# Fileset

[Epitaxial lateral overgrowth of r-plane α-Ga2O3 with stripe masks along .pdf](https://mdr.nims.go.jp/filesets/37e89325-de02-4c2a-a8b9-563880604614/download)

## Creator

[Yuichi Oshima](https://orcid.org/0000-0001-8293-4891), Shingo Yagyu, Takashi Shinohe

## Rights

This article may be downloaded for personal use only. Any other use requires prior permission of the author and AIP Publishing. This article appeared in Yuichi Oshima et al., J. Appl. Phys. 130, 175304 (2021) and may be found at https://doi.org/10.1063/5.0068097.[In Copyright](http://rightsstatements.org/vocab/InC/1.0/)

## Other metadata

[Epitaxial lateral overgrowth of r-plane α-Ga2O3 with stripe masks along -12-10>](https://mdr.nims.go.jp/datasets/a5782ca1-dfa8-4215-afd3-567d12925052)

## Fulltext

Epitaxial lateral overgrowth of r-plane α-Ga2O3 with stripe masks along \langle { \bar{1}2\bar{1}0}\rangle ViewOnlineExportCitationCrossMarkRESEARCH ARTICLE |  NOVEMBER 01 2021Epitaxial lateral overgrowth of r-plane α-Ga 2O3 with stripemasks along  Special Collection: Wide Bandgap Semiconductor Materials and DevicesYuichi Oshima   ; Shingo Yagyu; Takashi ShinoheJ. Appl. Phys. 130, 175304 (2021)https://doi.org/10.1063/5.0068097⟨1̄21̄⟩ 30 January 2024 07:38:50https://pubs.aip.org/aip/jap/article/130/17/175304/1063745/Epitaxial-lateral-overgrowth-of-r-plane-Ga2O3-withhttps://pubs.aip.org/aip/jap/article/130/17/175304/1063745/Epitaxial-lateral-overgrowth-of-r-plane-Ga2O3-with?pdfCoverIconEvent=citehttps://pubs.aip.org/aip/jap/article/130/17/175304/1063745/Epitaxial-lateral-overgrowth-of-r-plane-Ga2O3-with?pdfCoverIconEvent=crossmarkhttps://pubs.aip.org/jap/collection/1300/Wide-Bandgap-Semiconductor-Materials-and-Devicesjavascript:;https://orcid.org/0000-0001-8293-4891javascript:;javascript:;javascript:;https://doi.org/10.1063/5.0068097https://servedbyadbutler.com/redirect.spark?MID=176720&plid=2219938&setID=592934&channelID=0&CID=814978&banID=521401185&PID=0&textadID=0&tc=1&scheduleID=2141444&adSize=1640x440&data_keys=%7B%22%22%3A%22%22%7D&matches=%5B%22inurl%3A%5C%2Fjap%22%5D&mt=1706600330706938&spr=1&referrer=http%3A%2F%2Fpubs.aip.org%2Faip%2Fjap%2Farticle-pdf%2Fdoi%2F10.1063%2F5.0068097%2F15273152%2F175304_1_online.pdf&hc=7d2b063d0be7aa03d9614ae9fbc1e6b1fbfd4276&location=Epitaxial lateral overgrowth of r-plane α-Ga2O3with stripe masks along h�12�10iCite as: J. Appl. Phys. 130, 175304 (2021); doi: 10.1063/5.0068097View Online Export Citation CrossMarkSubmitted: 23 August 2021 · Accepted: 12 October 2021 ·Published Online: 1 November 2021Yuichi Oshima,1,a) Shingo Yagyu,2 and Takashi Shinohe2AFFILIATIONS1Optical Single Crystals Group, National Institute for Materials Science, 1-1 Namiki, 305-0044 Tsukuba, Japan2FLOSFIA, Inc., Kyodai-Katsura Venture Plaza, 615-8245 Kyoto, JapanNote: This paper is part of the Special Topic on: Wide Bandgap Semiconductor Materials and Devices.a)Author to whom correspondence should be addressed: OSHIMA.Yuichi@nims.go.jpABSTRACTWe demonstrated the epitaxial lateral overgrowth (ELO) of (�1012) (r-plane) α-Ga2O3 using a striped mask pattern along h�12�10i. α-Ga2O3stripes with an asymmetric cross-sectional shape were formed selectively on the windows at the initial growth stage. They grew verticallyand laterally to coalesce with each other, and a compact film was obtained. The film surface exhibited wave-like morphology with macro-scale inclined terraces and steps because of the asymmetric cross-sectional stripe shape. The macrosteps moved laterally like traveling wavesas the growth proceeded. Transmission electron microscopy revealed that a domain on a window grew toward the inclined c-axis directionto cover the adjacent domain after the coalescence. As a result, the dislocations, which propagated into the α-Ga2O3 stripe from the seedlayer through the window, bent toward the c axis and concentrated in a narrow area. This concentration should enhance the pair annihila-tion. Therefore, the dislocation density was markedly reduced on the top surface including the areas above the windows and coalescedboundaries in contrast to the cases of conventional c-, a-, and m-plane ELO.Published under an exclusive license by AIP Publishing. https://doi.org/10.1063/5.0068097I. INTRODUCTIONCorundum-structured α-Ga2O3 is one of the metastablephases of Ga2O3. This material is an ultra-wide bandgap semicon-ductor with a bandgap energy of approximately 5.3 eV1 and so ispromising for applications such as power devices and ultraviolet(UV) photodetectors. Indeed, Schottky barrier diodes (SBDs) withvery low on-resistance and an excellent ideality factor have beendemonstrated in spite of the high dislocation density.2,3 High-response UV detectors have also been demonstrated.4,5 In addition,α-(AlxGa1−x)2O3 solid solutions can be grown without limiting alu-minum content x,6 in contrast to β-(AlxGa1−x)2O3, where x islimited to being below ∼0.7 for bulk films (t > 50 nm) althoughhigher x is possible for very thin films (t < 10 nm).7 Therefore,band engineering using α-(AlxGa1−x)2O3 is possible over a widerange (5.3–8.6 eV) with higher freedom of the film thickness. It isalso possible to make a hetero pn junction with corundum-structured p-type oxides such as α-Ir2O3, which has a lattice cons-tant close to that of α-Ga2O3 (Δa/a∼ 0.1%).8–11 Normally offα-Ga2O3-based metal–oxide–semiconductor field-effect transistors(MOSFETs) with a p-type well have been demonstrated,12 and ahigh channel mobility of 72 cm2V−1 s−1, which is greater than thatof a typical commercially available SiC-based MOSFET, has beenreported.13A drawback of α-Ga2O3 is the absence of homoepitaxial bulksubstrates because this material is metastable under ambient pres-sure,14 and so melt-grown bulk α-Ga2O3 single crystals are notavailable, unlike β-Ga2O3.15–18 Accordingly, α-Ga2O3 films need tobe grown heteroepitaxially. Sapphire substrates are mainly used forthe growth because sapphire also has a corundum structure, andlarge-diameter substrates are available at a reasonable price. Mistchemical vapor deposition, halide vapor phase epitaxy (HVPE),and molecular beam epitaxy have been mainly reported as thegrowth methods.1,19–22 However, the dislocation density in a heter-oepitaxial α-Ga2O3 layer is as high as 1 × 1010 cm−2 because of thelarge lattice mismatch (Δa/a = 4.5%, Δc/c = 3.3%), if no measure istaken.23 Such a high density of dislocations can deteriorate the elec-trical properties and device performance. For example, a relativelylarge leakage current was reported for α-Ga2O3-based SBDs.2 Thisresult indicates that dislocations could be a current leakage path. Inaddition, a large drop in the electron mobility has been reportedJournal ofApplied Physics ARTICLE scitation.org/journal/japJ. Appl. Phys. 130, 175304 (2021); doi: 10.1063/5.0068097 130, 175304-1Published under an exclusive license by AIP Publishing 30 January 2024 07:38:50https://doi.org/10.1063/5.0068097https://doi.org/10.1063/5.0068097https://www.scitation.org/action/showCitFormats?type=show&doi=10.1063/5.0068097http://crossmark.crossref.org/dialog/?doi=10.1063/5.0068097&domain=pdf&date_stamp=2021-11-01http://orcid.org/0000-0001-8293-4891mailto:OSHIMA.Yuichi@nims.go.jphttps://doi.org/10.1063/5.0068097https://aip.scitation.org/journal/japfor n-type α-Ga2O3 in low concentration regions probably becauseof the scattering by dislocations.24 Such mobility drop would be abarrier to achieve high breakdown voltage and low on-resistance atthe same time. Furthermore, minority carrier lifetime in a pn junc-tion is likely to be limited because of the dislocations. This limita-tion could be an obstacle upon optimizing the switching speed andon-resistance. Therefore, the dislocation density needs to bereduced. To decrease the dislocation density, epitaxial lateral over-growth (ELO) has been investigated, and reported for the growth ofc-, a-, and m-plane α-Ga2O3 epilayers.23–29 The ELO techniquedramatically decreases the dislocation density in the laterally grownarea on the mask, and a very low density of less than 5 × 106 cm−2has been reported.23 However, the dislocation density above thewindow remains very high when the α-Ga2O3 stripe/island formedon the window has a flat top because the dislocations in the seedlayer go up straight to the top surface.23,28,29 It is possible to bendthe dislocations to prevent them from reaching the top surface andsuppress the formation of the defective areas by the facet-initiatedELO (FIELO) technique, which was originally developed toimprove the crystal quality of heteroepitaxial GaN layers.30 In theFIELO process, the growth conditions are chosen so that the toppart of a stripe/island consists of inclined facets. Then, the disloca-tions bend to be perpendicular to the facets to minimize the elasticenergy along the dislocation line. So far, for α-Ga2O3, the FIELOtechnique has been applied only to the c-plane,23 and the process issuccessful as long as the stripes/islands are isolated from eachother. However, a highly disordered morphology develops after thecoalescence leaving deep pits, which are not easily removed by pol-ishing, in contrast to the case of GaN where a smooth surface canbe easily recovered. As an alternative approach, the double-ELOtechnique has been demonstrated.31 In this technique, the secondmask is designed so that the window areas are located only on theFIG. 1. Bird’s-eye view SEM images of r-plane α-Ga2O3 grown on the striped mask. (a) Formation of isolated α-Ga2O3 stripes, (b) just after the coalescence, and(c) thick film growth after the coalescence.FIG. 2. (a) and (b) Cross-sectional SEM images of the α-Ga2O3 stripes along [�12�10] grown on r- and c-plane sapphire, respectively. (c) Schematic comparing the facetstructures.Journal ofApplied Physics ARTICLE scitation.org/journal/japJ. Appl. Phys. 130, 175304 (2021); doi: 10.1063/5.0068097 130, 175304-2Published under an exclusive license by AIP Publishing 30 January 2024 07:38:50https://aip.scitation.org/journal/japhigh-quality areas, and the formation of high dislocation densityareas is avoided. However, this method is costly because thisprocess requires two rounds of photolithography and epitaxialgrowth, as well as a polishing process prior to the second growth.In the present work, we report on the remarkable reduction ofthe high dislocation density areas by a single ELO process by per-forming the growth on an r-plane sapphire substrate with anappropriate mask design.II. EXPERIMENTALα-Ga2O3 was grown by the conventional ELO technique on a(�1012) (r-plane) sapphire substrate with an α-Ga2O3 seed layerwith a thickness of approximately 350 nm. We used a striped maskpattern along h�12�10i with mask/window widths of 5/5 μm.Conventional HVPE was used for the growth using GaClx and O2as the precursors. The partial pressures of GaClx and O2 supplywere 1.25 × 10 −1 and 1.25 kPa, respectively. N2 was used as thecarrier gas. The growth was performed at 520 °C under atmo-spheric pressure. The details of the growth conditions have beendescribed elsewhere.23 The morphologies of the grown crystals wereobserved by field-emission scanning electron microscopy(FE-SEM) with Hitachi SU8230. Transmission electron microscopy(TEM) was used to investigate the behavior of the dislocations. Themeasurement was performed by JEOL JEM-2100F operated at200 kV. The TEM samples were prepared by a focused ion beamwith a thickness of approximately 200 nm. The crystal orientationwas investigated by selective area electron diffraction (SAED) andx-ray pole figure measurement. The XRD measurement was per-formed by PANanalytical X’-pert MRD with Cu-Kα1 radiation.III. RESULTS AND DISCUSSIONFigures 1(a)–1(c) show bird’s-eye view SEM images of thegrowth evolution. Isolated α-Ga2O3 stripes were formed selectivelyon the windows [Fig. 1(a)]. The stripes had an asymmetricFIG. 3. (a) Cross-sectional bright-field TEM image of the coalesced α-Ga2O3 film grown for 2 h. The rectangle shows the depth from which the plan-view TEM specimenwas cut. The inset shows a magnified image of the seed layer. (b) and (c) SAED patterns of the circled areas in panel (a). (d) Plan-view bright-field TEM image of thefilm.Journal ofApplied Physics ARTICLE scitation.org/journal/japJ. Appl. Phys. 130, 175304 (2021); doi: 10.1063/5.0068097 130, 175304-3Published under an exclusive license by AIP Publishing 30 January 2024 07:38:50https://aip.scitation.org/journal/japcross-sectional shape with an inclined top surface, in contrast toconventional c-, a-, and m-plane α-Ga2O3. The stripes grew verti-cally and laterally to coalesce with each other [Fig. 1(b)].Macro-scale steps and inclined terraces were formed on the coa-lesced surface because of the asymmetric shape. The macro-stepand terrace structure was maintained throughout the growth afterthe coalescence, and they moved laterally like traveling waves as thegrowth proceeded [Fig. 1(c)]. Note that the initial acute shape ofthe macro-steps became smoother after the long-time growth. Thismorphological change took place probably to minimize the surfaceenergy. Further long-time growth could result in a much smoothersurface, which should be easily flattened by chemical mechanicalpolishing in contrast to the case of c-plane FIELO.Figure 2(a) shows a cross-sectional SEM image of theα-Ga2O3 stripe grown on r-plane sapphire. The top surfacewas (�1011), and the side facets were (10�11). On the other hand,Fig. 2(b) shows a cross-sectional SEM image of an α-Ga2O3 stripesalong [�12�10] grown on c-plane sapphire. The stripes were com-posed of (�1011) and (10�11) planes. Figure 2(c) schematicallydescribes these facet structures. The r-plane stripes and c-planestripes are drawn along the common c axis. This schematic tells usthat the both stripes have basically the same facet structure. Notethat the facet structures close to the bottom parts were different.This difference is probably because the conditions to minimize thesum of the surface energy and interface energy are differentdepending on the crystal orientation.Figure 3(a) shows a cross-sectional bright-field TEM image ofthe coalesced film shown in Fig. 1(c). A high density of dislocationswas observed in the seed layer as shown in the inset. The densitywas estimated to be approximately 1.4 × 1010 cm−2. The dislocationspropagated into the regrown α-Ga2O3 through the window as wasobserved for the cases of c-, a-, and m-plane α-Ga2O3. However,the dislocations bent toward the [0001] direction, and the densitylargely decreased at the surface. The kinks of the dislocation lineswere located along the dashed line, which should be the locus ofthe bottom edge of the macro-step. Note that belt-like contrastswere observed along the c axis. SAED patterns of these parts[Figs. 3(b) and 3(c)] revealed that the dark-contrast area includedtwinned domains, which were rotated by 180° around the c axis. InFig. 2(c), the diffraction spots from both the α-Ga2O3 matrix andtwinned domains were observed at the same time while only dif-fraction spots from the α-Ga2O3 matrix were observed in Fig. 3(b).Figure 3(d) shows a plan-view bright-field TEM image of a speci-men cut from the depth indicated by the rectangle in Fig. 3(a).Again, twinned domains were observed as the dark-contrast areas.One of the twinned domains is indicated by the arrow as anexample. Dislocations were observed virtually only in the dashedframe with a liner density of approximately 7 × 104 cm−1, asexpected from the cross-sectional observation. The areal dislocationdensity of the ELO-grown layer was estimated to be approximately5 × 107 cm−2 from the plan-view TEM image. The identification ofthe dislocation character would be carried out as our future work.Figures 4(a) and 4(b) show the x-ray pole figures of the sap-phire substrate and ELO-grown α-Ga2O3 film, respectively. InFig. 4(a), three diffraction spots of sapphire �1012 were observed atthe positions that were expected for a single crystalline corundumstructure. However, in Fig. 4(b), three additional weak spots(marked by dashed circles) were observed. The intensities of theFIG. 4. X-ray pole figures (log-scale) of (a) sapphire �1012 and (b) α-Ga2O3�1012.Journal ofApplied Physics ARTICLE scitation.org/journal/japJ. Appl. Phys. 130, 175304 (2021); doi: 10.1063/5.0068097 130, 175304-4Published under an exclusive license by AIP Publishing 30 January 2024 07:38:50https://aip.scitation.org/journal/japadditional peaks were approximately an order of magnitude lowerthan those of the main peaks. These additional spots wereexplained consistently by the existence of the twinned domainsaround the c axis. This result agreed with that of the TEM investi-gation. Note that such twinned domains were also found in ther-plane α-Ga2O3 seed layer by x-ray pole figure measurement (notshown). Therefore, the twinned domains in the regrown layer wereproposed to originate from those in the seed layer. Hence, thegrowth conditions of the seed layer need to be improved to sup-press the twinning.Figure 5 shows the epitaxial relationship between the α-Ga2O3matrix, twinned α-Ga2O3 domains, and sapphire substrate deter-mined by the SAED/pole figure measurements. The α-Ga2O3matrix and sapphire substrate have the same orientation, i.e.,(0001)Ga2O3 || (0001)sap and [�12�10]Ga2O3|| [�12�10]sap. On the otherhand, the twinned domain was rotated around the c axis by 180°,i.e., (0001)Ga2O3 || (0001)sap and [1�210]Ga2O3|| [�12�10]sap.Figures 6(a) and 6(b) are the cross-sectional schematicsshowing the growth models of conventional c-plane ELO andr-plane ELO in the present work, respectively. In the case of c-, a-,and m-plane ELO [Fig. 6(a)], a high dislocation density arearemained on the surface above the window because the dislocationsin the seed layer went up straight through the window. However, inthe case of the r-plane [Fig. 6(b)], each domain grew toward theinclined direction so as to cover the adjacent domain. In thisgrowth process, the dislocations first propagated so that they wereapproximately perpendicular to the (Å1011) top surface to minimizethe elastic energy along the dislocation lines. After that, when theymet the bottom edge of the macro-step on the dashed line, theybent toward the c axis to follow the macro-step. As a result, the dis-locations were concentrated in a narrow area along the c axis, andthen pair annihilation should be enhanced. Although dislocationsstill remained on the top surface, the density was much lower thanthose on the windows for the case of ELO on the conventionalplanes.IV. SUMMARYWe demonstrated the ELO of r-plane α-Ga2O3 using a stripedmask pattern along h�12�10i, and a coalesced compact film was suc-cessfully achieved. The α-Ga2O3 stripes formed at the initial growthstage exhibited an asymmetric cross-sectional stripe shape. Theasymmetric shape led to a wave-like morphology of the coalescedfilm, and this was maintained throughout the growth.Cross-sectional TEM observation revealed that each domain grewtoward the inclined [0001] direction so as to cover the adjacentdomain. As a result, although the dislocations propagated into theregrown α-Ga2O3, they bent toward the c axis to concentrate in anarrow area, leading to the effective pair annihilation. Thus, thistechnique enabled a much-improved crystal quality at the surfacein contrast to the case of conventional c-, a-, and m-plane ELOwhere dislocations go up straight to the top surface through thewindow to form high dislocation density areas.ACKNOWLEDGMENTSThis work was supported by Innovative Science andTechnology Initiative for Security (JPJ004596), ATLA, Japan.AUTHOR DECLARATIONSConflict of InterestThe authors have no conflicts to disclose.FIG. 5. Schematic of the epitaxial relationship between (a) the α-Ga2O3 matrix,(b) twinned α-Ga2O3 domains, and (c) sapphire substrate.FIG. 6. Cross-sectional schematics of α-Ga2O3 films grown on (a) a c-planesubstrate with a striped mask along [1�100] and (b) an r-plane substrate with astriped mask along [�12�10].Journal ofApplied Physics ARTICLE scitation.org/journal/japJ. Appl. Phys. 130, 175304 (2021); doi: 10.1063/5.0068097 130, 175304-5Published under an exclusive license by AIP Publishing 30 January 2024 07:38:50https://aip.scitation.org/journal/japDATA AVAILABILITYThe data that support the findings of this study are availablewithin the article.REFERENCES1D. Shinohara and S. Fujita, Jpn. J. Appl. Phys. 47, 7311 (2008).2M. Oda, R. Tokuda, H. Kambara, T. Tanikawa, T. Sasaki, and T. Hitora, Appl.Phys. Express 9, 021101 (2016).3T. Maeda, M. Okigwa, Y. Kato, I. Takahashi, and T. Shinohe, AIP Adv. 10,125119 (2020).4M. Lee, M. Yang, H. Lee, H. U. Lee, H. Lee, H. Son, and U. J. Kim, Mater. Sci.Semicond. Process. 123, 105565 (2021).5J. Bae, D. Jeon, J. Park, and J. Kim, J. Vac. Sci. Technol. A 39, 033410 (2021).6S. Fujita and K. Kaneko, J. Cryst. Growth 401, 588 (2014).7R. Wakabayashi, K. Yoshimatsu, M. Hattori, J. Lee, O. Sakata, and A. Ohtomo,Cryst. Growth Des. 21, 2844 (2021).8K. Kaneko, S. Fujita, and T. Hitora, Jpn. J. Appl. Phys. 57, 02CB18 (2018).9S. Kan, S. Takemoto, K. Kaneko, I. Takahashi, M. Sugimoto, T. Shinohe, andS. Fujita, Appl. Phys. Lett. 113, 212104 (2018).10J. G. Hao, H. H. Gong, X. H. Chen, Y. Xu, F. Ren, S. L. Gu, R. Zhang,Y. D. Zheng, and J. D. Ye, Appl. Phys. Lett. 118, 261601 (2021).11K. Kaneko, Y. Masuda, S. Kan, I. Takahashi, Y. Kato, T. Shinohe, and S. Fujita,Appl. Phys. Lett. 118, 102104 (2021).12FLOSFIA and Kyoto University, News release, July 13, 2018, see http://flosfia.com/20180713/.13FLOSFIA and Kyoto University, News release, July 13, 2019, see https://flosfia.com/20191202-2/.14D. Machon, P. F. McMillan, B. Xu, and J. Dong, Phys. Rev. B 73, 094125 (2006).15A. Kuramata, K. Koshi, S. Watanabe, Y. Yamaoka, T. Masui, andS. Yamakoshi, Jpn. J. Appl. Phys. 55, 1202A2 (2016).16Z. Galazka, K. Irmscher, R. Uecker, R. Bertram, M. Pietsch, A. Kwasniewski,M. Naumann, T. Schulz, R. Schewski, D. Klimm, and M. Bickermann, J. Cryst.Growth 404, 184 (2014).17K. Hoshikawa, E. Ohba, T. Kobayashi, J. Yanagisawa, C. Miyagawa, andY. Nakamura, J. Cryst. Growth 447, 36 (2016).18E. G. Víllora, K. Shimamura, Y. Yoshikawa, K. Aoki, and N. Ichinose, J. Cryst.Growth 270, 420 (2004).19Y. Oshima, E. G. Víllora, and K. Shimamura, Appl. Phys. Express 8, 055501(2015).20A. I. Pechnikov, S. I. Stepanov, A. V. Chikiryaka, M. P. Scheglov,M. A. Odnobludov, and V. I. Nikolaev, Semiconductors 53, 780 (2019).21H. Son and D. W. Jeon, J. Alloys Compd. 773, 631 (2019).22Z. Cheng, M. Hanke, P. Vogt, O. Bierwagen, and A. Trampert, Appl. Phys.Lett. 111, 162104 (2017).23Y. Oshima, K. Kawara, T. Shinohe, T. Hitora, M. Kasu, and S. Fujita, APLMater. 7, 022503 (2019).24K. Akaiwa, K. Kaneko, K. Ichino, and S. Fujita, Jpn. J. Appl. Phys. 55, 1202BA(2016).25H. Son, Y. Choi, J. Ha, S. Jung, and D. Jeon, Cryst. Growth Des. 19, 5105(2019).26Y. Oshima, K. Kawara, T. Oshima, M. Okigawa, and T. Shinohe, Jpn. J. Appl.Phys. 59, 025512 (2020).27A. N. Cha, S. Bang, H. Rho, H. Bae, D. W. Jeon, J. W. Ju, S. K. Hong, andJ. S. Ha, Appl. Phys. Lett. 115, 091605 (2019).28R. Jinno, N. Yoshimura, K. Kaneko, and S. Fujita, Jpn. J. Appl. Phys. 58,120912 (2019).29G. T. Dang, T. Yasuoka, and T. Kawaharamura, Appl. Phys. Lett. 119, 041902(2021).30A. Usui, H. Sunakawa, A. Sakai, and A. A. Yamaguchi, Jpn. J. Appl. Phys. 36,L899 (1997).31K. Kawara, Y. Oshima, M. Okigawa, and T. Shinohe, Appl. Phys. Express 13,075507 (2020).Journal ofApplied Physics ARTICLE scitation.org/journal/japJ. Appl. Phys. 130, 175304 (2021); doi: 10.1063/5.0068097 130, 175304-6Published under an exclusive license by AIP Publishing 30 January 2024 07:38:50https://doi.org/10.1143/JJAP.47.7311https://doi.org/10.7567/APEX.9.021101https://doi.org/10.7567/APEX.9.021101https://doi.org/10.1063/5.0028985https://doi.org/10.1016/j.mssp.2020.105565https://doi.org/10.1016/j.mssp.2020.105565https://doi.org/10.1116/6.0000940https://doi.org/10.1016/j.jcrysgro.2014.02.032https://doi.org/10.1021/acs.cgd.1c00030https://doi.org/10.7567/JJAP.57.02CB18https://doi.org/10.1063/1.5054054https://doi.org/10.1063/5.0047710https://doi.org/10.1063/5.0027297http://flosfia.com/20180713/http://flosfia.com/20180713/http://flosfia.com/20180713/https://flosfia.com/20191202-2/https://flosfia.com/20191202-2/https://flosfia.com/20191202-2/https://flosfia.com/20191202-2/https://doi.org/10.1103/PhysRevB.73.094125https://doi.org/10.7567/JJAP.55.1202A2https://doi.org/10.1016/j.jcrysgro.2014.07.021https://doi.org/10.1016/j.jcrysgro.2014.07.021https://doi.org/10.1016/j.jcrysgro.2016.04.022https://doi.org/10.1016/j.jcrysgro.2004.06.027https://doi.org/10.1016/j.jcrysgro.2004.06.027https://doi.org/10.7567/APEX.8.055501https://doi.org/10.1134/S1063782619060150https://doi.org/10.1016/j.jallcom.2018.09.230https://doi.org/10.1063/1.4998804https://doi.org/10.1063/1.4998804https://doi.org/10.1063/1.5051058https://doi.org/10.1063/1.5051058https://doi.org/10.7567/JJAP.55.1202BAhttps://doi.org/10.1021/acs.cgd.9b00454https://doi.org/10.35848/1347-4065/ab6fafhttps://doi.org/10.35848/1347-4065/ab6fafhttps://doi.org/10.1063/1.5100246https://doi.org/10.7567/1347-4065/ab55c6https://doi.org/10.1063/5.0057704https://doi.org/10.1143/JJAP.36.L899https://doi.org/10.35848/1882-0786/ab9fc5https://aip.scitation.org/journal/jap