{"id":"4030acfc-e6bf-4e5f-ae3d-f232ddbffa4f","arxiv_id":"2411.16986","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":7,"one_line_summary":"The ground state of neutral graphene in a strong magnetic field switches from antiferromagnetic to Kekulé-distorted as screening and field decrease, with the Dirac sea providing the decisive energy difference.","lead":"This paper calculates how the magnetic ordering of graphene in a strong magnetic field changes with the dielectric environment, using a combination of renormalization group and mean-field theory. It finds that weaker screening and lower field favor a Kekulé-distorted state over an antiferromagnet, and that the filled 'Dirac sea' of negative-energy Landau levels drives this switch.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The isotropic Dirac fluid assumption (Eq. 71) is the load-bearing constraint: it is untested precisely in the strong-coupling KD regime where the Dirac sea is claimed to dominate the magnetic anisotropic energy.","rationale":"The reader's weakest-assumption identification matches my own reading: Eq. (71) is the pivot on which the phase diagram and the Dirac-sea energy budget turn. The paper is honest in flagging this assumption and its possible breakdown into a nematic phase, but it does not test that breakdown in the parameter regime where the headline claims are made. Since the KD phase is predicted precisely at large κ0 and small B, and since Fig. 10 shows the Dirac sea becoming the dominant energy contribution in that same regime, the validity of Eq. (71) is not a peripheral technicality but a necessary condition for the central claim. The concrete test I propose would settle the question: an unrestricted HF calculation (or a nematic-stability check) at representative points in the KD region would reveal whether the isotropic solution is the true mean-field ground state. A finding of a lower-energy nematic state would require re-evaluating the phase diagram and the Dirac-sea decomposition; a null finding would significantly strengthen the paper. I therefore keep the reader's CONDITIONAL verdict unchanged, with the condition being this missing stability check.","tokens_in":25437,"tokens_out":7441,"duration_ms":79457,"concrete_test":"Perform unrestricted self-consistent Hartree-Fock at representative points in the KD phase (e.g., κ0 = 0.8 at B = 5 T and B = 10 T for ν = 0 and ν = ±1), allowing off-diagonal density-matrix elements ρ_{nξ,n′ξ′} with n ≠ n′ (or at least nematic Δn = 2 perturbations). If an anisotropic/nematic solution has lower energy than the isotropic solution satisfying Eq. (71), the central claim fails in the regime where it is made. As a lighter check, compute the Hartree-Fock stability matrix (nematic susceptibility) around the converged isotropic solution: negative eigenvalues at any of these points would invalidate Eq. (71).","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central phase diagrams (Figs. 1 and 2) and the claim that the Dirac sea dominates the magnetic anisotropic energy (Fig. 10) rest on the converged Hartree-Fock density matrix being diagonal in the Landau-level index, with particle-hole mixing only within the same index (Eq. 71). This 'isotropic Dirac fluid' property is what allows the decomposition ρ = ρ0 ⊕ ρDS (Eq. 72) and the attribution of energy differences to the Dirac sea via Eq. (89). The authors explicitly state in Section IV.A that for very strong interactions the isotropic fluid may become nematic, in which case Eq. (71) and the decomposition would fail. The parameter region where the KD state wins and where the Dirac-sea contribution exceeds the zero-LL contribution (κ0 ≳ 0.8, small B, bottom right of Figs. 1 and 2) is exactly the strong-coupling regime where this assumption is most questionable. Because the order-parameter search is restricted to translation-invariant, k-independent states and the ansatz of Eq. (71) is imposed (or at least heavily favored), there is no evidence that the true Hartree-Fock ground state in the predicted KD region is isotropic. If a nematic or other LL-index-mixing state has lower energy, the phase boundaries and the quantitative Dirac-sea decomposition in Fig. 10 would change.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This manuscript proposes a two-step microscopic theory of the ν=0 and ν=±1 quantum Hall states in monolayer graphene. The authors first run a renormalization group from the carbon lattice scale to the magnetic length to obtain renormalized Fermi velocity and short-range valley-sublattice couplings, using bare values from a previous LCAO calculation and literature electron-phonon estimates. They then feed these couplings into self-consistent Hartree-Fock calculations with up to 50 Landau levels. The central results are: (i) at ν=0, a transition from canted antiferromagnetic to Kekulé-distorted (and sublattice-polarized) states as the fine-structure constant grows or the magnetic field falls, in qualitative agreement with the experimental screening dependence; (ii) at ν=±1, a transition from spin-polarized charge-density wave to spin-polarized KD; and (iii) a decomposition of the magnetic anisotropic energy into zero-Landau-level and Dirac-sea parts, with the latter dominant at large κ0. The paper emphasizes that these transitions arise without fine-tuning parameters to the target experiments.","tokens_in":25761,"tokens_out":6919,"duration_ms":71144,"significance":"If correct, the results provide a parameter-free-from-fitting explanation for why STM sees KD order in open-surface devices while transport sees AF order in encapsulated devices, and they make concrete predictions for ν=±1 and for Landau-level coherence visible in STM. The paper should be credited for giving explicit formula-level derivations (self-energies in Appendix C), a convergence check in NLL (Appendix B), and candid statements of the limitations of the RG and of the isotropic-fluid assumption. The main caveat is that the Dirac-sea attribution is built on an assumption whose validity is least tested exactly in the strong-coupling region where that attribution matters.","major_comments":[{"comment":"The 'isotropic Dirac fluid' ansatz is load-bearing. The decomposition ρ = ρ0 ⊕ ρDS (Eq. 72), the energy splitting ϵ = ϵ0 + ϵDS (Eq. 89), and the central Fig. 10 all require the converged density matrix to be diagonal in the Landau-level index. The authors note in the same section that for very strong interactions the isotropic fluid may become nematic, in which case Eq. (71) would fail. The parameter region where the Dirac sea dominates (κ0 ≳ 0.8 at B=10 T in Fig. 10) and where KD is stabilized (bottom right of Figs. 1 and 2) is precisely that strong-coupling regime. The numerical search is restricted to translation-invariant states and, as described, does not attempt states with off-diagonal Landau-level-index coherence; random seeds within the restricted subspace do not test Eq. (71). I request a stability check against nematic LL-index mixing, or a clear statement that the predicted KD region and the Dirac-sea decomposition are conditional on this untested assumption.","section":"Section IV.A, Eq. (71)"},{"comment":"The central claim that the Dirac sea dominates the magnetic anisotropic energy at large κ0 rests on separately converged values of ϵ0 and ϵDS, but Appendix B verifies convergence only of the total energy difference between AF and KD. Since the two components are individually divergent before background subtraction (Appendix C) and are regulated by the Landau-level cutoff, the relative weight of ϵDS in Fig. 10 should be shown to be stable as NLL grows. Please provide the NLL dependence of the separate zero-Landau-level and Dirac-sea energy differences.","section":"Section IV.B, Appendix B, Eq. (89), Fig. 10"},{"comment":"The phase diagrams in Figs. 1 and 2 label the transitions as first-order, yet the RG vertex calculation neglects momentum-dependent vertices, and the authors state in Section II that these neglected terms 'could potentially alter the nature of the transitions from first to second order.' If the first-order label is part of the paper's claims, it is not supported; otherwise the figures should be relabeled or the text should explicitly state that the order of the transition is not determined by the present calculation.","section":"Section II and Section III.B"}],"minor_comments":[{"comment":"The text says 'anisotropic relativistic fluid' when describing the converged solution that satisfies Eq. (71); this should presumably read 'isotropic relativistic fluid'. In addition, Eq. (71) is described as 'diagonal with respect to the Landau level index', but it allows particle-hole mixing with the same index; 'block-diagonal in n' would be clearer.","section":"Section IV.A"},{"comment":"The line 'Next thing to do' followed by two bullet items appears to be an editing remnant and should be removed from the published text.","section":"Before Fig. 2"},{"comment":"There are several grammatical slips, including 'we predict a transitions', 'the Zeorth Landau level', and 'groudnstates'; a careful proofread is needed.","section":"Abstract and throughout"},{"comment":"The condition excluding the (0,0) term is written as 'exclude u = v = 0', which is ambiguous; it should be written as '(u,v) ≠ (0,0)' or equivalent.","section":"Eq. (6)"},{"comment":"The caption calls the quantity an 'absolute energy difference' while the text refers to 'energy difference per particle'; please unify the terminology and specify the units.","section":"Fig. 10"},{"comment":"The statement that the Dirac sea 'contributes to one electron per graphene unit cell' is not defined or elaborated in the main text; please clarify what this statement means.","section":"Abstract"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is a competent and largely self-contained calculation with an honest discussion of its limitations. The main risk is the untested isotropic-fluid assumption in the exact parameter region where the headline Dirac-sea effect appears. I would recommend that the revision include a concrete stability test or a correspondingly cautious rewrite of the central claims. The stray 'Next thing to do' note before Fig. 2 suggests the manuscript was not fully finalized, but this should not affect the scientific assessment."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe short version: this is a serious attempt to go beyond the usual zeroth-Landau-level projection for graphene quantum Hall states. The genuinely new content is the κ0–B phase diagrams for ν=0 and ν=±1 (Figs. 1 and 2) and the explicit decomposition of the magnetic anisotropic energy into zero-Landau-level and Dirac-sea contributions. The claim that the Dirac sea dominates the anisotropy at small screening and small field (Fig. 10) is concrete and in principle testable.\n\nCredit where it's due: the authors are unusually honest about their approximations. They say up front that the neglect of momentum-dependent vertices in the RG could change the order of the transition, that the first-order conclusion is tentative, and in Sec. IV.A that the isotropic Dirac fluid may become nematic when interactions are strong. Appendix B shows the LL-cutoff convergence. The input couplings come from an earlier LCAO calculation and literature electron-phonon values, not from fitting the target experimental phase diagram, so the circularity burden is low.\n\nThe soft spot is the one the stress-test flags. Equation (71), the converged density matrix being diagonal in the Landau-level index, is what legitimizes the ρ0 ⊕ ρDS decomposition and the Dirac-sea energy attribution. But the regime where KD wins and the Dirac sea dominates is the strong-coupling, small-B corner, exactly where the authors suspect Eq. (71) might break down. The order-parameter search is restricted to translation-invariant, k-independent states, so a nematic or other LL-mixing competitor is not ruled out. Without released code or data, I can't independently check whether the ansatz is stable there. That's a real hole, not a manufactured one. It means the phase boundaries in Figs. 1 and 2 are plausible but not yet quantitatively robust.\n\nOne minor presentation issue: the Fig. 2 caption region contains a leftover author note (\"Next thing to do…\"). Sloppy, easily fixed.\n\nBottom line: within the stated model the central argument holds up, and the qualitative reconciliation with STM versus transport experiments is credible. This deserves a serious referee. I'd send it to review, with the expectation that the referees press on Eq. (71) in the KD regime and ask for a reproducibility artifact. I'd cite it for the Dirac-sea decomposition, though with a cautionary flag on the ansatz.","headline":"A credible non-perturbative RG+HF phase diagram for graphene QH states, with the Dirac sea doing real work, but the load-bearing isotropic-fluid ansatz (Eq. 71) needs independent confirmation before trusting the quantitative phase boundaries.","tokens_in":26315,"tokens_out":3480,"would_cite":true,"duration_ms":31484,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Ground states at charge neutrality and at filling ±1 in monolayer graphene switch between magnetic and Kekulé-distorted order as dielectric screening and magnetic field change.","keywords":["monolayer graphene","quantum Hall ferromagnetism","Kekulé distortion","Dirac sea","Landau level mixing","magnetic anisotropic energy","self-consistent Hartree-Fock","renormalization group"],"falsifier":"In an open-surface graphene sample with weak dielectric screening, image the ν=0 quantum Hall state at a field where the paper's phase diagram predicts the Kekulé-distorted phase: observing the honeycomb pattern of the canted antiferromagnet rather than the threefold bond-ordered Kekulé pattern would contradict the central claim.","tokens_in":25209,"feed_emoji":"🧲","tokens_out":9865,"duration_ms":91219,"temperature":0.7,"pith_summary":"The paper tries to explain why scanning tunneling experiments see Kekulé-distorted (bond-ordered) states in graphene under strong magnetic fields while transport experiments on more screened devices see antiferromagnetic order. It claims the deciding physics is the Dirac sea: the filled negative-energy Landau levels renormalize the valley-sublattice interactions and, through first-order exchange, supply a substantial part of the small magnetic anisotropic energy when the dielectric constant is small. The authors combine renormalization group flow from the carbon-lattice scale down to the magnetic length with self-consistent Hartree-Fock calculations over many Landau levels, and obtain phase diagrams in which the $\\nu=0$ ground state changes from a canted antiferromagnet to a spin-singlet Kekulé-distorted state, and the $\\nu=\\pm1$ ground state changes from a spin-polarized charge-density wave to a spin-polarized Kekulé state, as screening and field strength decrease.","feed_headline":"Weak screening flips graphene's ν=0 state to Kekulé order","feed_subtitle":"Dirac-sea calculation says open-surface samples should show Kekulé order where transport devices show antiferromagnetism.","key_machinery":"The central object is the self-consistent Hartree-Fock density matrix of an \"isotropic Dirac fluid\" in a magnetic field, which the paper argues is diagonal in the Landau level index: $\\rho^{vv',ss'}_{n\\xi,n'\\xi'}=0$ for $n\\neq n'$. That block structure allows the density matrix to be decomposed as $\\rho=\\hat{\\rho}_0\\oplus\\rho_{\\mathrm{DS}}$, with the zeroth Landau level a $4\\times4$ spin-valley matrix and the Dirac sea a direct sum of $8\\times8$ matrices parameterized by angles $\\theta_n$ that describe particle-hole mixing within each Landau level. Because the short-range self-energy is independent of $n$ while the Coulomb self-energy decays with $n$, the magnetic anisotropic energy separates into a zeroth-Landau-level piece and a Dirac-sea piece, and the two can be compared directly. The other half of the machinery is the two-step procedure: RG flow of the Fermi velocity and of the four short-range couplings $g_{zz},g_{\\perp z},g_{z\\perp},g_{\\perp\\perp}$ from lattice scale to magnetic length, then nonperturbative Hartree-Fock with dozens of Landau levels.","core_discovery":"On its own terms, the central discovery is a quantitative $(\\kappa_0,B)$ phase diagram—where $\\kappa_0$ is graphene's fine-structure constant, inversely tied to the dielectric screening $\\epsilon_r$—showing a first-order transition from AF/CAF to KD order at $\\nu=0$ and from spin-polarized CDW to spin-polarized KD at $\\nu=\\pm1$ as $\\kappa_0$ grows and $B$ falls. The mechanism is not a single-particle Zeeman or substrate effect. The renormalization group makes the inter-valley, sublattice-flipping coupling $g_{\\perp z}$ increasingly attractive, favoring the Kekulé state, while Landau-level mixing with the Dirac sea changes the Hartree and Fock potentials so that their zeroth-Landau-level cancellation at $\\nu=\\pm1$ is lifted. The paper reports that once $\\kappa_0$ exceeds roughly 0.8, the Dirac sea contributes more to the magnetic anisotropic energy than the zeroth Landau level, so the effect is nonperturbative in character.","pith_inferences":["As an extension, the same two-step scheme should give concrete phase boundaries for bilayer graphene in the $\\nu=5/2$ regime, where the Dirac sea has also been invoked; a test would be whether the predicted boundaries shift with dielectric screening.","The isotropic-fluid Ansatz is the point where the theory has a natural failure mode: if interactions are strong enough for a nematic state with mixing between different Landau-level indices, Eq. (71) and the whole $\\rho_0 \\oplus \\rho_{\\mathrm{DS}}$ energy decomposition would need revision.","A reader could also test the screening dependence directly: the paper's central parameter is the dielectric constant of the environment, so varying the encapsulation material should move the critical field at which KD order appears in the same sample."],"forward_implications":["Open-surface STM devices should predominantly show KD order at low magnetic field, while double-encapsulated devices should remain canted-antiferromagnetic, reconciling the two experimental observations within one parameter set.","At $\\nu=\\pm1$, the ground state should be a spin-polarized charge-density wave over most of the screened phase diagram, with a switch to spin-polarized KD only when screening is weak and the field is small.","Quantitative ground-state predictions that project only onto the zeroth Landau level are missing a comparable part of the magnetic anisotropic energy at small dielectric screening.","The KD phase should appear in STM as a threefold bond-density pattern, with valley phase $\\phi$ distinguishing the Kekulé-O from the symmetry-broken Kekulé variant.","The transition between the ordered states is first order, although the authors mark that conclusion as tentative because momentum dependence of the vertex function is neglected."],"supporting_citations":[{"why":"Supplies the RG flow equations and scheme the paper uses to integrate out states from lattice scale to magnetic length.","marker":"[29]"},{"why":"Provides the microscopic bare coupling constants and the earlier perturbation-theory prediction that KD wins over CAF.","marker":"[27]"},{"why":"Defines the zeroth-Landau-level (u⊥, uz) phase diagram that the paper goes beyond with full Landau-level mixing.","marker":"[30]"},{"why":"Reports the STM observations of Kekulé-distorted ordering that this paper aims to explain.","marker":"[17–19]"},{"why":"Reports transport evidence for canted-antiferromagnetic and sublattice-polarized phases in double-encapsulated devices, the discrepancy this work resolves.","marker":"[22–26]"},{"why":"Establishes the isotropic Dirac fluid property and the background-subtraction scheme used to define Dirac sea energies.","marker":"[50]"},{"why":"Gives the RPA/RG calculation of the renormalized Fermi velocity and quasiparticle residue used in the first step.","marker":"[38]"},{"why":"Show that projecting onto the zeroth Landau level yields four coupling parameters rather than two, motivating the nonperturbative treatment.","marker":"[34,35]"}],"fun_headline_variants":["Dirac sea drives graphene to Kekulé order at low screening","Dirac sea flips graphene's AF state to Kekulé as screening drops","Nonperturbative Dirac sea steers graphene phase transitions","Strong B reveals Dirac sea's decisive role in graphene order","Dirac sea tilts graphene's phase diagram toward Kekulé order"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The calculation assumes the quantum Hall ground state is an isotropic Dirac fluid whose density matrix is block-diagonal in the Landau level index, so particle and hole states mix only within the same index; if the true state mixes different Landau level indices, the phase boundaries and the Dirac sea energy contributions would change.","fun_headline_variants_meta":{"raw":{"variants":["Dirac sea drives graphene to Kekulé order at low screening","Dirac sea flips graphene's AF state to Kekulé as screening drops","Nonperturbative Dirac sea steers graphene phase transitions","Strong B reveals Dirac sea's decisive role in graphene order","Dirac sea tilts graphene's phase diagram toward Kekulé order"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001184,"raw_usage":{"total_tokens":4953,"prompt_tokens":1074,"completion_tokens":3879,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":690,"completion_tokens_details":{"reasoning_tokens":3785}},"tokens_in":690,"tokens_out":3879,"duration_ms":27507,"temperature":1.0,"reasoning_tokens":3785,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T12:40:04.075645+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"In an open-surface graphene sample with weak dielectric screening, image the ν=0 quantum Hall state at a field where the paper's phase diagram predicts the Kekulé-distorted phase: observing the honeycomb pattern of the canted antiferromagnet rather than the threefold bond-ordered Kekulé pattern would contradict the central claim.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the RG flow equations and scheme the paper uses to integrate out states from lattice scale to magnetic length."},{"cited_title":"Absence of heat flow in {\\nu} = 0 quantum Hall ferromagnet in bilayer graphene","cited_arxiv_id":"2409.09663","evidence_quote":"Provides the microscopic bare coupling constants and the earlier perturbation-theory prediction that KD wins over CAF."},{"cited_title":"Kharitonov, Physical Review B85, 155439 (2012)","cited_arxiv_id":null,"evidence_quote":"Defines the zeroth-Landau-level (u⊥, uz) phase diagram that the paper goes beyond with full Landau-level mixing."},{"cited_title":"Zibrov, E","cited_arxiv_id":null,"evidence_quote":"Establishes the isotropic Dirac fluid property and the background-subtraction scheme used to define Dirac sea energies."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the RPA/RG calculation of the renormalized Fermi velocity and quasiparticle residue used in the first step."}],"review_version":1}