{"id":"cca0142a-e86f-467a-a1fa-aa243bcd6a13","arxiv_id":"2412.10888","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"A dipolar-coupled magnetic multilayer model shows that including antimagnons produces topological bands with nonzero Chern numbers and tunable spin-wave chirality.","lead":"Researchers propose a model where left-handed spin waves, called antimagnons, join ordinary magnons in magnetic multilayers to create topologically protected surface states. The result points toward new ways to control spin-wave chirality and coupling in future low-power magnonic devices.","discovery_kind":"first_principles","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Dissipative coupling regions yield imaginary frequencies, so the computed Chern number—and hence the claimed bulk-boundary correspondence—is potentially ill-defined.","rationale":"The reader's weakest assumption concerned the nearest-neighbor dipolar truncation and the phenomenological thickness d_ph in the MSSW comparison. That is real, but it affects mainly the quantitative link to classic MSSW. The more fundamental risk is internal: the model's own 'dissipative coupling' mechanism produces imaginary-frequency bands, and the paper still applies the standard Hermitian Chern formula. This threatens the central assertion of nonzero Chern numbers and topological protection, not just the MSSW identification. However, the micromagnetic simulations (Sec. D) do show nonreciprocal surface states, so the concern is not that the states are absent, but that their topological character is unproven in the complex-gap regime. The paper could fix this by restricting claims to parameter regions with a real gap, or by adopting a non-Hermitian invariant. I therefore keep the reader's conditional verdict (conditions are added, not removed).","tokens_in":23456,"tokens_out":16388,"duration_ms":158520,"concrete_test":"For the nontrivial field values of Figs. 2(b), 3(b), and 6(b)-(c), diagonalize ηH_FF(kx,kz) and ηH_AF(kx,kz) on a dense 2D grid and plot Im ω. If Im ω ≠ 0 anywhere, recompute the plaquette sum in Eq. (S57): a non-integer or grid-dependent result would confirm the invariant is undefined. Then compute a non-Hermitian (biorthogonal) Chern number or include the damping/SOT terms in a finite-strip calculation to see whether the surface states survive with a real gap.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim rests on nonzero Chern numbers computed in Sec. A from the η-metric Berry connection of the BdG Hamiltonian ηH_FF_BdG (η=diag(1,1,−1,−1)). A Chern number is a valid topological invariant only when bands are separated by a real gap over the entire BZ. The paper explicitly states that the magnon–antimagnon coupling (the 'dissipative' ϕ=π channel in Eq. S41) leads to level attraction and 'imaginary-frequency states between the anticrossing' (Sec. A). In such regions the eigenvalues of ηH_FF are complex, so the bands are not gapped in the real-frequency sense. The paper does not show that the parameter sets used in Figs. 2, 3 and 6 avoid these complex regions, nor does it use a non-Hermitian/complex-frequency topological invariant. If Im ω ≠ 0 over a finite area of the BZ, Eq. (S57) does not define an integer bulk invariant, and the 'topologically protected surface states' could instead be ordinary (non-topological) edge modes. This is load-bearing because the novelty claim—antimagnon-mediated topology—depends entirely on the Chern number being meaningful.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper revisits the spin-wave bands of ferromagnetic multilayers with dipolar interactions, treating magnon and antimagnon sectors on an equal footing in a Bogoliubov–de Gennes formalism. The authors introduce the '2D-SSH4 chain', a nearest-neighbor SSH-like model in which interlayer and intralayer dipolar couplings connect magnon and antimagnon states. They compute band structures and Chern numbers from the η-metric Berry connection, identify nonreciprocal surface states in antiparallel and parallel FM multilayers, and argue that these states share a common topological origin with magnetostatic surface spin waves (MSSW). They further demonstrate tunable coherent/dissipative interlayer coupling and layer-resolved chirality, and extend the model to AFM/FM multilayers. Micromagnetic COMSOL simulations for small stacks and a proposed propagating spin-wave spectroscopy measurement are used to support the band-structure and detectability claims.","tokens_in":23707,"tokens_out":7733,"duration_ms":66373,"significance":"If the central topological claim is valid, the work offers a compact phenomenological model connecting antimagnon physics to spin-wave topology, and it makes a concrete, testable prediction that MSSW-type nonreciprocal surface states are bulk-Chern protected. The paper's strengths include a careful second-quantization derivation of the BdG Hamiltonian in Section A, a gauge-invariant plaquette formula for the η-metric Chern number (Eqs. S55–S57), and COMSOL simulations that include the full dipolar interaction rather than the nearest-neighbor approximation. The propagating spin-wave spectroscopy calculation (Fig. S7) provides a falsifiable experimental signature. The significance is tempered by the fact that the Chern-number computation is only meaningful for gapped real-frequency bands, and the manuscript does not demonstrate that the parameter regimes used for the topological figures avoid the complex-frequency regions it explicitly associates with dissipative coupling.","major_comments":[{"comment":"The Chern number is computed from eigenvectors of the non-Hermitian matrix ηH_BdG using the η-metric Berry connection, but this invariant requires a real line gap separating the band from all others over the entire Brillouin zone. The manuscript itself states that the dissipative (ϕ=π) coupling channel produces 'level attraction and imaginary-frequency states between the anticrossing' (Eq. S41 and the following discussion), and the captions of Figs. 2 and 3 place the topological surface states 'between the level-attraction anticrossing bulk states'. In such regions the eigenvalues of ηH_BdG are complex and the eigenvectors can coalesce at exceptional points; the normalization ⟨χ_m|η|χ_m⟩ = ±1 used in Eq. (S50) is then not guaranteed, and the plaquette sum in Eq. (S57) does not define an integer invariant. The paper does not show that the parameter sets used for Ch_m avoid complex eigenvalues over the entire BZ, nor does it use a non-Hermitian topological invariant appropriate for complex line gaps. This is load-bearing because the bulk-boundary correspondence claim rests entirely on Ch_m being well-defined.","section":"Sec. A (Supplemental), Eqs. (S41) and (S57); Figs. 2 and 3"},{"comment":"The identification of the multilayer surface states with MSSW relies on a nearest-neighbor truncation of dipolar interactions together with a phenomenological thickness d_ph = 0.1 μm introduced to compensate for the truncation. The paper states that the truncation underestimates the effective film thickness and that d_ph ≈ 4d restores the missing dipolar weight, but no independent derivation or convergence test is provided. Since d_ph is adjusted so that the 40-layer calculation matches the classical MSSW dispersion, the quantitative identification is not parameter-free. The topological-origin claim would be substantially stronger if the authors showed that the Chern numbers and the surface-state spectrum are robust when the long-range dipolar interaction is included, for example by comparing directly with the full-dipolar COMSOL results or by varying the truncation range and d_ph systematically.","section":"Sec. B (Supplemental), MSSW comparison"},{"comment":"The AFM/FM multilayer model inherits the same complex-gap issue: the 6×6 Hamiltonian is treated with the same η-metric Chern-number formalism, and the surface states in Fig. 6 are identified without specifying whether the relevant bands have real eigenvalues across the BZ. In addition, the model excludes same-sublattice FM exchange within the AFM layer (only J_bc is retained), an approximation whose effect on the topological phase is not discussed. Please clarify whether the topological conclusions for Figs. 6(b,c) are independent of these two assumptions.","section":"Sec. C (Supplemental), AFM/FM model"}],"minor_comments":[{"comment":"The text reads 'we show that MWWS has the same topological origin as the 2D-SSH model' — 'MWWS' should be 'MSSW'.","section":"Main text, Sec. B (page 4)"},{"comment":"The caption for Fig. 3(b) says 'three antiparallel FM-bilayer unit cells', but the configuration is the parallel one (Fig. 3(a) and the surrounding text); it should read 'parallel'.","section":"Fig. 3 caption"},{"comment":"In Eq. (5), the term \\hat H_bc^0 is used without being defined in the main text; please define it explicitly or refer the reader to Eq. (S60) in the Supplemental Material.","section":"Eq. (5) and Sec. C"},{"comment":"The biorthonormalization ⟨χ_m|η|χ_m⟩ = ±1 is stated, but the paper does not explain how the sign is fixed for a band that changes from magnon-like to antimagnon-like across the BZ; this matters for the Berry-connection definition and should be clarified.","section":"Sec. A (Supplemental), Eq. (S50)"},{"comment":"The simulation details do not specify the size of the air domain or the boundary conditions beyond 'Magnetic Insulation'; adding these parameters would improve reproducibility.","section":"Sec. D (Supplemental), Fig. S5"}],"recommendation":"major_revision","confidential_remarks":"The central issue is the unaddressed complex band gap in the Chern-number calculation. If the authors cannot demonstrate a genuine real-frequency gap for the parameter sets used in Figs. 2, 3 and 6, or alternatively adopt a well-defined non-Hermitian/complex-gap invariant, the topological claims should be substantially revised or withdrawn."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Antimagnons are the actual new ingredient here: the paper explicitly includes the left-handed spin-wave sector and shows that interlayer dipolar coupling between magnons and antimagnons can reorganize the band topology of a ferromagnetic multilayer. The 2D-SSH4 model is a natural extension of Hu et al. and it does explain the nonreciprocal surface states in the COMSOL simulations. That part is solid and worth taking seriously.\n\nThe soft spots are real, though. The Chern number is computed from the η-metric Berry connection of the BdG Hamiltonian ηH_BdG. The paper itself states that the dissipative (ϕ = π) coupling channel gives level attraction and imaginary-frequency states between the anticrossings. In those regions the eigenvalues of ηH are complex, so the bands are not separated by a real gap everywhere in the BZ. The paper never shows that the parameter sets used in Figs. 2, 3, and 6 avoid those complex regions, nor does it switch to a non-Hermitian topological invariant. If Im ω ≠ 0 over some area, the plaquette construction in Eq. (S57) does not necessarily give an integer Chern number, and the claimed bulk-boundary correspondence is not established. This is not a nitpick; it hits the central novelty claim.\n\nThe other soft spot is the phenomenological thickness d_ph = 0.1 μm in Sec. B. The nearest-neighbor truncation requires this fitted parameter to reproduce the MSSW dispersion, and the argument that the multilayer surface states share the MSSW topological origin rests on that compensation. Long-range dipolar interactions could, in principle, do more than renormalize the thickness. The authors are transparent about the approximation, which is good, but it leaves the MSSW identification conditional.\n\nWhat the paper does well: the BdG derivation is careful, the COMSOL workflow is described in enough detail to be reproduced, and the chirality tuning predictions are concrete and testable. The propagating spin-wave spectroscopy simulation is a nice addition. No code or data are shipped, so verification of the Chern numbers is on the referee.\n\nBottom line: this is a promising and genuinely novel idea, but the topological invariant needs to be checked in the complex-frequency regime before the central claim is reliable. A serious referee should push on that point. I would not cite it yet, but I would bring it to a reading group and I would definitely send it out for review.","headline":"Clear, inventive antimagnon-based topological magnonics, but the Chern number may be ill-defined in the dissipative-coupling regime.","tokens_in":24226,"tokens_out":2124,"would_cite":false,"duration_ms":19523,"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":"By coupling ordinary magnons to left-handed antimagnons via layer-to-layer dipolar forces, the authors construct a 2D-SSH4 chain whose bands carry nonzero Chern numbers and argue that the resulting surface states share the topological…","keywords":["antimagnons","left-handed spin waves","dipolar interaction","topological magnonics","Chern number","Su-Schrieffer-Heeger model","magnetic multilayer","magnetostatic surface spin waves"],"falsifier":"A micromagnetic simulation of a 40-layer YIG film or a YIG/permalloy multilayer that keeps all dipolar couplings, with no nearest-neighbor truncation and no d_ph compensation, would settle the question: if the bands claimed to have Chern number plus or minus one become trivial, or the nonreciprocal spectral peak near 5 GHz disappears, the central claim is falsified. Equivalently, an experiment measuring transmission in the propagating spin-wave geometry at kx = +-k0 should show a one-way peak at the predicted frequency.","tokens_in":23266,"feed_emoji":"🧲","tokens_out":5810,"duration_ms":51829,"temperature":0.7,"pith_summary":"This paper argues that the usually neglected left-handed spin-wave branch, the antimagnon, is not a passive decay product but an active partner in band topology. In a dipolar-coupled ferromagnetic multilayer, the authors show, magnon and antimagnon sectors hybridize through long-range dipolar interactions, producing a two-dimensional Su-Schrieffer-Heeger model (the 2D-SSH4 chain) with nonzero Chern numbers. That result matters because it gives a single framework for the topological surface states of ferromagnetic multilayers and for the classic magnetostatic surface spin wave, and it points to experimentally tunable control of spin-wave chirality and interlayer coupling via external fields and spin-orbit torques.","feed_headline":"Antimagnons unlock topological spin-wave bands","feed_subtitle":"Dipolar coupling to left-handed spin waves gives nonzero Chern numbers, tunable chirality, and a unifying origin for surface spin waves.","key_machinery":"The central object is the 2D-SSH4 chain: a two-dimensional Su-Schrieffer-Heeger model with four internal states per magnetic bilayer unit cell (magnons a and c, antimagnons a-bar and c-bar) stacked along the film normal. A Bogoliubov-de Gennes Hamiltonian with metric eta = diag(1,1,-1,-1) doubles the magnon sector to include antimagnons; the interlayer dipolar matrix elements acquire a phase pi when coupling magnon to antimagnon, making that coupling dissipative (level attraction) rather than coherent (level repulsion). The Chern number computed with the eta-metric Berry connection is the topological invariant, and it becomes nonzero when the dipolar coupling breaks chiral and time-reversal symmetries.","core_discovery":"The central claim is that incorporating left-handed spin waves (antimagnons) into an enlarged bosonic Hamiltonian fundamentally reorganizes band topology. In the 2D-SSH4 model, interlayer dipolar interactions connect magnon states in one layer to antimagnon states in the next, generating dissipative (level-attractive) couplings alongside coherent ones; together with intralayer dipolar terms, these couplings open topological gaps with Chern number plus or minus one. The paper identifies the resulting nonreciprocal surface states with the same topological origin as magnetostatic surface spin waves, and shows that the hybridized bands can be tuned from trivial to nontrivial by external magnetic fields and spin torques, with all four layer-resolved chirality combinations accessible.","pith_inferences":["A natural extension the paper leaves implicit: the same antimagnon-mediated, dipolar mechanism might generate topological bands in other bosonic systems with a negative-frequency branch, such as coupled photonic or phononic waveguides with analogous left-handed modes.","Because the dissipative coupling strength decays exponentially with layer separation, engineering interlayer gaps could tune the level-attractive gap size, shifting surface-state frequency without changing external fields.","The empirical thickness compensation (d_ph) suggests that a fully long-range treatment might alter band topology for very thin films; a systematic study of surface-state onset versus layer count would test where the nearest-neighbor truncation breaks down.","The claim that antimagnons connect otherwise disconnected magnon bands implies that future magnon-only models could miss entire classes of topological transitions; direct observation of the predicted nonreciprocal transmission peak would strengthen the case for always coupling to the negative-frequency sector."],"forward_implications":["If the central claim is correct, a dipolar-coupled YIG/permalloy multilayer should display nonreciprocal topological surface states at GHz frequencies, detectable by propagating spin-wave spectroscopy.","The same model applies to AFM/FM multilayers, where two coupled 2D-SSH4 chains produce tunable surface states, with extra band crossings from the three atom types.","All four chirality combinations (RH-LH, LH-RH, LH-LH, RH-RH) are achievable in a single bilayer by tuning external fields and wavevector, with some states protected by the topological surface states.","Coherent coupling gives level-repulsive anticrossings while dissipative magnon-antimagnon coupling gives level-attractive anticrossings, and both can be controlled by fields and spin torques.","Magnetostatic surface spin waves in a single ferromagnetic film appear as the single-chain limit of this model, unifying a classic surface wave with modern topological magnonics."],"supporting_citations":[{"why":"Supplies the dipolar-coupled FM multilayer Hamiltonian with second quantization and the magnonic Chern band framework that this paper extends to include antimagnons.","marker":"[29]"},{"why":"Establishes the antimagnonics concept, the enlarged Hamiltonian with magnon and antimagnon sectors, and the need for spin-orbit torque to sustain antimagnon states.","marker":"[17]"},{"why":"Prior proof that magnetostatic surface spin waves are topological when time-reversal symmetry is preserved; this paper extends the argument to broken time-reversal symmetry.","marker":"[60]"},{"why":"Provides the classic magnetostatic surface spin wave dispersion used as the benchmark for the model's surface-state frequencies.","marker":"[61]"},{"why":"Motivates the Chern-number characterization of chiral spin-wave edge modes in dipolar magnetic thin films.","marker":"[58]"},{"why":"Demonstrates level attraction from dissipative magnon-photon coupling, supporting the paper's interpretation of magnon-antimagnon coupling as dissipative.","marker":"[46]"},{"why":"Supplies the bosonic Chern-number formalism with the eta-metric Berry connection used for the non-Hermitian BdG Hamiltonian.","marker":"[15]"}],"fun_headline_variants":["Antimagnons twist spin-wave bands into topology","Left-handed spin waves flip magnon band topology","Antimagnons give spin waves a topological twist","Tunable chirality via antimagnons in magnetic stacks","Antimagnons link topology to tunable spin-wave chirality"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The model's predictions rest on truncating layer-to-layer dipolar interactions to nearest neighbors and correcting the resulting thickness error with a fitted 0.1-micrometer scaling; if full long-range dipolar forces alter the band topology rather than merely renormalize the thickness, the predicted surface states and their identification with magnetostatic surface spin waves would not hold as stated.","fun_headline_variants_meta":{"raw":{"variants":["Antimagnons twist spin-wave bands into topology","Left-handed spin waves flip magnon band topology","Antimagnons give spin waves a topological twist","Tunable chirality via antimagnons in magnetic stacks","Antimagnons link topology to tunable spin-wave chirality"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000559,"raw_usage":{"total_tokens":2631,"prompt_tokens":891,"completion_tokens":1740,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":507,"completion_tokens_details":{"reasoning_tokens":1659}},"tokens_in":507,"tokens_out":1740,"duration_ms":11422,"temperature":1.0,"reasoning_tokens":1659,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T15:31:01.298342+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A micromagnetic simulation of a 40-layer YIG film or a YIG/permalloy multilayer that keeps all dipolar couplings, with no nearest-neighbor truncation and no d_ph compensation, would settle the question: if the bands claimed to have Chern number plus or minus one become trivial, or the nonreciprocal spectral peak near 5 GHz disappears, the central claim is falsified. Equivalently, an experiment measuring transmission in the propagating spin-wave geometry at kx = +-k0 should show a one-way peak at the predicted frequency.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the dipolar-coupled FM multilayer Hamiltonian with second quantization and the magnonic Chern band framework that this paper extends to include antimagnons."},{"cited_title":"Harms, H","cited_arxiv_id":null,"evidence_quote":"Establishes the antimagnonics concept, the enlarged Hamiltonian with magnon and antimagnon sectors, and the need for spin-orbit torque to sustain antimagnon states."},{"cited_title":"Yamamoto, G","cited_arxiv_id":null,"evidence_quote":"Prior proof that magnetostatic surface spin waves are topological when time-reversal symmetry is preserved; this paper extends the argument to broken time-reversal symmetry."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the classic magnetostatic surface spin wave dispersion used as the benchmark for the model's surface-state frequencies."},{"cited_title":"Shindou, J.-i","cited_arxiv_id":null,"evidence_quote":"Motivates the Chern-number characterization of chiral spin-wave edge modes in dipolar magnetic thin films."},{"cited_title":"Wang and X","cited_arxiv_id":null,"evidence_quote":"Supplies the bosonic Chern-number formalism with the eta-metric Berry connection used for the non-Hermitian BdG Hamiltonian."}],"review_version":1}