{"id":"43f57301-1f08-440b-a0fa-783bfff2eaec","arxiv_id":"2509.09903","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"By topochemical potassium deintercalation, KxNi4S2 is continuously tuned from a topological Dirac metal (x=1) to a flat-band-induced antiferromagnet (x=0), with a Z2 index of 1;(000) and TN up to 10.1 K.","lead":"KxNi4S2, a layered nickel sulfide, can have its potassium content removed in a controlled way, shifting its electronic behavior from a high-mobility Dirac metal to an antiferromagnet with heavy, flat-band electrons. The work demonstrates a single crystalline platform where two qualitatively different quantum states can be tuned continuously, which is rare in bulk materials.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The x=0 antiferromagnetic order is not microscopically confirmed: NPD was collected only for x=1, no heat-capacity lambda anomaly is seen, and the susceptibility-based assignment relies on subtracting a metallic Ni impurity. The flat-band causality also depends on an unvaried Hubbard U=5 eV.","rationale":"The paper makes a compelling case for tunable electronic structure: the high mobility, large MR, Hall data, enhanced Sommerfeld coefficient, and DFT band evolution all support a Fermi-level shift with K deintercalation. The reader's weakest-assumption identification is exactly the load-bearing point: the AFM order at x=0 is inferred from magnetization with an impurity subtraction and is not confirmed by a microscopic probe. I agree with this assessment. My own reading of the manuscript finds no internal inconsistency, but the absence of NPD for x=0 and the explicit lack of a heat-capacity anomaly at TN leave the intrinsic-AFM claim open. The U=5 eV dependence of the flat band adds a second, but subordinate, vulnerability. These are addressable concerns, not fundamental errors, so the CONDITIONAL verdict is appropriate and should remain unchanged.","tokens_in":17956,"tokens_out":5113,"duration_ms":53402,"concrete_test":"Collect neutron powder diffraction on the x=0 (Ni2S) sample at T=1.5 K and T=20 K (below and above the claimed TN≈10.1 K) and search for additional magnetic Bragg peaks; index any peaks to the Ni2S lattice or to an impurity phase. A null result would demonstrate the absence of long-range magnetic order and directly falsify the intrinsic-AFM claim. If magnetic peaks are found, also recompute the x=0 band structure with U=0, 3, 4, 5, and 6 eV to test whether the flat band remains near the Fermi level.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim requires the AFM state at x=0 (and x=0.7) to be intrinsic to the KxNi4S2 lattice and induced by the flat band approaching the Fermi level. This is not settled. The manuscript reports neutron powder diffraction only for x=1, showing no magnetic reflections between 260 K and 5 K (Fig. S12); no NPD is presented for x=0, where TN=10.1 K is claimed. The authors explicitly state that 'the heat capacity reveals no clear 2nd order phase transition across the TN' (just after Fig. 4g), so the AFM assignment rests on the susceptibility kink, FC/ZFC splitting, and AC susceptibility alone. The raw susceptibility contains a large temperature-independent term attributed to metallic Ni impurity, which is subtracted using x=0.7 as a constant χ0 baseline (Supplemental Note S4). If x=0 crystals contain more Ni impurity or a secondary nickel-sulfide magnetic phase, the observed transition could be extrinsic, decoupling it from the band-structure picture. Moreover, the DFT flat-band position for x=0 is computed with U=5 eV on Ni d orbitals, and no U-sensitivity test is reported; the 'flat-band-induced' causality is therefore not robust to the chosen Hubbard U. If the AFM is extrinsic, the switchable Dirac-metal / flat-band-AFM claim loses its experimental foundation.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports layered nickel subsulfide KxNi4S2 (0 ≤ x ≤ 1) as a single bulk system whose Fermi level can be shifted by topochemical K deintercalation between a Dirac-cone-dominated regime (x = 1) and a flat-band-dominated regime (x = 0). DFT calculations show Dirac cones near the Fermi level at x = 1 and flat bands approaching the Fermi level at x = 0, and report a nontrivial Z2 index of 1;(000) for KNi4S2. Transport, heat capacity, and magnetization measurements on single crystals are presented: T-linear resistivity, large magnetoresistance, carrier mobility up to 1471 cm2/Vs at x = 0.7 decreasing to 9.4 cm2/Vs at x = 0, a Sommerfeld coefficient increasing from 32.9 to 76 mJ/mol K2, and a susceptibility kink at TN = 8.6–10.1 K attributed to canted antiferromagnetism. The paper concludes that the ground state can be fine-tuned from a non-magnetic topological Dirac metal to a flat-band-induced antiferromagnetic metal.","tokens_in":18262,"tokens_out":4321,"duration_ms":53884,"significance":"If the central claim holds, this is a valuable platform: a single crystalline material in which a simple chemical knob (K content) continuously moves the Fermi level between a Dirac-cone-dominated state and a flat-band-dominated state with an accompanying magnetic transition, without relying on kagome/honeycomb lattices or Moiré engineering. The study combines independent experimental inputs—transport, Hall effect, heat capacity, and magnetization—that are not fitted to the DFT band positions, and the topochemical deintercalation provides a reproducible compositional series. The main shortfall is that the 'topological Dirac metal' label rests on a DFT Z2 calculation, and the 'flat-band-induced antiferromagnet' endpoint currently lacks microscopic magnetic confirmation for x = 0; the AFM assignment also depends on a single Hubbard U value. These gaps are load-bearing for the headline claims.","major_comments":[{"comment":"The intrinsic bulk antiferromagnetic order for x = 0 and x = 0.7 is not microscopically established. Neutron powder diffraction is presented only for x = 1 (Fig. S12), which shows no magnetic reflections; no NPD or other magnetic diffraction is shown for x = 0, where TN = 10.1 K is the key endpoint. The authors also state that the heat capacity shows no clear second-order phase transition across TN, and the susceptibility kink is observed after subtracting a temperature-independent Ni-impurity baseline calibrated against x = 0.7 (Supplemental Note S4). The AC susceptibility excludes a canonical spin glass, but it does not exclude an extrinsic secondary nickel-sulfide phase or a larger Ni impurity content in the x = 0 crystals. Because the flat-band-induced AFM is one of the two endpoints of the central claim, this needs direct magnetic diffraction (NPD or resonant X-ray scattering) for x","section":"Flat-Bands-Induced Magnetism, Fig. 4 and Supplemental Note S4"},{"comment":"The 'flat-band-induced' causality rests on the DFT position of the flat band at x = 0, computed with a single value of the Hubbard U = 5 eV on Ni d orbitals, with no sensitivity test reported. The proximity of the flat band to EF (Eflat = -82 meV for x = 0) and the stability of the AFM configuration (Fig. S15) both depend on the correlation correction. The authors should vary U over a reasonable range (e.g., 3–7 eV) or cross-check with another method (hybrid functional or DFT+DMFT) and report the resulting EF - Eflat and the AFM/FM energy difference. Without this, the flat-band-induced AFM mechanism is not robust and could be an artifact of the chosen U.","section":"Theoretical methods and Fig. 2"},{"comment":"The paper repeatedly labels KNi4S2 a 'topological Dirac metal' and emphasizes the nontrivial Z2 index in the abstract. The Z2 calculation is a valid theoretical result, but the experimental evidence (high mobility, large MR, T-linear resistivity) is consistent with a Dirac metal and does not probe the topological invariant. No ARPES, surface-state transport, or quantum-oscillation experiment is presented. Since the topological classification is a headline claim, the authors should either temper the wording to 'Dirac metal with a predicted nontrivial Z2 index' or provide an experimental probe of the topological surface state; they should also report the stability of the Z2 index with respect to U and to the magnetic configurations considered.","section":"Topological Dirac Metal, Fig. 3 and Fig. S3"}],"minor_comments":[{"comment":"There is a typo in the Methods section: 'heat capacuty measurements' should be 'heat capacity measurements.'","section":"Methods"},{"comment":"The notation is inconsistent: 'K0Ni4S2 (x = 0)' is used, while the compound is referred to as Ni2S elsewhere. Please use one convention throughout.","section":"Photoemission Yield Spectroscopy"},{"comment":"The magnetism section mentions a TN ~ 10 K feature in the x = 1 specimen but attributes it to a minor K-deintercalated phase, labeling the sample x = 1-δ. However, transport and heat capacity for x = 1 are reported without this caveat. The possible δ in nominally x = 1 crystals should be stated in the main text when presenting those data, since it affects the interpretation of the x = 1 endpoint.","section":"Fig. 4 and x = 1"},{"comment":"The Curie-Weiss fits are described only briefly in the main text, with no fit residuals or uncertainty estimates. Given the impurity subtraction and the limited fitting range (above 200 K), the fitted θCW and μeff should be reported with errors and the fit range justified.","section":"Fig. 4h and Curie-Weiss analysis"},{"comment":"The impurity subtraction procedure is described only in the Supplemental Information but is central to the AFM claim. A concise description of the baseline removal should be included in the main text or at least summarized with the key figure (Fig. S11) referenced in the main text.","section":"Supplemental Note S4"}],"recommendation":"major_revision","confidential_remarks":"The paper reports a potentially important material system and the experimental/computational work is largely sound, but the headline claims currently outrun the evidence: the AFM state at x = 0 is not microscopically confirmed, and the topological/flat-band assignments depend on a single DFT Hubbard U. These are fixable, in my view, with additional experiments (magnetic diffraction for x = 0, impurity quantification) and a U-sensitivity analysis. I recommend major revision rather than rejection. The fit of the paper to a broad materials science journal is good, but the authors should be asked to either strengthen or soften the 'topological Dirac metal' and 'flat-band-induced AFM' wording in the abstract and conclusions."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: the core claim holds up. A single layered system where K-deintercalation continuously moves the Fermi level between Dirac-cone-dominated and flat-band-dominated regimes, with transport, Hall, heat capacity, and susceptibility all trending the same way, is a genuine platform advance. That part is solid.\n\nWhat's new: the continuous tuning across KxNi4S2, the Z2=1;(000) calculation for KNi4S2, the high mobility (1471 cm2/Vs) at the K-rich end, and the clear mobility/magnetism evolution. The MO analysis linking Ni-Ni bonding to Dirac cones is a useful design hint. Credit to the authors for transparency: they state no heat capacity lambda anomaly, show NPD only for x=1 with no magnetic reflections, and attribute the x=1 AFM-like signal to minor deintercalated phases.\n\nSoft spots, in proportion. First, \"topological Dirac metal\" rests entirely on a DFT Z2 index with WannierTools and a chosen Hubbard U=5 eV; there is no ARPES, no surface-state probe, no quantum oscillation signature. The transport is consistent with high mobility but does not pin topology. The text hedges by saying \"supported by first-principles calculations,\" but the title and abstract overreach.\n\nSecond, \"flat-band-induced antiferromagnet\" is inferred, not demonstrated. No NPD on x=0, where TN=10.1 K is claimed; only x=1 was measured. The AFM assignment rests on a susceptibility kink, FC/ZFC splitting, AC susceptibility, and Curie-Weiss fits, after subtracting a metallic Ni impurity using x=0.7 as baseline. If x=0 has more Ni impurity, the kink could be extrinsic. The heat capacity shows no clear second-order transition, as the authors concede. The AC frequency-independence and the systematic evolution across the series argue for intrinsic order, so I would not call this fatal, but \"flat-band-induced\" goes beyond what is shown. Also, the flat-band position is computed at a single U value with no sensitivity check.\n\nThese are addressable issues, not fatal flaws. The central tunability result is well-supported, and the paper is honest about its limitations. I would cite the tunability result and bring it to a reading group. A serious editor should send this to peer review rather than desk reject, with a clear request: either soften the topological and flat-band-induced claims to match the evidence, or add the missing probes (NPD on x=0, U variation, ideally ARPES or a direct flat-band probe).","headline":"The continuous K-tunability between Dirac-cone and flat-band regimes is real and well-supported; the 'topological' and 'flat-band-induced' labels outrun the evidence, but the paper deserves peer review.","tokens_in":18873,"tokens_out":2510,"would_cite":true,"duration_ms":29867,"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 single layered material can be chemically tuned between two exotic electronic states: a topological Dirac metal and a flat-band antiferromagnet.","keywords":["Dirac metal","flat-band-induced magnetism","topochemical deintercalation","non-Fermi liquid","topological band structure","layered nickel sulfide","antiferromagnetism","strongly correlated electrons"],"falsifier":"Neutron powder diffraction on fully deintercalated Ni2S below 10 K: if no magnetic Bragg peaks appear with a propagation vector matching the predicted antiferromagnetic configuration, the claimed flat-band-induced antiferromagnetism collapses. Alternatively, ARPES on Ni2S showing the flat band more than roughly 100 meV below the Fermi level would falsify the Fermi-level alignment that drives the argument.","tokens_in":1417,"feed_emoji":"🧲","tokens_out":2156,"duration_ms":68294,"temperature":0.7,"pith_summary":"This paper claims that the layered compound KxNi4S2, by losing potassium through a topochemical deintercalation, continuously moves its Fermi level between two distinct electronic regimes: at x = 1 it is a non-magnetic topological Dirac metal with a nontrivial Z2 index, and at x = 0 it is a metal with a flat band near the Fermi level that develops antiferromagnetic order below about 10 K. Both Dirac cones and flat bands coexist at different energies in the same crystal, without needing a Kagome or honeycomb lattice. If true, this provides a single bulk material platform where a chemical knob—potassium content—switches between massless Dirac fermion physics and flat-band-induced correlated magnetism, and the wide-range Fermi level tuning also opens a route to in-situ control via electrochemical intercalation.","feed_headline":"Removing potassium flips crystal from Dirac metal to flat-band magnet","feed_subtitle":"One layered compound shows both massless and heavy electrons—and a chemical knob switches between their ground states.","key_machinery":"The central object is the Ni9 cluster formed by extensive Ni–Ni bonding inside the Ni4S2 layers. In a molecular-orbital picture, the cluster's HOMO has dz2 character at the center and dx2−y2 on surrounding nickel, while the LUMO reverses the roles, so the protruding dz2 orbitals mimic graphene's pz orbitals and give rise to Dirac cones on a square Ni net—without Kagome or honeycomb geometry. A flat band near the Fermi level arises from the dominant Ni–Ni dx2−y2 and Ni–S bonding. The chemical knob is topochemical potassium deintercalation, which shifts the Fermi level over hundreds of meV and thereby selects which electronic feature dominates.","core_discovery":"The core discovery is a bulk crystalline system in which the ground state can be fine-tuned by potassium deintercalation from a non-magnetic topological Dirac metal (KNi4S2, x=1) to a flat-band-induced antiferromagnetic metal (Ni2S, x=0). First-principles calculations place Dirac cones just above the Fermi level for x=1, with a Z2 invariant of 1;(000) similar to Bi2Se3, and flat bands below the Fermi level; as potassium is removed, the Fermi level drops toward the flat bands. Experimentally, x=0.7 crystals show high Hall mobility (1471 cm2 V−1 s−1) and large magnetoresistance characteristic of Dirac electrons, while x=0 crystals show a two-fold increase in carrier density, a roughly 150-fold","pith_inferences":["A direct testable extension would be angle-resolved photoemission (ARPES) on both end members: if the flat band is not found near the Fermi level for x=0, or the Dirac cone is not visible for x=1, the DFT-based picture would need revision.","The flat-band-induced magnetism argument implicitly predicts that intermediate x values should show intermediate Néel temperatures; this could be tested with the same deintercalation method and would sharpen the Fermi-level-to-magnetism link.","If the canted antiferromagnetic order is intrinsic, KxNi4S2 at low K may exhibit metamagnetic transitions or spin-flop behavior under magnetic field, a regime the paper does not explore but which would clarify the magnetic ground state.","The coexistence of Dirac cones, flat bands, and non-Fermi liquid transport suggests that tuning x may access quantum criticality without external pressure or doping; checking whether the linear resistivity continues to lower temperatures at optimal x would extend the claim."],"forward_implications":["If correct, KxNi4S2 is a rare bulk material where Dirac cone and flat band physics coexist without Kagome or honeycomb structure, providing a natural laboratory for studying the interplay of massless and heavy electrons.","Potassium content acts as a continuous magnetic switch: removing potassium fills the flat band and triggers antiferromagnetic order, so the same crystal can be tuned between non-magnetic and magnetic ground states by chemical means alone.","The linear-in-T resistivity observed across all compositions indicates persistent strange-metal behavior in both regimes, implying strong correlations that survive the transition between Dirac-dominated and flat-band-dominated states.","Demonstrating ex-situ topochemical control of the Fermi level establishes a concrete pathway for electrochemical in-situ tuning of quantum materials, potentially enabling reconfigurable electronics and multi-state memory devices.","The absence of superconductivity up to 10 GPa in KNi4S2 narrows the expected correlated phases, suggesting that pressure tuning first acts on lattice degrees of freedom before any electronic instability."],"fun_headline_variants":["Chemical switch toggles between Dirac metal and flat-band antiferromagnet","Remove potassium to turn a Dirac metal into a flat-band magnet","One layered crystal: tune from topological Dirac to antiferromagnet","Potassium deintercalation flips ground state from Dirac to flat-band","A single material: Dirac metal and flat-band magnet, switchable"],"cache_read_input_tokens":20096,"weakest_assumption_plain":"The antiferromagnetic order seen in susceptibility is intrinsic to the KxNi4S2 lattice and caused by the flat band, not by metallic nickel impurities or a secondary nickel sulfide phase, and the DFT-predicted flat band position (with U = 5 eV) is accurate enough to place it near the Fermi level for x = 0.","fun_headline_variants_meta":{"raw":{"variants":["Chemical switch toggles between Dirac metal and flat-band antiferromagnet","Remove potassium to turn a Dirac metal into a flat-band magnet","One layered crystal: tune from topological Dirac to antiferromagnet","Potassium deintercalation flips ground state from Dirac to flat-band","A single material: Dirac metal and flat-band magnet, switchable"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00075,"raw_usage":{"total_tokens":3189,"prompt_tokens":769,"completion_tokens":2420,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":513,"completion_tokens_details":{"reasoning_tokens":2340}},"tokens_in":513,"tokens_out":2420,"duration_ms":19426,"temperature":1.0,"reasoning_tokens":2340,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-04T18:29:05.567021+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Neutron powder diffraction on fully deintercalated Ni2S below 10 K: if no magnetic Bragg peaks appear with a propagation vector matching the predicted antiferromagnetic configuration, the claimed flat-band-induced antiferromagnetism collapses. Alternatively, ARPES on Ni2S showing the flat band more than roughly 100 meV below the Fermi level would falsify the Fermi-level alignment that drives the argument.","supporting_citations":[],"review_version":1}