{"id":"be228c70-a275-4717-b668-0869521e035b","arxiv_id":"1908.08680","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Collinear antiferromagnetic order alone can produce momentum-dependent spin splitting without atomic spin-orbit coupling when the magnetic pattern and the inter-site hopping share the same symmetry representation.","lead":"This paper shows that collinear antiferromagnets can split the spin-up and spin-down electronic bands even when atomic spin-orbit coupling is negligible, by generating an effective spin-orbit coupling from the magnetic order and the lattice geometry. It supplies a symmetry-based classification over the 32 crystallographic point groups, which can guide searches for light-element spintronic materials.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Irrep condition is not sufficient: the paper's own t^(2)=0 tetragonal example satisfies the stated condition via T_x T_y but shows no spin splitting, so the 32-point-group classification overpredicts for bipartite lattices.","rationale":"The reader's weakest_assumption concerned the negligible-SOC limit; that is a legitimate scope restriction and not the main weakness. The more load-bearing issue is internal to the zero-SOC framework: the central selection rule stated in Sec. III B and summarized in the abstract is not sufficient, as the paper's own t^(2)=0 tetragonal calculation demonstrates. Appendix A proves that chiral symmetry in bipartite systems forbids spin splitting, but the general 32-point-group classification (Tables V and VI) is organized by irrep alone and does not mark these forbidden cases. This matters because the paper's headline contribution is the systematic point-group correspondence, and the material predictions depend on it. The pyrochlore and non-bipartite tetragonal examples are likely correct and the mechanism is real, so the paper should not be rejected; however, the claim of a complete correspondence for 32 point groups needs revision: the condition should be restated as a nonzero effective coupling in the perturbation series, and the tables should flag bipartite/chiral-symmetry exclusions. This supports the reader's CONDITIONAL verdict, though for a different and more central reason than the SOC caveat.","tokens_in":16571,"tokens_out":8220,"duration_ms":92753,"concrete_test":"Diagonalize the full 8x8 Hamiltonian of the tetragonal model of Sec. III B at a generic k, e.g., k = (0.3 pi, 0.7 pi), with parameters t^(1)_a = 1.1, t^(1)_b = 0.8, t^(2)_a = t^(2)_b = 0, h_xy = 0.5, and confirm that up- and down-spin bands are exactly degenerate even though T_x^(1b)(k)T_y^(1b)(k) ~ sin k_x sin k_y is B2. If the degeneracy holds, then the irrep condition alone is not sufficient, and Tables V and VI need a qualifier excluding bipartite lattices (or requiring sufficiently long-range hoppings) before they can be used as a material guide.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"In Sec. III B the paper states the necessary condition for k-dependent spin splitting: the hopping multipole Q^(n)(k) or a higher-order product such as T_x^(1)(k)T_y^(1)(k) must belong to the same irrep as the mean-field multipole Q^(0). However, in the same section the model with t^(2)_a = t^(2)_b = 0 and finite t^(1)_b has T_x^(1b)T_y^(1b) ~ sin k_x sin k_y, which is B2 under C4v, the same irrep as h_xy, yet the bands are spin-degenerate (dashed lines in Fig. 2(d)). The paper attributes this to the chiral symmetry of the bipartite lattice (Appendix A), which proves that no spin splitting occurs in any bipartite system. Thus the stated condition is insufficient, and the general Tables V and VI, which list spin-splitting types purely by irrep, do not encode the bipartite/chiral-symmetry exclusion. As a result, the claimed correspondence between ordering patterns and spin splitting for 32 point groups is incomplete: a collinear AFM on a bipartite lattice with only nearest-neighbor (odd) hoppings will show no splitting even when the irrep condition is satisfied. The candidate material list includes simple bipartite systems (e.g., MnF2, FeBO3) where this caveat is directly relevant, so the classification can overpredict the effect. The paper acknowledges the chiral-symmetry mechanism only in the specific tetragonal example and does not revise the general statement or tables accordingly.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript develops a multipole-based symmetry analysis of momentum-dependent spin splitting in collinear antiferromagnets without atomic spin-orbit coupling. After establishing that spin-rotation symmetry restricts the splitting to even-in-k (symmetric) forms, the authors show in two tight-binding models (pyrochlore and tetragonal) that a collinear AFM mean field can generate such a splitting via hopping multipoles in the same irreducible representation. They then propose a classification of second- and higher-order spin-splitting types for the 32 crystallographic point groups and discuss consequences for spin-current generation and piezomagnetic responses.","tokens_in":16919,"tokens_out":8078,"duration_ms":83885,"significance":"The paper is potentially useful and, for the worked examples, clean: the pyrochlore and tetragonal band calculations are explicit, Appendix A gives a simple proof of the chiral-symmetry exclusion, and the multipole tables are a convenient reference. The central idea—that anisotropic hopping combined with collinear AFM order can mimic spin-orbit coupling—is timely and relevant. The main weakness is that the general point-group tables are presented as a correspondence between ordering irrep and spin splitting without incorporating the bipartite chiral-symmetry exclusion demonstrated in the authors' own Appendix A; as a result the tables can overpredict spin splitting for important material classes.","major_comments":[{"comment":"The condition stated in Sec. III B ('the hopping matrix ... belongs to the same irrep. Γ of the MF multipole') is presented as the necessary condition, and the paper is careful to call it necessary; nevertheless the surrounding presentation and Tables V and VI invite a sufficiency reading, and the authors' own example shows the condition is not sufficient. In the tetragonal model with t^(2)_a = t^(2)_b = 0 and finite t^(1)_b, Table IV gives T_x^(1b)T_y^(1b) ~ sin k_x sin k_y, which is B2 under C4v, the same irrep as h_xy; yet Fig. 2(d) shows spin-degenerate bands. Appendix A proves the general reason: any bipartite lattice with opposite sublattice spin polarization has chiral symmetry and hence no spin splitting. Tables V and VI list the multipole/irrep combinations that give spin-split forms without encoding this bipartite exclusion. Consequently the classification overpredicts for bipartite materials, including several in the candidate list (MnF2, FeBO3). Please amend the tables and text to state explicitly that the irrep condition is not sufficient, and incorporate the chiral-symmetry exclusion into the general classification.","section":"Sec. III B; Appendix A; Tables V and VI"},{"comment":"Section III C introduces Tables V and VI with 'Similar analysis can be straightforwardly extended to any other point groups' and does not provide the derivation or a proof that the listed irreps are actually associated with nonzero spin splitting in a lattice model. Since the irrep condition is only necessary, the tables cannot be read as a constructive classification without additional conditions. I request either a derivation of the tables (or a reference to a complete derivation), or an explicit statement of the assumptions under which the SS column is realized, together with the bipartite caveat.","section":"Sec. III C"}],"minor_comments":[{"comment":"The text 'Sec. ??he discussion is generalized' appears to be a typesetting error; it should likely read 'Sec. III C.'","section":"Sec. III B"},{"comment":"Equations (B9) and (B10) contain denominators such as Q'^(1)_0(k) and [Q'^(1)_0(k) ± Q^(2)_0(k)]^2; the validity of the perturbative expansion at parameter values where these vanish should be stated.","section":"Appendix B"},{"comment":"The caption should define the SS column explicitly as the form of ε_σ(k) − ε_{−σ}(k) for a given irrep, and note that it lists allowed forms, not guaranteed splittings.","section":"Tables V and VI"}],"recommendation":"major_revision","confidential_remarks":"The paper's central examples are sound, but the unqualified nature of Tables V and VI is the main obstacle. I would not reject: the issue can be addressed by reframing the classification as necessary-only and adding the bipartite exclusion. The lack of derivation for the point-group tables should also be remedied at least by a sketch or by a clear statement of the assumed sufficient conditions."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Read the paper with the stress-test in hand. The test lands. The paper's central claim—same irrep between hopping multipole and MF multipole—is presented as the condition, but Appendix A's chiral-symmetry argument shows it is not sufficient. In the tetragonal example, the T_x T_y product ~ sin kx sin ky sits in B2, same as h_xy, yet the bands are spin degenerate when t2=0 because the bipartite lattice enforces ε_σ(k)=ε_−σ(k). That is the paper's own result, so the general tables overpredict for any collinear AFM on a bipartite lattice whose hoppings are only odd (nearest-neighbor). MnF2 and FeBO3 in the candidate list are exactly that kind of system. This is a genuine gap, though not a fatal one: the symmetry machinery is sound and the explicit models are clean.\n\nWhat is actually new: a systematic classification of symmetric k-dependent spin splitting for 32 point groups, and a demonstration that collinear AFM order can induce effective spin-orbit coupling without atomic SOC through anisotropic hopping in sublattice clusters. The pyrochlore and tetragonal tight-binding calculations are transparent, use no fitted parameters, and the multipole condition is checked against explicit band structures. Appendix B's perturbation derivation shows phase-convention independence. That is real and reproducible.\n\nSoft spots beyond the bipartite issue: Tables V and VI appear without derivation; saying 'similar analysis can be straightforwardly extended' is not a proof, and for a classification paper a reader should not have to take the table on faith. The zero-SOC restriction is stated up front, so not a flaw, but it means the material list is qualitative—real SOC will add antisymmetric terms and can change the spin polarization. The paper says as much, only briefly.\n\nOverall, this deserves serious refereeing. It would be accepted after revision if the authors (1) state the irrep condition as necessary but not sufficient, (2) add the bipartite/chiral exclusion to the general classification, and (3) qualify the candidate materials accordingly. I would cite it for the symmetry tables and the demonstration.","headline":"Useful symmetry classification of spin splitting in collinear AFMs without SOC, but the central 'necessary condition' is only necessary: bipartite lattices with odd hoppings satisfy it and still show no splitting, so the 32-point-group tables need a caveat.","tokens_in":17388,"tokens_out":2741,"would_cite":true,"duration_ms":27518,"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":"Collinear antiferromagnetic order alone can generate momentum-dependent spin splitting without atomic spin-orbit coupling.","keywords":["collinear antiferromagnet","spin splitting","effective spin-orbit interaction","multipole expansion","irreducible representation","spin current","piezomagnetic effect","tight-binding model"],"falsifier":"Take any collinear antiferromagnet with negligible atomic spin-orbit coupling and a known ordered multipole $Q^{(0)}_{\\Gamma\\gamma}$; compute the tight-binding hopping multipoles of its crystal structure and compare irreducible representations. If the condition predicts a splitting of a given form (say $k_xk_y\\sigma$) but spin-resolved angle-resolved photoemission or quantum-oscillation measurements show no such even-in-$\\mathbf{k}$ spin splitting, the central claim fails; conversely, observing spin splitting in a collinear antiferromagnet whose hopping multipoles do not share an irrep with the order would also disprove it.","tokens_in":16418,"feed_emoji":"🧲","tokens_out":14043,"duration_ms":122827,"temperature":0.7,"pith_summary":"This paper aims to show that momentum-dependent spin splitting of electronic bands does not require atomic spin-orbit coupling. It argues that a collinear antiferromagnetic order in a crystal with several sublattice sites acts as an effective spin-orbit interaction through the anisotropic hopping between sites, producing band dispersions of the form $\\varepsilon_\\sigma(\\mathbf{k})=\\sum_{\\Gamma\\gamma} X_{\\Gamma\\gamma} Q_{\\Gamma\\gamma}(\\mathbf{k})\\sigma$ that are even in $\\mathbf{k}$. The central criterion is an irreducible-representation match: a hopping multipole $Q^{(n)}_{\\Gamma\\gamma}(\\mathbf{k})$ (or a product such as $T_x^{(1)}(\\mathbf{k})T_y^{(1)}(\\mathbf{k})$) must belong to the same representation $\\Gamma$ as the multipole $Q^{(0)}_{\\Gamma\\gamma}$ created by the magnetic order. If true, this opens a route to spin-orbit-like physics in light-element antiferromagnets, including spin-current generation by charge or thermal currents and strain-induced magnetization.","feed_headline":"Collinear antiferromagnets split spin bands without atomic spin-orbit","feed_subtitle":"A symmetry condition shows when magnetic order alone creates spin-orbit-type band splitting.","key_machinery":"The load-bearing mechanism is the microscopic multipole decomposition of a tight-binding Hamiltonian on a cluster of magnetic sites. The ordered moments are represented by onsite electric multipoles $Q^{(0)}_{\\Gamma\\gamma}$ (with Pauli-matrix products $\\rho_\\mu,\\tau_\\nu$ encoding the sublattice degrees of freedom), and the kinetic part by wave-vector-dependent bond multipoles $Q^{(n)}_{\\Gamma\\gamma}(\\mathbf{k})$ of the $n$-th neighbor hopping plus magnetic toroidal multipoles $T^{(n)}_{\\Gamma\\gamma}(\\mathbf{k})$ (odd-parity, time-reversal-odd bond objects). Because interactions in the Hamiltonian couple only objects in the same irreducible representation, the multiplication of bond multipoles effectively generates the conjugate field of the ordering multipole; Appendix B shows, for the tetragonal model, how products of toroidal hopping terms produce an effective quadrupole $\\tilde{Q}^{(0)}_{xy}$ of order $k_xk_y$ even when no direct quadrupole hopping exists. The classification tables for 32 point groups then list which ordering irreps activate which symmetric spin-splitting polynomials, up to sixth order in $\\mathbf{k}$.","core_discovery":"The central claim is that spin-split bands can appear in collinear antiferromagnets even when atomic spin-orbit coupling is negligible and the Hamiltonian has exact SU(2) spin-rotation symmetry. In that limit, time reversal combined with a spin rotation enforces $\\varepsilon_\\sigma(\\mathbf{k})=\\varepsilon_\\sigma(-\\mathbf{k})$, so the only possible spin splitting is symmetric in $\\mathbf{k}$, expressed by even-parity electric multipoles in momentum space: $\\varepsilon_\\sigma(\\mathbf{k})=\\sum_{\\Gamma\\gamma}X_{\\Gamma\\gamma}Q_{\\Gamma\\gamma}(\\mathbf{k})\\sigma$. The paper identifies the microscopic condition for a nonzero anisotropic field $X_{\\Gamma\\gamma}$: in a cluster tight-binding model, the hopping part contains bond multipoles $Q^{(n)}_{\\Gamma\\gamma}(\\mathbf{k})$ and magnetic toroidal multipoles $T^{(n)}_{\\Gamma\\gamma}(\\mathbf{k})$, and the collinear order creates an onsite mean-field multipole $Q^{(0)}_{\\Gamma\\gamma}$; when a hopping multipole, or a higher-order product of toroidal multipoles, transforms as the same irreducible representation of the point group as the mean-field multipole, the order activates a k-dependent spin splitting of the form $Q_{\\Gamma\\gamma}(\\mathbf{k})\\sigma$. The argument is demonstrated in four-sublattice pyrochlore and tetragonal models and then organized into a classification of the allowed second-, fourth-, and sixth-order spin-splitting forms under all 32 crystallographic point groups.","pith_inferences":["Inference: the irreducible-representation match can be used as a quick computational screen: for any candidate collinear antiferromagnet, compare the irrep of the ordered-moment multipole with the irreps of its hopping multipoles, and only systems with a match need be examined in detail.","Inference: because the splitting scale from the perturbation analysis is roughly $(t^{(1)})^2/h$, materials with large inter-site hopping and moderate exchange constants should show the effect most strongly; selecting for that ratio, rather than for heavy elements, is a testable material-design rule.","Inference: if spin currents generated this way have a well-defined spin quantization axis (the ordered moment), then antiferromagnetic spintronic devices based on these materials could avoid the spin-relaxation channels associated with atomic spin-orbit coupling; the paper notes the current is well-defined but does not develop the device consequence."],"forward_implications":["Collinear antiferromagnets with weak atomic spin-orbit coupling can exhibit spin-split Fermi surfaces in zero field, with the splitting even in momentum (for example proportional to $k_xk_y\\sigma$).","The symmetric spin-split dispersion produces a symmetric spin conductivity tensor, so an electric current or temperature gradient can generate a pure spin current along the ordered-moment direction without atomic spin-orbit coupling.","A shear strain can induce a uniform magnetization in the direction of the collinear antiferromagnetic order, realizing a piezomagnetic (magnetostriction) response.","The 32-point-group classification lists which collinear antiferromagnetic ordering patterns activate which spin-splitting forms, providing a direct guide for searching candidate materials among light-element and 3d transition-metal compounds.","In strictly bipartite lattices, chiral symmetry forbids the spin splitting, so the effect requires multisublattice clusters or hoppings that break the bipartite condition."],"supporting_citations":[{"why":"Provides the microscopic multipole description and the even-parity electric multipole expression for band dispersions in momentum space.","marker":"[34-36]"},{"why":"Supplies the cluster multipole basis sets used to construct ordered patterns and identify their irreducible representations.","marker":"[38]"},{"why":"Demonstrates spin-current generation in an organic antiferromagnet without atomic spin-orbit coupling, the prototype this paper generalizes.","marker":"[28]"},{"why":"Shows the anomalous Hall effect in collinear antiferromagnets, evidence that antiferromagnetic order can induce spin-dependent band structure without spin-orbit coupling.","marker":"[18]"},{"why":"Shows spin-split Fermi surfaces from electric toroidal order, supporting the idea that electronic order can mimic spin-orbit coupling.","marker":"[25,26]"}],"fun_headline_variants":["Antiferromagnetic order alone splits spin bands","Spin splitting without spin-orbit: collinear antiferromagnets","Magnetic order alone yields spin-split bands without spin-orbit","Momentum-dependent spin splitting without spin-orbit in antiferromagnets"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument assumes atomic spin-orbit coupling is weak enough to ignore, so the only spin splitting that can appear under collinear order is symmetric in momentum; if spin-orbit coupling is not negligible, extra antisymmetric terms appear and the classification is incomplete.","fun_headline_variants_meta":{"raw":{"variants":["Antiferromagnetic order alone splits spin bands","Spin splitting without spin-orbit: collinear antiferromagnets","Magnetic order alone yields spin-split bands without spin-orbit","Momentum-dependent spin splitting without spin-orbit in antiferromagnets"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001277,"raw_usage":{"total_tokens":5247,"prompt_tokens":997,"completion_tokens":4250,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":613,"completion_tokens_details":{"reasoning_tokens":4176}},"tokens_in":613,"tokens_out":4250,"duration_ms":30445,"temperature":1.0,"reasoning_tokens":4176,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T11:32:33.063259+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take any collinear antiferromagnet with negligible atomic spin-orbit coupling and a known ordered multipole $Q^{(0)}_{\\Gamma\\gamma}$; compute the tight-binding hopping multipoles of its crystal structure and compare irreducible representations. If the condition predicts a splitting of a given form (say $k_xk_y\\sigma$) but spin-resolved angle-resolved photoemission or quantum-oscillation measurements show no such even-in-$\\mathbf{k}$ spin splitting, the central claim fails; conversely, observing spin splitting in a collinear antiferromagnet whose hopping multipoles do not share an irrep with the order would also disprove it.","supporting_citations":[{"cited_title":"Suzuki , author T","cited_arxiv_id":null,"evidence_quote":"Supplies the cluster multipole basis sets used to construct ordered patterns and identify their irreducible representations."}],"review_version":1}