{"id":"2e96914a-1f9c-42c9-a676-85390b82c6b8","arxiv_id":"1908.01264","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"A ferromagnetic domain wall driven by an AC field or current above the Walker threshold moves at a speed set by the AC frequency, not its amplitude, via phase locking of the wall's internal angle.","lead":"A magnetic domain wall can be pushed along a wire not by how strong an alternating field or current is, but by how fast it oscillates, because the wall's internal direction of magnetization synchronizes with the oscillation. This could let small oscillating signals drive magnetic memory devices at low power, and it repositions a previously ignored internal motion as a useful control.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Current-drive Eq. (8) appears to drop a nonzero time-averaged AC SOT term from Eq. (7), so the current-driven amplitude-independence claim may fail.","rationale":"The reader identified the time-averaging step as the weakest point, which is the correct general region, but the specific load-bearing assumption is not the hard-axis anisotropy averaging in Eq. (3); it is the averaging of the direct AC SOT term in the position equation (7) when forming Eq. (8). The field-driven central result, Eq. (4), is supported by the perturbative Adler treatment and by the micromagnetic simulations, and I do not object to it. The current-driven result, however, is derived from an equation that visibly contains a term whose phase-locked average is generically nonzero and proportional to AC amplitude. If that average is not negligible, the paper's headline claim of amplitude-independent velocity fails for the current-drive case, and Fig. 4b would need to show a systematic \\(J_{\\mathrm{AC}}\\) dependence rather than collapse onto a single line. The proposed check directly measures the disputed average in the same simulation setup, so it settles whether the omission is physical or harmless. The verdict is CONDITIONAL rather than REJECT because the field-driven mechanism and the synchronization framework are likely sound; what is needed is either a corrected Eq. (8) with the extra term, an explicit demonstration that \\(\\cos\\psi_0=0\\) in the locked regime, or a quantitative argument that the prefactor is below the simulation's resolution.","tokens_in":9330,"tokens_out":31935,"duration_ms":321771,"concrete_test":"Run the Fig. 4 micromagnetic setup at a fixed frequency inside the phase-locked band (for example near the center of the locking range with \\(J_{\\mathrm{DC}}=2\\times10^{12}\\,\\mathrm{A/m^2}\\)) for two AC amplitudes, \\(J_{\\mathrm{AC}}=2\\times10^{10}\\) and \\(8\\times10^{10}\\,\\mathrm{A/m^2}\\). Compute the time-averaged velocity difference \\(\\Delta V\\) and independently compute \\(\\frac{m}{2}\\gamma\\theta_{\\mathrm{SH}}\\langle\\cos(\\omega t)\\cos\\phi\\rangle\\) from the simulated trajectory. If \\(\\Delta V\\) matches this direct-SOT average to within numerical error, then Eq. (8) is missing a real AC-amplitude-dependent term; if \\(\\Delta V\\approx0\\), either \\(\\cos\\psi_0\\approx0\\) or the direct term is unexpectedly suppressed, and the claim can stand.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"In the current-driven case, the collective-coordinate position equation (7) contains a direct AC spin-orbit-torque term proportional to \\(\\cos(\\omega t)\\cos\\phi\\) in \\(\\dot X\\), written as \\(\\frac{m}{2}\\gamma\\theta_{\\mathrm{SH}}\\cos(\\omega t)\\cos\\phi\\). The paper then states that in the phase-locked regime, where \\(\\phi=\\omega t+\\psi\\) with \\(\\langle\\dot\\psi\\rangle=0\\), the time-averaged velocity is Eq. (8), whose displayed terms depend only on \\(\\omega\\) and on the DC STT \\(u_{\\mathrm{DC}}\\). But the time average of the direct SOT term is \\(\\frac{m}{4}\\gamma\\theta_{\\mathrm{SH}}\\cos\\psi_0\\) to leading order, which is generally nonzero for a locked Adler phase and is proportional to the AC amplitude through \\(\\theta_{\\mathrm{SH}}\\propto J_{\\mathrm{AC}}\\). Eq. (8) contains no such term, so the derivation from Eq. (7) to Eq. (8) appears to omit an AC-amplitude-dependent velocity contribution unless \\(\\cos\\psi_0\\) vanishes or the prefactor is numerically negligible. This is load-bearing because the abstract's \"independent of bias strength\" claim and the current-drive demonstration in Fig. 4b rest on Eq. (8). The field-driven result, Eq. (4), is not affected because Eq. (2) has no direct AC field term in \\(\\dot X\\), but the current-driven headline is.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper studies AC-bias-driven domain wall (DW) motion in a ferromagnetic nanowire with perpendicular magnetic anisotropy. The authors consider a DC field (or DC spin-transfer torque) strong enough to exceed the Walker breakdown, so that the DW in-plane angle precesses, together with a transverse AC field (or AC spin-orbit torque). Using collective-coordinate equations, they derive an Adler-type equation for the DW angle and show that, when the angle phase-locks to the AC drive, the time-averaged DW velocity is proportional to the AC frequency and independent of the AC amplitude: Eq. (4) for the field-driven case and Eq. (8) for the current-driven case. They also derive the critical frequencies of the phase-locking-unlocking transition (Eq. (5)) and support the analytical results with micromagnetic simulations.","tokens_in":9653,"tokens_out":14103,"duration_ms":135114,"significance":"The field-driven mechanism is a clever and clearly presented idea: the internal angle degree of freedom, usually associated with Walker breakdown and reduced velocity, is used as a resonant handle that moves the DW at a frequency-controlled velocity. Equation (4) is parameter-free in the sense that no quantity in it is fitted to simulations; material parameters and applied fields are direct inputs. The micromagnetic simulations for the field-driven case reproduce the predicted linear velocity plateau and the locking-unlocking windows for H_z=5 and 20 mT and H_x=5-40 mT, which is convincing. If the current-driven result is placed on a sound footing, the predicted DW-type-independent unidirectional motion is practically important for racetrack memory. The paper is well structured and builds on standard collective-coordinate and Adler-equation tools, and the 'internal degree of freedom as a control handle' perspective is likely to stimulate further work.","major_comments":[{"comment":"The derivation of Eq. (8) from Eq. (7) appears to omit a nonzero time-averaged contribution. Equation (7) contains a direct AC spin-orbit-torque term in \\dot{X} proportional to cos(\\omega t)cos\\phi. In the phase-locked regime \\phi=\\omega t+\\psi with \\langle\\dot{\\psi}\\rangle=0, the time average is \\langle\\cos(\\omega t)\\cos(\\omega t+\\psi)\\rangle=\\frac{1}{2}\\langle\\cos\\psi\\rangle, which is generally nonzero. Since the SOT magnitude is proportional to the AC current amplitude, this term contributes a current-amplitude-dependent velocity, and if the prefactor carries the DW-type label F it also breaks the claimed DW-type independence. Equation (8) contains no such term, so the statement that Eq. (8) follows from Eq. (7) is not justified, and the current-driven claims of amplitude independence and F-independence are not established. The authors should either demonstrate that cos\\psi=0 throughout the relevant locking range using the phase equation, or include the term in Eq. (8) and in the comparison with the Fig. 4b simulations.","section":"Current-driven case, Eqs. (6)-(8)"}],"minor_comments":[{"comment":"Several figure captions reference incorrect equation numbers: Fig. 3a,b should cite Eq. (4) rather than Eq. (6); Fig. 3c should cite Eq. (5) rather than Eq. (8); Fig. 4b should cite Eq. (8) rather than Eq. (12); Fig. 4c should cite the current-driven analog of Eq. (5).","section":"Figure captions"},{"comment":"The phase difference \\delta\\phi is used in the discussion of the phase-locking condition but is not explicitly defined before the ansatz; please define it formally as \\delta\\phi=\\phi-\\omega t.","section":"Main text, paragraph after Eq. (4)"},{"comment":"There are typographical errors such as 'prcession' instead of 'precession'; the manuscript should be carefully proofread.","section":"Main text, paragraph after Eq. (5)"},{"comment":"The figure caption lists results for H_z=5, 10, and 20 mT, but the main text only mentions H_z=5 and 20 mT for the velocity plots; please clarify the status of the H_z=10 data in the text.","section":"Fig. 3c"}],"recommendation":"major_revision","confidential_remarks":"The field-driven part of the paper is solid and the central idea is attractive. The main obstacle is the derivation of Eq. (8), which currently drops a nonzero time-averaged SOT term. This is a fixable issue, but it affects the current-driven central claim, so the revision should be substantial. The figure captions also contain several wrong equation-number citations, which should be corrected before resubmission."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: the field-driven part of this paper is a genuinely clever result, but the current-driven half has a suspicious dropped term that could break the headline claim for AC currents. It deserves peer review, but the authors should fix or defend Eq. (8).\n\nWhat's new: they treat the DW's internal angle as the oscillator being synchronized by an AC bias, and the gyrotropic coupling then converts phase-locked rotation of that angle into translational motion. That's a nice twist on the Adler equation. The field-driven velocity, V = FΔ(ω/2 − γH_z), is derived cleanly without any fitting, and the micromagnetic simulations in Figs. 2 and 3 support the linear plateau and the locking–unlocking windows across H_z = 5 and 20 mT and H_x = 5–40 mT. The absence of fitting is real credit; material parameters and biases are direct inputs.\n\nNow the soft spot. In the current-driven case, Eq. (7) contains a direct AC spin–orbit-torque term, (m/2)γθ_SH cos(ωt)cosφ, in Ẋ. In the phase-locked regime, φ = ωt + ψ, so this term's time average is (m/4)γθ_SH cosψ. That is generically nonzero for an Adler-type locked solution. Eq. (8), as printed, has no such term, so the derivation from Eq. (7) to Eq. (8) appears to omit an AC-amplitude-dependent contribution. The stress-test note is on target. I don't see any argument in the text that ψ sits at ±π/2 across the entire locking window, or that the projection coefficient m is zero. If the term survives, the velocity in the current-driven case depends on J_AC through θ_SH, and the abstract's 'independent of bias strength' claim is wrong for currents. The simulations in Fig. 4b supposedly match Eq. (8) for four different J_AC values; if the omitted term were significant, those curves should separate. Either the coefficient is genuinely tiny, or the overlap needs a physical explanation. Either way, this is load-bearing and must be addressed.\n\nMinor issue: the figure captions reference equations with inconsistent numbering (Eq. (6) vs Eq. (4), Eq. (12) vs Eq. (8)), which is cosmetic but should be cleaned up.\n\nThe reader's ACCEPT verdict is reasonable for the field-driven mechanism, but a bit generous overall because the current-driven claim is part of the abstract and the demonstrations. I'd send this to a good referee rather than desk reject: the core idea is worth engaging, and the current-driven issue is fixable. The field-driven result alone would make a solid short paper, but as written, the current-driven section needs revision before it's reliable.","headline":"Neat idea and a solid field-driven result, but the current-driven extension appears to drop a time-averaged AC SOT term, so the 'independent of bias strength' claim for AC current is shaky.","tokens_in":10165,"tokens_out":7983,"would_cite":true,"duration_ms":83421,"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":"A magnetic domain wall can be driven by the frequency of an AC field or current: once phase-locked, its mean velocity is linear in frequency and independent of amplitude.","keywords":["domain wall","phase locking","AC field","AC current","Walker breakdown","spin-transfer torque","spin-orbit torque","micromagnetic simulation"],"falsifier":"Measure the time-averaged domain-wall velocity inside the predicted lock-in band while holding the AC frequency fixed and sweeping the AC amplitude over a severalfold range, including values where the DC bias is just above the Walker breakdown. The central claim predicts the velocity is flat in amplitude; a clear amplitude dependence, or the failure of the linear-in-frequency plateau to appear at all, would falsify the phase-locking explanation.","tokens_in":9132,"feed_emoji":"🧲","tokens_out":8006,"duration_ms":73540,"temperature":0.7,"pith_summary":"The paper shows that a magnetic domain wall can be moved not by how hard an AC bias pushes, but by how fast it oscillates. In a ferromagnet biased above the Walker breakdown, the wall's in-plane angle precesses; the authors find that an AC field (or an AC spin-orbit-torque current) can phase-lock to that precession, and the gyrotropic coupling between angle and position then converts the locked rotation into a steady drift. The resulting time-averaged wall velocity is proportional to the AC frequency and independent of the AC amplitude, a result the authors derive analytically and confirm in micromagnetic simulations. The mechanism matters because it turns the previously neglected internal angle of the wall into a useful control handle, with potential low-power spintronic operation.","feed_headline":"AC bias moves domain walls at a speed set by frequency alone","feed_subtitle":"Phase-locking between wall precession and AC drive gives velocity linear in frequency and independent of amplitude.","key_machinery":"The load-bearing object is the Adler-type phase equation obtained by time-averaging the LLG angle dynamics above the Walker breakdown: $\\dot{\\phi} = \\omega_0 + \\omega_a\\sin(\\phi-\\omega t) + \\omega_a\\sin(\\phi+\\omega t)$ (Eq. 3). Its steady phase-locked solution $\\dot{\\phi}=\\omega$ feeds into the gyrotropically coupled position equation $\\dot{X}=F\\Delta(\\dot{\\phi}/\\alpha-\\gamma H_z)$ (Eq. 2), turning synchronized precession into wall translation. The locking range itself comes from a small-amplitude ansatz for the phase mismatch, which yields the upper and lower critical frequencies of Eq. (5). For the current drive, the same Adler structure emerges from the angle equation containing DC spin-transfer torque and AC spin-orbit torque.","core_discovery":"The central claim is that an AC bias, despite having zero time average, can move a ferromagnetic domain wall through synchronization of the wall's internal angle with the drive. Above the Walker breakdown the angle precesses at rate $\\omega_0$; when an AC field of frequency close to $\\omega_0$ is added, the angle locks, $\\dot{\\phi} = \\omega$, and the position equation then gives a time-averaged velocity $V=F\\Delta(\\omega-\\gamma H_z)/2$ (Eq. 4). The AC-induced part grows linearly with frequency and does not depend on the AC strength. The same structure appears for current drive, where DC spin-transfer torque plus AC spin-orbit torque yields Eq. (8), with velocity independent of the wall type $F=\\pm1$. The phase-locked regime is bounded by critical frequencies (Eq. 5), and both the velocity and the critical frequencies match micromagnetic simulations.","pith_inferences":["Inside the locked band, the amplitude independence makes the drive naturally insensitive to current or field fluctuations, so a practical device's speed would be set by the stability of the AC source rather than by its power.","The synchronization mechanism relies only on a gyrotropic coupling between an internal phase and position, so similar phase-locked translation might be engineered for skyrmions or other solitons; antiferromagnetic walls, lacking this coupling, would not move this way.","The authors leave fractional synchronization as an open possibility: at drive frequencies near rational submultiples of the precession frequency, one might see velocity plateaus at fractions of the main locked value, a signature available to future experiments.","Because the lock-in bandwidth grows with AC amplitude via Eq. (5), sweeping frequency at fixed amplitude maps the band edges; checking whether those edges follow the predicted form of Eq. (5) would test the theory beyond the velocity plateau itself."],"forward_implications":["For field drive in the locked regime, the time-averaged wall velocity is $V=F\\Delta(\\omega-\\gamma H_z)/2$: the AC-induced component is exactly linear in frequency and independent of AC amplitude.","Locking occurs only between the two critical frequencies of Eq. (5); outside this band the wall is unlocked and its velocity is no longer set by the AC frequency.","For current drive, a DC spin-transfer torque plus an AC spin-orbit torque yields the same linear-in-frequency, amplitude-independent velocity, and the result holds for both wall chiralities, so alternating walls in a train move in the same direction.","The analytical velocity and critical-frequency formulas agree with micromagnetic simulations over the tested parameter sets, including different DC field strengths and AC amplitudes."],"supporting_citations":[{"why":"Supplies the collective-coordinate model of a 180-degree domain wall and defines the Walker-breakdown regime that the AC drive is added to.","marker":"[4]"},{"why":"Provide the low-energy equations of motion for the wall position and angle used to derive Eqs. (1)-(2).","marker":"[7,8]"},{"why":"Supplies the Adler equation and its phase-locked solution, the mathematical core of the synchronization argument.","marker":"[13]"},{"why":"Gives the Pt/NiO/CoTb heterostructure as a material platform where the proposed AC spin-orbit-torque drive could be realized.","marker":"[14]"},{"why":"Provide the Landau-Lifshitz-Gilbert equation that underlies both the field-driven and current-driven dynamics.","marker":"[15,16]"},{"why":"Supply the spin-transfer-torque terms used in the augmented LLG equation for the current-driven calculation.","marker":"[20,21]"},{"why":"Supply the spin-orbit-torque terms used in the augmented LLG equation for the AC-current-driven calculation.","marker":"[22-26]"}],"fun_headline_variants":["AC bias moves domain walls; speed set by frequency alone","Zero-average AC drive still propels domain walls","Phase-locking tunes AC-driven wall velocity to frequency","Wall speed from AC bias: frequency matters, amplitude doesn't","AC bias: wall speed from phase-locking, not amplitude"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"Everything rests on the assumption that above the Walker breakdown the rapidly oscillating hard-axis anisotropy term in the angle equation averages out over each precession cycle, leaving a simple Adler-type phase equation; if that averaging is not valid, the phase-locked solution and the frequency-linear velocity formula do not follow.","fun_headline_variants_meta":{"raw":{"variants":["AC bias moves domain walls; speed set by frequency alone","Zero-average AC drive still propels domain walls","Phase-locking tunes AC-driven wall velocity to frequency","Wall speed from AC bias: frequency matters, amplitude doesn't","AC bias: wall speed from phase-locking, not amplitude"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000879,"raw_usage":{"total_tokens":3764,"prompt_tokens":875,"completion_tokens":2889,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":491,"completion_tokens_details":{"reasoning_tokens":2809}},"tokens_in":491,"tokens_out":2889,"duration_ms":21528,"temperature":1.0,"reasoning_tokens":2809,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T15:18:52.613640+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the time-averaged domain-wall velocity inside the predicted lock-in band while holding the AC frequency fixed and sweeping the AC amplitude over a severalfold range, including values where the DC bias is just above the Walker breakdown. The central claim predicts the velocity is flat in amplitude; a clear amplitude dependence, or the failure of the linear-in-frequency plateau to appear at all, would falsify the phase-locking explanation.","supporting_citations":[{"cited_title":"Rotating magnetic field driven antiferromagnetic domain wall motion: Role of Dzyaloshinskii-Moriya interaction","cited_arxiv_id":"1904.00870","evidence_quote":"Supplies the collective-coordinate model of a 180-degree domain wall and defines the Walker-breakdown regime that the AC drive is added to."}],"review_version":1}