{"id":"761bf617-7844-42e8-aacd-88a51d062d16","arxiv_id":"2504.20225","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"Strain that modulates both Heisenberg and Kitaev couplings can make topological magnon edge modes nonreciprocal (uniaxial) or flat and non-propagating (twist) in a honeycomb ferromagnet.","lead":"This paper derives the magnon Hamiltonian of a strained Heisenberg-Kitaev magnet and shows numerically how different lattice deformations change the magnon bands. It offers a path to controlling magnon propagation, including nonreciprocal edge modes and flat bands, by engineering strain.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Twist-deformation model is internally inconsistent: Eq. (7) modulates the vertical bond and not the slanted z-bond, contrary to Eq. (6); and λ = 0.2–0.5 gives expansion parameter ≫ 1 at the ribbon edge, so the twist flat-band claim is not established.","rationale":"The paper's central twist result depends on Eq. (7), but Eq. (7) does not follow from Eq. (6) for the stated displacement field: under the twist, the vertical bond has zero x-component and should be unaffected, while the two slanted bonds are strained. The printed modulation X_x = X_y, X_z = X is therefore not the model defined by the geometry. Even if the bond labels were merely swapped in the text, the numerical values λ = 0.2–0.5 used in Figs. 5–7 make the expansion parameter in Eq. (6) large at the upper edge of a 20-unit-cell ribbon, so the linearized bond modulation and the ferromagnetic ground state are not reliable. This does not necessarily invalidate the uniaxial results or the general nonlocal Hamiltonian (Eqs. 3–4), but it undermines the most novel claim. The reader's weakest_assumption pointed to scalar strain derivatives; the sharper issue is the bond assignment and the strain magnitude. A targeted rerun of the twist calculation settles it. Hence the conditional verdict is retained, subject to this specific check.","tokens_in":15698,"tokens_out":21622,"duration_ms":221222,"concrete_test":"Re-derive Eq. (7) from Eq. (6) for u = (0, y cos λx − y, y sin λx), first identifying which bonds have nonzero δᵢ,ₓ. Then rerun the twist diagonalization of Sec. IV B for Fig. 5c with (i) the corrected bond assignment (X_x = X_z = X − c(y² + a y/2), X_y = X) and (ii) parameters satisfying (3/4)λ² y_max² ≪ 1, e.g. λ = 0.05. If flat bands and v_m ≈ 0 across the BZ disappear in either case, the twist claim is an artifact of the printed bond modulation or of the large-strain regime. Additionally, check classical stability of the ferromagnetic state for the modulated couplings at the ribbon edge.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The most load-bearing problem is in the twist-strain calculation (Sec. IV B), which supports the flat-band/non-propagating edge-state claim. Eq. (6) gives the strained bond length as δ̃ᵢ² = δᵢ² + λ²δᵢ,ₓ²(y² + yδᵢ,ᵧ); with the lattice vectors of Fig. 1, δ_y = (0, −1) has δᵢ,ₓ = 0 and is unmodulated, while the two slanted bonds δ_x and δ_z are modulated. Eq. (7), however, states X_x = X_y = X − c(y² + a y/2), X_z = X, i.e. the vertical bond is modulated and one slanted bond is not. The numerical twist results therefore do not implement the model defined by Eq. (6) for the stated displacement field. Compounding this, the small-twist expansion is used at λ = 0.2–0.5 with y_max ≈ 29 (N_y = 20). The argument inside the square root of Eq. (6) is (3/4)λ²(y² + a y/2), about 25 at the upper edge for λ = 0.5, so Eq. (7) is not a small correction: local couplings change sign, and the assumed ferromagnetic ground state is likely unstable there. Reported flat bands and edge localization may be artifacts of this unphysical regime. Because the twist flat bands are the most novel qualitative claim, this inconsistency is load-bearing.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper develops a general linear spin-wave theory for weakly strained Heisenberg-Kitaev magnets, starting from a magnetoelastic coupling that modulates the Heisenberg, Kitaev, and anisotropic exchange terms according to the strain tensor. The resulting magnon Hamiltonian is nonlocal in momentum space for generic deformation fields and becomes local for displacement fields that are linear in position. The theory is then applied numerically to finite nanoribbons under two types of deformation: uniaxial strain, which is reported to shift edge-mode crossings and induce nonreciprocal in-gap magnon modes near the Brillouin zone boundary, and twist strain, which is reported to produce flat bands with non-propagative, edge-localized magnon states across the whole Brillouin zone. The central claims of the twist section rest on a particular assignment of which bonds are modulated by the twist, and this assignment is internally inconsistent with the derived bond-length formula.","tokens_in":16077,"tokens_out":7187,"duration_ms":68210,"significance":"If the uniaxial results stand, the paper provides a useful and self-contained framework for studying strained Heisenberg-Kitaev magnets, going beyond Dirac-point expansions and treating the full Brillouin zone. A clear strength is that the strain strengths are control parameters rather than fitted quantities, and the numerical diagonalization implements the stated model without circular fitting. The predicted strain-tunable edge-mode crossings and changes in magnon velocity are testable ingredients for future experiments. The twist flat-band claim, if correct, would be a genuinely novel qualitative effect. However, the twist analysis is undermined by an explicit contradiction between the bond-modulation formula (Eq. (6)) and the implementation used in the numerics (Eq. (7)), and by the use of a small-deformation expansion far outside its regime of validity. The nonreciprocity claim also needs sharpening because the presented velocity comparison is between opposite edges rather than opposite momenta. These issues are load-bearing for the two most distinctive conclusions of the paper.","major_comments":[{"comment":"The bond modulation in Eq. (7) contradicts the bond-length formula of Eq. (6). For the lattice vectors given in Fig. 1, δ_y = a0(0,−1) has δ_{y,x} = 0, so Eq. (6) predicts that the vertical y bond is unmodulated (δ̃_y = δ_y), while the two slanted bonds δ_x and δ_z acquire length changes. Eq. (7), however, states X_x = X_y = X − c(y² + (a/2)y) and X_z = X, which modulates the vertical bond and leaves one slanted bond unchanged. Since the numerical twist calculations implement Eq. (7), they do not realize the deformation field defined by Eq. (6). The flat bands and non-propagative edge states shown in Figs. 5–7 are therefore computed from a different model than the one derived, so the central twist-strain claim is not supported.","section":"§IV B, Eqs. (6)–(7)"},{"comment":"The small-twist expansion leading to Eq. (7) is uncontrolled for the parameter values used. With Ny = 20, the maximum vertical index is y_max = 29; the quantity q = (3/4)λ²(y² + (a/2)y) inside the square root of Eq. (6) is already about 26 at the upper edge for λ = 0.2 and exceeds 100 for λ = 0.5. The approximation exp(1 − √(1+q)) ≈ 1 − q/2, which underlies Eq. (7), is therefore invalid across most of the ribbon. Moreover, the linearized couplings X − c(y² + (a/2)y) become negative near the upper edge for the displayed λ values, indicating that the assumed ferromagnetic ground state is likely unstable in exactly the region where the flat-band modes localize. Thus the flat-band and controlled-localization results of Figs. 5–7 are not established within the stated weak-strain approximation.","section":"§IV B, Eq. (7), Figs. 5–7"},{"comment":"The claim that uniaxial strain makes magnons 'strongly non-reciprocal' is not demonstrated by the plotted group velocities. The solid and dashed curves in Fig. 4 correspond to the 'upper' and 'lower' edge states, which are localized at opposite edges of the ribbon. Nonreciprocity of an edge magnon means that a mode on a given edge has group velocity v(k) ≠ −v(−k); comparing two distinct edge modes at the same k is a comparison of opposite edges, not of opposite momenta. To support the nonreciprocity conclusion, the authors should present v(k) and v(−k) for each edge separately, or rephrase the result as a strain-induced asymmetry between the two edge-localized modes.","section":"§IV A, Fig. 4"},{"comment":"The paper repeatedly refers to the edge modes as 'topologically protected' in the presence of strain, but no topological invariant is computed for the strained system. For position-dependent strain, translation invariance is broken and the bulk Chern number is not immediately defined; the edge localization shown in Figs. 6 and 7 is necessary but not sufficient evidence of topological protection. The authors themselves state in Section V that 'a comprehensive study on topological invariants ... is needed to get further conclusions,' which is in tension with the earlier claim of topologically protected flat bands. This point should either be addressed by a suitable definition of topological protection in the strained system or by softening the claims.","section":"§IV B and §V"}],"minor_comments":[{"comment":"The caption reads 'λK = 085' for panel (b); this should presumably be 'λK = 0.8' with a decimal point.","section":"Fig. 5 caption"},{"comment":"The caption for panel (b) states 'λJ = λJ = 0.5'; the second entry should be λK, giving 'λJ = λK = 0.5'.","section":"Fig. 6 caption"},{"comment":"There are two word-level typos: 'apparition' in the abstract should be 'appearance', and 'Landu Levels' in the Conclusions should be 'Landau Levels'.","section":"Abstract and Conclusions"},{"comment":"The notation Xᵢⱼˡ = X − γⁱʲ_η uⁱʲ is not clearly defined: γⁱʲ_η and uⁱʲ should be spelled out explicitly, and the relation to the earlier c_X and h_η(r) notation of Eq. (2) should be made explicit.","section":"§IV A"},{"comment":"The uniaxial displacement field in the caption is written as 'u(r) = cχyA2ŷ', where the symbol A2 is not introduced in the text; the displacement field for the uniaxial case should be defined explicitly in Section IV A.","section":"Fig. 1 caption"}],"recommendation":"major_revision","confidential_remarks":"The paper contains a promising general derivation and a careful numerical implementation for the uniaxial case, but the twist-strain section—which supplies the most novel qualitative claims—has a direct inconsistency between the derived bond-length formula and the bond modulation implemented numerically, and it uses an expansion far outside its validity range. The nonreciprocity claim also appears to compare the wrong quantities. These are correctable within the manuscript's scope, so major revision rather than rejection seems appropriate. If the twist results are recomputed with the correct bond assignment and restricted to genuinely small λ, the flat-band conclusion may or may not survive."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here's the quick take. The general strained Heisenberg-Kitaev magnon Hamiltonian in Sec. III is genuinely useful — I don't think that expression appears elsewhere. The uniaxial strain part is largely believable. But the twist section, which carries the most novel claim (flat bands across the whole Brillouin zone), is internally inconsistent and runs the calculation where the small-twist expansion is not valid. The twist part needs to be redone before the headline result can be trusted.\n\nWhat is new: they strain-modulate the Kitaev coupling on the same footing as the Heisenberg one, over the full Brillouin zone, instead of expanding near the Dirac points. The observation that a weakly strained HK magnet has a momentum-nonlocal Hamiltonian, which becomes local for linear deformation fields, is clean and useful. The Landau-level structure under uniaxial strain is consistent with earlier work on strained honeycomb magnets. The literature coverage is good, and the authors are honest that they never computed topological invariants under strain.\n\nWhere it falls down. Most seriously, Eq. (6) and Eq. (7) describe different models. From the stated displacement field, the vertical bond (δ_y, with δ_x = 0) is unmodulated and the two slanted bonds are modulated. Eq. (7) does the reverse: X_x and X_y are modulated, X_z is not. The twist numerics implement Eq. (7), so the flat bands and non-propagating edge states don't correspond to the stated twist. That is load-bearing for the most novel claim.\n\nSecond, the small-twist expansion is used at λ = 0.2–0.5 with y up to about 29. The argument (3/4)λ²(y²+ay/2) is around 25 at the upper edge for λ = 0.5. Eq. (7) is not a small correction there; the linearized couplings change sign, so the assumed ferromagnetic ground state is likely unstable in that region. The reported flat bands and edge localization may be artifacts of that unphysical regime.\n\nThird, softer: the nonreciprocity claim for uniaxial strain compares the velocities of edge modes localized on opposite edges. Nonreciprocity normally means v(k) ≠ −v(−k) on the same branch; that is a different statement and should be reframed. The topological-protection language is also asserted rather than demonstrated — the authors concede this in the conclusions.\n\nBottom line: the Sec. III derivation is worth keeping and citing; the twist section as written is not reliable. Readers working on magnon straintronics or Heisenberg-Kitaev materials will want the general Hamiltonian. I would send this to peer review: a serious referee can push the authors to fix the bond assignment, restrict the calculation to a valid strain regime, and reframe the nonreciprocity. Conditional acceptance after major revision is the right outcome.","headline":"Useful general strained Heisenberg-Kitaev magnon Hamiltonian and plausible uniaxial results, but the twist flat-band claim rests on an inconsistency between Eq. (6) and Eq. (7) and on an invalid small-twist regime.","tokens_in":16585,"tokens_out":10337,"would_cite":true,"duration_ms":96294,"reading_group":"maybe","serious_thinker":"no","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Weak strain is enough to make Heisenberg-Kitaev magnon edge modes nonreciprocal, and twist strain can flatten them into edge-localized non-propagating bands across the Brillouin zone.","keywords":["Heisenberg-Kitaev magnets","magnons","strain engineering","topological edge modes","nonreciprocal magnons","flat bands","magnon Landau levels","honeycomb nanoribbons"],"falsifier":"Compute or measure the strain derivatives of the Heisenberg and Kitaev couplings in a candidate material and diagonalize the strained nanoribbon Hamiltonian: if the edge-mode crossing does not shift with the sign of the uniaxial strain, or if combined $J$-$K$ twist strain does not flatten the edge bands at the predicted strengths, the central claim is falsified.","tokens_in":15508,"feed_emoji":"🧲","tokens_out":11264,"duration_ms":103012,"temperature":0.7,"pith_summary":"This paper sets out to establish how mechanical strain changes the magnon excitations of a two-dimensional Heisenberg-Kitaev ferromagnet, a honeycomb magnet whose spin interactions combine ordinary exchange ($J$) with bond-dependent Kitaev coupling ($K$). The authors derive a weakly strained magnon Hamiltonian in which the symmetric strain tensor enters through deformation fields, and they show the Hamiltonian is nonlocal in momentum space but becomes local for deformation fields linear in position. In nanoribbon calculations, uniaxial strain makes the bulk magnon bands more dispersive and shifts the crossing of the topologically protected in-gap edge modes, making them nonreciprocal near the Brillouin-zone boundary. For twist strain, simultaneous modulation of $J$ and $K$ produces flat bands with non-propagating, edge-localized topological magnon states across the Brillouin zone. A careful reader would care because the paper suggests strain offers a practical control knob for magnon velocity, density of states, and localization without destroying topological protection.","feed_headline":"Strain twists topological magnons into flat, frozen edge bands","feed_subtitle":"Uniaxial strain shifts edge-mode crossings; twist strain with J and K modulation gives non-propagating edge states.","key_machinery":"The load-bearing object is the weak-strain expansion of each magnetic coupling, $X_{ij} \\approx X - c_X h_\\eta(r)$, where $h_\\eta(r) = \\delta^a_\\eta \\delta^b_\\eta u_{ab}(r)$ projects the symmetric strain tensor onto the bond direction $\\eta$ and $c_X$ is the strain derivative of that coupling. Inserting this expansion into a linear spin-wave magnon Hamiltonian generates the momentum-nonlocal blocks that carry all strain effects, including Kitaev-specific contributions that do not appear when only the exchange coupling is modulated. For displacement fields linear in position the Hamiltonian becomes local in momentum space and can be diagonalized exactly; for nanoribbons the bands are obtained numerically and read out through the spectral function. The machinery works by redistributing magnon states in momentum space rather than just shifting energies uniformly, acting like an elastic gauge field near the Dirac points.","core_discovery":"The central claim is that weak lattice deformations, through a linear modulation of both Heisenberg and Kitaev couplings, qualitatively retune the topological magnon spectrum of a honeycomb Heisenberg-Kitaev ferromagnet. The strained magnon Hamiltonian (Eqs. 3-4) couples magnons with different momenta through strain-dependent fields built from the symmetric strain tensor; this momentum nonlocality disappears for linear displacement fields, making the Hamiltonian local and in principle exactly diagonalizable for an infinite sample. Uniaxial strain ($u \\propto y^2 \\hat{y}$) leaves topological protection intact but shifts the crossing of the in-gap edge modes and induces nonreciprocal group velocities tuned by strain strength. Twist strain, with separate parameters $\\lambda_J$ and $\\lambda_K$, flattens the edge modes at the zone boundary when only the Kitaev coupling is strained, and produces flat bands across the entire Brillouin zone when both couplings are strained together; the associated edge magnons remain strongly localized at the sample ends. The paper claims this establishes a general strain-based route to controlling magnon nonreciprocity, flat bands, and edge-state localization in Heisenberg-Kitaev magnets.","pith_inferences":["Beyond the paper: the same linear-strain machinery should extend to other ordered phases of the Heisenberg-Kitaev model, such as zigzag or stripy order, after a local spin rotation, since the strain modulation enters the coupling constants rather than the magnetic order.","Beyond the paper: because the edge-mode crossing shifts with the sign of $c_J$ and $c_K$, reversing from compression to tension should reverse the nonreciprocity, offering a directional control for magnon transport that could be probed by inelastic light scattering before any full topological-invariant calculation.","Beyond the paper: the twist-induced flat bands with edge localization suggest a strain-tunable magnon flat-band platform; a natural next test is whether disorder or edge roughness preserves the localization at large twist strength, a question the paper leaves open.","Beyond the paper: since the strained Hamiltonian is nonlocal for general deformation fields, exact diagonalization for curved or nonlinear strain profiles might reveal momentum-mixing effects beyond Landau-level physics, a numerical extension the paper does not pursue."],"forward_implications":["Uniaxial strain shifts the crossing point of the two in-gap topological edge modes and makes their group velocities unequal near the Brillouin-zone boundary, with the imbalance controlled by the strain strength and sign.","Straining only the Kitaev coupling tends to flatten edge modes and push one mode toward hybridization with bulk bands, while straining only the Heisenberg coupling mainly changes band dispersion; topological protection keeps the mode from fully hybridizing by re-opening a gap.","Twist strain with both $J$ and $K$ modulated produces flat magnon bands over the entire Brillouin zone, yielding non-propagating edge states whose localization at the ribbon ends strengthens as the twist parameter increases.","The spectral function shows strain-induced magnon Landau levels and a strain-dependent redistribution of magnon weight between upper and lower bulk bands, indicating that strain controls the magnon density of states.","Because the localization and velocity of topological edge magnons survive under both strain types, the bulk-boundary correspondence can be tested under lattice deformation rather than only in pristine lattices."],"supporting_citations":[{"why":"Defines the topological magnon bands and Chern numbers of the unstrained ferromagnetic Heisenberg-Kitaev model that the strain results perturb.","marker":"[58]"},{"why":"Establishes topological magnons in a van der Waals Heisenberg-Kitaev candidate material, providing the material context for the model.","marker":"[50]"},{"why":"The earlier strain study of magnon topology in a Kitaev-type monolayer that this paper extends by isolating Heisenberg and Kitaev modulations.","marker":"[46]"},{"why":"Supplies the linear spin-wave transformation used to derive the bilinear magnon Hamiltonian from the spin model.","marker":"[70]"},{"why":"Supplies the diagonalization method for the bosonic magnon Hamiltonian used in the nanoribbon spectra.","marker":"[74]"},{"why":"Provides the spectral-function framework and twisted-honeycomb Landau level results that the spectral calculations build on.","marker":"[71]"},{"why":"Establishes magnon Landau levels in strained magnets, the phenomenon invoked for the dispersive bulk bands under uniaxial strain.","marker":"[39]"}],"fun_headline_variants":["Twist strain freezes magnon edge modes into flat bands","Uniaxial strain skews edge magnons; twist strain flattens them","Strain control: twist for flat magnon bands, stretch for nonreciprocal edges","Magnons obey strain: uniaxial steers, twist flattens","Topological magnons under strain: flat bands from twist, skewed edges from stretch"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The calculations assume each magnetic coupling responds to strain as a single slope multiplied by a smooth, small deformation field, with no higher-order or direction-dependent terms; if the actual Kitaev or Heisenberg couplings respond differently at the strains used, the predicted nonreciprocity and flat bands need not appear.","fun_headline_variants_meta":{"raw":{"variants":["Twist strain freezes magnon edge modes into flat bands","Uniaxial strain skews edge magnons; twist strain flattens them","Strain control: twist for flat magnon bands, stretch for nonreciprocal edges","Magnons obey strain: uniaxial steers, twist flattens","Topological magnons under strain: flat bands from twist, skewed edges from stretch"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000837,"raw_usage":{"total_tokens":3664,"prompt_tokens":970,"completion_tokens":2694,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":586,"completion_tokens_details":{"reasoning_tokens":2594}},"tokens_in":586,"tokens_out":2694,"duration_ms":20264,"temperature":1.0,"reasoning_tokens":2594,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T05:35:35.819321+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute or measure the strain derivatives of the Heisenberg and Kitaev couplings in a candidate material and diagonalize the strained nanoribbon Hamiltonian: if the edge-mode crossing does not shift with the sign of the uniaxial strain, or if combined $J$-$K$ twist strain does not flatten the edge bands at the predicted strengths, the central claim is falsified.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Defines the topological magnon bands and Chern numbers of the unstrained ferromagnetic Heisenberg-Kitaev model that the strain results perturb."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes topological magnons in a van der Waals Heisenberg-Kitaev candidate material, providing the material context for the model."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"The earlier strain study of magnon topology in a Kitaev-type monolayer that this paper extends by isolating Heisenberg and Kitaev modulations."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the linear spin-wave transformation used to derive the bilinear magnon Hamiltonian from the spin model."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the diagonalization method for the bosonic magnon Hamiltonian used in the nanoribbon spectra."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the spectral-function framework and twisted-honeycomb Landau level results that the spectral calculations build on."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes magnon Landau levels in strained magnets, the phenomenon invoked for the dispersive bulk bands under uniaxial strain."}],"review_version":1}