{"id":"279da21c-002e-4d64-b9a8-29a98f8840e7","arxiv_id":"2506.03262","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Infinite-layer nickelate superconductors show a spin-glass magnetic state with an onset temperature several times Tc, and its strength depends on the substrate while Tc does not.","lead":"Magneto-optical measurements on two superconducting nickelate films show the hallmarks of a spin glass: irreversible magnetization, slow aging, and memory effects, with the magnetic onset temperature far above the superconducting transition. The magnetic signal depends strongly on the substrate while the superconducting temperature barely changes, which the authors interpret as a loose coupling between the glassy magnetism and the superconductivity.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Intrinsic interpretation of the Kerr spin-glass signal rests on a degraded-sample muSR correlation, while the reported wavelength sign reversal mirrors elemental Ni; element-specific magnetization data are needed.","rationale":"The reader's weakest_assumption—that the magnetism is intrinsic rather than dominated by NiOx particles—is exactly the load-bearing concern. The paper's qualitative spin-glass fingerprints (irreversibility, aging, memory) are well presented and the substrate comparison is a useful new result, so the central claim is not obviously false. However, the step from 'a magneto-optical signal with glassy dynamics' to 'an intrinsic nickelate spin-glass coexisting with superconductivity' depends on excluding a known extrinsic source. The paper's own disclosure of a wavelength-dependent sign reversal similar to elemental nickel, combined with the heavy reliance on a prior muSR study on a degraded, same-authored sample, leaves this exclusion insufficiently supported. An element-specific XMCD measurement would directly settle whether the magnetic moments reside on the nickelate Ni sites or on metallic Ni clusters. Because the reader already made acceptance conditional on this kind of verification, no change to the verdict is needed.","tokens_in":12730,"tokens_out":6731,"duration_ms":88216,"concrete_test":"Perform element-specific X-ray magnetic circular dichroism (XMCD) at the Ni L2,3 edges on the same LSNO/STO film used for the Kerr measurements. The XMCD lineshape and L3/L2 branching ratio distinguish Ni0 (metallic Ni, e.g., particles) from Ni2+ in the infinite-layer nickelate. If the XMCD signal exhibits a Ni2+ spectrum with field-cooling history matching the Kerr data, the intrinsic interpretation is confirmed; if it shows a Ni0 signature, the spin-glass signal is extrinsic. As a complement, spatially map the Kerr rotation and compare to scanning SQUID NiOx maps of Ref. [24].","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that the measured magneto-optical signal reflects an intrinsic nickelate spin glass requires that the signal be dominated by the nickelate film rather than by the known NiOx particles at the nickelate/capping-layer interface (Ref. [24] reports ~37 particles/um^2). The paper's only direct support for the intrinsic interpretation is the statement in the Discussion ('Importantly, present magneto-optical measurements agree with the muSR study [23]...') that same-sample muSR proves bulk magnetism. That link is weakened by two facts disclosed in the paper itself: (i) the LSNO sample had degraded (Tc decreased) between the muSR run and the present measurements (Fig. 6), so the magnetic state may have changed; (ii) the muSR study shares authors with the present work, so it is not independent confirmation. Furthermore, the Methods section notes that the 1550-nm Kerr signal reverses sign at 830 nm, 'similar to elemental nickel'—precisely the behavior expected if metallic Ni particles dominate the magneto-optical response. No quantitative estimate of the Ni-particle contribution is provided, and the LSAT substrate effect is not tied to a measured particle density. Thus, if the muSR signal is also influenced by Ni clusters or the sample state changed, the conclusion that the glassy dynamics is intrinsic and coexists with superconductivity in the nickelate is not secured.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports magneto-optical Kerr and Faraday measurements on superconducting infinite-layer nickelate thin films, La0.8Sr0.2NiO2 on SrTiO3 and Nd0.825Sr0.175NiO2 on SrTiO3 and LSAT substrates. The authors observe a remanent Kerr rotation after field cooling that reverses with training-field direction, irreversible susceptibility between zero-field-cooled and in-field-cooled protocols, slow aging at 15 K, a small hysteresis loop, and a memory feature at previously visited temperatures. They interpret these as hallmarks of a spin-glass phase that coexists with superconductivity and persists to temperatures well above Tc, and they report a strong substrate dependence of the signal magnitude and onset temperature. The central interpretive claim is that this magneto-optical signal is intrinsic to the nickelate film rather than dominated by NiOx particles at the nickelate/capping-layer interface.","tokens_in":12811,"tokens_out":4132,"duration_ms":49844,"significance":"If the intrinsic interpretation is correct, this paper provides the first direct magneto-optical evidence that infinite-layer nickelate superconductors host a spin-glass state that coexists with superconductivity and extends to temperatures much higher than Tc, in contrast to the typical cuprate phase diagram where spin-glass order and superconductivity are more exclusive. The paper uses a well-established zero-area-loop Sagnac interferometer with sub-microradian sensitivity and presents several independent glassy fingerprints (aging, memory, irreversibility) in at least one sample. The substrate dependence and the explicit comparison of wavelength-dependent Kerr response are valuable experimental constraints. However, the load-bearing step from the observed signals to an intrinsic bulk nickelate spin glass rests on a correlation with a prior muSR study on a sample that the authors acknowledge has degraded, and the wavelength dependence resembles elemental nickel; these concerns are not fully resolved by the data as presented.","major_comments":[{"comment":"The conclusion that the glassy signal is intrinsic and not due to Ni clusters is supported only by the agreement with the muSR study of Ref. [23], yet the manuscript itself states in the Methods that the LSNO sample has degraded (Tc decreased) since the muSR measurements, and the muSR study shares several co-authors with the present work. Consequently, the muSR correlation does not independently establish that the present Kerr signal arises from the bulk nickelate. A quantitative estimate of the expected contribution from the reported ~37 Ni particles per µm^2 (Ref. [24]), or an element-specific control, is needed to rule out a dominant NiOx contribution.","section":"Discussion, paragraph beginning 'Importantly, present magneto-optical measurements agree with the muSR study...'"},{"comment":"The manuscript notes that the 830-nm Kerr signal reverses sign relative to 1550 nm and explicitly states that this behavior is 'similar to elemental nickel.' If the signal were dominated by metallic Ni particles, exactly this sign reversal would be expected. The paper provides no quantitative estimate of the Ni-particle magneto-optical response, so the wavelength dependence weakens rather than strengthens the intrinsic interpretation. This is a central point because the title and abstract attribute the spin-glass state to the nickelate superconductor, not to interfacial particles.","section":"Methods, paragraph on wavelength dependence and 'Relationship between MOKE and magnetization'"},{"comment":"For NSNO/LSAT the authors state that the magnetic signal is so small that aging could not be demonstrated, and the evidence is limited to irreversible susceptibility and a gradual onset. In the Results for LSNO they explicitly acknowledge that irreversible susceptibility and gradual onset 'could also be explained with domain physics alone.' For NSNO/LSAT no aging or memory data are presented, so the attribution of a spin-glass phase in that sample is not supported by the same dynamical fingerprints, and the claim of a dramatic substrate-dependent spin-glass onset is correspondingly weaker.","section":"Results, paragraph on Nd0.825Sr0.175NiO2 on LSAT (Fig. 4c,d)"},{"comment":"Given the concerns above, the statement that the measurements 'clearly demonstrate the presence of a spin-glass phase' is too strong. The data are consistent with glassy dynamics, but the intrinsic-bulk attribution is not uniquely determined; the manuscript itself discloses the sample-degradation and wavelength-dependence caveats. The conclusions should be tempered to match the level of evidence or be backed by additional controls.","section":"Conclusions, first paragraph"}],"minor_comments":[{"comment":"There is a typographical error: 'The La0.8Sr0.2NiO2 sample was was grown on SrTiO3' repeats 'was'.","section":"Methods, Samples section"},{"comment":"The contrast with cuprates is overstated: the authors themselves cite studies of La2−xSrxCuO4 where spin-glass order coexists with superconductivity and extends into the superconducting dome, so the phrase 'where magnetism and superconductivity remain largely exclusive' is imprecise.","section":"Abstract and Introduction"},{"comment":"The criteria for Tcusp and Tonset are not defined in the main text or table caption; defining how these temperatures are extracted from the Kerr and susceptibility data would improve reproducibility.","section":"Table I"},{"comment":"The notation V_RMS_omega and V_RMS_2omega is introduced without defining the lock-in voltage readings; a brief definition would improve readability.","section":"Methods, Eq. (A.6)"}],"recommendation":"major_revision","confidential_remarks":"The paper addresses an important question and contains high-quality measurements, but the central claim of intrinsic spin-glass behavior in the nickelate is currently supported mainly by a correlation with a muSR study that is neither independent nor performed on the same pristine sample. The wavelength dependence resembling elemental Ni is a red flag that should be confronted head-on with a quantitative model or a control experiment. I would encourage the editor to request a revision that either provides such evidence or substantially tempers the intrinsic claim. The manuscript fits the scope of the journal, but the present version is not yet convincing on its strongest claim."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nHere's my read of arXiv:2506.03262. The work is worth engaging with. The new measurements are real: three spin-glass fingerprints (irreversible susceptibility, aging, memory) in LSNO/STO, and a striking STO/LSAT comparison in NSNO where the Kerr signal drops about fivefold and onset from ~70 K to ~14 K while Tc rises modestly. That substrate knob is new and bears on how disorder and crystallinity shape the magnetic state relative to superconductivity. The ZALSI technique is well-suited, the zero-field protocols avoid the worst Faraday backgrounds, and the authors are careful to measure substrate-only contributions.\n\nThe soft spots are mostly about the strength of the 'intrinsic' conclusion. The paper leans on the muSR agreement on the same LSNO mosaic, but the sample had degraded (lower Tc) between the muSR run and the present measurements, and the muSR study shares authors. That is not independent confirmation. The Methods note that the Kerr sign reverses between 1550 and 830 nm 'similar to elemental nickel' — exactly the behavior you'd expect if Ni clusters contribute. The authors don't provide a quantitative estimate of the Ni-particle contribution. To their credit, they point out that the thin-film signal magnitude is comparable to Ni-free polycrystalline LaNiO2, which is a partial counter, but it doesn't close the issue. I'd also flag the overclaim in the last discussion paragraph: Kerr sensitivity to net moment does not exclude an AFM interaction as a component of the spin glass; a frozen glass with AFM exchange can still produce a net moment. That's a logical slip, not a fatal one. The LSAT sample is classified as a spin glass without the aging test that the paper itself uses to distinguish glass from domain dynamics — that's a weaker classification, and the authors admit they couldn't run the aging measurement.\n\nMissing error bars and single-run memory are real but not fatal; they're typical for this kind of ultra-sensitive measurement. The paper is honest about its caveats, which I respect.\n\nWho is this for? Anyone working on nickelate magnetism and superconductivity, and the magneto-optics community. It deserves a serious referee: the data are novel and the substrate-dependent behavior will need to be explained by any theory of nickelate superconductivity. I'd send it to review with the expectation that the intrinsic versus extrinsic question gets sharpened, not necessarily resolved. Yes, I'd accept it for peer review; I'd probably cite the substrate comparison in my own work.","headline":"Solid new Kerr data on nickelate films with a genuinely new substrate comparison, but the intrinsic spin-glass claim leans on a degraded-sample muSR correlation and the Ni-particle alternative is not fully closed.","tokens_in":13515,"tokens_out":2223,"would_cite":true,"duration_ms":26876,"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":"Spin glass lives in nickelate superconductors far above Tc","keywords":["spin glass","nickelate superconductors","infinite-layer nickelates","polar Kerr effect","magneto-optical","Sagnac interferometer","muon spin relaxation","substrate dependence"],"falsifier":"Map the polar Kerr rotation spatially at 10 K on the NSNO/STO film and overlay it with the known distribution of interfacial NiO$_x$ particles: if the $\\sim 6~\\mu$rad signal appears only where the particles sit, or if a film grown with a capping scheme that eliminates those particles shows no signal, the intrinsic-bulk interpretation fails. Alternatively, measuring muSR and Kerr simultaneously on one sample as $T_c$ degrades would show whether both probes track the same magnetic population.","tokens_in":12373,"feed_emoji":"🧲","tokens_out":13487,"duration_ms":124409,"temperature":0.7,"pith_summary":"The paper reports magneto-optical measurements on two infinite-layer nickelate superconductors, La$_{0.8}$Sr$_{0.2}$NiO$_2$ and Nd$_{0.825}$Sr$_{0.175}$NiO$_2$, showing that a spin-glass magnetic state with onset at 70–80 K coexists with superconductivity that sets in near $T_c \\approx 9$–15 K. This places the nickelate phase diagram in sharp contrast with the cuprates, where spin-glass order generally overlaps superconductivity only in a narrow doped region and the glass temperature is anti-correlated with $T_c$. The authors also find that switching the substrate from SrTiO$_3$ to LSAT shrinks the Kerr signal roughly sixfold and lowers its onset to about 14 K while $T_c$ changes only modestly, indicating that disorder tied to crystalline quality controls the glass while superconductivity remains robust. They conclude that the glass is intrinsic to the nickelate layers, not dominated by interfacial NiO$_x$ particles, and that spin-glass order and superconductivity are not directly linked.","feed_headline":"Spin glass lives in nickelate superconductors far above Tc","feed_subtitle":"Glass magnetism onsets near 80 K, superconductivity near 12 K, and the substrate tunes the gap.","key_machinery":"The central instrument is the zero-area-loop Sagnac interferometer (ZALSI), a fiber-optic device that measures the polar Kerr rotation $\\theta_K$ with a noise floor near 100 nrad/$\\sqrt{\\rm Hz}$. Because $\\theta_K \\propto \\sigma_{xy}(\\omega) \\propto M_z$, the Kerr angle is a direct probe of out-of-plane magnetization, and since a nonzero Kerr signal requires time-reversal symmetry breaking, it serves as a magnetic signature. The argument rests on three dynamical hallmarks of spin glasses observed in the Kerr signal—irreversible susceptibility under field-cooled and zero-field-cooled protocols, slow aging relaxation after a field step, and a memory effect after partial thermal cycles—plus the substrate comparison (STO versus LSAT) that tunes the disorder strength.","core_discovery":"Using a zero-area-loop Sagnac interferometer (ZALSI), a fiber-optic polarimeter with sub-microradian sensitivity, the authors observe in La$_{0.8}$Sr$_{0.2}$NiO$_2$/STO a spontaneous polar Kerr rotation appearing below $T_{\\rm onset} \\approx 80$ K, and in Nd$_{0.825}$Sr$_{0.175}$NiO$_2$/STO below $\\approx 70$ K. The signal shows the classic spin-glass signatures: irreversible magnetic susceptibility between zero-field-cooled and field-cooled protocols, slow aging of the magnetization on minute timescales, and a memory effect when the system is warmed through previously visited temperatures. In Nd$_{0.825}$Sr$_{0.175}$NiO$_2$/LSAT the signal is much weaker, about 1 $\\mu$rad, and onsets near 14 K, while $T_c$ rises from 12.5 to 15.3 K. The authors take these observations, together with agreement with an earlier muon spin-relaxation study on the same La-based sample, to demonstrate a bulk spin-glass phase coexisting with superconductivity up to temperatures far above $T_c$.","pith_inferences":["If the glass arises from remnant apical-oxygen disorder, then chemical routes that reduce such disorder should suppress $T_{\\rm onset}$ and could raise $T_c$ further; this is a testable prediction of the disorder-based picture.","The wavelength-dependent sign reversal of the Kerr signal between 0.8 and 1.5 eV, reminiscent of elemental nickel, suggests the magnetism lives on Ni orbitals; resonant Kerr spectroscopy could identify which optical transitions generate $\\sigma_{xy}$.","The memory effect implies a hierarchical free-energy landscape; multi-stop temperature protocols could map its timescales and distinguish droplet-like from hierarchical glass dynamics.","The substrate dependence implies that disorder, not just doping, is a control parameter for the magnetism; comparing films with identical composition but different Ruddlesden-Popper fault densities would isolate the role of each disorder source."],"forward_implications":["Superconductivity in infinite-layer nickelates emerges from a normal state that is already magnetically glassy, with $T_g$ well above $T_c$.","The glassy state is not a byproduct of the superconducting condensate, since it persists to temperatures several times $T_c$.","Crystallinity, not doping alone, controls the strength and onset of the glass: better-ordered films on LSAT show a roughly sixfold weaker magnetic signal with only about a 20% change in $T_c$.","If the muSR correspondence holds, the glass is a bulk intrinsic property of the nickelate layers rather than a cluster-driven surface effect.","The weak coupling between glass and $T_c$ suggests magnetic disorder is not the primary pairing mechanism, possibly by analogy to the cuprates."],"supporting_citations":[{"why":"Polycrystalline LaNiO2 study that first reported the spin-glass hallmarks (irreversible susceptibility, aging, memory) with T_cusp near 13 K and T_onset near 85 K, giving the paper its comparison baseline.","marker":"[10]"},{"why":"A second polycrystalline nickelate study confirming glassy dynamics in the parent compound, strengthening the claim that the glass is intrinsic to the nickelate layers.","marker":"[21]"},{"why":"The muSR study of R1-xSrxNiO2 that reported intrinsic magnetism coexisting with superconductivity; the paper uses agreement with this bulk probe to argue the Kerr signal is intrinsic.","marker":"[23]"},{"why":"SQUID microscopy identifying about 37 NiOx particles per square micrometer in similar films, the competing extrinsic explanation the paper argues against.","marker":"[24]"},{"why":"Original description of the zero-area-loop Sagnac interferometer, the measurement apparatus the entire dataset rests on.","marker":"[27]"},{"why":"Shows films on LSAT have higher crystallinity and fewer stacking faults than on STO, the basis for attributing the reduced glass signal to improved crystalline quality.","marker":"[34]"},{"why":"Establishes that nickelate films on LSAT have higher Tc than on STO, used to quantify the weak response of superconductivity to the substrate change.","marker":"[36]"}],"fun_headline_variants":["Spin glass at 80 K in nickelate superconductors","Nickelate superconductors host spin glass far above Tc","Nickelates break cuprate rule: spin glass above Tc","Nickelate superconductors flip into spin glass at 80 K"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The conclusion that the glass is intrinsic and bulk rests on the assumption that the Kerr signal comes from the nickelate film itself and that the earlier muSR measurement, taken on the same sample before it degraded, probes the same magnetic population that the Kerr effect sees.","fun_headline_variants_meta":{"raw":{"variants":["Spin glass at 80 K in nickelate superconductors","Nickelate superconductors host spin glass far above Tc","Nickelates break cuprate rule: spin glass above Tc","Nickelate superconductors flip into spin glass at 80 K"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00162,"raw_usage":{"total_tokens":6451,"prompt_tokens":953,"completion_tokens":5498,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":569,"completion_tokens_details":{"reasoning_tokens":5427}},"tokens_in":569,"tokens_out":5498,"duration_ms":39216,"temperature":1.0,"reasoning_tokens":5427,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T11:08:56.833660+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Map the polar Kerr rotation spatially at 10 K on the NSNO/STO film and overlay it with the known distribution of interfacial NiO$_x$ particles: if the $\\sim 6~\\mu$rad signal appears only where the particles sit, or if a film grown with a capping scheme that eliminates those particles shows no signal, the intrinsic-bulk interpretation fails. Alternatively, measuring muSR and Kerr simultaneously on one sample as $T_c$ degrades would show whether both probes track the same magnetic population.","supporting_citations":[{"cited_title":"von Rohr, and A","cited_arxiv_id":null,"evidence_quote":"Polycrystalline LaNiO2 study that first reported the spin-glass hallmarks (irreversible susceptibility, aging, memory) with T_cusp near 13 K and T_onset near 85 K, giving the paper its comparison baseline."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"A second polycrystalline nickelate study confirming glassy dynamics in the parent compound, strengthening the claim that the glass is intrinsic to the nickelate layers."},{"cited_title":"Fowlie, M","cited_arxiv_id":null,"evidence_quote":"The muSR study of R1-xSrxNiO2 that reported intrinsic magnetism coexisting with superconductivity; the paper uses agreement with this bulk probe to argue the Kerr signal is intrinsic."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"SQUID microscopy identifying about 37 NiOx particles per square micrometer in similar films, the competing extrinsic explanation the paper argues against."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Original description of the zero-area-loop Sagnac interferometer, the measurement apparatus the entire dataset rests on."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes that nickelate films on LSAT have higher Tc than on STO, used to quantify the weak response of superconductivity to the substrate change."}],"review_version":1}