{"id":"75ae66d2-d60e-4bb2-9d36-7cc7b40dcfd9","arxiv_id":"2504.13564","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"A Kagome metal exhibits h/e flux-periodic magnetoresistance oscillations at temperatures and length scales far beyond the single-particle mean free path, pointing to interaction-stabilized coherence.","lead":"Experiments on tiny pillars of the layered metal CsV3Sb5 show resistance oscillations whose period is set exactly by one quantum of magnetic flux threading between adjacent layers. The oscillations persist above 20 kelvin and across micrometer distances, far beyond the expected electron mean free path, suggesting a collective state may protect quantum coherence.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The coherence claim rests on mean-free-path estimates from different geometries (Supplement E/F) that do not exclude a long-lived minority Fermi-surface region; a semiclassical Bloch-Lorentz explanation therefore remains viable, so the central claim is not yet established.","rationale":"The empirical content of the paper is strong: the width-dependent h/e period, the parameter-free scaling Delta_B * w * c = Phi_0 with c' = 9.11 +/- 0.36 Angstrom, the strain suppression, and the sharp 45-degree switching are difficult to dismiss as artifacts. The reader's conditional verdict reflects this. My stress-test focuses on the single assumption that the mean free paths extracted from B-parallel-c SdH (Supplement F) and in-plane Drude analysis (Supplement E) transfer to the B-in-plane, current-along-c geometry. This assumption is load-bearing because the abstract's 'well exceeding the single-particle mean free path' is the entire basis for ruling out semiclassical Bloch-Lorentz oscillations and invoking a cooperative mechanism. The supplement itself flags the loophole (a quasi-ballistic subset) and only labels it 'highly unusual' without a quantitative multiband analysis. The existing semiclassical simulations are limited to model Fermi surfaces with uniform tau, so they cannot exclude a momentum-dependent lifetime on the CDW-reconstructed Fermi surface. A clean causal test—dose-controlled irradiation comparing the survival of SdH and h/e oscillations—would settle whether the relevant lifetime is the short, geometry-mismatched one or a decoupled long-lived channel. Pending that check, the claim of long-range coherence is plausible but conditional; the reader's verdict should stand unchanged.","tokens_in":19221,"tokens_out":17752,"duration_ms":189879,"concrete_test":"Perform a controlled electron-irradiation series on one c-axis pillar (monitoring T_CDW and T_c to ensure the correlated state is not destroyed) and compare the dose dependence of the SdH amplitude (Dingle lifetime) with that of the h/e oscillation amplitude at T = 2 K and 20 K. If the h/e oscillations are suppressed at the same rate as SdH, the oscillations require the same single-particle lifetime, so the estimated short mean free paths are relevant and the 'beyond mean free path' claim is falsified; if they survive doses that suppress SdH substantially, the oscillations are decoupled from the normal quasiparticle channel, directly supporting the interaction-stabilized coherence claim. This test separates the two interpretations without relying on cross-geometry mean-free-path estimates.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central inference—that the h/e oscillations occur 'well exceeding the single-particle mean free path' and thus demand interaction-stabilized long-range coherence—is supported only by two estimates that are not made in the oscillation geometry. The Dingle analysis (Supplement F) uses SdH oscillations measured with B parallel to c and assigns l_q < 200 nm above 1 K; the transport mean free path (Supplement E) is a Drude estimate from the in-plane resistivity rho_a with a single-band carrier density, giving l_t ~ 500 nm at T_c and ~ 150 nm at 20 K. The h/e oscillations, however, are measured with B in the plane and current along c; the relevant carriers are those that can traverse the in-plane width w coherently. Supplement E explicitly concedes that 'one may still argue for substantially suppressed scattering for some regions of one of the Fermi surfaces to reach a quasi-ballistic limit for these few select states only,' dismissing this as 'highly unusual.' That concession marks the exact loophole: a small, long-lived portion of a reconstructed multiband Fermi surface could carry the c-axis current and produce the same Bloch-Lorentz type B-periodic oscillations as in PdCoO2 without any cooperative coherence. The semiclassical simulations in Supplement H test only cylindrical and hexagonal model Fermi surfaces with uniform tau, so they do not rule out a momentum-dependent tau on the real CDW-reconstructed Fermi surface. Since no direct phase-coherence measurement is presented, the 'long-range coherence' conclusion is contingent on an assumption about unmeasured quasiparticle lifetimes.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports magnetic-field-periodic magnetoresistance oscillations in micron-sized pillars of CsV3Sb5 with current along the c-axis and magnetic field applied in the ab plane. The oscillation period follows ΔB = h/e/(w c) with the pillar width w and the interlayer spacing c, and a fit across eight devices yields c' = 9.11 ± 0.36 Å against the known c = 9.28 Å. The oscillations persist to temperatures above 20 K, are strongly suppressed when the field tilts out of plane by more than about 5 degrees, switch abruptly at 45-degree in-plane rotation, and are suppressed in a strained control device. The authors argue that, because the single-particle quantum and transport mean free paths are far below the device width, the oscillations cannot be explained by standard ballistic or semiclassical transport and instead imply interaction-stabilized long-range electron coherence linked to the correlated state below T'.","tokens_in":19601,"tokens_out":5955,"duration_ms":62993,"significance":"If the coherence interpretation is correct, this would be a striking result: h/e-periodic interference-like magnetotransport at high temperature and on micrometer length scales in a strongly correlated Kagome metal, with a period set only by geometry and the flux quantum rather than by Fermi-surface details. The paper has real strengths: the width-period scaling is convincingly documented across eight devices by a parameter-free relation, two independent background-subtraction methods agree on the oscillatory signal, the extracted interlayer spacing matches the lattice constant, and the temperature dependence tracks a broad set of independent experimental probes. The authors are also candid that the microscopic mechanism is currently unexplained. However, the central claim that the oscillations require long-range electron coherence is not yet established: the mean-free-path estimates are not made in the oscillation geometry, and the semiclassical simulations do not cover the actual reconstructed Fermi surface or momentum-dependent scattering.","major_comments":[{"comment":"The Dingle analysis is performed with B parallel to the c-axis and current along c, so the extracted quantum lifetimes characterize closed cyclotron orbits in the ab plane. The h/e oscillations, by contrast, are measured with B in the plane and current along c; the relevant carriers traverse the in-plane width, and the lifetime of the states responsible for these oscillations need not equal the lifetime of the γ and δ pockets observed in SdH. The manuscript's central assertion that the oscillations occur 'well exceeding the single-particle mean free path' therefore depends on transferring scattering rates between two different geometries and, potentially, between different Fermi-surface sheets. Supplement E explicitly concedes the key loophole: one could invoke 'substantially suppressed scattering for some regions of one of the Fermi surfaces to reach a quasi-ballistic limit for these few select states only.' That possibility is dismissed as 'highly unusual,' but no quantitative argument is given to exclude it, either from the CDW-reconstructed Fermi surface or from measurements of quantum oscillations with B in-plane. This is the load-bearing point for the coherence claim, and it needs to be addressed directly.","section":"Supplement F / Fig. 2a"},{"comment":"The semiclassical Bloch-Lorentz simulations test only cylindrical and hexagonal model Fermi surfaces with a uniform relaxation time and fully diffusive boundaries. The real system has a reconstructed, multiband CDW Fermi surface, and the simulations therefore do not rule out a semiclassical mechanism operating on a minority Fermi-surface sheet with weak scattering or on a restricted region of k-space. The Discussion's statement that it is 'highly implausible that a non-interacting band structure can account for the observed phenomenology' is accordingly stronger than the evidence presented. Closing this gap requires either a semiclassical calculation on the reconstructed Fermi surface with momentum-dependent τ or an experimental measurement that directly constrains the lifetime of the states involved in the in-plane-field geometry.","section":"Supplement H / Fig. S10"},{"comment":"The strained-device comparison (S9 versus S6) is used to argue for a 'fundamental link between long-range coherence and correlated electronic order.' However, the strained device differs from the strain-free device not only in the correlated state (the charge-order temperature is raised) but potentially also in defect density, inhomogeneous strain, and scattering rates introduced by the rigid mounting and glue. The argument that a lattice-constant change below 0.5% cannot by itself account for the suppression through inhomogeneous dephasing does not exclude strain-induced disorder scattering. As presented, this control experiment is suggestive but does not isolate the correlated-order mechanism.","section":"Supplement D / Fig. S4"}],"minor_comments":[{"comment":"The phrase 'periodicity is independent of materials parameters' is imprecise because the period explicitly depends on the lattice constant c; the intended meaning is presumably independence from electronic parameters such as Fermi-surface geometry or carrier density, and this should be stated more carefully.","section":"Abstract"},{"comment":"The word 'deconstructive interference' should be 'destructive interference.'","section":"Supplement D"},{"comment":"The claim that all plotted quantities fall on a 'universal temperature scaling' would be more convincing if the comparison were shown on normalized axes or with a quantitative collapse analysis; the current figure is qualitative.","section":"Fig. 4 / text"},{"comment":"The manuscript states that data and code 'will be deposited' with a link 'displayed here,' but the v1 manuscript does not yet provide the link; the revised version should include a working repository identifier.","section":"Data and code availability"},{"comment":"The color bar indicating Fermi-velocity anisotropy in the inset is difficult to read at the printed size; a larger inset or a separate panel would improve clarity.","section":"Fig. 2a"}],"recommendation":"major_revision","confidential_remarks":"The experimental dataset—particularly the width-period scaling across eight devices and the cross-probe temperature correlation—is solid and likely of broad interest. My main concern is that the headline interpretation ('interaction-stabilized long-range electron coherence') is not supported by the mean-free-path arguments as currently presented, since those arguments rely on scattering estimates from a different geometry and on model Fermi surfaces that exclude the suspected loophole. This is fixable in principle: a semiclassical simulation on the reconstructed Fermi surface with momentum-dependent scattering, or a direct measurement of in-plane-field quantum oscillations or phase coherence, would materially strengthen the claim. Without such an addition, the paper should present the oscillations as an empirical discovery with an open theoretical explanation rather than as a demonstration of interaction-stabilized coherence."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Read the Kagome pillar paper. Bottom line: the empirical h/e oscillations are on solid footing, but the paper's headline interpretation—that they demand interaction-stabilized long-range coherence—outruns the evidence. I'd send it to review, expecting the coherence claim to get pushed back hard.\n\nWhat's genuinely new: first observation of flux-periodic h/e oscillations in CsV3Sb5, in a regime far from the ultra-clean PdCoO2 case. The width-period scaling across eight FIB devices gives c' = 9.11 ± 0.36 Å against the known 9.28 Å interlayer spacing; the fit is parameter-free. The strain experiment (oscillations washed out on a rigid substrate) and the two background-subtraction methods agreeing are good control work. And the abrupt 45-degree in-plane period switch, consistent across different device aspect ratios, is a real puzzle that no existing semiclassical model predicts. That alone makes the paper worth refereeing.\n\nWhere it's soft: the leap from oscillations to long-range coherence rests on the claim that single-particle mean free paths are far shorter than the pillar width. The two estimates are made in different geometries—Dingle analysis with B parallel to c, Drude estimate from in-plane resistivity—while the oscillations are measured with B in-plane and current along c. The supplement explicitly concedes that one could argue for a small, long-lived portion of the Fermi surface reaching a quasi-ballistic limit, and calls that 'highly unusual.' Unusual isn't impossible. A minority pocket with a long lifetime on the CDW-reconstructed Fermi surface could produce the same Bloch-Lorentz type periodicity without any cooperative coherence. The semiclassical simulations in Supplement H only use cylindrical and hexagonal Fermi surfaces with uniform tau, so they do not rule that out. There is no direct phase-coherence measurement, and the data and code are promised but not yet posted.\n\nSo the central claim is not established. The empirical result is. If I were refereeing, I'd ask for either a direct coherence-sensitive probe, a realistic Fermi-surface calculation with momentum-dependent lifetimes, or a sharpened mechanism—and a serious engagement with the minority-pocket scenario.\n\nWho gets value: anyone working on CsV3Sb5, Kagome metals, or mesoscopic interference in moderately clean metals. It deserves a serious referee, but the referee should treat the long-range coherence conclusion as a hypothesis, not a demonstrated fact.","headline":"Robust h/e oscillations in CsV3Sb5, but the long-range coherence interpretation outruns the evidence—worth refereeing with the coherence claim treated as an open hypothesis.","tokens_in":20246,"tokens_out":1751,"would_cite":true,"duration_ms":16930,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["72.15.Gd","73.23.-b"],"model":"deepseek-v4-flash","headline":"Flux-quantum magnetoresistance oscillations in the kagome metal CsV3Sb5 persist on micrometer scales and above 20 K, where single-particle coherence should be dead.","keywords":["h/e oscillations","Aharonov-Bohm effect","Kagome metal","CsV3Sb5","long-range coherence","magnetoresistance","charge order","quantum interference"],"falsifier":"A decisive experiment would tune the correlated electronic state near 30 K (for example, by uniaxial strain or controlled disorder) while monitoring the h/e oscillation amplitude: the cooperative-mechanism claim predicts the oscillations vanish together with that state even if the mean free path barely changes, whereas a single-particle interference origin predicts they survive whenever the mean free path is long enough.","tokens_in":19030,"feed_emoji":"🧲","tokens_out":10558,"duration_ms":89890,"temperature":0.7,"pith_summary":"The paper attempts to establish that electrons in the layered kagome superconductor CsV3Sb5 can interfere coherently over distances and temperatures that are far beyond the limits set by single-particle scattering. The evidence is a set of magnetoresistance oscillations in micron-sized pillars whose period is set by one magnetic flux quantum $h/e$ threading between adjacent kagome layers, with no adjustable material parameter. If correct, this means a cooperative or collective electronic state, not ballistic quasiparticles, is carrying the quantum phase information. The result would tie the long-debated $T'\\sim 30$ K electronic crossover in CsV3Sb5 to a macroscopic quantum-interference phenomenon and make kagome metals a promising platform for interaction-stabilized electron coherence.","feed_headline":"Flux-quantum oscillations survive far beyond the mean free path","feed_subtitle":"In CsV3Sb5, the oscillation period is set only by h/e and device width, hinting at interaction-stabilized coherence.","key_machinery":"The central object is the flux-quantum box: a rectangular loop of width $w$ (the pillar width) and height $c$ (the spacing between adjacent kagome planes) threaded by the in-plane magnetic field. The identity $\\Delta B = \\Phi_0/(w c)$ with $\\Phi_0 = h/e$ turns every measured oscillation period into a parameter-free geometric prediction, and the systematic variation of $w$ across devices tests that identity directly. The enabling mechanism on the experimental side is the suspended microstructure: focused-ion-beam carved pillars mounted on soft silicon-nitride membranes, because strain from a rigid substrate suppresses the oscillations by more than 90 percent. The non-analytic angular response, including the 45-degree switching that is independent of the device aspect ratio, is the signature that identifies the field-selected interfering path.","core_discovery":"The central claim is that out-of-plane transport through micron-sized pillars of CsV3Sb5, with magnetic field applied in the kagome planes, shows oscillations whose period obeys $\\Delta B \\cdot w \\cdot c = h/e = \\Phi_0$, where $c$ is the interlayer spacing and $w$ is the pillar width chosen by fabrication. Because this relation contains no material-specific parameter, the authors interpret it as an atomic-scale Aharonov-Bohm effect: a flux quantum is enclosed between adjacent kagome layers. The oscillations survive above 20 K and in devices far wider than the measured quantum and transport mean free paths, collapse when the field tilts more than about 5 degrees out of plane, jump discontinuously at 45 degrees for in-plane rotation, and track the $T'\\sim 30$ K temperature scale seen in many other probes of CsV3Sb5. The authors conclude that single-particle interference and semiclassical Bloch-Lorentz orbits both fail to account for the data, and that a correlated many-body state establishes the long-range coherence.","pith_inferences":["If the cooperative picture is right, the h/e signal is carried by degrees of freedom distinct from the Landau quasiparticles that produce Shubnikov-de Haas oscillations; a testable consequence is that the h/e amplitude should not follow the Lifshitz-Kosevich thermal envelope governing those oscillations.","A natural extension is to search for the same parameter-free h/e periodicity in other layered kagome or frustrated metals that possess a $T'$-like correlated state, with the period predicted solely by device width and interlayer spacing.","Because the 45-degree switching is independent of the device aspect ratio, the interfering paths appear to be selected by the magnetic field's orientation relative to the crystal lattice rather than by the sample boundaries; a pillar with a non-rectangular cross-section would test this.","If the coherence is interaction-stabilized, small out-of-plane magnetic fields may tune the correlated state and should suppress the oscillation amplitude in the same manner as the field-switchable diode effect reported for this material."],"forward_implications":["The period obeys $\\Delta B \\cdot w \\cdot c = h/e$ with no fit parameters, so the same experiment in another layered metal should show the same period for the same width and spacing if the mechanism is generic.","The survival of oscillations above 20 K and on micrometer scales implies an effective coherence length exceeding the single-particle mean free path, so the coherence must be carried by a collective or interaction-stabilized degree of freedom.","The amplitude tracks the $T'\\sim 30$ K scale observed in many other probes of CsV3Sb5, linking the interference to the correlated electronic order rather than to the normal Fermi surface.","The sharp suppression beyond 5 degrees of out-of-plane tilt and the 45-degree in-plane switching rule out simple semiclassical Bloch-Lorentz orbits and indicate a non-analytic response of the correlated state.","Strain suppresses the oscillations, so the effect requires the unstrained, delicately suspended environment and is not a generic ballistic artifact of clean microstructures."],"supporting_citations":[{"why":"Reports the analogous h/e oscillations in (Pd,Pt)CoO2, the previous case this work extends to a kagome metal.","marker":"[4]"},{"why":"Provides the semiclassical Bloch-Lorentz model of B-periodic oscillations that the authors test and rule out as an explanation for their angular dependence.","marker":"[49]"},{"why":"Establishes the strain-sensitive magneto-chiral transport and the T' scale whose angular and temperature behavior the h/e oscillations mirror.","marker":"[20]"},{"why":"Supplies the muon-spin-rotation temperature dependence that falls on the same universal T' scaling as the oscillation amplitude.","marker":"[24]"},{"why":"Supplies the NMR temperature dependence used to anchor the T' crossover tied to the oscillation onset.","marker":"[25]"},{"why":"Provides the STM temperature dependence of the 2a0 charge order used in the comparison of the oscillation onset with T'.","marker":"[40]"}],"fun_headline_variants":["Kagome metal shows atomic-scale Aharonov-Bohm oscillations","Electron coherence survives far beyond mean free path in kagome metal","Flux quantum controls transport in CsV3Sb5 pillars","Interaction-stabilized interference in a kagome metal","Quantum oscillations defy single-particle limits in kagome metal"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument that the oscillations outlive single-particle coherence assumes that the mean free paths extracted from c-axis Shubnikov-de Haas oscillations and from in-plane Drude resistivity apply to the actual out-of-plane, in-plane-field transport in the pillars; if some carrier population is effectively ballistic in that geometry, the semiclassical explanation would not be excluded.","fun_headline_variants_meta":{"raw":{"variants":["Kagome metal shows atomic-scale Aharonov-Bohm oscillations","Electron coherence survives far beyond mean free path in kagome metal","Flux quantum controls transport in CsV3Sb5 pillars","Interaction-stabilized interference in a kagome metal","Quantum oscillations defy single-particle limits in kagome metal"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00045,"raw_usage":{"total_tokens":2308,"prompt_tokens":1024,"completion_tokens":1284,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":640,"completion_tokens_details":{"reasoning_tokens":1198}},"tokens_in":640,"tokens_out":1284,"duration_ms":8917,"temperature":1.0,"reasoning_tokens":1198,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T12:05:55.346677+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A decisive experiment would tune the correlated electronic state near 30 K (for example, by uniaxial strain or controlled disorder) while monitoring the h/e oscillation amplitude: the cooperative-mechanism claim predicts the oscillations vanish together with that state even if the mean free path barely changes, whereas a single-particle interference origin predicts they survive whenever the mean free path is long enough.","supporting_citations":[{"cited_title":"Putzke , author M","cited_arxiv_id":null,"evidence_quote":"Reports the analogous h/e oscillations in (Pd,Pt)CoO2, the previous case this work extends to a kagome metal."},{"cited_title":"Vilkelis , author L","cited_arxiv_id":null,"evidence_quote":"Provides the semiclassical Bloch-Lorentz model of B-periodic oscillations that the authors test and rule out as an explanation for their angular dependence."},{"cited_title":"Guo , author C","cited_arxiv_id":null,"evidence_quote":"Establishes the strain-sensitive magneto-chiral transport and the T' scale whose angular and temperature behavior the h/e oscillations mirror."},{"cited_title":"Nie , author K","cited_arxiv_id":null,"evidence_quote":"Supplies the NMR temperature dependence used to anchor the T' crossover tied to the oscillation onset."},{"cited_title":"Zhao , author H","cited_arxiv_id":null,"evidence_quote":"Provides the STM temperature dependence of the 2a0 charge order used in the comparison of the oscillation onset with T'."}],"review_version":1}