{"id":"ebcc7cd4-ebc5-48b7-af00-37cadc606e5e","arxiv_id":"2607.22283","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"A Gutzwiller derivation shows cavity-induced shifts of the Brinkman–Rice Mott boundary are controlled by the bond-gradient-projected Pauli–Fierz spectral density, making single normalized modes thermodynamically inert while structured multimode continua can shift Uc.","lead":"Vacuum photons can shift a bulk Mott transition only if the electromagnetic environment carries finite thermodynamic spectral weight with bond-scale spatial variation; a single normalized bright mode cannot. A Gutzwiller derivation gives a joint frequency–spatial spectral-density rule and predicts a measurable d⁻³-to-d⁻⁵ crossover for a surface-phonon-polariton geometry.","discovery_kind":"first_principles","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The quantitative kernel in Eq. (4) is proven only for the Brinkman–Rice variational manifold; the abstract's unconditional 'only when' overstates the result because the exact Hubbard response kernel is untested.","rationale":"The paper is internally consistent within its stated variational manifold: the mode-extensivity theorem is proven under explicit normal-state assumptions, Eq. (4) follows from the degenerate ensemble expansion in SM Appendix F, and the finite-coordination VMC is a genuine independent check of the central coefficient at r=1. The main risk is external validity. The Gutzwiller/Brinkman-Rice description is known to miss Hubbard bands and magnetic order; the authors acknowledge this but do not test how much the response kernel changes in a more accurate solver. Because Eq. (4) is the quantitative bridge to the surface-polariton d^-3-to-d^-5 prediction and the numerical shifts, the correctness risk is concentrated exactly there. The reader's CONDITIONAL verdict already captures this, so no change in verdict is needed. The abstract's unconditional phrasing is the clearest manifestation of the concern and should be amended to include the Gutzwiller/normal-state qualifiers.","tokens_in":20540,"tokens_out":23128,"duration_ms":208314,"concrete_test":"Perform single-site DMFT for the half-filled Hubbard model coupled to an extensive ensemble of site-local modes (mode density νm=1, r=1) with the same Pauli–Fierz self-polarization term, and compute Uc(ω) for ω/A0 ∈ {0.1, 0.25, 0.5, 1, 2, 4, 8} at weak coupling. Compare with the Gutzwiller prediction ΔUc = −νmλ² 4A0/(ω+4A0) (Eq. 4 with r=1). If the DMFT kernel deviates by more than ~20% in any frequency window, the quantitative criterion and the surface-polariton shifts need to be re-derived with the correct response kernel, even if the M/N extensivity law survives.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central quantitative statement, Eq. (4), with kernel K(ω,r)=4A0r/(ω+4A0r), is derived from the infinite-coordination Gutzwiller energy functional (SM Eq. A15) and the Brinkman–Rice quasiparticle weight Z(D)=8D(1−2D). The paper itself states in SM Appendix B that the Brinkman–Rice transition 'is not the exact Hubbard-model transition,' and the VMC validation section concedes that the trial state 'omits Hubbard bands and long-range magnetic order.' The VMC check in Fig. 3 is valuable but it samples the same Gutzwiller determinant family; it does not test whether the exact Hubbard-model response kernel has the same frequency and bond-gradient dependence. In DMFT/QMC the Mott boundary has incoherent Hubbard bands, a dynamical self-energy, and magnetic fluctuations, so the coefficient 4A0r/(ω+4A0r) and the d^-5 surface-polariton asymptotics are not guaranteed. The abstract states without qualification that a Mott transition shifts only under the spectral-weight/bond-variation condition, omitting the normal-state non-superradiant caveat and the 'within Gutzwiller theory' qualifier that the body and appendices carefully retain. This is the load-bearing weakness: the criterion's necessity may survive, but its quantitative content and the experimental d^-3-to-d^-5 prediction rest on an untested proxy for the actual Mott transition.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper asks whether vacuum electromagnetic fluctuations can shift a bulk Mott transition, and answers this question within a cavity-extended Gutzwiller/Brinkman-Rice variational description. It defines a thermodynamic mode weight ηα and a bond-gradient factor rα for each normalized electromagnetic mode, proves a normal-state mode-extensivity theorem (SM Theorem 1) stating that a fixed number of normalized modes shifts the energy density only by O(M/N), and derives the central quantitative criterion, Eq. (4): ΔUc = −∫ JPF(ω,r) 4A0r/(ω+4A0r) + O(JPF²), with JPF the joint frequency–spatial Pauli–Fierz spectral density. The criterion is first obtained from a solvable degenerate mode ensemble and then extended to a dilute spectral functional. The paper applies it to a 4H-SiC surface-phonon-polariton environment, obtaining a d⁻³-to-d⁻⁵ crossover for the shift of the Mott boundary and numerical estimates (≈0.37 meV at d=1 nm, ≈5 meV at d=0.5 nm for the stated parameters). Finite-coordination determinant variational Monte Carlo on random regular graphs is used to test the infinite-coordination critical coefficient and the M/N scaling, with agreement at the 5–7% level. The paper is careful to distinguish spectroscopic hybridization from thermodynamic control and repeatedly notes the variational nature of the calculation.","tokens_in":20898,"tokens_out":17565,"duration_ms":186630,"significance":"If the criterion holds, it is a genuinely useful organizing principle: the relevant quantity for thermodynamic cavity control of a correlated phase is not the Rabi splitting or the local density of states, but the bond-projected, frequency-resolved electromagnetic spectral density. The derivation is explicit and self-contained within the stated variational manifold; the mode-extensivity theorem is simple but powerful; the VMC section provides a nontrivial finite-coordination check with graph-resolved error bars; and the connection to the recent QMC result [15] strengthens confidence in the single-mode no-go. The paper is also unusually candid about its limitations, stating in SM Appendix B that the Brinkman–Rice transition is not the exact Hubbard transition and in the VMC section that Hubbard bands and long-range magnetic order are omitted. The main weakness is that the quantitative kernel K(ω,r)=4A0r/(ω+4A0r) is derived from the infinite-coordination Gutzwiller functional, so the numerical predictions for the surface-polariton geometry inherit that approximation; the authors acknowledge this, but the abstract states the criterion without the same qualifiers.","major_comments":[{"comment":"The abstract states without qualification that \"a Mott transition shifts only when the electromagnetic environment supplies finite thermodynamic spectral weight with bond-scale variation.\" This is established only within the normal-state, non-superradiant, infinite-coordination Brinkman–Rice/Gutzwiller manifold. SM Appendix B explicitly says that the Brinkman–Rice transition is not the exact Hubbard-model transition, and the VMC section says the trial state omits Hubbard bands and long-range magnetic order. Since the title correctly carries the qualifier \"within Gutzwiller Theory,\" the abstract and the closing summary should carry the same qualifier, e.g., \"within this variational description.\" Without this, the reader may take the criterion as a proven theorem for the actual Hubbard-model boundary rather than a controlled variational-manifold result.","section":"Abstract and Conclusion"},{"comment":"The spectral functional is derived to first order in JPF; the O(JPF²) corrections from the self-consistent renormalization of Ac are identified in SM but not estimated. The paper should state the relevant small parameter (e.g., Λloc/A0 or νmλ²/(8A0)) and verify it for the SiC values in Table I. At d=0.5 nm, Λloc≈8.5 meV against A0=0.2 eV, so the ratio is about 4%, but this bound is not given in the main text. Relatedly, Eq. (8) substitutes a continuous spectral density J_surf(Ω,q) into a formula derived for a discrete mode ensemble with independent variational displacements; the additivity of the first-order kernel is plausible, but the conditions under which the continuum limit commutes with the variational minimization should be stated. This would materially strengthen confidence in the quantitative d⁻³-to-d⁻⁵ numbers.","section":"Eq. (4) and SM Appendices F,H"}],"minor_comments":[{"comment":"The sentence \"The resulting coefficients are The finite-coordination estimate...\" has a grammatical break; the table reference and sentence should be cleaned up.","section":"SM Appendix J.5"},{"comment":"Reference [56] is a duplicate of reference [5]; consolidate.","section":"References"},{"comment":"The legend distinguishes \"M=1, varying N\" and \"N=16, varying M\" only by symbols. Please make the fixed parameters explicit in the caption or axis labels, since both data sets are plotted against the same horizontal axis label.","section":"Fig. 3(c)"},{"comment":"\"M/Nscaling\" should read \"M/N scaling\" (missing space).","section":"Abstract"},{"comment":"\"Theorem1intheSM\" appears with missing spaces; similarly, a few other LaTeX artifacts remain (e.g., in the abstract).","section":"Main text, mode-extensivity paragraph"},{"comment":"The phrase \"the difference between two uncoupled energy densities\" is unclear; it should read \"the difference between the energy densities of the two phases of the matter Hamiltonian\" or similar.","section":"SM Theorem 1 proof"}],"recommendation":"minor_revision","confidential_remarks":"The paper is a solid theory contribution and suits the journal's scope. The central derivation is sound within its stated manifold, and the authors' open listing of limitations is a strength rather than a defect. The only substantive issue is scope wording in the abstract and the lack of an explicit small-parameter estimate for the dilute spectral expansion; both are fixable without changing the main results. I found no evidence of circularity or of fitted constants being recycled as predictions."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The useful result here is the criterion: for a normal-state Mott transition, a normalized single cavity mode contributes O(1/N) to the energy density and cannot shift the boundary, while an extensive continuum can, provided it has finite local spectral weight and a field profile that varies across the electronic bond. That statement is derived cleanly — the joint Pauli–Fierz density J_PF(ω,r) and the bond-gradient filter rα are genuinely new — and the paper earns its main claim.\n\nThe algebra is explicit and the checks are honest. Eq. (4) comes from expanding the variational energy, not from an assumption. The degenerate-ensemble solution reproduces it in the dilute limit, and the finite-coordination VMC gives the critical coefficient within 5–7% of the infinite-coordination contraction without imposing Z(D)=8D(1−2D). The M/N scaling test is nice. The VMC sits inside the same Gutzwiller variational family, so it tests the 1/z contraction, not the exact Hubbard boundary — the authors say exactly that. The load-bearing caveat is that the quantitative kernel 4A0r/(ω+4A0r) and the d−3-to-d−5 surface-polariton asymptotics are proven only within the Brinkman–Rice manifold. A DMFT or QMC response kernel could change the coefficient even if the spectral-weight/bond-gradient criterion survives. That is a real limitation, but it is stated in the body and SM, and it does not undercut the structural result.\n\nThe only thing I'd press on is the abstract. \"A Mott transition shifts only when...\" is too crisp; it should carry \"within Gutzwiller theory\" and the normal-state, non-superradiant caveat in the same sentence. That is wording, not substance. The body lists the exceptions (superradiance, divergent susceptibility, superextensive coupling) and the theorem assumes them away by name.\n\nThis is a paper for the cavity-quantum-materials crowd, and it deserves peer review: the criterion is testable in electrodynamic solvers, the VMC is reproducible, and the open question — what the exact Hubbard kernel does — is stated plainly rather than hidden. I'd bring it to reading group and cite it for the spectral-density design rule.","headline":"A clean variational answer to when a cavity can move a Mott transition — single modes can't, bond-varying continua can — with the honest caveat that the quantitative kernel is proven only inside Gutzwiller theory.","tokens_in":21405,"tokens_out":2994,"would_cite":true,"duration_ms":28047,"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":"A cavity moves a Mott boundary only if its vacuum field carries bond-scale spectral weight.","keywords":["Mott transition","cavity quantum electrodynamics","Gutzwiller variational method","Brinkman–Rice transition","Pauli–Fierz spectral density","surface phonon polaritons","mode extensivity","variational Monte Carlo"],"falsifier":"Measure or compute with an unbiased many-body method the critical interaction of a correlated layer at several spacer thicknesses d above a surface-phonon-polariton medium: if the shift does not show the predicted d^-3-to-d^-5 crossover with the stated coefficient, or if one properly normalized mode moves the phase boundary by an O(1) amount, the criterion fails.","tokens_in":20397,"feed_emoji":"🔬","tokens_out":6198,"duration_ms":62044,"temperature":0.7,"pith_summary":"Can empty-space quantum fluctuations move a bulk metal–insulator transition? This paper answers with a precise condition: a cavity environment shifts a Mott boundary only when its vacuum field carries finite thermodynamic spectral weight that varies across the electronic bond. Built inside the Gutzwiller–Brinkman–Rice variational description, the argument produces a joint frequency–spatial Pauli–Fierz spectral density whose first moment controls the shift. A single normalized bright mode changes the energy density only as O(1/N) and cannot move a regular phase boundary, no matter how large its Rabi splitting; an extensive mode continuum or lossy environment with bond-scale field variation can. The criterion is evaluated for a surface-phonon-polariton geometry, giving an experimentally testable d^-3-to-d^-5 crossover, and supported by finite-coordination variational Monte Carlo checks of the coefficient and M/N scaling.","feed_headline":"Cavity shifts Mott boundary only with bond-scale fields","feed_subtitle":"Most bright modes are thermodynamically inert; only fields that resolve an electronic bond can shift a phase transition.","key_machinery":"The central objects are the joint Pauli–Fierz spectral density JPF(ω,r) and the bounded response kernel K(ω,r)=4A0r/(ω+4A0r). JPF collects each mode's thermodynamic weight per site, coupling strength, frequency, and bond-gradient factor r, while K applies the two physical filters: modes slow enough to leave a residual self-polarization penalty, and fields that actually vary across the electronic bond. The product integrated against JPF gives the leading shift of Uc. The companion normal-state mode-extensivity theorem supplies the counting: M normalized modes change an extensive energy density by O(M/N), with exceptions only for divergent susceptibilities, superradiant order, or superextensiv","core_discovery":"Within the infinite-coordination Gutzwiller variational manifold, the paper separates collective spectroscopic hybridization from thermodynamic phase control. It defines a joint Pauli–Fierz density JPF(ω,r)=Σα ηα λα² δ(ω−ωα)δ(r−rα), where ηα is the mode's thermodynamic weight per site and rα measures the field's variation across a nearest-neighbor bond. The leading shift of the critical interaction is ΔUc = −∫ dω dr JPF(ω,r) 4A0r/(ω+4A0r) + O(JPF²): the bounded kernel 4A0r/(ω+4A0r) suppresses modes that are too fast and fields that are uniform across the bond. A conventionally normalized single mode has ηα=1/N, so its thermodynamic effect vanishes; an environment with finite local spectral w","pith_inferences":["The same joint spectral-density criterion plausibly extends beyond the single-band Hubbard–Brinkman–Rice case: replacing the Gutzwiller response kernel with a more accurate many-body kernel should preserve the conditions of finite thermodynamic weight and r>0, while changing the numerical coefficient.","The d^-3-to-d^-5 crossover is a direct experimental target: if a correlated layer's Uc shifts independently of whether the field resolves the bond, or fails to show the predicted distance law, the criterion is falsified.","The criterion reframes nanophotonic engineering for phase control as a classical electromagnetism task: computing the bond-correlated field-difference spectrum, rather than maximizing mode confinement or quality factor.","The no-go statement for a single mode is a normal-state result; the paper's own caveats imply that at a critical point with a divergent susceptibility, or with macroscopic photon occupation, the counting can break, so the criterion should not be read as a universal no-go theorem."],"forward_implications":["A single bright cavity mode, even with a large collective Rabi splitting, cannot move a normal-state Mott boundary in the thermodynamic limit; its energy-density effect scales as 1/N.","An extensive set of modes or a lossy continuum with finite local Pauli–Fierz spectral weight and bond-scale field variation shifts Uc downward, making localization easier at fixed bare interaction.","The shift follows the mode-density law M/N, directly verified in finite-coordination variational Monte Carlo within roughly 5–7% of the infinite-coordination coefficient.","For a planar surface-phonon-polariton environment, the shift crosses from d^-3 to d^-5 with increasing spacer thickness, giving a measurable signature by varying the gap.","Cavity design should target the bond-projected spectral density, equivalently the first inverse-frequency moment, rather than Rabi splitting or the local density of optical states alone."],"fun_headline_variants":["Cavity shifts Mott only with bond-resolving fields","Mott moves only for bond-scale cavity fields","Bond-scale fields alone move Mott transitions","Cavity thermodynamic control needs bond-scale modes","Mott shift requires bond-resolving vacuum modes"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The load-bearing premise is that the infinite-coordination Gutzwiller–Brinkman–Rice variational description is a faithful proxy for the actual Hubbard-model Mott boundary; the paper itself notes this description omits Hubbard bands and long-range magnetic order.","fun_headline_variants_meta":{"raw":{"variants":["Cavity shifts Mott only with bond-resolving fields","Mott moves only for bond-scale cavity fields","Bond-scale fields alone move Mott transitions","Cavity thermodynamic control needs bond-scale modes","Mott shift requires bond-resolving vacuum modes"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000206,"raw_usage":{"total_tokens":1191,"prompt_tokens":660,"completion_tokens":531,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":404,"completion_tokens_details":{"reasoning_tokens":459}},"tokens_in":404,"tokens_out":531,"duration_ms":4925,"temperature":1.0,"reasoning_tokens":459,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-01T05:13:59.545037+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure or compute with an unbiased many-body method the critical interaction of a correlated layer at several spacer thicknesses d above a surface-phonon-polariton medium: if the shift does not show the predicted d^-3-to-d^-5 crossover with the stated coefficient, or if one properly normalized mode moves the phase boundary by an O(1) amount, the criterion fails.","supporting_citations":[],"review_version":1}