{"id":"46f6b8b9-7e83-4dd8-baeb-107d6d100e0b","arxiv_id":"2506.07788","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"The authors derive Kubo-type linear response and Fermi's golden rule for Floquet systems in the wave-function picture, showing that photoemission current is proportional to the l=0 component of the Floquet Green function.","lead":"This paper develops a linear response theory for many-electron systems in a time-periodic drive, with a weak non-periodic probe added on top. It derives a Floquet Fermi golden rule and connects photoemission spectra to a positive-definite Floquet spectral function.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The asserted equivalence of Eq. (60) to the imaginary part of the l=0 Floquet Green function is not derived: the golden-rule photocurrent is a particle-removal rate, while the retarded Green function also contains addition contributions.","rationale":"The reader's weakest assumption focuses on the sudden approximation and the pure-Floquet initial state. Those are real limitations, but the paper explicitly identifies the sudden approximation and states that a Floquet mode is chosen as the initial state. My concern is different and more directly about the central claim: even within those assumptions, the step from the golden-rule removal rate in Eq. (60) to the l=0 component of the retarded Floquet Green function is asserted, not derived. The retarded Green function includes both removal and addition spectral weight, while the golden-rule expression is manifestly removal-only. In equilibrium photoemission, the addition part is Pauli-blocked below the chemical potential, but for a generic Floquet state the corresponding separation is not demonstrated and can fail because quasienergies repeat with period Omega. This is a load-bearing gap because the paper's headline result is exactly the proportionality of the photocurrent to the positive-definite l=0 Floquet spectral function. The proposed Lehmann-representation check on the Appendix A model would settle whether the identification holds or whether an additional occupation/filtering assumption is needed. Taking the concern seriously does not require abandoning the paper; it requires adding a proof or restricting the claim, which is consistent with the existing CONDITIONAL verdict.","tokens_in":9098,"tokens_out":26230,"duration_ms":331814,"concrete_test":"Use the analytically solvable driven homogeneous electron gas in Appendix A (or a one-band Floquet model) with a fixed N-particle Floquet state |Phi_alpha>. Compute I(epsilon_k) from Eq. (60) and separately compute the l=0 component of the retarded Floquet Green function -Im G^R_{00}(k,omega) from its Lehmann representation, keeping both N -> N-1 and N -> N+1 terms. If the two functions differ in any frequency window, Eq. (60) is not proportional to -Im G^R_{00}; the paper must then either prove an occupation-factor identity or restrict the claim to the removal (lesser) Green function. This test is parameter-free and can be done analytically for Appendix A.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section III.C derives the photocurrent as a golden-rule transition rate between an N-particle Floquet mode and a factorized (N-1)-electron Floquet mode plus a free photoelectron. The resulting Eq. (60) is a sum over removal matrix elements |f^{k alpha gamma}_m|^2 with delta functions. The paper then states that this right-hand side is proportional to the imaginary part of the l=0 component of the Floquet Green function. That identification is not shown. For a retarded/anti-commutator Floquet Green function, -Im G^R_{00}(k,omega) has both removal (N -> N-1) and addition (N -> N+1) Lehmann contributions; Eq. (60) contains only the removal part. In an equilibrium ground state the addition part vanishes for omega below the chemical potential, which justifies the photoemission identification. For a generic pure Floquet initial state, no such separation is established; Floquet copies repeated by Omega can put addition and removal resonances at the same omega. Thus the central claim 'photocurrent proportional to -Im G^R_{00}' is an unproven assertion unless the occupation/addition filtering is supplied. The sudden approximation and the assumption of a pure Floquet initial state are further restrictions, but even granting them, the Green-function identification in Eq. (60) does not follow from the preceding derivation.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper develops a linear-response formalism for many-body Floquet systems using an interaction (Floquet) picture in which the time-periodic Hamiltonian H0(t) is treated exactly and a non-periodic probing field Ξ(t) is treated to first order. It derives a retarded density-response function (Eq. 33), a Floquet Fermi golden rule (Eqs. 49-51), and, under the sudden approximation, a formula for the photocurrent from an N-electron Floquet state (Eq. 60). The authors then claim that this photocurrent is proportional to the imaginary part of the l=0 component of the Floquet Green function, and they illustrate the formalism with calculations for a noninteracting electron gas driven by a traveling-wave potential.","tokens_in":9319,"tokens_out":8550,"duration_ms":103509,"significance":"If the central identification between the photocurrent and the Floquet Green function is valid, the paper would provide a useful formal bridge between time-resolved photoemission of driven materials and the Floquet spectral function, whose positivity was established in Ref. 16. The linear-response derivation in Sec. III is internally consistent and parameter-free, and the Floquet golden rule is a clean extension of standard results. However, the key link to the retarded Green function is asserted rather than derived, and this gap is load-bearing for the main conclusion. The numerical application in Sec. IV is also too terse to reproduce. The paper is promising, but the central claim needs to be either proven or explicitly qualified.","major_comments":[{"comment":"The identification of Eq. (60) with the imaginary part of the l=0 component of the Floquet Green function is not established by the preceding derivation. Equation (60) sums over the removal amplitudes f^{kαγ}_m, i.e., transitions from the N-particle Floquet state |Φ_α(t)> to (N−1)-particle states. The retarded Floquet Green function, however, contains both removal (N→N−1) and addition (N→N+1) Lehmann contributions. In a generic pure Floquet state there is no equilibrium-like particle-hole separation, because quasienergies are only defined modulo Ω; an addition resonance at ω = E^{(N+1)}_δ − E^{(N)}_α + nΩ can coincide with a removal resonance at ω = E^{(N)}_α − E^{(N−1)}_γ + mΩ for suitable integers n,m. Therefore, without an additional occupation/filtering argument or an explicit restriction to a state in which the addition terms vanish, the claim that the photocurrent is proportional to −Im G^R_{00} does not follow. The authors should derive the relation to the lesser (particle-removal) Floquet Green function and state under which conditions it equals −Im G^R_{00}, or amend the claim.","section":"Sec. III.C, Eq. (60)"},{"comment":"The sudden-approximation factorization |Φ_β(t)> = c_k^† |Φ_γ^{N−1}(t)> is an extra physical input beyond the perturbation expansion in Sec. III.B. It is valid only when the photoelectron kinetic energy is large compared to the binding energy; the paper acknowledges this in the text but does not state the resulting quantitative limitation on Eq. (60). Because this factorization is what converts the general golden rule into a single-particle removal rate, the scope of the photoemission result — for example, its failure at low photoelectron kinetic energies or for strongly correlated initial states — should be stated explicitly in the conclusion.","section":"Sec. III.C, Eq. (57)"},{"comment":"The illustrative calculation for the traveling-wave driving field V(x,t)=V0 cos(κx−Ωt) is described only verbally. The 'combined Bloch-Floquet formalism' is not presented, and no equations are given for the Green function, the photocurrent, or the numerical procedure used to produce Fig. 2. As a result, the numerical results cannot be reproduced or checked. Please provide the defining equations, the Brillouin-zone parameters, and the discretization/integration scheme, and clarify how Eq. (60) is used to compute I(ε_k).","section":"Sec. IV"}],"minor_comments":[{"comment":"There is a typo: 'sates' should be 'states'. Also, the time t0 in Eq. (34) is not explicitly defined; it should be stated that the initial Floquet mode is taken at t0.","section":"Sec. III.A, Eq. (34)"},{"comment":"The Fourier index for the b amplitude is written as n in Eqs. (44)-(47), although the conjugate expansion in Eq. (43) carries −m. The reindexing should be made explicit; otherwise the cross-term condition n = −m − 2M in Eq. (51) is difficult to follow.","section":"Sec. III.B, Eqs. (40)-(51)"},{"comment":"The condition for non-vanishing cross terms should be re-derived. As written, the two delta-function conditions in the long-time limit appear to require simultaneous resonance conditions that are inconsistent with the Fourier indices in Eqs. (40)-(47).","section":"Sec. III.B, Eq. (51)"},{"comment":"There is a typo: 'band gand' should be 'band gap'. Also, the statement that the low-frequency limit 'returns to a usual time-independent Bloch system' should be qualified, since the drive amplitude V0 remains nonzero and only the spatial periodicity dominates.","section":"Sec. IV.B"},{"comment":"The vertical axis is labeled only as I(a.u.); please specify the normalization of the photocurrent and, if possible, the units of the horizontal axis relative to the Fermi energy.","section":"Sec. IV, Fig. 2"},{"comment":"The sentence 'exp(−iεαT) can be regarded as a one-dimensional irreducible representation of the element T of the Abelian time translation group' is standard but would benefit from a more precise statement, since the representation is of the cyclic subgroup generated by T rather than the full time-translation group.","section":"Sec. II"}],"recommendation":"major_revision","confidential_remarks":"The paper has a sound perturbation-theoretic core, and the Floquet golden rule is a useful contribution. The main concern is that the central photoemission claim is not derived: Eq. (60) is a removal-only rate, while the retarded Floquet Green function also contains addition terms, and the Floquet Brillouin zone makes removal and addition resonances potentially degenerate. This is fixable by deriving the appropriate lesser Green function relation, but it is a load-bearing point and cannot be waved through by citing Ref. 16 alone. The numerical section is also too under-specified for a quantitative application. If the authors supply the missing derivation and clarify the scope, the paper would be suitable for publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things to know. First, the linear-response and golden-rule derivations in Sections III.A and III.B are sound and clearly presented; they give a wave-function path to results that are otherwise usually derived in Green-function language. Second, the photoemission punchline—Eq. (60) being proportional to the imaginary part of the l=0 Floquet Green function—is asserted, not proved, and the gap is load-bearing.\n\nWhat's new: the wave-function derivation itself is a legitimate alternative to Refs 17,18, and the explicit photoemission identification (if it can be made to work) would be useful for trARPES. The paper also correctly reproduces the sideband structure and the Umklapp terms when 2Δβα/Ω is an integer. Those parts deserve credit.\n\nThe soft spot: Section III.C derives a removal-only golden-rule rate. The right-hand side of Eq. (60) has no addition contributions. The paper then jumps to −Im G^R_00. For an equilibrium ground state, the addition channel is empty below the chemical potential and the identification works; for a generic pure Floquet state with quasienergy in the first Floquet zone, there is no argument that separates removal and addition resonances, and Floquet copies can bring them to the same ω. The cited positive-definiteness result (Ref 16) does not fill that gap, because it concerns the spectral function itself, not the photocurrent's removal-only content. The sudden approximation and pure-state assumption are secondary—minor if the identification were proven, but they narrow the experimental claim.\n\nThe numerical illustration is also not reproducible from the text: the 'combined Bloch-Floquet formalism' is never defined, so Fig. 2 is a black box.\n\nOverall: the formal spine is correct; the photoemission identification needs real work, either a derivation or a careful statement that the result is a removal spectral function that equals −Im G^R only when addition is forbidden. The paper deserves a serious referee—not a desk reject, but a major-revision request.","headline":"Sound wave-function derivation of Floquet linear response and golden rule, but the photoemission identification with -Im G^R_00 is asserted rather than derived, and the numerics are not reproducible.","tokens_in":9941,"tokens_out":3495,"would_cite":false,"duration_ms":35650,"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 photocurrent in a Floquet photoemission experiment is proportional to the imaginary part of the $l=0$ component of the Floquet Green function.","keywords":["Floquet theory","linear response","Fermi golden rule","photoemission spectrum","Floquet Green function","spectral function","periodically driven systems","trARPES"],"falsifier":"A direct calculation of the full photoemission current for a small driven system, for example a few-site driven Hubbard chain, that keeps the photoelectron coupled to the remaining electrons during escape, compared with Eq. (60), would settle the sudden approximation: any sideband that shifts or gains weight beyond the predicted positions $\\omega - \\epsilon_k - m\\Omega - (E_\\gamma^{N-1} - E_\\alpha)$ would falsify the reduction of the photocurrent to the $l=0$ Floquet Green-function component.","tokens_in":8846,"feed_emoji":"⚛️","tokens_out":9518,"duration_ms":103392,"temperature":0.7,"pith_summary":"This paper aims to put linear response of periodically driven many-electron systems on the same footing as equilibrium Kubo theory. It treats a strong time-periodic drive exactly and a second, weak, non-periodic probe to first order, and derives a Floquet Fermi golden rule: transition rates peak at frequency differences $\\omega + n\\Omega + \\Delta$. Applied to photoemission under the sudden approximation, the result ties the measured photocurrent to the imaginary part of a specific component ($l=0$) of the Floquet Green function, so time-resolved ARPES spectra of driven materials can be read as a positive spectral function. The paper illustrates the formalism on a driven electron gas and shows how drive parameters control the sideband structure.","feed_headline":"Photocurrent of a driven solid maps to one Floquet spectral component","feed_subtitle":"Floquet golden rule puts photoemission sidebands at ω+nΩ+Δ; trARPES reads the driven spectrum directly.","key_machinery":"The working tool is the Floquet picture, an interaction picture built from the evolution operator of the periodic Hamiltonian alone, so that the weak probe enters as $\\hat{\\Xi}_F(t) = \\hat{U}_0(t_0,t)\\hat{\\Xi}(t)\\hat{U}_0(t,t_0)$. The argument is carried by expanding the periodic Floquet-mode matrix elements $\\langle\\Phi_\\beta(t)|\\hat{\\Xi}|\\Phi_\\alpha(t)\\rangle$ in Fourier series $\\sum_n f^{\\beta\\alpha}_n e^{in\\Omega t}$; the time integrals then collapse to delta functions at $\\omega + n\\Omega + \\Delta_{\\beta\\alpha}$. The photoemission result additionally uses the sudden approximation, factorizing the final Floquet mode as $\\hat{c}^\\dagger_k |\\Phi^{N-1}_\\gamma(t)\\rangle$, which reduces the golden rule to Eq. (60). This decomposition is what makes the photocurrent equal to a definite component of the Floquet Green function.","core_discovery":"In the Floquet picture, the paper's central result is Eq. (60): the photocurrent $I(\\epsilon_k)$ is proportional to $\\sum_{\\gamma m} |f^{k\\alpha\\gamma}_m|^2 \\delta(\\omega - \\epsilon_k - m\\Omega - (E_\\gamma^{N-1} - E_\\alpha))$, and this expression is proportional to the imaginary part of the $l=0$ component of the Floquet Green function. Since that component is positive definite, the photoemission spectrum is a genuine spectral function rather than a signed quantity. The Floquet generalization of Fermi's golden rule that precedes it shows transitions occur at $\\omega + n\\Omega + \\Delta_{\\beta\\alpha}$, with extra cross-term resonances when $2\\Delta_{\\beta\\alpha}/\\Omega$ is an integer, which the paper calls Umklapp in frequency space. The application to a periodically driven noninteracting electron gas illustrates how the ratio of the drive phase velocity $\\Omega/\\kappa$ to the Fermi velocity $v_F$ selects between Floquet sidebands and Bloch gap physics.","pith_inferences":["Iterating the Floquet-picture perturbation expansion beyond first order would yield nonlinear response tensors for periodically driven systems, connecting this Kubo analogue to pump-probe and harmonic-generation experiments.","Because only the $l=0$ component of the Floquet Green function enters the photoemission formula, angle-integrated spectra may hide Floquet physics living in off-diagonal frequency sectors; phase-resolved or two-time measurements would be needed to expose those sectors.","A density-matrix version of the Floquet golden rule would extend the result to thermal and mixed-state initial conditions, a natural next step since the paper explicitly starts from a pure Floquet mode."],"forward_implications":["Time-resolved ARPES on a periodically driven material can be interpreted directly: the measured photocurrent at each momentum is proportional to the positive $l=0$ Floquet spectral function.","The Floquet Fermi golden rule places photoemission sidebands at $\\omega - \\epsilon_k - m\\Omega - (E_\\gamma^{N-1} - E_\\alpha)$; drive frequency and amplitude therefore tune sideband positions and weights, a route to Floquet engineering of spectra.","When $2\\Delta_{\\beta\\alpha}/\\Omega$ is an integer, interference between absorption and emission amplitudes produces additional Umklapp resonances in frequency space that should be observable as extra lines.","In the high-frequency limit the spectrum shows Floquet sidebands, in the low-frequency limit a Bloch gap of size approximately $2V_0$ appears, and in the intermediate regime the traveling-wave drive breaks Kramers degeneracy; the ratio $\\Omega/\\kappa$ versus $v_F$ controls which regime is realized."],"supporting_citations":[{"why":"Establishes positivity of the $l=0$ component of the Floquet Green function, which the paper's photoemission result relies on for a positive-definite spectrum.","marker":"16"},{"why":"Provides the Floquet-theory background and the expression for the evolution operator in terms of Floquet modes used to define the Floquet picture.","marker":"3"},{"why":"Supplies the minimal-coupling replacement of the momentum operator by $p - (e/c)A$, which generates the light-matter perturbation treated by the golden rule.","marker":"19"},{"why":"Introduces the sudden approximation in which the photoelectron is decoupled from the remaining $N-1$-electron system, the load-bearing step for Eq. (60).","marker":"20"},{"why":"Presents the nonequilibrium Green-function formalism that the paper's wave-function linear response is an alternative to.","marker":"17"},{"why":"Provides an earlier Floquet many-body Green-function formulation that motivates the response-theoretic setup.","marker":"18"}],"fun_headline_variants":["Floquet golden rule yields positive photoemission spectral function","Driven solid photocurrent is one Floquet spectral piece","Floquet linear response: photoemission is positive spectral function","Floquet golden rule: photocurrent is one spectral component","Photocurrent of driven solid: one Floquet spectral function"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is the sudden approximation, that the outgoing photoelectron's kinetic energy is large enough that it separates cleanly from the remaining $N-1$-electron Floquet system, together with the assumption that the initial state is a single pure Floquet mode, so mixed states and finite temperature are outside the claimed result.","fun_headline_variants_meta":{"raw":{"variants":["Floquet golden rule yields positive photoemission spectral function","Driven solid photocurrent is one Floquet spectral piece","Floquet linear response: photoemission is positive spectral function","Floquet golden rule: photocurrent is one spectral component","Photocurrent of driven solid: one Floquet spectral function"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000691,"raw_usage":{"total_tokens":3104,"prompt_tokens":899,"completion_tokens":2205,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":515,"completion_tokens_details":{"reasoning_tokens":2115}},"tokens_in":515,"tokens_out":2205,"duration_ms":17458,"temperature":1.0,"reasoning_tokens":2115,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T05:26:11.037371+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A direct calculation of the full photoemission current for a small driven system, for example a few-site driven Hubbard chain, that keeps the photoelectron coupled to the remaining electrons during escape, compared with Eq. (60), would settle the sudden approximation: any sideband that shifts or gains weight beyond the predicted positions $\\omega - \\epsilon_k - m\\Omega - (E_\\gamma^{N-1} - E_\\alpha)$ would falsify the reduction of the photocurrent to the $l=0$ Floquet Green-function component.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes positivity of the $l=0$ component of the Floquet Green function, which the paper's photoemission result relies on for a positive-definite spectrum."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the minimal-coupling replacement of the momentum operator by $p - (e/c)A$, which generates the light-matter perturbation treated by the golden rule."},{"cited_title":"Fadley ,\\ https://doi.org/10.1016/0009-2614(74)89123-2 journal journal Chemical Physics Letters \\ volume 25 ,\\ pages 225 ( year 1974 ) NoStop","cited_arxiv_id":null,"evidence_quote":"Introduces the sudden approximation in which the photoelectron is decoupled from the remaining $N-1$-electron system, the load-bearing step for Eq. (60)."}],"review_version":1}