{"id":"c36b7eac-6999-4fa6-9c73-c49609c2350d","arxiv_id":"2608.09909","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"PtNi/Zr films on sapphire grow larger faceted crystallites during annealing, while on langasite they roughen more at 600 C before partially smoothing.","lead":"This paper anneals 100-nanometer films of a platinum-nickel alloy on two different crystal surfaces and tracks how their surfaces change up to 800 C. The substrate choice changes how fast the film coarsens and how large the final crystallites become, which matters for high-temperature sensors.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The substrate-selection mechanism is not yet established: without Zr interfacial chemical data or replicated film pairs, the observed sapphire/langasite difference could be driven by adhesion-layer reactions or initial-state differences rather than intrinsic PtNi/substrate energetics.","rationale":"The reader's CONDITIONAL verdict is reasonable. The descriptive statistics are internally consistent and the AFM segmentation is documented, but the interpretation of substrate dependence as energetic/mobility selection rests on an unmeasured buried interface. I agree with the reader's identified weakest assumption. Rather than invoking sample-number or segmentation concerns as primary, the Zr chemical-state gap is the single check that would most directly decide whether the central mechanism claim holds. If future Zr-interface data are identical across substrates, the claim stands; if not, the paper should be revised to attribute differences to adhesion-layer chemistry. This does not require changing the verdict from CONDITIONAL, since the concern was already the basis for that verdict.","tokens_in":11510,"tokens_out":5943,"duration_ms":59998,"concrete_test":"Take the existing 600 C and 800 C samples (or a fresh identically prepared pair) and acquire cross-sectional TEM with EELS/EDX line scans plus Zr 3d XPS depth profiles at the PtNi/Zr and Zr/substrate interfaces on both sapphire and langasite. Quantify Zr oxidation state, Zr-oxide thickness, and Pt/Ni interdiffusion. If the Zr chemical state and interfacial profiles are statistically indistinguishable between substrates, the Zr confound is ruled out and the substrate-selection claim is supported; if they differ, the observed morphology is at least partly Zr-mediated and the conclusion should be reworded accordingly.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is that sapphire promotes larger late-stage crystallites via substrate-dependent morphological selection. Because the full stack includes a 10-nm Zr adhesion layer, the substrate dependence must be shown to originate at the PtNi/substrate interface rather than in the Zr/oxide chemistry. The paper itself lists \"interfacial reactions involving the Zr layer and the oxide substrate\" as a possible cause (Section IV), but provides no Zr 3d XPS, no post-anneal interface spectroscopy, and no cross-sectional analysis. Langasite is documented to react under high-temperature vacuum annealing (Ref. 14), and Zr is known to oxidize and interdiffuse, so a substrate-dependent Zr reaction product could dominate the 400-600 C transition. The as-deposited films already differ strongly between substrates (mean area 784 vs 332 nm^2; mean height 3.02 vs 5.52 nm), and only one pair of films is annealed; measured feature-level standard errors do not capture sample-to-sample variation. Consequently the data support a substrate-dependent outcome but not the causal attribution to alloy mobility, interfacial energetics, and substrate selection; the Zr/interface confound remains unresolved.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports an AFM-based study of the morphological evolution of 100-nm Pt0.9Ni0.1 films with a 10-nm Zr adhesion layer on sapphire and langasite substrates during sequential vacuum annealing from room temperature to 800 °C. The authors find that both films remain granular through 400 °C, undergo a large coarsening and roughening change between 400 and 600 °C, and form faceted crystallites by 800 °C. Mean projected feature areas increase from 784±17 to 108,529±4,362 nm2 on langasite and from 332±7 to 166,992±4,739 nm2 on sapphire, with height and roughness peaking at 600 °C and partially relaxing at 800 °C. Area distributions and shape descriptors indicate a population-level restructuring rather than uniform growth. An effective Arrhenius-type analysis based on stepwise increments of <r^2>=<A>/pi identifies the 400–600 °C interval as the dominant coarsening window. The paper concludes that the thermal stability of PtNi/Zr films is governed by the coupled influence of alloy mobility, interfacial energetics, and substrate-dependent morphological selection.","tokens_in":11762,"tokens_out":6115,"duration_ms":59127,"significance":"If the descriptive results hold, this is a useful quantitative dataset for the high-temperature SAW and thin-film electrode community, where the full PtNi/Zr/substrate stack rather than the alloy alone determines device stability. The study's strengths include a direct side-by-side comparison of two technological substrates, systematic manual segmentation with documented criteria, and explicit disclaimers that the effective coarsening energies are not microscopic diffusion barriers. The supplementary material provides per-temperature histograms and shape analyses that support the main morphological narrative. However, the causal interpretation—that the substrate differences arise from intrinsic PtNi/substrate energetic selection—is not established by the present data because of the Zr adhesion-layer confound and the absence of sample replication. The paper is honest about many of its limitations, but the abstract and conclusions overstate the mechanistic insight relative to what the measurements can support.","major_comments":[{"comment":"The causal attribution to substrate-dependent energetic selection is not supported by the data. In Section IV the authors list 'interfacial reactions involving the Zr layer and the oxide substrate' as a possible cause of the observed differences, and the Abstract and Section V conclude that thermal stability is 'governed by the coupled influence of alloy mobility, interfacial energetics, and substrate-dependent morphological selection.' However, no Zr XPS, post-anneal interface spectroscopy, or cross-sectional analysis is presented, and only one film pair was annealed. Because langasite is documented to react under high-temperature vacuum treatment (Ref. 14) and Zr is prone to oxidation and interdiffusion, the sapphire-versus-langasite differences could originate in the Zr/oxide interface or in the already different as-deposited states (Table I: RT mean areas 784 vs 332 nm2 and mean heights 3.02 vs 5.52 nm) rather than in intrinsic PtNi/substrate selection. The standard errors in Table I are feature-level, not sample-to-sample, so the substrate comparison lacks replication. This is a load-bearing point for the mechanistic interpretation, even though the descriptive substrate dependence is well documented. I recommend either adding interface chemical/structural data (e.g., Zr 3d XPS before and after annealing, cross-sectional TEM/EDS) or limiting the conclusions to full-stack, substrate-dependent behavior without claiming the dominant mechanism.","section":"Section IV; Abstract; Section V"},{"comment":"The claim of a 'sharp coarsening transition between 400 and 600 °C' is inferred from only two temperature endpoints (400 and 600 °C) in a cumulative annealing sequence. No intermediate temperature or replicate sample is reported, and the high-temperature feature counts are small (N=70 and 28 on langasite; N=65 and 36 on sapphire). A monotonic increase in mobility with temperature would also produce a large area jump over this interval, so the current data support 'a large change between 400 and 600 °C' but not the word 'sharp' as a kinetic statement. Please add an intermediate annealing point or a second film pair, or revise the wording and state explicitly that the temperature resolution of the transition is limited to this single step.","section":"Section III, Table I and Fig. 2"},{"comment":"The effective coarsening energies Q_eff=1.79 eV (langasite) and 1.23 eV (sapphire) are quoted without uncertainties. Because each value is a local slope between only two assigned points in Fig. 5, and because the underlying mean areas carry standard errors, the numerical comparison between substrates is not yet quantified. The paper's disclaimer that Q_eff is not a microscopic activation energy is appropriate, and I do not regard the analysis as circular given the explicit statement in Appendix A that n=2 is an analytical convenience. However, the two-significant-figure precision is not supported by the data. Either propagate the feature-level standard errors into the slopes or present Q_eff only as qualitative descriptors of the dominant coarsening interval.","section":"Section III, Fig. 5 and Appendix A"}],"minor_comments":[{"comment":"The claim that XPS confirms a Pt:Ni ratio of approximately 90:10 is not backed by high-resolution peak fits or quantification details; the supplement shows only survey spectra. Please provide the fitting procedure and uncertainty for the composition.","section":"Section II and Supplemental Fig. S1"},{"comment":"The label 'effective Arrhenius-like analysis' may still suggest an Arrhenius rate law despite the disclaimers; a name such as 'local temperature-sensitivity slopes' would more accurately describe the plotted quantity.","section":"Section II"},{"comment":"The caption contains the apparent typo 'overleaping' (likely 'overlapping'), and the statement that the rescaled distributions are 'broadly comparable' is not supported by a quantitative similarity test.","section":"Supplemental Fig. S3 caption"},{"comment":"Circularity is reported without an uncertainty. Adding standard errors, or explicitly stating that the value is a mean without error, would aid comparison across temperatures.","section":"Table I"},{"comment":"The crack-like surface features are mentioned but not followed up; a sentence on whether these features persist after annealing or affect the segmentation results would clarify their role.","section":"Section III, first paragraph"}],"recommendation":"major_revision","confidential_remarks":"The descriptive dataset is valuable and within the scope of cond-mat.mtrl-sci. The main risk is overinterpretation of the substrate-dependent mechanism given the Zr adhesion-layer confound and the lack of sample replication. The revision can likely be satisfied either by adding interface-sensitive measurements or by carefully limiting the causal claims in the abstract and conclusions."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"What you should know: this is a careful AFM study of how 100-nm Pt0.9Ni0.1 films with a 10-nm Zr adhesion layer evolve on sapphire and langasite under sequential vacuum annealing. The descriptive results are solid, and the paper is honest about what its effective Arrhenius analysis can and cannot mean.\n\nThe new content is a quantitative morphology dataset for this specific stack, and it does establish a substrate-dependent outcome: both films stay granular through 400 C, then undergo a sharp 400-600 C coarsening transition that replaces the small-feature population with a distinct ensemble of large crystallites, with a roughness peak at 600 C and partial smoothing at 800 C. Sapphire ends with larger features than langasite. The analysis is careful: manual segmentation criteria are documented, multiple locations and tips were compared, and the Q_eff values are explicitly labeled as regime descriptors, not activation barriers. That discipline matters.\n\nThe soft spots are real but not fatal. The biggest one is the missing interfacial chemistry. There is a 10-nm Zr adhesion layer, and the paper itself lists Zr/oxide interfacial reactions as a possible cause of substrate differences, but no post-anneal Zr chemical-state data (no XPS, no cross-section) are provided. The as-deposited films already differ substantially between substrates, and only one pair of films was annealed, so sample-to-sample variation is not captured. The high-temperature feature counts are small (N=28-70), Q_eff values come without uncertainties, and raw data are only available on request. These limitations mean the data support a substrate-dependent morphology, but the causal language about 'substrate-dependent morphological selection' is doing more work than the evidence supports.\n\nThe stress-test note is on point: the Zr/interface confound could dominate the observed difference. That said, the paper does not overclaim at every turn; it flags the possible mechanisms. So I would not call this a load-bearing flaw, but a limitation that should be addressed in revision.\n\nWho this is for: device designers working on high-temperature SAW sensors or Pt-alloy electrodes, and researchers studying metal-on-oxide dewetting. It deserves a serious referee; I would send it to review and ask for either additional Zr interface characterization or a moderated interpretation of the substrate-selection mechanism.","headline":"A careful, well-caveated experimental study; the descriptive morphology results are trustworthy, but the substrate-dependent mechanism claim is not fully separated from Zr/oxide interfacial chemistry.","tokens_in":12287,"tokens_out":3583,"would_cite":true,"duration_ms":34741,"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":"Sapphire promotes larger late-stage crystallites than langasite in Pt0.9Ni0.1 films annealed to 800 °C, while langasite shows the larger roughness peak at 600 °C.","keywords":["PtNi thin films","sapphire substrate","langasite substrate","thermal coarsening","solid-state dewetting","atomic force microscopy","thin-film morphology","high-temperature stability"],"falsifier":"Measure the chemical state and depth distribution of the zirconium layer after 400, 600, and 800 °C anneals on both substrates, for example by X-ray photoelectron spectroscopy depth profiling or cross-sectional transmission electron microscopy, and compare them; if zirconium oxidizes or interdiffuses markedly differently on sapphire versus langasite in a way that tracks the morphology differences, the paper's attribution of those differences to substrate-dependent PtNi behavior is not established.","tokens_in":11334,"feed_emoji":"🔥","tokens_out":14569,"duration_ms":112055,"temperature":0.7,"pith_summary":"The paper shows that 100-nm-thick Pt$_{0.9}$Ni$_{0.1}$ films with a 10-nm Zr adhesion layer follow different morphological paths on sapphire versus langasite when annealed identically from room temperature to 800 °C. Both films stay densely granular through 400 °C, then undergo a sharp coarsening transition between 400 and 600 °C, and finish as connected faceted crystallites at 800 °C. The quantitative outcome is substrate-controlled: the mean projected feature area grows to about $1.67\\times 10^5$ nm$^2$ on sapphire but only $1.09\\times 10^5$ nm$^2$ on langasite, while roughness peaks higher on langasite at 600 °C. The 400–600 °C window is identified as the dominant coarsening regime, and area histograms show the transition is a population replacement rather than a uniform enlargement of existing grains. This matters because the thermal stability of such films, used in high-temperature acoustic-wave devices and sensors, cannot be predicted from alloy composition alone.","feed_headline":"Sapphire grows larger PtNi crystallites than langasite","feed_subtitle":"Identical anneals to 800 °C: they coarsen sharply at 400–600 °C, but final crystal size is substrate-controlled.","key_machinery":"The central analytical device is an effective coarsening metric built from the mean projected feature area: each segmented feature is treated as a circle of radius $r$ with $A=\\pi r^2$, so $\\langle r^2\\rangle=\\langle A\\rangle/\\pi$. Plotting the stepwise increment $\\Delta\\langle r^2\\rangle$ between annealing steps against $1000/T$ yields local effective slopes that locate the dominant coarsening window without claiming microscopic activation energies. This is combined with population-resolved area histograms, which show that the 400–600 °C transition is a population replacement rather than a uniform shift, and with vertical metrics (average height and RMS roughness) to separate lateral coarsening from vertical restructuring. The paper explicitly treats the exponent $n=2$ in the classical grain-growth law as an analytical convenience, not as a mechanistic claim.","core_discovery":"The central finding is that identical thermal histories produce substrate-selected morphologies in Pt$_{0.9}$Ni$_{0.1}$/Zr films. Both films occupy a dense granular state up to 400 °C, with only modest changes in mean projected area. Between 400 and 600 °C they undergo the dominant transformation: mean projected area jumps from about $1.1\\times 10^3$ to $6.2\\times 10^4$ nm$^2$ on langasite and from $1.3\\times 10^3$ to $1.26\\times 10^5$ nm$^2$ on sapphire, while average height and RMS roughness peak at 600 °C. The area distributions show that the low-temperature granular population is replaced, not uniformly shifted, by a new population of large crystallites. At 800 °C, both films partially smooth while continuing to coarsen laterally into faceted, connected crystallites, with sapphire reaching $\\langle A\\rangle = 1.67\\times 10^5$ nm$^2$ versus $1.09\\times 10^5$ nm$^2$ on langasite. An effective Arrhenius analysis based on $\\Delta\\langle r^2\\rangle$ with $\\langle r^2\\rangle = \\langle A\\rangle/\\pi$ assigns local effective coarsening energies of 1.79 eV (langasite) and 1.23 eV (sapphire) to the 400–600 °C step, and negative local slopes above 600 °C are interpreted as saturation or faceting-limited growth, not negative barriers.","pith_inferences":["Editorial inference: if the substrate effect is energetic in origin, analogous differences should appear for other noble-metal alloy films on sapphire versus langasite, and the same segmentation-based area metrics could screen for them.","Editorial inference: a control experiment with PtNi deposited directly on the oxides, or with a different adhesion metal, would determine how much of the substrate dependence is intrinsic to the PtNi/oxide interface rather than mediated by zirconium.","Editorial inference: the 600 °C roughness maximum suggests that device processing or operation that lingers near 600 °C may be the worst case for electrode quality, even though 800 °C appears smoother in the planar AFM metrics.","Editorial inference: because the annealing sequence is cumulative and stepwise, the quoted effective energies should be validated with fixed-temperature isothermal series before being used in kinetic models; the paper itself flags this limitation."],"forward_implications":["Thermal stability of PtNi films is a property of the full PtNi/Zr/substrate stack, not of the alloy alone; substrate and adhesion layer must be part of any stability prediction.","The 400–600 °C window is the critical thermal budget risk: passing through or dwelling in this range triggers the main coarsening and roughening event.","Sapphire-supported films progress farther along the dewetting and coarsening pathway, producing larger connected faceted crystallites, while langasite retains smaller features but roughens more at intermediate temperature.","The 800 °C state is not simply rougher than the 600 °C state; it is partially smoothed by faceting-limited growth, so the worst roughness may occur before the highest operating temperature.","Area distributions imply that tracking only the average feature size is insufficient; the sudden appearance of a distinct large-crystallite population should be monitored separately."],"supporting_citations":[{"why":"This reference supplies the solid-state dewetting framework used to interpret the 400–600 °C coarsening transition.","marker":"[5]"},{"why":"This reference provides the agglomeration behavior of Pt films on dielectric substrates that motivates comparing the two oxide supports.","marker":"[6]"},{"why":"This reference gives the Pt/Ta-on-langasite film breakup sequence against which the connected 800 °C crystallite state is compared.","marker":"[26]"},{"why":"This reference identifies ZrO2 precipitates and grain-boundary pinning as the stabilization mechanism proposed for the Zr adhesion layer.","marker":"[3]"},{"why":"This reference supplies the classical grain-growth rate law used in the Appendix to define the effective area-based coarsening relation.","marker":"[27]"},{"why":"This reference provides the normal grain-growth exponent n=2 adopted as an analytical convenience for the effective Arrhenius analysis.","marker":"[28]"},{"why":"This reference defines the equilibrium-shape framework used to frame final crystallite shape as a balance of surface and interface energies.","marker":"[19]"},{"why":"This reference underlies the free-energy decomposition into surface, interface, grain-boundary, and stress contributions used throughout the discussion.","marker":"[8]"},{"why":"This reference documents langasite surface chemical changes under high-temperature vacuum, supporting the substrate-dependent interpretation.","marker":"[14]"}],"fun_headline_variants":["Substrate dictates final PtNi crystallite size","Sapphire yields larger PtNi crystals than langasite","Same anneal, different morphology: PtNi on sapphire vs langasite","Heat reveals substrate-selected PtNi film coarsening","Sapphire PtNi grains outsize langasite after 800°C"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the 10-nm zirconium adhesion layer behaves the same way on both substrates, so the differing film morphologies can be attributed to the sapphire and langasite surfaces rather than to chemistry involving zirconium.","fun_headline_variants_meta":{"raw":{"variants":["Substrate dictates final PtNi crystallite size","Sapphire yields larger PtNi crystals than langasite","Same anneal, different morphology: PtNi on sapphire vs langasite","Heat reveals substrate-selected PtNi film coarsening","Sapphire PtNi grains outsize langasite after 800°C"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000535,"raw_usage":{"total_tokens":2744,"prompt_tokens":1292,"completion_tokens":1452,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":908,"completion_tokens_details":{"reasoning_tokens":1362}},"tokens_in":908,"tokens_out":1452,"duration_ms":11863,"temperature":1.0,"reasoning_tokens":1362,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T04:34:16.764321+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the chemical state and depth distribution of the zirconium layer after 400, 600, and 800 °C anneals on both substrates, for example by X-ray photoelectron spectroscopy depth profiling or cross-sectional transmission electron microscopy, and compare them; if zirconium oxidizes or interdiffuses markedly differently on sapphire versus langasite in a way that tracks the morphology differences, the paper's attribution of those differences to substrate-dependent PtNi behavior is not established.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"This reference supplies the solid-state dewetting framework used to interpret the 400–600 °C coarsening transition."},{"cited_title":"Galinski, T","cited_arxiv_id":null,"evidence_quote":"This reference provides the agglomeration behavior of Pt films on dielectric substrates that motivates comparing the two oxide supports."},{"cited_title":"Aubert, O","cited_arxiv_id":null,"evidence_quote":"This reference gives the Pt/Ta-on-langasite film breakup sequence against which the connected 800 °C crystallite state is compared."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"This reference identifies ZrO2 precipitates and grain-boundary pinning as the stabilization mechanism proposed for the Zr adhesion layer."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"This reference supplies the classical grain-growth rate law used in the Appendix to define the effective area-based coarsening relation."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"This reference provides the normal grain-growth exponent n=2 adopted as an analytical convenience for the effective Arrhenius analysis."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"This reference defines the equilibrium-shape framework used to frame final crystallite shape as a balance of surface and interface energies."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"This reference underlies the free-energy decomposition into surface, interface, grain-boundary, and stress contributions used throughout the discussion."},{"cited_title":"Seifert, G","cited_arxiv_id":null,"evidence_quote":"This reference documents langasite surface chemical changes under high-temperature vacuum, supporting the substrate-dependent interpretation."}],"review_version":1}