{"id":"5371792f-35f2-4215-ae61-e63078393a29","arxiv_id":"2509.06832","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"Strong tilting of a time-modulated superlattice activates Landau-Zener tunneling that equally populates several bands, making the per-cycle drift equal to the average Chern number, yielding fractional drifts such as 1/3 and 1/2.","lead":"A particle in a specially shaped, tilted optical lattice can drift a fractional amount of a lattice period per driving cycle, even though it is alone and non-interacting. The fraction is the average of the Chern numbers of the energy bands it mixes together, and can be tuned to 1/3 or 1/2 by changing the lattice periods.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Eq. (B7) factorizes the time average of P_n(t)F_n(k0,t) into a product of averages; this is unjustified for Rabi-oscillating occupations and is the load-bearing step linking equal occupations to the fractional drift.","rationale":"The reader's weakest assumption correctly identifies Eq. (B7)'s factorization as the central load-bearing step. My reading agrees: the derivation of Eq. (7) requires replacing the average of P_n(t)F_n(k0,t) by the product of averages, which is not justified and is generally false for correlated oscillatory functions. The paper's independent numerical evidence for specific fractions (1/3 and 1/2) is real support for the phenomenon, but it does not validate the general formula. Because the reader's CONDITIONAL verdict already asks the authors to fix Eq. (B7), justify or drop the dispersion-term cancellation, and quantify the equal-population regime, my concern does not move the verdict. I would keep the verdict at CONDITIONAL: the paper is plausible and numerically supported, but the central analytic relation is not established as stated.","tokens_in":13283,"tokens_out":6586,"duration_ms":75415,"concrete_test":"Numerically evaluate both sides of Eq. (B7) for the paper's 1/3 example: τ1=τ2=25, d1=1/2, d2=2/3, ω=0.01, F=1.8, k0=0. Time-step the Schrödinger equation over 3T; at each t compute P_n(t) by projecting onto instantaneous eigenstates, and compute F_n(k0,t) from the static 2D Hamiltonian with Kx=k0−Ft, Ky=ωt. Form I_n=∫_0^{3T}P_nF_ndt and J_n=P̄_n(3T)∫_0^{3T}F_ndt for n=1,2,3,4. If ΣI_n≠ΣJ_n, Eq. (B7) is invalid; then check whether the full simulated Δx(3T) equals ΣI_n plus the neglected dispersion integral. This calculation settles whether Eq. (7) is a derived identity or an accidental numerical coincidence.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The decisive step is in Appendix B, Eq. (B7), mirrored by Eq. (6) of the main text. The drift after ξ periods is Δx(ξT) = Σ_n ∫_0^{ξT} P_n(t)F_n(k0,t)dt. Eq. (B7) then asserts this equals Σ_n P̄_n(ξT) ∫_0^{ξT} F_n(k0,t)dt, i.e. it replaces the time average of a product by the product of time averages. The stated justification—common period ξT of P_n and F_n—only makes each factor periodic; it does not make their correlation vanish. In fact P_n(t) oscillates strongly via Rabi/Landau-Zener transitions precisely where F_n(k0,t) is also nonuniform, so ∫(P_n−P̄_n)(F_n−F̄_n)dt is generically nonzero. Thus even with exactly equal average occupations P̄_n=1/ξ, the drift need not equal (1/ξ)ΣC_n. A second bundled assumption is the exact cancellation of Σ∫P_n ∂ε_n/∂k dt; the text asserts this from periodicity, but P_n(t) is not proven T-periodic after non-adiabatic transitions and this term is never computed. Consequently Eq. (7) is not derived from the stated semiclassical dynamics. The numerical examples are consistent with 1/3 and 1/2, but they do not by themselves establish the general factorization or the general formula.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies a single noninteracting particle in a one-dimensional time-modulated superlattice with a constant tilt, maps the driven dynamics to a static two-dimensional Hamiltonian, and computes Chern numbers. It claims that for weak tilt the one-cycle drift is an integer equal to the Chern number of the occupied band, while for strong tilt Landau-Zener tunneling and Rabi oscillations among the lowest ξ bands produce equal average occupations and a fractional drift Δx(ξT)/(qLξ) = (1/ξ)∑_{n=1}^{ξ} C_n. Numerical wave-packet simulations are presented for a 1/3 drift (Fig. 2) and a 1/2 drift (Fig. 4), and the occupation probabilities are shown in Figs. 3 and 5. The paper concludes that this is a new non-interacting realization of fractional quantized response controlled by tilt strength.","tokens_in":13687,"tokens_out":8642,"duration_ms":110226,"significance":"If the central formula Eq. (7) were rigorously established, the result would be significant: it would show that fractional quantized drift can arise in a single-particle, non-interacting system via Landau-Zener mediated multi-band occupation, with the fraction determined by average Chern numbers. The numerical simulations are concrete and the Chern-number computation from the static two-dimensional Hamiltonian is standard. The direction and plateaus in Figs. 2(f) and 4(c) are cleanly demonstrated. However, the general claim is currently overreaching: the derivation of Eq. (7) contains an unjustified factorization of a time integral, an unproven cancellation of the dispersion term, and an equal-occupation input extracted from the same dynamics that are meant to be explained. These are not presentation issues; they are load-bearing for the paper's central claim.","major_comments":[{"comment":"The step ∫_0^{ξT} P_n(t) F_n(k0,t) dt = P̄_n(ξT) ∫_0^{ξT} F_n(k0,t) dt is not justified. Common periodicity of P_n and F_n makes each factor periodic, but it does not make their product average equal to the product of averages. The cross term ∫ (P_n - P̄_n)(F_n - F̄_n) dt is generically nonzero, and near avoided crossings P_n changes rapidly precisely where F_n is also nonuniform. Without an estimate showing this correlation term vanishes in the stated limit, Eq. (7) does not follow from the semiclassical equations. The numerical examples may still be consistent with the formula, but they do not prove the general factorization.","section":"Appendix B, Eq. (B7); main-text Eq. (6)"},{"comment":"The paper asserts that the integral of the dispersion velocity term Σ_n P_n(t) ∂ε_n(k0,t)/∂k over the period T is exactly zero because the energy is periodic. This is only true if the occupation probabilities are constant (or have a special symmetry). For time-dependent P_n(t), the weighted integral is not zero in general; an integration by parts gives boundary terms plus ∫ Ṗ_n ε_n dt, and P_n is not proved to be T-periodic after non-adiabatic Landau-Zener transitions. This term is never computed in the manuscript, yet its exact cancellation is needed for Eq. (6).","section":"Main text, text before Eq. (6) and Eq. (5)"},{"comment":"The fractional value (1/ξ)Σ C_n relies on the input P̄_n(ξT)=1/ξ, which is read off from the numerical time evolution of the same wave packet whose drift is then predicted. The equal-occupation condition is not derived from the lattice parameters and tilt strength, nor is its basin of validity specified. Thus Eq. (7) is partly circular as a predictive formula: the fraction is an input obtained from the simulation, not an independent prediction. A complete argument should either derive P̄_n=1/ξ from the Landau-Zener dynamics or clearly state it as an additional empirical assumption, with the resulting statement limited to the parameter regime where the assumption holds.","section":"Eq. (7) and Figs. 3(b), 5(c), 5(d)"},{"comment":"The proof that the one-period integral of the Berry curvature is independent of the initial quasi-momentum k0 is not sound. In Eq. (C1), shifting the integration variable and the momentum argument amounts to an identity of the form I(k)=I(k+Δk) after relabeling; it does not establish the claimed relation unless one already assumes the integral is independent of k. The Hamiltonian has two arguments, Kx=k-Ft and Ky=ωt, so the invariance of ∫_0^T F_n(k,t)dt under k shifts requires a careful statement about simultaneous shifts in k and t and about the periodicity in Ky. The argument as written does not provide that. Since the k0-independence is used to replace the k-integral in the Chern number by the time integral at a single k0, this gap also affects Eq. (B4) and hence Eq. (B9).","section":"Appendix C, Eqs. (C1)-(C3)"}],"minor_comments":[{"comment":"The text says weak tilt is shown in Figs. 2(a) and (c), and large tilt in Figs. 2(b) and (d); the caption assigns (a)/(b) to weak tilt and (c)/(d) to large tilt. Please correct the cross-reference.","section":"Fig. 2 and caption"},{"comment":"The relation F_n(Kx,Ky)=(1/ω)F_n(k,t) deserves a few more steps and a sign check, since it is central to converting the static Chern number into the time-integral formula. The notation ∂t H in the Berry-curvature expression should also be defined explicitly.","section":"Appendix B, Eq. (B2)"},{"comment":"There are typographical errors and slightly nonstandard phrasing: 'enegy' in the abstract, 'weal' before 'tilt strength', 'consisting' for 'consistent', and 'tapped' in the introduction. The authors should also explain why 'one-cycle-averaged' is used for the drift over ξT, since ξ can be larger than 1; this is potentially confusing.","section":"Throughout"},{"comment":"The parameter q is introduced in Eq. (B3) and then set to q=1 in the main text. Since the section on Chern numbers uses both q and p, the precise integer relations (q=p, Tm/TF=p/q, etc.) should be stated more carefully to avoid ambiguity in the final formula.","section":"Eqs. (B3)-(B4) and main text near Eq. (7)"}],"recommendation":"major_revision","confidential_remarks":"The numerical evidence for the specific fractional drifts is credible and the paper addresses an interesting question. My main reservation is not the numerics but the paper's framing: Eq. (7) is presented as a derived formula, while the derivation contains an uncontrolled factorization (Eq. B7) and relies on an equal-occupation condition that is taken from the simulation. These are fixable in principle, but only if the authors either provide a rigorous justification (or a controlled limit in which the correlation terms vanish) or explicitly reposition the claim as a numerical observation with a clearly stated empirical condition. I would not reject outright, because the central phenomenon may well be correct, but the current manuscript overclaims its theoretical status."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe thing to know about this paper: it claims fractional quantized drift in a non-interacting tilted superlattice, via Landau-Zener tunneling that equally populates several bands whose Chern numbers average to a fraction. The numerics support the specific fractions (1/3 and 1/2), and the Chern numbers are computed independently from the static 2D Hamiltonian. That part is solid and is the paper's real asset.\n\nWhat's genuinely new is the mechanism rather than the framework. The machinery—semiclassical group velocity, Berry curvature, (k,t) to 2D mapping—is standard. But using tilt strength to balance band occupations and get fractional drift without interactions or nonlinearity is a nice twist and an experimentally plausible knob.\n\nThe soft spot is precisely where the reader put the finger: Eq. (B7) replaces the time average of P_n(t) F_n(k,t) with the product of averages, citing common period ξT. Periodicity doesn't kill the correlation; the Rabi oscillations are strongest where the Berry curvature is changing fastest, so the cross term is generically nonzero. The dispersion term's claimed exact cancellation over a period is also asserted, not proved, and it matters because P_n(t) is not T-periodic in the non-adiabatic regime. So Eq. (7) is not actually derived from the stated semiclassical dynamics. The numerics could still be right, and the mechanism may survive a proper treatment, but as written the central formula is a conjecture supported by two examples.\n\nAdditional concerns: the equal occupation P̄ = 1/ξ is extracted from the same numerics that produce the drift, so part of the claim is circular. There are no error bars or plateau-width robustness checks. The paper would be much stronger if the authors showed the factorization holds approximately, with a bound on the neglected correlation term, and if they demonstrated the plateau is reasonably flat against parameter variation.\n\nBottom line: this is a plausible and interesting mechanism, with genuine numerical evidence for the specific fractions, but the derivation needs repair. It deserves a serious referee, not a desk reject. I'd send it out, but with the expectation of major revision.","headline":"Plausible mechanism for non-interacting fractional quantized drift, but the central factorization in Eq. (B7) is unjustified and needs repair before this can be trusted.","tokens_in":14208,"tokens_out":2233,"would_cite":false,"duration_ms":23186,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["03.65.Vf","67.85.-d"],"model":"deepseek-v4-flash","headline":"A single non-interacting particle in a tilted modulated lattice can drift a fraction of a lattice constant per cycle, with the fraction equal to the average Chern number of the bands it tunnels through.","keywords":["fractional quantized drift","non-interacting topological systems","Landau-Zener tunneling","Rabi oscillations","Chern number","time-modulated superlattice","tilt-induced transport","ultracold atoms in optical lattices"],"falsifier":"Compute the exact one-cycle displacement for the parameters of Fig. 2 using the full time-dependent wave packet without the factorization of Eq. (B7), but with equal average occupations enforced artificially (e.g., by post-processing the trajectory); if Δx(3T)/(3L) deviates measurably from (C_1+C_2+C_3)/3 = 1/3, the factorization is invalid. Alternatively, choose an initial quasimomentum k0 where F_n(k0, t) varies strongly within one period and check whether the drift remains 1/3; if it does not, the claimed k0-independence fails.","tokens_in":13109,"feed_emoji":"⚛️","tokens_out":3816,"duration_ms":47677,"temperature":0.7,"pith_summary":"This paper claims that fractional quantized drift, typically associated with interacting topological systems, can occur in a completely non-interacting system: a quantum particle in a time-modulated optical superlattice subjected to a static tilt. When the tilt is weak, the particle remains in one band and its per-cycle drift is an integer Chern number; when the tilt is strong, Landau-Zener tunneling induces Rabi oscillations between several low-lying bands, and the one-cycle-averaged drift becomes a fraction equal to the arithmetic mean of those bands' Chern numbers. The authors demonstrate this with concrete examples producing fractional drifts of 1/3 and 1/2, controlled only by lattice periods and tilt strength. If correct, this would mean interactions are not required to produce fractional quantized responses, and the drift could serve as a direct, tunable topological marker.","feed_headline":"Tilt alone makes drift fractional in a non-interacting lattice","feed_subtitle":"Landau-Zener tunneling between bands makes a single particle drift by the average of multiple Chern numbers.","key_machinery":"The key object is the reduced Chern number, defined as the ratio of the one-cycle-averaged drift displacement to the lattice period. The underlying mechanism couples a time-modulated superlattice (with periods d1 and d2) to a static gradient potential (tilt F). The tilt and modulation make the two-dimensional Brillouin-zone parameters (K_x, K_y) become time-dependent, so the instantaneous band structure sweeps out a 2D parameter space. Strong tilt drives Landau-Zener tunneling between nearly degenerate low-lying bands, producing Rabi oscillations whose time-averaged occupations equalize to 1/ξ. The central identity equating the averaged drift to the average Chern number is derived by factori","core_discovery":"The central claim is captured by Eq. (7): the reduced Chern number C^red_sum = Δx(ξT)/(qLξ) equals (1/ξ) Σ_{n=1}^{ξ} C_n, where Δx(ξT) is the displacement of a wave packet after ξ periods of the time-modulated lattice, and C_n are the Chern numbers of the ξ bands involved in Landau-Zener tunneling. This holds when the average occupation probability of each participating band is equal, P̄_n = 1/ξ, a condition the authors verify numerically for strong tilt. For their representative parameters, weak tilt gives C^red_sum = -1 (integer drift), while strong tilt gives C^red_sum = (C_1 + C_2 + C_3)/3 = 1/3 (fractional drift). The paper further shows that by changing the lattice periods (e.g., d1=2/","pith_inferences":["One direct testable extension is to probe the crossover between the integer and fractional plateaus: the value of the drift in the transition region should reflect the incomplete Rabi oscillation averaging, which the paper does not characterize analytically.","The factorization assumption (Eq. B7) implies that if the occupation probabilities and Berry curvatures are engineered to be strongly correlated in time, the fractional quantization may break down even when the equal-occupation condition P̄_n = 1/ξ is satisfied; this is a regime not explored in the paper.","If this mechanism generalizes beyond the specific superlattice model, other driven non-interacting platforms (photonic waveguides, acoustic lattices) might display fractional quantized responses, potentially broadening the experimental landscape for topological metrology.","The equal-occupation condition is the essential ingredient; any perturbation that biases the Rabi oscillations (e.g., asymmetric coupling between bands) would shift the fractional value, making the plateau width in tilt strength a sensitive probe of the tunneling dynamics."],"forward_implications":["Fractional quantized drift can be realized in non-interacting ultracold atoms, removing the need for strong particle interactions or mean-field nonlinearity.","The tilt strength acts as a switch: below a threshold the drift is an integer Chern number, above it the drift becomes a fractional average, offering an experimentally controllable topological transport knob.","By adjusting the periods d1 and d2 of the superlattice, the denominator ξ of the fraction can be changed, allowing design of fractional drifts such as 1/3 and 1/2.","The measured drift provides a quantitative route to extract Chern numbers of multiple bands: the fractional value equals the average of the Chern numbers of all bands participating in Landau-Zener tunneling.","Because the effect persists for an arbitrary initial quasimomentum (under the stated adiabatic sampling condition), the drift is robust to band-structure details and initial-state preparation."],"supporting_citations":[{"why":"Provides the method of detecting fractional Chern insulator physics through center-of-mass drifts in weakly interacting gases.","marker":"[43]"},{"why":"Shows how quantized displacement can detect fractional Chern insulators in optical lattices, the baseline this work extends to non-interacting systems.","marker":"[44]"},{"why":"Reports fractional Thouless pumping of solitons via nonlinearity, the effect this paper shows can be produced without nonlinearity.","marker":"[45]"},{"why":"Establishes the experimental platform of a moving lattice superimposed on a static one, which the present Hamiltonian directly uses.","marker":"[46]"},{"why":"A theoretical treatment of fractional topological pumping, providing contrast to the non-interacting mechanism proposed here.","marker":"[50]"},{"why":"Shows preparation of a wave packet with a chosen quasimomentum in an optical lattice, supporting the initial-state preparation used in the numerical simulations.","marker":"[55]"},{"why":"Supplies the Landau-Zener-Stückelberg transition probability formula (T_lz) used to characterize when tunneling occurs between low-lying bands.","marker":"[56]"}],"fun_headline_variants":["Tilt alone makes drift fractional without interactions","Landau-Zener tunneling gives fractional drift in noninteracting lattice","Strong tilt induces fractional quantized drift in single-particle system","Fractional drift in a tilted lattice without particle interactions","Tilt-induced Landau-Zener tunneling leads to fractional drift"],"cache_read_input_tokens":2688,"weakest_assumption_plain":"The central result relies on the step where the time-varying occupation probability P_n(t) is pulled out of the integral with the Berry curvature F_n(t), treating them as effectively constant or uncorrelated; this approximation is least trustworthy near avoided crossings, where Landau-Zener tunneling actually happens, and the accompanying claim that the dispersion contribution integrates to exactly zero also rests on a periodic equilibrium assumption.","fun_headline_variants_meta":{"raw":{"variants":["Tilt alone makes drift fractional without interactions","Landau-Zener tunneling gives fractional drift in noninteracting lattice","Strong tilt induces fractional quantized drift in single-particle system","Fractional drift in a tilted lattice without particle interactions","Tilt-induced Landau-Zener tunneling leads to fractional drift"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000247,"raw_usage":{"total_tokens":1374,"prompt_tokens":730,"completion_tokens":644,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":474,"completion_tokens_details":{"reasoning_tokens":564}},"tokens_in":474,"tokens_out":644,"duration_ms":6969,"temperature":1.0,"reasoning_tokens":564,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-04T23:02:13.466999+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the exact one-cycle displacement for the parameters of Fig. 2 using the full time-dependent wave packet without the factorization of Eq. (B7), but with equal average occupations enforced artificially (e.g., by post-processing the trajectory); if Δx(3T)/(3L) deviates measurably from (C_1+C_2+C_3)/3 = 1/3, the factorization is invalid. Alternatively, choose an initial quasimomentum k0 where F_n(k0, t) varies strongly within one period and check whether the drift remains 1/3; if it does not, the claimed k0-independence fails.","supporting_citations":[{"cited_title":"Fractional chern insulators of few bosons in a box: Hall plateaus from center-of-mass drifts and density profiles,","cited_arxiv_id":null,"evidence_quote":"Provides the method of detecting fractional Chern insulator physics through center-of-mass drifts in weakly interacting gases."},{"cited_title":"Detecting fractional Chern insula- tors in optical lattices through quantized displacement,","cited_arxiv_id":null,"evidence_quote":"Shows how quantized displacement can detect fractional Chern insulators in optical lattices, the baseline this work extends to non-interacting systems."},{"cited_title":"Quantized fractional thouless pumping of soli- tons,","cited_arxiv_id":null,"evidence_quote":"Reports fractional Thouless pumping of solitons via nonlinearity, the effect this paper shows can be produced without nonlinearity."},{"cited_title":"Nonlinear thouless pumping: solitons and transport breakdown,","cited_arxiv_id":null,"evidence_quote":"Establishes the experimental platform of a moving lattice superimposed on a static one, which the present Hamiltonian directly uses."},{"cited_title":"Quantum theory of fractional topological pumping of lattice solitons,","cited_arxiv_id":null,"evidence_quote":"A theoretical treatment of fractional topological pumping, providing contrast to the non-interacting mechanism proposed here."},{"cited_title":"Geometrical pumping with a Bose-Einstein condensate,","cited_arxiv_id":null,"evidence_quote":"Shows preparation of a wave packet with a chosen quasimomentum in an optical lattice, supporting the initial-state preparation used in the numerical simulations."},{"cited_title":"Landau- Zener-St¨ uckelberg interferometry,","cited_arxiv_id":null,"evidence_quote":"Supplies the Landau-Zener-Stückelberg transition probability formula (T_lz) used to characterize when tunneling occurs between low-lying bands."}],"review_version":1}