{"id":"62f099f7-26c4-461a-a8ee-c9b363543b6e","arxiv_id":"2607.02822","paper_version":1,"verdict":"ACCEPT","confidence":"HIGH","novelty_score":7.0,"correctness_risk":"low","formal_verification":"none","parameter_count":4,"one_line_summary":"Global twist configuration (helical vs alternate) selects moiré–moiré stacking registry via local rotation matching of the shared graphene layer, producing diverse commensurate double-moiré phases and topological flat bands below the magic angle.","lead":"Twisted bilayer graphene on h-BN forms diverse locked double-moiré patterns whose local stacking is fixed by whether the twists are helical or alternate. The finding gives a practical rule for designing multilayer moiré devices with topological flat bands even below the magic angle.","discovery_kind":"extension","skeptic_critique":{"model":"grok-4.5","headline":"No significant objection identified","rationale":"The paper’s strongest claim is structural and is corroborated by direct real-space imaging plus matching continuum simulations for both twist configurations and for both C3z-symmetric and strained domains. The free parameters noted by the reader affect the quantitative extent of the phase diagram but not the qualitative selection rule that is already fixed by the sign of the local rotational fields. Because that selection rule is the load-bearing element of the strongest claim, and because it survives the most obvious parameter variation, no adjustment to the ACCEPT verdict is warranted. The proposed bare-Lamé re-minimization is a low-cost check that would further confirm robustness without requiring new experiments.","tokens_in":20470,"tokens_out":517,"duration_ms":4930,"concrete_test":"Re-run the continuum energy minimization of SI §II.C for the helical (0.60°, 0.62°) and alternate (0.39°, −0.97°) cases using the bare (unenhanced) Lamé parameters for layers 2–5; if the AA-on-AB′ (helical) and AA-on-AA′/BA′ (alternate) registries remain the energy minima and the simulated current-contrast patterns still match Figs. 1d,g and 2b,f, the rotation-matching claim is robust to the ad-hoc scaling.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central structural claim—that global twist configuration (helical vs alternate) uniquely selects the local tBG–Gr/h-BN stacking registry via local rotation matching of the shared graphene layer, and that this registry plus twist/strain produces the observed family of commensurate domains—is directly supported by C-AFM images that match continuum simulations across multiple samples and configurations (Figs. 1–3). The reader’s weakest assumption (the |ΔL|/max < 0.1 window and the 2.4× Lamé enhancement) is a free-parameter choice that can shift phase-diagram boundaries, but it is not load-bearing for the registry selection itself: the rotation-matching argument is already visible in the experimental contrast patterns and in the sign-reversed rotational fields of the isolated-moiré decompositions (Fig. 2), independent of the precise numerical threshold. Electronic flat-band predictions are presented as theory-only and do not underwrite the structural claim. No internal inconsistency or experimental–theory mismatch that would overturn the strongest claim is apparent.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.5","summary":"The manuscript reports that in twisted bilayer graphene on h-BN, the global twist configuration (helical versus alternate) uniquely selects the local spatial registry between the coexisting tBG and Gr/h-BN moiré lattices through local rotation matching of the shared graphene layer. Combining C-AFM imaging with continuum elasticity simulations, the authors show that this registry, together with twist angle and strain, stabilizes a family of commensurate double-moiré domains—from C3z-symmetric structures at quantized period ratios (1:1, 2:1, 3:1, √7:1, 2:√3) to strained, symmetry-broken yet registry-preserving phases—organized into sub-micrometer domains with collective boundary sliding. Continuum band-structure calculations further predict that the configuration-dependent registries open gaps and stabilize topological flat bands with distinct valley Chern numbers at twist angles below the magic angle.","tokens_in":20742,"tokens_out":1444,"duration_ms":18350,"significance":"If the structural claims hold, the work supplies a concrete, experimentally grounded organizing principle—local rotation matching mediated by a shared layer—for multi-moiré reconstruction, going beyond single-interface moiré physics and prior limited observations of local commensuration in tBG/h-BN. The systematic mapping of helical versus alternate registries, the commensurate-domain phase diagrams in twist and strain, and the mesoscale domain morphology constitute a predictive framework that is transferable in principle to other multilayer van der Waals stacks. The continuum simulations reproduce experimental C-AFM contrast and domain patterns across multiple samples and twist conditions (Figs. 1–3), which is a clear strength. The electronic flat-band and Chern-number predictions are falsifiable by STS/Landau-level spectroscopy and, if confirmed, would open a design route to topological flat bands below the magic angle controlled by twist configuration rather than angle alone.","major_comments":[{"comment":"SI Section II.C states that effective Lamé parameters for layers 2–5 are enhanced by a factor 2.4 relative to bare graphene/h-BN values “because the graphene/h-BN moiré pattern is relatively rigid.” This factor is free and load-bearing for the quantitative match of domain morphology and for the strained configurations in Fig. 3e,g,h and SI Fig. S4. The manuscript should either (i) provide an independent estimate or literature bound for the enhancement, or (ii) show a sensitivity analysis demonstrating that the registry selection (AA on AB′ vs AA on AA′/BA′) and the topology of the phase diagrams in Fig. 3l,m survive under bare or moderately varied Lamé parameters. Without this, the claimed generality of the phase boundaries remains under-constrained.","section":null},{"comment":"SI Section II.B defines the commensurate-domain window by |ΔL|/max(|LtBG|,|LGr/BN|)<0.1, motivated by prior twisted-trilayer work where domains appeared near ~0.20. The threshold directly paints the colored regions of Fig. 3l and the strain windows of Fig. 3m. The text should state how the predicted windows change if the cutoff is varied (e.g., 0.05–0.20) and whether any experimentally observed domain (notably the √7:1 structure of Fig. 3f, already noted as slightly outside the window) would fall in or out. A short robustness check would make the phase diagrams predictive rather than post-hoc.","section":null},{"comment":"Abstract, Introduction, and Conclusion present “theoretically predicted topological flat bands below the magic angle” and configuration-dependent Chern numbers as a central outcome of moiré–moiré reconstruction. Fig. 5 and SI Fig. S7 are continuum calculations only; no STS, Landau-level, or transport data are shown. The structural claim does not depend on these bands, but the framing does. Either (i) clearly separate the electronic results as theoretical predictions with specified experimental tests (as briefly suggested in the Conclusion), or (ii) temper the abstract/title-level language so that the primary, experimentally supported result remains the structural registry and domain formation.","section":null}],"minor_comments":[{"comment":"Fig. 1d,g captions and main text use mixed notation for angles and period ratios (e.g., (θtBG, θGr/BN) and LtBG:LGr/BN); ensure consistent symbols between main text, figure labels, and SI Tables S1–S2.","section":null},{"comment":"Fig. 2 bottom panels: the color scale for the rotational component Ω is described in the caption but not shown as a color bar; adding a bar would aid quantitative reading of clockwise/counterclockwise magnitudes.","section":null},{"comment":"SI Section V and Fig. 3i: the claim that near θGr/BN≈0° rotational matching is suppressed is important for the limits of the mechanism; a short quantitative estimate of the rotational versus dilatational energy scales (or a reference to Krisna & Koshino) in the main text would help non-specialist readers.","section":null},{"comment":"Methods and SI Table S3 list contact forces and biases; a brief statement that tip-induced strain was checked not to alter the observed registry (beyond the 90 nN sliding experiment of Fig. 4c–d) would strengthen the experimental section.","section":null},{"comment":"References 26–28 and related double-moiré STM/C-AFM works are cited; a one-sentence comparison in the Introduction clarifying what is new relative to Lai et al. (Nat. Mater. 2025) and Li et al. (PRL 2024)—namely the helical/alternate registry dichotomy and the rotation-matching mechanism—would sharpen novelty for the reader.","section":null},{"comment":"Typographical/encoding artifacts appear in several places (e.g., “-!\"#:-#$/\"&=2:1”, “|∆?|/max”, “ê#5-symmetric” in SI). Clean these for production.","section":null}],"recommendation":"minor_revision","confidential_remarks":"The structural core is solid and the C-AFM–theory agreement is convincing; the free-parameter issues (Lamé factor, |ΔL| threshold) affect quantitative phase boundaries more than the registry mechanism itself. I would not block acceptance over them if the authors add a short sensitivity discussion. The electronic flat-band claims are interesting but purely theoretical; depending on journal scope, the editor may want them de-emphasized in the abstract so the paper is judged primarily on the structural advance. Fit for a high-quality condensed-matter/materials journal is good."},"author_rebuttal":null,"desk_editor":{"model":"grok-4.5","letter":"The real advance here is simple and useful: helical versus alternate twist of tBG on h-BN deterministically picks which local stacking registry the two moirés lock into, via rotational relaxation of the shared graphene layer. Prior STM work saw local self-alignment in limited windows; this paper maps both configurations systematically, shows the registry is preserved even under strain, and builds phase diagrams for C3z-symmetric and strained commensurate domains.\n\nWhat they do well is the C-AFM–continuum match. Figs. 1–3 and the SI decompositions make the rotation-matching argument concrete: the sign of the local rotation field in the shared layer flips with twist sense, and the observed AA-on-AB′ versus AA-on-AA′/BA′ registries follow. Domain morphology at larger scales and the collective sliding of AA chains along boundaries are clean extras. The electronic calculations are standard continuum models; they show configuration-dependent topological flat bands below the magic angle as a prediction, not a claim of measurement. That is honest.\n\nSoft spots are real but secondary. The 2.4× Lamé boost for the Gr/h-BN layers and the |ΔL|/max < 0.1 window are free parameters that can move the phase-diagram edges; they are not required for the registry selection itself, which is already in the images and the isolated-moiré rotation fields. Twist angles are estimated from the images rather than independently measured, and there is no public code or raw data. None of that overturns the structural claim.\n\nThis is for people building or modeling multi-moiré stacks who need a concrete rule for registry and domain formation. The math and citation pattern look solid (Koshino-group continuum machinery plus the relevant STM literature). I would send it to referees; it deserves a serious read and will be cited by anyone working on double-moiré devices or reconstruction.","headline":"Solid experimental–theory package that turns helical vs alternate twist into a usable design rule for double-moiré registry and domains.","tokens_in":21363,"tokens_out":477,"would_cite":true,"duration_ms":4689,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.5","headline":"Global twist configuration selects how two moiré lattices stack by matching local rotations in the shared graphene layer, producing diverse commensurate double-moiré phases and topological flat bands below the magic angle.","keywords":["moiré–moiré reconstruction","twisted bilayer graphene","graphene/h-BN","local rotation matching","commensurate double-moiré","topological flat bands","helical vs alternate twist","conductive AFM"],"falsifier":"Conductive-AFM or STM maps of helical and alternate devices at the same tBG twist angle but opposite Gr/h-BN twist signs that show identical (rather than opposite) AA-to-AB′ versus AA-to-AA′ registries, or scanning-tunneling spectroscopy that fails to find the predicted gap openings and configuration-dependent Chern flat bands below the magic angle.","tokens_in":21360,"feed_emoji":"⚛️","tokens_out":698,"duration_ms":6219,"temperature":0.7,"pith_summary":"When two different moiré patterns share a graphene sheet, their lattice relaxations are not independent. This paper shows that the global twist geometry—helical (same sense) or alternate (opposite sense)—forces the two patterns into one unique local stacking registry so that the local rotations they induce in the shared layer reinforce each other. That registry, together with the twist angles and any strain, locks the system into extended commensurate double-moiré domains that can be C3-symmetric or deliberately symmetry-broken. At larger scales the domains form well-ordered sub-micrometer tiles whose boundaries slide collectively. The same registries open gaps and isolate topological flat bands with configuration-dependent Chern numbers even below the magic angle of twisted bilayer graphene alone. The claim is that moiré–moiré reconstruction, driven by local rotation matching, is a general design principle for multilayer van der Waals stacks.","feed_headline":"Twist sense locks two moiré lattices into unique stacking","feed_subtitle":"Local rotation matching of the shared graphene creates commensurate domains and flat bands below the magic angle","key_machinery":"Local rotation matching: the two moiré lattices arrange so that the local rotational displacements they induce in the shared graphene layer point in the same direction, making that registry energetically preferred and locking the commensurate domains.","core_discovery":"In tBG/h-BN the global twist configuration (helical versus alternate) uniquely selects the local spatial registry between the triangular tBG moiré and the hexagonal graphene/h-BN moiré through local rotation matching of the shared graphene layer. That registry, combined with twist angle and strain, stabilizes a family of commensurate double-moiré domains ranging from C3z-symmetric period-ratio phases to strained, symmetry-modified structures, and produces topological flat bands with distinct Chern numbers below the magic angle.","pith_inferences":[],"forward_implications":[],"fun_headline_variants":["Global twist locks twin moirés into unique local stacking registry","Helical vs alternate twist selects commensurate double-moiré phases","Shared graphene rotation matches tBG and graphene/h-BN moirés","Twist sense and strain stabilize diverse double-moiré domains","Moiré-moiré reconstruction yields flat bands below magic angle"],"cache_read_input_tokens":16512,"weakest_assumption_plain":"The continuum model assumes that a simple near-commensurability window plus an ad-hoc stiffening of the graphene/h-BN elastic constants are enough to decide which twist and strain combinations form stable domains; if the true elastic or binding energies differ, the predicted phase boundaries and the claimed generality of rotation matching shift.","fun_headline_variants_meta":{"raw":{"variants":["Global twist locks twin moirés into unique local stacking registry","Helical vs alternate twist selects commensurate double-moiré phases","Shared graphene rotation matches tBG and graphene/h-BN moirés","Twist sense and strain stabilize diverse double-moiré domains","Moiré-moiré reconstruction yields flat bands below magic angle"]},"model":"grok-4.5","effort":"low","cost_usd":0.00738,"raw_usage":{"total_tokens":1802,"prompt_tokens":762,"num_sources_used":0,"completion_tokens":91,"cost_in_usd_ticks":73800000,"prompt_tokens_details":{"text_tokens":762,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":949,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":762,"tokens_out":91,"duration_ms":6581,"temperature":1.0,"reasoning_tokens":949,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-12T06:48:39.779469+00:00","model_set":{"reader":"grok-4.5"},"falsifier":"Conductive-AFM or STM maps of helical and alternate devices at the same tBG twist angle but opposite Gr/h-BN twist signs that show identical (rather than opposite) AA-to-AB′ versus AA-to-AA′ registries, or scanning-tunneling spectroscopy that fails to find the predicted gap openings and configuration-dependent Chern flat bands below the magic angle.","supporting_citations":[],"review_version":1}