{"id":"d8c2cd3f-c097-4af3-83d3-64b23779eb26","arxiv_id":"2607.25128","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"Long-range photocarrier transport in bulk-insulating Sb-doped Bi2Se3 nanoribbons is carried by charge-neutral excitons, not free carriers, as shown by a ~50× elevated transverse dissociation field and Einstein-relation violation.","lead":"A transverse electric field must be ~50× stronger than free-carrier theory predicts to kill long-range photocurrent in insulating Bi2Se3 nanoribbons, and diffusivity far exceeds the Einstein value from drift mobility. That points to charge-neutral excitons, not free charges, carrying the mm-scale photoresponse.","discovery_kind":"new_method","skeptic_critique":{"model":"moonshotai/kimi-k3","headline":"The 47× threshold anomaly is parameter-free only if the transverse field inside the channel equals Vt/d; the paper's own model invokes strong band bending at the side contacts, which is precisely where an unmeasured fraction of Vt could drop.","rationale":"The paper is, on the whole, careful: the free-carrier exclusion via Eq. (1) is genuinely parameter-free given field calibration; the Poole–Frenkel scaling (linear √Eth vs T, three decades of sT), polarity independence, and the high-fluence/high-T reversion to free-carrier behavior are real internal controls that cut against trivial artifacts; and the TPV Einstein violation is an independent second probe whose order-of-magnitude excess survives the stated fit uncertainty and the acknowledged factor-of-2 geometric systematic. I read the artifact exclusions (heating, misalignment) as plausible given the polarity symmetry. The one place where a load-bearing quantity is assumed rather than measured is the transverse field actually present in the channel — the quantity to which the entire 47× factor is anchored. The reader instead flagged the contact-as-probe model (Eq. 3, capture neglected). That assumption is indeed soft, but it primarily supports the multiple-pass global-shift narrative and the IQE reconciliation; the neutrality conclusion would stand on the threshold + TPV evidence even if the contact model were revised. Hence partial agreement: both concerns are about how device interfaces translate applied/measured quantities into inferred physics, but the field-calibration issue is the one tied directly to the strongest claim. I do not adjust the verdict: CONDITIONAL remains right, and I would add \"direct verification of the transverse potential profile (or quantitative mobility cross-check at high fluence)\" to the reader's conditions. If the potential mapping showed ≥90% of Vt across the channel, the free-carrier exclusion would be about as solid as transport evidence of this kind gets.","tokens_in":11798,"tokens_out":3908,"duration_ms":71111,"concrete_test":"Map the electrostatic potential across the T1–T2 gap under operating bias at 20 K (scanning Kelvin probe force microscopy, or a companion nanoribbon with floating potential-sensing fingers along the gap) and verify that ≥90% of Vt drops across the 5 µm channel. Internal cross-check: at high fluence, the drift mobility implied by the TPVM drift rate (~460 cm²/Vs, computed with E=Vt/d) should agree within a factor of ~2 with the mobility from the TPV slope (Eq. 5) at the same fluence; a larger discrepancy rescales the 47× anomaly proportionally.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The strongest claim rests on two legs; the load-bearing soft spot is in leg one. The free-carrier benchmark Eth = 2kBT/qd ≈ 0.7 mV/µm (Eq. 1) kills the free-carrier picture only if the field experienced by photocarriers in the channel interior is the nominal Vt/d with d = 5 µm. The paper assumes this throughout (simulations in Fig. 1(e), threshold extraction, TPVM mobility), but never measures the potential distribution across the gap. This is not a pedantic worry: the paper itself states that \"the metal–TI interface at each side contact presents a region of strong band bending and a high density of states,\" and such interfaces are exactly where a substantial fraction of an applied bias can drop (contact resistance, depletion regions). If, say, half of Vt dropped at the T1/T2 interfaces, the true threshold field would halve; if the drops were larger still, the 47× anomaly could shrink toward the free-carrier scale. The claim \"confirmed experimentally\" refers only to the ±Vt/2 self-referencing of the floating supply relative to ground, not to the field profile across the channel. Two mitigations exist: (i) at high fluence the system reverts to free-carrier-like behavior with a sensible drift mobility (~460 cm²/Vs) computed using E = Vt/d, which is an implicit field-calibration check; (ii) the polarity symmetry and temperature trends are unaffected by a common-mode rescaling. But mitigation (i) is never quantitatively cross-checked against an independent mobility at the same fluence, and a factor-of-several field error would still leave a \"sensible-looking\" mobility. The second leg (TPV Einstein violation, 48±35) is internally robust to geometric systematics, as the authors themselves argue, and does not depend on Vt/d calibration — so the neutrality conclusion would survive on TPV alone, but the headline \"50× larger than needed to deflect free carriers\" would not.","agreement_with_reader":"partial"},"referee_report":{"model":"moonshotai/kimi-k3","summary":"The manuscript reports a charge-neutrality test of the long-range photoresponse in bulk-insulating Sb-doped Bi2Se3 nanoribbons. A transverse electric field applied across a 5 µm gap between side contacts is used as a discriminator: free carriers should be deflected into the side-contact recombination sinks once drift beats diffusion, giving a threshold Eth = 2kBT/qd ≈ 0.7 mV/µm at 20 K (Eq. 1), independent of mobility; excitons should transit until the field ionizes them. Experimentally the photocurrent survives to ~10 mV/µm and collapses by ~60 mV/µm (50% crossing ≈ 33 mV/µm, ~47× the benchmark), consistent with an estimated exciton ionization field Eb/(qaB) ≈ 27 mV/µm. SPCM shows the suppression is a polarity-symmetric, position-independent global shift with a matching enhancement on the far side of the collection contacts, reproduced by a multiple-pass 1D exciton diffusion model (Eqs. 2–3) in which the biased region acts as a field-assisted (Poole–Frenkel) dissociator (Eq. 4); sqrt(Eth) decreases linearly with T as predicted, opposite to the free-carrier trend. Independently, ultrafast TPV yields (D/µ)/(kBT/q) = 48±35 at 12 K and low fluence — an order-of-magnitude Einstein-relation violation even at the lower uncertainty edge — relaxing toward unity at high T or fluence. The authors conclude the long-range photocarriers are a charge-neutral correlated e–h state, consistent with an exciton condensate.","tokens_in":12238,"tokens_out":6973,"duration_ms":45696,"significance":"If the result holds, this is an important contribution to the exciton-condensate search: a transport diagnostic of charge neutrality that complements Hall/counterflow and Coulomb-drag signatures, and is portable to platforms where spectroscopy is hard. The strengths deserve explicit credit: the free-carrier null benchmark of Eq. (1) is parameter-free by the Einstein cancellation and is corroborated by 2D drift-diffusion simulations shown to be mobility-insensitive; the temperature dependence of the threshold (decreasing, and linear in sqrt(Eth)) is opposite to the free-carrier prediction, a genuinely falsifiable discriminator; and two independent probes (steady-state SPCM and ultrafast TPV) converge on the same conditions for the anomaly. The TPV bound 48±35, while broad, exceeds unity by an order of magnitude even at its lower edge, and the degeneracy counter-explanation is argued away by its wrong fluence trend. The claim of outside-consensus novelty (excitonic, possibly condensed, photocarriers in a TI) is appropriately hedged in the body.","major_comments":[{"comment":"Eq. (1), Fig. 1(e): the headline '~50x free-carrier benchmark' assumes the interior field is Et = Vt/d with d = 5 um. Yet the manuscript itself invokes 'strong band bending and a high density of states' at the metal-TI side contacts (p. 6); such interface regions are precisely where a fraction of Vt could drop, reducing the interior field and shrinking the 47x anomaly proportionally. The cited confirmation ('self-references to +/-Vt/2... as confirmed experimentally') fixes the floating supply relative to ground, not the potential profile across the gap. The temperature trend, polarity symmetry, and Poole-Frenkel behavior are robust to a common-mode field rescaling, but the quantitative threshold claim is not. Please provide a direct bound on the potential drop: e.g., threshold *voltage* vs gap d across the five devices of Table S1, electrostatic simulation including the TI dielectric and","section":"§'Transverse field dependent photocurrent', Eq. (1), Fig. 1(e)"},{"comment":"Eq. (1) applies the Einstein relation at the *lattice* temperature T. A free-carrier reading survives if the photocarriers retain an elevated effective temperature over the transit time, since Eth = 2 kB Teff/(q d) then rises; matching 33 mV/um at 20 K would require Teff ~ 900-1000 K sustained over um-scale, ps-to-ns transit, which is implausible but not excluded by anything presented. The SI excludes Joule heating of the lattice, not photocarrier heating. The same loophole bears more strongly on the TPV leg (ps timescales, Fig. 4): the 12 K ratio (D/mu)/(kBT/q) = 48 could partly reflect hot-carrier diffusion (D/mu = kBTeff/q) rather than charge neutrality. Please bound the carrier cooling time against the measured transit times (e.g., using the cooling dynamics of Ref. 12) and discuss why the flat low-fluence field response excludes a hot-but-charged population.","section":"Eq. (1) and §TPV analysis, Fig. 4"},{"comment":"At the two lowest fluences at 12 K the fitted slopes of 1/tau vs Et are 'consistent with zero,' so mu is an upper bound rather than a measurement, and the reported 48 +/- 35 inherits a near-zero denominator. The abstract's claim that the bound 'holds even at the lower end of the fitting uncertainty' is only as strong as the one-sided upper limit on mu; this should be stated explicitly, with the raw slopes, their uncertainties, and the D-mu fit covariance reported. Also, Eq. (5) is a 1D escape-rate heuristic (4D/r^2 + muEt/r) applied to a 2D Gaussian spot; the escape coefficient for 2D Gaussian spreading differs numerically. The authors attribute the deviations from unity at 150 K to this prefactor; please quantify the systematic and show that it is common-mode across temperature and fluence, as asserted.","section":"Fig. 4(c),(f), Eq. (5)"},{"comment":"The Poole-Frenkel analysis is presented qualitatively but can be made overdetermined. Please (i) report the fitted beta from sT(Et) in Fig. 2(c) and compare it to the theoretical beta_PF = sqrt(q^3/(pi eps eps0)) for the known dielectric constant of Bi2Se3; (ii) from the intercept and slope of sqrt(Eth) vs T in Fig. 3(b), sqrt(Eth) = Eb/beta - C kB T/beta, extract Eb/beta and check consistency with the prior Eb = 3 meV, aB = 110 nm used for Ediss ~ 27 mV/um. Currently Eb and aB are rough priors from Ref. 13, so the 'consistent with the observed onset' statement is a weak check; supplying the numbers would either sharpen it into an independent determination or expose an inconsistency. Fit parameters for Fig. 3(b) should be given regardless.","section":"Fig. 2(c), Eq. (4), Fig. 3(b)"},{"comment":"The multiple-pass narrative requires that C1/C2 sense exciton density via electrochemical potential with negligible capture, while the chemically similar Cr/Au-T1/T2 interfaces act as perfect sinks; this asymmetry is assumed, not justified. Two requests: (i) a physical argument (geometry, band-bending orientation, contact area) for why the two interface types differ so strongly; (ii) a clarification of the IQE accounting: if Iph = Delta_mu_ec/(q Rch) is a potential-probe current, the operational meaning of '30% internal quantum efficiency (electrons collected per absorbed photon)' needs to be reconciled with the sensing picture. This does not carry the central claim, but the polarity-symmetric global shift and the equal suppression/enhancement -- the paper's most striking spatial evidence -- are interpreted entirely through Eqs. (2)-(3), so the assumption deserves explicit defense.","section":"Eqs. (2)-(3), multiple-pass model"}],"minor_comments":[{"comment":"Eq. (2) is dimensionally inconsistent as written: the left side has units of n/length^2 while the source term -(1/D) delta(x - x0) has units of time/length^3. Presumably a generation-rate amplitude is absorbed; please restore it or note the normalization.","section":"Eq. (2)"},{"comment":"Typo: 'modeled as localized loss of strengths T (units of um^-1)' should read 'strength s_T'.","section":"Text after Eq. (2)"},{"comment":"The claim that the mobility cancellation 'leav[es] no parameter that could account for the deviation' is too strong as written; field-distribution and carrier-heating caveats remain. Suggest softening in line with the revisions requested above.","section":"p. 7, paragraph after Eth estimate"},{"comment":"State which experimental quantity Ediss ~ 27 mV/um is being compared to: the onset (~10 mV/um), the 50% crossing (33 mV/um), or full suppression (60 mV/um). The text currently says 'consistent with the observed onset of suppression' while the threshold is defined by the 50% crossing.","section":"Exciton dissociation estimate"},{"comment":"Fig. 1(e): list the mobility values used for the simulated curves in the caption; the claim of mobility-insensitivity rests on these curves. In Fig. 4(f), mark the low-fluence points as upper bounds on mu. Table S1: report each device's gap d and threshold *voltage* in addition to threshold field.","section":"Figs. 1(e), 4(f), Table S1"},{"comment":"The abstract's closing sentence and the keyword 'exciton condensation' go beyond what is measured (charge-neutral correlated transport; no coherence or gap observable). The body hedges appropriately ('consistent with'); please align the abstract and keywords.","section":"Abstract/keywords"}],"recommendation":"major_revision","confidential_remarks":"The manuscript frames the result toward \"exciton condensation\" (keywords, abstract's final sentence), but the measurements establish charge-neutral correlated transport, not condensation: no coherence, gap, or order-parameter observable is accessed. The authors mostly hedge correctly (\"consistent with\"); the editor may nonetheless wish to check that the framing matches the evidence. Much of the motivating context (mm-scale transport, superdiffusion) rests on this group's own prior work; that is acceptable, but the independent-probe character of the present work should be weighed against that pedigree."},"author_rebuttal":null,"desk_editor":{"model":"grok-4.5","letter":"The load-bearing result here is real: at 20 K they need ~33 mV/µm to kill the photocurrent, against a parameter-free free-carrier benchmark Eth = 2kBT/qd ≈ 0.7 mV/µm that does not depend on mobility, and the temperature trend of √Eth goes the wrong way for free carriers. The TPV leg at 12 K/low fluence independently shows (D/µ)/(kBT/q) still >10 even at the bottom of 48±35. That combination is new relative to their earlier mm-transport and superdiffusion papers, and the high-fluence controls recovering free-carrier-like drift (~460 cm²/Vs) and Einstein ratios near one are the right sanity checks.\n\nWhat they do well is the discriminator geometry itself—side contacts as a field-tunable sink, polarity symmetry, SPCM maps that show a global shift rather than a local deflection, and the Poole–Frenkel-style √Eth(T). The free-carrier 2D drift-diffusion sims being mobility-insensitive is clean. Citation pattern is normal for a group extending its own platform; the bilayer QH / TMD condensate literature is engaged honestly.\n\nSoft spots, in proportion. The stress-test on Vt/d is fair but not fatal: they never map the potential across the 5 µm gap, and they themselves invoke strong band bending at T1/T2, so some fraction of Vt could drop at the interfaces. High-fluence mobility looking sensible under E = Vt/d is only a partial calibration. Even a factor-of-two field error leaves a large anomaly and does not touch the TPV Einstein violation, which does not need absolute E. The bigger interpretive soft spot is the contact-as-probe model (Iph ∝ density difference, capture neglected) needed for 30% IQE plus the multiple-pass global shift; if C1/C2 are strong sinks the spatial story changes. Eb and aB are rough priors, and “correlated state / condensate” is a step past “charge-neutral excitons.” Free parameters (η, sT, Ld, geometric prefactors) are acknowledged and mostly shape-fitting, not the null test.\n\nThis is for people working TI photocarriers, excitonic transport, or neutrality diagnostics. Math and data look solid enough for a serious referee; tighten field calibration, contact-capture bounds, and language separating neutral excitons from condensate. I would send it to review and read the SI carefully if I were on the panel.","headline":"Strong free-carrier exclusion via a clean transverse-field null test, with a softer leap from neutral excitons to condensate.","tokens_in":13137,"tokens_out":617,"would_cite":true,"duration_ms":22620,"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":"Long-range photoresponse in bulk-insulating Bi2Se3 nanoribbons is carried by charge-neutral excitons, not free carriers.","keywords":["topological insulator","nanoribbon","ultrafast photocurrent","exciton condensation","scanning photocurrent microscopy","transient photovoltage","Bi2Se3","charge neutrality"],"falsifier":"A device in which the side-contact gap and temperature still set Eth ≈ 2kBT/qd, yet the photocurrent collapses at that free-carrier field (or the low-T, low-fluence TPV ratio D/µ stays at kBT/q), would rule out neutral long-range transport under the paper’s own benchmark.","tokens_in":12734,"feed_emoji":"⚡","tokens_out":1073,"duration_ms":28112,"temperature":0.7,"pith_summary":"Photoexcited carriers in topological insulators can travel millimeter distances at low temperature, but whether that transport is free charge or something neutral has been hard to settle. This paper uses a sideways electric field as a charge filter on Sb-doped Bi2Se3 nanoribbons: free charges should be swept out of the channel at a small field set only by geometry and temperature, while neutral excitons should keep going until the field is strong enough to break them apart. Scanning photocurrent maps show the photocurrent survives until fields nearly fifty times that free-carrier benchmark, matching the expected exciton dissociation field, and the spatial pattern is a global, polarity-symmetric shift explained by excitons that repeatedly pass a localized dissociation zone. Ultrafast transient photovoltage independently finds that at 12 K and low fluence the diffusivity far exceeds what the Einstein relation would allow from the measured drift mobility. If correct, the long-range photoresponse is a charge-neutral correlated electron-hole state, and the transverse-field test becomes a practical transport probe of neutrality in other candidate excitonic systems.","feed_headline":"Sideways fields unmask neutral excitons in Bi2Se3 ribbons","feed_subtitle":"Photocurrent survives fields 50× too strong for free charges; diffusivity breaks the Einstein link at 12 K.","key_machinery":"Transverse-field discrimination: a side-contact bias deflects free carriers out of the channel at Eth set only by geometry and temperature, while charge-neutral excitons feel no center-of-mass force until field-assisted (Poole–Frenkel) dissociation; paired with a multiple-pass exciton diffusion model in which collection contacts act as electrochemical-potential probes of local exciton density.","core_discovery":"In bulk-insulating Sb-doped Bi2Se3 nanoribbons, the long-range photocurrent is inconsistent with free charge carriers and consistent with a charge-neutral correlated state. Suppressing the photocurrent requires a transverse field of order 33 mV/µm—about 47 times the free-carrier threshold Eth = 2kBT/qd set by drift-versus-diffusion across the side-contact gap—yet comparable to the estimated exciton ionization field. At 12 K and low fluence, transient photovoltage yields a diffusivity-to-mobility ratio more than an order of magnitude above the Einstein value even at the lower edge of the fit uncertainty; both the high threshold and the Einstein violation relax toward free-carrier behavior as","pith_inferences":["If the contacts truly do not capture most excitons, contact engineering that deliberately increases exciton dissociation at one end could turn the same nanoribbons into directional exciton-to-charge converters.","The method is geometry-first: any quasi-1D or strip geometry with a well-defined transverse gap can host the same Eth benchmark without requiring bilayer counterflow or Hall drag.","A direct spectroscopic search for a condensate gap in the mid-infrared on the same Sb-doped ribbons would be the natural next cross-check of the transport claim."],"forward_implications":["The millimeter-scale photoresponse and superdiffusion reported earlier in this material are reassigned to neutral excitonic transport rather than free-carrier diffusion.","Raising temperature or excitation density should continuously convert the photoresponse from excitonic to free-carrier behavior, lowering the transverse threshold and restoring Einstein agreement.","The transverse-field threshold and diffusivity–mobility decoupling become portable transport diagnostics of charge neutrality for other candidate excitonic platforms.","Field-ionized free carriers at the side contacts act as a localized sink, so the same geometry can map exciton dissociation rates versus field, temperature, and density."],"fun_headline_variants":["Transverse fields reveal excitons drive Bi2Se3 photocurrent","Excitons not free carriers explain long-range Bi2Se3 transport","50x free-carrier field needed to quench Bi2Se3 photocurrent","Einstein link broken at 12 K points to neutral state in Bi2Se3","Charge-neutral correlated state carries Bi2Se3 photocurrent"],"cache_read_input_tokens":128,"weakest_assumption_plain":"The longitudinal contacts mainly sense local exciton density as voltage probes and do not strongly absorb excitons or free carriers—needed both for the high collected efficiency with millimeter neutral diffusion and for the polarity-symmetric global photocurrent shift.","fun_headline_variants_meta":{"raw":{"variants":["Transverse fields reveal excitons drive Bi2Se3 photocurrent","Excitons not free carriers explain long-range Bi2Se3 transport","50x free-carrier field needed to quench Bi2Se3 photocurrent","Einstein link broken at 12 K points to neutral state in Bi2Se3","Charge-neutral correlated state carries Bi2Se3 photocurrent"]},"model":"grok-4.5","effort":"low","cost_usd":0.005237,"raw_usage":{"total_tokens":1498,"prompt_tokens":829,"num_sources_used":0,"completion_tokens":80,"cost_in_usd_ticks":52368000,"prompt_tokens_details":{"text_tokens":829,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":589,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":829,"tokens_out":80,"duration_ms":12431,"temperature":1.0,"reasoning_tokens":589,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-31T00:33:59.646004+00:00","model_set":{"reader":"grok-4.5"},"falsifier":"A device in which the side-contact gap and temperature still set Eth ≈ 2kBT/qd, yet the photocurrent collapses at that free-carrier field (or the low-T, low-fluence TPV ratio D/µ stays at kBT/q), would rule out neutral long-range transport under the paper’s own benchmark.","supporting_citations":[],"review_version":1}