{"id":"de53538c-3320-4c56-aaed-b86aedc63be4","arxiv_id":"2504.13715","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"Excitonic electron-hole interactions strongly reshape the K-point optical phonon of graphene, producing a ~150 cm^-1 red-shift, a 10x group-velocity increase, and a 5x linewidth increase.","lead":"This paper computes how electron-hole interactions (excitonic effects) alter the vibrations of graphene, finding a large red-shift and a fivefold broadening of the optical phonon near the K point. It matters because those phonons control the Raman 2D peak and the intrinsic electrical resistivity of graphene.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The dropped double-counting term required by the GDFT functional is the load-bearing assumption; its magnitude at K is untested, so the 150 cm^-1 shift and 5x linewidth are conditional.","rationale":"The paper is doing something genuinely useful: it formulates phonon response from the two-particle Green's function with ladder diagrams and applies it to graphene. The framework is coherent and the gauge-field argument at Γ supplies a nontrivial internal consistency check: the FWHM at Γ is protected when the full ladder series is summed, and this is verified numerically. I therefore do not see a reason to reject the framework. The load-bearing weak point is elsewhere. The central quantitative claims—150 cm^-1 redshift at K, 5x linewidth, 26x e-ph enhancement—all come from the correction ΔC in Eq. (8), and in the implementation the term containing ΔfHxc is set to zero. This is not a minor detail: the authors' own GDFT functional includes ¯Exc specifically to avoid double counting the Fock interaction, and its omission changes the vertex from the GDFT vertex to the DFT vertex. The magnitude of this omitted local term at K is never estimated. The paper's appeal to experimental agreement is the only evidence, and that comparison is weakened by the 20 cm^-1 rigid shift and the reliance on graphite data except for the single Raman point of Ref. 13. Figure 5 shows large cancellations between self-energy and ladder contributions; a missing term of similar scale would easily move the final numbers. The other possible concerns—fitted tight-binding W, static W, spectral-function peak versus pole—are real but secondary: they affect the size of the ladder correction, but the dropped double-counting term is a formal inconsistency in the very functional used to define the calculation, and it is explicitly admitted in the text. A direct computation of the omitted term is feasible and would settle the issue. Because the reader's verdict is already conditional and this concern is precisely what makes it conditional, the verdict should remain unchanged pending that check.","tokens_in":24940,"tokens_out":7422,"duration_ms":70295,"concrete_test":"Recompute ΔC(q,ω) at the K point with the omitted term restored. Implement the LDA form of ¯Exc given in the SM (Eq. 13), take its second functional derivative ΔfHxc, and add the local term -ρ_DFT ΔfHxc ρ_DFT to Eq. (8), using the same tight-binding basis, W, and grids. Redo the phonon spectral function for the TO mode at K at T=70 K. If the shift changes by less than ~15 cm^-1 and the FWHM by less than ~10%, the neglect is harmless; if changes are comparable to the claimed 150 cm^-1 and 100 cm^-1, the headline claims fail.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central numerical claim—~150 cm^-1 redshift, ~5x FWHM, 26x |g_K|^2—is computed from Eq. (8) with ΔfHxc set to zero. The formal GDFT functional (Eq. 4) includes ¯Exc to cancel the Fock-like term in the uniform limit; its second density derivative enters Eq. (8) as -ρ_DFT ΔfHxc ρ_DFT and modifies the GDFT vertex (Eq. 9). The SM states this term is neglected because of the reduced tight-binding Hilbert space, so ~V_GDFT reduces to V_DFT. The only in-paper justification is the 'excellent agreement' with experiment, but that comparison applies a 20 cm^-1 rigid shift and uses graphite data for most of the BZ; the freestanding-graphene data point is a single 2D Raman measurement (Ref. 13). Nothing demonstrates that ΔfHxc is small at K, where the exchange-correlation kernel can be large. If this term contributes at the scale of tens of cm^-1 or more, the headline values change, because the cancellation between self-energy and ladder contributions is delicate (Fig. 5). Thus the central claim rests on an unverified neglect of a term required by the authors' own functional.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper develops a generalized Kohn-Sham (GDFT) framework for phonon dispersions and linewidths that includes electron-hole (excitonic) interactions, and applies it to graphene. The formal part expresses the phonon dynamical matrix as a DFT contribution plus a correction built from the BSE-ladder susceptibility χW and the GDFT vertex. Numerically, the correction is evaluated in a five-nearest-neighbor π-band tight-binding model fitted to DFT, with the double-counting term ΔfHxc set to zero and with a static screened interaction W. The central claims near K are a TO-phonon redshift of about 150 cm^-1, a tenfold group-velocity enhancement, a fivefold FWHM increase, and a 26-fold enhancement of |g|^2; near Γ the excitonic effects are claimed to be small due to a gauge-field protection. The validation compares the computed dispersion with graphite data over most of the Brillouin zone and with a single freestanding-graphene Raman dataset (Ref. 13) shifted rigidly by 20 cm^-1.","tokens_in":25271,"tokens_out":6790,"duration_ms":65482,"significance":"If the central claims are correct, the paper would reshape the standard DFT picture of the K-point Kohn anomaly and identify zone-boundary phonons as the dominant source of intrinsic resistivity in graphene at room temperature. The formal framework is a useful extension of the variational formulation of Ref. 23, and the ladder-convergence analysis in Fig. 5 plus the gauge-field consistency check are valuable methodological contributions. However, the numerical predictions are conditional on the untested neglect of the double-counting term and on the fitting/validation choices, so the result is a promising but not yet fully established quantitative claim.","major_comments":[{"comment":"The central numerical claim rests on setting ΔfHxc=0. Equation (4) defines the GDFT functional with a double-counting term \\bar E_xc that cancels the Fock-like term in the uniform limit; its second density derivative enters Eq. (8) explicitly and modifies the GDFT vertex in Eq. (9). SM Sec. VIII states that this term is neglected because of the reduced tight-binding Hilbert space. The only in-paper justification is the 'excellent agreement' with experiment, but Fig. 1 compares mostly graphite data and Fig. 2(b) uses a single freestanding-graphene dataset (Ref. 13) shifted by 20 cm^-1. No estimate or bound on \\bar E_xc near K is provided. Since Fig. 5 shows that self-energy and ladder contributions have opposite signs and nearly cancel at Γ, an omitted term of similar scale could change the 150 cm^-1 redshift and the fivefold linewidth materially. Please compute or bound this term, or otherwise demonstrate that it is negligible at K.","section":"Eq. (8) and SM Sec. VIII"},{"comment":"The linewidth claim is not reproducible as written. The SM states that the 0+ limit in the phonon Green's function is replaced by a finite damping η=10 meV. Inserting η=10 meV into G^-1=ω^2-D(ω+iη) adds a Lorentzian broadening of roughly 2η≈160 cm^-1 to every spectral function, yet Fig. 3(e) reports a DFT FWHM of order 10 cm^-1 at Γ. Either the FWHMs are extracted after deconvolving or subtracting this broadening, or η is used only in the electronic response and not in the final phonon Green's function. The present wording is contradictory and needs clarification, because the fivefold enhancement near K is a headline result.","section":"SM Sec. VIII, spectral function"},{"comment":"The experimental validation is weaker than the text claims. The 'excellent agreement' that justifies neglecting \\bar E_xc is obtained by comparing the graphene calculation with graphite data over most of the Brillouin zone (Refs. 58-60), and the only freestanding-graphene data near K are the Ref. 13 Raman points, which are rigidly shifted by 20 cm^-1 before comparison. A rigid shift applied to the reference data cannot validate a momentum-dependent 150 cm^-1 redshift, and graphite data cannot constrain monolayer-specific excitonic effects. Please provide a quantitative comparison with residuals and error bars, either for the unshifted graphene data or after a clearly stated and justified shift.","section":"Figs. 1-2, validation"}],"minor_comments":[{"comment":"The phrase 'partially-screen partially-screen formulation' appears to be a duplicated phrase and should read 'partially screened formulation'.","section":"Main text before Eq. (5)"},{"comment":"The word 'programe' should be 'program'.","section":"Acknowledgments"},{"comment":"The x-axis label 'K M0' is unclear; the momentum path should be labeled consistently with the high-symmetry points used in the text.","section":"Fig. 1"},{"comment":"The columns labeled ℏΓ_Γ and ℏΓ_K are reported in cm^-1; the symbol ℏΓ suggests an energy, so the units should be stated explicitly and the notation made consistent.","section":"SM Table I"}],"recommendation":"major_revision","confidential_remarks":"The paper is suitable for the journal if the numerical issues are resolved. I would not reject it because the formal framework is coherent and the missing double-counting contribution is, in principle, computable within the same tight-binding model. However, the construction and validation both rely heavily on the authors' own previous work (the TB model of Ref. 22 and the Raman data of Ref. 13), so an independent benchmark or an explicit estimate of the neglected term is important before the quantitative claims can be accepted."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things you should know. First, this is the first calculation of phonon dispersions and lifetimes from the interacting two-particle Green's function with an infinite ladder sum of electron-hole interactions. That is a real methodological step. Second, the central quantitative claim—~150 cm^-1 redshift, ~5x FWHM at K—is conditional on an approximation the paper states plainly but does not justify numerically: the double-counting term ΔfHxc required by their GDFT functional is set to zero.\n\nWhat is genuinely good: the framework itself. Combining GDFT with a range-separated hybrid and BSE ladder diagrams, and using the variational formulation from Caldarelli et al., gives a formally clean route to phonon self-energies beyond finite-order perturbation theory. The physics result is also plausible. Earlier analytic work by Basko and Aleiner predicted a divergent electron-phonon coupling at K, and measured graphite TO branches are redshifted relative to DFT. Their finding that self-energy and ladder corrections have opposite signs and compensate at Γ is a nice consistency check, and the gauge-field protection at zone center is elegant. They are also honest about what they dropped.\n\nWhere I worry: the dropped ΔfHxc is not a minor detail. It enters directly in the vertex and in Eq. (8) as -ρ_DFT ΔfHxc ρ_DFT. The paper's only justification is 'excellent agreement' with experiment, but that agreement is with same-group Raman data (Ref. 13) shifted by 20 cm^-1, and most of the comparison is to graphite rather than freestanding graphene. Nothing shows ΔfHxc is small at K, where the exchange-correlation kernel can be large. The cancellation between self-energy and ladder terms is delicate (Fig. 5), so if ΔfHxc contributes at the tens of cm^-1 scale, the headline numbers shift. The tight-binding model fitted to DFT and the self-consistent W from their own Ref. 22 add model dependence, though that is not circular on the target observable. No code or data is shipped, so reproduction requires re-implementation.\n\nBottom line: this is a serious paper for people working on electron-phonon coupling in 2D materials. It deserves a serious referee. The right outcome is probably acceptance after the authors either quantify ΔfHxc in the model or at least discuss its expected magnitude in the low-energy subspace. I would not desk-reject, and I would not treat the 150 cm^-1 number as established.","headline":"Genuinely new framework for excitonic effects in phonons, with a plausible central physics but a load-bearing approximation (dropped double-counting term) that the paper does not quantify, so the headline numbers remain conditional.","tokens_in":25763,"tokens_out":2522,"would_cite":true,"duration_ms":23881,"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 finds that excitonic electron-hole interactions red-shift graphene's K-point TO phonon by ~150 cm⁻¹, raise its group velocity tenfold, and quintuple its linewidth via a 26x enhancement of the electron-phonon coupling.","keywords":["excitonic effects","phonon self-energy","Kohn anomaly","graphene","electron-phonon coupling","Bethe-Salpeter equation","generalized Kohn-Sham","phonon linewidth"],"falsifier":"Recompute the K-point TO phonon self-energy including the double-counting correction (nonzero $\\Delta f_{Hxc}$ and full GDFT vertex) within the same tight-binding model; if the ~150 cm$^{-1}$ red-shift and ~5x FWHM enhancement are not reproduced, the excitonic claim rests on the neglected term.","tokens_in":24749,"feed_emoji":"⚛️","tokens_out":12949,"duration_ms":103457,"temperature":0.7,"pith_summary":"The paper develops a first-principles framework that puts excitonic (electron-hole) interactions into the phonon self-energy, and applies it to graphene's optical phonons. It claims that near the Brillouin-zone corner K these interactions dramatically reshape the Kohn anomaly: the transverse optical phonon red-shifts by ~150 cm⁻¹, its group velocity increases tenfold, and its linewidth grows about fivefold, driven by a 26-fold increase in the squared electron-phonon matrix element. Near the zone center Γ the excitonic corrections are minor, because at long wavelengths the electron-phonon coupling acts as a gauge field that merely shifts the Dirac cone. If the claim is right, standard DFT phonon calculations miss the dominant physics of the K-point phonons and of graphene's intrinsic resistivity.","feed_headline":"Graphene's K-point phonons red-shift ~150 cm-1 and widen 5x","feed_subtitle":"Electron-hole pairs quintuple K-point phonon linewidths and boost group velocity 10x, reshaping the 2D Raman mode.","key_machinery":"The central object is the electron-hole ladder propagator $L_W$ (solution of the Bethe-Salpeter equation with static $W$), contracted to the density-density response $\\chi_W$, which enters the phonon self-energy as $\\Delta C = V_{\\mathrm{DFT}} (\\chi_W - \\chi_0) V_{\\mathrm{DFT}}$. The paper also uses the exact generalized Fermi golden rule from the variational formulation to extract effective dressed electron-phonon matrix elements, and the gauge-field argument that at $\\Gamma$ an atomic displacement merely shifts the Dirac cone, protecting the phonon linewidth.","core_discovery":"Working in a generalized Kohn-Sham (GDFT) framework, the authors compute the phonon dynamical matrix from the interacting electron density-density response $\\chi_W(z)$ that includes the full Bethe-Salpeter ladder of electron-hole interactions with a static screened Coulomb interaction $W$, evaluated in a tight-binding model fitted to DFT. The phonon correction takes the differential form $\\Delta C = V(\\chi_W - \\chi_0)V$, so that all deviation from DFT comes from the difference between the interacting and bare responses. They find that at the K point the excitonic ladder reverses the self-energy-driven blue-shift and produces a substantial red-shift of the TO phonon (~150 cm$^{-1}$ for the dynamical phonon), a tenfold increase of the group velocity, and a fivefold increase of the full-width at half-maximum (from ~20 to ~100 cm$^{-1}$), which they attribute to a 26-fold increase in the effective electron-phonon coupling squared. These effects persist while $2E_F < \\hbar\\omega_{ph}$ and are quenched at higher doping. At $\\Gamma$, the linewidth is protected because the electron-phonon coupling renormalizes the Fermi velocity, leaving the ratio $v_g/v_\\phi$ nearly unchanged, a 'gauge-field protection' that requires the full ladder series to be respected.","pith_inferences":["A quantitative transport calculation built on the enhanced $|g_K|^2$ would likely show that DFT-based estimates of phonon-limited resistivity in graphene systematically underestimate the K-point contribution; the paper hints at this but does not compute the resistivity.","The strong monolayer-vs-graphite contrast suggests that screening by a substrate (for example hBN encapsulation or multilayer stacking) would suppress the excitonic redshift and linewidth broadening, a testable difference between suspended and supported graphene.","A direct measurement of the TO branch near K in suspended graphene, via momentum-resolved EELS or high-resolution Raman of the 2D mode as a function of doping, could verify the predicted ~150 cm⁻¹ red-shift and ~100 cm⁻¹ FWHM, and should show quenching when $2E_F$ exceeds $\\hbar\\omega_{ph}$."],"forward_implications":["At K, the TO phonon frequency in freestanding graphene is about 150 cm⁻¹ lower than DFT predicts, and its group velocity is about ten times larger, which should show up in the dispersion of the 2D Raman mode.","The K-point FWHM grows from ~20 to ~100 cm⁻¹, meaning phonon lifetimes are shortened about fivefold by excitonic effects.","The squared electron-phonon matrix element at K is enhanced by a factor ~26, which the paper argues makes K-point phonons the likely dominant source of the intrinsic electrical resistivity of graphene at room temperature.","The excitonic enhancement survives up to doping $2E_F < \\hbar\\omega_{ph}$ and is quenched at higher doping, so it is most relevant in undoped or lightly doped graphene.","Near Γ, the phonon frequency and linewidth are essentially unchanged by excitonic effects, because the coupling acts as a gauge field; the full ladder sum is required to recover this cancellation."],"supporting_citations":[{"why":"Supplies the interacting π-band tight-binding model with self-consistent screened W used to compute χW and the ladder series.","marker":"[22]"},{"why":"Provides the variational formulation of the phonon response with nonlocal exchange and the exact generalized Fermi golden rule, giving the differential correction and the FWHM formula.","marker":"[23]"},{"why":"Provides the measured graphene Raman 2D-mode dispersion that the static GDFT phonon dispersion is compared to after a 20 cm⁻¹ rigid shift.","marker":"[13]"},{"why":"Gives experimental graphite phonon dispersions used to validate the overall DFT/GDFT dispersion in Fig. 1.","marker":"[58]"},{"why":"Provides additional experimental graphite phonon data used for the same validation.","marker":"[59]"},{"why":"Supplies the measured transverse optical branch near K in graphite that exhibits the anomalous redshift the theory aims to explain.","marker":"[60]"},{"why":"Predicts a logarithmic divergence of the electron-phonon coupling at K at zero temperature, which the 26x enhancement is compared with.","marker":"[25]"},{"why":"Extends the analytical prediction of the diverging K-point coupling, used to rationalize the |g_K|² enhancement.","marker":"[26]"},{"why":"Provides the DFT phonon dispersions and Kohn-anomaly slopes used as the no-excitonic baseline in Figs. 1 and 2.","marker":"[61]"}],"fun_headline_variants":["Excitons red-shift graphene K-phonons 150 cm-1, quintuple linewidth","Electron-hole pairs boost graphene K-phonon speed 10x, broaden 5x","Excitonic ladder red-shifts graphene K-phonons 150 cm-1, 5x wider","Graphene K-phonons: excitons cause 150 cm-1 redshift, 5x linewidth","K-point phonons in graphene: excitons red-shift 150 cm-1, speed 10x"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The entire claimed effect is computed while neglecting the double-counting correction required by the theory, using a model interaction fitted to DFT; if that neglected correction or the fitted interaction materially changes the result, the predicted redshift and linewidth enhancement would change.","fun_headline_variants_meta":{"raw":{"variants":["Excitons red-shift graphene K-phonons 150 cm-1, quintuple linewidth","Electron-hole pairs boost graphene K-phonon speed 10x, broaden 5x","Excitonic ladder red-shifts graphene K-phonons 150 cm-1, 5x wider","Graphene K-phonons: excitons cause 150 cm-1 redshift, 5x linewidth","K-point phonons in graphene: excitons red-shift 150 cm-1, speed 10x"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001471,"raw_usage":{"total_tokens":5945,"prompt_tokens":1009,"completion_tokens":4936,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":625,"completion_tokens_details":{"reasoning_tokens":4803}},"tokens_in":625,"tokens_out":4936,"duration_ms":32570,"temperature":1.0,"reasoning_tokens":4803,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T12:02:45.175658+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Recompute the K-point TO phonon self-energy including the double-counting correction (nonzero $\\Delta f_{Hxc}$ and full GDFT vertex) within the same tight-binding model; if the ~150 cm$^{-1}$ red-shift and ~5x FWHM enhancement are not reproduced, the excitonic claim rests on the neglected term.","supporting_citations":[{"cited_title":"Guandalini, G","cited_arxiv_id":null,"evidence_quote":"Supplies the interacting π-band tight-binding model with self-consistent screened W used to compute χW and the ladder series."},{"cited_title":"Excitonic effects in phonons: reshaping the graphene Kohn anomalies and lifetimes","cited_arxiv_id":"2504.13715","evidence_quote":"Provides the variational formulation of the phonon response with nonlocal exchange and the exact generalized Fermi golden rule, giving the differential correction and the FWHM formula."},{"cited_title":"Saito, A","cited_arxiv_id":null,"evidence_quote":"Provides the measured graphene Raman 2D-mode dispersion that the static GDFT phonon dispersion is compared to after a 20 cm⁻¹ rigid shift."},{"cited_title":"Massidda, A","cited_arxiv_id":null,"evidence_quote":"Gives experimental graphite phonon dispersions used to validate the overall DFT/GDFT dispersion in Fig. 1."},{"cited_title":"Rohlfing, Phys","cited_arxiv_id":null,"evidence_quote":"Provides additional experimental graphite phonon data used for the same validation."},{"cited_title":"Rocca, R","cited_arxiv_id":null,"evidence_quote":"Extends the analytical prediction of the diverging K-point coupling, used to rationalize the |g_K|² enhancement."}],"review_version":1}