Pith. sign in

REVIEW 3 major objections 4 minor 52 references

Universal giant spin Hall effect in moire metal

T0 review · 3 major / 4 minor · reviewed 2026-08-16 · deepseek-v4-flash

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

Pith's one-line read Twisted NbSe2 sets a spin Hall record: -5200 (ℏ/e) S/cm at a 5.09° twist, a value the authors say surpasses every known bulk material.

desk verdict Interesting computational work and credible t-MoTe2 predictions, but the headline NbX2 record SHC depends on an unstated thickness conversion that likely halves the quoted value. read the letter →

arxiv 2504.16179 v1 pith:EXXMA33G submitted 2025-04-22 cond-mat.mes-hall

Ning Mao , Cheng Xu , Ting Bao , Nikolai Peshcherenko , Claudia Felser , Yang Zhang This is my paper · ORCID
classification cond-mat.mes-hall PACS 72.25.-b73.22.-f
keywords spinHalleffectmoiremetalstwistedbilayerNbSe2FermisurfacereconstructionBerrycurvaturetopologicalflatbandsabinitiotransportMoTe2
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 claims that moiré patterning can produce a giant spin Hall effect not only in lightly doped semiconductors, where isolated topological flat bands give quantized spin Hall conductivity, but also in heavily doped metallic regimes, where the long-wavelength moiré potential reconstructs large Fermi surfaces and creates dense networks of band inversions. In twisted MoTe2 the authors find the peak spin Hall conductivity triples from 6 to 17 e/4π when going from 5.09° to 3.89°, without any flat bands. In the intrinsic moiré metals twisted NbS2 and NbSe2, they report a record spin Hall conductivity of -17 e/4π at the Fermi level, quoted as -5200 (ℏ/e) S/cm in three-dimensional units, about 2.6 times the benchmark value for platinum. If right, this makes a commercially accessible twist angle in a common transition-metal dichalcogenide a better spin-current generator than any known bulk metal.

What carries the argument

Spin Berry curvature $\Omega^S(\mathbf{k})$ integrated via the Kubo formula (Eqs. 13–14): the moiré potential folds the large Fermi surface of the monolayer into a mini Brillouin zone, gaps out crossings, and creates dense networks of band inversions whose interband matrix elements of spin current and momentum produce large spin Berry curvature. The argument is carried by the identification that these networks, not isolated flat bands, generate the giant response, so the enhancement is universal across semiconductors and metals.

What would settle it

Measure the spin Hall conductivity of twisted bilayer NbSe2 at 5.09° via spin-torque ferromagnetic resonance or nonlocal spin Hall detection, or recompute the 3D conversion using the explicit interlayer spacing (roughly 6 Å per layer) instead of an unspecified slab height, and check whether the peak at the Fermi level survives at broader or smaller broadening.

Watch

Extended reading notes

Core claim

The authors' central claim is that the spin Hall effect is universally amplified in moiré systems by Fermi-surface reconstruction rather than by topological flat-band physics. In twisted bilayer MoTe2, lightly doped regimes show quantized spin Hall conductivity steps of 4, 8, and 10 e/4π as the twist angle decreases, coming from an increasing count of isolated Chern bands. In heavily doped metallic regimes, where these bands are gone, they report a non-quantized peak of 17 e/4π at 3.89°. Extending to the intrinsic moiré metals NbX2, whose large Fermi surfaces cover about 48% of the Brillouin zone, they find the same mechanism produces -17 e/4π at the natural Fermi level at a 5.09° twist, which they convert to -5200 (ℏ/e) S/cm and state surpasses all known bulk spin Hall materials.

Load-bearing premise

The 2D spin Hall conductivity of the bilayer is converted to the bulk-like 3D value -5200 (ℏ/e) S/cm using a thickness that is not specified, and the comparison to bulk platinum assumes that conversion is the right metric.

Editorial extensions

If this is right

  • A device built from twisted NbSe2 at 5.09° should generate a spin current at its natural Fermi level without electrostatic doping, since the spin Hall conductivity peak lies there.
  • Metallic moiré systems are a more robust platform than topological flat bands for spin generation, because the giant spin Hall effect does not depend on quantization or narrow energy windows.
  • The mechanism is universal: any twisted metal whose large Fermi surface is reconstructed by a long-wavelength moiré potential should show amplified spin Hall effect.
  • In twisted MoTe2, heavy doping can outperform the multi-QSH state: the metallic peak reaches 17 e/4π at 3.89°, compared with the experimental 6 e/4π at a 2.1° twist.

Reading between the lines

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

  • The 3D-unit conversion (2D spin Hall conductivity divided by a thickness that the paper never specifies) is the fragile link; if the physical bilayer thickness is used rather than the implicit bulk-like slab height, the record claim may not hold.
  • A direct spin-torque or nonlocal spin Hall measurement on twisted NbSe2 at the reported twist angle would settle whether the peak is real and whether it sits exactly at the Fermi level.
  • The fixed 10 meV broadening used to read off the metallic peaks is untested; a sensitivity scan across broadenings would show whether the -17 e/4π value is robust or partly an artifact of the smearing.
  • The same mechanism should be sought in other 4d and 5d transition-metal dichalcogenides with large Fermi surfaces, where the density of moiré-induced band-inversion points could be used as a screening criterion.
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

3 major / 4 minor

Summary. The paper reports GPU-accelerated ab initio Kubo-formula calculations of the spin Hall conductivity (SHC) in twisted bilayer MoTe2 and twisted NbX2 (X = S, Se). For lightly doped t-MoTe2, the authors find quantized SHC plateaus of 4–10 e/4π arising from isolated Chern bands as the twist angle decreases from 5.09° to 1.89°. In heavily doped metallic regimes, they report an SHC peak of 17 e/4π at 3.89°, which they attribute to Fermi surface reconstruction under the long-wavelength moiré potential. For t-NbS2 and t-NbSe2 at 5.09°, they report a Fermi-level SHC of -17 e/4π, quoted as -5200 (ℏ/e)S/cm in 3D units, and claim this surpasses all known bulk spin Hall materials including Pt. The computations use transfer-learning structural relaxation, OpenMX pseudoatomic-orbital Hamiltonians, and a two-stage dense matrix diagonalization on GPUs, with a 100×100 k-mesh and a stated 200×200 convergence check.

Significance. If the central claims hold, the paper would establish a new mechanism for enhancing spin Hall effects in metallic moiré systems, with a concrete prediction at a commercially accessible twist angle and at the natural Fermi level. The methodological strengths are substantial: a dense k-mesh with a reported convergence check, symmetry reduction to 884 irreducible points, treatment of matrices of dimension about 45,000, and quantitative comparison with published experimental SHC values for lightly doped t-MoTe2. The prediction for t-NbX2 is falsifiable in principle and would be of immediate interest to the spintronics community. However, the headline record claim is currently not reproducible as stated because the conversion from the computed 2D SHC to the quoted 3D value is not specified, and the metallic peak values rest on a single broadening parameter. These issues are load-bearing for the central quantitative claim, so revision is required.

major comments (3)
  1. [Abstract and 'Metallic regimes' section; Eq. (13); Fig. 3(b)] The record claim of -5200 (ℏ/e)S/cm in 3D units from a 2D SHC of -17 e/4π is not reproducible because the conversion from 2D to 3D is never specified. Eq. (13) writes d^3k, but the system is a twisted bilayer, and no effective thickness or conversion formula is given in the main text or the Supplemental Material. The quoted value implicitly corresponds to dividing by a distance of order one monolayer thickness; if the full physical bilayer thickness is used instead, the converted value is reduced by roughly a factor of two, placing it close to the Pt benchmark rather than 2.6 times above it. Since the abstract and the discussion of Fig. 3 explicitly compare with bulk Pt, the authors must state the conversion formula, the assumed thickness, and justify that choice; the record statement should be revised accordingly.
  2. [Fig. 2 caption and Fig. 3(b)] All metallic SHC peaks, including the Fermi-level value of -17 e/4π in t-NbSe2, are computed with a single broadening parameter of 10 meV in the Kubo formula. No sensitivity study is reported, although metallic Berry-curvature transport coefficients can depend strongly on broadening when the Fermi level cuts sharp band-inversion features or narrow minibands. The authors should show the dependence of the peak SHC on the broadening parameter (for example, 2, 5, 10, 20, and 50 meV) for both t-MoTe2 in the metallic regime and for t-NbX2, and confirm that the record comparison survives this variation.
  3. [Title, abstract, and 'Metallic regimes' section] The claim of a 'universal' amplification of the spin Hall effect is broader than what is demonstrated: the metallic enhancement is established for two related material families, and for t-NbX2 only a single twist angle (5.09°) is studied. To justify the title and abstract, the authors should either provide additional twist-angle or material examples in the metallic regime, or explicitly temper the word 'universal' to reflect the scope of the calculations.
minor comments (4)
  1. [Acknowledgments] The sentence 'We thanks Mark Gates...' should be corrected to 'We thank Mark Gates...'.
  2. [Main text, discussion of Fig. 3] The phrase 'almost three time of current record SHE in platinum' should read 'almost three times the current record', and the sign convention for the SHC comparison with Pt should be stated explicitly.
  3. [Eq. (13)] The notation d^3k in Eq. (13) is inconsistent with a two-dimensional bilayer calculation; the authors should use d^2k and clarify the normalization factor (2π)^2, or explicitly define the 2D-to-3D conversion there.
  4. [Fig. 4] The spin Berry curvature maps in panels (b) and (d) lack color bars and axis tick labels, which makes it difficult to assess the magnitude and location of the curvature; adding color scales and labeled axes would improve the figure.

Circularity Check

0 steps flagged · score 1.0 of 10

No circular derivation; the 2D SHC values are independently computed, though the 3D 'record' conversion is under-specified.

full rationale

The derivation chain is: relaxed moiré structures are fed into OpenMX, the real-space Hamiltonian and overlap matrices are Fourier transformed, and the SHC is evaluated from the Kubo formula in Eqs. (12)-(14). The reported values (4, 6, 10, 17 e/4π) are outputs of this calculation, not inputs; no parameter is fitted to the claimed SHC, and the target result is not assumed in constructing the Hamiltonian. The self-cited transfer-learning relaxation method and prior t-MoTe2 SHC work are independent computational inputs rather than the result being tested, and the platinum benchmark is supported by multiple external references. The only significant gap is the unstated conversion from the 2D value -17 e/4π to the 3D value -5200 (ℏ/e)S/cm: the paper does not give the effective thickness used to convert between 2D and 3D units, and Eq. (13)'s d^3k notation obscures the 2D nature of the moiré Brillouin zone. This makes the 'surpassing all known bulk materials' comparison under-specified and unfalsifiable as written, but it is a missing-support or units-convention issue rather than a circular reduction: the 2D SHC is computed from first principles, and the conversion does not feed back into the Kubo calculation. No equation is satisfied by construction, no fitted parameter is renamed as a prediction, and no load-bearing argument reduces to a self-citation. Hence no significant circularity.

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

The central claim rests on a standard DFT plus Kubo linear-response pipeline with PBE functionals, a machine-learned relaxation potential from prior work, and a fixed broadening. No new physical entities are introduced. The most consequential free parameter is the implicit layer thickness used to convert the 2D spin Hall conductivity into a 3D record value.

free parameters (3)
  • Kubo broadening eta = 10 meV
    Chosen for the spin Hall conductivity in the metallic regime; no sensitivity study is shown, and the peak values that set the headline numbers may depend on it (main text, Fig. 2 caption).
  • Unfolding smearing eta = 0.02
    Gaussian broadening used in the spectral function for Fermi surface unfolding (main text, after Eq. 16); affects the visual Fermi surface plots but does not enter the SHC directly.
  • Layer thickness for 3D conversion = unstated (implicit)
    The 3D SHC value -5200 (ℏ/e) S/cm is obtained by converting the 2D bilayer value using an unspecified thickness; the comparison with bulk platinum depends on this choice (main text, Fig. 3 discussion).
assumptions (5)
  • domain assumption DFT-PBE with the specified PAO basis sets accurately describes the electronic structure and spin-orbit coupling of the twisted bilayers.
    The entire SHC calculation rests on the OpenMX Hamiltonian; PBE is an approximation and the PAO basis sizes are limited (Mo7.0-s3p2d1, Te7.0-s3p2d2).
  • domain assumption The transfer-learning relaxed structures faithfully represent the experimental moiré lattice reconstruction.
    The interatomic potential is trained on generated data from the authors' prior work (refs. 22-25) and could carry systematic errors that affect the band inversions producing large Berry curvature.
  • standard math The Kubo formula with the spin current operator defined in Eq. 14 is the correct linear-response expression for spin Hall conductivity in these systems.
    This is a standard formalism, but its implementation with non-orthogonal PAOs via S(k)*s_z requires further justification that the paper does not provide.
  • domain assumption Zero temperature and a single 10 meV broadening capture the relevant physics; finite temperature and disorder are neglected.
    Metallic SHC at finite temperature may be reduced, and the paper does not address temperature or disorder dependence.
  • domain assumption The twisted bilayer can be treated as a standalone two-dimensional system for direct comparison with three-dimensional bulk spin Hall conductors.
    The conversion to 3D units and the 'record' statement assume that a bilayer's 2D SHC divided by an interlayer spacing is directly comparable to bulk platinum, ignoring substrate, encapsulation, and screening effects.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Universal giant spin Hall effect in moire metal." pith.science (2026). https://pith.science/paper/EXXMA33G

@misc{pith2026250416179,
  author       = {Pith},
  title        = {Pith review of: Universal giant spin Hall effect in moire metal},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/EXXMA33G}},
  note         = {Machine review of arXiv:2504.16179}
}
abstract

While moir\'e phenomena have been extensively studied in low-carrier-density systems such as graphene and semiconductors, their implications for metallic systems with large Fermi surfaces remain largely unexplored. Using GPU-accelerated large-scale ab-initio quantum transport simulations, we investigate spin transport in two distinct platforms: twisted bilayer MoTe$_2$ (semiconductor, from lightly to heavily doping) and NbX$_2$ ($X$ = S, Se; metals). In twisted MoTe$_2$, the spin Hall conductivity (SHC) evolves from $4\tfrac{e}{4\pi}$ at $5.09^\circ$ to $10\tfrac{e}{4\pi}$ at $1.89^\circ$, driven by the emergence of multiple isolated Chern bands. Remarkably, in heavily doped metallic regimes--without isolated Chern bands--we observe a universal amplification of the spin Hall effect from Fermi surface reconstruction under long-wavelength potential, with the peak SHC tripling from $6\tfrac{e}{4\pi}$ at $5.09^\circ$ to $17\tfrac{e}{4\pi}$ at $3.89^\circ$. For prototypical moir\'e metals like twisted NbX$_2$, we identify a record SHC of $-17\tfrac{e}{4\pi}$ (-5200 $(\hbar / e)S/cm$ in 3D units), surpassing all known bulk materials. These results establish moir\'e engineering as a powerful strategy for enhancing spin-dependent transport, and advancing ab-initio methodologies to bridge atomic-scale precision with device-scale predictions in transport simulations.

Figures

Figures reproduced from arXiv: 2504.16179 by the authors.

Figure 1
Figure 1. FIG. 1. Band structures of t-MoTe [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. FIG. 2. Zoomed-in views of the SHC over a 0.15 eV energy [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. FIG. 3. (a) Band structure of t-NbSe [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: FIG. 4. (a) Folded Fermi surface within the Brillouin zone of [PITH_FULL_IMAGE:figures/full_fig_p004_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

52 extracted references · 33 canonical work pages

  1. [1]

    Bistritzer and A

    R. Bistritzer and A. H. MacDonald, Proc. Natl. Acad. Sci. U.S.A. 108, 12233 (2011)

    2011

  2. [2]

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

    2018

  3. [3]

    F. Wu, T. Lovorn, E. Tutuc, I. Martin, and A. MacDon- ald, Phys. Rev. Lett. 122, 086402 (2019)

    work page 2019

  4. [4]

    D. M. Kennes, M. Claassen, L. Xian, A. Georges, A. J. Millis, J. Hone, C. R. Dean, D. Basov, A. N. Pasupathy, and A. Rubio, Nat. Phys. 17, 155 (2021)

    work page 2021

  5. [5]

    K. F. Mak and J. Shan, Nature Nanotechnology 17, 686 (2022)

    2022

  6. [6]

    S. Wang, Y. Liu, Y. Gu, T. Bao, N. Mao, C. Li, S. Jiang, L. Liu, D. Guan, Y. Li, et al. (2024)

    work page 2024

  7. [7]

    M. Kang, L. Ye, S. Fang, J.-S. You, A. Levitan, M. Han, J. I. Facio, C. Jozwiak, A. Bostwick, E. Rotenberg, et al., Nature materials 19, 163 (2020)

    work page 2020

  8. [8]

    J. Cai, E. Anderson, C. Wang, X. Zhang, X. Liu, W. Holtzmann, Y. Zhang, F. Fan, T. Taniguchi, K. Watanabe, et al., Nature pp. 1–3 (2023)

    work page 2023

Show all 52 references
  1. [9]

    Y. Zeng, Z. Xia, K. Kang, J. Zhu, P. Kn¨ uppel, C. Vaswani, K. Watanabe, T. Taniguchi, K. F. Mak, and J. Shan, Nature pp. 1–2 (2023)

    2023

  2. [10]

    H. Park, J. Cai, E. Anderson, Y. Zhang, J. Zhu, X. Liu, C. Wang, W. Holtzmann, C. Hu, Z. Liu, et al., Nature 622, 74 (2023)

    2023

  3. [11]

    F. Xu, Z. Sun, T. Jia, C. Liu, C. Xu, C. Li, Y. Gu, K. Watanabe, T. Taniguchi, B. Tong, et al., Phys. Rev. X 13, 031037 (2023)

    2023

  4. [12]

    K. Kang, Y. Qiu, K. Watanabe, T. Taniguchi, J. Shan, and K. F. Mak, Nano Lett. 24, 14901 (2024)

    2024

  5. [13]

    K. Kang, B. Shen, Y. Qiu, Y. Zeng, Z. Xia, K. Watanabe, T. Taniguchi, J. Shan, and K. F. Mak, Nature pp. 1–5 (2024)

    2024

  6. [14]

    C. Xu, N. Mao, T. Zeng, and Y. Zhang, Phys. Rev. Lett. 134, 066601 (2025)

    2025

  7. [15]

    Z. Tao, B. Shen, W. Zhao, N. C. Hu, T. Li, S. Jiang, L. Li, K. Watanabe, T. Taniguchi, A. H. MacDonald, et al., Nat. Nanotechnol. 19, 28 (2024)

    2024

  8. [16]

    Zhang, C

    Y. Zhang, C. Felser, and L. Fu, Phys. Rev. B 110, L041407 (2024)

    2024

  9. [17]

    Finkelstein, C

    J. Finkelstein, C. F. Negre, and J.-L. Fattebert, J. Chem. Phys. 159 (2023)

    2023

  10. [18]

    Nguyen, D

    M.-H. Nguyen, D. Ralph, and R. Buhrman, Phys. Rev. Lett. 116, 126601 (2016)

    2016

  11. [19]

    Zhang, W

    W. Zhang, W. Han, X. Jiang, S.-H. Yang, and S. SP Parkin, Nat. Phys. 11, 496 (2015)

    2015

  12. [20]

    G.-Y. Guo, S. Murakami, T.-W. Chen, and N. Nagaosa, Phys. Rev. Lett. 100, 096401 (2008)

    2008

  13. [21]

    Zhang, Q

    Y. Zhang, Q. Xu, K. Koepernik, R. Rezaev, O. Janson, J. ˇZelezn` y, T. Jungwirth, C. Felser, J. van den Brink, and Y. Sun, npj Comput. Mater. 7, 167 (2021)

    2021

  14. [22]

    N. Mao, C. Xu, J. Li, T. Bao, P. Liu, Y. Xu, C. Felser, L. Fu, and Y. Zhang, Communications Physics 7, 262 (2024)

    2024

  15. [23]

    T. Bao, N. Mao, W. Duan, Y. Xu, A. Del Maestro, and Y. Zhang, arXiv preprint arXiv:2501.12452 (2025). 6

    2025 arXiv

  16. [24]

    Zhang, C

    X.-W. Zhang, C. Wang, X. Liu, Y. Fan, T. Cao, and D. Xiao, Nat. Commun. 15, 4223 (2024)

    2024

  17. [25]

    Y. Jia, J. Yu, J. Liu, J. Herzog-Arbeitman, Z. Qi, H. Pi, N. Regnault, H. Weng, B. A. Bernevig, and Q. Wu, Phys. Rev. B 109, 205121 (2024)

    2024

  18. [26]

    Ozaki and H

    T. Ozaki and H. Kino, Phys. Rev. B 69, 195113 (2004)

    2004

  19. [27]

    Ozaki, Phys

    T. Ozaki, Phys. Rev. B 67, 155108 (2003)

    2003

  20. [28]

    Gmitra and J

    M. Gmitra and J. Fabian, Phys. Rev. Lett. 119, 146401 (2017)

    2017

  21. [29]

    Bawden, S

    L. Bawden, S. P. Cooil, F. Mazzola, J. Riley, L. Collins- McIntyre, V. Sunko, K. Hunvik, M. Leandersson, C. Pol- ley, T. Balasubramanian, et al., Nat. Commun. 7, 11711 (2016)

    2016

  22. [30]

    Ezawa, Sci

    M. Ezawa, Sci. Rep. 3, 3435 (2013)

    2013

  23. [31]

    Ezawa, Phys

    M. Ezawa, Phys. Rev. Lett. 109, 055502 (2012)

    2012

  24. [32]

    Fukui, Y

    T. Fukui, Y. Hatsugai, and H. Suzuki, J. Phys. Soc. Jpn.. 74, 1674 (2005)

    2005

  25. [33]

    See Supplemental Material at [URL will be inserted by publisher] for the full details on the derivations

  26. [34]

    C. Wang, S. Zhao, X. Guo, X. Ren, B.-L. Gu, Y. Xu, and W. Duan, New J. Phys. 21, 093001 (2019)

    2019

  27. [35]

    G. Jin, D. Zheng, and L. He, J. Condens. Matter Phys. 33, 325503 (2021)

    2021

  28. [36]

    Lee, Y.-T

    C.-C. Lee, Y.-T. Lee, M. Fukuda, and T. Ozaki, Phys. Rev. B 98, 115115 (2018)

    2018

  29. [37]

    Nishi, Y.-i

    H. Nishi, Y.-i. Matsushita, and A. Oshiyama, Phys. Rev. B 95, 085420 (2017)

    2017

  30. [38]

    Zhang, J

    L. Zhang, J. Han, H. Wang, R. Car, and E. Weinan, Phys. Rev. Lett. 120, 143001 (2018)

    2018

  31. [39]

    Morrison, D

    I. Morrison, D. M. Bylander, and L. Kleinman, Phys. Rev. B 47, 6728 (1993)

    1993

  32. [40]

    R. Yu, X. L. Qi, A. Bernevig, Z. Fang, and X. Dai, Phys. Rev. B 84, 075119 (2011)

    2011

  33. [41]

    Alexandradinata, X

    A. Alexandradinata, X. Dai, and B. A. Bernevig, Phys. Rev. B 89, 155114 (2014)

    2014

  34. [42]

    C. L. Kane and E. J. Mele, Phys. Rev. Lett. 95, 226801 (2005)

    2005

  35. [43]

    C. L. Kane and E. J. Mele, Phys. Rev. Lett. 95, 146802 (2005)

    2005

  36. [44]

    Sheng, D

    L. Sheng, D. Sheng, C. Ting, and F. Haldane, Phys. Rev. Lett. 95, 136602 (2005)

    2005

  37. [45]

    Fukui and Y

    T. Fukui and Y. Hatsugai, Phys. Rev. B 75, 121403 (2007)

    2007

  38. [46]

    Fu and C

    L. Fu and C. L. Kane, Phys. Rev. B 74, 195312 (2006)

    2006

  39. [47]

    Thompson, K

    E. Thompson, K. T. Chu, F. Mesple, X.-W. Zhang, C. Hu, Y. Zhao, H. Park, J. Cai, E. Anderson, K. Watan- abe, et al., arXiv preprint arXiv:2405.19308 (2024)

    2024 arXiv

  40. [48]

    Kerelsky, L

    A. Kerelsky, L. J. McGilly, D. M. Kennes, L. Xian, M. Yankowitz, S. Chen, K. Watanabe, T. Taniguchi, J. Hone, C. Dean, et al., Nature 572, 95 (2019). 7 Supplemental Materials I. TRANSFER LEARNING LA TTICE RELAXA TION We adopted a two-step transfer learning approach to model la...

    2019

  41. [49]

    Due to the strain, the superlattice vectors cannot perfectly match the modified moir´ e wave vectors, resulting in an inherent error

    (S19) where a′ 1 and a′ 2 are strained lattice vectors. Due to the strain, the superlattice vectors cannot perfectly match the modified moir´ e wave vectors, resulting in an inherent error. To minimize this error, we employed a gradient descent method by setting the twist angl...

    2010

  42. [50]

    Because this is a proper rotation, Ω xy does not change sign but is evaluated at the rotated wavevector

    Threefold rotation about the z-axis, C3z: C3z : Ω xy(k) −→ Ωxy R3z(k) , (S21) whereR3z rotates the in-plane momentum by 120◦. Because this is a proper rotation, Ω xy does not change sign but is evaluated at the rotated wavevector

  43. [51]

    As an example, under 2 100, kx→kx, k y→−ky, k z→−kz, (S22) and the antisymmetric tensor component Ω xy changes sign: 2100 : Ω xy(kx,ky,kz) −→ −Ωxy kx,−ky,−kz

    Twofold rotations about in-plane axes, 2100, 2010, 2110: Each of these is a 180◦ rotation about a vector lying in the crystal plane. As an example, under 2 100, kx→kx, k y→−ky, k z→−kz, (S22) and the antisymmetric tensor component Ω xy changes sign: 2100 : Ω xy(kx,ky,kz) −→ −Ω...

    2010

  44. [52]

    (S24) Time-reversal thus reverses the sign of Ω xy and simultaneously inverts k

    Time-reversal symmetry ,T : Under time reversal, T : k−→− k, Ωxy(k)−→− Ωxy −k . (S24) Time-reversal thus reverses the sign of Ω xy and simultaneously inverts k. As before, time-reversal symmetry flip the spin channel, and give the Jx a minus sign. Overall, we can use all the f...

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.