# Fileset

[s41467-024-47099-6.pdf](https://mdr.nims.go.jp/filesets/c20747a4-6109-4cd1-94db-d07a99fcf077/download)

## Creator

N. Fang, Y. R. Chang, S. Fujii, D. Yamashita, M. Maruyama, Y. Gao, C. F. Fong, [D. Kozawa](https://orcid.org/0000-0002-0629-5589), K. Otsuka, K. Nagashio, S. Okada, Y. K. Kato

## Rights

[Creative Commons BY Attribution 4.0 International](https://creativecommons.org/licenses/by/4.0/)

## Other metadata

[Room-temperature quantum emission from interface excitons in mixed-dimensional heterostructures](https://mdr.nims.go.jp/datasets/5a3010b1-3588-4391-8434-bbf12019f7c9)

## Fulltext

Room-temperature quantum emission from interface excitons in mixed-dimensional heterostructuresArticle https://doi.org/10.1038/s41467-024-47099-6Room-temperature quantum emission frominterface excitons in mixed-dimensionalheterostructuresN. Fang 1 , Y. R. Chang 1, S. Fujii 2,3, D. Yamashita2,4, M. Maruyama 5,Y. Gao5, C. F. Fong 1, D. Kozawa1,2,6, K. Otsuka 1,7, K. Nagashio 8,S. Okada 5 & Y. K. Kato 1,2The development of van der Waals heterostructures has introduced uncon-ventional phenomena that emerge at atomically precise interfaces. Forexample, interlayer excitons in two-dimensional transition metal dichalco-genides show intriguing optical properties at low temperatures. Here wereport on room-temperature observation of interface excitons in mixed-dimensional heterostructures consisting of two-dimensional tungsten dis-elenide and one-dimensional carbon nanotubes. Bright emission peaks origi-nating from the interface are identified, spanning a broad energy range withinthe telecommunication wavelengths. The effect of band alignment is investi-gated by systematically varying the nanotube bandgap, and we assign the newpeaks to interface excitons as they only appear in type-II heterostructures.Room-temperature localization of low-energy interface excitons is indicatedby extended lifetimes as well as small excitation saturation powers, and pho-ton correlationmeasurements confirm antibunching.Withmixed-dimensionalvan derWaals heterostructures where band alignment can be engineered, newopportunities for quantum photonics are envisioned.The discovery of van der Waals (vdW) materials, including two-dimensional (2D) transition metal dichalcogenides (TMDs) and gra-phene, has brought about a revolution in the assembly of artificialheterostructures by allowing for the combination of two differentmaterials without the constraints of lattice matching. Such an unpre-cedented level of flexibility in heterostructure design has led to theemergence of novel properties not seen in individual materials. Aprime example is twisted bilayer graphene at magic angles, whichexhibits exotic phases such as correlated insulating states1 andsuperconductivity2. Another notable development is the stacking oftwo TMDs, resulting in the observation of unique excitons known asinterlayer excitons, characterized by electrons and holes located inseparate layers3–6. The spatially indirect nature of interlayer excitonsimparts them with distinct properties, including long excitonlifetimes3, extended diffusion lengths7, large valley polarization8, andsignificant modulation by moiré potentials9,10.The existing vdW heterostructures comprise 2D materials withsimilar lattice structure, excitonic characteristics, and inherentlyidentical dimensions. Development of vdW heterostructures thatencompass lower dimensional materials may give rise to uniqueinterface exciton states resulting from the mixed dimensionality.Carbon nanotubes (CNTs), a typical one-dimensional (1D)material, areideal for such heterostructures as they have all bonds confined to thetube itself11,12. CNTs interact with 2D materials through weak vdWReceived: 5 September 2023Accepted: 19 March 2024Check for updates1NanoscaleQuantumPhotonics Laboratory, RIKENCluster for Pioneering Research, Saitama, Japan. 2QuantumOptoelectronics Research Team, RIKENCenterfor Advanced Photonics, Saitama, Japan. 3Department of Physics, Keio University, Yokohama, Japan. 4Platform Photonics Research Center, National Instituteof Advanced Industrial Science and Technology (AIST), Ibaraki, Japan. 5Department of Physics, University of Tsukuba, Ibaraki, Japan. 6Research Center forMaterials Nanoarchitectonics, National Institute for Materials Science, Ibaraki, Japan. 7Department of Mechanical Engineering, The University of Tokyo,Tokyo, Japan. 8Department of Materials Engineering, The University of Tokyo, Tokyo, Japan. e-mail: nan.fang@riken.jp; yuichiro.kato@riken.jpNature Communications |         (2024) 15:2871 11234567890():,;1234567890():,;http://orcid.org/0000-0002-1053-1900http://orcid.org/0000-0002-1053-1900http://orcid.org/0000-0002-1053-1900http://orcid.org/0000-0002-1053-1900http://orcid.org/0000-0002-1053-1900http://orcid.org/0000-0003-1629-3525http://orcid.org/0000-0003-1629-3525http://orcid.org/0000-0003-1629-3525http://orcid.org/0000-0003-1629-3525http://orcid.org/0000-0003-1629-3525http://orcid.org/0000-0002-0998-366Xhttp://orcid.org/0000-0002-0998-366Xhttp://orcid.org/0000-0002-0998-366Xhttp://orcid.org/0000-0002-0998-366Xhttp://orcid.org/0000-0002-0998-366Xhttp://orcid.org/0000-0002-2872-5543http://orcid.org/0000-0002-2872-5543http://orcid.org/0000-0002-2872-5543http://orcid.org/0000-0002-2872-5543http://orcid.org/0000-0002-2872-5543http://orcid.org/0000-0003-1676-4665http://orcid.org/0000-0003-1676-4665http://orcid.org/0000-0003-1676-4665http://orcid.org/0000-0003-1676-4665http://orcid.org/0000-0003-1676-4665http://orcid.org/0000-0002-6694-0738http://orcid.org/0000-0002-6694-0738http://orcid.org/0000-0002-6694-0738http://orcid.org/0000-0002-6694-0738http://orcid.org/0000-0002-6694-0738http://orcid.org/0000-0003-1181-8644http://orcid.org/0000-0003-1181-8644http://orcid.org/0000-0003-1181-8644http://orcid.org/0000-0003-1181-8644http://orcid.org/0000-0003-1181-8644http://orcid.org/0000-0002-0783-3596http://orcid.org/0000-0002-0783-3596http://orcid.org/0000-0002-0783-3596http://orcid.org/0000-0002-0783-3596http://orcid.org/0000-0002-0783-3596http://orcid.org/0000-0002-9942-1459http://orcid.org/0000-0002-9942-1459http://orcid.org/0000-0002-9942-1459http://orcid.org/0000-0002-9942-1459http://orcid.org/0000-0002-9942-1459http://crossmark.crossref.org/dialog/?doi=10.1038/s41467-024-47099-6&domain=pdfhttp://crossmark.crossref.org/dialog/?doi=10.1038/s41467-024-47099-6&domain=pdfhttp://crossmark.crossref.org/dialog/?doi=10.1038/s41467-024-47099-6&domain=pdfhttp://crossmark.crossref.org/dialog/?doi=10.1038/s41467-024-47099-6&domain=pdfmailto:nan.fang@riken.jpmailto:yuichiro.kato@riken.jpforces, resulting in well-defined, atomically sharp interfaces13,14. Thechirality-dependent bandgap of CNTs can be utilized to tune the bandalignment as demonstrated in exciton transfer process15, allowing forunambiguous identification of excitonic states at the 1D-2D interface.Herewe report on the observation ofmultiple emergent excitonicpeaks in the 1D-2D CNT/tungsten diselenide (WSe2) heterostructuresat room temperature. These peaks appear exclusively at the interfaceregion with a broad energy range lower than CNT E11 states, and theirdependence on the chirality of CNTs and the layer number of WSe2 isinvestigated. The emergence of the peaks is found to be highly cor-related with the band alignment, and they are interpreted as interfaceexcitons. Prominent linear polarization, low excitation saturationpower, and a long lifetime are characteristic of low-energy interfaceexcitons, suggesting strong confinement. Through photon correlationmeasurements, room-temperature antibunching has been confirmed.These findings expand the existing concept of spatially indirect exci-tons based on 2D heterostructures to 1D systems, demonstrating sig-nificant potential of the interface excitons for nanophotonics andquantum information processing.Results and discussionEmerging peaks in 1D-2D mixed-dimensional heterostructuresThe CNT/WSe2 heterostructures under investigation are entirely free-standing to preclude substrate effects5,16, as depicted in Fig. 1a, b. High-quality CNTs are initially grown over trenches (see Methods and Sup-plementary Fig. 1), followed by placement of a WSe2 flake upon thetubes using the anthracene-assisted transfer technique17,18.Weperformphotoluminescence excitation (PLE) measurements to determine thechirality before transfer for all samples. A (9,4) CNT is selected as arepresentative case, forming type-II band alignment with WSe215. Suchalignment should establish emergent excitonic states between theCNT conduction bandminimumand theWSe2 valence bandmaximum(Fig. 1c). The formation of the indicated indirect excitons generallyrequires charge transfer,which is plausible as exciton transfer hasbeenobserved in similar heterostructures with type-I band alignment15.Room-temperature photoluminescence (PL) spectroscopy isemployed to investigate the excitonic states present within theheterostructure19–21. The PL spectrum of the pristine, suspended (9,4)CNT displays a singular peak at 1.143 eV, corresponding to the E11transition (Fig. 1d)11. After the transfer of themonolayerWSe2 flake, theCNT E11 peak is redshifted to 1.102 eV (Fig. 1e) as a result of thedielectric screening effect14, indicating intimate contact between thetwo materials. Notably, two excitonic peaks arise with energies lowerthan E11. These peaks cannot be attributed to the suspended mono-layer (1L) WSe2 emission as only the A exciton peak at 1.658 eV isexpected (see Supplementary Fig. 2). The transfer process is notexpected to introduce defects, since CNT/hexagonal boron nitrideheterostructures prepared in a similar manner does not exhibit suchlow-energy peaks14, suggesting the existence of emergent excitonicstates in the CNT/WSe2 heterostructure. The newly emerged peaks areinitially unstable and exhibit temporal blinking (inset in Fig. 1e). SuchPL evolution is not observed in sp3 defects, which exhibit more stableemission once formed22,23. Following a spectral development involvingfluctuations of the peaks, the unstable excitonic states vanish andstable states remain for which we perform the subsequent measure-ments (see Supplementary Fig. 3). Figure 1f presents the stable PLFig. 1 | Emergenceof interface excitons in 1D-2Dheterostructures. aAschematicof a suspended carbon nanotube (CNT)/WSe2 heterostructure. b An opticalmicroscope image of the (9,4) CNT/1LWSe2 heterostructure. The substrate is SiO2/Si. The red and white dashed lines indicate CNT and WSe2, respectively. The scalebar represents 5μm. cThe band diagramof the (9,4) CNT/1LWSe2 heterostructure,illustrating the excitons in CNT and WSe2 as well as the interface exciton. d Aphotoluminescence (PL) spectrum of the pristine, suspended (9,4) CNT prior toWSe2 transfer. e A PL spectrum immediately following the transfer of 1L WSe2 toform the heterostructure. Emergent peaks are denoted by IX. Arrows indicateunstable IX peaks. The inset depicts the time-trace of the integrated PL intensity forthe IX peak indicated by the orange arrow. The peaks at 1.011 and 0.962 eV ind, e are K-momentum exciton peaks. f, A PL spectrum from the same sample afterapproximately eight hours of optical measurements. E11 indicates the CNT E11 state,and IX1 and IX2 indicate the emergent excitonic states. g, h PL intensity maps of E11g and IX1 h. The edges of the trench are indicated by the green broken lines. Thescale bar represents 1μm. i A photoluminescence excitation (PLE) map of the (9,4)CNT/1L WSe2 heterostructure. The excitation power is 4.5 and 5μW for f andi, and 10μW otherwise. The excitation energy is 1.730 eV for d, e, 1.680 eV forf, and 1.653 eV for g, h.Article https://doi.org/10.1038/s41467-024-47099-6Nature Communications |         (2024) 15:2871 2spectrum, featuring two prominent peaks at 0.924 eV and 0.821 eV,denoted as IX1 and IX2, respectively. With an energy difference ofapproximately 0.278 eV from E11, IX2 is at a lower energy than anyreported dark or defect states in (9,4) CNTs24–26. Such low-energypeaks are also observed in other monolayer WSe2 heterostructureswith type-II band alignment (see Supplementary Fig. 4). We hypothe-size these states to be interface excitons, which could possess sub-stantially lower excitonic energies as determined by theheterostructure band alignment.The spatial and spectral correlationbetweenE11 excitons and IXs isinvestigated by conducting PL imaging measurement and PLE spec-troscopy. Figure 1g is an integrated PL image from E11, exhibitingstrong signal above the trench due to the quenching from the silicondioxide (SiO2)/silicon (Si) substrate. Both IX1 and IX2 peaks areobserved precisely at the position of the E11 peak, as demonstrated bythe IX1 and IX2 images displayed in Fig. 1h and Supplementary Fig. 5,respectively. Such spatial overlapwith the CNT emission indicates thatIX peaks cannot be explained by emission from randomly distributeddefect states within WSe2. In the PLE map (Fig. 1i), E11 shows a strongresponse to excitation energy, which is identified as CNT E22 transition(Supplementary Fig. 6). Similar response as E11 is observed for IX1 andIX2 peaks, implying that the carriers forming the IXs are supplied fromthe CNT. In comparison, we do not observe a clear signature of theWSe2 A exciton peak in the PLE map. Considering IXs only emergeupon transfer of the WSe2 flake, the spatial and spectral overlap withCNT supports the hypothesis that they originate from the interface.Band alignment effect on IX peaksSince interface excitons formbetween the twomaterials, manipulatingthe heterostructure band alignment should affect the IXs6,27. It is pos-sible to vary the CNT bandgap (Fig. 2a) by studying different CNTchiralities (Fig. 2b), whereas WSe2 bandgap can be altered (Fig. 2c)through the layer number (Fig. 2d, Supplementary Fig. 2).We first investigate the chirality dependence, which significantlymodulates the CNT bandgap. The band alignment is systematicallytuned by utilizing CNT/WSe2 heterostructures with different CNTchiralities as illustrated in Fig. 2b. WSe2 layer numbers ≤ 4 are used forheterostructures since IX peaks can be observed as shown in the casefor (9,4) CNTs with bilayer (2L) and quadlayer (4L) WSe2. It is notedthat the valance band maximum of WSe2 changes from the K point tothe Γ point with increasing the layer number, but correlation with thebehavior of IX peaks cannot be observed. Multiple IX peaks appear in(9,4), (12,1), (8,6), (8,7), and (14,0) CNTs, which have large E11 energiesand therefore large bandgaps. This is consistent with the expectationthat a largebandgap is favorable for type-II band alignment asdepictedin Fig. 2a. It is noted that sp3 defects in CNTs generally introducedoublet peaks23, different from the numerous IX peaks observed here(see Supplementary Fig. 3). The observed IXpeaks spana broad energyrange within the telecommunication wavelengths. Note that the low-energy peaks approach the edge of the spectral window, suggestingthe possibility of lower-energy IX peaks existing beyond our currentdetection capability. Meanwhile, the highest energy peak in eachchirality is located close to E11, with a difference of ~ 0.05 eV.Remarkably, a further decrease in the bandgap leads to the dis-appearance of IXs (Supplementary Fig. 7). The presence of the IX peaksis determined by chirality, consistent with the transition in bandalignment from type-II to type-I that is observed in a density functionaltheory simulation15. We note that for sp3 defects, there exists no suchtransition and any chirality CNT can form sp3 defects25. We thereforeidentify IXs as interface excitons.The IX peaks in the PL spectra from various heterostructures aresummarized in Fig. 2e by plotting the number of the peaks(N) observed during time-trace measurements as a function of E11(see Supplementary Table 1 for the list of samples). Two distinctregions can be seen below and above 0.94 eV, corresponding to type-Iand type-II alignment, respectively (see Supplementary Fig. 8). IXpeaks are absent for type-I band alignment, whereas numerous peaksappear for type-II alignment. Exciton transfer observed in similar het-erostructures show anticorrelation with the appearance of the IXpeaks, consistent with the band alignment transition (SupplementaryFig. 9)15.The dependence of IXs on the number of WSe2 layers (Fig. 2c) ismore subtle, since the number of layers does not significantly mod-ulate the bandgap in comparison to CNT chirality (SupplementaryFig. 10). For example, (9,4) CNT/WSe2 heterostructures are consistentwith complete type-II band alignment irrespective of the WSe2 layernumber (Figs. 1f and 2b). We therefore study (10,5) CNT/WSe2 het-erostructures located at theband alignment transition15, as they shouldbe sensitive to the small changes inWSe2 bandgap. The PL spectrumofthe 1L WSe2 heterostructure is presented in Fig. 2d, which does notshow any observable IX peaks. In contrast, the trilayer (3L) WSe2 het-erostructure reveals a discernible IX peak in between the E11 excitonand the trion (T) peaks. Two IX peaks appear for the 4L WSe2 hetero-structure, with increased PL intensity for the higher energy IXpeak andan additional lower energy IX peak besides the trion peak. This layer-number dependent behavior of the IX peaks can be explained by theband alignment transition, as shown in Fig. 2c.Optical properties of the interface excitonsThe interface excitons display several distinct features different fromthe E11 excitons. In Fig. 3a, we first examine the emission polarizationdependence of IX1 and IX2 for the (9,4) CNT/1L WSe2 sample used inFig. 1. PL from IXs exhibits near ideal linear polarization of > 95%,consistent with confinement in the 1D channel11,28. We also note thatthe polarization angle of IX1 and IX2 deviates from that of E11 polar-ization by 13.7°, potentially suggesting some distortion of the opticaldipolemoment in the IXs (see Supplementary Fig. 11). Owing to the useof a conventional normal-incidence photoluminescence setup whichpredominantly detects in-plane dipoles, the vertical componentremains unresolved in our experiments.Time-resolved PL is then performed for E11 and IX2 excitons, as thelifetime of IXs is expected to be long due to the spatially indirectnature3,7. PL decay curves corresponding to E11 and IX2 aremeasured asshown in Fig. 3b, and the lifetime is extracted by reconvolution fittingusing the instrument response function (IRF). Two decay componentsare obtained for the E11 PL decay curve, as is the case for suspendedCNTs: A main fast component with a lifetime of 59 ps associated withthe bright states, and a small slow component with a lifetime of 646 psassociated with the dark states12. In contrast, only one decay compo-nent is observed for IX2 with a long lifetime of 673 ps, consistent withthe reduced optical dipole moment.The interface excitons exhibit considerably bright emission at lowexcitation powers as shown in Fig. 3c. At a low power of 0.04μW, bothIX1 and IX2 display bright emission with the IX1 PL even exceeding theE11 PL. Such high intensity emission from interface excitons is unex-pected at room temperature. Generally, indirect excitons exhibit weakPL emission due to diminished dipole coupling, which is the case for2D-2D heterostructures where interlayer excitons can hardly beobserved at room temperature3,4,6. The evident PL from interfaceexcitons in our system can be ascribed to two primary factors. Firstly,we employ a fully suspended structure, which reduces the substrate-induced screening effect and helps sustain the dipole strength. Sec-ondly, the wavefunction of π-orbitals in CNTs, which extend sig-nificantly out of the tube, could reduce the spatially indirect nature ofthe interface excitons.Emission intensity of the interface excitons exhibit intriguingpower dependence as shown in Fig. 3d. Both IX1 and IX2 PL nearlysaturate with a low threshold power of approximately 0.6μW, whilethe E11 PL increases substantially. The saturation observed is muchmore pronounced compared to interlayer excitons in 2D-2DArticle https://doi.org/10.1038/s41467-024-47099-6Nature Communications |         (2024) 15:2871 3heterostructures3. It is suspected that the IXs are further confined to alower dimension, that is, 0D. Interface excitons in CNTs may be morereadily localized than interlayer excitons in 2D-2D heterostructuresbecause of the lower dimensionality. In general, localized states showmuch stronger saturation behavior than free states because of thestate-filling effects29,30. In other samples where interface excitons alsoemerge,wefind that the saturation behavior dependson their energies(see Supplementary Fig. 12). The IX peaks with energies substantiallylower than the E11 energy display stronger saturation behaviors. Thiscould be explained by the deeper trap potential in the confinement ofFig. 2 | Band alignment effect on IXs. a Band diagram of heterostructures con-structed fromCNTswith various chiralities. bChirality-dependent PL spectra of 1D-2D heterostructures. We define a heterostructure nomenclature where (9,4) CNT/2L WSe2 is represented by (9,4)/2L, with other samples following the same con-vention. Excitation energy is adjusted to CNT E22 state for each heterostructure.Excitation power values are 4, 6, 5, and 5 μW for (9,4)/2L, (12,1)/2L, (8,7)/2L, and(14,3)/3L heterostructures, respectively, and 10 μW for other samples. Arrowsindicate IX peaks. Samples exhibiting IX peaks may also host other unstable IXpeaks on a longer timescale, which is not indicated here. The peaks fromK-momentum states are carefully checked and excluded in the analysis of IXs.c Band diagram of heterostructures constructed from WSe2 with different layernumbers. d PL spectra from (10,5)/1L, (10,5)/3L, and (10,5)/4L samples. The exci-tation energy is adjusted to 1.653 eV and the excitation power is 5μW for the (10,5)/4L sample and 10 μW for others. Red arrows indicate IX peaks while the gray arrowdenotes trionpeaks T. eTheplot of the number of the IXpeaksN as a function of E11energy for all samples. It should be noted thatN encompasses all IX peaks observedduringmeasurements, and someof them are not illustrated inb. More spectra of IXpeaks are shown in Supplementary Figs. 3, 7, and 12. The energies of all IX peaks canbe found in Supplementary Fig. 8.Article https://doi.org/10.1038/s41467-024-47099-6Nature Communications |         (2024) 15:2871 4the interface excitons. The localization is also supported by theobserved blinking noise from the unstable IXs that host only “on” and“off” states (see Supplementary Figs. 3 and 13). The pure two-levelnoise is exclusively observed in quasi-0D systems, such as quantumdots and single molecules31,32, suggesting that the interface excitonstate could function as a single-photon emitter.Room-temperature quantum emission from interface excitonsTo elucidate the quantum nature of interface excitons, we conductphoton correlation measurements on a (9,4) CNT/2L WSe2 sampleunder continuous-wave laser excitation. Two primary peaks areobserved in this sample, originating from the stable low-energy IX andE11 (see Supplementary Figs. 3 and 14). The second-order correlationg(2)(τ) from the IX is shown in Fig. 4a, after background correction. Adistinct antibunching dip is observed at zero delay with g(2)(0) = 0.467.Over a longer timescale,we alsoobserve abunchingpeak (Fig. 4b).The bunching behavior is often observed in single-photon sources andis associated with the dynamics between excited states and other dark,charged, or meta-stable states33,34. We employ the equationgð2ÞðτÞ= ½1� α expð�jτj=τAÞ�*½1 +β expð�jτj=τBÞ� to fit the observedg(2)(τ) statistics, where factors α and β quantify the degree of anti-bunching and bunching with values of 0.720 and 0.666, respectively35.τA and τB indicate the timescales of the antibunching dip and thebunching peak, with values of 0.148 ns and 3.752 ns from the fitting,respectively. The value of 1 − α is 0.280, revealing high single-photonpurity by considering the effect of bunching behavior. It is noteworthythat most of the low-energy interface excitons display similar peaklinewidths and power saturation behavior (see Supplementary Fig. 12),implying that each of them acts as a quantum emitter. This is furthersupported by the high reproducibility of the antibunching behavior inother samples (see Supplementary Figs. 15 and 16). It is noted that anelectron in CNTwould sit at the Γpoint in theCNTBrillouin zone, and ahole will be at the K point and the Γ point in monolayer and 2-4LWSe2,respectively15. The corresponding interface excitons therefore couldbe momentum indirect. However, the strong localization observedwould relax the momentum selection rule, potentially explaining thebright PL emission from the interface excitons.For comparison, g(2)(τ) from E11 excitons does not exhibit anyantibunching or bunching behavior, as shown in Fig. 4c. Under pulsedlaser excitation, the E11 excitons are known to go through an efficientexciton-exciton annihilation (EEA) process that could result in SPE36.The exciton density is generally lower with continuous-wave excita-tion, hindering the SPE through EEA. The confinement effect is crucialfor SPE, and the absence of any antibunching behavior thereforeindicates the 1D free feature of E11 excitons.We now discuss the possible origins of the localized interfaceexciton states, which would require a potential depth exceeding a fewmultiples of the thermal energy. The confinement can be provided byFig. 3 | Optical properties of low-energy interface excitons. a Emission polar-ization dependence of PL emission from E11 (blue circles), IX1 (green dots), and IX2(red circles). PL emission is plotted as a function of anglewith respect to the trench.The lines are fits to a cosine squared function. The degree of polarization is esti-mated by ðImax � IminÞ=ðImax + IminÞ, where Iax (Imin) is the maximum (minimum) PLemission intensity. The excitation energy is adjusted to E22 of 1.653 eVwith a powerof 5μW. b PL decay curves taken from the CNT/WSe2 heterostructures for E11 (bluedots) and IX2 (red dots). Shortpass (1.033 eV) and longpass filters (0.919 eV) areused for themeasurements of E11 and IX2, respectively. Thepulsed laser is used herewith the power adjusted to 2 nW, and the excitation energy is 1.653 eV. The graycurves are the instrument response function (IRF), and the black curves are theexponential fitting curves convoluted with the IRF. c PL spectra at powers of 0.04,0.4, and 4μW from bottom to top. The excitation energy is 1.680 eV. d IntegratedPL intensity as a function of the laser power for E11 (blue), IX1 (green), and IX2 (red),respectively.Fig. 4 | Room-temperature quantum emission from low-energy interfaceexcitons. a–c Second-order correlation statistics of the IX peak during a short andb long time scales, and of c the E11 peak for the (9,4) CNT/2L WSe2 sample. Theexcitation energy is adjusted to E22 of 1.653 eV with a power of 2μW. Shortpass(1.033 eV) and longpass filters (0.886 eV) are used for the measurements of E11 andIX, respectively. The gray lines are experimental results, and the red lines are thefittings. See Supplementary Note 14 for the background correction procedure usedin a, b. For the second-correlation statistics data inb, c, four data points are binnedtogether to reduce the noise. See SupplementaryNote 15 and 16 for additional data.Article https://doi.org/10.1038/s41467-024-47099-6Nature Communications |         (2024) 15:2871 5defect states, for example in materials such as diamond37, siliconcarbide38, hexagonal boron nitride34, and CNTs39. It is unlikely thatdefects in CNTs play a role, since we use pristine suspended CNTscontaining negligible exciton quenching sites11,36. We note that theyhave been characterized under low-power conditions within drynitrogen gas environment, precluding the formation of defects in theCNTs11. In comparison, WSe2 flakes inherently encompass a range ofdefect states, spanning from single vacancies to complex vacancyclusters40,41. Among them, it is known that single tungsten vacanciesinduce defects states close to the valance band41, which could beresponsible for the confinement. The defect-bound interface excitonsmay be formed, similar to the case of interlayer excitons in 2D-2Dheterostructures42. The variability of the emission energy is consistentwith the picture in which defects play a role, as the location and thegeometry of the defect with respect to the CNT can influence theexcitonic states.As another possible explanation, inhomogeneous strain couldalso contribute to the localizationof interface excitons as in the caseofmonolayer WSe243. The morphology of the heterostructure is exam-ined with an atomic force microscope (Supplementary Fig. 17), and aclean CNT/WSe2 interface is confirmed which would facilitate theformation of interface excitons. In one heterostructure, we observe ashallow local dip with a depth of ~ 3 nm and a width of ~350nm, whichmay confine interface excitons nearby. Such nanoscale strain mightalso impact the sample through various mechanisms from van derWaals gap fluctuation44 to lattice reconstruction45, in a manner similarto 2D-2D heterostructures. While the exact impact of strain remainsunclear, the spatially indirect nature renders interface excitons moresusceptible to the aforementioned effects than intralayer excitons.In conclusion, we observe numerous IX peaks at the CNT/WSe2interface below the CNT E11 energy, spanning the telecommunicationwavelengths. By systematically varying the chirality of CNTs and thelayer number of WSe2, we are able to assign the peaks to interfaceexcitons as they only appear for type-II band alignment. The lowsaturationpower and the long lifetime indicate that low-energy interfaceexcitons are strongly confined, and photon antibunching is confirmed.The observation of interface excitons as room-temperature quantumemitters at telecommunication bands opens up new opportunities forapplications in quantum technologies and optoelectronics, under-scoring the emerging potential of mixed-dimensional heterostructures.MethodsAir-suspended carbon nanotubesWe prepare air-suspended CNTs using trenched SiO2/Si substrates11.First, we pattern alignment markers and trenches with lengths of900μm and widths ranging from 0.5 to 3.0μm onto the Si substratesusing electron-beam lithography, followed by dry etching. We thenthermally oxidize the substrate to form a SiO2 film, with a thicknessranging from60 to 70nm.Another electron-beam lithographyprocessis used to define catalyst regions along the edges of the trenches. A1.5Å thick iron (Fe) film is deposited as a catalyst for CNT growth usingan electron beam evaporator. CNTs are synthesized by alcohol che-mical vapor deposition at 800 °C for 1min. The Fe film thickness isoptimized to control the yield for preparing isolated CNTs. The PLimages and PL polarization measurements are performed to excludethe existence of any quenching sites in the CNTs, and only the tubesshowing a smooth profile along the length and a high polarizationdegree > 90% are selected and used for the preparation of theheterostructures11,12,36. We select isolated, fully suspended chirality-identified CNTs with lengths ranging from 0.5 to 2.0μm to form theheterostructures with WSe2.Anthracene crystal growthFor transferring WSe2 flakes onto CNTs, we grow anthracene crystalsthrough an in-air sublimationprocess17,18. Anthracenepowder is heatedto 80 °Con a glass slide, while another glass slide is placed 1mmabovethe anthracene source. Thin and large-area single crystals are thengrown on the glass surface. To promote the growth of large-area singlecrystals, we pattern the glass slides using ink from commercial mar-kers. The typical growth time for anthracene crystals is 10 h.Transfer of WSe2 by anthracene crystalsFirst, WSe2 (HQ graphene) flakes are prepared on 90-nm-thick SiO2/Sisubstrates using mechanical exfoliation, and the layer number isdetermined by optical contrast. An anthracene single crystal is pickedup with a glass-supported PDMS sheet to form an anthracene/PDMSstamp.Next, theWSe2 flakes arepickedupby pressing the anthracene/PDMS stamp against a substratewith the targetWSe2 flakes. The stampis quickly separated ( > 10mm/s) to ensure that the anthracene crystalremains attached to the PDMS sheet. The stamp is then applied to thereceiving substrate with the desired chirality-identified CNT, whoseposition is determined by a prior measurement. Precise positionalignment is accomplished with the aid of markers prepared on thesubstrate. By slowly peeling off the PDMS ( < 0.2μm/s), the anthracenecrystal with the WSe2 flake is released on the receiving substrate.Sublimation of anthracene in air at 110 °C for 10min removes theanthracene crystal, leaving behind a clean suspended CNT/WSe2 het-erostructure. This all-dry process eliminates contamination from sol-vents, and the solid single-crystal anthracene protects the 2D flakesand the CNT during the transfer, ensuring that the CNT/WSe2 het-erostructure experiences minimal strain17,18.PL measurementsA homebuilt confocal microscopy system is employed to perform PLmeasurements for interface excitons and E11 excitons at room tem-perature in dry nitrogen gas11,14. We utilize a wavelength-tunable con-tinuous-wave Ti:sapphire laser for excitation, with its power controlledby neutral density filters. The excitation polarization angle is adjustedto be parallel to the CNT axis and the emission polarization angledependence is measured using a half-wave plate followed by a polar-izer placed in front of a spectrometer. The laser beam is focused on thesamples with an objective lens that has a numerical aperture of 0.65and a working distance of 4.5mm. The 1/e2 spot diameter and thecollection spot size defined by a confocal pinhole are approximately1.2 and 5.4μm, respectively. PL is collected through the same objectivelens and detected using a liquid-nitrogen-cooled 1024-pixel InGaAsdiode array connected to the spectrometer. A 150-lines/mm grating isused to obtain a dispersion of 0.52 nm/pixel at a wavelength of1340 nm. For photoluminescencemeasurements ofWSe2 A excitons, a532-nm laser and a charge-coupled device camera are employed.Time-resolved and photon correlation measurementsApproximately 100 femtosecond pulses at a repetition rate of 76MHzfrom a Ti:sapphire laser is utilized for time-resolved measurements.The excitation laser beam is focused onto the sample using an objec-tive lens with a numerical aperture of 0.85 and a working distance of1.48mm. The PL from the center of the nanotube within the hetero-structure is coupled to a superconducting single-photon detector withan optical fiber, and a time-correlated single-photon counting moduleis used to collect the data. IRFs dependent on the detection wave-length are acquired by dispersing supercontinuum white light pulseswith a spectrometer. Photon correlationmeasurements are carried outusing a Hanbury-Brown-Twiss setup with a 50:50 fiber coupler underexcitation with a continuous-wave laser. The experiments are con-ducted at room temperature.Data availabilityAll the data generated in this study have been deposited into theR2DMS-GakuNinRDM database, and are accessible at https://dmsgrdm.riken.jp/42tsf/.Article https://doi.org/10.1038/s41467-024-47099-6Nature Communications |         (2024) 15:2871 6https://dmsgrdm.riken.jp/42tsf/https://dmsgrdm.riken.jp/42tsf/References1. Cao, Y. et al. Correlated insulator behaviour at half-filling in magic-angle graphene superlattices. Nature 556, 80 (2018).2. Cao, Y. et al. Unconventional superconductivity in magic-anglegraphene superlattices. Nature 556, 43 (2018).3. Rivera, P. et al. Observation of long-lived interlayer excitons inmonolayer MoSe2-WSe2 heterostructures. Nat. Commun. 6,6242 (2015).4. Xiong, R. et al. Correlated insulator of excitons in MoSe2-WSe2moiré superlattices. Science 380, 860 (2023).5. Sun, X. et al. Enhanced interactions of interlayer excitons in free-standing heterobilayers. Nature 610, 478 (2022).6. Ubrig, N. et al. Design of van der Waals interfaces for broad-spectrum optoelectronics. Nat. Mater. 19, 299 (2020).7. Unuchek, D. et al. Room-temperature electrical control of excitonflux in a van der Waals heterostructure. Nature 560, 340 (2018).8. Rivera, P. et al. Valley-polarized exciton dynamics in a 2D semi-conductor heterostructure. Science 351, 688 (2016).9. Jin, C. et al. Observation of moiré excitons in WSe2/WS2 hetero-structure superlattices. Nature 567, 76 (2019).10. Seyler, K. L. et al. Signatures of moiré-trapped valley excitons inMoSe2/WSe2 heterobilayers. Nature 567, 66 (2019).11. Ishii, A., Yoshida, M. & Kato, Y. K. Exciton diffusion, end quenching,and exciton-exciton annihilation in individual air-suspended carbonnanotubes. Phys. Rev. B 91, 125427 (2015).12. Ishii, A., Yoshida, M. & Kato, Y. K. High efficiency dark-to-bright exci-ton conversion in carbon nanotubes. Phys. Rev. X 9, 041048 (2019).13. Jariwala, D., Marks, T. J. & Hersam, M. C. Mixed-dimensional van derWaals heterostructures. Nat. Mater. 16, 170 (2016).14. Fang, N. et al. Hexagonal boron nitride as an ideal substrate forcarbon nanotube photonics. ACS Photonics 7, 1773 (2020).15. Fang, N. et al. Resonant exciton transfer in mixed-dimensionalheterostructures for overcoming dimensional restrictions in opticalprocesses. Nat. Commun. 14, 8152 (2023).16. Lefebvre, J., Homma, Y. & Finnie, P. Bright band gap photo-luminescence from unprocessed single-walled carbon nanotubes.Phys. Rev. Lett. 90, 217401 (2003).17. Otsuka, K. et al. Deterministic transfer of optical-quality carbonnanotubes for atomically defined technology. Nat. Commun. 12,3138 (2021).18. Fang,N. et al.Quantizationofmodeshifts in nanocavities integratedwith atomically thin sheets. Adv. Opt. Mater. 10, 2200538 (2022).19. Jiang, M. et al. Gate-controlled generation of optical pulse trainsusing individual carbon nanotubes. Nat. Commun. 6, 6335 (2015).20. Uda, T., Yoshida, M., Ishii, A. & Kato, Y. K. Electric-field inducedactivation of dark excitonic states in carbon nanotubes. Nano Lett.16, 2278 (2016).21. Otsuka, K., Ishii, A. & Kato, Y. K. Super-resolution fluorescenceimaging of carbon nanotubes using a nonlinear excitonic process.Opt. Express 27, 17463 (2019).22. Piao, Y. et al. Brightening of carbon nanotube photoluminescencethrough the incorporation of sp3 defects. Nat. Chem. 5, 840 (2013).23. He, X. et al. Tunable room-temperature single-photon emission attelecom wavelengths from sp3 defects in carbon nanotubes. Nat.Photon. 11, 577 (2017).24. Matsunaga, R., Matsuda, K. & Kanemitsu, Y. Origin of low-energyphotoluminescence peaks in single carbon nanotubes: K-momen-tum dark excitons and triplet dark excitons. Phys. Rev. B 81,033401 (2010).25. Kozawa, D. et al. Formation of organic color centers in air-suspended carbon nanotubes using vapor-phase reaction. Nat.Commun. 13, 2814 (2022).26. Yu, B. et al. ortho-substituted aryldiazonium design for the defectconfiguration-controlled photoluminescent functionalization ofchiral single-walled carbon nanotubes. ACS Nano 16, 21452 (2022).27. Tan, Q. et al. Layer-engineered interlayer excitons. Sci. Adv. 7,1 (2021).28. Bai, Y. et al. Excitons in strain-induced one-dimensional moirépotentials at transition metal dichalcogenide heterojunctions. Nat.Mater. 19, 1068 (2020).29. Settele, S. et al. Synthetic control over the binding configuration ofluminescent sp3-defects in single-walled carbon nanotubes. Nat.Commun. 12, 2119 (2021).30. Shinokita, K., Miyauchi, Y., Watanabe, K., Taniguchi, T. &Matsuda, K.Resonant coupling of amoiré exciton to a phonon in aWSe2/MoSe2heterobilayer. Nano Lett. 21, 5938 (2021).31. Nirmal, M. et al. Fluorescence intermittency in single cadmiumselenide nanocrystals. Nature 383, 802 (1996).32. Moerner, W. E. & Orrit, M. Illuminating single molecules in con-densed matter. Science 283, 1670 (1999).33. Sallen, G. et al. Exciton dynamics of a single quantum dotembedded in a nanowire. Phys. Rev. B 80, 085310 (2009).34. Tran, T. T., Bray, K., Ford, M. J., Toth, M. & Aharonovich, I. Quantumemission from hexagonal boron nitride monolayers. Nat. Nano-technol. 11, 37 (2016).35. Gao, K., Solovev, I., Holmes, M., Arita, M. & Arakawa, Y.Nanosecond-scale spectral diffusion in the single photon emissionof a gan quantum dot. AIP Adv. 7, 125216 (2017).36. Ishii, A., Yoshida, M. & Kato, Y. K. Room-temperature single-photonemission from micrometer-long air-suspended carbon nanotubes.Phys. Rev. Appl. 8, 054039 (2017).37. Babinec, T.M. et al. A diamond nanowire single-photon source.Nat.Nanotechnol. 5, 195 (2010).38. Castelletto, S. et al. A silicon carbide room-temperature single-photon source. Nat. Mater. 13, 151 (2014).39. Ma, X., Hartmann, N. F., Baldwin, J. K. S., Doorn, S. K. & Htoon, H.Room-temperature single-photon generation from solitary dopantsof carbon nanotubes. Nat. Nanotechnol. 10, 671 (2015).40. Linhart, L. et al. Localized intervalley defect excitons as single-photon emitters in WSe2. Phys. Rev. Lett. 123, 146401 (2019).41. Zhang, S. et al. Defect structure of localized excitons in a WSe2monolayer. Phys. Rev. Lett. 119, 046101 (2017).42. Zhao, H. et al. Manipulating interlayer excitons for near-infraredquantum light generation. Nano Lett. 23, 11006 (2023).43. Parto, K., Azzam, S. I., Banerjee, K. & Moody, G. Defect and strainengineering of monolayer WSe2 enables site-controlled single-photon emission up to 150 K. Nat. Commun. 12, 3585 (2021).44. Shin, B. G. et al. Indirect bandgap puddles in monolayer MoS2 bysubstrate-induced local strain. Adv. Mater. 28, 9378 (2016).45. Zhao, S. et al. Excitons in mesoscopically reconstructed moiréheterostructures. Nat. Nanotechnol. 18, 572 (2023).AcknowledgementsParts of this study are supported by JSPS (KAKENHI JP22K14624 to D.Y.,JP22K14625 to S.F., JP21K14484 to M.M., JP22K14623 to C.F.F.,JP22H01893 to D.K., JP21H05233 to S.O., JP23H00262, JP22F22350,JP20H02558 to Y.K.K.) and MEXT (ARIM JPMXP1222UT1135). Y.R.C. issupported by JSPS (International Research Fellow). N.F. and C.F.F. aresupported by RIKEN Special Postdoctoral Researcher Program. Wethank the AdvancedManufacturing Support Team at RIKEN for technicalassistance.Author contributionsN.F. carried out sample preparation and performed measurements onthe samples. Y.R.C. performed atomic forcemicroscopemeasurementsand assisted in sample preparation. Y.R.C., C.F.F., and K.N. assisted inoptical measurements. D.Y., S.F., and D.K. contributed to the time-resolved PL and photon correlationmeasurements. M.M., Y.G., and S.O.performed density functional theory calculations. K.O. aided in thedevelopment of the anthracene-assisted dry transfer method. Y.K.K.Article https://doi.org/10.1038/s41467-024-47099-6Nature Communications |         (2024) 15:2871 7supervised the project. N.F. and Y.K.K. co-wrote the manuscript, with allauthors providing input and comments on the manuscript.Competing interestsThe authors declare no competing interests.Additional informationSupplementary information The online version containssupplementary material available athttps://doi.org/10.1038/s41467-024-47099-6.Correspondence and requests for materials should be addressed toN. Fang or Y. K. Kato.Peer review information Nature Communications thanks the anon-ymous reviewers for their contribution to the peer review of this work. Apeer review file is available.Reprints and permissions information is available athttp://www.nature.com/reprintsPublisher’s note Springer Nature remains neutral with regard to jur-isdictional claims in published maps and institutional affiliations.Open Access This article is licensed under a Creative CommonsAttribution 4.0 International License, which permits use, sharing,adaptation, distribution and reproduction in any medium or format, aslong as you give appropriate credit to the original author(s) and thesource, provide a link to the Creative Commons licence, and indicate ifchanges were made. The images or other third party material in thisarticle are included in the article’s Creative Commons licence, unlessindicated otherwise in a credit line to the material. If material is notincluded in the article’s Creative Commons licence and your intendeduse is not permitted by statutory regulation or exceeds the permitteduse, you will need to obtain permission directly from the copyrightholder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.© The Author(s) 2024Article https://doi.org/10.1038/s41467-024-47099-6Nature Communications |         (2024) 15:2871 8https://doi.org/10.1038/s41467-024-47099-6http://www.nature.com/reprintshttp://creativecommons.org/licenses/by/4.0/http://creativecommons.org/licenses/by/4.0/ Room-temperature quantum emission from interface excitons in mixed-dimensional heterostructures Results and discussion Emerging peaks in 1D-2D mixed-dimensional heterostructures Band alignment effect on IX�peaks Optical properties of the interface excitons Room-temperature quantum emission from interface excitons Methods Air-suspended carbon nanotubes Anthracene crystal�growth Transfer of WSe2 by anthracene crystals PL measurements Time-resolved and photon correlation measurements Data availability References Acknowledgements Author contributions Competing interests Additional information