{"id":"06e033a9-80ff-4e80-9c89-62e7cae83cc3","arxiv_id":"2412.03638","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":8.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"One-dimensional mobility plus infinite on-site repulsion forces all particle exchanges to be even permutations, which exactly yields a fully spin-polarized metallic ground state in a Lieb-lattice model.","lead":"This paper proves that electrons constrained to move along one-dimensional lines, with strong repulsion blocking double occupancy, can form a half-metallic ferromagnet at any doping fraction. The proof is exact for a solvable Lieb-lattice model and suggests physical routes via the Emery model and Wigner-crystal vacancies.","discovery_kind":"first_principles","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Prop. II.3's even-permutation ergodicity is proved only for canonical configurations; the reduction from arbitrary configurations is not shown, and the induction can land on smaller blocks violating condition (a), so the Perron-Frobenius uniqueness argument in Theorem II.5 has a genuine gap.","rationale":"The reader's weakest assumption is exactly the even-permutation ergodicity proof in Propositions II.3 and II.4. My stress-test refines that concern: the written induction proves transitivity only for canonical configurations, while the proposition is stated for all configurations satisfying condition (a). The paper's brief appeal to positional ergodicity does not explain why the group of closed-loop 3-cycles at an arbitrary configuration is conjugate to the group at a canonical configuration; in fact the induction can construct smaller blocks whose row/column occupancies violate the two-vacancy condition, so the inductive hypothesis cannot be invoked. This is load-bearing because Theorem II.5 uses even-permutation ergodicity to establish irreducibility of the non-positive Hamiltonian matrix; without irreducibility, Perron-Frobenius gives only a ground state in each connected component, and spin degeneracies could survive. I do not claim the main theorem is false; the OBC non-positivity argument (Theorem II.2), the boson-fermion equivalence, and the metallic Fermi-surface construction are independent and credible. The issue is a proof gap in the uniqueness argument. A small-lattice exhaustive check of the generated 3-cycle groups would settle whether the gap is purely expositional or reflects a real counterexample. Since this is the same concern the reader identified, and the recommended disposition remains conditional rather than full acceptance or rejection, the verdict should be unchanged.","tokens_in":31643,"tokens_out":24585,"duration_ms":266632,"concrete_test":"Exhaustively enumerate all OBC electron configurations on a 3x3 and 4x4 Lieb lattice satisfying condition (a). For each configuration, list every elementary 3-cycle of the type shown in Fig. 2(b) that returns the charge configuration to itself, compute the subgroup of S_N generated by these 3-cycles, and check that it equals A_N. Be sure to include non-canonical representatives, not only configurations with all dangling bonds vacant. If any configuration fails, construct the two spin states whose connection requires the missing permutation; their disconnection would break Perron-Frobenius irreducibility. If all configurations pass, the gap is only in the exposition of the reduction, not in the mathematical claim.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The most load-bearing step is Proposition II.3, which supplies the irreducibility needed for the Perron-Frobenius uniqueness argument in Theorem II.5. The written proof establishes transitivity of the available elementary 3-cycles only for 'canonical charge configurations' with all dangling bonds vacant, and then asserts that arbitrary configurations satisfying condition (a) can be reduced to this case by positional ergodicity. That reduction is not automatic: positional ergodicity gives a path between charge configurations, but it does not by itself conjugate the group of closed-loop 3-cycles at an arbitrary configuration to the group at the canonical configuration, because the path can mix electron and vacancy labels. More concretely, in the induction step from (Lx+1) x Ly to Lx x Ly, the restricted smaller block can have X(y) = Lx, i.e., only one vacancy on a row, violating the hypothesis 0 < X(y) < Lx on which the inductive call depends; the same issue can occur for columns. If any non-canonical configuration satisfying condition (a) has a strictly smaller set of generated even permutations, the Hamiltonian matrix is reducible in that subsystem-symmetry sector, and Theorem II.5's claim of a unique half-metallic ground state fails for that sector.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper proposes a kinetic mechanism for half-metallic ferromagnetism in strongly interacting systems where doped particles have strictly one-dimensional mobility. The authors introduce a solvable U=∞ model on a Lieb lattice with correlated hopping, prove that all induced ring exchanges are even permutations (Prop. II.1), and show that the fully spin-polarized state is a ground state (Thm. II.2). They then claim uniqueness of the half-metallic ground state for any doping in the thermodynamic limit (Thm. II.5), based on an even-permutation ergodicity property (Props. II.3 and II.4) that is supposed to make the Hamiltonian matrix irreducible in each symmetry sector. The paper also discusses quasiparticles with approximate 1D mobility in the Emery model and Wigner-crystal vacancies, and proves an exact boson-fermion equivalence (Thm. V.1).","tokens_in":31813,"tokens_out":23917,"duration_ms":221772,"significance":"If Thm. II.5 is correct, it would provide a rare rigorous example of a metallic ferromagnetic ground state that is robust in the thermodynamic limit, going beyond the single-hole Nagaoka result. The proposed mechanism is physically appealing, and the model is exactly solvable in the ferromagnetic sector. The boson-fermion equivalence and the connection to Bose metals are interesting corollaries. The paper is clearly written, and the non-positivity/Perron-Frobenius strategy is standard and elegant. However, the central uniqueness theorem depends on the even-permutation ergodicity property, whose proof currently has significant gaps; these gaps are load-bearing and must be repaired before the main claim can be accepted.","major_comments":[{"comment":"The reduction to canonical charge configurations is not justified. The proof states that it is sufficient to prove transitivity for configurations in which all dangling bonds are vacant because positional ergodicity makes this possible. Positional ergodicity only provides a path from an arbitrary configuration to a canonical one; it does not imply that the group of electron permutations generated by closed loops at the arbitrary configuration is isomorphic to that at the canonical configuration. A path that changes the vacancy positions can mix electron and vacancy labels, and the available elementary 3-cycles at the two configurations can differ. Unless the generated group is shown to be invariant under such a path, the transitivity proof applies only to the canonical configuration, and the Perron-Frobenius irreducibility argument in Theorem II.5 may fail for non-canonical sectors.","section":"Appendix A, proof of Prop. II.3"},{"comment":"The induction step on (Lx+1) × Ly splits the electrons into X1 (on and left of column x = Lx) and X2 (right of that column) and requires at least one px-electron on each row to the left of x = Lx. This condition is not implied by the hypothesis 0 < X(y) < Lx. If a row has its only x-bond electron on the bond x = Lx + 1/2 (which lies in X2), then after moving the leftmost vacancy to x = Lx + 1/2 the reduced Lx × Ly block has X_block(y) = 0, so the inductive hypothesis (condition (a)) does not apply. The same issue afflicts case (2) and case (3), which depend on filling the column x = Lx + 1/2 on every row. The proof does not explain how rows with no electrons in X1 (or columns with no electrons in the analogous column split) are handled, so the transitivity of the generated group on the full electron index set is not established.","section":"Appendix A, induction step of Prop. II.3"},{"comment":"The PBC version of even-permutation ergodicity is stated without proof ('The details are omitted here'). This is a load-bearing component of the PBC part of Theorem II.5, and the PBC proof has additional complications (the gauge transformation (10) changes the boundary condition, and matrix elements across the periodic boundary acquire phase factors that require the fermion parity condition (a)). A statement without proof is not sufficient for a central proposition on which a main theorem rests.","section":"Appendix A, Prop. II.4"},{"comment":"The proof of even-permutation ergodicity for the vacancy model is also insufficient. For the base case the four 3-cycles are listed, but their transitivity is merely asserted. For larger systems the proof says that correlated movements of vacancies in blocks A, B and C 'induce an overlapping set of 3-cycles' and that 'it is straightforward to see' that these generate a transitive group, without specifying the actual sequences or verifying that the conditions (c-d) guarantee the required room for all electrons. Because Theorem IV.4 relies entirely on this proposition, the uniqueness claim for the vacancy model is not rigorously established.","section":"Appendix B, proof of Prop. IV.3"}],"minor_comments":[{"comment":"The claim that under OBC 'the vacancies in the final configuration cannot be permuted from their initial configuration' is too strong and is not generally true: with three or more vacancies on a row, the dynamics can induce even permutations of vacancies while electrons return to their original positions. The argument only needs the vacancy permutation to be even, which follows from the evenness of the total electron-vacancy permutation; the proof should be revised accordingly.","section":"Prop. II.1, OBC case"},{"comment":"The notation X(y) is defined in Eq. (4) for PBC as a sum over x = 1,...,Lx. For OBC there are Lx+1 x-bond sites per row, and the text says the conserved quantities are 'analogously defined,' but the condition 0 < X(y) < Lx in Proposition II.3 is only equivalent to 'two vacancies per row' if X(y) counts all Lx+1 bonds. Please clarify the OBC definition explicitly.","section":"Sec. II, notation for X(y)"},{"comment":"In Eqs. (38-39), the energies of the nematic and isotropic phases are expanded in powers of x, but the definitions of E0, Δc, and Veff are not fully specified in the text; please provide the definitions or a reference for these quantities.","section":"Sec. III, Eqs. (38-39)"},{"comment":"Figure 3 is not described in detail in the text; it would be helpful to state in the caption how the Fermi surfaces are obtained from the single-particle dispersions of the 2L wires, particularly the relation between the wire index and the momentum perpendicular to the wire.","section":"Fig. 3"}],"recommendation":"major_revision","confidential_remarks":"The manuscript addresses an important open problem and the proposed mechanism is elegant, but the proof of the central uniqueness theorem has a genuine gap in the even-permutation ergodicity argument, and the PBC and vacancy versions are even less complete. The paper is likely to be influential if the gap is fixed, but in its current form the main claim is not rigorously established. I recommend major revision rather than rejection because the gap is local to Appendix A and may be repairable, and the rest of the paper (Prop. II.1, Thm. II.2, boson-fermion equivalence) is sound."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Kyung-Su and Veit have written a paper that deserves to be taken seriously. The central idea is new: in a strong-coupling model on a Lieb lattice where electrons move only along rows or columns, every induced ring exchange is an even permutation, and that evenness, combined with a Perron-Frobenius argument, yields a unique fully spin-polarized ground state at any fixed doping in the thermodynamic limit. If the proof is correct, this is the first rigorous example of a metallic half-metal from purely repulsive, spin-independent interactions away from Nagaoka's single-hole setting. The OBC argument is mostly careful: the gauge transformation to a non-positive matrix, the Cauchy-Schwarz energy comparison, and the reduction to free fermions in the ferromagnetic sector are all clean. The boson-fermion equivalence (Theorem V.1) is a nice formal result and appears correct.\n\nThe soft spot is the proof of even-permutation ergodicity in Appendix A. The induction establishes transitivity only for canonical charge configurations, where all dangling bonds are vacant. The reduction of an arbitrary configuration to a canonical one is asserted through positional ergodicity, but that is not enough: making a smaller block canonical requires bringing vacancies into its boundary bonds, and the dynamics that does this can move electrons across the partition boundary, spoiling the inductive assumption. In the step from (Lx+1)xLy to Lx x Ly, a smaller block can end up with one vacancy on a row, violating condition (a). If any non-canonical configuration generates a strictly smaller set of even permutations, the Perron-Frobenius matrix is reducible in that sector and the uniqueness conclusion in Theorem II.5 fails. This is a genuine gap, though likely repairable with a more careful induction or a different transitivity argument. Two smaller issues: the PBC statements (Prop II.4 and Theorem II.5) are given without proof, which matters because the thermodynamic-limit claim should be boundary-condition independent; and the Emery-model and Wigner-crystal sections are explicitly heuristic, which is fine as long as they are not read as theorems.\n\nThe paper is for researchers working on rigorous many-body theory and kinetic magnetism. It should go to peer review; a good referee should scrutinize Proposition II.3 and ask the authors to either fix the induction or state the ergodicity condition explicitly. Even with the gap, the mechanism and the solvable model are worth publishing after revision.","headline":"Genuinely new mechanism and a substantial proof effort, but the even-permutation ergodicity induction has a real gap; worth a careful referee.","tokens_in":32458,"tokens_out":5462,"would_cite":true,"duration_ms":52650,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["71.10.Fd","75.10.-b"],"model":"deepseek-v4-flash","headline":"One-dimensional mobility forces even-parity multi-electron exchanges, making a repulsive model a unique half-metallic ferromagnet at any doping.","keywords":["half-metallic ferromagnetism","one-dimensional mobility","Lieb lattice","ring-exchange parity","even-permutation ergodicity","boson-fermion equivalence","Emery model","Wigner crystal vacancies"],"falsifier":"Enumerate all electron configurations satisfying the row/column condition on a small Lieb lattice with open boundaries, form the graph whose edges are the elementary 3-cycles, and check whether the induced permutations generate the full alternating group on the electrons; one counterexample would disprove even-permutation ergodicity and break the uniqueness theorem. The same exhaustive check on small periodic clusters would test the periodic-boundary version, which the paper states without a full proof.","tokens_in":31346,"feed_emoji":"🧲","tokens_out":12816,"duration_ms":114441,"temperature":0.7,"pith_summary":"The paper proposes a general kinetic mechanism for half-metallic ferromagnetism—a conducting state with full spin polarization—based only on strong on-site repulsion and constrained one-dimensional motion. On a solvable Lieb lattice model where electrons hop only along rows or columns and double occupancy is forbidden, the authors show that every multi-electron exchange generated by the dynamics is an even permutation of the electrons. By the parity rule for ring exchanges, even-parity permutations favor ferromagnetism, and the paper proves that the fully spin-polarized state is not only a ground state but, whenever each row and column contains at least one mobile electron and two empty sites, the unique ground state at every filling in the thermodynamic limit, with only spin-rotational and Fermi-surface degeneracies. This contrasts with the classic single-hole ferromagnetism theorem, which does not survive thermodynamic-limit scaling. The same evenness argument is applied to quasi-one-dimensional holes in the Emery model and to vacancies in a Wigner crystal, and it yields an exact spectral equivalence between fermionic and hard-core bosonic versions of these models.","feed_headline":"Constrained 1D motion forces a fully polarized metallic ground state","feed_subtitle":"A rigorous Lieb-lattice proof makes the half-metallic ferromagnet the unique ground state at any doping.","key_machinery":"The load-bearing object is the elementary 3-cycle: a correlated move in which two p-electrons and the d-electron at a shared vertex permute among three neighboring sites. The central identity is that every hopping term of (3) acts as such a 3-cycle, and every 3-cycle is an even permutation; a gauge transformation makes all matrix elements non-positive. The proof then needs even-permutation ergodicity: for configurations satisfying the row/column condition, the 3-cycles generate the full alternating group on the $N$ electrons, established by induction using a graph-puzzle lemma. Non-positivity plus irreducibility in each sector forces the unique ferromagnetic ground state, and the ferromagnetic reduction to decoupled wires supplies the Fermi surface.","core_discovery":"The paper's central result is Theorem II.5: on an $L\\times L$ Lieb lattice (a square lattice with an extra site on each bond) with open boundaries and Hamiltonian (3), the infinite-repulsion model hosts a half-metallic ferromagnet as its unique ground state for every filling $0<\\nu_p<2$, aside from the $(N+1)$-fold spin degeneracy from rotational symmetry and the $2L$-fold degeneracy of the gapless Fermi surface. The proof combines a gauge transformation that makes every off-diagonal matrix element negative, a proof that the allowed elementary 3-cycles generate the alternating group on the electrons, and an irreducibility argument that forces a unique fully polarized eigenstate in each symmetry sector. The metallic character is explicit: in the ferromagnetic sector the Hamiltonian reduces to $2L$ independent one-dimensional wires with spectrum $E_k=-2t\\cos k$, so the ground state fills each wire uniformly and has straight-line Fermi surfaces. The authors state an analogous theorem for Wigner-crystal vacancies under conditions that guarantee even-permutation ergodicity, and they note the same results hold on other lattices in any dimension.","pith_inferences":["If the evenness argument is the whole story, any lattice or experimental realization of line-constrained, infinite-repulsion particles should develop full spin polarization; an optical-lattice implementation of correlated hopping could test this directly by measuring spin-resolved densities and compressibility.","The periodic-boundary version of the ergodicity proposition is stated without a full proof; an exhaustive enumeration on small periodic clusters would settle whether the uniqueness claim is genuinely independent of boundary conditions.","The boson-fermion equivalence suggests a sharp diagnostic for the Bose metal: in the hard-core bosonic system, the straight-line Fermi surfaces should persist under weak interactions and show up in momentum-resolved noise correlations.","Because the parity argument is geometric, it likely extends to SU(N) spins and to three-dimensional lattices with line mobility; a natural next question is whether the nematic-to-isotropic transition in the two physical examples coincides with the onset of full spin polarization."],"forward_implications":["Half-metallic ferromagnetism is a thermodynamic phase in a spin-independent model: it survives at fixed doping in the thermodynamic limit, not just at single-hole doping.","The ground state is provably metallic, with straight-line Fermi surfaces obtained by uniformly filling $2L$ one-dimensional wires.","In the strong-coupling Emery model, intermediate hole doping should show full spin polarization, while dilute and near-full doping favor a fully nematic, spin-degenerate state.","In the Wigner-crystal vacancy model, sectors with equal vacancy numbers per line and at least two vacancy orientations have a unique ferromagnetic ground state with at least $3L$ low-energy excitations at the $2k_F$ wavevectors.","The fermionic and hard-core bosonic versions of these models have identical spectra, so the bosonic system is a ferromagnetic Bose metal with the same Fermi surfaces."],"supporting_citations":[{"why":"supplies the parity rule for ring exchanges: even-parity permutations mediate ferromagnetism.","marker":"[45]"},{"why":"provides the graph-puzzle lemma used to prove that the elementary 3-cycles generate the alternating group.","marker":"[50]"},{"why":"is the single-hole ferromagnetism theorem that this work generalizes to finite doping in the thermodynamic limit.","marker":"[5]"},{"why":"derives quasi-one-dimensional hole dynamics in the strong-coupling Emery model, one of the physical realizations of the mechanism.","marker":"[14]"},{"why":"establishes the one-dimensional t11 vacancy dynamics in a Wigner crystal used for the second physical realization.","marker":"[15]"},{"why":"gives the perturbative derivation of multi-spin ring exchange whose sign underlies the parity rule.","marker":"[7]"},{"why":"proves that even-parity ring-exchange permutations imply ferromagnetism, connecting the parity argument to ground-state magnetization.","marker":"[64]"}],"fun_headline_variants":["1D mobility drives itinerant ferromagnetism at any doping","Full spin polarization from purely 1D electron motion","Lieb-lattice proof: half-metal ground state for all dopings","Exact half-metallic ferromagnet from constrained 1D dynamics","1D motion forces full spin polarization at any filling"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The uniqueness proof assumes that from every allowed arrangement—with at least one mobile electron and two empty sites in every row and column—the allowed moves rearrange the electrons through every possible even permutation; if a single allowed arrangement fails this, the ferromagnetic ground state need not be unique.","fun_headline_variants_meta":{"raw":{"variants":["1D mobility drives itinerant ferromagnetism at any doping","Full spin polarization from purely 1D electron motion","Lieb-lattice proof: half-metal ground state for all dopings","Exact half-metallic ferromagnet from constrained 1D dynamics","1D motion forces full spin polarization at any filling"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000534,"raw_usage":{"total_tokens":2580,"prompt_tokens":967,"completion_tokens":1613,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":583,"completion_tokens_details":{"reasoning_tokens":1527}},"tokens_in":583,"tokens_out":1613,"duration_ms":11907,"temperature":1.0,"reasoning_tokens":1527,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T22:15:44.935854+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Enumerate all electron configurations satisfying the row/column condition on a small Lieb lattice with open boundaries, form the graph whose edges are the elementary 3-cycles, and check whether the induced permutations generate the full alternating group on the electrons; one counterexample would disprove even-permutation ergodicity and break the uniqueness theorem. The same exhaustive check on small periodic clusters would test the periodic-boundary version, which the paper states without a full proof.","supporting_citations":[{"cited_title":"Oshikawa, Topological approach to Luttinger’s theo- rem and the Fermi surface of a Kondo lattice, Physical Review Letters 84, 3370 (2000)","cited_arxiv_id":null,"evidence_quote":"supplies the parity rule for ring exchanges: even-parity permutations mediate ferromagnetism."},{"cited_title":"Yang and Y.-Q","cited_arxiv_id":null,"evidence_quote":"provides the graph-puzzle lemma used to prove that the elementary 3-cycles generate the alternating group."},{"cited_title":"Nagaoka, Ferromagnetism in a narrow, almost half- filled s band, Physical Review 147, 392 (1966)","cited_arxiv_id":null,"evidence_quote":"is the single-hole ferromagnetism theorem that this work generalizes to finite doping in the thermodynamic limit."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"derives quasi-one-dimensional hole dynamics in the strong-coupling Emery model, one of the physical realizations of the mechanism."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"establishes the one-dimensional t11 vacancy dynamics in a Wigner crystal used for the second physical realization."},{"cited_title":"Takahashi, Half-filled Hubbard model at low tempera- ture, Journal of Physics C: Solid State Physics 10, 1289 (1977)","cited_arxiv_id":null,"evidence_quote":"gives the perturbative derivation of multi-spin ring exchange whose sign underlies the parity rule."},{"cited_title":"Aizenman and E","cited_arxiv_id":null,"evidence_quote":"proves that even-parity ring-exchange permutations imply ferromagnetism, connecting the parity argument to ground-state magnetization."}],"review_version":1}