{"id":"168dd0cd-f2d9-4866-835c-2e4f9305ecdc","arxiv_id":"2608.10574","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"In MBE-grown Cr-NbSe2 films, post-growth annealing tunes carrier density independently of the chromium superstructure, and ferromagnetism appears at fixed sqrt3xsqrt3 moment geometry only above a critical annealing temperature.","lead":"Researchers grew two different ordered arrangements of chromium atoms inside the same NbSe2 crystal and then heated the films to change the number of electrons without moving the chromium atoms. They found that the magnetic state can change with electron count even when the atomic arrangement is the same, separating two controls that usually move together in crystals.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The carrier-density interpretation rests on unquantified structural equivalence between the 400C and 500C sqrt3 samples and on a Hall proxy in a multicarrier system; the 3/4 magnetization argument does not establish phase-fraction equivalence.","rationale":"The reader's weakest assumption identifies the same load-bearing soft spot: the operative variable separating the 400C and 500C samples is inferred, not directly established. The central observation itself, different magnetic ground states in samples assigned to the same sqrt3 x sqrt3 superstructure class, is valuable and experimentally supported by LEED and Raman. However, the interpretation that carrier density is the selecting axis requires ruling out differences in phase fraction, domain size, disorder, and local composition between the two samples. The 3/4 saturation-magnetization ratio is a single empirical point, not a quantitative phase-fraction measurement for the 400C sample. The Hall coefficient is a reasonable qualitative proxy but cannot uniquely determine carrier density in a multicarrier system with changing mobilities; the paper's own multi-band analysis shows that R_H depends on mobilities as well as densities. A direct comparison of phase fractions and carrier densities across replicate 400C and 500C samples would settle the issue. The RKKY framing is explicitly qualitative, so the quantitative exchange-coupling claim is appropriately modest; the conditional verdict already reflects the need for additional evidence. I therefore do not recommend changing the reader's CONDITIONAL verdict, and I agree that the experimental platform and core observation are likely sound.","tokens_in":18424,"tokens_out":4404,"duration_ms":43159,"concrete_test":"Prepare at least three 400C and three 500C pieces from the same parent wafer; on each, acquire plan-view STEM-EDX maps or scanning nano-LEED/STM data to quantify the Cr1/3NbSe2 phase fraction, coherent domain size, and local Cr concentration. Then extract carrier densities from multi-field Hall measurements fitted with the two-band model of Eqs. S1-S5, or from ARPES Fermi-surface areas if measurable. If the 400C and 500C pieces have statistically indistinguishable phase fraction and domain size but measurably different carrier density, the carrier-density interpretation is supported; if their microstructures differ substantially, or if their carrier densities are statistically the same, the non-bijective magnetism result must be reinterpreted as phase-separation- or disorder-driven rather than k_F-driven.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central claim is that carrier density, not structure, selects the magnetic ground state because the 400C and 500C annealed pieces both exhibit the sqrt3 x sqrt3 Cr superstructure yet only the 500C piece is ferromagnetic. For that conclusion to hold, the two pieces must be equivalent in every relevant respect except carrier density. This equivalence is not established. Supplemental S10 itself states that the 2x2-to-sqrt3 transformation should accompany phase separation into 75% Cr1/3NbSe2 and 25% NbSe2 if the Cr content is conserved, and the 3/4 saturation-magnetization ratio is shown only for the 500C sample; it is consistent with, but does not prove, that specific phase fraction. LEED and Raman establish superlattice periodicity, not phase fraction, coherent domain size, or chemical homogeneity. If the 400C piece contains less Cr1/3NbSe2, smaller ferromagnetic domains, or greater disorder, the absence of ferromagnetism may be structural rather than carrier-driven. Additionally, R_H is used as the carrier-density axis, but the system is explicitly multicarrier (Supplemental D, Table S1); R_H depends on mobilities as well as carrier densities, and annealing-induced Se vacancies or disorder can change mobilities. The systematic Hall evolution therefore does not by itself identify k_F as the operative tunable axis. Finally, the 400C-versus-500C comparison rests on one piece per annealing condition, so sample-to-sample variation is not excluded.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports MBE growth of two Cr-intercalated NbSe2 phases, Cr1/4NbSe2 with a 2x2R0° Cr superstructure and Cr1/3NbSe2 with a sqrt3 x sqrt3R30° superstructure, and characterizes them by STEM, XRD, LEED, RHEED, ARPES, Raman, magnetization, and transport, with neutron scattering on a bulk Cr1/4NbSe2 reference. The as-grown phases differ in magnetic ground state (AFM with TN≈50 K in 370-layer Cr1/4NbSe2; FM with TC=71 K in 9-layer Cr1/3NbSe2) and in electronic/transport signatures. The central experiment is a post-growth annealing series on 6-layer Cr1/4NbSe2 films: annealing at 400 °C and 500 °C both transform the superlattice to the sqrt3 x sqrt3R30° class, yet only the 500 °C sample becomes ferromagnetic, while the Hall coefficient evolves systematically across the series. The authors conclude that the carrier density, inferred from Hall measurements, selects the magnetic ground state at fixed moment geometry, and interpret the result within an RKKY framework as evidence of carrier-sensitive RKKY magnetism.","tokens_in":18705,"tokens_out":5514,"duration_ms":52233,"significance":"If the central claim were established, the work would constitute a notable advance: independent tuning of the RKKY variables r and kF within a single crystalline host, using a thin-film plus annealing approach that is in principle generalizable to other intercalated TMDCs. The paper's strengths are its extensive cross-checked structural characterization (LEED, Raman, STEM, XRD, RHEED), the clean distinction between the two as-grown superlattice phases, the neutron-scattering confirmation of the bulk Cr1/4NbSe2 magnetic structure, and the transparency of the transport fits (Supplemental Table S1). The empirical observation that two samples with the same LEED-resolved superlattice can have different magnetic ground states is interesting in its own right. However, the mechanistic conclusion that carrier density is the operative axis is currently under-supported: it relies on a Hall proxy in a multicarrier system and on an unquantified structural equivalence between the 400 °C and 500 °C samples, and the RKKY interpretation is not tested quantitatively. The paper would be suitable for a strong venue if these gaps are addressed or the claims are appropriately softened.","major_comments":[{"comment":"The load-bearing comparison between the 400 °C and 500 °C annealed samples does not establish that they are structurally equivalent in every relevant way except carrier density. LEED and Raman determine superlattice periodicity, not phase fraction, coherent domain size, or chemical homogeneity. Supplemental S10 itself states that, if Cr is conserved, the 2x2-to-sqrt3 transformation should be accompanied by phase separation into 75% Cr1/3NbSe2 plus 25% NbSe2; the 3/4 saturation-magnetization ratio is reported only for the 500 °C sample. Without phase-fraction evidence for the 400 °C sample, the absence of ferromagnetism at 400 °C could stem from a smaller volume fraction of the Cr1/3NbSe2 phase, smaller ferromagnetic domains, or greater disorder, rather than from a different carrier density. This ambiguity directly undermines the central claim that the sqrt3 structural class has a non-unique magnetic ground state selected by carrier density.","section":"Section IV, Fig. 4, and Supplemental S10"},{"comment":"The carrier-density axis is inferred entirely from the Hall coefficient R_H in a system that the paper explicitly identifies as multicarrier (Supplemental Section D). In the two-band model of Eqs. S1-S5, the linear Hall coefficient A depends on both carrier densities and mobilities, and annealing-induced Se vacancies can change mobilities as well as carrier numbers. The systematic evolution of R_H across the annealing series therefore does not uniquely identify k_F as the tunable axis; changes in mobility or in the relative weights of electron and hole pockets could produce the same Hall trend without a corresponding shift in k_F. Since the manuscript's title and conclusions assert carrier-tunable RKKY magnetism, a direct carrier-density probe (e.g., quantum oscillations, Seebeck coefficient, or ARPES-derived Fermi-surface volume) or an explicit modeling of R_H in terms of carrier densities is needed to support the operative-axis claim.","section":"Section IV, Fig. 4d; Supplemental Section D, Table S1"},{"comment":"The RKKY interpretation is presented without quantitative test. Equation (2) is introduced as an interpretive framework, but no parameter is fitted, no k_F values are extracted from the transport data, and the authors explicitly state in Fig. 5f that 'quantitative calibration to the Cr-NbSe2 band structure is not attempted in this work.' The empirical structure-magnetism decoupling in Fig. 4 could equally be explained by annealing-induced disorder, phase-fraction changes, or modifications of the Cr valence state. As it stands, the paper demonstrates a non-bijective correspondence between superlattice and magnetic ground state, but the conclusion that this correspondence reflects carrier-sensitive RKKY exchange goes beyond the evidence. A model calculation of J_RKKY(r,kF) for these two Cr geometries, or a control experiment that varies carrier density without changing the superlattice or disorder, would be required to substantiate the mechanistic claim.","section":"Section V, Eq. (2), and Fig. 5f"}],"minor_comments":[{"comment":"The sentence 'Controlled post-growth annealing ... modifies the carrier density while leaving the Cr superstructure intact below a structural-transition threshold' is slightly misleading, because the superstructure changes at 400 °C and 500 °C; consider rewording to make explicit that the structure is preserved only for low annealing temperatures.","section":"Abstract"},{"comment":"The '3/4 ratio' is stated as quantitatively consistent with a 75% Cr1/3NbSe2 + 25% NbSe2 phase mixture, but no error bar or propagation analysis is given; since this ratio is used to support phase conservation, it should be quantified.","section":"Section IV"},{"comment":"The phase diagram axes and the meaning of the colored bands are only described in the caption; adding labels directly to the figure would improve readability.","section":"Section V, Fig. 5c"},{"comment":"The table lists R² values but no uncertainties on the fitted coefficients A and C; reporting standard errors would strengthen the claim that the cubic term is robust across all temperatures.","section":"Supplemental Table S1"}],"recommendation":"major_revision","confidential_remarks":"The stress-test concern is well placed: the paper's most novel empirical observation (the 400 °C vs 500 °C decoupling) is precisely the point where the evidence is thinnest. The manuscript may still be publishable after a major revision if the authors either provide the missing structural and carrier-density quantification or reframe the conclusions to present the RKKY interpretation as a hypothesis rather than an established mechanism."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper is worth reading for the growth and the 400C/500C contrast. They stabilize two Cr superstructures in the same NbSe2 host by choosing MBE conditions, then anneal pieces of the same wafer and watch structure, transport, and magnetism evolve. The observation that two pieces share the sqrt3 x sqrt3 LEED pattern yet one is a ferromagnet and the other is not is a genuinely useful result, and it does challenge any simple structure-magnetism mapping.\n\nWhere the paper stands out: the structural characterizations are cross-checked (RHEED, LEED, STEM, XRD, Raman), ARPES distinguishes the two as-grown phases, and the AFM ground state of Cr1/4NbSe2 is anchored by neutron scattering on a bulk reference. The annealing series is systematic, and the transport data are analyzed with care, including the cubic Hall term that shows the system is multicarrier. The authors are also honest that the RKKY formula is qualitative; they explicitly say no quantitative calibration is attempted.\n\nThe soft spots are real but not fatal. One: each annealing condition is a single piece, so sample-to-sample variation isn't excluded. Two: LEED and Raman establish superlattice periodicity, not phase fraction. The supplement itself says the 2x2-to-sqrt3 transformation should be accompanied by phase separation into 75% Cr1/3NbSe2 + 25% NbSe2 if Cr is conserved, and the 3/4 saturation moment is shown only for the 500C sample. Without quantifying the 400C piece, the absence of ferromagnetism could be a structural or microstructural effect—smaller ferromagnetic domains, lower Cr1/3 fraction, more disorder—rather than a carrier-density effect. Three: the Hall coefficient is used as a proxy for carrier density, but in a multicarrier system R_H depends on mobilities as well as densities. The systematic Hall evolution is suggestive, not conclusive. There is no direct k_F measurement on the annealed films, no ARPES on them, no quantum oscillations.\n\nNone of this kills the paper. The platform is likely solid and the non-bijective observation is worth reporting. I'd send it to a serious referee, with the expectation of a major revision: more samples per condition, quantitative composition/phase-fraction analysis for the 400C and 500C pieces, and either direct carrier-density evidence or a softened causal claim. The title and abstract overstate the carrier-tuning conclusion relative to the evidence.\n\nWho this is for: experimentalists working on vdW magnets, intercalated TMDCs, or RKKY. They should engage with it. My own verdict is that the mechanistic claim is provisional, but the experimental core is strong.","headline":"Careful growth work and one striking 400C/500C contrast, but the central carrier-density claim is an assumption the data don't yet support.","tokens_in":19332,"tokens_out":4501,"would_cite":false,"duration_ms":39919,"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":"This paper shows that in epitaxial Cr-NbSe2 films the RKKY magnetic ground state is set by carrier density rather than by the chromium superstructure alone: two annealed samples with identical √3×√3R30° Cr ordering are respectively…","keywords":["RKKY interaction","intercalated transition-metal dichalcogenides","molecular-beam epitaxy","post-growth annealing","carrier-density tuning","itinerant magnetism","Cr-NbSe2 epitaxial thin films","Hall effect"],"falsifier":"Measure the Fermi wavevector directly (ARPES Fermi-surface mapping or quantum oscillations) on the 400 °C and 500 °C annealed samples: if the two have the same $k_\\mathrm{F}$ while one is ferromagnetic and the other is not, the carrier-selection claim fails. Electrostatic gating of a 400 °C sample that induces ferromagnetism without reheating would confirm the claim.","tokens_in":18212,"feed_emoji":"🧲","tokens_out":15302,"duration_ms":122415,"temperature":0.7,"pith_summary":"This paper reports a crystalline thin-film system in which the two variables that control RKKY magnetism—the spacing between magnetic moments and the Fermi wavevector of the mediating electrons—can be tuned separately in the same material. The authors grow Cr$_{1/4}$NbSe$_2$ and Cr$_{1/3}$NbSe$_2$ films with distinct Cr superstructures by molecular-beam epitaxy, then anneal pieces of the same wafer to change carrier density while leaving the Cr lattice intact below a structural-transition threshold. The key observation is that two samples sharing the $\\sqrt{3}\\times\\sqrt{3}\\mathrm{R}30^\\circ$ Cr superstructure, annealed at 400 °C and 500 °C, are respectively non-ferromagnetic and ferromagnetic ($T_\\mathrm{C}=71$ K), while the Hall response evolves continuously across the series. This is taken as evidence that carrier density, not moment geometry alone, selects the magnetic ground state, realizing carrier-sensitive RKKY magnetism in a single crystalline host. If correct, it makes Cr-NbSe$_2$ and similar intercalated dichalcogenides a platform for switching magnetism by doping instead of by structural change.","feed_headline":"Carrier density, not structure, flips Cr-NbSe2 magnetism","feed_subtitle":"Annealing tunes electron count while Cr spacing stays fixed, flipping magnetic order from AFM to FM.","key_machinery":"The load-bearing object is the RKKY exchange coupling $J_\\mathrm{RKKY}(r,k_\\mathrm{F}) \\propto F(2k_\\mathrm{F}r)/r^n$, where $F$ is an oscillatory function and $n$ depends on dimensionality; it ties magnetic-state selection to the joint coordinate $(r,k_\\mathrm{F})$. In this paper, the Cr superstructure realizes two discrete $r$ values—6.74 Å for the $2\\times2$ phase and 5.84 Å for the $\\sqrt{3}\\times\\sqrt{3}$ phase—and the annealing series realizes a continuum of $k_\\mathrm{F}$ values whose proxy is the Hall coefficient. The experimental lever is the annealing series itself: pieces of one wafer annealed from 200 °C to 600 °C, with LEED and Raman tracking structure while Hall and magnetization track carriers and magnetism. The non-bijective map, same superstructure but different magnetic ground state, is what does the argumentative work.","core_discovery":"The paper's central claim is that the Cr superstructure in Cr-NbSe$_2$ does not uniquely determine the magnetic ground state, and that the missing selecting variable is the carrier density of the NbSe$_2$ itinerant background. In the annealing series, the 400 °C and 500 °C samples share the same $\\sqrt{3}\\times\\sqrt{3}\\mathrm{R}30^\\circ$ Cr ordering, yet only the 500 °C sample is ferromagnetic ($T_\\mathrm{C}=71$ K), with saturation magnetization about 3/4 of the as-grown Cr$_{1/3}$NbSe$_2$ reference; the Hall coefficient meanwhile evolves monotonically across the series, including a sign change at 600 °C. The authors interpret this as a joint $(r,k_\\mathrm{F})$ selection rule for the RKKY exchange $J_\\mathrm{RKKY}(r,k_\\mathrm{F})\\propto F(2k_\\mathrm{F}r)/r^n$: the superstructure fixes $r$ at discrete values (2$a_{\\mathrm{NbSe}_2}$ = 6.74 Å and $\\sqrt{3}a_{\\mathrm{NbSe}_2}$ = 5.84 Å), while annealing-induced Se vacancies dope electrons and shift $k_\\mathrm{F}$. Ferromagnetism emerges only above a critical annealing temperature, placing the system in a low-carrier-density RKKY regime where modest $k_\\mathrm{F}$ tuning crosses an exchange sign reversal. The paper concludes that carrier density and moment geometry are experimentally disentangled in a single crystalline host, and that the MBE-plus-annealing route generalizes across intercalated transition-metal dichalcogenides.","pith_inferences":["Editorial inference: if the carrier-density picture is right, electrostatic gating of a 400 °C-annealed film should drive it ferromagnetic without any structural change, giving a reversible test that the irreversible annealing series cannot provide.","Editorial inference: the saturation magnetization ratio of about 3/4 suggests the 500 °C sample is a phase mixture of Cr$_{1/3}$NbSe$_2$ and NbSe$_2$; spatially resolved magnetometry could separate percolation of ferromagnetic ordered regions from uniform $k_\\mathrm{F}$ tuning.","Editorial inference: the Hall coefficient is only a proxy for carrier density in a multicarrier system; a direct Fermi-surface measurement on the 400 °C/500 °C pair would either harden the assignment or reveal a second variable at work.","Editorial inference: because $J_\\mathrm{RKKY}$ oscillates in $2k_\\mathrm{F}r$, a continuously tunable $k_\\mathrm{F}$ should produce more than one magnetic-state switch; finer annealing steps could look for further ferromagnet/antiferromagnet oscillations as a test of the RKKY sign-reversal mechanism."],"forward_implications":["If the central claim holds, the same $\\sqrt{3}\\times\\sqrt{3}\\mathrm{R}30^\\circ$ structural class can host both non-ferromagnetic and ferromagnetic ground states, so diffraction alone cannot predict the magnetism of an intercalated TMDC.","Carrier density becomes a practical tuning axis: post-growth annealing shifts the Hall response monotonically and, above a threshold, switches on ferromagnetic order with $T_\\mathrm{C}=71$ K at 500 °C and $T_\\mathrm{C}=84$ K at 600 °C.","The low-carrier-density, multicarrier Hall response implies that modest perturbations—annealing, gating, or doping—can move $k_\\mathrm{F}$ enough to cross an RKKY exchange oscillation, enabling magnetic-state control at fixed moment spacing.","The MBE-plus-annealing protocol is extensible to other intercalated transition-metal dichalcogenides, offering a general route to decouple moment geometry from carrier density in itinerant magnets.","The 400 °C intermediate window, where the superstructure has transformed but ferromagnetism has not yet emerged, shows the structural and magnetic transitions need not coincide and defines the experimentally accessible carrier-tuned regime."],"supporting_citations":[{"why":"Supplies the oscillatory RKKY coupling form whose locked dependence on moment spacing and Fermi wavevector the paper sets out to break.","marker":"[3–5,22]"},{"why":"Defines the distance-dominant RKKY tuning limit in metallic multilayers, the contrast for the carrier-density-dominant case.","marker":"[8]"},{"why":"Supplies the bulk intercalated-TMDC context, including RKKY ferromagnetism in Fe1/4TaS2, which the thin-film route extends.","marker":"[9–12]"},{"why":"Provides the epitaxial Cr1/3NbSe2 growth and band-structure reference for the √3×√3 phase.","marker":"[14]"},{"why":"Attributes annealing-driven carrier-density changes to Se vacancies, the mechanism that tunes the Fermi wavevector along the series.","marker":"[16,17]"},{"why":"Establishes the bulk Cr1/4NbSe2 120-degree antiferromagnetic ground state used to assign the as-grown thin-film magnetism.","marker":"[20]"},{"why":"Reports a carrier-induced antiferromagnet-to-ferromagnet transition in a low-carrier compound, the neighboring regime the paper distinguishes from RKKY.","marker":"[24]"},{"why":"Shows multiple superlattices and magnetic ground states coexisting in bulk V1/3NbS2, the in-host multiplicity precedent extended here to separately tunable films.","marker":"[25]"}],"fun_headline_variants":["Carrier tuning, not Cr spacing, sets magnetic order in Cr-NbSe2","Same Cr lattice, different magnetism: carrier density decides","Annealing tunes electrons to switch Cr-NbSe2 from AFM to FM","Carrier density selects magnetic ground state in Cr-NbSe2","RKKY magnetism flips with carrier density alone in Cr-NbSe2"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the 400 °C and 500 °C samples differ only in carrier density, with the Hall coefficient standing in for a direct density measurement in a multicarrier system and with phase fractions and disorder left unquantified.","fun_headline_variants_meta":{"raw":{"variants":["Carrier tuning, not Cr spacing, sets magnetic order in Cr-NbSe2","Same Cr lattice, different magnetism: carrier density decides","Annealing tunes electrons to switch Cr-NbSe2 from AFM to FM","Carrier density selects magnetic ground state in Cr-NbSe2","RKKY magnetism flips with carrier density alone in Cr-NbSe2"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001159,"raw_usage":{"total_tokens":4937,"prompt_tokens":1216,"completion_tokens":3721,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":832,"completion_tokens_details":{"reasoning_tokens":3623}},"tokens_in":832,"tokens_out":3721,"duration_ms":24096,"temperature":1.0,"reasoning_tokens":3623,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T21:30:51.844752+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the Fermi wavevector directly (ARPES Fermi-surface mapping or quantum oscillations) on the 400 °C and 500 °C annealed samples: if the two have the same $k_\\mathrm{F}$ while one is ferromagnetic and the other is not, the carrier-selection claim fails. Electrostatic gating of a 400 °C sample that induces ferromagnetism without reheating would confirm the claim.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Defines the distance-dominant RKKY tuning limit in metallic multilayers, the contrast for the carrier-density-dominant case."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the epitaxial Cr1/3NbSe2 growth and band-structure reference for the √3×√3 phase."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Reports a carrier-induced antiferromagnet-to-ferromagnet transition in a low-carrier compound, the neighboring regime the paper distinguishes from RKKY."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Shows multiple superlattices and magnetic ground states coexisting in bulk V1/3NbS2, the in-host multiplicity precedent extended here to separately tunable films."}],"review_version":1}