{"id":"43e6d83c-79c5-4380-92cb-20a09d3dd534","arxiv_id":"2507.05879","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":7,"one_line_summary":"Oxygen vacancies in (LaPrNdSmEu)NiO3-δ films first lower the metal-insulator transition temperature and then, at higher vacancy levels, drive the films into a fully insulating and disordered state.","lead":"This paper grows thin films of a high-entropy nickelate oxide with different amounts of missing oxygen atoms and measures how their electrical behavior changes. It finds that small amounts of oxygen vacancies first make the metal-insulator transition colder, while larger amounts eliminate the metallic state entirely, a two-faced response not seen in simpler oxides.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Quantitative δ values appear to overcount oxygen vacancies by a factor of two (the fitted Ni2+ fraction is used directly as δ rather than divided by the two electrons donated per vacancy), so the phase-diagram x-axis is uncalibrated; the non-monotonic PO2 trend itself is not in question.","rationale":"The reader's weakest assumption is the uncalibrated XAS-based δ scale, and I agree that this is the most load-bearing assumption. I add a sharper, internal consistency point: for NiO3−δ, charge neutrality requires each oxygen vacancy to produce two Ni2+ ions, so a fitted Ni2+ fraction f should correspond to δ = f/2, not δ = f. The paper's reported values appear to ignore this factor of two, which would compress the claimed δ range from 0–0.2 to roughly 0–0.1. Even with this correction, the qualitative Janus-faced behavior (TMIT first decreasing, then the metallic phase being suppressed) remains as a function of PO2, so the paper's central observation is not destroyed. The CONDTIONAL verdict already issued by the reader is appropriate: the phase diagram plot versus δ and the quantitative comparison with electron antidoping require an independent, charge-consistent calibration of the oxygen-vacancy concentration. A secondary concern is the use of three-dimensional weak-localization theory in 5–6 nm films, but that does not affect the non-monotonic TMIT trend itself.","tokens_in":11475,"tokens_out":7897,"duration_ms":98322,"concrete_test":"Reanalyze the published Ni L2-edge spectra: extract the NiO coefficient c for each film from the same two-component fit, and independently estimate the Ni2+ fraction f_ref in the 150 mTorr reference from the 870 eV feature. Compute the charge-neutral δ as δ = [f_ref + c(1 − f_ref)]/2. If this returns roughly 0.0075, 0.07, and 0.1 instead of 0.015, 0.14, and 0.2, the δ axis in Fig. 5(c) must be corrected and the manuscript should state whether a nonstandard definition of δ was used. Optionally cross-check with calibrated Ni 2p XPS or RBS on a stoichiometric film to fix the zero of the scale.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is a phase sequence as a function of oxygen-vacancy concentration δ, so the δ scale is load-bearing. The paper estimates δ by linear-combination fitting of Ni L2-edge XAS with NiO (Ni2+) and the 150 mTorr film (assumed δ≈0) as references. If c is the fitted NiO coefficient, it is effectively a Ni2+ fraction. Each missing oxygen atom in NiO3−δ donates two electrons and reduces two Ni3+ to Ni2+, so charge conservation gives Ni2+ fraction f = 2δ for a two-valence Ni manifold. The reported values ('δ around 0.015, 0.14, 0.2') are consistent with treating the fitted coefficients directly as δ, not as f/2. If so, the actual δ values are roughly 0.0075, 0.07, and 0.1, halving the claimed 0–0.2 range. The problem is compounded because the 150 mTorr reference already contains a Ni2+ feature near 870 eV, so the zero of the δ scale is not established, and the O K-edge indicates square-planar Ni2+ environments, whose XAS line shape differs from octahedral NiO; a two-component linear combination with NiO is therefore not a clean valence meter. The non-monotonic transport trend as a function of PO2 is internally consistent and would survive a corrected δ axis, but the specific 'δ from 0 to 0.2' phase diagram in Fig. 5(c) and the comparison to electron antidoping in RENiO3 (where δ has a stoichiometric meaning) rest on this unverified and numerically inconsistent calibration.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports a systematic study of oxygen-vacancy effects in single-crystalline high-entropy nickelate films [La0.2Pr0.2Nd0.2Sm0.2Eu0.2]NiO3−δ grown by pulsed laser deposition at oxygen pressures from 150 mTorr down to 10 mTorr. Structural characterization (RHEED, synchrotron XRD with Laue oscillations) indicates single-crystalline, perovskite-phase films, and XAS at the Ni L2 and O K edges is used to infer increasing oxygen-vacancy content with decreasing PO2. Temperature-dependent sheet resistance shows a monotonic increase of room-temperature resistance with decreasing PO2, while the MIT temperature first decreases from about 185 K (150 mTorr) to a hysteresis-free, reduced transition (25 mTorr), then increases to 285 K (15 mTorr), and finally the metallic phase is absent (10 mTorr). Low-temperature transport is analyzed with 2D Mott VRH for the 50 mTorr film, 3D weak localization for the 25 and 15 mTorr films, and nearest-neighbor hopping for the 10 mTorr film. The central claim is a 'Janus-faced' influence of oxygen vacancies, with a proposed phase diagram in δ spanning 0 to 0.2.","tokens_in":11877,"tokens_out":3485,"duration_ms":37723,"significance":"If the quantitative vacancy scale is correct, the non-monotonic evolution of the MIT in a high-entropy nickelate would be an interesting counterpoint to the monotonic effects reported for VO2, V2O3, and electron-doped RENiO3, and it would extend the study of disorder-driven phases into HEOs. The paper has clear strengths: the film series is well characterized structurally, the transport trend with PO2 is presented transparently, the magnetotransport analysis is detailed, and the comparison with a NdNiO3−δ film grown under the same conditions provides a useful control. The main caveat is that the quantitative δ axis, which is load-bearing for the phase diagram and for the comparison to electron antidoping, rests on a single linear-combination XAS analysis without error bars and with an assumed δ≈0 reference that itself shows a Ni2+ feature.","major_comments":[{"comment":"The quantitative oxygen-vacancy scale δ is load-bearing for the claimed phase diagram, but the linear-combination analysis in the XAS section does not establish it. Each oxygen vacancy in NiO3−δ formally donates two electrons and reduces two Ni3+ to Ni2+, so the fitted NiO (Ni2+) coefficient c is a Ni2+ fraction f with f ≈ 2δ, not δ itself. The reported values (δ ≈ 0.015, 0.14, 0.2) appear to set δ = c, which would overestimate δ by roughly a factor of two; the corrected scale would be about 0.0075, 0.07, and 0.1. Because Fig. 5(c) and the comparison with electron antidoping in RENiO3 rest on this specific calibration, the authors need to justify the mapping (or divide by the electron count), report error bars from the fitting, and provide an independent check (e.g., XPS quantification or lattice-parameter calibration). The non-monotonic TMIT vs PO2 trend would survive, but its mapping to δ is what supports the vacancy-concentration phase diagram.","section":"XAS analysis and Fig. 5(c)"},{"comment":"The zero of the δ scale is not established. The 150 mTorr film is assumed to have δ≈0, yet the text notes a minor Ni2+ feature around 870 eV; if this film already contains some Ni2+, all δ values shift. Additionally, the O K-edge shows a square-planar Ni2+ feature around 529.4 eV, and such Ni2+ sites have a different L-edge line shape than octahedral NiO; a two-component NiO + reference linear combination is therefore not a clean valence meter. The authors should quantify the reference Ni2+ content, test a three-component fit, or soften the quantitative δ claims in the abstract and in Fig. 5(c).","section":"XAS reference and O K-edge discussion"}],"minor_comments":[{"comment":"There is a typo 'due to to the 3d8L → c3d8 transition'; it should read 'due to the 3d8L → c3d8 transition.'","section":"O K-edge paragraph"},{"comment":"In the sentence defining the Frydman model, 'a is constant and order of unity' is missing a verb; it should read 'a is a constant of order unity.'","section":"Magnetoconductance model description"},{"comment":"The notation 'P O2' appears with an unintended space in the text near Fig. 1; it should be PO2 for consistency.","section":"Fig. 2 caption and main text"},{"comment":"Reference [7] lacks a volume and page number ('Advanced Materials n/a, 2415351'); please update if available.","section":"Reference list"},{"comment":"The phrase 'Janus-faced' is used in the abstract and summary but not explicitly defined; consider adding one sentence specifying that it refers to opposite responses of TMIT at low versus high vacancy concentrations.","section":"Abstract and text"}],"recommendation":"major_revision","confidential_remarks":"The central experimental trend is likely robust, but the quantitative δ scale is calibrated in a way that appears to overcount vacancies by a factor of two and lacks error bars. Since the headline phase diagram and the comparison to electron antidoping depend on that scale, I recommend major revision rather than rejection; the fix is within the manuscript's scope."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe thing to know: this is a solid experimental report with a genuinely interesting central observation, but the δ axis is not as clean as the paper presents. The non-monotonic TMIT versus PO2 trend — first down, then up, then insulating — looks robust and is new for rare-earth nickelates. Prior OV work showed monotonic increases; this initial decrease is a real finding. The transport data are clearly presented: smooth curves, monotonic sheet resistance increase with lower PO2, hysteresis suppression, and plausible phase classification.\n\nThe main soft spot is the calibration of δ. The XAS linear combination of NiO and the 150 mTorr film gives a Ni2+ fraction, not directly δ. Each oxygen vacancy reduces two Ni3+ to Ni2+, so Ni2+ fraction f ≈ 2δ. The reported values (0.015, 0.14, 0.2) look like f, not f/2, meaning the actual vacancy concentrations are roughly half. The XPS 'confirmation' may suffer the same confusion. Also, the 150 mTorr reference already has a Ni2+ peak, so the zero point is not well established, and the O K-edge shows square-planar Ni2+, which has a different line shape than octahedral NiO. So the quantitative δ scale is uncertain, likely off by a factor of two, and possibly skewed nonlinearly. The qualitative trend survives as a function of PO2, but comparing quantitatively to electron antidoping in RENiO3 is not yet justified.\n\nSecondary issues: applying 3D weak localization to 5–6 nm films without discussing the thickness crossover is a stretch; a 2D model is natural at that thickness unless the dephasing length is shorter. The assignment of the 10 mTorr film to a Mott-Anderson insulator is plausible but not proven; the NNH fit and a comparison to one NdNiO3 film is suggestive, not conclusive.\n\nThe paper ships no code or raw data, but the measurements are reproducible in principle and the fits are standard. I think the central claim will survive with a corrected δ calibration. The paper is worth a serious referee, but it needs a major revision on the quantification of OV content and a more careful treatment of the localization dimensionality.","headline":"Solid new experimental observation of non-monotonic MIT response to oxygen vacancies in high-entropy nickelates, but the δ calibration is likely off by a factor of two.","tokens_in":12514,"tokens_out":2664,"would_cite":true,"duration_ms":30111,"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":"Oxygen vacancies act as a two-faced control knob for the metal–insulator transition in high-entropy nickelate films, first lowering the transition temperature and then driving the system into a Mott–Anderson insulator.","keywords":["high entropy oxide","oxygen vacancy","metal-insulator transition","rare-earth nickelate","Mott-Anderson insulator","weak localization","variable-range hopping","pulsed laser deposition"],"falsifier":"Measure the oxygen content of identically grown films by an independent method, such as ion-beam analysis with elastic recoil detection, and compare the results with the Ni L2-edge linear-combination estimates; if the 10 mTorr film does not sit near $\\delta \\approx 0.2$, or if films grown at fixed total pressure but different oxygen partial pressure do not reproduce the same phase sequence, the claimed vacancy-controlled phase diagram would need revision.","tokens_in":11233,"feed_emoji":"🕳️","tokens_out":9016,"duration_ms":85614,"temperature":0.7,"pith_summary":"This paper tries to establish that oxygen vacancies—sites where oxygen atoms are missing from the perovskite lattice—act as a two-faced control knob for electronic behavior in a high-entropy nickelate (LPNSE)NiO$_{3-\\delta}$, a five-cation rare-earth nickelate film. Starting from a film that undergoes a first-order metal–insulator transition near 185 K, adding a small number of vacancies lowers the transition temperature and removes its thermal hysteresis, while adding more vacancies first raises the transition back to about 285 K and then eliminates the metallic phase entirely. The authors trace the underlying sequence from a bond-disproportionated metal through a weakly localized phase to a Mott–Anderson insulator as $\\delta$ grows from roughly 0 to 0.2. The interest is that vacancies deliver electron doping and oxygen-sublattice disorder at the same time, so they offer a growth-pressure-only route to tune correlated-electron phases in a high-entropy oxide platform relevant for correlated-electron and optoelectronic devices.","feed_headline":"Oxygen vacancies first lower, then erase a nickelate metal transition","feed_subtitle":"Tuning growth oxygen alone sweeps the film from metal to weakly localized to fully insulating.","key_machinery":"The central object is the oxygen vacancy inside the Ni–O network of a negative-charge-transfer nickelate, where the ground state is dominated by the $d^8\\underline{L}$ configuration. Each missing oxygen both dopes electrons into the Ni sites and removes a bonding oxygen, creating random local disorder and reduced Ni coordination. In the pristine film the MIT is a bond-disproportionation transition ($d^8\\underline{L}+d^8\\underline{L}\\to d^8+d^8\\underline{L}^2$) often described as polaron condensation; vacancies disrupt this order, which the paper invokes to explain the initial lowering of $T_{\\mathrm{MIT}}$. At higher vacancy content, weakened Ni–O hybridization (a diminishing O K-edge pre-peak), lowered charge-transfer energy, and disorder on both cation and anion sublattices drive the system into a Mott–Anderson insulating state described by nearest-neighbor hopping. The vacancy concentration itself is estimated by fitting Ni L2-edge X-ray absorption spectra as a linear combination of NiO and the 150 mTorr film.","core_discovery":"The paper claims that in single-crystalline films of the high-entropy rare-earth nickelate [La$_{0.2}$Pr$_{0.2}$Nd$_{0.2}$Sm$_{0.2}$Eu$_{0.2}$]NiO$_{3-\\delta}$, the oxygen-vacancy concentration is a non-monotonic driver of the metal–insulator transition. As growth oxygen pressure is lowered from 150 to 25 mTorr, the MIT temperature falls from about 185 K to a hysteresis-free, weakly localized state; further lowering to 15 mTorr pushes $T_{\\mathrm{MIT}}$ up to 285 K, and at 10 mTorr the film is fully insulating below room temperature. X-ray absorption spectroscopy yields $\\delta \\approx 0.015$, 0.14, and 0.2 for the 50, 25, and 10 mTorr films, while room-temperature sheet resistance rises monotonically throughout. The authors interpret the initial drop as disruption of the bond-disproportionation and polaron-condensation mechanism that drives the pristine transition, and the final, fully insulating state as a Mott–Anderson insulator arising from combined disorder on the rare-earth and oxygen sublattices, going beyond the earlier \"electron antidoping\" picture.","pith_inferences":["If the linear-combination calibration of $\\delta$ is off, the qualitative non-monotonic shape as a function of growth pressure would likely survive, but the claim that vacancy concentration is the controlling variable would need re-anchoring; a direct oxygen-content measurement would settle which axis is real.","The two-faced response should not be limited to this one compound: any negative-charge-transfer nickelate with a bond-disproportionation MIT and intrinsic A-site disorder could show the same initial drop followed by localization, so strain or A-site composition may be used to search for it elsewhere.","The collapse of hysteresis at intermediate vacancy content suggests quenched disorder destroys the first-order character of the MIT before the metallic phase disappears; heating- and cooling-rate-dependent measurements could probe how sharp the remaining transition is.","A series grown at fixed total pressure with varying O$_2$/Ar ratio would separate the effect of oxygen vacancies from other growth-pressure effects, a testable extension the paper does not carry out."],"forward_implications":["Varying only the oxygen pressure during growth sweeps the same film through four electronic regimes: a first-order MIT metal, a hysteresis-free weakly localized metal, a strongly localized insulator, and a fully Anderson-localized insulator.","The initial decrease of $T_{\\mathrm{MIT}}$ gives experimental access to a low-vacancy regime previously predicted but hard to reach in simpler nickelates, with the response controlled by reduced charge-transfer energy rather than ordinary band filling.","Since sheet resistance at room temperature rises monotonically while $T_{\\mathrm{MIT}}$ is non-monotonic, vacancy engineering can separate these two transport properties, which matters for device design.","The fully insulating state of the high-entropy film is not the site-selective Ni$^{2+}$ Mott state seen in LaNiO$_{3-\\delta}$, but a Mott–Anderson insulator that needs disorder on both rare-earth and oxygen sublattices; a NdNiO$_{3-\\delta}$ comparison film does not show the same localization."],"supporting_citations":[{"why":"Defines the negative charge-transfer ground state and bond disproportionation of rare-earth nickelates that the pristine film's MIT is based on.","marker":"[19]"},{"why":"Supplies the bipolaron-condensation picture whose disruption by vacancies the authors use to explain the initial $T_{\\mathrm{MIT}}$ drop.","marker":"[23]"},{"why":"Provides the prior oxygen-vacancy carrier-localization study and the theoretical prediction of a non-monotonic gap the authors say they access at low doping.","marker":"[25]"},{"why":"Documents electron antidoping in a perovskite nickelate, the competing interpretation the paper argues it goes beyond.","marker":"[24]"},{"why":"Shows the monotonous MIT response to oxygen content in another nickelate, the baseline against which the non-monotonic behavior is contrasted.","marker":"[30]"},{"why":"Supports the attribution of hysteresis suppression and hopping transport to quenched disorder from oxygen vacancies.","marker":"[43]"},{"why":"Gives the site-selective Ni$^{2+}$ Mott insulating picture for LaNiO$_{3-\\delta}$ that the authors distinguish from their Mott–Anderson state.","marker":"[60]"},{"why":"Provides first-principles electron-doping results used to compare vacancy doping with rare-earth-site chemical substitution.","marker":"[45]"}],"fun_headline_variants":["Oxygen vacancies lower then erase a nickelate MIT","In high-entropy nickelates, vacancies bend MIT down, up, off","Janus-faced vacancies: nickelate MIT cedes then vanishes","Vacancy-doped HEO films: MIT dips, rises, then disappears","Non-monotonic vacancies control the nickelate MIT"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing assumption is that the Ni$^{2+}$ spectral weight in Ni L2-edge X-ray absorption is a linear measure of oxygen-vacancy concentration, calibrated against NiO and a reference film assumed to be nearly stoichiometric, with no error bars or independent measurement of $\\delta$.","fun_headline_variants_meta":{"raw":{"variants":["Oxygen vacancies lower then erase a nickelate MIT","In high-entropy nickelates, vacancies bend MIT down, up, off","Janus-faced vacancies: nickelate MIT cedes then vanishes","Vacancy-doped HEO films: MIT dips, rises, then disappears","Non-monotonic vacancies control the nickelate MIT"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000695,"raw_usage":{"total_tokens":3224,"prompt_tokens":1105,"completion_tokens":2119,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":721,"completion_tokens_details":{"reasoning_tokens":2032}},"tokens_in":721,"tokens_out":2119,"duration_ms":22581,"temperature":1.0,"reasoning_tokens":2032,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T19:16:38.240703+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the oxygen content of identically grown films by an independent method, such as ion-beam analysis with elastic recoil detection, and compare the results with the Ni L2-edge linear-combination estimates; if the 10 mTorr film does not sit near $\\delta \\approx 0.2$, or if films grown at fixed total pressure but different oxygen partial pressure do not reproduce the same phase sequence, the claimed vacancy-controlled phase diagram would need revision.","supporting_citations":[{"cited_title":"Middey, J","cited_arxiv_id":null,"evidence_quote":"Defines the negative charge-transfer ground state and bond disproportionation of rare-earth nickelates that the pristine film's MIT is based on."},{"cited_title":"Shamblin, M","cited_arxiv_id":null,"evidence_quote":"Supplies the bipolaron-condensation picture whose disruption by vacancies the authors use to explain the initial $T_{\\mathrm{MIT}}$ drop."},{"cited_title":"Kotiuga, Z","cited_arxiv_id":null,"evidence_quote":"Provides the prior oxygen-vacancy carrier-localization study and the theoretical prediction of a non-monotonic gap the authors say they access at low doping."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Documents electron antidoping in a perovskite nickelate, the competing interpretation the paper argues it goes beyond."},{"cited_title":"Guo and B","cited_arxiv_id":null,"evidence_quote":"Shows the monotonous MIT response to oxygen content in another nickelate, the baseline against which the non-monotonic behavior is contrasted."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supports the attribution of hysteresis suppression and hopping transport to quenched disorder from oxygen vacancies."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the site-selective Ni$^{2+}$ Mott insulating picture for LaNiO$_{3-\\delta}$ that the authors distinguish from their Mott–Anderson state."},{"cited_title":"Iglesias, M","cited_arxiv_id":null,"evidence_quote":"Provides first-principles electron-doping results used to compare vacancy doping with rare-earth-site chemical substitution."}],"review_version":1}