{"id":"c1ccfc95-3cc4-4013-b8bf-db62fe4d3ea3","arxiv_id":"2607.28752","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"The nearest-neighbor manganese fraction on the B site controls oxygen-vacancy formation energy by 1.0-1.5 eV, enabling composition maps that predict ceria-competitive hydrogen cycling at 1350°C.","lead":"This paper maps how the local arrangement of atoms around an oxygen site controls the energy needed to remove that oxygen in Ca-Ce-Ti-Mn perovskite materials used for solar hydrogen production. The authors find that the nearby manganese content dominates, and identify compositions predicted to match ceria's hydrogen output at lower temperatures, with initial experiments supporting the trend.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"One KDE-selected vacancy per 1NN class cannot constrain the low-Ev tail that exponentially drives Δδ; the 2.2x dGNN-vs-SCAN+U spread at CCTM2112 (0.098 vs 0.044, measured 0.048–0.067) shows this tail is unverified, and the screening claims rest on it.","rationale":"Good-faith reading: the paper resolves the 15 1NN oxygen environments, shows B-site 1NN chemistry dominates Ev, and adds interpretable CFM + dGNN screening with three experimental compositions. The mechanism claim is well supported: Fig. 4 is direct DFT; all 15 environments are realized within each composition, so the B-site effect is identified within-composition; CFM LOGO MAE (0.174 eV) matching random-split MAE is genuine evidence; the measured monotonic XCe trend under Protocol C is not contradicted by the fitted models.\n\nThe least secure link is the screening layer: it is exponentially sensitive to the low-Ev tail, and one KDE-typical vacancy per 1NN class cannot see that tail. The paper's own Discussion names 'a nonrepresentative low-Ev tail in the 15-site DFT sample' as a live explanation of the XCe = 0.29 overprediction, and the Methods state plainly 'we created one neutral oxygen vacancy per environment class.' The dGNN-vs-SCAN+U spread at the flagship composition (0.098 vs 0.044; measured 0.048–0.067 as lower bounds) is the empirical signature of tail blindness; the CFM inherits it by construction. The bias is composition-dependent (under at CCTM2112, over at XCe = 0.29), so in-window fractions and Fig. 5 are miscalibrated in shape, not merely offset.\n\nWhy not REJECT: the mechanism claim is internally consistent, parameter-light, backed by direct DFT residuals, and the experimental data show the expected qualitative trend. Why not ACCEPT: the headline ceria-level Δδ claim depends on the unverified tail, and the paper's own cross-check model disagrees by 2.2x at the flagship composition. The reader's CONDITIONAL verdict already captures this; I sharpen the condition — multi-site-per-class tail bracketing with Δδ re-derived — but do not move the verdict.\n\nPer the review rule, I weighed the manuscript's self-flagged limitations as evidence: Methods' one-vacancy-per-class statement, the Discussion's nonrepresentative-tail admission, and the explicit lower-bound character of the TGA/LSFR Δδ all point to the same soft spot rather than away from it.","tokens_in":39354,"tokens_out":23553,"duration_ms":245620,"concrete_test":"Compute SCAN+U Ev for the 10 sites per composition that the fine-tuned dGNN ranks lowest (the tail that drives Δδ), plus ~5 additional sites per high-Mn 1NN class spanning the 2NN Ce/Mn distribution, in the existing SQS supercells. Re-evaluate eqn (S14) for all six compositions. If any re-sampled Δδ shifts by more than 0.01 absolute — e.g., CCTM2112 moving toward the dGNN's 0.098 or the measured 0.048–0.067 — then single-representative-per-class sampling is inadequate and Table 3, Fig. 5, and the ceria-comparison claim must be recomputed.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's screening layer — in-window Ev fractions, Fig. 5 Δδ maps, and the 'matches or exceeds ceria at 1350 °C' claim — feeds per-environment Ev values through an exponential Boltzmann average (Methods eqn (4); ESI eqn (S12)–(S14)). That average is dominated by the low-Ev tail of the distribution. But each of the 15 1NN environment classes is represented by exactly one vacancy, chosen as 'statistically typical of its class' (Methods: KDE over second-shell Ce/Mn fractions, site nearest the density maximum). A mode-selected site cannot determine a tail.\n\nThe paper's own numbers show the tail is unverified. At CCTM2112, the 15-site SCAN+U sample gives Δδ = 0.044; the dGNN — the only model that scores every O site and sees beyond 1NN — gives 0.098; measured TGA-C/LSFR values are 0.048/0.067 and are admitted lower bounds. At XCe = 0.29 the same SCAN+U pipeline overshoots (0.069 vs measured 0.044–0.047), which the Discussion explicitly leaves open as 'a nonrepresentative low-Ev tail in the 15-site DFT sample.' Tail error appears in both directions within the same pipeline, composition-dependently; the CFM, fitted to those same 90 typical sites, inherits the sampling bias. Neither the 3-composition experimental check (kinetic lower bounds) nor the dGNN (positive 0.14 eV bias; MAE 0.25 eV, worse than CFM's 0.175 eV) can adjudicate the tail.\n\nThe central mechanism claim — 1NN B-site chemistry dominates Ev (Fig. 4, 1.0–1.5 eV vs 0.2–0.6 eV) — survives: it rests on class central tendencies, which the typical-site scheme estimates reasonably, and all 15 environments are realized within each composition, cleanly separating bulk from local effects. The load-bearing weakness is the exponential tail beneath the screening payoff that the abstract foregrounds. The authors themselves flag this in the Discussion; multi-vacancy-per-class re-sampling is the cheap decisive test.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper combines SCAN+U DFT, a coverage-constrained special quasirandom structure (SQS) scheme, a linear crystal-feature model (CFM), a fine-tuned defect graph neural network (dGNN), and TGA/LSFR experiments to map oxygen-vacancy formation energies (Ev) and redox cycle capacities (Δδ) in Ca–Ce–Ti–Mn perovskites. The central mechanistic claim is that the local first-nearest-neighbor B-site Mn fraction controls Ev (1.0–1.5 eV shifts) more strongly than the local A-site Ce fraction (0.2–0.6 eV), with the six bulk compositions showing transferable local-environment trends. The screening layer identifies Ce/Mn-balanced compositions (XCe ≈0.29–0.33, XMn ≈0.58–0.67) whose predicted Δδ at 1350 °C matches or exceeds ceria, and three experimental compositions show Δδ increasing with XCe, qualitatively consistent with the models.","tokens_in":39806,"tokens_out":5421,"duration_ms":65161,"significance":"If the central mechanism claim holds, the paper provides a concrete design rule—short-range B-site cation order, not just bulk stoichiometry, can tune redox performance—and an interpretable framework transferable to other perovskite families. The coverage-constrained SQS methodology that realizes all fifteen 1NN environments is a useful technical contribution, and the public data release strengthens reproducibility. The CFM coefficients recovering a Born–Haber-like decomposition (βb ≈2.0, βr ≈−1.3) is a nice interpretability result. However, the quantitative screening claims (Fig. 5, the 'matches or exceeds ceria' benchmark in the abstract) rest on a 15-site DFT sample per composition with one vacancy per 1NN environment class, selected to be statistically typical. This sampling cannot determine the low-Ev tail that exponentially dominates the Boltzmann average underlying Δδ (Methods eqn (4)–(5); ESI S7). The paper's own numbers (dGNN 0.098 vs SCAN+U 0.044 vs measured 0.048–0.067 at CCTM2112; SCAN+U 0.069 vs measured 0.044–0.047 at XCe=0.29) show that tail errors appear in both directions and are composition-dependent. The central B-site-over-A-site hierarchy is plausible and survi","major_comments":[{"comment":"The Δδ evaluations in Fig. 5 and Table 3 are exponentially sensitive to the low-Ev tail of each composition's Ev distribution, yet each of the fifteen 1NN environment classes is represented by exactly one vacancy chosen as the KDE mode of the second-shell composition. A mode-selected site is, by design, not informative about the tail. The paper's own cross-checks demonstrate this: at CCTM2112, the SCAN+U pipeline gives Δδ=0.044 while the dGNN, which scores every O site and sees beyond 1NN, gives 0.098; measured values are 0.048–0.067. At XCe=0.29 the SCAN+U pipeline overshoots (0.069 vs 0.044–0.047 measured) and the Discussion explicitly leaves open 'a nonrepresentative low-Ev tail in the 15-site DFT sample.' The tail is therefore unverified in both directions. Because the abstract's central performance claim ('matches or exceeds ceria at 1350 °C') and the Fig. 5 maps rely on these Δδ va","section":"Methods: Supercell construction and vacancy site selection; Methods eqn (4)–(5); ESI S7"},{"comment":"The CFM is a Huber regression fitted to all 90 DFT Ev values, so the 'predictions' in Fig. 5a and the CFM column of Table 3 are in-sample interpolations, not out-of-sample predictions. The leave-one-composition-out test in ESI S3 (MAE 0.174 eV) shows generalization to new bulk compositions under the assumed linear form, but it does not validate the form itself—in particular, it cannot test whether the 1NN-averaged descriptors with a composition-independent coefficient set capture the tail behavior that drives Δδ. Similarly, the dGNN was fine-tuned on 18 of the 90 CCTM sites, so its hold-out validation on the remaining 72 is better but still shares the same DFT reference. I recommend clearly labeling the CFM and dGNN results as fitted-model interpolations within the sampled composition window, and distinguishing the mechanistic interpretation of the CFM coefficients (which is robust) from","section":"Model construction; ESI S3; Table 3"},{"comment":"The claim that the local-environment dependence is 'largely transferable across bulk compositions' is based on the tight clustering of the six bulk markers in each (xCe, xMn) cell. But each marker is the Ev of a single vacancy chosen to be the KDE mode of the second-shell environment; there is no within-class, within-composition spread to assess. The clustering could therefore reflect the selection rule rather than physical transferability. This matters because the central hierarchy—B-site 1NN chemistry dominates Ev—is quantified from these single representatives, and the dGNN distributions (ESI Fig. S8) show continuous site-to-site variation beyond the 1NN description. Please provide a within-class variance estimate (e.g., from the dGNN's all-site predictions or from additional SCAN+U vacancies for a few classes) or explicitly state that the 1.0–1.5 eV and 0.2–0.6 eV ranges are single-s","section":"Fig. 4 and caption; Results: Local cation control of vacancy energetics"}],"minor_comments":[{"comment":"The column headings 'Ev ≤3.4 eV' and 'Ev ≤3.9 eV' are overlapping and not mutually exclusive; the 'In target range' column (3.4–3.9 eV) is the informative one. Consider renaming to 'Ev <3.4 eV' and '3.4–3.9 eV' or adding an explicit note that the first two columns are cumulative.","section":"Table 2"},{"comment":"The coverage penalty k=2000 in eqn (2) is introduced without a sensitivity check. Since the paper claims coverage is obtained at no cost in the SQS objective, please report the range of k for which all fifteen environments are realized, or at least state that the result is insensitive to k above some threshold.","section":"Methods: Supercell construction and vacancy site selection"},{"comment":"The fine-tuning dataset is described as a 'randomly drawn 20%' of the 90 CCTM sites. For exact reproducibility, specify the random seed and include the list of the 18 fine-tuning sites in the data release.","section":"Model construction: Defect graph neural network model; ESI S8.2"},{"comment":"The interpretation of the CFM coefficient βb ≈2.0 as 'approximately two bonds' worth' is slightly loose, because the descriptor ⟨Eb⟩ is averaged over all six nearest-neighbor cations; a single coefficient of 2.0 on the average does not uniquely identify two bonds. Consider rephrasing to 'the fitted coefficient corresponds to a total contribution of 2×⟨Eb⟩, which the authors associate with the two B-site M–O bonds.'","section":"Results: Local cation control of vacancy energetics"}],"recommendation":"major_revision","confidential_remarks":"The paper is a solid candidate for publication if the authors address the sampling-tail issue for the Δδ predictions and reframe the screening claims accordingly. The central B-site hierarchy is well supported by the DFT data as presented, and the experimental trend is a useful qualitative check, but the current abstract overstates the quantitative match to ceria given the one-vacancy-per-environment sampling. The authors are clearly aware of the tail problem (Discussion), which makes the needed revision a matter of either additional calculations or more measured claims, not a fundamental flaw. I would not reject the manuscript; the mechanistic contribution and methodological novelty justify a major-revision opportunity."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things to know before you read it. First, the central mechanistic claim — local B-site Mn fraction controls oxygen-vacancy formation energy far more than A-site Ce — is well supported by the DFT and is a genuinely new systematic mapping. Second, the screening layer that claims ceria-level Δδ at 1350 °C is shakier than the abstract implies, because it depends on one KDE-selected vacancy per 1NN environment class, and the Boltzmann average that produces Δδ is tail-dominated.\n\nWhat the paper does well: the coverage-constrained SQS method that forces all fifteen 1NN environments into each supercell is a real methodological advance. The crystal-feature model is transparent and its coefficients (βb ≈ 2.0, βr ≈ −1.3) encode sensible Born–Haber physics. The experimental TGA/LSFR data on three compositions are genuine, independent checks, and the authors are unusually candid in the Discussion about the deviations among SCAN+U, CFM, and dGNN. Code and data are released. That is reproducible work and should count.\n\nThe soft spots, in proportion: the one-typical-site-per-class sampling cannot constrain the low-Ev tail that exponentially drives Δδ. The dGNN — which scores every oxygen site — predicts 0.098 at CCTM2112 versus 0.044 from the SCAN+U pipeline, and the measured values (0.048–0.067) sit in between, so the tail is simply not pinned down. The authors themselves flag a possible nonrepresentative low-Ev tail at XCe = 0.29. The CFM and dGNN were fitted on the same 90 sites (dGNN fine-tuned on 18), so their Δδ maps are in-sample, not out-of-sample predictions. And the \"matches or exceeds ceria\" phrasing is too strong: on a per-atom basis the measured Δδ/5 ≈ 0.010 is comparable to ceria, not clearly above it.\n\nThe important point is that these weaknesses do not sink the central claim. The B-site-over-A-site hierarchy rests on class central tendencies, which typical-site sampling estimates reasonably, and the tight clustering in Fig. 4 shows the local-environment dependence transfers across bulk compositions. This is a good paper that needs a revision.\n\nWho benefits: experimentalists and theorists working on perovskite thermochemical water splitting, and anyone building structure–defect models for disordered oxides. It deserves serious peer review — the referee should push on the tail sampling and the ceria comparison, not on the mechanism. With a few more vacancies per class (especially in low-Ev environments) and softened screening claims, this could become a solid contribution.","headline":"The B-site-dominates-Ev claim is solid and new; the Δδ screening payoff rests on an unverified low-Ev tail, so treat the ceria-beating numbers as provisional.","tokens_in":40491,"tokens_out":2239,"would_cite":true,"duration_ms":26138,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Local B-site chemistry, not bulk stoichiometry, dominates oxygen-vacancy energetics in Ca–Ce–Ti–Mn perovskites for thermochemical hydrogen production.","keywords":["perovskite","oxygen vacancy","thermochemical water splitting","hydrogen production","CCTM","special quasirandom structure","defect graph neural network","short-range cation order"],"falsifier":"Compute Ev for two oxygen vacancies that have identical (xCe, xMn) first-nearest-neighbor environments but different second-shell Ti/Mn arrangements; if the energy difference exceeds about 0.2 eV, the single-shell model underpinning all reported distributions and Δδ values fails.","tokens_in":39214,"feed_emoji":"☀️","tokens_out":7653,"duration_ms":75438,"temperature":0.7,"pith_summary":"This paper tries to establish that, in Ca–Ce–Ti–Mn perovskite oxides, the energy needed to create an oxygen vacancy is set mainly by the two B-site metal cations directly bonded to that oxygen, not by the material's overall composition. Varying the local Mn fraction among nearest-neighbor B sites shifts the vacancy formation energy by 1.0–1.5 eV depending on local Ce content, whereas changing the nearby A-site Ce fraction shifts it by only 0.2–0.6 eV. If true, short-range ordering of B-site cations becomes a practical way to tune redox behavior without changing bulk stoichiometry. The authors support this with first-principles calculations that sample all fifteen distinct local oxygen environments, an interpretable two-descriptor model whose coefficients recover the bond-breaking and electron-stabilization picture, and thermogravimetric measurements on three compositions that track the predicted trend. A sympathetic reader would care because it suggests a concrete path toward lower-temperature solar hydrogen production: compositions near a Ce/Mn-balanced region are predicted to match or exceed ceria's per-cycle oxygen capacity at 1350 °C instead of roughly 1600 °C.","feed_headline":"Nearest B-site metals set oxygen-vacancy energy in CCTM perovskites","feed_subtitle":"Swapping the two B-site metals shifts vacancy energy by 1.0–1.5 eV; balanced Ce/Mn compositions rival ceria at 1350 °C.","key_machinery":"The central object is the coverage-constrained special quasirandom structure: supercells constructed so that all fifteen symmetry-distinct oxygen first-nearest-neighbor environments—defined by the Ce fraction among four A-site neighbors and the Mn fraction among two B-site neighbors—are realized. The interpretable crystal-feature model then fits Ev to two local descriptors, the average metal–oxygen bond-dissociation energy and average crystal reduction potential over the six neighboring cations. Its fitted coefficients (approximately two bonds' worth of dissociation energy and stabilization of two released electrons) directly encode the thermochemical bond-breaking/reduction-stabilization pi","core_discovery":"The central claim is that oxygen-vacancy formation energy Ev in CCTM perovskites is controlled by the local first-nearest-neighbor cation environment—the identity of the two B-site cations (Ti vs Mn) and, secondarily, the four A-site cations (Ca vs Ce)—rather than by bulk stoichiometry alone. Increasing local Mn fraction lowers Ev by 1.0–1.5 eV depending on local Ce, while A-site Ce variation shifts Ev by only 0.2–0.6 eV. Because the B-site effect spans 2–3 times the width of the thermochemically favorable 3.4–3.9 eV window, the authors identify short-range B-site order as a candidate tuning lever. Composition-space maps then show that Ce/Mn-balanced compositions (XCe ≈ 0.29–0.33, XMn ≈ 0.58","pith_inferences":["If B-site ordering can be established, even compositions outside the identified Ce/Mn-balanced region could in principle be engineered to exceed the ideal-solution-predicted in-window fraction; this is testable by comparing ordered-motif DFT calculations with the SQS baseline.","The paper's worked example shows that using the mean Ev underestimates Δδ by an order of magnitude relative to the full environment distribution; by extension, the low-Ev tail of the site distribution governs performance, so synthesis and characterization should target tail properties, not just the average.","The disagreement between the linear crystal-feature model and the nonlinear graph-neural-network surface in mixed Ce–Mn regimes suggests some environment coupling beyond the first shell may exist; expanding the DFT dataset to intermediate XCe compositions would determine whether the ridge is physical or a sampling artifact.","If cerium is predominantly Ce3+ with electrons delocalized over many Ce 4f states rather than a discrete Ce4+→Ce3+ redox couple, then A-site Ce acts as an electron reservoir; this would change how the entropic and capacity benefits of Ce are modeled in other Ce-containing perovskites."],"forward_implications":["Short-range B-site order, if kinetically retained, could narrow the Ev distribution and raise the fraction of sites inside the 3.4–3.9 eV target window at fixed bulk composition.","Composition-space screening identifies Ce/Mn-balanced compositions whose predicted per-cycle oxygen exchange at 1350 °C matches or exceeds ceria, which requires roughly 1600 °C.","Measured Δδ on three compositions increases monotonically with Ce content under protocols close to the model conditions, confirming the qualitative predicted trend and giving an experimental anchor for the screening.","Because the fitted model encodes bond-breaking and reduction-stabilization physics, the design rule transfers: dopants with weaker metal–oxygen bonds or more favorable reduction couples should lower Ev in related perovskite families.","The dominance of B-site chemistry implies that bulk stoichiometry screening alone is insufficient; local environment sampling is necessary to predict redox performance.","The find that B-site effects dominate implies bulk optimization averages over local environments, so targeted B-site ordering is a more effective lever than composition tuning alone."],"fun_headline_variants":["Oxygen vacancies follow nearest B-site metals","Mn vs Ti at B-site swings vacancy energy up to 1.5 eV","Balanced Ce/Mn perovskite rivals ceria at 1350°C","B-site neighbors decide oxygen vacancy cost"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The whole screening pipeline assumes that the energy to create an oxygen vacancy depends only on its six immediately neighboring cations, with one representative vacancy standing in for each environment and ideal-mixing probabilities weighting them; if cations farther away also shift the energy, every reported distribution and cycle capacity is miscalibrated.","fun_headline_variants_meta":{"raw":{"variants":["Oxygen vacancies follow nearest B-site metals","Mn vs Ti at B-site swings vacancy energy up to 1.5 eV","Balanced Ce/Mn perovskite rivals ceria at 1350°C","B-site neighbors decide oxygen vacancy cost"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.002599,"raw_usage":{"total_tokens":9892,"prompt_tokens":967,"completion_tokens":8925,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":711,"completion_tokens_details":{"reasoning_tokens":8856}},"tokens_in":711,"tokens_out":8925,"duration_ms":55635,"temperature":1.0,"reasoning_tokens":8856,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-03T00:28:56.952577+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute Ev for two oxygen vacancies that have identical (xCe, xMn) first-nearest-neighbor environments but different second-shell Ti/Mn arrangements; if the energy difference exceeds about 0.2 eV, the single-shell model underpinning all reported distributions and Δδ values fails.","supporting_citations":[],"review_version":1}