{"id":"1257b3f1-f9c0-424a-b5f2-bccd3791a0a5","arxiv_id":"2509.02789","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Low-angle grain boundaries can decompose into two boundaries separated by a new grain when the applied stress moves one dislocation type but not the other.","lead":"Simulations show that a single low-angle grain boundary in nickel can split into two separate boundaries with a new grain growing between them when stress acts on only one of its dislocation types. The finding suggests a reverse path to the usual grain-merging process, which could give materials scientists a new tool for patterning microstructures.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Internal contradiction in strain-rate/temperature dependence of mixed-GB decomposition undermines the reported parameter-space analysis and mechanism.","rationale":"The central claim—that a low-angle GB can decompose into two independent GBs with a new grain under differential Peach-Koehler forces—is directly supported by MD simulations for two distinct GBs. The most load-bearing concern is not the immobility assumption (which the simulations appear to uphold) but the internal contradiction in the reported strain-rate/temperature dependence of mixed-GB decomposition. This contradiction is located in the main text and affects the interpretation of the parameter-space study, which is a key part of the paper's support for the separability condition. If the data reporting is erroneous, the mechanism (node-reaction reversal) and the boundaries of the phenomenon are unreliable. However, the existence of decomposition at 100 K is documented, so the central claim is not invalidated. The reader also flagged this contradiction in the rationale, though the stated weakest assumption was different. Therefore, the appropriate verdict remains CONDITIONAL: the central claim is plausible and partially demonstrated, but the inconsistency and lack of data sharing prevent full acceptance.","tokens_in":8546,"tokens_out":10043,"duration_ms":102369,"concrete_test":"Re-examine the raw simulation outputs for the mixed GB at 300 K for each of the six strain rates (1×10^7 to 5×10^8 s^-1). For each case, classify the outcome as decomposition (screw network fully separates) or partial mobility (remains pinned), and record the node structure (double vs single loop) at the same strain as in Fig. 4b1/S2. If the 300 K low-strain-rate cases show single-loop nodes and no separation, then the statement 'decomposition is observed at low strain rates' is a typo, and the conclusions should be revised to 'decomposition only at low temperature/high strain rate,' consistent with the hypothesis. If those cases actually decompose despite single-loop nodes, the proposed mechanism is falsified and the temperature/strain-rate analysis requires substantial rework.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"In the Results section 'Strain rate and temperature effects', the text states two mutually exclusive claims about the low-angle mixed GB: first, 'decomposition is only observed at low temperatures and high strain rates', and then, two sentences later, 'At 300 K, decomposition is observed at low strain rates but not at high strain rates.' This is a direct logical contradiction. Moreover, the following hypothesis says a slower strain rate provides more time for the double-loop node to evolve into the single-loop node, which is explicitly described as 'unable to break away.' Thus the proposed mechanism predicts decomposition should be suppressed at low strain rates, not observed there. Either the reported data or the mechanistic explanation is wrong. Because the paper uses this temperature/strain-rate dependence to support the role of node reactions in separability (conclusion 2), the inconsistency makes the conditions under which decomposition occurs indeterminate and calls into question the interpretation of Figure 4b and Supplementary Figures S2–S4.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports atomistic simulations demonstrating that a low-angle grain boundary can decompose into two separate grain boundaries when the applied stress state exerts sufficiently different Peach-Koehler forces on the different dislocation types composing the boundary. Two cases are studied in Ni with an EAM potential: a [001] low-angle asymmetric tilt GB (two edge dislocation types) and a (001) low-angle mixed tilt-twist GB (an edge dislocation array plus a screw dislocation network). In both cases, the mobile dislocation component separates from the stationary one, leaving a new grain between two GBs. The paper also examines strain-rate and temperature effects, identifies double-loop versus single-loop node reaction products at intersections in the mixed GB, and reports an apparent stress barrier for separating the twist component. Three conditions for GB decomposition are proposed.","tokens_in":8734,"tokens_out":9584,"duration_ms":99498,"significance":"If correct, the claimed phenomenon is a genuinely new elementary process in grain-boundary microstructure evolution: a single boundary splitting into two boundaries with a new grain between them, i.e., the inverse of GB coalescence. The central evidence is direct MD visualization, and the Peach-Koehler model is used as a prior design principle rather than fitted to the simulations, which is a strength. The three-condition framework is concrete and falsifiable, and the stress-barrier quantification is a useful target for future work. However, the parameter-space interpretation is currently compromised by a direct internal contradiction in the temperature/strain-rate narrative and by the definition of 'critical RSS'; these issues must be resolved before the conclusions can be accepted.","major_comments":[{"comment":"This section contains a direct contradiction. It first states 'For the low-angle mixed GB, decomposition is only observed at low temperatures and high strain rates,' and two paragraphs later states 'At 300 K, decomposition is observed at low strain rates but not at high strain rates.' The proposed mechanism then argues that at 300 K the slower strain rate provides more time for the double-loop node to evolve into the single-loop node, which is described as 'unable to break away.' That mechanism predicts suppression, not decomposition, at low strain rates. Because this passage underpins the separability discussion and conclusion (2), and is used to interpret Figure 4b and Supplementary Figures S2–S4, the reported parameter-space behavior is indeterminate as written. The text and the data/figure must be reconciled; if the intended statement is 'decomposition at low temperatures and high st","section":"Strain rate and temperature effects"},{"comment":"For the mixed GB, the authors define the critical RSS as 'the highest RSS achieved during each simulation,' considered as the critical RSS to activate decomposition or partial mobility. This conflates a peak stress with a threshold. For decomposition, the meaningful quantity is the RSS at the moment the screw network breaks away, which may be lower than the later peak if stress continues to increase; for non-decomposing cases, there is no single critical event and the highest RSS is just an endpoint value. This affects the interpretation of Figure 4b and the quantitative barrier of 134 ± 10 MPa in Figure 4d. Please report the RSS at the onset of separation (or justify that the maximum coincides with it) and specify a criterion for partial mobility.","section":"Strain rate and temperature effects / Figure 4b-d"}],"minor_comments":[{"comment":"The displayed formula for RSS is garbled in the provided text. Please define all symbols (unit Burgers vector, slip-plane normal, summation convention) and correct the typography so that Eq. (2) is unambiguous.","section":"Methodology, Eq. (2)"},{"comment":"The phrase 'At higher temperatures and lower simulation times (slower strain rates)' is internally inconsistent: lower simulation times correspond to higher strain rates, not slower ones. This typo should be corrected to avoid compounding the confusion in the Results section.","section":"Conclusions"},{"comment":"The text refers to 'the stress-strain curve in Figure 2b,' but the caption lists (b) as 'Velocities' and (c) as 'RSS-strain/time curve.' Please align the references with the caption.","section":"Figure 2 caption"},{"comment":"The caption states 'at 100 K and 107 strain rate'; the exponent appears to be missing and should read 10^7 s^-1.","section":"Figure 3 caption"},{"comment":"The reversibility of the asymmetric-tilt decomposition is asserted from the smooth stress-strain curve, but no unloading or reverse-loading simulation is shown. Either add such a test or soften the reversibility claim.","section":"Asymmetric tilt GB / reversibility"},{"comment":"The three proposed conditions are supported by only two low-angle GBs in one EAM Ni potential. A sentence noting that node reactions and core energetics may differ for other potentials or materials would help calibrate the scope of the claim.","section":"Generality"}],"recommendation":"major_revision","confidential_remarks":"The core two-case demonstration appears sound and worth publishing after revision. The main blocker is the direct logical contradiction in the temperature/strain-rate narrative; I suspect it is a typo, but as submitted the text, Figure 4b, and Supplementary S2–S4 cannot be interpreted consistently. The 'critical RSS equals highest RSS' definition also needs revisiting because it bears directly on the reported stress barrier. These are fixable issues, hence major revision rather than rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"You should know two things. First, the core simulation result is genuinely new: the authors design stress states from the Peach-Koehler model and show in MD that a low-angle asymmetric tilt GB and a low-angle mixed tilt-twist GB each split into two boundaries with a new grain between them. That goes beyond the dissociation reported in refs 21,22, where the boundaries do not separate into independent GBs. The simulations are direct, the DXA analysis is credible, and the P-K model is used a priori, not fitted to the MD output. For these two specific GBs, the demonstration is solid.\n\nSecond, the temperature/strain-rate analysis contains a direct self-contradiction that strikes at their mechanism. The text says mixed-GB decomposition is 'only observed at low temperatures and high strain rates,' then two sentences later says 'At 300 K, decomposition is observed at low strain rates but not at high strain rates.' Both cannot be true. The hypothesis that follows—slower strain rates give the double-loop node time to evolve into the single-loop node, which is 'unable to break away'—predicts the opposite of the second sentence. So either the reported data or the proposed mechanism is wrong. The conclusion's parenthetical 'lower simulation times (slower strain rates)' has the causality backwards as well. Since conclusion 2 about separability rests on this parameter-space map, a referee should require the contradiction to be resolved.\n\nTwo smaller concerns. One, generality rests on a single EAM potential and two geometries, with no shared code or data. That keeps the three conditions as a plausible but unproven generalization. Two, the critical RSS for the mixed GB is defined as the highest RSS reached in each simulation, which is a measured outcome rather than a physically determined threshold. The asymmetric tilt GB definition at the onset of migration is cleaner.\n\nThis paper is for people who care about grain boundary migration mechanisms and dislocation patterning. The central phenomenon is worth taking seriously, and the two simulations are evidence for its existence. But the strain-rate/temperature narrative needs rewriting, and the data should be available if the generality claims are to be checked.\n\nRecommendation: send to peer review—it deserves a serious referee. I'd also bring it to a reading group; the contradiction is a useful case study in how easily a simulation-based narrative can slip.","headline":"Core result is a real MD demonstration, but the mixed-GB temperature/strain-rate story is self-contradictory and needs rewriting.","tokens_in":9185,"tokens_out":5587,"would_cite":false,"duration_ms":56606,"reading_group":"yes","serious_thinker":"no","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"A grain boundary can split into two boundaries with a new grain between them when stress drives one dislocation type away from another.","keywords":["grain boundary decomposition","Peach-Koehler force","low-angle grain boundary","dislocation separation","molecular dynamics simulation","mixed tilt-twist grain boundary","FCC nickel","grain boundary reversibility"],"falsifier":"In-situ transmission electron microscopy of a nickel bicrystal with an 8.8 degree asymmetric tilt boundary loaded in the direction that should move only one edge dislocation array: if the two arrays do not separate into two boundaries with a new grain between them, or if the supposedly stationary array also glides, the decomposition claim is falsified for that regime.","tokens_in":1622,"feed_emoji":"🧊","tokens_out":2862,"duration_ms":124892,"temperature":0.7,"pith_summary":"This paper argues that grain boundaries are not only capable of merging; under the right stress, a single low-angle boundary can decompose into two independent boundaries separated by a newly formed grain. The effect is traced to the Peach-Koehler force, which lets a chosen stress state push one dislocation type while leaving another localized. The authors demonstrate this by atomistic simulation in two nickel bicrystals, an asymmetric tilt boundary and a mixed tilt-twist boundary. They identify three requirements: two different Burgers vectors, differential Peach-Koehler forces, and dislocation separability. The result matters because it suggests a mechanical handle on dislocation patterns, with possible uses in film patterning and defect management.","feed_headline":"Stress can split a grain boundary in two","feed_subtitle":"Simulations show low-angle boundaries decomposing when differently oriented dislocations feel different forces.","key_machinery":"The central object is the Peach-Koehler force F = (sigma dot b) cross l acting on a dislocation with Burgers vector b and line direction l. By designing a stress state sigma such that this force is nonzero for one dislocation type and zero for another, the authors make only one array glide. The paper's three stated conditions for decomposition at least two Burgers vectors, differential forces, and dislocation separability determine when this selective glide results in boundary splitting. In the mixed boundary, separability is governed by reactions at tilt-twist intersection nodes, where the two dislocation types form loops and can become sessile; this node behavior is what makes the process","core_discovery":"The central claim is that low-angle grain boundaries can decompose when their dislocation content responds non-uniformly to an applied stress. In a simulated [001] asymmetric tilt boundary, a 1/2[110] edge array glides away from a stationary [010] edge array, leaving two tilt boundaries with a new grain between them. In a mixed tilt-twist boundary, a [001] edge array stays pinned while a 1/2[110] screw network bows, then breaks away after overcoming a stress barrier, yielding separate tilt and twist boundaries. The paper gives three conditions for decomposition: the boundary has at least two different Burgers vectors, the stress state produces sufficiently different Peach-Koehler forces, and","pith_inferences":["Editorial inference: because the three criteria are stated in terms of local stress and dislocation character, the same decomposition should be realizable in other easy-glide metals and alloys, not only nickel; the paper's material choice is presented as a modeling convenience.","Editorial inference: the stress barrier seen in the mixed boundary suggests that engineering the intersection nodes through alloying, precipitate pinning, or temperature could tune whether a boundary decomposes or merely bows, making the effect a controllable processing variable.","Editorial inference: a direct extension would be unloading-reloading cycles on the asymmetric tilt case to test whether the two boundaries re-merge reversibly, which would turn decomposition into a mechanically driven switch for dislocation arrangements."],"forward_implications":["If the three criteria hold, any low-angle grain boundary with at least two separable dislocation types should be inducible to decompose, not just the two simulated cases.","For asymmetric tilt boundaries, decomposition occurs at every temperature and strain rate tested, with the critical stress rising with strain rate and falling with temperature, consistent with a thermally activated process.","For mixed tilt-twist boundaries, decomposition occurs only at low temperatures and high strain rates; otherwise the screw network bows out but remains pinned by edge dislocations, producing partially mobile boundaries.","The extra stress needed to separate the twist component from a mixed boundary is about 134 +/- 10 MPa in this model and is nearly independent of strain rate.","Decomposition offers a mechanical route to control dislocation arrays in thin films and could be used to sweep dislocation arrays through grains to gather defects or impurities."],"supporting_citations":[{"why":"Supplies the force equation F=(sigma dot b) cross l that predicts differential dislocation motion under stress.","marker":"[19]"},{"why":"Provides the dislocation theory background: Burgers vectors, resolved shear stress, and sessile dislocation types used in the node analysis.","marker":"[20]"},{"why":"Describes the previously reported grain-boundary dissociation process that this work distinguishes from decomposition.","marker":"[21]"},{"why":"Prior study of stress-driven migration of low-angle mixed grain boundaries, giving context for the mixed-boundary result.","marker":"[30]"},{"why":"Supplies the adaptive common neighbor analysis used to identify grain-boundary atoms and track boundary positions in the simulations.","marker":"[31]"},{"why":"Establishes nodal effects in dislocation mobility, the basis for pinning at tilt-twist intersection nodes.","marker":"[32]"},{"why":"Extends understanding of dislocation node mobility and is used to interpret screw dislocations breaking away from edge dislocations.","marker":"[33]"},{"why":"Supplies the dislocation extraction algorithm used to identify Burgers vectors and line directions in the simulated grain boundaries.","marker":"[35]"},{"why":"Supplies the embedded-atom nickel interatomic potential used for all molecular dynamics simulations.","marker":"[45]"},{"why":"Supplies the sampling method used to generate the grain-boundary structures under zero temperature and pressure.","marker":"[47]"}],"fun_headline_variants":["Stress splits grain boundaries in two","Differential forces disassemble grain boundaries","Simulations show boundaries breaking into two","When dislocations diverge, boundaries decompose"],"cache_read_input_tokens":11136,"weakest_assumption_plain":"The load-bearing premise is that the imposed constant-strain loading really does keep one dislocation type immobile while the other glides; if local stress, lattice rotation, or the interatomic potential lets the supposedly stationary dislocations move or pins the mobile ones differently, selective separation will not occur.","fun_headline_variants_meta":{"raw":{"variants":["Stress splits grain boundaries in two","Differential forces disassemble grain boundaries","Simulations show boundaries breaking into two","When dislocations diverge, boundaries decompose"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000427,"raw_usage":{"total_tokens":1964,"prompt_tokens":627,"completion_tokens":1337,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":371,"completion_tokens_details":{"reasoning_tokens":1297}},"tokens_in":371,"tokens_out":1337,"duration_ms":14722,"temperature":1.0,"reasoning_tokens":1297,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-05T11:23:12.210282+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"In-situ transmission electron microscopy of a nickel bicrystal with an 8.8 degree asymmetric tilt boundary loaded in the direction that should move only one edge dislocation array: if the two arrays do not separate into two boundaries with a new grain between them, or if the supposedly stationary array also glides, the decomposition claim is falsified for that regime.","supporting_citations":[{"cited_title":"The forces exerted on dislocations and the stress fields produced by them","cited_arxiv_id":null,"evidence_quote":"Supplies the force equation F=(sigma dot b) cross l that predicts differential dislocation motion under stress."},{"cited_title":"M.; Cohen, D","cited_arxiv_id":null,"evidence_quote":"Describes the previously reported grain-boundary dissociation process that this work distinguishes from decomposition."},{"cited_title":"Stress-driven migration of simple low-angle mixed grain boundaries","cited_arxiv_id":null,"evidence_quote":"Prior study of stress-driven migration of low-angle mixed grain boundaries, giving context for the mixed-boundary result."},{"cited_title":"Structure identification methods for atomistic simulations of crystalline materials","cited_arxiv_id":null,"evidence_quote":"Supplies the adaptive common neighbor analysis used to identify grain-boundary atoms and track boundary positions in the simulations."},{"cited_title":"Nodal effects in dislocation mobility","cited_arxiv_id":null,"evidence_quote":"Establishes nodal effects in dislocation mobility, the basis for pinning at tilt-twist intersection nodes."},{"cited_title":"Enhanced mobility of dislocation network nodes and its effect on dislocation multiplication and strain hardening","cited_arxiv_id":null,"evidence_quote":"Extends understanding of dislocation node mobility and is used to interpret screw dislocations breaking away from edge dislocations."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the dislocation extraction algorithm used to identify Burgers vectors and line directions in the simulated grain boundaries."},{"cited_title":"Computation of grain boundary stiffness and mobility from boundary fluctuations","cited_arxiv_id":null,"evidence_quote":"Supplies the embedded-atom nickel interatomic potential used for all molecular dynamics simulations."},{"cited_title":"Foiles, S.M., Holm, E.A","cited_arxiv_id":null,"evidence_quote":"Supplies the sampling method used to generate the grain-boundary structures under zero temperature and pressure."}],"review_version":1}