{"id":"14d223cc-7652-445f-88ae-09f6f7edd933","arxiv_id":"2501.07518","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":7,"one_line_summary":"In epitaxial graphene, the ν=6 quantum Hall plateau widens with carrier density and narrows with disorder, as shown by storage-controlled experiments and Anderson-disorder simulations.","lead":"This paper shows that the width of a quantum Hall plateau in graphene depends on both carrier density and disorder, not density alone. By controlling surface molecules to change carrier density in two device regions with different mobilities, the authors connect plateau widths to mobility and impurity density, supported by tight-binding simulations.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The oxygen-adsorption protocol cannot separate carrier-density from disorder effects because O2 may itself scatter; a quantum-lifetime/Dingle analysis of the existing SdH data, or a gated control at fixed adsorbate coverage, would settle it.","rationale":"The reader's weakest_assumption identifies exactly the same load-bearing premise: the oxygen-adsorption protocol is assumed to change only carrier density, not scattering, and the two regions are assumed to share the same scatterer type. My reading of the manuscript confirms that this premise is both central and untested. If O2 adsorption/desorption alters the disorder potential, the within-region plateau-width growth in Fig. 3b could be driven by disorder changes rather than by density, and the quantitative linear fits and extrapolated critical densities would lose their stated interpretation. The between-region comparison still supports a role for disorder, so the manuscript's qualitative conclusion is not overturned, but the density contribution is not cleanly established. A Dingle/quantum-lifetime analysis on the already-measured SdH oscillations would directly test the scatterer-neutrality of O2 without new fabrication, and a gated control would be the cleanest experimental separation. This does not change the reader's conditional verdict; it sharpens the condition that should be met before the central claim is treated as established.","tokens_in":12371,"tokens_out":4732,"duration_ms":53931,"concrete_test":"Extract the Shubnikov–de Haas oscillation amplitude as a function of magnetic field from the existing ρxx(B) data at each storage time and fit the Dingle factor to obtain the single-particle (quantum) lifetime τq; then compare τq/τt with the Drude transport lifetime τt inferred from μ and n. If τq/τt varies systematically across the storage sequence at overlapping carrier densities, the assumption that oxygen changes only n fails and the ΔB-vs-n dependence in Fig. 3b is confounded. A complementary control would be to electrostatically gate one region to sweep n while keeping surface adsorbate coverage fixed and verify that the ν=6 plateau width still increases with n.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The experimental protocol modulates carrier density by oxygen adsorption/desorption (Section III.1). The interpretation that the ν=6 plateau width increases with n relies on the stated assumption that molecular oxygen is weakly physisorbed and \"only play[s] the role of modulating the carrier density and have a minor effect in the scattering rates\" (Section III.1). This is load-bearing because the same process changes both n and the surface environment. If desorbed or adsorbed O2 changes the scattering rate or introduces charged impurities, then the within-region trends in Fig. 3b cannot be uniquely assigned to carrier density; they could reflect a correlated disorder change. The UV treatment is not a clean control: it raises n but also heats the sample, and the observed mobility increase is attributed to annealing, so it does not demonstrate that O2 is scatterer-neutral. The between-region comparison in Fig. 3b is the only direct handle on disorder, but it relies on the additional, untested assumption in Section III.2 that both regions contain the same type of scatterers, so that the mobility ratio τHigh/τLow ≈ 12 can be read as an impurity-density ratio. No independent probe of the scattering mechanism (quantum lifetime, temperature dependence, Dingle analysis) is provided, and the Anderson-disorder simulations, while qualitatively supportive, do not by themselves rule out a purely density-driven explanation in the experimental device.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents an experimental study of the width of the ν=6 quantum Hall plateau (QHP) in a single epitaxial graphene Hall bar with two regions of markedly different mobility. The carrier density is varied not by a gate but by storage in vacuum or air, which adsorbs/desorbs molecular oxygen from the graphene surface, with one additional UV-treatment step. The authors report that the QHP width ΔB increases with carrier density in both regions, but that at similar densities the high-mobility region shows broader plateaus than the low-mobility region. They interpret this as evidence that both carrier density and disorder/impurity density control the plateau extension. Supporting tight-binding transport simulations (Landauer-Büttiker with Anderson disorder, implemented in Kwant) reproduce the qualitative trends of mobility decreasing with density and disorder, and of plateau width increasing with density and decreasing with disorder. The central claim is that the plateau width in graphene is set by a combination of carrier density and impurity density/strength, not by carrier density alone.","tokens_in":12651,"tokens_out":4333,"duration_ms":45154,"significance":"If the central claim is correct, the paper provides a useful experimental dataset on how environment-driven carrier-density changes and local mobility variations affect the QHE in epitaxial graphene, with direct relevance to quantum metrology. The study's strengths include the long-term storage protocol that tunes carrier density without a gate, the two-region internal comparison, and the independent tight-binding simulations, which are documented and presumably reproducible through Kwant. The qualitative agreement between simulated and experimental trends is a positive feature. However, the paper's central attribution of the plateau-width differences to disorder rests on two explicit assumptions (oxygen physisorption without scattering, and equal scatterer strength in both regions) that are plausible but not independently verified. The absence of experimental error bars and the lack of a quantitative definition of the experimental ΔB further weaken the quantitative claims. The contribution is potentially significant, but the evidence currently falls short of conclusively separating density effects from disorder effects.","major_comments":[{"comment":"The within-region interpretation relies on the statement that oxygen adatoms \"only play the role of modulating the carrier density and have a minor effect in the scattering rates of the system.\" This is load-bearing because the same storage process changes both n and the surface environment. No quantum-lifetime (τ_q) or Dingle analysis of the SdH oscillations shown in Figs. 2b and 2e is presented, and the UV control is not clean: it increases n but also heats the sample, and the observed mobility increase is attributed to annealing. Without an independent probe of the scattering rate, the ΔB vs. n trends in Fig. 3b could be partly due to correlated changes in disorder rather than solely to carrier density. Please add a Dingle/quantum-lifetime analysis of the existing data, or a control experiment with gating at fixed adsorbate coverage, to support the assumption.","section":"Section III.1"},{"comment":"The experimental plateau widths are plotted without error bars, and the criterion used to define ΔB for the experimental data is not stated in the main text (a definition is given only for the simulations, via σ(Rxy)>10^-3 R0/3). The critical densities n_c^High ≈ 3×10^11 cm^-2 and n_c^Low ≈ 2.3×10^11 cm^-2 are derived by extrapolating linear fits through very few points, but no fit uncertainties or goodness-of-fit measures are reported. Consequently, the key between-region comparison—that high-mobility plateaus are broader and have a different intercept—is not quantitatively supported. Please report the plateau-width extraction procedure, error bars on ΔB, the number of data points per fit, and the fit parameters with uncertainties.","section":"Fig. 3b, Section III.2"},{"comment":"The argument that the mobility ratio τHigh/τLow ≈ 12 reflects an impurity-density difference assumes U0|Low ≈ U0|High and identical scattering mechanisms in the two regions. The authors themselves acknowledge that \"it is not experimentally possible to disentangle the nature of the observed mobility differences.\" The Anderson-disorder simulations, while qualitatively supportive, are not fitted to the experimental ΔB values and use an arbitrary threshold; they therefore do not independently confirm that the between-region plateau-width difference is caused by impurity density rather than by a difference in scatterer type or scattering strength. An independent experimental probe of the scattering mechanism (e.g., temperature dependence of the quantum lifetime, or comparison of SdH amplitudes at matched densities) is needed to substantiate the disorder interpretation.","section":"Section III.2"}],"minor_comments":[{"comment":"The term \"absorption\" is used where \"adsorption\" is meant (e.g., \"absorption/desorption\" in the abstract and Section III.1); please correct this terminology.","section":"Throughout"},{"comment":"The sentence \"In this case, the increase in carrier density accompanied by an increase in mobility could be attributed to thermal annealing effects due to laser heating\" is somewhat vague; please specify the magnitude of the mobility change (the text later cites Δµ ≈ 50 cm^2/Vs) and how it compares to the overall mobility drift.","section":"Section III.1"},{"comment":"In Fig. 3a, the two regions are plotted with different mobility scales in the main panel and the inset; the fitting model µ(n) ∝ n^β is mentioned in the text but the explicit functional form and the uncertainties on the exponents β should be stated in the caption or in the text.","section":"Section III.2"},{"comment":"Reference [11] is cited as \"(in press)\" without volume or page numbers; please update to the published version if available.","section":"References"},{"comment":"The \"scaled tight-binding model\" is only referenced via Supplementary Information Section 1; a sentence summarizing the scaling procedure (e.g., how the magnetic field and disorder are rescaled) would improve the readability of the main text.","section":"Simulation section"}],"recommendation":"major_revision","confidential_remarks":"The manuscript reports a single-device, two-region study. The central conclusion rests on two explicit assumptions—oxygen neutrality and equal scatterer strength—that are plausible but not independently verified. The lack of error bars and of a stated experimental plateau-width definition is a serious quantitative gap. The paper is within the scope of the journal, and the simulation component is a positive element. If the authors can provide a quantum-lifetime or Dingle analysis, or another independent test of the oxygen-physisorption assumption, the central claim would be much better supported. As it stands, the manuscript is suggestive rather than conclusive."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"One thing to know: this is a solid qualitative experimental study—plateau width grows with carrier density in both regions, and the lower-mobility region shows narrower plateaus at comparable densities—but the protocol cannot fully separate density from disorder within a region, and the paper gives no error bars on the widths. The central qualitative point, that disorder as well as doping sets the QHP extension, is supported by the two-region comparison even if the density slope is partly confounded.\n\nWhat's new: a systematic, months-long tracking of the ν=6 QHP width in two regions of the same epitaxial graphene Hall bar, with carrier density tuned by oxygen adsorption/desorption rather than by a gate. The mobility–density exponents are fitted, and the Anderson-disorder tight-binding simulations are independent and qualitatively reproduce both the density trend and the disorder-induced narrowing. That is genuine value, especially for the graphene metrology community.\n\nThe main soft spot is the oxygen assumption. The within-region density trend rests on the claim that molecular oxygen only dopes and does not scatter. The paper states this but does not test it, and the UV treatment is not a clean control because it heats the sample. A Dingle or quantum-lifetime analysis of the existing SdH data would settle it, and the authors should either supply that or soften the claim. The between-region comparison also assumes the same scatterer type in both regions so that the mobility ratio can be read as an impurity-density ratio; that is reasonable but untested. The other soft spots are smaller but real: the plateau widths in Fig. 3b have no error bars, the critical densities come from extrapolating a few points, and no data or code are released. None of this destroys the qualitative conclusion, but it does prevent quantitative claims.\n\nWho this is for: researchers working on graphene QHE metrology or disorder physics. It is a useful experimental reference and a plausible simulation study, not a new law or a definitive method.\n\nI would send it to peer review. The referee should ask for error analysis on the plateau widths, an explicit test or caveat on oxygen-induced scattering, and ideally a Dingle analysis. With those additions the paper would be convincing; without them it is still a citable observation but not an established scaling result.","headline":"A useful qualitative dataset on QHP widths in epitaxial graphene, but the density/disorder separation is incompletely controlled and the manuscript needs error bars before the quantitative claims can be trusted.","tokens_in":661,"tokens_out":744,"would_cite":true,"duration_ms":30057,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["73.43.Qt","73.63.-b","72.80.Vp"],"model":"deepseek-v4-flash","headline":"The paper establishes that carrier density and disorder jointly set the magnetic-field width of the ν=6 quantum Hall plateau in epitaxial graphene.","keywords":["epitaxial graphene","quantum Hall effect","plateau width","carrier density","disorder","mobility","Anderson disorder","Landauer-Büttiker"],"falsifier":"Measure the width of the ν=6 plateau on an encapsulated graphene Hall bar where carrier density is tuned continuously with a gate while the surface is kept clean, so disorder cannot change with doping. If $\\Delta B$ fails to grow with density, or if an oxygen-exposed device shows a different width at the same gate-set density, the claim that density and disorder act jointly would be contradicted.","tokens_in":12159,"feed_emoji":"🧲","tokens_out":5223,"duration_ms":48705,"temperature":0.7,"pith_summary":"The paper asks what controls the magnetic-field window over which the ν=6 quantum Hall plateau survives in epitaxial graphene. By letting oxygen adsorb and desorb on two regions of the same Hall bar, the authors changed carrier density without a gate, while the two regions kept clearly different mobilities and hence different disorder levels. They report that plateau width grows with carrier density in both regions, but for equal densities the cleaner region always shows wider plateaus, so density alone cannot explain the widths. Tight-binding transport simulations with Anderson disorder reproduce both trends, supporting the conclusion that disorder and carrier density jointly set the plateau width.","feed_headline":"Both doping and disorder set graphene's quantum Hall plateau width","feed_subtitle":"Oxygen on/off a single Hall bar shows the ν=6 plateau widens with density and narrows with disorder, so density alone is not enough.","key_machinery":"The experimental machinery is a gate-free doping knob: molecular oxygen physisorbs weakly on graphene and acts as an electron acceptor, so vacuum storage desorbs it and raises electron density, while air storage re-adsorbs it and lowers density. The two device regions provide two disorder realizations within one wafer. The theoretical machinery is a scaled tight-binding model of a graphene Hall bar with nearest-neighbor hopping plus an Anderson on-site disorder potential $\\epsilon_i \\in [-\\omega_A/2, \\omega_A/2]$; conductance is computed with the Landauer-Büttiker formalism, and the plateau width is defined as the field range where the standard deviation of $R_{xy}$ stays below a threshold. Comparing experiment and simulation maps carrier density to plateau width for two disorder strengths.","core_discovery":"The central claim is that the width $\\Delta B$ of the $\\nu=6$ quantum Hall plateau in epitaxial graphene is controlled by carrier density and by disorder together, not by carrier density alone. The evidence is a single Hall bar with two regions whose mobilities differ by an order of magnitude; storage in vacuum or air shifts the carrier density via oxygen adsorption and desorption. In the high-mobility region, $\\Delta B$ grows from about 1 T to nearly 3 T as density rises; in the low-mobility region, $\\Delta B$ shrinks and vanishes near a critical density $n_c \\approx 2.3 \\times 10^{11}\\,\\text{cm}^{-2}$, while the high-mobility region extrapolates to a larger $n_c \\approx 3 \\times 10^{11}\\,\\text{cm}^{-2}$. Landauer-Büttiker simulations on a scaled tight-binding model with Anderson disorder show the same qualitative behavior: $\\Delta B$ increases almost linearly with carrier density, and stronger disorder reduces $\\Delta B$ at every density. The paper therefore identifies impurity density and strength, as reflected in mobility, as an independent factor that must be included to predict the plateau width.","pith_inferences":["A natural extension the paper does not report is a gate-tunable experiment on a single region, which would separate the density effect from adsorbate-induced disorder completely; if $\\Delta B$ still grows linearly with gate density in an encapsulated device, the density effect is intrinsic.","The same oxygen adsorption/desorption protocol could map the spatial disorder landscape of a wafer by measuring $\\Delta B$ region by region, turning plateau width into a local diagnostic.","The relation between $n_c$, mobility, and impurity density suggests that plateau-width measurements could complement Hall mobility as a faster characterization of scattering in graphene devices.","Because molecular oxygen only weakly perturbs the graphene lattice, these results may extend to other weakly adsorbed species, implying that ambient history is a general control knob for quantum transport in air-sensitive 2D materials."],"forward_implications":["At fixed disorder, raising carrier density widens the ν=6 plateau because the Landau levels move apart in field, so a density-tunable graphene device has a predictable operating window for quantized Hall resistance.","Sample storage history, by changing adsorbate coverage, is a hidden variable in quantum Hall metrology: identical devices aged differently will have different plateau widths at the same nominal density.","The critical density at which the plateau disappears, $n_c$, is itself a disorder metric: cleaner regions sustain the plateau down to lower carrier densities.","Devices designed with controlled, weak disorder should exhibit broader plateaus and therefore allow quantum resistance standards to operate at lower magnetic fields.","The Anderson-disorder simulation, despite being a scaled model, can be used to screen materials or treatments for their expected plateau width before fabrication."],"supporting_citations":[{"why":"Supplies the Kwant software used for the Landauer-Büttiker conductance calculations in the tight-binding simulations.","marker":"[20]"},{"why":"Provides the mobility-versus-density model used to fit the experimental mobility data with the exponent β.","marker":"[38]"},{"why":"Gives the scattering-rate relation τ ∝ n_imp^{-1}(A_sr U_0^2)^{-1} that connects mobility differences to impurity density and strength.","marker":"[39]"},{"why":"Provides the scalable tight-binding model for graphene used in the numerical Hall bar simulations.","marker":"[40]"},{"why":"Supports the use of Anderson disorder as a model for uncorrelated short-range disorder in quantum transport in graphene.","marker":"[41]"},{"why":"Reports the adsorption energy of oxygen molecules on graphene, supporting the weakly physisorbed electron-acceptor picture.","marker":"[23]"},{"why":"Shows atmospheric oxygen binding and hole doping in graphene, supporting the storage-induced carrier density changes.","marker":"[25]"},{"why":"Demonstrates the detection of individual gas molecules adsorbed on graphene, underpinning the sensitivity of graphene transport to adsorbates.","marker":"[36]"}],"fun_headline_variants":["Graphene quantum Hall plateau width: density and disorder both matter","Disorder, not just density, controls graphene quantum Hall plateaus","Quantum Hall plateau width in graphene hinges on disorder plus doping","Two factors set graphene's quantum Hall plateau width, not just density","Graphene's quantum Hall plateau width: both doping and disorder decide"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument assumes that storing the sample in vacuum or air changes only how many electrons the graphene holds, not how strongly it scatters them: molecular oxygen is taken to sit on the surface without directly becoming a scatterer, and the two device regions are assumed to contain the same kind of impurities, just at different densities.","fun_headline_variants_meta":{"raw":{"variants":["Graphene quantum Hall plateau width: density and disorder both matter","Disorder, not just density, controls graphene quantum Hall plateaus","Quantum Hall plateau width in graphene hinges on disorder plus doping","Two factors set graphene's quantum Hall plateau width, not just density","Graphene's quantum Hall plateau width: both doping and disorder decide"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000513,"raw_usage":{"total_tokens":2568,"prompt_tokens":1094,"completion_tokens":1474,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":710,"completion_tokens_details":{"reasoning_tokens":1384}},"tokens_in":710,"tokens_out":1474,"duration_ms":10664,"temperature":1.0,"reasoning_tokens":1384,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T20:39:35.423374+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the width of the ν=6 plateau on an encapsulated graphene Hall bar where carrier density is tuned continuously with a gate while the surface is kept clean, so disorder cannot change with doping. If $\\Delta B$ fails to grow with density, or if an oxygen-exposed device shows a different width at the same gate-set density, the claim that density and disorder act jointly would be contradicted.","supporting_citations":[{"cited_title":"Kwant: a soft- ware package for quantum transport","cited_arxiv_id":null,"evidence_quote":"Supplies the Kwant software used for the Landauer-Büttiker conductance calculations in the tight-binding simulations."},{"cited_title":"Quan- tum transport in two-dimensional graphite sys- tem","cited_arxiv_id":null,"evidence_quote":"Provides the mobility-versus-density model used to fit the experimental mobility data with the exponent β."},{"cited_title":"Gosling, Oleg Makarovsky, Feiran Wang, Nathan D","cited_arxiv_id":null,"evidence_quote":"Gives the scattering-rate relation τ ∝ n_imp^{-1}(A_sr U_0^2)^{-1} that connects mobility differences to impurity density and strength."},{"cited_title":"Scalable tight-binding model for graphene","cited_arxiv_id":null,"evidence_quote":"Provides the scalable tight-binding model for graphene used in the numerical Hall bar simulations."},{"cited_title":"Quantum transport in disor- dered graphene: A theoretical perspective","cited_arxiv_id":null,"evidence_quote":"Supports the use of Anderson disorder as a model for uncorrelated short-range disorder in quantum transport in graphene."},{"cited_title":"Ajayan, Junichiro Kono, Masayoshi Tonouchi, and Iwao Kawayama","cited_arxiv_id":null,"evidence_quote":"Reports the adsorption energy of oxygen molecules on graphene, supporting the weakly physisorbed electron-acceptor picture."},{"cited_title":"Flynn, and Louis E","cited_arxiv_id":null,"evidence_quote":"Shows atmospheric oxygen binding and hole doping in graphene, supporting the storage-induced carrier density changes."},{"cited_title":"Schedin, A","cited_arxiv_id":null,"evidence_quote":"Demonstrates the detection of individual gas molecules adsorbed on graphene, underpinning the sensitivity of graphene transport to adsorbates."}],"review_version":1}