{"id":"fe6fb0b4-7601-4b1c-b8a7-335548c64b47","arxiv_id":"2502.01592","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Graphene intercalation stabilizes the 2/3 monolayer bismuthene phase on SiC, protects it from air oxidation, and preserves its topological edge states and band structure.","lead":"Researchers grew a protective graphene layer over bismuthene, a two-dimensional topological insulator, by sliding bismuth atoms between the graphene and its silicon carbide support. The cap keeps bismuthene from oxidizing in air while preserving its electronic edge states, which could enable real-world devices based on this room-temperature quantum spin Hall material.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Topological-integrity claim rests on an indirect gap lower bound; ARPES cannot distinguish a trivial gap from the inverted bismuthene gap without direct conduction-band or spin-texture evidence.","rationale":"After reading the paper, I find the intercalation and air-stability claims convincing: LEED, STEM, bias-dependent STM, and XPS provide convergent structural and chemical evidence. The central remaining uncertainty is whether the intercalated phase retains the topological character of pristine bismuthene. The authors support this with two indirect signatures: the K-doping saturation of the VBM at −230 meV (Fig. 4) and the STM edge modulation (Fig. 5). Both are necessary but not sufficient. The K-doping experiment is particularly load-bearing because it is the only quantitative gap estimate; however, it never measures the conduction band and does not rule out trivial explanations for the saturation, such as K-induced work-function changes or Fermi-level pinning at in-gap states. The non-dispersive K 4s state at −760 meV highlights the presence of alkali-derived states that could alter the electronic structure. The STM edge signature is also not uniquely topological. The paper's own language ('strongly suggesting', 'likely just an extremely conservative understatement') acknowledges the indirectness. To make the central claim of 'preserved topological integrity' robust, a direct observation of the conduction band or a spin-resolved/texture measurement of the inverted bands is needed. This does not invalidate the work, but it supports the CONDITIONAL verdict already given.","tokens_in":7170,"tokens_out":6198,"duration_ms":60935,"concrete_test":"Perform a control K-dosing series on graphene/SiC without bismuthene, tracking the graphene Dirac-point shift and sample work function. If the bismuthene VBM saturates at −230 meV at the same K dose at which the graphene Dirac point (or work function) saturates, then the saturation is a doping/interface limit, not the bulk CBM, and the gap lower bound is invalid. Separately, use spin-resolved ARPES at KBi on the K-doped intercalated sample: the topologically inverted px/py bands must exhibit the same spin-momentum locking as pristine bismuthene; a trivial gap would show a different or absent spin texture.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The most load-bearing step is the inference in Fig. 4 that saturation of the bismuthene valence-band maximum at −230 meV under potassium dosing yields a valid lower bound on the topological gap and thus demonstrates preserved topological integrity. This inference assumes (i) that K electrons transfer into the bismuthene bulk bands rather than into graphene, interface states, or the K layer; (ii) that saturation marks the Fermi level reaching the conduction-band minimum, not a work-function or band-bending limit; and (iii) that no in-gap states pin EF. The paper never observes the conduction band, so a trivial gap of 230 meV, a defect-pinned EF, or a K-induced potential shift would produce the identical ARPES signature. The non-dispersive K 4s feature at −760 meV (Fig. 4c) shows that K-derived states are present near the valence band, making K–bismuthene hybridization a real possibility. The STM edge modulations in Fig. 5, cited as evidence for helical edge states, are not unique: trivial step-edge states or Friedel oscillations can also produce a √3-modulated signal. The structural intercalation and oxidation protection are well supported; the weak link is the topological-integrity inference, which is indirect.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This manuscript reports an intercalation protocol that places the 2/3-monolayer bismuthene phase between SiC(0001) and an epitaxial graphene overlayer. The structural identity of the intercalated film is established through LEED, cross-sectional STEM, and bias-dependent STM, and the valence band is compared with literature DFT calculations. Potassium dosing in ARPES is used to infer a 230 meV lower bound for the band gap from saturation of the bismuthene valence band, and STM images at SiC step edges are presented as evidence for conserved topological edge states. Finally, XPS shows that the graphene cap suppresses bismuth oxidation upon oxygen and air exposure. The authors conclude that graphene intercalation preserves the structural and topological integrity of bismuthene while making it air-stable.","tokens_in":7503,"tokens_out":5938,"duration_ms":56680,"significance":"The structural intercalation and air-stability results are well supported and practically valuable: the hydrogen-assisted protocol appears to solve a previously elusive preparation problem, and the XPS comparison between pristine and capped bismuthene directly demonstrates oxidation protection. The paper also benefits from comparison with independent literature DFT calculations rather than fitting its own model, and the LEED/STEM/STM evidence for the 2/3-monolayer phase is convincing. The main weakness is that the topological-integrity claim rests on an indirect gap estimate and on edge-state signatures that are not unique; those parts of the argument need strengthening or explicit reframing.","major_comments":[{"comment":"The inference that valence-band saturation at about -230 meV gives Egap >= 230 meV assumes that potassium electrons are transferred into the bismuthene conduction band and that no in-gap states pin the Fermi level. The conduction band is never observed, so the identical ARPES signature would arise from a trivial 230 meV gap, from a defect-pinned Fermi level, or from a potassium-induced electrostatic shift. The non-dispersive K 4s feature at -760 meV in Fig. 4c shows that K-derived states are occupied near the valence band, so the simple rigid-band charge-transfer picture is not established. Please either provide direct evidence of conduction-band occupation or spin texture, or explicitly state the assumptions under which the 230 meV value is a valid lower bound.","section":"Fig. 4 and accompanying text"},{"comment":"The enhanced tunneling signal with sqrt(3) periodicity at SiC step edges is not unique to helical topological edge states; trivial step-edge states, Friedel oscillations, or boundary standing waves can produce similar constant-current modulations. The comparison with pristine bismuthene in Fig. 5a is helpful, but a single bias condition in Fig. 5b cannot establish spin-momentum locking. Please add energy-dependent or wavevector-resolved evidence, or soften the wording from 'strongly suggesting the conservation' to 'consistent with the conservation' of topological edge states.","section":"Fig. 5 and surrounding text"},{"comment":"The red DFT lines in Fig. 3a are calculated for pristine bismuthene, not for the graphene-intercalated stack studied here. Since the central claim is that topological integrity is preserved under intercalation, a DFT calculation of the actual graphene/bismuthene/SiC heterostructure would close the loop by checking that the band inversion and gap survive the presence of the graphene overlayer and the observed doping. If such a calculation is already available, it should be shown; if not, the manuscript should state that the topological conclusion is inferred from structural identity and the pristine-phase literature rather than computed for the intercalated system.","section":"Comparison with DFT in Fig. 3"}],"minor_comments":[{"comment":"The text refers to the saturation as occurring in Fig. 4c, while the caption describes the doping saturation in panel d); please correct the cross-reference.","section":"Fig. 4"},{"comment":"The sentence 'bismuth occupying the the T1 site' contains a duplicated article; please remove the second 'the'.","section":"Page 3"},{"comment":"The statement that hydrogen is the critical missing component in previous intercalation attempts is not supported by a control experiment without hydrogen; please either add such a control or soften the claim to a working hypothesis.","section":"Preparation protocol"},{"comment":"The high-symmetry path notation 'MBi gamma KBi MBi' is not defined in the text; a brief definition would help readers unfamiliar with the bismuthene Brillouin zone.","section":"Notation"},{"comment":"The manuscript mixes 'two-third monolayer' in the abstract with '2/3 monolayer' in the main text; please use one consistent convention.","section":"Abstract and text"},{"comment":"The Fermi-surface pocket used to extract bismuthene hole doping may overlap with the graphene replica discussed in the same figure; a sentence quantifying any possible contamination would strengthen the doping estimate.","section":"Fig. 3b"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is well within the scope of a condensed-matter/materials journal, and the structural and oxidation-protection results are publishable contributions if the topological-integrity claims are either strengthened or appropriately qualified. I have no concerns about citation practice or novelty disclosure; the main risk is overclaiming topological integrity from indirect evidence. If the authors cannot provide direct gap or edge-state measurements, softening the relevant statements should be sufficient, so I see this as a major revision rather than a rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Best to start with the concrete achievement: the authors have intercalated the 2/3 monolayer bismuthene phase below epitaxial graphene on SiC and shown that the graphene cap protects it from oxidation. The structural evidence is strong—LEED and STM show the expected (√3×√3)R30° bismuthene superstructure, STEM places the bismuth at T1 sites, and the ARPES px/py bands match the pristine bismuthene calculation. The oxidation test is clean and directly demonstrates the practical payoff. The hydrogen desorption step is a useful new protocol detail, and it explains why earlier bismuth intercalation runs got trivial phases. That is a real advance, and I would expect the experimental core to hold up.\n\nThe soft spot is the topological-integrity claim. The 230 meV lower bound from K-doped VBM saturation is an indirect inference. ARPES never shows the conduction band, so a trivial gap, a defect-pinned Fermi level, or a K-induced potential shift would produce the same saturation signature. The K 4s feature at -760 meV is a reminder that potassium is present near the bismuthene bands, so hybridization is not excluded. On its own, Figure 4 does not prove that the gap is inverted. The edge-state modulation in Figure 5 is also not unique—trivial step-edge states could give a √3 pattern.\n\nThat said, the claim is not riding on that lower bound alone. The band dispersion across the measured range matches the pristine topological phase, including Rashba splitting, and the structural parameters match the known bismuthene phase. Together these make the topological-integrity conclusion plausible, and the gap lower bound is honestly labeled as a lower bound. I would only ask the authors to soften the wording—Fig. 4d's 'conserves its topological gap' is stronger than the data justify—and to acknowledge the alternative explanations for saturation and edge modulations. A control study for the hydrogen step would also be good, but that is a minor point.\n\nWho is this for: anyone working on intercalation of 2D topological materials or on protecting air-sensitive monolayers. The paper merits serious review; the structural and oxidation results are solid, and the topological caveat is fixable with revised language. I would send it to a good referee, with the expectation of a revised version that is more careful about the gap inference.","headline":"Convincing intercalation and oxidation protection; the topological gap claim is indirect but the circumstantial case is strong.","tokens_in":7935,"tokens_out":3499,"would_cite":true,"duration_ms":32089,"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":"Graphene intercalation preserves the topological state of bismuthene and shields it from oxidation in air.","keywords":["bismuthene","quantum spin Hall insulator","graphene intercalation","topological edge states","oxidation protection","scanning tunneling microscopy","angle-resolved photoemission spectroscopy","two-dimensional topological insulator"],"falsifier":"Measure nonlocal or two-terminal transport through the buried edge channels at room temperature: quantized helical edge conductance that persists after air exposure would confirm the preserved topological phase, while a bulk-dominated conductivity or the complete absence of edge conductance would show that the STM modulation and 230 meV lower bound were not sufficient evidence.","tokens_in":6955,"feed_emoji":"🛡️","tokens_out":9444,"duration_ms":73169,"temperature":0.7,"pith_summary":"Bismuthene, a two-thirds monolayer of bismuth on SiC(0001), is a two-dimensional topological insulator with a large gap and helical edge states, but it oxidizes readily in air, which has confined experiments to ultrahigh vacuum. This paper reports a way around that barrier: intercalating bismuthene between the SiC substrate and a single graphene sheet, so the graphene acts as a protective cap. The authors show that with the correct growth protocol, which includes a hydrogen annealing step, the 2/3 monolayer bismuthene phase forms intact under graphene, retains its topological band structure and edge-state signatures, and no longer oxidizes even when exposed to air. If correct, this turns bismuthene from a vacuum-only curiosity into a material that can be handled, patterned, and measured outside the growth chamber.","feed_headline":"Graphene cap shields bismuthene from air, preserving its topology","feed_subtitle":"Hydrogen-assisted intercalation yields the 2/3 monolayer bismuthene phase under graphene, opening ex-situ devices.","key_machinery":"The key object is bismuthene, the two-thirds monolayer bismuth phase on SiC(0001), a (√3×√3)R30° superstructure whose flat honeycomb geometry, strong spin-orbit coupling, and substrate-mediated orbital filtering open an 800 meV topological gap. The enabling step in this work is hydrogen-assisted intercalation: annealing with hydrogen passivates the silicon dangling bonds and decouples the graphene overlayer, and a final 550 °C desorption drives the bismuth coverage to the 2/3 monolayer phase rather than the 1/3 or full-monolayer structures seen in earlier attempts. The evidence chain then rests on four measurements: LEED/STEM for the atomic registry, STM for the moiré pattern and edge modulations, ARPES plus potassium doping for the band structure and gap lower bound, and XPS for oxidation resistance.","core_discovery":"The authors claim to stabilize the 2/3 monolayer bismuthene phase as an intercalant between 4H-SiC(0001) and zero-layer graphene. After depositing bismuth and annealing at 350 °C, a subsequent desorption at 550 °C in a hydrogen atmosphere removes excess bismuth and leaves a single bismuth layer sitting at the T1 sites directly above silicon, in the same (√3×√3)R30° arrangement as pristine bismuthene. LEED and cross-sectional STEM confirm the vertical stacking; STM resolves both the graphene overlayer and the bismuthene lattice, together with a moiré pattern of about 3.2 nm. ARPES shows the characteristic px/py bands of bismuthene, matching density functional calculations for the pristine phase, and potassium doping saturates the valence band at about –230 meV below the Fermi level, giving a conservative lower bound for the preserved topological gap. STM at SiC step edges shows the same enhanced signal and √3 modulation seen at pristine bismuthene edges, which the authors interpret as conserved topological edge states. XPS shows that the Bi 4f core levels remain unchanged after exposure to oxygen and ambient air, unlike the pristine, uncapped film, which develops oxidation satellites.","pith_inferences":["Extending this result, one could map the bismuth phase diagram against hydrogen partial pressure and annealing time; the paper identifies hydrogen as the missing ingredient but does not show how sharp the 2/3 monolayer window is.","Because the finished stack has n-doped graphene and p-doped bismuthene, it behaves as a built-in heterojunction; gating the stack would let one tune the relative doping, possibly reaching bismuthene charge neutrality without potassium deposition.","The edge-state evidence here is topographic rather than transport-based; a direct nonlocal conductance measurement through the buried helical edge would test topological protection more stringently than STM modulation alone."],"forward_implications":["The 2/3 monolayer bismuthene phase can be intercalated beneath one graphene sheet on SiC(0001), giving a buried quantum spin Hall insulator with the same (√3×√3)R30° registry as the pristine film.","The preserved valence-band saturation at about –230 meV places the topological gap at or above room-temperature requirements, so the buried edge states should survive thermal excitation.","The graphene cap blocks oxidation of bismuthene in oxygen and ambient air, so samples grown in ultrahigh vacuum can be taken out for ex situ experiments and device processing.","The hydrogen desorption step is the decisive control that selects the 2/3 monolayer phase instead of the metallic or 1/3 monolayer trivial films reported in earlier intercalation attempts."],"supporting_citations":[{"why":"defines pristine bismuthene's 2/3 monolayer structure and 800 meV topological gap that this work aims to preserve.","marker":"[14]"},{"why":"supplies the DFT band structure used as the reference for the measured px/py dispersion.","marker":"[13]"},{"why":"demonstrates graphene intercalation for the related topological insulator indenene, the methodological template extended here.","marker":"[20]"},{"why":"introduces hydrogen intercalation to decouple epitaxial graphene from SiC, the key step adapted for bismuth.","marker":"[15]"},{"why":"provides the hydrogen etching and passivation procedure for the SiC substrate preparation.","marker":"[24]"},{"why":"documents earlier bismuth intercalation attempts that yielded only trivial phases, the failure the new protocol overcomes.","marker":"[21]"},{"why":"shows another prior bismuth intercalation that produced a topologically trivial structure, establishing the difficulty.","marker":"[22]"},{"why":"establishes the √3-periodic STM modulation at bismuthene armchair edges as the signature of topological edge states.","marker":"[32]"},{"why":"gives the room-temperature gap threshold, Egap≥3.5 kBT, used to argue the 230 meV lower bound is sufficient.","marker":"[11]"}],"fun_headline_variants":["Graphene cap shields bismuthene from air, keeping topology intact","Air-stable bismuthene: graphene intercalation preserves topological gap","Hydrogen unlocks air-stable bismuthene under a graphene layer","Bismuthene's 800 meV gap survives graphene capping in air","Graphene wraps bismuthene, blocking oxidation, preserving spin Hall"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing assumption is that the valence-band stop at about 230 meV below the Fermi energy gives a true lower bound for the topological gap, and that the √3-periodic bright lines seen at silicon-carbide step edges belong to helical topological edge states; if either signal is actually a trivial edge effect or a potassium-doping artifact, the claim that bismuthene's topological order survives the graphene cap is not proven.","fun_headline_variants_meta":{"raw":{"variants":["Graphene cap shields bismuthene from air, keeping topology intact","Air-stable bismuthene: graphene intercalation preserves topological gap","Hydrogen unlocks air-stable bismuthene under a graphene layer","Bismuthene's 800 meV gap survives graphene capping in air","Graphene wraps bismuthene, blocking oxidation, preserving spin Hall"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00105,"raw_usage":{"total_tokens":4453,"prompt_tokens":1029,"completion_tokens":3424,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":645,"completion_tokens_details":{"reasoning_tokens":3323}},"tokens_in":645,"tokens_out":3424,"duration_ms":21247,"temperature":1.0,"reasoning_tokens":3323,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-09T14:51:40.270189+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure nonlocal or two-terminal transport through the buried edge channels at room temperature: quantized helical edge conductance that persists after air exposure would confirm the preserved topological phase, while a bulk-dominated conductivity or the complete absence of edge conductance would show that the STM modulation and 230 meV lower bound were not sufficient evidence.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"defines pristine bismuthene's 2/3 monolayer structure and 800 meV topological gap that this work aims to preserve."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"supplies the DFT band structure used as the reference for the measured px/py dispersion."},{"cited_title":"et al.Achieving environmental stability in an atomically thin quantum spin Hall insulator via graphene intercalation","cited_arxiv_id":null,"evidence_quote":"demonstrates graphene intercalation for the related topological insulator indenene, the methodological template extended here."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"introduces hydrogen intercalation to decouple epitaxial graphene from SiC, the key step adapted for bismuth."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"provides the hydrogen etching and passivation procedure for the SiC substrate preparation."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"documents earlier bismuth intercalation attempts that yielded only trivial phases, the failure the new protocol overcomes."},{"cited_title":"Intercalation of graphene on SiC(0001) via ion implantation","cited_arxiv_id":null,"evidence_quote":"shows another prior bismuth intercalation that produced a topologically trivial structure, establishing the difficulty."},{"cited_title":"S., Yang, S","cited_arxiv_id":null,"evidence_quote":"gives the room-temperature gap threshold, Egap≥3.5 kBT, used to argue the 230 meV lower bound is sufficient."}],"review_version":1}