# Fileset

[Ohtake_2024_Jpn._J._Appl._Phys._63_03SP10.pdf](https://mdr.nims.go.jp/filesets/a5b90d77-458b-4f5c-b869-f3fdd51bb81d/download)

## Creator

[Akihiro Ohtake](https://orcid.org/0000-0002-3519-4613), [Takaaki Mano](https://orcid.org/0000-0002-6955-260X)

## Rights

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

## Other metadata

[Initial stage of InSb heteroepitaxial growth on GaAs (111)A: effect of thin InAs interlayers](https://mdr.nims.go.jp/datasets/c4dc5d2f-25ca-4b44-bee0-e4b19fd2cf4e)

## Fulltext

Initial stage of InSb heteroepitaxial growth on GaAs (111)A: effect of thin InAs interlayersJapanese Journal of AppliedPhysics     REGULAR PAPER • OPEN ACCESSInitial stage of InSb heteroepitaxial growth onGaAs (111)A: effect of thin InAs interlayersTo cite this article: Akihiro Ohtake and Takaaki Mano 2024 Jpn. J. Appl. Phys. 63 03SP10 View the article online for updates and enhancements.You may also likeGrowth and Characterization of InSb ThinFilms on GaAs (001) without Any BufferLayers by MBEXiao-Meng Zhao,  , Yang Zhang et al.-Step Hall Measurement of InSb FilmsGrown on Si(111) Substrate Using InSbBilayerKoji Nakayama, Kimihiko Nakatani, SaraKhamseh et al.-Ion irradiation-induced polycrystalline InSbfoamR Giulian, J B Salazar, W Just et al.-This content was downloaded from IP address 144.213.253.16 on 13/02/2024 at 00:49https://doi.org/10.35848/1347-4065/ad2032https://iopscience.iop.org/article/10.1088/0256-307X/34/7/076105https://iopscience.iop.org/article/10.1088/0256-307X/34/7/076105https://iopscience.iop.org/article/10.1088/0256-307X/34/7/076105https://iopscience.iop.org/article/10.1143/JJAP.50.01BF01https://iopscience.iop.org/article/10.1143/JJAP.50.01BF01https://iopscience.iop.org/article/10.1143/JJAP.50.01BF01https://iopscience.iop.org/article/10.1088/1361-6463/aa920fhttps://iopscience.iop.org/article/10.1088/1361-6463/aa920fhttps://googleads.g.doubleclick.net/pcs/click?xai=AKAOjssvNuST6iWbmkhKaf0Z4r2Rt7GGV5RE8aPm9PGHXhpi-fCJDlb20GPERXD4S_CHSxzXBV8fw9TabzM4hXfwTuGtneZAC-I5a8Of0uKiNHBucqi5tck091waPtIEMKOC1aQgtSHZHnpp8ABiRQMo6aFT4FQNyNT_KW8xjixn4excQ-_Z3ViJRvf3fm9eaF3H_3lgY99SJRbPvCvOaePHsQR_PiDIODXt4s0pRWxykjl6tdMqL2YfpoiAH7xpFhhYuntM_IzccnPalulq4u0zCAoXlQnVB3lbbOWPUY1HzL20iYzqavKFXvyuAFQA1E8gduvnIocwaA&sai=AMfl-YQ_BRQdqJfCfonA6lxQlShGUym2I7186KaJM8UsMs01yIGGoqr6FYfvUutxKwd4zY3cqthzuIbNAt4Gi-4&sig=Cg0ArKJSzIPG7hMgpro2&fbs_aeid=%5Bgw_fbsaeid%5D&adurl=https://ecs.confex.com/ecs/prime2024/cfp.cgi%3Futm_source%3DIOP%26utm_medium%3Dbanner%26utm_campaign%3Dprime_abstract_submissionInitial stage of InSb heteroepitaxial growth on GaAs (111)A: effect of thin InAsinterlayersAkihiro Ohtake* and Takaaki Mano*National Institute for Materials Science, 1-1 Namiki, Tsukuba, Ibaraki 305-0044, Japan*E-mail: OHTAKE.Akihiro@nims.go.jp; MANO.Takaaki@nims.go.jpReceived December 11, 2023; revised January 15, 2024; accepted January 18, 2024; published online February 8, 2024MBE of InSb on the (111)A-oriented GaAs substrates has been studied using electron diffraction, X-ray diffraction, and scanning probemicroscopy. The direct heteroepitaxial growth of InSb on GaAs(111)A results in a cracked morphology with flat terraces and deep gaps, whichcould be attributed to the extremely large lattice mismatch between InSb and GaAs (14.6%). When thin (5–30 monolayer thickness) InAs films areused as interlayers, more continuous and flat InSb films are obtained. The proposed growth technique using (111)A-oriented GaAs substrates andthin InAs interlayers are effective in improving the surface morphology and the structural quality of InSb films in highly lattice-mismatched systems.© 2024 The Author(s). Published on behalf of The Japan Society of Applied Physics by IOP Publishing Ltd1. IntroductionIII–V semiconductors with narrow band gaps, such as InAs,GaSb, and InSb, are essential materials for various deviceapplication due to their high mobility characteristic andoptical properties at IR region. As the examples, Hall sensors,IR-photosensors, and IR LEDs have been developed andcommercialized.1–7) The wafers of InAs, GaSb, and InSb arenormally very expensive and large-size wafers are notavailable. In addition, the narrow-gap semiconductor sub-strates can be a source of parallel conduction when used assubstrates for lateral conductive devices.8) Therefore, hetero-epitaxial growth of these narrow gap III–V semiconductorson versatile lattice-mismatched substrates (Si, GaAs, or InP)with wider band gap has been important research topic formore than three decades.8–12) InSb has a highest electronmobility (∼78 000 cm2 V−1 s−1) and a very narrow band gap(0.17 eV at 300 K) among the III–V semiconductors, makingitself promising for applications in magnetic field sensing andmid-IR light detection with high sensitivity. However, a largelattice constant of InSb (0.6479 nm) causes difficulty ingrowing high quality layers on GaAs substrates whose latticeconstant is 14.6% smaller than that of InSb. On the standardGaAs (100) substrates, several techniques have been devel-oped to improve the quality, such as anion exchange (Sbsoaking)12–14) and/or two-step growth (temperaturevariation).7,15–18) It has been reported that most of thedislocations induced by the lattice mismatch could beconfined at the interface of InSb/GaAs by utilizing thesegrowth technique,16) and that high mobility characteristicshave been realized in relatively thick InSb layers grown onGaAs(100).8,15,18,19) However, three-dimensional islands areoften formed at the initial stage of the InSb growth, whichdegrade the structural and electrical properties of thinner InSbfilms.20,21) Since device structures with thin InSb layer areeffective in reducing the cost, and are also preferred forspecific applications using magnetoresistance,8,20) the initialisland growth should be suppressed.Kanisawa et al. have proposed to use the (111)A-orientedGaAs substrate to obtain smooth surface from the beginning ofthe InSb growth.21) Earlier studies have shown that in thelattice-mismatched system of InAs on GaAs(111)A, theformation of three-dimensional islands is effectively inhibitedby introducing misfit-dislocation network at the interface.22–25)The layer-by-layer growth continues throughout the growth,which is in stark contrast to that for the (100) orientation whereStranki-Krastanow growth occurs.26–28) The layer-by-layergrowth was also realized for InSb on GaAs(111)A bycombining the two-step growth technique, and the 200 nmthick InSb layer exhibits mobility close to ∼10 000cm2 V−1 s−1.21) On the other hand, atomic force microscopy(AFM) observations showed that three-dimensional structures(density: ∼108 cm−2) were actually formed on the 30 nm thickInSb layers.21) This means that further studies are required forthe growth of thin InSb layers on GaAs (111)A in order toimprove the crystal quality.In this study, we investigate the initial growth stage ofInSb on GaAs(111)A in detail. We focus on the low-temperature growth of InSb which corresponds to the firststep of the two-step growth. By studying strain relaxationprocesses and surface morphology of InSb, we found that thedirect growth of InSb on GaAs(111)A causes a rough surfacewhich consists of terraces and deep gaps. To suppress the gapformation, we propose the growth method using InAsinterlayers, which has been developed for growing InGaAson GaAs(111)A, GaSb on GaAs(111)A, and GaSb onSi(111).29–31) By introducing thin InAs interlayers, flatsurfaces of InSb(111)A could be formed even at the earlystage of the growth.2. Experimental methodsThe growth experiments were carried out in a system ofinterconnecting ultrahigh vacuum (UHV) chambers for MBEand for on-line surface characterization by means of scanningtunneling microscopy (STM).32) The InSb films were grownon the GaAs(111)A substrates with and without thin InAslayers. The samples were also grown on the InAs(111)Asubstrate for comparison. The clean surfaces of GaAs(111)Aand InAs(111)A were prepared by growing undoped homo-epitaxial layers at 450 °C on the thermally cleanedsubstrates.33) The GaAs (InAs) layers were grown with anAs4/Ga (As4/In) flux ratio of ∼50. The cleaned GaAs andContent 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.03SP10-1© 2024 The Author(s). Published on behalf ofThe Japan Society of Applied Physics by IOP Publishing LtdJapanese Journal of Applied Physics 63, 03SP10 (2024) REGULAR PAPERhttps://doi.org/10.35848/1347-4065/ad2032https://crossmark.crossref.org/dialog/?doi=10.35848/1347-4065/ad2032&domain=pdf&date_stamp=2024-02-08https://orcid.org/0000-0002-3519-4613https://orcid.org/0000-0002-3519-4613https://orcid.org/0000-0002-6955-260Xhttps://orcid.org/0000-0002-6955-260Xmailto:OHTAKE.Akihiro@nims.go.jpmailto:MANO.Takaaki@nims.go.jphttps://creativecommons.org/licenses/by/4.0/https://doi.org/10.35848/1347-4065/ad2032InAs surfaces show a (2 × 2) reconstruction. Thin InAslayers on GaAs(111)A were also grown under otherwiseidentical conditions.InSb films were grown at substrate temperatures of 300 °Cand 320 °C. The beam-equivalent pressures (BEPs) of In andSb were measured using a beam flux monitor at the sampleposition. The BEP of In was fixed at ∼9 × 10−7 Pa. On theother hand, because of the lower sticking probability of Sbmolecules at a higher temperature of 320 °C, the BEPs for Sbwas controlled to ∼5 × 10−6 Pa and ∼1.5 × 10−5 Pa for thegrowth at 300 °C and 320 °C, respectively. Prior to the InSbgrowth at 320 °C, the initial substrate surface was exposed tothe Sb molecular beam for 20 s. Under either growthcondition, the growing surface shows an In-stabilized(2 × 2) reconstruction at the initial stage of the growth,and an Sb-stabilized (2√3 × 2√3)R30° phase begins tocoexist above ∼20 monolayer (ML)-thickness. The appear-ance of the (2√3 × 2√3)R30° reconstruction indicates that theInSb film grows with a (111)A orientation.34) The samplesshowed (2 × 2) surfaces after interrupting the In and Sbmolecular beams at growth temperatures, and were thentransferred via UHV transfer modules to another chamber forthe on-line characterizations by STM. The growth rate ofInSb was approximately 0.03ML s−1, which was calibratedby reflection high-energy electron diffraction (RHEED)intensity oscillation measurements for the (001)-orientedInAs growth. Here, 1 ML of InSb is defined as 5.5 × 1014atoms cm−2, which is the site-number density of unrecon-structed InSb(111)A surface.The growth processes were monitored by in situ RHEEDwith electron- beam energy of 15 keV. All the STM imageswere collected at RT in the constant current mode with atunneling current of 0.1 nA and a sample voltage of −3 V.The samples were also characterized by AFM, and X-raydiffraction (XRD). High resolution XRD measurements werecarried out using a monochromatic Cu Kα1 radiation. Achannel-cut analyzer crystal was used for X-ray rockingcurve (XRC) measurements.Results and discussionFigure 1 shows RHEED patterns taken before and after thegrowth of InSb (50ML thickness) on GaAs(111)A (a and b),10 ML-InAs/GaAs(111)A (c and d), and InAs(111)A (e andf) substrates at 320 °C. Since sharp and intense streaks wereobserved, one may expect that high-quality InSb films withflat surfaces were grown on all substrates. However, thesurface morphology critically depends on the type ofsubstrate, as we will show later. From the spacing of thestreaks, the in-plane lattice constant of the InSb(111)A film isroughly estimated to be 0.46 nm, which is quite close to thebulk value (0.458 nm).Figure 2(a) shows the in-plane lattice constant (d110) of theInSb film grown on the GaAs(111)A substrate at 300 °C and320 °C plotted as a function of the InSb film thickness. Thed110 values were measured from the distance between the 1 1and ̅11 reflections in the RHEED patterns taken along the[ ̅112] direction. The reflections from InSb film appeared at1.5 ML thickness, in addition to those from the GaAssubstrate. The d110 values of the InSb film are quite closeto the value of bulk InSb, and remains almost unchangedthroughout the growth. This indicates that the InSb film wasnucleated with its inherent lattice constant, and that pseudo-morhic InSb layers are not formed. Thus, it is likely that thelattice mismatch of InSb/GaAs(111)A (14.6%) is too large tobe accommodated by elastic deformation of thin InSb films,similarly to the case for InAs/Si (11.5%).24)To reduce the lattice mismatch between InSb and GaAs, thegrowth experiments were carried out using thin InAs inter-layers. Figures 2(b)–2(e) show the variation of d110 value forthe InSb growths on 5 ML-, 10 ML-, 20 ML-, and 30 ML-thickInAs layers on GaAs(111)A. As already reported,23,24) the in-plane lattice constants of InAs on GaAs(111)A increases withInAs film thickness: the d110 values of the 5 ML-, 10 ML-,20 ML-, and 30 ML-thick InAs films are roughly estimated tobe 0.418, 0.421, 0.422, and 0.423 nm, respectively.As can be seen in Figs. 2(b)–2(e), the insertion of InAslayer drastically changes the strain relaxation process of InSb:the d110 value begins to change at the very early stage of thegrowth (0.2–0.5 ML thickness), and gradually increase as thegrowth proceeds. In the InSb/InAs system with a smallerlattice mismatch of 7.0%, as shown in Fig. 2(f), the d110value begins to change at ∼1ML, and increases with InSbthickness: the observed variation in d110 is similar to that forInAs/GaAs(111)A (7.1%).22,23) On the other hand, strainstate is different between the two systems: the strain in theInAs film on GaAs(111)A has relaxed by only ∼80% afterthe 50 ML-growth, while the 50 ML-InSb film on InAs(111)A is mostly relaxed.Here, it is interesting to note that the strain relaxationproceeds more rapidly for thinner InAs thickness. Thus, it is(a)(b)(c)(d)(e)(f)Fig. 1. RHEED patterns of before and after the growth of 50 ML-InSb films on GaAs(111)A (a) and (b), 10 ML-InAs/GaAs(111)A (c) and (d), and InAs(111)A (e) and (f) substrates. The growth temperature was 320 °C. The electron-incidence azimuths are [ ̅110] and [ ̅112]. Vertical lines indicate the position ofthe streaks.03SP10-2© 2024 The Author(s). Published on behalf ofThe Japan Society of Applied Physics by IOP Publishing LtdJpn. J. Appl. Phys. 63, 03SP10 (2024) A. Ohtake and T. Manopossible that thin InAs interlayer is elastically deformed toaccumulate compressive strain in InSb layers, as alreadyfound in InGaAs on InAs/GaAs(111)A.30) Another inter-esting finding is that the strain relaxation is delayed at a lowertemperature of 300 °C. Since, as mentioned earlier, the InSbgrowth at 320 °C was carried out under more Sb-richconditions: the higher Sb flux of 1.5 × 10−7 Pa was usedand the initial surface was exposed to the Sb flux prior to thegrowth. Thus, it is suggested that Sb atoms are morepreferentially incorporated in InAs layers at the initial stageof the InSb growth at 320 °C to form an InAsSb-like phase,which promotes the strain relaxation of InSb.Figures 3(a) and 3(b) show STM and AFM images,respectively, observed for the 50 ML thick InSb film directlygrown on GaAs(111)A at 320 °C. While the surface consistsof terraces with a typical width of several tens of nm, deepgaps are frequently observed. The depth of the gap is roughlyestimated to be ∼20 nm, being consistent with the thicknessof the 50 ML-InSb film. The rms roughness (Rq) values are2.56 and 3.36 nm for STM (a) and AFM (b) images,respectively. The existence of the gap indicates that thedirect InSb growth on GaAs(111)A proceeds with theformation of large two-dimensional islands, but not with alayer-by-layer mode.The insertion of thin InAs layer is effective to improve thesurface morphology of InSb films. The STM and AFMimages of InSb films grown with InAs interlayers (5 ML and30ML) are shown in Figs. 3(c)–3(f). The surfaces showatomically flat terraces separated by steps with the height of∼0.37 nm, which corresponds to the ML height of InSb(111):while shallow holes are formed on terraces, no deep gap wasobserved. The Rq values are significantly deceased, and aresmaller than 0.5 nm in the whole range of InAs thicknessbetween 5 and 30ML. As shown in Figs. 3(g) and 3(h), theInSb film grown on the InAs substrate also shows a flatsurface with small Rq values of 0.20 nm (g) and 0.39 nm (h).These results indicate that the reduction of the latticemismatch by inserting thin InAs layers is crucial in obtainingbetter surface morphology of epitaxial InSb films. Here wenote that the difference in surface morphologies could not beassessed by RHEED observations: as shown in Figs 1(b),1(d), and 1(f), the InSb films grown on GaAs(111)A (a), 10(a)(b)(c)(d)(e)(f)Fig. 2. Variation of the in-plane lattice constant (d110) of InSb films grown on GaAs(111)A (a), InAs/GaAs(111)A (b)–(e), and InAs(111)A (f) substrates.The values were measured from the distance between the 1 1 and ̅11 reflections in the RHEED patterns. Red and blue circles represent the results for the growthat 300 °C (BEP of Sb: ∼5 × 10−6 Pa) and 320 °C (BEP of Sb: ∼1.5 × 10−5 Pa), respectively. The horizontal dashed lines indicate the d110 values of thesubstrates and bulk InSb.03SP10-3© 2024 The Author(s). Published on behalf ofThe Japan Society of Applied Physics by IOP Publishing LtdJpn. J. Appl. Phys. 63, 03SP10 (2024) A. Ohtake and T. ManoML-InAs/GaAs(111)A (b), and InAs(111)A (c) substratesshow quite similar RHEED patterns.To investigate the crystalline quality of the InSb film, wecarried out XRD measurements for the 100 nm thick InSb filmsprepared under identical growth conditions at 300 °C.Figures 4(a)–4(f) compares XRCs of the symmetric 111reflection measured from 100 nm InSb films. The FWHMvalue of the InSb film directly grown on GaAs(111)A is339 arcsec. For the InSb films grown using InAs interlayers,the FWHM values are less than 200 arcsec, showing a bettercrystallinity, indicating that the thin InAs layer is also effectivein improving the structural properties of InSb. We have reportedthat the residual strain in epitaxial films gives rise to the peakbroadening in XRC profiles.24,29) However, since the 100 nm-InSb film on GaAs(111)A is almost relaxed (99.7%), theobserved peak broadening could not be explained on the basisof residual strain. On the other hand, the broader XRC profile iscompatible with the STM and AFM results: as can be seen inFigs. 3(a) and 3(b), the InSb film directly grown on GaAssubstrate consists of two-dimensional islands separated by gaps,while those grown on InAs are more continuous. Since the XRCwidth generally increases as the domain size decreases, theimprovement in the film morphology is likely to be responsiblefor the narrower FWHM width in Figs. 4(b)–4(f).4. ConclusionsWe have studied the strain relaxation processes InSb onGaAs(111)A with a large lattice mismatch of 14.6%. Thefully-relaxed InSb film starts to grow directly on the GaAs(111)A substrate, resulting in the rather defective filmquality. The insertion of the thin InAs layers (5–30ML inthickness) significantly changes the strain relaxation pro-cesses and improve the surface morphology and the structuralquality of the InSb films. The heteroepitaxial growth tech-nique using thin InAs layers is promising for the realizationof InSb-based devices on the low-cost GaAs wafers.AcknowledgmentsThe authors are grateful to Drs. T. Kawazu, H. T. Miyazaki,and Y. Sakuma for their fruitful discussion.ORCID iDsAkihiro Ohtake https://orcid.org/0000-0002-3519-4613Takaaki Mano https://orcid.org/0000-0002-6955-260X1) G. Singh, E. Michel, C. Jelen, S. Slivken, J. Xu, P. Bove, I. Ferguson, andM. Razeghi, J. Vac. Sci. Technol. B 13, 782 (1995).2) I. Shibasaki, J. Cryst. Growth 175-176, 13 (1997).3) A. Okamoto, A. Ashihara, T. Akaogi, and I. Shibasaki, J. Cryst. Growth227-228, 619 (2001).4) Rogalski, Opto-Electron. Rev. 20, 279 (2012).5) A. Krier, H. H. Gao, V. V. Sherstnev, and Y. Yakovlev, J. Phys. D: Appl.Phys. 32, 3117 (1999).(a)(b)(c)(d)(e)(f)(g)(h)Fig. 3. Typical filled-state STM images and AFM images of 50 ML-InSb films grown on GaAs(111)A (a) and (b), InAs/GaAs(111)A (c)–(f), and InAs(111)A (g) and (h) substrates. The growth temperature was 320 °C. Image dimensions of STM and AFM images are 500 nm × 500 nm and 5 μm × 5 μm,respectively. Black to white represents the height difference of 23 nm (a), 25 nm (b), 2 nm (c), 5 nm, (d), 3 nm (e), 5 nm (f), 3 nm (g), and 3 nm (h).(a)(b)(c)(d)(e)(f)Fig. 4. Symmetric 111 XRCs of 100 nm InSb films grown on GaAs(111)A(a), 5 ML-InAs/GaAs(111)A (b), 10 ML-InAs/GaAs(111)A (c), 20 ML-InAs/GaAs(111)A (d), 30 ML-InAs/GaAs(111)A (e), and InAs(111)A (f)substrates. The growth temperature was 300 °C.03SP10-4© 2024 The Author(s). Published on behalf ofThe Japan Society of Applied Physics by IOP Publishing LtdJpn. J. Appl. Phys. 63, 03SP10 (2024) A. Ohtake and T. Manohttps://orcid.org/0000-0002-3519-4613https://orcid.org/0000-0002-3519-4613https://orcid.org/0000-0002-3519-4613https://orcid.org/0000-0002-6955-260Xhttps://orcid.org/0000-0002-6955-260Xhttps://orcid.org/0000-0002-6955-260Xhttps://doi.org/10.1116/1.588163https://doi.org/10.1016/S0022-0248(96)00924-4https://doi.org/10.1016/S0022-0248(01)00784-9https://doi.org/10.1016/S0022-0248(01)00784-9https://doi.org/10.2478/s11772-012-0037-7https://doi.org/10.1088/0022-3727/32/24/304https://doi.org/10.1088/0022-3727/32/24/3046) H. Fujita, K. Ueno, O. Morohara, E. Camargo, H. Geka, Y. Shibata, andN. Kuze, Phys. Status Solidi A 215, 1700449 (2018).7) H. Lin, Z. Zhou, H. Xie, Y. Sun, X. Chen, J. Hao, S. Hu, and N. Dai, Phys.Status Solidi A 218, 2100281 (2021).8) T. Zhang, S. K. Clowes, M. Debnath, A. Bennett, C. Roberts, J. J. Harris, R.A. Stradling, L. F. Cohen, T. Lyford, and P. F. Fewster, Appl. Phys. Lett. 84,4463 (2004).9) H. Yamaguchi, M. R. Fahy, and B. A. Joyce, Appl. Phys. Lett. 69, 776(1996).10) K. Murata, N. B. Ahmad, Y. Tamura, M. Mori, C. Tatsuyama, andT. Tambos, J. Cryst. Growth 301–302, 203 (2007).11) A. Ohtake and K. Mitsuishi, J. Vac. Sci. Technol. B 29, 031804 (2011).12) M. D. Nordstrom, T. A. Garrett, P. Reddy, J. McElearney, J. R. Rushing,K. D. Vallejo, K. Mukherjee, K. A. Grossklaus, T. E. Vandervelde, andP. J. Simmonds, Cryst. Growth Des. 23, 8670 (2023).13) B. W. Jia, K. H. Tan, W. K. Loke, S. Wicaksono, and S. F. Yoon, J. Appl.Phys. 120, 035301 (2016).14) S. Huang, G. Balakrishnan, and D. L. Huffaker, J. Appl. Phys. 105, 103104(2009).15) M. C. Debnath, T. Zhang, C. Roberts, L. F. Cohen, and R. A. Stradling, J.Cryst. Growth 267, 17 (2004).16) T. W. Kim, H. C. Bae, and H. L. Park, Appl. Phys. Lett. 74, 380 (1999).17) J. L. Davis and P. E. Thompson, Appl. Phys. Lett. 54, 2235 (1989).18) E. Michel, G. Singh, S. Slivken, C. Besikci, P. Bove, I. Ferguson, andM. Razeghi, Appl. Phys. Lett. 65, 3338 (1994).19) S. Fujikawa, T. Taketsuru, D. Tsuji, T. Maeda, and H. I. Fujishiro, J. Cryst.Growth 425, 64 (2015).20) S. A. Solin, T. Thio, D. R. Hines, and J. J. Heremans, Science 289, 1530(2000).21) K. Kanisawa, H. Yamaguchi, and Y. Hirayama, Appl. Phys. Lett. 76, 589(2000).22) H. Yamaguchi, J. G. Belk, X. M. Zhang, J. L. Sudijono, M. R. Fahy,T. S. Jones, D. W. Pashley, and B. A. Joyce, Phys. Rev. B 55, 1337 (1997).23) A. Ohtake, M. Ozeki, and J. Nakamura, Phys. Rev. Lett. 84, 4665 (2000).24) A. Ohtake, T. Mano, and Y. Sakuma, Sci. Rep. 10, 4606 (2020).25) T. Mano, A. Ohtake, T. Kawazu, H. T. Miyazaki, and Y. Sakuma, ACSAppl. Mater. Interfaces 15, 29636 (2023).26) L. Goldstein, F. Glas, J. Y. Marzin, M. N. Charasse, and G. Le Roux, Appl.Phys. Lett. 47, 1099 (1985).27) D. Leonard, M. Krishnamurthy, C. M. Reaves, S. P. Denbaars, andP. M. Petroff, Appl. Phys. Lett. 63, 3203 (1993).28) Y. Nabetani, T. Ishikawa, S. Noda, and A. Sasaki, J. Appl. Phys. 76, 347(1994).29) A. Ohtake, T. Mano, K. Mitsuishi, and Y. Sakuma, ACS Omega 3, 15592(2018).30) T. Mano, K. Mitsuishi, N. Ha, A. Ohtake, A. Castellano, S. Sanguinetti,T. Noda, Y. Sakuma, T. Kuroda, and K. Sakoda, Cryst. Growth Des. 16,5412 (2016).31) A. Ohtake, T. Mano, N. Miyata, T. Mori, and T. Yasuda, Appl. Phys. Lett.104, 032101 (2014).32) A. Ohtake, Surf. Sci. Rep. 63, 295 (2008).33) A. Ohtake, N. Ha, and T. Mano, Cryst. Growth Des. 15, 485 (2015).34) A. J. Noreika, M. H. Francombe, and C. E. C. Wood, J. Appl. Phys. 52,7416 (1981).03SP10-5© 2024 The Author(s). Published on behalf ofThe Japan Society of Applied Physics by IOP Publishing LtdJpn. J. Appl. Phys. 63, 03SP10 (2024) A. Ohtake and T. Manohttps://doi.org/10.1002/pssa.201700449https://doi.org/10.1002/pssa.202100281https://doi.org/10.1002/pssa.202100281https://doi.org/10.1063/1.1748850https://doi.org/10.1063/1.1748850https://doi.org/10.1063/1.117888https://doi.org/10.1063/1.117888https://doi.org/10.1016/j.jcrysgro.2006.11.080https://doi.org/10.1116/1.3589807https://doi.org/10.1021/acs.cgd.3c00812https://doi.org/10.1063/1.4958863https://doi.org/10.1063/1.4958863https://doi.org/10.1063/1.3129562https://doi.org/10.1063/1.3129562https://doi.org/10.1016/j.jcrysgro.2004.03.033https://doi.org/10.1016/j.jcrysgro.2004.03.033https://doi.org/10.1063/1.123077https://doi.org/10.1063/1.101134https://doi.org/10.1063/1.112384https://doi.org/10.1016/j.jcrysgro.2015.02.047https://doi.org/10.1016/j.jcrysgro.2015.02.047https://doi.org/10.1126/science.289.5484.1530https://doi.org/10.1126/science.289.5484.1530https://doi.org/10.1063/1.125826https://doi.org/10.1063/1.125826https://doi.org/10.1103/PhysRevB.55.1337https://doi.org/10.1103/PhysRevLett.84.4665https://doi.org/10.1038/s41598-020-61527-9https://doi.org/10.1021/acsami.3c05725https://doi.org/10.1021/acsami.3c05725https://doi.org/10.1063/1.96342https://doi.org/10.1063/1.96342https://doi.org/10.1063/1.110199https://doi.org/10.1063/1.358483https://doi.org/10.1063/1.358483https://doi.org/10.1021/acsomega.8b02359https://doi.org/10.1021/acsomega.8b02359https://doi.org/10.1021/acs.cgd.6b00899https://doi.org/10.1021/acs.cgd.6b00899https://doi.org/10.1063/1.4862542https://doi.org/10.1063/1.4862542https://doi.org/10.1016/j.surfrep.2008.03.001https://doi.org/10.1021/cg501545nhttps://doi.org/10.1063/1.328732https://doi.org/10.1063/1.328732 1. Introduction 2. Experimental methods Results and discussion 4. Conclusions Acknowledgments A6