{"id":"e53beb0a-d62e-49f9-b535-86b0f6707aff","arxiv_id":"2507.11307","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"A slow annealing ramp at high precursor coverage yields 17-atom-wide armchair graphene nanoribbons averaging about 17 nm in length, which are then transferred and integrated into field-effect transistors with graphene electrodes.","lead":"This paper reports a recipe for growing unusually long, 17-atom-wide graphene nanoribbons, about 17 nm on average, on gold, and shows they survive transfer onto device substrates and can be wired into transistor-like structures. The result matters because long, narrow-bandgap ribbons are needed for carbon-based nanoelectronics.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The device-transport claim is not established: no control devices without GNRs are shown, and the gate-independent, multi-channel signatures are also consistent with parasitic leakage or contamination paths.","rationale":"The reader's weakest-assumption analysis correctly identifies the absence of control devices and the non-diagnostic electrical signatures as the key gap in the device-transport claim. My independent reading of the manuscript confirms that the synthesis result (mean length ~17 nm from STM) and the Raman/DFT analysis are reasonably supported, so I do not object to the core materials-science finding. However, the paper's abstract and conclusion explicitly claim that electronic transport through the GNRs is confirmed; that claim is not supported by the presented evidence because no GNR-free control is shown and the measured gate-independent, multi-channel behavior is equally consistent with leakage or contamination paths. The reader's proposed conditional verdict is appropriate: the device-integration statement should be weakened or supplemented with control measurements before the full abstract claim is accepted. My verdict_should_be is UNCHANGED because this concern matches the reader's and does not alter the recommended conditional acceptance.","tokens_in":78,"tokens_out":2370,"duration_ms":40834,"concrete_test":"Fabricate a matched control set of devices using the identical graphene electrode nanogap and full transfer workflow but with no GNRs present (e.g., transfer from a bare Au(111)/mica film through the same HCl and gold-etch steps, or selectively remove GNRs in the gap of a transferred device by a gentle oxygen plasma or RIE step after initial electrical characterization). Measure I-V and dI/dV at room temperature and at 9 K under the same vacuum and bias conditions. If control devices show currents comparable to the ~nA level, resistances in the 1-10 GOhm range, or similar gate-independent behavior, then the observed transport cannot be attributed to 17-AGNRs and the device claim fails.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that 'electronic transport occurs through the GNRs' (abstract; Device integration and transport measurements, Fig. 4) rests on an uncontrolled attribution. All 13 measured devices contain transferred 17-AGNRs, and no negative control without GNRs is reported. The room-temperature I-V curves reach nanoamperes at 1 V with resistances peaking in the 1-10 GOhm range; such values could also arise from leakage through the 20 nm Al2O3 gate dielectric, residual gold/etchant contamination from the polymer-free transfer, or unintentional percolation paths in the graphene electrode patterning process. The absence of gate dependence at room temperature, while consistent with previous low-bandgap GNR reports, does not discriminate between GNR-mediated transport and a parasitic shunt. At low temperature, the authors explicitly state that no clear Coulomb diamonds with well-defined crossing points are observed and that transport likely occurs simultaneously through multiple 17-AGNRs in parallel and in series; localized electrode states or oxide charge traps are also invoked. The paper itself thus concedes that the measured current cannot be unambiguously assigned to individual GNRs. Because the device-integration claim is the paper's distinguishing advance beyond prior 17-AGNR synthesis work, this missing control is the most load-bearing weakness.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports an optimized on-surface synthesis protocol for 17-atom-wide armchair graphene nanoribbons (17-AGNRs) on Au(111), using slow temperature ramping and high precursor coverage to increase the average ribbon length to approximately 17 nm. The ribbons are characterized in UHV by STM/STS and by Raman spectroscopy, and the experimental Raman spectra are compared with DFT and REBOII simulations. The authors also transfer the ribbons to device substrates, demonstrate their stability through the transfer and chemical processing, and fabricate field-effect transistors with graphene electrodes. The central claims are that the optimized synthesis yields sufficiently long 17-AGNRs for device integration and that electrical transport measurements confirm current flow through the ribbons.","tokens_in":18675,"tokens_out":5495,"duration_ms":67441,"significance":"If fully supported, the synthesis advance would be significant: 17-AGNR is the widest experimentally realized member of the 3p+2 armchair family, and reaching lengths of about 17 nm would bridge the 15-20 nm electrode gaps needed for transistor integration. The Raman analysis is a particular strength: the low-frequency mode assignments are checked against independent REBOII and DFT calculations rather than being fitted to the measured spectra, and the comparison between simulated and measured LCM positions provides a concrete, falsifiable length diagnostic. The stability of the ribbons under transfer and chemical processing is also valuable for the device-engineering community. The principal weakness is the electrical transport claim, which currently lacks the control experiments needed to attribute the measured current to the GNRs rather than to parasitic paths.","major_comments":[{"comment":"The abstract claims that electronic transport occurs through the GNRs, but this is not established because all 13 measured devices contain transferred ribbons and no control devices without GNRs are reported. The room-temperature I-V curves with 1-10 GOhm resistances, the absence of gate dependence, and the low-temperature lack of clear Coulomb diamonds are also consistent with leakage through the 20 nm Al2O3 dielectric or with transport through contaminants from the transfer and electrode processing. The text itself states that transport likely occurs through multiple ribbons in parallel and in series and that oxide charge traps may contribute, which prevents an unambiguous assignment of the current to individual GNRs. I request either control devices without GNRs (or with intentionally blocked electrode gaps) or a revision that limits the claim to 'transport consistent with GNR-mediated current' rather than 'confirmed'.","section":"Device integration and transport measurements; Figure 4"},{"comment":"The central quantitative claim of an average length of about 17 nm, and the comparisons to 6.2 nm and 9.8 nm for the two earlier growth conditions, are reported without histograms, standard deviations, or the number of ribbons counted. Without these statistics the reader cannot judge whether the length increase is statistically significant or reproducible across samples. Please add the length distributions for all three growth conditions and report the number of measured ribbons and the statistical uncertainty on each average.","section":"Synthesis and length optimization; Figure 1"},{"comment":"The reported experimental STS bandgap of approximately 0.45 eV is inconsistent with both the DFT-PBE value of 0.14 eV and the previously published experimental value of 0.19 eV cited from ref. 26. The discrepancy is not commented on in the text. Since the low-bandgap nature of 17-AGNRs is part of the motivation for device integration, the authors should explain the possible origins of this difference, for example substrate doping, measurement position, or the finite length of the measured ribbons.","section":"Synthesis and length optimization; Figure S3"}],"minor_comments":[{"comment":"In the inset description of Figure 1b, the scale bar is given as 10 nm, but the high-resolution image of the ribbon ends should presumably have a much smaller scale bar; please check whether this should be 1 nm.","section":"Figure 1 caption"},{"comment":"The sentence 'the width of the G mode centered at 1593 cm-1 decreases by 4 cm-1, and those of the CH/D modes at 1305, 1333, and 1375 cm-1 increase by 3 cm-1 for the G modes' is confusing; please clarify which widths change and for which excitation wavelength.","section":"Raman characterization"},{"comment":"The abstract refers to a 'gradual temperature ramping during an extended annealing period', but the Methods section only specifies a constant heating ramp of roughly 2 °C/min until 400 °C; the exact temperature profile and any dwell time should be stated.","section":"Abstract and Methods"},{"comment":"The Methods state that all Raman spectra were collected at pressures of 1-30 x 10^-6 mbar, while the text claims stability under ambient conditions; please clarify whether the samples were exposed to ambient conditions before measurement and how the vacuum measurement supports the ambient-stability claim.","section":"Methods, Raman spectroscopy"}],"recommendation":"major_revision","confidential_remarks":"The stress-test concern about the transport claim is valid and lands on a load-bearing point. The synthesis and Raman sections are credible and should not be unduly delayed, but the device conclusion needs either additional control experiments or a carefully weakened formulation. I do not see grounds for rejection; the weakness is fixable within a revision."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"I read the paper as a solid synthesis and Raman study wrapped in an overreaching transport section. The growth result is the real meat: STM shows 17-AGNRs with an average length around 17 nm, up from 6 nm under the initial conditions, and the images support the claim. The Raman low-frequency analysis, using REBOII to assign the sub-150 cm-1 peaks to LCMs of 5- to 12-mer ribbons, is a genuinely useful length probe for this width, and the transfer-stability data are practically valuable.\n\nThe transport claim is the soft spot. The abstract says electronic transport through the GNRs is confirmed, but the electrical data do not establish that. There are no control devices without ribbons, so leakage through the Al2O3 gate dielectric or through residue from the transfer/etch process is not ruled out. The room-temperature gate independence could mean either GNR transport or a parasitic shunt. At low temperature, the paper itself admits there are no clear Coulomb diamonds and attributes the features to multiple ribbons plus localized states and oxide traps. That is honest, but it means the measured current cannot be unambiguously assigned to 17-AGNRs. This needs either a negative control experiment or a reduced claim, along the lines of 'consistent with GNR-mediated transport.' Given the difficulty of clean controls, the second option is realistic.\n\nSome smaller issues: length averages are quoted without standard deviations or ribbon counts; the improvement from slow ramp and high coverage is inferred from three samples, not isolated; and the STS gap of 0.45 eV is not reconciled with the DFT gap of 0.14 eV or the 0.19 eV reported earlier for the same ribbons. None of these are fatal, but they should be addressed.\n\nThis paper deserves a serious referee. The synthesis and Raman portion is a solid step for the GNR community, and the transport section has a clear, fixable weakness. I would cite the growth and Raman results, but I would not cite the transport claim as established. A revision that adds a control or softens the language would make this a much stronger paper.","headline":"Solid 17-AGNR synthesis and Raman work with an overreaching transport claim that needs controls or a softer conclusion.","tokens_in":19340,"tokens_out":3174,"would_cite":true,"duration_ms":37285,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["73.63.-b","73.23.-b","78.30.-j","81.07.-b"],"model":"deepseek-v4-flash","headline":"The paper reports that optimized on-surface synthesis yields 17-atom-wide armchair graphene nanoribbons with an average length of about 17 nm, and that these ribbons can be transferred and integrated into graphene-electrode field-effect…","keywords":["graphene nanoribbon","17-AGNR","armchair edge","on-surface synthesis","narrow bandgap","field-effect transistor","Raman spectroscopy","substrate transfer"],"falsifier":"Fabricate identical graphene-electrode devices without any GNR transfer and measure the same current-voltage and differential-conductance maps; if comparable nanoampere currents or similar low-temperature resonances appear, the claim that transport occurs through the ribbons fails. A complementary check would be to correlate each device's resistance with ribbon coverage or length measured by Raman or atomic force microscopy on the same gap.","tokens_in":18229,"feed_emoji":"⚡","tokens_out":6590,"duration_ms":78782,"temperature":0.7,"pith_summary":"The paper seeks to show that 17-atom-wide armchair graphene nanoribbons, strips of graphene only 17 carbon atoms across, can be grown long enough to bridge the gaps between electrodes, survive transfer to device substrates, and carry current in a transistor. The authors report that a slow annealing ramp combined with near-monolayer precursor coverage raises the average ribbon length from about 6 nm to about 17 nm, with some ribbons exceeding 50 nm. This length gain makes the 17-AGNR the widest member of the narrow-bandgap 3p+2 ribbon family to be electrically measured in a field-effect transistor geometry. A sympathetic reader would care because the 17-AGNR has a low predicted bandgap that could make it a practical transistor channel, but only if the ribbons are long, stable, and integrable.","feed_headline":"17-atom-wide graphene ribbons grow to 17 nm, bridge transistor gaps","feed_subtitle":"Slow heating and near-monolayer coverage make 17-AGNRs long enough to carry current across graphene-electrode gaps.","key_machinery":"The load-bearing mechanism is the thermally sequenced on-surface reaction of the BADBB monomer (1,2-bis-(anthracenyl)-3,6-dibromobenzene): dehalogenation starting around 150 °C couples monomers into polymers, and cyclodehydrogenation above 350 °C planarizes them into the 17-carbon-wide ribbon. The optimization keeps these steps separated with a 2 °C/min ramp and uses nearly complete precursor monolayers so that the monomers assemble into ordered islands that template elongation; the remaining failure mode is a flipped anthracene unit at the ribbon end that forms pentagons and stops growth. The second supporting mechanism is the Raman longitudinal compressive mode (LCM), whose frequency decreases with ribbon length and lets the authors read the length distribution after transfer and on device substrates.","core_discovery":"On its own terms, the paper's central discovery is that the length bottleneck for 17-AGNRs can be broken by process control rather than by new chemistry. Subliming the BADBB precursor to near-monolayer coverage and annealing at a slow 2 °C/min ramp from 150 to 400 °C separates the dehalogenative polymerization from the cyclodehydrogenation step, so the ribbons grow to an average of 17.3 nm, compared with 6.2 nm for the initial protocol. STM and STS confirm the expected 2.45 nm width and a bandgap of about 0.45 eV on Au(111). Raman spectroscopy, including length-assigned longitudinal compressive modes, shows that the ribbons survive polymer-free transfer and harsh chemical treatments. In graphene-electrode FETs with 15–20 nm gaps, the devices pass nanoampere currents at 1 V, show Schottky-like current-voltage curves and room-temperature gate independence, and display low-temperature resonances; the authors conclude that electronic transport occurs through multiple 17-AGNRs bridging the electrodes.","pith_inferences":["If the flipped-anthracene termination is the limiting defect, redesigning the precursor to suppress that flip could push average lengths well past 17 nm; the paper's own images imply this mechanism is what stops growth.","The room-temperature gate independence could mean the measured current is dominated by injection barriers or by parallel ribbons rather than by the intrinsic 0.45 eV gap, so single-ribbon or aligned-array devices would be the next test of whether that gap controls switching.","The Raman LCM calibration against ribbon length could be transferred to other AGNR widths as a fast ex-situ length metrology, provided the mode-frequency versus length relation is computed for each width.","Because the transport data suggest multiple ribbons in parallel and in series, a direct correlation of device resistance with ribbon coverage and orientation on the same gap would clarify whether the nanoampere currents come from few ribbons or many."],"forward_implications":["Longer 17-AGNRs remove the main length bottleneck for device integration, making the 17-AGNR the widest 3p+2 AGNR to be electrically measured in a FET geometry.","Raman spectroscopy can serve as a post-transfer and post-processing length and quality check, since LCM frequencies map to ribbon lengths in the short-ribbon range and the spectra remain stable under ambient and harsh chemical conditions.","The polymer-free transfer process preserves the narrow-bandgap ribbons well enough for device fabrication, indicating that low-bandgap 3p+2 ribbons are not too fragile for practical integration.","At low temperature, the devices show gate-tunable resonant transport features consistent with single-electron tunneling through multiple ribbons in parallel and in series, even though clear Coulomb diamonds are not observed.","The absence of a beneficial thermal annealing step for 17-AGNRs, unlike for 9-AGNRs, means the device optimization path must avoid high-temperature treatments that degrade the ribbons."],"supporting_citations":[{"why":"Provides the BADBB precursor design and the first synthesis and bandgap characterization of 17-AGNRs that this work extends.","marker":"[26]"},{"why":"Establishes the 15–20 nm graphene-electrode nanogap platform and the annealing protocol baseline for contacting graphene nanoribbons.","marker":"[29]"},{"why":"Supplies the universal length-dependent vibrational mode concept, the LCM, that the paper uses to fingerprint ribbon length by Raman spectroscopy.","marker":"[13]"},{"why":"Provides the polymer-free transfer and ex-situ characterization methodology that the paper applies to verify ribbon stability after transfer.","marker":"[27]"},{"why":"Reports growth optimization and device integration of narrow-bandgap 5-AGNRs, the prior 3p+2-family benchmark this work surpasses in width.","marker":"[22]"},{"why":"Supplies the GNR field-effect transistor transport signatures, including gate-dependent quantum dot features, against which the 17-AGNR transport is compared.","marker":"[23]"},{"why":"Supports the template-like effect of ordered precursor assemblies that the paper invokes to explain enhanced ribbon elongation at high coverage.","marker":"[34]"}],"fun_headline_variants":["Slow heating yields 17-nm graphene ribbons for working transistors","Tripled graphene ribbon length via optimized synthesis","Process control enables 17-atom-wide ribbons to bridge FET gaps","Longer 17-AGNRs: synthesis fix moves ribbons into devices"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The electrical measurements assume that the measured source-drain current flows through 17-AGNR chains bridging the graphene electrodes, rather than through leakage paths, solvent residues, or metallic contamination from transfer, yet no control devices without ribbons are reported.","fun_headline_variants_meta":{"raw":{"variants":["Slow heating yields 17-nm graphene ribbons for working transistors","Tripled graphene ribbon length via optimized synthesis","Process control enables 17-atom-wide ribbons to bridge FET gaps","Longer 17-AGNRs: synthesis fix moves ribbons into devices"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000496,"raw_usage":{"total_tokens":2485,"prompt_tokens":1053,"completion_tokens":1432,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":669,"completion_tokens_details":{"reasoning_tokens":1361}},"tokens_in":669,"tokens_out":1432,"duration_ms":12985,"temperature":1.0,"reasoning_tokens":1361,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T17:11:15.260995+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Fabricate identical graphene-electrode devices without any GNR transfer and measure the same current-voltage and differential-conductance maps; if comparable nanoampere currents or similar low-temperature resonances appear, the claim that transport occurs through the ribbons fails. A complementary check would be to correlate each device's resistance with ribbon coverage or length measured by Raman or atomic force microscopy on the same gap.","supporting_citations":[],"review_version":1}