{"id":"1fcc9570-6784-4581-abc1-e14456e1c1b5","arxiv_id":"2507.10887","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Bilayer graphene nanoribbons are predicted to host long-lived, optically active interlayer excitons whose energies and radiative lifetimes can be tuned by stacking order.","lead":"Using quantum chemistry simulations, this paper predicts that stacking two narrow graphene ribbons creates long-lived interlayer excitons, with radiative lifetimes from about one nanosecond to nearly ten microseconds. Because the lifetimes depend on how the ribbons are stacked, the findings suggest a route to tuning the optical response of carbon-based ribbon devices.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Reported 0.25% oscillator strength for the 6-10AGNR interlayer exciton and the reported 9.4 µs lifetime are mutually inconsistent under Eq. (4); one of the headline numbers must be wrong.","rationale":"The reader's CONDITIONAL verdict is appropriate, and I agree that the dipole moments and radiative-lifetime model are the crux. However, the most load-bearing issue is more specific than 'the model is unbenchmarked for interlayer excitons': the manuscript's own reported inputs and outputs for the 6-10AGNR system are numerically inconsistent under the model it uses. This does not necessarily invalidate the broader physics—the GW-BSE workflow is standard, the intralayer lifetime benchmark to experiment is a point in favor, and the inconsistency may be a typo—but it means the headline '9.4 µs' cannot currently be taken at face value. The proposed check settles which side of the inconsistency is wrong, so I keep the reader's CONDITIONAL verdict rather than escalating to REJECT without confirmation.","tokens_in":13399,"tokens_out":14055,"duration_ms":158527,"concrete_test":"Recompute μ_IX and f_IX/f_X1 from the stored BSE exciton amplitudes A_{vck} and single-particle dipoles d_{vck} for 6-10AGNR, then re-evaluate Eq. (4) with Ω_IX=1.18 eV and M_S=0.377 m0. If the recomputed τ0 is ≈1.9 ns, the '0.25%' statement is consistent but Table I's 83.7 ns is wrong; if it is ≈83.7 ns, the oscillator-strength statement is wrong by about a factor of 40; if it is ≈830 ns, the lifetime column or the 9.4 µs claim is wrong. Repeat the same check for 6β-stacking with the quoted '13%' strength to see whether the control case reproduces Table I's interlayer τ0 of 32 ps.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's headline lifetime range is anchored by the 6-10AGNR interlayer exciton (IX): Sec. III.B says its oscillator strength is '~0.25% of exciton X1', while Table I lists τ0(IX)=83.7 ns and τ_RT=9.4 µs with M_S=0.377 and Ω_IX=1.18 eV. Under Eq. (4), γ(0) is proportional to Ω_S^2 μ_S^2, so τ0(IX)/τ0(X1) should equal (Ω_X1/Ω_IX)^2 / (μ_IX^2/μ_X1^2). If '0.25%' is an oscillator-strength ratio (f ∝ Ω μ^2), this predicts τ0(IX) ≈ 4.34 ps × (1.29/1.18)^2 / 0.0025 ≈ 1.9 ns and τ_RT ≈ 213 ns, not 83.7 ns and 9.4 µs—a factor of roughly 40 discrepancy. If '0.25%' were instead a linear dipole-moment ratio, the predicted τ0 would be ≈830 ns, an order of magnitude above the table value. A similar but smaller tension exists for 6β-stacking, where the quoted '13%' also does not quantitatively match the tabulated τ0 ratio. Because the abstract's upper lifetime bound is precisely the 9.4 µs value, either the oscillator-strength statement, the Table I entries, or the extraction of the exciton transition dipole is erroneous; the manuscript as written does not allow the reader to tell which.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports G0W0+BSE calculations for bilayer armchair graphene nanoribbons in α- and β-stackings and for 3-6 and 6-10 heterobilayers. It claims that interlayer excitons appear with type-I or type-II band alignments, oscillator strengths up to 13% of the brightest intralayer exciton, and room-temperature radiative lifetimes from 1 ns to 9.4 µs, tunable by stacking order. The authors use the Spataru model for radiative lifetimes, project exciton wavefunctions to classify intralayer/interlayer character, and explicitly state limitations such as defect-free and freestanding assumptions.","tokens_in":13726,"tokens_out":19213,"duration_ms":196472,"significance":"If the results hold, this is a useful prediction of tunable interlayer excitons in one-dimensional van der Waals systems and would extend the interlayer-exciton paradigm to graphene nanoribbons, with stacking order as a control parameter. Strengths include the state-of-the-art G0W0+BSE pipeline with no fitted parameters, wavefunction-based exciton classification, and room-temperature predictions that are experimentally testable. The credibility of the central lifetime claim, however, is compromised by numerical inconsistencies in the reported oscillator strengths and lifetimes.","major_comments":[{"comment":"The quoted oscillator strengths and the tabulated lifetimes are mutually inconsistent. For 6-10AGNR the text states that IX has ~0.25% of the oscillator strength of X1, with energies Ω_IX=1.18 eV and Ω_X1=1.29 eV. Since f ∝ Ω|μ|^2, Eq. (4) gives τ0(IX)/τ0(X1) = (Ω_X1/Ω_IX)/(f_IX/f_X1) ≈ 437. The table gives τ0(X1)=4.34 ps and τ0(IX)=83.7 ns, i.e., a ratio of ~1.9×10^4, a factor ~44 discrepancy that directly affects the headline 9.4 µs lifetime (if the tabulated intralayer τ0 is retained, the predicted τ_RT for IX is ~0.2 µs; if the tabulated IX τ0 is correct, the oscillator-strength ratio should be ~0.006%, not 0.25%). For 6β-stacking the same check fails: with f_IX/f_X1=0.13, Ω_X1=1.30 eV and Ω_IX=1.62 eV, Eq. (4) predicts τ0(IX)/τ0(X1) ≈6.2, whereas Table I implies 0.032 ns / 0.79 ps ≈40. The 6-10 intralayer row is also internally inconsistent: τ_RT/τ0 = 180 ps / 4.34 ps ≈41, while Eq. (4) with M_S=0.398 and Ω=1.29 eV gives ≈106. The authors need to provide the raw transition dipoles and reconcile these numbers; as written, the central lifetime range is not self-consistent.","section":"III.B, Table I, Eq. (4)"},{"comment":"The radiative model of Spataru et al. is applied to interlayer excitons without validation. Equation (4) assumes a parabolic exciton dispersion and a single effective mass M_S, and for the interlayer states reported here (charge-transfer-like, 80–92% electron–hole separation, small dipoles) this assumption is not established. The 9.4 µs lifetime scales as sqrt(M_S)/μ_S^2, so uncertainties in the dipole or mass—which are computed from a BSE truncated to six valence and six conduction bands—propagate directly to the central claim. Please provide convergence tests of μ_S and M_S with respect to BSE band number and k-grid, and, if possible, compute the exciton dispersion Ω_S(Q) to justify the parabolic approximation.","section":"II, Eq. (4); III.B"}],"minor_comments":[{"comment":"The column headers are inconsistent: intralayer τ_RT entries are in ps (e.g., 6β-stacking is 140 ps in the text), while interlayer τ_RT entries are in ns; the τ0 columns similarly mix ps and ns. Please use uniform units or separate columns.","section":"Table I"},{"comment":"The Introduction states that heterobilayer lifetimes range from 258 ns to 9.4 µs, while the Abstract says the range is from 1 ns to 9.4 µs; please clarify which systems define the stated range.","section":"I and Abstract"},{"comment":"The phrase 'oscillator strength that is ~0.25% of exciton X1 as depicted in the inset of Fig. 3e' is ambiguous because the inset caption says it shows normalized transition dipoles, not oscillator strengths; please define the quantity actually plotted.","section":"III.B"},{"comment":"There are several typos: 'diserable' in Sec. III.C, 'zoon-in' in the Fig. 2 caption, 'resamble' in the Fig. 1 caption, and 'the, suggesting' in Sec. III.A.","section":"General"}],"recommendation":"major_revision","confidential_remarks":"To the editor: the manuscript is within scope and the methodology is appropriate. The main problem is internal: the reported oscillator strengths, Table I, and the abstract lifetime range cannot all be correct, and one of the two main quantitative claims (the microsecond lifetime) may be off by an order of magnitude. This is fixable in revision if the authors recompute and report the raw dipoles and masses; if the 9.4 µs value is retained, the oscillator-strength statements must be revised accordingly. I would not recommend rejection, but the authors must address the inconsistency before publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear [Colleague],\n\nHere's my read of arXiv:2507.10887, the GW-BSE study of interlayer excitons in bilayer armchair graphene nanoribbons (AGNRs). The genuinely new result is the prediction of bright interlayer excitons in heterobilayer AGNRs—type-II aligned 6-10AGNR in particular—with room-temperature radiative lifetimes from ~1 ns up to 9.4 µs depending on stacking. If correct, that would make bilayer AGNRs a tunable 1D platform for long-lived interlayer excitons.\n\nThe paper does several things well. The methodology is state of the art: G0W0 + BSE with a truncated Coulomb interaction, and the single-layer lifetimes (~140–180 ps) are consistent with the experimental lower bound of >100 ps for sub-2 nm ribbons. The layer-projected band structures and real-space exciton wavefunctions give a clear picture of intralayer vs interlayer character, and the type-I/type-II alignment discussion is plausible.\n\nThe problem is a serious internal inconsistency that the stress-test note surfaces. In Section III.B, the interlayer exciton in 6-10AGNR is said to have an oscillator strength of about 0.25% of exciton X1. Table I lists τ0(IX) = 83.7 ns and τ_RT = 9.4 µs. Using their own Eq. (4) and the usual relation f ∝ Ω μ^2, that oscillator strength ratio gives τ0(IX) ≈ 1.9 ns—a factor of ~40 shorter. If \"0.25%\" is instead a linear dipole-moment ratio, the predicted τ0 is ~830 ns, still an order of magnitude off. For 6β stacking, the quoted 13% of X1 similarly gives τ0 ~5 ps if taken as an oscillator strength, versus the tabulated 32 ps (it only works if 13% is a dipole ratio). The manuscript as written is therefore not self-consistent: either the oscillator-strength statements, the lifetimes, or the dipole extraction is wrong.\n\nThis is not a minor detail—the abstract's headline upper bound of 9.4 µs comes directly from that number. The paper needs a careful re-examination of the lifetime calculation and the oscillator-strength statements before the central claim can be trusted. The reader's \"conditional\" verdict is right, and the stress-test concern lands.\n\nOverall, this is a serious computational study with a plausible physical narrative, but the internal inconsistency is load-bearing. I'd send it to a thoughtful referee, with an explicit request to verify the lifetime numbers and oscillator strengths independently. Researchers working on 1D excitons in van der Waals heterostructures will find it useful once the numbers are straightened out. I wouldn't cite it in its current form.\n\nRecommendation: engage with it enough to send to peer review, but flag the inconsistency as a mandatory fix.\n\nBest.","headline":"Serious GW-BSE study of bilayer GNR interlayer excitons, but the reported oscillator strengths and the tabulated lifetimes are internally inconsistent by up to a factor of 40; the 9.4 µs headline needs verification.","tokens_in":14265,"tokens_out":6030,"would_cite":false,"duration_ms":60267,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Stacking two graphene nanoribbons can create long-lived interlayer excitons whose room-temperature radiative lifetime is tunable from about one nanosecond to nearly ten microseconds by changing the stacking order and ribbon widths.","keywords":["interlayer excitons","graphene nanoribbons","van der Waals heterostructures","GW-BSE calculations","radiative lifetimes","type-II band alignment","stacking order","Bethe-Salpeter equation"],"falsifier":"A time-resolved photoluminescence experiment on synthesized bilayer AGNRs with beta-like stacking should show radiative components near the predicted near-infrared energies with room-temperature lifetimes of roughly 1.66 ns for 6-beta, 258 ns for 3-6AGNR, and 9.4 microseconds for 6-10AGNR; absence of the microsecond component or of the predicted interlayer absorption peaks would falsify the central claim. A temperature sweep should also follow the model's square-root scaling if the parabolic-dispersion assumption is right.","tokens_in":13225,"feed_emoji":"🔬","tokens_out":9027,"duration_ms":101235,"temperature":0.7,"pith_summary":"Vertically stacking two armchair graphene nanoribbons is predicted to create interlayer excitons—bound electron-hole pairs whose electron and hole sit in different layers—whose brightness and radiative lifetime are controlled by the stacking order. The paper's GW-BSE calculations find room-temperature radiative lifetimes ranging from about 1 ns for a moderately bright interlayer state in a beta-stacked homobilayer to 9.4 microseconds for a very dark interlayer state in a 6-10 heterobilayer, with the interlayer absorption reaching up to 13% of the main intralayer peak. Depending on which ribbon widths are paired, the bilayer forms either a type-I or type-II band alignment, which decides whether pure interlayer excitons are favored. If these predictions hold, bilayer graphene nanoribbons would be a one-dimensional, synthetically accessible platform for tunable exciton physics in the near-infrared.","feed_headline":"Stacked graphene ribbons tune exciton lifetimes from 1 ns to 9 µs","feed_subtitle":"Stacking order controls how bright and how long-lived these electron-hole pairs are, from 1 ns to 9.4 µs.","key_machinery":"The load-bearing machinery is the GW-BSE many-body perturbation chain: G0W0 quasiparticle corrections to the DFT eigenvalues set the single-particle gaps, and the Bethe-Salpeter equation, truncated to six valence and six conduction bands, produces the exciton energies and envelope wavefunctions. Interlayer character is diagnosed by projecting the exciton's real-space electron density onto the layer opposite the hole, giving percentages such as 78%, 80%, and 92% for the interlayer states. Radiative lifetimes are computed with the temperature-dependent model of Eq. (4), which converts the exciton energy, effective mass, and transition dipole matrix element into a radiative rate; the same model has been benchmarked for semiconducting carbon nanotubes.","core_discovery":"On the paper's own terms, bilayer AGNRs are a working one-dimensional van der Waals platform for interlayer excitons. In 6-AGNR homobilayers, beta-stacking preserves the monolayer optical spectrum and adds an interlayer exciton at about 1.62 eV with 13% of the intralayer oscillator strength and a 1.66 ns room-temperature radiative lifetime, while alpha-stacking produces hybridized excitons spread over both layers. In heterobilayers, fully relaxed structures adopt beta-like stacking: 3-6AGNR gives a type-I alignment with an interlayer exciton at 1.57 eV living 258 ns, and 6-10AGNR gives a type-II alignment whose lowest-energy interlayer exciton sits at 1.18 eV, carries only 0.25% of the intralayer oscillator strength, and radiates in 9.39 microseconds. The paper interprets these as long-lived radiative upper bounds, noting that defects and substrates would introduce nonradiative channels and shorten measured lifetimes.","pith_inferences":["A natural next knob the paper does not explore is an out-of-plane electric field or twist angle, which in 2D homobilayers continuously shifts interlayer exciton energy and lifetime; the same tunability could plausibly be engineered in these 1D bilayers.","The 0.25% oscillator strength of the 9.4 microsecond state suggests a brightness-lifetime trade-off: pairing ribbons with intermediate quasiparticle gap differences might yield interlayer excitons with lifetimes in the tens to hundreds of nanoseconds while keeping them bright enough for detection.","The radiative model assumes a parabolic exciton dispersion; measuring the temperature dependence of the lifetime would test that assumption, since non-parabolic or multiple-mass effects would break the predicted square-root temperature scaling.","Because the predicted lifetimes are radiative upper bounds, a direct comparison with experiment requires time-resolved photoluminescence on synthesized quasi-free-standing bilayer ribbons, where substrate and defect effects should shorten but not eliminate the microsecond component."],"forward_implications":["Stacking order is a functional knob: beta-stacked homobilayers host a moderately bright interlayer exciton, whereas alpha-stacking hybridizes the states, and heterobilayer width combinations select type-I versus type-II alignment.","Interlayer radiative lifetimes at room temperature are predicted to span 1.66 ns (6-beta), 258 ns (3-6AGNR), and 9.39 microseconds (6-10AGNR), orders of magnitude longer than the 140-180 ps intralayer excitons in the same systems.","Interlayer excitons remain visible in absorption, with peak strengths up to 13% of the maximum absorption in the homobilayer, making them usable for near-infrared detectors or emitters.","Exciton binding energies above 1 eV and singlet-triplet splittings of 0.14-0.26 eV make the lowest excitons stable against thermal fluctuations and suppress intersystem crossing.","Because the predicted lifetimes are radiative upper bounds, clean and defect-free samples are the target for experiments, since even modest defect concentrations directly shorten measured lifetimes."],"supporting_citations":[{"why":"Supplies the converged GW-BSE parameter set and the monolayer quasiparticle gaps used to choose ribbon pairs with type-I or type-II alignment.","marker":"[7]"},{"why":"Defines the Bethe-Salpeter Hamiltonian and its diagonalization that yields the exciton energies and envelope functions.","marker":"[46]"},{"why":"Provides the radiative lifetime formula in Eq. (4), benchmarked on carbon nanotubes, from which all reported lifetimes follow.","marker":"[50]"},{"why":"Defines the alpha- and beta-stacking arrangements whose difference drives the tunability.","marker":"[51]"},{"why":"Documents the hybridization mechanism between intralayer and dark charge-transfer excitons in non-centrosymmetric bilayer MoS2, the analogue invoked for homobilayer AGNRs.","marker":"[17]"},{"why":"Provides the comparison class for interlayer exciton oscillator strengths and lifetimes in TMDC homobilayers.","marker":"[16]"},{"why":"Reports experimental exciton lifetimes exceeding 100 ps in sub-2-nm GNRs, used to validate the intralayer lifetime calculations.","marker":"[35]"},{"why":"Demonstrates synthesis of quasi-free-standing bilayer graphene nanoribbons with AB stacking, the experimental route the paper points to.","marker":"[40]"}],"fun_headline_variants":["Stacking order tunes interlayer exciton lifetimes in bilayer ribbons","Bilayer graphene ribbons: exciton lifetimes span 1 ns to 9.4 µs","Stacking controls exciton lifetime in bilayer graphene ribbons","Interlayer excitons in graphene nanoribbons live up to 9.4 µs","Tunable interlayer excitons in stacked graphene ribbons"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The predicted lifetimes rest on the radiative model of Eq. (4), which assumes a parabolic exciton dispersion with a single effective mass and uses dipole matrix elements from a Bethe-Salpeter calculation truncated to six valence and six conduction bands; if those dipole strengths or masses are inaccurate, the whole 1 ns to 9.4 microsecond range shifts.","fun_headline_variants_meta":{"raw":{"variants":["Stacking order tunes interlayer exciton lifetimes in bilayer ribbons","Bilayer graphene ribbons: exciton lifetimes span 1 ns to 9.4 µs","Stacking controls exciton lifetime in bilayer graphene ribbons","Interlayer excitons in graphene nanoribbons live up to 9.4 µs","Tunable interlayer excitons in stacked graphene ribbons"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000729,"raw_usage":{"total_tokens":3262,"prompt_tokens":943,"completion_tokens":2319,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":559,"completion_tokens_details":{"reasoning_tokens":2221}},"tokens_in":559,"tokens_out":2319,"duration_ms":18797,"temperature":1.0,"reasoning_tokens":2221,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T17:22:58.633337+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A time-resolved photoluminescence experiment on synthesized bilayer AGNRs with beta-like stacking should show radiative components near the predicted near-infrared energies with room-temperature lifetimes of roughly 1.66 ns for 6-beta, 258 ns for 3-6AGNR, and 9.4 microseconds for 6-10AGNR; absence of the microsecond component or of the predicted interlayer absorption peaks would falsify the central claim. A temperature sweep should also follow the model's square-root scaling if the parabolic-dispersion assumption is right.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the converged GW-BSE parameter set and the monolayer quasiparticle gaps used to choose ribbon pairs with type-I or type-II alignment."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the radiative lifetime formula in Eq. (4), benchmarked on carbon nanotubes, from which all reported lifetimes follow."},{"cited_title":"Sahu , author H","cited_arxiv_id":null,"evidence_quote":"Defines the alpha- and beta-stacking arrangements whose difference drives the tunability."},{"cited_title":"Deilmann and author K","cited_arxiv_id":null,"evidence_quote":"Documents the hybridization mechanism between intralayer and dark charge-transfer excitons in non-centrosymmetric bilayer MoS2, the analogue invoked for homobilayer AGNRs."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the comparison class for interlayer exciton oscillator strengths and lifetimes in TMDC homobilayers."},{"cited_title":"Tries , author S","cited_arxiv_id":null,"evidence_quote":"Reports experimental exciton lifetimes exceeding 100 ps in sub-2-nm GNRs, used to validate the intralayer lifetime calculations."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Demonstrates synthesis of quasi-free-standing bilayer graphene nanoribbons with AB stacking, the experimental route the paper points to."}],"review_version":1}