{"id":"5195b071-cfa2-4709-a17e-a7516670a57e","arxiv_id":"2506.03416","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"Spin gapped metallic half-Heusler compounds can be grouped into localized, itinerant, and mixed magnets, with predicted Curie temperatures up to about 684 Kelvin for several members.","lead":"This paper maps which 24 half-Heusler compounds with near-gap metallic band structures become magnetic, using density functional theory and a Stoner criterion. It also predicts Curie temperatures and spin-wave stiffnesses for the magnetic members, which are candidate materials for spintronic transistors and low-power logic.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Stoner classification is internally inconsistent: bare I·N(E_F)>1 for several nonmagnetic gapped metals, and the α=0.6 renormalization is applied without a material-specific decision rule.","rationale":"The reader correctly identified sensitivity of the Stoner product to the PBE Fermi-level DOS as a weakness, but the more immediate and load-bearing problem is internal inconsistency in the Stoner classification itself. Table II directly contradicts the Section IV C claim that nonmagnetic gapped metals have both bare and renormalized Stoner products below unity. The α=0.6 factor is the only thing separating these compounds from the Stoner criterion, and it is inserted without a material-specific derivation or a decision rule. This makes the central predictive framework unfalsifiable as stated: one can always choose α to make the nonmagnetic cases fall below threshold, and the same α leaves some magnetic cases with one sublattice below threshold while still assigning moments to that sublattice. The concern is not merely about consensus disagreements but about whether the paper's own tables support its headline claim. Because the issue can be corrected by specifying and testing the renormalization rule, I would keep the reader's CONDITIONAL verdict rather than escalate to rejection: the broad DFT dataset, magnetic moments, exchange constants, and Curie temperatures remain useful, but the Stoner-based predictor should be reframed as an α-parameterized empirical classification until the rule is justified and tested. Agreement with the reader is partial because the reader's weakest assumption centered on N(E_F) sensitivity, whereas the stronger concern is the alpha-dependent circularity and the explicit contradiction in Section IV C.","tokens_in":22577,"tokens_out":3095,"duration_ms":35890,"concrete_test":"Recompute the Section IV C classification for all 24 compounds using the unrenormalized products and also with α=0.5, 0.7, and 1.0, tabulating which compounds change classification across thresholds. Specifically, verify the eight 'Gapped metals' rows in Table II: if any unrenormalized I·N(E_F) exceeds 1 (CoNbSb, CoTaSb, NiTiSb, and NiNbSn already do), the claim that both products remain below unity is false, and the paper must either derive α from first principles, specify an a priori decision rule, or restrict the Stoner-based prediction to the renormalized criterion with a stated sensitivity range.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central predictive claim in Section IV C is undercut by an internal contradiction. The text states that for the eight compounds classified as nonmagnetic gapped metals, 'both the unrenormalized and renormalized Stoner products, I·N(E_F) and α·I·N(E_F), remain below the critical threshold of unity.' Table II shows otherwise: CoNbSb has bare products 1.39/1.16, CoTaSb 1.11/0.65, NiTiSb 0.86/1.50, and NiNbSn 1.40/1.29. Each of these satisfies the unrenormalized Stoner criterion on at least one sublattice, yet is computed to be nonmagnetic. Only after multiplying by the ad hoc factor α=0.6 do these products drop below 1. The factor is justified solely by a generic 40% correlation reduction cited from Ref. 51, with no material-specific derivation and no sensitivity analysis. Consequently, the classification depends on an unprotected parameter: a slightly different α changes which compounds are deemed nonmagnetic, so the claimed criterion is not a falsifiable predictor unless α is fixed a priori. The inconsistency also affects magnetic compounds: FeVSn has α·I·N(E_F)=1.32/0.80, but V carries a 1.03 μB moment even though its renormalized product is below unity, showing that the stated decision rule is not applied consistently.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports a systematic density-functional-theory (DFT) study of magnetism in 24 spin gapped metallic half-Heusler compounds. For each compound the authors compute magnetic moments in ferromagnetic/ferrimagnetic and antiferromagnetic configurations, identify itinerant versus localized moment character using moment collapse and spin-density isosurfaces, and obtain cRPA-based Hubbard U and Hund's exchange J parameters. From U and J they construct the Stoner parameter I=(U+6J)/5 and evaluate a sublattice-resolved Stoner criterion I·N(E_F) against nonmagnetic projected densities of states. They further compute Heisenberg exchange parameters via the LKAG formalism, estimate mean-field Curie temperatures, and report magnon dispersions and spin-wave stiffness constants. The central claim is that the cRPA-based Stoner criterion, with a renormalization factor α=0.6, distinguishes compounds that remain nonmagnetic gapped metals from those that order magnetically, and that the magnetic compounds with high Curie temperatures are promising for spintronic applications.","tokens_in":22796,"tokens_out":4670,"duration_ms":51956,"significance":"If the Stoner-based classification were reliable, the paper would provide a computationally inexpensive, material-specific predictor for magnetism in an underexplored family of half-Heusler compounds, and the reported Curie temperatures and spin-wave stiffnesses would constitute useful design inputs. The study has commendable breadth: 24 compounds, two independent DFT codes cross-validated, cRPA interaction parameters, LKAG exchange, magnon spectra, and a clear presentation of structural and electronic data. The authors are also appropriately cautious about the limitations of the Heisenberg mapping for itinerant magnets and about the neglect of Landau damping. However, the key predictive claim is undermined by an internal inconsistency in the Stoner analysis (see Major Comment 1), and the robustness of the criterion with respect to the DFT gap problem is not demonstrated. Because the classification is the paper's central new message, the manuscript needs substantive revision.","major_comments":[{"comment":"The text states that for the eight compounds classified as gapped metals, 'both the unrenormalized and renormalized Stoner products, I·N(E_F) and α·I·N(E_F), remain below the critical threshold of unity.' Table II directly contradicts this for four of the eight compounds: CoNbSb has bare products 1.39 (Co) and 1.16 (Nb), CoTaSb 1.11 (Co), NiTiSb 1.50 (Ti), and NiNbSn 1.40 (Ni) and 1.29 (Nb). For these compounds it is only after multiplying by the ad hoc α=0.6 that the products fall below 1. Thus the classification as nonmagnetic depends entirely on this renormalization factor, which is imported from a generic 40% correlation reduction cited from Ref. 51 without any material-specific justification or sensitivity analysis. This is load-bearing because the abstract and conclusions present the Stoner criterion as the framework that 'successfully accounts for the absence of magnetic order' in these compounds.","section":"§IV C, Table II"},{"comment":"The decision rule is not applied consistently to the magnetic members. FeVSn is listed as satisfying the Stoner condition on both sublattices, but its renormalized products are 1.32 (Fe) and 0.80 (V), with V carrying a 1.03 μB moment. If α·I·N(E_F) is the relevant criterion, V does not satisfy it, contradicting the text; if the unrenormalized product is used for magnetic compounds, then the criterion is applied differently to magnetic and nonmagnetic compounds, and the classification is post hoc. The manuscript must state explicitly whether the bare or renormalized Stoner product is the predictor, apply it uniformly to all 24 compounds, and discuss the resulting misclassifications (including FeVSn, and also NiTiIn where the renormaized Ti product 1.56 exceeds 1 yet the compound is classified in the 'only one sublattice satisfies' group without comment).","section":"§IV C, Table II"},{"comment":"The projected N(E_F) values used in the Stoner product are obtained from PBE non-spin-polarized band structures, but in these gapped metals the Fermi level sits at or very near the edge of a narrow gap, so the projected DOS at E_F is extremely sensitive to the accuracy of the band structure. The authors do not quantify this sensitivity; a band-edge shift of order 0.1 eV—well within PBE's expected gap error—could push several of the borderline products (e.g., CoNbSb Ni 1.16, NiTiSb Ti 1.50) across the I·N(E_F)=1 threshold, or pull magnetic cases below it. Since the central claim is a predictive threshold criterion, the manuscript should include an explicit sensitivity test, for example a rigid scissor shift applied to the conduction or valence band edges, or results from a hybrid functional for a representative subset, to show that the separation between magnetic and nonmagnetic compounds is robust.","section":"§III, §IV C"}],"minor_comments":[{"comment":"The abstract and conclusion describe the Stoner analysis as establishing a 'predictive framework,' but the analysis is applied post hoc to a dataset already known to contain magnetic and nonmagnetic members, and the α=0.6 factor is not fixed a priori. The wording should be softened unless the authors provide a genuinely predictive validation (e.g., leave-one-out or a test on compounds not used to set α).","section":"Abstract and §V"},{"comment":"There is a typographical error in the opening sentence: 'spin gapped metsllic' should read 'spin gapped metallic.'","section":"§III"},{"comment":"The column headings 'FM (001)' and 'AFM (111)' are not defined in the table caption; the reader must infer the magnetic ordering from the text and Fig. 2. Please define the FM and AFM configurations explicitly in the caption and state that the AFM column gives moments in the doubled [111] cell; otherwise the meaning of 'mtotal' in the AFM case is ambiguous.","section":"Table I"},{"comment":"The caption describes CoTiSn as ferromagnetic and FeVSb as ferrimagnetic, but the text in §IV D says FeVSn and CoVSb are ferrimagnets; the label 'FeVSb' in Fig. 5(b) appears to be a typo for 'FeVSn', and the caption should be checked for consistent compound names.","section":"Fig. 5(b)"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper is worth reading as a computational survey, but the central predictive claim does not survive contact with its own table. The text in Section IV C says that for the eight nonmagnetic gapped metals, both the unrenormalized and renormalized Stoner products stay below unity. Table II shows that four of them (CoNbSb, CoTaSb, NiTiSb, NiNbSn) have bare I·N(EF) above 1 on at least one sublattice. Only after multiplying by α=0.6 do they drop below threshold. That α is imported from a generic correlation reduction argument and applied uniformly, with no material-specific justification and no sensitivity analysis. FeVSn makes the inconsistency worse: the V sublattice has a renormalized product of 0.80, below 1, yet V carries a 1.03 μB moment. So the stated decision rule is not actually the rule being used.\n\nWhat is genuinely new and useful: the authors provide a systematic set of cRPA U and J values, sublattice-resolved magnetic moments in FM and AFM configurations, LKAG exchange parameters, Curie temperatures, and spin-wave stiffnesses for 24 spin-gapped half-Heusler compounds. That is real data, computed with standard methods, and the qualitative classification of localized versus itinerant magnetism based on moment collapse in the AFM state plus spin-density isosurfaces is plausible and mostly convincing. The exchange-constant analysis and magnon spectra for the representative compounds look like solid, serviceable work. If I worked in this area, I would want this dataset in the literature.\n\nThe soft spots are concentrated in the Stoner analysis. Aside from the internal contradiction, the N(EF) values come from non-spin-polarized PBE calculations, and for these narrow-gap metals a small band-edge shift could move several compounds across the I·N=1 threshold. The paper does not quantify that uncertainty. Data availability on request is a minor annoyance, not a flaw in the science.\n\nBottom line: this is a solid survey with a flawed framing section. The Stoner claim should be either corrected, substantially qualified, or removed in favor of a more honest statement that the classification is empirical. The underlying DFT data and magnetic interaction results still deserve referee time. Send it to review, but insist the authors fix the internal inconsistency and provide sensitivity information on α and on the band-structure sensitivity of N(EF).","headline":"Useful half-Heusler magnetism survey, but the Stoner-based predictor is undercut by an internal inconsistency with its own Table II.","tokens_in":23438,"tokens_out":2532,"would_cite":true,"duration_ms":29603,"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 sublattice-resolved Stoner product separates magnetic from nonmagnetic spin-gapped half-Heuslers, with cRPA-derived I and PBE density of states as inputs.","keywords":["half-Heusler compounds","spin gapped metals","itinerant magnetism","localized magnetism","Stoner criterion","constrained random phase approximation","Curie temperature","spin-wave stiffness"],"falsifier":"A decisive check would be to measure the magnetization of a predicted nonmagnetic gapped metal such as NiTiSb or CoNbSb; if it orders magnetically below some temperature, the sublattice Stoner criterion would be falsified for that member. A cheaper calculation-based check would be to recompute N(E_F) with a hybrid functional or GW: if the classification of any of the borderline compounds flips across I·N(E_F)=1, the predictor is not stable to the band-structure method.","tokens_in":22325,"feed_emoji":"🧲","tokens_out":6997,"duration_ms":67807,"temperature":0.7,"pith_summary":"This paper aims to establish a quick material-specific predictor for magnetism in spin-gapped metallic half-Heusler compounds: the sublattice-resolved Stoner product I·N(E_F), where the Stoner parameter I is built from cRPA-computed Hubbard U and Hund exchange J through I=(U+6J)/5 and N(E_F) is the nonmagnetic PBE density of states at the Fermi level. The claim is that compounds in which neither transition-metal sublattice satisfies I·N(E_F)>1 stay nonmagnetic, while compounds in which one or both sublattices satisfy the criterion order magnetically, with the moment localized on the satisfying site or spread over both. The paper further classifies Co- and Ni-based members as mostly itinerant magnets and Fe-, Ti-, and V-based members as localized, itinerant, or mixed, and it reports exchange parameters, Curie temperatures, and spin-wave stiffnesses for the magnetic members. If the criterion holds, it gives a low-cost design rule for choosing magnetic versus nonmagnetic electrodes for spin-gapped-metal spintronics.","feed_headline":"Stoner product predicts which half-Heuslers order magnetically","feed_subtitle":"Stoner product I×N(EF) separates itinerant, localized, and nonmagnetic spin-gapped half-Heuslers.","key_machinery":"The load-bearing object is the sublattice-resolved Stoner criterion. For a compound XYZ with two transition-metal sites X and Y, the paper checks I_X N_X(E_F)>1 and I_Y N_Y(E_F)>1 separately, with I=(U+6J)/5 from cRPA-calculated on-site Coulomb repulsion U and Hund exchange J, and N_X(E_F), N_Y(E_F) the site-projected nonmagnetic densities of states. The mechanism does the sorting: both sublattices unstable gives moments on both, one unstable gives a localized moment on that site with an induced moment on the other, and neither unstable gives a nonmagnetic gapped metal. A many-body renormalization factor α=0.6, taken from the literature, is also reported as a more conservative version of the threshold.","core_discovery":"On the paper's own terms, the central finding is that the Stoner criterion, applied separately to the two transition-metal sublattices of each half-Heusler, correctly separates the magnetic from the nonmagnetic members of the spin-gapped-metal family. The authors estimate I from cRPA values of U and J via I=(U+6J)/5, evaluate N(E_F) in the non-spin-polarized state, and find that compounds with I·N(E_F)>1 on at least one sublattice develop moments, while compounds with the product below unity on both sublattices—NiHfIn, CoNbSb, CoTaSb, NiTiSb, NiZrSb, NiHfSb, NiNbSn, NiTaSn—remain nonmagnetic. They also show that the stability of the moment when switching from ferromagnetic to antiferromagnetic alignment, together with real-space spin-density isosurfaces, distinguishes itinerant from localized behavior, and that for the magnetic members Heisenberg exchange parameters yield Curie temperatures and magnon dispersions with stiffnesses between 165 and 936 meV Ų.","pith_inferences":["This criterion should transfer to other gapped-metal families, such as 18-electron half-Heusler semiconductors doped by one electron or hole; the same nonmagnetic-DOS plus cRPA calculation would predict where magnetism appears, which the paper does not test.","A useful stress test would be to compute N(E_F) with a hybrid functional or GW corrections for the borderline compounds; because the Fermi level hugs a gap edge, small band shifts could move a compound across the I·N(E_F)=1 line, and knowing which compounds are near the boundary would show how robust the classification is.","The α=0.6 renormalized criterion, taken literally, would call NiZrIn nonmagnetic even though the DFT ground state has a small moment; deciding whether the renormalization factor should be material-dependent is a concrete question the paper leaves open.","If the gapped character survives above T_C as argued, the same compounds could serve as spin-filtering electrodes in both ordered and paramagnetic regimes, which is an experimentally testable transport prediction."],"forward_implications":["A DFT plus cRPA screen using only the nonmagnetic density of states and on-site U, J can label a candidate half-Heusler as magnetic or nonmagnetic before any spin-polarized or Heisenberg calculation is run.","The eight compounds with the Stoner product below unity on both sublattices are predicted to stay nonmagnetic, making them candidate nonmagnetic electrodes for spin-gapped-metal transistors.","FeVSn, CoVSb, and NiVSb have estimated Curie temperatures of 532 K, 419 K, and 684 K, respectively, putting them above room temperature and in range for spintronic and magnonic use.","Magnon spectra of the two-sublattice magnets contain an acoustic branch and an optical branch, with spin-wave stiffnesses from 165 to 936 meV Ų, giving quantitative input for magnon-transport modeling.","Above T_C, compounds whose spin channels share the same p- or n-type gapped character may remain gapped metals in the paramagnetic state rather than becoming ordinary metals."],"supporting_citations":[{"why":"Defines the spin-gapped-metal family and supplies the compound list, lattice constants, and spin-gap classification this study extends.","marker":"[26]"},{"why":"Gives the Stoner-parameter relation I=(U+6J)/5 and the many-body renormalization factor used for the α-scaled criterion.","marker":"[51]"},{"why":"Provides the cRPA methodology for computing Hubbard U and Hund exchange J in Heusler compounds, as used here for the Stoner products.","marker":"[50]"},{"why":"Earlier cRPA application to half-metallic Heusler magnets that establishes the reliability of the computed U and J.","marker":"[49]"},{"why":"LKAG formalism used to map total energies onto Heisenberg exchange parameters.","marker":"[42]"},{"why":"Multi-sublattice spin-wave formalism used to compute magnon dispersions and stiffness constants D.","marker":"[43]"},{"why":"Conceptual justification for site-resolved Stoner-like instabilities via a combined Stoner-Heisenberg-Hubbard spin-fluctuation framework.","marker":"[61]"},{"why":"Single-sublattice spin-wave stiffness and Curie-temperature method generalized by Ref. [43] for the present multi-sublattice magnets.","marker":"[44]"},{"why":"Supplies the PBE exchange-correlation functional used for all band structures and densities of states.","marker":"[36]"}],"fun_headline_variants":["Stoner product decides magnetism in half-Heuslers","Stoner criterion separates magnetic and nonmagnetic half-Heuslers","Stoner rule predicts itinerant vs localized magnetism in half-Heuslers","Spin-gapped half-Heuslers: Stoner product tells who orders magnetically"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The Fermi level of these gapped metals sits at or near the edge of a narrow gap, so the nonmagnetic density of states used in the Stoner product is very sensitive to the accuracy of the PBE band structure, and the paper does not quantify how GGA gap errors shift the products.","fun_headline_variants_meta":{"raw":{"variants":["Stoner product decides magnetism in half-Heuslers","Stoner criterion separates magnetic and nonmagnetic half-Heuslers","Stoner rule predicts itinerant vs localized magnetism in half-Heuslers","Spin-gapped half-Heuslers: Stoner product tells who orders magnetically"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000485,"raw_usage":{"total_tokens":2472,"prompt_tokens":1104,"completion_tokens":1368,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":720,"completion_tokens_details":{"reasoning_tokens":1292}},"tokens_in":720,"tokens_out":1368,"duration_ms":12504,"temperature":1.0,"reasoning_tokens":1292,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T11:04:16.313704+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A decisive check would be to measure the magnetization of a predicted nonmagnetic gapped metal such as NiTiSb or CoNbSb; if it orders magnetically below some temperature, the sublattice Stoner criterion would be falsified for that member. A cheaper calculation-based check would be to recompute N(E_F) with a hybrid functional or GW: if the classification of any of the borderline compounds flips across I·N(E_F)=1, the predictor is not stable to the band-structure method.","supporting_citations":[{"cited_title":"S ¸a¸ sıo˘ glu, M","cited_arxiv_id":null,"evidence_quote":"Defines the spin-gapped-metal family and supplies the compound list, lattice constants, and spin-gap classification this study extends."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the Stoner-parameter relation I=(U+6J)/5 and the many-body renormalization factor used for the α-scaled criterion."},{"cited_title":"S ¸a¸ sıo˘ glu, I","cited_arxiv_id":null,"evidence_quote":"Provides the cRPA methodology for computing Hubbard U and Hund exchange J in Heusler compounds, as used here for the Stoner products."},{"cited_title":"Friedrich, M","cited_arxiv_id":null,"evidence_quote":"Earlier cRPA application to half-metallic Heusler magnets that establishes the reliability of the computed U and J."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Multi-sublattice spin-wave formalism used to compute magnon dispersions and stiffness constants D."},{"cited_title":"S ¸a¸ sıo˘ glu, C","cited_arxiv_id":null,"evidence_quote":"Conceptual justification for site-resolved Stoner-like instabilities via a combined Stoner-Heisenberg-Hubbard spin-fluctuation framework."}],"review_version":1}