{"id":"314e6e8e-34f1-4236-8572-44621038fb77","arxiv_id":"2607.01997","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"Thick NiTi shape-memory coatings that recover strain when heated were made on steel by high-velocity air-fuel spraying followed by annealing — a first for this method.","lead":"HVAF spraying plus a 500 °C anneal produced thick nickel–titanium coatings on steel that change phase and recover strain when heated — a shape-memory behavior that plasma-sprayed NiTi coatings usually lose. The result points to a practical way to put functional smart coatings on ordinary engineering parts, though the current coatings are porous, brittle, and the claimed adhesion is not yet measured.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Functional SME claim is not established: recovery is attributed to martensitic transformation, but the paper's own tests state nanoindentation 'cannot be classified ... as superelastic nor as shape memory' and scratch recovery is explicitly on as-fabricated coatings that 'shows neither superelastici","rationale":"The reader's weakest assumption identifies exactly the load-bearing gap: recoverable strain is attributed to martensitic transformation without a direct test. The paper's own internal statements undermine the shape-memory interpretation: scratch tests were on as-fabricated coatings that 'shows neither superelasticity nor shape memory,' and nanoindentation responses are explicitly not classifiable as superelastic or shape memory. The only positive functional evidence is DSC peaks in annealed coatings (which prove transformation occurs) and one annealed sample's non-standard bending hysteresis. The latter could be ordinary thermoelastic or viscoelastic response of a porous composite; no control experiment separates these. The adhesion claim in the abstract is also contradicted by the body's admission that adhesion was not evaluated, but the functional claim is more central to the novelty. The appropriate verdict remains CONDITIONAL: the feasibility of HVAF NiTi coatings with MT is plausible and DSC data provide independent support, but the paper must either supply direct recovery-on-heating evidence or soften the headline claim. No change to the reader's verdict is needed; the concern reinforces it.","tokens_in":15902,"tokens_out":3337,"duration_ms":38501,"concrete_test":"Perform a free-standing thermal shape-memory test on an annealed coating (at least coating 4 and one replicate): detach the coating, deform it in bending at T < Mf (or in the R-phase regime at room temperature) to introduce a measurable residual strain, then heat through Af (~55 °C) in a TMA/DMA without applied load and record strain recovery. If the residual strain does not substantially recover on heating, the functional SME claim collapses to ordinary elastic accommodation of a damaged composite. Repeating the 3-point bending thermal cycle on a non-transforming control (e.g., Ni-rich NiTi with Ms < -60 °C) would further confirm that the hysteresis is transformation-induced rather than viscoelastic/elastic response.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim — 'first ever successful fabrication of thick NiTi coatings that display functional thermomechanical properties' — requires that the measured recoverable strains genuinely arise from the B2–R–B19' martensitic transformation, not from ordinary elastic accommodation of a porous, oxide-laden, stress-affected composite. The manuscript does not establish this. (1) The only bending test with recoverable strain is one annealed sample (coating 4, Fig. 11), and the authors state the strain-temperature hysteresis is 'not standard.' (2) In §3.4.3, scratch tests on as-fabricated coatings show ~60% elastic recovery, but the text explicitly says the as-fabricated coating 'shows neither superelasticity nor shape memory'; the recovery is 'somewhere in between' literature SE/SM values, so ordinary elasticity is a plausible explanation. (3) In §3.4.4, nanoindentation responses 'can be classified neither as superelastic nor as shape memory.' (4) No free-standing recovery-on-heating test (deform below Mf, heat above Af) is reported — the standard proof of shape memory effect. Thus the central functional novelty is inferred, not demonstrated. Adhesion is also asserted in the abstract ('very good adherence'), but §3.3.2 states 'we did not evaluate the adhesion.'","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports fabrication of thick NiTi coatings (100–300 µm) on mild steel by high-velocity air-fuel (HVAF) spraying using four processing-parameter sets, followed by microstructural, compositional, phase-transformation, and mechanical characterization. The authors find as-sprayed coatings are austenitic with no clear DSC peaks; after annealing at 500 °C for 1 h, the coatings show B2–R–B19′ transformations with small, broad DSC peaks. One annealed coating (coating 4) displays hysteretic recoverable bending strain in 3-point bending, while scratch tests on as-fabricated coatings show ~60% elastic recovery and multicycle nanoindentation shows stable recovered depths. The abstract claims 'first ever successful fabrication' of thick NiTi coatings with functional thermomechanical properties and 'very good adherence' to the substrate. The central functional claim is that these recoverable strains arise from the martensitic transformation, i.e., genuine shape-memory behavior.","tokens_in":16127,"tokens_out":3766,"duration_ms":42905,"significance":"If substantiated, the work would identify HVAF as a promising route for thick NiTi coatings, avoiding melting/oxidation issues of plasma spray and adhesion issues of cold spray. The paper is commendable for its direct experimental characterization of composition, porosity, phases, transformation temperatures, and mechanical responses, and for making data available on Zenodo. However, the central claim is not yet established: the only bending demonstration is on one annealed sample with a 'non-standard' hysteresis; the scratch and nanoindentation responses are explicitly stated not to be classifiable as superelastic or shape-memory; adhesion is explicitly not evaluated; and no free-recovery-on-heating test is provided. The reported recoveries may be ordinary elastic accommodation of a porous, oxide-laden, cold-worked composite, in which case the novelty collapses. The paper is a useful process-development and characterization study, but the functional-SMA claim requires substantial additional evidence or significant rewriting.","major_comments":[{"comment":"The abstract claims the coatings 'simultaneously show very good adherence to the substrate,' but §3.3.2 states 'we did not evaluate the adhesion.' The absence of detachment in bending/tensile tests is not a quantitative adhesion measurement. The adhesion claim must be removed or supported by a proper test (e.g., pull-off, scratch adhesion, or interfacial indentation).","section":"Abstract and §3.3.2"},{"comment":"The central claim of functional shape-memory behavior is not demonstrated. The only recoverable strain in bending is reported for one annealed sample (coating 4), and the authors state the hysteresis is 'not standard.' Scratch tests were performed on as-fabricated coatings, and §3.4.3 says the coating 'shows neither superelasticity nor shape memory'; nanoindentation (§3.4.4) 'can be classified neither as superelastic nor as shape memory.' No free-standing recovery-on-heating test (deform below Mf, heat above Af) or constrained-recovery work-output test is reported. The observed recoveries could be ordinary elasticity of a porous, oxide-containing, stressed composite. The authors should either provide direct evidence that recovery is caused by the B2–R–B19′ transformation, or substantially weaken the functional claims.","section":"§3.4.2, Fig. 11; §3.4.3; §3.4.4"},{"comment":"The functional-property evidence is thin relative to the breadth of the claim. The only bending functional test is on one annealed coating; the scratch and nanoindentation tests are on as-fabricated coatings; the as-sprayed DSC curves show no clear transformation peaks; and the annealed peaks are small and broad. There is no replication or error quantification for the bending result. The conclusion that HVAF NiTi coatings 'display functional thermomechanical properties' is therefore not supported for the coating family as a whole, only suggestive for one annealed sample.","section":"§3.3.4, Table 5, §3.4"},{"comment":"The 'first ever successful fabrication' claim in the abstract needs careful benchmarking. The same group's earlier work (Refs. 13, 14, 18) reports thermal-plasma-sprayed NiTi coatings 'possessing functional properties,' which appears to overlap with the current claim. If the novelty is specifically HVAF, or thickness, or simultaneous adhesion, that must be stated precisely. Otherwise the 'first ever' claim is overreaching and may be inconsistent with the prior literature cited in the paper itself.","section":"Introduction, Abstract, Refs. 13,14,18"}],"minor_comments":[{"comment":"Typo: 'compar6' should be 'compared.' Also in §3.2, 'temeperatre and tine' and 'the the' should be corrected.","section":"§3.3.4"},{"comment":"The text references 'DSC peaks ... in Fig. 1f' when comparing powder and coating DSC peaks, but Fig. 1f appears to show EDX chemical-composition results, not DSC. Please correct the cross-reference.","section":"Fig. 1 and text"},{"comment":"There are two subsections numbered '4.1' ('Microstructure and functional mechanical properties' and 'Characterization of mechanical properties by scratch and nanoindentation'); the second should be renumbered '4.2.'","section":"Discussion, §4.1"},{"comment":"The Discussion says 'Ni content ... (Table 5)' and 'Ms temperature ... (Table 6).' Table 5 lists transformation temperatures and Table 6 lists Young's moduli/strengths; chemical composition is in Table 4. Check all table cross-references for consistency.","section":"Discussion, text near Table 5/Table 6"}],"recommendation":"major_revision","confidential_remarks":"The paper is an interesting process-development study, but the claims in the abstract and conclusions outrun the evidence. The authors' own previous publications (Refs. 13, 14, 18) appear to undercut the 'first ever' novelty statement, so I recommend the editor ask for a precise novelty positioning. The functional-SMA claim needs either a free-recovery test on annealed coatings or a clear statement that the recoveries are not yet proven to be transformation-related. With those changes, the paper could be acceptable as a materials-characterization contribution."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Worth a look if you work on thermal spray or NiTi coatings. The genuinely useful result: HVAF can produce 100–300 μm NiTi coatings on mild steel that, after a 500 °C anneal, show DSC peaks for B2-R-B19' transformation and a recoverable bending strain in one annealed sample. That is more than prior plasma-spray NiTi coatings achieved, and the paper presents it as a feasibility study, not a finished process. The data are empirical, the authors openly report porosity, low tensile strength, oxides, and residual stress, and they deposit the datasets on Zenodo. Credit where due: no invented mechanism, no fitting, and the limitations section is unusually candid.\n\nThe soft spots are mostly in the gap between abstract and body. The abstract claims \"very good adherence\" and \"shape memory effects in nanoindentation and scratch tests,\" but the text says adhesion was not evaluated and the as-fabricated scratch/nanoindentation responses \"can be classified neither as superelastic nor as shape memory.\" That is an internal mismatch, and it matters because the central functional claim—that the recoverable strain originates from the martensitic transformation—is inferred rather than demonstrated. The bending proof rests on a single annealed sample (coating 4) with a hysteresis curve the authors themselves call \"not standard.\" No free-standing recovery-on-heating test is reported, so ordinary elastic accommodation of a porous, oxide-laden, cold-worked composite remains a live alternative explanation. These are fixable gaps, but they need to be fixed before the \"first ever successful fabrication of thick NiTi coatings that display functional thermomechanical properties\" claim can stand.\n\nMinor issues: quantitative tables (porosity, moduli, transformation temperatures) have no error bars or sample counts in the text, and the four processing conditions differ in multiple parameters at once, so no processing-property conclusion should be drawn. The citation pattern is heavy on the same group's prior work, but the relevant comparison papers (plasma spray, cold spray, HVOF NiTi) are there, and the novelty against those is plausible.\n\nWho should read it: anyone trying to make smart coatings by thermal spray, and experimentalists looking for a benchmark dataset. It deserves a serious referee, but only after the authors either demonstrate recovery-on-heating on a free-standing annealed coating or soften the functional claims. I'd send it to review with a request for major revision rather than desk reject it.","headline":"Solid fabrication study with an overreach: the HVAF NiTi coatings are real, but the abstract's functional shape-memory claim is ahead of the evidence.","tokens_in":16795,"tokens_out":2088,"would_cite":true,"duration_ms":23127,"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":"Thick NiTi shape-memory coatings on steel can be made by warm-spray deposition and restored by annealing.","keywords":["NiTi shape memory alloy","HVAF thermal spray","martensitic transformation","B2-R-B19'","shape memory coating","nanoindentation","scratch test","mild steel substrate"],"falsifier":"Heat a plastically deformed, free-standing annealed coating above A_f (about 55 °C) under no load and measure whether it recovers its original shape; if it does not, the observed 'shape memory' is ordinary elasticity. Alternatively, run the 3-point bending cycle at zero stress and see whether the strain-temperature hysteresis persists.","tokens_in":15700,"feed_emoji":"🔥","tokens_out":6298,"duration_ms":66651,"temperature":0.7,"pith_summary":"Shape-memory NiTi is hard to deposit as a thick coating: melting-based spray routes oxidize and destabilize it, and cold spray doesn't stick. This paper claims that high-velocity air-fuel (HVAF) spraying, a 'warm spray' that keeps the alloy solid, can lay down 100–300 µm NiTi coatings on mild steel that, after a 500 °C anneal, undergo the B2–R–B19' martensitic transformation and show recoverable strain in 3-point bending. It also reports that scratch tests recover about 60% of the deformation elastically and that multicycle nanoindentation shows stable recovered depth, suggesting functional shape-memory behavior. If the claim holds, HVAF gives a practical route to large-area smart coatings on cheap steel substrates for wear, corrosion, and actuation applications.","feed_headline":"First thick NiTi shape-memory coatings on steel, made by warm spray","feed_subtitle":"After a one-hour anneal they transform martensitically and recover strain in bending, scratch, and indentation tests.","key_machinery":"The mechanism that carries the argument is the martensitic transformation of the NiTi matrix, in particular the B2→R→B19' sequence that emerges after annealing. The process that makes this possible is HVAF's 'warm spray' regime: particles are heated to roughly 1000 °C and accelerated to 600–900 m/s, enough to plastically deform and compact them into a thick adherent layer without melting, which limits oxidation and preserves the NiTi matrix. The annealing step at 500 °C/1 h is the key post-treatment: it partially relieves the dislocation density and internal stress that suppress the transformation in the as-sprayed state.","core_discovery":"The paper's central claim is that high-velocity air-fuel spraying can produce thick NiTi shape-memory coatings on mild steel that are adherent and, after post-annealing, functionally thermomechanically active. Using pre-alloyed Ni50Ti50 powder and four parameter sets, the authors deposited 100–300 µm coatings. As-sprayed coatings are austenitic, chemically homogeneous on the macroscale but nanoscale-inhomogeneous, with pores, TiO2 oxides, high dislocation density, and internal stress; DSC shows no clear transformation peaks. Annealing at 500 °C for 1 h produces B2–R–B19' transformation peaks on DSC. In 3-point bending under 300 MPa, one annealed coating (coating 4) shows a hysteretic strain–","pith_inferences":["The transformation origin of the observed recoverable strains is not yet directly demonstrated; a free-standing recovery-on-heating test above A_f would confirm whether the 'shape memory' is genuine martensitic recovery or ordinary elasticity of a damaged composite.","The abstract's 'very good adherence' is an inference: the paper states adhesion was not measured, so the claim rests on the absence of delamination in bending and scratch tests rather than on bond-strength data.","The scratch-recovery values (57–61%) fall between typical superelastic (69–81%) and shape-memory (37–52%) ranges, suggesting the as-sprayed coatings behave like an intermediate, partially transforming material — annealing studies could tune which regime they enter.","If the B2–R–B19' transformation is confirmed to be load-bearing, the R-phase (which is absent in Ti-rich NiTi) could itself be used as a sensitive diagnostic for internal stress and Ni-enrichment remaining after HVAF."],"forward_implications":["If the claim holds, HVAF becomes a viable industrial route for placing thick, functional NiTi layers on steel components without the chemical inhomogeneity that defeats plasma spray.","Annealing at 500 °C for 1 h is sufficient to restore martensitic transformation in as-sprayed coatings, giving a simple post-processing recipe for functional behavior.","Scratch and multicycle nanoindentation are sensitive to HVAF processing parameters and can serve as substrate-attached screening tools for optimizing the spray process.","Because the coatings can be separated from the substrate, conventional thermomechanical tests can be run on coating material itself, enabling quantitative property optimization.","Reducing porosity and oxide content in HVAF coatings should increase the current low tensile strength (~200 MPa) and unlock the full functional thermomechanical performance."],"fun_headline_variants":["HVAF spray makes first thick NiTi shape-memory coatings on steel","Warm-sprayed NiTi coatings on steel get shape memory after anneal","Thick NiTi coatings on steel via HVAF: shape memory after 1h anneal","First thick NiTi SMA coatings on steel, HVAF-sprayed and annealed","HVAF-sprayed NiTi: thick, adherent, shape-memory after anneal"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The central claim depends on the assumption that the recoverable strains measured in bending, scratch, and nanoindentation come from martensitic transformation of the NiTi matrix rather than from ordinary elastic accommodation of a porous, oxide-laden, cold-worked composite.","fun_headline_variants_meta":{"raw":{"variants":["HVAF spray makes first thick NiTi shape-memory coatings on steel","Warm-sprayed NiTi coatings on steel get shape memory after anneal","Thick NiTi coatings on steel via HVAF: shape memory after 1h anneal","First thick NiTi SMA coatings on steel, HVAF-sprayed and annealed","HVAF-sprayed NiTi: thick, adherent, shape-memory after anneal"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00017,"raw_usage":{"total_tokens":1150,"prompt_tokens":834,"completion_tokens":316,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":578,"completion_tokens_details":{"reasoning_tokens":206}},"tokens_in":578,"tokens_out":316,"duration_ms":3479,"temperature":1.0,"reasoning_tokens":206,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-03T01:59:23.181184+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Heat a plastically deformed, free-standing annealed coating above A_f (about 55 °C) under no load and measure whether it recovers its original shape; if it does not, the observed 'shape memory' is ordinary elasticity. Alternatively, run the 3-point bending cycle at zero stress and see whether the strain-temperature hysteresis persists.","supporting_citations":[],"review_version":2}