{"id":"58e4d5ac-26bf-45db-8363-831a717a6c55","arxiv_id":"1908.11361","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"In the nematic state of FeSe, B1g Raman intensity is suppressed at low frequencies because the orbital content of the Fermi pockets becomes nearly mono-orbital, and charge-conservation vertex corrections cancel the resulting s-wave component of the Raman vertex.","lead":"This theory paper explains the puzzling drop in a specific Raman signal in the nematic metal FeSe as a consequence of the electron pockets becoming nearly mono-orbital, which lets ordinary charge-conservation vertex corrections suppress the signal. A smart generalist would read it because it turns a seemingly gap-like spectrum into a diagnostic of orbital reconstruction in metals without a gap.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The predicted B1g suppression requires the physical orbital composition to follow the bare eigenstates of Eq. (2); the paper itself concedes that orbital-selective spectral weight would keep a d-wave vertex and eliminate the drop, so the central claim is conditional on a disputed renormalization.","rationale":"The reader's conditional verdict identifies the correct soft spot. The entire suppression mechanism is driven by Γ_d^2 → 0 in Eq. (5), and Γ_d is computed from the bare eigenstates of the two-orbital Hamiltonian in Eq. (2), with no orbital-dependent quasiparticle weight renormalization. The paper itself states in the Summary and Discussion that under the orbital-selective spectral-weight scenario of Refs. [11,19,48-50], the outer pocket would not become mono-orbital, the B1g vertex would retain its d-wave form, and the predicted suppression would not occur. That is not a hidden inconsistency; it is an uncomputed dependence on a disputed external input. The cited polarized ARPES data [14,16] support strong orbital content change but are surface-sensitive and do not by themselves determine the low-energy Z factors that control the physical Raman vertex. I considered whether a more internal flaw exists, such as whether the s-wave component of the B1g vertex is truly the conserved density in a multiband system. In the effective low-energy sector used for the numerical calculation, the inner hole pocket is below the Fermi level and the outer pocket is the only active band, so the single-band Ward-identity cancellation of the angle-averaged vertex is defensible. The analytical part is internally consistent, and the frequency-dependent numerical treatment follows the same logic. Therefore the appropriate action is to keep the reader's CONDITIONAL verdict: the mechanism is coherent and the derivation from the model is sound, but the central claim is conditional on the absence of strong orbital-selective spectral weight. A direct calculation of R_B1g with Z factors, as proposed in the concrete test, would settle whether the concern actually lands.","tokens_in":46156,"tokens_out":16794,"duration_ms":184859,"concrete_test":"Compute the B1g vertex and Γ_d using the same band parameters as Table I but with orbital-dependent quasiparticle weights Z_xz and Z_yz representative of the orbital-selective scenario (e.g., from Refs. [11,19] or a slave-spin/DMFT calculation for FeSe, with Z_yz substantially larger than Z_xz on the outer hole pocket). Replace the bare weight factors by the physical spectral weights Z_xz u_k^2 and Z_yz |v_k|^2 in Eqs. (6)-(7), and re-evaluate R_B1g from Eq. (5). If Γ_d remains O(1) at Ω ≲ Δ_h and the low-frequency drop in R_B1g disappears, the central claim fails in that scenario; if Γ_d remains small, the mechanism survives orbital selectivity.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The gap-like drop in Eqs. (4)-(5) is controlled entirely by Γ_d^2, the non-constant part of the band-basis B1g vertex cos2θ̄_k=(h3−Δh)/h(k), computed from the bare eigenstates of Hamiltonian (2). This identification assumes there is no orbital-dependent quasiparticle weight renormalization. If dxz and dyz carry different Z factors, the physical intraband vertex is Z_xz u_k^2 − Z_yz |v_k|^2 rather than u_k^2 − |v_k|^2, and the physical minority-orbital spectral weight is Z_yz |v_k|^2 rather than |v_k|^2. In the orbital-selective scenario of Refs. [11,19,48–50], even when the bare eigenvector is nearly pure dxz, the minority dyz spectral weight need not be small; the B1g vertex then retains a d-wave component and Γ_d stays of order one. The manuscript explicitly recognizes this in the Summary and Discussion: under that scenario the outer pocket would not become mono-orbital, the vertex would retain d-wave form, and the Raman response would remain largely unchanged, in disagreement with the data. Thus the central claim is not robust to a specific competing renormalization that is actively advocated for FeSe; it is a condition imposed by the model, not a consequence established within the model.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper addresses the observed rapid decrease of the low-frequency B1g Raman response in FeSe below the nematic transition at T_n ~ 85 K. The authors argue that this drop does not signal a gap, but rather a change in the orbital composition of the Fermi pockets: in the nematic phase the outer hole pocket becomes nearly mono-orbital (mostly dxz). The B1g Raman vertex, which in the tetragonal phase is purely d-wave (cos 2θ), acquires an angle-independent s-wave component in the band basis. Because the s-wave component of the Raman susceptibility is cancelled by impurity-scattering vertex corrections that enforce charge conservation, the remaining response is proportional to Γ_d^2, the squared d-wave part of the vertex, which becomes small as the pocket becomes mono-orbital. The authors support this with an analytical derivation (Eqs. (4)-(5)) and with numerical calculations using ARPES-derived band parameters, showing a gap-like suppression that recovers at higher frequencies. They also argue that the effect is inconsistent with orbital-selective spectral-weight scenarios.","tokens_in":46500,"tokens_out":11841,"duration_ms":110592,"significance":"If correct, the paper offers a new and falsifiable explanation for a long-standing puzzle in FeSe, connecting Raman spectroscopy to orbital reconstruction. The analytical derivation is transparent and the numerical implementation uses frequency-dependent vertices rather than a constant fit. The paper makes a concrete prediction: the suppression is controlled by Γ_d^2, and it would not occur if orbital-selective quasiparticle weights strongly renormalize the vertex. This gives a bulk spectroscopic test for orbital selectivity. The paper is clearly written and the main calculation is self-contained, with the supplementary providing the diagrammatic details.","major_comments":[{"comment":"The central prediction of a suppressed B1g response relies on the orbital content of the pockets being given by the non-interacting eigenvectors of Eq. (2), with no orbital-dependent quasiparticle weight renormalization. As the paper acknowledges in the Summary and Discussion, if the orbital-selective spectral-weight scenario of Refs. [11,19,48–50] is realized, the physical intraband vertex becomes Z_xz u_k^2 − Z_yz |v_k|^2 rather than u_k^2 − |v_k|^2, and the minority yz spectral weight need not be small even when the bare eigenvector is nearly pure xz. The statement that the orbital-selective scenario would leave the Raman response unchanged 'in disagreement with the data' is qualitative; to make the central claim robust, the authors should quantify how large the Z_yz/Z_xz anisotropy must be to eliminate the predicted Γ_d^2 suppression and discuss whether the polarized ARPES data of Refs. [14,16] constrain this anisotropy. Without this, the mechanism is conditional on a disputed renormalization.","section":"Summary and Discussion; §The Raman response in the nematic phase (Eq. (5))"},{"comment":"The numerical R_B1g presented in Fig. 3 includes only the hole pockets at Γ; the electron pockets are dismissed with the remark that they can be analyzed along the same lines, but no calculation or estimate is provided. The measured B1g Raman response is the sum over all Fermi pockets, and if the electron-pocket contribution is not suppressed to the same degree, the total drop would be weaker than shown. Since the paper claims 'full agreement' with the experimental data of Refs. [22,24], the authors should either compute the electron-pocket contribution explicitly or provide a quantitative argument for its neglect (e.g., relative spectral weight or a similar mono-orbital suppression).","section":"Numerical calculations (Fig. 3) and Summary"}],"minor_comments":[{"comment":"The term 'gap-like' is used throughout, but Eq. (5) shows R_B1g(Ω) → 0 as Ω → 0 already in the tetragonal phase; the nematic effect is a suppression by Γ_d^2, not the appearance of a true excitation gap. A brief clarification in the introduction would avoid confusion.","section":"Introduction and Eq. (5)"},{"comment":"The recovery to the tetragonal value is shown as a function of Ω/Δ_h0, but no experimental data are overlaid. Given the claim of full agreement, plotting the data of Refs. [22,24] on the same axis would make the comparison quantitative.","section":"Fig. 3"},{"comment":"The symbol 'Ø' appears in place of Ω in several places; this should be corrected to avoid ambiguity.","section":"Supplementary, Eq. (27) and surrounding text"},{"comment":"The sentence 'The result can be straightforwardly extended to the case when the damping rate has both s-wave and d-wave components, γ_s and γ_d... γ = γ_s − γ_d' would benefit from a derivation, as the sign convention for γ_d is not obvious.","section":"After Eq. (5)"}],"recommendation":"major_revision","confidential_remarks":"The paper is well suited for the journal and likely to attract attention. The main concern is that the central claim is conditional on the absence of orbital-selective quasiparticle renormalization, a scenario actively debated for FeSe; the authors state this limitation but do not quantify it. The missing electron-pocket contribution weakens the quantitative comparison with experiment. I believe these issues are addressable and recommend major revision."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThis is a good paper and worth a serious referee. The new physics is the claim that the gap-like drop of the B1g Raman response in nematic FeSe comes from the orbital content of the pockets changing so drastically that the B1g vertex acquires a large s-wave component, which is then killed by the same vertex corrections that enforce charge conservation. What remains is the small d-wave piece Γ_d, so R_B1g drops. That is distinct from the earlier damping-reduction explanation, and the formal derivation in the main text plus the supplementary ladder summation is coherent. I particularly appreciate that Δ_h and the band parameters come from ARPES-derived inputs, not from fitting the target Raman data; the circularity burden is genuinely low.\n\nThe weak point is real but openly admitted. The prediction only goes through if the physical orbital composition of the outer hole pocket follows the bare eigenstates of Eq. (2). If dxz and dyz carry different quasiparticle weights, the minority orbital retains spectral weight, the B1g vertex keeps its d-wave form, and the drop does not happen. The authors state this in the Summary and Discussion when they say their results are inconsistent with the orbital-selective spectral weight scenario. So the central claim is conditional on a disputed renormalization, not a consequence that follows within the model. That does not sink the paper, but it means the title-level claim is stronger than what is established. A careful referee should ask for a computation with orbital-dependent Z factors to see how robust the suppression actually is, or at least for a clear statement of what experimental fingerprint would distinguish the two scenarios beyond the Raman drop itself.\n\nThe more minor soft spots are the qualitative comparison to the measured spectra—Figure 3 shows the right shape but there is no direct overlay with data from Refs. [22,24]—and the simplifying assumptions of isotropic impurity scattering and a constant Γ_s at low frequency. Those are not damaging; the authors do include frequency-dependent vertices in the numerics.\n\nBottom line: this deserves peer review. It gives the FeSe Raman community a concrete, falsifiable mechanism and it frames a real dispute with the orbital-selective scenario. I would encourage the editor to send it out, and I would want the referee reports to press on the Z-factor sensitivity.","headline":"A genuinely new mechanism for the B1g Raman drop in FeSe—orbital transmutation plus charge-conservation vertex corrections—with one main caveat: it presumes the bare eigenstate orbital composition, a condition the authors themselves flag.","tokens_in":47004,"tokens_out":2347,"would_cite":true,"duration_ms":24236,"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":"The gap-like drop in FeSe's $B_{1g}$ Raman response comes from the Fermi pockets becoming nearly mono-orbital.","keywords":["FeSe","nematic order","B1g Raman response","orbital reconstruction","mono-orbital pockets","vertex corrections","charge conservation","iron-based superconductors"],"falsifier":"A decisive test is to compute $\\Gamma_d^2(\\Omega)$ from Eqs. (6)-(7) using independently measured $\\Delta_h(T)$ and band parameters, and compare the depth and temperature dependence of the measured low-frequency $B_{1g}$ drop with the predicted $\\Gamma_d^2$ scaling; the mechanism fails if the drop does not follow that scaling while the damping rate stays constant.","tokens_in":45954,"feed_emoji":"🔬","tokens_out":14710,"duration_ms":118740,"temperature":0.7,"pith_summary":"Below the 85 K nematic transition, FeSe's $B_{1g}$ Raman response drops at low frequencies, a behavior usually read as a sign of a gap. The paper argues that FeSe remains a metal and no gap opens: the drop comes from the nematic order changing the orbital composition of the Fermi pockets. Deep in the nematic phase the outer hole pocket becomes almost entirely $d_{xz}$, so the $B_{1g}$ Raman form factor loses its pure $d$-wave angular dependence and develops an angle-independent ($s$-wave) piece. That $s$-wave piece is eliminated by the same vertex corrections that enforce charge conservation, leaving a strongly reduced response with the same functional form as above $T_n$. If the argument is right, Raman spectroscopy becomes a bulk probe of orbital reconstruction in FeSe.","feed_headline":"A pocket of one orbital explains FeSe's Raman drop","feed_subtitle":"If true, Raman becomes a bulk probe of orbital reconstruction and the 'gap' in a metal dissolves","key_machinery":"The load-bearing object is the orbital-to-band unitary transformation of the two-orbital Hamiltonian for the hole pockets at $\\Gamma$, and in particular the angular average of $\\cos 2\\bar\\theta_k$ -- the $B_{1g}$ Raman vertex in the band basis -- decomposed as $\\cos 2\\bar\\theta_k = \\Gamma_s + \\Gamma_d\\cos 2\\theta$. In the tetragonal phase $\\Gamma_s = 0$, and impurity vertex corrections vanish because $\\int d\\theta\\, \\cos 2\\theta = 0$. In the nematic phase $\\Gamma_s \\neq 0$, and the ladder summation of vertex corrections cancels the $s$-wave piece, leaving $\\chi_{B1g}(\\Omega) = N_F \\Gamma_d^2\\, 2i\\gamma/(\\Omega + 2i\\gamma)$. The suppression follows because $\\Gamma_d \\sim 1/\\lambda_F$ once the pocket is nearly mono-orbital.","core_discovery":"The paper's central claim is that the low-frequency suppression of $R_{B1g}(\\Omega)$ below $T_n$ is not due to quasiparticle damping or to a gap but to the orbital transmutation of the pockets. In the tetragonal phase, the $B_{1g}$ Raman vertex in the band basis is proportional to $\\cos 2\\theta$; below $T_n$ it acquires an angle-independent component $\\Gamma_s$ because the nematic order parameter $\\Delta_h$ reorganizes the orbital weights $u_k^2$ and $|v_k|^2$. Writing $\\cos 2\\bar\\theta_k = \\Gamma_s + \\Gamma_d\\cos 2\\theta$, the ladder vertex corrections cancel the $\\Gamma_s$ part exactly, as charge conservation requires, and the full response becomes $R_{B1g}(\\Omega) \\propto \\Gamma_d^2\\, \\Omega\\gamma/[\\Omega^2 + 4\\gamma^2(1-U\\Gamma_d^2/U_{cr})^2]$. At large $\\lambda_F = \\Delta_h/(b k_F^2)$, the outer hole pocket is almost pure $d_{xz}$, giving $\\Gamma_s \\approx -1$ and $\\Gamma_d \\sim 1/\\lambda_F \\ll 1$; hence the low-frequency intensity drops by $\\Gamma_d^2$ and recovers only at $\\Omega \\gtrsim 2{-}3\\Delta_h$, matching the measured gap-like behavior.","pith_inferences":["A quantitative signature the paper leaves implicit: since $\\Gamma_d \\sim b k_F^2/\\Delta_h$, the low-frequency Raman intensity should keep deepening roughly as $1/\\Delta_h^2$ as $T$ falls below $T_n$, which can be checked against independently measured $\\Delta_h(T)$.","A testable extension is to measure the $B_{1g}$ Raman response in another nematic iron-based compound with larger Fermi pockets; the same logic predicts a much weaker or absent low-frequency suppression because $\\Gamma_d$ stays of order one.","An analogous calculation for other zero-momentum quadrupolar probes that acquire an angle-independent form-factor component under nematic order should show the same conservation-law cancellation, so the mechanism is not specific to Raman scattering."],"forward_implications":["Below $T_n$ the low-frequency $B_{1g}$ Raman intensity scales as $\\Gamma_d^2(\\Omega)\\, \\Omega\\gamma/[\\Omega^2 + 4\\gamma^2(1-U\\Gamma_d^2/U_{cr})^2]$, so it can be strongly suppressed without any change in the damping rate $\\gamma$ and without opening a quasiparticle gap.","The suppression is confined to $\\Omega \\lesssim 2{-}3\\Delta_h$; at larger frequencies the vertex returns to its tetragonal $d$-wave form and $R_{B1g}(\\Omega)$ recovers its normal-metal value.","The measured drop is therefore a bulk signature of the same orbital reconstruction observed by polarized ARPES, rather than evidence for a gap in a metal.","The mechanism is generic to any nematic metal, but its strength is band-structure dependent; FeSe's unusually small Fermi energy makes $\\lambda_F$ large, which is why the effect is so pronounced there.","If the orbital-selective spectral-weight scenario is realized instead, the outer pocket would not become mono-orbital, the $B_{1g}$ vertex would keep its $d$-wave form, and no suppression is expected -- in disagreement with the data."],"supporting_citations":[{"why":"supplies the bulk Raman data whose low-frequency drop below $T_n$ is the effect being explained.","marker":"[22]"},{"why":"corroborates the gap-like $B_{1g}$ Raman drop in the nematic phase of FeSe.","marker":"[24]"},{"why":"polarized ARPES showing the outer hole pocket carries over 80% $d_{xz}$ weight, the mono-orbital input to the argument.","marker":"[14]"},{"why":"polarized ARPES evidence for orbital-origin reconstruction in nematic FeSe, supporting the orbital-composition change.","marker":"[16]"},{"why":"provides the two-orbital low-energy Hamiltonian whose eigenvectors give the orbital weights.","marker":"[42]"},{"why":"source of the FeSe band parameters and the orbital-dependent Fermi-surface shrinking used in the numerics.","marker":"[8]"},{"why":"method for summing the impurity ladder vertex corrections that cancel the $s$-wave part of the Raman vertex.","marker":"[41]"},{"why":"establishes the charge-conservation and vertex-correction constraint in Raman response that forces the cancellation.","marker":"[33, 34]"},{"why":"gives the relaxation form of $R_{B1g}$ above $T_n$ that the nematic-phase formula generalizes.","marker":"[4, 29, 36]"}],"fun_headline_variants":["FeSe's Raman 'gap' is orbital reshuffling, not a real gap","Vertex corrections, not gaps, explain FeSe's Raman downturn","Mono-orbital pockets make FeSe's Raman response plummet","Orbital transmutation, not a gap, drives FeSe's Raman drop","FeSe's Raman 'gap' is just pockets going mono-orbital"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument depends on the orbital mix on the pockets being set entirely by the two-orbital band Hamiltonian; if strong correlations renormalize the orbital weights so the outer hole pocket never becomes nearly mono-orbital, the predicted Raman suppression would not occur.","fun_headline_variants_meta":{"raw":{"variants":["FeSe's Raman 'gap' is orbital reshuffling, not a real gap","Vertex corrections, not gaps, explain FeSe's Raman downturn","Mono-orbital pockets make FeSe's Raman response plummet","Orbital transmutation, not a gap, drives FeSe's Raman drop","FeSe's Raman 'gap' is just pockets going mono-orbital"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000637,"raw_usage":{"total_tokens":2964,"prompt_tokens":1004,"completion_tokens":1960,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":620,"completion_tokens_details":{"reasoning_tokens":1861}},"tokens_in":620,"tokens_out":1960,"duration_ms":14887,"temperature":1.0,"reasoning_tokens":1861,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T10:17:39.543272+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A decisive test is to compute $\\Gamma_d^2(\\Omega)$ from Eqs. (6)-(7) using independently measured $\\Delta_h(T)$ and band parameters, and compare the depth and temperature dependence of the measured low-frequency $B_{1g}$ drop with the predicted $\\Gamma_d^2$ scaling; the mechanism fails if the drop does not follow that scaling while the damping rate stays constant.","supporting_citations":[{"cited_title":"Benfatto, B","cited_arxiv_id":null,"evidence_quote":"supplies the bulk Raman data whose low-frequency drop below $T_n$ is the effect being explained."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"corroborates the gap-like $B_{1g}$ Raman drop in the nematic phase of FeSe."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"polarized ARPES showing the outer hole pocket carries over 80% $d_{xz}$ weight, the mono-orbital input to the argument."},{"cited_title":"Hashimoto, Y","cited_arxiv_id":null,"evidence_quote":"polarized ARPES evidence for orbital-origin reconstruction in nematic FeSe, supporting the orbital-composition change."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"source of the FeSe band parameters and the orbital-dependent Fermi-surface shrinking used in the numerics."},{"cited_title":"Onari, Y","cited_arxiv_id":null,"evidence_quote":"method for summing the impurity ladder vertex corrections that cancel the $s$-wave part of the Raman vertex."}],"review_version":1}