Pith. sign in

REVIEW 4 major objections 5 minor 70 references

Origin of Suppressed Ferroelectricity in k-Ga$_2$O$_3$: Interplay Between Polarization and Lattice Domain Walls

T0 review · 4 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read The suppressed ferroelectricity of κ-Ga2O3 arises because its 120° lattice domain walls topologically block the in-plane sliding that would reverse polarization, leaving a pinned network of polarization domain walls.

desk verdict A plausible mechanism for the κ-Ga2O3 polarization gap, with a testable prediction—but the 'topological' blocking claim outruns the evidence. read the letter →

arxiv 2507.16167 v1 pith:6YDEOTOP submitted 2025-07-22 cond-mat.mtrl-sci

Yonghao Zhu , Zhi Wang , Junwei Luo , Lin-Wang Wang This is my paper · ORCID
classification cond-mat.mtrl-sci
keywords κ-Ga2O3ferroelectricpolarizationdomainwalllatticeslidingferroelectricitymachinelearningpotentialcoercivefieldremanent
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper addresses a long-standing mismatch: first-principles theory gives κ-Ga2O3 a remanent polarization near 23 μC/cm2 and a coercive field near 3 MV/cm, while experiments measure less than 8.6 μC/cm2 and under 0.5 MV/cm. It argues that the missing physics is not defects but the material's own lattice texture. Using a machine-learned interatomic potential trained on ab initio molecular dynamics, the authors find that polarization reversal proceeds by in-plane sliding and shear of Ga–O layers, and that the 120° lattice domain walls present in epitaxial films act as topological barriers that pin polarization domain walls. The result is a stable network of residual walls that suppresses the observable polarization and coercive field, with the predicted polarization decreasing as lattice domain size shrinks, matching experiment.

What carries the argument

The load-bearing structure is the coupling between the out-of-plane polarization $P$ and the in-plane relative sliding vector $u$ between adjacent Ga–O sublayers; reversal is carried by sliding and shear, not by rigid vertical displacement. At a 120° lattice domain wall, the continuity condition $L_{A-\parallel B-} = L_{A+\parallel B+} \pm \sqrt{3}a/3$ shows that a passing polarization wall would have to insert or remove a finite slice of crystal, which is topologically forbidden when the lattice wall is atomically sharp. Two wall types complete the mechanism: PDW(100), where $u$ changes continuously and the wall moves fast, and PDW(010), where $u$ jumps across structural voids, giving slower motion that obeys Merz's creep law at low fields. This pair sets the anisotropic switching rates, and the topological blocking sets which fraction of the sample can switch.

What would settle it

Prepare a κ-Ga2O3 sample with an isolated 120° lattice domain wall and drive switching across it under a field below the intrinsic nucleation field; the model predicts polarization domain walls stop at the wall and leave residual domains, so observing complete reversal through the boundary would falsify the topological argument. A second check is to measure remanent polarization versus controlled lattice-domain size: the model predicts a clear decrease with smaller domains, as described by its equation (5).

Watch

Extended reading notes

Core claim

The paper's central claim is that the suppressed ferroelectricity of κ-Ga2O3 arises from the interplay between polarization domain walls (PDWs) and lattice domain walls (LDWs), not from defects alone. In the primitive cell, polarization reversal follows a sliding-like path: layers α and β translate along [100] while layers 1 and 2 shear, with the order parameter being the relative sliding vector $u$, and the energy barrier is 0.10 eV per formula unit. In large-scale molecular dynamics, this path makes PDW(100) propagate about twice as fast as PDW(010), and a pre-nucleated wall moves under fields as low as 0.2 MV/cm. The decisive step is a topological argument: a PDW cannot pass through a 120° LDW because continuity would require the wall width to change by $\Delta = \sqrt{3}a/3$, equivalent to inserting or removing a finite slice of crystal. In the simulated 120° multidomain supercell, PDWs stop at the LDWs, leaving about 60% of the lattice reversed; the remanent polarization is therefore $0.6P_0 = 14.7\ \mu\text{C}/\text{cm}^2$, falling to about 5 μC/cm2 for 12 nm domains, close to measured values.

Load-bearing premise

The topological-blocking argument assumes the 120° lattice wall is atomically sharp and rigid; if real lattice walls have finite width, strain, or defect-mediated flexibility, the claimed permanent pinning could fail, and the predicted suppression of polarization and coercivity would need revision.

Editorial extensions

If this is right

  • Remanent polarization in κ-Ga2O3 should be tunable through lattice-domain size, approaching the intrinsic value of about 24.6 μC/cm2 in large domains and falling to a few μC/cm2 for domains of 5–15 nm.
  • The experimentally low coercive field is a wall-propagation value, not a nucleation value, so switching can proceed fast below 0.5 MV/cm once a reversed nucleus exists.
  • Samples with pre-existing polarization walls at lattice boundaries, such as A+ ∥ B− ∥ C+, switch both polarization directions within about 2.5 ps under a few MV/cm, which is relevant for ferroelectric memory speed.
  • The roughly 2:1 anisotropy between PDW(100) and PDW(010) velocities means switching speed depends on lateral crystallographic direction and could be exploited in device layout.
  • Residual polarization is set by the non-switchable polyhedra near triple junctions rather than by the intrinsic polarization, giving a testable, microstructure-based scaling law.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Editorial inference: the same topological blocking should apply to isostructural Pna21 oxides with a/b ≈ 1/√3 and 120° rotational domains, so ε-Fe2O3, AlFeO3, and GaFeO3 may show similar microstructure-limited polarization that has been attributed to defects or stoichiometry.
  • Editorial inference: if finite-width or strained lattice walls permit partial transmission, the barrier could be tuned by strain engineering, giving a path to control the trade-off between switching speed and retained polarization.
  • Editorial inference: growing single-domain κ-Ga2O3 should recover near-intrinsic polarization, providing a direct fabrication test of the mechanism.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 5 minor

Summary. The manuscript combines DFT, a long-range machine-learned interatomic potential (DPLR), and large-scale molecular dynamics to study ferroelectric switching in κ-Ga2O3. It identifies an in-plane sliding and shear mechanism for polarization reversal with a NEB barrier of 0.10 eV/f.u., and it shows in single-crystal MD that PDW(100) propagates roughly twice as fast as PDW(010). When a reversed domain is pre-nucleated, low-field propagation (0.2 MV/cm) reproduces the experimental coercive-field range. The central claim is that 120° lattice domain walls observed in synthesized samples topologically block PDW propagation, leaving a stable network of residual domain walls that suppresses both the remanent polarization and the coercive field. The paper further proposes a domain-size-dependent expression for remanent polarization and suggests lattice-domain engineering as a tuning strategy.

Significance. If the central claim holds, this is an important contribution to a long-standing discrepancy between first-principles and experimental ferroelectric quantities in κ-Ga2O3, and it opens a promising avenue for lattice-domain engineering in Pna2_1 ferroelectrics. The work has concrete strengths: the ML potential is trained on DFT data rather than on the experimental polarization or coercive field, so the switching-pathway finding is not circular; the DPLR validation (energy RMSE 0.19 meV/atom, force RMSE 36.61 meV/Å, WC-position RMSE 0.0015 Å) is carefully benchmarked; the 24,000-atom MD simulation demonstrates a plausible pinning mechanism; and the low-field Merz-law behavior is a falsifiable prediction. The main risk is that the ``topologically forbidden'' argument is stronger than the evidence supports, and the quantitative extrapolation of remanent polarization depends on a fitted quantity.

major comments (4)
  1. [Eq. (4), Fig. 6] The proof that PDW propagation across a 120° LDW is topologically forbidden assumes that the LDW is atomically sharp and rigid, with fixed registry and no freedom to shift or reconstruct. The DPLR simulation in Fig. 5d,e shows finite-width walls in which the sliding vector rotates continuously. A finite-width or movable wall can accommodate the required registry change by local wall displacement, partial-dislocation emission, or elastic strain; these would convert the claimed infinite barrier into a finite, possibly large, activation barrier. The MD observation that a PDW halts at an LDW within about 2 ps demonstrates pinning, not topological impossibility. Please either provide a more rigorous topological treatment (for example, in terms of lattice commensuration and defect conservation) or soften the claim to ``extremely strong pinning'' and support it with barrier estimates or longer-time simulations.
  2. [Figs. 7e,f and Eq. (5)] The quantity N_ns, the number of non-switchable polyhedra per lattice domain, is inferred from the same simulation that gives the 60% switched fraction; the prediction P(L) then reuses this fitted quantity. This is a circular element in the quantitative prediction. Please supply an independent determination of N_ns (for example, from the geometry of junction 2 alone) or show that the 60% result is insensitive to simulation time and field history. The extrapolation to about 5 μC/cm2 at 12 nm domain size is an important quantitative claim and needs a sensitivity analysis.
  3. [Fig. 7 and Discussion, second paragraph] The claim that residual PDWs ``bypass slow nucleation'' is demonstrated only for the A+||B-||C+ configuration (Fig. 8), where pre-existing PDWs are present. In the A+||B+||C+ supercell (Fig. 7), nucleation still has to occur in each lattice domain; the text does not state the electric-field magnitude used in Fig. 7, and if it is the same high field (24 MV/cm) as in the single-crystal run, the nucleation bottleneck is not actually bypassed for this configuration. Please state the fields explicitly and clarify under which experimental conditions the residual-domain-wall network supplies nuclei.
  4. [Fig. 5b and Generalization] Only one multi-domain supercell geometry is simulated. The authors correctly note that other constructions are possible, and the XRD comparison selects the model for the static structure but does not validate the dynamics. The conclusion that LDWs are impenetrable should be tested for other junction geometries and LDW orientations, since strain relaxation and step formation may differ. Without such tests, the universal ``origin of suppressed ferroelectricity'' claim is tied to a single structural motif.
minor comments (5)
  1. [Fig. 1 caption] The caption lists the intermediate and final states both as ``(d)''; the final state should be labeled (e).
  2. [Paragraph above Eq. (3)] The sentence ``the experimentally observed coercive field in κ-Ga2O3, its experimentally observed coercive field can be well described'' contains a duplicated phrase and should be rewritten.
  3. [Fig. 4b caption] ``Insert'' should be ``inset,'' and the scatter-color description (``light blue'') is inconsistent with the text's description of the field ranges for the Merz-law fit.
  4. [Introduction, refs 7 and 12] The statement that previous first-principles and Landau-Ginzburg results give ~3 MV/cm cites Ref. 12, which is a general paper on the intrinsic coercive field rather than a κ-Ga2O3-specific calculation; please clarify the provenance of this value.
  5. [Methods, DFT section] The energy convergence criterion is stated as 0.001 meV; this is unusually strict and may be a typo for 0.001 eV. Please check and correct.

Circularity Check

0 steps flagged · score 0.0 of 10

Derivation is self-contained; no load-bearing step reduces to its inputs by construction.

full rationale

The paper's central claims are supported by an independent chain: the deep-learning potential is trained on ab initio MD/DFT data and benchmarked against DFT energies, forces, Wannier-center positions, phonons, and Born effective charges; no experimental remanent polarization or coercive field enters the training set or the switching simulations. The sliding-driven reversal pathway is obtained from NEB and MD rather than being imposed by the order-parameter definition, and the reported velocities, Merz-law behavior, and PDW pinning are simulation outputs, not fits to experiment. The topological-blocking argument in Eq. (4) is an analytic geometric statement with an explicitly stated sharp-LDW limit, and its conclusion is additionally corroborated by the DPLR MD observation that PDWs halt at LDWs. The domain-size extrapolation in Eq. (5) does read the non-switchable count N_ns from the same simulation's 60% switched fraction, so the Fig. 7f agreement with experiment is a post-hoc comparison rather than an independent test; however, N_ns is not fitted to the experimental polarization, so this is a limitation of the extrapolation, not circularity. No load-bearing self-citation or imported uniqueness theorem is used. The derivation chain therefore does not reduce to its own inputs.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The central claim rests on a trained ML potential (whose weights and data are not public), a sharp-LDW assumption in the topological proof, and a representative-domain assumption for the supercell. The quantitative reconciliation with experiment also uses N_ns inferred from the authors' own simulation.

free parameters (3)
  • DPLR neural network weights = Trained on 21,700 ab initio MD configurations; weights not released
    The central simulations depend entirely on the surrogate potential's accuracy, but the trained model and training set are not available for independent assessment.
  • N_ns (non-switchable polyhedra per lattice domain around junction 2) = Inferred from 60% switched fraction in Fig. 7e (roughly 40% non-switchable)
    Used in Eq. (5) to extrapolate remanent polarization versus domain size; not derived from first principles, but inferred from a single simulation.
  • Merz law parameters (v0, εa, μ) = μ = 1 at low fields; v0 and εa fit to velocity-field data in Fig. 4b
    Standard creep-law fit to simulation data; not independently predicted.
assumptions (5)
  • domain assumption PBEsol DFT accurately describes the energy landscape of κ-Ga2O3, including the sliding switching path and domain wall energies.
    All training data and the NEB barrier (0.10 eV/f.u.) come from PBEsol calculations; functional errors would propagate to the ML potential and the predicted switching dynamics.
  • domain assumption The DPLR potential trained on 21,700 configurations generalizes to 24,000-atom multi-domain supercells, long-time MD, and electric-field-driven switching.
    Validation covers energies, forces, Wannier centers, and phonons for configurations similar to training data; extrapolation to rare events and junction geometries is assumed.
  • ad hoc to paper The 120° LDW is atomically sharp and its physical width cannot adjust, making PDW crossing topologically forbidden (Eq. 4).
    The proof in Fig. 6 considers the sharp-wall limit; real walls have finite width and could accommodate the required displacement via strain or defects.
  • domain assumption The chosen rhombohedral A|B|C multi-domain supercell reproduces the experimental XRD pattern and is representative of real samples.
    The authors state this structure best reproduces XRD, but other domain arrangements could alter junction statistics and the residual PDW network.
  • standard math The modern theory of polarization (Berry phase) gives the spontaneous polarization P0 = 24.58 μC/cm2.
    Standard first-principles method; not a point of contention.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Origin of Suppressed Ferroelectricity in k-Ga$_2$O$_3$: Interplay Between Polarization and Lattice Domain Walls." pith.science (2026). https://pith.science/paper/6YDEOTOP

@misc{pith2026250716167,
  author       = {Pith},
  title        = {Pith review of: Origin of Suppressed Ferroelectricity in k-Ga$_2$O$_3$: Interplay Between Polarization and Lattice Domain Walls},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/6YDEOTOP}},
  note         = {Machine review of arXiv:2507.16167}
}
read the original abstract

The large discrepancy between experimental and theoretical remanent polarization and coercive field limits the applications of wide-band-gap ferroelectric materials. Here, using a machine-learning potential trained on ab-initio molecular dynamics data, we identify a new mechanism of the interplay between polarization domain wall (PDW) and lattice domain wall (LDW) in ferroelectric k-phase gallium oxide (Ga2O3), which reconciles predictions with experimental observations. Our results reveal that the reversal of out-of-plane polarization is achieved through in-plane sliding and shear of the Ga-O sublayers. This pathway creates strong anisotropy in PDW propagation, and crucially leads to topologically forbidden PDW propagation across the 120 degree LDWs observed in synthesized samples. The resulting stable network of residual domain walls bypasses slow nucleation and suppresses the observable polarization and coercive field. These insights highlight the potential for tailoring the ferroelectric response in k-Ga2O3 from lattice-domain engineering.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

70 extracted references · 70 canonical work pages

  1. [1]

    Detecting single viruses and nanoparticles using whispering gallery microlasers

    He L, Özdemir ŞK, Zhu J, Kim W, Yang L. Detecting single viruses and nanoparticles using whispering gallery microlasers. Nat. Nanotechnol. 6, 428-432 (2011)

    work page 2011

  2. [2]

    Ferroelectric photosensor network: an advanced hardware solution to real -time machine vision

    Cui B, et al. Ferroelectric photosensor network: an advanced hardware solution to real -time machine vision. Nat. Commun. 13, 1707 (2022)

    work page 2022

  3. [3]

    Tuning the interfacial spin-orbit coupling with ferroelectricity

    Fang M, et al. Tuning the interfacial spin-orbit coupling with ferroelectricity. Nat. Commun. 11, 2627 (2020)

    work page 2020

  4. [4]

    Ferroelectric memories

    Scott JF, Paz de Araujo CA. Ferroelectric memories. Science 246, 1400-1405 (1989)

    work page 1989

  5. [5]

    Tin-assisted heteroepitaxial PLD-growth of κ-Ga2O3 thin films with high crystalline quality

    Kneiß M, et al. Tin-assisted heteroepitaxial PLD-growth of κ-Ga2O3 thin films with high crystalline quality. APL Mater. 7, 022516 (2018)

    work page 2018

  6. [6]

    Silane-mediated expansion of domains in Si -doped κ-Ga2O3 epitaxy and its impact on the in-plane electronic conduction

    Mazzolini P, et al. Silane-mediated expansion of domains in Si -doped κ-Ga2O3 epitaxy and its impact on the in-plane electronic conduction. Adv. Funct. Mater. 33, 2207821 (2023)

    work page 2023

  7. [7]

    Epitaxial engineering of polar ε -Ga2O3 for tunable two -dimensional electron gas at the heterointerface

    Cho SB, Mishra R. Epitaxial engineering of polar ε -Ga2O3 for tunable two -dimensional electron gas at the heterointerface. Appl. Phys. Lett. 112, 162101 (2018)

    work page 2018

  8. [8]

    Phase diagram and polarization of stable phases of (Ga1−xInx)2O3

    Maccioni MB, Fiorentini V. Phase diagram and polarization of stable phases of (Ga1−xInx)2O3. Appl. Phys. Express 9, 041102 (2016)

    work page 2016

Show all 70 references
  1. [9]

    First -principles study of crystal structure, elastic stiffness constants, piezoelectric constants, and spontaneous polarization of orthorhombic Pna2 1-M2O3 (M = Al, Ga, In, Sc, Y)

    Shimada K. First -principles study of crystal structure, elastic stiffness constants, piezoelectric constants, and spontaneous polarization of orthorhombic Pna2 1-M2O3 (M = Al, Ga, In, Sc, Y). Mater. Res. Express 5, 036502 (2018)

    2018

  2. [10]

    Investigation of ferrimagnetism and ferroelectricity in Al xFe2−xO3 thin films

    Rao BN, et al. Investigation of ferrimagnetism and ferroelectricity in Al xFe2−xO3 thin films. J. Mater. Chem. C 8, 706-714 (2020)

    2020

  3. [11]

    Chemical tuning of room -temperature ferrimagnetism and ferroelectricity in ε-Fe2O3-type multiferroic oxide thin films

    Katayama T, Yasui S, Hamasaki Y, Osakabe T, Itoh M. Chemical tuning of room -temperature ferrimagnetism and ferroelectricity in ε-Fe2O3-type multiferroic oxide thin films. J. Mater. Chem. C 5, 12597-12601 (2017)

    2017

  4. [12]

    Intrinsic ferroelectric coercive field

    Ducharme S, et al. Intrinsic ferroelectric coercive field. Phys. Rev. Lett. 84, 175-178 (2000)

    2000

  5. [13]

    Comprehensive Raman study of orthorhombic κ/ε -Ga2O3 and the impact of rotational domains

    Janzen BM , et al. Comprehensive Raman study of orthorhombic κ/ε -Ga2O3 and the impact of rotational domains. J. Mater. Chem. C 9, 14175-14189 (2021)

    2021

  6. [14]

    A climbing image nudged elastic band method for finding saddle points and minimum energy paths

    Henkelman G, Uberuaga BP, Jónsson H. A climbing image nudged elastic band method for finding saddle points and minimum energy paths. J. Chem. Phys. 113, 9901-9904 (2000)

    2000

  7. [15]

    Theory of polarization of crystalline solids

    King-Smith RD, Vanderbilt D. Theory of polarization of crystalline solids. Phys. Rev. B 47, 1651-1654 (1993)

    1993

  8. [16]

    Ferroelectric polarization and magnetic structure at domain walls in a multiferroic film

    Tao A , et al. Ferroelectric polarization and magnetic structure at domain walls in a multiferroic film. Nat. Commun. 15, 6099 (2024)

    2024

  9. [17]

    Origin of ferrimagnetism and ferroelectricity in room-temperature multiferroic ε-Fe2O3

    Xu K, Feng JS, Liu ZP, Xiang HJ. Origin of ferrimagnetism and ferroelectricity in room-temperature multiferroic ε-Fe2O3. Phys. Rev. Appl. 9, 044011 (2018)

    2018

  10. [18]

    VESTA3 for three-dimensional visualization of crystal, volumetric and morphology data

    Momma K, Izumi F. VESTA3 for three-dimensional visualization of crystal, volumetric and morphology data. J. Appl. Crystallogr. 44, 1272-1276 (2011)

    2011

  11. [19]

    Intrinsic ferroelectric switching from first principles

    Liu S, Grinberg I, Rappe AM. Intrinsic ferroelectric switching from first principles. Nature 534, 360-363 (2016)

    2016

  12. [20]

    Nucleation and growth mechanism of ferroelectric domain -wall motion

    Shin Y-H, Grinberg I, Chen IW, Rappe AM. Nucleation and growth mechanism of ferroelectric domain -wall motion. Nature 449, 881-884 (2007)

    2007

  13. [21]

    Machine learning force fields

    Unke OT, et al. Machine learning force fields. Chem. Rev. 121, 10142-10186 (2021)

    2021

  14. [22]

    A deep potential model with long -range electrostatic interactions

    Zhang L, Wang H, Muniz MC, Panagiotopoulos AZ, Car R, E W. A deep potential model with long -range electrostatic interactions. J. Chem. Phys. 156, 124107 (2022)

    2022

  15. [23]

    On the role of long -range electrostatics in machine-learned interatomic potentials for complex battery materials

    Staacke CG, Heenen HH, Scheurer C, Csányi G, Reuter K, Margraf JT. On the role of long -range electrostatics in machine-learned interatomic potentials for complex battery materials. ACS Appl. Energy Mater. 4, 12562- 12569 (2021)

    2021

  16. [24]

    Anisotropic collective variables with machine learning potential for ab initio crystallization of complex ceramics

    Deng Y, Fu S, Guo J, Xu X, Li H. Anisotropic collective variables with machine learning potential for ab initio crystallization of complex ceramics. ACS Nano 17, 14099-14113 (2023)

    2023

  17. [25]

    Thermodynamics and dielectric response of BaTiO3 by data-driven modeling

    Gigli L, Veit M, Kotiuga M, Pizzi G, Marzari N, Ceriotti M. Thermodynamics and dielectric response of BaTiO3 by data-driven modeling. npj Comput. Mater. 8, 209 (2022)

    2022

  18. [26]

    Challenges, opportunities, and prospects in metal halide perovskites from theoretical and machine learning perspectives

    Myung CW, Hajibabaei A, Cha J -H, Ha M, Kim J, Kim KS. Challenges, opportunities, and prospects in metal halide perovskites from theoretical and machine learning perspectives. Adv. Energy Mater. 12, 2202279 (2022)

    2022

  19. [27]

    First-principles theory of ferroelectric phase transitions for perovskites: The case of BaTiO3

    Zhong W, Vanderbilt D, Rabe KM. First-principles theory of ferroelectric phase transitions for perovskites: The case of BaTiO3. Phys. Rev. B 52, 6301-6312 (1995)

    1995

  20. [28]

    Ultrafast switching dynamics of the ferroelectric order in stacking -engineered ferroelectrics

    He R, et al. Ultrafast switching dynamics of the ferroelectric order in stacking -engineered ferroelectrics. Acta Mater. 262, 119416 (2024)

    2024

  21. [29]

    Deep learning of accurate force field of ferroelectric HfO 2

    Wu J, Zhang Y, Zhang L, Liu S. Deep learning of accurate force field of ferroelectric HfO 2. Phys. Rev. B 103, 024108 (2021)

    2021

  22. [30]

    Unconventional ferroelectric domain switching dynamics in CuInP2S6 from first principles

    He R, Wang H, Liu F, Liu S, Liu H, Zhong Z. Unconventional ferroelectric domain switching dynamics in CuInP2S6 from first principles. Phys. Rev. B 108, 024305 (2023)

    2023

  23. [31]

    Strain -aided room -temperature second -order ferroelectric phase transition in monolayer PbTe: Deep potential molecular dynamics simulations

    Gong Z, Liu JZ, Ding X, Sun J, Deng J. Strain -aided room -temperature second -order ferroelectric phase transition in monolayer PbTe: Deep potential molecular dynamics simulations. Phys. Rev. B 108, 134112 (2023)

    2023

  24. [32]

    DeePMD -kit: A deep learning package for many -body potential energy representation and molecular dynamics

    Wang H, Zhang L, Han J, E W. DeePMD -kit: A deep learning package for many -body potential energy representation and molecular dynamics. Comput. Phys. Commun. 228, 178-184 (2018)

    2018

  25. [33]

    DeePMD-kit v2: A software package for deep potential models

    Zeng J, et al. DeePMD-kit v2: A software package for deep potential models. J. Chem. Phys. 159, 054801 (2023)

    2023

  26. [34]

    Self -consistent determination of long -range electrostatics in neural network potentials

    Gao A, Remsing RC. Self -consistent determination of long -range electrostatics in neural network potentials. Nat. Commun. 13, 1572 (2022)

    2022

  27. [35]

    Generalized neural -network representation of high -dimensional potential -energy surfaces

    Behler J, Parrinello M. Generalized neural -network representation of high -dimensional potential -energy surfaces. Phys. Rev. Lett. 98, 146401 (2007)

    2007

  28. [36]

    Deep neural network for the dielectric response of insulators

    Zhang L, Chen M, Wu X, Wang H, E W, Car R. Deep neural network for the dielectric response of insulators. Phys. Rev. B 102, 041121 (2020)

    2020

  29. [37]

    Ab initio molecular dynamics in a finite homogeneous electric field

    Umari P, Pasquarello A. Ab initio molecular dynamics in a finite homogeneous electric field. Phys. Rev. Lett. 89, 157602 (2002)

    2002

  30. [38]

    First -principles approach to insulators in finite electric fields

    Souza I, Íñiguez J, Vanderbilt D. First -principles approach to insulators in finite electric fields. Phys. Rev. Lett. 89, 117602 (2002)

    2002

  31. [39]

    A beginner's guide to the modern theory of polarization

    Spaldin NA. A beginner's guide to the modern theory of polarization. J. Solid State Chem. 195, 2-10 (2012)

    2012

  32. [40]

    Non -Kolmogorov-Avrami switching kinetics in ferroelectric thin films

    Tagantsev AK, Stolichnov I, Setter N, Cross JS, Tsukada M. Non -Kolmogorov-Avrami switching kinetics in ferroelectric thin films. Phys. Rev. B 66, 214109 (2002)

    2002

  33. [41]

    Note on ferroelectric domain switching

    Ishibashi Y, Takagi Y. Note on ferroelectric domain switching. J. Phys. Soc. Jpn. 31, 506-510 (1971)

    1971

  34. [42]

    Theoretical lower limit of coercive field in ferroelectric hafnia

    Yang J, et al. Theoretical lower limit of coercive field in ferroelectric hafnia. Phys. Rev. X 15, 021042 (2025)

    2025

  35. [43]

    Nonlinear dynamics of domain-wall propagation in epitaxial ferroelectric thin films

    Jo JY, et al. Nonlinear dynamics of domain-wall propagation in epitaxial ferroelectric thin films. Phys. Rev. Lett. 102, 045701 (2009)

    2009

  36. [44]

    Domain wall creep in epitaxial ferroelectric Pb(Zr0.2Ti0.8)O3 thin films

    Tybell T, Paruch P, Giamarchi T, Triscone JM. Domain wall creep in epitaxial ferroelectric Pb(Zr0.2Ti0.8)O3 thin films. Phys. Rev. Lett. 89, 097601 (2002)

    2002

  37. [45]

    Domain formation and domain wall motions in ferroelectric BaTiO3 single crystals

    Merz WJ. Domain formation and domain wall motions in ferroelectric BaTiO3 single crystals. Phys. Rev. 95, 690- 698 (1954)

    1954

  38. [46]

    Mechanism for the sidewise motion of 180° domain walls in barium titanate

    Miller RC, Weinreich G. Mechanism for the sidewise motion of 180° domain walls in barium titanate. Phys. Rev. 117, 1460-1466 (1960)

    1960

  39. [47]

    The real structure of ε-Ga2O3 and its relation to κ-phase

    Cora I, et al. The real structure of ε-Ga2O3 and its relation to κ-phase. CrystEngComm 19, 1509-1516 (2017)

    2017

  40. [48]

    Heteroepitaxial growth of Ga2O3 on 4H-SiC by liquid-injection MOCVD for improved thermal management of Ga2O3 power devices

    Hrubišák F, et al. Heteroepitaxial growth of Ga2O3 on 4H-SiC by liquid-injection MOCVD for improved thermal management of Ga2O3 power devices. J. Vac. Sci. Technol. A 41, 042708 (2023)

    2023

  41. [49]

    In situ TEM study of κ→β and κ→γ phase transformations in Ga2O3

    Cora I, Fogarassy Z, Fornari R, Bosi M, Re č nik A, P écz B. In situ TEM study of κ→β and κ→γ phase transformations in Ga2O3. Acta Mater. 183, 216-227 (2020)

    2020

  42. [50]

    Plan -view TEM observation of a single - domain κ -Ga2O3 thin film grown on ε -GaFeO3 substrate using GaCl 3 precursor by mist chemical vapor deposition

    Nishinaka H, Ueda O, Ito Y, Ikenaga N, Hasuike N, Yoshimoto M. Plan -view TEM observation of a single - domain κ -Ga2O3 thin film grown on ε -GaFeO3 substrate using GaCl 3 precursor by mist chemical vapor deposition. Jpn. J. Appl. Phys. 61, 018002 (2022)

    2022

  43. [51]

    Control of crystal -domain orientation in multiferroic Ga 0.6Fe1.4O3 epitaxial thin films

    Katayama T, Yasui S, Hamasaki Y, Itoh M. Control of crystal -domain orientation in multiferroic Ga 0.6Fe1.4O3 epitaxial thin films. Appl. Phys. Lett. 110, 212905 (2017)

    2017

  44. [52]

    Ferroelectric and magnetic properties in ε -Fe2O3 epitaxial film

    Hamasaki Y, Yasui S, Katayama T, Kiguchi T, Sawai S, Itoh M. Ferroelectric and magnetic properties in ε -Fe2O3 epitaxial film. Appl. Phys. Lett. 119, 182904 (2021)

    2021

  45. [53]

    Domain topology and domain switching kinetics in a hybrid improper ferroelectric

    Huang FT , et al. Domain topology and domain switching kinetics in a hybrid improper ferroelectric. Nat. Commun. 7, 11602 (2016)

    2016

  46. [54]

    Ferroelectric switching dynamics of topological vortex domains in a hexagonal manganite

    Han M-G, et al. Ferroelectric switching dynamics of topological vortex domains in a hexagonal manganite. Adv. Mater. 25, 2415-2421 (2013)

    2013

  47. [55]

    Developing fatigue-resistant ferroelectrics using interlayer sliding switching

    Bian R, et al. Developing fatigue-resistant ferroelectrics using interlayer sliding switching. Science 385, 57-62 (2024)

    2024

  48. [56]

    Operando electron microscopy investigation of polar domain dynamics in twisted van der Waals homobilayers

    Ko K, et al. Operando electron microscopy investigation of polar domain dynamics in twisted van der Waals homobilayers. Nat. Mater. 22, 992-998 (2023)

    2023

  49. [57]

    Effect of film microstructure on domain nucleation and Intrinsic switching in ferroelectric Y:HfO 2 thin film capacitors

    Buragohain P, Erickson A, Mimura T, Shimizu T, Funakubo H, Gruverman A. Effect of film microstructure on domain nucleation and Intrinsic switching in ferroelectric Y:HfO 2 thin film capacitors. Adv. Funct. Mater. 32, 2108876 (2022)

    2022

  50. [58]

    Efficient iterative schemes for ab initio total -energy calculations using a plane -wave basis set

    Kresse G, Furthmüller J. Efficient iterative schemes for ab initio total -energy calculations using a plane -wave basis set. Phys. Rev. B 54, 11169-11186 (1996)

    1996

  51. [59]

    CP2K: An electronic structure and molecular dynamics software package -Quickstep: Efficient and accurate electronic structure calculations

    Kühne TD, et al. CP2K: An electronic structure and molecular dynamics software package -Quickstep: Efficient and accurate electronic structure calculations. J. Chem. Phys. 152, 194103 (2020)

    2020

  52. [60]

    Fast plane wave density functional theory molecular dynamics calculations on multi-GPU machines

    Jia W, et al. Fast plane wave density functional theory molecular dynamics calculations on multi-GPU machines. J. Comput. Phys. 251, 102-115 (2013)

    2013

  53. [61]

    The analysis of a plane wave pseudopotential density functional theory code on a GPU machine

    Jia W, et al. The analysis of a plane wave pseudopotential density functional theory code on a GPU machine. Comput. Phys. Commun. 184, 9-18 (2013)

    2013

  54. [62]

    Restoring the density-gradient expansion for exchange in solids and surfaces

    Perdew JP, et al. Restoring the density-gradient expansion for exchange in solids and surfaces. Phys. Rev. Lett. 100, 136406 (2008)

    2008

  55. [63]

    Separable dual-space Gaussian pseudopotentials

    Goedecker S, Teter M, Hutter J. Separable dual-space Gaussian pseudopotentials. Phys. Rev. B 54, 1703-1710 (1996)

    1996

  56. [64]

    Gaussian basis sets for accurate calculations on molecular systems in gas and condensed phases

    VandeVondele J, Hutter J. Gaussian basis sets for accurate calculations on molecular systems in gas and condensed phases. J. Chem. Phys. 127, 114105 (2007)

    2007

  57. [65]

    Maximally localized Wannier functions: Theory and applications

    Marzari N, Mostofi AA, Yates JR, Souza I, Vanderbilt D. Maximally localized Wannier functions: Theory and applications. Rev. Mod. Phys. 84, 1419-1475 (2012)

    2012

  58. [66]

    Projector augmented-wave method

    Blöchl PE. Projector augmented-wave method. Phys. Rev. B 50, 17953-17979 (1994)

    1994

  59. [67]

    Optimized norm-conserving Vanderbilt pseudopotentials

    Hamann DR. Optimized norm-conserving Vanderbilt pseudopotentials. Phys. Rev. B 88, 085117 (2013)

    2013

  60. [68]

    LAMMPS - a flexible simulation tool for particle-based materials modeling at the atomic, meso, and continuum scales

    Thompson AP, et al. LAMMPS - a flexible simulation tool for particle-based materials modeling at the atomic, meso, and continuum scales. Comput. Phys. Commun. 271, 108171 (2022)

    2022

  61. [69]

    Deep potential molecular dynamics: A scalable model with the accuracy of quantum mechanics

    Zhang L, Han J, Wang H, Car R, E W. Deep potential molecular dynamics: A scalable model with the accuracy of quantum mechanics. Phys. Rev. Lett. 120, 143001 (2018)

    2018

  62. [70]

    Berry-phase treatment of the homogeneous electric field perturbation in insulators

    Nunes RW, Gonze X. Berry-phase treatment of the homogeneous electric field perturbation in insulators. Phys. Rev. B 63, 155107 (2001)

    2001

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.