{"id":"89130ce5-60f5-48e9-b9e0-f2b501460a61","arxiv_id":"1908.08723","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Intersite Coulomb repulsion V drives d3+d1 valence-skipping charge order in a three-orbital Hund's metal, and the critical V is lowered inside the spin-freezing crossover regime.","lead":"This paper simulates a three-orbital electron model and shows that repulsion between neighboring sites drives a 'valence-skipping' checkerboard charge order. The order appears at much lower interaction strength inside the spin-freezing crossover regime of Hund's metals, a result relevant to iron-pnictide superconductors.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Density-density-only impurity polarizability may miss spin-flip vertex corrections that could shift the Vc downturn in the spin-freezing regime.","rationale":"The reader's weakest_assumption identifies the density-density truncation of the impurity two-particle correlations as the key soft spot. My analysis agrees: this is the single most load-bearing concern because the quantitative deviation from the atomic limit—the Vc downturn—is obtained from a charge susceptibility that is built from a density-density-only impurity polarizability. The paper explicitly acknowledges this restriction and justifies it by citing that non-monopole screening is weak, but the spin-freezing regime is precisely where spin fluctuations are strong and the coupling between spin and charge channels could matter. Without a full-vertex benchmark or at least a test at one representative parameter point, the possibility that the enhancement is an artifact remains open. Other concerns (no error bars, no deposited code, self-referential wavefunction model) are secondary and do not independently threaten the central mechanism. Given this, the verdict stays CONDITIONAL: the concern is real but not yet demonstrated to break the claim, and a targeted numerical check could settle it. No change to the reader's verdict is needed.","tokens_in":17316,"tokens_out":4650,"duration_ms":45705,"concrete_test":"At the representative point U=4, J/U=0.2, V=0.9Vc, repeat the GW+EDMFT calculation using an impurity solver that measures the full two-particle Green's function (all spin-flip and pair-hopping components) and feeds the resulting full local polarizability into Eq. (S10), instead of the density-density-only χimp. If the (π,π) divergence of χ(k,iν0) still occurs at a Vc within, say, 10% of the density-density result, the truncation is not responsible for the downturn. If Vc shifts by more than 20% or the divergence disappears, the central claim is an artifact of the density-density approximation.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that Vc drops below the atomic-limit estimate U/4(1-3J/U) in the spin-freezing regime rests on the divergence of the static charge susceptibility χ(k,iν0) at (π,π) computed from GW+EDMFT. In the Methods, the authors state that only density-density two-particle correlation functions are measured from the impurity, χimp(τ)=⟨Tτ n_{γσ}(τ)n_{γ'σ'}(0)⟩, and that non-density-density channels are neglected because they are 'ill-screened' (ref. [55]). This justification is plausible for the monopole part of the charge response, but the spin-freezing regime is defined by large dynamical spin fluctuations; in the Kanamori model the spin-flip and pair-hopping terms couple the spin and charge channels through the local vertex. By truncating the impurity polarizability to density-density components, the calculation omits vertex corrections that could either enhance or suppress the effective nonlocal interaction in the charge channel. Since the Vc downturn is seen only at J/U=0.2 and U≥3, exactly where spin fluctuations are strongest, the possibility that the downturn is an artifact of this truncation cannot be excluded from the data presented. No convergence test or full-vertex benchmark is shown. This is the weakest link in the chain: the atomic-limit Vc formula is exact for t=0, and the multiplet-population argument is consistent, but the actual GW+EDMFT Vc values that deviate from atomic limit are obtained with the restricted susceptibility. If the neglected channels shift the charge susceptibility divergence significantly, the headline conclusion (spin-freezing enhancement of valence-skipping CO) would not survive.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript studies a three-orbital Kanamori model on a square lattice at 1/3 filling (two electrons per site) with a nearest-neighbor Coulomb interaction V. The authors first derive the atomic-limit phase diagram, finding a valence-skipping d3+d1 charge-ordered state above Vc = U/4 (1 - 3J/U). They then perform self-consistent GW+EDMFT calculations, identifying the charge order transition from the divergence of the static charge susceptibility at (π,π). The central numerical claim is that at J/U = 0.2 and U ≥ 3, the critical V drops significantly below the atomic-limit estimate, with the strongest downturn at U = 4; this is attributed to the spin-freezing crossover regime, where maximum-spin multiplet states dominate and the populations of one- and three-electron states become nearly equal. A phenomenological wavefunction with these dominant states is used to rationalize the enhancement of charge order.","tokens_in":17632,"tokens_out":4727,"duration_ms":50054,"significance":"If the numerical result is correct, the paper establishes a concrete mechanism by which Hund's coupling and the spin-freezing crossover enhance the propensity to valence-skipping charge order in multiorbital systems, which is of direct relevance to Hund's metals and the charge-ordered iron-pnictide superconductors. The atomic-limit calculation is exact and provides a clean benchmark (Vc = U/4(1 - 3J/U)). The study also carefully distinguishes GW+EDMFT, EDMFT, and GW results, and the analysis of the multiplet population profile is physically transparent. However, the headline result rests entirely on GW+EDMFT with a density-density-only impurity polarizability and is not backed by error bars, convergence tests, or independent benchmarks; the plausibility argument for the truncation is not sufficient to exclude a vertex-correction artifact in exactly the regime where the new effect appears.","major_comments":[{"comment":"The statement 'we measured only the density-density type of two-particle correlation functions from the impurity' is the weakest load-bearing point of the paper. The central claim that Vc is lowered below the atomic-limit estimate U/4(1 - 3J/U) at J/U = 0.2 and U ≥ 3 (Fig. 2(a)) is obtained from the divergence of the static charge susceptibility computed with this restricted polarizability. The spin-freezing regime is precisely where dynamical spin fluctuations are strongest, and in the Kanamori model the spin-flip and pair-hopping terms couple spin and charge channels through the local vertex. The cited justification (ref. [55]) that non-monopole charge terms are ill-screened is plausible for long-range screening but does not by itself rule out sizeable local vertex corrections to the charge response at these parameters. Please provide numerical evidence — for example a benchmark with the full local vertex in the single-orbital limit, a small-cluster exact calculation, or a direct estimate of the neglected spin-flip contribution to the charge vertex at U = 4, J/U = 0.2 — that the Vc downturn is not an artifact of this truncation.","section":"Methods, first paragraph"},{"comment":"No statistical or systematic error estimates are reported for Vc, χ(k,iν0), α, or Γ. Given that the central result is a quantitative deviation of Vc from the atomic-limit line (Fig. 2(a), J/U = 0.20), the reader cannot judge whether the downturn at U = 4 is significant or within Monte Carlo noise. Please report CTQMC error bars on the charge susceptibility and on the inferred Vc, and convergence checks with respect to the 32×32 k-grid, the number of Matsubara frequencies used in the GW summations, the mixing parameter Rmix, and the frequency range used to extract α and Γ.","section":"Fig. 2(a) and Methods (Supplemental Note 2)"},{"comment":"The wavefunction ψ = √p1|1,1,1/2⟩ + √(1-2p1)|2,1,1⟩ + √p1|3,0,3/2⟩ imposes p3 = p1 by construction and keeps only maximum-spin states. The Vc(p1) curve in Fig. 5(d) is built from populations measured in the same GW+EDMFT calculation, so it is a consistency check rather than an independent confirmation of the mechanism. The sentence 'This result confirms the role of maximum S states in N = 3 subspace in enhancing the CO instability' therefore overstates the evidential weight. Please rephrase this as an interpretive consistency check, or provide an independent test — for example, artificially suppressing the spin-flip and pair-hopping terms and showing that the Vc downturn disappears.","section":"Results and Discussion, Fig. 5(d)"}],"minor_comments":[{"comment":"The summation limits for the terms with γ ≠ γ′ and γ < γ′ are not specified; presumably they run over 1 ≤ γ, γ′ ≤ 3. Please make the index ranges explicit.","section":"Eq. (2)"},{"comment":"The horizontal axis label in Fig. 4(a) appears garbled ('Δχs/χs' rendered as '_6r/r(ii0)' in the available figure). Please ensure all axis labels and legends are legible and correctly typeset.","section":"Fig. 4(a) and caption"},{"comment":"The phrase '54C18∼O(10^13) manipulations' is unclear; if it refers to a binomial coefficient, please write it unambiguously (e.g., '54 choose 18 ≈ 10^13') and explain the counting.","section":"Supplemental Note 1"},{"comment":"The identification of the spin-freezing regime uses 0.4 ≲ α ≲ 0.5 and Γ ≈ 0, with α and Γ obtained from a fit to only the three lowest Matsubara frequencies. Please provide the fit uncertainty or a sensitivity check with respect to the number of frequencies included.","section":"Methods, last paragraph"},{"comment":"The abstract states that the transition 'is shown to be driven by V' and the enhancement is 'significantly enhanced'; given the unresolved truncation concern, a more cautious phrasing (e.g., 'we find evidence that') would better match the level of numerical support.","section":"Abstract and Conclusions"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is technically serious and the atomic-limit part is solid, but the headline GW+EDMFT result depends on an unbenchmarked approximation in precisely the parameter region where the new physics appears. I would ask the editor to require the authors to provide either a full-vertex benchmark or a direct estimate of the neglected spin-flip/pair-hopping contribution before publication. The data and code are only 'available upon reasonable request'; a concrete data-availability statement would also strengthen reproducibility."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nHere's my take on 1908.08723. The headline is real: in a three-orbital Kanamori model at 1/3 filling, nonlocal repulsion V drives a d3+d1 valence-skipping charge order, and the critical V is reduced below the atomic-limit estimate precisely in the spin-freezing crossover regime. That combination is new. The atomic-limit Vc = U/4(1-3J/U) is straightforward and correct; the GW+EDMFT downturn at J/U=0.2 and U>=3 is not in the prior work, and the multiplet-population analysis (p1≈p3, maximum spin states) gives a plausible local picture for why the charge instability is enhanced.\n\nWhat I like: the calculation is careful, with GW+EDMFT, EDMFT, and GW compared side by side; the susceptibility divergence is shown to sit at (π,π); the self-energy analysis locates the spin-freezing crossover cleanly; and the paper is honest about what is and isn't included. The atomic-limit phase diagram is a nice anchor for the numerics. The citation pattern is also fine: prior valence-skipping work by Strand and Isidori is credited, and the claim that this route is distinct from the anisotropic orbital-multipole scattering mechanism is accurate.\n\nThe soft spots are real but not fatal. The biggest one is the restriction to density-density two-particle correlation functions in the impurity solver. The stress-test concern—that spin-flip and pair-hopping vertex corrections could shift the charge response in the spin-freezing regime—is plausible. The authors defend the truncation by citing the ill-screening of non-monopole channels; that is a reasonable argument but it is not a benchmark. Since the Vc downturn appears exactly where spin fluctuations are strongest, a referee should ask for a full-vertex calculation in at least one representative point, or a documented test of the truncation. The absence of error bars and convergence tests is minor but worth noting; CTQMC statistical errors are likely small, but showing them would help.\n\nThe p1-based wavefunction estimate is a consistency check, not an independent confirmation; the paper doesn't oversell it, so I don't hold that against it. Also, the calculation is paramagnetic and isotropic, so ordered magnetic states and orbital order are out of scope; that's stated.\n\nOverall, the central claim holds up as a plausible mechanism. It deserves peer review. I'd send it to a journal and ask for a revision that addresses the density-density truncation and adds convergence data.\n\nYours,","headline":"A solid GW+EDMFT study that makes a plausible case for nonlocal-V-driven valence-skipping charge order with spin-freezing enhancement, worth refereeing with a request for full-vertex and convergence checks.","tokens_in":18203,"tokens_out":2859,"would_cite":true,"duration_ms":28310,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper shows that intersite Coulomb repulsion V drives valence-skipping d3+d1 charge order in a three-orbital Hund metal, and that spin-freezing lowers the critical V below the atomic-limit estimate.","keywords":["valence-skipping charge order","Hund's metal","spin-freezing crossover","GW+EDMFT","nonlocal Coulomb interaction","Kanamori model","charge susceptibility","iron pnictides"],"falsifier":"Recompute the charge susceptibility at U = 4, J/U = 0.2 with the impurity polarizability built from all two-particle vertex components, including spin-flip and pair-hopping channels, and see whether χ((π,π),0) still diverges at a Vc below the atomic-limit line. If the downturn vanishes, the spin-freezing enhancement is an artifact of the density-density-only measurement; if it survives, the mechanism is robust.","tokens_in":17050,"feed_emoji":"⚡","tokens_out":7519,"duration_ms":70720,"temperature":0.7,"pith_summary":"This paper asks what makes a multiorbital metal at one-third filling, with two electrons per site, develop a valence-skipping charge order in which neighboring sites carry three and one electrons while the two-electron state is skipped. Using GW plus extended dynamical mean-field theory on a three-orbital Kanamori model with intersite Coulomb repulsion V, the authors show that the transition to this d3+d1 order is driven by V. The central result is that in the spin-freezing crossover regime, reached at strong Hund's coupling and intermediate-to-large U, the critical V drops well below the atomic-limit estimate Vc = U/4(1-3J/U). The mechanism is the local multiplet population: maximum-spin states dominate each charge sector, and the one- and three-electron maximum-spin states become nearly equally populated, making the charge-ordered state energetically cheap. If the paper is right, nonlocal Coulomb interactions are a generic route to valence-skipping order in Hund metals, distinct from previously proposed single-site mechanisms.","feed_headline":"Spin freezing lowers the threshold for valence-skipping charge order","feed_subtitle":"Critical nonlocal repulsion falls below the atomic-limit estimate in the spin-freezing regime of a Hund's metal.","key_machinery":"The load-bearing diagnostic is the static charge susceptibility χ(k,iν0): the d3+d1 transition is identified where this quantity diverges at the (π,π) wave vector as V grows. The explanatory mechanism is the local multiplet population profile of the impurity, classified by charge N, orbital L, and spin S as |N,L,S⟩; in the spin-freezing regime the maximum-spin states |1,1,1/2⟩, |2,1,1⟩, and |3,0,3/2⟩ dominate, with p3 ≈ p1, so the local Hilbert space itself is nearly an equal mixture of the one- and three-electron valences that the ordered phase requires. A phenomenological wavefunction built from these maximum-spin states reproduces the Vc downturn qualitatively, while the full calculation uses GW+EDMFT, a method in which a local impurity model with a dynamically screened interaction is solved self-consistently together with nonlocal GW corrections.","core_discovery":"At one-third filling of a three-orbital square-lattice Kanamori model, the paper establishes a d3+d1 valence-skipping charge-ordered phase with ordering wave vector (π,π) whose boundary is controlled by the nonlocal Coulomb repulsion V. In the atomic limit the boundary is Vc = U/4(1-3J/U), but the GW+EDMFT calculation finds that for J/U = 0.2 and U ≥ 3, in the spin-freezing crossover regime, the instability is significantly enhanced and Vc falls below this estimate, with the most pronounced downturn at U = 4 and a rapid upturn at U = 5 as the system enters the frozen-moment regime. The enhancement is tied to the impurity multiplet populations: maximum-spin states dominate in each charge subspace with substantial one- and three-electron weights, p3 ≈ p1, so the local state is already nearly an equal superposition of the two valences that the charge order requires. This route to valence-skipping is presented as distinct from the anisotropic orbital-multipole scattering mechanism proposed earlier.","pith_inferences":["If the spin-freezing enhancement is generic, real Hund metals with appreciable nonlocal Coulomb interactions may have an intrinsic tendency toward valence-skipping charge fluctuations even without an effective negative U; this could connect to the charge order observed in iron-pnictide superconductors.","The p3 ≈ p1 condition could be used as a local diagnostic for charge-order propensity: materials whose multiplet populations show nearly equal one- and three-electron maximum-spin weights should be checked for d3+d1 order.","A direct testable extension would be to include spin-flip and pair-hopping two-particle vertices in the impurity polarizability; if those channels appreciably screen the nonlocal interaction, the Vc lowering would shrink, which would separate the spin-freezing population effect from a truncation artifact.","The rapid upturn at U = 5 suggests an optimal correlation strength for valence-skipping order, so tuning through the spin-freezing-to-frozen-moment crossover via pressure or doping should make charge-order propensity peak and then fall."],"forward_implications":["In the spin-freezing crossover, a modest nonlocal repulsion below U/4 can nucleate valence-skipping charge order in a three-orbital metal.","EDMFT alone or GW alone does not capture the downturn of Vc, so the coupling between local spin fluctuations and nonlocal screening is essential for the effect.","The downturn should be absent for weak Hund's coupling, approximately J/U ≤ 0.15, and should disappear as the system enters the frozen-moment regime at larger U, where the two-electron population dominates.","Even at V = 0, the system near U = 4 and J/U = 0.2 already has substantial nonlocal charge susceptibility, indicating proximity to charge order before the intersite interaction is switched on.","The mechanism is different from anisotropic orbital-multipole scattering and relies on the local populations of maximum-spin one- and three-electron states."],"supporting_citations":[{"why":"Supplies the self-consistent GW+EDMFT scheme whose impurity-plus-nonlocal-GW action is used for all the method comparisons in the paper.","marker":"[49]"},{"why":"Defines the Hund's metal regime, the spin-freezing crossover, and the atomic multiplet classification used to analyze the results.","marker":"[3]"},{"why":"Characterizes spin freezing and the non-Fermi-liquid self-energy power law with parameters α and Γ used to identify the regime.","marker":"[8]"},{"why":"Supplies the Δχs/χs frozen-moment diagnostic and the association of spin freezing with emerging magnetic moments.","marker":"[25]"},{"why":"Prior slave-boson identification of the d3+d1 and 2d3+d0 phases in the Kanamori model, which the present work places in a nonlocal-V phase diagram.","marker":"[27]"},{"why":"Establishes valence-skipping and negative-U from local orbital-multipole scattering, the alternative route this paper distinguishes from.","marker":"[28]"},{"why":"Atomic-limit charge-order boundaries for the extended Hubbard model that anchor the Vc reference line.","marker":"[39, 40]"},{"why":"Shows that interaction energy dominates over kinetic energy in setting the charge-order boundary, explaining why atomic-limit estimates survive in many GW+EDMFT regimes.","marker":"[43]"},{"why":"Multitier GW+EDMFT discussion justifying the density-density two-particle vertex measurement, since non-density-density channels are argued to be poorly screened.","marker":"[55]"}],"fun_headline_variants":["Nonlocal repulsion plus spin freezing triggers valence-skipping order","Valence-skipping charge order aided by spin-freezing crossover","Critical V drops in spin-freezing regime for charge order","Spin-freezing crossover boosts valence-skipping charge order","How spin freezing aids valence-skipping charge order"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The calculation measures only the density-density part of the two-particle response and stays in the paramagnetic isotropic phase; if the omitted spin-flip and pair-hopping channels significantly screen the nonlocal interaction, the spin-freezing-induced drop in critical V could disappear.","fun_headline_variants_meta":{"raw":{"variants":["Nonlocal repulsion plus spin freezing triggers valence-skipping order","Valence-skipping charge order aided by spin-freezing crossover","Critical V drops in spin-freezing regime for charge order","Spin-freezing crossover boosts valence-skipping charge order","How spin freezing aids valence-skipping charge order"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000386,"raw_usage":{"total_tokens":2064,"prompt_tokens":993,"completion_tokens":1071,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":609,"completion_tokens_details":{"reasoning_tokens":987}},"tokens_in":609,"tokens_out":1071,"duration_ms":8250,"temperature":1.0,"reasoning_tokens":987,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T11:31:42.218471+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Recompute the charge susceptibility at U = 4, J/U = 0.2 with the impurity polarizability built from all two-particle vertex components, including spin-flip and pair-hopping channels, and see whether χ((π,π),0) still diverges at a Vc below the atomic-limit line. If the downturn vanishes, the spin-freezing enhancement is an artifact of the density-density-only measurement; if it survives, the mechanism is robust.","supporting_citations":[{"cited_title":"& Mravlje, J","cited_arxiv_id":null,"evidence_quote":"Defines the Hund's metal regime, the spin-freezing crossover, and the atomic multiplet classification used to analyze the results."},{"cited_title":"& Werner, P","cited_arxiv_id":null,"evidence_quote":"Supplies the Δχs/χs frozen-moment diagnostic and the association of spin freezing with emerging magnetic moments."}],"review_version":1}