{"id":"6541d882-ce2a-49d4-94bb-ac273a5c7a5e","arxiv_id":"1909.02277","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"C3-symmetric K2IrO3 is predicted to have off-diagonal exchange couplings about ten times larger than Na2IrO3, and exact diagonalization shows such couplings can stabilize spin liquid phases.","lead":"Ab initio calculations for a proposed honeycomb iridate, K2IrO3, reveal unusually large off-diagonal magnetic exchange couplings that are about ten times larger than in the related compound Na2IrO3. The authors argue these couplings, rooted in the crystal's high C3 symmetry, can stabilize quantum spin liquid states and suppress magnetic order.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"MRCI-only NN couplings place K2IrO3 in a 3-fold SDW ordered phase; the no-order conclusion is obtained only after adding J2,J3 fitted to the same experiments, so the claim that large Gamma suppresses ordering is not established.","rationale":"The reader's conditional verdict is appropriate, but the decisive soft spot is the internal gap between the computed NN couplings and the claimed experimental consistency. Fig. 3(a) shows the MRCI point in a 3-fold SDW ordered phase; only after adding J2/J3 chosen to reproduce the experimental specific heat does the model sit near a phase boundary with suppressed order. The paper's own text in Section II.D makes this fitting step explicit, so the no-order statement is not a consequence of the ab initio couplings alone. This is an internal-consistency concern, not a disagreement with the community consensus on Gamma terms. We credit the central ab initio result: the MRCI values and the phi-twist trend are concrete, and the similar large off-diagonal couplings reported in Ref. [23] provide independent support for the magnitude of Gamma. The structural-disorder caveat noted by the reader is real and is acknowledged by the authors, but it is secondary: even for the idealized C3 structure, the NN parameters alone do not establish the suppression of ordering. A conditional acceptance with a request for independent J2/J3 determination is therefore the right posture; the stress-test does not move the verdict.","tokens_in":17023,"tokens_out":7486,"duration_ms":73042,"concrete_test":"Compute J2 and J3 for the same DFT-optimized C3 structure with an independent method, e.g., the same embedded-cluster MRCI scheme on clusters that include second- and third-neighbor Ir pairs, or constrained DFT. Then rerun the ED phase determination (Fig. 3b) at the independent (J2,J3) values. If those values do not fall close to the zigzag/3-fold SDW boundary, or if the neighboring no-order window is narrow, the claim that the calculated couplings explain the absence of magnetic order in K2IrO3 is not supported. As a minimal control, also state explicitly that the MRCI-only point (J2=J3=0) is ordered, confirming that the no-order result is carried by the fitted longer-range couplings rather than by the large Gamma terms.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The most load-bearing problem is internal to the ED argument rather than only the structural idealization. Section II.D states that the pure MRCI parameter set (Gamma_xy = 5.2 meV, Gamma_yz = -8.9 meV) stands on the 3-fold SDW ordered phase (Fig. 3a). Hence the ab initio NN Hamiltonian by itself does not predict suppression of magnetic order. To connect to the measured absence of order, the authors introduce extended-range couplings: 'To estimate realistic values of J2 and J3 for K2IrO3, we turn towards the recent experimental observations,' and choose J2~2, J3~2-3 to reproduce the 30 K specific-heat maximum. The resulting point lies near the zigzag/3-fold SDW boundary, and the inferred TN<2 K follows from that tuned location. The headline statement that large Gamma couplings 'imply lack of magnetic ordering consistent with the experiments' is therefore not an ab initio prediction; it is achieved by fitting longer-range couplings to the same data being explained. A compounding limitation, acknowledged in the same section, is that real KxIryO2 samples are non-stoichiometric, with K-layer vacancies and Ir/K occupancy at hexagon centers, and that such disorder can 'significantly influence' the magnetic couplings. Both the fitted J2/J3 and the idealized C3 structure must hold for the comparison to experiments to stand.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper reports ab initio quantum-chemistry (MRCI) calculations of the nearest-neighbor magnetic couplings for the proposed C3-symmetric honeycomb iridate K2IrO3, finding unusually large off-diagonal exchange terms, Γxy≈5.2 meV and Γyz≈−8.9 meV, roughly an order of magnitude larger than in Na2IrO3. The authors attribute the enhancement to the C3 point-group symmetry at the Ir sites, which constrains the O-O link orientations, and support this with reduced-cluster twist-angle calculations. Exact diagonalization of the fully anisotropic K-J-Γ model yields a phase diagram with a Γ-driven spin-liquid region, but the pure MRCI parameter set falls in a 3-fold SDW ordered phase. By adding extended-range Heisenberg couplings J2~2 meV and J3~2-3 meV chosen to reproduce the experimental specific heat, the system is placed near the zigzag/3-fold SDW boundary, with a calculated ordering temperature below 2 K, which the authors take to be consistent with the absence of magnetic order in KxIryO2 samples.","tokens_in":17284,"tokens_out":9928,"duration_ms":97825,"significance":"The ab initio finding of large off-diagonal exchange couplings in a C3-symmetric honeycomb iridate is timely and, if correct, identifies a new parameter regime for spin-liquid candidates. The ED phase diagram for the fully anisotropic K-J-Γ model is a useful reference for future studies. The authors are also transparent about the qualitative nature of the twist-angle tests and the disorder in the real samples. However, the central claim that the large Γ terms suppress magnetic ordering is not directly demonstrated by the ab initio model; the paper's experimental link rests on fitted J2 and J3 values. This limits the strength of the conclusions as they stand.","major_comments":[{"comment":"The pure MRCI nearest-neighbor parameter set (Γxy=5.2 meV, Γyz=−8.9 meV) is stated in Section II.D to lie in the 3-fold SDW ordered phase in Fig. 3(a). The abstract's claim that \"large quantum fluctuations imply lack of magnetic ordering consistent with the experiments\" is therefore not a prediction of the ab initio NN model. The no-order conclusion is achieved only after adding J2~2 meV and J3~2-3 meV, which are explicitly fitted to reproduce the experimental specific-heat maximum at ~30 K. This makes the explanation of the absence of order circular with respect to the data it is meant to explain. To make the central claim load-bearing, the authors should either obtain J2 and J3 from independent ab initio calculations or demonstrate that the MRCI Γ values suppress order for a robust range of J2,J3 values without fitting to the target observable.","section":"II.D, Fig. 3(a)"},{"comment":"The comparison to experiment depends on the idealized stoichiometric C3-symmetric K2IrO3 structure, while the synthesized KxIryO2 samples are acknowledged to be non-stoichiometric with K-layer vacancies and Ir/K occupancy at hexagon centers, and the paper notes that such disorder can significantly influence the magnetic couplings. Since the headline result is that the calculated couplings are consistent with the measured absence of order, the authors should quantify the effect of the dominant disorder configurations on the NN couplings (for example, by additional cluster calculations with representative local environments) or explicitly restrict the experimental comparison to the idealized end member. Without this, the structural idealization remains an unquantified caveat that could invalidate the application of the calculated Γ values to the measured compound.","section":"II.D, final paragraph"},{"comment":"The statement that \"the zigzag state is the most probable ground state\" is made without presenting the energy difference between the zigzag and 3-fold SDW states at the fitted J2,J3 point, nor any finite-size scaling analysis. Given that the point is chosen to be near the phase boundary, this claim needs quantitative support to justify the conclusion that no long-range order is expected. The paper should also discuss how robust the TN<2 K result is to small variations of J2 and J3 around the chosen values.","section":"II.D, phase diagram discussion"}],"minor_comments":[{"comment":"The MRCI couplings are reported without error bars or an estimate of the uncertainty from the mapping procedure; given the sensitivity of the phase diagram to the coupling values, a statement of the expected accuracy is needed.","section":"Table I"},{"comment":"The wording \"large quantum fluctuations imply lack of magnetic ordering\" overstates the result because Fig. 3(a) shows the pure MRCI point in an ordered phase; the role of extended couplings should be acknowledged explicitly in the abstract.","section":"Abstract and Introduction"},{"comment":"The comparison with the couplings quoted in Ref. [23] is only qualitative (\"similar\"); a table listing the values from that work would allow the reader to assess the claimed agreement.","section":"II.B, Table I"},{"comment":"The filled symbols in Fig. 2 correspond to the reduced-cluster calculations, not the full embedded-cluster values in Table I; the caption should state this distinction explicitly to avoid confusion.","section":"Fig. 2"},{"comment":"The computed Curie-Weiss temperature θ≈−135 K differs from the experimental −180 K by about 25%; describing this as \"somewhat smaller\" underplays the discrepancy, and its implications for the fitted model should be discussed.","section":"II.D, Fig. 3(e)"},{"comment":"The specific-heat fits in Fig. 3(c,d) use a 12-site cluster while the phase diagram in Fig. 3(b) is computed on a 24-site cluster; the effect of this cluster-size mismatch on the extracted J2 and J3 values should be commented on.","section":"II.D and Methods D"}],"recommendation":"major_revision","confidential_remarks":"The paper contains an acknowledged circular step: the no-order conclusion is obtained only after fitting J2 and J3 to the specific heat that the paper then claims to explain. The ab initio NN result (large Γ) is interesting in its own right, but the experimental link is not yet established. The structural disorder and the θ mismatch further weaken the claim of consistency with experiments. A major revision that either computes J2,J3 from first principles or systematically explores the extended-coupling space would strengthen the paper considerably."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The genuinely new piece here is the MRCI calculation for the C3-symmetric K2IrO3 model: Gamma_yz = -8.9 meV and Gamma_xy = 5.2 meV, roughly ten times the Na2IrO3 values, with smaller K and J. The authors trace this to the 120-degree arrangement of O-O links imposed by the large K ions, and the twist-angle scan in Fig. 2 backs that up with a clear trend. That is a real result, and it gives the honeycomb-iridate community a concrete alternative to the usual Kitaev-Heisenberg tuning: a structural mechanism for making off-diagonal exchange dominant. The fully anisotropic K-J-Gamma phase diagram is also useful, especially the spin-liquid region for negative Gamma and the Kitaev-SL sliver near the FM-stripy boundary.\n\nThe soft spot is exactly where the stress-test lands. In Fig. 3(a), the pure MRCI NN parameter set sits in the 3-fold SDW ordered phase, not in a spin-liquid or even a frustration boundary. The no-order conclusion only emerges after adding J2 and J3 chosen to reproduce the measured 30 K specific-heat maximum. So the abstract's claim that large Gamma implies lack of magnetic ordering is not an ab initio prediction; it is a model result with extended-range couplings fitted to the very data being explained. The authors are transparent about this in Section II.D, but the abstract and conclusions overstate the implication. The second issue is the structural idealization: the synthesized KxIryO2 is non-stoichiometric with K vacancies and Ir/K occupancy at hexagon centers, and the authors themselves note disorder can significantly influence couplings. Both the fitted J2/J3 and the idealized C3 structure must hold for the experimental comparison to stand.\n\nWhat holds up is the core chemistry. The MRCI method is established, the embedding is described in enough detail to be reproducible, and the agreement with the independent couplings in Ref. [23] is a genuine check. The twist-angle calculations are explicitly qualitative but the direction of the effect is consistent. The citation pattern is honest, and the limitations are acknowledged rather than hidden.\n\nThis paper deserves a serious referee. The quantum chemistry result and the C3 mechanism are the value; the experimental narrative needs to be decoupled from the fitted J2/J3, and ideally the authors would add some estimate of how disorder affects the computed Gamma's. I would recommend peer review with a request to rewrite the abstract and conclusions to separate prediction from fitting.","headline":"The MRCI result is genuinely new and the C3 mechanism is worth taking seriously, but the paper's experimental spin-liquid claim is weaker than the abstract suggests because the no-order conclusion comes from fitted J2/J3, not from the ab initio couplings alone.","tokens_in":17849,"tokens_out":1875,"would_cite":true,"duration_ms":22665,"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":"In C3-symmetric K2IrO3, off-diagonal exchange dominates magnetism and stabilizes a spin liquid.","keywords":["spin liquid","honeycomb lattice","iridates","off-diagonal exchange","Kitaev model","C3 symmetry","magnetic frustration","exact diagonalization"],"falsifier":"A stoichiometric K2IrO3 crystal that shows long-range magnetic order below about 2 K in neutron or muon-spin-rotation measurements would contradict the predicted suppression of ordering; alternatively, a structural refinement showing that the local O-O twist angle deviates by more than a few degrees from the 120-degree C3 arrangement, together with measured $\\Gamma$ couplings an order of magnitude smaller, would undercut the symmetry-origin claim.","tokens_in":16812,"feed_emoji":"🌀","tokens_out":14698,"duration_ms":127520,"temperature":0.7,"pith_summary":"The paper tries to establish that in the C3-symmetric honeycomb iridate K2IrO3, the nearest-neighbor magnetic couplings are dominated by off-diagonal exchange terms (the $\\Gamma$ couplings, which mix different spin components on neighboring sites), with $\\Gamma_{xy}\\approx 5.2$ meV and $\\Gamma_{yz}\\approx -8.9$ meV, roughly ten times larger than in Na2IrO3. It traces this dominance to the threefold rotational symmetry at each iridium site, which locks the oxygen-oxygen edges into a 120-degree arrangement that constrains the exchange paths. The consequence is a strongly anisotropic and frustrated magnet in which long-range order is suppressed, matching the absence of magnetic order observed in experiments down to 1.8 K. Exact diagonalization of the fully anisotropic $K$-$J$-$\\Gamma$ model then shows that negative off-diagonal couplings stabilize a spin-liquid state over a wide parameter region, even when the Kitaev coupling is ferromagnetic. If true, this provides a symmetry-based alternative to the usual strategy of tuning the Kitaev-to-Heisenberg ratio in the search for spin liquids.","feed_headline":"Off-diagonal exchange keeps K2IrO3 from ordering","feed_subtitle":"New calculations trace the compound's missing magnetism to C3 symmetry and open a new route to spin liquids.","key_machinery":"The carrying object is the fully anisotropic nearest-neighbor spin Hamiltonian $H_{ij}=J\\,\\tilde{\\mathbf{S}}_i\\cdot\\tilde{\\mathbf{S}}_j + K\\,\\tilde{S}_i^\\gamma \\tilde{S}_j^\\gamma + \\sum_{\\alpha\\neq\\beta}\\Gamma_{\\alpha\\beta}(\\tilde{S}_i^\\alpha \\tilde{S}_j^\\beta+\\tilde{S}_i^\\beta \\tilde{S}_j^\\alpha)$, written in a local Kitaev frame for each Ir-Ir bond, with the $z$ axis perpendicular to each Ir2O2 plaquette. The couplings are obtained by mapping spin-orbit multi-reference configuration-interaction data onto the lowest four pseudospin states of an Ir2O10 dimer. The structural mechanism is the C3 point-group symmetry at each Ir site: the projections of O-O links on the honeycomb plane are arranged at 120 degrees, a constraint imposed by the large interlayer K ions, and the deviation from this arrangement is quantified by a single twist angle $\\varphi$. In the exact-diagonalization analysis, the Kitaev-type spin liquid is identified through the hexagonal plaquette operator, whose expectation value is large only in the spin-liquid region of the $\\Gamma_{xy}$-$\\Gamma_{yz}$ plane.","core_discovery":"The central discovery is that K2IrO3, taken in its proposed C3-symmetric structure, is a $\\Gamma$-dominated rather than a Kitaev-dominated magnetic material. Spin-orbit multi-reference configuration-interaction calculations on the Ir2O10 cluster yield nearest-neighbor couplings $K=-6.3$ meV, $J=1.3$ meV, $\\Gamma_{xy}=5.2$ meV, and $\\Gamma_{yz}=-8.9$ meV, with the off-diagonal terms about ten times larger than the corresponding values in Na2IrO3. The paper identifies the origin in the C3 point-group symmetry: the in-plane projections of the O-O edges on neighboring Ir2O2 plaquettes are locked at 120 degrees by the large interlayer K ions, and rotating these edges away from that arrangement (a twist angle $\\varphi$) shrinks the $\\Gamma$ terms and grows $K$. Exact diagonalization of the resulting fully anisotropic $K$-$J$-$\\Gamma$ Hamiltonian on a 24-site cluster places the pure ab initio parameter set near a boundary between a threefold spin-density-wave state and a zigzag state, and adding small second- and third-neighbor Heisenberg couplings reproduces the broad specific-heat maximum near 30 K and the finite $C/T$ down to 1.8 K seen in experiment.","pith_inferences":["If the C3 oxygen-edge arrangement is generic across the KxIryO2 family, then the whole family, not just the K2IrO3 end member, should be $\\Gamma$-dominated and magnetically disordered; this is a testable family-wide prediction the paper only hints at.","The twist-angle sweep suggests a concrete control knob: epitaxial strain or K-vacancy disorder that rotates O-O links by a few degrees should shrink the $\\Gamma$ couplings and restore ordered phases, which could be checked by combining structural refinement with magnetic measurements on the same samples.","One could extend the exact-diagonalization analysis to a distribution of $\\Gamma$ values representing the local disorder noted in Section II.D; such a distribution, rather than the single clean parameter set, may be what actually produces the gapless spin-liquid-like response seen in the synthesized material.","The sign tendency seen in the calculated phase diagram (negative $\\Gamma$ favors spin liquid, positive $\\Gamma$ favors ferromagnetism) gives a screening criterion: look for C3-symmetric honeycomb iridates whose off-diagonal exchange is negative, perhaps by choosing interlayer cations large enough to lock the 120-degree oxygen arrangement."],"forward_implications":["If the calculated couplings are correct, K2IrO3 is a $\\Gamma$-dominated honeycomb magnet in which magnetic order is suppressed below 2 K, explaining the experimentally observed absence of ordering and spin freezing down to 1.8 K.","A spin liquid can be stabilized for ferromagnetic Kitaev coupling $K<0$ when the off-diagonal couplings are negative, so tuning $\\Gamma$ rather than $K/J$ becomes a practical design route.","The ferromagnet-to-Kitaev-spin-liquid-to-stripy sequence known from the $K$-$J$ model reappears in the $\\Gamma_{xy}$-$\\Gamma_{yz}$ plane, so off-diagonal couplings can effectively control the ratio $K/J$.","The pure ab initio couplings place the system at a competing threefold spin-density-wave/zigzag boundary, implying that modest changes in environment, strain, or stoichiometry could switch the ground state between different ordered states."],"supporting_citations":[{"why":"Provides the proposed C3-symmetric structural model of K2IrO3 on which all calculations are based.","marker":"[23]"},{"why":"Supplies the experimental susceptibility and specific-heat data showing no magnetic order down to 1.8 K, the observations the calculations are compared with.","marker":"[24]"},{"why":"Establishes the mapping scheme used to extract the J, K, and Gamma couplings from spin-orbit quantum-chemistry data.","marker":"[12]"},{"why":"Gives the Na2IrO3 couplings used as the baseline that makes the roughly tenfold larger Gamma claim concrete.","marker":"[33]"},{"why":"Provides the K-J phase diagram whose ferromagnet-Kitaev-spin-liquid-stripy path is recovered at finite Gamma.","marker":"[5]"},{"why":"Shows that a K-Gamma honeycomb model can host a spin liquid, the result extended here to fully anisotropic Gamma terms.","marker":"[26]"},{"why":"Defines the hexagonal plaquette operator used to identify the Kitaev-type spin liquid in the exact-diagonalization phase diagram.","marker":"[48]"}],"fun_headline_variants":["C3 symmetry magnifies off-diagonal exchange, kills order","Iridate spin liquid via giant Gamma couplings","Large off-diagonal couplings stabilize iridate spin liquid","K2IrO3: Gamma-driven frustration yields spin liquid","C3 iridate's big Gamma terms open spin liquid route"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the real synthesized KxIryO2 material is faithfully represented by the idealized stoichiometric C3-symmetric K2IrO3 structure; the paper itself notes that the samples are non-stoichiometric, with K-layer vacancies and Ir/K occupancy at hexagon centers, and that such disorder can significantly change the magnetic couplings.","fun_headline_variants_meta":{"raw":{"variants":["C3 symmetry magnifies off-diagonal exchange, kills order","Iridate spin liquid via giant Gamma couplings","Large off-diagonal couplings stabilize iridate spin liquid","K2IrO3: Gamma-driven frustration yields spin liquid","C3 iridate's big Gamma terms open spin liquid route"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000567,"raw_usage":{"total_tokens":2726,"prompt_tokens":1024,"completion_tokens":1702,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":640,"completion_tokens_details":{"reasoning_tokens":1619}},"tokens_in":640,"tokens_out":1702,"duration_ms":14206,"temperature":1.0,"reasoning_tokens":1619,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T04:54:21.762771+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A stoichiometric K2IrO3 crystal that shows long-range magnetic order below about 2 K in neutron or muon-spin-rotation measurements would contradict the predicted suppression of ordering; alternatively, a structural refinement showing that the local O-O twist angle deviates by more than a few degrees from the 120-degree C3 arrangement, together with measured $\\Gamma$ couplings an order of magnitude smaller, would undercut the symmetry-origin claim.","supporting_citations":[{"cited_title":"Yadav, R","cited_arxiv_id":null,"evidence_quote":"Provides the proposed C3-symmetric structural model of K2IrO3 on which all calculations are based."},{"cited_title":"Koitzsch, C","cited_arxiv_id":null,"evidence_quote":"Supplies the experimental susceptibility and specific-heat data showing no magnetic order down to 1.8 K, the observations the calculations are compared with."},{"cited_title":"Abragam and B","cited_arxiv_id":null,"evidence_quote":"Gives the Na2IrO3 couplings used as the baseline that makes the roughly tenfold larger Gamma claim concrete."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the K-J phase diagram whose ferromagnet-Kitaev-spin-liquid-stripy path is recovered at finite Gamma."},{"cited_title":"Quantum Spin Liquid in a depleted triangular lattice Iridate K$_x$Ir$_y$O$_2$","cited_arxiv_id":"1908.08475","evidence_quote":"Shows that a K-Gamma honeycomb model can host a spin liquid, the result extended here to fully anisotropic Gamma terms."},{"cited_title":"Fuentealba, H","cited_arxiv_id":null,"evidence_quote":"Defines the hexagonal plaquette operator used to identify the Kitaev-type spin liquid in the exact-diagonalization phase diagram."}],"review_version":1}