{"id":"8c509d13-5e6d-4a52-9b12-053651ef35de","arxiv_id":"2502.02987","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"An anisotropic thermoelastic damping model predicts a dissipation peak at the magnetic phase transition of FePS3, but quantitative agreement requires dividing the thermal conductivity by factors of 100 to 300.","lead":"This paper models how heat flow inside a suspended magnetic flake damps its vibrations, extending thermoelastic damping theory to materials where heat travels much faster in-plane than out-of-plane. The model predicts a peak in mechanical dissipation at the magnetic ordering temperature, but only matches experiments on FePS3 after shrinking the thermal conductivity by 100 to 300 times.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Attribution of the FePS3 damping peak to thermoelastic damping is not established: with measured thermal conductivity the model is ~10x too low, and matching data requires an unconstrained 100-300x conductivity reduction while magnetoelastic damping near TN remains unquantified.","rationale":"I read the paper in good faith. The anisotropic TED derivation is systematic, the thermodynamic model is internally coherent, and the qualitative prediction of a dissipation peak at the magnetic phase transition is plausible because the thermal expansion anomaly, and hence the relaxation strength, is expected to peak there. The specific proposal of two in-plane relaxation times is also falsifiable via higher-order modes. However, the central claim that the observed FePS3 peak is TED is only as strong as the assumption that other mechanisms are negligible. The paper itself provides two warning signs: the quantitative comparison with experiment fails by an order of magnitude when using measured thermal conductivity, and the only way to match the data is to scale thermal conductivity down by 100-300, which has no independent support. Separately, the frequency-mismatch argument against magnetoelastic damping is exactly the kind of argument that weakens near a critical point, where the relevant coupling is enhanced by critical fluctuations and slow magnetic modes; Ref. [19] explicitly flags this. None of this indicates fraud or a flawed derivation; it means the validation of the central claim is conditional. The proposed calculation of the magnetoelastic contribution would settle whether the neglect is safe. In the meantime, keeping the reader's CONDITIONAL verdict is the honest assessment.","tokens_in":18994,"tokens_out":13069,"duration_ms":123631,"concrete_test":"Compute the magnetoelastic contribution to Q-1 for the same 45 nm FePS3 drum using the spin-phonon Hamiltonian and coupling parameters from Ref. [19] (arXiv:2309.09672), evaluated at T = TN with the measured fundamental mode shape and frequency. If this contribution is within one order of magnitude of the measured dissipation, or of the residual after using unmodified thermal conductivity, then the Section II neglect is unjustified and the observed peak cannot be assigned to TED; if it is orders of magnitude smaller, the attribution survives this attack.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim requires that TED dominates the measured dissipation in FePS3 and that the model quantitatively captures it. The comparison in Fig. 6(a) shows that with published bulk thermal conductivity values all model variants underestimate the data by about an order of magnitude; the agreement in Fig. 6(b) is obtained only after dividing the calculated thermal conductivity by 100-300, a factor the authors acknowledge is not supported by measurements. This leaves the magnitude of the predicted peak unvalidated. The other pillar of the attribution is the neglect of magnetoelastic damping in Section II, justified only by the MHz versus GHz-THz frequency mismatch. That argument is weakest exactly at the phase transition: Ref. [19] states that nonlinear spin-mechanical coupling can become significant near TN, and Table IV lists magnetoelastic damping as 'Unknown' with no numerical estimate. Critical spin fluctuations can enhance the spin-phonon response even at MHz frequencies, so without an independent estimate the observed broad peak above and below TN could be dominated by a non-TED channel. Thus the paper's central quantitative claim rests on an untested dominance assumption.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper develops a theory of thermoelastic damping (TED) in suspended van der Waals membranes with anisotropic thermal conduction, and applies it to FePS3 drum resonators. Starting from the thin-plate equation with tension and the coupled thermoelastic heat equation, the authors derive an analytic expression for Q^{-1} that includes in-plane and out-of-plane heat flow, showing regimes governed by the thermal relaxation times in the two directions. They combine this with a model of the thermal expansion anomaly near the magnetic phase transition, in which magnetoelastic coupling enters through an effective Grüneisen parameter, and they predict a dissipation peak at the Néel temperature. The predictions are compared with existing FePS3 resonator data; with published bulk thermal conductivity the model underestimates the measured dissipation by about an order of magnitude, and agreement is obtained only after reducing the calculated conductivity by factors of 100–300.","tokens_in":19246,"tokens_out":2434,"duration_ms":32280,"significance":"If validated, the anisotropic TED model would be a useful and broadly applicable tool for vdW nanomechanics, since it goes beyond the standard Zener and Lifshitz–Roukes treatments in a direction that is physically important for layered materials. The derivation in Section III is systematic and analytic, and the model makes concrete, falsifiable predictions about the temperature and radius dependence of dissipation and about the role of the in-plane/out-of-plane conductivity ratio. The paper also clearly identifies the key limitation—the large conductivity reduction needed to match experiment—in Section VI. However, the central claim that the observed FePS3 dissipation peak is quantitatively captured by TED is not yet established, because the agreement hinges on an ad hoc rescaling of thermal conductivity and on an unquantified neglect of competing magnetic dissipation channels.","major_comments":[{"comment":"The validation of the central claim is not quantitative. In Fig. 6(a), all model variants using experimentally reported thermal conductivity values underestimate the measured dissipation by about a factor of 10, and the model curves are scaled by 10 only for visual comparison. The agreement in Fig. 6(b) is obtained after dividing the calculated thermal conductivity by 100–300, a factor that the authors themselves note is not supported by measurements. Since Q^{-1} in this regime depends on the thermal relaxation times and hence on conductivity, this rescaling is load-bearing; without a physical justification for such a strong reduction, the paper does not demonstrate that TED quantitatively explains the measured peak.","section":"Section V, Fig. 6"},{"comment":"The attribution of the measured dissipation to TED rests on the neglect of magnetoelastic damping, but that neglect is not quantitatively justified. The argument in Section II is only the frequency mismatch between MHz mechanics and GHz–THz spin waves, and the text itself notes that nonlinear spin–mechanical coupling can become important close to the phase transition (ref. [19]). Table IV lists magnetoelastic damping as “Unknown” with no estimate, and no calculation is provided for its magnitude near T_N. Since the experimental peak is broad and extends into the antiferromagnetic phase, the possibility that a non-TED channel contributes significantly is not excluded by the evidence presented.","section":"Section II and Table IV"},{"comment":"The predictive power of the phase-transition peak is weakened by fitted parameters that control its magnitude. The magnetic Grüneisen parameter γ_M is fitted to the thermal expansion data and enters α_T through Eq. (28), and since Q^{-1} depends on α_T^2, the height of the predicted peak is set by this fit. Similarly, the pretension N_0 in Appendix B is a fitting parameter that affects the resonance frequency and mode shape entering the dissipation. The peak position is robust because it follows from the c_V anomaly, but the quantitative agreement shown in Fig. 6(b) is not a parameter-free test of the theory.","section":"Sections IV and V, Eqs. (28)–(29), Appendix B"},{"comment":"The two in-plane relaxation peaks in Fig. 7 are explained by a hypothesized renormalized length scale λ ≈ 0.1, but this hypothesis is not derived from the model; the calculation in Section III contains only one in-plane thermal mode set indexed by the Bessel zeros. The statement that the strain generates heat in two localized regions that travel different distances is plausible but is not demonstrated from Eq. (10) or the modal sum. Since this feature is one of the paper’s claimed new qualitative results, it should either be derived from the modal structure or clearly presented as a speculation rather than a result.","section":"Section V, Fig. 7"}],"minor_comments":[{"comment":"Ref. [23] is cited as “Prabhakar and Vanglatore” but the correct name is Vengallatore; this appears also in the introduction text.","section":"Introduction and Acknowledgments"},{"comment":"Ref. [38] is written as “A. Halmund” but the thesis author is A. Haglund; please correct the citation.","section":"Section V and Fig. 6"},{"comment":"The symbol ΔE is used for both the energy loss per cycle in Eq. (15) and the dimensionless relaxation strength in Eq. (20). This reuse of notation makes the proportionality in Eq. (20) confusing; a separate symbol such as Δ_TED would improve readability.","section":"Section III, Eq. (20)"},{"comment":"The term “∇²” in the first line of Eq. (C1) appears to be a typographical leftover, since the full Laplacian is already written out in the following terms; please remove it for clarity.","section":"Appendix C, Eq. (C1)"}],"recommendation":"major_revision","confidential_remarks":"The paper is a solid theoretical contribution with a clear derivation, but its central quantitative claim is currently supported only by a large ad hoc rescaling of thermal conductivity. The authors should be encouraged to either identify a physical mechanism for the reduced conductivity in suspended flakes (e.g., finite-size or interface effects) or reframe the comparison as qualitative. I would also urge them to provide at least an order-of-magnitude estimate of magnetoelastic damping near T_N, since that channel is listed as “Unknown” and is the main competitor to their attribution. The paper fits the journal’s scope well and with these additions could become a strong contribution."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nThe one thing to know: this paper has a genuinely new and useful model—thermoelastic damping with anisotropic heat conduction for a circular plate—but its central quantitative claim about FePS3 isn't supported. The match with experiment only appears after dividing the thermal conductivity by 100–300, a factor the authors concede is unjustified.\n\nThe new piece is the solution of the 2D heat equation with separate κ∥ and κ⊥, including the full thermal-mode expansion. That's a real gap in the TED literature, and the derivation is careful. It reduces to the known isotropic results when κ∥=κ⊥, and the qualitative prediction—a dissipation peak at the magnetic phase transition via α_T²/c_v—is physically reasonable. The two in-plane relaxation times they identify are a concrete, testable consequence: higher mechanical modes should show extra peaks as a function of κ∥.\n\nThe soft spots are substantial. With measured bulk thermal conductivity, all model variants underpredict the experimental dissipation by roughly an order of magnitude. The \"agreement\" in Fig. 6b is obtained by reducing κ by factors of 100, 200, and 300—an unconstrained fit. On top of that, γ_M (the magnetic Grüneisen parameter) is fitted to the thermal expansion data, and N0 to the frequency data. So the quantitative comparison uses three fitted parameters to force a match. The attribution to TED is also insecure: magnetoelastic damping is dismissed because MHz mechanical frequencies are far below GHz–THz spin waves, but that argument is weakest at TN, where critical fluctuations can enhance spin-phonon coupling. The authors' own ref [19] says exactly this, and Table IV lists magnetoelastic damping as 'Unknown' with no estimate. Without an independent bound, a broad experimental peak above and below TN could be due to another channel.\n\nThat said, the model itself is worth publishing. The thermodynamic modeling of FePS3—specific heat without fit parameters, thermal expansion via a generalized Grüneisen relation—is thoughtful, and the discussion is transparent about the conductivity problem. A revision that reframes the experimental comparison as qualitative, adds sensitivity analysis, and provides at least a rough estimate of the magnetoelastic contribution would greatly strengthen the paper.\n\nWho should read it: people working on dissipation in vdW resonators or on TED in anisotropic materials. It deserves a serious referee, but with the expectation of major revision.","headline":"The anisotropic TED model is a genuine extension of Zener-Lifshitz-Roukes and worth having, but the experimental validation is not quantitative—it's propped up by a 100–300x conductivity rescaling and an unquantified magnetoelastic channel.","tokens_in":19831,"tokens_out":4667,"would_cite":true,"duration_ms":70017,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Thermoelastic damping in van der Waals magnets peaks at the magnetic phase transition, with dissipation set by the square of the thermal-expansion anomaly divided by specific heat, and by which anisotropic heat-flow path resonates.","keywords":["thermoelastic damping","van der Waals magnets","nanomechanical resonators","FePS3","thermal expansion anomaly","magnetoelastic coupling","anisotropic thermal conductivity","Néel transition"],"falsifier":"Measure $Q^{-1}(T)$ on a suspended FePS3 drum while independently determining the thermal expansion anomaly of the same flake: if the dissipation peak does not scale with $\\alpha_T^2/c_v$ across the Néel temperature, or if it remains when magnetic order is suppressed and the expansion anomaly disappears, the thermoelastic explanation is falsified.","tokens_in":18815,"feed_emoji":"🧲","tokens_out":10932,"duration_ms":203148,"temperature":0.7,"pith_summary":"Van der Waals magnets are promising nanomechanical resonators, yet the microscopic origin of their mechanical losses has remained unclear. This paper proposes that thermoelastic damping—irreversible heat flow generated when bending strains alternately compress and expand parts of a resonator—is the dominant loss channel and that it is sharply enhanced at the magnetic phase transition. The mechanism is magnetoelastic coupling: as magnetic order develops, the thermal expansion coefficient acquires an anomaly, and the thermoelastic relaxation strength scales as that coefficient squared divided by the specific heat. The paper extends standard thermoelastic-damping theory to anisotropic heat conduction, a defining property of van der Waals materials, and shows that the in-plane versus out-of-plane conductivity ratio selects which thermal relaxation path resonates with the mechanical vibration. Applied to FePS3 drums, the model predicts a dissipation peak at the Néel temperature on top of the usual Debye peak, in qualitative agreement with existing measurements.","feed_headline":"Thermoelastic damping spikes at FePS3's Néel temperature","feed_subtitle":"A magnetoelastic expansion anomaly sets the loss; heat-flow anisotropy decides which relaxation path dominates.","key_machinery":"The central object is an anisotropic thermoelastic-damping model for a clamped circular drum, which generalizes the standard through-thickness theory by keeping both out-of-plane ($\\kappa_\\perp$) and in-plane ($\\kappa_\\parallel$) heat conduction. The temperature field induced by the oscillating flexural strain is expanded in Bessel modes $J_0(j_n^0 r/a)$ along the radius with a $\\cosh$-profile across the thickness, and the dissipated energy is computed from the out-of-phase part of the resulting thermal strain. The magnetic side enters through an effective Grüneisen parameter that merges magnetoelastic coupling into the thermal expansion coefficient, so that the anomaly in $\\alpha_T$ at the phase transition becomes an anomaly in the relaxation strength. Two thermal relaxation times, $\\tau_z = h^2 \\rho c_V/(\\pi \\kappa_\\perp)$ and $\\tau_r = a^2 \\rho c_V/(\\mu^2 \\kappa_\\parallel)$, control whether the through-plane or radial heat flow resonates with the mechanical period; the model's $Q^{-1}$ is the sum over thermal modes of these resonant overlaps.","core_discovery":"The central claim is that the inverse quality factor of a suspended van der Waals magnet carries a direct imprint of the magnetic phase transition through the thermoelastic relaxation strength, which the paper derives as proportional to $\\alpha_T^2/c_v$ (Eq. 20). Magnetoelastic coupling makes the thermal expansion coefficient $\\alpha_T$ peak at the magnetic ordering temperature, so the dissipation $Q^{-1}$ should show a pronounced maximum there, superimposed on the conventional Debye peak governed by the thermal relaxation time. The paper further claims that standard through-thickness-only thermoelastic models miss essential physics in van der Waals materials: with anisotropic thermal conductivity, heat also relaxes radially, and $Q^{-1}$ acquires extra resonance conditions when the radial thermal time constant matches the mechanical period. For FePS3 drums the model predicts a temperature-dependent peak at the Néel temperature $T_N \\approx 118$ K and a radius-dependent Debye peak whose position shifts with temperature; comparison with published resonator data reproduces the shape of the dissipation, while matching its magnitude requires reducing the out-of-plane thermal conductivity by roughly two orders of magnitude relative to bulk values.","pith_inferences":["If the predicted peak is confirmed to scale with $\\alpha_T^2/c_v$, resonator damping measurements would become a local, quantitative probe of magnetoelastic coupling and magnetic order in two-dimensional magnets, complementing magnetometry and optical techniques.","The two-order-of-magnitude conductivity reduction needed to match experiment could indicate that boundary scattering or disorder in suspended flakes suppresses heat transport far below bulk values; direct thermal transport measurements on the same devices would separate that uncertainty from the thermoelastic mechanism itself.","The model points to a testable mode-number signature: higher-order mechanical modes should excite additional thermal modes and yield extra dissipation peaks in $Q^{-1}$ versus $\\kappa_\\parallel$, so a multimode drum experiment could confirm the anisotropic heat-flow picture.","Near the transition, nonlinear spin-mechanical coupling that the paper intentionally neglects may become significant; if so, the dissipation peak would acquire an amplitude dependence, a clean experimental way to distinguish thermoelastic from spin-mediated loss."],"forward_implications":["A suspended FePS3 drum should show a dissipation peak at the Néel temperature whose height tracks the square of the thermal expansion anomaly divided by the specific heat.","At fixed temperature, $Q^{-1}$ as a function of radius is a Debye peak, and its position shifts with temperature because the through-plane thermal relaxation time changes, so geometry selects the dominant loss regime.","In anisotropic van der Waals materials, radial heat conduction creates additional dissipation resonances when the in-plane thermal relaxation time matches the mechanical period, producing a second peak in $Q^{-1}$ as a function of $\\kappa_\\parallel$.","Matching the measured magnitude of dissipation in FePS3 requires an out-of-plane thermal conductivity about two orders of magnitude smaller than bulk values, implying that effective heat transport in suspended nanoscale flakes is strongly suppressed.","For any anisotropic two-dimensional material with known thermodynamic properties, the model supplies quantitative $Q^{-1}$ predictions without magnetic fitting parameters."],"supporting_citations":[{"why":"Supplies the experimental FePS3 resonator dissipation data and the thermal expansion anomaly that the model is compared against.","marker":"[16]"},{"why":"Introduces the thermoelastic damping mechanism and the resonance condition between mechanical and thermal time constants on which the model is built.","marker":"[21]"},{"why":"Provides the standard quality-factor formalism for thermoelastic damping in thin resonators that this work extends to anisotropic conduction.","marker":"[22]"},{"why":"Establishes the two-dimensional heat-conduction correction to the through-thickness-only approximation underlying the RZ model.","marker":"[23]"},{"why":"Supplies the circular-plate mode shape and frequency equations used to compute strain and stored energy.","marker":"[24]"},{"why":"Gives the measured in-plane and out-of-plane thermal conductivities of FePS3 used to set the anisotropy ratio in the model.","marker":"[28]"},{"why":"Supplies the thermal relaxation time constants $\\tau_z$ and $\\tau_r$ and the geometry factor used to explain the dissipation regimes.","marker":"[34]"},{"why":"Provides bulk specific heat measurements of FePS3 used to validate the computed phonon, magnon, and Ising contributions.","marker":"[37]"},{"why":"Provides temperature-dependent bulk thermal conductivity data used to extrapolate the anisotropic conductivities across temperature.","marker":"[38]"}],"fun_headline_variants":["Phase transition spikes thermoelastic damping in van der Waals magnets","Anisotropic heat conduction dictates damping regimes in 2D magnets","Neel point leaves fingerprint on mechanical dissipation in FePS3","Thermoelastic damping peak at magnetic transition in layered magnets","Magnetoelastic expansion anomaly sets resonator loss in vdW magnets"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The measured dissipation in the FePS3 resonators is dominated by thermoelastic damping, with magnetoelastic and other loss channels too small to matter near the phase transition.","fun_headline_variants_meta":{"raw":{"variants":["Phase transition spikes thermoelastic damping in van der Waals magnets","Anisotropic heat conduction dictates damping regimes in 2D magnets","Neel point leaves fingerprint on mechanical dissipation in FePS3","Thermoelastic damping peak at magnetic transition in layered magnets","Magnetoelastic expansion anomaly sets resonator loss in vdW magnets"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000897,"raw_usage":{"total_tokens":3876,"prompt_tokens":972,"completion_tokens":2904,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":588,"completion_tokens_details":{"reasoning_tokens":2817}},"tokens_in":588,"tokens_out":2904,"duration_ms":21075,"temperature":1.0,"reasoning_tokens":2817,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-09T10:18:09.326625+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure $Q^{-1}(T)$ on a suspended FePS3 drum while independently determining the thermal expansion anomaly of the same flake: if the dissipation peak does not scale with $\\alpha_T^2/c_v$ across the Néel temperature, or if it remains when magnetic order is suppressed and the expansion anomaly disappears, the thermoelastic explanation is falsified.","supporting_citations":[{"cited_title":"ˇSiˇ skins, M","cited_arxiv_id":null,"evidence_quote":"Supplies the experimental FePS3 resonator dissipation data and the thermal expansion anomaly that the model is compared against."},{"cited_title":"Zener, Phys","cited_arxiv_id":null,"evidence_quote":"Introduces the thermoelastic damping mechanism and the resonance condition between mechanical and thermal time constants on which the model is built."},{"cited_title":"Lifshitz and M","cited_arxiv_id":null,"evidence_quote":"Provides the standard quality-factor formalism for thermoelastic damping in thin resonators that this work extends to anisotropic conduction."},{"cited_title":"Prabhakar and S","cited_arxiv_id":null,"evidence_quote":"Establishes the two-dimensional heat-conduction correction to the through-thickness-only approximation underlying the RZ model."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the circular-plate mode shape and frequency equations used to compute strain and stored energy."},{"cited_title":"Kargar, E","cited_arxiv_id":null,"evidence_quote":"Gives the measured in-plane and out-of-plane thermal conductivities of FePS3 used to set the anisotropy ratio in the model."},{"cited_title":"Baglioni, M","cited_arxiv_id":null,"evidence_quote":"Supplies the thermal relaxation time constants $\\tau_z$ and $\\tau_r$ and the geometry factor used to explain the dissipation regimes."},{"cited_title":"Takano, N","cited_arxiv_id":null,"evidence_quote":"Provides bulk specific heat measurements of FePS3 used to validate the computed phonon, magnon, and Ising contributions."},{"cited_title":"Haglund, Thermal Conductivity of MXY3 Magnetic Layered Trichalcogenides, Ph.D","cited_arxiv_id":null,"evidence_quote":"Provides temperature-dependent bulk thermal conductivity data used to extrapolate the anisotropic conductivities across temperature."}],"review_version":1}