⇦Chinese Physics Letters, 2017, Vol. 34, No. 7, Article code 077801 Fluorescence Intermittency in Monolayer WSe$_{2}$ * Yan-Xia Ye(叶艳霞)1,2, Xiu-Ming Dou(窦秀明)1,2, Kun Ding(丁琨)1, Fu-Hua Yang(杨富华)1,2, De-Sheng Jiang(江德生)1, Bao-Quan Sun(孙宝权)1,2** Affiliations 1State Key Laboratory for Superlattices and Microstructures, Institute of Semiconductors, Chinese Academy of Sciences, Beijing 100083 2College of Materials Science and Opto-Electronic Technology, University of Chinese Academy of Sciences, Beijing 100049 Received 13 April 2017 *Supported by the National Basic Research Program of China under Grant No 2013CB922304, and the National Natural Science Foundation of China under Grant Nos 11474275, 61674135 and 91536101.
**Corresponding author. Email: bqsun@semi.ac.cn Citation Text: Ye Y X, Dou X M, Ding K, Yang F H and Jiang D S et al 2017 Chin. Phys. Lett. 34 077801 Abstract Fluorescence intermittent dynamics of single quantum emitters in monolayer WSe$_{2}$ are investigated via measuring spectrally resolved time traces and time-dependent fluorescence intensity trajectories. Analysis of fluorescence trajectories and spectral shifting reveal a correlation between the fluorescence intermittency and spectral diffusion. A model of an inverse power law can be used to understand the observed blinking dynamics. DOI:10.1088/0256-307X/34/7/077801 PACS:78.67.-n, 61.72.-y, 62.50.-p © 2017 Chinese Physics Society Article Text Single photon emission in emerging two-dimensional (2D) materials was reported in recent years,[1-5] which has attracted a great deal of interest and exploration.[6-10] Single photon sources are crucial for quantum information processing and quantum communication applications. Note that single photon emissions can be found in an isolated quantum state, such as atoms,[11,12] molecules,[13] single quantum dots,[14,15] and impurity centers.[16,17] As Bohr predicted the quantum jump of electrons between discrete energy levels of atoms, universal emission intermittency has been observed and several models were proposed for understanding the fluorescence intermittency (blinking). A power-law distribution of on- and off-times, $P_{\rm on/off}(t)\sim t^{-m}$, is proposed to explain the blinking events due to a fluctuating environment of isolated quantum states.[18-20] In this case, spectral diffusion was observed in molecules[21] or semiconductor dots[22,23] due to the perturbing interactions induced a stochastic spectrum shift or the localized charges in the vicinity of a quantum dot induced spectrum shift through the quantum confined Stark effect. For a charged exciton (trion), the Efros–Rosen model first predicts characteristic on/off rates through the Auger recombination process,[24] leading to fluorescence intermittency and corresponding exponential on-time/off-time distributions, $P_{\rm on/off}(t)\sim \exp(-t/t_{\rm c})$. A third model of considering non-radiative relaxation of the exciton through multiple recombination centers was presented and the blinking effect can be described by a truncated power-law distribution,[25] $P_{\rm on/off}(t)\sim t^{-m}\exp(-t/t _{\rm c})$. Recently, fluorescence intermittency was observed in 2D vertically stacked bilayer heterostructures,[26] which can be explained by the models mentioned above. Note that spectral diffusion will reduce the coherence characteristic time of the quantum dots, resulting in an increase of quantum dot emission linewidth from μeV to a few meV,[27] which is a great obstacle to the single photon source applications. However, to study the correlation between fluorescence intermittency and spectral diffusion in isolated quantum states can help us to better understand the basic quantum jump process and to improve the quality of the quantum emitters. In this Letter, we report on observations of fluorescence intermittency (blinking), and spectral shifting and forking of single quantum emitters in monolayer WSe$_{2}$ by measuring spectrally resolved time traces and time-dependent fluorescence trajectories. Single quantum emitters relevant to discrete quantum levels are observed at low temperature by applying pressure. It is noticed that blinking events are correlated with large spectral diffusion during the observation time (integration time), resulting in a broadening fluorescence line width of a few meV. An inverse power law in both on/off time can well describe the blinking dynamic processes. The monolayer WSe$_{2}$ flakes were prepared by micromechanical exfoliation from a bulk WSe$_{2}$ (2D Semiconductors supplied) on a thinned SiO$_{2}$/Si substrate, which are identified by microscopic image contrast and typical photoluminescence (PL) spectra at room temperature, as shown in Figs. 1(a) and 1(b). A high pressure can be applied to the measured samples by using a diamond anvil cell (DAC) device, as shown in Fig. 1(c). Condensed argon was used as the pressure-transmitting medium, and the ruby R$_{1}$ fluorescence line shift was used to determine pressure. The calibrated temperature of the cryogenic DAC sample chamber is 20 K by measuring the PL intensity ratio of the ruby R$_{2}$/R$_{1}$ line. PL was collected by a 20$\times$ objective (NA: 0.35) and spectrally analyzed using a 0.5 monochromator equipped with a silicon charge-coupled device (CCD) at an excitation of tens of μW by a cw 640 nm semiconductor laser. An avalanche photodiode (APD) with a temporal response of 380 ps was used for measuring time-dependent PL intensities (trajectories) with a multichannel scaler (MCS) with an integration time (bin width) of 1 s.
cpl-34-7-077801-fig1.png
Fig. 1. (Color online) (a) Microscopic image of monolayer (1L) and bilayer (2L) WSe$_{2}$ flakes. (b) PL spectra of both 1L and 2L WSe$_{2}$ at room temperature. (c) Schematic drawing of the core component of the DAC device. (d) PL spectra of the monolayer WSe$_{2}$ measured at a temperature of 20 K at zero pressure (black line) and 2.4 GPa (red line). (e) Excitation power dependence of PL spectra at 0.2 GPa for another sample. With increasing the excitation power, peak 1 is gradually saturated, as indicated by dashed gray lines.
It is known that typical PL peaks in monolayer WSe$_{2}$ at low temperature refer to 2D neutral exciton (X$^{0}$), 2D charge exciton (X$^{-}$) and defect-related L excitons, as shown in Fig. 1(d) at zero pressure.[6] When pressure is applied and increased, 2D exciton peak energies show blue-shift and corresponding peak intensities decrease rapidly and turn to become too weak to detect.[28,29] By contrast, many discrete emission lines emerge, as shown in Fig. 1(d) at 2.4 GPa. Here the full width at half maximum (FWHM) of the narrower one is approximately 0.3 nm. In our previous work, these discrete lines have been identified with single photon characteristics, which are generated by deep levels originating from local defects of the flakes, and the corresponding distribution of the discrete lines are changing with different monolayer samples or under pressures.[29] Figure 1(e) presents the excitation power-dependent PL spectra at pressure of 0.2 GPa for another monolayer sample with increasing the excitation power from $P_{0}$ to 27$P_{0}$, where $P_{0}$ is only a few μW. It is shown that discrete lines are predominant at low excitation power. However, with increasing the excitation power, a broadening PL peak or an ensemble of discrete emission lines, emerges due to an increase of the laser excitation spot and the appearance of much more emission centers. It is worth noting that the discrete line of peak 1 turns to be saturate at higher excitation of 9$P_{0}$, showing a quantum dot like characteristic of the power-dependent exciton populations on the quantum levels. Figure 2(a) presents the time evolution of the PL peak intensity at the wavelength of 757.7 nm for lasting 360 s obtained by measuring 180 snapshot spectra, as shown in Fig. 2(b) for the first snapshot spectrum, and each PL spectrum integration time and interval time are 1 s. From the snapshot PL spectra, it clearly demonstrates a remarkable PL intensity fluctuation. The spectrally resolved time traces corresponding to the full PL spectrum (see Fig. 2(b)) summarize the spectral shifting and blinking for discrete emission lines, as shown in Fig. 2(c). Zooming into the time traces of Fig. 2(c) shows spectral shifting and forking, in which the spectral changes are related with the PL peak intensity fluctuation shown in Fig. 2(a). This reveals a correlation of fluorescence intermittency and spectral shifting or forking.
cpl-34-7-077801-fig2.png
Fig. 2. (Color online) (a) Time evolution of the PL peak intensity at a wavelength of 757.7 nm for 0–360 s, taking 180 snapshot spectra with an integration time of each spectrum of 1 s and interval time of 1 s. (b) Full PL spectrum of the first snapshot. (c) Spectrally dispersed image for 0–360 s (180 snapshot spectra), showing PL peak energy wandering. The marked areas are illustrated on an enlarged scale, clearly showing spectral forking and shifting.
Figure 3(a) shows the PL spectra taken at the on-time period (black line) and the off-time period (red line), clearly demonstrating the PL intensity stochastic effect. It is noted that, different from spectral forking or shifting due to the spectral competition shown in Fig. 2(c), hereafter the detected PL line with a wavelength of 786 nm (black line) disappears, no other emission lines are replaced. In this case, it is more likely that the emission centers can be quenched by non-radiative recombination centers in the vicinity of the emission centers. To study this blinking dynamics in detail, time-dependent PL trajectories are taken with an MCS with an integration time of 1 s for lasting 2 h, as shown in Fig. 3(b). The trajectories clearly reveal the blinking effect with turning on-time and off-time abruptly, despite continuous laser excitation. The observed blinking dynamics is comparable with the fluorescence intermittency of the colloid quantum dots.[20,30] In the inset of Fig. 3(b), a typical time tracing PL intensity of the GaAs/AlGaAs quantum dot is presented, showing that the QD emissions and PL measurement setup are stable. Thus the observed fluorescence intermittency in monolayer WSe$_{2}$ reflects the intrinsic characteristics of the emission lines.
cpl-34-7-077801-fig3.png
Fig. 3. (Color online) (a) PL spectra taken at on-time (black line) and off-time (red line) periods. (b) PL intensity trajectory of peak 786.3 nm marked in (a). Integration time is 1 s/bin and a threshold level of 500 count/s is set for off/on time statistical data analysis. Inset: GaAs quantum dot time tracing PL intensity. (c) and (d) Log-log plots of the experimental off-time/on-time events with the slopes of 1.38 and 1.4, respectively.
Note that the typical models of the power-law, Efros–Rosen and truncated power-law distributions are proposed for understanding the fluorescence intermittency dynamics. The log-log plots can provide a more quantitatively useful representation of this on-time/off-time statistical data. These data are shown in Figs. 3(c) and 3(d) for both off and on distributions considering the threshold level of 500 count/s (see the red dotted lines) and reveal the remarkably linear characteristic of an inverse power-law $P_{\rm on/off} (t)\sim t^{-m}$. Results from a linear least-square fit to the data are shown by solid lines, which also reveal slopes representing best estimates of the inverse power law distributions with the exponents of 1.38(0.08) for off-time and 1.4(0.1) for on-time. Similar power law exponent characteristics of fluorescence intermittency were observed and the exponent ranges are from 1.2 to 2.0.[22] Here the exponent value of on-time/off-time is closer to 3/2, which was also predicted based on a model of a multiple stochastic two-level system.[18,31,32] It is found that in most cases the fluorescence intermittency emerges as the emission peak of single discrete lines has a broadening FWHM, as noted in Figs. 2(b) and 3(a) with the FWHM of 1 nm ($\sim$2 meV) and 4.3 nm ($\sim$8.7 meV), respectively. In this case, spectral diffusion under illumination is relevant to a fluctuating environment, leading to an increase of the FWHM from a few μeV to a few meV. From the experimental findings shown in Figs. 2 and 3, we can provide representative schematic diagrams in Fig. 4 for understanding the fluorescence intermittency processes occurring in monolayer WSe$_{2}$. Many discrete emission lines corresponding to the discrete optical transition levels were proved easier to be observed as the pressure is applied.[29] Thus two competing activation recombination centers may be responsible for spectral shifting or the forking effect as Bohr predicted a quantum jump of electrons between discrete energy levels, as schematically shown in Fig. 4(a). However in Fig. 4(b), non-radiative recombination centers (levels) capture electrons occupying activation levels and the trapped electrons can recombine non-radiatively with a valence-band hole before the next capturing event, leading to an off-time period. If the trap centers become populated, as shown in Fig. 4(c), the electron relaxation process from activation levels to deactivation levels is completely suppressed owing to the Coulomb blockade and on-time period of PL emissions recovers again. All these processes are stochastic events, as observed in many quantum systems.[22,25,33]
cpl-34-7-077801-fig4.png
Fig. 4. (Color online) Schematic diagrams for the fluorescence intermittency processes (a) for the forking state, (b) for off state, and (c) for on state.
In summary, we have observed the blinking effect in monolayer WSe$_{2}$, in which the blinking dynamic is related to intermittent and random electron transfer between two optical activation levels or optical activation and deactivation levels. The former one is responsible for spectral shifting and forking, while the latter one causes fluorescence intermittency. An inverse power law with an exponent value of on-time/off-time closing to a value of 3/2 can be used to well describe the blinking dynamic processes. We find a correlation between the dynamics of fluorescence intermittency and spectral diffusion in multiple radiative or non-radiative centers of the quantum emitters. References Optically active quantum dots in monolayer WSe2Single-photon emission from localized excitons in an atomically thin semiconductorVoltage-controlled quantum light from an atomically thin semiconductorSingle quantum emitters in monolayer semiconductorsSingle photon emitters in exfoliated WSe2 structuresStrain-Induced Spatial and Spectral Isolation of Quantum Emitters in Mono- and Bilayer WSe 2Discrete quantum dot like emitters in monolayer MoSe 2 : Spatial mapping, magneto-optics, and charge tuningNanoscale Positioning of Single-Photon Emitters in Atomically Thin WSe 2Quantum emission from hexagonal boron nitride monolayersCascaded emission of single photons from the biexciton in monolayered WSe2Controlled Single-Photon Emission from a Single Trapped Two-Level AtomDeterministic Generation of Single Photons from One Atom Trapped in a CavitySingle photons on demand from a single molecule at room temperatureA highly efficient single-photon source based on a quantum dot in a photonic nanowireSingle-mode solid-state single photon source based on isolated quantum dots in pillar microcavitiesSingle Photon Emission from Individual Nitrogen Pairs in GaPPhoton antibunching in the fluorescence of individual color centers in diamondModel of Fluorescence Intermittency of Single Colloidal Semiconductor Quantum Dots Using Multiple Recombination CentersNonexponential “blinking” kinetics of single CdSe quantum dots: A universal power law behavior?On?/?off? fluorescence intermittency of single semiconductor quantum dotsFluorescence spectroscopy and spectral diffusion of single impurity molecules in a crystalUniversal emission intermittency in quantum dots, nanorods and nanowiresCorrelation between Fluorescence Intermittency and Spectral Diffusion in Single Semiconductor Quantum DotsRandom Telegraph Signal in the Photoluminescence Intensity of a Single Quantum DotFluorescence Intermittency in Single InP Quantum DotsCorrelated fluorescence blinking in two-dimensional semiconductor heterostructuresUnconventional motional narrowing in the optical spectrum of a semiconductor quantum dotPressure-induced K–Λ crossing in monolayer WSe 2Single Photon Emission from Deep Level Defects in Monolayer WSe2Blinking statistics in single semiconductor nanocrystal quantum dotsRenewal aging as emerging property of phase synchronizationPower Law Blinking Quantum Dots: Stochastic and Physical ModelsTwo types of luminescence blinking revealed by spectroelectrochemistry of single quantum dots
[1] Srivastava A, Sidler M, Allain A V, Lembke D S, Kis A and Imamoglu A 2015 Nat. Nanotechnol. 10 491
[2] Tonndorf P, Schmidt R, Schneider R, Kern J, Buscema M, Steele G A, Castellanos-Gomez A, van der Zant H S J, Michaelis de Vasconcellos S and Bratschitsch R 2015 Optica 2 347
[3] Chakraborty C, Kinnischtzke L, Goodfellow K M, Beams R and Vamivakas A N 2015 Nat. Nanotechnol. 10 507
[4] He Y M, Clark G, Schaibley J R, He Y, Chen M C, Wei Y J, Ding X, Zhang Q, Yao W, Xu X, Lu C Y and Pan J W 2015 Nat. Nanotechnol. 10 497
[5] Koperski M, Nogajewski K, Arora A, Cherkez V, Mallet P, Veuillen J Y, Marcus J, Kossacki P and Potemski M 2015 Nat. Nanotechnol. 10 503
[6] Kumar S, Kaczmarczyk A and Gerardot B D 2015 Nano Lett. 15 7567
[7] Branny A, Wang G, Kumar S, Robert C, Lassagne B, Marie X, Gerardot B D and Urbaszek B 2016 Appl. Phys. Lett. 108 142101
[8] Kern J, Niehues I, Tonndorf P, Schmidt R, Wigger D, Schneider R, Stiehm T, Michaelis de Vasconcellos S, Reiter D E, Kuhn T and Bratschitsch R 2016 Adv. Mater. 28 7101
[9] Tran T T, Bray K, Ford M J, Toth M and Aharonovich I 2015 Nat. Nanotechnol. 11 37
[10] He Y M, Iff O, Lundt N, Baumann V, Davanco M, Srinivasan K, Hofling S and Schneider C 2016 Nat. Commun. 7 13409
[11] Darquié B, Jones M P, Dingjan J, Beugnon J, Bergamini S, Sortais Y, Messin G, Browaeys A and Grangier P 2005 Science 309 454
[12] McKeever J, Boca A, Boozer A, Miller R, Buck J, Kuzmich A and Kimble H 2004 Science 303 1992
[13] Lounis B and Moerner W E 2000 Nature 407 491
[14] Claudon J, Bleuse J, Malik N S, Bazin M, Jaffrennou P, Gregersen N, Sauvan C, Lalanne P and Gérard J M 2010 Nat. Photon. 4 174
[15] Moreau E, Robert I, Gérard J, Abram I, Manin L and Thierry-Mieg V 2001 Appl. Phys. Lett. 79 2865
[16] Ikezawa M, Sakuma Y and Masumoto Y 2007 Jpn. J. Appl. Phys. 46 L871
[17] Brouri R, Beveratos A, Poizat J P and Grangier P 2000 Opt. Lett. 25 1294
[18] Frantsuzov P A, Volkan-Kacso S and Janko B 2009 Phys. Rev. Lett. 103 207402
[19] Kuno M, Fromm D P, Hamann H F, Gallagher A and Nesbitt D J 2000 J. Chem. Phys. 112 3117
[20] Kuno M, Fromm D P, Hamann H F, Gallagher A and Nesbitt D J 2001 J. Chem. Phys. 115 1028
[21] Ambrose W and Moerner W 1991 Nature 349 225
[22] Frantsuzov P, Kuno M, Janko B and Marcus R A 2008 Nat. Phys. 4 519
[23] Neuhauser R G, Shimizu K T, Woo W K, Empedocles S A and Bawendi M G 2000 Phys. Rev. Lett. 85 3301
[24] Efros A L and Rosen M 1997 Phys. Rev. Lett. 78 1110
[25] Kuno M, Fromm D P, Gallagher A, Nesbitt D J, Micic O I and Nozik A J 2001 Nano Lett. 1 557
[26] Xu W, Liu W, Schmidt J F, Zhao W, Lu X, Raab T, Diederichs C, Gao W, Seletskiy D V and Xiong Q 2016 Nature 541 62
[27] Berthelot A, Favero I, Cassabois G, Voisin C, Delalande C, Roussignol P, Ferreira R and Gérard J M 2006 Nat. Phys. 2 759
[28] Ye Y, Dou. X, Ding K, Jiang D, Yang F and Sun B 2016 Nanoscale 8 10843
[29] Ye Y, Dou. X, Ding K, Chen Y, Jiang D, Yang F and Sun B 2016 arXiv:1612.06438 [cond-mat.mtrl-sci]
[30] Shimizu K T, Neuhauser R G, Leatherdale C A, Empedocles S A, Woo W K and Bawendi M G 2001 Phys. Rev. B 63 205316
[31] Bianco S, Geneston E, Grigolini P and Ignaccolo M 2008 Physica A 387 1387
[32] Margolin G, Protasenko V, Kuno M and Barkai E 2005 arXiv:cond-mat/0506512 [cond-mat.stat-mech]
[33] Galland C, Ghosh Y, Steinbruck A, Sykora M, Hollingsworth J A, Klimov V I and Htoon H 2011 Nature 479 203