{"id":"73227aa0-c327-43b3-9a78-9fe4b03791db","arxiv_id":"2608.09644","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"The polar decomposition of the spin-dependent hopping bond, into a unitary phase part and a Hermitian amplitude part, unifies even-parity altermagnets and odd-parity p-wave magnets and predicts a new non-commuting sector with an even-in-momentum transverse spin polarization.","lead":"The paper proposes that all non-relativistic spin splitting in antiferromagnets can be understood through the polar decomposition of the quantum mechanical hopping bond between atomic sites. This provides a unified language for altermagnets and p-wave magnets, and predicts a new non-coplanar spin texture with a distinctive photoemission signature.","discovery_kind":"unification","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Even-transverse spin polarization is not unique to non-commuting bonds: a pure Hermitian altermagnet with in-plane amplitude axis (U=σ0, p⊥z) also gives ⟨s⊥⟩even≠0, contradicting Eq. (32) as stated.","rationale":"The reader's weakest_assumption focused on the model restrictions (reciprocal bonds, bipartite lattice, uniform collinear exchange), but their rationale separately noted that the iff criterion is proven only for the minimal bond pattern and that a pure Hermitian bond with an in-plane axis would generate even-transverse texture. Our concern sharpens that point into an explicit counterexample within the paper's own formalism, making it the most load-bearing issue: it directly undermines the advertised spectroscopic fingerprint, a central novelty claim, rather than merely delimiting the exact-solvable regime. The algebraic polar decomposition and the exact block factorization remain correct for the stated model, and the counterexample is addressable by narrowing the fingerprint claim to Hermitian axes along the Néel vector or by adding a complementary observable (e.g., simultaneous longitudinal/transverse components or band asymmetry). Therefore the paper merits conditional acceptance with revisions, matching the reader's verdict. We partially agree with the reader because they identified the fingerprint risk but did not present the concrete in-plane Hermitian counterexample or elevate it above the model-assumption concern.","tokens_in":26694,"tokens_out":16178,"duration_ms":142205,"concrete_test":"Add a control model to the Methods verification: set Tx=e^{β x̂·σ}, Ty=e^{-β x̂·σ}, and Uδ=σ0, with parameters β=0.5, Δ/t=2. Compute S(k)=Tx e^{ikx}+Tx† e^{-ikx}+Ty e^{iky}+Ty† e^{-iky}, extract h(k), use Eq. (20) for the s=+1 lower band, decompose ⟨s_x⟩(k) into even/odd parts via Eq. (31), and report max_k |⟨s_x⟩even|. If this is nonzero, Eq. (32) as a general fingerprint is falsified. Repeat with p̂=ŷ to confirm the effect is generic for in-plane Hermitian axes.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's flagship observable is the iff criterion of Eq. (32): ⟨s⊥⟩even(k)≠0 ⇔ [Uδ,Pδ]≠0, advertised as a unique spin-ARPES fingerprint of the non-commuting sector. This is false as stated. Take a pure Hermitian bond pattern with Uδ=σ0 and Pδ=e^{β x̂·σ} (in-plane amplitude axis), e.g., P_x=e^{β x̂·σ}, P_y=e^{-β x̂·σ}. This satisfies the reciprocal condition T_{-δ}=Tδ†, so it lies inside the paper's exact-solvable family, and it is a bond-structured altermagnet (Eq. (24) with p̂=x̂). The Bloch field is h(k)=2t sinhβ (cos kx − cos ky) x̂, even in k and purely transverse to the Néel axis ẑ. Inserting into the sector-resolved texture Eq. (20), the lower-band projection ⟨s_x⟩(k) is even under k→−k and nonzero, so ⟨s⊥⟩even≠0 while [U,P]=0. The paper's numerical verification only tests p̂=ẑ, so it misses this case. The Discussion's claim that even-transverse polarization is 'strictly forbidden in pure altermagnets' is therefore too broad: a pure altermagnet with in-plane spin-amplitude axis produces the same fingerprint as the non-commuting sector.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper introduces a spin-bond theory of non-relativistic spin splitting in compensated magnets. The central object is the spin-dependent hopping matrix T_δ, decomposed as T_δ = U_δ P_δ into a unitary spin-phase factor and a positive Hermitian spin-amplitude factor. Under assumed conditions (bipartite lattice, inter-sublattice hopping only, uniform collinear exchange, and reciprocal bonds T_{-δ}=T_δ†), the Bloch Hamiltonian is shown to factorize into two 2x2 blocks. The unitary sector is claimed to generate odd-in-momentum p-wave/SOC-like textures, the Hermitian sector even-in-momentum Γ-split/altermagnetic textures, and the non-commuting sector [U_δ,P_δ]≠0 to generate a non-coplanar texture with a distinctive even-in-momentum transverse spin polarization ⟨s⊥⟩even(k)≠0. This fingerprint, Eq. (32), is advertised as a unique spin-ARPES signature of non-commuting spin bonds. The authors also derive explicit limits (Rashba, Dresselhaus, p-wave, d-altermagnet, Γ-split), discuss crossovers, and estimate consequences for EDSR-based spin qubits.","tokens_in":26858,"tokens_out":9004,"duration_ms":93559,"significance":"If the central classification and fingerprint claim were correct as stated, the paper would provide a useful organizing principle: an exact two-block solution, explicit parity locking of unitary versus Hermitian bonds, and analytic textures for all known non-relativistic spin-split limits. The derivations in the appendices are explicit and the numerical control models are clearly described. However, the flagship observable claim—that even-in-momentum transverse spin polarization is unique to the non-commuting sector—is false as stated: a purely Hermitian altermagnet with an in-plane amplitude axis produces the same type of signal. This is a load-bearing issue for the advertised spin-ARPES fingerprint, requiring a substantive revision of Eq. (32) and the related Discussion statements.","major_comments":[{"comment":"The iff criterion ⟨s⊥⟩even(k)≠0 ⇔ [U_δ,P_δ]≠0 is false as stated. Consider a purely Hermitian bond pattern on the square lattice with T_x=e^{βσ_x}, T_y=e^{-βσ_x}, and U_x=U_y=σ0. This pattern satisfies T_{-δ}=T_δ†, so it lies inside the exact-solvable family of the section 'Exact solution on bipartite lattices'. The Bloch field is h(k)=2t sinhβ (cos k_x − cos k_y) x̂, which is even in k and transverse to the exchange axis ẑ. From the sector-resolved texture in Eq. (20), the lower-band polarization ⟨s_x⟩(k) is even under k→−k and nonzero, while [U_δ,P_δ]=0. The Methods control tests check only Hermitian bonds with amplitude axis p̂=ẑ, so they miss this case. This contradicts the abstract's and Discussion's claims that even-in-momentum transverse polarization is strictly forbidden in pure altermagnets and is a unique fingerprint of the non-commuting sector. If 'transverse' is instead intended to mean 'along the specific axis u×p', that definition must be stated explicitly and the criterion restricted accordingly; a pure Hermitian bond has no intrinsic u×p axis, so the advertised spin-ARPES test is not well defined in that case.","section":"Results, Eq. (32); Methods, 'Verification against control models'; Discussion"}],"minor_comments":[{"comment":"The approximation θspin(k) ≃ arctan(t sinα/Δ sinθ) is ambiguous: the θ dependence should be written as arctan[(t sinα/Δ) sinθ] (or the equivalent explicit form), since the preceding text states that the in-plane component grows as sinθ.","section":"Results, Eq. (22)"},{"comment":"The caption states that in panels (a) and (b) 'h⊥ = 0' for the p-wave and altermagnet cases; this is true only for the specific axes shown (p̂=ẑ, û=ẑ). The caption should clarify that the vanishing transverse component is a statement about the chosen bond axes, not a general property of the unitary or Hermitian sectors.","section":"Fig. 3 caption"},{"comment":"The numerical verification should include a pure Hermitian control with in-plane amplitude axis (for example, p̂=x̂) in addition to the p̂=ẑ case, since this is precisely the configuration that tests the claimed uniqueness of the non-commuting fingerprint.","section":"Methods, 'Verification against control models'"},{"comment":"The manuscript relies heavily on the assumptions (i)–(iv) before Eq. (18); Appendix G explicitly notes that canted or staggered exchange breaks the exact factorization. The abstract and the opening of the Discussion should state this scope condition more prominently to avoid overgeneralization.","section":"General presentation"}],"recommendation":"major_revision","confidential_remarks":"The central spin-ARPES fingerprint claim needs a substantive rewrite: the counterexample with an in-plane Hermitian amplitude axis is within the paper's own exact-solvable family and invalidates Eq. (32) as written. The rest of the framework—the polar decomposition, parity locking, and the exact block factorization—appears sound and would be a useful contribution once the fingerprint statement is properly qualified or reformulated. No concerns about misconduct or novelty disclosure beyond the usual need to distinguish the authors' classification from existing symmetry-based classifications."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe spin-bond polar decomposition is a good idea, and the paper executes it carefully within its stated assumptions. The unitary/Hermitian split with parity locking is derived cleanly, and the exact block factorization on reciprocal bipartite lattices is a genuine technical contribution. The control checks—pure altermagnet, p-wave, Rashba—vanish in the even-transverse channel, which gives real support for the mixed sector's existence in the minimal model. That part deserves credit.\n\nThe problem is the fingerprint claim. Eq. (32) asserts ⟨s⊥⟩even(k)≠0 iff [Uδ,Pδ]≠0, and the abstract and discussion promote this as a unique spin-ARPES signature 'strictly forbidden in pure altermagnets.' That is too broad. A pure Hermitian bond pattern with an in-plane amplitude axis—say P_x=e^{β x̂·σ}, P_y=e^{-β x̂·σ}—satisfies the reciprocal condition and lies inside the exact-solvable family. The resulting field h(k)=2t sinhβ (cos kx − cos ky) x̂ is even, transverse to the Néel axis, and gives a nonzero ⟨s⊥⟩even with [U,P]=0. So the criterion only holds for the specific minimal pattern tested (p̂ along ẑ). The numerical verification doesn't cover the in-plane Hermitian case. This is fixable by narrowing the claim to the minimal model or proving conditions for uniqueness, but as stated it's wrong.\n\nSecondary issue: the novelty boundary with Ref. [41] is not delineated. The text acknowledges that mixed-parity and non-abelian altermagnetic states were proposed before; the abstract's 'uncovers' needs to be reconciled with that. A clear comparison of what the polar-decomposition classification adds would resolve it.\n\nThe quantum-computing section is speculative and could be trimmed. The exact derivations, the microscopic realization in App. C, and the generator expansion in App. H are solid and well worth keeping.\n\nWho should read this: anyone working on altermagnet classification, exchange-generated spin-orbit textures, or compensated spin-split magnets. The framework is useful once the fingerprint claim is corrected. I'd send it to a serious referee, not desk-reject, but the referee should demand a revision that narrows Eq. (32) and states the overlap with Ref. [41] honestly.","headline":"Useful spin-bond classification, but the even-transverse spin fingerprint is not unique to non-commuting bonds as claimed.","tokens_in":27515,"tokens_out":4294,"would_cite":true,"duration_ms":35647,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"One spin-bond algebra unifies all spin-split magnet classes","keywords":["spin-bond theory","polar decomposition","non-relativistic spin splitting","altermagnetism","p-wave magnetism","emergent spin-orbit coupling","spin-ARPES","compensated magnets"],"falsifier":"Perform spin-ARPES with the detection axis along the predicted emergent direction $\\hat{u}\\times\\hat{p}$ on a compensated magnet whose bond construction has non-collinear unitary and Hermitian axes; observing an identically zero even-in-momentum transverse polarization while even and odd longitudinal components are present would contradict the paper's criterion $\\langle s_\\perp\\rangle_{\\rm even}(k)\\neq0\\Leftrightarrow[U_\\delta,P_\\delta]\\neq0$. A first-principles band calculation for a specific candidate material with non-commuting bonds could settle the same point numerically.","tokens_in":26351,"feed_emoji":"🧲","tokens_out":9612,"duration_ms":76782,"temperature":0.7,"pith_summary":"Non-relativistic spin splitting in compensated magnets—altermagnets, $p$-wave magnets, and exchange-generated spin-orbit-like textures—has been treated as separate phenomena. This paper argues that all of them are facets of one microscopic object: the spin-dependent hopping bond, whose polar decomposition $T_\\delta=U_\\delta P_\\delta$ splits it into a unitary spin phase and a Hermitian spin amplitude. That split fixes the momentum parity of the spin texture: unitary bonds give odd-in-momentum $p$-wave and emergent spin-orbit textures, Hermitian bonds give even-in-momentum $\\Gamma$-split and altermagnetic textures. When the two factors do not commute, a third, mixed sector appears with a non-coplanar texture and an even-in-momentum transverse spin polarization that none of the pure classes can produce. This matters because it supplies a single algebraic classification with a direct experimental fingerprint, and because the synthetic spin-orbit coupling it predicts is geometrically tunable, pointing toward field-free spin-qubit control.","feed_headline":"One spin-bond algebra unifies all spin-split magnet classes","feed_subtitle":"A polar decomposition of the hopping bond predicts a new mixed sector that spin-ARPES can detect.","key_machinery":"The central object is the spin-bond operator $T_\\delta$, the $2\\times2$ spin matrix an electron experiences hopping along bond $\\delta$. Its polar decomposition $T_\\delta=t_\\delta e^{i\\phi_\\delta}U_\\delta P_\\delta$ separates the unitary spin phase $U_\\delta=e^{i\\alpha_\\delta\\hat{u}_\\delta\\cdot\\sigma}$ from the Hermitian spin amplitude $P_\\delta=e^{\\beta_\\delta\\hat{p}_\\delta\\cdot\\sigma}$. With reciprocal bonds $T_{-\\delta}=T_\\delta^\\dagger$, the Bloch hopping matrix $S(k)$ is Hermitian and the Hamiltonian splits into two independent two-level blocks labelled by sublattice parity $s=\\pm1$; each block has effective field $\\Delta\\hat{z}-s h(k)$, whose competition with exchange dictates the spin direction. The non-commutator $[U_\\delta,P_\\delta]\\sim-2\\alpha_\\delta\\beta_\\delta(\\hat{u}_\\delta\\times\\hat{p}_\\delta)\\cdot\\sigma$ is the seed of the mixed sector, and the invariant $\\chi_{\\rm mix,\\delta}=|a_\\delta\\times b_\\delta|\\simeq\\alpha_\\delta\\beta_\\delta|\\hat{u}_\\delta\\times\\hat{p}_\\delta|$ measures its local strength. This machinery converts a symmetry classification into a calculation of bands and spin textures from a single bond-level input.","core_discovery":"The central claim is that the spin-dependent bond operator $T_\\delta$, the $2\\times2$ spin matrix an electron experiences when hopping along bond $\\delta$, can be written as $T_\\delta=U_\\delta P_\\delta$, with $U_\\delta$ unitary (spin phase) and $P_\\delta$ positive Hermitian (spin amplitude). On reciprocal bipartite lattices with uniform collinear exchange, the Bloch Hamiltonian factorizes exactly into sublattice-parity sectors $s=\\pm1$, $H_s(k)=-s h_0(k)\\sigma_0+[\\Delta\\hat{z}-s h(k)]\\cdot\\sigma$, so the spin texture is set by the competition between uniform exchange $\\Delta\\hat{z}$ and the momentum-dependent bond field $h(k)$. Parity is locked to the polar character: unitary links produce sine form factors, Hermitian links cosine form factors. If $[U_\\delta,P_\\delta]\\neq0$, the bond cannot be diagonalized on one spin axis and the texture becomes non-coplanar, with the minimal-model fingerprint $\\langle s_\\perp\\rangle_{\\rm even}(k)\\neq0$ along $\\hat{u}_\\delta\\times\\hat{p}_\\delta$, a combination absent in pure altermagnets, pure $p$-wave magnets, and Rashba/Dresselhaus textures. The paper presents this as a unification of known non-relativistic spin-split phases and the identification of a previously unrecognized mixed sector.","pith_inferences":["The classification is exactly proven only under the stated reciprocal-bipartite uniform-exchange assumptions; extending it to non-reciprocal bonds or canted exchange would require a generalized analysis, though the polar decomposition may still organize those cases.","The even-in-momentum transverse polarization could be used as a screening observable in first-principles searches: compensated magnets with non-collinear bond axes should be checked for this component before being assigned to a pure class.","Beyond spin-ARPES, strain-dependent spin-orbit torque or magnetotransport measurements could test the predicted on/off switching of synthetic spin-orbit coupling in a single device.","For quantum dots in materials like MnTe, the theory indicates sub-nanosecond EDSR gate times at a few percent strain, but whether switching the synthetic spin-orbit coupling off improves charge-noise coherence requires a separate operating-point analysis, as the paper itself notes."],"forward_implications":["Every non-relativistic spin-split phase of a compensated magnet belongs to one of three sectors: unitary (odd), Hermitian (even), or mixed non-commuting.","Altermagnetism appears as the bond-structured Hermitian limit, while $p$-wave magnets and exchange Rashba, Dresselhaus, radial, and out-of-plane textures appear as unitary limits.","Continuous rotation of bond axes interpolates between classes, and any interpolation with non-parallel spin axes necessarily passes through the mixed non-coplanar sector.","Spin-ARPES measuring the even-in-momentum transverse polarization along $\\hat{u}\\times\\hat{p}$ can distinguish the mixed sector from every pure class.","The synthetic spin-orbit coupling scales as $\\lambda_{\\rm SOC}=2t\\alpha\\beta|\\hat{u}\\times\\hat{p}|$, is tunable by strain, and can in principle be switched on and off, supporting field-free EDSR qubit control."],"supporting_citations":[{"why":"Defines altermagnetism and its even-parity non-relativistic spin splitting, the Hermitian-sector target the theory must reproduce.","marker":"[1–6]"},{"why":"Defines p-wave magnets with odd-parity non-relativistic spin splitting, the unitary-sector target.","marker":"[7,8]"},{"why":"Reports nonrelativistic spin splitting at the Brillouin-zone centre, the uniform Gamma-split limit.","marker":"[38]"},{"why":"Proposes mixed-parity states bridging even and odd regimes, the context for the mixed non-commuting sector.","marker":"[41]"},{"why":"Supplies the spin-ARPES techniques and measurements (including MnTe) used to detect band-resolved spin textures.","marker":"[15–18]"},{"why":"Motivates field-free spin qubits and EDSR control in altermagnets, the application for tunable synthetic SOC.","marker":"[19–23]"}],"fun_headline_variants":["Spin-bond theory unifies altermagnets and p-wave magnets","One spin-bond algebra predicts a new non-coplanar texture","Unified spin-bond algebra exposes tunable spin-orbit coupling","Quantum bonds: unify spin splitting, tune spin-orbit","New mixed spin-bond sector detectable by spin-ARPES"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The exact results rest on three structural assumptions: the hopping bond is reciprocal ($T_{-\\delta}=T_\\delta^\\dagger$), the lattice is bipartite with only inter-sublattice hopping, and the exchange is uniform and collinear; if any of these fail, the factorization, the closed-form bands, and the clean criterion $\\langle s_\\perp\\rangle_{\\rm even}\\neq0\\Leftrightarrow[U_\\delta,P_\\delta]\\neq0$ are not established.","fun_headline_variants_meta":{"raw":{"variants":["Spin-bond theory unifies altermagnets and p-wave magnets","One spin-bond algebra predicts a new non-coplanar texture","Unified spin-bond algebra exposes tunable spin-orbit coupling","Quantum bonds: unify spin splitting, tune spin-orbit","New mixed spin-bond sector detectable by spin-ARPES"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000864,"raw_usage":{"total_tokens":3815,"prompt_tokens":1079,"completion_tokens":2736,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":695,"completion_tokens_details":{"reasoning_tokens":2650}},"tokens_in":695,"tokens_out":2736,"duration_ms":19698,"temperature":1.0,"reasoning_tokens":2650,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T13:31:34.979196+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Perform spin-ARPES with the detection axis along the predicted emergent direction $\\hat{u}\\times\\hat{p}$ on a compensated magnet whose bond construction has non-collinear unitary and Hermitian axes; observing an identically zero even-in-momentum transverse polarization while even and odd longitudinal components are present would contradict the paper's criterion $\\langle s_\\perp\\rangle_{\\rm even}(k)\\neq0\\Leftrightarrow[U_\\delta,P_\\delta]\\neq0$. A first-principles band calculation for a specific candidate material with non-commuting bonds could settle the same point numerically.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Reports nonrelativistic spin splitting at the Brillouin-zone centre, the uniform Gamma-split limit."}],"review_version":1}