{"id":"be29f204-04fb-4908-b96e-f03d10a68438","arxiv_id":"1908.04076","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"First-principles transport calculations predict that rotating the middle graphene layer in a trilayer spacer between metallic leads changes conductance by up to 1600%.","lead":"This paper uses computer simulations to show that twisting the middle graphene layer in a three-layer spacer between metal contacts can dramatically change how much current flows through it. The effect, called twisting resistance, reaches more than 1000% for copper contacts and can be tuned by switching the magnetization of nickel contacts.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Central TwR percentages rest on conductance at only two commensurate twist angles (21.8°, 27.8°); no evidence these bracket the continuous RTA dependence or the peak value.","rationale":"The reader identified the sparse sampling of twist angles as the weakest assumption, and I agree. The paper's headline numbers (TwR >1000% for Cu, 1600% for Ni/Cu) are computed at exactly two commensurate angles, selected for computational convenience, not for physical representativeness. The authors explicitly acknowledge this limitation in the text. If the conductance–angle relationship is non-monotonic, the maximum TwR could lie elsewhere, and the 'valve' claim—which implies tunable resistance versus angle—would be quantitatively wrong even if the effect exists at the sampled angles. The physical mechanism they propose (Dirac-state alignment) could actually produce a non-monotonic response because the band folding and density of states vary strongly with angle, especially across the range from 0° to 30°. Thus the two-point curve in Fig. 1(d) is not sufficient to establish the central quantitative claim. In addition, the use of a single-ζ basis is a further quantitative risk, but the primary load-bearing issue is the angle sampling. Since the reader's conditional verdict already reflects this uncertainty, I recommend no change: the paper merits conditional acceptance pending broader angular sampling (or at least theoretical justification that the chosen angles are representative) and basis-set convergence checks.","tokens_in":8195,"tokens_out":10642,"duration_ms":97426,"concrete_test":"Compute the zero-bias conductance for Cu/TTLG/Cu at an additional commensurate twist angle with a small unit cell, e.g., θ=13.17° (√19×√19), using the identical SIESTA/TranSIESTA settings and SZ basis as in the paper. If the TwR relative to θ=0° is not positive or if it exceeds the reported 3600% peak, the claim that the two sampled angles bracket the twist-angle dependence and the peak value fails.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central quantitative claim—TwR >1000% for Cu/TTLG/Cu and up to ~1600% for Ni/Cu/TTLG/Cu/Ni—derives from conductance values computed at only two commensurate twist angles, 21.787° and 27.8°, chosen solely because they have the smallest possible unit cells. The text explicitly states: 'Due to computational limitation we can only study two relative rotation angles.' From these three points (θ=0°, 21.8°, 27.8°), Fig. 1(d) plots a smooth RTA dependence and the abstract/conclusion generalize to 'depending on the angle of twist' and 'could reach more than 1000%'. The purported mechanism—alignment of the twisted layer's Dirac states with metal transmission states near K/M—does not imply a monotonic or single-peaked response; at other angles the conductance could be larger, smaller, or non-monotonic, and the maximum TwR is not established. Additionally, the calculations use a single-ζ (SZ) basis, which is of limited accuracy for graphene's π bands and tunneling decay; no basis-set convergence test is provided, so the absolute conductance values underpinning every TwR percentage are not quantitatively reliable. An internal inconsistency in the text—'almost two order of magnitude' increase yet TwR values of 3600% and 1000% (factors of ~37 and ~11)—underscores the fragility of these numbers.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript uses first-principles NEGF transport calculations (SIESTA/TranSIESTA) to study vertical conductance through tri-layer graphene in which the middle layer is rotated by 21.787° or 27.8° relative to the outer layers, sandwiched between Cu, Ni, or Ni/Cu leads. It defines a 'twisting resistance' TwR as the relative change in conductance between twisted and untwisted spacers, and reports TwR values above 1000% for Cu/TTLG/Cu, about -40% to 130% for Ni/TTLG/Ni depending on magnetization alignment, and up to about 200% (parallel) and 1600% (anti-parallel) for Ni/Cu/TTLG/Cu/Ni. The proposed mechanism is that twisting moves the middle graphene layer's Dirac states relative to the metal Fermi-surface transmission channels near the K and M points, thereby opening or closing vertical current paths.","tokens_in":8499,"tokens_out":6962,"duration_ms":63890,"significance":"If the claimed magnitudes hold, this would be an interesting and potentially useful mechanical-control mechanism for vertical spin transport: a rotationally actuated 'spinmechatronic' valve with large conductance switching and strong interplay with magnetoresistance. The study is genuinely predictive rather than fitting: the TwR percentages are computed directly from NEGF conductances with no parameter tuned to reproduce them, and the calculations build on established structural inputs from Karpan et al. The main limitation, however, is that the quantitative predictions rest on a very small number of computational geometries and on a minimal basis set; the central percentages should therefore be read as point estimates awaiting convergence and angle-sampling checks.","major_comments":[{"comment":"The central quantitative claims (abstract: TwR 'more than 1000%' for Cu/TTLG/Cu and 'up to 1600%' for Ni/Cu/TTLG/Cu/Ni anti-parallel; main text: TwR of 3600% and 1000% for the two angles) are computed at only two commensurate angles, 21.787° and 27.8°, selected solely because they have the smallest possible unit cells. The text explicitly states: 'Due to computational limitation we can only study two relative rotation angles.' With only three points (θ=0°, 21.787°, 27.8°), Fig. 1(d) cannot establish the claimed smooth RTA dependence or locate the maximum of TwR; the proposed mechanism (Dirac-state alignment near K/M) does not by itself rule out non-monotonic behavior at intermediate angles. Please supply additional angle points (e.g., 13.17°, 9.43°, or an approximate rigid-band model) or explicitly reframe the conclusions as point values rather than as a full RTA dependence.","section":"Device setup and Fig. 1(d)"},{"comment":"The transport calculations use a single-zeta (SZ) numerical atomic-orbital basis for all species, and no convergence tests with respect to basis size or transmission k-grid are reported. The manuscript's quantitative claims depend on absolute conductances that span several orders of magnitude (down to 10^-5 e2/h in Fig. 1(d)); single-zeta bases are known to be inadequate for accurately describing graphene's π bands and exponential tunneling decay. Please provide at least one basis-set convergence check (e.g., DZP for the θ=0° and θ=21.787° Cu junctions) and corresponding transmission-k-grid convergence, to demonstrate that the reported TwR values are not artifacts of basis-set incompleteness.","section":"Computational methods (basis set and convergence)"},{"comment":"The text in the Cu/TTLG/Cu paragraph states that the conductance 'increases of almost two order of magnitude' for θ=21.787° relative to the untwisted case, but then quotes TwR of about 3600% and 1000% for the two angles; those percentages correspond to conductance ratios of roughly 37 and 11, not factors near 100. This is internally inconsistent, and it matters because the abstract and conclusion advertise 'more than 1000%' and '1600%' as the headline effect. Please correct the statement or the values and ensure the TwR definition (ΔG/G) is applied consistently in Figs. 1(d), 2(b), and 3(a).","section":"Results, Cu/TTLG/Cu and TwR definition"},{"comment":"The calculations use unrelaxed structures: the in-plane lattice constant and interlayer distances are taken from ref. [6], with no relaxation described for the twisted interfaces. For twist angles with significant moiré registry variation, interlayer distances and local hybridization can change, and the vertical conductance is typically very sensitive to such changes. At minimum, please test the sensitivity of the reported conductance and TwR to a ±1-2% variation of the interlayer distance at one representative angle.","section":"Atomic structure and relaxation"}],"minor_comments":[{"comment":"Typos and grammatical errors: 'herterostructure' (Intro), 'starttofeel' and 'anitparallel' (main text), 'difference ,called' (Abstract), 'Fm/graphite/FM' (Intro), 'magento-resistance' (Fig. 3 paragraph), and 'TTGL spacer' (magnetic case) should be 'TTLG'. A careful proofread is needed.","section":"Throughout"},{"comment":"The angle label is inconsistent: the main text quotes TwR for 'RTA of 21.787 and 28.7' while all other instances use 27.8°; also Fig. 1(d) caption writes 'θ = 21°' and 'θ = 27°' where the text gives 21.787° and 27.8°.","section":"Results, Cu/TTLG/Cu paragraph"},{"comment":"The displayed definition of TwR uses G_α(θ) and G_α′(θ′), but the subsequent definitions of MR(θ) mix magnetization configuration and angle arguments; please make the notation for spin channels and angle arguments uniform across the equations, the figures, and the text.","section":"TwR definition and Figs. 2-3"},{"comment":"The caption of Fig. 1(d) states 'Quantum conductance at room temperature' but the Methods section does not specify how temperature broadening is included in the NEGF/TBTrans conductance; please clarify the temperature treatment.","section":"Methods and Fig. 1(d) caption"},{"comment":"Several references are incomplete (e.g., refs. 20, 23, 26, 31 lack journal/volume/page details); please update them to the standard citation format.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is a single-author computational Letter. The idea of a mechanical twist-controlled spin valve is attractive, but the computational evidence is currently too sparse for the headline percentages: only two commensurate angles, an SZ basis without convergence tests, and unrelaxed interfaces. If the authors can add at least one basis-set and k-grid convergence check and one or two additional twist angles (or visibly soften the claims to point estimates), the paper could become publishable. The practical feasibility discussion (e.g., manipulation speeds with AFM tips or optical tweezers) is speculative but appropriately brief."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This paper is a first-principles prediction that twisting the middle layer of a trilayer graphene spacer between metal leads gives a large, spin-dependent change in vertical conductance. The specific junction and the TwR definition are new, and the mechanism—moving the middle layer's Dirac states relative to the metal's transmission hot spots around K and M—is a natural and testable idea. That is the real contribution: it adds a mechanical degree of freedom to vertical graphene spintronics, and the effect sizes are claimed to be huge.\n\nThe paper does several things right. It uses established DFT/NEGF methods, builds on Karpan et al. without making that citation load-bearing, and honestly admits the main computational constraint: \"Due to computational limitation we can only study two relative rotation angles.\" The transmission maps in Figs. 1(e) and 2(c) give a plausible microscopic picture for the conductance changes.\n\nThe quantitative claims, however, rest on a shallow foundation. Only two commensurate angles (21.787° and 27.8°) are sampled, chosen because they have the smallest unit cells. Three points (including θ=0) do not establish a smooth RTA dependence or a maximum, and nothing in the mechanism rules out non-monotonic behavior or a peak at an unsampled angle. The abstract and conclusion generalize to \"more than 1000%\" as if the full angular sweep had been done. The single-zeta basis is minimal for graphene's π bands and tunneling decay, and no convergence test is reported. Since absolute conductances go down to 10^-5 e2/h, the TwR ratios of small numbers could be numerically fragile. There are also minor internal inconsistencies: \"almost two order of magnitude\" conflicts with TwR values of 3600% and 1000% (factors of ~37 and ~11), and the Ni/Cu TwR-AP is quoted as 1600% in the abstract but 1542% in the Fig. 3 caption. These are fixable, but they suggest the manuscript was not tightly checked.\n\nOverall, the idea is good and the paper is worth pursuing, but as written it does not establish the quantitative TwR values at the claimed level. It reads like an early computational report, not a definitive calculation. A serious referee could ask for more angles, a basis-set check, and a more measured interpretation. I would send it to peer review with the expectation of heavy revision, not desk-reject it.","headline":"A plausible twist-controlled vertical graphene junction with first-principles numbers that are under-sampled and under-converged; worth refereeing, not worth quoting yet.","tokens_in":8991,"tokens_out":1948,"would_cite":false,"duration_ms":21197,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["72.25.-b","73.40.-c","75.70.-i","85.75.-d"],"model":"deepseek-v4-flash","headline":"Twisting a graphene layer in a stack changes conductance by over a thousand percent.","keywords":["twistronics","twisted trilayer graphene","twisting resistance","spin-dependent transport","magnetoresistance","graphene spin filter","spinmechatronics","NEGF quantum transport"],"falsifier":"A third data point would settle it: compute or measure the vertical conductance for a twist angle outside the two used here, such as $30^\\circ$ or a small-angle moiré stack with lattice relaxation; if the twisting resistance does not appear at the reported scale, the proposed mechanism is incomplete.","tokens_in":1600,"feed_emoji":"🔄","tokens_out":3511,"duration_ms":111243,"temperature":0.7,"pith_summary":"This paper claims that rotating the middle layer of a three-layer graphene spacer between metal contacts acts as a mechanical valve for electric current. First-principles transport calculations find a twisting resistance above 1000% for copper contacts, and for nickel contacts the effect depends on whether the magnetizations are parallel or antiparallel. Inserting a thin copper layer at the nickel interfaces pushes the antiparallel twisting resistance to about 1600%. If correct, rotation of a single two-dimensional layer becomes a practical switch for charge and spin current.","feed_headline":"Twist one graphene layer, swing current by 1000%","feed_subtitle":"Rotating the middle layer of a trilayer graphene stack between metal leads alters conductance and magnetoresistance on a huge scale.","key_machinery":"The central object is the commensurate twisted trilayer graphene (TTLG) spacer, in which the middle layer is rotated by $21.787^\\circ$ ($\\sqrt{7}\\times\\sqrt{7}$ unit cell) or $27.8^\\circ$ ($\\sqrt{13}\\times\\sqrt{13}$ unit cell). The argument is carried by spin-resolved transmission maps across the Brillouin zone computed with non-equilibrium Green's function transport. These maps show where in reciprocal space current can flow and how rotation displaces the middle layer's Dirac states relative to the copper or nickel transmission states. This momentum-space overlap is the mechanism behind the large conductance changes.","core_discovery":"The central claim is that the relative rotation angle (RTA) of the middle graphene layer in a trilayer spacer controls vertical conductance and magnetoresistance between metallic leads. The paper quantifies this with a twisting resistance, $\\mathrm{TwR} = (G(\\theta)-G(\\theta'))/G(\\theta')$, and reports values above 1000% for Cu/TTLG/Cu, about $-40$% and up to 130% for Ni/TTLG/Ni in parallel and antiparallel configurations, and about 200% and 1600% for Ni/Cu/TTLG/Cu/Ni. The mechanism is momentum alignment: at zero twist, graphene's available states near the $K$ point do not overlap the metal transmission states, while twisting the middle layer moves populated states near the $M$ point into play and shifts the Dirac states relative to the lead states. In the nickel case the twist also increases the majority-spin conductance and reduces the minority-spin conductance.","pith_inferences":["If the momentum-overlap mechanism is generic, the effect should be tunable at other rotation angles too, and a full angle sweep including lattice relaxation might show an even sharper peak.","The same twist-valve logic could extend to other layered spacers, such as hexagonal boron nitride or transition-metal dichalcogenides.","Practical devices will need a faster way to rotate the layer than the AFM-tip method cited here, so optical or torque-based actuation is a natural next test for this valve."],"forward_implications":["Twisting the middle graphene layer by $21.787^\\circ$ raises the vertical conductance of Cu/TTLG/Cu by roughly two orders of magnitude relative to the untwisted stack.","In Ni/TTLG/Ni, the spin-filtering effect is angle-dependent: minority-spin conductance falls while majority-spin conductance rises as the twist angle increases.","The magnetoresistance of the Ni junction drops from about 1000% at zero twist to about 163% at $27.8^\\circ$.","A three-monolayer copper insertion at each Ni/graphene interface changes the parallel twisting resistance from negative to about 200% and raises the antiparallel value to roughly 1600%.","Mechanical rotation becomes a control parameter for both charge and spin transport in vertical graphene junctions, alongside magnetic and electric control."],"supporting_citations":[{"why":"Establishes the spin-filtering behavior of graphene on nickel and supplies the interlayer geometry and lattice constants used in the junctions.","marker":"[6]"},{"why":"Defines commensurate twisted graphene structures from which the two studied rotation angles come.","marker":"[33]"},{"why":"Provides the classification of commensurate and non-commensurate twisted layers, justifying the choice of the $\\sqrt{7}\\times\\sqrt{7}$ and $\\sqrt{13}\\times\\sqrt{13}$ cells.","marker":"[34]"},{"why":"Supplies the first-principles electronic-structure method used for the self-consistent ground state.","marker":"[35]"},{"why":"Provides the non-equilibrium Green's function transport formalism used to compute the spin-dependent conductance.","marker":"[38]"},{"why":"Demonstrates experimental in-situ control of the twist angle, the control knob the proposed valve relies on.","marker":"[28]"}],"fun_headline_variants":["Twist graphene layer, swing current by 1000%","Graphene twist valve: 1600% magnetoresistance swing","Spinmechatronic valve: rotate graphene to set current","One twist of graphene layer controls spin and current"],"cache_read_input_tokens":11136,"weakest_assumption_plain":"The load-bearing premise is that the two commensurate twist angles studied, $21.787^\\circ$ and $27.8^\\circ$, represent the full twist-angle behavior; the paper itself says computational limits allowed only those two smallest unit cells.","fun_headline_variants_meta":{"raw":{"variants":["Twist graphene layer, swing current by 1000%","Graphene twist valve: 1600% magnetoresistance swing","Spinmechatronic valve: rotate graphene to set current","One twist of graphene layer controls spin and current"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000386,"raw_usage":{"total_tokens":2042,"prompt_tokens":951,"completion_tokens":1091,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":567,"completion_tokens_details":{"reasoning_tokens":1022}},"tokens_in":567,"tokens_out":1091,"duration_ms":12034,"temperature":1.0,"reasoning_tokens":1022,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T13:52:19.146610+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A third data point would settle it: compute or measure the vertical conductance for a twist angle outside the two used here, such as $30^\\circ$ or a small-angle moiré stack with lattice relaxation; if the twisting resistance does not appear at the reported scale, the proposed mechanism is incomplete.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes the spin-filtering behavior of graphene on nickel and supplies the interlayer geometry and lattice constants used in the junctions."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Defines commensurate twisted graphene structures from which the two studied rotation angles come."},{"cited_title":"Suárez Morell, M","cited_arxiv_id":null,"evidence_quote":"Provides the classification of commensurate and non-commensurate twisted layers, justifying the choice of the $\\sqrt{7}\\times\\sqrt{7}$ and $\\sqrt{13}\\times\\sqrt{13}$ cells."},{"cited_title":"Artacho, D","cited_arxiv_id":null,"evidence_quote":"Supplies the first-principles electronic-structure method used for the self-consistent ground state."},{"cited_title":"Brandbyge, J.-L","cited_arxiv_id":null,"evidence_quote":"Provides the non-equilibrium Green's function transport formalism used to compute the spin-dependent conductance."}],"review_version":1}