{"id":"e28d5933-7031-4499-bedd-7fe275d5551e","arxiv_id":"2412.04957","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"A strained oxide film with a newly identified rippled atomic structure is both ferromagnetic and insulating, with an extrapolated Curie temperature near 1200 K.","lead":"A new atomic arrangement in a strontium iron oxide film is reported to be both a strong magnet and an electrical insulator, with magnetism that may last up to very high temperatures. If confirmed, the material could enable more efficient spintronic devices that work well above room temperature.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 1200 K Curie temperature is extrapolated from M(T) only up to 900 K, where re-oxidation already begins; if that extrapolation is invalid the ultrahigh-temperature claim is unsupported, though ferromagnetism above 900 K remains.","rationale":"The paper's core contribution is the simultaneous ferromagnetism and insulation in a new RIL SrFeO2 phase. The structural data (XRD, iDPC-STEM, XAS) and the thickness/transfer experiments provide plausible support for ferromagnetism and insulation. The weakest point is exactly the one the reader flagged: the TC near 1200 K value is an extrapolation beyond the thermal-stability window of the phase. The text itself says the RIL phase begins to re-oxidize near 900 K, so the measured M(T) decline above that temperature may be a chemical-transformation artifact rather than the intrinsic demagnetization curve. Without a specified functional form, uncertainty band, or direct high-temperature measurement, the headline number is not robust. This does not undermine the demonstrated 900 K ferromagnetism, the 3.27 eV gap, or the SMR result; it only affects the quantitative ultrahigh-temperature claim. Therefore the reader's CONDITIONAL verdict is appropriate and no further adjustment is needed.","tokens_in":8801,"tokens_out":5497,"duration_ms":61383,"concrete_test":"Measure M(T) on a 20-nm RIL-SFO film protected by a hermetic cap (e.g., 10 nm Al2O3 or the reported LaAlO3-delta encapsulation) in flowing Ar or vacuum from 300 K to 1250 K, with post-run XRD to confirm the RIL phase did not reoxidize, and extract TC via Arrott plots at several temperatures. If a clear transition to paramagnetism is observed near 1200 K, the extrapolation is confirmed; if the film cannot be measured above about 900 K without chemical change, or if Arrott analysis gives TC well below 1000 K, the TC claim should be revised to ferromagnetism stable to at least 900 K.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The most load-bearing assumption is that the RIL phase has an intrinsic Curie temperature near 1200 K. The evidence is an extrapolation of the M-T curve, but the curve is measured only to about 900 K because, in the authors' own words, the RIL phase begins to be partially re-oxidized there (main text, Fig. 2). Above 900 K the measured moment may therefore decline because the sample is progressively converting to a less-magnetic phase, not because the ferromagnetic order is approaching its thermodynamic transition. No fitting function, confidence interval, or control for the phase transformation is provided for the extrapolation. Moreover, M(T) is measured in a 6000 Oe field, so without Arrott-plot or zero-field data the field-induced tail cannot be separated from a true Curie transition. If this concern lands, the paper still demonstrates robust ferromagnetism up to at least 900 K, but the quantitative TC near 1200 K and the greater-than-1000 K headline would be unsupported.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports a new structural phase of SrFeO2, termed ripple-infinite-layer (RIL), obtained by lengthy topotactic reduction of brownmillerite SrFeO2.5 films under high compressive strain. The authors claim that RIL-SrFeO2 is a ferromagnetic insulator with magnetization of about 180 emu/cc, resistivity above 10^6 Ω·cm, a band gap near 3.27 eV, an extrapolated Curie temperature of about 1200 K, ferromagnetism retained down to 1.0 nm thickness on the substrate and 2.0 nm in freestanding films, and spin Hall magnetoresistance of about 2.6‰ in Pt/RIL devices. DFT+U calculations are presented in support of the claim that the IL phase favors antiferromagnetism while the RIL phase favors ferromagnetism.","tokens_in":9024,"tokens_out":3608,"duration_ms":41537,"significance":"If the claims hold, this is a notable result: an insulating ferromagnet with a Curie temperature extrapolated above 1000 K, combined with thickness scalability to the nanometer range and transferrability to arbitrary substrates, would be a strong advance for spintronics and topological electronics. The paper's strengths are its extensive structural characterization (XRD, HAADF/ABF-STEM, iDPC, EDS, TOF-SIMS), the XAS-based valence assignment, the high-temperature and thickness-dependent magnetometry, the freestanding-film transfer demonstration, the SMR device results, and the independent DFT+U explanation of the magnetism. The central quantitative claim of an ultrahigh Curie temperature, however, rests on an extrapolation that the manuscript does not document, and the resistivity claim is a measurement-limit lower bound; these points need clarification before the strongest conclusions can be accepted.","major_comments":[{"comment":"The 1200 K Curie temperature is obtained by extrapolating the magnetization-temperature curve, but the measured data appear to extend only to about 900 K, where the authors state that the RIL phase begins to be partially re-oxidized. Because the decline of M above 900 K could be caused by progressive conversion to a less-magnetic phase rather than by approach to the intrinsic magnetic transition, the extrapolation is not self-evidently valid. The manuscript does not disclose the functional form used for the fit, the fitting range, or a confidence interval, and the M(T) data are measured in a 6000 Oe field, so a field-induced tail cannot be distinguished from a true Curie transition without additional zero-field or Arrott-plot data. Please provide a detailed description of the extrapolation, show the fit against the measured points, report uncertainty, and ideally support the intrinsic TC with an alternative method such as high-field magnetization analysis, Mössbauer spectroscopy, or specific-heat measurements. If this is not possible, the abstract and conclusions should be revised to state robust ferromagnetism at or above 900 K without the quantitative 1200 K claim.","section":"Fig. 2A and main text, page 6"},{"comment":"The claim of resistivity above 10^6 Ω·cm is presented as a measured material property, but the text says the resistivity 'has gone beyond the maximum limit of our measurement setup,' so this is a lower bound, not a measured value. This distinction should be explicit in the main text and abstract, and the measurement geometry, temperature, and setup limit should be reported. Similarly, the band gaps from optical absorption and the magnetization values are quoted without error bars; given that the central comparison in Fig. 3B depends on these quantities, quantitative uncertainties should be provided.","section":"Fig. 3 and Methods"},{"comment":"The thickness-dependent and high-temperature magnetic data are central to the claim of robust ferromagnetism down to 1.0 nm and in freestanding films, but the reported values lack error bars and the hysteresis loops are not quantified in terms of coercivity or remanence. For the 1.0 nm film, the inset shows a loop at 400 K but no saturation magnetization value or measurement uncertainty is given; for transferred films the text mentions 'minor weakening' without a number. Please add quantitative magnetization, coercivity, and error estimates, and specify how many samples were measured for the thickness series.","section":"Fig. 2B, 2C, 2F"}],"minor_comments":[{"comment":"Please state the field and temperature at which 180 emu/cc was determined; the M(T) curve is measured at about 6000 Oe, so the magnetization value should be reported with the same field and temperature conditions.","section":"Page 3, 'desirable magnetization'"},{"comment":"The statement that the 1200 K TC is reconfirmed in the mixed IL & RIL phase refers to fig. S11, but the extrapolation procedure and fit should be shown for that system as well, or the claim should be softened if the data are not as complete.","section":"Page 6, 'mixed IL & RIL' reconfirmation"},{"comment":"The definition of δ is confusing because the brownmillerite phase is SrFeO2.5 and the reduced phases are SrFeO2; please define δ explicitly for each phase or use a less ambiguous notation.","section":"Page 5, 'SrFeO3-δ (0.5 ≤ δ ≤1.0)'"},{"comment":"The comparison plot of TC versus resistivity for known ferromagnetic/ferrimagnetic insulators should include error bars or at least a data table in the supplement, since the position of RIL-SrFeO2 in that plot is a central visual claim.","section":"Fig. 3B"}],"recommendation":"major_revision","confidential_remarks":"The stress-test concern about the Curie-temperature extrapolation is genuine and lands on the paper's headline claim. The manuscript otherwise contains a coherent set of structural, magnetic, and transport observations, and the 900 K ferromagnetism claim would still be strong if the 1200 K extrapolation were removed. I see no citation or novelty concerns; the self-citations (refs. 39 and 40) are peripheral to the main results. The paper fits the scope of cond-mat.mtrl-sci. If the authors can supply a fully documented extrapolation with uncertainty, or revise the central claim to 'ferromagnetism at temperatures up to 900 K' while presenting the 1200 K value only as a rough estimate, the paper would be publishable in my view."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague—this is a serious experimental paper with one number that overreaches. The new thing is a ripple-infinite-layer phase of SrFeO2, obtained by long topotactic reduction of brownmillerite under high compressive strain. The structural work is careful: XRD, iDPC-STEM, XAS, EDS, TOF-SIMS all support a distinct RIL phase with Fe2+ and no hydrogen. The claim that this phase is a ferromagnetic insulator is backed by magnetization, hysteresis up to 750 K, thickness down to 1 nm, and resistivity >10^6 ohm-cm. The DFT+U argument that RIL favors FM while IL favors AFM is independent, not a fit to the measured moment. That is real evidence and I trust it. The SMR demonstration on Pt/RIL is a useful add-on.\n\nThe soft spot is exactly where the reader put it: the 1200 K Curie temperature is an extrapolation of M(T) measured only to about 900 K, and the authors say the phase begins to re-oxidize there. No fitting function, no confidence interval, no control for the phase transformation. With 6000 Oe applied, you cannot separate a field-induced tail from a true transition. So the 'ultrahigh-temperature' headline is not established. What is established is strong ferromagnetism up to at least 900 K, which is already unusual for an insulating oxide film. The resistivity claim is also a lower bound from measurement setup, not a measured value. Minor.\n\nI'd treat the 1200 K figure as a hypothesis, not a result. If it falls, the paper still stands as a new FM insulating phase with TC > 900 K and strong thickness scaling. The citations look fine; the two self-citations are peripheral. No circularity.\n\nThis deserves peer review. A serious referee should push on the magnetic extrapolation and ask for Arrott plots or another method to estimate TC, and error bars on the key quantities. I would not cite the 1200 K number in my own work until the extrapolation is justified. Reading group: maybe, if anyone is interested in oxide topotactic phases.","headline":"A new ripple-infinite-layer SrFeO2 phase gives a plausible ferromagnetic insulator up to 900 K, but the 1200 K Curie temperature is an unsupported extrapolation and the headline overreaches.","tokens_in":9631,"tokens_out":2023,"would_cite":false,"duration_ms":20461,"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 heavily reduced strontium ferrite phase, ripple-infinite-layer SrFeO2, is claimed to be a ferromagnetic insulator with an extrapolated Curie temperature of about 1200 K and ferromagnetism down to 1 nm thickness.","keywords":["ripple-infinite-layer","SrFeO2","ferromagnetic insulator","topotactic reduction","brownmillerite","spin Hall magnetoresistance","strain engineering","high Curie temperature"],"falsifier":"Measure the magnetization-versus-temperature curve of RIL SrFeO2 in an oxygen-free atmosphere from 900 K upward: if the magnetization does not smoothly approach zero near 1200 K (or if the RIL phase decomposes before reaching it), the ultrahigh Curie temperature is not supported.","tokens_in":8616,"feed_emoji":"🧲","tokens_out":7601,"duration_ms":66263,"temperature":0.7,"pith_summary":"The paper claims that a heavily reduced phase of strontium ferrite, SrFeO2, arranged in a newly identified ripple-infinite-layer (RIL) structure, is simultaneously a robust ferromagnet and an electrical insulator, two properties usually considered mutually exclusive. The authors report a magnetization of roughly 180 emu/cc at low temperature, resistivity above $10^6$ Ω·cm, an optical bandgap of about 3.27 eV, and a Curie temperature extrapolated to about 1200 K. Ferromagnetic order is retained down to about 1.0 nm thickness on its growth substrate and about 2.0 nm in freestanding membranes, and Pt/RIL devices show spin Hall magnetoresistance up to about 2.6 parts per thousand at room temperature. The broader significance is that a single material with both high-temperature ferromagnetism and insulating behavior is a long-sought building block for spintronic and topological devices.","feed_headline":"Rippled atomic layers yield a 1200-K ferromagnetic insulator","feed_subtitle":"Rippling the FeO2 planes keeps SrFeO2 ferromagnetic and insulating down to 1-nm films.","key_machinery":"The load-bearing object is the ripple-infinite-layer (RIL) structure of SrFeO2, in which the FeO2 layers run out of plane and ripple instead of lying flat. It is generated by a long topotactic reduction (over 120 hours) of brownmillerite SrFeO2.5 films grown on YAlO3 substrates that impose compressive strains above 6%, and it is visualized with integrated differential phase contrast scanning transmission electron microscopy. The mechanism the authors propose is a change in exchange balance: the rippled Fe-O-Fe bond angles weaken super-exchange and simultaneously enhance direct Fe-Fe exchange, flipping the ground state from antiferromagnetic to ferromagnetic while preserving the square-planar FeO4 coordination and the insulating gap. Density functional theory plus Hubbard U calculations are used to support the energetic preference for ferromagnetism in the RIL structure.","core_discovery":"The central discovery is that the magnetic ground state of SrFeO2 switches from antiferromagnetic to ferromagnetic when the flat FeO2 planes of the conventional infinite-layer phase are deformed into the out-of-plane rippled layers of the RIL phase. The authors show that this structural deformation, rather than any change in iron valence or oxygen hybridization, is responsible for the magnetism: rippling reduces the Fe-O-Fe super-exchange interaction and strengthens Fe-Fe direct exchange, favoring ferromagnetism. The RIL phase is obtained by topotactic reduction of brownmillerite SrFeO2.5 films under compressive strain above 6%, and it combines a large magnetization with a bandgap near 3.27 eV and resistivity above $10^6$ Ω·cm. Extrapolating the measured magnetization-temperature curve gives a Curie temperature near 1200 K, and magnetization is detectable even in 1.0-nm-thick films at 400 K.","pith_inferences":["If the 1200 K extrapolation is taken at face value, RIL SrFeO2 would be among the highest-temperature ferromagnetic insulators known, and the rippling mechanism could be deliberately sought in other infinite-layer oxides such as nickelates.","The reported resistivity is only a lower bound set by the measurement limit, so the true insulating character (e.g., Mott vs. band insulator) remains untested by transport alone.","Because the RIL phase forms only under very high compressive strain, an in-plane testable prediction is that relaxing the strain should suppress ferromagnetism, which would confirm the exchange-deformation mechanism.","The freestanding films retain ferromagnetism with minor weakening, suggesting that the magnetic order is intrinsic to the RIL structure; comparing films transferred to substrates with different lattice parameters would separate strain effects from chemical effects."],"forward_implications":["The combination of extrapolated $T_C$ near 1200 K with resistivity above $10^6$ Ω·cm places RIL SrFeO2 outside the usual inverse correlation between magnetic ordering temperature and electrical insulation.","Ferromagnetism down to 1.0 nm on substrate and 2.0 nm freestanding makes the phase a candidate for ultrathin magnetic tunnel barriers, spin filters, and spin-orbit torque devices.","Pt/RIL Hall bars show a spin Hall magnetoresistance ratio up to about 2.6‰ at 300 K, an improvement of more than 200% over previously reported heavy-metal/ferromagnet or ferrimagnet devices.","The films can be released from the growth substrate and transferred to flexible PET, silicon, or glass while preserving crystal structure and magnetic properties, enabling flexible or silicon-integrated spintronics."],"supporting_citations":[{"why":"Supplies a strain-induced high-temperature perovskite ferromagnetic insulator used as the state-of-the-art comparison for Tc.","marker":"[4]"},{"why":"Provides the sacrificial water-soluble layer method used to lift off and transfer the RIL films.","marker":"[27]"},{"why":"Establishes the spin Hall magnetoresistance effect in Pt/YIG that the Pt/RIL devices are compared against.","marker":"[32]"},{"why":"Describes the square-planar FeO2 coordination family of which the RIL structure is a rippled variant.","marker":"[33]"},{"why":"Reports the parent infinite-layer SrFeO2 phase whose flat FeO2 planes the RIL structure deforms.","marker":"[34]"},{"why":"Documents the low-temperature anisotropic oxygen diffusion in iron oxides that underlies the topotactic reduction route.","marker":"[37]"},{"why":"Gives the quantitative framework for extracting spin Hall magnetoresistance ratios in ferromagnetic insulator/normal metal hybrids.","marker":"[53]"}],"fun_headline_variants":["Rippling FeO2 planes creates a 1200-K ferromagnetic insulator","Ultrathin magnetic insulator: rippled SrFeO2 reaches 1200 K","Rippled atomic layers yield 1200-K ferromagnetism in 1-nm insulator","Ferromagnetic and insulating: rippled SrFeO2 stays magnetic at 1 nm"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The 1200 K Curie temperature is an extrapolation of a magnetization curve measured only up to about 900 K, because the RIL phase starts to re-oxidize at that point; if the assumed shape of the magnetization falloff is wrong, the ultrahigh temperature claim fails, although ferromagnetism up to 900 K would still stand.","fun_headline_variants_meta":{"raw":{"variants":["Rippling FeO2 planes creates a 1200-K ferromagnetic insulator","Ultrathin magnetic insulator: rippled SrFeO2 reaches 1200 K","Rippled atomic layers yield 1200-K ferromagnetism in 1-nm insulator","Ferromagnetic and insulating: rippled SrFeO2 stays magnetic at 1 nm"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00074,"raw_usage":{"total_tokens":3297,"prompt_tokens":933,"completion_tokens":2364,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":549,"completion_tokens_details":{"reasoning_tokens":2271}},"tokens_in":549,"tokens_out":2364,"duration_ms":19141,"temperature":1.0,"reasoning_tokens":2271,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T21:06:55.637249+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the magnetization-versus-temperature curve of RIL SrFeO2 in an oxygen-free atmosphere from 900 K upward: if the magnetization does not smoothly approach zero near 1200 K (or if the RIL phase decomposes before reaching it), the ultrahigh Curie temperature is not supported.","supporting_citations":[{"cited_title":"Meng et al","cited_arxiv_id":null,"evidence_quote":"Supplies a strain-induced high-temperature perovskite ferromagnetic insulator used as the state-of-the-art comparison for Tc."},{"cited_title":"Lu et al","cited_arxiv_id":null,"evidence_quote":"Provides the sacrificial water-soluble layer method used to lift off and transfer the RIL films."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes the spin Hall magnetoresistance effect in Pt/YIG that the Pt/RIL devices are compared against."},{"cited_title":"Tassel, H","cited_arxiv_id":null,"evidence_quote":"Describes the square-planar FeO2 coordination family of which the RIL structure is a rippled variant."},{"cited_title":"Tsujimoto et al., Infinite-layer iron oxide with a square-planar coordination","cited_arxiv_id":null,"evidence_quote":"Reports the parent infinite-layer SrFeO2 phase whose flat FeO2 planes the RIL structure deforms."},{"cited_title":"Inoue et al., Anisotropic oxygen diffusion at low temperature in perovskite - structure iron oxides","cited_arxiv_id":null,"evidence_quote":"Documents the low-temperature anisotropic oxygen diffusion in iron oxides that underlies the topotactic reduction route."},{"cited_title":"Althammer et al., Quantitative study of the spin Hall magnetoresistance in ferromagnetic insulator/normal metal hybrids","cited_arxiv_id":null,"evidence_quote":"Gives the quantitative framework for extracting spin Hall magnetoresistance ratios in ferromagnetic insulator/normal metal hybrids."}],"review_version":1}