{"id":"a31e07fe-6fd7-432d-b025-1342b3613c34","arxiv_id":"2607.06703","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Magnetic fluctuations above the Curie temperature of ultrathin CoFeB films enhance damping-like spin-orbit torque and suppress field-like torque via mixing of longitudinal and transverse interfacial spin conductances.","lead":"Near the Curie point of ultrathin ferromagnets, magnetic fluctuations strongly enhance damping-like spin-orbit torque while suppressing the field-like component. The effect offers a route to more efficient spintronic switching by deliberately using heat or short-wavelength magnons, analogous to heat-assisted magnetic recording.","discovery_kind":"new_method","skeptic_critique":{"model":"grok-4.5","headline":"The short-length-scale averaging assumption (Eq. 5) is load-bearing for the microscopic claim, and the paper's own data leave it under-constrained.","rationale":"The Reader correctly isolates the short-length-scale assumption of Eq. 5 as the weakest link and correctly judges the experimental observation itself to be solid (two techniques, two material systems, GMS method). My stress-test confirms that this is the single most load-bearing concern for the microscopic claim: without it the predicted signs of the DL and FL changes reverse. The paper supplies only indirect support (Curie–Weiss ΔH and decreasing effective moment), and its own model (Fig. 6) does not reproduce the TC2-linked FL drop. Because the experimental claim stands independently of the length-scale interpretation, the Reader’s CONDITIONAL verdict is appropriate and needs no adjustment. The concrete scattering test would settle the issue cleanly.","tokens_in":18521,"tokens_out":648,"duration_ms":6498,"concrete_test":"Measure the magnetic correlation length ξ(T,H) above TC1 by small-angle neutron scattering or resonant soft-x-ray scattering on identically prepared CoFeB(0.65)/Pt and CoFeB(1.1)/Ta films, and compare ξ to the known spin-diffusion lengths of Pt (~1–2 nm) and Ta (~1 nm). If ξ ≳ λs in the regime where ξ DL rises and ξ FL falls, the short-length-scale premise of Eq. 5 is falsified and the mixing interpretation must be revised.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central interpretive claim is that the observed DL enhancement / FL suppression above TC arise from angular averaging of the interfacial spin-conductance tensor (Sec. II, Eq. 5). That derivation requires the characteristic fluctuation length to be shorter than the spin-diffusion length so that the interface is described by the configuration average of G0 rather than by a macroscopic average of local torques. The paper itself notes (Sec. VI) that if fluctuations are longer-ranged, the same averaging would suppress rather than enhance the measured DL torque. The only evidence offered for the short-length-scale regime is the Curie–Weiss form of ΔH1(T) above TC1 and the inference that the effective moment μ ~ kBT/ΔH decreases with T (Sec. IV). That is necessary but not sufficient: a decreasing moment does not by itself fix the spatial correlation length relative to λs of Pt or Ta. The semi-phenomenological Langevin calculation (Fig. 6) further fails to place the FL drop at TC2 rather than TC1, indicating that the length-scale assumption alone does not fully account for the data. Thus the microscopic mechanism remains the least secure link between the solid experimental observation and the claimed fluctuation-driven mixing.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.5","summary":"The manuscript reports that in ultrathin CoFeB/Pt and CoFeB/Ta bilayers with confinement-suppressed Curie temperatures, the damping-like spin-orbit field is strongly enhanced above TC while the field-like field is suppressed, with opposite in-plane field dependences. The authors introduce a generalized magnetoelectronic susceptibility (GMS) reformulation of second-harmonic Hall analysis that does not assume magnetic saturation or a unique anisotropy field, validate it against standard HHV on a thicker control film, and corroborate the temperature trends with ST-FMR. They interpret the divergent DL/FL behaviors as fluctuation-driven mixing of longitudinal and transverse interfacial spin conductances (Sec. II, Eq. 5), and discuss a possible spintronic analog of heat-assisted recording via engineered fluctuations or short-wavelength magnons.","tokens_in":18830,"tokens_out":1344,"duration_ms":24080,"significance":"If the experimental trends hold, the work is a useful contribution on two levels. Methodologically, the GMS approach is a practical tool for quantifying effective SOFs in unsaturated, fluctuating ultrathin magnets where standard HHV assumptions fail; the control-sample cross-check (Fig. 1) and dual-technique consistency (HHV and ST-FMR) strengthen that claim. Physically, opposite DL/FL temperature and field trends near TC in two HM systems with opposite spin-Hall signs are nontrivial and application-relevant, since high-current SOT devices inevitably heat the free layer. The paper is appropriately cautious in framing the microscopic picture as consistency rather than a unique proof, and it openly discusses REE and magnetization-scaling alternatives. The main value is therefore the robust phenomenology plus a clear, testable geometric mechanism, even if the length-scale assumption remains under-constrained.","major_comments":[{"comment":"Sec. II, Eq. (5): The central microscopic claim (enhancement of real Gmix and suppression of Im Gmix) requires that the characteristic fluctuation length be shorter than the spin-diffusion length so that the interface is described by the configuration average of G0. The paper itself notes (Sec. VI) that longer-ranged fluctuations would instead suppress the measured DL torque. The only support offered is the Curie–Weiss form of ΔH1(T) and the inference that μ ~ kBT/ΔH decreases above TC1 (Sec. IV). That constrains the effective moment size but does not fix the spatial correlation length relative to λs of Pt or Ta. This assumption is load-bearing for the interpretive claim and should be either better constrained (e.g., by correlation-length estimates, thickness/λs trends, or a clear falsification test) or the abstract/conclusions should more sharply separate the robust experimental observa","section":null},{"comment":"Sec. IV–V and Eq. (12): SOT efficiencies ξDL,FL depend on a semi-quantitative M(H,T) reconstructed from AHE on the Hall bars plus SQUID on a different stack (different buffer, CoFeB thickness, and TC1). The paper acknowledges the uncertainty in ΔHin and that the estimate is only semi-quantitative below TC. Because the claimed non-trivial DL enhancement is largely carried by ξDL (Fig. 4c), the manuscript should quantify how plausible variations in ΔHin and stack mismatch propagate into ξDL,FL, and show that the rise of ξDL above TC1 survives those bounds rather than being an artifact of the M proxy.","section":null},{"comment":"Sec. VI and Fig. 6: The Langevin–Brillouin calculation based on Eq. (5) places both the DL upturn and the FL drop at TC1, whereas the data show the FL collapse only above TC2 (and more abruptly). The paper notes this mismatch but still presents the model as reproducing the “overall anomalous behaviors.” Either the model should be extended to incorporate the TC2/Rashba-related feature that the authors themselves associate with the FL drop, or the claim that Eq. (5) accounts for the FL suppression should be narrowed, with TC2-related physics treated as a distinct, possibly coexisting channel rather than absorbed into the same free-parameter fit.","section":null}],"minor_comments":[{"comment":"Fig. 1(d) and related text: The control comparison is valuable; please state explicitly the field range and percentage agreement used to claim “within 10%,” and note whether Oersted subtraction was applied consistently in both methods.","section":null},{"comment":"Eq. (7) and the definition of dRH/dHIP, dRH/dHOP: Clarify whether these susceptibilities are evaluated at the same H used for each V2ω(ϕ) point, and how nonlinearity of RH(H) near TC is handled when the small-tilt linearization (Eq. 8) becomes marginal.","section":null},{"comment":"ST-FMR section (Sec. V.B): The limitations of applying Eqs. (14)–(15) in the fluctuating regime are stated; a brief quantitative estimate of systematic error from field-dependent α and Ha near TC would help the reader weight Fig. 5(d) relative to the HHV results.","section":null},{"comment":"Notation: TC, TC1, and TC2 are used somewhat interchangeably early on; a single consistent convention after their introduction in Sec. IV would reduce ambiguity in Figs. 4–5.","section":null},{"comment":"References and context: A short comparison to prior reports of SOT or SHE enhancement near magnetic critical points (beyond the cited FexPt1−x and NiO-spacer works) would help place the magnitude of the observed ξDL change.","section":null}],"recommendation":"major_revision","confidential_remarks":"The experimental phenomenology looks publishable and is the main strength; the mechanism section is the soft point. I would not reject on the length-scale issue alone if the authors reframe and bound it, but I would not accept the present abstract-level claim that the data “show” the mixing mechanism without that work. Scope fits a solid condensed-matter / spintronics journal; novelty is incremental but real (GMS + dual-system near-TC phenomenology)."},"author_rebuttal":null,"desk_editor":{"model":"grok-4.5","letter":"The real news here is experimental: above the suppressed Curie point of ultrathin CoFeB, the damping-like spin-orbit field rises while the field-like one drops, and the two have opposite field dependences. They see this in both Pt and Ta stacks, with harmonic Hall (via a new GMS reformulation) and ST-FMR, plus a thicker control film that behaves normally. That pattern is new relative to the usual SOT literature and is not an artifact of one material or one technique.\n\nWhat they do well is the metrology. Standard second-harmonic analysis assumes saturation and a single Ha; near TC that fails. GMS replaces those prefactors with measured Hall susceptibilities, so you can extract effective fields without assuming a uniform state. The control comparison at room temperature shows the two methods agree once the film is reasonably ordered. Magnetization is handled carefully enough (AHE as proxy, SQUID on a related stack, sigmoid widths) that the efficiency trends survive the 1/M scaling. Data and Zenodo deposit are there.\n\nThe soft spot is the microscopic claim, not the observation. Equation 5 (angular average of the spin-conductance tensor) requires fluctuation lengths shorter than the spin-diffusion length; if they are longer, the same averaging would suppress rather than enhance DL torque. The only support is the Curie–Weiss form of ΔH and a decreasing effective moment. That is necessary but does not pin the correlation length relative to λs of Pt or Ta. Their Langevin sketch (Fig. 6) also puts the FL drop at TC1 instead of the observed TC2, so the model is semi-phenomenological and incomplete. Rashba or orbital contributions are discussed honestly but not ruled out. The heat-assisted-recording analogy is suggestive, not demonstrated.\n\nThis is for people who actually measure or model SOT in ultrathin films and care about efficiency near criticality. The experimental claim and the GMS tool deserve a serious referee; the length-scale assumption and device pitch can be tightened in revision. I would send it out.","headline":"Solid experimental observation of opposite DL/FL SOT trends above TC in ultrathin CoFeB, with a useful GMS method; the mixing interpretation is plausible but rests on an under-constrained length-scale assumption.","tokens_in":19465,"tokens_out":520,"would_cite":true,"duration_ms":6744,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.5","headline":"Magnetic fluctuations above the Curie point boost damping-like spin-orbit torque while suppressing the field-like part.","keywords":["spin-orbit torque","Curie temperature","magnetic fluctuations","spin-mixing conductance","ultrathin ferromagnets","CoFeB/Pt","CoFeB/Ta","heat-assisted spintronics"],"falsifier":"Measure damping-like and field-like efficiencies in the same bilayers while independently tuning the magnetic correlation length (for example by controlled interfacial roughness or an antiferromagnetic spacer) so that it becomes longer than the spin-diffusion length; if the DL enhancement and FL suppression both disappear, the short-scale-mixing mechanism is ruled out.","tokens_in":19386,"feed_emoji":"🧲","tokens_out":625,"duration_ms":6497,"temperature":0.7,"pith_summary":"This paper shows that when an ultrathin ferromagnet is driven above its Curie temperature by confinement, magnetic fluctuations do not simply dilute the magnetization. They reorganize how the ferromagnet absorbs spin current from a neighboring heavy-metal layer. The damping-like spin-orbit field that drives useful switching and magnon generation rises sharply, while the field-like field falls, and the two quantities move in opposite directions when an external field is applied. The authors attribute the divergence to geometric averaging of the interfacial spin-conductance tensor: fluctuations mix the large longitudinal conductance into the transverse channel, so more of the transversely polarized spin current is absorbed. Because the same mixing can be engineered statically or by short-wavelength magnons generated by the spin current itself, the result points to a practical route for raising spin-orbit-torque efficiency without new materials, analogous to heat-assisted magnetic recording.","feed_headline":"Fluctuations above TC boost damping-like spin-orbit torque","feed_subtitle":"Mixing of longitudinal and transverse spin conductances raises absorption of spin current","key_machinery":"The angular average of the interfacial spin-conductance tensor over short-scale magnetization fluctuations (Eq. 5): the real (damping-like) mixing conductance is increased by a term proportional to the longitudinal conductance times the mean-square transverse angle, while the imaginary (field-like) part is reduced by the average cosine of the fluctuation angle.","core_discovery":"Above the Curie temperature of an ultrathin ferromagnet, the damping-like spin-orbit field is strongly enhanced and the field-like field is suppressed, with opposite field dependences; both behaviors match the prediction of fluctuation-driven mixing between the longitudinal and transverse channels of the interfacial spin conductance.","pith_inferences":[],"forward_implications":[],"fun_headline_variants":["Fluctuations mix spin conductances to enhance damping-like torque above TC","Above TC damping-like spin-orbit field rises as field-like contribution falls","Fluctuation-driven channel mixing raises transverse spin current absorption","Ultrathin magnets show opposite SOT field trends from magnetic fluctuations","Spin-current magnons and engineered fluctuations can activate the effect below TC"],"cache_read_input_tokens":16512,"weakest_assumption_plain":"The paper assumes that magnetic fluctuations occur on length scales shorter than the spin-diffusion length, so the interface can be replaced by a simple angular average of the spin-conductance tensor.","fun_headline_variants_meta":{"raw":{"variants":["Fluctuations mix spin conductances to enhance damping-like torque above TC","Above TC damping-like spin-orbit field rises as field-like contribution falls","Fluctuation-driven channel mixing raises transverse spin current absorption","Ultrathin magnets show opposite SOT field trends from magnetic fluctuations","Spin-current magnons and engineered fluctuations can activate the effect below TC"]},"model":"grok-4.5","effort":"low","cost_usd":0.006988,"raw_usage":{"total_tokens":1634,"prompt_tokens":656,"num_sources_used":0,"completion_tokens":98,"cost_in_usd_ticks":69880000,"prompt_tokens_details":{"text_tokens":656,"audio_tokens":0,"image_tokens":0,"cached_tokens":128},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":880,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":656,"tokens_out":98,"duration_ms":11663,"temperature":1.0,"reasoning_tokens":880,"cache_read_input_tokens":128,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-10T23:02:30.499171+00:00","model_set":{"reader":"grok-4.5"},"falsifier":"Measure damping-like and field-like efficiencies in the same bilayers while independently tuning the magnetic correlation length (for example by controlled interfacial roughness or an antiferromagnetic spacer) so that it becomes longer than the spin-diffusion length; if the DL enhancement and FL suppression both disappear, the short-scale-mixing mechanism is ruled out.","supporting_citations":[],"review_version":1}