{"id":"38245da9-13e3-43ff-b352-6ea11532b129","arxiv_id":"2411.16520","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"A computed SiC-biphenylene monolayer is predicted to be a stable direct-gap semiconductor with a 2.89 eV quasi-particle gap and a 0.82 eV exciton binding energy.","lead":"This paper predicts a new two-dimensional silicon carbide phase, SiC-biphenylene, with a direct band gap of 2.89 eV and a strongly bound exciton at 2.07 eV. It uses standard first-principles methods (DFT, G0W0, BSE) to claim the material is dynamically, thermally, and mechanically stable.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Monolayer stability rests on unverified dismissal of imaginary Γ-point acoustic phonons; no supercell or q-mesh convergence is shown.","rationale":"The reader identified the same weakest assumption: the dynamic stability claim depends on dismissing slight imaginary acoustic phonon frequencies at Γ as numerical artifacts. The paper explicitly acknowledges these imaginary frequencies (Sec. IIIA) but offers no supercell-size convergence or q-mesh convergence data to support the dismissal. The Methods section specifies only a 9×9×1 electronic k-grid for the DFPT calculation, not the supercell dimensions or the q-mesh used for force constants. In 2D materials, long-wavelength acoustic modes are the most sensitive to finite-size effects, but they are also the modes that determine dynamic stability. If these imaginary modes are physical, the Pmma monolayer is not a local minimum on the potential-energy surface, invalidating the entire structural and electronic characterization. This is more load-bearing than secondary issues such as the hydrogenic derivation of the Bohr radius (which depends on an unstated dielectric constant) or the overlap with Ref. [13], because those do not undermine the existence of the material itself. A concrete convergence test, as proposed, would settle whether the imaginary modes vanish. If they do, the central claim stands and the paper needs only minor revision; if they persist, the structural prediction fails. Thus the reader's CONDITIONAL verdict remains appropriate, with no change needed.","tokens_in":20859,"tokens_out":4100,"duration_ms":39696,"concrete_test":"Perform phonon calculations with supercells of increasing size (e.g., 2×2×1, 3×2×1, and 3×3×1) and q-meshes (e.g., 4×4×1, 6×6×1) using the same PAW/PBE settings; also test a finer electronic k-grid (e.g., 13×13×1). If the maximum imaginary acoustic frequency at or near Γ decreases below ~0.05 THz or vanishes with increasing supercell size, the artifact interpretation is confirmed. If it remains sizeable (≳0.1–0.2 THz) and independent of supercell size, the structure is dynamically unstable, and the paper's central claim fails.","verdict_should_be":"UNCHANGED","load_bearing_attack":"In Sec. IIIA (Fig. 1c), the authors report 'slight imaginary frequencies near the Γ-point' in the acoustic branches and state they 'in all likelihood, are numerical artifacts' (citing Refs [50,61]) without any convergence study. No supercell size, q-point mesh, or force-constant tolerance is given; the Methods section only states a 9×9×1 electronic k-grid for the DFPT/PHONOPY calculation, which is not the parameter that controls finite-size errors in the long-wavelength acoustic limit. The central claim—that SiC-biphenylene is a dynamically stable 2D phase—depends entirely on these modes being artifacts. If they are physical, the Pmma monolayer is dynamically unstable, and the predicted band gap and excitonic properties have no stable ground state to attach to. Because the manuscript provides no evidence (e.g., convergence of the imaginary frequency magnitude with supercell size or q-mesh density), the stability claim is unsubstantiated as written.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports a first-principles study of a predicted SiC monolayer with the biphenylene network (SiC-biphenylene), along with bilayer and bulk forms. Using PBE and HSE06 for ground states, G0W0 for quasiparticle corrections, and BSE for optical spectra, the authors report a direct HSE06 gap of 2.16 eV, a G0W0 gap of 2.89 eV, and a first bright exciton at 2.07 eV with a binding energy of 0.82 eV. They also report phonon dispersions, AIMD simulations, elastic constants, a melting point of about 3475 K, effective masses, and stacking-dependent bilayer and bulk results. The central claim is that SiC-biphenylene is a dynamically, thermally, and mechanically stable 2D semiconductor with strong, polarization-dependent excitonic effects.","tokens_in":20983,"tokens_out":11395,"duration_ms":98380,"significance":"If the stability and excitonic results hold, the paper identifies a new 2D SiC polymorph with a direct gap in the visible range, strong exciton binding, and marked optical anisotropy, which would be of interest for optoelectronic applications. The methodology is internally consistent and the paper provides a systematic progression from monolayer to bilayer to bulk, with quantitative predictions that can be checked by future experiments or calculations. The main caveats are that the dynamic-stability claim depends on an unverified dismissal of imaginary acoustic phonons, and the exciton-radius/Frenkel classification depends on an unreported dielectric constant in a hydrogenic model. These points are correctable and do not by themselves invalidate the electronic structure results, but they need to be substantiated before the central claims are accepted.","major_comments":[{"comment":"The dynamical stability claim rests on dismissing the 'slight imaginary frequencies near the Γ-point' as numerical artifacts, but the manuscript provides no convergence evidence for this dismissal. The Methods section reports only a 9×9×1 electronic k-grid for the DFPT/PHONOPY calculation; it does not state the supercell size, q-mesh density, or force-constant tolerance. For long-wavelength acoustic modes, finite-size errors are controlled by the supercell and q-mesh, not by the electronic k-grid, so the quoted grid does not address the issue. Since a physical imaginary mode would make the Pmma monolayer dynamically unstable, this is a load-bearing point. Please provide a convergence study (for example, phonon dispersions for increasing supercell sizes or q-mesh densities, with the magnitude of the imaginary frequencies reported) or revise the stability claim accordingly.","section":"Sec. IIIA, Fig. 1(c)"},{"comment":"The exciton effective mass and Bohr radius are obtained from the hydrogenic formulas in Eq. (8) using an effective dielectric constant ε_r that is never stated. With μ_exciton = (E_b/R_h) ε_r^2 m0 and a_exciton = a_h ε_r m0/μ_exciton, the reported μ_exciton = 1.01 m0 and a_exciton = 2.14 Å are algebraic rearrangements of the BSE binding energy E_b = 0.82 eV plus an assumed ε_r of about 4.1. The 'Frenkel exciton' classification, and the corresponding 'Mott-Wannier' classification for the AA-stacked bilayer, are therefore not independent first-principles outputs. Please report ε_r explicitly and, ideally, the real-space exciton probability distribution from the BSE calculation to support the spatial-extent classification.","section":"Sec. IIIB, Eq. (8)"},{"comment":"The melting point claim (approximately 3475 K) is inferred from AIMD trajectories of 6 ps in NVE plus 6 ps in NVT with a 1 fs timestep. This is a short simulation time for a melting transition, and no system-size convergence or heating-rate analysis is reported. The abstract and conclusion advertise the melting point as a quantitative result; as presented, the evidence supports only a qualitative statement of stability to roughly 3400 K on a 6-ps timescale. Please add longer or larger simulations or temper the claim.","section":"Sec. IIID, Fig. 5"}],"minor_comments":[{"comment":"As printed, Eq. (8) is dimensionally unclear; please define all symbols (ε_r, a_h, R_h) and write the hydrogenic relations explicitly.","section":"Sec. II, Eq. (8)"},{"comment":"No convergence tests are reported for the G0W0 calculations (160 empty bands, 7×7×1 k-mesh) or the BSE calculations (48 valence bands, 24 conduction bands); a brief convergence statement would strengthen confidence in the quantitative values of the 2.89 eV gap and the 0.82 eV binding energy.","section":"Sec. II / Sec. IIIB"},{"comment":"The caption labels the last MD snapshot as '(c) 3600 K', duplicating the label of the 3475 K panel; it should be labeled (f).","section":"Fig. 5 caption"},{"comment":"The AB-stacked electron effective mass is 0.05 m0 in Table III but 0.054 m0 in the text; please reconcile these values and use consistent significant figures.","section":"Table III / Sec. IIIE"},{"comment":"The conclusion calls SiC-biphenylene a 'SiC allotrope'; since an allotrope is a form of a single element, this should be rephrased as a polymorph or phase of SiC.","section":"Sec. IV"},{"comment":"There are minor typographical issues: Table I's caption contains 'comaprison', and the formation energy of biphenylene is quoted as 0.46 eV in Sec. IIIA while Table I lists 0.46 eV/atom; please make the units consistent.","section":"Table I / Sec. IIIA"}],"recommendation":"major_revision","confidential_remarks":"The phonon-convergence issue is the most important; I would make a supercell/q-mesh convergence test a condition of acceptance. The exciton-classification issue is also not optional because the paper uses the inferred Bohr radius to assign Frenkel versus Mott-Wannier character. The G0W0/BSE convergence and the melting-point protocol are secondary but should be addressed."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nQuick take: this is a standard but solid computational characterization of a predicted 2D SiC phase, and the G0W0-BSE results are the useful part. The authors report a direct HSE06 gap of 2.16 eV, a G0W0 gap of 2.89 eV, and a first bright exciton at 2.07 eV with a binding energy of 0.82 eV. The polarization dependence of the absorption and the bilayer-stacking trends are interesting, and the calculations look internally consistent. They even validate their method against known biphenylene, which is a good sign.\n\nThe soft spots are in the supporting claims, not in the main numbers. The most serious one is the dynamical stability argument. The paper shows a slight imaginary acoustic frequency near Γ and dismisses it as a numerical artifact, citing earlier work, but provides no convergence study (no supercell-size check, no q-mesh check) to back that up. If those modes are real, the Pmma monolayer is unstable and the electronic structure is moot. This is fixable: a 2x2 and 3x3 supercell phonon calculation would settle it. As written, the stability claim is unproven.\n\nThe second issue is novelty. Reference [13] (Demirci et al., PRB 105, 035408, 2022) covered group-IV and compound biphenylene networks. The paper never states whether SiC-biphenylene is in that set. If it was already predicted there, then the structural novelty disappears and the contribution is the many-body results, which the authors should say explicitly. If it wasn't, they should say that too. The ambiguity is easy to remove.\n\nThird, the exciton Bohr radius (2.14 Å) and the Frenkel classification are not independent facts. They are algebraic transforms of the same binding energy via the hydrogenic formula, with an effective dielectric constant that is never stated. That doesn't undercut the 0.82 eV BSE binding energy, but it does mean the 'strongly localized exciton' language is a restatement, not a separate result. Also, the 3475 K melting point comes from a 6 ps AIMD run; I would treat it as a rough estimate.\n\nOverall, the central electronic and excitonic results are plausible and the paper deserves a serious referee. I'd send it to review with a request for phonon convergence, an explicit positioning against Ref. [13], and a sentence on the hydrogenic model.","headline":"A competent G0W0-BSE study of a plausible SiC-biphenylene monolayer; the electronic numbers are believable, but the dynamical stability claim and the novelty positioning need tightening before I'd take the structure for granted.","tokens_in":21603,"tokens_out":5554,"would_cite":true,"duration_ms":50270,"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":"A computational design predicts a stable planar silicon carbide phase in the biphenylene network, with a direct 2.89 eV quasiparticle gap and a strongly bound 0.82 eV exciton.","keywords":["silicon carbide","biphenylene network","two-dimensional materials","G0W0","Bethe-Salpeter equation","excitons","density functional theory","direct band gap"],"falsifier":"Compute the phonon dispersion with systematically larger supercells (e.g., 3×3, 4×4, 5×5) and denser k-point sampling; the dynamic-stability claim collapses if the imaginary acoustic modes near $\\Gamma$ persist rather than vanishing with improved convergence. A second check is to grow or exfoliate the monolayer and measure the absorption edge: the model predicts a sharp y-polarized exciton peak at 2.07 eV, well below the 2.89 eV G0W0 gap.","tokens_in":20563,"feed_emoji":"🔬","tokens_out":7211,"duration_ms":62363,"temperature":0.7,"pith_summary":"The paper claims that a silicon carbide version of the biphenylene network, a planar 2D lattice of octagon, hexagon, and tetragon rings, is a stable semiconductor suitable for optoelectronics. It predicts a direct band gap of 2.16 eV at the HSE06 level and 2.89 eV at the G0W0 level, with a strongly bound bright exciton at 2.07 eV whose 0.82 eV binding energy indicates pronounced excitonic effects. The same framework predicts that stacking such monolayers into bilayers and a bulk phase changes the band gap character and exciton binding, giving a route to tune optical response. If correct, this would add a new member to the small family of experimentally accessible 2D silicon carbides, with polarization-dependent absorption and potentially long-lived excitons.","feed_headline":"New 2D SiC phase predicted with a 2.89 eV direct gap","feed_subtitle":"The monolayer is predicted to be stable, with its first bright exciton at 2.07 eV bound by 0.82 eV.","key_machinery":"The central object is the SiC-biphenylene network: a stoichiometric 1:1 Si:C planar allotrope with orthorhombic Pmma symmetry, whose unit cell contains two fused octagons plus hexagons and tetragons and has anisotropic lattice vectors $|\\vec a|=5.61$ Å and $|\\vec b|=9.44$ Å. The argument is carried by the hierarchy HSE06 $\\to$ G0W0 $\\to$ BSE: HSE06 supplies the starting eigenvalues and wave functions, the single-shot G0W0 self-energy gives the quasiparticle corrections, and the BSE kernel produces the exciton spectrum whose bright states and binding energies are the main results.","core_discovery":"The paper predicts that silicon carbide arranged in the biphenylene network, a planar Pmma lattice of fused octagon, hexagon, and tetragon rings, is a dynamically, thermally, and mechanically stable 2D phase. Its HSE06 band gap is direct at $\\Gamma$ with a value of 2.16 eV, and the G0W0 quasiparticle gap widens to 2.89 eV, a self-energy correction of 0.73 eV. Solving the Bethe-Salpeter equation on top of G0W0 yields a first bright exciton at 2.07 eV for y-polarized light, with a binding energy of 0.82 eV, a Bohr radius of 2.14 Å, and an exciton effective mass of 1.01 $m_0$, placing it in the Frenkel exciton regime. AIMD simulations support thermal stability and indicate a melting point near 3475 K. Different bilayer stackings alter the gap from direct to indirect and shift the exciton binding energy from 0.17 eV (AA) to 0.64 eV (AA' and AB), while the bulk ABA-stacked phase is predicted to have a direct gap of 3.01 eV with a smaller self-energy correction.","pith_inferences":["As an extension, the most direct synthesis route would be epitaxial growth on a metal or SiC substrate, following the bottom-up chemistry used for monolayer honeycomb SiC; an experimental absorption measurement at the predicted 2.07 eV edge would test the whole prediction chain.","As an extension, because the valence and conduction band edges sit on different atomic species (C versus Si), the material may form spatially separated excitons with long lifetimes; this is testable by time-resolved photoluminescence, though the paper does not compute lifetimes.","As an extension, the small calculated Bohr radius (2.14 Å) places the first exciton in the Frenkel regime, an unusual regime for an inorganic 2D semiconductor that could be probed by magneto-optical experiments.","As an extension, the predicted noble-gas intercalation in AA'-stacked bilayers suggests a route to tune the interlayer distance and electronic coupling, but this goes beyond the paper's own calculations."],"forward_implications":["A freestanding monolayer of SiC-biphenylene would be a direct-gap 2D semiconductor with a quasiparticle gap near 2.9 eV and its first strong optical absorption near 2.1 eV, placing excitonic features well below the single-particle gap.","Optical absorption is strongly polarization-dependent: the first bright exciton appears for y-polarized light, while x-polarized absorption starts near 2.97 eV, so polarized spectroscopy can distinguish the phase.","Charge carriers are highly anisotropic: electron and hole effective masses along $\\Gamma \\to Y$ are roughly five times smaller than along $\\Gamma \\to X$, predicting faster transport along one lattice direction.","Bilayer stacking is a tuning knob: AA stacking keeps a direct gap (2.06 eV G0W0) with a weakly bound exciton (0.17 eV), while AA' and AB stacking produce indirect gaps of 3.04 and 3.43 eV with more localized excitons.","The bulk AB-stacked phase is predicted to be a direct-gap semiconductor at 3.01 eV with reduced self-energy corrections and a Mott-Wannier exciton (binding 0.20 eV), extending the material family from monolayer to three dimensions."],"supporting_citations":[{"why":"Reports the large-scale synthesis of monolayer honeycomb SiC, establishing that 2D SiC phases can be grown epitaxially and motivating the search for other 2D SiC allotropes.","marker":"[69]"},{"why":"Reports the experimental realization of atomic monolayer Si9C15, providing a second synthesized 2D SiC phase that this work extends.","marker":"[16]"},{"why":"Reports the synthesis of biphenylene nanoribbons, establishing the biphenylene network as an experimentally accessible structural motif.","marker":"[14]"},{"why":"The original theoretical prediction of the biphenylene carbon sheet and its electronic properties, which the SiC analogue is built from.","marker":"[31]"},{"why":"Provides earlier excitonic binding energies for SiC sheets and nanotubes, used as a comparison for the monolayer exciton binding energy.","marker":"[30]"},{"why":"Supplies the many-body Green's-function framework (G0W0 and BSE) on which the quasiparticle and exciton calculations are based.","marker":"[65]"},{"why":"Establishes the Bethe-Salpeter equation approach for computing electron-hole excitations and optical spectra from first principles.","marker":"[70]"},{"why":"Provides structural, mechanical, electronic, and catalytic data for the carbon biphenylene network, used as a comparison for stability and elastic properties.","marker":"[55]"}],"fun_headline_variants":["2D SiC biphenylene phase predicted with 2.89 eV direct gap","Strongly bound excitons predicted in novel 2D SiC phase","2D SiC biphenylene: stable, 2.89 eV gap, strong excitons","New 2D SiC phase: direct 2.89 eV gap and 0.82 eV exciton"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The stability claim assumes that slight imaginary acoustic phonon frequencies near the $\\Gamma$ point are numerical artifacts, but no convergence study is provided to show they disappear with larger supercells or denser k-points.","fun_headline_variants_meta":{"raw":{"variants":["2D SiC biphenylene phase predicted with 2.89 eV direct gap","Strongly bound excitons predicted in novel 2D SiC phase","2D SiC biphenylene: stable, 2.89 eV gap, strong excitons","New 2D SiC phase: direct 2.89 eV gap and 0.82 eV exciton"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000895,"raw_usage":{"total_tokens":3957,"prompt_tokens":1142,"completion_tokens":2815,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":758,"completion_tokens_details":{"reasoning_tokens":2718}},"tokens_in":758,"tokens_out":2815,"duration_ms":19048,"temperature":1.0,"reasoning_tokens":2718,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T13:04:00.317809+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the phonon dispersion with systematically larger supercells (e.g., 3×3, 4×4, 5×5) and denser k-point sampling; the dynamic-stability claim collapses if the imaginary acoustic modes near $\\Gamma$ persist rather than vanishing with improved convergence. A second check is to grow or exfoliate the monolayer and measure the absorption edge: the model predicts a sharp y-polarized exciton peak at 2.07 eV, well below the 2.89 eV G0W0 gap.","supporting_citations":[{"cited_title":"Biphenylene network: A nonbenzenoid carbon allotrope.Science, 372(6544):852–856, 2021","cited_arxiv_id":null,"evidence_quote":"Reports the experimental realization of atomic monolayer Si9C15, providing a second synthesized 2D SiC phase that this work extends."},{"cited_title":"Tokatly, and Angel Rubio","cited_arxiv_id":null,"evidence_quote":"Reports the synthesis of biphenylene nanoribbons, establishing the biphenylene network as an experimentally accessible structural motif."},{"cited_title":"Hudspeth, Brandon W","cited_arxiv_id":null,"evidence_quote":"The original theoretical prediction of the biphenylene carbon sheet and its electronic properties, which the SiC analogue is built from."},{"cited_title":"Hybrid functionals based on a screened Coulomb potential","cited_arxiv_id":null,"evidence_quote":"Provides earlier excitonic binding energies for SiC sheets and nanotubes, used as a comparison for the monolayer exciton binding energy."},{"cited_title":"A unified formulation of the constant temperature molecular dynamics methods","cited_arxiv_id":null,"evidence_quote":"Supplies the many-body Green's-function framework (G0W0 and BSE) on which the quasiparticle and exciton calculations are based."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes the Bethe-Salpeter equation approach for computing electron-hole excitations and optical spectra from first principles."},{"cited_title":"The electronic characterization of biphenylene—experimental and theoretical insights from core and valence level spectroscopy","cited_arxiv_id":null,"evidence_quote":"Provides structural, mechanical, electronic, and catalytic data for the carbon biphenylene network, used as a comparison for stability and elastic properties."}],"review_version":1}