{"id":"00d9aef1-4292-4e87-a856-a1a301e286c0","arxiv_id":"1908.04931","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"A gate-controlled Dzyaloshinskii-Moriya interaction step produces a large, controllable transverse skyrmion motion that can encode multi-bit states and Boolean functions.","lead":"This paper proposes a magnetic skyrmion field-effect transistor in which a gate-controlled Dzyaloshinskii-Moriya interaction step deflects skyrmions sideways. The authors argue that this deflection can serve as a multi-bit and Boolean logic device, a low-power alternative to conventional transistors.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The gated-FET claim relies on reading a transient boundary deflection that a second, opposite DMI interface may cancel; the paper never simulates this geometry.","rationale":"The paper has independent strengths: the micromagnetic simulations show a reproducible large transverse excursion whose sign correlates with Q·δ_D, and gate-controlled DMI is supported by cited experiments. Those facts make the underlying boundary-kick mechanism plausible and worth testing. The reader's conditional verdict already captures the key gap: the device-level claim requires a transient deflection to be read before a compensating opposite kick at the exit interface, but no end-to-end simulation is provided. My stress-test does not find a reason to reject the central physics claim, but it reinforces that the multi-bit field-effect transistor is an extrapolation from a two-boundary defect simulation. The missing derivation of F_DMI and the undefined λ are genuine quantitative weaknesses, but they are secondary to the absence of a full gated-device simulation showing that the displaced state is readable. Because the reader's verdict already conditioned acceptance on exactly this evidence, no change in verdict is needed.","tokens_in":8285,"tokens_out":7579,"duration_ms":87077,"concrete_test":"Run a micromagnetic simulation of the actual Fig. 3(a) gated geometry with detectors inside region 1: inject an STT-driven skyrmion at the left edge, set δ_D to −0.2, −0.1, +0.1, +0.2 erg/cm² for both Q = +1 and Q = −1, and record y(t) while the skyrmion is inside region 1 and after it exits the right edge. If the maximum y-deflection inside region 1 is not monotonic in δ_D, or does not persist for a realistic TMR integration time before the skyrmion reaches the exit interface, then the multi-bit FET claim is unsupported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central device claim converts Eq. (7) into a persistent multi-bit y-position inside the gate region, but the evidence in Fig. 2 is for a planar defect that the skyrmion crosses completely. A DMI gate region has two interfaces: the gradient of the DMI is +δ_D on entry and −δ_D on exit, so the additional field in Eq. (4) reverses sign at the second interface. Therefore, under the paper's own Thiele model, a skyrmion that traverses the entire gate receives a second large transverse kick opposing the first, and the residual displacement is the small motion whose sign depends only on Q, not on δ_D. The FET concept instead requires reading the transient deflection inside region 1 before the compensating exit kick occurs. The paper never simulates this gated geometry, never specifies how long the displaced y-position persists, and never shows a detector sampling window compatible with TMR readout. Additionally, Eq. (7) itself is asserted rather than derived and λ is never defined, which makes quantitative device thresholds untestable. The boundary-induced transverse kick may be real, but the load-bearing multi-bit transistor claim is not established by the presented simulations.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a gate-controlled skyrmion field-effect transistor based on spatially inhomogeneous Dzyaloshinskii-Moriya interaction (DMI). The authors show by micromagnetic simulations that both point and planar DMI defects deflect a current-driven skyrmion transversally, producing a large transient deflection whose sign is set by the product of topological charge Q and DMI difference δ_D, plus a small residual shift whose sign depends only on Q. They model the effect via a Thiele equation with an effective damping-like torque, derive velocities in Eqs. (6) and (7), and propose a multi-bit transistor geometry (Fig. 3) with tunnel-magnetoresistance readout.","tokens_in":8473,"tokens_out":4466,"duration_ms":47377,"significance":"The central physics—an equilibrium damping-like torque from a DMI gradient that yields a sign-controlled transverse skyrmion deflection—is supported by direct micromagnetic simulations for both point and planar defects. This provides a concrete, falsifiable prediction (the Q·δ_D sign rule) and a plausible route to gate-controlled skyrmion logic. The paper is explicit that the large motion is a new extrinsic contribution to the skyrmion Hall effect, and the simulations are independent of the approximate Thiele model. However, the quantitative Thiele treatment is incomplete (λ undefined, F_DMI asserted) and the transistor-level claim is not tested in the actual two-interface geometry, so the device demonstration is significantly weaker than the defect-physics demonstration.","major_comments":[{"comment":"The force term F_DMI = (-(δ_D/2aM_S)πλ, 0) is introduced without derivation from the effective field in Eq. (4), and the characteristic length λ is never defined. Because Eq. (7) is the quantitative basis for the Δy1 vs δ_D relation in Fig. 3(c) and for the claim that only a 4% DMI change produces 30 nm of deflection, the missing derivation and undefined λ make the central quantitative prediction untestable.","section":"Eqs. (5)–(7) and the planar-defect discussion"},{"comment":"The transistor geometry has two DMI interfaces (entry into the gated region and exit from it), but the planar-defect simulation in Fig. 2 shows that a skyrmion crossing a single DMI step receives a large transverse kick at the entry edge and an opposite large kick at the exit edge, leaving only the small displacement Δy2 after full traversal. The manuscript never simulates the full gated region with its two interfaces, never specifies that the multi-bit readout must occur during the transient excursion inside region 1, and provides no detector-timing analysis to show that a tunnel-magnetoresistance readout can capture that transient. The multi-bit FET operation is therefore not established by the presented simulations.","section":"Fig. 3(a)–(c) and Fig. 2"},{"comment":"The paper states that 'the sign of large motion is determined by the sign of Q·δ_D', but for a point defect the trajectory is deflected upward on one side and downward on the other (Fig. 1(a)); the same should occur for the two boundaries of the planar defect. The manuscript does not define whether the sign of the large motion refers to the first deflection or to the maximum excursion, which is important because the proposed FET state relies on the y-position inside the gated region during a transient deflection.","section":"Fig. 2 and Fig. 3(c)"}],"minor_comments":[{"comment":"The abstract contains a typo: 'The propose device' should be 'The proposed device'.","section":"Abstract"},{"comment":"The symbols in Eq. (4) and the surrounding text are garbled (e.g., '𝒎ෝ ௜' and '𝛿' in the displayed equations); all vectors and indices should be typeset consistently so that the derivation can be followed.","section":"Eq. (4)"},{"comment":"In Eq. (1), the force is written as F = Q v × B, but the vector character of F and v is implied rather than shown, and the relationship B = -4π s_net z-hat with the gyromagnetic ratio should be stated with consistent notation.","section":"Eq. (1)"},{"comment":"For point defects, the text says the trajectory is 'largely deflected in the −y direction on the left side' and '+y direction on the right side'; define which of these is called the sign of the large motion, otherwise the statement 'the sign of large motion is determined by the sign of Q·δ_D' is ambiguous.","section":"Section 4, point-defect paragraph"},{"comment":"The extraction of Δy1 from simulations is not described (e.g., at what time or after which deflection event it is measured); since the device proposal depends on this quantitative relation, a clearer description of the measurement procedure would help.","section":"Fig. 3(c) and device discussion"}],"recommendation":"major_revision","confidential_remarks":"The core physics appears sound and the simulations provide direct support for the sign rule, which is a strength of the paper. The main risk is that the transistor claim is not backed by a simulation of the actual two-interface gated geometry; the authors need either to simulate that geometry or to explicitly reframe the contribution as only the single-interface defect physics. Also, the paper relies on two self-citations (Refs. [25] and [28]) for the DMI-field formalism; I see no circularity because the micromagnetic simulations are independent, but the derivation should be made self-contained in the main text."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"What you should know: the paper identifies a clean effect — a skyrmion crossing a DMI step receives a large transient transverse kick with sign set by Q·δ_D. The micromagnetic simulations for point and planar defects back that up, and the effect is new in the cited literature. That part deserves credit.\n\nThe soft spot is the device. The FET claim needs the skyrmion to land at a displaced y-position inside the gate region and stay there long enough for TMR readout. But a finite gate region has two DMI interfaces. Under the paper's own model, the second interface delivers an equal and opposite transverse kick, so a skyrmion that crosses the whole gate ends with only the small Q-dependent shift, not the large δ_D-dependent one. The simulation in Fig. 2 is for a planar defect the skyrmion traverses completely; it shows the transient deflection, not a persistent multi-bit state. The paper never simulates the gated geometry, never specifies a readout window, and never shows the skyrmion remaining at the deflected position. That is a load-bearing gap.\n\nOther issues are more minor. Eq. (7) is asserted rather than derived from Eq. (4), λ is never defined, and there is no quantitative comparison between the Thiele prediction and the simulated trajectories. The reliance on two self-cited papers for the effective DMI field is a bit circular, but the micromagnetics independently corroborate the sign rule, so that concern doesn't sink anything.\n\nNet: the core physics is plausible and the simulations are direct evidence for it. The device section overreaches relative to the evidence, but the overreach is addressable. A serious referee should engage — the paper deserves revision, not rejection. If I were editing, I'd send it out and ask for either a full device simulation with a defined readout sequence or a reframed claim that sticks to a controllable transient deflection rather than a stable multi-bit transistor.","headline":"The DMI-step transverse kick is real and well demonstrated, but the FET concept needs to read a transient deflection that a second interface cancels; the paper never simulates that geometry.","tokens_in":9032,"tokens_out":2873,"would_cite":true,"duration_ms":31221,"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 gate-controlled DMI step exerts an equilibrium damping-like torque that deflects skyrmions in a direction set by the product of topological charge and DMI difference, enabling multi-bit field-effect transistors.","keywords":["magnetic skyrmion","Dzyaloshinskii-Moriya interaction","skyrmion Hall effect","spin-orbit torque","field-effect transistor","multi-bit logic","micromagnetic simulation","topological charge"],"falsifier":"In a clean film with a single planar DMI step, drive a skyrmion across the step and measure the transverse displacement while independently reversing the sign of the DMI difference and the skyrmion's topological charge; the claim fails if the large deflection does not flip with the product $Q\\delta_D$, or if the same deflection is not observed when the skyrmion is moved through the step without current (for example, by an oscillating magnetic field gradient).","tokens_in":8035,"feed_emoji":"🧲","tokens_out":9100,"duration_ms":82873,"temperature":0.7,"pith_summary":"The paper proposes that a spatial step in the Dzyaloshinskii-Moriya interaction (DMI) acts like an equilibrium damping-like spin-orbit torque on a magnetic skyrmion, producing a large sideways deflection whose direction is set by the product of the skyrmion's topological charge $Q$ and the DMI difference $\\delta_D$. Micromagnetic simulations with both point and planar DMI defects show this large transverse motion on top of a small motion that depends only on $Q$. Because the DMI can be changed by a gate voltage, the deflection can be tuned continuously, which the authors use to design a skyrmion field-effect transistor with multi-bit states and Boolean AND/OR logic. The interest is a low-power logic and memory scheme based on magnetic solitons that exploits, rather than suppresses, the skyrmion Hall effect.","feed_headline":"Gate-controlled DMI steers skyrmions sideways for multi-bit logic","feed_subtitle":"Varying a gate voltage changes a DMI step, deflecting skyrmions by 30 nm with just a 4 percent DMI change.","key_machinery":"The Dzyaloshinskii-Moriya interaction (DMI) is an antisymmetric exchange coupling that stabilizes chiral spin textures like skyrmions. The load-bearing object is the effective equilibrium field at a DMI step, $H_{\\rm DMI,eff} = (1/M_S)[D\\,\\hat{\\boldsymbol{y}}\\times(\\partial\\boldsymbol{m}/\\partial x) + (\\delta_D/2a)(\\hat{\\boldsymbol{y}}\\times\\boldsymbol{m})]$, obtained from the interface DM energy for a planar defect. The second term is independent of current and has the symmetry of a damping-like spin-orbit torque; in the collective-coordinate equation of motion it enters as a force $F_{\\rm DMI} = (-(\\delta_D/2aM_S)\\pi\\lambda, 0)$. Combined with the gyroscopic and dissipative terms, this force yields Eqs. (6)-(7), the quantitative statement that the large transverse motion has sign set by $Q\\delta_D$.","core_discovery":"The central claim is that a spatial step in the Dzyaloshinskii-Moriya interaction (DMI) creates an additional equilibrium field with the same symmetry as a current-driven damping-like spin-orbit torque, and that this field deflects a passing skyrmion sideways by a large amount whose sign is set by the product $Q\\delta_D$ of the skyrmion topological charge and the DMI difference. For $\\alpha=\\beta$, the collective-coordinate equations of motion give $v_x = v_s - \\frac{\\alpha\\mathcal{D}}{\\mathcal{G}^2+(\\alpha\\mathcal{D})^2}\\left(\\frac{\\delta_D}{2aM_S}\\right)\\pi\\lambda$ and $v_y = \\frac{\\mathcal{G}}{\\mathcal{G}^2+(\\alpha\\mathcal{D})^2}\\left(\\frac{\\delta_D}{2aM_S}\\right)\\pi\\lambda$, so the transverse velocity is proportional to $\\delta_D$ and to the gyromagnetic coupling $\\mathcal{G}$ (hence to $Q$), while the longitudinal velocity is reduced. The trajectory consequently splits into a large deflection while crossing the step and a small residual displacement that depends only on $Q$. The authors then propose placing a gate over one region to change its DMI, making a controlled DMI step; varying the gate voltage selects the skyrmion's final $y$-position in the gated region, giving multi-bit states, and series/parallel connections of such gates implement AND and OR logic.","pith_inferences":["The authors treat the large motion as a crossing effect; whether the deflected position persists long enough for tunnel-magnetoresistance readout under thermal noise is not simulated, so the stability of the multi-bit state is an open question.","Because $v_y$ in Eq. (7) is independent of the driving current while $v_x$ grows with current, the skyrmion Hall angle shrinks at high drive, implying a current-tunable trade-off between speed and deflection that could matter for racetrack designs.","The same gate-controlled DMI-step mechanism may extend to other chiral textures such as domain walls or antiskyrmions, and possibly to antiferromagnetically coupled systems where the small $Q$-only displacement could be suppressed while the large $\\delta_D$-driven deflection remains.","The linear relation between deflection and $\\delta_D$, combined with trajectory readout, suggests that skyrmion paths could serve as a local probe of interfacial DMI differences in a film."],"forward_implications":["A gate voltage that changes the DMI by roughly 4% deflects a skyrmion about 30 nm, so multilevel positioning is feasible without large voltage swings.","The large deflection direction reverses when either the skyrmion's topological charge or the sign of the DMI difference is flipped, giving a binary control knob.","Connecting gated DMI steps in series implements AND logic; parallel connections implement OR logic, so the same building block supports Boolean functions.","Because the effect is linear in $\\delta_D$, increasing the DMI difference gives proportionally larger deflection, easing detector placement.","The equilibrium nature of the torque implies the deflection should also appear when the skyrmion is moved through the DMI step without current, for example by an oscillating magnetic field gradient, offering a current-free test."],"supporting_citations":[{"why":"Provides the collective-coordinate equation of motion with gyroscopic and dissipative terms used to derive the velocity formulas.","marker":"[15]"},{"why":"Shows that the sign of the DMI difference controls domain wall motion, motivating the same sign dependence for skyrmions.","marker":"[25]"},{"why":"Demonstrates gate-controlled interfacial DMI with up to 100% tunability, grounding the proposed gate operation experimentally.","marker":"[26]"},{"why":"Derives the effective DMI field at an inhomogeneous DMI interface, the source of the equilibrium damping-like torque.","marker":"[28]"},{"why":"Proposes moving skyrmions with oscillating magnetic field gradients, the current-free method suggested to identify the equilibrium torque.","marker":"[29]"}],"fun_headline_variants":["Skyrmion FET uses gate-controlled DMI to deflect bits","Multi-bit skyrmion logic from a gate-tuned DMI step","DMI gradient gives skyrmions a giant sideways kick","Gate voltage steers skyrmions via DMI step for logic","Skyrmion transistor: gate-controlled DMI enables multi-bit states"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The device concept assumes that the skyrmion's large sideways excursion while crossing the gated DMI step leaves it at a well-defined, stable $y$-position inside the gated region long enough for a detector to read it; the paper does not simulate the full gated device or the readout step.","fun_headline_variants_meta":{"raw":{"variants":["Skyrmion FET uses gate-controlled DMI to deflect bits","Multi-bit skyrmion logic from a gate-tuned DMI step","DMI gradient gives skyrmions a giant sideways kick","Gate voltage steers skyrmions via DMI step for logic","Skyrmion transistor: gate-controlled DMI enables multi-bit states"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000932,"raw_usage":{"total_tokens":4009,"prompt_tokens":985,"completion_tokens":3024,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":601,"completion_tokens_details":{"reasoning_tokens":2935}},"tokens_in":601,"tokens_out":3024,"duration_ms":22219,"temperature":1.0,"reasoning_tokens":2935,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T13:28:15.100609+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"In a clean film with a single planar DMI step, drive a skyrmion across the step and measure the transverse displacement while independently reversing the sign of the DMI difference and the skyrmion's topological charge; the claim fails if the large deflection does not flip with the product $Q\\delta_D$, or if the same deflection is not observed when the skyrmion is moved through the step without current (for example, by an oscillating magnetic field gradient).","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the collective-coordinate equation of motion with gyroscopic and dissipative terms used to derive the velocity formulas."},{"cited_title":"Hong, S.-W","cited_arxiv_id":null,"evidence_quote":"Shows that the sign of the DMI difference controls domain wall motion, motivating the same sign dependence for skyrmions."},{"cited_title":"Srivastava, M","cited_arxiv_id":null,"evidence_quote":"Demonstrates gate-controlled interfacial DMI with up to 100% tunability, grounding the proposed gate operation experimentally."},{"cited_title":"Lee, J.-H","cited_arxiv_id":null,"evidence_quote":"Derives the effective DMI field at an inhomogeneous DMI interface, the source of the equilibrium damping-like torque."},{"cited_title":"Psaroudaki and D","cited_arxiv_id":null,"evidence_quote":"Proposes moving skyrmions with oscillating magnetic field gradients, the current-free method suggested to identify the equilibrium torque."}],"review_version":1}