# Fileset

[【#4293】Mano_2010_Appl._Phys._Express_3_065203.pdf](https://mdr.nims.go.jp/filesets/01d5d37a-662d-4828-845c-b3ec7291ca03/download)

## Creator

[Takaaki Mano](https://orcid.org/0000-0002-6955-260X), Marco Abbarchi, [Takashi Kuroda](https://orcid.org/0000-0001-6445-7673), Brian McSkimming, [Akihiro Ohtake](https://orcid.org/0000-0002-3519-4613), [Kazutaka Mitsuishi](https://orcid.org/0000-0002-9361-4057), [Kazuaki Sakoda](https://orcid.org/0000-0002-5530-3020)

## Rights

https://openpolicyfinder.jisc.ac.uk/id/publication/16994 参照[Creative Commons BY Attribution 4.0 International](https://creativecommons.org/licenses/by/4.0/)

## Other metadata

[Self-Assembly of Symmetric GaAs Quantum Dots on (111)A Substrates: Suppression of Fine-Structure Splitting](https://mdr.nims.go.jp/datasets/e477d70f-ee66-4d84-91fa-7aafe1fee22e)

## Fulltext

Self-Assembly of Symmetric GaAs Quantum Dots on (111)A Substrates: Suppression of Fine-Structure SplittingApplied PhysicsExpress      OPEN ACCESSSelf-Assembly of Symmetric GaAs Quantum Dotson (111)A Substrates: Suppression of Fine-Structure SplittingTo cite this article: Takaaki Mano et al 2010 Appl. Phys. Express 3 065203 View the article online for updates and enhancements.You may also likeReducing “slice cross-talk” effect inmetamaterial assisted fast spin-echo MRIE Brui, S Rapacchi, D Bendahan et al.-Spectroscopic Studies of Cation StructureDependence on Lithium-Ion Conductivityin Phosphonium Ionic Liquid ElectrolytesShoki Nawate, Abbas Alshehabi, MitsuhiroMatsumoto et al.-Comparative study between conventionaland mechanical technology on fecalsludge treatment plants (FSTP) inIndonesiaA M Stevani and P Soewondo-This content was downloaded from IP address 144.213.253.16 on 05/12/2025 at 07:42https://doi.org/10.1143/APEX.3.065203https://iopscience.iop.org/article/10.1088/1742-6596/2015/1/012023https://iopscience.iop.org/article/10.1088/1742-6596/2015/1/012023https://iopscience.iop.org/article/10.1149/MA2024-02674495mtgabshttps://iopscience.iop.org/article/10.1149/MA2024-02674495mtgabshttps://iopscience.iop.org/article/10.1149/MA2024-02674495mtgabshttps://iopscience.iop.org/article/10.1088/1755-1315/896/1/012031https://iopscience.iop.org/article/10.1088/1755-1315/896/1/012031https://iopscience.iop.org/article/10.1088/1755-1315/896/1/012031https://iopscience.iop.org/article/10.1088/1755-1315/896/1/012031https://pagead2.googlesyndication.com/pcs/click?xai=AKAOjsslOu7bjP4QGWCRyBc7QAcn-TJk2xaoyaNpfLWTtg6zWKRt-HZENK7afb3JTCsI4W9RWZPeaNbTohVd5FUIj9Vu2DnlVn10Q1WuyHOgJ10eIxYW-JYtgXmPukQIRJ97MkYRrxwXJrslRhtrdmO5EBIaOWwFUiAcmTDfDLzaBUOtKT6WdB-kjy7b9M0mvNlpbEOw5GFBYR9Ey0ULVeK8aCsxqk1D0eQURY8cEPxfih5F4MmIlqbA8uvp6NzSXNTeyazsNq2YyhwrK7INpZbE09nOqZZYdy1wcIvAZkdCaT9TwGe_5wDoiKH6D9Tfr14mxZc-D9uz6ZWTb4GhWRXPoBQTKfiO5GsZpxf_ZEf0csXmxHrsyZnQAkiG&sig=Cg0ArKJSzHfH_mN3UzYA&fbs_aeid=%5Bgw_fbsaeid%5D&adurl=https://www.electrochem.org/249%3Futm_source%3DIOP%26utm_medium%3Dbanners%26utm_campaign%3DIOP_249_abstract_submission%26utm_id%3DIOP%2B249%2BAbstract%2BSubmissionSelf-Assembly of Symmetric GaAs Quantum Dots on (111)A Substrates:Suppression of Fine-Structure SplittingTakaaki Mano, Marco Abbarchi, Takashi Kuroda, Brian McSkimming,Akihiro Ohtake, Kazutaka Mitsuishi, and Kazuaki SakodaNational Institute for Materials Science, 1-2-1 Sengen, Tsukuba, Ibaraki 305-0047, JapanReceived March 24, 2010; accepted May 6, 2010; published online May 28, 2010Great suppression of fine-structure splitting (FSS) is demonstrated in self-assembled GaAs quantum dots (QDs) grown on AlGaAs(111)A surface.Due to the three-fold rotational symmetry of the growth plane, highly symmetric excitons with significantly reduced FSS are achieved. Scanningtunneling microscopy and cross-sectional transmission microscopy demonstrate a laterally symmetric dot shape with abrupt interface. Polarizedphotoluminescence spectra confirm excitonic transition with FSS smaller than �20 �eV, a substantial reduction from that of QDs grown on (100).# 2010 The Japan Society of Applied PhysicsDOI: 10.1143/APEX.3.065203Semiconductor quantum dots (QDs) are attractingmuch interest because of their potential applicationin entangled photon sources, which would serve as akey unit for quantum information technology. The genera-tion of entangled photons in the QDs relies on the twin-photon emission associated with the transition cascade fromthe biexciton state to the exciton state, and to the groundstate. Thanks to the spinor nature of excitons, this transitionpath becomes two-fold. When these two paths are indis-tinguishable, the two photons become entangled on apolarization basis.1)Many studies have focused on Stranski–Krastanov (SK)growth of QDs on (100) surfaces. However, these QDsnormally exhibit significant fine-structure splitting (FSS) dueto the elongated morphology, strain-induced piezoelectricfield, and intermixing between InAs and GaAs.2) Severalapproaches have been reported for solving these problems,such as high-temperature annealing3) and the application ofexternal fields.4–6) In the first case, however, only smallpercentage of total QDs exhibited small FSS. The latterapproach requires a complicated setup, which is notpractical. It is therefore desirable to establish an alternativetechnique for the self-assembly of symmetric QDs.Very recently, the reduction of FSS using (111) surfacewas predicted theoretically,7,8) and demonstrated experimen-tally.9,10) Note that, via SK growth, it is difficult to obtainQDs on the (111) substrates. Thus, they used patternedsubstrates (inverted pyramids)9) or droplet epitaxy (DE),10)both producing InGaAs QDs on GaAs(111)B. In the formercase, however, highly complex processes, such as electronbeam lithography, are required for preparing the patternedsubstrates. Also, a large number of impurities might beincorporated during these processes, which would shortenthe carrier lifetime. In the latter case, they applied DE tothe lattice-mismatched system, implying the lowering ofmorphological quality compared to the lattice-matchedcase.11)In this study, we demonstrate the self-assembly of GaAsQDs on AlGaAs(111)A surfaces via DE.12–14) In the caseof DE in the lattice-matched GaAs/AlGaAs system, theformation of strain-free, nearly pure GaAs QDs was realized.Owing to the three-fold rotational symmetry of the growthplane, well-defined, highly symmetric QDs were formed.Moreover, excitons with significantly reduced FSS wereachieved in all the investigated QDs.The sample was grown using a conventional molecularbeam epitaxy system. After the growth of a GaAs bufferand AlGaAs barrier layer on the GaAs(111)A substrateat 500 �C, followed by annealing at 600 �C, Ga dropletswere formed by supplying a 0.45 monolayer (ML) of Ga(0.09ML/s) at 350 �C. Then, the substrate was cooled downto 200 �C, and crystallized into GaAs by supplying an As4flux of 2� 10�6 Torr beam equivalent pressure. The samplewas then annealed at 400 �C (or 500 �C), followed bycapping with an Al0:3Ga0:7As layer. Finally, rapid thermalannealing was performed to improve the optical properties(800 �C for 4min).The structural properties were studied by in vacuoscanning tunneling microscopy (STM),15) and atomic forcemicroscopy (AFM) in an ambient environment. The QDsembedded in AlGaAs were further studied by cross-sectional high angle annular dark field scanning transmis-sion electron microscopy (HAADF-STEM). Optical prop-erties were analyzed by low-temperature photolumines-cence (PL) measurement based on both ensemble andmicro objectives. A continuous wave laser emitting at532 nm was used for excitation. Excitation polarizationwas set to be linear to avoid the Overhauser effect.16) �PLexperiments were performed using a confocal setup withlateral resolution of about 1 �m. The collected PL wasdispersed by polychromator and detected by chargecoupled device (CCD) camera with spectral resolution of130 �eV.Figure 1(a) shows an AFM image of bare GaAs QDsbefore capping. Well-defined QDs are present with a spatialdensity of 2:5� 1010 cm�2. They exhibit a disk-like shapewith a base diameter of 38 nm and a height of 1.0 nm onaverage. The base size shows a rather uniform distribution(17%), while the height is highly distributed (56%). This factsuggests that carrier quantization and its energy distributionare governed by vertical confinement along the growthdirection.Figure 1(b) summarizes the cross-sectional AFM profilesof three QDs differing in size. For each QD, the crosssections along two orthogonal directions ([211] and ½01�1�)are of identical shape. Such isotropic feature is in starkcontrast to QDs previously grown on the (100) surface;those were elongated by 5–20% along the ½1�10� direction.14)Thus, QD symmetry was substantially improved by usingthe (111)A substrate.Applied Physics Express 3 (2010) 065203065203-1 # 2010 The Japan Society of Applied PhysicsOriginal content from this work may be used under the terms of the Creative Commons Attribution 4.0 licence.Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.http://dx.doi.org/10.1143/APEX.3.065203http://dx.doi.org/10.1143/APEX.3.065203http://dx.doi.org/10.1143/APEX.3.065203http://dx.doi.org/10.1143/APEX.3.065203http://creativecommons.org/licenses/by/4.0/Note that in the DE growth, Ga droplets have circularsymmetry independent of their bottom surface due to theloss of bonding at the liquid-solid interface. Elongation musttake place at the crystallization and annealing stages, wherethe final QD shape is determined by the complex interplaybetween Ga flow (from droplets) and As flow (from As4flux), which is quite sensitive to the surface anisotropy. Forthe (111)A plane, on the other hand, equivalent directionsappear with respect to every 120� rotation, thus removingin-plane anisotropy in an ideal case.Figures 1(c) and 1(d) show in vacuo STM images ofGaAs QDs. Again, elongation is not evidenced, but randomorientation is exhibited, which reflects the microscopicenvironment at the surface. Such orientation randomnessmight have been emphasized by the formation of an atomicstep/terrace structure at the side surface, rather than theformation of preferential facets. The morphology is nearlytruncated pyramid with an atomically-flat top.Figures 1(e) and 1(f) present cross-sectional STEMimages of a GaAs QD embedded in AlGaAs. A buried QDwith truncated shape is again visible. The presence of a clearinterface suggests an abrupt boundary between the matrixand QDs, free from intermixing. There is no indication ofdefects, supporting high crystal quality. We also note theabsence of a two-dimensional GaAs wetting layer under-neath the QDs. This is consistent with the present growthconditions where the nominal amount of Ga deposition was0.45ML, much less than 1ML.Figure 2 shows a low-temperature PL spectrum of theensemble of GaAs QDs. High-yield PL was observed at awavelength of 720 nm (1.72 eV) with 180meV FWHM. ThePL spectrum consists of multiple peaks, rather than a broadband. Such spectral signature was independent of excitationintensity (not shown) and thus not ascribed to the state-filling of QD levels. We attribute the spectral multiplet toground state emissions from different GaAs QD familieswhose heights vary by a ML step, reflecting a flat shape withatomically-smoothed top, as was proven in Figs. 1(c) and1(e). The vertical lines plotted in Fig. 2 represent thetransition energies calculated within the framework of theeffective mass model.17) In this calculation we assumed atruncated cone shape with constant top and bottom diameters(15 and 38 nm, respectively, as deduced by AFM analysis),and changed the height by a ML step (0.32 nm). As shown inFig. 2, this model perfectly reproduces the spectral multi-plets, thus, confirming the validity of our model.Figure 3(a) shows an example of a �PL spectrum in the(111)A GaAs QDs.18) It consists of several sharp peaks,for which the linewidth, limited by the resolution of ourequipment, is less than 50 �eV. Figure 3(b) displays thedependence of the energy of two representative lines, whichare denoted by X and T in Fig. 3(a), as a function of theangle of analyzer polarization. By fitting the data with asinusoidal dependence we extract the amplitude of theoscillation (which represents the FSS) and its phase (the PLpolarization axis). It was found that X showed a significantpolarization-dependent shift having an amplitude of 27 �eV,while T did not exhibit a shift. The observation of sinusoidaldependence naturally led us to attribute X to a neutralexciton that gave rise to anisotropic FSS, and T to a trion.Note that all PL lines were categorized as either of twogroups, one with polarization dependence (like X), andthe other without (like T), allowing FSS statistics to becharacterized.Figure 3(c) presents a comparison between the absolutevalue of the FSS [as determined by the analysis shownin Fig. 3(b)] in GaAs(111)A QDs and that in (100) QDsreported as a function of the corresponding X emissionenergy. The latter series was previously reported.14,19,20)For (111)A, the magnitude of FSS was 17 �eV on average.Note that this value must be overestimated, because QDswith FSS smaller than the fit accuracy (�10 �eV) werenot included in our statistical analysis, even though, themagnitude of FSS in (111)A was much smaller than that of(100), for which the average was 54 �eV. This is the directconsequence of the improved symmetry of QDs grown on(111)A with respect to the (100) case.Height (nm)0.01.02.00.01.02.0-50 -25 0 25 500.01.02.0[211][011]Lateral distance (nm)100 nm (a)10 nm 5 nm10 nm 5 nm(b)(c) (d)(e) (f)Fig. 1. (a) AFM image of GaAs QDs formed on AlGaAs(111)A surface.The surface QDs were annealed at 400 �C for 10min. Scan area is250� 500 nm2. (b) Cross-sectional AFM profiles of three QDs differingin size along the [211] and ½01�1� directions. (c, d) In vacuo STM imagesof GaAs QDs formed on GaAs(111)A surface. (e, f) Cross-sectionalHAADF-STEM images of GaAs QD [shown in (a)] capped with AlGaAs.PL intensity (arb. unit)6 K12 ML 2 MLHeight38 nm15 nmHeight1.6 1.7 1.8 1.9 2.0Energy (eV)Fig. 2. PL spectrum of the ensemble of GaAs QDs, measured at6K. The vertical lines show the calculated energy levels, assuming atruncated cone shape with constant top (15 nm) and bottom (38 nm)diameters.T. Mano et al.Appl. Phys. Express 3 (2010) 065203065203-2 # 2010 The Japan Society of Applied PhysicsFigure 3(d) shows the dependence of the linear polariza-tion axis, which was determined by the polar plot analysis[Fig. 3(b)], as a function of X emission energy. The (111)AQDs showed completely disordered behavior of thepolarization axis, confirming the absence of preferentialin-plane directions. On the other hand, in the (100) QDsthe polarization axis was mostly aligned along the ½1�10�direction, while it became randomly oriented for energieshigher than 1.8 eV. The elongation of QDs on (100) wasenhanced for larger dots, which suffered more fromanisotropic atomic diffusion during growth.14) Thus, evenfor (100), sufficiently small QDs would possibly givenegligible FSS, which has been confirmed in In(Ga)As/GaAs QDs grown via the SK method.5) In contrast, (111)AQDs present the complete absence of preferential polariza-tion axes, suggesting a much higher probability of findingQDs with zero FSS in a wide range of wavelengths. Suchcharacteristics are highly favorable for QDs application toentangled photon sources.In conclusion, we demonstrated highly symmetric andstrain-free GaAs QDs grown on a (111)A substrate. Theabsence of elongation and that of the facet formation wereconfirmed by morphological analysis. FSS was substantiallyreduced in comparison with the QDs grown on (100). Webelieve that the present system assured the removal ofevery macroscopic origin that induced QD asymmetry,while residual FSS due to microscopic randomness could beremoved by post-selection procedures.Acknowledgment This work was partially supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports,Science and Technology of Japan.1) O. Benson, C. Santori, M. Pelton, and Y. Yamamoto: Phys. Rev. Lett.84 (2000) 2513.2) R. Seguin, A. Schliwa, S. Rodt, K. Pötschke, U. W. Pohl, and D.Bimberg: Phys. Rev. Lett. 95 (2005) 257402.3) W. Langbein, P. Borri, U. Woggon, V. Stavarache, D. Reuter, andA. D. Wieck: Phys. Rev. B 69 (2004) 161301(R).4) K. Kowalik, O. Krebs, A. Lemaı̂tre, S. Laurent, P. Senellart, P. Voisin,and J. A. Gaj: Appl. Phys. Lett. 86 (2005) 041907.5) R. M. Stevenson, R. J. Young, P. Atkinson, K. Cooper, D. A. Ritchie,and A. J. Shields: Nature (London) 439 (2006) 179.6) S. Seidl, M. Kroner, A. Högele, K. Karrai, R. J. Warburton, A.Badolato, and P. M. Petroff: Appl. Phys. Lett. 88 (2006) 203113.7) R. Singh and G. Bester: Phys. Rev. Lett. 103 (2009) 063601.8) A. Schliwa, M. Winkelnkemper, A. Lochmann, E. Stock, and D.Bimberg: Phys. Rev. B 80 (2009) 161307(R).9) A. Mohan, M. Felici, P. Gallo, B. Dwir, A. Rudra, J. Faist, and E.Kapon: Nat. Photonics 4 (2010) 302.10) E. Stock, T. Warming, I. Ostapenko, S. Rodt, A. Schliwa, J. A.Töfflinger, A. Lochmann, A. I. Toropov, S. A. Moshchenko, D. V.Dmitriev, V. A. Haisler, and D. Bimberg: Appl. Phys. Lett. 96 (2010)093112.11) T. Noda and T. Mano: Appl. Surf. Sci. 254 (2008) 7777.12) N. Koguchi, S. Takahashi, and T. Chikyow: J. Cryst. Growth 111(1991) 688.13) T. Mano, T. Kuroda, S. Sanuginetti, T. Ochiai, T. Tateno, J. S. Kim, T.Noda, M. Kawabe, K. Sakoda, G. Kido, and N. Koguchi: Nano Lett. 5(2005) 425.14) M. Abbarchi, C. A. Mastrandrea, T. Kuroda, T. Mano, K. Sakoda, N.Koguchi, S. Sanguinetti, A. Vinattieri, and M. Gurioli: Phys. Rev. B78 (2008) 125321.15) A. Ohtake: Surf. Sci. Rep. 63 (2008) 295.16) T. Belhadj, T. Kuroda, C.-M. Simon, T. Amand, T. Mano, K. Sakoda,N. Koguchi, X. Marie, and B. Urbaszek: Phys. Rev. B 78 (2008)205325.17) T. Kuroda, T. Mano, T. Ochiai, S. Sanguinetti, K. Sakoda, G. Kido,and N. Koguchi: Phys. Rev. B 72 (2005) 205301.18) For single QD studies, we prepared a sample with low QD density,where the Ga droplets were formed by supplying 0.043ML of Ga(0.009ML/s) at 400 �C. The density of QDs was 8:5� 108 cm�2 asshown in the inset of Fig. 3(a).19) M. Abbarchi, T. Kuroda, T. Mano, K. Sakoda, and M. Gurioli: Phys.Rev. B 81 (2010) 35334.20) J. D. Plumhof, V. Křápek, L. Wang, A. Schliwa, D. Bimberg, A.Rastelli, and O. G. Schmidt: Phys. Rev. B 81 (2010) 121309(R).(111)APolarization angle (degrees)(100) (a)|FSS| (µeV)X emission energy (meV)02010-10-1020100Energy shift (µeV)04590180225270315  XT(b)1736 1738 1740 1742 1744PL intensity (arb. unit)Photon energy (meV)6 KTX100 nmX emission energy (meV)045901351801700 1800 1900045901351801700 1800 1900(c) (d)(111)A(100) 020406080100120Fig. 3. (a) Typical PL spectrum of a single GaAs QD at 6K. (b) PLenergy shift as a function of the detected polarization angle for X (squaresin black) and T (triangles in green) of the QD shown in (a). The dottedblack (green) line is a sinusoidal fit to X (T) energy shift. (c) Summary ofthe FSS of QDs grown on (100) (squares in blue) and on (111)A (circles inred). The lines are guides for the eye. (d) Top (bottom) panel displays thepolarization axes of the QDs grown on the (111)A [(100)] surface as afunction of X emission energy (90� corresponds to the [211] (½1�10�) in-plane direction). The inset of (a) shows an AFM image of low-densityGaAs QDs used for the �PL study.T. Mano et al.Appl. Phys. Express 3 (2010) 065203065203-3 # 2010 The Japan Society of Applied Physicshttp://dx.doi.org/10.1103/PhysRevLett.84.2513http://dx.doi.org/10.1103/PhysRevLett.84.2513http://dx.doi.org/10.1103/PhysRevLett.95.257402http://dx.doi.org/10.1103/PhysRevB.69.161301http://dx.doi.org/10.1063/1.1855409http://dx.doi.org/10.1038/nature04446http://dx.doi.org/10.1063/1.2204843http://dx.doi.org/10.1103/PhysRevLett.103.063601http://dx.doi.org/10.1103/PhysRevB.80.161307http://dx.doi.org/10.1038/nphoton.2010.2http://dx.doi.org/10.1063/1.3337097http://dx.doi.org/10.1063/1.3337097http://dx.doi.org/10.1016/j.apsusc.2008.02.026http://dx.doi.org/10.1016/0022-0248(91)91064-Hhttp://dx.doi.org/10.1016/0022-0248(91)91064-Hhttp://dx.doi.org/10.1021/nl048192+http://dx.doi.org/10.1021/nl048192+http://dx.doi.org/10.1103/PhysRevB.78.125321http://dx.doi.org/10.1103/PhysRevB.78.125321http://dx.doi.org/10.1016/j.surfrep.2008.03.001http://dx.doi.org/10.1103/PhysRevB.78.205325http://dx.doi.org/10.1103/PhysRevB.78.205325http://dx.doi.org/10.1103/PhysRevB.72.205301http://dx.doi.org/10.1103/PhysRevB.81.035334http://dx.doi.org/10.1103/PhysRevB.81.035334http://dx.doi.org/10.1103/PhysRevB.81.121309