Pith. sign in

REVIEW 5 major objections 5 minor 29 references

Thermodynamic model of twisted bilayer graphene: Configuration entropy matters

T0 review · 5 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read A thermodynamic derivation traces structural superlubricity to configurational entropy, giving interlayer shear stress $\tau = 3kT\,\cot(\theta/2)/(2\pi R^3)$ and a self-rotation branch beyond 30 degrees.

desk verdict The new friction law rests entirely on an unproven entropy-area postulate, so the formula is not yet established, but the question is worth debating. read the letter →

arxiv 1908.04666 v2 pith:NVAQXJ67 submitted 2019-08-13 cond-mat.mes-hall physics.app-ph

Weidong Yan , Langquan Shui , Wengen Ouyang , Ze Liu This is my paper · ORCID
classification cond-mat.mes-hallphysics.app-ph MSC 82C0580A10 PACS 81.40.Pq
keywords structuralsuperlubricitytwistedbilayergrapheneconfigurationalentropymoiréunitcellinterlayershearstressthermodynamicmodelnegativefrictionsizescaling
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

This paper tries to establish a thermodynamic origin for structural superlubricity, the near-zero friction between atomically flat layers whose lattices are twisted out of commensurate registry. Its claim is that the effect comes from configurational entropy: a twisted bilayer has more accessible atomic arrangements inside its moiré unit cell, and that entropy difference dominates the free energy during quasi-static rotation. Working from this premise with rigid layers and constant internal energy, the paper derives a closed-form torque $M = kT\,\cot(\theta/2)$ and an interlayer shear stress $\tau = 3kT\,\cot(\theta/2)/(2\pi R^3)$. It argues that these formulas reproduce measured friction-versus-angle and size-dependent data, and that they imply superlubricity persists to large scales and low temperatures.

What carries the argument

The load-bearing machinery is the configuration-entropy ansatz of Eq. (2): the number of atomic microstates in a moiré unit cell is taken proportional to the cell area, $\Omega_\theta/\Omega_0 = S_\theta/S_0$. Combined with the moiré side-length law $a_m = \sqrt{3}\,a/(2\sin(\theta/2))$ and the quasi-static thermodynamic relation $dQ = T\,dS$, it yields the entropy difference $\Delta S = 2k\ln(a_m/a)$ and then the torque. The argument also relies on the rigid-layer approximation, justified by the large ratio of in-plane stiffness to interlayer shear modulus, and on the observation that the interlayer potential energy is essentially unchanged during rotation, which sets $dU=0$ and makes the torque purely entropic.

What would settle it

Measure, for a rigid, defect-free circular graphene flake of radius R twisted against graphite at fixed temperature, the torque as a function of twist angle from 0 to 60 degrees. The paper's claim predicts $M = kT\,\cot(\theta/2)$: divergence near zero angle, linear temperature dependence, no dependence on R beyond the entropy area factor, and a sign reversal at 30 degrees. Observing no sign reversal, or a torque that is not linear in temperature, would falsify Eq. (6); a direct enumeration of finite-flake microstates that fails to reproduce $k\ln(S_\theta/S_0)$ would falsify Eq. (2).

Watch

Extended reading notes

Core claim

The central claim is that structural superlubricity is an entropic phenomenon. Relative to AA-stacked bilayer graphene, a twisted bilayer is assigned a configuration entropy $\Delta S = 2k\ln(a_m/a)$, where $a_m = \sqrt{3}\,a/(2\sin(\theta/2))$ is the moiré side length and $a$ is the carbon bond length; the entropy is taken to dominate the Helmholtz free energy. The rotation torque follows as $M = kT\,\cot(\theta/2)$, and dividing by the flake area gives $\tau = 3kT\,\cot(\theta/2)/(2\pi R^3)$. Thus the shear stress is nonzero but vanishingly small for large rigid flakes, grows linearly with temperature, and becomes negative for twist angles between 30 and 60 degrees, meaning the flake should spontaneously rotate toward a commensurate registry. The paper backs the derivation with molecular-dynamics evidence that interlayer potential energy is nearly independent of twist angle even under normal loads up to 10 GPa, and with comparison to existing graphite-flake friction, nanotube pullout, and torque-induced reorientation experiments.

Load-bearing premise

Equation (2) assumes that the number of possible atomic configurations inside a moiré unit cell is proportional to the area of that cell; if the microstate count does not scale with moiré cell area, the entropy difference and the derived cotangent law do not follow.

Editorial extensions

If this is right

  • Shear stress in a twisted bilayer contact should vanish linearly as temperature approaches zero and grow linearly with temperature, so superlubricity should persist at low temperatures.
  • For fixed angle and temperature the stress scales as $R^{-3}$; ideal large-area rigid contacts should have immeasurably small shear stress, consistent with measured nanotube pullout strengths.
  • Between 30 and 60 degrees the torque reverses sign: a free flake should spontaneously rotate toward AB stacking, an entropy-driven self-rotation the paper identifies with observations of torque-induced reorientation.
  • The formula should remain valid under uniformly distributed normal loads up to 10 GPa, as long as potential energy stays nearly independent of twist angle.
  • Because the stress expression contains no material-specific interaction strength, the same law should describe other rigid layered contacts with the same geometry.

Reading between the lines

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

  • An extension of the model is to measure the full torque-angle curve for other layered materials such as MoS2 or hexagonal boron nitride; if the $kT\cot(\theta/2)$ law holds unchanged, the entropic origin is generic rather than specific to graphene.
  • The predicted sign change near 30 degrees implies a thermodynamic tendency for twisted contacts to anneal toward commensurate registry; this could mean that preferred twist angles observed in devices partly reflect entropy maximization rather than preparation history alone.
  • A computational test of the weakest premise is direct enumeration of distinct rigid registries for finite flakes: if the microstate count does not track moiré cell area, the law would acquire finite-size corrections beyond the simple $R^{-3}$ dependence.
  • At very low temperatures quantum zero-point motion should eventually replace the classical accessible-volume estimate of entropy, so the strictly linear $T$ dependence may saturate; measuring that saturation would delimit the classical regime.
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

5 major / 5 minor

Summary. The manuscript proposes a thermodynamic origin for structural superlubricity in twisted bilayer graphene. The authors define a configurational entropy change on the basis of Eq. (2), where the ratio of microstates is set equal to the ratio of moiré unit-cell areas, leading to ΔS = 2k ln(a_m/a). Combining this with the first law under the assumption dU = 0, they derive a torque M = kT cot(θ/2) and a shear stress τ = 3kT cot(θ/2)/(2πR^3) in Eqs. (5) and (6). The paper claims this formula explains experimental observations of the twist-angle dependence of friction, size-dependent shear stress, temperature-dependent sliding, spontaneous reorientation, and persistence under high normal loads, and it reports MD simulations indicating that the internal energy is nearly independent of twist angle.

Significance. If the central relation Eq. (2) were derivable from statistical mechanics, the model would offer a remarkably concise analytic formula for interlayer shear stress in twisted bilayer contacts, and the predicted τ ~ 1/R^3 scaling and temperature dependence would be useful design rules for superlubric systems. The paper is clearly written and identifies a potentially important, nonstandard contribution (configurational entropy) to the energetics of twisted interfaces. The MD data in Fig. 1(d) and Fig. 4, showing near-constancy of the potential energy with twist angle for rigid layers, are useful supporting evidence for the dU ≈ 0 assumption. However, the load-bearing step, Eq. (2), is asserted without a microscopic counting argument, and the treatment of entropy as a per-unit-cell quantity without accounting for the number of moiré cells undermines the physical meaning of the derived torque. As a result, the central quantitative prediction is not established, and the experimental comparisons, though suggestive, do not independently validate the entropy mechanism.

major comments (5)
  1. [Eq. (2)] Equation (2) is the central assumption of the paper, yet it is asserted without derivation: the ratio of microstate numbers is set equal to the ratio of moiré unit-cell areas, Ωθ/Ω0 = Sθ/S0. For rigid crystalline layers at a fixed twist angle, the atomic registry is deterministic, and it is not obvious why the number of microstates should scale with the moiré cell area. No partition function, phase-space argument, or counting procedure is provided. Because Eq. (2) is the sole source of the cotangent law in Eq. (5) and of the entropy term in Eq. (6), the central quantitative claim is unsupported unless this identification can be justified from a statistical-mechanical model.
  2. [Eq. (2) to Eq. (5): extensivity] The entropy ΔS in Eq. (2) is defined for a single moiré unit-cell, but Eq. (5) applies this entropy to the entire graphene flake without multiplying by the number of moiré cells. For a flake of radius R, the number of cells scales as N_cells ~ (R/a_m)^2, and the total entropy should be roughly N_cells times the per-cell entropy. If this extensivity is restored, the torque M would acquire an R-dependence and a different θ-dependence through the θ-dependence of N_cells; the size-independent torque in Eq. (5) is an artifact of treating the per-cell entropy as the system entropy. The statement in the text that the entropy is determined by one unit-cell 'as long as R > am' is not justified and is a critical flaw.
  3. [Eq. (4) and dU = 0] The derivation assumes dU = 0 during quasi-static rotation, supported only by MD simulations with the AIREBO potential for rigid layers. The authors themselves acknowledge in the paragraph following Eq. (5) that the model breaks down when energy contributions become comparable to entropy contributions, which is precisely the regime of standard registry-dependent interlayer interactions. Since real interfaces have finite stiffness and may undergo relaxation, the rigidity assumption is restrictive, and the extension of the dU ≈ 0 conclusion to normal loads up to 10 GPa in Fig. 4 is based on the same rigid-layer approximation. The validity of Eq. (6) beyond the idealized rigid case is therefore not established.
  4. [Negative friction prediction and experimental comparison] The prediction of a sign reversal of τ for 30° < θ < 60°, i.e., spontaneous rotation toward commensurate registry, is striking and falsifiable, but it rests entirely on the derivative of Eq. (3). The cited experimental support (Refs. [9] and [21]) is qualitative; no quantitative comparison with the predicted magnitude or angle dependence is attempted. In Fig. 2, the agreement with the experimental data of Ref. [8] is only approximate, and the statement that the theoretical curve is 'a little larger' is not a quantitative validation, especially because the plotted range and the experimental conditions (static vs. sliding friction) are not matched in detail.
  5. [Comparisons with experiments] The paper claims that Eq. (6) 'universally explains' a wide range of experiments, but most comparisons are qualitative. For the DWCNT pull-out (Ref. [13]), the twist angle is unknown, so the apparent agreement at R = 3663 nm is a consistency check with an adjustable geometric parameter, not an independent test of the entropy model. Given that Eq. (6) has no free parameters, a more stringent test would require experiments with known twist angles, sizes, and temperatures; the present set of comparisons does not achieve this.
minor comments (5)
  1. [Abstract and introduction] The abstract says the configuration entropy is 'directly derived from the Helmholtz free energy,' but the full text does not derive Eq. (2) from a free energy; it postulates the microstate-area proportionality. Please adjust the wording to avoid overstatement.
  2. [Fig. 1(c) and 1(d)] The y-axis of Fig. 1(c) is missing units, and Fig. 1(d) uses 'internal energy' where the text describes potential energy; please clarify the notation.
  3. [Terminology] The phrase 'micro-status' is nonstandard; the paper should use 'microstates' consistently.
  4. [Reference [13]] The text mentions a '9-mm-long inner shell,' but the original reference is about a micron-scale tube; this appears to be a typo and should be corrected.
  5. [MD simulation details] The Methods section is absent; the MD simulations lack details on system size, boundary conditions, equilibration, and whether the bottom layer is truly infinite. These details are necessary to assess the dU ≈ 0 conclusion.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the cotangent law follows from an explicit entropy postulate, but the postulate is not fitted to, or defined by, the predicted friction data.

full rationale

The derivation chain is transparent: Eq. (1) is standard moiré geometry; Eq. (2) introduces a physical postulate that the microstate ratio equals the moiré-cell area ratio; Eqs. (3)–(6) are algebraic consequences. This makes the cotangent law dependent on Eq. (2), but dependence on a postulate is not circularity. No parameter is fitted to the experimental data cited; the comparisons in Figs. 2 and 3 are post hoc and qualitative, and the paper explicitly notes that the DWCNT comparison is limited by the unknown twist angle. The extensivity issue (using one moiré cell instead of a number of cells) is a correctness or validity concern, not a circular reduction. There are no load-bearing self-citations: the rigidity and moiré-geometry citations are external standard results, and the authors' own prior experimental papers are not used to justify the entropy postulate. Therefore no circular step meets the evidentiary bar for flagging.

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

The model rests on a small number of explicit assumptions. The most important is the identification of microstate number with moire cell area, which is not derived from statistical mechanics and effectively pre-encodes the friction law. The other assumptions (rigid layers, constant internal energy, uniform stress) are standard approximations in the field. No free parameters are fitted to data, and no new entities are introduced.

assumptions (5)
  • ad hoc to paper The number of microstates in a moire unit-cell is proportional to the area of the unit-cell: Omega_theta/Omega_0 = S_theta/S_0.
    This is the foundational assumption of the entropy model, stated without derivation in Eq. (2). It is not a standard statistical mechanics result and is not justified from any microscopic model.
  • domain assumption The two graphene layers are rigid during rotation.
    Invoked after Ref [24] in the model setup; used to ignore elastic deformations. This is an approximation, but not the most fragile.
  • domain assumption The internal energy is unchanged during quasi-static rotation, dU = 0.
    Assumed in the first-law balance and supported by MD simulations at equilibrium interlayer distance; the paper argues this holds up to 10 GPa normal load, but the MD evidence is only qualitative.
  • domain assumption The interlayer shear stress is uniformly distributed over a circular flake.
    Used to convert torque M to shear stress tau in Eq. (6); real contacts have inhomogeneous stress.
  • standard math Classical thermodynamics with dQ = T dS applies to a rigid twisted bilayer.
    The framework of classical statistical mechanics is used, but applying thermodynamic entropy to a single finite rigid flake with a deterministic twist angle is nonstandard.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Thermodynamic model of twisted bilayer graphene: Configuration entropy matters." pith.science (2026). https://pith.science/paper/NVAQXJ67

@misc{pith2026190804666,
  author       = {Pith},
  title        = {Pith review of: Thermodynamic model of twisted bilayer graphene: Configuration entropy matters},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/NVAQXJ67}},
  note         = {Machine review of arXiv:1908.04666}
}
read the original abstract

Twisted bilayer materials have attracted tremendous attention due to their unique and novel properties. Here, we derive a thermodynamic model for twisted bilayer graphene (tBLG) within the framework of the classical statistical mechanics, based on which, the configuration entropy reflecting the number of micro-status in moire unit-cells, is directly derived from the Helmholtz free energy with a clear physical interpretation. More importantly, we show the configuration entropy of tBLG relative to the AB-stacked bilayer graphene is proportional to the logarithmic function of the ratio of moire period and the atomic lattice constant, which we found dominates the Helmholtz free energy of tBLG and can well explain experimental observations in superlubric contacts. Our work provides a theoretical foundation for studying moire effect of incommensurate contact interfaces and could facilitate twisting based applications such as superlubricity.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

29 extracted references · 26 canonical work pages

  1. [9]

    X. Feng, S. Kwon, J. Y. Park, and M. Salmeron, ACS Nano 7, 1718 (2013)

    work page 2013

  2. [21]

    A. E. Filippov, M. Dienwiebel, J. W. M. Frenken, J. Klafter, and M. Urbakh, Physical Review Letters 100, 046102 (2008)

    work page 2008

  3. [8]

    Dienwiebel, G

    M. Dienwiebel, G. S. Verhoeven, N. Pradeep, J. W. M. Frenken, J. A. Heimberg, and H. W. Zandbergen, Physical Review Letters 92, 126101 (2004)

    work page 2004

  4. [13]

    Zhang, Z

    R. Zhang, Z. Ning, Y. Zhang, Q. Zheng, Q. Chen, H. Xie, Q. Zhang, W. Qian, and F. Wei, Nature Nanotechnology 8, 912 (2013)

    work page 2013

  5. [1]

    Y. Cao, V. Fatemi, S. Fang, K. Watanabe, T. Taniguchi, E. Kaxiras, and P . Jarillo- Herrero, Nature 556, 43 (2018)

    work page 2018

  6. [2]

    Xu and L

    C. Xu and L. Balents, Physical Review Letters 121, 087001 (2018)

    work page 2018

  7. [3]

    Yankowitz, S

    M. Yankowitz, S. Chen, H. Polshyn, Y. Zhang, K. Watanabe, T. Taniguchi, D. Graf, A. F. Young, and C. R. Dean, Science 363, 1059 (2019). 12

    work page 2019

  8. [4]

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

    2018

Show all 29 references
  1. [5]

    Hirano and K

    M. Hirano and K. Shinjo, Physical Review B 41, 11837 (1990)

    1990

  2. [6]

    J. M. Martin, C. Donnet, T. Le Mogne, and T. Epicier, Physical Review B 48, 10583 (1993)

    1993

  3. [7]

    Hirano, K

    M. Hirano, K. Shinjo, R. Kaneko, and Y. Murata, Physical Review Letters 78, 1448 (1997)

    1997

  4. [10]

    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

  5. [11]

    J. Yang, Z. Liu, F. Grey, Z. Xu, X. Li, Y. Liu, M. Urbakh, Y. Cheng, and Q. Zheng, Physical Review Letters 110, 255504 (2013)

    2013

  6. [12]

    Berman, S

    D. Berman, S. A. Deshmukh, S. K. R. S. Sankaranarayana n, A. Erdemir, and A. V. Sumant, Science 348, 1118 (2015)

    2015

  7. [14]

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

    2018

  8. [15]

    Liu et al., Nature Communications 8, 14029 (2017)

    S.-W. Liu et al., Nature Communications 8, 14029 (2017)

    2017

  9. [16]

    Cihan, S

    E. Cihan, S. İpek, E. Durgun, and M. Z. Baykara, Nature Communications 7, 12055 (2016). 13

    2016

  10. [17]

    Dietzel, M

    D. Dietzel, M. Feldmann, U. D. Schwarz, H. Fuchs, and A. Schirmeisen, Physical Review Letters 111, 235502 (2013)

    2013

  11. [18]

    Yaniv and E

    R. Yaniv and E. Koren, Advanced Functional Materials 0, 1901138 (2019)

    2019

  12. [19]

    M. Z. Baykara, M. R. Vazirisereshk, and A. Martini, Appl ied Physics Reviews 5, 041102 (2018)

    2018

  13. [20]

    W. Wang, J. Shen, and Q. C. He, Physical Review B 99, 054103 (2019)

    2019

  14. [22]

    G. S. Verhoeven, M. Dienwiebel, and J. W. M. Frenken, Physical Review B 70, 165418 (2004)

    2004

  15. [23]

    Bosak, M

    A. Bosak, M. Krisch, M. Mohr, J. Maultzsch, and C. Thomsen, Physical Review B 75, 153408 (2007)

    2007

  16. [24]

    Hod, Physical Review B 86, 075444 (2012)

    O. Hod, Physical Review B 86, 075444 (2012)

    2012

  17. [25]

    Z. Y. Rong and P . Kuiper, Physical Review B 48, 17427 (1993)

    1993

  18. [26]

    Bistritzer and A

    R. Bistritzer and A. H. MacDonald, Proceedings of the National Academy of Sciences 108, 12233 (2011)

    2011

  19. [27]

    Wu et al., ACS Nano 9, 7440 (2015)

    J.-B. Wu et al., ACS Nano 9, 7440 (2015)

    2015

  20. [28]

    Y. Wang, Z. Qin, M. J. Buehler, and Z. Xu, Nature Communi cations 7, 12854 (2016)

    2016

  21. [29]

    C. C. Vu, S. Zhang, M. Urbakh, Q. Li, and Q. Zheng, Phys.rev.b 94, 081405 (2016)

    2016

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.