{"id":"820eb1a3-66ed-4b81-9c3c-782c384a8bd7","arxiv_id":"2607.28296","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"An iterative DFT/TB/SCTMA/experiment loop yields an effective disordered model of Au-intercalated graphene that matches ARPES VHS broadening and kinks after phenomenological parameter refinement.","lead":"The authors build an iterative DFT–tight-binding–SCTMA workflow and fit it to ARPES on Au-cluster intercalated graphene. It attributes the broadened van Hove singularity and kink features to selected Au d-orbital hybridization plus a local carbon-ring scattering potential.","discovery_kind":"new_method","skeptic_critique":{"model":"grok-4.5","headline":"The single-site dilute SCTMA idealization of multi-atom Au clusters is the load-bearing gap between the fitted spectra and the claimed microscopic identification.","rationale":"The reader correctly isolates the modeling step that converts a DFT-guided orbital filter into a claim that SCTMA “reproduces” cluster-phase ARPES and thereby identifies the essential microscopic ingredients. That step is load-bearing: without the dilute single-site representation, the fitted U sign flip and the visual match no longer license the same microscopic story. I find no stronger internal inconsistency—the pristine 6NN TB, Wannier truncation checks, and SCTMA equations are standard and internally coherent—so the verdict remains CONDITIONAL rather than REJECT. An explicit multi-atom-cluster SCTMA (or equivalent) is the single check that would decide whether the concern lands; until then the paper is a useful calibrated effective description, not yet a morphology-robust identification.","tokens_in":17081,"tokens_out":649,"duration_ms":11855,"concrete_test":"Replace the single-Au impurity by an explicit three-Au cluster motif (geometry from the experimental ~2.2 nm network / (2,1) intra-cluster vector) inside the same TB+SCTMA (or a small supercell T-matrix) pipeline, keeping total Au/C≈2%. Recompute A(k,ω) along the k_⊥ cut and the −2.35 eV constant-energy map of Fig. 8. If the kink and extended-VHS contours disappear or require qualitatively different |t_α|, ε_α, U to recover the ARPES match, the single-site idealization is load-bearing and the essential-ingredient claim weakens.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central claim equates SCTMA spectra of independent hollow-site Au atoms (n_imp≈2%, Eqs. 36–39, Sec. V D) with ARPES of the experimental “ostrich leather” cluster phase (triangular network d≈2.2 nm, ~3 Au per motif; Sec. V A). SCTMA sums repeated scattering on one impurity and explicitly neglects crossed diagrams and coherent inter-impurity scattering (Sec. III D). Intra-cluster Au–Au separations are only ~0.65 nm (the (2,1) vector), so local multipole structure, shared carbon rings, and coherent cluster form factors are outside the model. The headline match—VHS broadening/extension and kinks near −2.7 eV—is obtained only after large phenomenological retuning, most strikingly U from Wannier ≈−0.2 eV to +2 eV (Table II) and ε_α fixed to the experimental intensity maximum. Thus the “essential ingredients” (m=±1,±2 ring hybridization plus local U) are identified inside a single-site effective theory whose validity for the actual clustered morphology is assumed rather than demonstrated. If cluster-scale coherence or multi-atom potentials produce the same ARPES features without the same local channels, the microscopic attribution does not follow from the visual agreement.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.5","summary":"The manuscript proposes an iterative workflow that combines periodic DFT, Wannier truncation into a tight-binding model, disorder averaging in the self-consistent T-matrix approximation (SCTMA), and comparison with ARPES to build an effective model of dilute intercalated 2D systems. Applied to the Au-cluster (“ostrich leather”) phase of intercalated graphene on SiC, DFT/Wannier is used to identify hollow-site Au 5d channels (m=±1, ±2) hybridizing to the six-carbon ring; remaining parameters (ε_α, |t_α|, local ring potential U, broadenings) are refined against ARPES. The resulting SCTMA spectral function is reported to reproduce the main experimental signatures—broadening/extension of the occupied van Hove region and kink-like features near the impurity resonance—and the authors identify hybridization plus a local scattering potential as the essential microscopic ingredients.","tokens_in":17456,"tokens_out":1571,"duration_ms":33935,"significance":"If the workflow is robust, it offers a practical bridge between first-principles local chemistry and large-scale configurational disorder for intercalated van der Waals systems, where pure periodic DFT cannot capture ARPES averaging and pure phenomenology lacks orbital content. The pristine-graphene 6NN TB benchmark against DFT and ARPES is careful, and the Wannier truncation is cross-checked with a periodic TB unfolding. The application targets a concrete experimental puzzle (VHS reconstruction with only mild Dirac-point shift). The main value is methodological and transferable to other sparse intercalants (alkali, rare earth, etc.) linked to flat-band and Lifshitz physics. Strengths include explicit symmetry-channel projection of Au–C couplings and a transparent statement of which parameters DFT constrains versus which are phenomenological.","major_comments":[{"comment":"Sec. V A and V D equate SCTMA spectra of a dilute random gas of single hollow-site Au atoms (n_imp≈2%, Eqs. 36–39) with ARPES of the experimental cluster phase (triangular network d≈2.2 nm, ~3 Au per motif, intra-cluster separation ~√7 a_gr≃0.65 nm). SCTMA explicitly neglects crossed diagrams and coherent inter-impurity scattering (Sec. III D). Shared carbon rings, cluster multipoles, and coherent form factors are therefore outside the model. The microscopic attribution of VHS broadening and kinks to single-site m=±1,±2 ring hybridization plus local U is load-bearing for the central claim, yet the single-site idealization of multi-atom clusters is assumed rather than tested. A minimal check—e.g., a small multi-Au cluster impurity in T-matrix/SCTMA, a comparison of periodic multi-Au supercells vs single-Au, or a clear statement of which spectral features are robust to cluster form factors","section":"Sec. V A, V D; Eqs. 36–39; Sec. III D"},{"comment":"Table II and Sec. V C show large phenomenological retuning relative to DFT/Wannier: U from ≈−0.2 eV to +2 eV, ε_α fixed at −2.7 eV (the experimental intensity maximum), and |t_α| adjusted within a distance-dependent range. Sec. II(v) and the Abstract state that comparison with the same ARPES dataset guides refinement, then present the refined SCTMA maps as reproducing that ARPES (Fig. 8). Without a sensitivity analysis (which features survive when U is kept near the Wannier value; when ε_α is varied off the intensity peak; one-parameter scans) or a quantitative goodness-of-fit metric, it is unclear how much of the agreement is forced by the free parameters versus predicted by the DFT-selected channels. The sign and magnitude change in U is especially consequential for the claim that a local scattering potential is an essential ingredient; the QPI citation (Ref. 36, “in preparation”) and","section":"Table II; Sec. V C–D; Sec. II(v); Fig. 8"},{"comment":"The Abstract and Sec. VI state that the method “reproduces the main ARPES signatures” and identifies essential microscopic ingredients. Given the points above, the wording should be tightened to what is actually demonstrated: that a single-site effective impurity model with DFT-selected d-channels and phenomenologically adjusted ε_α, |t_α|, and U can match the main visual features of the cluster-phase ARPES. Claims that the experimental cluster morphology is thereby microscopically explained should be caveated unless additional evidence (cluster-resolved modeling or robustness tests) is supplied.","section":"Abstract; Sec. VI"}],"minor_comments":[{"comment":"Abstract typo: “V12an Hove singularity” should be “van Hove singularity”.","section":"Abstract"},{"comment":"Fig. 1 is referenced as an experimental image of the cluster phase but the caption in the text is incomplete (“Experimental image of the cluster phase and the SCTMA formalism”); ensure the figure and caption fully identify scale and what is shown.","section":"Fig. 1"},{"comment":"Eq. (12) and η_α=0.17 eV: clarify whether the same η is used for pristine graphene, impurity levels, and SCTMA plots, and how it relates to experimental resolution versus lifetime broadening.","section":"Sec. III B 1; Sec. V C"},{"comment":"Sec. IV: the −0.55 eV shift aligning theory to ARPES for pristine graphene (changing μ from −377 meV to +173 meV) should be stated once as a global energy reference convention when comparing intercalated spectra as well.","section":"Sec. IV"},{"comment":"Table I lists 6s and dz² parameters that are later dropped; a short explicit sentence on why m=0 channels do not affect the occupied VHS window (beyond “minor effect”) would help readers.","section":"Table I; Sec. V B"},{"comment":"Ref. 36 is “in preparation” and is used to support the positive U; if unavailable, weaken that citation or supply the essential QPI observation in the text/SI.","section":"Sec. V C"}],"recommendation":"major_revision","confidential_remarks":"The methodological framing is publishable in a solid specialized journal (e.g. PRB) after the cluster-vs-single-site and fit-robustness issues are addressed; I would not recommend rejecting on novelty grounds. The circularity concern is real but partly disclosed by the authors; the decisive fix is robustness tests and more precise claim language rather than a new theory. Watch that Ref. 36 (in preparation, overlapping authorship) is not doing heavy lifting for U without accessible evidence."},"author_rebuttal":null,"desk_editor":{"model":"grok-4.5","letter":"Punchline: this is a careful methods-plus-application paper that turns standard pieces (DFT, Wannier, ring-harmonic impurities, SCTMA) into an explicit iterative workflow and a concrete effective model for the Au-cluster ARPES reconstruction. That synthesis is the real product, not a new formalism.\n\nWhat they do well is clear. The pristine graphene baseline is done properly—6NN TB, chemical-potential alignment, constant-energy maps against ARPES. Wannier truncation is checked against periodic TB and DFT in the window that matters. They are unusually explicit about what DFT can fix (hollow site, which d channels, m=±1/±2, order of |t|) and what it cannot (ε_α, U, exact hybridizations once substrate/buffer/distance are missing). The SCTMA maps then do show the two experimental signatures they care about: VHS broadening/extension and kinks near the impurity resonance. For people who model dilute intercalants or need a minimal disordered TB for this phase, that is useful.\n\nSoft spots, in proportion. The headline “reproduces ARPES” is after hand-tuning, most visibly U from Wannier ~−0.2 eV to +2 eV, and ε_α parked on the experimental intensity maximum. That is disclosed in Table II and the text, but it raises the circularity burden: same dataset guides refinement, then is cited as agreement. The load-bearing modeling choice is representing the ostrich-leather cluster network (~2.2 nm spacing, multi-Au motifs, intra-cluster ~0.65 nm) as a dilute random gas of single hollow-site impurities in SCTMA. SCTMA drops coherent inter-impurity scattering by construction; cluster multipoles and shared rings are outside the model. So the “essential ingredients” (selected d-ring hybridization plus local U) are identified inside that effective theory, not proven unique for the real morphology. No code, no error bars, visual match only—fine for this genre, but keep it in mind.\n\nWho it is for: epitaxial-graphene / ARPES / disordered-intercalant modelers, and anyone hunting a practical route from supercell DFT to disorder-averaged spectra. Flat-band/high-Tc framing is aspirational; the result does not resolve that. Math and citations look standard and honest. I would send it to peer review. Engage if you work this subfield; treat the microscopic attribution as a calibrated effective description pending cluster-explicit or out-of-sample checks.","headline":"Solid, usable effective-model pipeline for disordered intercalants; the ARPES match is real but calibrated, and the single-site stand-in for Au clusters is the main caveat.","tokens_in":18142,"tokens_out":620,"would_cite":false,"duration_ms":17277,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.5","headline":"An iterative DFT–tight-binding–disorder–experiment loop builds an effective model that explains how Au clusters reshape graphene’s bands near the van Hove singularity.","keywords":["intercalated graphene","ARPES","DFT","tight-binding","SCTMA","van Hove singularity","Au clusters","disorder averaging"],"falsifier":"ARPES (or a controlled calculation) on a phase with the same average Au density but truly isolated single hollow-site atoms versus explicit multi-atom clusters: if the VHS broadening and −2.7 eV kinks appear only for clusters, or if SCTMA with single-site impurities fails when cluster multipoles are required, the mapping fails.","tokens_in":17883,"feed_emoji":"🔬","tokens_out":998,"duration_ms":18778,"temperature":0.7,"pith_summary":"Intercalation can rewrite a host material’s electronic bands, but disordered intercalants make first-principles modeling hard to connect to ARPES. This paper proposes a general iterative workflow: DFT and Wannierization pick the relevant orbitals and constrain hoppings; a tight-binding impurity model plus self-consistent T-matrix disorder averaging produces the momentum-resolved spectral function; comparison with experiment refines the remaining parameters. Applied to Au-cluster intercalated graphene, the method reproduces the main measured signatures—broadening and extension of the occupied van Hove region and kink-like features in the dispersion. The analysis isolates two essential ingredients: hybridization of selected Au 5d orbitals to the six-carbon hollow-site ring, and a local intercalation-induced scattering potential on those carbons. The result is a compact effective model for dilute intercalated systems, aimed especially at cases where intercalation creates flat or van-Hove-related bands.","feed_headline":"How Au clusters rewrite graphene’s bands, in one loop","feed_subtitle":"DFT picks the orbitals; disorder averaging plus ARPES fixes the rest and isolates two ingredients.","key_machinery":"The iterative DFT/Wannier → truncated ring-hybridized TB impurity → SCTMA disorder average → experiment refinement loop. SCTMA builds a momentum-dependent self-energy from repeated scattering off the energy-dependent six-site ring potential V_eff(ω), producing the configurationally averaged graphene spectral function compared to ARPES.","core_discovery":"A dilute random gas of hollow-site Au impurities, hybridized to graphene through the m=±1 and m=±2 Au 5d channels on the six-carbon ring and dressed by a positive local ring potential, yields—after SCTMA disorder averaging—the ARPES signatures of the Au-cluster phase: VHS broadening/extension and kink-like renormalizations near the impurity resonance. DFT supplies the orbitals, symmetries, and order-of-magnitude couplings; experiment fixes the remaining energies and potential.","pith_inferences":["If single-site SCTMA already matches cluster-phase ARPES, much of the spectral reconstruction may be local impurity physics rather than long-range cluster-superlattice band folding.","The large fitted U versus the small Wannier onsite shift suggests electrostatic and substrate screening dominate over bare DFT local potentials in real devices.","Extending the loop to magnetic or spin–orbit-active intercalants could test whether the same ring-channel truncation still captures ARPES gaps and spin textures.","Materials engineered for high-Tc candidates via intercalation-driven flat bands may be screened faster by this DFT-guided SCTMA fit than by large disordered supercells alone."],"forward_implications":["Dilute intercalated graphene can be described by a compact effective TB model built from DFT-selected orbitals plus a local ring potential and SCTMA averaging.","For Au clusters, only Au 5d orbitals with m=±1 and ±2 hybridized to the first carbon ring, plus a positive U on that ring, are needed for the main occupied-band ARPES features.","The same workflow is offered for other sparse intercalants (alkali, rare-earth, halide) where disorder and van Hove or flat-band physics matter.","Hybridization sets the localized kink near the impurity level; the local potential controls broader linewidth, Dirac-point shift, and VHS–Dirac separation."],"fun_headline_variants":["DFT-SCTMA loop captures Au-cluster graphene ARPES kinks","Hybridization plus ring potential reshapes Au-intercalated bands","Iterative method links DFT, disorder, ARPES in Au-graphene","Dilute Au impurities broaden VHS and add dispersion kinks","Two microscopic ingredients yield Au-cluster graphene signatures"],"cache_read_input_tokens":128,"weakest_assumption_plain":"The real Au-cluster network can be treated, for spectral purposes, as a dilute random gas of single hollow-site Au atoms, with substrate, buffer, and clustering effects absorbed into a few fitted parameters.","fun_headline_variants_meta":{"raw":{"variants":["DFT-SCTMA loop captures Au-cluster graphene ARPES kinks","Hybridization plus ring potential reshapes Au-intercalated bands","Iterative method links DFT, disorder, ARPES in Au-graphene","Dilute Au impurities broaden VHS and add dispersion kinks","Two microscopic ingredients yield Au-cluster graphene signatures"]},"model":"grok-4.5","effort":"low","cost_usd":0.004297,"raw_usage":{"total_tokens":1286,"prompt_tokens":746,"num_sources_used":0,"completion_tokens":72,"cost_in_usd_ticks":42968000,"prompt_tokens_details":{"text_tokens":746,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":468,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":746,"tokens_out":72,"duration_ms":6930,"temperature":1.0,"reasoning_tokens":468,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-31T11:58:33.937883+00:00","model_set":{"reader":"grok-4.5"},"falsifier":"ARPES (or a controlled calculation) on a phase with the same average Au density but truly isolated single hollow-site atoms versus explicit multi-atom clusters: if the VHS broadening and −2.7 eV kinks appear only for clusters, or if SCTMA with single-site impurities fails when cluster multipoles are required, the mapping fails.","supporting_citations":[],"review_version":1}