{"id":"ab377b28-724b-408b-9a1b-cb13320bbf9c","arxiv_id":"2502.07899","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Yb-intercalated graphene shows spatially varying van Hove singularity positions caused by inhomogeneous intercalation and substitutional Yb atoms in the graphene lattice.","lead":"A scanning tunneling microscopy study maps how the electron energy landscape of ytterbium-doped graphene varies across the surface, revealing two distinct types of ytterbium doping with opposite local effects. The result matters because heavy doping toward a van Hove singularity is a route to new electronic phases, and this paper shows that the doping is spatially uneven.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Substitutional-Yb assignment in Fig. 3B is indirect and load-bearing; without chemical identification the two-mechanism vHS-heterogeneity claim lacks support.","rationale":"The reader's weakest assumption matches the load-bearing concern: the bright spots in the differential conductance map are assigned to substitutional Yb atoms without direct chemical verification. This assignment is essential to the claimed two-mechanism picture, because the second mechanism (substitutional Yb causing a local vHS shift and quasiparticle renormalization) exists only if those spots are actually Yb in the graphene basal plane. The alternative—that they are contaminants, carbon vacancies, or other defects—would reduce the central claim to the first mechanism alone, namely spatial inhomogeneity of intercalated Yb with a ~3 nm screening length. The theoretical modeling does not rescue the assignment: it uses the disputed spot map as input, fits a Gaussian potential with W estimated from the same data and σ chosen to maximize agreement, and then reports a cross-correlation of only ~0.16, which the authors themselves describe as not strong. Therefore the quantitative confirmation is circular and weak, though the qualitative spatial heterogeneity of the vHS itself appears well supported by the experimental EvHS map. Because the reader already rendered a CONDITIONAL verdict that appropriately reflects this unverified assignment, my stress-test does not move the verdict; it reinforces the need for direct chemical or structural evidence before the two-mechanism claim can be accepted.","tokens_in":11730,"tokens_out":4090,"duration_ms":39840,"concrete_test":"Perform direct chemical/structural identification of the bright spots by preparing a cross-sectional or transferred lamella from the same Yb-graphene sample and imaging with aberration-corrected scanning transmission electron microscopy combined with electron energy loss spectroscopy (STEM-EELS) to map Yb atoms. If Yb atoms are found incorporated in the graphene basal plane at the exact locations corresponding to the bright spots in Fig. 3B, the substitutional assignment is confirmed; if the bright spots contain no Yb or are not in the basal plane, the second mechanism collapses and the central claim reduces to a single intercalation-driven inhomogeneity mechanism.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central two-mechanism claim depends on identifying the ~50 bright spots in g(r, E=-800 meV) (Fig. 3B) as substitutional Yb atoms. The paper's evidence is circumstantial: (i) the spot count cannot account for the total ARPES doping density, which only shows that not all Yb are substitutional; (ii) undoped graphene on SiC shows no such states, which does not exclude other Yb-induced defects or processing-related contamination. No direct chemical or atomic-structural identification is provided. This matters because the second mechanism—substitutional Yb causing a strong local vHS shift and quasiparticle renormalization—rests entirely on that assignment. The theoretical confirmation is also weakened by circularity and low cross-correlation: the disorder profile in Fig. 5A is derived from the same g(r,-800 meV) map used to define the impurities, W is estimated from the observed maximum LDOS shift, σ is chosen by maximizing the cross-correlation against the experimental vHS map, and the resulting peak cross-correlation is only ~0.16, which the authors themselves call weak. Thus the quantitative support for the substitutional-Yb mechanism is much weaker than the qualitative narrative suggests.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This manuscript reports an in-situ ARPES and SI-STM study of Yb-intercalated quasi-freestanding monolayer graphene grown on SiC. The authors map the local energy position of the van Hove singularity (vHS) and observe significant spatial heterogeneity, which they attribute to two mechanisms: (i) spatial variations in the doping caused by inhomogeneously distributed intercalated Yb atoms, with a characteristic screening length of about 3 nm, and (ii) strong local perturbations from substitutional Yb atoms in the graphene basal plane. A tight-binding model with local Gaussian potentials, whose parameters are matched to the experimental data, is used to reproduce the vHS map and the ARPES band broadening.","tokens_in":12123,"tokens_out":5277,"duration_ms":45248,"significance":"The direct spatial mapping of vHS energy variations is a valuable and relatively novel experimental contribution, and the in-situ combination of ARPES and STM on the same surface is a clear strength of the paper. If the two-mechanism picture is confirmed, the work would establish that heavily doped graphene near a vHS should be treated as a spatially inhomogeneous electronic system, with implications for interpreting transport, spectroscopic, and potential correlated-phase experiments. However, the chemical identification of substitutional Yb is indirect, and the theoretical confirmation is weakened by parameter fitting against the same data from which the disorder profile is constructed. The core qualitative observation is interesting, but the paper's strongest claim is not yet fully supported.","major_comments":[{"comment":"The assignment of the ~50 bright spots in g(r, E = -800 meV) to substitutional Yb atoms is not directly established. The arguments presented - the spot count being far below the total doping density and the absence of similar features in undoped graphene - exclude neither other Yb-induced defects nor contamination. Because the second mechanism (substitutional Yb in the basal plane) carries a large part of the paper's conclusion, this identification is load-bearing. The authors should provide direct chemical or atomic-structural evidence (e.g., atomically resolved STM simulations of substitutional Yb, X-ray photoelectron spectroscopy, or a characteristic signature in the topographic image) or clearly reframe the claim as tentative.","section":"Fig. 3B and 'In STM...'"},{"comment":"The model confirmation is partly circular. The disorder profile in Fig. 5A is constructed from the same g(r, -800 meV) map used to identify the impurities, W is estimated from the maximum LDOS shift generated by the same Gaussian potential, and sigma is selected by maximizing the cross-correlation between the simulated and experimental vHS maps. The resulting peak cross-correlation is only ~0.16, which the authors themselves call weak. This does not support the statement that the theoretical approach 'well captures' the experimental map. Please provide an out-of-sample test or a null-model comparison (e.g., random disorder profiles with the same density) to quantify the predictive power of the model.","section":"Section II (Theoretical Model Description) and Fig. 6"},{"comment":"The abstract states that the chemical potential is shifted to within 250 meV of the vHS, while Fig. 2E shows the vHS peak at 400 meV above the Fermi level and the Dirac point 1.4 eV below it. This discrepancy should be reconciled: either the 400 meV peak is not the vHS, or the 'within 250 meV' statement is incorrect.","section":"Abstract and Sec. II (sample characterization)"},{"comment":"The claim that the vHS peak amplitude increases with peak position, indicating interaction effects beyond a rigid band shift, is interesting; however, the correlation is reported only qualitatively. Please provide a quantitative correlation coefficient and error bars, and specify how the vHS position was extracted (e.g., fitting procedure, energy resolution) so the reader can assess the significance.","section":"Fig. 3D and accompanying text"},{"comment":"The model assumes the same Gaussian potential amplitude W and width sigma for both substitutional and intercalated Yb atoms. This is an ad hoc assumption with no physical justification, and since W and sigma are fit to the data, the model cannot separately validate the two mechanisms. A sensitivity analysis or separate treatment of the two species would strengthen the claim.","section":"Section II (Theoretical Model Description)"}],"minor_comments":[{"comment":"There are several typos: 'substitional' and 'subsitutional' should be 'substitutional', 'inhomogenity' should be 'inhomogeneity', and 'Inhonomgenous' in the caption of Fig. 3 should be corrected.","section":"Throughout"},{"comment":"Reference [36] is cited as 'Emstev et al.' in the text; the correct spelling is 'Emtsev et al.'.","section":"References"},{"comment":"The text referring to Figure 2 is inconsistent: 'Figure 2D represents a zoomed-in image of the area in Figure 2C marked by the white box' appears to refer to the topography panel, not the Fermi surface shown in Fig. 2C. Please correct the figure callouts.","section":"Fig. 2 callouts"},{"comment":"The sentence 'We take a supercell consisting of 82 x 82 unit cells which cover some of the substitutional/intercalated Yb profile shown in Figure 5A, upper panel of Figure 7, 8, and 9' is confusing because Figures 7-9 are not called out in order elsewhere. Please clarify which panels are being referenced.","section":"Fig. 7-9 callouts"},{"comment":"The caption of Fig. 5 states that the nearest-neighbor hopping parameter t is set to 3 eV and the finite strip is 200 x 200 unit cells, but the text in Section II mentions both 200 x 200 and 82 x 82 supercells; please ensure the caption is consistent with the methods.","section":"Fig. 5 caption"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is within the scope of a condensed matter physics journal, but the evidence for the substitutional-Yb mechanism is not yet strong enough for publication as the central claim. The authors should either provide direct chemical identification or substantially temper the conclusions and reframe the substitutional-Yb interpretation as a hypothesis. The vHS heterogeneity observation itself is solid and could be published with more careful theoretical support."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The first thing to know: this paper delivers the first real-space maps of the van Hove singularity position in heavily Yb-intercalated graphene, and the maps show genuine, sizeable spatial variation. That result stands even if the substitutional-Yb story weakens. The second thing: the paper's second mechanism—substitutional Yb in the basal plane—rests on indirect evidence and a theoretical model that is partly fit to the same data it is used to confirm.\n\nThe SI-STM data are the clear strength. The vHS map in Fig. 3C shows a spatially textured electronic structure, with higher vHS energy near the bright spots in the -800 meV conductance map. The correspondence is visible by eye. The in-situ ARPES on the same surface anchors the doping level and places the vHS within ~250 meV of EF, and the spatially averaged STM/ARPES comparison is clean. The ~3 nm screening length is a concrete, testable number. The tight-binding model with local Gaussian potentials reproduces the qualitative feature that the vHS moves away from EF near impurities, and the computed spectral broadening echoes the ARPES linewidths.\n\nThe substitutional-Yb assignment is the weakest link. It is inferred from the count of bright spots (about 50 in a 23 nm FOV) being much less than the total doping density, plus the absence of such states in undoped graphene. Neither observation rules out other Yb-induced defects or processing-related contamination that could also produce localized states. Without atomically resolved chemical identification or a convincing structural signature, the two-mechanism picture remains plausible but not proven. The theory also has a circular flavor: the disorder profile in Fig. 5A is generated from the same g(r,-800 meV) map used to locate the impurities, W is estimated from the maximum LDOS shift in that map, and σ is chosen by maximizing the cross-correlation, which peaks at ~0.16—weak by the authors' own admission. So the calculation is better described as a consistency check than as confirmation. The central vHS heterogeneity observation does not depend on these fits, but the specific causal decomposition does.\n\nMinor issues: the text inconsistently refers to a 20 nm vs 23 nm field of view in a few places, and some figure callouts are sloppy. These are cosmetic.\n\nBottom line: this is a genuinely new application of SI-STM to a heavily doped graphene system and yields a robust qualitative picture of inhomogeneous doping. It deserves a serious referee. I would send it out, but I would push for a revision that either provides direct evidence for substitutional Yb or reframes that mechanism as a phenomenological model, and that reports the cross-correlation honestly without overclaiming confirmation.","headline":"First real-space vHS maps in overdoped graphene are the real deal; the substitutional-Yb claim is plausible but under-supported.","tokens_in":12534,"tokens_out":1983,"would_cite":true,"duration_ms":19381,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Two distinct Yb dopant species make the van Hove singularity position in overdoped graphene vary on a 3-nm scale, so the material is electronically patchy rather than uniformly doped.","keywords":["graphene","van Hove singularity","ytterbium intercalation","substitutional doping","doping inhomogeneity","scanning tunneling microscopy","angle-resolved photoemission spectroscopy","electronic structure"],"falsifier":"Element-specific imaging of the same sample—for example, synchrotron-based X-ray scanning tunneling microscopy or atom probe tomography that chemically identifies the bright spots at -800 meV—would settle the claim. If those spots turned out to be carbon vacancies, residual contaminants, or silicon from the substrate, the two-mechanism picture would lose its substitutional-Yb component.","tokens_in":11531,"feed_emoji":"🔬","tokens_out":6230,"duration_ms":53678,"temperature":0.7,"pith_summary":"The paper uses spectroscopic-imaging scanning tunneling microscopy together with angle-resolved photoemission to test whether ytterbium-intercalated graphene, a material tuned close to its van Hove singularity, is electronically uniform. It finds it is not: the van Hove singularity position and amplitude vary over a length scale of about 3 nm, with two causes. Intercalated Yb atoms lying between the graphene and the silicon carbide substrate dope the sheet unevenly, while a smaller number of Yb atoms sitting inside the graphene basal plane shift the local doping strongly and renormalize the quasiparticle amplitude. The authors argue that heavily doped graphene near a van Hove singularity should be viewed as a spatially inhomogeneous electronic system, and that a rigid-band-shift description is insufficient.","feed_headline":"Doped graphene's key electronic feature is a 3-nm patchwork","feed_subtitle":"Intercalated and substitutional Yb atoms shift the van Hove peak differently, creating nanoscale regions with distinct electronic character.","key_machinery":"The paper's central objects are the spatially resolved van Hove singularity position $E_{\\text{vHS}}(\\mathbf{r})$, extracted pixel-by-pixel from $dI/dV$ spectra, and the differential conductance map at the Yb $4f$-state energy, $g(\\mathbf{r}, E=-800\\,\\text{meV})$, which marks the substitutional Yb sites. The mechanism that carries the argument is a local Gaussian potential model: each Yb atom, whether intercalated or substitutional, is represented by a potential $W \\exp(-r^2/(2\\sigma^2))$, and the tight-binding local density of states computed with these potentials reproduces both the measured vHS shifts and the ARPES broadening.","core_discovery":"The central claim is that overdoped graphene produced by Yb intercalation on SiC is not a uniformly doped metal but a patchwork of regions with different van Hove singularity energies. Two populations of Yb atoms are responsible. Most Yb atoms intercalate between graphene and the SiC substrate and donate electrons, moving the vHS toward the Fermi level, but their spatial distribution is uneven, and the resulting doping variations are screened over roughly 3 nm. A minority of Yb atoms become substitutional, embedded in the graphene basal plane; these act as strong local scatterers that push the vHS away from the Fermi level, suppress the quasiparticle amplitude, and broaden the spectral function, reproducing the ARPES linewidths. The spatially resolved vHS map correlates directly with the positions of the bright impurity states seen at -800 meV, and a tight-binding model with local Gaussian potentials placed at those sites reproduces the observed vHS shifts.","pith_inferences":["A similar two-species coexistence may occur in other lanthanide-intercalated graphene systems (for example, Er or Tb) and in other two-dimensional materials where metal atoms can both intercalate and substitute, potentially making electronic inhomogeneity a general feature rather than a Yb-specific quirk.","If nanoscale regions with the vHS close to the Fermi level are real, they are natural places to search for emergent superconductivity or charge order; local probes at lower temperature could test this.","The positive correlation between vHS amplitude and position hints that interactions, not just disorder, are at play; momentum-resolved STM or quasiparticle interference measurements could reveal the interaction channel.","The Gaussian-potential disorder model reconciles ARPES and STM, but its parameters were fit to a single field of view; testing it on multiple areas and on samples with different Yb coverage would show whether the 3 nm scale is universal."],"forward_implications":["Studies that treat Yb-intercalated graphene as uniformly doped will miss the nanoscale patchwork, because transport and spectroscopic averages mix regions with different vHS energies.","The vHS peak amplitude is positively correlated with its energy position, so a rigid-band-shift description of overdoped graphene is insufficient; interaction and disorder effects must be included.","The coexistence of intercalated and substitutional dopants means the total dopant density alone does not determine the local electronic structure.","Theoretical calculations of correlated phases near the vHS should use disorder profiles rather than uniform doping to predict where such phases could nucleate.","The roughly 3 nm screening length sets the scale over which local doping variations are felt, providing a concrete length scale for nanoscale electronic texture in graphene."],"supporting_citations":[{"why":"Establishes Yb intercalation as a route to dope graphene near the van Hove singularity and identifies the Yb 4f states used to locate the substitutional Yb atoms.","marker":"[20]"},{"why":"Shows overdoping beyond the van Hove singularity and the extended van Hove scenario that the present work's ARPES data builds on.","marker":"[21]"},{"why":"Provides the buffer-layer growth procedure for producing quasi-free-standing monolayer graphene on SiC that this study follows.","marker":"[36]"},{"why":"Supplies the tight-binding model of graphene used to compute the local density of states and van Hove shifts.","marker":"[38]"},{"why":"Provides the band-structure unfolding procedure used to compare the theoretical spectral functions with ARPES data.","marker":"[39–41]"},{"why":"Documents simultaneous intercalation and substitutional doping in epitaxial graphene, supporting the interpretation that both Yb species can coexist.","marker":"[47]"}],"fun_headline_variants":["Van Hove singularity shifts on a 3-nm scale in Yb-doped graphene","Yb dopants create nanoscale patchwork of electronic states in graphene","Overdoped graphene shows 3-nm electronic patchwork from Yb","Intercalated and substitutional Yb yield nanoscale electronic patches"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The assignment of the bright spots in the conductance map at -800 meV to substitutional Yb atoms in the graphene basal plane rests on indirect evidence, because no atomically resolved chemical identification distinguishes Yb from other possible impurities.","fun_headline_variants_meta":{"raw":{"variants":["Van Hove singularity shifts on a 3-nm scale in Yb-doped graphene","Yb dopants create nanoscale patchwork of electronic states in graphene","Overdoped graphene shows 3-nm electronic patchwork from Yb","Intercalated and substitutional Yb yield nanoscale electronic patches"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000879,"raw_usage":{"total_tokens":3850,"prompt_tokens":1043,"completion_tokens":2807,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":659,"completion_tokens_details":{"reasoning_tokens":2725}},"tokens_in":659,"tokens_out":2807,"duration_ms":17695,"temperature":1.0,"reasoning_tokens":2725,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-08T11:28:11.649980+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Element-specific imaging of the same sample—for example, synchrotron-based X-ray scanning tunneling microscopy or atom probe tomography that chemically identifies the bright spots at -800 meV—would settle the claim. If those spots turned out to be carbon vacancies, residual contaminants, or silicon from the substrate, the two-mechanism picture would lose its substitutional-Yb component.","supporting_citations":[{"cited_title":"Briggs, Z","cited_arxiv_id":null,"evidence_quote":"Establishes Yb intercalation as a route to dope graphene near the van Hove singularity and identifies the Yb 4f states used to locate the substitutional Yb atoms."},{"cited_title":"Rosenzweig, H","cited_arxiv_id":null,"evidence_quote":"Shows overdoping beyond the van Hove singularity and the extended van Hove scenario that the present work's ARPES data builds on."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the buffer-layer growth procedure for producing quasi-free-standing monolayer graphene on SiC that this study follows."},{"cited_title":"Zhang, V","cited_arxiv_id":null,"evidence_quote":"Supplies the tight-binding model of graphene used to compute the local density of states and van Hove shifts."},{"cited_title":"Schiros, D","cited_arxiv_id":null,"evidence_quote":"Documents simultaneous intercalation and substitutional doping in epitaxial graphene, supporting the interpretation that both Yb species can coexist."}],"review_version":1}