{"id":"ee376281-f94a-4912-ad94-5fc40ada71df","arxiv_id":"1908.03103","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"DFT simulations predict two stable, strong porous carbon-nitride monolayers, C7N6 (a direct-gap semiconductor) and C9N4 (a metal), plus an unstable third, C10N3.","lead":"Using density functional theory, the authors predict three new two-dimensional carbon-nitride materials and calculate their stability and properties. The work suggests two of them, C7N6 and C9N4, are stable and may be useful for optoelectronics and energy applications.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Stability claim is local only: without a global structure search or formation energies relative to known C-N phases, the 'prediction' of C7N6/C9N4 as stable phases remains conditional, as the paper itself concedes.","rationale":"The paper provides solid local-stability evidence: the POSCAR files are supplied, phonon dispersions show no imaginary frequencies, AIMD shows C7N6 surviving 1000 K and C9N4 surviving 500 K over 20 ps, and the mechanical and electronic calculations follow standard DFT practice. I see no internal inconsistency in these calculations, and I credit the explicit admission in Section 3 that global minimality is not established. The reader's weakest assumption correctly identifies the lack of a global structure search. My concern sharpens this: what is missing is not only a global search, but also a thermodynamic comparison against known competing phases. The paper's own Summary overstates the evidence by calling the phases 'synthesizable' without formation-energy data. This does not invalidate the local stability claims, but it does make the central 'prediction' conditional. The reader's CONDITIONAL verdict is the right category; the concern reinforces that verdict rather than moving it. The optical and excitonic issues raised by the reader are secondary, because the stability and mechanical claims are the core of the paper and are not affected by them.","tokens_in":17504,"tokens_out":9389,"duration_ms":108427,"concrete_test":"Perform a 2D structure search (e.g., USPEX, AIRSS, or active-learning methods) for the C7N6 and C9N4 stoichiometries at monolayer thickness using the same PBE functional, and compute the formation energy of the reported lattices relative to the convex hull of known C-N phases (graphene, g-C3N4, C2N, C3N, and molecular N2). If the search finds a phase lower in energy by more than roughly 50 meV/atom, or if the reported phases sit substantially above the convex hull, the synthesizability claim fails; if the reported phases remain lowest and are on or near the hull, the conditional verdict can be upgraded.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that C7N6 and C9N4 are stable, synthesizable nanosheets rests on local stability evidence: phonon spectra with no imaginary modes and short AIMD trajectories. Section 3 explicitly acknowledges that 'these observations cannot confirm that the predicted lattices are the global minimum structures.' Yet the Summary concludes that C7N6 and C9N4 'were estimated to be synthesizable,' based only on this local evidence. No global structure search for these stoichiometries is reported, and no formation energies or energy-above-hull values relative to known competing phases (graphene, g-C3N4, C2N, C3N, or N2) are given. The only energetic data are the per-atom energies -8.418, -8.464 and -8.436 eV for C7N6, C9N4 and C10N3, which are not compared with any reference phase. If a lower-energy 2D phase exists for either stoichiometry, or if these compositions are thermodynamically unstable with respect to phase separation, then the reported electronic, optical and mechanical properties would not be those of the material that could actually be synthesized. The title's 'Prediction of C7N6 and C9N4' therefore describes plausible metastable candidates, not established ground-state phases.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript predicts three porous carbon-nitride monolayer structures, C7N6, C9N4, and C10N3, built from pentagon cores bridged by N atoms. Using DFT (PBE and HSE06), DFPT phonon calculations, 20 ps AIMD simulations, and RPA optical calculations, the authors report that C7N6 and C9N4 are dynamically and thermally stable, while C10N3 is thermally unstable. They compute elastic moduli in the range 202–212 N/m, tensile strengths up to 22.4 N/m, a direct HSE06 band gap of 2.25 eV for C7N6, metallic character for C9N4 and C10N3, and optical absorption peaks in the visible for C7N6 and infrared for the metallic sheets. The paper concludes that C7N6 and C9N4 are synthesizable candidates with attractive electronic and optical properties and provides the optimized structures in POSCAR format in the supplementary information.","tokens_in":17753,"tokens_out":5314,"duration_ms":50805,"significance":"The paper proposes new carbon-nitride nanosheet stoichiometries and characterizes them with standard first-principles methods. If the structures are the relevant phases, the results are of interest because they identify a direct-gap semiconductor (C7N6 with a HSE06 gap of 2.25 eV) alongside metallic carbon nitrides, which are uncommon, and the mechanical properties are competitive with known porous carbon nitrides. The inclusion of the optimized coordinates in the supplementary information is a concrete asset that enables independent verification. The electronic-structure and mechanical results are internally consistent and are not fitted to any target property, so circularity is not a concern. The main weaknesses are that the stability evidence is local only and that the optical claims rest on RPA without excitonic corrections.","major_comments":[{"comment":"The conclusion that C7N6 and C9N4 are 'synthesizable' is not supported by the evidence presented. The paper demonstrates only local stability through phonon dispersions and short AIMD trajectories, and it explicitly concedes in §3 that 'these observations cannot confirm that the predicted lattices are the global minimum structures.' No formation energies relative to competing phases (e.g., graphene, g-C3N4, C2N, C3N, and N2) and no energy-above-hull analysis are reported; the per-atom energies of -8.418, -8.464, and -8.436 eV are quoted without any reference values. To support the title's claim of 'Prediction' and the Summary's 'synthesizable', the authors should add thermodynamic stability data or restrict the claim to locally stable, metastable candidates.","section":"§3 (Stability discussion) and §4 (Summary)"},{"comment":"The optical absorption and conductivity results are obtained with RPA using PBE eigenvalues, without excitonic (BSE) or self-energy (GW) corrections. For C7N6, a semiconductor with a 2.25 eV HSE06 gap, excitonic effects are expected to shift the absorption edge and modify peak intensities, which directly affects the claim that its absorption coefficient and optical conductivity exceed those of graphene in the visible range. The authors should either perform BSE calculations for C7N6 or moderate the comparative claims to reflect the RPA level of theory.","section":"§3 (Optical properties, Figs. 5–7)"},{"comment":"The thermal-stability distinction between C9N4 (stable at 500 K) and C10N3 (unstable) rests on single 20 ps AIMD trajectories on 2×2×1 supercells. While such short runs are common in the literature, the paper uses this evidence to declare C10N3 outside the set of 'synthesizable' materials. The authors should either extend the AIMD simulations or explicitly acknowledge the sensitivity of the stability assignment to the short simulation time and small cell size.","section":"§2 (AIMD) and §3 (Thermal stability)"}],"minor_comments":[{"comment":"In the sentence describing the metallic systems, 'C10N13' should be 'C10N3'.","section":"§2 (Methods, after Eq. (4))"},{"comment":"The sentence about the C-C bond length in the 12-membered rings compares it to 'the corresponding C-N bond in the C7N6, 1.464 Å'; this is unclear because C7N6 is not described as having 12-membered rings, and the comparison should be spelled out explicitly.","section":"§3 (Structural analysis)"},{"comment":"The caption of Fig. 7 says 'comparison of optical absorption spectra' in the inset, but the plotted quantity is the real part of the optical conductivity, Re σαβ; please harmonize the terminology.","section":"§3 (Fig. 7 inset)"},{"comment":"Reference [48] is a cross-sectional study of sexual desire in married women and is unrelated to the topic of two-dimensional materials; this appears to be a citation error and should be replaced or removed.","section":"References (Ref. [48])"},{"comment":"Equation (1) is garbled in the typeset text, with the integrand and summation symbols not legible; it should be re-typeset correctly.","section":"Eq. (1)"}],"recommendation":"major_revision","confidential_remarks":"The manuscript fits the scope of Computational Materials Science, and the central ideas are worth publishing after revision. The main issue is that the 'prediction'/'synthesizable' framing overstates the thermodynamic evidence; the authors should either add formation-energy analysis or explicitly reframe the claims as concerning locally stable metastable structures. The unrelated reference [48] should be corrected, as it may indicate careless citation handling."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"What you should know: this is a routine but solid DFT screening paper that introduces three new carbon-nitride stoichiometries, C7N6, C9N4, and C10N3. Two of them (C7N6 and C9N4) survive phonon and AIMD checks, and one of those, C7N6, is a direct-gap semiconductor at 2.25 eV (HSE06). That is a real addition to the 2D carbon-nitride zoo, which has been dominated by C3N4, C2N, C3N, and close relatives.\n\nThe paper does several things well. The structural data are in the SI as POSCAR files, which makes the work reproducible. Phonon dispersions have no imaginary modes, and the AIMD tests distinguish stable from unstable compositions rather than pretending all three are good. The mechanical characterization is complete enough: elastic moduli around 200 N/m and tensile strengths comparable to or better than other porous C-N sheets. The authors also deserve credit for stating plainly in Section 3 that these lattices are not confirmed to be global minima. That is the key limitation, and they put it on the record.\n\nThe soft spots are real but proportionate. The biggest one is exactly what the stress-test note says: no global structure search and no formation energies or convex-hull context relative to known competing phases. So the title's \"prediction\" should be read as \"prediction of a plausible metastable candidate,\" not \"prediction of the ground state.\" The abstract and summary oversell this a little by calling the sheets \"synthesizable.\" That is a word I would soften. Second, the optical comparison against graphene uses RPA without excitonic effects; since graphene is treated at the same level it is a fair relative benchmark, but the claim of \"larger than graphene\" should explicitly carry that caveat. Third, and I almost missed it: reference [48] is a medical sexology paper. That is a clear citation-mixing error and should be fixed. The 20 ps, 2x2 AIMD runs are on the light side but within the field's normal practice.\n\nOverall, the central stability claim holds at the local level, and the authors did not hide the global-minimum caveat. This is not a game-changing paper, but it is honest, useful, and citable work for people hunting new 2D semiconductors and metals. I would send it to peer review, with the expectation that the authors tone down the synthesis language, add any available energetic comparison to competing phases, and clean up the references. A careful referee can make this better without needing to sink it.","headline":"Two plausible porous C-N monolayers, locally stable by phonon/AIMD, with the authors openly conceding they are not proven ground states; solid screening work, slightly oversold in the conclusion, and one citation is botched.","tokens_in":18312,"tokens_out":1836,"would_cite":false,"duration_ms":21479,"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":"C7N6 and C9N4 are predicted to be stable, strong carbon-nitride nanosheets.","keywords":["two-dimensional materials","carbon nitride nanosheets","density functional theory","porous lattices","direct band gap","metallic nanosheet","optical absorption","mechanical properties"],"falsifier":"Run a crystal-structure search at the C7N6 and C9N4 stoichiometries; if any sampled phase has lower energy per atom than the reported values of -8.418 and -8.464 eV, the predicted properties belong to a metastable arrangement rather than the ground state. Experimentally, synthesizing the sheets and measuring the optical gap would directly test the predicted 2.25 eV direct-gap character of C7N6.","tokens_in":17316,"feed_emoji":"🧪","tokens_out":6326,"duration_ms":60968,"temperature":0.7,"pith_summary":"This paper predicts three porous carbon-nitride monolayers, C7N6, C9N4, and C10N3, built from five-membered carbon-nitrogen rings linked by single nitrogen atoms. It argues that C7N6 and C9N4 are the stable, synthesizable members: their phonon dispersions contain no imaginary frequencies, and molecular dynamics shows C7N6 surviving to 1000 K and C9N4 to 500 K, while C10N3 disintegrates at all tested temperatures. The central result is electronic contrast: C7N6 is a direct-gap semiconductor with a 2.25 eV HSE06 gap, whereas C9N4 is metallic, an unusual character for carbon-nitride frameworks. The paper also reports elastic moduli near 200 N/m, tensile strengths among the highest for porous carbon-nitride sheets, and visible or infrared absorption with coefficients exceeding graphene's in parts of the visible spectrum. A sympathetic reader would take these predictions as evidence that C7N6 and C9N4 merit experimental attention for optoelectronic and energy applications.","feed_headline":"C7N6 and C9N4 nanosheets predicted stable and strong","feed_subtitle":"DFT predicts a 2.25 eV direct-gap semiconductor and a metallic sheet, with visible absorption beyond graphene.","key_machinery":"The load-bearing construction is a repeating motif of three pentagon cores joined by single nitrogen atoms, forming porous hexagonal sheets with alternating 12-membered and 9-membered rings. The short C-N bridge bonds, about 1.29 angstroms, carry much of the lattice rigidity, and electron localization function maps confirm the covalent bonding that explains the high elastic moduli; failure analysis identifies the edge bonds of the pentagon cores as the first to rupture. Computationally, the argument rests on phonon dispersions from density-functional perturbation theory, showing no imaginary frequencies, 20-picosecond ab initio molecular dynamics trajectories for thermal stability, the screened hybrid functional HSE06 for the band gap, and random-phase-approximation optical spectra with intraband contributions added for the metallic cases.","core_discovery":"The central claim is that two previously unstudied stoichiometries of porous carbon-nitride monolayers are viable two-dimensional materials: C7N6 is a direct-band-gap semiconductor with a 2.25 eV gap at the HSE06 level, and C9N4 is a metal, even though most carbon-nitride nanosheets are semiconductors. Both are dynamically stable, C7N6 remains intact in ab initio molecular dynamics at 1000 K and C9N4 at 500 K, and both combine elastic moduli of 212 and 202 N/m with maximum tensile strengths of 14.1 and 22.4 N/m, respectively. The structures are hexagonal lattices made of three pentagon cores connected by single nitrogen atoms, producing pores bounded by 12-membered and 9-membered rings. C7N6's first absorption peak lies in the visible range and C9N4's in the infrared, and in the 370-500 nm window both show larger absorption coefficients and optical conductivities than graphene. The paper explicitly notes that these observations do not confirm the predicted lattices are the global minimum structures.","pith_inferences":["Because the paper's lattices are hand-constructed rather than found by global search, a more stable phase at the same stoichiometry could supersede the reported properties; an evolutionary or random crystal-structure search is the direct test.","The metallic, porous C9N4 lattice is a natural candidate for alkali-ion storage or electrocatalysis, but the paper does not evaluate electrochemistry; computing ion adsorption energies would be a concrete extension.","The optical spectra are computed at the random-phase-approximation level without excitonic corrections, so a GW-BSE treatment could shift absorption edges and would determine whether the visible-light advantage over graphene survives many-body effects.","If the metal-semiconductor contrast between C9N4 and C7N6 can be realized in a single junction, the interface may exhibit built-in charge separation; the paper only hints at this possibility without exploring transport."],"forward_implications":["C7N6 offers a direct band gap of 2.25 eV, placing it in the visible-light range and making it a candidate for optoelectronic and photovoltaic components.","C9N4 provides a porous metallic carbon-nitride membrane, a combination that could be useful as a conductive catalyst support or electrode material.","Both sheets show larger absorption coefficients and optical conductivities than graphene across the violet-to-green part of the visible spectrum, which is directly relevant to light-harvesting devices.","With elastic moduli around 202-212 N/m and tensile strengths up to 22.4 N/m, the predicted sheets rank among the strongest known porous carbon-nitride monolayers, suitable for flexible and load-bearing nanodevices.","The near coincidence of the C7N6 and C9N4 lattice constants, differing by about one percent, makes lateral or planar heterostructures between a semiconductor and a metal geometrically feasible."],"supporting_citations":[{"why":"Supplies the graphene baseline whose zero band gap motivates the search for semiconducting two-dimensional materials.","marker":"[1]"},{"why":"Documents the experimentally realized triazine-based graphitic carbon nitride that the new porous sheets extend.","marker":"[24]"},{"why":"Provides the wet-chemical synthesis precedent for nitrogenated holey two-dimensional C2N.","marker":"[25]"},{"why":"Defines the PBE exchange-correlation functional used for structural, mechanical, and band-structure calculations.","marker":"[52]"},{"why":"The screened hybrid functional used to establish the 2.25 eV gap of C7N6.","marker":"[55]"},{"why":"The code used with density-functional perturbation theory to produce the phonon dispersions that establish dynamical stability.","marker":"[56]"},{"why":"The all-electron code used to compute the random-phase-approximation optical spectra and dielectric functions.","marker":"[57]"},{"why":"Supplies graphene's elastic modulus and ideal strength values of 350.7 and 40.4 N/m used as dense-lattice benchmarks.","marker":"[68]"},{"why":"Provides elastic moduli and tensile strengths of triazine-based graphitic carbon nitride and C2N that bracket the new sheets' porous-lattice performance.","marker":"[70]"},{"why":"Supplies the computed graphene optical absorption curve used as the comparison in the visible-range insets.","marker":"[77]"}],"fun_headline_variants":["C7N6 semiconductor and C9N4 metal: predicted 2D sheets","Strong porous carbon-nitrides: gapped and metallic","Visible absorption from C7N6, IR from C9N4: predicted","New C-N sheets: high strength, direct gap, metal","C7N6, C9N4: stable porous sheets with unique electronics"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the hand-built pentagon-core lattices are the right structural candidates for these stoichiometries; the paper itself says the calculations cannot confirm these are the global minimum structures, so a more stable phase could replace them and change every predicted property.","fun_headline_variants_meta":{"raw":{"variants":["C7N6 semiconductor and C9N4 metal: predicted 2D sheets","Strong porous carbon-nitrides: gapped and metallic","Visible absorption from C7N6, IR from C9N4: predicted","New C-N sheets: high strength, direct gap, metal","C7N6, C9N4: stable porous sheets with unique electronics"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000591,"raw_usage":{"total_tokens":2832,"prompt_tokens":1069,"completion_tokens":1763,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":685,"completion_tokens_details":{"reasoning_tokens":1666}},"tokens_in":685,"tokens_out":1763,"duration_ms":15160,"temperature":1.0,"reasoning_tokens":1666,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T14:23:41.833143+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run a crystal-structure search at the C7N6 and C9N4 stoichiometries; if any sampled phase has lower energy per atom than the reported values of -8.418 and -8.464 eV, the predicted properties belong to a metastable arrangement rather than the ground state. Experimentally, synthesizing the sheets and measuring the optical gap would directly test the predicted 2.25 eV direct-gap character of C7N6.","supporting_citations":[{"cited_title":"Mahmood, E.K","cited_arxiv_id":null,"evidence_quote":"Provides the wet-chemical synthesis precedent for nitrogenated holey two-dimensional C2N."},{"cited_title":"Schwarz, P","cited_arxiv_id":null,"evidence_quote":"The all-electron code used to compute the random-phase-approximation optical spectra and dielectric functions."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies graphene's elastic modulus and ideal strength values of 350.7 and 40.4 N/m used as dense-lattice benchmarks."},{"cited_title":"Rahaman, B","cited_arxiv_id":null,"evidence_quote":"Provides elastic moduli and tensile strengths of triazine-based graphitic carbon nitride and C2N that bracket the new sheets' porous-lattice performance."},{"cited_title":"Shahrokhi, C","cited_arxiv_id":null,"evidence_quote":"Supplies the computed graphene optical absorption curve used as the comparison in the visible-range insets."}],"review_version":1}