{"id":"6ace7d15-3df8-4f5e-9d74-9bd4657d3fb0","arxiv_id":"2412.00619","paper_version":2,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"A periodically driven stacked-kagome lattice is shown to host a Chern vector (1,1,-1) with unidirectional surface states on all surfaces.","lead":"The authors propose a three-dimensional lattice model with time-periodic hopping that breaks time-reversal symmetry and predicts a topological Chern vector. They report chiral surface states on all faces, suggesting a Floquet route to three-dimensional topological insulators without static magnetic fields.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The claimed Chern vector is never computed for the actual time-periodic system; at λ^(1)/ω_F=0.48 the 1/ω_F Floquet-Magnus and first-harmonic truncations are uncontrolled, and the paper's Δϕ=π/3 versus π/6 ambiguity makes the prediction non-unique.","rationale":"I read the paper as making two separable claims: the original time-periodic model realizes a Chern-vector insulator with C=(1,1,-1), and its surfaces host unidirectional states. The surface-state simulation is direct evidence for the second claim, but the quantitative invariant is derived solely from approximate effective Hamiltonians. The reader's weakest assumption points at the same gap. I agree with that assessment and sharpen it: the decisive missing calculation is a converged Floquet Chern-number computation for the actual H(t), because the high-frequency criterion fails by a factor of two and the phase convention is internally inconsistent. The Fig. 4 propagation does not substitute for this, since a finite-lattice wave packet can be chiral-looking without the invariant being the stated one. I therefore keep the reader's conditional verdict: the paper should be accepted only after a full-Floquet Chern computation and a clarification of the phase-delay convention.","tokens_in":14844,"tokens_out":11644,"duration_ms":112940,"concrete_test":"Build the full Floquet Hamiltonian of Eq. (10) truncated to harmonics n=-N_h,...,+N_h and, for N_h=10,20,40, compute the quasienergy spectrum at the Sec. V parameters (m=100, λ^(0)=40, λ^(1)=120, λ_z=4, ω_F=250). For the gap near ε≈140, compute the three Chern numbers from Eq. (17) on the modes below the gap, first with Δϕ=π/6 and then with Δϕ=π/3 as Eq. (5) literally specifies. Also recompute Eq. (16) using the exact high-frequency coefficient (1/ω_F)[H^(-1),H^(1)] without any extra 1/2 rescaling. If the converged Chern vector is (1,1,-1) for the convention actually used in the code, the concern is resolved; if it differs or fails to converge, the central claim is not established.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim C=(1,1,-1) for the time-periodic model in Eq. (3) is established only through two approximate Hamiltonians: the 3×3 first-order Floquet-Magnus Hamiltonian in Eqs. (12)-(13) and the 9×9 first-harmonic truncation in Eq. (11). Neither is the full Floquet Hamiltonian of Eq. (10), and the paper never checks convergence of the Chern numbers with respect to the harmonic cutoff. The chosen parameters in Sec. V give λ^(1)/ω_F=120/250=0.48 and λ^(0)/ω_F=0.16, outside the regime ω_F ≫ λ used to justify the expansion, so omitted O(1/ω_F^2) terms and neglected diagonal corrections in the commutator could shift the gap or the Berry curvature. The two approximations shown in Fig. 2(b,c) already produce visibly different band structures, and no calculation ties either to the exact quasienergy spectrum. The phase-delay convention is also not uniquely specified: Eq. (5) with N=6 gives Δϕ=π/3, while Sec. IV and Fig. 1 use Δϕ=π/6; since Eq. (16) contains |sinΔϕ|, these are different models with different effective hoppings. The direct time-domain propagation in Fig. 4 of the original H(t) is good evidence for unidirectional surface dynamics, but it does not by itself certify the three Chern numbers, so the quantitative invariant in the abstract remains unsupported.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript proposes a three-dimensional tight-binding model on a modified stacked kagome lattice with phase-delayed time-periodic hopping amplitudes. The authors argue that this time modulation breaks time-reversal symmetry and induces an effective gauge field with nonzero flux projections in all three spatial directions, yielding a Floquet Chern vector C=(1,1,-1). The paper analyzes the model through a first-harmonic (9x9) truncation and a Floquet-Magnus (3x3) effective Hamiltonian, computes Chern numbers for each, and reports unidirectional chiral surface states on all open surfaces, including a direct time-domain simulation with defects. The central quantitative claim is that the time-periodic model hosts a Chern vector topological insulator with a nonzero invariant in every spatial direction.","tokens_in":15246,"tokens_out":3265,"duration_ms":34252,"significance":"If the central claim is correct, the work offers a potentially significant route to three-dimensional Chern vector insulators in classical wave systems using time modulation instead of external magnetic fields or rotation. The direct numerical propagation in Fig. 4 provides concrete evidence of chiral surface dynamics and defect-immune transport, and the manuscript is clearly written with useful schematic figures. However, the paper's central invariant is not presently established for the actual time-periodic system: the analytic derivation contains a phase convention inconsistency and an incomplete Floquet-Magnus commutator, and the Chern numbers are computed only for approximate Hamiltonians under parameter conditions outside the stated high-frequency regime. These issues make the quantitative prediction C=(1,1,-1) unsupported as it stands.","major_comments":[{"comment":"The phase delay convention is internally inconsistent. Equation (5) defines Δϕ = 2π/N with N=6, giving Δϕ = π/3, but the text in Section IV and the subsequent discussion set Δϕ = π/6 and state that the effective hopping phase is π/2. Because Eq. (16) depends on |sin Δϕ|, these are distinct models with different effective hoppings, so the computed Chern vector C=(1,1,-1) is not uniquely assigned to the Hamiltonian in Eq. (3). The authors should adopt a single convention, state it unambiguously, and verify that the topology is unchanged if π/3 is intended.","section":"§II, §IV, Eq. (5) and Eq. (16)"},{"comment":"The first-order Floquet-Magnus correction is written as (1/ω_F)[H^{(-1)}, H^{(1)}] in Eq. (13), but the standard expansion contains a prefactor 1/2, so there is a missing factor of 1/2 in the effective Hamiltonian. In addition, H^{(-1)} and H^{(1)} are off-diagonal 3x3 matrices, and their commutator generically has nonzero diagonal entries; Eq. (16) displays only the off-diagonal nearest-neighbor and next-nearest-neighbor parts and gives no argument that the diagonal (on-site) terms vanish for this model. Both omissions change the effective Hamiltonian quantitatively, and the paper does not show that the Chern numbers are unaffected by these corrections.","section":"§IV, Eq. (13) and Eq. (16)"},{"comment":"The Chern vector C=(1,1,-1) is computed only for the 3x3 Floquet-Magnus Hamiltonian and the 9x9 first-harmonic truncation, never for the full Floquet Hamiltonian of Eq. (10). With the parameters in Section V, λ^(1)/ω_F = 120/250 = 0.48 and λ^(0)/ω_F = 0.16, which are not in the high-frequency regime ω_F >> λ used to justify the Floquet-Magnus expansion. The two approximate band structures in Figs. 2(b) and 2(c) visibly differ, and no convergence check with respect to the harmonic cutoff is reported. The direct time-domain simulation in Fig. 4 demonstrates chiral surface propagation but does not establish the bulk Chern invariant for the actual time-periodic Hamiltonian in Eq. (3).","section":"§V, Figs. 2(b)-(c) and Eq. (10)"}],"minor_comments":[{"comment":"There is a typo in 'the Plank's constant is set to ℏ = 1' — it should be 'Planck's constant.'","section":"§II"},{"comment":"The caption labels two panels as '(c)': one describing the Brillouin zone and one illustrating the closed loop with effective magnetic flux; the latter should be '(d)' and subsequent panel references adjusted.","section":"Fig. 1 caption"},{"comment":"The expression for the surface local density of states is ambiguous: the factor Γ should be placed clearly, e.g., as Γ/[π((ω-ω_i)^2 + Γ^2)], rather than having Γ appear in the numerator of a large fraction with unclear grouping.","section":"Eq. (18)"},{"comment":"The manuscript uses 'time-reversal' and 'time-reversal symmetry' interchangeably; for precision, the symmetry should be referred to consistently as 'time-reversal symmetry' where a symmetry operation is meant.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"The paper cites an arXiv preprint (Ref. [26]) that appears directly relevant to Chern vector insulators; the authors should clarify in the revision how their Floquet construction differs from or extends that work. The load-bearing issues are the Δϕ ambiguity, the missing 1/2 and diagonal terms in the Floquet-Magnus commutator, and the absence of a direct computation of Chern numbers for the full Floquet Hamiltonian with controlled convergence. These are fixable within the manuscript's scope, so I recommend major revision rather than rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is a genuine new model—phase-delayed Floquet driving on a modified stacked kagome lattice, engineered so the effective gauge field has nonzero projections in all three directions, with direct time-domain simulations of chiral surface states that go around defects. If the Floquet analysis were tight, it would be a solid contribution to topological metamaterials. As it stands, the headline Chern vector C=(1,1,-1) is not actually computed for the time-dependent system; it's computed for two truncated effective Hamiltonians, and neither truncation is controlled at the parameters used.\n\nWhat the paper does well: the lattice construction is clear, the closed-loop flux argument is intuitive, and the direct simulation of H(t) in Fig. 4 is the right check—you see unidirectional surface propagation and no backscattering at the defects. That is real evidence, more than many Floquet papers show. The fact that the two truncated approximations give the same Chern vector is mildly reassuring, but both are truncations of the same type.\n\nThe soft spots are all on the Floquet side. First, the phase convention is inconsistent: Eq. (5) with N=6 gives Δϕ=π/3, while Sec. IV and Fig. 1 use π/6, and the effective Hamiltonian depends on |sin Δϕ|, so these are different models. Second, the Floquet-Magnus correction in Eq. (13) lacks the 1/2 factor that appears in the standard expansion, and the commutator in Eq. (16) drops diagonal terms that a commutator of off-diagonal matrices generally produces. Third, the parameters λ^(1)/ω_F = 120/250 = 0.48 are not deep in the high-frequency regime, so the 1/ω expansion is not obviously controlled, and the 9×9 first-harmonic truncation is not tested against higher harmonics. The paper never shows that the Chern number converges with harmonic cutoff or matches the exact quasienergy spectrum. The surface-state propagation is good evidence for chirality, but it does not by itself certify three bulk Chern numbers.\n\nThese are fixable: compute the Chern numbers from the full Floquet Hamiltonian with sufficient harmonic truncation, check convergence, fix the phase convention and the missing factor. Who is this for? People working on Floquet topological phases in classical metamaterials will find the model attractive and the simulation useful. It deserves a serious referee, but not acceptance as is. I would send it out with a request to tighten the Floquet calculation before publication.","headline":"A promising Floquet Chern-vector model whose headline invariant is computed only in uncontrolled truncations; the direct surface-state simulation is real, but the bulk topology needs to be checked against the full Floquet spectrum.","tokens_in":15696,"tokens_out":2458,"would_cite":false,"duration_ms":31354,"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":"This paper proposes that phase-delayed temporal-periodic hopping in a three-dimensional modified stacked kagome lattice produces a Floquet Chern vector C=(1,1,-1), so the insulator should host unidirectional, backscattering-free surface…","keywords":["Chern vector","Floquet topological insulator","time-reversal symmetry breaking","three-dimensional tight-binding model","stacked kagome lattice","chiral surface states","time-periodic hopping","gauge field"],"falsifier":"Compute the exact quasienergy spectrum of the full time-dependent Hamiltonian H(t) over one driving period in the same parameter regime and integrate the Berry curvature of the five bands below the target gap; if the resulting triplet is not (1,1,-1), or if the gap closes anywhere in the Brillouin zone, the effective-Hamiltonian prediction fails. A simpler check is to repeat the supercell surface-state calculation at omega_F = 2500, deep in the assumed high-frequency regime, and compare the Chern numbers and surface dispersions with those at omega_F = 250.","tokens_in":14665,"feed_emoji":"🌀","tokens_out":5719,"duration_ms":124614,"temperature":0.7,"pith_summary":"The paper argues that a three-dimensional tight-binding lattice with time-periodically modulated hopping, whose phases advance by a fixed angle bond by bond, can effectively produce a magnetic-flux-like gauge field with nonzero projections in all three spatial directions. In the Floquet description this yields three Chern numbers that together form the Chern vector C=(1,1,-1). If the argument is right, the system is a three-dimensional Chern vector topological insulator, distinct from a trivial stack of two-dimensional Chern insulators, with chiral surface states on every face rather than only on horizontal surfaces. The authors support this with Bloch-Floquet analysis, a high-frequency Floquet-Magnus effective Hamiltonian, and numerical wave-packet simulations of the original time-dependent model.","feed_headline":"Phase-delayed hopping yields Chern vector (1,1,-1) on all faces","feed_subtitle":"A time-modulated lattice breaks time-reversal symmetry and predicts chiral, defect-proof surface transport in all directions.","key_machinery":"The central object is the phase-delayed time-periodic hopping amplitude lambda_i(t) = $\\lambda$^(0) + $\\lambda$^(1) cos(omega_F t + phi_i) with phi_{i+1} = phi_i + 2pi/6, placed on six bonds of a modified stacked kagome lattice. The argument turns on the Floquet-Magnus commutator [H_k^(-1), H_k^(1)], which converts the time modulation into an effective time-reversal-breaking gauge field: it produces nearest-neighbor and next-nearest-neighbor complex hoppings with phase i|sin Delta_phi|, giving a flux of pi/2 per shaded plaquette with nonzero projections in all three spatial directions. This commutator is what turns periodic driving into a Chern vector, and the Chern numbers are then obtained by integrating the Berry curvature of the occupied bands over each two-dimensional surface Brillouin zone.","core_discovery":"The central discovery is that phase-delayed periodic driving alone can break time-reversal symmetry in three dimensions and produce a gapped Floquet band structure whose topological invariants form a Chern vector C=(1,1,-1). The driving assigns each of six modulated bonds a phase phi_i = phi_1 + (i-1)Delta_phi with Delta_phi = 2pi/6, and the first-order Floquet-Magnus commutator [H_k^(-1), H_k^(1)] generates complex nearest-neighbor and next-nearest-neighbor hoppings with phase pi/2, equivalent to a gauge flux of pi/2 per shaded plaquette with nonzero projections along all three primitive directions. The authors compute Chern numbers by integrating Berry curvature over two-dimensional slices of the Brillouin zone and find the same Chern vector for both the first-order truncated Floquet Hamiltonian and the zeroth-order effective Hamiltonian, indicating stability of the topological indices. Supercell and finite-lattice simulations show in-gap chiral surface states whose group velocities match the sign of each Chern component, and these states propagate unidirectionally past three structural defects without backscattering.","pith_inferences":["Going beyond the paper, the same phase-delay prescription could be applied to other three-dimensional lattices whose bond loops have nonzero projections in all directions, potentially generating Chern vectors other than (1,1,-1) such as (2,1,-1).","The paper's numerical regime has omega_F = 250 rad/s compared to lambda^(1) = 120 rad/s, a ratio of about two rather than deep high frequency; a direct test at much larger omega_F would show whether the first-order Floquet-Magnus term captures the full time-dependent topology or whether higher harmonics matter.","The surface Fermi loops form torus knots whose winding numbers equal the Chern vector components, which suggests an experimental signature: measuring constant-frequency surface Fermi loops in a time-modulated photonic, mechanical, or electrical metamaterial would directly reveal the Chern vector.","Because the simulations use the original time-dependent Hamiltonian H(t) rather than only the effective Hamiltonian, the predicted surface-state robustness could persist even where the high-frequency expansion is only marginally valid, provided the bulk gap remains open."],"forward_implications":["A purely time-modulated lattice, without static magnetic fields or mechanical rotation, can realize a three-dimensional Chern insulator.","Chiral surface transport exists on all faces of a finite sample, with propagation direction for each surface determined by the corresponding Chern vector component.","Surface states are expected to remain unidirectional when encountering structural defects, because the in-gap dispersion offers no opposite-velocity channel for backscattering.","The effective next-nearest-neighbor hopping generated by the Floquet mechanism is absent from the bare lattice, so time modulation expands the connectivity and topological possibilities of a given lattice geometry.","The matching Chern numbers from the first-order and zeroth-order truncated Floquet Hamiltonians indicate that the topological prediction is robust against moderate changes in the effective Hamiltonian."],"supporting_citations":[{"why":"Supplies the Haldane-model mechanism by which complex hopping phases in closed loops produce a time-reversal-breaking gauge field, the analog the present driving scheme aims to imitate.","marker":"[10]"},{"why":"Introduces the extension of the Chern number to three dimensions and the notion of a Chern vector, the central topological object of this work.","marker":"[26]"},{"why":"Reports the experimental observation of Chern vector photonic states in a three-dimensional metamaterial, providing the prior context and benchmark that motivates a time-modulation route.","marker":"[27]"},{"why":"Provides the stacked kagome lattice network connectivity on which the present tight-binding model is built.","marker":"[40]"},{"why":"Gives the Floquet-Magnus expansion method used to derive the effective time-reversal-breaking Hamiltonian in Eq. (13).","marker":"[41]"},{"why":"Supplies the Berry-phase and Chern-number formalism used to compute the Chern vector from the occupied-band Berry curvature.","marker":"[60]"}],"fun_headline_variants":["Periodic driving yields 3D Chern vector (1,1,-1)","Floquet phases create topological Chern insulator in 3D","Chiral surface states from phase-delayed periodic driving","Time-modulated bonds yield robust 3D Chern vector","Floquet driving creates topologically protected surface states"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument relies on the first-order Floquet-Magnus effective Hamiltonian correctly reproducing the band topology of the actual time-dependent system, which requires the driving frequency to be much larger than all energy scales; in the numerics omega_F = 250 rad/s is only about twice $\\lambda$^(1) = 120 rad/s, and the commutator in Eq. (13) omits diagonal on-site contributions, so if higher harmonics or the omitted terms close the band gap the Chern vector and surface-state predictions would not follow.","fun_headline_variants_meta":{"raw":{"variants":["Periodic driving yields 3D Chern vector (1,1,-1)","Floquet phases create topological Chern insulator in 3D","Chiral surface states from phase-delayed periodic driving","Time-modulated bonds yield robust 3D Chern vector","Floquet driving creates topologically protected surface states"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000719,"raw_usage":{"total_tokens":3205,"prompt_tokens":897,"completion_tokens":2308,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":513,"completion_tokens_details":{"reasoning_tokens":2224}},"tokens_in":513,"tokens_out":2308,"duration_ms":92889,"temperature":1.0,"reasoning_tokens":2224,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T05:11:10.199081+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the exact quasienergy spectrum of the full time-dependent Hamiltonian H(t) over one driving period in the same parameter regime and integrate the Berry curvature of the five bands below the target gap; if the resulting triplet is not (1,1,-1), or if the gap closes anywhere in the Brillouin zone, the effective-Hamiltonian prediction fails. A simpler check is to repeat the supercell surface-state calculation at omega_F = 2500, deep in the assumed high-frequency regime, and compare the Chern numbers and surface dispersions with those at omega_F = 250.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Reports the experimental observation of Chern vector photonic states in a three-dimensional metamaterial, providing the prior context and benchmark that motivates a time-modulation route."},{"cited_title":"Baardink, A","cited_arxiv_id":null,"evidence_quote":"Provides the stacked kagome lattice network connectivity on which the present tight-binding model is built."},{"cited_title":"Vanderbilt, Berry phases in electronic structure theory: electric polarization, orbital magnetization and topological insulators (Cambridge University Press, 2018)","cited_arxiv_id":null,"evidence_quote":"Supplies the Berry-phase and Chern-number formalism used to compute the Chern vector from the occupied-band Berry curvature."}],"review_version":1}