{"id":"adb9ba35-f933-4bdd-86fb-b5b315394ff9","arxiv_id":"2607.09977","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"A size-dependent chemo-thermo-mechanical enthalpy model for coherent Ni/NiH interfaces predicts enthalpy and volume changes from ~1 nm to bulk and attributes anisotropic growth to elastic anisotropy.","lead":"Researchers built a continuum enthalpy model for nickel–nickel-hydride interfaces that folds in chemistry, heat, strain, size, and interface energy. It explains why hydride growth prefers different crystal faces as hydrogen loading rises, which matters for Ni catalysts and solid-state hydrogen storage.","discovery_kind":"extension","skeptic_critique":{"model":"grok-4.5","headline":"The G≈H approximation is the load-bearing soft spot for the claimed {111}→{100} crossover and equilibrium dimensions.","rationale":"The Reader correctly isolates the G≈H substitution as the weakest assumption that directly underwrites both the equilibrium dimensions and the anisotropic-growth explanation. The MD–model enthalpy agreement (Fig. 11) remains solid under the paper’s own metric, the elastic-constant anisotropy is real, and the pure-end-member / coherent-lattice approximations are secondary for the nano-scale regime the authors target. No stronger internal inconsistency appears. Because the entropy check is feasible with the same trajectories already used for enthalpy, the verdict stays CONDITIONAL rather than being upgraded or rejected; the concrete free-energy comparison would settle whether the soft spot actually moves the crossover.","tokens_in":18739,"tokens_out":538,"duration_ms":4978,"concrete_test":"From the existing NPT MD trajectories (or short additional runs) of the same {100} and {111} supercells at x=0.25, 0.5, 0.75 and T=300–500 K, compute the vibrational free-energy difference via the phonon density of states (or quasi-harmonic approximation) of the strained α and β regions plus the interface. If |TΔS| exceeds ~3–5 meV/Ni atom and changes sign between the two orientations, the enthalpy-only ranking and the claimed crossover are unreliable.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central claim rests on minimizing H(x,T,P,{l},λ) (Eqs. 11–12) rather than G, with the justification that “entropy terms were shown to be small in our earlier work [22]” (Section 4.3). That earlier work treated bulk, unstrained Ni–H; it does not quantify configurational or vibrational entropy differences between coherently strained α and β layers, nor the interface excess entropy. Because the reported molar-enthalpy difference between {100} and {111} orientations is only 1–7 meV per Ni atom (Fig. 9), even a modest orientation- or strain-dependent –TΔS term of a few meV can reverse the sign of ΔG and therefore move or eliminate the predicted crossover at x≈0.25. The same approximation also determines the equilibrium lattice dimensions that the model claims to reproduce. Thus the anisotropy explanation and the size-dependent enthalpy predictions are only as reliable as the untested G≈H substitution under the precise strain states of the coherent interfaces.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.5","summary":"The manuscript develops a continuum chemo-thermo-mechanical (CTM) enthalpy model H(x, T, P, {l}, λ) for coherent Ni/NiH (α/β) interfaces in thin films. Starting from bulk 0 K properties, heat capacities, thermal expansion, and temperature- and strain-dependent elastic constants (fitted to single-phase NVT MD and transformed for orientation), the model adds an independently extracted interface energy λ and minimizes H with respect to the layer dimensions. It is shown to reproduce NPT MD molar enthalpies and equilibrium lattice parameters for both {100} and {111} interfaces from ~1 nm to tens of nm (and by construction to the bulk limit). The authors attribute the observed switch in preferred growth direction—from {111} at low x to {100} at higher x—to anisotropic elastic constants rather than interface energy, and argue that the same framework can be transferred to other nanostructured metal hydrides.","tokens_in":19062,"tokens_out":1398,"duration_ms":24404,"significance":"If the predictions hold, the work supplies a practical, size-aware continuum enthalpy model that bridges atomistic MD to continuum scales for a technologically relevant hydride system, correctly capturing the strong chemo-thermo-mechanical coupling that is usually omitted from continuum treatments. The demonstration that elastic anisotropy alone can reverse the preferred interface orientation with hydride fraction is a concrete, falsifiable explanation for anisotropic growth previously seen in the authors’ Monte Carlo nanoparticle simulations and in related Pd systems. The systematic protocol (single-phase fitting → tensor transformation → excess-enthalpy extraction of λ → free-dimension minimization) is reusable for other solid-state hydrogen-storage materials. Strengths include quantitative parity with the underlying MD (R^{2} improving to 0.99 once coupling terms are retained) and explicit coverage of the full length-scale range claimed in the abstract.","major_comments":[{"comment":"Section 4.3 and Eqs. (11)–(12): equilibrium dimensions and the orientation ranking are obtained by minimizing H rather than G, with the sole justification that “entropy terms were shown to be small in our earlier work [22]”. That earlier work examined bulk, unstrained Ni–H; it does not quantify configurational or vibrational entropy differences between coherently strained α and β layers or the interface excess entropy. Because the molar-enthalpy difference between {100} and {111} orientations is only 1–7 meV per Ni atom (Fig. 9), a modest orientation- or strain-dependent –TΔS of a few meV can reverse the sign of ΔG and therefore move or eliminate the claimed {111}→{100} crossover at x ≈ 0.25. The same approximation underpins the lattice-parameter predictions that the model claims to reproduce. Either a free-energy calculation (or at least an estimate of the relevant ΔS under the actual i","section":"Section 4.3, Eqs. (11)–(12), Fig. 9"},{"comment":"Section 2.1 and the construction of H: the model treats α as pure Ni and β as stoichiometric NiH, thereby discarding both the finite H solubility of the α phase and all configurational entropy of the interstitial solid solution. While this simplifies the continuum description, the neglected terms are of the same order as the elastic and interface contributions that drive the reported orientation switch. A quantitative bound on the error introduced by the pure-phase approximation (e.g., by comparing to a few mixed-composition MD cells or to the authors’ own earlier free-energy calculations) is needed to confirm that the predicted crossover and size dependence remain intact.","section":"Section 2.1"}],"minor_comments":[{"comment":"Table 4 lists C44 = 0 for NiH. If this is the value returned by the EAM potential it should be stated explicitly and its consequences for the {111} transformation discussed; if it is a typographical omission the correct number should be supplied.","section":"Table 4"},{"comment":"The abstract and title emphasize nanoparticles, yet all continuum calculations are performed on periodic thin-film geometries. A short paragraph clarifying how the film results map onto faceted nanoparticles (or an explicit caveat) would improve accessibility.","section":"Abstract / Section 5"},{"comment":"Surface energies are computed (Section 4.6) but never inserted into the working enthalpy expression used for the size-dependent comparisons. Either drop them or show a representative calculation that includes free surfaces.","section":"Section 4.6"},{"comment":"Several figure captions and the TOC image are missing or incomplete in the supplied manuscript; ensure all panels of Figs. 7–11 are fully labeled with units and that the supplementary figures referenced in the text are available.","section":"Figures 7–11"},{"comment":"Minor typographical issues: “s ize-dependent” (title), “α/β interface” vs. “α/β” inconsistency, and occasional missing spaces around mathematical symbols.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"The central technical contribution (a transferable CTM enthalpy model that matches its parent MD) is solid and publishable after the free-energy justification is strengthened. The anisotropy claim is the main selling point and currently rests on an untested G ≈ H substitution at an energy scale of a few meV/atom; that is the only load-bearing weakness. The heavy self-citation to the authors’ prior nanoparticle MC work is understandable for motivation but should be balanced by more independent experimental literature on anisotropic hydride growth. Scope is appropriate for a materials-theory journal."},"author_rebuttal":null,"desk_editor":{"model":"grok-4.5","letter":"The useful new piece is a fully parameterized size-dependent chemo-thermo-mechanical enthalpy H(x,T,P,l,λ) for coherent Ni/NiH thin-film interfaces, with explicit {100} vs {111} anisotropy and temperature–strain coupling corrections extracted from MD. Bulk numbers (lattice, cohesive energy, Cij, Cp, thermal expansion) sit close to experiment and prior EAM work; parity plots improve once the coupling terms are added; and the continuum minimization recovers NPT enthalpies and z-dimensions from ~1 nm to tens of nm. Interface energy λ is measured independently from excess enthalpy vs 1/N, not fitted into the bulk model. That is clean, transferable protocol work for people who need continuum inputs for Ni hydride or similar systems.\n\nWhat it does well is the bookkeeping. Single-phase strained MD supplies the hyper-elastic corrections (Table 5); those are then used, with tensor rotation, on two-phase cells that were not in the fit. The elastic-anisotropy story for why {111} is preferred at low x and {100} at higher x is physically plausible and matches the authors’ earlier MC observation on nanoparticles. Surface energies are shown to be secondary, which is honest.\n\nThe soft spot that actually matters is the G≈H substitution used for the equilibrium dimensions and the crossover. Justification is a citation to their earlier bulk, unstrained work. The enthalpy difference between orientations is only 1–7 meV/Ni (Fig. 9). A modest orientation- or strain-dependent –TΔS of a few meV can flip the sign of ΔG and move or erase the x≈0.25 switch. Pure end-members and the always-coherent assumption are secondary approximations for the nano regime they target, but they are still approximations. No public code or raw trajectories, so independent checks cost more than they should.\n\nThis is for people building continuum or multi-scale models of metal-hydride interfaces, not for a broad materials audience. The math and data are solid enough that a serious editor should send it to referees; the G≈H issue is exactly the kind of thing referees can force into the open. I would cite the parameterized model and the λ(T,x) numbers if I needed Ni/NiH continuum inputs; I would not treat the growth-direction switch as settled without an entropy check.","headline":"Solid, carefully parameterized CTM enthalpy for coherent Ni/NiH {100}/{111} interfaces that matches MD; the G≈H step is the real soft spot for the claimed growth switch.","tokens_in":19659,"tokens_out":581,"would_cite":true,"duration_ms":5498,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.5","headline":"A chemo-thermo-mechanical enthalpy model explains anisotropic Ni hydride growth from elastic anisotropy and size.","keywords":["α/β interface","strain energy","nickel hydride","hydrogen storage","enthalpy of hydride","molecular dynamics","volume expansion","interface energy"],"falsifier":"Measure the relative populations of (111) versus (100) coherent Ni/NiH interfaces in thin films or nanoparticles as a function of hydride fraction x at fixed temperature; if the crossover near x = 0.25 is absent or reversed, the elastic-anisotropy explanation fails.","tokens_in":19620,"feed_emoji":"⚗️","tokens_out":898,"duration_ms":7252,"temperature":0.7,"pith_summary":"Nickel nanoparticles can form hydride at modest hydrogen pressure, but the growing Ni/NiH interface is coherent and anisotropic: growth prefers one crystal plane early and another later. This paper builds a simplified molar enthalpy model H(x, T, P, l, λ) that folds chemical composition, temperature, elastic strain, film thickness, and interface energy into one expression for thin-film Ni/NiH systems. The model is fitted to molecular-dynamics data for bulk heat capacity, thermal expansion, elastic constants (with temperature- and strain-dependent corrections), and interface energies of the (100) and (111) orientations. It recovers the enthalpy and lattice dimensions measured in MD from roughly 1 nm to tens of nanometres and beyond. The central physical insight is that anisotropic elastic constants alone reverse the relative stability of the two interfaces once the hydride fraction exceeds about 0.25, thereby explaining the observed switch in preferred growth direction without invoking surface or interface energy as the dominant driver.","feed_headline":"Elastic anisotropy flips Ni hydride growth plane with loading","feed_subtitle":"A single enthalpy model captures size, heat and strain from 1 nm to bulk and explains the switch.","key_machinery":"The size-dependent chemo-thermo-mechanical enthalpy H(x, T, P, l, λ) obtained by adding bulk thermo-elastic contributions of the α and β phases to a positive interface term λ A_int and then minimising with respect to the four independent lengths of the bilayer.","core_discovery":"The authors show that a continuum enthalpy function H(x, T, P, l, λ) that includes temperature- and strain-corrected elastic constants, thermal expansion, heat capacity, and a size-dependent interface term accurately reproduces MD enthalpies and equilibrium lattice parameters of coherent Ni/NiH bilayers for both (100) and (111) orientations across length scales from ~1 nm to the continuum limit; the same elastic anisotropy causes the preferred interface to switch from (111) at low hydride fraction to (100) at higher fraction.","pith_inferences":["If the G ≈ H approximation holds for other fcc metals, the same elastic-anisotropy switch should appear in Pd and Pt hydrides and could be checked by TEM.","The model supplies a cheap continuum surrogate that can replace repeated MD runs when screening particle-size or temperature effects on hydride thermodynamics.","Once dislocations become probable (larger particles), the coherent-lattice assumption will break and the predicted crossover may shift or disappear."],"forward_implications":["Preferred hydride growth direction in Ni can be predicted from elastic constants alone once x is known.","The same construction supplies a transferable enthalpy model for other coherent metal/hydride systems that exhibit anisotropic growth.","Interface and surface energies matter only below ~20 nm; above that scale continuum elastic anisotropy dominates.","Corrections that couple temperature and strain into the elastic constants and heat capacity are required for quantitative enthalpy predictions."],"fun_headline_variants":["Elastic anisotropy switches Ni hydride growth from (111) to (100)","Enthalpy model shows size and strain flip preferred Ni/NiH plane","Coherent Ni/NiH interface orientation flips with hydride fraction","Anisotropic elasticity drives Ni hydride growth-plane switch","Single H(x,T,P,l,λ) model predicts Ni/NiH plane flip across scales"],"cache_read_input_tokens":16512,"weakest_assumption_plain":"Free-energy minimisation is replaced by pure enthalpy minimisation (G ≈ H), on the claim that entropy differences between the strained phases remain small.","fun_headline_variants_meta":{"raw":{"variants":["Elastic anisotropy switches Ni hydride growth from (111) to (100)","Enthalpy model shows size and strain flip preferred Ni/NiH plane","Coherent Ni/NiH interface orientation flips with hydride fraction","Anisotropic elasticity drives Ni hydride growth-plane switch","Single H(x,T,P,l,λ) model predicts Ni/NiH plane flip across scales"]},"model":"grok-4.5","effort":"low","cost_usd":0.00618,"raw_usage":{"total_tokens":1626,"prompt_tokens":798,"num_sources_used":0,"completion_tokens":101,"cost_in_usd_ticks":61800000,"prompt_tokens_details":{"text_tokens":798,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":727,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":798,"tokens_out":101,"duration_ms":5905,"temperature":1.0,"reasoning_tokens":727,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-14T01:18:40.468695+00:00","model_set":{"reader":"grok-4.5"},"falsifier":"Measure the relative populations of (111) versus (100) coherent Ni/NiH interfaces in thin films or nanoparticles as a function of hydride fraction x at fixed temperature; if the crossover near x = 0.25 is absent or reversed, the elastic-anisotropy explanation fails.","supporting_citations":[],"review_version":1}