# Fileset

[04-出版-Open-APEX.pdf](https://mdr.nims.go.jp/filesets/ba62ea81-14ac-451b-ac84-67dc5e6883cb/download)

## Creator

[Neul Ha](https://orcid.org/0000-0002-7695-2193), [Takaaki Mano](https://orcid.org/0000-0002-6955-260X), Samuel Dubos, [Takashi Kuroda](https://orcid.org/0000-0001-6445-7673), [Yoshiki Sakuma](https://orcid.org/0000-0001-6804-7217), [Kazuaki Sakoda](https://orcid.org/0000-0002-5530-3020)

## Rights

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

## Other metadata

[Single photon emission from droplet epitaxial quantum dots in the standard telecom window around a wavelength of 1.55 μm](https://mdr.nims.go.jp/datasets/b923d763-d926-4843-98a1-2d2a450b16e3)

## Fulltext

Single photon emission from droplet epitaxial quantum dots in the standard telecom window around a wavelength of 1.55 μmApplied Physics Express     LETTER • OPEN ACCESSSingle photon emission from droplet epitaxial quantum dots in thestandard telecom window around a wavelength of 1.55 μmTo cite this article: Neul Ha et al 2020 Appl. Phys. Express 13 025002 View the article online for updates and enhancements.This content was downloaded from IP address 144.213.253.16 on 17/02/2020 at 08:52https://doi.org/10.35848/1882-0786/ab6e0fSingle photon emission from droplet epitaxial quantum dots in the standardtelecom window around a wavelength of 1.55μmNeul Ha*, Takaaki Mano , Samuel Dubos, Takashi Kuroda* , Yoshiki Sakuma , and Kazuaki SakodaNational Institute for Materials Science, 1-1 Namiki, Tsukuba 305-0044, Japan*E-mail: ha.neul@nims.go.jp; kuroda.takashi@nims.go.jpReceived December 14, 2019; revised January 17, 2020; accepted January 20, 2020; published online February 3, 2020We study the luminescence dynamics of telecom wavelength InAs quantum dots grown on InP(111)A by droplet epitaxy. The use of the ternaryalloy InAlGaAs as a barrier material leads to photon emission in the 1.55 μm telecom C-band. The luminescence decay is well described in termsof the theoretical interband transition strength without the impact of nonradiative recombination. The intensity autocorrelation function shows clearanti-bunching photon statistics. The results suggest that our quantum dots are useful for constructing a practical source of single photons andquantum entangled photon pairs. © 2020 The Japan Society of Applied PhysicsAsource of single photons and quantum entangledphoton pairs is a key device in vast quantumtechnologies. Semiconductor quantum dots are ex-pected to serve as photon sources that can be operated veryefficiently and deterministically. Numerous efforts havealready been made to develop a practical quantum dot photonsource. However, photon emission in the standard telecomband, particularly around a wavelength of 1.55 μm, which isthe maximum transmission window of silica optical fibers, isa material challenge. Careful growth optimization is requiredto achieve a 1.55 μm emission.1–10) Nevertheless, the well-known quantum dot growth based on the Stranski–Krastanowmode leads to an asymmetric dot shape, which is notfavorable for entangled pair generation.The problem is ideally solved by using droplet epitaxy,which offers considerable freedom regarding the choice ofmaterials and substrates.11) The application of a C3v sym-metric (111)A surface to the growth substrate results in thecreation of almost perfectly symmetric quantum dots, whichcan work in both the visible wavelength region12,13) and theinfrared telecom wavelength region.14,15) Recently, the emis-sion wavelength has been extended beyond 1.5 μm for InAsdots embedded in InAlGaAs on InP(111)A.16) However, theprevious samples were not sufficiently optimized: the dotdensity was too high for a single quantum dot to be isolatedusing standard micro optics. Moreover, the dot size distribu-tion is relatively large so that careful dot selection is requiredto find a dot that emits at 1.55 μm.Here, we extend the droplet epitaxy scheme to achieve apurely 1.55 μm photon emission. We introduce the hightemperature crystallization protocol, which has recently beenapplied to the GaAs material system,17,18) to the InAs/InPmaterial system in order to improve the dot morphologyproperty. The use of a state-of-the-art superconductingphoton detector, together with an efficient dot sample, allowsus to investigate single photon emission dynamics in thestandard telecom C-band.The quantum dot sample is grown on Fe-doped semi-insulating InP(111)A using a solid source molecular beamepitaxy machine. After depositing a lattice-matched In0.52Al0.12 Ga0.36 As barrier with a thickness of 200 nm at 490 °C,we grow 0.5 of a monolayer (ML) of InAs at 490 °C.We then supply 0.25ML of indium at a growth rate of0.16 ML s−1 at 400 °C, which leads to the formation ofindium droplets on InAlGaAs(111)A. Next, an As4 flux(9× 10−5 Torr) is supplied at 400 °C to crystallize InAsdots from indium droplets. After in vacuo annealing at 450 °C for 5 min, the dots are capped by In0.52 Al0.12 Ga0.36 Aswith a thickness of 100 nm. A notable point in this sequenceis the crystallization temperature, which we set higher thanthat of the standard protocol. The small diffusion length ofgroup-III adatoms suppresses the transformation from dots tolayers even at 400 °C, leading to the formation of dots with ahigh crystalline quality.We measure the stationary- and time-resolved responses ofphotoluminescence from single InAs dots. For the stationarystudy we use a semiconductor laser diode at a wavelength of980 nm as a cw excitation source. For the time-resolved studywe use a ps mode-locked titanium sapphire laser whosewavelength is tuned to 900 nm as a pulsed source. The laserlight is focused on the sample using a microscope objectivelens with a numerical aperture of 0.65 (OlympusLCPlan50xIR). The luminescence signal is collected by thesame lens, passed through a dichroic beam splitter, andcoupled to a single mode optical fiber that has a mode fielddiameter of 9 μm at a wavelength of 1.3 μm. The fiber outputis fed into a 50 cm spectrometer that consists of a 600line mm−1 grating.The luminescence signal is spectrally analyzed using acooled InGaAs photodiode array (Andor iDus 491) andtemporally resolved using a superconducting single photondetector (Single Quantum Eos) with a fast-response time-to-digital converter (PicoQuant PH300). The polarization stateof the input light is adjusted to maximize the detectionefficiency. Note that we attach a high-index hemisphericallens (n= 2) to the sample surface to increase the lightcollection efficiency.19) Thanks to the efficient setup togetherwith the bright sample we achieve count rates up to 50 kHz.The sample is cooled using a closed cycle cryostat. All theexperiments are performed at 8 K unless otherwise noted.The inset in Fig. 1(b) shows an atomic force microscopeimage of the quantum dot surface. It reveals the formation ofnearly circular dots without significant elongation. The dotisotropy arises due to the use of the C3v symmetric {111}surface as a growth substrate. The quantum dots have a disk-like shape with a diameter of 48 ± 8 nm and a height ofContent from this work may be used under the terms of the Creative Commons Attribution 4.0 license. Any further distribution of thiswork must maintain attribution to the author(s) and the title of the work, journal citation and DOI.© 2020 The Japan Society of Applied Physics025002-1Applied Physics Express 13, 025002 (2020) LETTERhttps://doi.org/10.35848/1882-0786/ab6e0fhttps://crossmark.crossref.org/dialog/?doi=10.35848/1882-0786/ab6e0f&domain=pdf&date_stamp=2020-02-03https://orcid.org/0000-0002-6955-260Xhttps://orcid.org/0000-0002-6955-260Xhttps://orcid.org/0000-0001-6445-7673https://orcid.org/0000-0001-6445-7673https://orcid.org/0000-0001-6804-7217https://orcid.org/0000-0001-6804-7217mailto:ha.neul@nims.go.jpmailto:kuroda.takashi@nims.go.jphttps://creativecommons.org/licenses/by/4.0/https://doi.org/10.35848/1882-0786/ab6e0f1.9 ± 0.3 nm. These shape parameters are similar to those ofdroplet epitaxy GaAs dots on AlGaAs(111)A.12) The dotdensity is ∼6× 108 cm−2 thus making it easy to isolate asingle dot without any post-growth processing such as thefabrication of small mesas or apertures.Figure 1(a) shows the photoluminescence spectra of thequantum dot ensemble. They were measured using standardlong focus optics. For low excitation, the spectrum has aGaussian-like single peak centered at a wavelength of1550 nm. Its full width at half maximum is ∼100 nm, whichis more than two times smaller than that of our previoussample targeting a 1.55 μm emission.16) Hence, the majorityof the dots in the present sample can emit in the telecom C-band. For high excitation, the spectrum shows another broadband that originates from the excited states, as well as anadditional peak at 1350 nm due to carrier recombination inthe barrier layer.Figure 1(b) shows a typical luminescence spectrum for asingle isolated dot. The observed split lines are attributed toneutral excitons (X, 1572.4 nm), positively charged excitons(X+, 1573.8 nm), and neutral biexcitons (XX, 1578.6 nm).The spectral assignment was based on the large numberstatistics of the multiexciton binding energies in InAs/InAlAsdroplet dots.15) Note that the present sample frequently showsX+, but rarely shows a negatively charged line. This impliesthat our sample is slightly p-doped possibly due to theresidual presence of carbon impurities. (The hypothesis alsoaccounts for the observation of relatively weak XX signalseven for saturation conditions in Fig. 1(b): with opticalinjections, charge neutral impurities tend to be ionized, thenholes are more injected into dots, resulting in the lowerpopulation of neutral XX).Figure 2 shows the luminescence decay signal of the X lineshown in Fig. 1(b) after short pulsed excitation. The decaysignal of a GaAs quantum dot embedded in Al0.3 Ga0.7 As(111)A, which we studied previously,20) is also shown forcomparison. Both decay curves are well approximated bystraight lines in the semilogarithmic plot, which implies thatthey follow single exponent functions. The decay timeconstant of the InAs dot is estimated to be 1.56 ns, whichis significantly longer than that of the GaAs dot (0.56 ns).The large difference in the luminescence decay times arisesdue to the frequency dispersion of the photonic density ofstates, as discussed below.The spontaneous emission rate for atomic transitions(Einstein’s A coefficient) is expressed as21)tmmwp=- ne pm c3, 1102 20 02 3⎛⎝⎜⎞⎠⎟∣ ¯ ∣ ( )where ò0 and ò (μ0 and μ) are the vacuum and relativepermittivities (permeabilities), respectively, n is the refractiveindex given by m m 0 0 , ω is the angular frequency ofemitted light, and p is the matrix element of the momentumoperator = - ip . The above formula was deducedby including the interaction Hamiltonian =Hint+ e mA p p A 2 0( · · )  , where A is the quantized vectorFig. 1. (Color online) (a) The luminescence spectra of a large ensemble of InAs quantum dots in In0.52 Al0.12 Ga0.36 As /InP(111)A at 12 K at differentexcitation powers. (b) The luminescence spectra of a single isolated InAs dot with a cw excitation of 40 nW. We focus on this dot in our time-resolved study.The inset shows an atomic force microscope image of the dot surface.Fig. 2. (Color online) The luminescence decay of the neutral exciton lineof an InAs quantum dot emitting at a wavelength of 1572 nm with anexcitation power of 7 nW (red circles). The black square line is theluminescence decay of the neutral exciton line of a GaAs quantum dotembedded in Al0.3 Ga0.7 As(111)A, with an emission wavelength of 674 nm.The inset shows the decay rate dependence on the emission frequency for themeasured GaAs and InAs dots, and the theoretical prediction of Eq. (1) forn = 3.5 and 2p2/m0 = 20 eV.© 2020 The Japan Society of Applied Physics025002-2Appl. Phys. Express 13, 025002 (2020) N. Ha et al.potential, and the three-dimensional photonic density of statesρ(ω)=Vω2/π2c3, where V is the normalization volume, to theFermi’s golden rule.Note that the matrix element p serves as a band mixingsource in the k · p perturbation theory, and it is more or lessconstant for most group-IV, III–V, and II–VI semiconduc-tors, with the Kane energy 2p2/m0≈ 20 eV.22) Consequently,the material dependence in Eq. (1) appears only in the ω-proportional factor if we assume a constant n value. The solidline in the inset of Fig. 2 is the model dependence of theemission decay rate on the photon frequency, where weassume that n= 3.5 and 2p2/m0= 20 eV. The ω-lineardependence agrees with the measured decay rates of telecomwavelength InAs dots and visible wavelength GaAs dots.Thus, the radiative process of our quantum dots is purelydescribed by the atomic description in Eq. (1) free from theimpact of nonradiative recombination.Figure 3(a) shows the luminescence transients for differentexcitation powers. The signal observed at 7 nW, i.e. thelowest excitation condition, is identical to that shown by thesemilogarithmic plot in Fig. 2. Hence, the decay curvefollows a single exponent. When the excitation power isincreased to 25 nW, the signal deviates from a monotonicdecay, and reveals a significant rise after t= 0. The risesignature is more evident for 80 nW, where the intensitymaximum is substantially delayed by more than 1 ns afterexcitation. The observed power-dependent evolution arisesdue to the multiexciton relaxation cascade. We analyze thesingle exciton spectral line, which is generated only when asingle electron and hole pair remains in the dot, following therecombination of all the other pairs. The same luminescencebehavior is reported in Refs. 23,24. Figure 3(b) shows thenumerical simulation results, where we assume a Poissoniandistribution for the initial number of excitons.25) For simpli-city, we deal only with the cascade evolution from XX to X,and we assume that XX decays twice as fast as X, i.e. XXdecays like noninteracting two excitons. The simple modelreproduces the measured behavior qualitatively.Figure 4 shows the intensity autocorrelation function g(2)(t)of the X luminescence line in Fig. 1(b). Here, we adopt theHanbury Brown and Twiss setup to measure the coincidenceof two photons as a function of delay time.26) The sample isilluminated by cw light. With low excitation at 7 nW, thesignal shows a clear antibunching dip, which yields nearly noprobability of emitting two photons at the same time. As thedelay time is increased the signal recovers from ∼0 to theequilibrium value, i.e. the accidental coincidence number,with a time constant given by the luminescence decay timeanalyzed in Fig. 2 (τ= 1.56 ns). The dashed line shows amodel function t- - t1 exp( ∣ ∣ ), which agrees with theobserved signal. With increasing excitation power, the dipwidth is observed to decrease, and the signal quickly recoversto the equilibrium level. This is due to the acceleration of theX population recovery for strong excitations.27) Note thatseveral researchers have reported the emergence of positivebunching correlations superimposed on the antibunchingdip.28–30) However, we do not observe such a signaturepossibly due to the lower population of neutral XX in oursample. Nevertheless, the value of g(2)(0) is lower than theclassical limit of 0.5 at least over the present excitation range,supporting the emission of single photons from this dot.Fig. 3. (Color online) (a) Comparison of the transient responses of theluminescence signals of the X line for different excitation powers. (b) Modelcalculation results for the luminescence transients. N is the average initialexciton number.Fig. 4. (Color online) The intensity autocorrelation function of the X line for different excitation powers. The coincidence number was integrated with a timebin of 256 ps for 3600 min (7 nW), 80 min (25 nW), and 20 min (80 nW). The simulation results are also plotted by the gray broken line.© 2020 The Japan Society of Applied Physics025002-3Appl. Phys. Express 13, 025002 (2020) N. Ha et al.In conclusion, we used droplet epitaxy to fabricate InAsquantum dots that emit single photons at wavelengths around1.55 μm. Careful growth optimization enabled us to reducethe dot size distribution, and so the majority of the dots couldemit photons in the telecom C-band. The exciton lifetime wasnearly the same as the theoretically ideal value free from theimpact of nonradiative recombination. The use of a trigonallysymmetric InP(111)A substrate led to the formation of nearlycircular dots. Thus, our dots can be expected to serve asbright single photon and entangled photon pair sources thatwill be useful for practical quantum communication applica-tions.Acknowledgments We acknowledge support from a Grant-in-Aid fromthe Japan Society for the Promotion of Science (Grant number 16H02203), andInnovative Science and Technology Initiative for Security, ATLA, Japan.ORCID iDs Takaaki Mano https://orcid.org/0000-0002-6955-260X Takashi Kuroda https://orcid.org/0000-0001-6445-7673 Yoshiki Sakuma https://orcid.org/0000-0001-6804-72171) K. Takemoto, M. Takatsu, S. Hirose, N. Yokoyama, Y. Sakuma, T. Usuki,T. Miyazawa, and Y. Arakawa, J. Appl. Phys. 101, 081720 (2007).2) K. Takemoto, Y. Nambu, T. Miyazawa, Y. Sakuma, T. Yamamoto,S. Yorozu, and Y. Arakawa, Sci. Rep. 5, 14383 (2015).3) T. Miyazawa et al., Appl. Phys. Lett. 109, 132106 (2016).4) M. D. Birowosuto, H. Sumikura, S. Matsuo, H. Taniyama, P. J. vanVeldhoven, R. Nötzel, and M. Notomi, Sci. Rep. 2, 321 (2012).5) J. Skiba-Szymanska et al., Phys. Rev. Appl. 8, 014013 (2017).6) T. Müller, J. Skiba-Szymanska, A. B. Krysa, J. Huwer, M. Felle,M. Anderson, R. M. Stevenson, J. Heffernan, D. A. Ritchie, andA. J. Shields, Nat. Commun. 9, 862 (2018).7) M. Benyoucef, M. Yacob, J. P. Reithmaier, J. Kettler, and P. Michler, Appl.Phys. Lett. 103, 162101 (2013).8) M. Paul, F. Olbrich, J. Höschele, S. Schreier, J. Kettler, S. L. Portalupi,M. Jetter, and P. Michler, Appl. Phys. Lett. 111, 033102 (2017).9) Ł. Dusanowski, M. Syperek, P. Mrowiński, W. Rudno-Rudziński,J. Misiewicz, A. Somers, S. Höfling, M. Kamp, J. P. Reithmaier, and G. Seķ,Appl. Phys. Lett. 105, 021909 (2014).10) Ł. Dusanowski, M. Syperek, J. Misiewicz, A. Somers, S. Höfling, M. Kamp,J. P. Reithmaier, and G. Seķ, Appl. Phys. Lett. 108, 163108 (2016).11) M. Gurioli, Z. Wang, A. Rastelli, T. Kuroda, and S. Sanguinetti, Nat. Mater.18, 799 (2019).12) T. Mano, M. Abbarchi, T. Kuroda, B. McSkimming, A. Ohtake,K. Mitsuishi, and K. Sakoda, Appl. Phys. Express 3, 065203 (2010).13) N. Ha, T. Mano, T. Kuroda, Y. Sakuma, and K. Sakoda, Appl. Phys. Lett.115, 083106 (2019).14) N. Ha, X. Liu, T. Mano, T. Kuroda, K. Mitsuishi, A. Castellano,S. Sanguinetti, T. Noda, Y. Sakuma, and K. Sakoda, Appl. Phys. Lett. 104,143106 (2014).15) X. Liu, N. Ha, H. Nakajima, T. Mano, T. Kuroda, B. Urbaszek, H. Kumano,I. Suemune, Y. Sakuma, and K. Sakoda, Phys. Rev. B 90, 081301(2014).16) N. Ha, T. Mano, Y.-N. Wu, Y.-W. Ou, S.-J. Cheng, Y. Sakuma, K. Sakoda,and T. Kuroda, Appl. Phys. Express 9, 101201 (2016).17) M. Jo, T. Mano, M. Abbarchi, T. Kuroda, Y. Sakuma, and K. Sakoda, Cryst.Growth Des. 12, 1411 (2012).18) F. Basso Basset, S. Bietti, M. Reindl, L. Esposito, A. Fedorov, D. Huber,A. Rastelli, E. Bonera, R. Trotta, and S. Sanguinetti, Nano Lett. 18, 505(2018).19) N. Ha et al., Phys. Rev. B 92, 075306 (2015).20) T. Kuroda et al., Phys. Rev. B 88, 041306 (2013).21) R. Loudon, The Quantum Theory of Light (Clarendon, Oxford, 1973),Chaps. 3 and 8.22) P. Y. Yu and M. Cardona, Fundamentals of Semiconductors (Springer,Berlin, 1996), Chap. 2.23) E. Dekel, D. V. Regelman, D. Gershoni, E. Ehrenfreund, W. V. Schoenfeld,and P. M. Petroff, Phys. Rev. B 62, 11038 (2000).24) T. Kuroda, S. Sanguinetti, M. Gurioli, K. Watanabe, F. Minami, andN. Koguchi, Phys. Rev. B 66, 121302 (2002).25) M. Abbarchi, C. Mastrandrea, T. Kuroda, T. Mano, A. Vinattieri, K. Sakoda,and M. Gurioli, J. Appl. Phys. 106, 053504 (2009).26) T. Kuroda et al., Appl. Phys. Express 1, 042001 (2008).27) H. Sasakura, S. Muto, and H. Kumano, Appl. Phys. Express 8, 112002(2015).28) D. V. Regelman, U. Mizrahi, D. Gershoni, E. Ehrenfreund, W.V. Schoenfeld, and P. M. Petroff, Phys. Rev. Lett. 87, 257401 (2001).29) A. Kiraz, S. Fälth, C. Becher, B. Gayral, W. V. Schoenfeld, P. M. Petroff,L. Zhang, E. Hu, and A. Imamoğlu, Phys. Rev. B 65, 161303 (2002).30) S. Odashima, H. Sasakura, H. Nakajima, and H. Kumano, J. Appl. Phys.122, 223104 (2017).© 2020 The Japan Society of Applied Physics025002-4Appl. Phys. Express 13, 025002 (2020) N. Ha et al.https://orcid.org/0000-0002-6955-260Xhttps://orcid.org/0000-0002-6955-260Xhttps://orcid.org/0000-0002-6955-260Xhttps://orcid.org/0000-0002-6955-260Xhttps://orcid.org/0000-0002-6955-260Xhttps://orcid.org/0000-0001-6445-7673https://orcid.org/0000-0001-6445-7673https://orcid.org/0000-0001-6445-7673https://orcid.org/0000-0001-6445-7673https://orcid.org/0000-0001-6445-7673https://orcid.org/0000-0001-6804-7217https://orcid.org/0000-0001-6804-7217https://orcid.org/0000-0001-6804-7217https://doi.org/10.1063/1.2723177https://doi.org/10.1038/srep14383https://doi.org/10.1063/1.4961888https://doi.org/10.1038/srep00321https://doi.org/10.1103/PhysRevApplied.8.014013https://doi.org/10.1038/s41467-018-03251-7https://doi.org/10.1063/1.4825106https://doi.org/10.1063/1.4825106https://doi.org/10.1063/1.4993935https://doi.org/10.1063/1.4890603https://doi.org/10.1063/1.4947448https://doi.org/10.1038/s41563-019-0355-yhttps://doi.org/10.1038/s41563-019-0355-yhttps://doi.org/10.1143/APEX.3.065203https://doi.org/10.1063/1.5103217https://doi.org/10.1063/1.5103217https://doi.org/10.1063/1.4870839https://doi.org/10.1063/1.4870839https://doi.org/10.1103/PhysRevB.90.081301https://doi.org/10.1103/PhysRevB.90.081301https://doi.org/10.7567/APEX.9.101201https://doi.org/10.1021/cg201513mhttps://doi.org/10.1021/cg201513mhttps://doi.org/10.1021/acs.nanolett.7b04472https://doi.org/10.1021/acs.nanolett.7b04472https://doi.org/10.1103/PhysRevB.92.075306https://doi.org/10.1103/PhysRevB.88.041306https://doi.org/10.1103/PhysRevB.62.11038https://doi.org/10.1103/PhysRevB.66.121302https://doi.org/10.1063/1.3197848https://doi.org/10.1143/APEX.1.042001https://doi.org/10.7567/APEX.8.112002https://doi.org/10.7567/APEX.8.112002https://doi.org/10.1103/PhysRevLett.87.257401https://doi.org/10.1103/PhysRevB.65.161303https://doi.org/10.1063/1.4995225https://doi.org/10.1063/1.4995225 Acknowledgments A2