{"id":"9c5d2195-e964-4bd9-ab07-b62896c9b9ee","arxiv_id":"2411.16125","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":7,"one_line_summary":"Amorphous V2O3+p thin films (p = 0.05, 0.23, 0.49) show room-temperature ferromagnetism with about 14 kA/m saturation magnetization, tunable by oxygen content.","lead":"Vanadium oxide films grown with just the right amount of oxygen show a small but real ferromagnetic signal at room temperature. The magnetic strength and its coupling to a cobalt layer can be tuned by changing the oxygen flow during growth, pointing toward a new tunable oxide for spintronics.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The intrinsic-ferromagnetism claim hinges on element-specific proof: absent XMCD, trace impurities or interfacial CoO can account for every signature reported.","rationale":"The single most load-bearing assumption in the paper is that the hysteresis measured in Fig. 1 originates from VOx itself. Everything downstream - the partial-volume ferromagnetic model, the 'control of ferromagnetism by oxidation states', the antiferromagnetic coupling with Co, and the proposed spintronic applications - inherits this assumption. The paper provides no element-specific magnetic probe. The susceptibility analysis only elevates the signal above a paramagnetic substrate background; it says nothing about a weak ferromagnetic contaminant. Because the observed magnetization is only ~1% of a typical ferromagnet, trace magnetic impurities are a quantitatively viable alternative. The Co/Pt inverted loop is likewise ambiguous because CoO is a known source of exchange bias and inverted hysteresis, and no control with a nonmagnetic oxide spacer is provided. These are not internal inconsistencies but missing external validity checks. The micromagnetic simulations, with parameters explicitly acknowledged as non-unique and fitted to reproduce the data, cannot resolve the ambiguity. The reader's conditional verdict is therefore the appropriate one: accept provisionally if XMCD and suitable CoO-eliminating controls confirm the intrinsic origin; otherwise the claim should be rejected. No change to the reader's verdict is warranted based on this stress-test pass.","tokens_in":16939,"tokens_out":4101,"duration_ms":38273,"concrete_test":"Perform XMCD at the V L2,3 edges at 300 K on a 20-nm VOx film identical to Sample 1, measuring field-dependent XMCD hysteresis loops in grazing incidence (±1 T). If the V-edge XMCD signal is absent or non-hysteretic, the measured 14 kA/m cannot be intrinsic VOx ferromagnetism and the central claim fails; if present, the signal identifies the V valence (V3+/V4+) contributing to the moment.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central claim is that amorphous V2O3+p films (Samples 1-3) are intrinsically ferromagnetic at 300 K with Ms ≈ 14 kA/m (Abstract; Fig. 1). The evidence is macroscopic magnetization after subtracting a large paramagnetic background from the 550-μm Si/SiO2 substrate (Section 1, Fig. 1) and an inverted M(H) loop in VOx/Co/Pt attributed to antiferromagnetic VOx/Co coupling (Fig. 4c). No element-specific magnetometry is presented. The measured saturation moment is only 1-2% of a typical ferromagnet (Section 1), precisely the magnitude expected from trace Fe/Co/Ni impurities (at ppm level) or from a spurious ferromagnetic minor phase. The paper itself acknowledges that 'more direct evidence, such as detailed measurements of the atomic moments and spin configurations within the VOx layer, would be necessary' and suggests XMCD or neutron scattering, but does not provide them. The micromagnetic simulations (Section 3, Fig. 6) do not close this gap: the authors state the parameters were varied over a wide range to reproduce the experiments, and that they 'may not represent the actual material properties of VOx.' Similarly, the VOx/Co inverted loop could arise from interfacial CoO (cobalt oxide) formation during VOx deposition or after air exposure, producing exchange bias or inverted hysteresis without any VOx moment. A Co/Pt control without VOx, or with a nonmagnetic oxide interlayer, is not shown. Therefore the central claim is conditionally supported at best.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper reports a study of reactive-sputtered 20-nm amorphous VOx films with nominal compositions V2O3+p (p = 0.05, 0.23, 0.49) and claims room-temperature ferromagnetism with saturation magnetization ≈14 kA/m, based on SQUID magnetometry, TEM-EELS determination of V3+/V4+ ratios, transport Hall measurements in VOx/Co/Pt trilayers, and micromagnetic simulations. The authors attribute the ferromagnetism to a partial volume fraction of ferromagnetic phases embedded in an antiferromagnetic matrix, and they interpret an inverted hysteresis loop and azimuthal Hall phase shifts as evidence of antiferromagnetic exchange coupling between VOx and Co layers.","tokens_in":17288,"tokens_out":5786,"duration_ms":51504,"significance":"If the claim is correct, this would be a notable result: amorphous mixed-valent V2O3+p films would host a room-temperature ferromagnetic phase with a moment an order of magnitude below conventional ferromagnets, tunable by oxidation state, and capable of coupling to a Co overlayer—potentially relevant for spintronic applications. The paper's strengths include a systematic control of oxygen flow, a three-method EELS analysis of the V3+/V4+ ratio, structural characterization by HR-TEM/FFT, and a reasonable micromagnetic modeling framework that reproduces the qualitative features of the data. However, the evidence currently falls short of the claim of 'clear evidence' because the magnetization is not element-specific, the coupling signature lacks a critical control, and the simulations are fitted rather than predictive.","major_comments":[{"comment":"The claim of intrinsic room-temperature ferromagnetism with Ms ≈ 14 kA/m rests exclusively on macroscopic SQUID magnetometry after subtraction of a large paramagnetic substrate background; no error bars, no repeated-sample statistics, and no element-specific magnetic probe (XMCD or neutron scattering) are provided. The measured saturation moment is only 1–2% of a typical ferromagnet, precisely the magnitude expected for ppm-level FM impurities, and the paper itself acknowledges in Sec. 1 that 'more direct evidence, such as detailed measurements of the atomic moments and spin configurations within the VOx layer, would be necessary.' This missing element-specific evidence is load-bearing: without it, the central claim of intrinsic VOx ferromagnetism is not established.","section":"Sec. 1, Fig. 1, Table I"},{"comment":"The inverted hysteresis loop in VOx/Co/Pt at 300 K (Fig. 4c) is attributed to antiferromagnetic exchange coupling between VOx and Co, but no control sample without VOx (e.g., Co/Pt on the same substrate) or with a nonmagnetic oxide interlayer (e.g., AlOx/Co/Pt) is presented. Interfacial CoO formation during VOx deposition or after air exposure can produce exchange bias or inverted hysteresis without any magnetic moment in VOx. Because the coupling signature is the authors' second pillar of evidence for VOx ferromagnetism, the absence of this control is a load-bearing gap.","section":"Sec. 1, Fig. 4"},{"comment":"The micromagnetic simulations are parameterized to reproduce the observations: only 2% of VOx grains are assigned ferromagnetic order with Ms = 1.0×10^6 A/m (yielding ~15 kA/m net magnetization), and the coupling energy is adjusted to fit the AHE phase shift. The authors state that the parameters 'may not represent the actual material properties of VOx' and were 'varied over a wide range to ensure the simulations could reproduce the experimental observations.' As presented, the simulation is a fitted restatement of the partial-volume assumption rather than an independent test, and therefore it does not close the evidential gap for intrinsic ferromagnetism.","section":"Sec. 3 (Experimental), Fig. 6"},{"comment":"The reported susceptibility χ ~ 10 is not adequately defined. If χ = ∂M/∂H with M in kA/m and H in T, the value is not dimensionless and cannot be compared with |χ| < 10^-5 for paramagnetic phases; if a dimensionless susceptibility is intended, the conversion and the field range over which the slope is extracted should be specified. The manuscript does not show the raw subtraction procedure or the uncertainty on χ, which is essential because the ferromagnetic signal is comparable to the substrate background that must be removed.","section":"Sec. 1, Fig. 1"}],"minor_comments":[{"comment":"In several places the manuscript uses nonstandard or inconsistent notation: Sample 1A/2A/3A are introduced as trilayers, but Figure 5b is described in the text with colors 'red, blue, and olive' while the caption says 'red, blue, and green'; please harmonize.","section":"Fig. 5b and main text"},{"comment":"Eq. (1) is written for the untilted geometry, while the actual measurement uses a tilted geometry; the reader must consult Note S1 to understand the sin φ_H dependence. Please state in the main text that Eq. (1) is generalized to tilted geometry as Eq. (S1).","section":"Eq. (1) and Note S1"},{"comment":"The Conclusions speculate about 'gating with bias voltage' and voltage-controlled phenomena, but no gating experiment is reported in the manuscript; this should be clearly labeled as a future outlook or removed.","section":"Conclusions"},{"comment":"Typos and minor grammatical errors appear throughout, e.g., 'In ordered to confirm' (Sec. 1, paragraph on exchange coupling) and 'AHH' (Fig. 5b caption).","section":"Throughout"},{"comment":"Reference [28] is a GitHub URL without a full citation; please provide an archival reference for MuView.","section":"References"},{"comment":"The notation V2O3+p in Table I should be defined explicitly: p is the fraction of V4+ per V site, not a conventional oxygen off-stoichiometry parameter, to avoid confusion.","section":"Table I"}],"recommendation":"major_revision","confidential_remarks":"This manuscript overstates its conclusions. The phrase 'clear evidence of ferromagnetism' in the abstract should be tempered unless XMCD or equivalent element-specific data are provided. I would encourage the editor to require the additional experiments described in the major comments: element-specific magnetometry, control trilayers without VOx or with a nonmagnetic spacer, and error bars for the magnetization and susceptibility. With those additions, the paper could become publishable."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Worth knowing: this is not a claim that amorphous V2O3+p is a strong ferromagnet; it is a claim that a small volume fraction (on the order of 1%) is ferromagnetic at 300 K, and that this ferromagnetic phase exchange-couples to an adjacent Co layer. The paper itself is refreshingly explicit that the evidence is not decisive and says XMCD or neutron scattering would be needed.\n\nWhat is new: prior reports of ferromagnetism in vanadium oxides came from doped nanotubes, twin boundaries in epitaxial VO2, or oxygen-deficient nanowires. Here they show a nominally undoped, predominantly amorphous 20-nm film, made by reactive sputtering, with Ms ≈ 14 kA/m at 300 K and a susceptibility of about 10. The oxidation-state trend (Sample 1 > 2 > 3) and the fact that the Hall-effect phase shift scales with the magnetization of the VOx layer give some internal consistency. The EELS work is careful: three independent methods give the V3+/V4+ ratios, and the control measurements on V2O3 and VO2 samples are appropriate.\n\nWhere it is soft: the magnetization is only 1–2% of a typical ferromagnet, which is exactly the regime where trace Fe/Co/Ni impurities can masquerade as intrinsic. There is no element-specific measurement, and the paper admits this. The inverted M-H loop in the VOx/Co/Pt trilayer is interpreted as antiferromagnetic coupling, but interfacial CoO could produce a similar signature, and no control with a nonmagnetic spacer is shown. The micromagnetic simulations are explicitly a fit: parameters were varied to reproduce the experiments, so they cannot independently confirm the partial-volume model. The authors say as much.\n\nNone of this kills the paper. The internal consistency with oxidation state is real evidence, and the authors are honest about the gap. But the central claim remains conditionally supported, not proven. A serious referee would ask for XMCD or a careful impurity analysis, and for a control bilayer with an MgO or Al2O3 interlayer. Those are doable experiments.\n\nFor a reading group: this is a good example of a readable, honest paper about a tentative but potentially interesting effect. For me, I would not yet build on the result as established. That said, the paper deserves full peer review, not desk rejection. Send it out.","headline":"A credible and honest report of possible room-temperature ferromagnetism in amorphous V2O3+p, held back from proof by the lack of element-specific magnetometry.","tokens_in":17883,"tokens_out":2676,"would_cite":false,"duration_ms":24691,"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":"Amorphous, mixed-valence vanadium oxide films made by reactive sputtering are ferromagnetic at room temperature, with about 14 kA/m magnetization that decreases as oxygen content rises.","keywords":["vanadium oxide thin films","ferromagnetism","oxidation state control","mixed-valence vanadium","V2O3","reactive sputtering","antiferromagnetic exchange coupling","spintronics"],"falsifier":"Perform X-ray magnetic circular dichroism at the vanadium L2,3 edges on Sample 1 at 300 K: if the vanadium dichroic signal is absent or too small to account for the 14 kA/m saturation magnetization, the intrinsic-ferromagnetism claim is falsified. A complementary control, inserting a 1 nm nonmagnetic spacer between VOx and Co in the trilayer and checking whether the inverted hysteresis loop and Hall phase shift vanish together, would rule out interfacial CoO as the source of the antiferromagnetic coupling.","tokens_in":16734,"feed_emoji":"🧲","tokens_out":8231,"duration_ms":72297,"temperature":0.7,"pith_summary":"This paper claims that amorphous, non-stoichiometric vanadium oxide thin films, written $V_2O_{3+p}$ with p = 0.05, 0.23, or 0.49, are intrinsically ferromagnetic at room temperature. The 20-nm films, made by reactive sputtering with different oxygen flows, show saturation magnetization about 14 kA/m and magnetic susceptibility around 10, far above the paramagnetic or antiferromagnetic response of crystalline V2O3 and VO2. The authors attribute the magnetism to a small volume fraction, roughly one to two percent, of ferromagnetic V4+-bearing regions embedded in an antiferromagnetic V3+ matrix, rather than to weak ferromagnetism or altermagnetism, which require ordered crystal structures. They support the intrinsic claim by showing that the VOx layer couples antiferromagnetically to an adjacent cobalt layer, producing an inverted hysteresis loop and a tunable phase shift in Hall measurements that strengthens as the magnetic moment grows. If correct, the result matters because it adds a room-temperature magnetic oxide whose properties can be tuned by oxidation state and, potentially, by voltage.","feed_headline":"How oxygen controls room-temperature ferromagnetism in vanadium oxide","feed_subtitle":"Mixed V3+/V4+ films reach 14 kA/m and couple to cobalt, giving a new spintronic control knob.","key_machinery":"The load-bearing object is the disordered mixed-valence film $V_2O_{3+p}$, produced by reactive sputtering, whose V3+:V4+ ratio is quantified by three EELS methods and whose microstructure is amorphous. The central mechanism proposed is a partial-volume ferromagnetic phase: isolated, low-density V4+ regions form ferromagnetic grains inside an antiferromagnetic V3+ matrix; the small ferromagnetic volume fraction explains the low net magnetization, while the surrounding antiferromagnetic matrix produces exchange pinning and slanted hysteresis loops. In bilayers, exchange coupling between the VOx ferromagnetic grains and the Co layer is what generates the inverted hysteresis loop at 300 K and the azimuthal phase shift in tilted Hall measurements; micromagnetic simulations that set only 2% of the VOx grains ferromagnetic reproduce both the 15 kA/m loop and the Co phase shifts.","core_discovery":"On the paper's own terms, the central discovery is that ferromagnetism in vanadium oxide is not limited to impurity- or defect-engineered nanotubes, twin boundaries, or nanowires; it is a property of simple 20-nm reactive-sputtered VOx films with mixed V3+/V4+ valence. The films have Curie temperatures above 300 K, a saturation magnetization of roughly 14 kA/m, and a susceptibility of order 10, and their magnetization falls monotonically as oxygen content increases. Electron energy loss spectroscopy assigns the films to $V_2O_{3+p}$ with p = 0.05, 0.23, and 0.49, i.e., predominantly V3+ with increasing V4+; transmission electron microscopy shows a mainly amorphous structure. The authors' partial-volume model says that only about 1-2% of the film is ferromagnetic, in V4+-rich regions, while the V3+ matrix remains antiferromagnetic, which reconciles the small net magnetization with atomic moments of order one Bohr magneton and with the strong exchange coupling seen in VOx/Co/Pt trilayers. They explicitly decline to call the effect weak ferromagnetism or altermagnetism, because both require well-defined crystalline symmetry, which these amorphous films lack.","pith_inferences":["Editorial inference: Because the ferromagnetic fraction is small and sits in an antiferromagnetic matrix, the films may exhibit internal exchange bias and training effects similar to granular ferromagnet/antiferromagnet composites; measuring field-cooled loop shifts at low temperature would test this.","Editorial inference: If oxidation can be moved by ionic gating, VOx-based devices could offer voltage-controlled magnetic anisotropy, exchange coupling, or domain-wall motion without write currents; the authors mention gating only as a possible direction, not a demonstrated result.","Editorial inference: An element-specific X-ray magnetic circular dichroism measurement at the vanadium L-edges would be the decisive next experiment; a positive vanadium dichroic signal would convert the indirect coupling evidence into direct proof of vanadium-derived moments."],"forward_implications":["Room-temperature ferromagnetism in amorphous, non-stoichiometric VOx is achievable by reactive sputtering, so the material can be deposited onto arbitrary substrates without epitaxy.","Oxidation state, meaning the V3+:V4+ ratio, is a practical control knob: the saturation moment, susceptibility, and the strength of coupling to Co all decrease as p increases from 0.05 to 0.49.","VOx/Co/Pt stacks show antiferromagnetic interlayer coupling strong enough to reverse the Co spin configuration, so the VOx layer can act as a magnetic control layer in spintronic devices.","The measured 14 kA/m magnetization requires only 1-2% of the volume to be ferromagnetic, meaning local atomic moments are near one Bohr magneton, consistent with V4+ moments.","VOx introduces an in-plane anisotropy into the adjacent ferromagnet that shows up as a phase shift in Hall measurements, allowing the oxide to tailor the magnetic anisotropy of a neighboring layer."],"supporting_citations":[{"why":"reports room-temperature ferromagnetism in doped vanadium-oxide nanotubes, the prior evidence this work extends from impurity-doped to nominally pure films.","marker":"[4]"},{"why":"documents defect-mediated room-temperature ferromagnetism with about 18 kA/m saturation in VO2 twin boundaries, a key comparison for the magnitude of the observed moment.","marker":"[5]"},{"why":"provides the V3+–V4+ dimer picture by which oxygen vacancies produce ferromagnetic chains in VO2 nanowires.","marker":"[6]"},{"why":"DFT work cited by the authors as the prediction that oxygen vacancies can induce a ferromagnetic phase in vanadium oxides.","marker":"[7]"},{"why":"ab initio calculation showing vanadium moments of one to two Bohr magnetons that nonetheless order antiferromagnetically, the basis for invoking vacancies to destabilise that order.","marker":"[14]"},{"why":"defines the linear relation between the V-L3 edge energy loss and vanadium oxidation state used in the third EELS method.","marker":"[19]"},{"why":"documents inverted hysteresis loops arising from antiferromagnetic coupling, the mechanism invoked for the room-temperature VOx/Co loop.","marker":"[22]"},{"why":"supplies the micromagnetic solver used to test the partial-volume ferromagnetic model against the measured hysteresis and Hall phase shifts.","marker":"[27]"}],"fun_headline_variants":["Oxygen tuning gives vanadium oxide room-T ferromagnetism","Mixed-valence VOx films show ferromagnetism at 300 K","Oxygen content controls ferromagnetism in vanadium oxide","Ferromagnetic VOx: oxygen stoichiometry as a spintronic knob"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The identification of the measured magnetization as intrinsic to the oxide assumes that, after subtracting the substrate's paramagnetic background, all remaining signal comes from VOx rather than from trace magnetic impurities, and that the inverted loop in VOx/Co/Pt comes from VOx-Co coupling rather than interfacial CoO; if either assumption fails, the claim collapses.","fun_headline_variants_meta":{"raw":{"variants":["Oxygen tuning gives vanadium oxide room-T ferromagnetism","Mixed-valence VOx films show ferromagnetism at 300 K","Oxygen content controls ferromagnetism in vanadium oxide","Ferromagnetic VOx: oxygen stoichiometry as a spintronic knob"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001068,"raw_usage":{"total_tokens":4534,"prompt_tokens":1060,"completion_tokens":3474,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":676,"completion_tokens_details":{"reasoning_tokens":3400}},"tokens_in":676,"tokens_out":3474,"duration_ms":24554,"temperature":1.0,"reasoning_tokens":3400,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T13:31:30.911386+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Perform X-ray magnetic circular dichroism at the vanadium L2,3 edges on Sample 1 at 300 K: if the vanadium dichroic signal is absent or too small to account for the 14 kA/m saturation magnetization, the intrinsic-ferromagnetism claim is falsified. A complementary control, inserting a 1 nm nonmagnetic spacer between VOx and Co in the trilayer and checking whether the inverted hysteresis loop and Hall phase shift vanish together, would rule out interfacial CoO as the source of the antiferromagnetic coupling.","supporting_citations":[{"cited_title":"The magnetic moment as a function of temperature was recorded starting from a saturated state at 10 K under a magnetic field of 1 T","cited_arxiv_id":null,"evidence_quote":"reports room-temperature ferromagnetism in doped vanadium-oxide nanotubes, the prior evidence this work extends from impurity-doped to nominally pure films."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"documents defect-mediated room-temperature ferromagnetism with about 18 kA/m saturation in VO2 twin boundaries, a key comparison for the magnitude of the observed moment."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"provides the V3+–V4+ dimer picture by which oxygen vacancies produce ferromagnetic chains in VO2 nanowires."},{"cited_title":"Blümer, E","cited_arxiv_id":null,"evidence_quote":"DFT work cited by the authors as the prediction that oxygen vacancies can induce a ferromagnetic phase in vanadium oxides."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"ab initio calculation showing vanadium moments of one to two Bohr magnetons that nonetheless order antiferromagnetically, the basis for invoking vacancies to destabilise that order."},{"cited_title":"Dzyaloshinsky, J","cited_arxiv_id":null,"evidence_quote":"defines the linear relation between the V-L3 edge energy loss and vanadium oxidation state used in the third EELS method."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"documents inverted hysteresis loops arising from antiferromagnetic coupling, the mechanism invoked for the room-temperature VOx/Co loop."},{"cited_title":"Maity, D","cited_arxiv_id":null,"evidence_quote":"supplies the micromagnetic solver used to test the partial-volume ferromagnetic model against the measured hysteresis and Hall phase shifts."}],"review_version":1}