{"id":"12650f2d-f338-46f2-8081-8420a0850053","arxiv_id":"2505.09257","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":0.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A review of electron- and magnon-mediated torques, covering spin-orbit, orbital, and magnon torque mechanisms and materials.","lead":"This paper is a review of recent progress in spin-orbit torque, orbital torque, and magnon torque for magnetic memory devices. It summarizes the materials and mechanisms in each class and highlights open controversies, making it a useful entry point to a fast-moving field.","discovery_kind":"review","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The orbital-torque section presents the Ta/Ni sign reversal as direct OT evidence, yet Section 5(4) of the same paper says it may be a self-induced ST-FMR artifact; this internal tension undercuts the systematic-accuracy claim.","rationale":"The reader's weakest assumption — that the review faithfully reports and interprets the cited primary literature — maps directly onto the Ta/Ni orbital-torque example, and I agree that this is the central risk for a review with no new measurements. My concern is sharper than a general 'fidelity to the literature' caveat: the manuscript itself contains the counter-evidence in Section 5(4), yet Section 3.2 presents the challenged result as direct proof without qualification. This is an internal inconsistency that a reader of Section 3.2 alone could not detect, and it affects a major pillar of the review's coverage. The RuO2 ground-state controversy is also real but is a single material subsection, whereas the orbital-torque issue spans the entire third section and its quantitative conclusions. The review deserves credit for including Section 5(4), which is why this is not a REJECT; the gap is editorial, not fundamental. The required change is to add inline caveats or explicit cross-references whenever Section 3.2 relies on results that Section 5(4) identifies as contested, and to soften claims of 'direct evidence' accordingly. With that revision, the review would be acceptably accurate and balanced. Hence CONDITIONAL rather than UNCHANGED or REJECT.","tokens_in":32843,"tokens_out":3955,"duration_ms":41381,"concrete_test":"Re-analyze the Ta/Ni ST-FMR data from Ref. [79] (Fig. 8(a)) using the self-induced ST-FMR correction derived in Liu et al. (Ref. [261], arXiv:2501.10260), and check whether the extracted anti-damping torque sign flips from positive to negative. If the sign reverses, the Section 3.2 claim that Ta/Ni provides 'direct evidence of OT' is falsified, and the orbital-torque section must be revised to mark this result as disputed, with a cross-reference to Section 5(4), before the review can be accepted as a reliable synthesis.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"For the review's central claim — an accurate, systematic synthesis of electron- and magnon-mediated torques — to hold, the body text must not present contested experimental interpretations as established fact. The most load-bearing weakness is in Section 3.2. There, the Ta/Ni sign reversal of the anti-damping torque (Ref. [79]) is cited as 'direct evidence of OT' and is used to anchor the orbital-Hall-effect picture, alongside quantitative claims such as ξOT ≈ 0.78 for Zr/[Co/Pt]3, λOH ≈ 60 nm for Ti/Ni, and λOH ≈ 47 nm for Ti/Ni from ST-FMR. However, Section 5(4) of the same manuscript acknowledges that Liu et al. attribute the positive Ta/Ni torque to a self-induced ST-FMR artifact, and that first-principles scattering calculations (Rang et al.) find orbital angular momentum relaxing within 1–2 atomic layers, while Song et al. find OHE contributions overshadowed or canceled in Ni. These caveats are not carried into Section 3.2, so the body presents as established several results the authors themselves later mark as disputed. If Liu et al. and Rang et al. are correct, the 'direct evidence' anchor of the orbital-torque narrative is invalid, and some reported efficiencies and relaxation lengths would need re-evaluation. The explicit acknowledgement in Section 5 is a positive feature, but it is isolated from the section where the claims are used, creating an internal inconsistency that weakens the 'systematic and accurate' claim from the abstract.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This manuscript is a review of three torque mechanisms for magnetization manipulation: conventional spin-orbit torque (SOT), orbital torque (OT), and magnon torque (MT). It surveys the relevant material families, including heavy metals, topological insulators, low-crystal-symmetry materials, non-collinear antiferromagnets, altermagnets, orbital Hall and orbital Rashba-Edelstein materials, and NiO- and BiFeO3-based magnon devices. The review emphasizes generation mechanisms, switching demonstrations, and open questions, and it closes with a list of six remaining challenges, including unresolved disputes over the orbital-torque interpretation and the magnetic ground state of RuO2.","tokens_in":33093,"tokens_out":5258,"duration_ms":49666,"significance":"If accurate, this review provides a useful, current synthesis across three active subfields, assembling the key material classes and referencing the literature through 2024–2025. Its strengths include a balanced discussion of several controversies in the concluding section, a generally accurate reproduction of standard formulas, and explicit acknowledgment of conflicting reports on the Ta/Ni orbital torque and on the RuO2 magnetic ground state. The value is primarily bibliographic and pedagogical for a general condensed-matter and materials-science readership, rather than a novel technical contribution. The review's central claim, that it systematically and accurately covers electron- and magnon-mediated torques, is defensible but requires that contested experimental interpretations presented in the body be reconciled with the caveats stated in the conclusion.","major_comments":[{"comment":"Sections 3.1 and 3.2 present the Ta/Ni sign reversal and the long orbital diffusion length as established facts: the text near Fig. 8(a) states that the sign reversal is \"direct evidence of OT,\" and Section 3.1 states that the orbital diffusion length is much longer than the spin diffusion length and that this has been used to separate OT from SOT. However, Section 5, item (4) of the same manuscript reports that Liu et al. attribute the positive Ta/Ni torque to a self-induced ST-FMR artifact, that Rang et al. find orbital angular momentum relaxing within 1–2 atomic layers in transition-metal/FM bilayers, and that Song et al. find OHE contributions often overshadowed or canceled in ferromagnets such as Ni. These positions directly contradict the body-text claims. The review should qualify the body statements as one interpretation among several, or at minimum add an explicit cross-reference to Section 5(4) at the point where the Ta/Ni result is used as evidence. As written, the unresolved controversy is acknowledged in the conclusion but isolated from the section where the claim is load-bearing, which weakens the \"systematic and accurate\" synthesis asserted in the abstract.","section":"§3.1–3.2 vs. §5(4)"},{"comment":"Section 2.5 describes RuO2 as a d-wave altermagnet and presents the spin-split effect as the mechanism for generating spin currents, citing experimental field-free switching results without qualification. Section 5, item (2), however, states that the exact magnetic ground state of RuO2 remains controversial, citing multiple recent studies that question the altermagnetic order. The reader encountering Section 2.5 is not told that the material's altermagnetic ground state is disputed. The caveat should be integrated into Section 2.5 where RuO2 is first discussed, not deferred to the final section, so that the body does not present a contested experimental interpretation as established fact.","section":"§2.5 vs. §5(2)"}],"minor_comments":[{"comment":"The text refers to \"Weishengzhao's group\"; this should be \"Weisheng Zhao's group.\"","section":"§4.2"},{"comment":"Several equations and symbols appear with garbled subscripts and superscripts (for example, the expression for the total torque efficiency and the definitions of θ_OH^eff, σ_OHE, and ξ_DL). These are likely font-conversion artifacts and need to be typeset correctly in the final version.","section":"§2.1, §3.1, §3.2"},{"comment":"The units for critical current density are inconsistent: the text uses \"A·m^-2\" in places (e.g., for the Bi2Se3/NiO/Py device) where \"A·cm^-2\" is presumably intended. Please standardize the units throughout.","section":"§4.2"},{"comment":"Reference [148] is cited with a DOI instead of journal volume and page information; please update it to the published record if available.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is a suitable review for Chinese Physics B in scope and level. The main concern is the internal tension between the body text and the concluding caveats on orbital torque and RuO2; once those cross-references and qualifications are added, the review would be acceptable for publication. No concerns about citation norms or novelty disclosure."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick read of arXiv:2505.09257. It's a review of electron- and magnon-mediated torques, and I think it's a genuinely useful synthesis of the 2024–2025 literature. There are no new results, which is fine for a review. Coverage is broad and organized: conventional SOT materials (heavy metals, TIs, low-symmetry materials, non-collinear AFMs, altermagnets), orbital torque from transition metals, and magnon torque via NiO and BiFeO3. Figures are reproduced with permissions, formulas are standard, and the conclusion explicitly lists open controversies. That last part is a real strength.\n\nThe soft spot is internal and worth taking seriously. Section 3.2 presents the Ta/Ni sign reversal as 'direct evidence of OT' and uses it as an anchor for the orbital-Hall picture, along with numbers like ξOT ≈ 0.78 in Zr/[Co/Pt]3 and λOH ≈ 60 nm in Ti/Ni. But Section 5(4) of the same manuscript says Liu et al. propose the positive Ta/Ni torque is a self-induced ST-FMR artifact, and first-principles work by Rang et al. and Song et al. predicts short orbital relaxation and cancellation in Ni. Those caveats are not carried into Section 3.2. So the body treats as established exactly the results the authors later mark as disputed. That's a genuine internal inconsistency, and it undercuts the 'systematic and accurate' claim in the abstract. It doesn't sink the whole review, but the orbital torque section needs restructuring or explicit hedging before I'd trust it as a reference. The RuO2 issue is milder: Section 2.5 describes the spin-split effect generation, and Section 5(2) says the magnetic ground state is controversial, but the body doesn't mention the debate.\n\nCitation pattern is reasonable. Self-citations are to original experimental papers, and the review doesn't depend on them. No invented entities.\n\nWho this is for: newcomers or device-oriented readers who want a map of recent torque physics. It deserves serious peer review, because the scope is useful and the authors know the field. The referee should demand consistency between the OT materials section and the controversies in Section 5(4). I would not desk reject.","headline":"Useful, current review of electron- and magnon-mediated torques, but the orbital-torque section presents as established the very results the authors later flag as disputed.","tokens_in":33696,"tokens_out":2501,"would_cite":false,"duration_ms":24797,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["75.70.Tj","75.60.Jk","75.76.+j","72.25.-b"],"model":"deepseek-v4-flash","headline":"A review maps the three torque mechanisms—spin-orbit, orbital, and magnon—to the materials that enable them, and flags which reported efficiencies remain contested.","keywords":["spin-orbit torque","orbital torque","magnon torque","altermagnet","field-free switching","spin Hall effect","orbital Hall effect","magnetic random-access memory"],"falsifier":"A re-measurement of Ta/Ni and Zr/Ni bilayers with the self-induced spin-torque ferromagnetic resonance contribution subtracted, or with harmonic-Hall detection that does not suffer that artifact, would settle whether the sign reversal in torque efficiency is truly orbital in origin. Independently, a decisive measurement of RuO2's magnetic ground state—for example by muon spin rotation or neutron diffraction on well-characterized films with controlled stoichiometry—would settle whether its reported spin-split-effect torques are genuine.","tokens_in":32605,"feed_emoji":"🧲","tokens_out":8179,"duration_ms":73638,"temperature":0.7,"pith_summary":"This review sets out to organize current-induced magnetization switching into three routes distinguished by the carrier of angular momentum: electron spins (conventional spin-orbit torque), electron orbital motion (orbital torque), and magnons in insulating magnets (magnon torque). For each route it ties the generation mechanism—spin Hall and Rashba-Edelstein effects for spins, orbital Hall and orbital Rashba-Edelstein effects for orbitals, interfacial spin-to-magnon conversion for magnons—to the material families that have demonstrated switching. The practical stake is SOT-MRAM as a high-speed, low-power memory technology, where the key bottleneck is field-free switching of perpendicular magnetization. The review's own conclusion, however, is that the field is not settled: it lists contested claims, notably some positive orbital-torque efficiencies and the altermagnetic ground state of RuO2, as open problems rather than solved ones.","feed_headline":"Three torque mechanisms, one roadmap for magnetic memory","feed_subtitle":"The paper ties each switching mechanism to its materials and flags which efficiency claims are still disputed","key_machinery":"The organizing quantity is the torque efficiency ξ, defined separately for each mechanism: for conventional SOT the spin Hall angle θ_SH and spin Hall conductivity σ_SH; for orbital torque the effective orbital Hall angle θ_OH^eff = θ_OH · η_L-S, where θ_OH is the orbital Hall angle of the source metal and η_L-S the orbital-to-spin conversion coefficient of the ferromagnet or an inserted conversion layer; and for magnon torque the spin torque ratio θ_i = J_i / J_c transmitted through the antiferromagnetic spacer. These ratios serve as a common comparison frame: sign reversals of ξ with layer thickness or ferromagnet identity are used as fingerprints that distinguish orbital-dominated from spin-dominated torque, and thickness dependence of θ_i is used to separate electron-mediated from magnon-mediated contributions.","core_discovery":"The paper's central claim, on its own terms, is that electron- and magnon-mediated torques now form a coherent map of mechanisms and materials: heavy metals and topological insulators supply conventional spin-orbit torque; low-crystal-symmetry materials, non-collinear antiferromagnets, and altermagnets supply z-polarized spin currents that enable field-free switching of perpendicular magnetization; light 3d, 4d, and 5d metals can act as orbital-current sources through the orbital Hall and orbital Rashba-Edelstein effects; and NiO- or BiFeO3-based sandwiches transmit torque as magnon currents through insulating antiferromagnets. The review argues that each branch has reached device-relevant demonstrations but that the underlying physics is still being settled, and it explicitly flags cases where the reported picture may be wrong: the positive orbital-torque efficiencies of Ta/Ni, Zr/Ni, and Cr-based stacks, and the altermagnetic classification of RuO2.","pith_inferences":["If the self-induced ST-FMR artifact criticism is correct, the strongest experimental signature of orbital torque (the Ta/Ni sign reversal) would lose its evidential weight, and the orbital-torque field would need a new anchor experiment in a system with negligible background torques.","The short orbital relaxation lengths predicted by first-principles scattering calculations (1–2 atomic layers) suggest that many 'orbital current' observations could instead be interfacial effects; a layer-resolved measurement with a nonmagnetic spacer of varying thickness inside the ferromagnet could test this directly.","The review's emphasis on 2D antiferromagnetic insulators as future magnon spacers suggests a concrete comparison: if interface smoothness rather than bulk magnon lifetime dominates transmission, then WTe2/NiO/CoFeB should outperform polycrystalline-NiO stacks with the same nominal thickness.","For altermagnets, the RuO2 controversy implies the field is one decisive experiment away from either consolidating or reclassifying the spin-split effect; strain- and stoichiometry-controlled sample series would be the natural test."],"forward_implications":["If the review's synthesis holds, field-free switching of perpendicular magnetization no longer requires external assist fields or complex symmetry-breaking geometries; it can be engineered into the spin source itself, via low-crystal-symmetry, non-collinear antiferromagnetic, or altermagnetic materials that emit z-polarized spin currents.","Orbital torque would make cheap and abundant 3d and 4d metals (Ti, Cr, Zr) viable torque sources, but the paper's own Section 5 shows this branch is conditional: if the Ta/Ni and Zr/Ni interpretations are artifacts, the claimed advantages of orbital currents would need revision.","Magnon torque through antiferromagnetic insulators would allow switching of a ferromagnet without passing charge current through it, and the BiFeO3 results add ferroelectric polarization as a voltage knob for torque efficiency.","The review implies that the immediate next step is methodological: a unified means of separating orbital from spin torque in the same device is needed before reported efficiencies can be compared.","Material integration constraints—thermal stability above 350 °C and compatibility with CMOS and MgO-MTJs—are the gating obstacle for industrial adoption, not the existence of a working torque mechanism."],"supporting_citations":[{"why":"Demonstrates current-induced perpendicular magnetization switching in Pt/Co, the foundational SOT experiment the review builds on.","marker":"[13]"},{"why":"Establishes the spin Hall effect as the origin of SOT in Ta, the baseline for conventional SOT materials.","marker":"[14]"},{"why":"Shows topological-surface-state spin-momentum locking can drive SOT, anchoring the topological insulator section.","marker":"[16]"},{"why":"First observation of out-of-plane anti-damping torque in WTe2/Py, the anchor for low-crystal-symmetry z-spin sources.","marker":"[49]"},{"why":"Demonstrates the magnetic spin Hall effect in Mn3Sn with time-reversal-odd spin accumulation, the anchor for non-collinear antiferromagnets.","marker":"[62]"},{"why":"Shows z-polarized spin current from the altermagnet RuO2, the key experimental support for the altermagnet section.","marker":"[74]"},{"why":"Reports the Ta/Ni sign anomaly attributed to orbital torque, the central evidence for the orbital torque mechanism.","marker":"[79]"},{"why":"Demonstrates orbital torque in Ti/Ni with a long orbital diffusion length, anchoring the 3d-metal orbital Hall effect.","marker":"[26]"},{"why":"Demonstrates magnon-torque switching in Bi2Se3/NiO/Py, the foundational experiment for NiO-based magnon torque.","marker":"[24]"},{"why":"Shows ferroelectric-polarization control of magnon torque in BiFeO3 heterostructures, anchoring the multiferroic magnon section.","marker":"[28]"}],"fun_headline_variants":["Torque trio: electrons, orbitals, magnons for memory","Magnetic memory: one review, three torque paths","Spin-orbit torque: materials map with disputed numbers","From NiO to RuO2: torques that could rewire memory","The torque review: what's settled, what's not in SOT"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The review contains no new measurements; its value rests entirely on the accuracy of the published results it summarizes, most critically the disputed positive orbital-torque efficiencies in Ta/Ni, Zr/Ni, and Cr-based stacks and the altermagnetic identification of RuO2.","fun_headline_variants_meta":{"raw":{"variants":["Torque trio: electrons, orbitals, magnons for memory","Magnetic memory: one review, three torque paths","Spin-orbit torque: materials map with disputed numbers","From NiO to RuO2: torques that could rewire memory","The torque review: what's settled, what's not in SOT"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000995,"raw_usage":{"total_tokens":4215,"prompt_tokens":944,"completion_tokens":3271,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":560,"completion_tokens_details":{"reasoning_tokens":3184}},"tokens_in":560,"tokens_out":3271,"duration_ms":24606,"temperature":1.0,"reasoning_tokens":3184,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T21:35:25.549142+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A re-measurement of Ta/Ni and Zr/Ni bilayers with the self-induced spin-torque ferromagnetic resonance contribution subtracted, or with harmonic-Hall detection that does not suffer that artifact, would settle whether the sign reversal in torque efficiency is truly orbital in origin. Independently, a decisive measurement of RuO2's magnetic ground state—for example by muon spin rotation or neutron diffraction on well-characterized films with controlled stoichiometry—would settle whether its reported spin-split-effect torques are genuine.","supporting_citations":[],"review_version":1}