Structure and Electro-Optical Properties of Thin Films Grown by ...

Materials Transactions, Vol. 51, No. 2 (2010) pp. 219 to 226 Special Issue on Development and Fabrication of Advanced Materials Assisted by Nanotechnology and Microanalysis #2010 The Japan Institute of Metals

Structure and Electro-Optical Properties of Thin Films Grown by Alternate Atomic Layer Deposition of ZnO and Al2 O3 on the Sapphire Substrate ˇ eh1 , Hsing-Chao Chen2; *1 , Miin-Jang Chen2; *2 , Jer-Ren Yang2 and Makoto Shiojiri3 Miran C 1

Nanostructured materials, Jozˇef Stefan Institute, Jamova cesta 39, 1000 Ljubljana, Slovenia Institute of Materials Science and Engineering, National Taiwan University, Taipei, Taiwan 106, R. O. China 3 Professor Emeritus of Kyoto Institute of Technology, 1-297 Wakiyama, Kyoto 618-0091, Japan 2

Al-doped ZnO thin films were prepared on the (0001) sapphire (c-Al2 O3 ) substrates by atomic layer deposition (ALD) using alternating pulses of Zn(C2 H5 )2 , Al(CH3 )3 and H2 O precursors and post-deposition high-temperature annealing. Photoluminescence (PL) spectroscopy showed that the threshold of stimulated emission decreases with increasing Al concentration, from 49.2 kW/cm2 of the ZnO film to 12.2 kW/ cm2 of the ZnO film nominally containing 4% Al. This reduction is attributable to the increase in the optical scattering resulting from segregation of excess Al in heavily Al-doped ZnO films. The structure of these films was investigated by analytical scanning-transmission electron microscopy (STEM) as well as X-ray diffraction (XRD). It was revealed that a single crystal ZnO layer containing a small amount of Al is formed with the orientation relation with respect to the c-Al2 O3 : [0001]ZnO == [0001]Al2 O3 and [011 0]ZnO == [21 1 0]Al2 O3 , and that a polycrystalline ZnAl2 O4 layer is formed between the ZnO layer and the c-Al2 O3 substrate. The electron microscopy observation accounts for the results of the electro-optical experiments. The growth mechanism of the observed two layers is discussed. [doi:10.2320/matertrans.MC200902] (Received July 29, 2009; Accepted October 15, 2009; Published December 2, 2009) Keywords: atomic layer deposition, ultra-violet light emission diode, zinc-oxide, X-ray diffraction, photoluminescence spectroscopy, analytical scanning-transmission electron microscopy

1.

Introduction

Zinc oxide is one of promising materials for ultraviolet (UV) photonic devices because of the high direct band-gap energy of 3.37 eV at room temperature and the large exciton binding energy up to 60 meV, exhibiting various chemical, environmental and economical advantages.1–13) Many techniques have been utilized for growing high-quality ZnO films, such as molecular beam epitaxy,1–3) metal-organic chemical vapor deposition,4,5) hybrid beam deposition,6) chemical vapor deposition7,8) and pulsed laser deposition (PLD).9,10) Another prospective method for preparing ZnO films is atomic layer deposition (ALD).11–13) ALD is a surface-controlled process for depositing materials with atomic-layer accuracy. One of the presenting authors and his colleagues prepared high-quality ZnO thin films on the (0001)-sapphire (c-Al2 O3 ) substrates by ALD, and observed stimulated emission from the ZnO films.14) They have also fabricated light-emitting diodes (LEDs) composed of n-ZnO/ p-GaN heterojunction,15) n-ZnO:Al/SiO2 -ZnO nanocomposite/p-GaN:Mg16,17) and n-ZnO/ZnO nanodots-SiO2 composite/p-AlGaN by ALD,18) all of which emitted significant UV electroluminescence from the n-ZnO layers. In addition to the application in UV photonic devices, ZnO is also known as a transparent conductive oxide. The doping of ZnO with group-III impurities such as Al improves electrical conductivity. Recently, Wang et al.,19) who investigated undoped ZnO thin films grown on the c-Al2 O3 substrate at 600 C by PLD, have reported an increase in electron concentration and lowering of the threshold of optical-pumped stimulated *1Graduate

Student, National Taiwan University author, E-mail: [email protected]

*2Corresponding

emission from ZnO by thermal annealing (900 C, 2 h). They attributed this phenomenon to diffusion of Al from Al2 O3 into ZnO, assuming the formation of a ZnAl2 O4 crystalline layer at the ZnO/Al2 O3 interface. In our work, we employed an ALD technique followed by high-temperature annealing to prepare heavily Al-doped ZnO (ZnO:Al) films. The electrical and optical properties, as well as the structure of the prepared films were investigated to elucidate the effect and behavior of Al doping. 2.

Experiment

Heavily Al-doped ZnO thin films were deposited on the c-Al2 O3 substrate at a temperature as low as 180 C by ALD using alternating pulses of Zn(C2 H5 )2 (Diethylzinc, DEZn), Al(CH3 )3 (Trimethylaluminum, TMA) and H2 O vapor in an N2 carrier gas flow. The ALD process consisted of two kinds of cycles; cycle I contained the following sequence: DEZn, 0:01 s ! N2 purge, 5 s ! H2 O, 0:1 s ! N2 purge, 5 s for the growth of ZnO, and cycle II contained TMA, 0:03 s ! N2 purge, 5 s ! H2 O, 0:1 s ! N2 purge, 5 s for the doping of Al. The total number of the ALD cycles was 1500, where one cycle of II was performed every 49 cycles of I to prepare nominal ZnO:Al (2%) film, as shown in Fig. 1. To prepare nominal ZnO:Al (4%) film, one cycle of II was performed every 24 cycles of I. After deposition, the films were annealed at 1000 C for 2 h in N2 atmosphere to homogenize the composites and improve the crystallinity. The post-annealed ZnO:Al films were analyzed by means of photoluminescence (PL), Hall effect measurement with the van der Pauw configuration, X-ray diffraction (XRD) and analytical scanning-transmission electron microscopy (STEM). The room temperature PL spectra of the ZnO:Al

ˇ eh, H.-C. Chen, M.-J. Chen, J.-R. Yang and M. Shiojiri M. C

220

Fig. 1 Atomic layer deposition (ALD) of heavily Al-doped ZnO thin films. The ALD for the ZnO:Al (2%) film is indicated.

films were measured using a fourth harmonic Q-switched Nd:YAG laser (266 nm, 10 ns, 15 Hz) as the excitation source, in the standard backscattering configuration. For analytical STEM observation, the specimens with a lowindex zone axis were embedded into Ti rings, ground, dimpled, and ion-milled in two steps: until perforation they were ion-milled at 4 keV, beam current of 1.0 mA, incident angle of 10 (Bal-Tec RES010) and additionally ion-milled at low-energy on both sides under oscillation at 300 eV (Technoorg Linda, model IV5).20) The additional low-energy ion-milling significantly reduced the amorphous layer and artifacts caused by the sputtering. STEM, high-resolution transmission electron microscopy (HRTEM) and energy dispersive X-ray spectroscopy (EDXS) were performed in a JEM-2010F TEM/STEM, operated at 200 keV, equipped with a pole piece of CS ¼ 0:48 mm. 3.

Results and Discussion

Before structural investigation, we examined the electrooptical properties of the prepared specimens. PL spectra from the ZnO film and ZnO:Al (4%) film are shown in Figs. 2(a) and 2(b), respectively. The spontaneous emission, which is associated with the free excitons or bond excitons,21,22) was observed at 383 nm with a low excitation intensity. Increase in excitation intensity causes the exciton concentration to reach the Mott density, where the excitons overlap with each other, lose their individual character and become electronhole plasma (EHP). Hence, spectral peaks attributed to EHP appear around 394 nm at high excitation intensities. With further increase in the excitation intensity, the EHP peaks gradually dominate the emission spectrum and shift toward lower energy due to the band-gap renormalization, which are similar to the behavior previously reported.23–25) The insets show the plots of the integrated PL intensity versus the excitation intensity. The integrated PL intensity increases rapidly as the excitation intensity is greater than a threshold value. The appearance of the spectral peaks, as well as significant increase in the emission intensity after the threshold indicate that the stimulated emission takes place in the ZnO:Al film. The thresholds for stimulated emission of the ZnO film and the ZnO:Al (4%) film are 49.2 kW/cm2 and 12.2 kW/cm2 , respectively. Since the diameter of the

Fig. 2 Photoluminescence (PL) spectra of (a) the ZnO film and (b) ZnO:Al (4%) film. The insets are the integrated PL intensity as a function of the excitation intensity.

incident laser beam used in this study was about 1 mm, which is much larger than the film thickness, it is likely that the stimulated emission was caused by the closed-loop paths in the direction parallel to the film surface via multiple scattering between optical scattering centers and crystalline grains. The scattering of the stimulated emission propagating in the in-plane direction might lead to the observed stimulated emission in the direction perpendicular to the film surface. Table 1 summarizes electrical properties of the ZnO:Al films characterized by the Hall effect measurement, together with the threshold for stimulated emission. All samples exhibited n-type conductivity. The electron concentrations are of the same order of magnitude (1018 cm3 ) and only slightly decrease with increasing Al content, suggesting that Al concentration is almost the same among the samples and that a large amount of Al atoms may segregate in heavily doped ZnO:Al films due to the solid solubility limit of Al in ZnO. The solid solubility limit was estimated to be 2 mol% from XRD and Raman spectroscopy by Yoo et al.26) The segregation of Al might cause optical scattering centers in the ZnO:Al films. The increase in optical scattering centers due to Al doping may facilitate the formation of closed-loop paths for light through multiple optical scattering, leading to the decrease in the threshold of stimulated emission.27,28) Additionally, it is clearly seen from Fig. 2(b) that there are

Structure and Electro-Optical Properties of Thin Films Grown by Alternate ALD of ZnO and Al2 O3

221

Table 1 Hall effect measurement of the ZnO:Al films and threshold for PL stimulate emission. Sample

Carrier Mobility Resistivity Threshold concentration (cm2 /Vs) ( cm) (kW/cm2 ) 3 (cm )

ZnO

2:76 1018

16.1

0.14

49.2

*

ZnO:Al (2%)

2:37 1018

48.5

0.054

29.9

*

ZnO:Al (4%)

1:16 1018

13.5

0.40

12.2

*

ZnO (900 C, 2 h)

1:6 1019

79

32

Ref. 19)

The present samples heat-treated at 1000 C for 2 h.

Fig. 3 (a) X-ray diffraction patterns and (b) FWHM of the (0002) peaks of the ZnO, ZnO:Al (2%), and ZnO:Al (4%) films.

Fig. 4 (a) Bright-field (BF) and (b) high-angle dark-field (HAADF) STEM images of ZnO:Al(2%). Arrowheads indicate dislocations.

many narrow spectral peaks appearing upon the emission spectra of the ZnO:Al (4%) film. On the other hand, the undoped ZnO exhibits relatively smooth emission spectra (Fig. 2(a)). The appearance of the narrow spectral peaks upon the spectra of the ZnO:Al (4%) may be attributed to the constructive closed-loop paths due to optical scattering provided by the Al segregation in the heavily doped ZnO:Al films.29) In the Table 1 the data reported by Wang et al.19) are also included. It is not obvious why ZnO layer prepared by Wang et al. had a higher carrier concentration of 1:6 1019 cm3 as compared to our samples, while it had the same order of the threshold with the present samples. In order to explain the above-mentioned electro-optical and electrical results, XRD and HRTEM/STEM investigations were performed. Figure 3(a) shows XRD patterns of the post-annealed ZnO:Al films. The appearance of strong ZnO (0002) peaks reveals the preferential (0002)-orientation of the ZnO:Al films in respect with c-Al2 O3 crystal of the substrate. The full-width at half-maximum (FWHM) of the ZnO (0002) K1 peaks is shown in Fig. 3(b). The crystallinity becomes worse with increasing Al concentration that is from 0.08 to 0.12 . Incidentally, the FWHM for the (0002) reflection of the ZnO sample of Wang et al., grown on the cAl2 O3 substrate at 600 C by PLD and post annealed at 900 C

for 2 h, was 0.15 .19) The FWHM of our X-ray measurements was obtained from the line profile of -2 scanning which was recorded along the reciprocal vector g in the reciprocal space, while their FWHM was of the !-scanning rocking curve which was recorded along the direction perpendicular to both, the reciprocal vector and the incident X-ray wave vector. Although the scanning manner is different between the two measurements, Fig. 3(b) seems to confirm the good quality of these ZnO crystals. The broadening of the ZnO (0002) peak might result from the distortion of ZnO crystal lattice due to heavy doping of Al. Since the solid solubility limit of Al in ZnO is only 2 mole%,26) a large amount of Al atoms might segregate from the surrounding ZnO, leading to the dramatic broadening of ZnO (0002) peak in the ZnO:Al (4%). Figures 4(a) and 4(b) show bright-field (BF) and highangle dark-field (HAADF) STEM images of the ZnO:Al (2%), respectively. We can clearly distinguish three different layers, namely, A, B and C. Figure 5 shows BF-STEM image from an area in the ZnO:Al (2%) with overall EDXS spectrum and the corresponding X-ray mappings using characteristic X-rays emitted from Al, Zn, and O and C. The carbon signal originates from adhesive that was used for STEM specimen preparation and which was deposited on the

222


Fig. 5 BF-STEM image, the overall energy EDXS spectrum and the corresponding X-ray mappings, from an area in the ZnO:Al (2%) specimen.

specimen surface during the ion-milling, in spite of careful specimen preparation.20) The mappings reveal that the layer A, characterized only with Al and O emission, is the Al2 O3 layer, the layer B is composed of Zn, Al and O, while the layer C is a ZnO layer. This was additionally confirmed by semi-quantitative EDXS analysis shown in Fig. 6. The probe size that was used for the EDX spectroscopy was 0:5 nm. However, all the EDX spectra were acquired from frames and not from points (spot mode) in order to obtain an average composition from a relatively larger area.29) The beambroadening effect in this case did not affect the peak intensities and/or the ratios in the EDXS spectra. The EDXS analysis was performed from an area of 10 10 nm2 , indicated by regions p1p4. The spectra sp1, sp2, sp3 and sp4 were acquired from the area p1 in layer A, p2 in layer B, and p3 and p4 in layer C, respectively. The spectra sp3 and sp4 reveal that the ZnO in layer C contains a small amount of Al atoms. Figure 7 shows an HRTEM image of an area including layer B and layer C. Fourier transformation (FT) images from the inset squares are also presented. The FT image from the

layer C (lower part) reveals that the corresponding image comprises the (0002) and (112 0) lattice fringes of the ZnO crystal with the wurtzite structure. Thus, layer C turns out to be ZnO crystal with a little Al content. The images in Fig. 4, especially the HAADF image, reveal that the whole layer C is more or less perfect single crystal containing some inclusions. Dislocations (indicated by arrowheads) trapped by these inclusions exhibit dark contrast in the BF image and bright contrast in the HAADF image. As shown in our previous paper,14) the as-deposited ZnO films grown on the c-Al2 O3 at 180 C by ALD are composed of grains with an average size as small as 34 nm, which was estimated from X-ray diffraction peak using Scherrer’s equation. The contours in layer C in Figs. 7 and 8 indicate strain field, which can be regarded as traces of boundaries between the initial small grains coalesced during the heat treatment at 1000 C for 2 h. The strain is accumulated around the boundaries. The observed dislocations may form in the ZnO during the coalescence process between the initial small grains. The dislocations occurring in the coalescence process are indicated in Fig. 8.


223

Fig. 6 EDXS analysis of the ZnO:Al (2%) specimen. The sp1, sp2, sp3 and sp4 are EDXS spectra acquired from regions p1 in layer A, p2 in layer B, and p3 and sp4 in layer C, respectively.

The layer B is composed of grains of several hundred nm in size, as seen in Figs. 4–6. The FT image from B in Fig. 7 can be indexed as ZnAl2 O4 crystal with the spinel structure with the [011 ] axis along the incident electron beam and the (1 11) plane parallel to the (0001) plane of the ZnO. Apart from electron diffraction results, the chemical composition of the layer B was additionally determined by EDXS. Quantification of the EDXS spectra acquired from the layer B confirmed that the chemical composition of the layer B corresponds to the ZnAl2 O4 spinel-type compound. And finally, the XRD spectrum of the ZnO:Al (2%) presented in a log scale (Fig. 9) also reveals the presence of small intensity peaks that are characteristic for the ZnAlO4 compound. Based on these results it was concluded that the layer B is composed of ZnAl2 O4 grains.

It is undoubted that the layer B is the film formed by the reaction of excess Al with ZnO which was assumed from the Hall effect measurements and the strong PL scattering from the ZnO:Al films. The thickness of the layer B is 100 150 nm in the ZnO:Al (2%) and is even larger in the ZnO:Al (4%). Wang et al.19) reported the formation of ZnAl2 O4 layer (6070 nm thick) at the ZnO/Al2 O3 interface heated for 2 h at 900 C. However, our HRTEM observation has not detected any ZnAl2 O4 layer in a test sample of ZnO/Al2 O3 , prepared by ALD and treated by a rapid thermal annealing (5 min, 950 C). Therefore, the formation of the ZnAlO4 phase was caused by the long time heat treatment (1000 C for 2 h). In the ALD, ZnO (and also Al2 O3 ) molecules are chemisorbed under an orientative influence of the lattices of the c-Al2 O3 substrate or the

224


Fig. 7 HRTEM image of layers B and C in the ZnO:Al (2%) and FT images from the areas indicated by squares in the HRTEM image. The diffraction spots from B and C can be indexed as a ZnAl2 O4 crystal (gahnite) with the [011 ] axis and a ZnO crystal (zincite) with the [1 100] axis, respectively, along the incident electron beam.

underlying ZnO. This is supposed from our previous X-ray measurement of the as-deposited ZnO film on the cAl2 O3 ,14) which exhibited a broad ZnO 0002 peak with a FWHM of 0.245 that decreased to 0.084 (as seen in Fig. 3(b)) by the post annealing. Therefore, it is assumed that in the early stage of the heat treatment ZnO crystalline embryos nucleate, coalesce with each other, and grow to large grains, keeping the following orientation relation with respect to the c-Al2 O3 : [0001]ZnO == [0001]Al2 O3 and [011 0]ZnO == [21 1 0]Al2 O3 . The ZnO gradually becomes almost a perfect single crystal (layer C) by successive coalescence. It is disordered only near the interface with the Al2 O3 substrate because of a lattice misfit as large as 18% with Al2 O3 . Al atoms diffuse into the ZnO layer not only from the interleaved Al layers but also from the substrate

Al2 O3 . Since the solid solubility of Al in ZnO is limited, excess Al atoms react with the ZnO to form ZnAl2 O4 crystals, especially near the ZnO/Al2 O3 interface. The nucleation of the ZnAl2 O4 crystals with the spinel-type structure can occur with several orientations (as seen in Fig. 9) on the natal ZnO crystals. The Al atoms are successively supplied and consequently the ZnAl2 O4 crystals grow to form layer B. During the coalescence process, dislocations can occur in layer C, as seen in Fig. 8. ZnAl2 O4 crystals may nucleate on these dislocations or remained boundaries in the C, by supplying Al atoms from sandwiched Al layer. They become inclusions appearing in C in Fig. 4. The inclusions can be identified as ZnAl2 O4 because they have almost the same contrast as that of the layer B in the Z-contrast HAADF-STEM image in Fig. 4(b).30)


225

for electrical devices can be obtained by the optimization of thickness of layer B and the repression of ZnAl2 O4 inclusions in layer C, which are controllable by Al cycles in ALD and the heat annealing. 4.

Fig. 8 Dislocations occurring during the coalescence process to form layer C. The dislocations are enclosed by circles.

Fig. 9 X-ray diffraction pattern of the ZnO:Al (2%) film, reproduced in log scale.

The thickness of layer B and the number (and perhaps size) of the inclusions in the layer C increase with increasing nominal Al concentration in the ALD. The substantial broadening of the ZnO (0002) peak for the ZnO:Al (4%) (as seen in Fig. 3(b)) is ascribed to strain of ZnO lattices in the layer C, which is accumulated around the grain boundaries, as a result of coalescence inhibited by Al excess, and also around the ZnAl2 O4 inclusions. The effect of Aldoping in ZnO in layer C on PL (as seen in Fig. 2) and electrical properties (as seen in Table 1) are significant. The Hall mobility, which increased once with Al-doping, however, reduced in the ZnO:Al (4%) film (as seen in Fig. 3). This is ascribed to the thickened ZnAl2 O4 layer (B) and the increase of inclusions in layer C. The concentration of Al solved in the ZnO crystals in layer C might be the same between ZnO:Al (2%) and ZnO:Al (4%) films due to the solid solubility limit of Al in ZnO, resulting in almost the same carrier concentration. Thus, analytical electron microscopy has elucidated the electrical and optical properties of the ZnO:Al films acquired by PL and Hall effect measurements. This indicates that the ZnO:Al film useful

Conclusion

We investigated the structure of Al-doped ZnO films using X-ray diffraction (XRD) and analytical scanningtransmission electron microscopy (STEM) to elucidate their electro-optical properties that were also examined in the present study. The conclusions can be summarized as follows: (1) ZnO, ZnO:Al (2%) (nominally) and ZnO:Al (4%) (nominally) films were prepared on c-Al2 O3 substrate by alternate atomic layer deposition (ALD) of ZnO and Al2 O3 together with the high temperature postdeposition annealing (1000 C for 2 h). (2) PL measurements revealed that an increase in Al concentration reduces the threshold of stimulated emission, from 49.2 kW/cm2 for the ZnO film to 12.2 kW/cm2 for ZnO:Al (4%) film, which is attributed to the segregation of excess Al in the heavily doped ZnO:Al films. The carrier (electron) concentration is almost the same between the prepared films, while the Hall mobility increases due to Al-doping in the ZnO:Al (2%) film but then decreases in the ZnO:Al (4%) films. (3) XRD revealed the (0002)-oriented growth of ZnO:Al films on the c-Al2 O3 substrate. The FWHM of the ZnO (0002) K1 peaks indicated high-crystallinity of formed ZnO films, however the crystallinity becomes worse with increasing Al concentration. (4) Analytical STEM observations, including EDXS analysis, as well as BF- and HAADF-STEM, showed that ZnO:Al films comprise two layers. One is a single crystal ZnO layer with the following orientation relation with respect to the c-Al2 O3 : [0001]ZnO == [0001]Al2 O3 and [011 0]ZnO == [21 1 0]Al2 O3 . This ZnO layer also contains dislocations and small inclusions of ZnAl2 O4 phase. The other layer is a polycrystalline ZnAl2 O4 layer that is formed between the ZnO layer and c-Al2 O3 substrate. The ZnO layer is doped with a limited amount of Al, and excess Al atoms are expelled from the ZnO initial grains to form the ZnAl2 O4 layer or inclusions in the ZnO layer. The STEM observations consistently accounted for the electro-optical results shown in (2). The growth of the observed layers was also elucidated. The results of our study identified the most important parameters that determine the properties of ZnO:Al films used for photonic devices. Furthermore, by tailoring processing parameters, it is evident that ALD is one of the most powerful techniques for preparation of thin films with desirable physical properties. Acknowledgement We thank Ms Medeja Gec, Jozˇef Stefan Institute, for the preparation of samples for STEM observations.

226


REFERENCES 1) A. Tsukazaki, A. Ohtomo, T. Onuma, M. Ohtani, T. Makino, M. Sumiya, K. Ohtani, S. F. Chichibu, S. Fuke, Y. Segawa, H. Ohno, H. Koinuma and M. Kawasaki: Nat. Mater. 4 (2005) 42–46. 2) S. J. Jiao, Z. Z. Zhang, Y. M. Lu, D. Z. Shen, B. Yao, J. Y. Zhang, B. H. Li, D. X. Zhao, X. W. Fan and Z. K. Tang: Appl. Phys. Lett. 88 (2006) 031911 (3 pages). 3) Y. Chen, D. Bagnall and T. Yao: Mater. Sci. Eng. B 75 (2000) 190– 198. 4) W. Z. Xu, Z. Z. Ye, Y. J. Zeng, L. P. Zhu, B. H. Zhao, L. Jiang, J. G. Lu and H. P. He: Appl. Phys. Lett. 88 (2006) 173506 (3 pages). 5) W. Liu, S. L. Gu, J. D. Ye, S. M. Zhu, S. M. Liu, X. Zhou, R. Zhang, Y. Shi and Y. D. Zheng: Appl. Phys. Lett. 88 (2006) 092101 (3 pages). 6) Y. Ryu, T. S. Lee, J. A. Lubguban, H. W. White, B. J. Kim, Y. S. Park and C. J. Youn: Appl. Phys. Lett. 88 (2006) 241108 (3 pages). 7) Ya. I. Alivov, E. V. Kalinina, A. E. Cherenkov, D. C. Look, B. M. Ataev, A. K. Omaev, M. V. Chukichev and D. M. Bagnall: Appl. Phys. Lett. 83 (2003) 4719–4721. 8) Ya. I. Alivov, J. E. Van Nostrand, D. C. Look, M. V. Chukichev and B. M. Ataev: Appl. Phys. Lett. 83 (2003) 2943–2945. 9) D. J. Rogers, F. Hosseini Teherani, A. Yasan, K. Minder, P. Kung and M. Razeghi: Appl. Phys. Lett. 88 (2006) 141918 (3 pages). 10) X. Li. Guo, J. H. Choi, H. Tabata and T. Kawai: Jpn. J. Appl. Phys. 40 (2001) L177–L180. 11) M. Ritala and M. Leskela: Nanotechnology 10 (1999) 19–24. 12) L. Niinisto¨, J. Païva¨saari, J. Niinisto¨, M. Putkonen and M. Nieminen: Phys. Stat. Sol. (a) 201 (2004) 1443–1452. 13) J. Lim, K. Shin, H. W. Kim and C. Lee: J. Lumin. 109 (2004) 181– 185. 14) H. C. Chen, M. J. Chen, M. K. Wu, Y. C. Cheng and F. Y. Tsai: IEEE J. Selected Topics Quantum Electronics 14 (2008) 1053–1057.

15) H. C. Chen, M. J. Chen, M. K. Wu, W. C. Li, H. L. Tsai, J. R. Yang, H. Kuan and M. Shiojiri: IEEE J. Quantum Electron. (in print). 16) M. K. Wu, Y. T. Shih, M. J. Chen, H. Kuan, J. R. Yang and M. Shiojiri: IEEE Photonics Technology Lett. 20 (2008) 1772–1774. 17) M. K. Wu, Y. T. Shih, M. J. Chen, J. R. Yang and M. Shiojiri: Phys. Stat. Sol. Rapid Res. Lett. 3 (2009) 88–90. 18) Y. T. Shih, M. K. Wu, W. C. Li, H. Kuan, J. R. Yang, M. Shiojiri and M. J. Chen: Nanotechnology 20 (2009) 165201 (8 pages). 19) Y. G. Wang, N. Ohashi, Y. Wada, I. Sakaguchi, T. Ohgaki and H. Haneda: J. Appl. Phys. 100 (2006) 023524 (6 pages). 20) M. Gec, M. Cˇeh and M. Shiojiri: Proc. Seventh Multinational Congress on Mocroscopy, (Portoroz, Slovenia, 2005) pp. 385–386. 21) X. Q. Zhang, Ikuo Suemune, H. Kumano, J. Wang and S. H. Huang: J. Appl. Phys. 96 (2004) 3733–3736. 22) B. K. Meyer, H. Alves, D. M. Hofmann, W. Kriegseis, D. Forster, F. Bertram, J. Christen, A. Hoffmann, M. Straßburg, M. Dworzak, U. Haboeck and A. V. Rodina: Phys. Stat. Sol. (b) 241 (2004) 231–260. 23) Y. Segawa, A. Ohtomo, M. Kawasaki, H. Koinuma, Z. K. Tang, P. Yu and G. K. L. Wong: Phys. Stat. Sol. (b) 202 (1997) 669–672. 24) Z. K. Tang, G. K. L. Wong, P. Yu, M. Kawasaki, A. Ohtomo, H. Koinuma and Y. Segawa: Appl. Phys. Lett. 72 (1998) 3270–3272. 25) D. M. Bagnall, Y. F. Chen, Z. Zhu, T. Yao, M. Y. Shen and T. Goto: Appl. Phys. Lett. 73 (1998) 1038–1040. 26) M. H. Yoon, S. H. Lee, H. L. Park, H. K. Kim and M. S. Jang: J. Mat. Sci. Lett. 21 (2002) 1073–1074. 27) F. A. Pinheiro and L. C. Sampaio: Phys. Rev. A 73 (2006) 013826 (4 pages). 28) H. Cao, J. Y. Xu, S. H. Chang and S. T. Ho: Phys. Rev. E 61 (2000) 1985–1989. 29) H. Cao, Y. G. Zhao, H. C. Ong, S. T. Ho, J. Y. Wu and T. P. H. Chang: Appl. Phys. Lett. 73 (1998) 3656–3658. ˇ eh, S. Sˇturm, C. C. Chuo, J. T. Hsu, J. R. Yang and 30) M. Shiojiri, M. C H. Saijo: J. Appl. Phys. 100 (2006) 013110 (7 pages).