{"id":"4a29e3e7-1fe3-4771-8b44-5feb59eec21d","arxiv_id":"2501.15815","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Atomistic LLGS simulations of all in-plane grain rotations show configuration II switches Mn3Sn to a single +z octupole state, while configuration I sub-configurations switch in opposite directions and cancel.","lead":"A simulation study of polycrystalline Mn3Sn finds that only one crystallographic orientation, with the Kagome plane perpendicular to the spin polarization, reliably switches its octupole moment in one polarity under spin-orbit torque, while the other orientation's contributions cancel. This explains the robust switching seen in experiments and points to grain orientation as a design handle for Mn3Sn memory devices.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Configuration I cancellation requires an even distribution of sub-configurations and full coverage of grain orientations; neither is demonstrated, so the claim that configuration II dominates the observed AHE is conditional.","rationale":"The reader identified the even-distribution and independence assumptions as the weakest link. I agree. The paper's own text makes the cancellation conditional on even distribution, and no evidence is supplied. The additional concern about incomplete coverage of grain orientations strengthens the reader's point: even an even distribution of the simulated in-plane rotations would not represent a random polycrystalline film, because tilts are excluded. The atomistic LLGS framework and the 360-point sweep are credible, and the paper correctly reproduces its prior results, giving some independent support. However, the step from single-grain dynamics to polycrystalline AHE is the load-bearing transition, and it depends on unverified assumptions about the orientation distribution and grain independence. A direct multi-grain simulation or texture measurement would settle whether the cancellation occurs.","tokens_in":8893,"tokens_out":7800,"duration_ms":74677,"concrete_test":"Perform a multi-grain LLGS simulation of a polycrystalline Mn3Sn film with ~10^4 grains, sampling grain orientations uniformly from SO(3) (including tilts between the Kagome plane and σ/Hext), include inter-grain exchange coupling at boundaries, apply the experimental Jc and Hext, and compute the net AHE. If the net signal from configuration-I-like grains is nonzero or tilted grains switch with a preferred polarity, the paper's conclusion that configuration II dominates fails.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central conclusion is that configuration II supplies the measured AHE because configuration I sub-configurations switch with opposite polarities that cancel. The authors explicitly condition this cancellation on even distribution of sub-configurations (Results, Fig. 5(b) discussion). No experimental texture data or statistical justification is provided for this distribution. More fundamentally, the 360 sub-configurations are generated only by in-plane rotations of the two ideal configurations I and II; real grains have arbitrary tilts of the Kagome plane relative to σ and Hext, so the simulated set does not cover all possible orientations as claimed. If the orientation distribution is textured, or if tilted grains switch with a net polarity, the configuration I contribution need not vanish and could offset configuration II. Since the conclusion that configuration II is the primary source of the measured AHE rests on this, the claim is not fully secured.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies SOT-driven switching in polycrystalline Mn3Sn by constructing 360 sub-configurations for each of two ideal device configurations, obtained by rotating the crystal in-plane in 1° increments. Atomistic LLGS simulations show that all configuration II sub-configurations switch the octupole moment to a narrow +z easy-axis range (φFinal ∈ [−14°, 46°]), giving a stable switching polarity, while configuration I sub-configurations switch in opposite directions and are claimed to cancel when evenly distributed. The authors propose a 'dynamic balance model' based on the six-fold anisotropy of Mn3Sn to explain the six-period behavior, and compare their results with experiments to conclude that configuration II contributes the dominant measured AHE signal.","tokens_in":9098,"tokens_out":8077,"duration_ms":70298,"significance":"If the central claim holds, the paper offers a plausible resolution of a longstanding puzzle: why polycrystalline Mn3Sn shows robust SOT switching even though the earlier three-configuration picture predicted switching mainly in configuration I. The systematic 360-orientation scan is a methodological strength, and the identification of a stable polarity interval for configuration II is a concrete, falsifiable prediction. The work also highlights the importance of grain-orientation distributions, a point often neglected in single-crystal-based models. However, the significance is tempered by two load-bearing gaps: the experimental comparison rests on a miscitation, and the cancellation in configuration I depends on an even grain-orientation distribution for which no evidence is provided. The dynamic balance model is descriptive rather than quantitative and partly presupposes the six-fold symmetry it is invoked to explain.","major_comments":[{"comment":"The experimental comparison is based on incorrect citations. The Introduction states that experimental studies have demonstrated robust SOT-induced switching in polycrystalline samples and cites Refs. [7,9,17]; however, Ref. [7] (Lonsky & Hoffmann) is about skyrmion breathing modes, Ref. [9] (Roschewsky et al.) is about GdFeCo, and Ref. [17] (J. Lu et al.) is about IrMn-based perpendicular magnetic tunnel junctions. The final paragraph of Results and Discussion further claims 'Ref. [17] studied SOT switching of Mn3Sn by rotating the Hall bar device,' which is not what Ref. [17] reports. This miscitation undermines the claim that the simulations explain previous experimental results. Please identify the correct experimental paper(s) (likely Refs. [22], [37], [41], or [45]) and accurately describe their findings, or revise the comparative claim.","section":"Introduction; Results and Discussion (final paragraph)"},{"comment":"The central conclusion that configuration I contributes negligibly to the measured AHE relies on the assumption that the sub-configurations are 'evenly distributed' in the polycrystalline film. This condition is stated only parenthetically in the discussion of Fig. 5(b), and no experimental texture data, grain-size distribution, or statistical argument is provided to justify it. Moreover, the text in Fig. 2(a) asserts that the 360 sub-configurations 'cover all the possible orientations' in the polycrystalline sample, but the construction only performs in-plane rotations of two ideal configurations; real grains have arbitrary tilts of the Kagome plane relative to σ and Hext, which are not sampled. If the orientation distribution is textured, or if tilted grains switch with a net polarity, the configuration I contribution need not vanish and could offset the configuration II signal. Please provide evidence for the assumed distribution, or weaken the conclusion accordingly.","section":"Results and Discussion, Fig. 5(b) and Fig. 2(a)"},{"comment":"The dynamic balance model is presented as explaining the six-period behavior and the stable switching polarity, but it is a descriptive vector sketch rather than a quantitative derivation. It postulates a six-fold easy-axis anisotropy with evenly distributed easy axes, which is the same symmetry input that already resides in the LLGS Hamiltonian (via the anisotropy term). The agreement between the model and the simulations is therefore expected by construction and does not independently validate the switching mechanism. To support the claim that the model 'perfectly explains' the variation of φDiff, the authors should provide an explicit energy or effective-field formulation that predicts φMid(φsub_conf), φFinal(φsub_conf), and the stability interval quantitatively.","section":"Results and Discussion, Fig. 3(b)-(d)"},{"comment":"The abstract and conclusion state that 'the signals from various sub-configurations in configuration I cancel each other out' as a definitive result, whereas the Results section shows the cancellation holds only 'when these sub-configurations are evenly distributed.' This overstatement is load-bearing because the paper's main conclusion about the origin of the measured AHE rests on the cancellation. Please ensure the abstract and conclusion carry the same qualification as the results, or provide evidence that the distribution is indeed even.","section":"Abstract; Conclusion"}],"minor_comments":[{"comment":"The symbol φ is used without definition at its first occurrence; define it as the in-plane rotation angle of the crystal lattice relative to a reference direction.","section":"Introduction"},{"comment":"The Hamiltonian contains a term −Σ(K_i·m_i)^2, which is the form of a uniaxial anisotropy, but the text later describes a six-fold easy-axis anisotropy in the Kagome plane; please clarify how the six-fold symmetry is encoded in the Hamiltonian or in the effective anisotropy field.","section":"Methodology"},{"comment":"The phrase 'Noted that' appears twice; it should be 'Note that.'","section":"Results and Discussion, Fig. 2(a) and Fig. 4(a)"},{"comment":"The switching current for configuration I is Jc = 2.2×10^14 A/m², which is four orders of magnitude larger than the Jc = 5×10^10 A/m² used for configuration II; a brief comment on the practical relevance of this value for realistic devices would be helpful.","section":"Results and Discussion, Fig. 4(b)"},{"comment":"There are occasional grammar inconsistencies, e.g., 'the switching results in configuration II is insensitive' should be 'are insensitive.'","section":"Results and Discussion"}],"recommendation":"major_revision","confidential_remarks":"The miscitation of experimental references is a serious accuracy issue that should be corrected before publication. The even-distribution assumption is the key physics assumption but is presented almost as an aside; the authors should either provide supporting evidence (e.g., texture measurements) or substantially soften the central claim. The paper is within the scope of cond-mat.mes-hall and the LLGS methodology is sound, but the overstatement in the abstract and conclusion, combined with the mis-cited experimental basis, currently makes the central claim insufficiently secured."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Honestly, the new thing here is the 360° sweep. Prior work, including their own [43], used three configurations; this paper maps every in-plane rotation and shows that configuration II always switches to the same easy-axis window, while configuration I splits into opposing directions. That's a real step and the polarity diagrams in Fig. 5 make it clear.\n\nThe simulations are standard LLGS with stated parameters, and they check that their previous result is a subset. The dynamic balance model is qualitative, but it accounts for why φMid and φFinal both rotate with the sample, which the old explanation couldn't.\n\nThe soft spots are in the experimental comparison and the step to polycrystalline reality. First, the text says Ref. [17] studied SOT switching by rotating a Hall bar, but [17] is an IrMn MTJ paper. That's a direct miscitation; it's the only experimental support for 'switching at all rotation angles' and it doesn't hold up. Second, the paper claims the 360 sub-configurations cover all possible orientations. They cover in-plane rotations only. Real grains have arbitrary tilts of the Kagome plane relative to σ and Hext, so the coverage claim is overbroad. Third, the cancellation in configuration I rests on an even distribution of sub-configurations. The authors state that as a condition, but give no texture data or robustness check. If the film is textured, the cancellation can fail and the conclusion that configuration II is the measured AHE source weakens. Fourth, the dynamic balance model is a vector sketch, not a derivation; it assumes the six-fold anisotropy it is trying to explain.\n\nNone of this kills the core simulation result. The stable polarity in configuration II is plausible and well-presented. But the paper's central claim about experiments is conditional, and the miscitation should not survive review.\n\nThis paper is for the antiferromagnetic spintronics people, and they'll want to see these diagrams. I'd send it to peer review, but with a referee who checks the citation and demands either texture data or a robustness analysis for the distribution. If those are addressed, the result could be citable; as it stands, treat the experimental conclusion with caution.","headline":"A useful simulation sweep of Mn3Sn sub-configurations that needs a citation fix and a texture caveat before its central claim is trustable.","tokens_in":9553,"tokens_out":2802,"would_cite":true,"duration_ms":27183,"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":"Spin-orbit torque switching in polycrystalline Mn3Sn gets its stable polarity from configuration II; configuration I cancels.","keywords":["Mn3Sn","spin-orbit torque","noncollinear antiferromagnet","anomalous Hall effect","polycrystalline","octupole moment","Kagome lattice","LLGS simulation"],"falsifier":"Measure the anomalous Hall switching polarity of a Mn3Sn film with a deliberately biased grain-orientation texture, or simulate a pair of coupled grains with differing sub-configurations: the model predicts configuration I contributes nothing only for an even orientation distribution, so a skewed texture or inter-grain coupling that produces a net configuration I signal would falsify the central claim.","tokens_in":8709,"feed_emoji":"🧲","tokens_out":9845,"duration_ms":85556,"temperature":0.7,"pith_summary":"This paper argues that the robust spin-orbit-torque (SOT) switching seen in polycrystalline Mn3Sn comes almost entirely from one class of grain orientation: configuration II, where the Kagome plane (the plane of corner-sharing Mn triangles) is perpendicular to the spin polarization. By rotating the lattice in 1-degree increments to create 360 sub-configurations and simulating each with the Landau-Lifshitz-Gilbert-Slonczewski equations, the authors find that every sub-configuration of configuration II sends the magnetic octupole moment to the same +z easy-axis range $\\varphi \\in [-14^\\circ, 46^\\circ]$, so its switching polarity is stable. For configuration I, the Kagome plane parallel to the spin polarization, switching occurs only in four intervals and with opposite polarities, so under an even grain distribution those contributions cancel and add little to the measured anomalous Hall signal. If this is right, experiments on polycrystalline Mn3Sn should be interpreted through configuration II, and device design should favor grains with the Kagome plane perpendicular to the spin polarization.","feed_headline":"Configuration II alone yields stable Mn3Sn switching polarity","feed_subtitle":"All 360 grain orientations in that alignment switch to the same +z range; the other cancels.","key_machinery":"The load-bearing object is the sub-configuration scan: 360 atomic environments generated by rotating the Kagome plane in 1-degree increments, meant to cover all possible grain orientations in a polycrystalline film. The simulation method is atomistic LLGS dynamics for the three Mn sublattices, with exchange, Dzyaloshinskii-Moriya, anisotropy, Zeeman, and damping-like spin-orbit torque terms. The explanatory mechanism is the dynamic balance model: the stable octupole state is the vector balance of the SOT effective field $\\mathbf{H}_{DL} = \\boldsymbol{\\sigma} \\times \\mathbf{m}_{oct}$, the fixed external field $\\mathbf{H}_{ext}$, and the six-fold-anisotropy field $\\mathbf{H}_{an} = (\\mathbf{m}_{oct}\\cdot \\mathbf{e})\\mathbf{e} - \\mathbf{m}_{oct}$. The model explains both the 60-degree periodicity of the final state and why the final octupole orientation is confined to $[-14^\\circ, 46^\\circ]$: outside the comfortable region the required anisotropy field becomes too large, so the octupole resets to a nearby easy axis.","core_discovery":"The central claim is that in a polycrystalline Mn3Sn/Pt bilayer the measured SOT-driven anomalous Hall switching is carried by configuration II, not configuration I. The evidence is a 1-degree rotation scan of 360 atomic environments: for configuration II every sub-configuration, under $J_c = 5\\times10^{10}$ A/m² and $\\mathbf{H}_{ext} = 100$ Oe along +y, reaches a final octupole orientation in the range $-14^\\circ$ to $46^\\circ$, all with $m_{z,\\mathrm{oct}} > 0$, and reversing the current or field sends it to the opposite z sign. Configuration I, scanned the same way at its own switching condition, switches only in four angular intervals, with two switching from +z to -z and two from -z to +z, so the net contribution vanishes if those sub-configurations are evenly distributed. The paper proposes a dynamic balance model in which the equilibrium is set by the damping-like SOT field $\\mathbf{H}_{DL}$, the fixed applied field $\\mathbf{H}_{ext}$, and the six-fold anisotropy field $\\mathbf{H}_{an}$; because one of the six easy axes always lies in a comfortable region where the required $\\mathbf{H}_{an}$ is small, the octupole always relaxes into the same polarity range, giving configuration II a stable switching polarity.","pith_inferences":["The paper's cancellation prediction for configuration I is conditional on an even distribution of sub-configurations; if real films have texture bias, configuration I could contribute a net anomalous Hall signal, a testable consequence the paper does not develop.","The simulations treat grains as independent; including grain-boundary exchange or inter-grain coupling could alter the cancellation and the stability range, especially for small grains.","The dynamic balance model should transfer to other six-fold-symmetric noncollinear antiferromagnets such as Mn3Ge or Mn3Pt, predicting similar robust switching intervals for the analogous configuration II orientations.","A direct device-level prediction is that a Mn3Sn memory cell should tolerate plus or minus 30 degrees of in-plane crystal misalignment in configuration II without losing deterministic polarity; this could be probed with patterned single-crystal islands of controlled orientation."],"forward_implications":["In polycrystalline Mn3Sn devices, the anomalous Hall signal from SOT switching should be dominated by configuration II grains, so optimizing the texture toward the Kagome-plane-perpendicular orientation should strengthen the switching signal.","Configuration I grains can be treated as a near-zero background in the measured anomalous Hall effect, which explains why many experiments see robust switching despite having many grain orientations.","Because every configuration II sub-configuration ends in the same +z range and reverses with current or field, the switching polarity is stable across essentially arbitrary in-plane crystal rotation, a useful robustness property for memory cells.","The 60-degree periodicity and the restricted final-state range are direct predictions of the six-fold anisotropy balance, so measurements of the final octupole orientation as a function of crystal rotation can test the model."],"supporting_citations":[{"why":"The authors' prior atomistic study showing deterministic switching in representative configuration II states, which the full 360-sub-configuration scan extends.","marker":"[43]"},{"why":"The experimental demonstration of SOT switching in polycrystalline Mn3Sn that motivates the configuration classification and supplies damping parameters.","marker":"[37]"},{"why":"The experiment reporting perpendicular full switching of Mn3Sn, used for simulation parameters and as evidence for configuration II switching.","marker":"[41]"},{"why":"The report of the handedness anomaly under SOT, which the earlier two-field balance incorporated and the new three-field model extends.","marker":"[42]"},{"why":"The discovery of the large room-temperature anomalous Hall effect in Mn3Sn, the measured signal the paper attributes mainly to configuration II.","marker":"[24]"},{"why":"The rotating-Hall-bar experiment reporting deterministic switching at all rotation angles, which the paper compares with its sub-configuration results.","marker":"[17]"}],"fun_headline_variants":["Why Mn3Sn switching never flips: config II wins","Mn3Sn mystery solved: one crystal alignment locks polarity","Octupole switching stabilized by a single sub-configuration","Config II flips Mn3Sn polarity; Config I cancels out"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the 360 simulated sub-configurations cover all grain orientations in the polycrystalline film, that they are evenly distributed, and that each grain switches independently with no grain-boundary coupling; if the real distribution is biased or grains interact, the cancellation of configuration I signals may fail.","fun_headline_variants_meta":{"raw":{"variants":["Why Mn3Sn switching never flips: config II wins","Mn3Sn mystery solved: one crystal alignment locks polarity","Octupole switching stabilized by a single sub-configuration","Config II flips Mn3Sn polarity; Config I cancels out"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000831,"raw_usage":{"total_tokens":3665,"prompt_tokens":1015,"completion_tokens":2650,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":631,"completion_tokens_details":{"reasoning_tokens":2579}},"tokens_in":631,"tokens_out":2650,"duration_ms":19209,"temperature":1.0,"reasoning_tokens":2579,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T13:55:09.015844+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the anomalous Hall switching polarity of a Mn3Sn film with a deliberately biased grain-orientation texture, or simulate a pair of coupled grains with differing sub-configurations: the model predicts configuration I contributes nothing only for an even orientation distribution, so a skewed texture or inter-grain coupling that produces a net configuration I signal would falsify the central claim.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"The authors' prior atomistic study showing deterministic switching in representative configuration II states, which the full 360-sub-configuration scan extends."},{"cited_title":"Tsai et al., Electrical manipulation of a topological antiferromagnetic state, Nature 580, 608 (2020)","cited_arxiv_id":null,"evidence_quote":"The experimental demonstration of SOT switching in polycrystalline Mn3Sn that motivates the configuration classification and supplies damping parameters."},{"cited_title":"Higo et al., Perpendicular full switching of chiral antiferromagnetic order by current, Nature 607, 474 (2022)","cited_arxiv_id":null,"evidence_quote":"The experiment reporting perpendicular full switching of Mn3Sn, used for simulation parameters and as evidence for configuration II switching."},{"cited_title":"Yoon et al., Handedness anomaly in a non-collinear antiferromagnet under spin –orbit torque, Nat","cited_arxiv_id":null,"evidence_quote":"The report of the handedness anomaly under SOT, which the earlier two-field balance incorporated and the new three-field model extends."},{"cited_title":"Lu et al., V oltage -gated spin -orbit torque switching in IrMn -based perpendicular magnetic tunnel junctions, Appl","cited_arxiv_id":null,"evidence_quote":"The rotating-Hall-bar experiment reporting deterministic switching at all rotation angles, which the paper compares with its sub-configuration results."}],"review_version":1}