Pith. sign in

REVIEW 3 major objections 3 minor 46 references

Orbital correlations in bilayer nickelates: roles of doping and interlayer coupling

T0 review · 3 major / 3 minor · reviewed 2026-08-09 · deepseek-v4-flash

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

Pith's one-line read The paper claims that transverse orbital correlations near wavevector (π/2, π/2) dominate in bilayer nickelates and may help stabilize the weakly insulating state.

desk verdict A transparent, modest RPA study of orbital correlations in bilayer nickelates, but the claim that transverse orbital correlations dominate is not backed by a spin-susceptibility comparison. read the letter →

arxiv 2502.00701 v2 pith:6LKHVZV5 submitted 2025-02-02 cond-mat.str-el cond-mat.supr-con

Garima Goyal , Aastha Jain , Dheeraj Kumar Singh This is my paper · ORCID
classification cond-mat.str-elcond-mat.supr-con
keywords bilayernickelatesLa3Ni2O7orbitalcorrelationstransversesusceptibilityrandom-phaseapproximationFermi-surfacenestinginterlayercouplingweaklyinsulatingstate
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 asks whether orbital fluctuations, not just spin fluctuations, are strong in the bilayer nickelate superconductor La$_3$Ni$_2$O$_7$. Within a two-orbital tight-binding model and static random-phase-approximation susceptibilities, it finds that the dominant orbital response is transverse: the transverse orbital susceptibility develops a sharp, near-diverging peak near $\sim(\pi/2,\pi/2)$ for realistic hole doping and interlayer coupling, while the longitudinal channel stays broad. The peak positions and intensities move systematically with carrier concentration and with the interlayer hopping $t_z^\perp$, tracking changes in Fermi-surface nesting and orbital content. If correct, this makes transverse orbital fluctuations a plausible partner in stabilizing the weakly insulating state seen at ambient pressure, and a candidate source of pairing glue.

What carries the argument

The load-bearing object is the static orbital susceptibility matrix $\hat{\chi}^{\mathrm{orb}}(\mathbf{q}) = \hat{\chi}(\mathbf{q})[\hat{1} + \hat{U}\hat{\chi}(\mathbf{q})]^{-1}$ evaluated in the random-phase approximation, with $\hat{\chi}(\mathbf{q})$ the noninteracting susceptibility of a two-orbital ($d_{x^2-y^2}$, $d_{3z^2-r^2}$) bilayer tight-binding model. The response is split into longitudinal orbital correlations, from $O^{\mathrm{long}}_{il} = n_{i\mu} - n_{i\nu}$, and transverse orbital correlations, from $O^{\mathrm{trans}}_{il} = d^\dagger_{i\mu} d_{i\nu} + d^\dagger_{i\nu} d_{i\mu}$; strong transverse correlations would order the rotated orbitals $d_+ = (d_{x^2-y^2} + d_{3z^2-r^2})/\sqrt{2}$ and $d_- = (d_{x^2-y^2} - d_{3z^2-r^2})/\sqrt{2}$. The sharp peaks in the transverse channel trace the interpocket nesting between the straight legs of the hole pockets around $M$, which is why doping and the interlayer hopping $t_z^\perp$ move the peak positions.

What would settle it

A calculation of the orbital susceptibility with vertex corrections beyond RPA—for instance dynamical mean-field theory or diagrammatic Monte Carlo on the same two-orbital model—that shows no near-divergence at $\sim(\pi/2,\pi/2)$ would falsify the RPA-level claim; so would an experiment such as orbital-resolved RIXS finding no enhancement of transverse orbital fluctuations or $d_+$/$d_-$ orbital order near that wavevector in La$_3$Ni$_2$O$_7$ at low temperature and pressures around 1 GPa.

Watch

Extended reading notes

Core claim

The paper's central claim is that transverse orbital correlations at wavevector $\sim(\pi/2,\pi/2)$ dominate the orbital response of La$_3$Ni$_2$O$_7$. The authors compute static orbital susceptibilities in the random-phase approximation and find that the transverse channel—the response of $d^\dagger_\mu d_\nu + d^\dagger_\nu d_\mu$, which would order the rotated orbitals $d_+$ and $d_-$—shows sharp peaks on the verge of divergence, whereas the longitudinal channel shows only broad peaks. The sharp peaks are tied to interpocket nesting between the straight legs of the two hole pockets around $M=(\pi,\pi)$; as doping or the interlayer hopping $t_z^\perp$ changes, the legs straighten or bend and the peak wavevectors shift accordingly. The paper concludes that these transverse orbital fluctuations, through their interplay with spin degrees of freedom, are expected to play an important role in stabilizing the weakly insulating state, and that orbital-lattice coupling could push them to mediate superconductivity.

Load-bearing premise

The calculation assumes the two-orbital tight-binding model with its quoted hoppings and on-site energies faithfully represents the low-energy electronic structure of La$_3$Ni$_2$O$_7$, and that the static RPA susceptibility accurately captures the orbital correlations; if either fails, the claimed dominance of the $\sim(\pi/2,\pi/2)$ transverse channel could shift or disappear.

Editorial extensions

If this is right

  • The transverse orbital susceptibility is near divergence at $\sim(\pi/2,\pi/2)$ for realistic parameters, so a staggered orbital order of $d_+$/$d_-$ orbitals is a plausible instability of the paramagnetic metal.
  • Hole doping moves the orbital-correlation peaks toward smaller wavevectors, while electron doping flattens the transverse response and shifts its peaks toward $(\pi,\pi)$, making orbital correlations strongly doping-tunable.
  • Increasing the interlayer coupling $t_z^\perp$ sharpens the transverse response and can push it to divergence at $\sim(\pi/2,\pi/2)$ for $x_h=0.1$, so pressure acts as a direct control knob for orbital fluctuations.
  • If transverse orbital fluctuations are as strong as the RPA indicates, they should be weighed alongside spin fluctuations in explaining the weakly insulating state and the pairing mechanism of bilayer nickelates.
  • Jahn-Teller coupling of the NiO$_6$ octahedra could further enhance transverse orbital fluctuations, possibly turning them into pairing glue.

Reading between the lines

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

  • Editorial: the near-divergence implies a real ordered phase with staggered $d_+$/$d_-$ orbital order at $\sim(\pi/2,\pi/2)$; orbital-resolved RIXS or resonant x-ray scattering could look for that order directly.
  • Editorial: if transverse orbital fluctuations mediate pairing, the superconducting gap symmetry should differ from the spin-fluctuation-driven $(\pi,0)$ scenario, a distinction experiments can probe.
  • Editorial: the strong $t_z^\perp$ dependence suggests that uniaxial c-axis strain, not only hydrostatic pressure, would tune orbital correlations and possibly $T_c$.
  • Editorial: the static-RPA treatment neglects vertex corrections and finite-frequency dynamics; a dynamical calculation could confirm whether the near-divergence survives.
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

3 major / 3 minor

Summary. The paper studies orbital correlations in a minimal two-orbital tight-binding model of bilayer nickelates (dx2−y2 and d3z2−r2 orbitals), computing static longitudinal and transverse orbital susceptibilities within the random-phase approximation (RPA) as functions of hole/electron doping and interlayer coupling tz⊥. The authors present Fermi-surface maps showing how orbital content and nesting change with doping and tz⊥, and they plot the orbital susceptibilities along high-symmetry lines. The central qualitative findings are that the longitudinal orbital susceptibility exhibits broad peaks while the transverse orbital susceptibility shows sharper, nearly divergent peaks near wavevectors such as (π/2, π/2), and that these peak positions shift systematically with doping and interlayer coupling. The conclusion claims that transverse orbital correlations with wavevector ~(π/2, π/2) are dominant in this class of materials and, through interplay with spin degrees of freedom, may play an important role in the weakly insulating state.

Significance. If the central claim were fully established, the paper would provide a useful counterpoint to the prevailing spin-fluctuation focus in bilayer nickelate research, pointing to orbital degrees of freedom as a potentially relevant actor. The calculations are internally consistent: all model parameters are stated, the RPA formulas are standard, and the Fermi-surface analysis is clear and well illustrated. The paper also discloses its parameter choices (e.g., U=0.44 eV, J=0.1U) and uses a two-orbital model from the literature, making the computations reproducible. However, the key interpretive claim that transverse orbital correlations are 'dominant' goes beyond the evidence presented, because the spin susceptibility is never computed and because the plotted quantity from the 16×16 RPA matrix is not explicitly identified. These omissions weaken the significance of the paper as it stands, although they are fixable with additional analysis.

major comments (3)
  1. [Summary and conclusion] The conclusion states that 'transverse orbital correlations with wavevector ~(π/2, π/2) in this class of superconducting materials are dominant,' but the paper never computes the spin susceptibility, so 'dominant' is not established over the spin channel. Within the same two-orbital model and interaction matrix (Eqs. (7)-(10)), the RPA spin susceptibility can be computed straightforwardly; the authors should do so at the same parameter sets (U=0.43/0.44 eV, J=0.1U, and the same tz⊥ values) and compare peak heights to the transverse orbital susceptibility. Without such a comparison, the claim of dominance and the proposed spin-orbital interplay for the weakly insulating state rest on an uncomputed competition.
  2. [Model and Method, Eq. (9), and Figs. 4-7] The manuscript does not specify whether the plotted susceptibility is the largest eigenvalue of the 16×16 RPA matrix χ^orb(q) or a particular matrix element (e.g., a component diagonal in layer and orbital indices). This distinction matters because the RPA divergence condition is det(1 + Ûχ̂) = 0, and a single element may not reflect the leading instability. The authors should state explicitly which quantity is plotted in Figs. 4-7; if only a component is plotted, the leading eigenvalue should also be reported, especially for the claim of 'diverging behavior' near (π/2, π/2).
  3. [Results and discussion, choice of U (before Fig. 4 and before Fig. 6)] The interaction strength U is chosen as U=0.44 (and U=0.43 for the tz⊥ dependence) explicitly because the transverse orbital susceptibility diverges beyond that value. This makes the observation that the transverse susceptibility is 'on the verge of divergence' partly a consequence of the parameter choice, not an independent finding. To give physical weight to the near-instability statement, the authors should anchor U to independent estimates from first-principles or DMFT studies of bilayer nickelates, or at least show how the susceptibility evolves over a broader range of U and discuss the uncertainty in U. Without this, the near-criticality is an artifact of the chosen U.
minor comments (3)
  1. [Figures 4 and 5] There is an inconsistency between the in-figure labels and the captions: the captions state U=0.44, while the in-figure labels read U=0.45. Please correct one of them.
  2. [Manuscript header] The header contains the placeholder '*** Missing PACS ***'; the authors should supply the appropriate PACS numbers for the journal submission.
  3. [Introduction, first paragraph] There are several typos and awkward phrasings, e.g., 'differerent' and 'Ni +2. 5 shows mixed valency' appears garbled. A careful proofread is recommended throughout.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the orbital susceptibility peaks emerge from the RPA calculation and are not equivalent to the model inputs.

full rationale

The claimed derivation chain runs from an external two-orbital tight-binding model (Eqs. 1-6, parameters from Luo et al. [40]) through the bare susceptibility (Eq. 10) and the RPA resummation (Eq. 9) to the longitudinal and transverse orbital susceptibilities shown in Figs. 4-7. The peak positions and relative peak heights are emergent outputs of the band structure, orbital content, and RPA denominator; they are not imposed by any fitting step. The interaction U is scanned and set near the RPA divergence (U=0.44), but U is an input parameter, not a quantity the paper claims to predict, so this parameter choice does not constitute circularity. The only self-citations ([41], [43]) supply standard operator definitions and RPA interaction-matrix elements; neither is load-bearing, and no uniqueness theorem or prior result by the authors is invoked to forbid alternatives. The statement that transverse orbital correlations are 'dominant' compares longitudinal and transverse orbital channels and does not reduce to an input by construction; whether it should have been benchmarked against the spin susceptibility is a completeness concern, not a circularity one. Accordingly, no circular step is identified.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The paper introduces no new particles, forces, or conserved quantities. Its free parameters are the interaction scales U and J, chosen partly to sit near the RPA critical point, and the inherited tight-binding parameters. The axioms are the usual RPA-model assumptions: the two-orbital model represents the material, the RPA susceptibility captures the leading correlations, and the rotation-invariant interaction form holds with a fixed Hund's ratio.

free parameters (3)
  • On-site Coulomb interaction U = 0.43 to 0.44 eV
    Chosen just below the value where the transverse RPA orbital susceptibility diverges, so the near-criticality of the transverse channel is partly imposed by this choice rather than predicted independently.
  • Hund's coupling J = 0.1 U
    Set ad hoc to 10% of U throughout; this affects the relative strength of the longitudinal and transverse orbital channels.
  • Tight-binding parameters from ref [40] = See Eqs. (4)-(6)
    All hopping integrals and on-site energies are inherited from a DFT-fitted model, and the Fermi-surface topology and orbital contents, hence all susceptibility peak positions, depend on these values.
assumptions (4)
  • domain assumption The two-orbital tight-binding model of ref [40] captures the low-energy electronic structure of La3Ni2O7, with t2g orbitals fully occupied and inert.
    The model omits other orbitals and correlation effects beyond the two eg orbitals; this is central to the Fermi surfaces and all susceptibility results.
  • domain assumption The static RPA susceptibility of Eq. (9), built from the bare bubble of Eq. (10), is a reliable approximation for orbital correlations in this material.
    RPA neglects vertex corrections and finite-frequency effects; the near-divergence predictions depend on this approximation.
  • ad hoc to paper Interaction parameters satisfy the rotation-invariant condition U = U' + 2J with J = 0.1U.
    The paper fixes the Hund's coupling ratio without independent justification, and U' is then determined by the relation, affecting both orbital channels.
  • domain assumption The undoped filling n = 1.5 corresponds to chemical potential mu = 0, assuming a Ni oxidation state with 7.5 d electrons including fully occupied t2g orbitals.
    This filling defines the doping axis in Figs. 1-7, and the susceptibility peak trends depend on it.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Orbital correlations in bilayer nickelates: roles of doping and interlayer coupling." pith.science (2026). https://pith.science/paper/6LKHVZV5

@misc{pith2026250200701,
  author       = {Pith},
  title        = {Pith review of: Orbital correlations in bilayer nickelates: roles of doping and interlayer coupling},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/6LKHVZV5}},
  note         = {Machine review of arXiv:2502.00701}
}
read the original abstract

We study the nature of orbital correlations present in the bilayer nickelate within a minimal two-orbital tight-binding model to gain insights into their possible role in stabilizing the less-known weakly-insulating state. The latter has been observed experimentally at ambient pressure. In order to achieve this objective, we examine the static orbital susceptibilities within the random-phase approximation. Our study highlights the sensitivity of orbital correlations to various factors including the interlayer coupling, carrier concentration, band-structure details such as the orbital contents, the number of bands contributing at the Fermi level etc. We relate this sensitiveness to the modification of the Fermi surfaces as well as their orbital contents dependent on aforementioned factors.

Figures

Figures reproduced from arXiv: 2502.00701 by the authors.

Figure 1
Figure 1. Fermi surfaces in the paramagnetic state with orbita [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Fermi surfaces in the paramagnetic state with orbita [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 5
Figure 5. Transverse-orbital susceptibility along the high- [PITH_FULL_IMAGE:figures/full_fig_p004_5.png] view at source ↗
Figures from the paper (1 more)
Figure 6
Figure 6. Figure 6: Longitudinal orbital susceptibility along the high [PITH_FULL_IMAGE:figures/full_fig_p005_6.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

46 extracted references · 42 canonical work pages

  1. [1]

    Stewart G. R., Adv. in Phys., 66 (2017) 75

    work page 2017

  2. [2]

    Bednorz J. G. and Miller K. A., Z. Physik B - Condensed Matter , 64 (1986) 189

    work page 1986

  3. [3]

    Fernandes R M, Coldea A I, Ding H, Fisher I R, Hirschfeld P J, Kotliar G., Nature, 601 (2022) 35

    work page 2022

  4. [4]

    S., Norman M

    Karp J., Botana A. S., Norman M. R., Park H., Zingl M., and Millis A., Phys. Rev. X, 10 (2020) 021061

    work page 2020

  5. [5]

    and Hoshino S., Phys

    Werner P . and Hoshino S., Phys. Rev. B, 101 (2020) 041104(R)

    work page 2020

  6. [6]

    Phys., 3 (2020) 84

    Gu Y ., Zhu S., Wang X., Hu J., and Chen H., Commun. Phys., 3 (2020) 84

    work page 2020

  7. [7]

    Taillefer L., Annu. Rev. Condens. Matter Phys. , 1 (2010) 51

    work page 2010

  8. [8]

    Martinelli A., Bernardini F., and Massidda S., C. R. Physique, 17 (2016) 5

    work page 2016

Show all 46 references
  1. [9]

    H., Cadogan J

    Ryan D. H., Cadogan J. M., Ritter C., Canepa F., Palenzona A., and Putti M., Phys. Rev. B, 80 (2009) 220503(R)

    2009

  2. [10]

    Y ., Lee K., Harvey S

    Li D., Wang B. Y ., Lee K., Harvey S. P ., Osada M., Goodge B. H., Kourkoutis L. F., and Hwang H. Y ., Phys. Rev. Lett., 125 (2020) 027001

    2020

  3. [11]

    and Arita R., Rep

    Nomura Y . and Arita R., Rep. Prog. Phys., 85 (2022) 052501

    2022

  4. [12]

    Botana A. S. and Norman M. R., Phys. Rev. X, 10 (2020) 011024

    2020

  5. [13]

    et al., Sci

    Liu Z. et al., Sci. China Phys. Mech. Astron. , 66 (2023) 217411

    2023

  6. [14]

    et al., Nature, 621 (2023) 493

    Sun H. et al., Nature, 621 (2023) 493

    2023

  7. [15]

    et al., Chin

    Hou J. et al., Chin. Phys. Lett. , 40 (2023) 117302

    2023

  8. [16]

    and Yang Y

    Cao Y . and Yang Y . -F., Phys. Rev. B, 109 (2024) L081105

    2024

  9. [17]

    Zhang, Y . Lin L. F., Moreo A., Maier T. A., and Dagotto E., Phys. Rev. B, 108 (2023) 165141

    2023

  10. [18]

    Zhang, Y . Lin L. F., Moreo A., and Dagotto E., Phys. Rev. B, 108 (2023) L180510

    2023

  11. [19]

    Commun., 15 (2024) 7570

    Liu Z et al., Nat. Commun., 15 (2024) 7570

    2024

  12. [20]

    Y .et al., Nat

    Peng Y . Y .et al., Nat. Phys. 13 (2017) 1201

    2017

  13. [21]

    P ., Cohn J

    Neupane K. P ., Cohn J. L., Terashita H., and Neumeier J. J ., Phys. Rev. B, 74 (2006) 144428

    2006

  14. [22]

    and Tokura Y ., Lecture Notes in Physics , 595 (2002) 361

    Kimura T. and Tokura Y ., Lecture Notes in Physics , 595 (2002) 361

    2002

  15. [23]

    Rep., 344 (2001) 1

    Dagotto E., Hotta T., and Moreo A., Phys. Rep., 344 (2001) 1

    2001

  16. [24]

    Rao C. N. R., J. Phys. Chem. B , 104 (2000) 5877

    2000

  17. [25]

    K., Kamble B., and Singh A., Phys

    Singh D. K., Kamble B., and Singh A., Phys. Rev. B 81 (2010) 064430

    2010

  18. [26]

    K., Kamble B

    Singh D. K., Kamble B. and Singh A., J. Phys.: Condens. Matter 22 (2010) 396001

    2010

  19. [27]

    and Tomioka Y ., J

    Tokura Y . and Tomioka Y ., J. Magn. Magn. Mater ., 200 (1999) 1

    1999

  20. [28]

    and Singh D

    Singh A. and Singh D. K., J. Phys.: Condens. Matter 24 (2012) 086003

    2012

  21. [29]

    et al., Nat

    Chen X. et al., Nat. Commun., 15 (2024) 9597. p-5 Garima Goyal Aastha Jain Dheeraj Kumar Singh

    2024

  22. [30]

    et al., arXiv:2311.15717 (2023)

    Chen K. et al., arXiv:2311.15717 (2023)

    2023 arXiv

  23. [31]

    et al., arXiv:2402.10485 (2024)

    Khasanov R. et al., arXiv:2402.10485 (2024)

    2024 arXiv

  24. [32]

    et al., arXiv:2402.03952 (2024)

    Dan Z. et al., arXiv:2402.03952 (2024)

    2024 arXiv

  25. [33]

    et al., Sci

    Xie T. et al., Sci. Bull., 69 (2024) 3221

    2024

  26. [34]

    F., Moreo A., Maier T

    Zhang Y ., Lin L. F., Moreo A., Maier T. A., and Dagotto E., Nat. Comm. 15 (2024) 2470

    2024

  27. [35]

    M., Phys

    Lechermann F., Gondolf J., B¨ otzel S., and Eremin I. M., Phys. Rev. B, 108 (2023) L201121

    2023

  28. [36]

    Yang Q.-G., Wang D., and Wang Q.-H., Phys. Rev. B, 108 (2023) L140505

    2023

  29. [37]

    Gu Y ., Le C., Yang Z., Wu X., and Hu J., arXiv:2306.07275 (2023)

    2023 arXiv

  30. [38]

    -C., and Hu J., arXiv:2401.15097

    Wang Y ., Jiang K., Wang Z., Zhang F. -C., and Hu J., arXiv:2401.15097

    arXiv

  31. [40]

    -X., Phys

    Luo Z., Hu X., Wang M., W´ u W., and Yao D. -X., Phys. Rev. Lett., 131 (2023) 126001

    2023

  32. [41]

    K., Lee K

    Singh D. K., Lee K. L., and Takimoto T., J. Phys. Soc. Jpn. 84 (2015) 064709

    2015

  33. [42]

    Takimoto T., Hotta T., and Ueda K., Phys. Rev. B, 69 (2004) 104504

    2004

  34. [43]

    K., EPL 112 (2015) 27004

    Singh D. K., EPL 112 (2015) 27004

    2015

  35. [44]

    Dagotto E., Hotta T., and Moreo A., Phys. Rep. 344 (2001) 1

    2001

  36. [45]

    and Onari S., Phys

    Kontani H. and Onari S., Phys. Rev. Lett. 104 (2010) 157001

    2010

  37. [46]

    et al., Nature 634, (2024) 579

    Wang N. et al., Nature 634, (2024) 579

    2024

  38. [47]

    et al., arXiv:2501.09255 (2025)

    Li P . et al., arXiv:2501.09255 (2025). p-6

    2025 arXiv

Pith tools

Reviewed August 9, 2026 · model on record in the stance chip above.