holt, Z. Zhang, J. F. Mitchell, L. W. Martin, Y. Cao,Science2024,384, 6703 1447

work page
[10]

A. A. Bokov, H. Guo, M. Bari, Z.-G. Ye,Acta Materialia2026,309122095

work page
[11]

P´ erez-Moyet, W

R. P´ erez-Moyet, W. A. Kaminsky, Y. Cardona-Quintero, W. A. Visser, A. A. Heitmann, H. C. Robinson,Journal of the American Ceramic Society2026,109, 3 e70643

work page
[12]

Luo, W.-Y

C. Luo, W.-Y. Chang, M. Gao, C.-H. Chang, J. Li, D. Viehland, J. Tian, X. Jiang,Acta Materialia 2020,18210

work page 2020
[13]

Patel, A

S. Patel, A. Chauhan, R. Vaish,Journal of Applied Physics2015,117, 8 084102

work page
[14]

Bencan, G

A. Bencan, G. Drazic, H. Ursic, M. Makarovic, M. Komelj, T. Rojac,Nature communications2020, 11, 1 1762

work page
[15]

Schultheiß, G

J. Schultheiß, G. Picht, J. Wang, Y. Genenko, L. Chen, J. Daniels, J. Koruza,Progress in Materials Science2023,136101101

work page
[16]

E. Sun, W. Cao,Progress in Materials Science2014,65124

work page
[17]

M. L. Jablonski, S. Liu, C. R. Winkler, A. R. Damodaran, I. Grinberg, L. W. Martin, A. M. Rappe, M. L. Taheri,ACS Applied Materials & Interfaces2016,8, 5 2935

work page
[18]

Zhang, H

Y. Zhang, H. Lu, X. Yan, X. Cheng, L. Xie, T. Aoki, L. Li, C. Heikes, S. P. Lau, D. G. Schlom, L. Chen, A. Gruverman, X. Pan,Advanced Materials2019,31, 36 1902099

work page
[19]

M. W. Mervosh, C. A. Randall,Acta Materialia2026,302121627

work page
[20]

M. W. Mervosh, J.-H. Lee, C. A. Randall,Nano Convergence2025,12, 1 62

work page
[21]

Chen,Journal of the American Ceramic Society2008,91, 6 1835

L.-Q. Chen,Journal of the American Ceramic Society2008,91, 6 1835

work page
[22]

J.-J. Wang, B. Wang, L.-Q. Chen,Annual Review of Materials Research2019,49, 1 127

work page
[23]

Khakpash, H

N. Khakpash, H. Khassaf, G. Rossetti, S. Alpay,Applied Physics Letters2015,106, 8

work page
[24]

Hlinka,Ferroelectrics2007,349, 1 49

J. Hlinka,Ferroelectrics2007,349, 1 49. 13 Supporting Information Yuan-Jie Sun, Bo Wang, ∗ and Long-Qing Chen † Department of Materials Science and Engineering and Materials Research Institute, The Pennsylvania State University, University Park, PA 16802, USA I. ENERGY COMP ARISON TABLE I. Energy density differences (unit: J/(m 3)) between the final relax...

work page 2000
[25]

subsequently drives domain-wall elimination and produces the eliminated-wall configuration. 2 III. SIMULA TIONS OF SLAB SAMPLE 0 0.5 1.0 1.5 2.0 2.5 3.0 3.2 0 0.5 1.0 1.5 2.0 2.5 3.0 3.2 initial w/h ﬁnal ratio w/h slab ‹σ›=0 w/h=1.29 w/h=0.93 w/h=2.05 FIG. S3. Domain size analysis of a single slab similar to Fig. 4 with mechanical boundary condition ⟨σ⟩= ...

work page

[1] [1]

N. M. Brichta, L. Tegg, L. W. Giles, J. E. Daniels, J. M. Cairney,Journal of the American Ceramic Society2025,108, 5 e20346

work page

[2] [2]

Ushakov, Q

AD. Ushakov, Q. Hu, X. Liu, Z. Xu, X. Wei, V. Y. Shur,Applied Physics Letters2021,118, 23

work page

[3] [3]

F. Yang, L. Chen,arXiv preprint arXiv:2412.043082024

work page internal anchor Pith review Pith/arXiv arXiv

[4] [4]

F. Yang, X. Li, D. Talbayev, L. Chen,Physical review letters2025,135, 5 056901

work page

[5] [5]

H. Wan, C. Luo, C. Liu, W.-Y. Chang, Y. Yamashita, X. Jiang,Acta Materialia2021,208116759

work page

[6] [6]

Xiong, Z

J. Xiong, Z. Wang, X. Yang, R. Su, W. Zhang, X. Long, Y. Liu, C. He,Scripta Materialia2022,215 114694

work page

[7] [7]

M. J. Krogstad, P. M. Gehring, S. Rosenkranz, R. Osborn, F. Ye, Y. Liu, J. P. C. Ruff, W. Chen, J. M. Wozniak, H. Luo, O. Chmaissem, Z.-G. Ye, D. Phelan,Nature Materials2018,17718

work page

[8] [8]

Eremenko, V

M. Eremenko, V. Krayzman, A. Bosak, H. Y. Playford, K. W. Chapman, J. C. Woicik, B. Ravel, I. Levin,Nature Communications2019,102728

work page

[9] [9]

Zheng, T

H. Zheng, T. Zhou, D. Sheyfer, J. Kim, J. Kim, T. D. Frazer, Z. Cai, M. V. Holt, Z. Zhang, J. F. Mitchell, L. W. Martin, Y. Cao,Science2024,384, 6703 1447

work page

[10] [10]

A. A. Bokov, H. Guo, M. Bari, Z.-G. Ye,Acta Materialia2026,309122095

work page

[11] [11]

P´ erez-Moyet, W

R. P´ erez-Moyet, W. A. Kaminsky, Y. Cardona-Quintero, W. A. Visser, A. A. Heitmann, H. C. Robinson,Journal of the American Ceramic Society2026,109, 3 e70643

work page

[12] [12]

Luo, W.-Y

C. Luo, W.-Y. Chang, M. Gao, C.-H. Chang, J. Li, D. Viehland, J. Tian, X. Jiang,Acta Materialia 2020,18210

work page 2020

[13] [13]

Patel, A

S. Patel, A. Chauhan, R. Vaish,Journal of Applied Physics2015,117, 8 084102

work page

[14] [14]

Bencan, G

A. Bencan, G. Drazic, H. Ursic, M. Makarovic, M. Komelj, T. Rojac,Nature communications2020, 11, 1 1762

work page

[15] [15]

Schultheiß, G

J. Schultheiß, G. Picht, J. Wang, Y. Genenko, L. Chen, J. Daniels, J. Koruza,Progress in Materials Science2023,136101101

work page

[16] [16]

E. Sun, W. Cao,Progress in Materials Science2014,65124

work page

[17] [17]

M. L. Jablonski, S. Liu, C. R. Winkler, A. R. Damodaran, I. Grinberg, L. W. Martin, A. M. Rappe, M. L. Taheri,ACS Applied Materials & Interfaces2016,8, 5 2935

work page

[18] [18]

Zhang, H

Y. Zhang, H. Lu, X. Yan, X. Cheng, L. Xie, T. Aoki, L. Li, C. Heikes, S. P. Lau, D. G. Schlom, L. Chen, A. Gruverman, X. Pan,Advanced Materials2019,31, 36 1902099

work page

[19] [19]

M. W. Mervosh, C. A. Randall,Acta Materialia2026,302121627

work page

[20] [20]

M. W. Mervosh, J.-H. Lee, C. A. Randall,Nano Convergence2025,12, 1 62

work page

[21] [21]

Chen,Journal of the American Ceramic Society2008,91, 6 1835

L.-Q. Chen,Journal of the American Ceramic Society2008,91, 6 1835

work page

[22] [22]

J.-J. Wang, B. Wang, L.-Q. Chen,Annual Review of Materials Research2019,49, 1 127

work page

[23] [23]

Khakpash, H

N. Khakpash, H. Khassaf, G. Rossetti, S. Alpay,Applied Physics Letters2015,106, 8

work page

[24] [24]

Hlinka,Ferroelectrics2007,349, 1 49

J. Hlinka,Ferroelectrics2007,349, 1 49. 13 Supporting Information Yuan-Jie Sun, Bo Wang, ∗ and Long-Qing Chen † Department of Materials Science and Engineering and Materials Research Institute, The Pennsylvania State University, University Park, PA 16802, USA I. ENERGY COMP ARISON TABLE I. Energy density differences (unit: J/(m 3)) between the final relax...

work page 2000

[25] [25]

subsequently drives domain-wall elimination and produces the eliminated-wall configuration. 2 III. SIMULA TIONS OF SLAB SAMPLE 0 0.5 1.0 1.5 2.0 2.5 3.0 3.2 0 0.5 1.0 1.5 2.0 2.5 3.0 3.2 initial w/h ﬁnal ratio w/h slab ‹σ›=0 w/h=1.29 w/h=0.93 w/h=2.05 FIG. S3. Domain size analysis of a single slab similar to Fig. 4 with mechanical boundary condition ⟨σ⟩= ...

work page