{"id":"c54249ff-61ad-4d31-a09d-64118429c03b","arxiv_id":"2501.08530","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"In sputtered Ni-Al films with 11 to 24 at% Al, the disordered and ordered phases form a coherent, strain-coupled nano-solution with one lattice parameter that decomposes on annealing above 673 K.","lead":"Thin epitaxial nickel-aluminum films made by magnetron sputtering form an intermixed, strained mixture of two crystal phases, called a nano-solution, for aluminum contents between 11% and 24%. The structure is metastable: it survives annealing up to 673 K and hardens slightly, then decomposes into separate phases and softens at higher temperatures.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The central strained-nano-solution claim rests on an ambiguity the authors concede: a single (002) peak plus (001) superlattice spots cannot distinguish two phases strained to one lattice parameter from two phases whose local compositions make their unstrained lattice parameters equal, so the…","rationale":"The paper's complementary evidence (XRD, HRTEM, SAD, DF, EDS, and the annealing response) is a solid characterization dataset, and the existence of nanoscale γ/γ′ intermixing is well supported. The unresolved point is whether the intermixed phases are elastically strained to a common lattice parameter or are simply two phases whose local compositions render their unstrained lattice parameters equal. This matters because the hardness story is tied to coherency strain. The authors explicitly leave this open in Section 3.2. A decisive local composition and lattice-parameter measurement would settle it. The hardness comparison to x=0.24 is also within error, further weakening the \"enhanced hardness\" part of the conclusion. These are addressable with experiment, so a conditional accept remains appropriate.","tokens_in":18208,"tokens_out":4878,"duration_ms":49166,"concrete_test":"Perform atom probe tomography (APT) or high-resolution STEM-EDS on an as-deposited Ni0.86Al0.14 film to measure Al concentrations inside individual γ and γ′ domains; on the same regions, use nanobeam electron diffraction or geometric phase analysis to measure the local (002) d-spacing. Compare the measured d-spacings with the unstrained lattice parameters predicted for those local compositions from Fig. 4. If the measured d-spacing is uniform while the composition-predicted unstrained values differ by more than the measurement uncertainty, the strained coherent nano-solution is confirmed. If the measured d-spacings match the unstrained predictions for each local composition, the equal-lattice-parameter scenario holds and the \"strained\" and hardness interpretation should be revised.","verdict_should_be":"UNCHANGED","load_bearing_attack":"In Section 3.2 the authors state, \"Although our results do not unequivocally rule out the above-explained scenario...\" referring to the possibility that γ and γ′ have identical equilibrium lattice parameters because their local compositions differ. This is the load-bearing weakness. The central conclusion—that the as-deposited films are a coherent strained γ/γ′ nano-solution with enhanced hardness—requires that the observed single (002) XRD peak and matching (001) superlattice reflections reflect two phases with different intrinsic lattice parameters pulled into coherency by elastic strain. But the identical-peak data are equally consistent with the two phases simply having the same unstrained lattice parameter at their actual local compositions. The authors' own Fig. 4 computed lattice-parameter curves show that the phases would differ if both had the global composition, but no measurement of local phase compositions is provided; the \"fine tuning\" argument is not a quantitative exclusion. If the equal-lattice-parameter alternative is correct, the \"strained coherent lattice\" and the associated age-hardening mechanism are not established, and the microstructure is a conventional fine two-phase mixture. The hardness claim is also directly weakened by the overlapping error bars between x=0.14 (5.5±0.3 GPa) and x=0.24 (5.4±0.4 GPa), so \"enhanced hardness in relation to single-phase structures\" is not supported by the data as presented.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports on epitaxial Ni1-xAlx thin films (x up to 0.24) grown on MgO(001) by magnetron sputtering. XRD, TEM/HRTEM/SAD, EDS, and nanoindentation show that for x < 0.11 the films are a random FCC γ solid solution, while for 0.11 ≤ x ≤ 0.24 a single (002) XRD peak and superlattice (001) reflections coexist. The authors interpret this as a coherent, strained γ/γ′ nano-solution with one average lattice parameter, which is metastable upon annealing: at 673 K it persists with a slight hardness increase, whereas at 873 K and 1073 K it decomposes into distinct γ and γ′ domains with softening. DFT and machine-learned (tabGAP) potential calculations are used to compare computed lattice parameters of γ and γ′ phases with the experimental values.","tokens_in":18455,"tokens_out":2781,"duration_ms":28737,"significance":"If the central interpretation holds, the paper demonstrates a new metastable coherent two-phase nanostructure in a canonical intermetallic system, formed by far-from-equilibrium sputter deposition, with implications for non-equilibrium synthesis routes and for strain-based design of mechanical properties. The experimental work is broad and uses complementary, generally well-executed probes (XRD, HRTEM with image simulation, SAD, EDS, nanoindentation), and the computational comparison with DFT and a tabGAP potential is a useful addition. However, the key physical claim—that the observed single lattice parameter arises from two phases with different intrinsic lattice parameters that are strained into coherency—is not uniquely established, and the authors explicitly leave open a simpler alternative. The hardness evidence for the nano-solution's benefit is also weaker than the conclusions state.","major_comments":[{"comment":"The central claim that the films contain a coherent strained γ/γ′ nano-solution with a single average lattice parameter is not uniquely established. The authors state, 'Although our results do not unequivocally rule out the above-explained scenario,' referring to the possibility that γ and γ′ have identical equilibrium lattice parameters because their local compositions differ. In that alternative, the single (002) peak and matching (001) superlattice reflections arise from a conventional fine two-phase mixture at local equilibrium, not from coherency strain. The 'fine tuning' argument against this scenario is qualitative and depends on the computed lattice-parameter curves without local composition measurements. To make the strained-nano-solution claim load-bearing, the authors should provide quantitative evidence, for example atomic-scale composition mapping (STEM-EDS or APT) showing that the phase compositions are close to the global composition, or strain/lattice-parameter mapping (e.g., GPA on HRTEM images) showing continuous variation around the average value.","section":"Section 3.2, Fig. 4"},{"comment":"The conclusion that the nano-solution provides 'enhanced hardness in relation to single-phase structures' is not supported by the data as presented. The hardness at x = 0.14 is 5.5 ± 0.3 GPa and at x = 0.24 (γ′-phase dominated) is 5.4 ± 0.4 GPa, so the two values are statistically indistinguishable. The comparison with x = 0.07 (3.1 ± 0.3 GPa) is clear, but the claim that the nano-solution is harder than the single-phase γ′ structure at x = 0.24 requires either a statistical test or additional data. Similarly, the 'age-hardening' at 673 K from 5.5 ± 0.3 to 6.0 ± 0.3 GPa is within overlapping error bars and should not be stated as a definite increase without a significance assessment.","section":"Section 3.3, Fig. 5(b2)"},{"comment":"The lattice-parameter data extracted from XRD (Fig. 4) are plotted without any uncertainty estimates, which is problematic because the central discrimination between the strained-coherent and equal-lattice-parameter scenarios rests on the comparison between the experimental single lattice parameter and the computed γ/γ′ curves. Without error bars, the reader cannot judge whether the experimental values are actually inconsistent with a scenario in which each phase has its own equilibrium lattice parameter at its local composition. In addition, the authors state that the tabGAP potential was modified by tuning its hyperparameters to match DFT lattice-constant predictions; therefore the MS and DFT curves are not an independent prediction, and the statement that their agreement 'lends confidence to the accuracy of the machine-learned interatomic potential' should be moderated. This issue affects the quantitative basis for the 'fine tuning' argument, though it does not by itself invalidate the experimental observations.","section":"Section 2.2 and Fig. 4"}],"minor_comments":[{"comment":"In the text describing Fig. 1(c), the peak at 2θ = 51.24° is attributed to 'Ni3Al(001)'; this should be Ni3Al(002), as (001) is the superlattice reflection at about 25°.","section":"Section 3.1"},{"comment":"The superlattice reflection used for dark-field imaging is denoted (101̅) in the text but (011̅) in the Fig. 2 caption; please make the notation consistent.","section":"Section 3.1 and Fig. 2 caption"},{"comment":"The phrase 'reduced module Y' should be 'reduced modulus Y'; also, in the same paragraph, 'constat' should be 'constant'.","section":"Section 3.3"},{"comment":"The statement that 'the (002) fringes are intermittently visible' for the Ni0.86Al0.14 sample is vague; a more quantitative description (e.g., fraction of the image area with visible (001) fringes) would strengthen the claim of atomic-scale intermixing.","section":"Section 3.2"},{"comment":"The proposal that 'structure formation may proceed via surface-directed spinodal decomposition' is speculative and not tested by the present data; since the manuscript does not present time-resolved or composition-fluctuation measurements, this sentence should be explicitly labeled as a hypothesis, not a conclusion.","section":"Section 3.2"}],"recommendation":"major_revision","confidential_remarks":"The paper is likely of interest to the community and the experimental dataset is valuable, but the central 'nano-solution' claim needs to be backed by either local composition/strain measurements or a clearly presented statistical argument that rules out the equal-lattice-parameter alternative. The hardness claims also need error-aware treatment. A major revision is appropriate; the work is not beyond repair, but the current manuscript overstates what the data can support."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper is a solid experimental characterization of epitaxial sputtered Ni-Al films across the composition range x=0 to 0.24, and it does something genuinely useful: it provides a systematic map of phase formation and annealing response, showing a coherent, intermixed γ/γ′ state that the authors call a nano-solution. The combination of XRD, HRTEM, SAD, DF imaging, EDS, and nanoindentation is complementary and internally consistent. The annealing behavior—stable up to 673 K with a hardness increase, then decomposition into distinct phases with softening at higher temperature—is a clean and believable demonstration of metastability. The paper is also honest: it explicitly acknowledges that the data do not rule out an equal-lattice-parameter alternative, which is the right thing to do but also the crux of the interpretation problem.\n\nThe soft spots are real but proportionate. The central claim that the two phases are strained to a single average lattice parameter is not uniquely established. A single symmetric (002) peak plus (001) superlattice spots is exactly what you would expect if the γ and γ′ phases simply have the same unstrained lattice parameter at their actual local compositions. The authors try to argue against this by noting the computed lattice-parameter difference at the global composition, but they do not measure the local phase compositions. So the \"strained coherent nano-solution\" is plausible, but the equal-lattice-parameter fine two-phase mixture remains a live alternative. This is a load-bearing ambiguity because the hardening mechanism is tied to coherency strain. Also, hardness at x=0.14 is 5.5±0.3 GPa versus 5.4±0.4 GPa at x=0.24; the claim of \"enhanced hardness relative to single-phase structures\" is not supported by those overlapping values. Lattice parameters are plotted without error bars, which makes the subtle sub-linear trend harder to evaluate. The DFT/MS comparison is useful, but the ML potential was tuned to DFT, so its agreement is not an independent prediction—though that does not affect the experimental observation itself.\n\nNone of these issues break the paper. The phase coexistence and the annealing response are documented with standard, complementary probes, and the authors’ interpretation, while not unique, is reasonable and clearly stated. The ambiguities are addressable: local composition mapping by APT or EELS, or a more careful coherency strain analysis, would settle it. This is a good candidate for peer review and a conditional accept, not a reject. I would cite it for the composition-annealing map, and the discussion of the nano-solution ambiguity could be a useful reading-group topic.","headline":"Solid experimental map of sputtered Ni-Al films, with a real nameable ambiguity in the central 'nano-solution' claim that the authors themselves concede; worth peer review, but the interpretation needs sharpening.","tokens_in":19072,"tokens_out":1199,"would_cite":true,"duration_ms":13708,"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":"Magnetron-sputtered Ni-Al films with 11–24% aluminum form a coherent, strained two-phase \"nano-solution\" with a single lattice parameter, and annealing decomposes it while changing hardness.","keywords":["Ni-Al alloys","magnetron sputtering","nano-solution","L12 phase","coherent strain","metastable phases","thin films","hardness"],"falsifier":"Measure the local aluminum composition of individual $\\gamma$ and $\\gamma'$ domains in an as-deposited $\\mathrm{Ni}_{0.86}\\mathrm{Al}_{0.14}$ film with sub-nanometre probe analysis (for example atom-probe tomography or high-resolution STEM-EDS). If the domains share the global composition yet show one common lattice parameter, the strained nano-solution is confirmed; if they have different compositions chosen so that each domain's unstrained lattice parameter equals the measured value, the alternative equilibrium interpretation wins and the coherency-strain hardening story would need revision.","tokens_in":17988,"feed_emoji":"🧪","tokens_out":10906,"duration_ms":97247,"temperature":0.7,"pith_summary":"This paper reports that magnetron-sputtered nickel-aluminum films with aluminum fractions $x$ between 0.11 and 0.24 do not form a simple random alloy or a coarse phase mixture. Instead, nanometer-scale domains of the disordered $\\gamma$ phase and the ordered $\\mathrm{L1}_2$ $\\gamma'$ phase interlock coherently and share one lattice parameter, a metastable structure the authors call a nano-solution. The claim matters because it shows that far-from-equilibrium vapor deposition can self-organize ordered intermetallic phases into strain-coupled nanostructures with mechanical properties distinct from either single phase. Annealing a $\\mathrm{Ni}_{0.86}\\mathrm{Al}_{0.14}$ film shows this structure persists to 673 K while hardness rises from 5.5 to 6.0 GPa, then decomposes into separate $\\gamma$ and $\\gamma'$ domains at 873 K and above, with hardness falling to about 4 GPa.","feed_headline":"Sputtered Ni-Al films form a strained two-phase nano-solution","feed_subtitle":"At 11–24% Al, the γ and γ′ phases share one strained lattice; hardness peaks at 6 GPa before annealing decomposes it.","key_machinery":"The load-bearing object is the \"nano-solution\": a coherent interlocking of $\\gamma$ (disordered fcc) and $\\gamma'$ (ordered $\\mathrm{L1}_2$) domains that are too small for their interfacial strain fields to decay, so both phases are locked to a single average lattice parameter. The paper identifies this structure through the combination of a single symmetric (002) XRD reflection, (001) superlattice reflections, and HRTEM intensity profiles across a $\\gamma'/\\gamma$ interface that match a simulated two-phase supercell. The single lattice parameter, which expands with Al content and follows the computed $\\gamma'$ curve more closely as $x$ approaches 0.24, is the observable that carries the argument; annealing converts the single peak into two peaks, showing that the shared lattice is a metastable, strain-imposed state rather than an equilibrium one.","core_discovery":"The central claim is that for $0.11 \\le x \\le 0.24$ the films consist of coexisting $\\gamma$ and $\\gamma'$ domains intermixed at the atomic scale, forming a coherent strained lattice with a single lattice parameter that increases with global Al content. The evidence is a symmetric (002) XRD peak with no splitting, a (001) superlattice reflection from the $\\mathrm{L1}_2$ phase appearing only for $x \\ge 0.11$, and HRTEM images with matching simulations showing domains only a few nanometers across. The two phases meet through a transition zone about 1.4 nm wide, roughly eight (002) planes, and the authors argue that strain from coherent interfaces cannot relax inside such small domains, so the lattice adopts an average parameter between the computed equilibrium values of $\\gamma$ and $\\gamma'$. Annealing at 673 K leaves the nano-solution intact and hardens the film from 5.5 to 6.0 GPa, while annealing at 873 and 1073 K splits the (002) peak into distinct $\\gamma$ and $\\gamma'$ reflections, coarsens the domains, and softens the film to about 4 GPa, confirming the structure is metastable.","pith_inferences":["If coherency strain is the operative stabilizer, sputtering other alloy pairs with larger equilibrium lattice mismatches should produce even harder nano-solutions, and the hardness gain should scale with mismatch; this is testable without changing the deposition method.","The suggestion that formation proceeds by surface-directed spinodal decomposition implies a kinetic signature: varying substrate temperature or deposition rate should change domain size in a predictable diffusion-limited way, which would distinguish spinodal decomposition from nucleation and growth.","The fact that the nano-solution at $x=0.14$ matches the hardness of the near-stoichiometric $\\gamma'$ film at $x=0.24$ suggests coherency strain can substitute for raw $\\gamma'$ content, possibly reducing aluminum demand in hard coatings; the paper does not address this economic angle.","Extending the annealing study to $x=0.24$ would show whether the age-hardening window widens or narrows as the $\\gamma'$ fraction dominates, and whether the 673 K hardening step persists in films that are already mostly ordered."],"forward_implications":["Composition alone tunes the lattice parameter and hardness of as-deposited films across the two-phase window, because the nano-solution behaves like a substitutional solid solution but substitutes nanometer-scale phase domains for individual atoms.","A 673 K anneal of $\\mathrm{Ni}_{0.86}\\mathrm{Al}_{0.14}$ films raises hardness to about 6 GPa without dissolving the nano-solution, indicating a practical low-temperature strengthening step for sputtered Ni-Al coatings.","Annealing above 873 K decomposes the nano-solution into distinct $\\gamma$ and $\\gamma'$ domains and softens the film to about 4 GPa, so thermal exposure is a direct control for switching between hardened and softened states.","The measurable lattice parameter of the $\\gamma'$ phase can be systematically shifted by the surrounding $\\gamma$ phase, providing a structural-design route for metastable $\\gamma/\\gamma'$ intermetallic alloys."],"supporting_citations":[{"why":"Supplies the equilibrium Ni-Al phase diagram that defines the γ and γ′ phase fields the non-equilibrium films are compared against.","marker":"[5]"},{"why":"Documents the small coherent γ/γ′ lattice mismatch (<0.5%) that makes a shared coherent lattice plausible.","marker":"[6]"},{"why":"Provides the equilibrium γ′/γ interface transition width (1.23 ± 0.19 nm) used to benchmark the measured 1.4 nm transition zone.","marker":"[7]"},{"why":"Supplies the estimated cooling rate near 10^14 K/s in magnetron sputtering, the far-from-equilibrium premise for nano-solution formation.","marker":"[35]"},{"why":"Provides the algorithm used to simulate HRTEM images of a mixed γ′/γ supercell against which the experimental images are matched.","marker":"[40]"},{"why":"Provides the machine-learned interatomic potential used for molecular-statics lattice-parameter calculations of γ and γ′ phases.","marker":"[50]"},{"why":"Reports a similar 'supranano L12 particles in FCC matrix' structure in another alloy, supporting the nano-solution motif.","marker":"[58]"},{"why":"Establishes γ′ as a strengthening agent in Ni-Al alloys, used to interpret the hardness hierarchy.","marker":"[64]"},{"why":"Supplies the surface-directed spinodal decomposition mechanism invoked for the self-organization of the nano-solution.","marker":"[60-63]"}],"fun_headline_variants":["Sputtered Ni-Al films form a strained two-phase nano-solution","Metastable Ni-Al nano-solution keeps lattice till 673 K","Two-phase Ni-Al films share one strained lattice, hardness 6 GPa","Ni-Al nano-solution hardens to 6 GPa, then decomposes at 873 K","Strained Ni-Al two-phase mix: hardness up then down"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The interpretation depends on assuming that the single symmetric (002) XRD peak reflects two strained phases with different natural lattice constants locked to one average value, rather than two phases whose local compositions coincidentally give them the same equilibrium lattice parameter.","fun_headline_variants_meta":{"raw":{"variants":["Sputtered Ni-Al films form a strained two-phase nano-solution","Metastable Ni-Al nano-solution keeps lattice till 673 K","Two-phase Ni-Al films share one strained lattice, hardness 6 GPa","Ni-Al nano-solution hardens to 6 GPa, then decomposes at 873 K","Strained Ni-Al two-phase mix: hardness up then down"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000532,"raw_usage":{"total_tokens":2635,"prompt_tokens":1091,"completion_tokens":1544,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":707,"completion_tokens_details":{"reasoning_tokens":1443}},"tokens_in":707,"tokens_out":1544,"duration_ms":11582,"temperature":1.0,"reasoning_tokens":1443,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T20:23:52.162188+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the local aluminum composition of individual $\\gamma$ and $\\gamma'$ domains in an as-deposited $\\mathrm{Ni}_{0.86}\\mathrm{Al}_{0.14}$ film with sub-nanometre probe analysis (for example atom-probe tomography or high-resolution STEM-EDS). If the domains share the global composition yet show one common lattice parameter, the strained nano-solution is confirmed; if they have different compositions chosen so that each domain's unstrained lattice parameter equals the measured value, the alternative equilibrium interpretation wins and the coherency-strain hardening story would need revision.","supporting_citations":[{"cited_title":"Vander Voort, S.R","cited_arxiv_id":null,"evidence_quote":"Supplies the equilibrium Ni-Al phase diagram that defines the γ and γ′ phase fields the non-equilibrium films are compared against."},{"cited_title":"Liu, S.-L","cited_arxiv_id":null,"evidence_quote":"Documents the small coherent γ/γ′ lattice mismatch (<0.5%) that makes a shared coherent lattice plausible."},{"cited_title":"Forghani, J.C","cited_arxiv_id":null,"evidence_quote":"Provides the equilibrium γ′/γ interface transition width (1.23 ± 0.19 nm) used to benchmark the measured 1.4 nm transition zone."},{"cited_title":"Musil, J","cited_arxiv_id":null,"evidence_quote":"Supplies the estimated cooling rate near 10^14 K/s in magnetron sputtering, the far-from-equilibrium premise for nano-solution formation."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the machine-learned interatomic potential used for molecular-statics lattice-parameter calculations of γ and γ′ phases."},{"cited_title":"Yan, W.H","cited_arxiv_id":null,"evidence_quote":"Reports a similar 'supranano L12 particles in FCC matrix' structure in another alloy, supporting the nano-solution motif."},{"cited_title":"Lee, An investigation of thermal aging effects on the mechanical properties of a Ni3Al-based alloy by nanoindentation, J","cited_arxiv_id":null,"evidence_quote":"Establishes γ′ as a strengthening agent in Ni-Al alloys, used to interpret the hardness hierarchy."}],"review_version":1}