{"id":"d73ec7d6-bfa8-4a10-a825-ca80cec9f1d9","arxiv_id":"2607.13575","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Rotation subspaces carry independent Z2 invariants; time-reversal symmetry hides paired nontrivial subspaces from the conventional global Z2, as realized in CsCl with surface double Weyl points.","lead":"This paper shows that when a material has rotational symmetry, its electrons can be grouped into separate \"rotation subspaces,\" and each subspace can carry its own topological label even when the material as a whole looks topologically ordinary. The authors predict that the simple compound CsCl hosts protected surface states—double Weyl points—that ordinary topological diagnostics would miss.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Subspace bulk-boundary correspondence (Sec. II.C) is asserted, not derived: a nontrivial parity product along a 1D line need not, by itself, produce a protected surface DWP, so the CsCl prediction is not yet established.","rationale":"The reader's weakest assumption is exactly the subspace bulk-boundary correspondence in Sec. II.C. My independent read of the paper confirms that this is the load-bearing step: without it, the parity tables in Sec. III (which are internally consistent) only show that certain rotation-subspace parity products are odd; they do not by themselves prove that CsCl hosts surface double Weyl points. The concern is not that the parity counting is wrong, but that the inference from a 1D line invariant in a 3D BZ to a protected surface state is nontrivial and unproven. The paper's own caveat about the ¯M DWP coexisting with bulk bands further weakens the material claim. I therefore agree with the reader's conditional verdict rather than moving to accept or reject. The proposed slab-thickness and toy-model tests would settle whether the subspace bulk-boundary correspondence holds as a general principle or only as an artifact of the specific CsCl calculation.","tokens_in":10427,"tokens_out":10502,"duration_ms":127322,"concrete_test":"Using the Wannier tight-binding Hamiltonian that produced Figs. 3/4, compute the (111) and (001) slab spectra for thicknesses 5, 10, 20, and 40 unit cells. At each thickness, resolve the ¯Γ and ¯M states by C3/C4 eigenvalue and count states in the global gap; then repeat with a surface potential that preserves C_n and T but changes the surface environment. If the DWPs vanish, move off E_F, or cannot be distinguished from bulk states with increasing thickness, the line invariant in Eq. (3) does not control the surface point. For a sharper test, build a toy 3D model with the same C_n,P,T symmetries and parity tables but with the m=2 subspace VBM placed above the m=1,3 VBM; if the ¯M surface state disappears while (ν0,ν1,ν2,ν3)=(0,1,0,1), the asserted subspace bulk-boundary correspondence is falsified.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Sec. II.C states that 'the standard bulk-boundary correspondence naturally extends to the subspace level' and that each subspace with ν_m=1 contributes one protected boundary state at the surface point. This is the single step that turns the parity counting in Eq. (3) into the predicted double Weyl points in Figs. 3 and 4. It is not derived from any complete classification, and it is not a standard consequence of the 1D Fu-Kane parity formula. The invariant in Eq. (3) is the parity product at the two P-invariant endpoints of a line in a 3D BZ; for an isolated 1D inversion-symmetric insulator it quantizes the Zak phase/polarization, but without an additional symmetry (e.g., chiral) it does not guarantee a bound state inside the gap. Here the line is merely a 1D submanifold of a 3D BZ; the surface point receives bulk states from a 2D set of momenta, so even a nontrivial line invariant need not force a surface eigenstate. The numerical slab spectra in Figs. 3(c) and 4(c) could be finite-size surface resonances. The paper itself concedes that the ¯M DWP on the (001) surface lies below E_F and coexists with the m=2 bulk continuum, leaving unclear what protects it as a topological boundary state. Thus the central material claim is premature unless the subspace BBC is either proven from the 3D symmetry data or verified by a controlled slab/thickness and symmetry-breaking study.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a rotation-subspace topological classification for band insulators. Along a C_n-invariant 1D path, the occupied Hilbert space decomposes into n subspaces labeled by rotation eigenvalues, and the paper assigns each subspace an independent Z_2 invariant ν_m computed from inversion parity eigenvalues at the two endpoints (Eq. 3). The conventional global Z_2 is then the sum of the subspace invariants modulo 2 (Eqs. 4–5). The central formal claim is that time-reversal symmetry forces ν_m = ν_{n-m} for m ≠ 0, n/2 (Eq. 6), so a nontrivial pair of T-related subspaces always cancels in the global invariant, making the topology invisible to conventional diagnostics. The paper further postulates a 'subspace bulk-boundary correspondence' (Sec. II.C): each nontrivial subspace contributes one protected boundary state at the C_n-invariant surface point, so T-paired nontrivial subspaces produce a doubly degenerate surface double Weyl point. This framework is applied to bulk CsCl (space group Pm-3m, No. 221). Using DFT without spin-orbit coupling, the authors find Z_2^3 = (0,1,1) along Γ-R under C_3 and Z_2^4 = (0,1,0,1) along Γ-Z and M-R under C_4, while the global Z_2 is trivial. Surface calculations for the (111) and (001) surfaces show quadratic surface band crossings at the expected surface points, which are identified as double Weyl points.","tokens_in":10805,"tokens_out":2418,"duration_ms":30139,"significance":"If the central claim holds, the paper identifies a genuine blind spot in standard global-invariant diagnostics: for any spinless system with C_n symmetry and time reversal, a T-paired pair of nontrivial rotation subspaces is invisible to the conventional Z_2 invariant. This is a clean and potentially widely applicable observation, and the proposed Z_2^n refinement is a natural extension of mirror Chern number logic. The specific prediction for CsCl—a known, simple, experimentally synthesized material—is falsifiable by ARPES, and the paper ships concrete first-principles surface spectra and k·p models. The strengths are the transparent parity-counting derivation of Eqs. (3)–(5), the T-pairing argument of Eq. (6), and the independent verification of surface states from the Wannier model. However, the material realization rests on an unproven subspace bulk-boundary correspondence and on ignoring spin-orbit coupling for a heavy element (Cs), so the significance is conditional on those gaps being filled.","major_comments":[{"comment":"The subspace bulk-boundary correspondence is asserted, not derived. The paragraph states that 'the standard bulk-boundary correspondence naturally extends to the subspace level' and that each subspace with ν_m=1 contributes one protected boundary state at the surface point. This is the step that turns the parity product of Eq. (3) into the predicted double Weyl points. For an isolated 1D inversion-symmetric insulator, the parity product at the two P-invariant endpoints gives a Z_2 polarization/Zak phase; it does not by itself guarantee a bound state in the gap without additional conditions. Here the 1D path is only a submanifold of a 3D BZ, and the surface point receives Bloch states from a 2D set of momenta. A nontrivial line invariant therefore does not automatically force a surface eigenstate. The numerical slab spectra in Figs. 3(c) and 4(c) could be finite-size surface resonances. T","section":"Sec. II.C"},{"comment":"The material realization is computed entirely without spin-orbit coupling, yet Cs is a heavy element (Z=55) and CsCl has significant core SOC. The paper states that 'the low-energy bands remain almost unchanged upon the inclusion of spin-orbit coupling' (Fig. 2d), but no subspace-resolved parity analysis with SOC is given. The spinless C_n eigenvalues and the T-pairing argument in Sec. II.B are extended to spinful systems only by a single sentence ('the extension ... is straightforward'), and the actual C_n eigenvalues for spinors (e.g., e^{iπm/n} with m half-integer) are not used. Since the central claim is a material prediction, the authors must show that the subspace invariants (ν_0, ν_1, ν_2) and (ν_0, ν_1, ν_2, ν_3) are stable when SOC is included, or at least provide the spinful version of the invariants for the specific paths in CsCl.","section":"Sec. III.B, SOC neglect"},{"comment":"The paper itself concedes that the ¯M-point double Weyl point lies below E_F and coexists with the bulk continuum from the m=2 subspace. The subspace bulk-boundary correspondence is said to protect boundary states within the subspace gap, not the global gap. At ¯M, the m=2 bulk states project onto the same surface momentum, so the protection of the surface state against hybridization with these bulk states is not established; the claim that 'hybridization ... is forbidden by C_4 rotational symmetry' is only argued by the different C_4 eigenvalues of the surface and bulk states, but the surface state at ¯M may couple to bulk states with the same C_4 eigenvalue or to other surface resonances. This makes the predicted DWP at ¯M less decisive as evidence for the subspace BBC. The authors should either compute the thickness dependence of the ¯M feature, show that it remains a genuine surface","section":"Sec. III.B.2, ¯M-point DWP"}],"minor_comments":[{"comment":"The notation n_{k=0}^m and n_{k=π}^m is not defined precisely in the text; it should state that these are numbers of occupied valence states with negative inversion eigenvalue in the m-th rotation subspace at the two inversion-invariant momenta. Also, for spinful systems the rotation eigenvalues in Eq. (1) need the phase convention for spinors; please clarify.","section":"Eqs. (3)–(5)"},{"comment":"The panel labels in Fig. 2 are partially duplicated ('Energy (eV)' appears twice), and the color scales in Figs. 3(c,d) and 4(c,d) are not clearly labeled (units of k, color bar meaning). This makes the surface spectra hard to evaluate.","section":"Fig. 2"},{"comment":"The Wannier-function construction is described only briefly. Please report the number of Wannier orbitals, the projection centers, and the maximal spread; this is needed to assess the accuracy of the surface-state calculation.","section":"Sec. III.A"},{"comment":"The statement that 'any material with rotational symmetry can be analyzed within this framework' is too broad; the framework applies to a 1D C_n-invariant path with inversion at the endpoints, not to general rotational symmetry in 2D/3D. Please qualify.","section":"Sec. IV"},{"comment":"The paper cites arXiv:2607.13575 as its own submission; this is fine for a preprint, but for a journal version the citation format should be updated. Also, Ref. [35] appears to be an unrelated paper on antiferroelectricity; please verify the citation intent.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The central formal derivation (Eqs. 3–6) is correct and the T-pairing argument is a nice observation. The concern is the material claim: the subspace bulk-boundary correspondence is the load-bearing bridge, and it is asserted rather than proven, while the SOC neglect is a practical but non-negligible approximation for CsCl. These are fixable with additional calculations (spinful invariants, slab thickness scaling, or a symmetry-breaking benchmark), so the paper is not fatally flawed. I would recommend major revision rather than rejection, but I would not accept in the current form."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe core idea here is simple and probably sound: decompose occupied bands along a rotation-invariant line by rotation eigenvalues, compute the Fu-Kane parity product in each subspace, and note that time-reversal forces conjugate subspaces to have equal Z2 values. When two subspaces are nontrivial, the global Z2 vanishes, so the topology is invisible to standard diagnostics. The parity tables for CsCl are consistent, and the T-pairing argument is a clean corollary. The k·p models and first-principles surface spectra are competently done.\n\nWhat is genuinely new is the explicit Z_2^n classification per rotation subspace and the observation that T-paired nontrivial subspaces always cancel in the global sum. As far as I can tell, this is a legitimate refinement of the Fu-Kane criterion, though it overlaps substantially with existing symmetry-indicator and rotation-anomaly frameworks. The paper cites Fang-Fu surface rotation anomaly but never says how the new invariant relates to it. That comparison needs to be made.\n\nThe load-bearing weakness is the subspace bulk-boundary correspondence in Sec. II.C. A nontrivial parity product along a 1D line in a 3D BZ does not, by itself, guarantee a protected surface state. The surface point collects states from a 2D set of momenta, and the line invariant is not a complete 3D invariant. The numerical slab states in Figs. 3 and 4 could be finite-size resonances. The authors themselves concede that the M-bar DWP coexists with bulk bands and is protected only by rotation eigenvalues, which is not a direct test of topological origin. I would like to see a slab-thickness scaling study or a symmetry-breaking perturbation that gaps the surface states, or a derivation of the BBC from the full 3D data. Without that, the material prediction is not established.\n\nThere are smaller issues: SOC is neglected for Cs, which is heavy; the claim that low-energy bands are unchanged deserves a quantitative check. Also, no data or code are deposited.\n\nOverall, the theory core is worth refereeing; the material realization needs another round of work. I would send it to review with the expectation of major revision, asking specifically for a proof or controlled numerical verification of the subspace BBC and a comparison with existing classification frameworks.","headline":"Clean parity-counting idea, plausible T-pairing argument, but the CsCl surface prediction rests on an unproven subspace bulk-boundary correspondence.","tokens_in":11268,"tokens_out":4126,"would_cite":false,"duration_ms":43422,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"A material the standard toolkit calls trivial can still carry hidden topology, provided you sort its electron bands by rotation eigenvalue first: cesium chloride realizes this with double Weyl surface points.","keywords":["rotation topological states","subspace topology","Z2 invariant","time-reversal symmetry","cesium chloride","double Weyl point","symmetry indicators","bulk-boundary correspondence"],"falsifier":"Angle-resolved photoemission on cleaved CsCl (111) and (001) surfaces: absence of the predicted two-fold-degenerate quadratic surface states at Γ̄ and M̄ — or a tight-binding surface calculation including rotation-subspace coupling that gaps those states — would falsify the subspace bulk-boundary correspondence and the material prediction.","tokens_in":10312,"feed_emoji":"🌀","tokens_out":3482,"duration_ms":42149,"temperature":0.7,"pith_summary":"The paper argues that the usual way of diagnosing topological insulators — computing a single invariant from all occupied bands — can miss real topology when the crystal has rotational symmetry. Rotational symmetry splits the occupied Hilbert space along high-symmetry lines into several independent subspaces, each labeled by a rotation eigenvalue, and each subspace can carry its own Z2 invariant; the full classification is then Z2^n rather than a single Z2. Because the conventional global Z2 invariant is the sum of these subspace invariants modulo 2, an even number of nontrivial subspaces gives a zero global invariant, hiding the topology. Time-reversal symmetry makes this hiding systematic: it pairs conjugate rotation subspaces and forces their invariants to be equal, so nontrivial rotation topology is always invisible to global diagnostics. The paper makes the idea concrete with bulk CsCl, a simple cubic insulator that is globally trivial but carries nontrivial Z2^3 and Z2^4 subspace invariants, which surface calculations show produce double Weyl points on the (111) and (001) surfaces.","feed_headline":"Rotation symmetry hides CsCl's topological surface states","feed_subtitle":"Sorting bands by rotation eigenvalue reveals Z2^3 and Z2^4 invariants that the standard global Z2 misses.","key_machinery":"The key object is the subspace-resolved parity invariant: for the m-th rotation subspace on a 1D high-symmetry path, ν_m counts (mod 2) the number of valence-band states with negative inversion eigenvalue at the two P-invariant points of that path. Stacking these invariants for m = 0,1,…,n−1 gives the Z2^n classification. The load-bearing relation is the time-reversal pairing ν_m = ν_{n−m}, which forces nontrivial subspaces to come in pairs and makes the global invariant Σ_m ν_m mod 2 identically zero whenever any m ≠ 0, n/2 subspace is nontrivial. The subspace bulk-boundary correspondence then assigns one protected boundary state to each nontrivial subspace at the Cn-invariant surface point","core_discovery":"The central claim is that rotation symmetry enables a refined, subspace-resolved topological classification that the conventional global Z2 invariant cannot see. Along a path invariant under an n-fold rotation, the occupied bands decompose into n subspaces labeled by rotation eigenvalues e^{2πim/n}; each subspace has its own Z2 invariant ν_m, computed from parity eigenvalues at the two P-invariant points of that path, so the overall classification is Z2^n. The global invariant is the product of the subspace invariants, i.e. ν_global = Σ_m ν_m mod 2, so any even number of nontrivial subspaces yields ν_global = 0. Time-reversal symmetry pairs the m and n−m subspaces and, because T preserves pa","pith_inferences":["The argument likely extends to spinful systems with spin-orbit coupling, where time-reversal still pairs conjugate rotation subspaces; the paper states this extension is straightforward but does not carry it out, so subspace-resolved screening of known 'trivial' insulators with heavy elements is a natural next step.","The hidden topology here is 1D-in-character: each nontrivial path produces boundary states only at a single surface point rather than a full 3D bulk phase. This suggests the phenomenon is closer to weak or fragile topology than to a strong invariant, an interpretation the paper leaves open.","High-throughput topological-materials databases built on global symmetry indicators cannot see this class, so a re-analysis of rotation-symmetric insulators with subspace-resolved parity counting could uncover many more double Weyl surface points, possibly relevant to surface catalysis due to the higher density of states of a quadratic degeneracy."],"forward_implications":["Any rotation-symmetric insulator with an even number of nontrivial subspaces is systematically mislabeled trivial by standard Z2 and symmetry-indicator diagnostics, so existing topological-material catalogs may contain overlooked hidden-topology candidates.","Within the paper's own argument, bulk CsCl is a concrete, experimentally synthesized material predicted to host surface double Weyl points at Γ̄ on (111) and at Γ̄ and M̄ on (001), observable by angle-resolved photoemission.","The subspace bulk-boundary correspondence predicts that breaking the C3 or C4 rotation should lift the two-fold degeneracy at those surface points and gap the surface states, providing a direct symmetry test of the topological origin.","The M̄ double Weyl point coexists with bulk bands but is stabilized by C4 eigenvalue separation, meaning surface states can be symmetry-isolated from bulk projections even when not inside the global gap.","Because the construction works along any rotation-symmetric path, the framework applies broadly to crystals with C3, C4, C6, or other n-fold rotations, extending the mirror-subspace logic of mirror Chern insulators to a larger class of symmetries."],"fun_headline_variants":["Rotation subspaces hide topological states standard Z2 misses","CsCl's hidden topology: rotation eigenvalues expose Z2^3 and Z2^4","Rotation symmetry reveals hidden topological states in CsCl","Standard Z2 says trivial, rotation says nontrivial: CsCl's hidden topology","Rotation decomposition uncovers topological states invisible to standard Z2"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The subspace bulk-boundary correspondence — that each nontrivial 1D rotation-subspace invariant guarantees one protected boundary state even though the global invariant is trivial — is asserted rather than derived from a full classification; if inter-subspace coupling on the surface or level repulsion from nearby bulk bands destroys those modes, the predicted double Weyl points vanish.","fun_headline_variants_meta":{"raw":{"variants":["Rotation subspaces hide topological states standard Z2 misses","CsCl's hidden topology: rotation eigenvalues expose Z2^3 and Z2^4","Rotation symmetry reveals hidden topological states in CsCl","Standard Z2 says trivial, rotation says nontrivial: CsCl's hidden topology","Rotation decomposition uncovers topological states invisible to standard Z2"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000478,"raw_usage":{"total_tokens":2235,"prompt_tokens":804,"completion_tokens":1431,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":548,"completion_tokens_details":{"reasoning_tokens":1340}},"tokens_in":548,"tokens_out":1431,"duration_ms":11249,"temperature":1.0,"reasoning_tokens":1340,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-02T04:46:56.769889+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Angle-resolved photoemission on cleaved CsCl (111) and (001) surfaces: absence of the predicted two-fold-degenerate quadratic surface states at Γ̄ and M̄ — or a tight-binding surface calculation including rotation-subspace coupling that gaps those states — would falsify the subspace bulk-boundary correspondence and the material prediction.","supporting_citations":[],"review_version":1}