{"id":"620a9d2e-e037-479c-8045-b1d17a7fe428","arxiv_id":"2607.07092","paper_version":1,"verdict":"CONDITIONAL","confidence":"UNKNOWN","novelty_score":6.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":2,"one_line_summary":"TTM-MD simulations reveal that swift heavy ion irradiation of GaN produces two distinct hillock morphologies depending on electronic energy loss, with temperature-dependent enlargement and nanochannel formation above 1200 K driven by liquid Ga fluidity.","lead":"This paper uses atomistic simulations to show how swift heavy ions create two types of surface bumps (hillocks) on gallium nitride: smooth bell-shaped ones from lower-energy ions and crater-rim ones with central holes from higher-energy ions. It also shows that high temperatures make these bumps larger and can create penetrating nanochannels, which matters for predicting GaN device failure in space and nuclear environments.","discovery_kind":"unclear","skeptic_critique":{"model":"glm-5.2","headline":"The potential's accuracy for wurtzite→zincblende energetics and GaN→Ga+N2 decomposition under non-equilibrium surface conditions is unvalidated, yet these underpin the zincblende nanodomain and Ga-rich hillock predictions.","rationale":"The reader correctly identified the potential accuracy concern as the most load-bearing issue. I sharpen it by specifying the three quantities that most directly underpin the paper's distinctive predictions (zincblende nanodomains, Ga-rich composition, nanochannel threshold) and noting that validation is deferred to a self-citation of unclear status. The qualitative two-morphology mechanism (bell vs. crater-rim, driven by Se-dependent pressure/sputtering) is likely robust regardless of potential details, since it follows from energy deposition magnitude. But the microstructural claims — zincblende nanodomains, Ga-rich composition, 1200 K nanochannel threshold — are potential-dependent and currently unvalidated against independent benchmarks. This warrants the CONDITIONAL verdict: the work is plausible and internally consistent, but the distinctive predictions need potential benchmarking before they can be accepted as physical rather than computational artifacts. The finite-size concern is secondary but also testable and should be checked alongside the potential validation.","tokens_in":9810,"tokens_out":5459,"duration_ms":223296,"concrete_test":"Compute the wurtzite–zincblende energy difference and the GaN→2Ga+N2 decomposition energy using the simulation potential and compare against DFT benchmarks. If the potential misrepresents the wurtzite–zincblende gap by more than 50% relative to the DFT value (~15–20 meV/atom), the zincblende nanodomain prediction is unreliable. Additionally, re-run the 1800 K Kr and Ta simulations in a 60×60×100 nm³ cell; if hillock diameters shrink by >15%, the temperature-dependent lateral expansion is confounded by finite-size effects.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central microstructural predictions — Ga-rich hillocks composed of liquid Ga clusters and recrystallized GaN with embedded metastable zincblende nanodomains — depend on the interatomic potential correctly representing three quantities under extreme non-equilibrium conditions: (1) the wurtzite-zincblende energy difference (DFT gives ~15–20 meV/atom with wurtzite stable), (2) GaN decomposition thermodynamics into liquid Ga + N2, and (3) liquid Ga viscosity/surface tension at thermal-spike temperatures. The main text does not specify which potential is used (deferring to supplementary materials) and presents no independent benchmark of these quantities against DFT or experiment. The zincblende nanodomain prediction is the most fragile: if the potential artificially lowers the zincblende formation energy or reduces the kinetic barrier for the wurtzite-to-zincblende transition, the predicted nanodomains at hillock peripheries — a headline result — could be artifacts. The N2 formation/escape mechanism, which underpins the Ga-rich composition claim, similarly requires a reactive potential validated for bond-breaking under thermal-spike conditions. Validation is deferred to self-cited ref 22, whose status (published vs. preprint) is unclear from the reference list. A secondary concern: at 1800 K the predicted bell-shaped hillock diameter (17.7 nm) exceeds half the 30 nm lateral cell, so periodic-image interactions could inflate the temperature-dependent lateral expansion claim.","agreement_with_reader":"agree"},"referee_report":{"model":"glm-5.2","summary":"This manuscript employs a two-temperature model coupled with molecular dynamics (TTM-MD) to investigate the atomistic mechanisms of surface hillock formation in GaN under swift heavy ion (SHI) irradiation. Two ion species are studied: 430 MeV Kr (Se = 18.2 keV/nm) and 1171 MeV Ta (Se = 40.2 keV/nm), producing bell-shaped hillocks and crater-rim hillocks with central holes, respectively. The authors report that hillocks are Ga-rich, containing liquid Ga clusters and recrystallized GaN with metastable zincblende nanodomains at their peripheries. Temperature-dependent simulations (300–1800 K) show lateral expansion of hillocks and, above 1200 K under Ta irradiation, the formation of penetrating nanochannels attributed to reduced liquid Ga viscosity. The study bridges atomistic simulations with macroscopic surface degradation, which is relevant for GaN device reliability in radiation environments.","tokens_in":10609,"tokens_out":1303,"duration_ms":108519,"significance":"The paper addresses a practically important problem: the surface damage morphology of GaN under SHI irradiation and its temperature dependence, which is relevant for aerospace and nuclear applications. The identification of two distinct hillock morphologies as a function of Se, and the prediction of zincblende nanodomain formation at hillock peripheries, are potentially valuable contributions. The temperature-dependent channel-formation threshold at 1200 K is a falsifiable prediction that could be tested experimentally. However, the significance is tempered by the absence of direct experimental validation and the reliance on self-cited prior work for key mechanistic foundations.","major_comments":[{"comment":"The interatomic potential and TTM-MD framework are not specified in the main text, and no validation of the potential's accuracy for the key quantities underlying the central predictions is provided. Specifically, the zincblende nanodomain result depends on the wurtzite-to-zincblende energy landscape, the Ga-rich hillock composition depends on GaN decomposition into liquid Ga + N2, and the nanochannel threshold depends on liquid Ga viscosity/surface tension at thermal-spike temperatures. If the potential misrepresents any of these, the headline results could be simulation artifacts. The authors should state which potential is used in the main text and provide at least a brief benchmark against DFT or experimental data for these three quantities. (See simulation setup section and Fig. 3.)","section":null},{"comment":"Reference 22 (Liang, He, et al., 'Atomistic Mechanisms of Temperature-Dependent Ion Track Formation in Gallium Nitride under Swift Heavy Ion Irradiation') is cited for the melt-recrystallization mechanism, GaN decomposition into liquid Ga + N2, and zincblende nanodomain formation along ion tracks. These are foundational to the current paper's analysis, yet the publication status of ref 22 is unclear from the reference list (no journal/volume/year given). The authors should clarify whether ref 22 is published or a preprint, and if the latter, the key supporting evidence should be summarized in the present manuscript so that the reader can independently evaluate the claims.","section":null},{"comment":"There is no quantitative comparison with experimental hillock morphologies, dimensions, or ion track radii for GaN under comparable irradiation conditions. The paper reports, for example, a bell-shaped hillock diameter of 9.6 nm and height of 2.1 nm under Kr irradiation, and a crater depth of 19.7 nm under Ta irradiation, but does not discuss whether these are consistent with available AFM/TEM data (e.g., refs 12–15). Without at least an order-of-magnitude comparison, it is difficult to assess whether the simulations capture the correct physics or whether the results are internally consistent but unvalidated. A table or paragraph comparing simulated and experimental dimensions would substantially strengthen the paper.","section":null},{"comment":"At 1800 K, the predicted bell-shaped hillock diameter is 17.7 nm (Fig. 4a), which exceeds half the 30 nm lateral cell dimension. This raises the possibility that periodic-image interactions between adjacent simulation cells could artificially inflate the lateral expansion at high temperatures. The authors should address whether finite-size effects have been tested (e.g., by repeating key simulations with a larger lateral cell) or provide justification that the 30 nm cell is sufficient at the highest temperatures studied.","section":null}],"minor_comments":[{"comment":"Page 3: The phrase 'high bias voltages' is used to motivate Joule heating in GaN devices. A brief quantitative estimate of typical junction temperatures reached under operating conditions would help the reader assess the relevance of the 300–1800 K range studied.","section":null},{"comment":"Figure captions (e.g., Fig. 1) reference color scales and variables (r, Rd) that are defined in the caption text but could be more clearly labeled on the figures themselves for quick interpretation.","section":null},{"comment":"Page 9: The statement 'hillock formation is a pressure-driven response' is somewhat general. A more quantitative discussion of the pressure magnitudes involved would strengthen the mechanistic argument.","section":null},{"comment":"References 8 and 10 appear to be identical (Singh et al., J Mater Sci: Mater Electron 2025, 36(27), 1795). One should be removed.","section":null},{"comment":"The abstract states 'electronic energy loss (Se) values' but the main text uses both 'Se' and 'electronic energy loss' interchangeably. Consistent notation throughout would improve readability.","section":null}],"recommendation":"major_revision","confidential_remarks":"The paper is a reasonable simulation study but its impact is limited by the lack of potential validation and experimental comparison. The heavy reliance on ref 22 (same author group, unclear publication status) for foundational mechanisms is a concern. If the authors can address the potential validation and provide experimental benchmarks, the paper could be suitable for publication. The finite-size concern at 1800 K should be checked."},"author_rebuttal":null,"desk_editor":{"model":"glm-5.2","letter":"This paper uses TTM-MD to simulate surface hillock formation in GaN under swift heavy ion irradiation, predicting two distinct morphologies: bell-shaped hillocks at lower Se (18.2 keV/nm Kr) and crater-rim hillocks with central holes at higher Se (40.2 keV/nm Ta). The temperature-dependent evolution up to 1800 K, including a threshold above 1200 K for penetrating nanochannel formation driven by reduced liquid Ga viscosity, is the most novel output. The pressure-driven mechanism distinguishing the two morphologies is physically plausible and internally consistent with the simulation snapshots. The Ga-rich composition claim, supported by RDF analysis showing liquid Ga and recrystallized GaN peaks, is a reasonable deduction from N2 escape dynamics. The zincblende nanodomain distribution at hillock peripheries, correlated with screw dislocations, is a genuinely new microstructural prediction. The stress-test concern about potential validation is the right one to press. The paper defers all potential details and validation to supplementary materials and self-cited ref 22, whose publication status is unclear. The zincblende prediction is the most fragile result — if the potential artificially lowers the wurtzite-to-zincblende barrier or misrepresents the energy landscape under non-equilibrium surface conditions, the nanodomain distribution could be an artifact. The N2 formation mechanism requires a reactive potential validated for bond-breaking under thermal spike conditions, and the paper does not address this. The secondary concern about periodic-image effects at 1800 K is also legitimate: the predicted bell-shaped hillock diameter of 17.7 nm at 1800 K exceeds half the 30 nm lateral cell dimension, which could inflate the lateral expansion claim. The reader's conditional verdict and scores are about right. The novelty is real — nobody has done atomistic surface hillock formation in GaN before — but the load-bearing assumption about the potential is untested in the main text. This deserves a serious referee who checks the supplementary materials for potential validation and presses the authors on finite-size effects at the highest temperatures. The core mechanism is sound enough to warrant that level of scrutiny.","headline":"Two distinct hillock morphologies in GaN under SHI irradiation, predicted via TTM-MD, with a temperature threshold for nanochannel formation — but the interatomic potential is unvalidated for the surface phase transitions it predicts.","tokens_in":10842,"tokens_out":521,"would_cite":false,"duration_ms":94524,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["61.80.Az","61.72.Ff","64.70.Nd"],"model":"glm-5.2","headline":"Ion energy loss governs nanoscale hillock shape in GaN","keywords":[],"falsifier":"If experiments irradiating GaN with Kr and Ta ions at controlled temperatures fail to observe the predicted morphological dichotomy (bell-shaped vs. crater-rim), the zincblende nanodomains at hillock peripheries, or the nanochannel penetration onset above 1200 K, the simulation predictions would be contradicted. In particular, if high-resolution TEM of irradiated GaN surfaces shows no zincblende phase or shows hillocks with significantly different composition than Ga-rich, the core mechanistic claims would be undermined.","tokens_in":10077,"feed_emoji":"🔬","tokens_out":1070,"duration_ms":202652,"temperature":0.7,"pith_summary":"This paper uses coupled two-temperature-model molecular dynamics simulations to argue that swift heavy ion irradiation of gallium nitride produces two distinct surface hillock morphologies determined by the electronic energy loss (Se) of the incident ion. At moderate Se (18.2 keV/nm, Kr ions), pressure-driven extrusion of molten material produces bell-shaped hillocks roughly 10 nm wide and 2 nm tall. At high Se (40.2 keV/nm, Ta ions), intense sputtering excavates a central cavity, leaving crater-rim hillocks with a penetrating nanochannel. The hillocks are Ga-rich because nitrogen escapes as N2 gas during the thermal spike, leaving behind liquid gallium clusters and recrystallized GaN. Rapid recrystallization at the hillock periphery traps metastable zincblende-structure GaN nanodomains, which correlate spatially with screw dislocations. Elevated ambient temperatures enlarge hillock dimensions without changing their fundamental morphology; above 1200 K, the reduced viscosity of liquid gallium allows molten material to flow out more readily, producing deep penetrating nanochannels.","feed_headline":"Ion energy loss governs nanoscale hillock shape in GaN","feed_subtitle":"Simulations reveal two distinct surface damage morphologies in irradiated gallium nitride, with temperature above 1200 K triggering deep nan","key_machinery":"The argument rests on coupled two-temperature-model molecular dynamics (TTM-MD) simulations of 430 MeV Kr and 1171 MeV Ta ions incident along [0001] in GaN, with ambient temperatures varied from 300 K to 1800 K. The key mechanistic chain is: (1) electronic energy loss sets the thermal spike intensity, (2) the spike melts GaN and decomposes it into liquid Ga plus N2 gas, (3) nitrogen escapes the near-surface region leaving Ga-rich molten material, (4) pressure gradients extrude or sputter this material to form hillocks whose shape depends on sputtering intensity, (5) rapid recrystallization from the outer edge inward traps metastable zincblende nanodomains, and (6) elevated temperature lowers","core_discovery":"The central claim is that the electronic energy loss of a swift heavy ion dictates which of two surface damage morphologies forms on GaN, with the transition governed by whether the thermal spike is strong enough to cause sustained sputtering and central excavation rather than mere extrusion. The hillock material is inherently Ga-rich due to nitrogen loss, and the rapid cooling that follows the thermal spike metastably traps zincblende GaN nanodomains at hillock edges. Temperature acts as an amplifier: it enlarges existing features and, above a threshold near 1200 K, enables liquid gallium's low viscosity to drive nanochannel penetration.","pith_inferences":[],"forward_implications":["Device engineers can use the Se threshold separating bell-shaped from crater-rim morphology to predict whether a given radiation environment produces benign protrusions or penetrating structural damage in GaN components.","The identification of zincblende nanodomains as a recrystallization byproduct suggests that swift heavy ion irradiation may locally alter the electronic band structure of GaN at damage sites, since zincblende GaN has different optoelectronic properties from the wurtzite phase.","The 1200 K nanochannel penetration threshold provides a concrete design constraint: GaN devices operating above this temperature under radiation exposure face qualitatively different failure modes involving deep material loss rather than surface protrusion.","The correlation between zincblende nanodomains and screw dislocations implies that pre-existing dislocation networks could serve as nucleation sites for phase transformation under irradiation, linking crystal quality to radiation tolerance."],"fun_headline_variants":["Swift heavy ion energy dictates GaN surface damage morphology","Electronic energy loss controls nanoscale hillock shapes in GaN","Ion irradiation creates dual nanoscale hillock morphologies in GaN","GaN hillock shape depends on swift heavy ion energy loss","High temperatures drive nanochannel formation in irradiated GaN"],"cache_read_input_tokens":0,"weakest_assumption_plain":"The simulations assume that the interatomic potential and TTM-MD framework accurately capture coupled electronic-lattice dynamics, liquid gallium rheology, and the wurtzite-to-zincblende phase transition under the extreme non-equilibrium conditions of a thermal spike at a free surface. If the potential misrepresents liquid Ga viscosity or the energy landscape between wurtzite and zincblende phases, the temperature-dependent nanochannel threshold and zincblende nanodomain分布 (d","fun_headline_variants_meta":{"raw":{"variants":["Swift heavy ion energy dictates GaN surface damage morphology","Electronic energy loss controls nanoscale hillock shapes in GaN","Ion irradiation creates dual nanoscale hillock morphologies in GaN","GaN hillock shape depends on swift heavy ion energy loss","High temperatures drive nanochannel formation in irradiated GaN","Ion energy and heat dictate nanoscale surface damage shapes in GaN","Swift heavy ion energy loss governs two GaN nanoscale damage shapes","Electronic energy loss determines GaN hillock morphology"]},"model":"glm-5.2","effort":"low","cost_usd":0.0,"raw_usage":{"total_tokens":1132,"prompt_tokens":511,"completion_tokens":621,"prompt_tokens_details":null},"tokens_in":511,"tokens_out":621,"duration_ms":24178,"temperature":1.0,"reasoning_tokens":535,"cache_read_input_tokens":0,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-09T20:07:46.635722+00:00","model_set":{"reader":"glm-5.2"},"falsifier":"If experiments irradiating GaN with Kr and Ta ions at controlled temperatures fail to observe the predicted morphological dichotomy (bell-shaped vs. crater-rim), the zincblende nanodomains at hillock peripheries, or the nanochannel penetration onset above 1200 K, the simulation predictions would be contradicted. In particular, if high-resolution TEM of irradiated GaN surfaces shows no zincblende phase or shows hillocks with significantly different composition than Ga-rich, the core mechanistic claims would be undermined.","supporting_citations":[],"review_version":1}