Pith. sign in

REVIEW 4 major objections 4 minor 1 cited by

Superlubric sliding ferroelectricity

T0 review · 4 major / 4 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read The paper predicts that inserting a graphene or BN monolayer between the layers of 3R bilayer MoS2 reduces the ferroelectric switching barrier by about two or one orders of magnitude, to 0.0267 and 0.167 meV/atom respectively, making the…

desk verdict Credible DFT proposal for superlubric sliding ferroelectricity, but the practical claims need thermal-stability and methodology support before they can be taken at face value. read the letter →

arxiv 2501.16118 v1 pith:7B6ZA2OM submitted 2025-01-27 cond-mat.mtrl-sci cond-mat.mes-hall

Zihao Yang , Menghao Wu This is my paper · ORCID
classification cond-mat.mtrl-scicond-mat.mes-hall
keywords slidingferroelectricitystructuralsuperlubricityincommensurateinterfacesvanderWaalsheterostructuresMoS2bilayerferroelectricswitchingbarriersandwichtrilayerdensityfunctionaltheory
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 predicts a way to make sliding ferroelectrics switch with almost no energy cost: separate the two ferroelectric layers by a different monolayer whose lattice does not match theirs, creating an incommensurate interface. In 3R-stacked bilayer MoS2, inserting a graphene monolayer lowers the switching barrier from 2.38 meV/atom to 0.0267 meV/atom, about two orders of magnitude, while inserting BN lowers it to 0.167 meV/atom; the voltage needed to switch is reduced by about one order of magnitude. The vertical polarization survives the spacer: 0.04 pC/m with graphene, 0.53 pC/m with BN, compared with 0.5 pC/m in the bilayer. The same recipe works in SnS2/BN/SnS2, where the barrier drops from 10.7 to 0.65 meV/atom. If correct, this superlubric sliding ferroelectricity would make data writing, energy harvesting, and optoelectronic switching faster and cheaper in van der Waals ferroelectrics.

What carries the argument

The mechanism is structural superlubricity at incommensurate interfaces: when two crystal surfaces have mismatched periods, atomic lateral forces tend to cancel and the sliding energy landscape flattens. The central object is a sandwich trilayer, a ferroelectric homobilayer separated by a different monolayer with a distinct lattice, in which the polarization comes from symmetry breaking in the across-layer stacking configuration of the two outer identical layers. The flattened potential energy surface lowers the ferroelectric switching barrier, and the paper verifies this by showing that differential charge density variations along the switching path, as well as interlayer distance variations, are much smaller for the sandwich systems than for the commensurate bilayer.

What would settle it

Grow or fabricate a MoS2/Gr/MoS2 stack, pole it into one stacking state, and monitor its out-of-plane polarization over time at 300 K without an applied field; if it spontaneously reverts to a mixed AB/BA domain state on experimental timescales, the superlubric barrier is too low for practical ferroelectric switching.

Watch

Extended reading notes

Core claim

The central claim is that inserting a different monolayer between a sliding ferroelectric homobilayer preserves the across-layer stacking polarization while flattening the potential energy surface for sliding, because the two outer-layer/spacer interfaces are incommensurate. First-principles calculations show that the barrier for switching 3R bilayer MoS2 drops by about two orders of magnitude with a graphene spacer and one order of magnitude with a BN spacer, while the required switching voltage drops by about an order of magnitude. The paper further shows that the near-vanishing barrier is accompanied by much smaller differential charge density fluctuations and much smaller interlayer distance variations along the sliding path, and that the same mechanism transfers to SnS2/BN/SnS2 and to twisted variants. The authors argue that this is a general class of superlubric sliding ferroelectrics, not a special case of MoS2.

Load-bearing premise

The load-bearing premise is that the computed AB and BA stackings remain stable ferroelectric states at room temperature; at a barrier of 0.0267 meV/atom, thermal energy at 300 K is about 25 meV, so without an additional pinning mechanism the polarization could reverse on its own.

Editorial extensions

If this is right

  • The switching voltage in MoS2/Gr/MoS2 and MoS2/BN/MoS2 should be about an order of magnitude lower than in bilayer MoS2, enabling lower-energy ferroelectric memory and data writing.
  • Dragging the top MoS2 layer along the zigzag direction repeatedly reverses the polarization with the same ultralow sliding barrier, producing an alternating voltage suitable for nanogenerator energy harvesting.
  • MoS2/BN/MoS2 becomes a direct bandgap semiconductor with a gap of 0.92 eV and electrons and holes concentrated on opposite layers, which should favor photovoltaics and photodetection.
  • The recipe transfers to other sandwich systems, such as SnS2/BN/SnS2, whose barrier drops from 10.7 to 0.65 meV/atom, so the phenomenon should be general across van der Waals homobilayers.
  • Small twist angles between the outer layers produce ferroelectric moiré domains, and homotrilayers with small angles can host overlapping moiré patterns tunable by sliding, which the paper connects to neuromorphic computing.

Reading between the lines

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

  • The paper leaves implicit that the lowest predicted barrier, 0.0267 meV/atom, is far below thermal energy at 300 K, so practical stability would require additional pinning, contacts, or a lower operating temperature; the paper does not test this.
  • Because the spacer weakens interlayer hybridization while preserving polarization, choosing spacers by dielectric screening could tune both the switching barrier and the electronic bandgap in the same device.
  • The same sandwich geometry could be used to test whether the microscopic charge rearrangement that controls friction also controls ferroelectric switching, since both are tied to differential charge density fluctuations along the sliding path.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 4 minor

Summary. This manuscript proposes 'superlubric sliding ferroelectricity' in sandwich trilayers in which a sliding ferroelectric homobilayer (3R MoS2, SnS2, or BN) is separated by a monolayer with a different lattice constant (graphene or BN), so that the interfaces are incommensurate. Using DFT-PBE-D2, NEB, and dipole-correction calculations, the authors report that the switching barrier of bilayer MoS2 drops from 2.38 meV/atom to 0.0267 meV/atom in MoS2/Gr/MoS2 and to 0.167 meV/atom in MoS2/BN/MoS2, with vertical polarizations of 0.04 and 0.53 pC/m, respectively. They further claim that the required switching voltage is about one order of magnitude lower, that a similar barrier reduction occurs for SnS2/BN/SnS2, that twisted BN/Gr/BN can become superlubric, and that these systems could be used for low-voltage memory, nanogenerators, and optoelectronics.

Significance. The concept is a conceptually attractive extension of sliding ferroelectricity: using structural superlubricity to flatten the switching barrier while retaining an out-of-plane polarization. The qualitative DFT trends are plausible, and the differential charge density analysis provides a useful microscopic diagnostic for the reduced electrostatic corrugation. The MoS2/BN/MoS2 case, with a reduced barrier and slightly enhanced polarization, is especially interesting. A clear strength is that the barriers are computed from first principles rather than fitted to the target quantity, so the central numerical result is not circular. However, the significance of the ultralow barriers is double-edged: the same flattening that enables low-voltage switching also threatens room-temperature retention, and the voltage-reduction claim is asserted without a quantitative model. The practical memory and energy-harvesting claims are therefore not yet established, although the underlying prediction is worth reporting after revision.

major comments (4)
  1. [Fig. 2 and the following paragraph] The central claim of room-temperature switchable ferroelectricity is not supported by the data as presented. The reported barriers of 0.0267 meV/atom for MoS2/Gr/MoS2 and 0.167 meV/atom for MoS2/BN/MoS2 correspond to total supercell barriers of order 0.1–1 meV even for a modest cell, well below kBT ≈ 25 meV at 300 K. The manuscript provides no molecular dynamics, no estimate of the attempt frequency or thermal flipping rate, and no free-energy barrier calculation, so the system could spontaneously slide between AB and BA stackings and behave as a paraelectric rather than a ferroelectric. The per-atom normalization in Fig. 2 obscures this issue; the relevant retention quantity is the activation free energy for reversing a ferroelectric domain, which is not computed. Please provide a finite-temperature kinetic analysis or explicitly restrict the claims to low-temperature operation.
  2. [Paragraph beginning 'As a result, the required voltage for switching...'] The predicted reduction of the switching voltage by about one order of magnitude is asserted directly from the NEB barrier reduction, without a model of how voltage maps to barrier in a sliding ferroelectric. If the coercive field scales as ΔE/(2PA) or a similar expression, then the polarization drop in MoS2/Gr/MoS2 (0.5 to 0.04 pC/m) partially cancels the barrier reduction, and the net voltage change must be quantified with an explicit formula. Please provide the model and the numerical inputs used for the voltage estimate.
  3. [Computational methods and Fig. 1(b)] The proposal is built on incommensurate interfaces, but the DFT calculations never specify the supercell construction used for the purportedly incommensurate contacts. The reader cannot tell whether a finite commensurate supercell with a particular moiré pattern, a strained cell, or an averaged-potential approximation was used; the reported barrier values may depend on that choice. Please state the supercell dimensions, the relative rotation or lattice mismatch, the strain applied to each layer, and the convergence of the barrier with respect to cell size and k-point sampling.
  4. [NEB sentence in Methods] For barriers as small as 0.0267 meV/atom, numerical convergence of the NEB calculation is critical, yet the manuscript does not report the number of images, whether the climbing-image variant was used, or the residual forces on the minimum-energy path. Please report these details and demonstrate that the barrier values are converged with respect to NEB parameters and electronic relaxation; otherwise the extremely low numbers could be within numerical noise.
minor comments (4)
  1. [Introduction] There is a typo in the phrase 'their swi tching barriers'; it should read 'switching barriers'.
  2. [Fig. 5 caption] The caption reads 'Ferroelectric switching pathways of (a) bilayer SnS2' with a duplicated '(a)' label; please renumber the panels consistently.
  3. [Throughout] The manuscript uses 'per unitcell' and similar run-together terms; please add the missing spaces and check hyphenation of compound adjectives such as 'incommensurate-contact interfaces'.
  4. [Fig. 2 and Fig. 3 text] The claim that 'the almost overlapping red and black line for the MoS2/Gr/MoS2 system' indicates superlubricity would be strengthened by reporting the actual numerical difference in plane-averaged differential charge density between maximum and minimum positions, rather than only visual overlap.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the superlubric switching barriers are direct DFT/NEB outputs, not fitted to the claimed effect.

full rationale

The central claimed results are the switching barriers and polarizations for MoS2/Gr/MoS2 and MoS2/BN/MoS2, with SnS2/BN/SnS2 and twisted BN/Gr/BN as additional examples. These quantities are obtained directly from DFT total energies and NEB calculations: the barrier values (2.38, 0.0267, and 0.167 meV/atom for bilayer MoS2, MoS2/Gr/MoS2, and MoS2/BN/MoS2, respectively) are computational outputs, not parameters fitted to reproduce the claimed barrier reduction or voltage reduction. The lateral forces are defined via f_lateral = dE/dD, so they inherit the same computed energy surface, but this is a definitional derivative rather than a circular fit. The authors' self-citations (e.g., Refs. [1,2,42,43]) establish the earlier sliding-ferroelectricity framework and prior predictions of across-layer stacking polarization, but the present polarization and barrier numbers do not reduce to those citations: they are independently computed in this work. The ambient-stability claims for earlier homobilayer systems are additionally supported by the many external experimental references cited (e.g., Refs. [3-24]). The absence of finite-temperature or kinetic analysis for the ultralow barriers is a legitimate physical-support limitation for the practical ferroelectricity claim, but it is a correctness/validity concern, not an instance of a prediction reducing by construction to its input. No circular step can be exhibited from the paper's equations or definitions.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

The central computational predictions depend on standard DFT approximations and on two unstated assumptions: that a small supercell captures the incommensurate limit, and that the tiny computed barriers still permit a stable ferroelectric state. The voltage-reduction claim adds an extra proportionality assumption that is not justified in the text.

assumptions (4)
  • domain assumption PBE + DFT-D2 accurately describes sliding potential energy surfaces for van der Waals heterostructures.
    All barrier and polarization values rely on this approximation; no validation against experiments or higher-level theory is provided.
  • domain assumption A finite commensurate supercell represents the incommensurate interface limit.
    The paper does not specify the supercell sizes or lattice-matching construction, nor demonstrate convergence of the barrier to the true incommensurate limit.
  • ad hoc to paper The ultralow switching barriers still yield a stable, switchable ferroelectric state at room temperature.
    The paper claims performance enhancement without analyzing thermal flipping rates; barriers around 0.0267 meV/atom are far below kT at 300 K.
  • ad hoc to paper Switching voltage scales directly with the NEB energy barrier.
    The claim that voltage is reduced by about 1 order of magnitude assumes a proportionality that is not derived and ignores the reduced polarization in the graphene case.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Superlubric sliding ferroelectricity." pith.science (2026). https://pith.science/paper/7B6ZA2OM

@misc{pith2026250116118,
  author       = {Pith},
  title        = {Pith review of: Superlubric sliding ferroelectricity},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/7B6ZA2OM}},
  note         = {Machine review of arXiv:2501.16118}
}
read the original abstract

Sliding ferroelectricity may emerge in many van der Waals bilayers/multilayers and the low switching barriers render ultrafast data writing with low energy cost. We note that such barriers are still much higher compared with structural superlubricity, and in this paper we propose a type of superlubric sliding ferroelectricity in homobilayers separated by a different layer that leads to unprecedented low switching barriers due to incommensurate interfaces. For example, the switching barrier of 3R bilayer MoS2 will be respectively reduced by around 2 or 1 order of magnitudes if they are separated by a graphene or BN monolayer, and the required voltage for switching can be about 1 order of magnitude lower. Such superlubric sliding ferroelectricity widely exists in various similar sandwich trilayer systems where the polarizations stem from symmetry breaking in across-layer stacking configurations, and with ultralow barriers of superlubric sliding, their performances for various applications are greatly enhanced compared with homobilayer sliding ferroelectrics.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. The interplay of ferroelectricity and magneto-transport in non-magnetic moir\'{e} superlattices

    cond-mat.mtrl-sci 2025-07 conditional novelty 5.0 of 10

    Graphene/hexagonal-boron-nitride moiré stacks show ferroelectric polarization that magnetic fields enhance up to room temperature, with purely electronic origin, and this polarization can suppress Shubnikov-de Haas os...

Reference graph

Works this paper leans on

55 extracted references · 53 canonical work pages · cited by 1 Pith paper

  1. [1]

    Li and M

    L. Li and M. Wu, ACS Nano 11, 6382 (2017)

    work page 2017

  2. [2]

    Wu and J

    M. Wu and J. Li, Proceedings of the National Academy of Sciences 118, e2115703118 (2021)

    work page 2021

  3. [3]

    Yasuda, X

    K. Yasuda, X. Wang, K. Watanabe, T. Taniguchi, and P. Jarillo -Herrero, Science 372, 1458 (2021)

    work page 2021

  4. [4]

    Vizner Stern et al., Science 372, 1462 (2021)

    M. Vizner Stern et al., Science 372, 1462 (2021)

    work page 2021

  5. [5]

    Wang et al., Nature 629, 74 (2024)

    L. Wang et al., Nature 629, 74 (2024)

    work page 2024

  6. [6]

    M. Lv, J. Wang, M. Tian, N. Wan, W. Tong, C. Duan, and J. Xue, Nature Communications 15, 295 (2024)

    work page 2024

  7. [7]

    Sharma, F

    P. Sharma, F. -X. Xiang, D. -F. Shao, D. Zhang, E. Y. Tsymbal, A. R. Hamilton, and J. Seidel, Science Advances 5, eaax5080 (2019)

    work page 2019

  8. [8]

    Z. Fei, W. Zhao, T. A. Palomaki, B. Sun, M. K. Miller, Z. Zhao, J. Yan, X. X u, and D. H. Cobden, Nature 560, 336 (2018)

    work page 2018

Show all 55 references
  1. [9]

    Xiao et al., Nature Physics 16, 1028 (2020)

    J. Xiao et al., Nature Physics 16, 1028 (2020)

    2020

  2. [10]

    X. Wang, K. Yasuda, Y. Zhang, S. Liu, K. Watanabe, T. Taniguchi, J. Hone, L. Fu, and P. Jarillo - Herrero, Nature nanotechnology 17, 367 (2022)

    2022

  3. [11]

    Rogée, L

    L. Rogée, L. Wang, Y. Zhang, S. Cai, P. Wang, M. Chhowalla, W. Ji, and S. P. Lau, Science 376, 973 (2022)

    2022

  4. [12]

    Weston et al., Nature nanotechnology 17, 390 (2022)

    A. Weston et al., Nature nanotechnology 17, 390 (2022)

    2022

  5. [13]

    Y. Liu, S. Liu, B. Li, W. J. Yoo, and J. Hone, Nano Lett. 22, 1265 (2022)

    2022

  6. [14]

    Wan et al., Phys

    Y. Wan et al., Phys. Rev. Lett. 128, 067601 (2022)

    2022

  7. [15]

    Deb et al., Nature 612, 465 (2022)

    S. Deb et al., Nature 612, 465 (2022)

    2022

  8. [16]

    Meng et al., Nature Communications 13, 7696 (2022)

    P. Meng et al., Nature Communications 13, 7696 (2022)

    2022

  9. [17]

    Jindal et al., Nature 613, 48 (2023)

    A. Jindal et al., Nature 613, 48 (2023)

    2023

  10. [18]

    K. H. Y. Ko et al., Nature Materials 22, 992 (2023)

    2023

  11. [19]

    Yang et al., Nature Photonics 16, 469 (2022)

    D. Yang et al., Nature Photonics 16, 469 (2022)

    2022

  12. [20]

    F. Sui, M. Jin, Y. Zhang, R. Qi, Y.-N. Wu, R. Huang, F. Yue, and J. Chu, Nature Communications 14, 36 (2023)

    2023

  13. [21]

    Sui et al., Nature Communications 15, 3799 (2024)

    F. Sui et al., Nature Communications 15, 3799 (2024)

    2024

  14. [22]

    L. P. Miao, N. Ding, N. Wang, C. Shi, H. Y. Ye, L. L. Li, Y. F. Yao, S. Dong, and Y. Zhang, Nature Materials 21, 1158 (2022)

    2022

  15. [23]

    Yasuda et al., Science 385, 53 (2024)

    K. Yasuda et al., Science 385, 53 (2024)

    2024

  16. [24]

    Bian et al., Science 385, 57 (2024)

    R. Bian et al., Science 385, 57 (2024)

    2024

  17. [25]

    O. Hod, E. Meyer, Q. Zheng, and M. Urbakh, Nature 563, 485 (2018)

    2018

  18. [26]

    Hirano and K

    M. Hirano and K. Shinjo, Phys Rev B Condens Matter 41, 11837 (1990)

    1990

  19. [27]

    Dienwiebel, G

    M. Dienwiebel, G. S. Verhoeven, N. Pradeep, J. W. Frenken, J. A. Heimberg, and H. W. Zandbergen, Phys Rev Lett 92, 126101 (2004)

    2004

  20. [28]

    A. E. Fi lippov, M. Dienwiebel, J. W. Frenken, J. Klafter, and M. Urbakh, Phys Rev Lett 100, 046102 (2008)

    2008

  21. [29]

    Pierno, L

    M. Pierno, L. Bruschi, G. Mistura, G. Paolicelli, A. di Bona, S. Valeri, R. Guerra, A. Vanossi, and E. Tosatti, Nat Nanotechnol 10, 714 (2015)

    2015

  22. [30]

    Zheng et al., Phys Rev Lett 100, 067205 (2008)

    Q. Zheng et al., Phys Rev Lett 100, 067205 (2008)

    2008

  23. [31]

    Z. Liu, J. Yang, F. Grey, J. Z. Liu, Y. Liu, Y. Wang, Y. Yang, Y. Cheng, and Q. Zheng, Physical Review Letters 108, 205503 (2012)

    2012

  24. [32]

    Y. Long, X. Wang, W. Tan, B. Li, J. Li, W. Deng, X. Li, W. Guo, and J. Yin, Nano Lett 24, 7572 (2024)

    2024

  25. [33]

    Y. Song, D. Mandelli, O. Hod, M. Urbakh, M. Ma, and Q. Zheng, Nat Mater 17, 894 (2018)

    2018

  26. [34]

    Wang et al., Nanoscale 9, 10846 (2017)

    L. Wang et al., Nanoscale 9, 10846 (2017)

    2017

  27. [35]

    Li et al., Tribology Letters 71, 13 (2022)

    H. Li et al., Tribology Letters 71, 13 (2022)

    2022

  28. [36]

    Wang, T.-B

    L.-F. Wang, T.-B. Ma, Y.-Z. Hu, Q. Zheng, H. Wang, and J. Luo, Nanotechnology 25, 385701 (2014)

    2014

  29. [37]

    P. Li, W. Y. Wang, X. Sui, X. Fan, and J. Li, Journal of Materials Science & Technology 196, 125 (2024)

    2024

  30. [38]

    Liao et al., Nat Mater 21, 47 (2022)

    M. Liao et al., Nat Mater 21, 47 (2022)

    2022

  31. [39]

    Mandelli, I

    D. Mandelli, I. Leven, O. Hod, and M. Urbakh, Scientific Reports 7, 10851 (2017)

    2017

  32. [40]

    Cheng, H

    Z. Cheng, H. Feng, J. Sun, Z. Lu, and Q. C. He, Advanced Materials Interfaces 10 (2023)

    2023

  33. [41]

    H. Li, J. Wang, S. Gao, Q. Chen, L. Peng, K. Liu, and X. Wei, Adv Mater 29 (2017)

    2017

  34. [42]

    L. Yang, S. Ding, J. Gao, and M. Wu, Physical Review Letters 131, 096801 (2023)

    2023

  35. [43]

    Yang and M

    L. Yang and M. Wu, Advanced Functional Materials 33, 2301105 (2023)

    2023

  36. [44]

    Kresse and J

    G. Kresse and J. Furthmüller, Computational Materials Science 6, 15 (1996)

    1996

  37. [45]

    Kresse and J

    G. Kresse and J. Furthmuller, Phys Rev B Condens Matter 54, 11169 (1996)

    1996

  38. [46]

    Kresse and D

    G. Kresse and D. Joubert, Physical Review B 59, 1758 (1999)

    1999

  39. [47]

    J. P. Perdew, K. Burke, and M. Ernzerhof, Phys Rev Lett 77, 3865 (1996)

    1996

  40. [48]

    J. P. Perdew, J. A. Chevary, S. H. Vosko, K. A. Jackson, M. R. Pederson, D. J. Singh, and C. Fiolhais, Phys Rev B Condens Matter 46, 6671 (1992)

    1992

  41. [49]

    Grimme, J Comput Chem 27, 1787 (2006)

    S. Grimme, J Comput Chem 27, 1787 (2006)

    2006

  42. [50]

    H. J. Monkhorst and J. D. Pack, Physical Review B 13, 5188 (1976)

    1976

  43. [51]

    Neugebauer and M

    J. Neugebauer and M. Scheffler, Phys Rev B Condens Matter 46, 16067 (1992)

    1992

  44. [52]

    Henkelman, B

    G. Henkelman, B. P. Uberuaga, and H. Jónsson, The Journal of Chemical Physics 113, 9901 (2000)

    2000

  45. [53]

    J. Sun, Y. Zhang, Z. Lu, Q. Li, Q. Xue, S. Du, J. Pu, and L. Wang, J Phys Chem Lett 9, 2554 (2018)

    2018

  46. [54]

    Zhong, Y

    T. Zhong, Y. Ren, Z. Zhang, J. Gao, and M. Wu, Journal of Materials Chemistry A 9, 19659 (2021)

    2021

  47. [55]

    Chen et al., Nature Nanotechnology 19, 962 (2024)

    M. Chen et al., Nature Nanotechnology 19, 962 (2024)

    2024

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.