{"id":"9a86d915-12af-49d4-8e9d-79a57245cb9f","arxiv_id":"2505.05150","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Magnetotropic susceptibility measurements on CsV3Sb5 show a sharp transition at 30 K into a phase with an extremely small, c-axis orbital magnetic moment, slow dynamics, and a saturation field of about 0.2 T.","lead":"Tuning fork measurements of a kagome metal reveal a highly anisotropic, tiny magnetic response below 30 K, pointing to an exotic orbital (loop current) magnetic order. The result offers a thermodynamic signature for time-reversal symmetry breaking in CsV3Sb5, a long-debated phase.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The c-axis orbital-moment claim rests on an untested uniqueness assumption: an in-plane nematic or stripe order with domain reorientation could reproduce the sharp B∥ab dips in the ac-plane magnetotropic response.","rationale":"The reader's conditional verdict is appropriate, and our concern refines rather than overturns it. The paper provides a plausible thermodynamic signature of a small c-axis moment, supported by consistency with µSR, non-linear transport, and neutron upper limits, and it transparently acknowledges its inability to probe in-plane anisotropy. However, the load-bearing step is the uniqueness of the line-shape interpretation: the sharp B∥ab dips are taken as proof of a c-axis magnetic moment, but an in-plane ordered state with domain reorientation can generate the same qualitative angular features. The paper's own limitation statement ('we cannot discern electronic nematicity directly'; configurations not able to capture in-plane nematic phase) makes this the key unresolved alternative. A dedicated in-plane rotation experiment would settle whether the signal is truly uniaxial out-of-plane or contains an in-plane component. Since the central claim is not uniquely established but remains plausible, the verdict stays CONDITIONAL (UNCHANGED).","tokens_in":12165,"tokens_out":10137,"duration_ms":114816,"concrete_test":"Perform the same tuning-fork measurement with the field rotating in the ab-plane (θ=90°, varying azimuth φ) on the same crystals below 30 K and at 9 T, and also repeat the ac-plane rotation for two different azimuthal mountings of the sample. If the in-plane magnetotropic response shows a two-fold (cos 2φ) anisotropic signal or sharp features, an in-plane nematic/stripe order contributes to the observed ac-plane line shape and the c-axis moment assignment is ambiguous. If the in-plane response is isotropic and the ac-plane line shape is independent of azimuth, the c-axis moment interpretation is supported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central inference is that the sharp dips in Δf(θ) at θ=90°/270° (B∥ab) below T1=30 K are 'signatures for extremely anisotropic magnetic moment along the c-axis,' simulated by a 2D ferromagnetic-type model (Supplementary Eq. 6). For this inference to be valid, the line shape must be unique to a c-axis moment. That uniqueness is not established. An in-plane uniaxial order (nematic, stripe, or chiral) with domains can produce sharp features when the field is rotated in the ac-plane: at B∥ab the in-plane field component is maximal, driving domain reorientation, and the resulting free-energy curvature as a function of θ has cusps at exactly the in-plane orientations. The authors explicitly state they 'cannot discern electronic nematicity directly' and could not exclude in-plane TRS-breaking phases because their sample configurations cannot capture in-plane rotation. Since the model fit is the basis for assigning both the orientation (c-axis) and the order-parameter character (magnetic/TRS-breaking), the central claim is not yet uniquely supported by the data.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports tuning fork resonator measurements of the magnetotropic susceptibility of the kagome metal CsV3Sb5 as a function of temperature, magnetic field, and rotation angle over a broad parameter range. The authors identify a cascade of anomalies in the CDW phase, most notably a sharp feature at T1 ≈ 30 K, where the angular response deviates from the high-temperature cos(2θ) form and develops sharp dips when the magnetic field lies in the ab-plane. They interpret this as a thermodynamic phase transition that breaks time-reversal symmetry through a highly anisotropic magnetic moment along the c-axis, with an upper limit of 0.03 μB per vanadium ion, which they argue is too small for conventional spin order. They additionally report sweep-rate-dependent hysteresis below 30 K, taken as evidence of slow domain dynamics, and extract a saturation field of about 0.2 T that is compared with nonlinear conductivity data. The authors conclude that the results support an orbital (loop-current) origin of the low-temperature magnetic state.","tokens_in":12308,"tokens_out":6740,"duration_ms":69729,"significance":"If the interpretation holds, the paper would provide a thermodynamic signature of the proposed time-reversal-symmetry-breaking orbital magnetic state in CsV3Sb5, a long-sought quantity in a heavily debated system. The measurements are extensive, and the crystals are of high quality with residual resistance ratios of 300–500. The sharp 30 K onset and the sweep-rate-dependent hysteresis are valuable empirical constraints regardless of the model used. However, the significance is currently limited by the model dependence of the central inference: the assignment of the magnetic moment to the c-axis rests on a model presented only in the Supplementary Information and on a uniqueness assumption that the authors themselves do not fully test. The manuscript is honest about these limitations, but as written the central claim is not uniquely supported by the data.","major_comments":[{"comment":"The central claim that the sharp dips at B∥ab are 'signatures for extremely anisotropic magnetic moment along the c-axis' is not uniquely established. The line shape is simulated by a 2D ferromagnetic-type model (Supplementary Eq. 6), but no evidence is provided that an in-plane uniaxial order (nematic, stripe, or chiral) with domain reorientation could not produce the same sharp features at B∥ab in the ac-plane rotation. The authors explicitly state they cannot discern electronic nematicity directly and that their sample configurations cannot capture in-plane rotation, so they cannot exclude in-plane TRS-breaking phases. Because the c-axis assignment is the foundation for the orbital-moment interpretation and for the quantitative estimate of the moment, the manuscript should either provide a model comparison that rules out in-plane orderings or present a direct measurement (e.g., in-plane torque or magnetization) that distinguishes the two scenarios. Without this, the conclusion that the 30 K transition develops an orbital magnetic moment along the c-axis is not uniquely supported.","section":"Section II, paragraph 2 and last paragraph of Section II"},{"comment":"The extraction of the upper limit for the magnetic moment relies on a background subtraction and a model presented in Supplementary Note 2, neither of which is reproduced in the main text. Since the quantitative smallness of the moment is used to argue against a spin origin, the uncertainty in the absolute calibration must be quantified. If the model-dependent or background-related uncertainty permits a moment larger than a few hundredths of a Bohr magneton per vanadium, the argument that the moment is incompatible with spin order weakens. The authors should explicitly state the largest moment that is compatible with their data and derivation, and provide the full error budget in the main text or in a clearly referenced appendix.","section":"Section III, first paragraph (upper limit mc ≈ 0.03 μB/V)"},{"comment":"The sweep-rate-dependent hysteresis is interpreted as evidence of slow domain dynamics, but no control experiment or calibration is shown to exclude instrument-related lag (for example, PLL time constants or mechanical resonance effects). A control measurement on a nonmagnetic sample under identical conditions, or a field-sweep counterpart at fixed angle, would strengthen the conclusion that the hysteresis reflects an intrinsic slow dynamical process of the magnetic state below T1. If the hysteresis is instrumental in origin, the claim of 'extremely slow dynamics' would lose its evidential support for TRS breaking.","section":"Section II, paragraph 4 (Fig. 3)"}],"minor_comments":[{"comment":"The caption lists temperatures as '45K, 40K, 35K, ...' but the individual curves in the panel are not labeled with a legend, making it difficult for the reader to associate each curve with its temperature. A legend or explicit labels on the curves would improve clarity.","section":"Fig. 1(b) caption"},{"comment":"The phrase 'atonic ratio' should be corrected to 'atomic ratio'.","section":"Methods, sample preparation"},{"comment":"The word 'peizomagnetic' should be 'piezomagnetic'.","section":"Section III, last paragraph"},{"comment":"References [26] and [47] are identical (Denner, Thomale, and Neupert, Phys. Rev. Lett. 127, 217601 (2021)); one of them should be removed or renumbered.","section":"References"},{"comment":"The sentence 'the sample is either isotropic or possesses an exceedingly weak susceptibility' above TCDW uses 'or' where 'and' would be more precise, since a perfectly isotropic response is not expected for a crystal with this symmetry.","section":"Section II, last paragraph before Discussion"}],"recommendation":"major_revision","confidential_remarks":"The central concern is the uniqueness of the c-axis assignment; this is a correctness-risk issue that should be addressed with either a direct measurement or a more complete model comparison. The authors' own admission that in-plane orderings cannot be excluded is a strong indication that the current evidence is not conclusive. I would encourage the editor to request the Supplementary Information during review, since the model used for the central interpretation is not available in the main text. The paper is otherwise well written and the raw observations appear solid."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague—\n\nWhat you should know: this paper reports a tuning-fork magnetotropic susceptibility study of CsV3Sb5 that finds a clear thermodynamic anomaly at ~30 K, with a sharp onset, strongly anisotropic angular response (dips at B∥ab), hysteresis that depends on sweep rate, and a saturation field of ~0.2 T. If the anomaly is real—and it looks real—it is the first direct thermodynamic signature of the TRS-breaking phase that loop-current scenarios have been hunting for. The authors also see a cascade of transitions at 56 K and 70 K and reproduce the CDW and superconducting features.\n\nWhat is genuinely good: the crystals are high quality (RRR 300–500), the temperature/field/angle coverage is broad, and the raw data clearly show the 30 K feature. They are honest about limitations: they cannot directly see nematicity, cannot exclude in-plane TRS-breaking states, and they only provide an upper limit of ~0.03 μB/V for the moment. The comparison to prior muSR, non-linear resistivity, and Nernst data is sensible, and the loop-current interpretation is framed as a consistency check, not as an input to the measurement. No sign of circular reasoning.\n\nWhere the soft spots are: the central claim—that the sharp dips at B∥ab are 'signatures for extremely anisotropic magnetic moment along the c-axis'—rests on a model fit (Supplementary Eq. 6) whose uniqueness is not tested. As the stress-test note says, an in-plane nematic or stripe order with domain reorientation can produce cusps at exactly the in-plane orientations. The authors themselves say they cannot discern nematicity. So the orientation and order-parameter character are model-dependent. The moment magnitude extraction also depends on background subtraction and on setting Bc=0.2 T from the data rather than deriving it independently. These are not fatal: the qualitative evidence for a magnetic phase at 30 K is strong. But the specific c-axis orbital-moment conclusion is an interpretation, not a proof.\n\nWho this is for: anyone working on the AV3Sb5 family, loop-current order, or magnetotropic probes. The data deserve a serious referee. A good referee should ask for the supplementary fits and a clearer statement about the uniqueness of the line-shape inversion, and maybe a measurement that can distinguish in-plane vs c-axis order (e.g., different sample shape or a direct magnetization measurement). I would send it to review rather than desk-reject.","headline":"A careful magnetotropic study that likely finds a real 30 K thermodynamic anomaly in CsV3Sb5; the c-axis orbital-moment attribution is reasonable but not uniquely established, so the paper deserves review with a request for the supplementary model details.","tokens_in":12959,"tokens_out":3090,"would_cite":true,"duration_ms":31382,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Cooling CsV3Sb5 below roughly 30 K produces a tiny orbital magnetic moment along the c-axis, breaking time-reversal symmetry as expected for loop-current order.","keywords":["kagome metal","CsV3Sb5","orbital magnetism","time-reversal symmetry breaking","magnetotropic susceptibility","tuning fork resonator","charge density wave","loop current"],"falsifier":"Take the same crystal and remount it on the tuning fork with its c-axis rotated 90° about the rotation axis, then remeasure Δf(θ) at 10 K and 9 T: if the sharp dips still lie where the field is in the crystal's ab-plane, the c-axis orbital moment assignment is confirmed; if the dips vanish or shift to a different absolute angle, they are an artifact of the mounting geometry rather than a property of the sample. A companion check is to measure the c-axis magnetization on the same crystal in fields up to 0.5 T with a SQUID magnetometer: a resolved moment above 0.03 μB per vanadium, or a saturation field far from 0.2 T, would directly contradict the paper's estimate.","tokens_in":11900,"feed_emoji":"🧲","tokens_out":8784,"duration_ms":74860,"temperature":0.7,"pith_summary":"By measuring the magnetotropic susceptibility of CsV3Sb5 with a quartz tuning fork resonator, the paper establishes a thermodynamic phase transition near 30 K that develops an extremely anisotropic magnetic response. The angular dependence of the resonant frequency shift changes from a smooth cos(2θ) form to sharp dips when the magnetic field lies in the ab-plane, which can be reproduced by a simple model of a two-dimensional ferromagnetic-type structure with the moment along the c-axis. The inferred moment is smaller than 0.03 Bohr magneton per vanadium, saturates in a field of about 0.2 T, and shows sweep-rate-dependent hysteresis that indicates slow domain dynamics. The authors conclude that the transition breaks time-reversal symmetry through an orbital, loop-current-like magnetic order rather than a conventional spin moment, and they place this phase within a cascade of CDW-related transitions at 56 K, 70 K, and 94 K.","feed_headline":"Probe finds tiny orbital moment in kagome metal below 30 K","feed_subtitle":"Dips at B∥ab reveal a c-axis moment under 0.03 μB per vanadium, too small for spin order.","key_machinery":"The probe is a quartz tuning fork whose resonant frequency shift Δf(θ) is proportional to the magnetotropic susceptibility k = ∂²F/∂θ², the second derivative of the free energy with respect to the field angle θ. The load-bearing comparison is between the measured Δf(θ) and a simple model of a two-dimensional ferromagnetic-type magnetic structure with a c-axis moment, which reproduces the sharp dips observed at B∥ab. This model converts the depth and field dependence of the dips into a moment size (upper limit 0.03 μB/V) and a saturation field (~0.2 T). The resonator also serves as a low-frequency clock near 45 kHz: the appearance of hysteresis only above certain rotation speeds quantifies the slow domain dynamics of the ordered state.","core_discovery":"The central claim is that the phase transition at around 30 K in CsV3Sb5 breaks time-reversal symmetry by developing an orbital magnetic moment along the c-axis. This is inferred from tuning fork resonator measurements of the magnetotropic susceptibility, whose angular line shape develops sharp dips at B∥ab below T1, consistent with a model of a two-dimensional ferromagnetic-type magnetic structure. The moment is tiny, with an upper limit of about 0.03 μB per vanadium, too small to be of electron-spin origin, and it saturates in a magnetic field near 0.2 T. Additional evidence includes very slow relaxation dynamics, with hysteresis only appearing at rotation speeds above roughly 0.15°/sec, and the alignment of the 30 K boundary with anomalies seen by nonlinear conductivity, anomalous Nernst effect, muon spin resonance, and scanning tunneling spectroscopy. The paper interprets these observations as direct thermodynamic support for a loop-current (orbital) CDW phase, and reports a cascade of other transitions — at 56 K (first order), 70 K, and 94 K — that map onto previously observed CDW and symmetry-breaking events.","pith_inferences":["If the orbital moment is uniform along c, a magnetic field along c should induce a measurable anomalous Hall or Kerr response below 30 K with the same 0.2 T scale; the paper does not report such transport-optical cross-checks, and a future experiment could test this directly.","The angular line shape of Δf(θ) may be sensitive to the in-plane geometry of the ordered state: comparing the measured dips against simulations for staggered versus uniform loop-current patterns could discriminate between the candidate CDW patterns discussed in the literature.","The 70 K anomaly, if it is a TRS-breaking state with antiphase layers, is hidden from bulk magnetization but could be probed by second-harmonic generation or circular dichroism that couples to the layer-stacking chirality.","Because the tuning fork measures only the curvature of the free energy, a quantitative comparison between Δf(θ) and torque magnetometry on the same crystal (already partly reported in the supplementary) would provide an independent check that the dips are not a resonator artifact."],"forward_implications":["The 30 K transition becomes a thermodynamic reference point for the TRS-breaking phase in CsV3Sb5, reconciling muon spin resonance, nonlinear conductivity, and anomalous Nernst anomalies with a single free-energy signature.","Because the moment is too small for spin order and points along c, the low-temperature phase is a rare experimental realization of orbital magnetism in an itinerant kagome metal.","The small saturation field (~0.2 T) and slow domain dynamics imply the ordered state is highly susceptible to strain, field, and pressure, so experiments that detwin the sample should reveal larger effective moments.","The cascade of transitions at 94, 70, 56, and 30 K maps a hierarchy of CDW and symmetry-breaking orders, with the 70 K feature possibly corresponding to a staggered loop-current state with antiphase layers that cancels the net moment.","The tuning fork technique, applied here to a kagome metal, is a general tool for finding thermodynamic signatures of time-reversal symmetry breaking in materials with sub-0.1 μB moments."],"supporting_citations":[{"why":"Establishes the quartz tuning fork technique for detecting small anisotropy of susceptibility.","marker":"[40]"},{"why":"Defines magnetotropic susceptibility as the second angular derivative of the free energy and its proportionality to the tuning fork frequency shift.","marker":"[41]"},{"why":"Muon spin resonance evidence for time-reversal symmetry breaking below around 70 K and a large signal below T1, which the tuning fork data connect to.","marker":"[15]"},{"why":"Non-linear conductivity experiments that give a 30 K phase boundary and a 0.15 T saturation scale, used as a consistency check for the 0.2 T saturation field.","marker":"[17]"},{"why":"Anomalous Nernst effect measurements that show a magnetic anomaly consistent with the 30 K boundary.","marker":"[42]"},{"why":"X-ray diffraction showing the 2×2×4 to 2×2×2 CDW superlattice folding at T2, supporting the assignment of the 56 K first-order transition.","marker":"[45]"},{"why":"Neutron scattering upper limit of ~0.02±0.01 μB per vanadium triangle, consistent with the moment size inferred here.","marker":"[51]"},{"why":"Mean-field calculation of the LC2 state that produces an orbital moment ~0.01 μB per vanadium, matching the experimental estimate.","marker":"[32]"}],"fun_headline_variants":["Orbital moment flips time-reversal symmetry in kagome metal below 30 K","Kagome metal's loop current leaves a tiny magnetic trace at 30 K","30 K transition unveils orbital magnetism in kagome metal","Tuning fork detects time-reversal-breaking orbital moment below 30 K"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument rests on the assumption that the sharp dips in Δf(θ) at B∥ab are produced by a c-axis orbital moment in a two-dimensional ferromagnetic-type magnetic structure, rather than by an in-plane ordered state, sample shape effects, or a mechanical artifact of the tuning fork mount; the paper explicitly states it cannot detect electronic nematicity and cannot exclude in-plane time-reversal-symmetry breaking phases.","fun_headline_variants_meta":{"raw":{"variants":["Orbital moment flips time-reversal symmetry in kagome metal below 30 K","Kagome metal's loop current leaves a tiny magnetic trace at 30 K","30 K transition unveils orbital magnetism in kagome metal","Tuning fork detects time-reversal-breaking orbital moment below 30 K"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00044,"raw_usage":{"total_tokens":2239,"prompt_tokens":960,"completion_tokens":1279,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":576,"completion_tokens_details":{"reasoning_tokens":1196}},"tokens_in":576,"tokens_out":1279,"duration_ms":9474,"temperature":1.0,"reasoning_tokens":1196,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T23:10:44.834903+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take the same crystal and remount it on the tuning fork with its c-axis rotated 90° about the rotation axis, then remeasure Δf(θ) at 10 K and 9 T: if the sharp dips still lie where the field is in the crystal's ab-plane, the c-axis orbital moment assignment is confirmed; if the dips vanish or shift to a different absolute angle, they are an artifact of the mounting geometry rather than a property of the sample. A companion check is to measure the c-axis magnetization on the same crystal in fields up to 0.5 T with a SQUID magnetometer: a resolved moment above 0.03 μB per vanadium, or a saturation field far from 0.2 T, would directly contradict the paper's estimate.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes the quartz tuning fork technique for detecting small anisotropy of susceptibility."},{"cited_title":"Shekhter, R","cited_arxiv_id":null,"evidence_quote":"Defines magnetotropic susceptibility as the second angular derivative of the free energy and its proportionality to the tuning fork frequency shift."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Non-linear conductivity experiments that give a 30 K phase boundary and a 0.15 T saturation scale, used as a consistency check for the 0.2 T saturation field."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Anomalous Nernst effect measurements that show a magnetic anomaly consistent with the 30 K boundary."},{"cited_title":"Stahl, D","cited_arxiv_id":null,"evidence_quote":"X-ray diffraction showing the 2×2×4 to 2×2×2 CDW superlattice folding at T2, supporting the assignment of the 56 K first-order transition."},{"cited_title":"Li` ege, Y","cited_arxiv_id":null,"evidence_quote":"Neutron scattering upper limit of ~0.02±0.01 μB per vanadium triangle, consistent with the moment size inferred here."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Mean-field calculation of the LC2 state that produces an orbital moment ~0.01 μB per vanadium, matching the experimental estimate."}],"review_version":1}