{"id":"d5de4617-f8e1-425b-be06-99fde9ce9660","arxiv_id":"2412.07190","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Dense-k-mesh DFT calculations find a nonmagnetic ground state and weak spin-orbit effects in AV3Sb5, attributing a reported ferromagnetic state to coarse Brillouin zone sampling.","lead":"Using density functional theory, the authors show that the kagome metal AV3Sb5 has a nonmagnetic ground state and weak spin-orbit coupling, and that an earlier claim of ferromagnetism came from an insufficient k-point mesh. The result matters because it narrows the proposed explanations for charge density waves and superconductivity in this material family.","discovery_kind":"replication","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 'definitively NM' claim rests on only two k-meshes; without a convergence study, a denser mesh could restore magnetism.","rationale":"The reader's verdict was CONDITIONAL, and I agree with the condition: the central claim of a definitively nonmagnetic pristine phase rests on a single 'dense' k-mesh (18×18×12) without a systematic convergence check. My stress-test sharpens this into a concrete, testable failure mode: if the magnetic energy surface is not converged at 18×18×12, the FM state could reappear at even denser meshes, directly undermining the paper's explanation that the coarse 5×5×3 mesh alone creates the FM artifact. The paper gives no intermediate meshes, no smearing dependence, and no AFM search, so the phrase 'definitively NM' is overstrong relative to the evidence. However, this is a gap in support rather than an internal inconsistency or a demonstrated error: the direction of the mesh dependence (coarse mesh raises DOS(EF) and stabilizes FM, dense mesh lowers it) is plausible, the reproduction of the 5×5×3 FM state is a good control, and the experimental comparison with Pauli paramagnetism and ARPES is creditable independent evidence. The proposed check—a denser-mesh spin-polarized and FSM calculation with multiple magnetic initializations—would settle whether the concern lands. I do not see a reason to reject or to tighten the verdict beyond the reader's conditional accept; the condition should explicitly include the convergence test and AFM search. I chose 'partial' agreement because the reader's weakest_assumption emphasizes the PBE functional and absence of Hubbard U, whereas I view the k-mesh convergence and magnetic order search as the more decisive and immediately testable issue; the functional question is secondary given the paper's PBE-level comparative claim against Hasan et al.","tokens_in":10033,"tokens_out":4305,"duration_ms":75935,"concrete_test":"Perform spin-polarized PBE calculations for CsV3Sb5 with k-meshes 24×24×16, 30×30×18, and 36×36×24 using the same PAW potentials, 500 eV cutoff, and DFT-D3, with at least two smearing widths (e.g., 0.05 and 0.2 eV) and initial FM, checkerboard AFM, and noncollinear 120° spin configurations. Also rerun the fixed-spin-moment E(M) curves at 30×30×18. If any finite-M minimum appears or an AFM/noncollinear state is energetically favored at a denser mesh, the 'definitively NM' claim is falsified; if E(M) remains monotonic in M and DOS(EF) changes by less than 2% from 18×18×12 to 30×30×18, the concern is settled in the paper's favor.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central conclusion—that the pristine phase is 'definitively NM' and the FM state of Hasan et al. is purely a 5×5×3 k-mesh artifact—depends on the 18×18×12 mesh being sufficiently converged for the magnetic energy surface. However, Section III reports only two meshes, with no tetrahedron/smearing convergence or intermediate mesh. For a metal with a sharp van Hove singularity near EF, DOS(EF) can remain sensitive at 18×18×12, and the DOS values in Table I (6.58 vs 10.97 states/eV for CsV3Sb5) show a large mesh dependence without establishing that 18×18×12 is the converged limit. The FSM curves in Fig. 3 show E(M) decreasing monotonically to M=0 at this mesh, but they do not prove the curvature remains positive at yet denser meshes. Moreover, the Stoner criterion values (I·D(EF)=1.82, 1.44, 1.21) are quoted only for the coarse mesh; no dense-mesh product is given, and the Stoner I is read off the self-consistent FM state, leaving the quantitative argument against FM incomplete. The statement that 'any initial configuration of magnetic moments quickly converges to a NM solution' is also verified only for collinear FM starting states, not AFM or noncollinear orders, so the broader 'nonmagnetic pristine phase' claim exceeds the evidence.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports DFT and DFT+SOC calculations for the kagome metals AV3Sb5 (A = Cs, Rb, K). The authors show that the previously reported ferromagnetic (FM) state, obtained with a 5×5×3 k-mesh, is not reproduced with a denser 18×18×12 k-mesh: spin-polarized calculations converge to a nonmagnetic (NM) solution, fixed-spin-moment curves decrease monotonically to M = 0, and the coarse-mesh FM instability is attributed to an artificial enhancement of the density of states at the Fermi level satisfying the Stoner criterion. They further find that spin-orbit coupling (SOC) has a minor effect on the band structure, opening gaps of less than approximately 20 meV at the Dirac points, and that their DFT bands agree well with ARPES data, supporting weak electron correlations. They conclude that the pristine phase is nonmagnetic and that magnetism, SOC, and correlations do not play the decisive role in the CDW physics claimed by Hasan et al.","tokens_in":10424,"tokens_out":6444,"duration_ms":65088,"significance":"If correct, the paper resolves a current controversy by showing that a high-profile theoretical prediction of FM order in AV3Sb5 is a numerical artifact of insufficient k-point sampling. The two-mesh comparison and fixed-spin-moment curves are a useful diagnostic, and the comparison with ARPES and magnetic susceptibility provides external validation. The paper also reinforces the weak-correlation picture for these kagome metals. However, the strength of the conclusion is not fully matched by the evidence: the k-mesh convergence is not established, only one exchange-correlation functional is used, and antiferromagnetic configurations are not considered.","major_comments":[{"comment":"The central claim that the pristine phase is 'definitively NM' is supported only by a comparison between two k-meshes (5×5×3 and 18×18×12), with no systematic convergence study. The density of states at the Fermi level changes dramatically between these meshes (e.g., from 10.97 to 6.58 states/eV for CsV3Sb5 in Section III), which shows that neither mesh is demonstrably converged for this quantity. The fixed-spin-moment curves in Fig. 3 show a monotonic decrease of E(M) for the dense mesh, but they do not rule out the possibility that an even denser mesh or a different smearing/tetrahedron integration scheme would restore a magnetic solution. The authors should provide intermediate k-meshes (e.g., 9×9×6, 12×12×8, 15×15×10) and specify the Brillouin-zone integration method, or soften the 'definitively NM' claim.","section":"§II–III, Fig. 3, Table I"},{"comment":"The statement that 'any initial configuration of magnetic moments quickly converges to a NM solution' is only verified for collinear FM starting states. The abstract's broader claim of a 'nonmagnetic pristine phase' would require testing of antiferromagnetic (AFM) and possibly noncollinear spin configurations, which are not reported. Since the title is limited to the 'absence of ferromagnetic instability,' the authors should either test AFM orders or explicitly restrict their conclusion to the FM instability.","section":"§III, first paragraph; Abstract"},{"comment":"The Stoner product I·D(EF) is quoted only for the 5×5×3 mesh (1.82, 1.44, and 1.21 for CVS, RVS, and KVS, respectively). The corresponding values for the 18×18×12 mesh are not given, so the reader cannot verify that the dense mesh is on the nonmagnetic side of the Stoner criterion. Moreover, the Stoner parameter I is estimated from the exchange splitting and magnetic moment of the self-consistent FM state at the coarse mesh; this quantity may itself be mesh-dependent. Providing D(EF) and I at the dense mesh would complete the quantitative argument.","section":"§III, Stoner criterion paragraph"},{"comment":"The conclusion is based exclusively on the PBE exchange-correlation functional. While PBE is an appropriate choice for comparing with the previous PBE-based claim, the paper's broader statement that the pristine phase 'possesses a NM ground state' (Summary) goes beyond the evidence. A test with PBE+U or a hybrid functional, or at minimum a caveat that the result is functional-dependent, would make the claim robust.","section":"§II, §III"}],"minor_comments":[{"comment":"Typo: 'ad DFT' should be 'and DFT'.","section":"§II"},{"comment":"Typo: 'tough' should be 'though' in the sentence 'rough their predicted magnetic moment...'.","section":"§III"},{"comment":"The definition of exchange splitting is convoluted; it should be stated directly as the difference between the spin-up and spin-down Kohn-Sham eigenvalues at the same k-point or as an average.","section":"Fig. 2 caption"},{"comment":"The Brillouin-zone integration scheme (e.g., Methfessel-Paxton smearing, tetrahedron method with Blöchl corrections) and the smearing width are not specified; this matters for the DOS(EF) values in Table I and the text.","section":"§II"},{"comment":"The placeholder URL for the Supplemental Material (Ref. [31]) should be replaced with the actual link.","section":"References"},{"comment":"The phrase 'definitively NM' is too informal; consider 'nonmagnetic within the present DFT approach' or similar.","section":"§III"}],"recommendation":"major_revision","confidential_remarks":"This manuscript is a direct challenge to a recent PRL. The central message is plausible and the two-mesh comparison is suggestive, but the evidence is not yet conclusive for the strong claims made. The main missing pieces are a k-mesh convergence study and AFM tests; without those, the paper is better framed as a cautionary note than a definitive refutation. I recommend major revision."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe paper you asked about is a computational note that takes direct aim at a PRL by Hasan et al. claiming magnetism in AV3Sb5. The main new result is the diagnosis: a 5×5×3 k-mesh inflates the DOS at EF enough to satisfy the Stoner criterion, while an 18×18×12 mesh gives a nonmagnetic state. The authors reproduce the coarse-mesh FM state, show via fixed-spin-moment curves that the dense mesh has its minimum at M=0, and estimate a Pauli-like susceptibility that lands within a factor of ~2.4 of the measured value. That is solid work, and the qualitative conclusion is almost certainly right: pristine AV3Sb5 is not an intrinsic ferromagnet, and the earlier FM result is a numerical artifact.\n\nThe paper also does well with the SOC question. The SOC-induced gap at the Dirac points is <20 meV, consistent with earlier DFT+SOC studies, and the band structure matches ARPES, which undercuts the strong-correlation story in the same PRL.\n\nThe soft spots are real but mostly minor. The central claim rests on only two k-meshes; no convergence study with an intermediate mesh or a different smearing method is shown. That is the main gap. The wording 'definitively NM' overreaches—with only two meshes, and no search for antiferromagnetic or noncollinear orders, the evidence supports 'nonmagnetic at 18×18×12 within PBE' more strongly than 'definitively NM'. The Stoner parameter is extracted from the self-consistent FM state, which is not the cleanest route, though the FSM curves carry the argument. There is also no shipped input files, so full reproduction requires some effort.\n\nThe stress-test worry that an even denser mesh could bring the magnetism back is speculative. 18×18×12 is already dense for a 12-atom cell, and the FSM curvature at M=0 is smooth. I would not hold that against the paper. The AFM check, though, is a legitimate request for a referee.\n\nAll told, this paper deserves peer review. It addresses a contested experimental and theoretical point, the technical correction is plausible and well-supported, and the requested additions (one convergence curve, a quick AFM scan, softer wording) are all reasonable. I'd recommend accepting it with minor revisions.","headline":"A useful, mostly convincing computational correction to the claimed magnetism in AV3Sb5; the k-mesh artifact explanation holds up, though 'definitively' overreaches the two-mesh evidence.","tokens_in":10852,"tokens_out":3316,"would_cite":true,"duration_ms":38625,"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":"The pristine phase of the kagome metals AV$_3$Sb$_5$ (A = Cs, Rb, K) is nonmagnetic, and a previously reported ferromagnetic state is an artifact of an insufficient k-point mesh.","keywords":["kagome metals","AV3Sb5","ferromagnetic instability","Stoner criterion","spin-orbit coupling","k-point mesh convergence","Pauli paramagnetism","density functional theory"],"falsifier":"A spin-polarized DFT calculation with the same functional but a denser mesh (for example $24\\times24\\times16$) that recovers a ferromagnetic or antiferromagnetic solution with nonzero V moments would falsify the central claim, as would a measurement on ultra-clean single crystals showing an intrinsic Curie-Weiss magnetic susceptibility that cannot be attributed to impurity spins.","tokens_in":9845,"feed_emoji":"🧲","tokens_out":7587,"duration_ms":70061,"temperature":0.7,"pith_summary":"This paper argues that the pristine phase of the kagome superconductors AV$_3$Sb$_5$ (A = Cs, Rb, K) has a nonmagnetic ground state, and that the ferromagnetic state reported in a recent first-principles study is an artifact of an insufficiently dense k-point mesh. With a coarse $5\\times5\\times3$ mesh the authors reproduce the ferromagnetic solution, but a dense $18\\times18\\times12$ mesh makes spin-polarized calculations converge to a nonmagnetic state whose paramagnetic susceptibility matches the Pauli-like response measured in single crystals. They also find that spin-orbit coupling changes the geometry and band structure only slightly, opening a gap of less than about 20 meV at the Dirac points, and that the DFT bands align closely with angle-resolved photoemission data, indicating weak electron correlations. If the conclusion holds, magnetism, strong spin-orbit effects, and strong correlations drop out of the explanation of the charge-density-wave physics in these materials.","feed_headline":"Coarse k-mesh fakes ferromagnetism in kagome metals","feed_subtitle":"Dense sampling shows CsV3Sb5, RbV3Sb5, KV3Sb5 are nonmagnetic; spin-orbit effects stay below 20 meV.","key_machinery":"The key mechanism is the Stoner criterion $I\\cdot D(E_F) > 1$, evaluated using the density of states at the Fermi level and an exchange parameter $I$ estimated from the ratio of exchange splitting to magnetic moment. The paper demonstrates that this criterion is sensitive to Brillouin-zone sampling: the coarse $5\\times5\\times3$ mesh inflates $D(E_F)$ (to 10.97, 8.66, and 7.85 states/eV per unit cell), producing $I\\cdot D(E_F)$ values of 1.82, 1.44, and 1.21 and a spurious ferromagnetic minimum, while the dense $18\\times18\\times12$ mesh gives $I\\cdot D(E_F)$ below 1 and a nonmagnetic minimum. The fixed-spin-moment method is the supporting tool, tracing total energy as a function of magnetization for both meshes and yielding a paramagnetic susceptibility consistent with Pauli paramagnetism.","core_discovery":"The central claim is that the ground state of pristine AV$_3$Sb$_5$ (A = Cs, Rb, K) is nonmagnetic, contradicting a recent DFT+DMFT+SOC study that predicted local moments on V atoms. The paper shows that a $5\\times5\\times3$ k-point mesh overestimates the density of states at the Fermi level, pushing the Stoner product $I\\cdot D(E_F)$ above 1 and stabilizing a ferromagnetic state, whereas an $18\\times18\\times12$ k-mesh yields $D(E_F)$ values of 6.58, 6.42, and 6.16 states/eV per unit cell for CsV$_3$Sb$_5$, RbV$_3$Sb$_5$, and KV$_3$Sb$_5$ and no ferromagnetic solution. Total-energy versus magnetization curves from the fixed-spin-moment method confirm a nonmagnetic minimum with the dense mesh. Spin-orbit coupling produces geometry changes below 0.01 \\AA\\ and a Dirac-point gap smaller than about 20 meV. The DFT band structures reproduce the van Hove singularities and Dirac points measured by ARPES, which the paper interprets as evidence against strong correlations and against the notion that magnetism, spin-orbit coupling, and correlations drive the charge-density wave.","pith_inferences":["The same k-mesh artifact could affect other narrow-band or van Hove metals where a coarse sampling inflates the Fermi-level DOS; a mesh-convergence test should precede any Stoner-based magnetism claim.","The calculated susceptibility (roughly $6.5\\text{--}8.5 \\times 10^{-4}$ emu/mol) overshoots the experimental Pauli value by about a factor of two; a finite-temperature treatment of spin fluctuations within the same framework could test whether that gap closes.","If the nonmagnetic ground state is correct, the anomalous Hall effect and chiral charge order reported in these kagome metals likely arise from nonmagnetic time-reversal-symmetry-breaking mechanisms associated with the CDW itself, rather than from local V moments."],"forward_implications":["Charge-density-wave and superconductivity models for AV$_3$Sb$_5$ should proceed without intrinsic V moments, shifting weight to orbital or loop-current mechanisms.","Published first-principles results for AV$_3$Sb$_5$ obtained with a $5\\times5\\times3$ k-mesh should be re-examined, since the magnetic ground state flips with Brillouin-zone sampling.","The SOC-induced gap at the Dirac points is bounded by about 20 meV, constraining topological or Berry-curvature effects that rely on this splitting.","Standard PBE band structures, without Hubbard U or DMFT corrections, adequately describe the normal-state bands near the Fermi level."],"supporting_citations":[{"why":"The recent theoretical report claiming a ferromagnetic ground state and local V moments; the paper reproduces and then refutes this result with a denser mesh.","marker":"[14]"},{"why":"Provides the single-crystal magnetization data showing Pauli-like paramagnetic susceptibility at high temperatures, the experimental anchor for the nonmagnetic state.","marker":"[2]"},{"why":"The original magnetization and neutron scattering study reporting Curie-Weiss behavior and a possible local moment that later experiments question.","marker":"[1]"},{"why":"Muon spin rotation spectroscopy showing absence of local moments in KV$_3$Sb$_5$.","marker":"[13]"},{"why":"ARPES data on CsV$_3$Sb$_5$ used to compare DFT and DFT+SOC band structures.","marker":"[15]"},{"why":"Defines the PBE exchange-correlation functional used throughout.","marker":"[21]"},{"why":"Defines the DFT-D3 dispersion correction for van der Waals interactions.","marker":"[22]"},{"why":"Provides the fixed-spin-moment method used to compute energy-magnetization curves and susceptibility.","marker":"[27]"}],"fun_headline_variants":["Coarse k-mesh invents ferromagnetism in kagome metals","Kagome metals nonmagnetic: dense k-mesh clears the air","SOC effect below 20 meV in AV3Sb5 kagome metals","Dense k-mesh flips AV3Sb5 from magnetic to nonmagnetic","No ferromagnetism or strong SOC in kagome metals"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the PBE exchange-correlation functional with the DFT-D3 dispersion correction and an $18\\times18\\times12$ k-mesh correctly captures the magnetic ground state, untested against antiferromagnetic orders, Hubbard U, hybrid functionals, or intermediate k-mesh densities.","fun_headline_variants_meta":{"raw":{"variants":["Coarse k-mesh invents ferromagnetism in kagome metals","Kagome metals nonmagnetic: dense k-mesh clears the air","SOC effect below 20 meV in AV3Sb5 kagome metals","Dense k-mesh flips AV3Sb5 from magnetic to nonmagnetic","No ferromagnetism or strong SOC in kagome metals"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000557,"raw_usage":{"total_tokens":2747,"prompt_tokens":1140,"completion_tokens":1607,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":756,"completion_tokens_details":{"reasoning_tokens":1507}},"tokens_in":756,"tokens_out":1607,"duration_ms":11604,"temperature":1.0,"reasoning_tokens":1507,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T19:00:55.156846+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A spin-polarized DFT calculation with the same functional but a denser mesh (for example $24\\times24\\times16$) that recovers a ferromagnetic or antiferromagnetic solution with nonzero V moments would falsify the central claim, as would a measurement on ultra-clean single crystals showing an intrinsic Curie-Weiss magnetic susceptibility that cannot be attributed to impurity spins.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"The recent theoretical report claiming a ferromagnetic ground state and local V moments; the paper reproduces and then refutes this result with a denser mesh."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the single-crystal magnetization data showing Pauli-like paramagnetic susceptibility at high temperatures, the experimental anchor for the nonmagnetic state."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"The original magnetization and neutron scattering study reporting Curie-Weiss behavior and a possible local moment that later experiments question."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Muon spin rotation spectroscopy showing absence of local moments in KV$_3$Sb$_5$."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"ARPES data on CsV$_3$Sb$_5$ used to compare DFT and DFT+SOC band structures."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Defines the PBE exchange-correlation functional used throughout."},{"cited_title":"Grimme, J","cited_arxiv_id":null,"evidence_quote":"Defines the DFT-D3 dispersion correction for van der Waals interactions."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the fixed-spin-moment method used to compute energy-magnetization curves and susceptibility."}],"review_version":1}