{"id":"b649f3c0-80d2-463b-bfae-27f3977a2e98","arxiv_id":"2506.08425","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Across 19 transition metals, the orbital Hall contribution to spin-pumping charge conversion is generally larger than the spin Hall contribution in the measured heterostructures.","lead":"This paper measures how 19 different transition metals convert angular momentum currents into electrical voltage using a magnetic resonance technique. It finds that the orbital part of that conversion is often much larger than the spin part, which matters for designing new spintronic and orbitronic devices.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Beta=0.23 in Eq. (2) is internally inconsistent with the reported 90 nA/109 nA values and ignores material-dependent Pt/X interface conductances; the corrected subtraction may erase orbital dominance in strong-SOC metals.","rationale":"The paper is a broad experimental survey whose novel claim is orbital dominance. I read it in good faith: the sign and trend agreement with Ref. 30 for many elements, the Cu null control, and negative OHC in Au and Ag are meaningful independent supports. However, all quantitative conclusions pass through Eq. (2), and that equation's two inputs are underdetermined. The beta=0.23 value appears to conflict with the numbers in the text (90 nA versus 109 nA implies 0.83); even if it is a different quantity, no formula is given for its extraction. The assumption that V_ISHE^X measured in a bilayer can be rescaled by a single constant to represent the ISHE contribution in a trilayer conflicts with standard spin-pumping phenomenology because the injected spin current depends on the F/N interface mixing conductance and on the N/N' interface. A material-dependent interface would change the correction term most for the very metals where ISHE is largest, which are those whose orbital dominance is most surprising. Thus the central claim is conditional on a calibration that is not demonstrated. The recommended verdict remains conditional because the concern is checkable by re-analysis and additional thickness series; it does not require rejecting the entire dataset, nor is there evidence of bad faith.","tokens_in":10562,"tokens_out":7698,"duration_ms":95616,"concrete_test":"Digitize Fig. 1(c) to verify the claimed 23% value against the stated 90 nA at 2 nm and 109 nA saturation; then recompute the extracted IOHE values in Fig. 3(d) for W, Ta, Ir, and Pd using (i) beta=0.83 and (ii) a beta independently measured from a Pt-thickness series in YIG/Pt(t)/X for each metal. If any strong-ISHE metal's corrected IOHE changes sign or becomes smaller than its ISHE contribution, the 'overwhelming orbital dominance' claim fails for that material.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim rests on Eq. (2), where the IOHE signal is isolated as Vtot - V_ISHE^Pt(2) - beta*V_ISHE^X(5), with beta=0.23 said to be the fraction of the pumped spin current that reaches the Pt/X interface. Two defects make this subtraction load-bearing and currently insecure. First, the calibration is internally inconsistent. The text reports I_ISHE^Pt(2)=90 nA for YIG/Pt(2) and states that the charge current saturates at 109 nA as Pt thickness increases; that ratio is 0.83, not 0.23. If beta were 0.83, the subtracted ISHE term for strong-ISHE metals (W, Ta, Ir, Pd) would be more than three times larger than the value used, directly lowering the extracted IOHE and potentially reversing the orbital-dominance conclusion for those metals. No derivation of 0.23 from Fig. 1(c) is shown. Second, even if beta were correct as a Pt spin-current attenuation factor, Eq. (2) applies it to the ISHE voltage measured in the YIG/X(5) bilayer. That reference signal is set by the YIG/X spin-mixing conductance, whereas in YIG/Pt(2)/X the spin current entering X is set by the YIG/Pt and Pt/X interface conductances, including possible spin memory loss. The reference and the trilayer therefore scale differently; beta cannot absorb material-dependent interfaces unless those interfaces are characterized. The linewidth data and the Cu control support the qualitative picture, but they do not calibrate the subtraction used to claim dominance.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports spin-pumping ferromagnetic resonance measurements on YIG/X(5) bilayers and YIG/Pt(2)/X(5) trilayers for 19 transition metals, aiming to separate inverse spin Hall (ISHE) and inverse orbital Hall (IOHE) contributions. The central claim is that the orbital contribution overwhelmingly dominates over the spin response in the trilayer charge-to-current conversion, and that negative orbital Hall conductivities occur in some metals. The extraction uses Eq. (2), which subtracts the fixed Pt(2) ISHE signal and a fraction β=0.23 of the X-layer ISHE signal, with β interpreted as the fraction of the pumped spin current that reaches the Pt/X interface. The authors compare their experimental ISHE and IOHE values with first-principles calculations from Go et al. and report qualitative agreement in trends, with discrepancies for several elements.","tokens_in":10894,"tokens_out":5635,"duration_ms":63371,"significance":"If the central extraction were reliable, this would be a valuable systematic dataset: a single study spanning 3d, 4d, and 5d transition metals, with explicit comparison to first-principles spin and orbital Hall conductivities, inclusion of linewidth-broadening data, and a Cu control. The confirmation of negative orbital Hall conductivity in Au, Ag, and possibly other metals, and the demonstration that light metals such as Mo and Zr can produce strong IOHE, would be of genuine interest to the orbitronics community. However, the quantitative claim of orbital dominance rests entirely on the calibration of β in Eq. (2), and that calibration is neither derived nor internally consistent with the reported numbers. Until this is resolved, the headline claim is not supported by the presented analysis.","major_comments":[{"comment":"The calibration β=0.23 is not supported by the data shown. The text reports I_ISHE^Pt(2)=90 nA and states that the charge current saturates at 109 nA as the Pt thickness increases, so the ratio of these values is 0.83, not 0.23. More fundamentally, Fig. 1(c) plots the ISHE charge current generated within the Pt layer itself, which scales approximately as tanh(t/2λ_s) and is not equal to the spin current transmitted to the top Pt/X interface. The 23% value therefore does not follow from the thickness dependence as stated. Because Eq. (2) uses this unsupported β, the extracted IOHE values in Fig. 3(d) are not reliable; for strong-ISHE metals such as W, Ta, Ir, and Pd, using the ratio 0.83 would more than triple the subtracted ISHE term and could change the magnitude or even the sign of the extracted IOHE. The authors must provide a derivation of β from the thickness-dependent data, with the assumed spin-current profile stated explicitly, and validate the transferability of β across the 19 materials.","section":"II, Eq. (2) and Fig. 1(c)"},{"comment":"The reference signal V_ISHE^X(5) is measured in a YIG/X(5) bilayer, but in YIG/Pt(2)/X(5) the spin current that enters X is governed by the YIG/Pt and Pt/X interface spin conductances, including possible spin memory loss at the Pt/X interface. Those interface properties are material-dependent, so a single scalar β cannot convert the YIG/X bilayer ISHE signal into the X-layer ISHE contribution in the trilayer unless the interfaces are explicitly characterized. The Cu control demonstrates only that Cu has negligible IOHE and ISHE in this geometry; it does not calibrate β for other materials. Please provide an explicit spin-transport model for the trilayer and state the assumptions on Pt/X spin transmission and spin memory loss.","section":"II and III, Eq. (2) and reference YIG/X(5) samples"},{"comment":"No error bars, sample-to-sample reproducibility data, or uncertainties from the Lorentzian fits are reported. The abstract's claim that the orbital contribution 'overwhelmingly dominates' is based on point estimates, and Fig. 3(d) uses a vertical scale three times larger than Fig. 3(b), which visually exaggerates the orbital dominance. Quantitative confidence intervals for the extracted IOHE values are necessary before this claim can be assessed, especially given the sensitivity of the subtraction in Eq. (2) to the value of β.","section":"III, Fig. 3(d) and conclusions"},{"comment":"For the magnetic X elements (Fe, Co, Ni, Cr), the manuscript states that a Cu(5) spacer was introduced in series A to avoid magnetic coupling. The reference signal labeled V_ISHE^X(5) for these elements is therefore measured in YIG/Cu(5)/X(5), not in YIG/X(5), and the inset in Fig. 2 confirms this geometry. This reference does not represent the ISHE of a bare X layer and does not describe the spin current that would reach X in the YIG/Pt(2)/X trilayer. Using these values in Eq. (2) for Fe, Co, Ni, and Cr is inconsistent with the models used for the other 15 elements, and the extracted IOHE values for these four elements are therefore not directly comparable.","section":"II and III, magnetic 3d metals"}],"minor_comments":[{"comment":"Equation (1) contains a factor f (frequency) but the text defines h_rf as the amplitude of the microwave magnetic field; the equation appears to omit h_rf and to use the symbol L_pxz, while the text defines p_yz. Please clarify the notation and verify the dimensional consistency of Eq. (1).","section":"II, Eq. (1)"},{"comment":"The reference numbering jumps from [26] to [28], so reference [27] is missing; please renumber or insert the intended citation.","section":"References"},{"comment":"The sentence 'The left and right signals correspond the ISHE contributions' should read 'correspond to the ISHE contributions'.","section":"II, paragraph after Eq. (2)"},{"comment":"The phrase 'the vertical scale of the IOHE is three times larger' appears only in the caption of Fig. 3; please state this explicitly in the main text or adjust the figure so the comparison is not misleading.","section":"III, 3d transition metals paragraph"}],"recommendation":"major_revision","confidential_remarks":"The manuscript addresses a timely and important question, and the systematic experimental coverage of 19 transition metals is a strength. My main concern is the β=0.23 calibration in Eq. (2), which is internally inconsistent with the reported 90 nA and 109 nA values and is not derived from Fig. 1(c). This is a load-bearing point for the orbital-dominance claim. I would ask the editor to require the authors to supply a detailed derivation of β, an explicit trilayer spin-transport model, and uncertainty estimates for the extracted IOHE values before publication. If the calibration cannot be justified, the central conclusion may need to be substantially weakened."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Net: this is a useful experimental survey, but the headline claim is not yet as secure as the authors present it. The 19-metal SP-FMR dataset is a real addition, the Cu control is clean, and the sign and trend agreement with Go et al.'s first-principles conductivities gives the qualitative picture substance. The linewidth measurements and the candid discussion of disorder, texture, and band-structure sensitivity add credibility. The problem is the extraction in Eq. (2). The authors say beta = 0.23 is read off Fig. 1(c), where the Pt(2) ISHE signal is 90 nA and the saturated thick-Pt signal is 109 nA; that ratio is 0.83, not 0.23. No other derivation is given, so the central subtraction is not reproducible from the text. Second, beta is applied to V_ISHE^X measured in YIG/X bilayers, which assumes the spin current entering X in the trilayer differs only by a Pt attenuation factor. That ignores material-dependent interface conductances and spin memory loss. For weak-SOC metals the ISHE subtraction is tiny, so those IOHE values should be robust; for W, Ta, Ir, and Pd the extracted numbers shift noticeably with beta, which is exactly where the orbital-dominance claim needs to be firm. Missing error bars and an 'available on request' data statement do not help. These are fixable shortcomings, not a broken premise. The qualitative conclusion probably survives for much of the periodic table; the quantitative 'overwhelmingly dominates' line needs more work. I would send this to peer review, but a serious referee should ask for a proper beta calibration, per-material interface checks, and deposited data. It is a good reading-group paper for orbitronics and spintronics groups; I just would not build on the numbers until the subtraction is cleaned up.","headline":"Broad and useful survey, but the beta=0.23 subtraction in Eq. (2) does not square with the paper's own 90 nA/109 nA numbers, so the orbital-dominance claim needs support.","tokens_in":11438,"tokens_out":5857,"would_cite":false,"duration_ms":64022,"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":"Orbital Hall effect dominates spin conversion in transition-metal heterostructures","keywords":["orbital Hall effect","inverse orbital Hall effect","spin pumping","ferromagnetic resonance","spin Hall effect","transition metals","orbitronics","spin-to-charge conversion"],"falsifier":"A decisive check would be to fabricate YIG/Pt($t$)/X(5) for a range of Pt thicknesses $t$ for several X metals and test whether the extracted IOHE value is independent of $t$. If the extracted IOHE changes with $t$, the $\\beta = 0.23$ transmission model is wrong; a complementary check would be to insert a thin Cu layer at the Pt/X interface and see whether the orbital signal disappears while the spin signal remains.","tokens_in":10343,"feed_emoji":"🧲","tokens_out":5073,"duration_ms":59026,"temperature":0.7,"pith_summary":"This paper claims that in spin-pumping measurements on YIG/Pt(2)/X(5) stacks, where X is one of 19 transition metals, the measured charge current comes mostly from the inverse orbital Hall effect (IOHE) in the X layer, not from the inverse spin Hall effect (ISHE). The authors extract the IOHE contribution by subtracting the fixed Pt(2) ISHE signal and a scaled fraction of the X-layer ISHE signal. If correct, this means orbital angular momentum transport is the dominant route for spin-to-charge conversion in these systems, and materials previously considered weak spin-Hall metals can still be strong orbital-to-charge converters. The results also provide experimental confirmation of negative orbital Hall conductivities in several metals, matching first-principles calculations in most cases.","feed_headline":"Orbital Hall effect dominates spin conversion in 19 metals","feed_subtitle":"Spin-pumping measurements show the inverse orbital Hall effect supplies most of the charge current in YIG/Pt/metal stacks.","key_machinery":"The key object is Eq. (2), the two-layer decomposition $V_{\\mathrm{tot}} = V_{\\mathrm{ISHE}}^{\\mathrm{Pt(2)}} + (V_{\\mathrm{IOHE}}^{\\mathrm{X(5)}} + \\beta V_{\\mathrm{ISHE}}^{\\mathrm{X(5)}})$, in which $\\beta = 0.23$ is the fraction of spin current transmitted through the Pt(2)/X interface, calibrated from the Pt-thickness dependence of YIG/Pt($t$). This decomposition is what turns the total FMR-driven voltage into an isolated IOHE estimate for each metal. The physical carrier is the coupled spin-orbital current generated by Pt's strong spin-orbit coupling, which injects orbital angular momentum into the X layer.","core_discovery":"The central discovery is that the IOHE is the dominant mechanism converting the pumped spin-orbital current into charge in YIG/Pt(2)/X(5) heterostructures. For most of the 19 transition metals studied, the extracted IOHE current is larger than the ISHE current from the same layer, and in some cases, such as Mo, the orbital signal exceeds 260 nA while the spin signal is below 13 nA. The authors show that the experimental ISHE values largely track theoretical spin Hall conductivities, while the IOHE values show larger deviations, which they attribute to the sensitivity of orbital transport to crystal texture, disorder, and interfaces. They also report negative IOHE signals for Ag and Au, corroborating predictions of negative orbital Hall conductivities that earlier tight-binding models missed.","pith_inferences":["If the $\\beta = 0.23$ calibration is not transferable across the 19 metals, the relative ranking of IOHE signals could change; a direct test would be to repeat the measurement with several Pt thicknesses for a few X metals and check that the extracted IOHE is stable.","The dominance of orbital conversion suggests that optimizing orbital texture and Fermi-level position may matter more than maximizing atomic spin-orbit coupling for charge-current generation, which would shift materials-design priorities toward light and abundant elements.","A natural extension is to probe the same metals by terahertz emission or spin-torque ferromagnetic resonance to see whether orbital dominance persists when the detection mechanism is different."],"forward_implications":["The measured charge current in YIG/Pt(2)/X stacks cannot be used to infer spin Hall angles without subtracting the IOHE contribution.","Metals with weak spin-orbit coupling, such as Mo, Zr, and Nb, become viable orbital-to-charge converters for orbitronic devices.","Negative orbital Hall conductivities in Ag and Au are experimentally observed, not artifacts of a particular theoretical model.","Spin pumping with ferromagnetic resonance is a practical tool for screening orbital Hall materials across broad families of elements."],"supporting_citations":[{"why":"Provides the spin-pumping theory that the experiment relies on for generating spin currents from ferromagnetic resonance.","marker":"[31]"},{"why":"Supplies Eq. (1), the model for ISHE-induced charge current density used to analyze the YIG/X(5) reference series.","marker":"[32]"},{"why":"First-principles Wannier-interpolation calculations of spin and orbital Hall conductivities that the measured ISHE and IOHE values are compared against.","marker":"[30]"},{"why":"Alternative first-principles calculation used to explain the positive IOHE in Pd and to contrast with tight-binding model predictions.","marker":"[28]"},{"why":"Prior spin-orbital pumping study that established the methodology and reports negligible ISHE and IOHE in Cu, used as a control.","marker":"[11]"},{"why":"Observation of the orbital Hall effect in light metal Ti, supporting the claim that weak spin-orbit-coupling metals can have strong orbital response.","marker":"[13]"}],"fun_headline_variants":["Orbital Hall effect dominates spin in 19 transition metals","Orbitronics: orbital Hall effect beats spin in metals","Orbital Hall effect outshines spin in transition metals","Inverse orbital Hall effect dominates spin in 19 metals"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The whole extraction rests on assuming that the YIG/Pt(2)/X(5) signal is exactly the Pt(2) ISHE signal plus the X-layer IOHE signal plus 23% of the X-layer ISHE signal, with that 23% transmission fraction remaining the same for every one of the 19 metals.","fun_headline_variants_meta":{"raw":{"variants":["Orbital Hall effect dominates spin in 19 transition metals","Orbitronics: orbital Hall effect beats spin in metals","Orbital Hall effect outshines spin in transition metals","Inverse orbital Hall effect dominates spin in 19 metals"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000639,"raw_usage":{"total_tokens":2883,"prompt_tokens":826,"completion_tokens":2057,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":442,"completion_tokens_details":{"reasoning_tokens":1997}},"tokens_in":442,"tokens_out":2057,"duration_ms":17806,"temperature":1.0,"reasoning_tokens":1997,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T05:11:40.053029+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A decisive check would be to fabricate YIG/Pt($t$)/X(5) for a range of Pt thicknesses $t$ for several X metals and test whether the extracted IOHE value is independent of $t$. If the extracted IOHE changes with $t$, the $\\beta = 0.23$ transmission model is wrong; a complementary check would be to insert a thin Cu layer at the Pt/X interface and see whether the orbital signal disappears while the spin signal remains.","supporting_citations":[{"cited_title":"Tserkovnyak, A","cited_arxiv_id":null,"evidence_quote":"Provides the spin-pumping theory that the experiment relies on for generating spin currents from ferromagnetic resonance."},{"cited_title":"Arana, M","cited_arxiv_id":null,"evidence_quote":"Supplies Eq. (1), the model for ISHE-induced charge current density used to analyze the YIG/X(5) reference series."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"First-principles Wannier-interpolation calculations of spin and orbital Hall conductivities that the measured ISHE and IOHE values are compared against."},{"cited_title":"Salemi, and P","cited_arxiv_id":null,"evidence_quote":"Alternative first-principles calculation used to explain the positive IOHE in Pd and to contrast with tight-binding model predictions."},{"cited_title":"Santos, J","cited_arxiv_id":null,"evidence_quote":"Prior spin-orbital pumping study that established the methodology and reports negligible ISHE and IOHE in Cu, used as a control."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Observation of the orbital Hall effect in light metal Ti, supporting the claim that weak spin-orbit-coupling metals can have strong orbital response."}],"review_version":1}