Structural atomic-scale analysis of GaAs/AlGaAs quantum
wires and quantum dots grown by droplet epitaxy on a (311)A
substrate
Citation for published version (APA):
Keizer, J. G., Jo, M., Mano, T., Noda, T., Sakoda, K., & Koenraad, P. M. (2011). Structural atomic-scale analysis of GaAs/AlGaAs quantum wires and quantum dots grown by droplet epitaxy on a (311)A substrate. Applied Physics Letters, 98(19), 193112-1/3. [193112]. https://doi.org/10.1063/1.3589965
DOI:
10.1063/1.3589965 Document status and date: Published: 01/01/2011
Document Version:
Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers)
Please check the document version of this publication:
• A submitted manuscript is the version of the article upon submission and before peer-review. There can be important differences between the submitted version and the official published version of record. People interested in the research are advised to contact the author for the final version of the publication, or visit the DOI to the publisher's website.
• The final author version and the galley proof are versions of the publication after peer review.
• The final published version features the final layout of the paper including the volume, issue and page numbers.
Link to publication
General rights
Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain
• You may freely distribute the URL identifying the publication in the public portal.
If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, please follow below link for the End User Agreement:
www.tue.nl/taverne
Take down policy
If you believe that this document breaches copyright please contact us at:
openaccess@tue.nl
providing details and we will investigate your claim.
Structural atomic-scale analysis of GaAs/AlGaAs quantum wires and
quantum dots grown by droplet epitaxy on a
„311…A substrate
J. G. Keizer,1,a兲M. Jo,2T. Mano,2T. Noda,2K. Sakoda,2and P. M. Koenraad1
1Department of Applied Physics, Eindhoven University of Technology, P.O. Box 513, 5600 MB, Eindhoven, The Netherlands
2National Institute for Materials Science, 1-2-1 Sengen, Tsukuba, Ibaraki 305-0047, Japan
共Received 4 February 2011; accepted 15 April 2011; published online 12 May 2011; publisher error corrected 17 May 2011兲
We report the structural analysis at the atomic scale of GaAs/AlGaAs quantum wires and quantum dots grown by droplet epitaxy on a 共311兲A-oriented substrate. The shape, interfaces, and composition of these nanostructures and their surrounding matrix are investigated. We show that quantum wires can be created by annealing uncapped quantum dots. Substantial interface fluctuations, attributed to interface instability induced by the liquid Ga droplet, are observed. Despite the interface fluctuations, no intermixing of Al was found in either the quantum wires or quantum dots. A wetting layer connecting the quantum dots could not be observed.
© 2011 American Institute of Physics. 关doi:10.1063/1.3589965兴
Lattice-matched quantum dots共QDs兲 can be grown us-ing molecular beam epitaxy workus-ing in the droplet epitaxy 共DE兲 mode. In this technique, liquid droplets of a group III-element are crystallized at low temperature into QDs by the incorporation of group V-elements.1 Traditionally, QDs are grown in Stranski–Krastanow mode, which is strain driven and results in self-assembled strained QDs. Such QDs can show heavy intermixing with the surrounding matrix2 and are usually accompanied by a pronounced wetting layer 共WL兲 that softens the boundary of the QDs.3
Such structural imperfections can obscure the intrinsic properties of QD la-sers. Furthermore, it hinders the linking of experiment with a realistic QD model that predicts the emission properties.4In this respect, QDs grown by DE can provide a much simpler system. Ideally, QDs grown in DE mode are strain free, do not suffer from intermixing, and can be grown with an arbi-trary WL thickness.5 Besides the QD, the quantum wire 共QWR兲 has also attracted great interest for use in optoelec-tronic device applications. As in the case with the QDs, the GaAs/AlGaAs material system potentially offers strain free QWRs with little material intermixing. Traditionally, the techniques used to fabricate QWRs are selective growth on patterned substrates6 and growth on vicinal7 or high-index surfaces.8Here we report a new technique, the formation of QWRs on 共311兲A-oriented substrates by thermally induced redistribution and diffusion of constituent atoms of QDs grown by DE. Structural peculiarities of both the QDs and the QWRs are revealed by cross-sectional scanning tunneling microscopy共X-STM兲.
The sample used in the X-STM measurements consists of a GaAs/AlGaAs QWR layer and a GaAs/AlGaAs QD layer that were grown on an n-doped共311兲A-oriented GaAs substrate by means of DE. First, a 50 nm GaAs layer is grown at 610 ° C, the nominal growth temperature. Next, a 20 nm AlGaAs buffer layer is grown after which the tem-perature is lowered to 200 ° C and the As is evacuated from the growth chamber. Subsequently, 5 monolayers of Ga are deposited resulting in the formation of liquid Ga droplets on
the surface. These droplets are crystallized into GaAs QDs by the application of an As4 flux 共2⫻10−6 Torr beam
equivalent pressure兲. The resulting uncapped QDs are an-nealed at 550 ° C to form QWRs. Next, the QWRs are capped with a 20 nm AlGaAs layer grown at the annealing temperature. After the completion of this capping layer, an GaAs/AlGaAs multiple quantum well 共4⫻兲 is grown at the nominal growth temperature. The whole growth sequence is then repeated with an annealing and overgrowth temperature of 400 ° C, resulting in QDs instead of QWRs.
All X-STM measurements were performed at room tem-perature under UHV conditions共p⬍6⫻10−11 mbar兲 with an Omicron STM-1, TS2 Scanner. The STM was operated in constant current mode on in situ cleaved兵01-1其-surfaces. The X-STM measurements were all done at high negative bias voltages 共V⬇−2.5 V兲. At these tunnel conditions, GaAs 共AlAs兲 rich regions appear bright 共dark兲 in the topographic X-STM images.
Before we start an atomic scale analysis of the nano-structures, we will first investigate the process of QWR formation with atomic force microscopy 共AFM兲. For this purpose two samples with uncapped QDs were grown. In Fig. 1, two AFM images of the uncapped nanostructures after 10 min annealing at共a兲 450 and 共b兲 550 °C are shown. The density and height of the QDs is determined to be ⬇1⫻1011 cm−2 and 2.5 nm, respectively. Upon annealing
a兲Electronic mail: j.g.keizer@tue.nl. FIG. 1. 共Color online兲 2⫻2 m
2AFM images of the uncapped QDs after
10 min annealing at a兲 450 and b兲 550 °C.
APPLIED PHYSICS LETTERS 98, 193112共2011兲
0003-6951/2011/98共19兲/193112/3/$30.00 98, 193112-1 © 2011 American Institute of Physics Downloaded 19 May 2011 to 131.155.110.244. Redistribution subject to AIP license or copyright; see http://apl.aip.org/about/rights_and_permissions
the QDs elongate along the 关2¯33兴-direction. The driving force for this elongation is the strong ansitropic nature of the 共311兲A-surface.8
In the current case the QD density is so high that the QDs start to coalesce to form QWRs that are oriented along the 关2¯33兴-direction. In a previous study of QWR formation on an 共001兲-oriented substrate the achiev-able QD density was found to be too low to achieve QWRs and as a result quantum dashes were formed instead.9 The fact that higher density QDs can be formed on the 共311兲A-surface as compared to the共001兲-surface is attributed to the shorter migration distance of Ga adatoms on the 共311兲A-surface.10,11 The average height, width, and length of the QWRs is found to be 1.9 nm, 30 nm, ⬎200 nm, respec-tively. Low temperature共6 K兲 photoluminesence 共PL兲 mea-surements of the QWRs after capping show that they emit around 700 nm with a linewidth of 48 meV. Furthermore, a polarization anisotropy of 0.18 along the关2¯33兴-direction and perpendicular to the 关01¯1兴-direction was found.12 This last feature is desirable in the fabrication of lasers that use cleaved共011兲-surfaces as Fabry–Pérot mirrors. As an appli-cation of the current QWRs to optoelectronic devices, we recently demonstrated electrically pumped lasing in a GaAs/ AlGaAs QWR laser diode.12
We now start with the atomic scale analysis of the QWR layer. In Fig.2共a兲, a topographic X-STM image of a typical part of the QWR layer is shown. One complete QWR 共be-tween the dashed lines兲 is visible in the image. The length and the height variation along this QWR is found to be 220 nm and 1.3 nm–1.9 nm, respectively. Although not imaged in their totality, other QWRs were found to span even longer 共⬎250 nm兲, which is in agreement with the AFM measure-ment shown in Fig. 1共a兲. The inter-QWR distance in the 关2¯33兴-direction was found to be in the order of tens of na-nometers. In Figs. 2共b兲 and 2共c兲, two close-up images are shown to illustrate the surface fluctuations of the GaAs/ AlGaAs interface at the bottom of the QWRs. These fluctua-tions show up as AlGaAs protrusions in the GaAs QWRs and
vica versa as GaAs intrusions in the AlGaAs buffer layer.
Both are absent at the top interface of the QWRs and the marker layers, indicating that they are formed prior to and/or during the crystallization of the Ga droplets but not during growth of the capping layer. The formation of these features is attributed to the destabilization of the buffer layer interface by the liquid Ga droplet at low temperature and a subsequent
reconfiguration of the surface. A more detailed explanation of this process will be given in the next section, where the QD layer is analyzed. Despite the substantial interface fluc-tuations at the bottom of the QWRs, the homogeneity of the contrast, e.g., Figs 2共b兲 and2共c兲, indicates that no intermix-ing of Al with the GaAs QWRs has occurred.
We now turn our attention to the QD layer. A total of 29 QDs were imaged by X-STM. The average height of the QDs was found to be 2.3 nm with = 0.6 nm, which is in agreement with the AFM measurements and suggest that there was little structural change during capping. Recently, photopumped lasing of QDs grown under the same growth conditions was demonstrated.13 Low temperature 共5 K兲 PL measurements on a dedicate sample shows that the current QDs emit around 703 nm, with a linewidth of 48 meV.10This small linewidth indicates high uniformity, which is reflected by the low found in the X-STM measurements.
In Fig. 3共a兲, a typical part of the QD layer is shown. Three QDs can be distinguished. A WL connecting the QDs is not observed. In a previous study we suggested that a WL will not form under the current growth conditions because low temperature 共6 K兲 PL measurements on that sample showed no clear peak that could be attributed to WL emission.14This is confirmed in the current letter. Due to the high density approximately half of the observed QDs were found to overlap with other QDs. This illustrated in Fig.3共c兲, where the right side of the QD overlaps with another. The QDs are found to be bounded by 共211兲- and 共411兲-side fac-ets, resulting in a slight elongation of the QDs in the 关2¯33兴-direction. Emanating from the top of the QDs are AlAs rich regions 共darker兲. As a guide to the eye the boundary of one of these regions is marked in Fig. 3共a兲 by two lines, which are found to roughly follow the 关011兴-direction. The formation of these features can be explained by the differ-ence in mobility of the Ga and Al adatoms. During the growth of the capping layer, Ga adatoms, which have a higher mobility move over the surface. Since they favor the 共311兲A-surface over the 共211兲- and 共411兲-side facets of the QD,15and given the limited migration length of Al adatoms, this results in AlAs rich regions on top of the QDs. The observation that the AlAs rich regions extend all the way to
FIG. 2. 共Color online兲 共a兲 250⫻75 nm2 topographic X-STM image of
one complete QWR 共marked by the two dashed lines兲. 关共b兲 and 共c兲兴 30 ⫻16 nm2close ups of the interface fluctuation.
FIG. 3. 共Color online兲 共a兲 35⫻170 nm2topographic X-STM image of the
QD layer. An AlAs rich region emanating from the top of one of the QDs is marked by two lines.共b兲 16⫻65 nm2topographic image of a typical QD
with a smooth and共c兲 rough bottom AlGaAs/GaAs interface.
193112-2 Keizer et al. Appl. Phys. Lett. 98, 193112共2011兲
the GaAs capping layer, and that slight interface irregulari-ties are found at this position, indicates that to some extend the original QD shape is retained at the growth front during capping. The lateral position of the asymmetric QD shape at the growth front is continuously shifted in the 关2¯33兴-direction during overgrowth resulting in a tilt of the AlAs rich regions toward the关011兴-direction.
In approximately one third of the observed QDs, a GaAs intrusion in the AlGaAs buffer layer that is paired with an AlGaAs protrusion in the GaAs QD is found at the bottom interface, see e.g., Fig. 3共c兲. These paired protrusions/intrusions are better resolved in another sample were the QDs are overgrown with a GaAs capping layer, see Fig. 4, resulting in a clearer view of the interface. In both samples the protrusions were always found to form in the关2¯33兴-direction relative to the intrusions and are bounded by an 共100兲- and 共211兲-surface. The paired protrusions/ intrusions are most likely the result of an interface instability induced by the liquid Ga droplet. It is well known that 共311兲A-surfaces show temperature instabilities that break up the flat surface into facets that have a lower surface energy.8,16 We suggest that a similar process happens in the current case, where inside the droplets the共311兲A-surface is destabilized by the presence of the liquid Ga and subse-quently reconfigures into a more energetically favorable con-figuration consisting of共100兲- and 共211兲-facets. In previously studied GaAs/AlGaAs QDs grown on an 共001兲-oriented substrate17we observed GaAs intrusions that are attributed to local etching.18,19In that case As atoms present in the buffer layer, and to a lesser extend also the Al atoms, dissolve into the liquid Ga and recrystallize inside the QD upon the appli-cation of an As4flux. As a result these QDs posses an
intru-sion at their bottom interface and show some amount of Al intermixing. We would like to stress that this is not the case with the current QDs, where the homogeneity of the contrast
within the QDs, e.g., Figs 3共b兲and3共c兲, indicates that they consist of pure GaAs and that no Al intermixing has oc-curred.
To summarize, we have shown that QWRs can be cre-ated by annealing uncapped QDs at elevcre-ated temperatures. Our X-STM study reveals that substantial interface fluctua-tions are present between the nanostructures and the buffer layer. These fluctuations are attributed to the destabilization of the interface by the liquid Ga droplet and subsequent re-configuration of the surface in energetically more favorable facets. Despite the interface fluctuations, no evidence of Al intermixing was found. A WL connecting the QDs was not found. AlAs rich regions, which are brought on by the dif-ference in mobility of Al and Ga, were found emanating from the top of the QDs.
We thank STW-VICI under Grant No. 6631 for their financial support.
1N. Koguchi, K. Ishige, and S. Takahashi,J. Vac. Sci. Technol. B 11, 787
共1993兲.
2D. M. Bruls, J. W. A. M. Vugs, P. M. Koenraad, H. W. M. Salemink, J. H.
Wolter, M. Hopkinson, M. S. Skolnick, F. Long, and S. P. A. Gill,Appl. Phys. Lett. 81, 1708共2002兲.
3P. Offermans, P. M. Koenraad, R. Nötzel, J. H. Wolter, and K. Pierz,Appl.
Phys. Lett. 87, 111903共2005兲.
4V. Mlinar, M. Bozkurt, J. Ulloa, M. Ediger, G. Bester, A. Badolato, P.
Koenraad, R. Warburton, and A. Zunger,Phys. Rev. B 80, 165425共2009兲. 5S. Sanguinetti,J. Cryst. Growth 253, 71共2003兲.
6R. Nötzel, M. Ramsteiner, J. Menniger, A. Trampert, H.-P. Schönherr, L.
Däweritz, and K. H. Ploog,J. Appl. Phys. 80, 4108共1996兲.
7P. M. Petroff, A. C. Gossard, and W. Wiegmann,Appl. Phys. Lett.45, 620
共1984兲.
8R. Nötzel, N. Ledentsov, L. Däweritz, K. Ploog, and M. Hohenstein,Phys.
Rev. B 45, 3507共1992兲.
9M. Jo, T. Mano, and K. Sakoda,Appl. Phys. Express 3, 045502共2010兲. 10T. Mano, T. Kuroda, K. Mitsuishi, T. Noda, and K. Sakoda, J. Cryst.
Growth 311, 1828共2009兲.
11C. Heyn, A. Stemmann, R. Eiselt, and W. Hansen,J. Appl. Phys. 105,
054316共2009兲.
12M. Jo, J. G. Keizer, T. Mano, P. M. Koenraad, and K. Sakoda,Appl. Phys.
Express 4, 055501共2011兲.
13T. Mano, T. Kuroda, K. Mitsuishi, Y. Nakayama, T. Noda, and K. Sakoda,
Appl. Phys. Lett. 93, 203110共2008兲.
14T. Mano, T. Noda, T. Kuroda, S. Sanguinetti, and K. Sakoda,Phys. Status
Solidi C 8, 257共2011兲.
15K. Asano, Y. Kangawa, H. Ishizaki, T. Akiyama, K. Nakamura, and T. Ito,
Appl. Surf. Sci. 237, 206共2004兲.
16R. Notzel, L. Däweritz, and K. Ploog,Phys. Rev. B 46, 4736共1992兲. 17J. G. Keizer, J. Bocquel, P. M. Koenraad, T. Mano, T. Noda, and K.
Sakoda,Appl. Phys. Lett. 96, 062101共2010兲.
18Z. M. Wang, B. L. Liang, K. A. Sablon, and G. J. Salamo,Appl. Phys.
Lett. 90, 113120共2007兲.
19A. Stemmann, C. Heyn, T. Koppen, T. Kipp, and W. Hansen,Appl. Phys.
Lett. 93, 123108共2008兲.
FIG. 4. 共Color online兲 共a兲 10⫻22 nm2 topographic X-STM image of a
GaAs QD capped with GaAs. The interface is marked by the white trans-parent line.
193112-3 Keizer et al. Appl. Phys. Lett. 98, 193112共2011兲
