{"id":"06a9e88e-f3ef-442b-9190-25bc312d8a59","arxiv_id":"2608.12105","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Using angle-resolved cathodoluminescence, the authors show that dark excitons in monolayer WSe2 emit at large angles, providing a nanostructure-free route for directional light emission.","lead":"A team measured the direction of light emitted by atom-thin semiconductors after electron-beam excitation, and found that different types of electron-hole pairs (excitons) radiate in different directions. This suggests a way to steer light at the nanoscale without etching special patterns into materials.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Dark-exciton assignment of the 1.69 eV CL feature is load-bearing and underdetermined: the large-angle profile is both identification evidence and claimed phenomenon, with no independent confirmation or fit uncertainties.","rationale":"The reader and I identify the same soft spot, so agreement_with_reader is agree. The manuscript has real supporting evidence: three independently fabricated samples, angle-resolved data on two additional samples in the Supporting Information, temperature-dependent behavior, and Monte Carlo support for the hBN-thickness effect. These establish the experimental capability, but they do not independently pin the 1.69 eV feature to the dark exciton. The ~40 meV splitting is necessary but not sufficient: defect-bound excitons and localized out-of-plane emitters are common in WSe2 and can appear in this energy range, especially under electron-beam excitation. The in-plane magnetic-field test is decisive because the dark exciton's brightening and angular-profile change under an in-plane field are known distinctive signatures that defect states would not share. My read therefore leaves the conditional verdict unchanged rather than moving it.","tokens_in":9284,"tokens_out":8293,"duration_ms":84600,"concrete_test":"Apply an in-plane magnetic field to the same hBN-encapsulated WSe2 monolayer while recording angle-resolved cathodoluminescence spectra. For the intrinsic spin-forbidden dark exciton, an in-plane field mixes it with the bright exciton, transferring in-plane dipole character and brightening the transition; the 1.69 eV feature should then appear at smaller emission angles, increase in absolute intensity, and follow the known dark-exciton Zeeman behavior. If the feature remains confined to large angles and does not brighten, the dark-exciton assignment is not supported, and a defect-bound or localized out-of-plane emitter becomes the more likely explanation.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim requires that the 1.69 eV cathodoluminescence feature in WSe2 be the spin-forbidden dark exciton. In the current manuscript this identification rests on two observations: the ~40 meV splitting below X0, quoted as consistent with refs [10,17,21], and the fact that the feature is seen only for emission angles above about 25 degrees, which is read as the out-of-plane dipole pattern. The latter is the problem: the angular profile is used both to identify the species and as the phenomenon being claimed. The text near Fig. 2d states that spectral fitting still allows all excitonic contributions to be reliably identified, but the paper reports no error bars, confidence intervals, or covariance information for the three-Voigt decomposition. The fitted intensities enter the normalized weight w(j) defined near Fig. 3; if XD is a weak shoulder buried under X0 at small angles, its fitted amplitude there is poorly constrained, and the apparent rise of w(XD) at large angles could be a fitting artifact rather than a physical dipole pattern. Alternative out-of-plane emitters in the same spectral window (defect-bound excitons, electron-beam-induced localized states) are not experimentally excluded. Because the beam-routing claim depends on this being the intrinsic dark exciton, an alternative assignment would reduce the result to observation of large-angle emission from some out-of-plane emitter, not the claimed exciton-selective routing mechanism.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports low-temperature, angle-resolved cathodoluminescence (CL) measurements on hBN-encapsulated monolayer WSe2, MoSe2, and MoTe2. In WSe2 the spectra are decomposed into three Voigt lines plus a low-order polynomial background and assigned to the bright exciton X0, the trion XT, and the spin-forbidden dark exciton XD on the basis of a ~20 meV XT shift, a ~40 meV XD splitting, and the angular dependence of the fitted intensities. The authors show that X0 and XT emit predominantly near the surface normal while XD appears only at large emission angles, and they support this with a simplified electromagnetic simulation. They additionally demonstrate that the local hBN thickness and an adjacent graphene layer change the trion-to-exciton CL ratio, and they conclude that intrinsic excitonic transition dipoles can act as exciton-selective nanoscale beam routers without nanostructuring.","tokens_in":9526,"tokens_out":4313,"duration_ms":42077,"significance":"If the central assignment holds, the paper establishes a new and potentially general mechanism for directional emission from unpatterned 2D semiconductors: the out-of-plane transition dipole of the dark exciton produces a large-angle CL channel that is inaccessible under normal-incidence optical excitation. The angle-resolved CL methodology, the multi-sample consistency, the temperature dependence, and the dielectric-environment tuning are notable strengths, and the angular trends are qualitatively reproduced by an independent simulation. The work is falsifiable and could influence nanophotonics and 2D-material spectroscopy. However, the load-bearing identification of the 1.69 eV feature as the intrinsic dark exciton is not yet sufficiently supported because the angular profile is used both as evidence for the assignment and as the claimed phenomenology, and because the three-Voigt decomposition lacks quantified uncertainties.","major_comments":[{"comment":"The identification of the 1.69 eV feature as the spin-forbidden dark exciton is load-bearing for the central routing claim, but it is currently underdetermined. The text argues that this feature is XD because it appears only above ~25 degrees and sits ~40 meV below X0; the same large-angle behavior is then presented as the routing phenomenon. This creates a self-referential element: the angular dependence is simultaneously the evidence for the assignment and the effect being claimed. The three-Voigt decomposition is not accompanied by uncertainties, confidence intervals, or covariance information, so the small-angle amplitude of the 1.69 eV shoulder buried under X0 is poorly constrained, and the rise of w(XD) in Fig. 3a could be a fitting artifact. I ask the authors to report fit uncertainties and to test explicitly whether an alternative out-of-plane emitter (defect-bound exciton, phonon replica, or beam-induced localized state) can be excluded by additional diagnostics, for example polarization-resolved CL, excitation-density dependence, or sample-to-sample statistics on the 40 meV splitting.","section":"§2, Fig. 2b, and Eq. (w(j))"},{"comment":"The polar plots in Fig. 3 contain no error bars, and the dashed simulation is described only as 'simplified' with no parameters or goodness-of-fit measure in the main text. The normalized weight w(j)=I(j)/(I(X0)+I(XT)+I(XD)) removes common angle-dependent detection efficiency only if all three resonances are affected equally, an assumption that is not justified; for example, if the detection efficiency varies strongly with angle near the mirror cutoff, the apparent increase of w(XD) could be exaggerated. The authors should propagate fitting uncertainties through w(j), show the raw intensity trends before normalization, and provide a more transparent description of the simulation (stack geometry, dipole orientation, and inclusion of hBN and substrate) so the reader can judge whether the predicted large-angle pattern is unique to an out-of-plane dipole.","section":"Fig. 3 and the normalization w(j)"},{"comment":"The claim that hBN thickness tunes the trion-to-exciton balance is based on visual correlation in a single sample, and the spectra in Fig. 4b are normalized to their own maxima, which obscures absolute intensity changes. If this environmental-tuning result is intended as a quantitative finding, the authors should provide statistics across samples, a measure of the spatial correlation, and a discussion of possible systematic effects such as local beam-current variations or thickness-dependent carrier generation that are not captured by the simplified Monte Carlo argument.","section":"§4, Fig. 4"}],"minor_comments":[{"comment":"The axis label 'Voigt /f_it' appears corrupted and should be replaced with a readable label such as 'Intensity' or a legend describing the fit components.","section":"Fig. 2b"},{"comment":"The gray shaded area indicating the cutoff angle of the parabolic mirror is not quantified; please give the numerical cutoff angle and state how data beyond it are handled.","section":"Fig. 3 caption and main text"},{"comment":"The sentence claiming 'the first identification of individual excitonic species from the angular distribution of their cathodoluminescence' should be supported by an explicit comparison with prior angle-resolved CL studies of excitonic emitters, or softened, since 'first' claims require a documented literature search.","section":"Main text near 'first identification'"},{"comment":"The text says the normalization accounts for the instrument response, but Eq. (w(j)) only divides by the total fitted intensity; the actual instrument-response correction should be stated explicitly, including whether it was applied to each resonance before computing w(j).","section":"Fig. 3 normalization"},{"comment":"The assignment of the 1.71 eV shoulder to the trion is based on a ~20 meV shift with no independent diagnostic; a sentence giving the expected trion binding energy or a comparison with the local carrier density would strengthen the case.","section":"§2, trion assignment"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is within scope and the core idea is novel, but the central dark-exciton identification needs to be made robust before publication. I see no misconduct or citation concerns; the main risk is that the angular-dependence evidence is partly circular and the fitting uncertainties are absent. A major revision with additional quantitative support, rather than rejection, seems appropriate."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here's my take. The genuinely new thing is the first angle-resolved cathodoluminescence measurement that resolves different excitonic species by their angular emission profiles in a TMD monolayer. That is a real and useful step, not a revolution, and the WSe2 data showing a large-angle feature 40 meV below X0 are consistent with the known out-of-plane dark-exciton dipole. The paper also does several things well: multiple samples, temperature series, control of the trion/exciton balance via hBN thickness and graphene coupling, and a simplified but sensible dipole simulation that matches the observed trends.\n\nThe soft spot is the one the stress test flags. The ~1.69 eV feature is identified as the spin-forbidden dark exciton partly from the literature splitting and partly from the fact that it appears only at large angles. That second piece is the same angular behavior the paper claims as the phenomenon, so there is a mild circularity. The text says the spectral fitting allows reliable identification, but no fit uncertainties or confidence intervals are reported, and the polar plots carry no error bars. If the 1.69 eV emitter were a defect-bound exciton or a phonon replica with out-of-plane character, the central \"beam routing through the dark exciton\" narrative would shrink to \"some out-of-plane emitter radiates at large angles.\" I don't think that alternative is likely—the 40 meV splitting is well established for WSe2 and the high-angle shoulder is visible in the raw spectra, not just in the deconvolution—but it is not excluded. A magnetic-field dependence or time-resolved measurement would settle it.\n\nThe \"beam routing\" language is also a bit generous for a fixed angular pattern; the paper demonstrates species-selectable emission direction, not an actively tunable route. Minor.\n\nOverall this is a fair, careful experimental study. It deserves a serious referee. The referee should ask for error bars on the fits and for an independent confirmation of the dark-exciton assignment, ideally field-dependent CL. If those are added, I would be happy to see the paper published. I would cite it for the CL methodology even now.\n\nRecommendation: engage with it; send to peer review.","headline":"Genuinely new angle-resolved CL species identification, with a plausible dark-exciton large-angle emission claim that would be stronger with fit error bars and an independent spectroscopic check.","tokens_in":10088,"tokens_out":2327,"would_cite":true,"duration_ms":22369,"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":"In monolayer WSe2, the dark exciton's out-of-plane dipole sends cathodoluminescence sideways, making dipole orientation a routing mechanism without nanostructures.","keywords":["cathodoluminescence","angle-resolved spectroscopy","dark excitons","transition metal dichalcogenides","WSe2 monolayers","transition dipole orientation","beam routing","hBN encapsulation"],"falsifier":"Refit the measured angle-resolved spectra with realistic added noise using a two-peak model without the 1.69 eV feature, and also measure polarization at collection angles above 50 degrees: if the large-angle feature can be absorbed into the neutral-exciton tail or turns out to be s-polarized rather than p-polarized, the dark-exciton routing claim is falsified.","tokens_in":9081,"feed_emoji":"💡","tokens_out":9154,"duration_ms":77551,"temperature":0.7,"pith_summary":"This paper claims that directional light routing at the nanoscale can be achieved by the intrinsic excitonic transitions of a semiconductor, without any nanostructured surfaces. Using low-temperature angle-resolved cathodoluminescence on hBN-encapsulated monolayers of WSe2, MoSe2, and MoTe2, the authors resolve the bright exciton, the trion, and the spin-forbidden dark exciton of WSe2 by their characteristic angular emission profiles. The dark exciton, whose transition dipole points out of the monolayer plane, emits preferentially at large angles, whereas the in-plane dipoles of the bright exciton and trion radiate mainly toward the surface normal. If correct, this turns a normally 'dark' electronic state into a built-in directional light source and provides a material-level mechanism for nanoscale beam routing.","feed_headline":"Dark excitons route light sideways with no nanostructures","feed_subtitle":"Angle-resolved electron-beam spectroscopy tells exciton species apart, making the dark state a directional emitter.","key_machinery":"The mechanism is angle-resolved low-temperature cathodoluminescence spectroscopy of hBN-encapsulated monolayer TMDs, combined with the transition-dipole orientation of each excitonic species. An electron beam generates carriers in hBN that relax into the monolayer; a parabolic mirror with numerical aperture 0.97 collects emission as a function of emission angle, and each spectrum is decomposed into three Voigt peaks corresponding to the neutral exciton, trion, and dark exciton. The normalized spectral weight removes angle-dependent collection efficiency and reveals that the in-plane-dipole species peak near the surface normal while the dark exciton grows toward large angles, the signature of an out-of-plane dipole. The load-bearing identity is the spin-forbidden dark exciton's out-of-plane dipole, which converts a normally dark state into a directional emitter when excited by the electron beam.","core_discovery":"The central discovery is that cathodoluminescence can distinguish excitonic species in transition metal dichalcogenide monolayers not only by emission energy but also by angular emission profile, establishing transition dipole orientation as a mechanism for exciton-selective nanoscale beam routing. In WSe2, the neutral exciton at about 1.73 eV and the trion at about 1.71 eV have in-plane dipoles and emit mainly near the surface normal, while a feature at about 1.69 eV assigned to the spin-forbidden dark exciton grows with emission angle above roughly 25 degrees, matching an out-of-plane dipole. The paper further shows that the local dielectric environment, namely hBN thickness, substrate material, and a nearby graphene layer, acts as a passive control on the balance between neutral and charged exciton emission, reshaping the routed spectrum without electrostatic gating.","pith_inferences":["If the angular fingerprint is as robust as the three-peak fits suggest, the same angle-resolved cathodoluminescence could identify dark-exciton character in other monolayer semiconductors and in moire superlattices, where dark states are predicted but hard to access optically.","Coupling the out-of-plane dark-exciton channel to a waveguide or polariton structure could yield on-chip beam routing in which the emitting species itself selects the guided direction, an extension beyond the free-space angular measurement reported here.","A polarization-resolved variant would be a sharp test: an out-of-plane dipole should emit predominantly p-polarized light at large angles, while an in-plane defect state would not follow the same polarization-angle pattern.","Electrostatic gating of the same WSe2 stack should shift the trion-to-exciton balance in a way comparable to the passive graphene contact, providing a direct electrical check of the proposed environment control."],"forward_implications":["Dark-exciton states in TMD monolayers can act as directional emitters at large angles with no grating, antenna, or waveguide structuring.","Angle-resolved cathodoluminescence becomes a species-resolving probe: each exciton species carries an angular fingerprint set by its transition dipole orientation.","The hBN encapsulation thickness and adjacent graphene act as passive controls that shift weight between neutral and charged exciton emission, letting substrate engineering tune both emission energy and angular channel.","Because the dark exciton is the lowest-energy state in WSe2 and remains resolved to about 100 K, it provides a spectrally isolated directional channel for compact photonic devices.","Electron-beam excitation reaches out-of-plane dipoles that normal-incidence optical excitation cannot, extending the probe of dark states to spatially local regions."],"supporting_citations":[{"why":"Supplies the review-level framework of bright and dark exciton species in TMD monolayers, including selection rules and expected splittings.","marker":"[10]"},{"why":"Reports the dark exciton below the bright exciton in WSe2, anchoring the 40 meV energy assignment.","marker":"[17]"},{"why":"Measured out-of-plane transition dipole of excitons in monolayer TMDs, supporting the angular-profile interpretation.","marker":"[20]"},{"why":"Provides the established neutral-dark exciton splitting value used to identify the 1.69 eV feature as the dark exciton.","marker":"[21]"},{"why":"Theoretical basis for the spin-forbidden dark exciton carrying an out-of-plane transition dipole.","marker":"[22]"},{"why":"Review of cathodoluminescence mechanisms that justifies treating the electron-beam signal as incoherent excitonic emission.","marker":"[29]"},{"why":"Establishes the angle-resolved cathodoluminescence methodology that allows angular emission profiles to be measured.","marker":"[41]"},{"why":"Provides the stack simulation used to compare the measured angular profiles with expected dipole emission patterns.","marker":"[44]"}],"fun_headline_variants":["Dark excitons route light at large angles in monolayers","Excitonic species separated by angular emission profiles","No nanostructures needed for dark exciton beam routing","Cathodoluminescence reveals dark exciton directional emission"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The central claim rests on identifying the 1.69 eV emission as the spin-forbidden dark exciton, inferred from its roughly 40 meV separation below the neutral exciton and its appearance only at large angles; if that feature is instead a defect-bound state or phonon replica with out-of-plane character, the beam-routing story does not follow.","fun_headline_variants_meta":{"raw":{"variants":["Dark excitons route light at large angles in monolayers","Excitonic species separated by angular emission profiles","No nanostructures needed for dark exciton beam routing","Cathodoluminescence reveals dark exciton directional emission"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000551,"raw_usage":{"total_tokens":2605,"prompt_tokens":895,"completion_tokens":1710,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":511,"completion_tokens_details":{"reasoning_tokens":1646}},"tokens_in":511,"tokens_out":1710,"duration_ms":13249,"temperature":1.0,"reasoning_tokens":1646,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T00:15:21.855352+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Refit the measured angle-resolved spectra with realistic added noise using a two-peak model without the 1.69 eV feature, and also measure polarization at collection angles above 50 degrees: if the large-angle feature can be absorbed into the neutral-exciton tail or turns out to be s-polarized rather than p-polarized, the dark-exciton routing claim is falsified.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the review-level framework of bright and dark exciton species in TMD monolayers, including selection rules and expected splittings."},{"cited_title":"Zhang, T","cited_arxiv_id":null,"evidence_quote":"Reports the dark exciton below the bright exciton in WSe2, anchoring the 40 meV energy assignment."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Measured out-of-plane transition dipole of excitons in monolayer TMDs, supporting the angular-profile interpretation."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the established neutral-dark exciton splitting value used to identify the 1.69 eV feature as the dark exciton."},{"cited_title":"Scharf, G","cited_arxiv_id":null,"evidence_quote":"Theoretical basis for the spin-forbidden dark exciton carrying an out-of-plane transition dipole."},{"cited_title":"Polman, M","cited_arxiv_id":null,"evidence_quote":"Review of cathodoluminescence mechanisms that justifies treating the electron-beam signal as incoherent excitonic emission."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes the angle-resolved cathodoluminescence methodology that allows angular emission profiles to be measured."},{"cited_title":"Akerboom, H","cited_arxiv_id":null,"evidence_quote":"Provides the stack simulation used to compare the measured angular profiles with expected dipole emission patterns."}],"review_version":1}