{"id":"409d5677-43db-41b7-b30f-a1d52c150154","arxiv_id":"2411.14722","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"CeSiI keeps its heavy fermion and antiferromagnetic behavior down to 2 layers and shows a slow, isotropic, thickness-dependent magnetic hysteresis interpreted as glassy relaxation.","lead":"This paper measures electrical transport in atomically thin flakes of the heavy fermion antiferromagnet CeSiI down to two van der Waals layers. It reports that the Kondo and magnetic ordering temperatures barely change with thickness, and that a slow, time-dependent magnetic hysteresis appears, which the authors attribute to glassy relaxation.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 'glassy' hysteresis is not yet separated from the ~200 G instrument lag the paper itself acknowledges; without a time-constant control or field-stop relaxation measurement, the central claim remains unproven.","rationale":"Reading in good faith: the paper is a careful transport study with a plausible and well-motivated central observation, and the hermetic-sealing fabrication advance is independently valuable. The thickness-insensitive T* and TN, the sharp metamagnetic transitions down to 2 L, and the sign reversal of RH near T* are credible evidence for intact Kondo-lattice and AFM physics in thin flakes. The genuinely novel claim, however, is the time-dependent glassy relaxation. That claim requires the broad sweep-rate-dependent hysteresis to be intrinsic material dynamics rather than instrumental lag. The paper explicitly concedes the instrument can contribute up to ~200 G, and a finite lock-in time constant or magnet flux lag produces exactly the observed rate dependence: smaller FW-BW difference at slower sweep, persistent nonzero difference where dMR/dH is finite, and temperature evolution tied to dMR/dH. No control experiment varying the time constant, performing DC field-stop settling, or subtracting a calibrated lag is reported. The power-law ARMS fits, the temperature peak near 4 K, and the apparent disappearance at 0.3 K are all consistent with an instrumental origin because they follow the temperature dependence of the MR slope. The secondary '3D' claim based on thickness and angle dependence would be moot if the component is instrumental, so the instrument-lag question is the single most load-bearing issue. This matches the reader's weakest assumption, so I agree with the reader's assessment. The appropriate disposition is to keep the CONDITIONAL verdict: the paper should be accepted only after the proposed settling/time-constant control isolates a residual intrinsic relaxation exceeding the instrument-lag estimate, or the claims should be softened to report the hysteresis phenomenology without asserting glassy dynamics.","tokens_in":14047,"tokens_out":4850,"duration_ms":51871,"concrete_test":"At T=4 K on the 4 L device, sweep at 100 G/s to 7 T, stop the field, and record MR(t) for up to 10^4 s; repeat with the lock-in time constant reduced by a factor of 10 at the same sweep rate. If the broad FW-BW hysteresis collapses to the predicted instrument-lag contribution or MR settles within the lock-in time constant, the glassy component is not established; if a slow material relaxation persists with a timescale comparable to tau, the interpretation survives.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that CeSiI exhibits an intrinsic slow-relaxing 'glassy' component rests entirely on the broad, sweep-rate-dependent part of the FW-BW hysteresis. The paper's own Fig. 2 caption states that trapped flux in the superconducting magnet and finite lock-in time constants can account for artificial hysteresis up to ~200 G, yet no subtraction, calibration, or instrumental control is shown. This is critical because a finite measurement time constant produces exactly the reported phenomenology: the FW-BW difference is proportional to (dH/dt)*(dR/dH), so it decreases at 5 G/s, remains nonzero above Hm2 where dMR/dH is nonzero, and can track the temperature dependence of dMR/dH (including the apparent disappearance at 0.3 K). The ARMS power-law fits in Fig. 3a therefore cannot by themselves distinguish material slow dynamics from an RC-type instrument lag. Until the residual broad hysteresis is shown to exceed the computed instrument-lag contribution at the same lock-in settings and sweep rates, the glassy interpretation is not established.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports magnetotransport measurements on atomically thin CeSiI flakes (2, 4, 8, 15 layers) and bulk samples, using air- and solvent-free fabrication and hermetic sealing. The authors find thickness-independent Kondo-lattice and antiferromagnetic transitions (T* ~ 40 K, TN = 8 K), thickness-independent metamagnetic fields Hm1 ~ 2.5 T and Hm2 ~ 4.5 T, and a sign reversal of the Hall coefficient near T*. Their central claim is that the forward-backward hysteresis in MR and Hall resistance below TN has two components: a time- and thickness-independent component tied to the 2D AFM metamagnetic transitions, and a broad, sweep-rate-dependent, isotropic component that increases with thickness and is interpreted as a hallmark of glassy relaxation dynamics of 3D nature, possibly a spin glass or multipolar order. The absence of exchange bias under field cooling is reported as a constraint on possible magnetic phases.","tokens_in":14202,"tokens_out":7301,"duration_ms":63836,"significance":"If the glassy interpretation is correct, the paper establishes coexisting 2D antiferromagnetic order and a separate slowly relaxing magnetic component in a heavy-fermion van der Waals metal, which is of considerable interest for quantum criticality and frustrated magnetism. The paper has notable strengths: the thickness-dependent R(T) and MR data are clean and support the 2D nature of the Kondo lattice and AFM order; the new hermetic sealing method is a technical advance; the field-angle dependence cleanly shows the metamagnetic features scale with the out-of-plane field component; and the field-cooling experiment is a genuine null result that constrains future models. However, the central claim rests on identifying the broad hysteresis as intrinsic, and the paper's own figure caption acknowledges an instrumental contribution that is not quantitatively excluded.","major_comments":[{"comment":"The paper's central claim that the broad hysteresis is a signature of intrinsic glassy relaxation is not yet separated from the instrumental hysteresis acknowledged in the Fig. 2 caption, which states that trapped flux in the superconducting magnet and finite lock-in time constants can account for artificial hysteresis of up to ~200 G. No subtraction, calibration, or control measurement is presented. This is load-bearing because a linear low-pass (RC-type) instrument response produces a forward-backward difference proportional to (dH/dt) * (dR/dH) at the same lock-in settings; such a contribution would (i) diminish at the slower sweep rate of 5 G/s, (ii) remain nonzero above H_m2 because dMR/dH is nonzero there, and (iii) track the temperature dependence of dMR/dH, including the apparent suppression of the time-dependent component at 0.3 K (Fig. S5). The power-law fits in Fig. 3(a) cannot distinguish this from material relaxation. To support the glassy interpretation, the authors should either subtract a quantitatively estimated instrument-lag contribution at each sweep rate and temperature, or show a control on a non-hysteretic sample (e.g., a normal-metal or non-magnetic material) measured with identical magnet and lock-in settings, or perform a field-stop relaxation measurement in which R is monitored at fixed field after the sweep is paused.","section":"Fig. 2 caption; Fig. 2(e,f); Figs. 3–4 and Fig. S5"},{"comment":"The claim that the time-dependent hysteresis has a 3D nature is based on comparing the MR hysteresis of one 2 L and one 4 L device (Fig. 4a,b). Two thicknesses are insufficient to establish a dimensional crossover, and the SI Methods explicitly states that ill-defined channel geometry prohibits quantitative comparison of transport magnitudes between devices. The normalization by zero-field resistance reduces, but does not eliminate, the influence of different contact geometries and flake shapes across the two devices. Data from 8 L and 15 L devices exist (e.g., an 8 L device is used in Fig. S5), so the thickness dependence of the broad hysteresis should be shown for all available thicknesses at the same temperature and sweep rate, and the conclusion should be re-evaluated if the comparison is not quantitatively reliable.","section":"Fig. 4(a,b) and SI Methods"}],"minor_comments":[{"comment":"The Supporting Information contains several typographical errors: 'Figurre S5' in the Fig. S5 heading, 'ploted' in the Fig. S6 caption, and 'suggeting' in the Fig. S7 caption should be corrected.","section":"SI captions (Figs. S5–S7)"},{"comment":"The exponent values quoted in the Fig. 3(a) caption (–2.6 and –1.3) do not match the values in Table S1 (–2.64 and –1.28); please make these consistent.","section":"Fig. 3(a) and Table S1"},{"comment":"The phrase 'we have not observe d any exchange bias effect' contains a typo, and earlier in the Introduction 'the focus our study presented in this paper' is missing the word 'of'.","section":"Main text, page 9"},{"comment":"The three-parameter power-law fit in Fig. 3(a) is applied to only a few points per device, and no fit uncertainties are reported; please provide the number of data points, error bars for the fit parameters, and a discussion of whether a power-law decay is statistically justified.","section":"Fig. 3(a) and main text"}],"recommendation":"major_revision","confidential_remarks":"For the editor: The manuscript is interesting and likely publishable if the authors can address the instrument-lag concern with a control experiment or a quantitative subtraction. The dimensionality claim also needs more thickness data. The central phenomenon is plausible, but the evidence as presented does not yet rule out a mundane instrumental origin for the broad hysteresis."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"First, the useful core: the paper cleanly shows T*, TN, and the metamagnetic transitions holding up from bulk down to 2 layers of CeSiI, with a better hermetic sealing method than the previous Nature paper. That is a real confirmation of the 2D nature of the heavy fermion and AFM orders, and the field-angle collapse onto H⊥ for the metamagnetic peaks is nice. The field-cooling exchange-bias check is also a good negative control.\n\nThe new-looking claim, sweep-rate-dependent isotropic hysteresis below TN, is where I stop following. The authors themselves note in the Fig. 2 caption that trapped flux and lock-in time constants can produce artificial hysteresis up to ~200 G, but they never show the subtraction or a control. A finite measurement time constant gives exactly the reported phenomenology: hysteresis proportional to (dH/dt)*(dR/dH), so it shrinks at 5 G/s, stays nonzero above Hm2 where dMR/dH is nonzero, peaks at 3–4 K where dMR/dH is largest, and vanishes at 0.3 K where the MR is flat. The ARMS power-law fits in Fig. 3a cannot distinguish that from material slow relaxation. The stress-test note is on target.\n\nThe \"3D glassy\" interpretation is also thin: two thicknesses (2 L and 4 L), two angles, no error bars. The paper's own hedge, \"may suggest,\" is the right level of commitment. And the SI Hall symmetrization formula as written adds Vxy(±H) instead of subtracting for the antisymmetric part; if that is not a typo, it is a sign error that would contaminate Rxy with the symmetric component.\n\nSo the central claim is not yet established, but the paper is worth engaging with seriously. The thickness-independent 2D heavy-fermion result is solid and citable, the fabrication is careful, and the authors are capable of running the needed control—a field-stop relaxation measurement or a computed instrument-lag subtraction at the same lock-in settings and sweep rates. This should go to peer review, not desk rejection, with referees instructed to focus on that control and on the thickness/angle statistics.","headline":"Solid 2D heavy-fermion confirmation, but the glassy hysteresis claim is not yet separated from instrument lag.","tokens_in":14825,"tokens_out":2269,"would_cite":true,"duration_ms":25117,"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":"CeSiI flakes down to two layers host both antiferromagnetic order and a slow glassy relaxation.","keywords":["CeSiI","heavy fermion","antiferromagnetism","glassy relaxation","hysteresis","magnetotransport","van der Waals 2D","Kondo lattice"],"falsifier":"Record the same field sweeps while measuring the field with a calibrated Hall sensor at the sample position and while explicitly varying the lock-in time constant: if the broad hysteresis persists when the true sample field is used and is independent of lock-in lag, the glassy interpretation survives; if it collapses to the known roughly 200 G instrument lag, the central claim fails. A complementary check is an AC susceptibility measurement on a sealed bulk crystal searching for a frequency-dependent peak near 4 K.","tokens_in":13832,"feed_emoji":"🧲","tokens_out":6827,"duration_ms":66549,"temperature":0.7,"pith_summary":"This paper argues that the van der Waals heavy fermion metal CeSiI, thinned to just two layers, retains both its Kondo-lattice heavy fermion behavior and its antiferromagnetic order, making both orders effectively two-dimensional. It further reports that the magnetotransport hysteresis below the Néel temperature splits into two components: a sharp, thickness-independent part tied to the antiferromagnetic metamagnetic transitions, and a broad, sweep-rate-dependent part that relaxes slowly and grows with flake thickness. The authors interpret the second component as glassy relaxation dynamics, possibly a spin-glass phase or multipolar order coexisting with antiferromagnetism. If correct, CeSiI becomes a platform for studying Kondo physics, magnetic order, and slow glassy degrees of freedom in the two-dimensional limit.","feed_headline":"Down to 2 layers, CeSiI shows antiferromagnetism plus glassy relaxation","feed_subtitle":"Slow, sweep-rate-dependent hysteresis points to spin glass or multipolar order in thin flakes.","key_machinery":"The central object is the field-sweep hysteresis $\\Delta\\mathrm{MR} = \\mathrm{MR}_{\\mathrm{FW}} - \\mathrm{MR}_{\\mathrm{BW}}$, condensed into the root-mean-square quantity $A_{\\mathrm{RMS}} = \\sqrt{(1/H_R)\\int_0^{H_R} (\\Delta\\mathrm{MR})^2\\,dH}$, plotted against the sweep time $\\tau = H_R/(dH/dt)$ with $H_R = 7$ T. A power-law decay of $A_{\\mathrm{RMS}}$ with $\\tau$ separates the slow, time-dependent relaxation from the time-independent antiferromagnetic contribution; thickness and field-angle dependence then separate the two-dimensional ordered component from the isotropic three-dimensional glassy one. The hermetic gallium-solder sealing of the devices is the enabling mechanism that lets these measurements reach two-layer flakes without environmental degradation.","core_discovery":"On its own terms, the paper establishes that atomically thin CeSiI, protected from air by a hermetic gallium-solder seal, shows the same Kondo-lattice onset near 40 K and the same Néel temperature near 8 K from bulk down to 2 layers, with metamagnetic transition fields varying by less than 20%. Below $T_N$ the forward-minus-backward field sweep hysteresis contains a time-independent part pinned to the two metamagnetic transitions and a broad, isotropic, time-dependent part whose root-mean-square magnitude decays as a power law in the sweep time. The time-dependent part peaks near 4 K, vanishes at 0.3 K, grows with flake thickness, and is insensitive to field angle, while field cooling produces no exchange bias. The paper reads these observations as evidence for a two-dimensional antiferromagnetic phase coexisting with a separate three-dimensional glassy relaxation component, and it identifies spin-glass or multipolar order as the plausible microscopic origin.","pith_inferences":["If the broad component is a true spin glass, a frequency-dependent AC susceptibility peak near 4 K should appear in bulk crystals; transport alone cannot distinguish spin freezing from multipolar order. This is an extension beyond the paper.","The vanishing of the time-dependent component at 0.3 K suggests the relaxation time exceeds the measurement window; a wait-and-measure protocol at fixed field could extract the relaxation time directly. This is an extension beyond the paper.","The thickness dependence implies interlayer coupling plays a role; a monolayer of CeSiI, if it can be isolated, would test whether the glassy component requires three-dimensionality. This is an extension beyond the paper.","The reported power-law exponents differ between the 2 L and 4 L devices; if reproducible, that difference may indicate a thickness-dependent distribution of relaxation times rather than a simple spin-glass transition. This is an extension beyond the paper."],"forward_implications":["Heavy-fermion Kondo lattice formation and antiferromagnetic ordering survive down to the bilayer, so the two-dimensional limit of this correlated metal becomes experimentally accessible for quantum-criticality studies.","The sweep-rate-dependent hysteresis is a separate magnetic degree of freedom whose relaxation time falls in the seconds-to-minutes window probed by 5 to 100 G/s sweeps, implying slow dynamics absent from the conventional antiferromagnetic response.","The glassy component's growth with thickness and its isotropy in field angle point to a three-dimensional, isotropic origin, such as a spin glass or hidden multipolar order, rather than surface or substrate effects.","The absence of exchange bias under field cooling constrains any magnetic glass phase to be decoupled or weakly coupled from the antiferromagnetic order, a constraint future models must respect."],"supporting_citations":[{"why":"Supplies the bulk CeSiI baseline: Kondo lattice onset, Néel temperature, two metamagnetic transitions, and the prior thinner-device degradation that this work improves on.","marker":"[11]"},{"why":"Theoretical prediction of a multipolar magnetic structure from strong spin-orbit coupling, cited as a candidate hidden-order state.","marker":"[19]"},{"why":"Proposal that magnetic order and heavy fermions can coexist via decoupled cerium sites, used to interpret Fermi-liquid behavior below $T_N$.","marker":"[23]"},{"why":"Establishes magnetic frustration in CeSiI and the cycloid or spin-density-wave candidates, motivating spin-glass or glassy behavior.","marker":"[24]"},{"why":"Provides the air- and solvent-free stencil-mask fabrication method that makes atomically thin CeSiI devices possible.","marker":"[25]"},{"why":"Theory of the Hall effect in heavy-fermion compounds used to interpret the Hall coefficient sign change and skew scattering.","marker":"[29]"},{"why":"Documents exchange bias from coexisting antiferromagnetic and spin-glass orders, the comparison that motivates and interprets the field-cooling test.","marker":"[35]"}],"fun_headline_variants":["CeSiI's 2D antiferromagnet reveals glassy spin dynamics","2D CeSiI: antiferromagnetic order meets spin-glass relaxation","Thin CeSiI layers show antiferromagnetism and glassy hysteresis","Glassy relaxation in 2D heavy fermion antiferromagnet CeSiI","CeSiI down to 2 layers: antiferromagnetic with glassy twist"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The broad, sweep-rate-dependent part of the hysteresis is assumed to be an intrinsic magnetic relaxation rather than a lag of the superconducting magnet or lock-in electronics, and the thickness trend rests on single 2 L and 4 L devices.","fun_headline_variants_meta":{"raw":{"variants":["CeSiI's 2D antiferromagnet reveals glassy spin dynamics","2D CeSiI: antiferromagnetic order meets spin-glass relaxation","Thin CeSiI layers show antiferromagnetism and glassy hysteresis","Glassy relaxation in 2D heavy fermion antiferromagnet CeSiI","CeSiI down to 2 layers: antiferromagnetic with glassy twist"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000182,"raw_usage":{"total_tokens":1297,"prompt_tokens":920,"completion_tokens":377,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":536,"completion_tokens_details":{"reasoning_tokens":273}},"tokens_in":536,"tokens_out":377,"duration_ms":3617,"temperature":1.0,"reasoning_tokens":273,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T14:58:39.609399+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Record the same field sweeps while measuring the field with a calibrated Hall sensor at the sample position and while explicitly varying the lock-in time constant: if the broad hysteresis persists when the true sample field is used and is independent of lock-in lag, the glassy interpretation survives; if it collapses to the known roughly 200 G instrument lag, the central claim fails. A complementary check is an AC susceptibility measurement on a sealed bulk crystal searching for a frequency-dependent peak near 4 K.","supporting_citations":[],"review_version":1}