SiGe nanorings by ultrahigh vacuum chemical vapor deposition

(1)

SiGe nanorings by ultrahigh vacuum chemical vapor deposition

C.-H. Lee, Y.-Y. Shen, C. W. Liu, S. W. Lee, B.-H. Lin, and C.-H. Hsu

Citation: Applied Physics Letters 94, 141909 (2009); doi: 10.1063/1.3116619 View online: http://dx.doi.org/10.1063/1.3116619

View Table of Contents: http://scitation.aip.org/content/aip/journal/apl/94/14?ver=pdfcov Published by the AIP Publishing

Articles you may be interested in

Ge out diffusion effect on SiGe nanoring formation J. Appl. Phys. 111, 076103 (2012); 10.1063/1.3702812

Germanium-rich silicon-germanium films epitaxially grown by ultrahigh vacuum chemical-vapor deposition directly on silicon substrates

Appl. Phys. Lett. 91, 252111 (2007); 10.1063/1.2825410

Ultrathin low temperature SiGe buffer for the growth of high quality Ge epilayer on Si(100) by ultrahigh vacuum chemical vapor deposition

Appl. Phys. Lett. 90, 092108 (2007); 10.1063/1.2709993

Dependence of substitutional C incorporation on Ge content for Si 1xy Ge x C y crystals grown by ultrahigh vacuum chemical vapor deposition

Appl. Phys. Lett. 77, 3962 (2000); 10.1063/1.1332111

Deposition of polycrystalline Si and SiGe by ultra-high vacuum chemical molecular epitaxy J. Vac. Sci. Technol. A 18, 1196 (2000); 10.1116/1.582324

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 140.113.38.11 On: Wed, 30 Apr 2014 15:23:14

(2)

SiGe nanorings by ultrahigh vacuum chemical vapor deposition

C.-H. Lee,1Y.-Y. Shen,1C. W. Liu,2,a兲 S. W. Lee,3B.-H. Lin,4and C.-H. Hsu4

1_{Department of Electrical Engineering and Graduate Institute of Electronics Engineering,}

National Taiwan University, Taipei 106, Taiwan

2_{Department of Electrical Engineering and Graduate Institute of Electronics Engineering,}

National Taiwan University, Taipei 106, Taiwan and National Nano Device Laboratories, Hsinchu 300, Taiwan

3

Institute of Material Science and Engineering, National Central University, Jhong-Li 32001, Taiwan

4

National Synchrotron Radiation Research Center, Hsinchu 300, Taiwan and Department of Photonics and Institute of Electro-Optical Engineering, National Chiao Tung University, Hsinchu 300, Taiwan

共Received 23 January 2009; accepted 20 March 2009; published online 9 April 2009兲

Formation of SiGe nanorings from Si capped Si0.1Ge0.9quantum dots 共QDs兲 grown at 500 °C by

ultrahigh vacuum chemical vapor deposition was investigated. SiGe nanorings have average diameter, width, and depth of 185, 30, and 9 nm, respectively. Based on both Raman and x-ray diffraction results, the formation of SiGe nanorings can be attributed to Ge outdiffusion from central SiGe QDs during in situ annealing. Moreover, the depth of SiGe nanorings can be controlled by Si cap thickness. The Si cap is essential for nanorings formation. © 2009 American Institute of Physics. 关DOI:10.1063/1.3116619兴

Ringlike structures in III-V compound semiconductor system were widely reported recently.1–3 Nanorings exhibit many interesting phenomena, such as unusual excitation properties and Aharonov–Bohm effect, due to the unique ro-tational symmetry4and the doubly connected topology of the ring,5 respectively. SiGe nanorings grown by ultrahigh vacuum共UHV兲 chemical vapor deposition 共CVD兲 have been reported previously. However, ring structures in the previous reports can only be observed in very limited process window due to specific growth mechanism.6In this work, SiGe nano-rings with controllable depth and well-defined Ge content at edges are demonstrated. The ring structures were character-ized by atomic force microscopy共AFM兲. Strain and Ge con-tent in nanorings were analyzed through x-ray diffraction 共XRD兲 using synchrotron radiation source and high reso-lution Raman spectroscopy. Ge outdiffusion mechanism is found to play a crucial role in the ring formation.

SiGe nanorings were grown by UHV/CVD system at 500 ° C. The base pressure was ⬃10−9 torr. Silane 共SiH4兲

and germane 共GeH4兲 were used as reactant gases. The gas

flows of GeH4 and carrier gas共He兲 were fixed at 5 and 35

SCCM 共SCCM denotes standard cubic centimeters per minute at STP兲, respectively. Ge layers transform into SiGe alloys due to Si/Ge interdiffusion.7After Stranski–Krastanov mode growth of quantum dots 共QDs兲, Si cap with different thickness was deposited. The samples were in situ annealed for 1 h to form ring structures.

Surface morphology of as-grown SiGe QDs grown at 500 ° C is shown in Fig. 1共a兲. Dot density is ⬃6 ⫻108 _cm−2_{, and the average dot height and dot base are 15}

and 115 nm, respectively. After Si cap layer deposition, small dots disappeared from AFM image关Fig.1共b兲兴 and only those with large dot size can be observed. Figure 1共c兲shows the morphology of 4 nm Si capped QDs after 1 h of annealing. Ring structures are evident on the surface and shown by white rings. The density of nanorings from 4 nm Si capped

QDs is⬃1.3⫻108 _cm−2_{, which is similar to the dot density}

共⬃1.2⫻108 _cm−2_{兲 of the sample before annealing within}

10% error bar of measurement. Surface morphology of nano-rings from 2 nm Si capped QDs is also shown in Fig. 1共d兲, and the density and diameter of nanorings are similar to those from 4 nm Si capped QDs. Note that QDs without Si cap cannot be transformed into rings after annealing. After Si capping, large QDs共⬎150 nm兲 are still observable in AFM image, but the contrast of small dots 共⬍90 nm兲 becomes much weaker关Fig.1共b兲兴. Surprisingly, the large QDs have no Si capping, while small dots were buried in Si cap as shown by cross-sectional transmission electron microscopy 共TEM兲 image in Fig. 2. An additional 兵111其 facet appeared at the edge of QDs in Fig.2共a兲, probably due to Ge surface diffu-sion from small dots in neighborhood during long Si cap growth time.8 Note that the growth rate of Si cap is ex-tremely small共0.1 nm/min兲 at 500 °C and growth time is 40

a兲_{Author to whom correspondence should be addressed. Electronic mail:} [email protected].

FIG. 1. AFM images 共5⫻5 ␮m2_{兲 of 共a兲 uncapped, 共b兲 4 nm Si capped} QDs,共c兲 nanorings from 4 nm Si capped QDs, and 共d兲 nanorings from 2 nm Si capped QDs.

APPLIED PHYSICS LETTERS 94, 141909共2009兲

(3)

min. SiGe QDs were in situ annealed during Si cap growth. Both Ge content and strain relaxation in QDs may increase. Moreover, dot height of large QDs at 500 ° C is much larger than Si cap thickness. Si atoms adsorbed directly on QDs may suffer tensile strain due to the larger lattice constants of QDs and tend to diffuse to relatively strain-free region around dots.6 SiGe dots shown on the surface are therefore not covered by Si cap.

Outdiffusion mechanism of InGaAs nanorings has been extensively discussed. The relatively large diffusivity of In is responsible for the formation of nanorings.2,9 Surface diffu-sion of Ge is generally larger than that of Si.10 Figure 3 illustrates the outdiffusion model of SiGe nanorings. Si cap layer only covers the edge of SiGe QDs and can block Ge surface diffusion out of the region covered underneath. Ge surface diffusion is faster than Ge interdiffusion into Si cap,11,12and Ge atoms inside SiGe dots can diffuse outward to the top of Si cap over the edge of QDs 关Figs. 3共b兲 and 3共c兲兴. The Ge outdiffusion continued with annealing and left a hole near the center of former SiGe dots. Ring depth共the distance from the bottom of ring to the ridge at edge兲 can be controlled by Si cap thickness. Thicker Si cap thickness leads to a larger ring depth of nanorings. Note that the Ge surface diffusion of SiGe QDs was reportedly suppressed by the Si cap layer deposition.13 Moreover, the preserving shape and composition of SiGe QDs during Si capping at deposition temperature Tdeplower than the QDs growth temperature of

600 ° C was reported.14In principle, the outdiffusion can be slow and the ring shape should be preserved after Si capping if the growth temperature of Si cap is lower than the nano-rings formation temperature of 500 ° C.

Strain variation between nanorings and QDs can be mea-sured by XRD and Raman measurements. X-ray scattering radial scans along in-plane and surface normal directions were performed to determine the lateral and vertical lattice constants, respectively. Taking both the Poisson ratio and the unstrained SiGe lattice constant15 into account, the ⬃90% Ge content in SiGe QDs can be derived by the XRD results in this work. Figure4shows XRD lateral radial scans across Si 共400兲 and Ge 共400兲 reflections of the uncapped, 4 nm Si capped QDs, and nanorings. The peak centered at⬃67° and is contributed from SiGe islands. An increase of lateral lat-tice parameter, as revealed by observed shift of ⬃−0.3° in Fig. 4, upon Si capping is attributed to the increase of Ge content and/or strain relaxation in QDs. Ge surface diffusion from small dots to large dots may be responsible for the Ge content increase and strain relaxation. When QDs trans-formed into nanorings by annealing, SiGe peak shifts back to the higher angle and the intensity decreases. The increase of compressive strain or/and the decrease of Ge concentration are responsible for the peak shift. Nevertheless, these two effects cannot be distinguished by XRD results since no non-surface reflection can be detected to provide vertical lattice constant. Raman spectroscopy with 488 nm laser excitation on uncapped, Si capped QDs, and nanorings are shown in Fig. 5 with the resolution of 0.2 cm−1_{. Ge–Ge peak of Si}

capped QDs sample is⬃0.5 cm−1_{higher than the uncapped}

QDs sample due to the increase of Ge content in large is-lands, which is similar to XRD results. However, the peak of nanorings sample shifts more to higher wave number 共⬃0.5 cm−1_{兲 as compared with Si capped QDs sample. The}

FIG. 2. TEM images of 4 nm Si capped QDs:共a兲 large QDs and 共b兲 small QDs.

FIG. 3. Outdiffusion mechanism of SiGe nanorings:共a兲 SiGe QDs with Si cap,共b兲 Ge outdiffuse from 2 nm Si capped QDs, and 共c兲 Ge outdiffuse from 4 nm Si capped QDs. AFM profiles show 4 nm Si cap sample has a larger depth共9 nm兲 than the 2 nm Si cap sample 共5 nm兲.

FIG. 4. 共400兲 XRD in-plane radial scan of uncapped QDs, 4 nm Si capped QDs, and nanorings. X-ray wavelength used is 1.5406 Å.

FIG. 5. Raman spectroscopy of uncapped QDs, 4 nm Si capped QDs, and nanorings.

141909-2 Lee et al. Appl. Phys. Lett. 94, 141909共2009兲

(4)

increase of Ge content or/and compressive strain is respon-sible for the blueshift. However, combined with XRD results, the increase of compressive strain is the only possible origin. Ge diffuses gradually from the top of QDs and the left nano-ring has more compressive strain. Note that the bottom of SiGe dots is more compressively strained than the top of QDs according to the simulation work.16

In summary, SiGe nanorings with controllable depth have been investigated. The Ge outdiffusion mechanism is proposed to nanorings formation at 500 ° C. Note that the true quantum ring nature has been demonstrated by photolu-minescence and spectral response measurement in the III-V system.2,3However, in our case, the interband detector based on the quantum behavior of SiGe nanorings is still under investigation.

This work was supported by National Science Council of Republic of China under Grant No. 95-2221-E-002-370.

1_{J. M. Garcia, G. Medeiros-Ribeiro, K. Schmidt, T. Ngo, J. L. Feng, A.} Lorke, J. Kotthaus, and P. M. Petroff,Appl. Phys. Lett. 71, 2014共1997兲. 2_{D. Granados and J. M. Garcia,}_{Appl. Phys. Lett.} _{82, 2401}_共2003兲. 3_{J.-H. Dai, Y.-L. Lin, and S.-C. Lee,} _{IEEE Photonics Technol. Lett.} _19,

1511共2007兲.

4_{R. J. Warburton, C. Schulhauser, D. Haft, C. Schäflein, K. Karrai, J. M.} Garcia, W. Schoenfeld, and P. M. Petroff, Phys. Rev. B 65, 113303

共2002兲.

5_{A. Lorke, R. J. Luyken, A. O. Govorov, J. P. Kotthaus, J. M. Garcia, and} P. M. Petroff,Phys. Rev. Lett. 84, 2223共2000兲.

6_{S. W. Lee, L. J. Chen, P. S. Chen, M.-J. Tsai, C. W. Liu, T. Y. Chien, and} C. T. Chia,Appl. Phys. Lett. 83, 5283共2003兲.

7_{M. H. Liao, C.-H. Lee, T.-A. Hung, and C. W. Liu,}_{J. Appl. Phys.} _102, 053520共2007兲.

8_{T. I. Kamins, E. C. Carr, R. S. Willams, and S. J. Rosner,}_{J. Appl. Phys.} 81, 211共1997兲.

9_{A. Lorke, R. J. Luyken, J. M. Garcia, and P. M. Petroff,} _{Jpn. J. Appl.} Phys., Part 1 40, 1857共2001兲.

10_{L. Huang, F. Liu, G.-H. Lu, and X. G. Gong,}_{Phys. Rev. Lett.} _{96, 016103} 共2006兲.

11_{E. Chason, T. M. Mayer, B. K. Kellerman, D. T. Mcllroy, and A. J.} Howard,Phys. Rev. Lett. 72, 3040共1994兲.

12_{G. L. McVay and A. R. DuCharme,}_{Phys. Rev. B} _{9, 627}_共1974兲. 13_{H. Liu and R. Huang,}_{J. Appl. Phys.} _{97, 113537}_共2005兲.

14_{M. De Seta, G. Capellini, L. Di Gaspare, F. Evangelisti, and F. D’Acapito,} J. Appl. Phys. 100, 093516共2006兲.

15_{J. M. Hartmann, B. Gallas, J. Zhang, and J. J. Harris,} _{Semicond. Sci.} Technol. 15, 370共2000兲.

16_{A. Marzegalli, V. A. Zinovyev, F. Montalenti, A. Rastelli, M. Stoffel, T.} Merdzhanova, O. G. Schmidt, and L. Miglio,Phys. Rev. Lett.99, 235505

共2007兲.

141909-3 Lee et al. Appl. Phys. Lett. 94, 141909共2009兲