# Fileset

[Mahdikhanysarvejahany_et_al-2022-Nature_Communications.pdf](https://mdr.nims.go.jp/filesets/91f0983a-6844-48af-8ace-b724715cadb1/download)

## Creator

Fateme Mahdikhanysarvejahany, Daniel N. Shanks, Matthew Klein, Qian Wang, Michael R. Koehler, David G. Mandrus, [Takashi Taniguchi](https://orcid.org/0000-0002-1467-3105), [Kenji Watanabe](https://orcid.org/0000-0003-3701-8119), Oliver L. A. Monti, Brian J. LeRoy, John R. Schaibley

## Rights

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

## Other metadata

[Localized interlayer excitons in MoSe2–WSe2 heterostructures without a moiré potential](https://mdr.nims.go.jp/datasets/15fbb822-a52c-4648-9757-6f57df03925f)

## Fulltext

Localized interlayer excitons in MoSe2–WSe2 heterostructures without a moiré potentialArticle https://doi.org/10.1038/s41467-022-33082-6Localized interlayer excitons in MoSe2–WSe2heterostructures without a moiré potentialFateme Mahdikhanysarvejahany 1, Daniel N. Shanks 1, Matthew Klein1,Qian Wang2, Michael R. Koehler3, David G. Mandrus4,5,6, Takashi Taniguchi 7,Kenji Watanabe 8, Oliver L. A. Monti1,9, Brian J. LeRoy1 & John R. Schaibley 1Interlayer excitons (IXs) in MoSe2–WSe2 heterobilayers have generated inter-est as highly tunable light emitters in transition metal dichalcogenide (TMD)heterostructures. Previous reports of spectrally narrow (<1meV) photo-luminescence (PL) emission lines at low temperature have been attributed toIXs localized by the moiré potential between the TMD layers. We show thatspectrally narrow IX PL lines are present even when the moiré potential issuppressed by inserting a bilayer hexagonal boron nitride (hBN) spacerbetween the TMD layers.We compare the doping, electricfield,magneticfield,and temperature dependence of IXs in a directly contacted MoSe2–WSe2region to those in a region separated by bilayer hBN. The doping, electric field,and temperature dependence of the narrow IX lines are similar for bothregions, but their excitonic g-factors have opposite signs, indicating that theorigin of narrow IX PL is not the moiré potential.Localized excitons (Coulomb-bound electron–hole pairs) which canserve as single photon emitters have been investigated for decadesdue to their potential applications in quantum information science andoptoelectronics1–9. Recently there hasbeen significant interest inmoiréeffects in 2D material heterostructures that arise from the in-planesuperlattice potential that occurs between two twisted or lattice mis-matched layers7,8,10. IXs are spatially indirect excitons comprised of anelectron in the MoSe2 layer bound to a hole in the WSe2 layer2,9. Seyleret al. were the first to report spectrally narrow PL arising from IXs inMoSe2–WSe2 heterostructures. In contrast, Kha et al. observed spec-trallywide (~20meV) PL peaks with alternating polarization, attributedto excited states ofmoiré IX in theMoSe2–WSe2heterostructure7,8. Thenarrow PL lines were observable only at low (<15 K) temperatures andlow (<100 nW) excitation power and consisted of an inhomogeneousdistribution of narrow lines with a spread on the order of 20meV8. ThenarrowPL lineswere attributed to single IXs trapped tomoiré potentialsites, as evidenced by the spectrally narrow (<1meV) PL emission andcircularly polarized optical selection rules7,11. There have beennumerous papers studying these narrow lines since the initial report,including reports of single photon emission, charged IXs, and Cou-lomb staircase effects12–14. In this work, we show that the spectrallynarrow IX PL lines are still present in MoSe2-hBN-WSe2 hetero-structures, where the moiré potential is suppressed by an hBN spacerlayer (see Supplementary Figs. 1 and 2). We compare the physicalbehaviors of narrow IX PL emission from non-separated, directlycontacted (DC) and hBN-separated sample regions and show that thelocalization potential resulting in the narrow IX lines is likely due to anextrinsic disorder effect, not the moiré potential.ResultsThe sample structure is comprised of a twist-angle alignedMoSe2–WSe2 heterostructure as depicted in Fig. 1a (see “Methods”). InReceived: 7 May 2022Accepted: 1 September 2022Check for updates1Department of Physics, University of Arizona, Tucson, AZ85721, USA. 2Guangdong Provincial Key Laboratory of QuantumMetrology and Sensing& School ofPhysics and Astronomy, Sun Yat-Sen University (Zhuhai Campus), Zhuhai 519082, China. 3IAMM Diffraction Facility, Institute for Advanced Materials andManufacturing, University of Tennessee, Knoxville, TN 37920, USA. 4Department of Materials Science and Engineering, University of Tennessee, Knoxville, TN37996, USA. 5Materials Science and Technology Division, Oak RidgeNational Laboratory, Oak Ridge, TN 37831, USA. 6Department of Physics and Astronomy,University of Tennessee, Knoxville, TN 37996, USA. 7International Center for Materials Nanoarchitectonics, National Institute for Materials Science, 1-1 Namiki,Tsukuba 305-0044, Japan. 8Research Center for Functional Materials, National Institute for Materials Science, 1-1 Namiki, Tsukuba 305-0044, Japan.9Department of Chemistry and Biochemistry, University of Arizona, Tucson, AZ 85721, USA. e-mail: johnschaibley@email.arizona.eduNature Communications |         (2022) 13:5354 11234567890():,;1234567890():,;http://orcid.org/0000-0002-5163-7293http://orcid.org/0000-0002-5163-7293http://orcid.org/0000-0002-5163-7293http://orcid.org/0000-0002-5163-7293http://orcid.org/0000-0002-5163-7293http://orcid.org/0000-0002-6329-0338http://orcid.org/0000-0002-6329-0338http://orcid.org/0000-0002-6329-0338http://orcid.org/0000-0002-6329-0338http://orcid.org/0000-0002-6329-0338http://orcid.org/0000-0002-1467-3105http://orcid.org/0000-0002-1467-3105http://orcid.org/0000-0002-1467-3105http://orcid.org/0000-0002-1467-3105http://orcid.org/0000-0002-1467-3105http://orcid.org/0000-0003-3701-8119http://orcid.org/0000-0003-3701-8119http://orcid.org/0000-0003-3701-8119http://orcid.org/0000-0003-3701-8119http://orcid.org/0000-0003-3701-8119http://orcid.org/0000-0002-8024-9193http://orcid.org/0000-0002-8024-9193http://orcid.org/0000-0002-8024-9193http://orcid.org/0000-0002-8024-9193http://orcid.org/0000-0002-8024-9193http://crossmark.crossref.org/dialog/?doi=10.1038/s41467-022-33082-6&domain=pdfhttp://crossmark.crossref.org/dialog/?doi=10.1038/s41467-022-33082-6&domain=pdfhttp://crossmark.crossref.org/dialog/?doi=10.1038/s41467-022-33082-6&domain=pdfhttp://crossmark.crossref.org/dialog/?doi=10.1038/s41467-022-33082-6&domain=pdfmailto:johnschaibley@email.arizona.eduorder to probe the dependence of the IX trapping on the moirépotential, we fabricated a device where half of the heterostructure hasthe TMD layers separated by bilayer hBN and the other half has thelayers in direct contact (with no hBN spacer). The TMD layers areencapsulated in hBN, and a few layers graphene top gate (Vt) andgraphite back gate (Vb) serves to independently control the chargedensity and external electric field experienced by the TMD layers. Weused low temperature confocal PL spectroscopy tomeasure the spatialdependence of the IX emission photon energy (Fig. 1b). In the DCregion of the device the IX photon energy is centered around 1.34 eV,consistent with R-type (near 0° twist) MoSe2–WSe2 heterostructures15.The hBN separated region shows higher energy PL centered around1.42 eV. This 80meV increase in the energy of the PL is in agreementwith the suppression of the moiré potential by the insertion of bilayerhBN (see Supplementary Fig. 1). When the confocal pinhole wasremoved, PL from both regions could be detected when exciting onthehigher energyhBN-separated region. Figure 1c showsco- and cross-circularly polarized PL when exciting the hBN-separated region with a1.72 eV laser and detecting from both regions. Surprisingly, while theDC region shows the cross-circularly polarized PL, consistent withprevious studies on R-type structures5,7,15, the hBN-separated regionshows mostly co-circularly polarized PL (see Supplementary Fig. 3 forsimilar data from another sample).To understand the opposite selection rules in the DC and hBN-separated regions, we must consider both the effect of the moirépotential as well as the stacking dependent IX oscillator strength.The oscillator strength was calculated as a function of interlayerstacking and separation (see Supplementary Data Fig. 4). We findthat that the oscillator strength is actually dominated by recombi-nation at the Rhh site which results in primarily observing co-circularly polarized PL. However, in the DC region, the moirépotential traps IXs at the RXh stacking site (which has the oppositeselection rule) emitting cross-circularly polarized PL. Therefore, it isthe trapping of IXs at RXh sites that results in the cross-circularlypolarized PL from the DC contact region, whereas the larger oscil-lator strength of co-polarized emission results in co-circularlypolarized PL in the hBN-separated regions. Note that this does notmean that the IX is localized at the Rhh site in the hBN-separatedregion, rather that since this local stacking possesses the highestoscillator strength, it dominates the PL spectrum. We also empha-size that both the DC and hBN-separated regions show a distribu-tion of spectrally narrow (< 1meV) IX lines indicating that they donot solely originate from the moiré potential.In order to confirm that the narrow PL emission in the hBNseparated region originates from IXs, we investigated the electric fieldanddopingdependencewhichwas achievedby applying simultaneoustop and back gate voltages. In these measurements, we again excitedthe hBN separated region and measured the PL from both regions.Figure 2a shows the electric field dependence of the PL while keepingthe sample charge neutral16. In the DC region, we observe a Stark shiftof 0.6 eV/(V/nm) that matches with the known dipole moment of theIX16,17. By adding the hBN bilayer between the TMD layers, we increasethe separation between the electron and hole and consequently the IXpermanent dipole moment will increase by a factor of two. Perfectlymatching with our finding yielding a 1.2 eV/(V/nm) energy shift, con-sistent with previous reports on hBN separated IXs18. We also exploredthe IX doping dependence for both sample regions (Fig. 2b). Weidentify three doping regions corresponding to i-electron doped, ii-intrinsic, and iii-hole doped similar to previous reports of chargedIXs12,13,16,19. The relatively small intrinsic doping range (−0.1 to0.1 × 1012 cm−2) is consistent with a high quality device. We note that inboth regions, the IX energy increases with doping, which is consistentwith previous reports13,16. We also note that the DC region showsmoreprominent fine structure in its doping dependence which was pre-viously attributed to a Coulomb staircase effect13 (see Supplemen-tary Fig. 5).In order to probe the spin-valley physics of the IXs, we measuredσ� circularly polarized PL as a function of out-of-plane magnetic field(Fig. 3a). In theDC region, wemeasured an exciton g-factor of 7:0±0:4(Fig. 3b), consistent with numerous past works on R-typeMoSe2–WSe2heterostructures7,20–22. In the hBN separated region, we measured anexciton g-factor of −5:4± 1:0 (Fig. 3c), which has not been previouslyreported. We note that the opposite sign of the g-factor is consistentwith our zerofield circularly polarizedmeasurements showing that thehBN separated PL has co-circularly polarized PL when pumpingat 1.72 eV.Finally, we compared the temperature-dependent behavior ofboth IX species to investigate the origin of localization in the hetero-structure. Figure4a, b shows the temperature-dependent PL frombothregions of the heterostructure. In Fig. 4a, we see the IX emission fromthe DC region of the heterostructure. By increasing the temperature,the IX emission changes from a series of individual, narrow peaks tobecome a homogeneous broad peak around 13 K. The behavior of thehBN-separated region shows a disappearance of the narrow IX lines atalmost the same temperature (9 K); however, no broadpeakpersists tohigher temperature. In both cases, the narrow lines disappear aroundFig. 1 | Schematic of the sample and spatially resolved IX photoluminescence.a Cartoon depiction of the device shows the WSe2–MoSe2 heterostructure encap-sulated with hBN. Bilayer graphene is used for the top gate and graphite is used forthe bottom gate. Approximately half of the TMD heterostructure is separated bybilayer hBN to suppress the moiré potential. b Confocal PL spatial map of thisdevice, plotting the average center IX photon energy. The hBN separated region isshown in blue with 1.37 to 1.44 eV IX emission energy and the direct contact area isshown in red with the emission energy between 1.28 and 1.36 eV. c Co- and cross-circularly polarized PL spectrameasured by exciting the hBN separated regionwitha polarized 1.72 eV laser and collecting from both regions at the same time. Thesignal from the hBN separated region ~1.42 eV is mostly co-polarized, whereas thesignal from the DC region ~1.33 eV is mostly cross-polarized.Article https://doi.org/10.1038/s41467-022-33082-6Nature Communications |         (2022) 13:5354 210K, suggesting that the localization potential that gives rise to thenarrow lines is same for both regions and independent of the moirépotential. We note, however, that there are differences between thetwo sample regions. In the DC region, a broad PL peak persists totemperatures above 19 K, whereas in the hBN separated region the PLdisappears completely (Fig. 4c). See Supplementary Fig. 6 for highertemperatures. The temperature-dependent PL of both regions weremeasured in the same thermal cycle, using the same conditionsincluding exposure time, power and excitation wavelength. The mea-surement was repeated several times with similar results. We,therefore, present a simple picture that explains all of the observedbehaviors. In the DC region, a weak extrinsic localization potential sitson top of the deep ~50–100meV moiré potential (Fig. 4d). The moirépotential explains the lower IX energy in the DC region, the cross-circularly polarized PL, and the positive g-factor; even so, the narrowlines originate from the shallower extrinsic potential which disappearsabruptly at ~10K. However, the broader PL signal persists to highertemperature due to the trapping of IX by the wider, deeper moirépotential. Whereas, on the hBN-separated region, the moiré potentialis highly suppressed (Fig. 4d), explaining the higher IX energy, co-Fig. 3 | Photoluminescence as a function of magnetic field. a Magnetic fielddependent PL (detecting σ�) shows opposite sign of g-factors for the DC and hBNseparated regions. Example magnetic field dependence of a single IX line for bothDC (b) and hBN separated regions (c). The excitonic g-factors reported are averagevalues of the fitting to six single IX lines. The error bar shows one standarddeviation.Fig. 2 | IX photoluminescence as a function of electric field and doping level.a PL emission as a function of electric fieldmeasured by exciting the hBN separatedregionwhile collecting fromboth regions. The hBN separated IX lines showa largerdipole moment due to the increased electron-hole separation. b PL emission as afunction of doping level. The regions i, ii, and iii correspond to electron doping,neutral, and hole doping respectively.Article https://doi.org/10.1038/s41467-022-33082-6Nature Communications |         (2022) 13:5354 3polarized optical selection rule, and negative g-factor. In this region,localization is solely due to the extrinsic potential. As such, the narrowlines are observed below 10K, but the PL vanishes completely above19 K since the hBN-separated region does not have themoiré potentialto confine the IXs, and the strong dipole–dipole and exchange inter-action scatters the IXs outside of the detection area or light cone.DiscussionWe note that we considered the possibility that the trapping potentialresponsible for the narrow lines in the DC region is from the moirépotential with inhomogeneous broadening. If this were the case, thepotential depth of each moiré trap should be about the same withsome small variations. Theoretical predications and STM results haveshown that the depth of the moiré potential in direct contact R typeMoSe2–WSe2 heterostructures is on the order of ~50–100 meV5,23,whereas themoiré potential in the hBN separated region is consideredto be negligible as evidenced by our DFT calculation (SupplementaryFig. 1). Our temperature dependent data shows that in both regions,the narrow IX PL disappears at 9 K, indicating that the depth of thepotential causing the narrow IX lines is nearly the same in both regions.Based on this observation, we conclude that the nanoscale localizationpotential giving rise to the narrow lines in the DC region cannot be theintrinsic moiré potential. We also considered the possibility that thenarrow lines originate from atomic recontruction24. Previous workshave shown that atomic reconstruction in R-type heterostructuresexhibits two domains with RXh and RMh type stacking. Therefore, theselection rule of the DC region could be explained either by trappingvia the moiré potential or atomic reconstruction. Future studies per-forming low-frequency Raman scattering may be able to distinguishthese effects. However, the same arguments that apply to the moirépotential can be applied to atomic reconstruction since the hBN spacerwould suppress both effects.In summary, we have shown that the spectrally narrow IX linesthat were previously attributed to intrinsic trapping via the moirépotential are extrinsic in nature and originate from nanoscale defectsor nanobubbles formed during the 2D heterostructure fabricationprocess. In DC heterostructures this extrinsic potential sits on top ofthe moiré potential giving rise to narrow IX lines that exhibit thecharacteristics of intrinsic moiré IXs. Our result provides crucialinsights into future quantumdevice applications ofmoiré excitons andmotivates the development of improved 2D material fabricationtechniques to realize the goal of 2D quantum emitter arrays withhomogeneous energies. Our results demonstrate that spatially trap-ped IXs are required to obtain optically detectable PL. The dis-appearance of PL in the hBN separated region as the IXs becomedelocalized is evidence of this.MethodsSample fabricationThe layers of WSe2, MoSe2, hBN and graphene were exfoliated frombulk using the scotch tape method. The layer thickness was measuredby atomic forcemicroscopy and optical contrast. For aligning the TMDmonolayers near 0° degree precisely, we used polarization resolvedsecond harmonic generation spectroscopy25,26. The device was fabri-cated using the dry transfer technique27. The bottom and top hBNthicknesses were 22 and 8 nm respectively. The top gate was bilayergraphene, and the bottom gate was 2 nm thick graphite. 7 nm/40nmchrome gold contacts to the device were patterned using electronicbeam lithography and thermal evaporation.Optical measurementsFor the confocal and polarized PL measurements, we used a 1.72 eVphoton energy continuous wave laser (M Squared SOLSTIS) onresonance with WSe2. Unless otherwise noted the excitation powerwas 20 µW and the sample temperature was 1.6 K. The experimentswere performed in the reflection geometry focusing the laser andcollecting with a 0.81 NA attocube objective. In the confocal mea-surements, a detection area of 1 μm was achieved by spatially fil-tering the PL and using a 50 μm pinhole and a 50× magnificationconfocal setup. We used appropriate combinations of polarizersand achromatic waveplates to control the excitation and detectionpolarizations.Fig. 4 | Temperature-dependent PL and origin of narrow IX lines. a PL of thenarrow IX lines in the DC region as a function of temperature shows the width ofindividual lines are increasingby9 Kanddisappear completely by 17 K. The signal inthe DC region includes narrow emission on top of a wider PL plateau. The wideremission persists to higher temperature (see Supplementary Fig. 6) that is con-sistent with previous temperature-dependent measurements on R-type hetero-structures.b Temperature-dependent PL from the hBN separated region shows thenarrow lines are widening by 9K which is in good agreement with the DC region’stemperature-dependent PL. The hBN separated IX signal disappears fully by 13 Kand does not have a wider PL plateau. c Spectrally integrated PL shows IX emissionin the hBN separated region disappears completely by 19 Kwhereas the signal fromthe DC region is approximately constant. dDepiction of the IX energy as a functionof position in the moiré, for DC region (black) and hBN separated (Sep.) region(green). Both regions exhibit a weak extrinsic trapping potential denoted by thefluctuations. The DC region has both moiré trapping and extrinsic fluctuations.Article https://doi.org/10.1038/s41467-022-33082-6Nature Communications |         (2022) 13:5354 4Electronic controlThe electric field reported in Fig. 2a is calculated by Ehs =Vt�Vbtt + tb� �* ϵhBNϵhswhereVt (Vb) is the voltage applied to the top (bottom) gate, ttðbÞ is thethickness of the top (bottom) hBN, εhBN = 3:7 (εhs) is the relativedielectric constant of the hBN (heterostructure). The dielectric con-stant of the heterostructure was determined by taking the weightedaverage (weighted by layer thickness) of the dielectric constant of theTMD and hBN layers. This is calculated to be εhs = fðtWSe2*εWSe2Þ+ðthBN*εhBNÞ+ ðtMoSe2*εMoSe2Þg=ðtWSe2+ thBN + tMoSe2Þ = 6.26 where tWSe2,tMoSe2and thBN are the thickness of the WSe2, MoSe2, and hBN layersrespectively. The doping density is calculated using the parallel platecapacitor model where σ= εhBNðVb +VtÞ=ttotal.DFT CalculationWecreate a 1 × 1MoSe2/WSe2 unit cell with a lattice constant of 3.317 Å,which is the average of the MoSe2 and WSe2 lattice constants. All cal-culations of the K point bandgap (Eg) of MoSe2/WSe2 hetero-bilayerswith different translation r0 between layers are performed in the fra-mework of density functional theory by using a plane-wave basis set asimplemented in the Vienna ab initio simulation package. For eachgiven r0, we fix the interlayer distance d = 6.447 Å, which is the mini-mum value among all given r0 in our calculations. We calculate Eg forthis interlayer distance along with a series of increasing interlayerdistances of 6.547, 6.647, 6.747, 6.847, 6.947, 7.447, and 9.447 Å. Theseincreasing distances simulate the effect of separating the layers withthe addition of hBN.The electron-ion interactions are modeled using the projectoraugmented wave (PAW) potentials. The generalized gradient approx-imation (GGA) with the Perdew-Burke-Ernzerhof (PBE) functional withvan der Waals corrections (vdW) for the exchange-correlation inter-actions is used. In all of our calculations the spin-orbit coupling (SOC)is fully taken into account. A vacuum of 15 Å is used in all the calcula-tions to avoid interaction between the neighboring slabs. The plane-wave cutoff energy is set to 500 eV and the first Brillouin zone of theunit cell of MoSe2/WSe2 hetero-bilayers is sampled by using theMonkhorst–Pack scheme of k-points with the 12 × 12 × 1 mesh for thestructural optimization and the 24× 24× 1mesh for the band structure.The residual forces have converged to less than0.01 eV/Å and the totalenergy difference to less than 10−5 eV.Data availabilityThe data that support the findings of this study are available in theFigshare database at the following links: https://figshare.com/projects/Localized_Interlayer_Excitons_in_MoSe2-WSe2_Heterostructures_without_a_Moir_Potential/146136.Code availabilityThe code that support the findings of this study are available in theFigshare database at the following links: https://figshare.com/projects/Localized_Interlayer_Excitons_in_MoSe2-WSe2_Heterostructures_without_a_Moir_Potential/146136.References1. Rivera, P. et al. Observation of long-lived interlayer excitons inmonolayer MoSe2–WSe2 heterostructures. Nat. Commun. 6,6242 (2015).2. Schaibley, J. R. et al. Directional interlayer spin-valley transfer intwo-dimensional heterostructures. Nat. Commun. 7, 13747 (2016).3. Schaibley, J. R. et al. Valleytronics in 2Dmaterials.Nat. Rev.Mater. 1,16055 (2016).4. Rivera, P. et al. Valley-polarized exciton dynamics in a 2D semi-conductor heterostructure. Science 351, 688–691 (2016).5. Yu, H., Liu, G.-B., Tang, J., Xu, X. & Yao, W. Moiré excitons: Fromprogrammable quantum emitter arrays to spin-orbit–coupled arti-ficial lattices. Sci. Adv. 3, e1701696 (2017).6. Ross, J. S. et al. Interlayer exciton optoelectronics in a 2D hetero-structure p–n junction. Nano Lett. 17, 638–643 (2017).7. Seyler, K. L. et al. Signatures of moiré-trapped valley excitons inMoSe2/WSe2 heterobilayers. Nature 567, 66–70 (2019).8. Tran, K. et al. Evidence for moiré excitons in van der Waals het-erostructures. Nature 567, 71–75 (2019).9. Rivera, P. et al. Interlayer valley excitons in heterobilayers of transitionmetal dichalcogenides. Nat. Nanotechnol. 13, 1004–1015 (2018).10. Yankowitz, M. et al. Emergence of superlattice Dirac points in gra-phene on hexagonal boron nitride. Nat. Phys. 8, 382–386 (2012).11. Huang, D., Choi, J., Shih, C.-K. & Li, X. Excitons in semiconductormoiré superlattices. Nat. Nanotechnol. 17, 227–238 (2022).12. Liu, E. et al. Signatures of moiré trions in WSe2/MoSe2 hetero-bilayers. Nature 594, 46–50 (2021).13. Baek, H. et al. Optical read-out of Coulomb staircases in a moirésuperlattice via trapped interlayer trions. Nat. Nanotechnol. 16,1237–1243 (2021).14. Baek, H. et al. Highly energy-tunable quantum light from moiré-trapped excitons. Sci. Adv. 6, eaba8526 (2020).15. Mahdikhanysarvejahany, F. et al. Temperature dependent moirétrappingof interlayer excitons inMoSe2–WSe2heterostructures.npj2D Mater. Appl. 5, 67 (2021).16. Jauregui, L. A. et al. Electrical control of interlayer exciton dynamicsin atomically thin heterostructures. Science 366, 870–875 (2019).17. Shanks, D. N. et al. Nanoscale trapping of interlayer excitons in a 2Dsemiconductor heterostructure. Nano Lett. 21, 5641–5647 (2021).18. Unuchek, D. et al. Valley-polarized exciton currents in a van derWaals heterostructure. Nat. Nanotechnol. 14, 1104–1109 (2019).19. Wang, X. et al. Moiré trions in MoSe2/WSe2 heterobilayers. Nat.Nanotechnol. 16, 1208–1213 (2021).20. Ciarrocchi, A. et al. Polarization switching and electrical control ofinterlayer excitons in two-dimensional van der Waals hetero-structures. Nat. Photonics 13, 131–136 (2019).21. Woźniak, T., Faria Junior, P. E., Seifert, G., Chaves, A. & Kunstmann,J. Exciton g factors of van der Waals heterostructures from first-principles calculations. Phys. Rev. B 101, 235408 (2020).22. Joe, A. Y. et al. Electrically controlled emission from singlet andtriplet exciton species in atomically thin light-emitting diodes. Phys.Rev. B 103, 161411 (2021).23. Nieken, R. et al. Direct STM measurements of R-type and H-typetwisted MoSe2/WSe2. APL Mater. 10, 031107 (2022).24. Rosenberger, M. R. et al. Twist angle-dependent atomic recon-struction and Moiré patterns in transition metal dichalcogenideheterostructures. ACS Nano 14, 4550–4558 (2020).25. Kumar, N. et al. Second harmonic microscopy of monolayer MoS2.Phys. Rev. B 87, 161403(R) (2013).26. Malard, L. M., Alencar, T. V., Barboza, A. P. M., Mak, K. F. & de Paula,A. M. Observation of intense second harmonic generation fromMoS2 atomic crystals. Phys. Rev. B 87, 201401(R) (2013).27. Zomer, P. J., Guimarães, M. H. D., Brant, J. C., Tombros, N. & vanWees, B. J. Fast pick up technique for high quality heterostructuresof bilayer graphene and hexagonal boron nitride. Appl. Phys. Lett.105, 013101 (2014).AcknowledgementsWe acknowledge Hongyi Yu for useful discussions. J.R.S. and B.J.L.acknowledge support from theNational Science FoundationGrant. Nos.DMR-2003583 and ECCS-2054572. J.R.S. acknowledges support fromAir ForceOffice of Scientific ResearchGrant Nos. FA9550-20-1-0217 andFA9550-21-1-0219. B.J.L. acknowledges support from theArmyResearchOffice under Grant nos. W911NF-18-1-0420 and W911-NF-20-1-0215.D.G.M. acknowledges support from the Gordon and Betty MooreFoundation’s EPiQS Initiative, Grant GBMF9069. K.W. and T.T.acknowledge support from JSPS KAKENHI Grant Nos. 19H05790,20H00354 and 21H05233.Article https://doi.org/10.1038/s41467-022-33082-6Nature Communications |         (2022) 13:5354 5https://figshare.com/projects/Localized_Interlayer_Excitons_in_MoSe2-WSe2_Heterostructures_without_a_Moir_Potential/146136https://figshare.com/projects/Localized_Interlayer_Excitons_in_MoSe2-WSe2_Heterostructures_without_a_Moir_Potential/146136https://figshare.com/projects/Localized_Interlayer_Excitons_in_MoSe2-WSe2_Heterostructures_without_a_Moir_Potential/146136https://figshare.com/projects/Localized_Interlayer_Excitons_in_MoSe2-WSe2_Heterostructures_without_a_Moir_Potential/146136https://figshare.com/projects/Localized_Interlayer_Excitons_in_MoSe2-WSe2_Heterostructures_without_a_Moir_Potential/146136https://figshare.com/projects/Localized_Interlayer_Excitons_in_MoSe2-WSe2_Heterostructures_without_a_Moir_Potential/146136Author contributionsJ.R.S. and B.J.L. conceived and supervised the project. D.N.S. fabricatedthe structures and F.M. performed the experiments, assisted by D.N.S.and M.K. F.M. analyzed the data with input from D.N.S., J.R.S., and B.J.L.M.R.K. and D.G.M. provided and characterized the bulkMoSe2 andWSe2crystals. T.T. andK.W. provided the bulk hBNcrystals. F.M.,D.N.S., J.R.S.,and B.J.L. wrote the paper with input from O.L.A.M., T.S. and R.L.N. Allauthors discussed the results.Competing interestsThe authors declare no competing interests.Additional informationSupplementary information The online version containssupplementary material available athttps://doi.org/10.1038/s41467-022-33082-6.Correspondence and requests for materials should be addressed toJohn R. Schaibley.Peer review information Nature Communications thanks Yia-ChungChang and the other, anonymous, reviewer(s) for their contribution tothe peer review of this work. Peer reviewer reports are available.Reprints and permission information is available athttp://www.nature.com/reprintsPublisher’s note Springer Nature remains neutral with regard tojurisdictional 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 license, and indicate ifchanges were made. The images or other third party material in thisarticle are included in the article’s Creative Commons license, unlessindicated otherwise in a credit line to the material. If material is notincluded in the article’s Creative Commons license 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 license, visit http://creativecommons.org/licenses/by/4.0/.© The Author(s) 2022Article https://doi.org/10.1038/s41467-022-33082-6Nature Communications |         (2022) 13:5354 6https://doi.org/10.1038/s41467-022-33082-6http://www.nature.com/reprintshttp://creativecommons.org/licenses/by/4.0/http://creativecommons.org/licenses/by/4.0/ Localized interlayer excitons in MoSe2&#x02013;nobreakWSe2 heterostructures without a moiré potential Results Discussion Methods Sample fabrication Optical measurements Electronic control DFT Calculation Data availability Code availability References Acknowledgements Author contributions Competing interests Additional information