{"id":"2911d8c7-4ba3-4581-8b90-41b339b41f3a","arxiv_id":"2506.20264","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"NNMO films on STO show three coexisting octahedral rotation domains and layer-dependent out-of-plane lattice expansion correlated with manganese orbital polarization, unlike single-domain films on NGO.","lead":"This thin-film study compares the double perovskite Nd2NiMnO6 grown on two different substrates and finds that films on SrTiO3 develop three distinct octahedral rotation patterns, while films on NdGaO3 stay in a single pattern. The authors link these structural changes to the orbital shape of manganese, which matters for magnetic behavior in spintronic devices.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Positive Mn XLD assigned to interfacial Mn3+ lacks quantitative TEY depth-weighting analysis; a forward model is needed to support the central orbital-symmetry claim.","rationale":"The paper is a careful experimental study with a plausible central narrative, and several pieces of independent support exist: the COBRA structural profiles are consistent with prior STEM/EELS work (Ref. [33]), the half-order diffraction analysis is systematic, and the Ni XLD behavior follows the expected strain trend. The load-bearing step, however, is the assignment of the positive Mn XLD in 10 and 20 uc NNMO/STO films to interfacial Mn3+, while the negative XLD in the 5 uc film is assigned to surface Mn2+. This assignment is made by exclusion ('bulk Mn4+ would lack any XLD contribution') rather than by quantitative decomposition. The TEY detection mode weights the surface much more strongly than the interface, especially for the thicker films where the interface is several nanometers below the surface. The observed sign reversal between 5 and 10/20 uc is not obviously consistent with a simple depth-weighted sum, since the interfacial region remains within a few nanometers of the surface in all films. This is precisely the reader's weakest assumption, and we agree it is the most vulnerable point in the argument. A forward model using the known layer-resolved Mn valence profile from Ref. [33] and a realistic TEY escape depth would settle whether the sign sequence can be explained by interface Mn3+ dominance; if it cannot, the orbital-symmetry conclusion and the framework linking structure to magnetism would need to be revised. Because this concern is specific, quantitative, and addressable with existing data, the reader's CONDITIONAL verdict is appropriate: the paper should not be accepted without this analysis, but the concern does not invalidate the structural and rotational-domain findings. We therefore recommend UNCHANGED.","tokens_in":16453,"tokens_out":7330,"duration_ms":80360,"concrete_test":"Using the layer-resolved Mn valence profile from the companion PRL (Ref. [33]) or EELS/STEM, compute the expected Mn XLD for the 5, 10, and 20 uc NNMO/STO films as sum_i f_i(z) XLD_i exp(-z/lambda), with f_i from the valence profile and XLD_i representing the per-layer dichroism of surface Mn2+ (negative), interfacial Mn3+ (positive), and bulk Mn4+ (near zero). Vary lambda in the 2-5 nm range and compare the predicted sign and relative magnitude with the measured XLD at each thickness. If no reasonable lambda reproduces the observed sequence (negative at 5 uc; positive at 10 and 20 uc), the assignment of the positive XLD to interfacial Mn3+ is quantitatively unsupported. An even more direct check would be to remeasure the same films in fluorescence-yield mode, which suppresses the surface contribution.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that the positive Mn XLD in 10 and 20 uc NNMO/STO films is dominated by interfacial Mn3+ from polarity compensation rests on an unquantified decomposition of TEY-XLD contributions. TEY is not bulk-averaged: the signal from a layer at depth z is weighted by exp(-z/lambda) with lambda typically 2-5 nm for Mn L edges. For a 20 uc film (~8 nm), the surface Mn2+ layer sits at the top and is strongly weighted, while the interfacial Mn3+ region sits 3-4 uc below and is attenuated. The authors state that 'bulk Mn4+ would lack any XLD contribution' and 'affirm this positive XLD contribution to the Mn3+ species,' but provide no quantitative estimate of the relative weights or per-ion XLD magnitudes. The observed sign reversal (negative for 5 uc, positive for 10/20 uc) is asserted to arise from surface vs. interface dominance, yet a naive depth-weighted sum for a 5 uc film (interfacial Mn3+ within ~2 nm of the surface) would still favor the interface if per-ion XLDs were comparable. Without a quantitative model, the sign of the XLD cannot uniquely identify the interfacial Mn3+ contribution, and the paper's conclusion that structural alterations at the interface control Mn orbital symmetry is not secure.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper investigates ultrathin Nd2NiMnO6 (NNMO) films grown on SrTiO3 (STO) and NdGaO3 (NGO) substrates using synchrotron X-ray diffraction, coherent Bragg rod analysis (COBRA), and X-ray linear dichroism (XLD). The authors report that films on STO exhibit three octahedral rotational domains, whereas films on NGO show a single domain. COBRA analysis reveals an enhanced out-of-plane lattice parameter near the interface for STO films and near the surface for both substrates, attributed to Mn3+ formation at the polar interface and Mn2+ formation at the surface. XLD measurements show a thickness-dependent sign of Mn orbital polarization for STO films, which the authors interpret as a competition between interfacial and surface effects. The central claim is that these structural alterations control the Mn orbital symmetry.","tokens_in":16707,"tokens_out":8463,"duration_ms":83430,"significance":"The study provides a valuable comparative dataset on a ferromagnetic insulating double perovskite in ultrathin form, with a clear experimental design that isolates polar and structural mismatch effects by comparing STO and NGO substrates. The diffraction data appear carefully measured, and the observation of three octahedral rotational domains in NNMO/STO is an interesting structural result. The COBRA layer-resolved c_pc profiles, if properly quantified, would be a useful contribution. However, the strength of the paper's central conclusion linking structural alterations to orbital symmetry is limited by the qualitative XLD decomposition and the absence of error bars on the COBRA-derived lattice parameters.","major_comments":[{"comment":"The assignment of the positive Mn XLD in 10 and 20 uc NNMO/STO films to interfacial Mn3+ is not supported by a quantitative analysis of total-electron-yield (TEY) depth weighting. With the effective TEY probing depth of 9-10 nm stated in the Methods, the surface Mn2+ layer contributes exponentially more than the interfacial Mn3+ region; for a 20 uc film (~8 nm), the interface is attenuated by a factor of roughly exp(-(6-7)/lambda) relative to the surface, where lambda is the TEY escape depth (typically 2-5 nm). The authors state that 'bulk Mn4+ states would lack any XLD contribution' and 'affirm this positive XLD contribution to the Mn3+ species,' but no per-ion XLD magnitudes or depth profiles are provided. Consequently, the observed sign reversal (negative at 5 uc, positive at 10/20 uc) is not uniquely explained by the proposed competition between surface and interface effects. A quantitative forward model incorporating TEY depth weighting, oxidation-state depth profiles, and per-ion XLD spectra is required to support the claim that interfacial structural alterations control the Mn orbital symmetry.","section":"Orbital symmetry from X-ray linear dichroism"},{"comment":"The central quantitative claim of a significant modification of the out-of-plane lattice parameter c_pc within a few unit cells at the interface and surface is presented without error bars or confidence intervals. The manuscript only states that the refinement converged with R << 1 (Eq. 2), but does not report the R-factor values or any uncertainty analysis (e.g., via Monte Carlo or parameter covariance). Given that the c_pc profile in Fig. 4(c) is derived from inter-NdO distances in the reconstructed electron density, the magnitude and significance of the observed enhancements (especially the 3-4 uc interfacial region and the surface region) must be quantified before the structural-orbital coupling is asserted.","section":"Layer-resolved out of plane lattice constant from 1D COBRA"}],"minor_comments":[{"comment":"The phrase 'The interface exhibits immeasurable resistance in both cases' should be clarified to 'no measurable conductivity' or 'insulating interface,' as the intended meaning is that no electrical conduction was detected.","section":"Abstract and Introduction"},{"comment":"The correlation between the c_pc enhancement and the Mn oxidation states relies on ref [33]; the authors should briefly summarize the relevant XAS/EELS results from that prior work to make the argument more self-contained.","section":"Layer-resolved out of plane lattice constant from 1D COBRA"},{"comment":"The normalization procedure for the XAS spectra I_c and I_ab before computing the XLD difference is not described; please specify how the spectra were normalized (e.g., to the edge jump or to the incident beam intensity).","section":"Orbital symmetry from X-ray linear dichroism"},{"comment":"In Fig. 3(j) and the accompanying text, the statement that the two ORPs are 'nearly equally populated' is based on visual comparison of peak intensities; a quantitative intensity analysis with error bars would be more convincing.","section":"Determination of global octahedral rotational pattern (ORP) using half-order Bragg rods"},{"comment":"The numerical values of c_pc for the interface, bulk-like, and surface layers should be stated explicitly in the text or in a table, in addition to the plot in Fig. 4(c), to support the quantitative claims.","section":"Layer-resolved out of plane lattice constant from 1D COBRA"}],"recommendation":"major_revision","confidential_remarks":"The manuscript leans heavily on the authors' own prior work (ref [33]) for the Mn oxidation-state depth profile and STEM data. This is not circular, but the editors may wish to consider whether the present paper is sufficiently self-contained; the main new contribution appears to be the COBRA analysis and the XLD data. The XLD interpretation requires a quantitative model to be convincing. The paper fits the journal scope."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper is worth reading for the diffraction work. The half-order analysis showing three coexisting octahedral rotational domains in NNMO/STO is careful and convincing, and the NGO comparison is a good control that isolates the effects of polar and symmetry mismatch. The COBRA-derived layer-resolved c_pc profile, with expansion near the interface and the surface, is also plausible and gains independent support from the group's earlier STEM work [33]. That part is solid.\n\nThe soft spot is the XLD interpretation, and it is the load-bearing part of the orbital story. The claim that the positive Mn XLD in 10 and 20 uc films comes from interfacial Mn3+ requires a quantitative treatment of TEY depth weighting. TEY is not a bulk probe; a surface Mn2+ layer at the top of the film is strongly weighted, and the interface region sits a few nm deeper. Without a forward model that accounts for the exponential attenuation and the per-ion XLD magnitudes, the sign reversal between 5 uc (negative) and 10/20 uc (positive) cannot uniquely identify interfacial Mn3+ as the dominant contribution. The authors' assertion that surface Mn2+ gives negative XLD and bulk Mn4+ gives none is reasonable, but the decomposition is qualitative. The paper acknowledges the surface contribution but does not quantify its weight in the 10/20 uc films.\n\nAlso, the c_pc profile in Fig. 4(c) is presented without error bars. That is a quantitative claim, and the absence of uncertainties makes it hard to judge significance, though the STEM corroboration mitigates the concern.\n\nOn the whole, I think the structural findings will hold up. The orbital-symmetry conclusion is plausible but not fully demonstrated. The paper does not provide the custom COBRA code or raw data, which would help, but that alone would not block publication.\n\nWho is this for? Researchers working on double perovskite films, polar oxide interfaces, and strain/orbital coupling in manganites. They will get useful structural information and a clear open question about the XLD interpretation. It deserves a serious referee, and I would send it to peer review. The referee should request a quantitative XLD model with depth weighting and error estimates for the layer-resolved lattice parameters.","headline":"Solid structural work on rotational domains and lattice profiles, but the central orbital-symmetry claim needs a quantitative TEY depth-weighted analysis before it is secure.","tokens_in":17220,"tokens_out":2396,"would_cite":true,"duration_ms":29983,"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":"Polarity compensation at a double-perovskite interface creates Mn3+ layers that stretch the lattice and reorient Mn orbitals.","keywords":["double perovskite","Nd2NiMnO6","interfacial reconstruction","polarity compensation","orbital polarization","X-ray linear dichroism","octahedral rotation domains","ultrathin films"],"falsifier":"Grow the same 10 uc NNMO film on STO with a thin STO or other nonpolar capping layer to eliminate the surface Mn2+ contribution, and measure Mn L-edge XLD with depth-sensitive detection (partial or total electron yield at different kinetic energies). If the positive XLD remains after capping, interfacial Mn3+ dominates the orbital signal; if it flips negative or vanishes, the uncapped-film assignment was wrong. Alternatively, an XLD measurement on a 20 uc film with the top layers etched away would leave the interfacial contribution isolated.","tokens_in":16307,"feed_emoji":"🧲","tokens_out":6685,"duration_ms":70316,"temperature":0.7,"pith_summary":"This paper argues that the functional behavior of an ultrathin ferromagnetic insulator film is set not by its average strain but by what happens at its two boundaries. For Nd2NiMnO6 films on SrTiO3, the polar mismatch between alternating charged layers forces Mn ions near the interface into a Mn3+ state, which expands the out-of-plane lattice constant in the first few unit cells; at the free surface, missing apical oxygen produces Mn2+ ions and a second expansion near the surface. Comparing with films on NdGaO3, where neither mismatch exists, the paper isolates these two effects and shows that the sign of the Mn orbital polarization measured by X-ray linear dichroism reverses with thickness as surface and interface contributions trade dominance. A sympathetic reader would care because the ferromagnetism of this double perovskite is believed to be governed by orbital-dependent superexchange, so the paper offers a structural mechanism for the thickness-dependent magnetism seen in ultrathin films.","feed_headline":"Polar mismatch, not strain, controls Mn orbitals in thin films","feed_subtitle":"Layer-resolved X-ray measurements trace the sign change to Mn3+ at the interface and Mn2+ at the surface.","key_machinery":"The argument rests on three experimental tools working together. 1D coherent Bragg rod analysis (COBRA) converts the specular crystal truncation rod into a layer-resolved electron density profile, from which the out-of-plane lattice parameter of each atomic plane is extracted; this is what reveals the interfacial and surface expansions. Half-order Bragg rod diffraction maps the octahedral rotation pattern (Glazer tilt system) of the whole film, distinguishing the three rotational domains on STO from the single domain on NGO. X-ray linear dichroism (XLD), defined as the difference between in-plane and out-of-plane polarized X-ray absorption at the Mn and Ni L edges, reports the orbital occupancy: positive XLD means preferential occupation of the $d_{x^2-y^2}$ orbital, negative XLD means $d_{3z^2-r^2}$, and the comparison between the two substrates separates strain-driven from polarity-driven orbital effects.","core_discovery":"The central claim is that the layer-resolved out-of-plane lattice parameter $c_{\\mathrm{pc}}$ of Nd2NiMnO6 (NNMO) films carries the signature of electronic reconstruction at both the film/substrate interface and the free surface. Using 1D coherent Bragg rod analysis of specular synchrotron X-ray diffraction, the authors find that $c_{\\mathrm{pc}}$ is enlarged for the first 3–4 unit cells at the NNMO/SrTiO3 interface and again in the last few unit cells near the surface, while the middle of the film is relaxed. They attribute the interfacial expansion to Mn3+ ions created by polarity compensation—Mn3+ has a larger ionic radius than the Mn4+ it replaces—and the surface expansion to Mn2+ formed by symmetry breaking at the surface. The structural investigation also shows that NNMO/STO films contain three coexisting octahedral rotational domains ($a^-a^-c^+$, $a^-a^+c^-$, $a^+a^-c^-$), whereas NNMO/NdGaO3 films are single-domain $a^-b^+c^-$. The X-ray linear dichroism results then connect this structure to orbital physics: a negative XLD at the Mn L2 edge appears for films on NGO and for the 5 uc film on STO, where surface Mn2+ states dominate, while a positive XLD appears for 10 and 20 uc films on STO, which the authors assign to interfacial Mn3+ under tensile strain. The paper concludes that thickness-dependent Mn orbital polarization in these films is controlled by the competition among polar catastrophe, epitaxial strain, and surface symmetry breaking.","pith_inferences":["The TEY probing depth (~9–10 nm) exceeds the film thickness for all samples, so the measured XLD is a whole-film average; the paper's attribution of the positive XLD in 10 and 20 uc films specifically to interfacial Mn3+ would become directly testable with depth-resolved XLD, e.g., by varying the electron yield mode or by capping one film with STO to remove the surface Mn2+ contribution.","The same polarity-compensation argument predicts that other polar double perovskites grown on nonpolar cubic substrates will show a comparable interfacial Mn3+/lattice-expansion signature; the layer-resolved COBRA approach is transferable to any epitaxial oxide and could be used to search for such reconstructions.","If the $a^-a^-c^+$ domain is the one that carries the ferroelectric instability, then controlling the relative populations of the three rotational domains—for instance, by choosing a substrate with different lattice mismatch or by growing on vicinal surfaces—would be a way to tune the multiferroic response of NNMO films.","The authors' distinction between 'interfacial' and 'surface' reconstructions is drawn at the structural level; magnetometry on the same thickness series, ideally with a surface-sensitive probe, would test whether the Mn2+ surface layer is magnetically inert or actively participates in the exchange network."],"forward_implications":["Thickness-dependent ferromagnetism in NNMO/STO films can be understood as a consequence of the orbital reconstruction: the interfacial Mn3+ layers with $d_{x^2-y^2}$ occupation modify the Ni–Mn superexchange that governs the magnetic order.","In the very thin limit (5 uc), the surface effect wins, so any device that relies on the ferromagnetic interfacial state must either cap the film or engineer the surface termination to suppress Mn2+ formation.","Because the $a^-a^-c^+$ rotation domain can support hybrid improper ferroelectricity, selectively stabilizing this domain by strain engineering could turn the ferromagnetic insulator into a multiferroic.","The same layer-resolved structural analysis can be applied to other A2BB'O6 double perovskites on nonpolar substrates to predict where polarity compensation will localize charge and alter magnetism.","The absence of these reconstructions in NNMO/NGO films marks NGO as the cleaner substrate for studying intrinsic double-perovskite magnetism in ultrathin form."],"supporting_citations":[{"why":"Provides the prior site-selective polar compensation data: layer-resolved Mn oxidation states and STEM showing the same c_pc trend that the present XLD interpretation relies on.","marker":"[33]"},{"why":"Establishes that ferromagnetism in La2NiMnO6-type double perovskites is mediated by orbital-dependent Ni–Mn superexchange, the reason orbital polarization matters.","marker":"[28]"},{"why":"Describes the high-resolution crystal truncation rod scattering method used for the COBRA analysis, without which the layer-resolved lattice parameter cannot be extracted.","marker":"[35]"},{"why":"Supplies the 1D COBRA phase-retrieval algorithm used to simulate the (00L) rod and reconstruct electron density.","marker":"[42]"},{"why":"Defines how half-order Bragg peaks map to Glazer octahedral rotation patterns and connects the a-a-c+ pattern to hybrid improper ferroelectricity; the paper uses these assignments to identify the three rotational domains.","marker":"[51]"},{"why":"Demonstrates surface symmetry-breaking and strain effects on orbital occupancy in perovskite films, the basis for interpreting the negative XLD from surface Mn2+ states.","marker":"[71]"},{"why":"Supplies the Shannon ionic radii for Mn2+, Mn3+, and Mn4+ used to convert the observed valence changes into expected lattice-parameter changes.","marker":"[68]"},{"why":"Reports thickness-dependent ferromagnetism in NNMO thin films, the experimental phenomenon the paper's orbital-reconstruction framework is invoked to explain.","marker":"[78]"},{"why":"Gives the bulk crystal structure and Tc of Nd2NiMnO6 used as the reference for strain and lattice parameter calculations.","marker":"[30]"},{"why":"Introduces X-ray linear dichroism as a probe of orbital character at oxide interfaces, the method behind the orbital-polarization measurements.","marker":"[36]"}],"fun_headline_variants":["Polar mismatch reshapes Mn orbitals in oxide films","Interface and surface control Mn orbital symmetry","Layer-resolved lattice changes dictate Mn orbital tilt","Reconstruction at both sides tunes Mn orbitals"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The positive XLD signature in the 10 and 20 uc STO films is attributed to interfacial Mn3+ even though the measurement averages over the entire film depth, and the paper does not quantitatively separate the interfacial Mn3+, surface Mn2+, and bulk-like Mn4+ contributions.","fun_headline_variants_meta":{"raw":{"variants":["Polar mismatch reshapes Mn orbitals in oxide films","Interface and surface control Mn orbital symmetry","Layer-resolved lattice changes dictate Mn orbital tilt","Reconstruction at both sides tunes Mn orbitals"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000375,"raw_usage":{"total_tokens":2134,"prompt_tokens":1210,"completion_tokens":924,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":826,"completion_tokens_details":{"reasoning_tokens":879}},"tokens_in":826,"tokens_out":924,"duration_ms":8938,"temperature":1.0,"reasoning_tokens":879,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T22:52:05.951614+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Grow the same 10 uc NNMO film on STO with a thin STO or other nonpolar capping layer to eliminate the surface Mn2+ contribution, and measure Mn L-edge XLD with depth-sensitive detection (partial or total electron yield at different kinetic energies). If the positive XLD remains after capping, interfacial Mn3+ dominates the orbital signal; if it flips negative or vanishes, the uncapped-film assignment was wrong. Alternatively, an XLD measurement on a 20 uc film with the top layers etched away would leave the interfacial contribution isolated.","supporting_citations":[{"cited_title":"Bhattacharya, A","cited_arxiv_id":null,"evidence_quote":"Provides the prior site-selective polar compensation data: layer-resolved Mn oxidation states and STEM showing the same c_pc trend that the present XLD interpretation relies on."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes that ferromagnetism in La2NiMnO6-type double perovskites is mediated by orbital-dependent Ni–Mn superexchange, the reason orbital polarization matters."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Describes the high-resolution crystal truncation rod scattering method used for the COBRA analysis, without which the layer-resolved lattice parameter cannot be extracted."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the 1D COBRA phase-retrieval algorithm used to simulate the (00L) rod and reconstruct electron density."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Defines how half-order Bragg peaks map to Glazer octahedral rotation patterns and connects the a-a-c+ pattern to hybrid improper ferroelectricity; the paper uses these assignments to identify the three rotational domains."},{"cited_title":"Pesquera, G","cited_arxiv_id":null,"evidence_quote":"Demonstrates surface symmetry-breaking and strain effects on orbital occupancy in perovskite films, the basis for interpreting the negative XLD from surface Mn2+ states."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the Shannon ionic radii for Mn2+, Mn3+, and Mn4+ used to convert the observed valence changes into expected lattice-parameter changes."},{"cited_title":"Spring, G","cited_arxiv_id":null,"evidence_quote":"Reports thickness-dependent ferromagnetism in NNMO thin films, the experimental phenomenon the paper's orbital-reconstruction framework is invoked to explain."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the bulk crystal structure and Tc of Nd2NiMnO6 used as the reference for strain and lattice parameter calculations."},{"cited_title":"Chakhalian, J","cited_arxiv_id":null,"evidence_quote":"Introduces X-ray linear dichroism as a probe of orbital character at oxide interfaces, the method behind the orbital-polarization measurements."}],"review_version":1}