Single-photon sources are basic building blocks for quantum communications, processing, and metrology. Solid-state quantum emitters in semiconductors have the potential for robust and reliable generation of photons, and atomically thin transition metal dichalcogenides, such as , , , and , are a promising new class of two-dimensional semiconductors with a direct optical bandgap in the visible or near-IR. Here, we observe bright and stable single-photon emission from localized excitons in a monolayer of tungsten diselenide (). The emitters appear at the edges of the flakes and are linearly polarized. The spectral width of their emission is below 120 μeV in a freestanding monolayer. Photoluminescence excitation spectroscopy reveals the excitonic nature of the emitters and provides evidence that these single excitons originate from free excitons trapped in local potential wells at the edges of the atomically thin flakes. We find that the emitters can also be deterministically created by scratching the WSe2 monolayer. Their excellent spectral stability implies that these localized single-photon emitters could find application in optoelectronics. Our results light the way to single exciton physics and quantum optics with atomically thin semiconductors.
© 2015 Optical Society of America
Two-dimensional materials based on carbon (graphene), boron nitride, or transition metal dichalcogenides (TMDCs) have raised considerable attention in both basic research and technology [1,2]. In particular, group 6 TMDCs , , , and are promising materials for optoelectronics due to their unique optical properties. They have been used for building transistors, photodetectors, and light-emitting devices [3,4]. TMDCs appear in the form ( metal atom and atom) and crystallize in a strongly covalently bound sandwich structure . Consecutive layers are weakly bound by the van der Waals interaction. While bulk and multilayer , , , and are indirect semiconductors, monolayers have a direct optical bandgap, which occurs at the K and K’ points of the Brillouin zone [6–8]. The large spin-orbit splitting and absence of inversion symmetry in these materials provide optical access to the valley degree of freedom with circularly polarized light , rendering these materials as promising candidates for valleytronics. Their photoluminescence (PL) spectra are governed by the emission from strongly bound, delocalized excitons. In addition, a broad band of defect-activated emission is observed [10,11].
In this work, we report on the observation of single quantum emitters in atomically thin . We find that this single-photon emission arises from excitons trapped in local potential wells of the monolayers.
2. SAMPLE FABRICATION AND METHODS
Atomically thin is obtained by mechanical exfoliation of a synthetically grown single crystal. Monolayers are placed on three different substrates: either directly onto a substrate, on a thin hexagonal boron nitride (h-BN) flake on , or on a substrate with etched holes. Using a laser scanning confocal microscope equipped with a helium-flow cryostat, we measure the spatially resolved PL emitted from the monolayer at cryogenic temperatures. The monolayer is excited off-resonantly with a cw laser at 532 nm at excitation power densities between 250 and . PL emission is collected with an objective lens with a numerical aperture and sent to a monochromator with a focal length of 750 mm (Andor Shamrock SR-750). PL is detected using a thermoelectrically cooled, back-illuminated deep-depletion charge-coupled device (CCD) (Andor DU420A BR-DD) with integration times between 0.1 and 300 s. For photon statistics measurements a Hanbury Brown and Twiss (HBT) setup is used. The photon emission is filtered by the same monochromator as for the PL measurements and split into two arms for the HBT correlation setup using a nonpolarizing 50:50 beam splitter. The photons are detected by two single-photon counters (PicoQuant tau-SPAD-100 and Excelitas SPCM-AQRH-13). The time resolution is 0.53 ns for the tau-SPAD and 0.63 ns for the Excelitas SPCM. The correlation histogram is recorded with a time-correlated single-photon counting peripheral component interconnect card (Becker & Hickl SPC-130).
3. RESULTS AND DISCUSSION
Figure 1(a) presents the PL spectrum, measured at a temperature of in the center of a single-layer flake. We find weak neutral A exciton () and charged exciton emission ( or ) at 1.747 and 1.718 eV, respectively, and a strong broad emission band, which is attributed to defect excitons in the range 1.6–1.7 eV . The spectrum shows only small variations in the integrated PL intensity across most parts of the monolayer. Also, the relative intensity of the different spectral features is constant across the flake. However, at some locations close to the edge of the monolayer the characteristics of the emission change drastically. As illustrated in the spatially resolved PL image in Fig. 1(b), strongly localized emission centers become visible for the narrow spectral window of . In Fig. 1(c), the PL of one of these centers is mapped at high spatial resolution. A Gaussian-shaped emission spot appears with a full width at half-maximum (FWHM) of 690 nm, limited by the optical resolution of our laser scanning setup. The PL spectrum at this position exhibits several sharp spectral lines, with a prominent one at 1.707 eV [Fig. 1(d)]. At the same time, the emission from the neutral and charged exciton is substantially reduced.
B. Photon Statistics
To explore the nature of the photon emission from the localized center, we measure the photon statistics using the HBT setup and spectral selection of the energy range . The measured second-order correlation function exhibits a prominent dip at zero time delay with [Fig. 1(e)]. A value of proves that we observe a single-photon emitter [12,13]. From the spectrum in Fig. 1(d) we determine that 17% (1700 counts/s) of the total photon count rate (10,000 counts/s) at is due to uncorrelated background emission. Using this value and taking into account the finite response time of the single-photon detectors, we can calculate a background-corrected , demonstrating that the centers are excellent single-photon emitters.
The rise time of the second-order correlation function is determined by the excitation rate and luminescence decay time. Measuring the rise time provides a lower limit of the luminescence decay time of the light source. It amounts to 1.8 ns for this emitter in monolayer on the substrate. Other centers on the same monolayer at different locations and emission energies show rise times up to 6.5 ns (data not shown). These times are in the same order of magnitude as the decay time for free excitons in , which has been measured recently at room temperature () .
The usability of single-photon sources critically depends on the stability of photon emission [15–18]. For example, colloidal semiconductor quantum dots with diameters of a few nanometers typically suffer from off times (blinking), spectral jumps (spectral diffusion), and even disappearance of photon emission (photobleaching). These detrimental effects are intimately linked to their large surface-to-volume ratio and presence of surface defects and charges. For TMDC monolayers the surface-to-volume ratio is even higher. However, we have not observed photobleaching of the emitters for hours. Moreover, the same single center is still present after several days and cycling of the temperature between 300 and 10 K. The spectral stability of the single-photon emitter is also excellent, especially if excited near-resonantly. Over 1 h the spectral diffusion only amounts to 0.5 meV (data not shown). With off-resonant excitation, spectral wandering increases to 2.5 meV due to high pumping into the conduction band, which generates a large number of additional charges that affect the spectral stability of the localized excitons.
C. Influence of the Substrate
We therefore investigate whether these favorable emission properties are due to a fortunate choice of the underlying substrate. Figures 1(f) and 1(g) depict photoluminescence and photon statistical data for a localized exciton in monolayer , transferred on a 49 nm thick h-BN flake previously deposited on a substrate. Interestingly, the emission from the neutral exciton and the negatively charged exciton are almost absent for this substrate. A deep antibunching dip of again proves single-photon emission. Obviously, the effect of the two different substrate materials with regard to emitter stability is negligible. In contrast, the rise time of is 7 ns on , which is longer than for all the centers we have measured on on the substrate. Additional measurements from other localized emitters on the substrate yield decay times up to 9.6 ns (data not shown). This lengthening might be due to a change of radiative and/or nonradiative decay time. The radiative decay time on h-BN is expected to be shorter compared to , due to the higher photonic density of states of the h-BN surroundings (, ). Therefore, a change in the radiative decay rate due to different surroundings of the nanoemitter cannot account for the observed lengthening of the decay time. Hence, a modification of the nonradiative decay rate is probably the main contribution, which requires further investigation on a microscopic scale.
To fully eliminate the influence of a substrate on the nanoemitters, we investigate a freestanding monolayer. Again, we observe sharp emission lines at different locations on the monolayer. However, the width of the lines drastically narrows. We observe emission maxima with a FWHM limited by the spectral resolution of our setup of 120 μeV with near-resonant excitation [see Fig. 2(a)].
The observed linewidth is at least one order of magnitude smaller than for on the two different substrates (2.3 meV). This narrowing is probably due to significantly fewer charge fluctuations in the surroundings of the localized excitons, leading to smaller and slower spectral jitter. The spectral wandering over 8 min, shown in Fig. 2(b), amounts to 1 meV, which is in the same order of magnitude as for the localized excitons on or h-BN. The shift of the two lines shown in Fig. 2(b) is identical, which might indicate that they belong to the same emitter.
Compared to on the or substrate, the PL decay time of the localized excitons in freestanding monolayers significantly increases by about one order of magnitude. This behavior is in good agreement with previous experiments on excitons in freestanding carbon nanotubes  and can be explained by fewer PL quenching sites. However, the long decay times result in a weak intensity of the PL emission, which renders photon correlation measurements difficult for freestanding monolayers.
D. Origin of the Single-Photon Emitters
We now turn to the nature of the single-photon emitter. It is known that missing atoms (vacancies) in TMDC monolayers are exciton trapping sites [11,20]. Alternatively, excitons may bind to impurity atoms or can be trapped in a potential well created by structural defects or local strain. Interestingly, close inspection of our PL maps reveals that stable and sharp emission lines, i.e., the single-photon emitters, preferably occur at the edges of the flakes. The effect of one-dimensional edge states in [21,22] or graphene  on the electronic and optical properties has been recently pointed out. However, only in monolayer has strong PL from the edges of triangular flakes been reported . Figure 3(a) shows a PL intensity map integrated over the broad spectral range 1.55–1.77 eV. The brightest emission centers are all located at the edge of the monolayer and exhibit narrow spectral features [curves i–iv in Fig. 3(b)]. In the center of the flake they are absent [curve v in Fig. 3(b)]. Figure 3(c) depicts a PL intensity map for a narrow energy range far outside the defect band. The flake boundaries are barely visible. In contrast, an energy range well inside the defect band is selected in Fig. 3(d). Here, the shape of the flake is discernible. However, the PL emission intensity of the localized excitons is much brighter than the emission from the rest of the flake. Therefore, the quantum efficiency of these emission centers should be significantly higher than for the neutral exciton and for the broad defect exciton band. The PL map of the neutral exciton is displayed in Fig. 3(e). We find a uniform emission except for locations at the edge of the flake, where PL emission of the localized excitons dominates, substantially reducing the PL intensity. This observation indicates that neutral excitons are involved in the formation of the localized excitons.
Since localized emission centers only appear close to the edge of the monolayer, the question arises whether it is possible to deliberately create centers in a monolayer. Therefore, we use a needle (Berrylium Copper, ) to scratch a monolayer on substrate. The atomic force microscope image of the scratched flake in Fig. 4(a) reveals some folded edges, where enhanced PL emission appears [Fig. 4(b)]. Figure 4(c) shows the spectra of three of these folds containing one or more emission centers.
E. Saturation and Polarization
To gain further insight into the nature of the emission centers, we investigate the power dependence of the PL response. The emitter is excited from the ground state into a higher excited state or into the continuum and then relaxes radiatively or nonradiatively into the lowest excited state. The observed PL originates from the recombination back into the ground state. For a single quantum emitter a saturation of the PL intensity is expected, since the maximum number of emitted single photons is limited by the lifetime of the excited state . Increasing the optical excitation power therefore results in a nonlinear dependence of the PL intensity, which can be described by the equation , where is the measured PL intensity and the PL intensity at saturation. is the power density of laser excitation. Evidence for the saturation behavior of single localized excitons is depicted in Fig. 5(a). The saturation power density () for the emitter in Fig. 1 amounts to . We find that the saturation power density varies for different emitters, but saturation is always observed.
The polarization dependence of the emitted PL indicates the orientation of the emission dipole. The excitonic transitions in monolayer TMDCs exhibit an in-plane dipole . Under off-resonant excitation the PL emission from the neutral and charged excitons, as well as the broad emission of the defect exciton band, is unpolarized . We find that the emission of single localized excitons is linearly polarized. The polarization angle randomly varies from emitter to emitter. This might be caused by anisotropic strain at the position of the emitter, similar to self-assembled semiconductor quantum dots . The angular dependence of the PL [Fig. 5(b)] indicates that the emitter exhibits an in-plane dipole, similar to the neutral exciton .
F. Energy Level Structure
The energy level structure of the single-photon emitter is investigated by photoluminescence excitation (PLE) spectroscopy. We detect the intensity of the PL at the energy of the ground state emission while exciting the exciton with a tunable Ti:sapphire laser at excitation power densities between 2000 and and a photon energy between 50 and 200 meV above the ground state. The measured PLE spectrum reveals prominent absorption maxima (Fig. 6). The energetic positions are labeled with the quantum number . The energetically highest absorption maximum at 1.75 eV corresponds to the ground state of the free neutral exciton in monolayer [Fig. 1(a)].
We believe that the observed localized excitons originate from neutral excitons trapped in a confinement potential, similar to self-assembled (“natural”) semiconductor quantum dots [29,30]. However, the confining potential in the monolayers is probably caused by a structural defect or local strain, because localized excitons appear only at the edge of the flakes. Our interpretation is also consistent with the observation that the PL intensity from the neutral exciton is strongly reduced at locations where the localized exciton is found [Fig. 3(e)], because free neutral excitons relax into the trapping potential.
The depth of the potential well, i.e., the confinement energy, is given by the spacing between the PL emission energy of the localized exciton and the free exciton absorption (1.75 eV). Investigations of many emission centers yield PL energies varying between 1.55 and 1.71 eV and result in confinement energies between 40 and 200 meV.
The confinement energies and the spacing of the energy levels are significantly smaller compared to the binding energy and the spacing of the energy levels of the free neutral exciton in monolayer . Notably, the energy difference between the ground and first excited states of the free neutral exciton in monolayer (170 meV) is larger or comparable to the depth of the confinement potential of the localized excitons. This situation is in contrast to typical semiconductor quantum dot systems (e.g., AlAs/GaAs), where the confinement potential is much larger than the exciton binding energy [29,30]. Therefore, we believe that the observed PLE maxima represent discrete energy levels of the confinement potential, which are occupied by an exciton in its ground state.
In conclusion, we have observed single-photon emission from localized excitons in monolayers. We find that they only appear close to the edges of the flakes, probably due to structural defects or inhomogeneous strain. With photon correlation and spectral diffusion measurements, we demonstrate that these excitons are excellent single-photon emitters. The spectral width of the emission is below 120 μeV in a freestanding monolayer. On the substrate the emission line broadens to a few millielectron volts. However, the spectral stability is excellent, which renders these localized emitters in monolayer promising for applications in optoelectronics and quantum information. Our results also open the door to single-exciton physics and quantum optics with atomically thin semiconductors.
We thank Hartmut Bracht for granting access to the atomic force microscope.
1. A. K. Geim and I. V. Grigorieva, “Van der Waals heterostructures,” Nature 499, 419–425 (2013). [CrossRef]
2. F. Xia, H. Wang, D. Xiao, M. Dubey, and A. Ramasubramaniam, “Two-dimensional material nanophotonics,” Nat. Photonics 8, 899–907 (2014). [CrossRef]
3. Q. H. Wang, K. Kalantar-Zadeh, A. Kis, J. N. Coleman, and M. S. Strano, “Electronics and optoelectronics of two-dimensional transition metal dichalcogenides,” Nat. Nanotechnol. 7, 699–712 (2012). [CrossRef]
4. R. Bratschitsch, “Optoelectronic devices: monolayer diodes light up,” Nat. Nanotechnol. 9, 247–248 (2014). [CrossRef]
5. R. F. Frindt and A. D. Yoffe, “Physical properties of layer structures: optical properties and photoconductivity of thin crystals of molybdenum disulphide,” Proc. R. Soc. A 273, 69–83 (1963). [CrossRef]
6. A. Splendiani, L. Sun, Y. Zhang, T. Li, J. Kim, C.-Y. Chim, G. Galli, and F. Wang, “Emerging photoluminescence in monolayer MoS2,” Nano Lett. 10, 1271–1275 (2010). [CrossRef]
7. K. F. Mak, C. Lee, J. Hone, J. Shan, and T. F. Heinz, “Atomically thin MoS2: a new direct-gap semiconductor,” Phys. Rev. Lett. 105, 136805 (2010). [CrossRef]
8. P. Tonndorf, R. Schmidt, P. Böttger, X. Zhang, J. Börner, A. Liebig, M. Albrecht, C. Kloc, O. Gordan, D. R. T. Zahn, S. Michaelis de Vasconcellos, and R. Bratschitsch, “Photoluminescence emission and Raman response of monolayer MoS2, MoSe2, and WSe2,” Opt. Express 21, 4908–4916 (2013). [CrossRef]
9. K. Behnia, “Condensed-matter physics: polarized light boosts valleytronics,” Nat. Nanotechnol. 7, 488–489 (2012). [CrossRef]
10. T. Korn, S. Heydrich, M. Hirmer, J. Schmutzler, and C. Schüller, “Low-temperature photocarrier dynamics in monolayer MoS2,” Appl. Phys. Lett. 99, 102109 (2011). [CrossRef]
11. S. Tongay, J. Suh, C. Ataca, W. Fan, A. Luce, J. S. Kang, J. Liu, C. Ko, R. Raghunathanan, J. Zhou, F. Ogletree, J. Li, J. C. Grossman, and J. Wu, “Defects activated photoluminescence in two-dimensional semiconductors: interplay between bound, charged, and free excitons,” Sci. Rep. 3, 2657 (2013). [CrossRef]
12. L. Mandel and E. Wolf, Optical Coherence and Quantum Optics (Cambridge University, 1995).
13. P. Michler, A. Imamoglu, M. Mason, P. Carson, G. Strouse, and S. Buratto, “Quantum correlation among photons from a single quantum dot at room temperature,” Nature 406, 968–970 (2000). [CrossRef]
14. S. Mouri, Y. Miyauchi, M. Toh, W. Zhao, G. Eda, and K. Matsuda, “Nonlinear photoluminescence in atomically thin layered WSe2 arising from diffusion-assisted exciton–exciton annihilation,” Phys. Rev. B 90, 155449 (2014). [CrossRef]
15. H. J. Kimble, “The quantum internet,” Nature 453, 1023–1030 (2008). [CrossRef]
16. A. Beveratos, R. Brouri, T. Gacoin, A. Villing, J.-P. Poizat, and P. Grangier, “Single photon quantum cryptography,” Phys. Rev. Lett. 89, 187901 (2002). [CrossRef]
17. E. Knill, R. La, and G. J. Milburn, “A scheme for efficient quantum computation with linear optics,” Nature 409, 46–52 (2001). [CrossRef]
18. S. V. Polyakov and A. L. Migdall, “Quantum radiometry,” J. Mod. Opt. 56, 1045–1052 (2009). [CrossRef]
19. M. S. Hofmann, J. T. Glückert, J. Noé, C. Bourjau, R. Dehmel, and A. Högele, “Bright, long-lived and coherent excitons in carbon nanotube quantum dots,” Nat. Nanotechnol. 8, 502–505 (2013). [CrossRef]
20. D. Liu, Y. Guo, L. Fang, and J. Robertson, “Sulfur vacancies in monolayer MoS2 and its electrical contacts,” Appl. Phys. Lett. 103, 183113 (2013). [CrossRef]
21. M. V. Bollinger, J. V. Lauritsen, K. W. Jacobsen, J. K. Nørskov, S. Helveg, and F. Besenbacher, “One-dimensional metallic edge states in MoS2,” Phys. Rev. Lett. 87, 196803 (2001). [CrossRef]
22. X. Yin, Z. Ye, D. A. Chenet, Y. Ye, K. O’Brien, J. C. Hone, and X. Zhang, “Edge nonlinear optics on a MoS2 atomic monolayer,” Science 344, 488–490 (2014). [CrossRef]
23. K. Nakada, M. Fujita, G. Dresselhaus, and M. Dresselhaus, “Edge state in graphene ribbons: nanometer size effect and edge shape dependence,” Phys. Rev. B 54, 17954–17961 (1996). [CrossRef]
24. H. R. Gutiérrez, N. Perea-López, A. L. Elías, A. Berkdemir, B. Wang, R. Lv, F. López-Urías, V. H. Crespi, H. Terrones, and M. Terrones, “Extraordinary room-temperature photoluminescence in triangular WS2 monolayers,” Nano Lett. 13, 3447–3454 (2013). [CrossRef]
25. L. Allen and J. Eberly, Optical Resonance and Two-Level Atoms (Dover, 1988).
26. J. A. Schuller, S. Karaveli, T. Schiros, K. He, S. Yang, I. Kymissis, J. Shan, and R. Zia, “Orientation of luminescent excitons in layered nanomaterials,” Nat. Nanotechnol. 8, 271–276 (2013). [CrossRef]
27. A. M. Jones, H. Yu, N. J. Ghimire, S. Wu, G. Aivazian, J. S. Ross, B. Zhao, J. Yan, D. G. Mandrus, D. Xiao, W. Yao, and X. Xu, “Optical generation of excitonic valley coherence in monolayer WSe2,” Nat. Nanotechnol. 8, 634–638 (2013). [CrossRef]
28. D. Gammon, E. S. Snow, B. V. Shanabrook, D. S. Katzer, and D. Park, “Fine structure splitting in the optical spectra of single GaAs quantum dots,” Phys. Rev. Lett. 76, 3005–3008 (1996). [CrossRef]
29. K. Brunner, U. Bockelmann, G. Abstreiter, M. Walther, G. Böhm, G. Tränkle, and G. Weimann, “Photoluminescence from a single GaAs/AlGaAs quantum dot,” Phys. Rev. Lett. 69, 3216–3219 (1992). [CrossRef]
30. A. Zrenner, L. Butov, M. Hagn, G. Abstreiter, G. Böhm, and G. Weimann, “Quantum dots formed by interface fluctuations in AlAs/GaAs coupled quantum well structures,” Phys. Rev. Lett. 72, 3382–3385 (1994). [CrossRef]
31. K. He, N. Kumar, L. Zhao, Z. Wang, K. F. Mak, H. Zhao, and J. Shan, “Tightly bound excitons in monolayer WSe2,” Phys. Rev. Lett. 113, 026803 (2014). [CrossRef]
32. A. Srivastava, M. Sidler, A. V. Allain, D. S. Lembke, A. Kis, and A. Imamoglu, “Optically active quantum dots in monolayer WSe2,” arXiv:1411.0025v1 [cond-mat.mes-hall] (2014).
33. Y. He, G. Clark, J. R. Schaibley, Y. He, M. Chen, Y. Wei, X. Ding, W. Zhang, W. Yao, X. Xu, C.-Y. Lu, and J.-W. Pan, “Single quantum emitters in monolayer semiconductors,” arXiv: 1411.2449v1 (2014).
34. M. Koperski, K. Nogajewski, A. Arora, J. Marcus, P. Kossacki, and M. Potemski, “Single photon emitters in exfoliated WSe2 structures,” arXiv:1411.2774v2 (2014).