Pith. sign in

REVIEW 4 major objections 5 minor 5 references

Temperature dependent spin dynamics in La$_{0.67}$Sr$_{0.33}$MnO$_3$/Pt bilayers

T0 review · 4 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Thin La0.67Sr0.33MnO3/Pt bilayers pump their maximum spin current at 170 K, where magnetic damping reaches 0.0020, linewidth 12 Oe, and spin Hall angle 3.2%.

desk verdict A genuinely new device-level temperature-dependent spin-pumping dataset for thin LSMO/Pt, with a credible 170 K optimum but quantitative spin Hall angle claims that rest on an unverified all-spin-pumping assumption. read the letter →

arxiv 2411.15417 v3 pith:4CTEUUJ7 submitted 2024-11-23 cond-mat.mtrl-sci

Biswajit Sahoo , Akilan K , Katherine Matthews , Alexandre Pofelski , Alex Frano , Eric E Fullerton , Sebastien Petit-Watelot , Juan-Carlos Rojas Sanchez
show 1 more author
Sarmistha Das
This is my paper · ORCID
classification cond-mat.mtrl-sci
keywords ferromagneticresonancespinpumpingspintronicdevicescomplexperovskitethinfilmsmicro-devicesLa0.67Sr0.33MnO3HallangleGilbertdamping
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper sets out to show that a 15-nanometre film of the magnetic oxide La0.67Sr0.33MnO3 capped with 5 nanometres of platinum is an efficient spin-current source, and that its best operating point is 170 K. The authors measure ferromagnetic resonance in unpatterned films and spin-pumping voltages in patterned micro-devices from 300 K down to 90 K, comparing the bilayer with the standard permalloy (NiFe)/Pt device. At 170 K they report the maximum pumped spin current into Pt (about 3.5 nJ/m2), an unusually low magnetic damping (Gilbert damping constant 0.0020), a resonance linewidth of about 12 Oe, and a spin Hall angle (charge-to-spin conversion efficiency) of about 3.2%. They argue that a phase-separated magnetically active layer at the interfaces acts as an extra spin sink below about 150 K, which is why the spin current peaks at an intermediate temperature rather than at the coldest temperature. If correct, the result points to thin LSMO/Pt bilayers as a temperature-tunable, low-power building block for spin-based oscillators.

What carries the argument

The load-bearing experimental method is device-level spin-pumping ferromagnetic resonance: a 500 µm by 10 µm wire is patterned from the LSMO/Pt film, a radio-frequency current excites uniform precession, and the DC voltage generated by the inverse spin Hall effect in Pt is measured as a function of field. The quantitative conversion from that voltage to a spin current and a spin Hall angle runs through two equations: Eq. (6), which converts the damping increase $\Delta\alpha = \alpha_{\mathrm{LSMO/Pt}} - \alpha_{\mathrm{LSMO}}$ into the spin current density $J_s$ via the effective spin mixing conductance $g_{\mathrm{eff}}$, and Eq. (5), which converts the spin-pumping current into the spin Hall angle $\theta_{\mathrm{SHA}}$ using an assumed Pt spin diffusion length of 3.4 nm. Eq. (7) then reports the interfacial transparency from the same $g_{\mathrm{eff}}$. The named mechanism carrying the temperature interpretation is the PSMA layer—a phase-separated magnetically active region with uncompensated antiferromagnetically coupled spins, detected through Nd interdiffusion—which the paper invokes as an additional spin sink that grows in importance below 150 K.

What would settle it

Measure the damping enhancement in LSMO/Pt as a function of Pt thickness at 170 K while holding the LSMO layer fixed. If the spin-pumping-only picture is correct, the derived spin mixing conductance should be independent of Pt thickness and the spin-pumping voltage should scale with $\tanh(t_{\mathrm{Pt}}/2\lambda_{\mathrm{SF}})$; a systematic deviation—especially in samples that also show the low-temperature damping upturn—would indicate that the PSMA layer is a second spin sink and that the damping-based spin current and spin Hall angle values are biased.

Watch

Extended reading notes

Core claim

The central discovery is that the spin current injected into Pt from a 15 nm La0.67Sr0.33MnO3 film is not monotonic in temperature: it rises as the sample cools, peaks near 170 K, and then falls as the temperature approaches 90 K. The paper attributes this to a competition between two effects of cooling. On one side, the magnetization and exchange stiffness of LSMO grow, strengthening the uniform precession and therefore the spin pumping into Pt. On the other side, a phase-separated magnetically active (PSMA) layer—identified by electron energy-loss spectroscopy as a region of Nd interdiffusion at the NdGaO3/LSMO interface—becomes an increasingly effective spin sink below roughly 150 K, lowering the LSMO/Pt interfacial transparency from about 76% at room temperature to about 30% at 90 K and absorbing part of the pumped angular momentum. The quantitative yield of the study is the 170 K optimum: a pumped spin current of about 3.5 nJ/m2, a Gilbert damping of 0.0020 ± 0.0005, a 12 ± 1 Oe linewidth, and a spin Hall angle of 3.2 ± 0.4%, with a fourfold larger device signal than a comparably prepared permalloy/Pt bilayer.

Load-bearing premise

The quantitative numbers stand on the assumption that the entire increase in magnetic damping when Pt is added to LSMO comes from spin pumping into Pt; the paper also invokes a phase-separated magnetic layer that absorbs spin angular momentum, and if that layer contributes to the measured damping increase, the extracted spin current density and spin Hall angle are systematically biased.

Editorial extensions

If this is right

  • If correct, the LSMO/Pt spin-current source can be operated at a 170 K sweet spot where the output is maximized, which is useful for cryogenic spintronics and neuromorphic circuits that already run cold.
  • The 0.0020 damping and 12 Oe linewidth imply that the threshold current for sustained precession in an LSMO/Pt nano-oscillator would be substantially lower than in a permalloy/Pt device.
  • The fourfold signal advantage over permalloy/Pt at comparable device resistance, together with roughly eight times lower damping, identifies LSMO/Pt as a competitive replacement for metallic bilayers in the 150–200 K range.
  • The spin Hall angle of 3–4% measured between 250 K and 170 K indicates that the LSMO/Pt interface does not degrade Pt's intrinsic spin-charge conversion, so the bilayer combines a strong spin injector with a good spin detector.
  • The drop in interfacial transparency from 76% at 300 K to 30% at 90 K shows that the low-temperature limit of the device is set by interface spin scattering rather than by the magnetic film's intrinsic properties.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Editorial inference: if the PSMA-sink picture is correct, the 170 K optimum is not a fixed property of LSMO/Pt but a crossover set by interface quality, so suppressing Nd interdiffusion with a buffer layer or a different substrate should shift the spin-current peak toward room temperature.
  • Editorial inference: the paper's own caution that the 90 K spin Hall angle (about 9%) is 'seemingly increased' suggests the true Pt spin Hall angle in this stack is closer to the 3–4% value measured between 250 K and 170 K, and a thickness-series experiment would test whether the low-temperature rise is an artifact of the PSMA sink.
  • Editorial inference: an independent measurement of the spin Hall angle by spin-torque ferromagnetic resonance or harmonic Hall on the same devices would bypass the damping-enhancement assumption; agreement with 3.2% would strengthen the claim, while disagreement would quantify how much angular momentum the PSMA layer absorbs.
  • Editorial inference: the practical cost of cooling to 170 K is significant for applications, but the paper's logic implies that interface engineering, not the intrinsic oxide, is the limiting factor; a room-temperature LSMO/Pt device with comparable interface quality would be a more direct competitor to metallic spin Hall oscillators.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The paper reports temperature-dependent ferromagnetic resonance (FMR) and spin-pumping FMR (SP-FMR) measurements on 15-nm La0.67Sr0.33MnO3 (LSMO) films and LSMO(15)/Pt(5) bilayers on NdGaO3(110), from 300 K down to 90 K, with a Py(7)/Pt(5) reference. The authors find that the Gilbert damping and FMR linewidth of LSMO/Pt reach a minimum near 170 K, that the spin-pumping voltage and derived spin current density Js peak near 170 K, and that the derived spin Hall angle is about 3.2% at that temperature. They also report a roughly fourfold larger SP-FMR voltage for LSMO/Pt than for Py/Pt at comparable device resistance, and they attribute the low-temperature degradation of spin transmission to a phase-separated magnetically active (PSMA) layer at the LSMO interfaces, supported by EELS evidence at the NGO/LSMO interface.

Significance. If the quantitative extraction is correct, thin LSMO/Pt is an attractive temperature-tunable spin-current source for spin-orbit-torque oscillators, combining lower damping and linewidth than Py/Pt with a competitive spin Hall angle. The manuscript has real strengths: the raw resonance-field, linewidth, and spin-pumping voltage data are directly measured; the temperature-dependent comparison with Py/Pt is useful; and the EELS characterization adds microstructural information. However, the central quantitative claims—the 170 K maximum in Js, the 3.2% spin Hall angle, and the interface transparency values—are not direct measurements. They depend on Eqs. (5)–(7), in which the entire damping increase of LSMO/Pt relative to LSMO is attributed to spin pumping into Pt and in which literature parameters such as λ_SF are used. The qualitative picture of a 170 K optimum is credible, but the quantitative level of confidence claimed in the abstract is not yet supported.

major comments (4)
  1. [Eq. (6) and Fig. 5] The derivation of geff from Δα = α_LSMO/Pt − α_LSMO assumes that the entire Pt-induced damping increase is due to spin pumping into Pt. The manuscript simultaneously invokes a PSMA layer at the LSMO/Pt interface as a spin sink below about 150 K and, at 130–90 K, states that the additional damping makes θSHA 'seemingly increased' and 'may not reflect the true spin-charge conversion efficiency.' Because the same Δα enters geff, Js, θSHA, and the transparency in Eq. (7), any PSMA or interface-modification contribution to α_LSMO/Pt biases all of those quantities at every temperature, including the 170 K maximum and the 3.2% θSHA. The EELS evidence in Fig. S3 demonstrates Nd interdiffusion at the bottom NGO/LSMO interface, not at the top LSMO/Pt interface where the spin sink is invoked, so the assumed top-interface PSMA contribution is uncharacterized. A Pt-thickness series, or a control experiment with an inert cap, is needed to separate the spin-pumping channel from other interface damping mechanisms; without such a control the central quantitative claim is not established.
  2. [Eq. (6) and Figs. 3–4] In Eq. (6), Js scales as (1/α)^2, and the bare LSMO film has a damping minimum near 170 K (α ≈ 0.0013). The maximum in Js at 170 K could therefore be dominated by the small denominator α^2 in Eq. (6) rather than by a maximum in interfacial spin transport. The manuscript does not test whether the 170 K maximum survives when the device-level damping is used. This is a concrete concern because the SP-FMR linewidths on patterned devices are systematically larger than the unpatterned-film FMR linewidths (e.g., 24 Oe versus 11 Oe at 200 K), indicating that device fabrication changes the damping; the actual device-level α is not measured, so the precession cone angle entering Eq. (6) is not the one appropriate to the measured voltages.
  3. [Eq. (5) and Fig. 5(c)] The spin Hall angle is obtained from Eq. (5) using λ_SF = 3.4 nm taken from a single literature reference (ref. [7]). The spin diffusion length and interface spin-memory loss in Pt depend strongly on film microstructure, interface quality, and temperature, and the same λ_SF is also used in the transparency expression in Eq. (7). The quoted uncertainty of ±0.4% on θSHA reflects only the scatter in voltage and damping measurements, not the uncertainty in λ_SF or in the assumed gr = geff identification. A sensitivity analysis with respect to λ_SF and an explicit statement of its uncertainty are required before reporting 3.2 ± 0.4% as a robust value.
  4. [Fig. 5(a) and text after Eq. (6)] The text reports the maximum spin current density as '3.5 ± 0.1 nJ/m2', while Eq. (6) has the dimensions of a spin current density (equivalently a charge current density in A/m²). The units appear inconsistent between the equation and the quoted figure-axis value; this should be reconciled, since the numerical value of Js is one of the central quantitative claims.
minor comments (5)
  1. [Supplementary title] The supplementary header lists 'La0.67Sr0.33MO3' where the material is La0.67Sr0.33MnO3; the 'n' is missing.
  2. [Fig. S8 discussion] The text uses both 'Py(7)/Pt(5)' and 'Pt(7)/Py(5)' for the reference bilayer; the notation should be made consistent.
  3. [Eq. (5)] The subscript 'sw' in Eq. (5) appears to denote the product of spin current density Js and width w; the symbol w should be defined explicitly in the text and used consistently.
  4. [Fig. 5(a) caption] The caption and text state that V_SP at 300 K is not plotted because of the two-mode fitting; this explanation should also be included in the figure caption for self-containedness.
  5. [General notation] The samples are referred to variously as LSMO(15)/Pt(5), LSMO/Pt, S1, and S2; defining the sample labels once in a table would improve readability.

Circularity Check

0 steps flagged · score 2.0 of 10

No constructed circularity: the 170 K maximum rests on directly measured SP-FMR voltages; the spin-pumping analysis uses standard formulas with only minor self-cited parameters.

full rationale

The central claim, a maximum spin current pumped into Pt near 170 K, is anchored by the directly measured device-level SP-FMR voltage V_SP (Fig. 4(b)), an independent electrical signal rather than a function of fitted damping parameters. The conversion of V_SP to J_s and theta_SHA follows standard Tserkovnyak/Rogdakis expressions (Eqs. 5-6), with g_eff computed from Delta_alpha = alpha_LSMO/Pt - alpha_LSMO; this is a modeling attribution of the measured damping difference to spin pumping, not a circular reduction, and the paper itself flags the low-temperature inflation of theta_SHA ('a seemingly increased value of thetaSHA which may not reflect the true spin-charge conversion efficiency'). The interfacial transparency T (Eq. 7) is a derived function of the same g_eff and is used to motivate the PSMA spin-sink interpretation, which is the weakest, self-referential part of the analysis; however, it is not used to produce the headline V_SP maximum and is presented as an interpretation rather than an independent prediction. The EELS-detected PSMA phase is at the NGO/LSMO bottom interface, while the spin-sink role is invoked at the top LSMO/Pt interface, an evidence gap that is a correctness/assumption risk rather than circularity. Self-citations appear for lambda_SF = 3.4 nm (ref. [7] by a co-author) and for Eq. 7 (ref. [10] by co-authors), but these are externally established material/formula inputs, not uniqueness claims or ansatze unique to this paper; none of the central claims reduces to them by construction. Score 2 reflects these minor self-referential parameter choices, not a derivational circularity.

Assumptions & free parameters 3 free parameters · 4 assumptions · 2 invented entities

The main quantitative results depend on one literature-assumed parameter (λ_SF), on the geff extraction from Δα, and on the existence and behavior of the PSMA layer. The PSMA layer is the most consequential invented entity: it is used to explain the low-temperature reduction of spin current, but it is not directly observed at the top interface. The Kambersky fit parameters are included for completeness; they are not load-bearing for the central claim.

free parameters (3)
  • Platinum spin diffusion length λ_SF = 3.4 nm (assumed from ref [7])
    Used in Eqs. 5 and 7 to convert spin-pumping voltage to spin Hall angle and to compute interface transparency. The extracted SHA scales with this value; it is not measured in this work.
  • Kambersky coefficient α_inter (resistivity-like) = 0.00303 (S1), 0.00723 (S2)
    Fitted to damping vs temperature in the supplementary (Eq. SE1) to model the damping minimum.
  • Kambersky coefficient α_intra (conductivity-like) = 2.12e-4 (S1), 2.48e-4 (S2)
    Fitted to damping vs temperature in the supplementary (Eq. SE1). These do not affect the main spin-current result but are fitted model parameters.
assumptions (4)
  • domain assumption The damping enhancement Δα = α_LSMO/Pt - α_LSMO is entirely due to spin pumping into Pt.
    Used to compute geff via geff = 4πMs dFM Δα / g μB (around Eq. 6). The paper also invokes PSMA as an additional spin sink, so this assumption is internally inconsistent.
  • domain assumption No spin-memory loss at the LSMO/Pt interface; bulk Pt spin diffusion length (3.4 nm) fully describes spin transport.
    Eqs. 5 and 7 contain no interface spin-loss parameter, although the cited ref [7] emphasizes spin-memory loss at metallic interfaces.
  • standard math Kittel resonance equation and linear linewidth-frequency relation hold.
    Used to extract Meff, Hk, and α via Eqs. 2 and 3.
  • domain assumption Kambersky torque-correlation model describes the temperature dependence of damping.
    Used in supplementary to fit damping vs T (Eq. SE1); deviations are attributed to PSMA.
invented entities (2)
  • PSMA layer at the top LSMO/Pt interface (and bottom NGO/LSMO interface)
    purpose: Explains the decrease in interface transparency and spin current below 150K, and the extra damping at low temperature.
    EELS confirms a Nd-interdiffused phase at the bottom interface, but the top PSMA layer is inferred from the temperature dependence of model-derived quantities (damping, transparency), not directly observed. The 'direct evidence' claim in the text is therefore not fully independent.
  • Magnetically dead layer at the LSMO/Pt interface (~1 nm)
    purpose: Explains the slightly reduced areal moment of LSMO/Pt relative to bare LSMO.
    Inferred from the ratio of VSM areal moments at room temperature (Fig. S7a); no direct imaging of a dead layer is provided.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Temperature dependent spin dynamics in La$_{0.67}$Sr$_{0.33}$MnO$_3$/Pt bilayers." pith.science (2026). https://pith.science/paper/4CTEUUJ7

@misc{pith2026241115417,
  author       = {Pith},
  title        = {Pith review of: Temperature dependent spin dynamics in La$_0.67$Sr$_0.33$MnO$_3$/Pt bilayers},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/4CTEUUJ7}},
  note         = {Machine review of arXiv:2411.15417}
}
abstract

Complex ferromagnetic oxides such as La$_{0.67}$Sr$_{0.33}$MnO$_3$ (LSMO) offer pathways for creating energy efficient spintronic devices with new functionalities. LSMO exhibits high-temperature ferromagnetism, half metallicity, sharp resonance linewidth, low damping and a large anisotropic magnetoresistance response. Combined with Pt, a proven material with high spin-charge conversion efficiency, LSMO can be used to create robust nano-oscillators for neuromorphic computing. Ferromagnetic resonance (FMR) and device level spin-pumping FMR measurements are performed to investigate the magnetization dynamics and spin transport in NdGaO3(110)/LSMO(15 nm)/Pt(0 and 5 nm) thin films ranging from 300K to 90K and compare the device performance with Py(7 nm)/Pt(5 nm) sample. The spin current pumped into Pt is quantified to determine the temperature dependent influence of interfacial interactions. The generated spin current in the micro-device is maximum at 170K for the optimally grown LSMO/Pt films. Additionally, this bilayer system exhibits low magnetic Gilbert damping (0.002), small linewidth (12 Oe) and a large spin Hall angle ($\approx$ 3.2%) at 170K. By fine-tuning the LSMO/Pt interface quality and integrating it into the device structure, the system exhibits a fourfold enhancement in signal output for LSMO/Pt devices compared to the Pt/Py system. Such robust device level performance can pave way for energy-efficient spintronic based devices.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

5 extracted references · 5 canonical work pages

  1. [1]

    Extrinsic contributions to the ferromagnetic resonance response of ultrathin films,

    R. Arias and D. L. Mills, "Extrinsic contributions to the ferromagnetic resonance response of ultrathin films, " Physical Review B, vol. 60, p. 7395–7409, September 1999

    work page 1999

  2. [2]

    Anomalous Damping of Spin Waves in Magnetic Metals,

    V . Korenman and R. E. Prange, "Anomalous Damping of Spin Waves in Magnetic Metals, " Physical Review B, vol. 6, p. 2769–2777, October 1972

    work page 1972

  3. [3]

    On ferromagnetic resonance damping in metals,

    V . Kamberský, "On ferromagnetic resonance damping in metals, " Czechoslovak Journal of Physics, vol. 26, p. 1366–1383, December 1976

    work page 1976

  4. [4]

    Temperature dependence of the Gilbert damping of La0.7Sr0.3MnO3 thin films,

    V . Haspot, P . Noël, J.-P . Attané, L. Vila, M. Bibes, A. Anane and A. Barthélémy, "Temperature dependence of the Gilbert damping of La0.7Sr0.3MnO3 thin films, " Physical Review Materials, vol. 6, p. 024406, February 2022

    work page 2022

  5. [5]

    Thickness and temperature-dependent damping in La0.67Sr0.33MnO3 epitaxial films,

    Y . Wang, X. Fan, X. Feng, X. Gao, Y . Ke, J. Yao, M. Guo, T. Wang, L. Shen, M. Liu, D. Xue and X. Fan, "Thickness and temperature-dependent damping in La0.67Sr0.33MnO3 epitaxial films, " Applied Physics Letters, vol. 123, September 2023

    work page 2023

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.