{"id":"77d98b5f-9677-4605-b65d-edfef782ed0e","arxiv_id":"2501.08216","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Resonant Higgs drive generates a divergent negative superfluid density, driving homogeneous superconductors toward spatial phase separation.","lead":"Resonant driving of the Higgs mode in a superconductor can make its superfluid density negative, a sign that the uniform state is unstable. The instability manifests as spontaneous spatial patterns and topological defects, predicted here in a one-dimensional simulation.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The negative-superfluid-density divergence relies on a number-nonconserving Markovian bath (Eq. 10); a physical phonon bath may regularize it, and the conclusion 'generally unstable' is broader than the evidence.","rationale":"As a second-pass reviewer, I find the paper's internal logic coherent: the Floquet-Bogoliubov band overlap at ω0≈2Δ is a real resonance, the second-order calculation is a natural way to obtain the static correction, and the 1D simulation provides supporting dynamical evidence. The main quantitative prediction, however, is the 1/γ_inel divergence of n_s^(2). This divergence is obtained using Eq. (10), a constant-γ Markovian reservoir that the authors themselves flag as number-nonconserving and as 'effectively' a non-superconducting fermionic bath. The same reservoir underlies the heating estimate in Appendix D, so the two arguments do not independently corroborate the scaling. A physical phonon bath changes the self-energy in two relevant ways: it conserves fermion number, and its imaginary part is not a constant—it is set by α²F(ω) and is suppressed near the gap edge when 2Δ>ω_D, the regime identified in the quench numerics. It is therefore conceivable that the true Floquet response has a milder singularity and that the negative superfluid density is an artifact of the Dynes-like broadening. The proposed check—recomputing n_s^(2) with the full phonon self-energy, or simulating a Floquet-driven Kadanoff-Baym system with phonons—directly settles this. I do not see another assumption that is both load-bearing and more insecure; the perturbative expansion in θ and the quench-versus-Floquet mismatch are real but secondary, because the divergence is nonperturbative in γ_inel and the 1D simulation already shows correlated phase slips. Thus I agree with the reader's weakest assessment, and the conditional verdict is appropriate; it would move to accept only after the phonon-bath check is passed.","tokens_in":23034,"tokens_out":11591,"duration_ms":122356,"concrete_test":"Replace Eq. (10) in the Floquet-Usadel equation (Eq. 9) by the retarded phonon self-energy obtained from Eq. (5) with an Einstein or Debye spectral function α²F(ω) ∝ ω² θ(ω_D−ω), in the regime ω_D < 2Δ with phonon damping γ_ph, and recompute the second-order static correction n_s^(2) as γ_ph → 0. If the 1/γ_inel divergence is regularized or the sign changes, the central claim fails; if the negative divergence survives with the same parametric form, the reservoir approximation is not the origin. A complementary check is to time-integrate the full Kadanoff-Baym equations (Eq. 4) for a periodic drive Δ(t) = Δ + θ cos(2Δt) with phonons included and extract the time-averaged superfluid density after the transient, verifying that a homogeneous steady state with n_s < 0 actually exists.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central result n_s^(2)/n_s^(0) ∝ −θ²/(Δγ_inel) for ω0≈2Δ is derived from the Floquet-Usadel equation (Eq. 9) with the phenomenological self-energy of Eq. (10): a frequency-independent inelastic rate γ_inel with a Keldysh ansatz 2τ3 tanh(βω/2). This self-energy is not the physical phonon self-energy of Eq. (5); the authors themselves state that it 'effectively describes coupling to a non-superconducting fermionic reservoir' and 'explicitly breaks the fermion number conservation.' The divergent 1/γ_inel correction and the resultant negative superfluid density are direct consequences of the Dynes-like broadening this reservoir produces: it fills states inside the gap and adds a paramagnetic contribution. If the physical phonon bath—number-conserving, with spectral function α²F(ω), and with the recombination channel closed when 2Δ>ω_D—yields a self-energy whose imaginary part is not a constant but is suppressed or structured at the Floquet resonance, the 1/γ_inel pole could be replaced by a weaker divergence or cut off, and the negative n_s could disappear. The heating estimate in Appendix D uses the same constant-γ reservoir, so it cannot independently validate the scaling. The one-dimensional simulation is a quench, not a Floquet steady state, and therefore also cannot isolate the number-nonconserving-bath artifact.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies Higgs amplitude oscillations in BCS-type superconductors after a quench or under periodic modulation of the pairing amplitude. Using a diffusive Floquet-Usadel description with a Dynes-like inelastic self-energy, the authors report that for drive frequency ω0 ≈ 2Δ the static superfluid density acquires a negative correction scaling as n_s^(2)/n_s^(0) ∝ -θ²/(Δγ_inel), which they interpret as an instability of the homogeneous solution. They support this with an analytic estimate of heating, a diagrammatic expansion of the current response, and a self-consistent numerical simulation of a quasi-1D disordered BdG superconductor showing proliferation of phase slips when n_s becomes negative.","tokens_in":23351,"tokens_out":7911,"duration_ms":85421,"significance":"If correct, the result would establish a new Floquet mechanism for negative superfluid density in conventionally paired superconductors, with direct implications for time-resolved experiments and for the stability of spatially homogeneous driven superconducting states. The manuscript has genuine strengths: the Floquet-Usadel expansion is presented explicitly (Appendix A), the static correction and its γ_inel scaling are stated as concrete, falsifiable predictions, the response functions are computed without fitting the target instability, and the authors are transparent about the number-nonconserving nature of Eq. (10) and about the simplified band structure. The claimed effect is interesting and would be important for the driven-superconductivity community, but its predictive force is currently tied to a phenomenological reservoir self-energy and to a perturbative regime that may not include the instability threshold.","major_comments":[{"comment":"The central scaling n_s^(2)/n_s^(0) ∝ -θ²/(Δγ_inel) is derived using the frequency-independent reservoir self-energy Eq. (10), which the authors themselves state 'effectively describes coupling to a non-superconducting fermionic reservoir' and 'explicitly breaks the fermion number conservation.' This is not a minor modeling detail: the 1/γ_inel divergence comes from filling states at the Floquet resonance with a constant quasiparticle broadening, whereas the physical phonon self-energy Eq. (5) contains an Eliashberg spectral function α²F(ω) and does not provide a constant, number-nonconserving broadening, especially when 2Δ > ω_D makes single-phonon recombination kinematically forbidden. A concrete and necessary test is to solve Eq. (9) with Eq. (5), or with an energy-dependent γ(ω) extracted from α²F, and to check whether the negative static correction survives; until this is done, the negative-superfluid-density result is a property of the toy reservoir rather than of the phonon-coupled model advertised in the abstract.","section":"Floquet-Usadel description, Eq. (10)"},{"comment":"The second-order expansion in θ is internally consistent, but it is used outside its controlled regime. The condition for the static correction to become comparable to the equilibrium value is θ²/(Δγ_inel) ~ 1, i.e. γ_inel ~ θ²/Δ. Near the Floquet resonance the iterative solution of Eq. (A1) produces first-order sideband Green functions of order θ/γ_inel ~ Δ/θ ≫ 1, so the effective expansion parameter is not small even though θ/Δ is small. Thus negative n_s appears precisely where the perturbative series has already broken down. The manuscript needs either a nonperturbative Floquet solution, or a demonstration that a different small parameter controls the series at the instability threshold.","section":"Perturbative expansion, Appendix A"},{"comment":"The heating estimate is not independent evidence for the physical mechanism. The quoted scaling Δ^(2) ∝ θ²/√(Δγ_inel) and E_ex = θ²/√(Δγ_inel) are obtained from the same constant-γ_inel reservoir (Eq. (D1) introduces γ_inel as a pair-breaking parameter), so it cannot exclude a bath-induced regularization of the 1/γ_inel pole. Moreover, in the weak-dissipation limit the heating estimate itself diverges as γ_inel → 0, so the steady state is defined only within the same model. A physical phonon bath could suppress or restructure the singularity differently, and that scenario must be explicitly excluded before the conclusion is drawn.","section":"Appendix D"},{"comment":"The 1D simulation in Appendix C is a quench protocol (a sudden change of λ from 1.2 to 2) on a disordered tight-binding superconductor, not a Floquet steady state at ω0 ≈ 2Δ, and it does not include the reservoir self-energy Eq. (10). It therefore demonstrates that a sufficiently strong interaction quench can produce negative n_s and phase slips, but it cannot validate the specific Floquet-resonance mechanism or the abstract's 'generally unstable' conclusion. The extrapolation from a quench to periodic driving should either be stated as a conjecture or supported by a periodically driven spatial simulation.","section":"Quasi-one-dimensional case, Fig. 3"}],"minor_comments":[{"comment":"The figure references in the Floquet-Usadel section are inconsistent: the text below Eq. (10) cites Fig. 3(b), Fig. 3(a), and Fig. 3(c) for the time-averaged LDOS, the Floquet bands, and n_s^(1)(ω), but those panels appear in Fig. 2; the in-text references should be corrected to avoid confusion with the quasi-1D simulation in Fig. 3.","section":"Floquet-Usadel section, figure references"},{"comment":"The notation in the Keldysh block of Eq. (10) is typeset ambiguously; the argument of the hyperbolic tangent should be written explicitly as tanh(βω/2) rather than as tanh βω/2 with the factor attached to τ3, so that the matrix structure is unambiguous.","section":"Eq. (10)"},{"comment":"The renormalized velocity vertex θ(k) is introduced to remove the ultraviolet surface term, but the discussion would benefit from one sentence explaining why this renormalization does not change the second-order Floquet response in the energy window that controls the negative n_s correction.","section":"Appendix B"}],"recommendation":"major_revision","confidential_remarks":"The paper addresses a timely and interesting question, and the authors are commendably explicit about the limitations of their reservoir model. In my view, however, the advertised conclusion currently rests on the number-nonconserving constant-γ_inel self-energy and on a perturbative expansion that breaks down at the instability threshold. I would ask for a calculation with the physical phonon self-energy of Eq. (5), or at minimum an energy-dependent γ(ω), plus a nonperturbative consistency check, before publication. The 1D quench simulation is suggestive but does not settle the Floquet steady-state question."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things you should know before reading. The core claim is genuinely new: at drive frequency ω0 ≈ 2Δ, second-order Floquet-Usadel theory gives a static correction n_s^(2)/n_s^(0) ∝ −θ²/(Δγ_inel), diverging as the inelastic rate goes to zero. So the homogeneous driven state can acquire a negative superfluid density — a paramagnetic response and a spatial stiffness instability. Prior work had oscillatory n_s or parametric photon emission; this divergent static component is not there. The second thing: the result is only as solid as the bath it comes from, and the authors know it. Their Eq. (10) is a number-nonconserving fermionic-reservoir self-energy, not the physical phonon self-energy of Eq. (5); they say so explicitly in the text. The 1/γ_inel divergence is a direct consequence of that Dynes-like broadening.\n\nCredit where due. The calculation is internally consistent; the response functions are computed independently of the self-consistently determined order parameter, with no fitting to the target result. The Appendix D heating estimate (Δ^(2) ∝ θ²/√(Δγ_inel)) shows that within the model the negative n_s is not a heating artifact, and it does leave a window where the correction is small while θ²/(Δγ) ~ 1. The quench numerics using the physical phonon self-energy (Fig. 1b) also show negative n_s when 2Δ > ω_D, which is independent evidence the phenomenon is not purely a construction of Eq. (10). The 1D phase-slip simulation is a nice numerical experiment, and its time-correlation with negative n_s is suggestive. Minor: no code or data is shipped, so the 1D results are not independently checkable from the paper.\n\nSoft spots, in proportion. The bath-model worry is real but partial: the physical phonon model does produce negative n_s in the quench, so the fear that the effect disappears with a physical bath is not established — but it is also not refuted for the Floquet steady state, because the divergent scaling is only derived for the constant-γ reservoir. The perturbative expansion in θ is pushed to the regime θ²/(Δγ) ~ 1 where it is not controlled; the sign of the leading divergence is probably right, but the quantitative boundary of the instability is not fixed by this calculation. And the 1D simulation is a quench, not a Floquet steady state, so the abstract's 'generally unstable' runs ahead of the evidence — the paper supports instability in the studied quench regimes, not a general theorem.\n\nThis paper is for the ultrafast pump-probe community and people who worry about Floquet steady states in superconductors. It deserves a serious referee: the claim is important and the derivation is checkable. My recommendation: send it to review, and steer the referee toward one question — whether a physical, number-conserving phonon self-energy regularizes the divergence — and ask the authors to scale back the 'generally unstable' phrasing.","headline":"A genuinely new, internally consistent prediction of a divergent negative superfluid density under resonant Higgs drive, but the central scaling rests on a simplified number-nonconserving bath and the 'generally unstable' conclusion runs ahead of the evidence.","tokens_in":23861,"tokens_out":12121,"would_cite":true,"duration_ms":94508,"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":"Periodically driven superconductors can develop a negative superfluid density, which the authors read as an instability of the homogeneous state toward spatial order-parameter textures.","keywords":["superfluid density","Higgs mode","Floquet-Bogoliubov bands","time-dependent superconductivity","Usadel-Keldysh equations","phase slips","paramagnetic Meissner response","quench dynamics"],"falsifier":"Solve the same Floquet-Usadel equations with a microscopically computed, frequency-dependent phonon self-energy, such as the full Eliashberg function, instead of Eq. (10), and check whether the static correction $n_s^{(2)}/n_s^{(0)}$ still diverges as $-\\theta^2/(\\Delta\\gamma_{\\mathrm{inel}})$ at $\\omega_0=2\\Delta$; if it saturates or changes sign, the instability is an artifact of the reservoir model. Experimentally, drive a thin-film superconductor at $\\omega_0\\simeq2\\Delta$ and look for a paramagnetic Meissner response or spontaneous phase-slip textures while the drive is on.","tokens_in":1880,"feed_emoji":"🧲","tokens_out":2286,"duration_ms":104536,"temperature":0.7,"pith_summary":"The paper claims that a superconducting condensate whose pairing amplitude is periodically modulated, exciting the Higgs mode, can lose its phase stiffness while the drive is on. Near a drive frequency of twice the gap, the static correction to the superfluid density is negative and grows as $-\\theta^2/(\\Delta\\gamma_{\\mathrm{inel}})$, where $\\theta$ is the modulation amplitude and $\\gamma_{\\mathrm{inel}}$ the inelastic scattering rate, so in the weak-dissipation limit the homogeneous state has $n_s<0$. A negative superfluid density means the current responds paramagnetically rather than diamagnetically, which the authors interpret as a thermodynamic instability of the uniform solution. Their spatially resolved dynamical Bogoliubov-de Gennes simulations support this reading: order-parameter inhomogeneities and phase slips appear at the same times that the uniform calculation yields negative $n_s$. If correct, strongly driven Higgs oscillations in conventional superconductors, and analogous amplitude-driven Fermi gases, should break up into a patterned state instead of oscillating homogeneously.","feed_headline":"Driven superconductors can make superfluid density negative","feed_subtitle":"At drive frequency twice the gap, the uniform condensate loses stiffness and breaks into phase-slip textures.","key_machinery":"The machinery is the Floquet problem of the time-periodic Bogoliubov-de Gennes Hamiltonian, solved in quasiclassical Usadel-Keldysh form. The central object is the set of Floquet-Bogoliubov quasienergy bands: when the drive frequency $\\omega_0$ approaches $2\\Delta$, the first Floquet band overlaps the zeroth band near $\\xi_k=0$, producing a divergent density of Bogoliubov quasiparticle states. The superfluid density is extracted from the Keldysh current kernel, and expanding in the modulation amplitude $\\theta$ to second order, the static piece picks up the divergence and scales as $n_s^{(2)}/n_s^{(0)}\\propto -\\theta^2/(\\Delta\\gamma_{\\mathrm{inel}})$ at $\\omega_0\\approx2\\Delta$, with $\\gamma_{\\mathrm{inel}}\\to\\delta$ for finite detuning $\\delta=2\\Delta-\\omega_0>0$.","core_discovery":"The central claim is that in a superconductor driven at or near the Higgs resonance $\\omega_0\\approx 2\\Delta$, the static superfluid density acquires a negative correction that diverges as $-\\theta^2/(\\Delta\\gamma_{\\mathrm{inel}})$ in the limit of weak inelastic dissipation. The authors argue this is not a heating artifact: the heating scale is $\\theta^2/\\sqrt{\\Delta\\gamma_{\\mathrm{inel}}}$, whereas the static correction behaves as $n_s^{(2)}/n_s^{(0)}\\approx -E_{\\mathrm{ex}}/\\sqrt{\\Delta\\gamma_{\\mathrm{inel}}}$, so the negative sign tracks a divergent Floquet-Bogoliubov density of states rather than quasiparticle heating. They conclude that the homogeneous time-dependent superconducting solution is generally unstable when the Floquet bands overlap, and that the system relaxes by forming a complicated spatial landscape of topological excitations, specifically phase slips. This is new because negative superfluid density is usually associated with exotic states such as Fulde-Ferrell-Larkin-Ovchinnikov phases or odd-frequency pairing; the paper finds it in an amplitude-driven conventional $s$-wave superconductor.","pith_inferences":["If this mechanism is generic, other time-dependent mean-field ordered phases with resonant amplitude modes, such as charge-density waves or spin-density waves, could show an analogous Floquet-band-overlap instability; the paper studies only $s$-wave superconductors, so this is an extrapolation.","The Floquet analysis fixes $\\Delta(t)=\\Delta+\\theta\\cos\\omega_0 t$ while the spatially resolved numerics are self-consistent; a full self-consistent two-dimensional treatment could reveal whether the instability selects periodic stripe patterns or proceeds as a disordered avalanche of phase slips.","Replacing the phonon bath by a frequency-independent fermionic reservoir is the main modeling step, and computing $n_s$ with a microscopically derived, frequency-dependent self-energy would test whether the $1/\\gamma_{\\mathrm{inel}}$ divergence survives in a realistic phonon bath.","Appendix B itself flags an unphysical surface contribution to the polarization from the simplified parabolic band with a hard cutoff, removed by renormalizing the velocity vertices; repeating the calculation with a lattice dispersion would confirm the quoted $n_s$ scaling."],"forward_implications":["Near $\\omega_0\\simeq 2\\Delta$, a driven homogeneous superconductor cannot persist; the uniform amplitude-oscillating state should spontaneously develop spatial inhomogeneities and phase slips.","The instability is strongest in the weak-dissipation limit $\\gamma_{\\mathrm{inel}}\\to 0$, where the negative static correction to $n_s$ diverges, and increasing inelastic scattering suppresses it.","Finite detuning gives the same scaling with $\\gamma_{\\mathrm{inel}}\\to\\delta$, so the effect survives off resonance.","In ultracold Fermi gases with interaction-modulation-induced Higgs oscillations, analogous spatial patterning should occur rather than only uniform amplitude dynamics.","The sign of the static $n_s$ correction is a diagnostic for the stability of the time-dependent mean-field state: when it turns negative, the homogeneous solution breaks down."],"supporting_citations":[{"why":"Supplies the scenario in which electron-hole recombination is forbidden, the limit where the Floquet-Bogoliubov density of states can diverge.","marker":"[13]"},{"why":"Provides the conventional quasiclassical Keldysh representation of the Green's function used to write the Usadel equation, Eq. (9).","marker":"[18]"},{"why":"Supplies the simplified fermionic-reservoir self-energy form, Eq. (10), with the single inelastic rate $\\gamma_{\\mathrm{inel}}$ on which the divergent correction depends.","marker":"[19–22]"},{"why":"Gives the Keldysh-Usadel expression for the supercurrent from which the superfluid density is extracted.","marker":"[23]"}],"fun_headline_variants":["Superfluid density goes negative in driven superconductors","At Higgs drive, superfluid density flips sign and breaks order","Negative superfluid density triggers phase-slip textures","Uniform superconductor loses stiffness at twice-gap drive","Driven superconductor: negative density, then spatial breakup"],"cache_read_input_tokens":25856,"weakest_assumption_plain":"The load-bearing premise is that a phonon bath can be represented by an energy-independent fermionic reservoir with one inelastic rate $\\gamma_{\\mathrm{inel}}$ (Eq. 10); that model explicitly breaks number conservation, and if a realistic phonon bath does not produce the same divergence at the Floquet resonance, the negative superfluid density could disappear.","fun_headline_variants_meta":{"raw":{"variants":["Superfluid density goes negative in driven superconductors","At Higgs drive, superfluid density flips sign and breaks order","Negative superfluid density triggers phase-slip textures","Uniform superconductor loses stiffness at twice-gap drive","Driven superconductor: negative density, then spatial breakup"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000325,"raw_usage":{"total_tokens":1830,"prompt_tokens":963,"completion_tokens":867,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":579,"completion_tokens_details":{"reasoning_tokens":787}},"tokens_in":579,"tokens_out":867,"duration_ms":10683,"temperature":1.0,"reasoning_tokens":787,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T20:29:11.034105+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Solve the same Floquet-Usadel equations with a microscopically computed, frequency-dependent phonon self-energy, such as the full Eliashberg function, instead of Eq. (10), and check whether the static correction $n_s^{(2)}/n_s^{(0)}$ still diverges as $-\\theta^2/(\\Delta\\gamma_{\\mathrm{inel}})$ at $\\omega_0=2\\Delta$; if it saturates or changes sign, the instability is an artifact of the reservoir model. Experimentally, drive a thin-film superconductor at $\\omega_0\\simeq2\\Delta$ and look for a paramagnetic Meissner response or spontaneous phase-slip textures while the drive is on.","supporting_citations":[{"cited_title":"Barankov and L","cited_arxiv_id":null,"evidence_quote":"Provides the conventional quasiclassical Keldysh representation of the Green's function used to write the Usadel equation, Eq. (9)."},{"cited_title":"Virtanen, T","cited_arxiv_id":null,"evidence_quote":"Gives the Keldysh-Usadel expression for the supercurrent from which the superfluid density is extracted."}],"review_version":1}