ARTICLE pubs.acs.org/crystal
The Growth of an Epitaxial ZnO Film on Si(111) with a Gd2O3(Ga2O3) Buffer Layer B. H. Lin,†,‡ W. R. Liu,#,‡ S. Yang,† C. C. Kuo,† C.-H. Hsu,*,‡,† W. F. Hsieh,*,†,# W. C. Lee,^ Y. J. Lee,^ M. Hong,^ and J. Kwo§,|| †
Department of Photonics and Institute of Electro-Optical Engineering, National Chiao Tung University, Hsinchu 30010, Taiwan Division of Scientific Research, National Synchrotron Radiation Research Center, Hsinchu 30076, Taiwan # Department of Photonics & Institute of Electro-Optical Science and Engineering, National Cheng Kung University, Tainan 70101, Taiwan ^ Department of Materials Science & Engineering, National Tsing Hua University, Hsinchu 30013, Taiwan § Center for Condensed Matter Sciences, National Taiwan University, Taipei 10617, Taiwan Department of Physics, National Tsing Hua University, Hsinchu 30013, Taiwan
)
‡
)
)
ABSTRACT: High-quality (0001)-oriented ZnO epitaxial films have been grown by pulsed laser deposition on Si(111) buffered with a nanometer-thick Gd2O3(Ga2O3), GGO, layer. The structural characteristics of the ZnO epi-layers were studied by X-ray diffraction and transmission electron microscopy (TEM). The inplane epitaxial relationship between the wurtzite ZnO, cubic GGO, and cubic Si follows {1010}ZnO {422}Gd2O3 {422}Si and the ZnO/GGO interface can be described by the domain matching epitaxy. TEM contrast analysis reveals that the major defect structures in the ZnO films are edge-type threading dislocations and intrinsic basal plane stacking faults. All the ZnO epi-films studied are under a tensile biaxial strain, but no cracks were found. Temperature-dependent photoluminescence results show the excellent optical properties of the obtained ZnO layers, and the origins of the spectral features are discussed.
’ INTRODUCTION Wurtzite ZnO is a II VI semiconductor with a wide direct band gap of 3.37 eV at room temperature (RT), suitable for photonic applications in UV and blue spectral range. One of the most attractive features of ZnO is its large exciton binding energy (60 meV), which is almost three times larger than that of GaN (25 meV).1 Such a large exciton binding energy allows stable existence of excitons and efficient excitonic emission at RT. Different growth methods, including pulsed laser deposition (PLD),2,3 molecular beam epitaxy (MBE),4,5 atomic layer deposition (ALD),6,7 metal organic chemical vapor deposition (MOCVD),8,9 and vapor phase epitaxy (VPE)10 have been employed to grow ZnO films. Besides the most commonly adopted substrate, sapphire, other substrates or buffer layers with small lattice mismatch, such as γ-LiAlO2,11 ScAlMgO4,1 and GaN12 14 have been adopted to grow high-quality ZnO with low defect densities. ZnO-based nanomaterials with all kinds of morphology, varying from nanowires to nanobelts and complicated hierarchical nanostructures,15 17 have been fabricated, and their unique and novel applications in catalysis, piezoelectricity, and optoelectronics have been demonstrated. Recently, much attention has been paid to heteroepitaxially grown ZnO on Si substrates, because of low costs, excellent quality, and large-area availability of Si wafers and the unique possibility of integrating well-established Si electronics with ZnO-based optoelectronic devices. However, direct growth r 2011 American Chemical Society
of ZnO on Si usually results in polycrystalline or textured films due to the formation of amorphous oxides on the Si surface.18,19 The other troublesome issue for the growth of ZnO epi-films on Si is the formation of cracks. The large mismatches in lattice (aSi = 5.431 Å, aZnO = 3.2438 Å, and cZnO = 5.2036 Å) and thermal expansion coefficient between ZnO (Ra = 6.5 10 6 K 1) and Si (Ra = 2.6 10 6 K 1) lead to the development of significant tensile stress and consequently the formation of cracks as the layer thickness exceeds a critical value. Although significant efforts have been put forth to use various materials, such as Y2O3, γ-Al2O3, Lu2O3, and Sc2O3, as the buffer layers,20 23 the growth of high-quality and crack-free ZnO epi-films on Si is still regarded as a difficult task. Gd2O3(Ga2O3), GGO for short, with the cubic bixbyite structure is known as a high-k oxide24 (dielectric constant k ≈ 12) and a stable passivating oxide for III V compounds and Si. In this paper, we report the growth of high-quality epitaxial ZnO films on Si(111) substrates buffered with GGO. The structural properties, including crystalline perfection, structural defects, and interface structure, of the ZnO/GGO/Si(111) heteroepitaxial system was thoroughly examined by X-ray diffraction (XRD) Received: December 17, 2010 Revised: April 7, 2011 Published: May 04, 2011 2846
dx.doi.org/10.1021/cg1016774 | Cryst. Growth Des. 2011, 11, 2846–2851
Crystal Growth & Design
ARTICLE
Figure 2. Azimuthal φ-scan profiles across the off-normal ZnO(1010), GGO(400), and Si(220) reflections.
’ EXPERIMENTAL SECTION Si(111) wafers were initially cleaned with the Radio Corporation America (RCA) method followed by HF acid dip for 20 s before loading into the growth chamber. A ∼9 nm thick GGO buffer layer was deposited on the clean Si wafer using electron beam evaporation from a single-crystal Ga5Gd3O12 garnet source.25 The composite substrate was then transferred through air exposure into the PLD growth chamber. The beam out of a KrF excimer laser (λ = 248 nm) was focused to produce an energy density of 5 7 J/cm2 on a commercial hot-pressed stoichiometric ZnO (5N) target at a repetition rate of 10 Hz. ZnO was deposited at the growth rate of ∼0.42 Å/s with substrate temperature ranging from 200 to 500 C without introducing oxygen gas. For samples with 2 h deposition, ZnO thickness varies from 313 to 292 nm with increasing temperature. The structural characteristics of all the samples are the same; the crystalline quality, as judged from both the tilt and twist angels determined by XRD, of the ZnO layers improves with increasing growth temperature. All the ZnO layers are tensily stressed with the strain increasing monotonically with growth temperature except the one grown at 500 C, which has the lowest strain. The strain reduction of the 500 C sample indicates the change of its microscopic structure from the others. As judged from the PL spectra, the sample grown at 400 C shows the best optical properties. The worse optical properties of the 500 C sample than the 400 C one may be attributed to the loss of oxygen at higher substrate temperature, especially since our deposition was conducted under an oxygen-free ambient. Same phenomenon was also reported on ZnO grown by PLD on 4H-SiC.3 It is also noteworthy that no cracks were found in all the ZnO layers studied whose thickness varies from 300 to 700 nm. The data presented in this work are taken from the sample with a ∼700 nm thick ZnO layer grown at 400 C. X-ray measurements were conducted using a four-circle diffractometer at the beamline BL13A of National Synchrotron Radiation Research Center (NSRRC) with an incident wavelength 1.0247 Å. Cross-sectional TEM specimens with a thickness of ∼90 nm were prepared by focused ion beam (FIB). TEM images were taken with a Philips TECNAI-20 field emission gun type TEM. PL measurements were carried out using a He Cd laser with wavelength of 325 nm as the pumping source. The emitted light was dispersed by a Triax-320
’ RESULTS AND DISCUSSION A radial scan along the surface normal (θ 2θ scan) of the sample is illustrated in Figure 1. Only ZnO(0002), (0004), and (0006) reflections together with the Si(111), (222), and (333) reflections were observed, elucidating that the ZnO layer is c-plane oriented with its [0001] axis parallel to the Si[111] direction. Because the lattice constant of GGO (aGGO = 10.86 Å) matches almost perfectly with that of Si (2aSi = 10.862 Å), the broad peaks on the base of Si(111) and (222) reflections are attributed to the GGO(222) and (444) reflections, respectively. Moreover, pronounced interference fringes, known as the thickness fringes, were observed near the Si(111) Bragg peak. From the period, Δqz = 0.07 Å 1, we determined the thickness of the GGO buffer layer to be ∼9 nm. The presence of the thickness fringe is also an indication of sharp interface and good crystalline structure. To examine the in-plane orientation relationship, the azimuthal cone scans (φ-scans) across the off-normal ZnO{1011}, GGO{400}, and Si{220} reflections was performed and is illustrated in Figure 2. Six evenly spaced ZnO peaks with 60 separation reveal that the hexagonal ZnO film was epitaxially grown on the GGO/Si(111) composite substrate. In the φ-scans across GGO{400} reflections, two sets of peaks with 3-fold symmetry were observed; the differences in their peak width and intensity imply that they have different origins. The set of narrow peaks with their angular positions rotated 60 from the Si{220} peaks were ascribed to the second harmonic of Si{400} Bragg peaks. The other set of broad peaks with their angular positions aligned with the Si{220} peaks are the GGO{400} reflections, which reveals the B-type cube-on-cube growth of GGO on Si. These results delineate that the in-plane epitaxial relationship of this heteroepitaxial system follows {1010}ZnO {422}Gd2O3 {422}Si. Under this orientation, the ZnO lattice is aligned with the O sublattice in GGO(111), which consists of a two-dimensional defective hexagonal lattice. In other words, the ZnO{1120} planes are parallel the {1120} planes of the O sublattice in GGO, which are also the {440} planes of GGO. The two-dimensional hexagonal lattice of the ZnO(0001) plane is ∼18.4% smaller than the O 2847
)
and transmission electron microscopy (TEM). Excellent optical characteristics of the ZnO films were verified by temperaturedependent photoluminescence (PL) taken between 13 K and RT.
)
Figure 1. XRD radial scan along the surface normal of ZnO film grown on GGO/Si(111).
spectrometer and detected by an UV-sensitive photomultiplier tube. We have also performed electrical transport measurements on the sample using the van der Pauw Hall method and obtained a resistivity of 0.16 Ω 3 cm, a carrier concentration of 1.9 1018 cm 3, and a mobility of 20 cm2/(V 3 s) at room temperature.
dx.doi.org/10.1021/cg1016774 |Cryst. Growth Des. 2011, 11, 2846–2851
Crystal Growth & Design
ARTICLE
Figure 3. (a) Cross-sectional high-resolution TEM image of the ZnO/ GGO interface with the Si[112] projection. The inset is the Fourierfiltered image of the ZnO/GGO interface. The selected area electron diffraction (SAED) pattern of the interfacial region taken along the same pole is displayed in panel b, in which the diffraction peaks of GGO overlap with that of Si.
sublattice in GGO(111) plane and thus their interplanar spacing retains a ratio between 6/7 and 5/6. Shown in Figure 3a is the cross-section TEM image taken near the ZnO/GGO interface with the ZnO(1010) pole. The Fourier-filtered image reconstructed by using the first-order diffraction peaks parallel to the interface is depicted in the inset, where an extra ZnO(1120) half plane for every five or six GGO{440} planes can be clearly identified. Similar to the case of ZnO grown Si(111) with a Y2O3 buffer, seven(six) planes of ZnO match with six(five) planes of GGO across the interface along the ZnOÆ1120æ directions.20 The huge lattice mismatch is thus accommodated by the misfit dislocations localized at the interface with an average periodicity of ∼6.5 times of ZnO{1120} planes. Domain matching epitaxy (DME), where integral multiples of lattice planes containing densely packed rows are matched across the interface, provides a nice description of the interfacial structure of these systems.26 The selected area electron diffraction (SAED) pattern taken at the interfacial region is shown in Figure 3b, which confirmed the epitaxial relationship determined by X-ray scattering. Because the nearly perfect lattice match between GGO and Si, their diffraction peaks coincide and cannot be distinguished in this orientation. By fitting the angular positions of several Bragg reflections, we determined the lattice parameters of the ZnO layer to be a = 3.261 Å and c = 5.195 Å. As compared with the bulk values, a = 3.244 Å and c = 5.204 Å, determined from a ZnO wafer, the ZnO epitaxial film exhibited tensile strain (0.52%) in the lateral direction, which is consistent with the observed red shift of the E2-high mode in the Raman spectrum. The lattice along the growth direction is correspondingly compressed by 0.17%. The full-width at half-maximum (fwhm) of the θ-rocking curve of the (0002) reflection is 0.22 and the fwhm of φ-scan of the {1011} reflection is 0.48. These numbers are smaller than the widths obtained from ZnO layers of similar thickness grown on Si with other buffers, such as Y2O3 (0.27/0.52 for tilting and twisting angles, respectively) and γ-Al2O3 (0.32/1.4), manifesting that GGO serves nicely as a buffer layer to facilitate the growth of the ZnO epi-layers with high crystalline quality.20,21 The optical microscopy image of crack-free ZnO epi-layer with 700 nm thickness on GGO buffer is observed as shown in Figure 4a. As a comparison, the surface image of a ∼555 nm thick ZnO epi-layer grown under a similar condition but on a Y2O3
Figure 4. (a) Plane-view image of the ZnO surface obtained by an optical microscope. No cracks are observed on the 700 nm thick ZnO layer buffered with GGO. As a comparison, the surface image of a thinner ZnO layer (555 nm) grown under a similar conditions but with a Y2O3 buffer layer is also shown in panel b, in which cracks running along Æ1120æ directions are readily observed.
buffer is shown in Figure 4b, where high density, 3.5 102 cm 1, of cracking channels running along the Æ1120æ direction are found. The preferred cracking direction indicates a nonisotropic fracture toughness related to the free {1010} surfaces formed by cracks along Æ1120æ direction. Ding et al. has demonstrated that wurtzite ZnO with {1010} side surfaces have the lowest surface energy by simple bonding density calculation, implying the lowest energy of breaking bonds at the {1010} surface.27 The absence of cracks on the ∼700 nm thick ZnO epi-layer grown on GGO buffer is not anticipated because the lattice mismatch of ZnO/GGO system is larger than that of ZnO/Y2O3 system. The smaller difference in thermal expansion coefficients (Ra(Gd2O3) = 7.6 10 6 K 1 and Ra(Y2O3) = 8.1 10 6 K 1) and the absence of two rotational variants in GGO layer may result in the suppression of the crack generation. TEM contrast analysis was conducted to characterize the nature of the structure defects developed in the ZnO layers.28 Bright-field cross-sectional TEM images taken under a two-beam contrast condition with diffraction vectors g equal to (0002), (1010), and (1011) are shown in Figure 5. In all three micrographs, the threading dislocations (TDs), seen as dark lines, stem from the ZnO/Si interface with the dislocation lines primarily along the [0001] direction. Three types of TDs with vertical [0001] line can be observed in hexagonal compact structure; they 2848
dx.doi.org/10.1021/cg1016774 |Cryst. Growth Des. 2011, 11, 2846–2851
Crystal Growth & Design
ARTICLE
Figure 7. TEM bright-field images of the ZnO film taken under a twobeam contrast conditions in which the diffraction vector g used is labeled in each image. The images a c were taken near the film surface and image d was taken near the ZnO/GGO interface.
Figure 5. TEM bright-field image of the ZnO film under a two-beam contrast condition with g equals (a) (0002), (b) (1010), and (c) (1011), respectively.
Figure 6. The HRTEM image shows the existence of a partial dislocation (PD) associated with a stacking fault (SF) in the (0001) plane of the ZnO epi-film on the GGO layer.
are pure screw dislocations, pure edge dislocations, and mixed dislocations with the corresponding Burgers vectors bc = Æ0001æ, bE = Æ1120æ/3, and bMÆ1123æ/3, respectively. According to the extinction criterion g 3 b = 0, the edge- and screw-type TDs are out of contrast in Figure 5a with g = (0002) and in Figure 5b with g =(1010), respectively. Taking the TEM specimen thickness of 90 nm into account, the densities of edge, screw, and mixed TDs were calculated to be approximately 5.2 109 cm 2 (∼38%), 5.7 108 cm 2 (∼4%), and 8.0 109 cm 2 (∼58%), respectively. Thus, the dislocations with edge component dominate over that
of the screw type. This result is similar to that of other c-plane ZnO films grown by PLD, for example, ZnO on sapphire,29 Y2O3/ Si,20γ-Al2O3/Si,21 Lu2O3/Si,22 and Sc2O3/Si,23 in which the edge component is the dominant type of TDs. Moreover, the high density of dislocations is observed near the ZnO/GGO interface due to large lattice mismatch and/or thermal strain, as shown in Figure 5c. Numerous interactions between dislocations, which resulted in defect reduction via half-loop formation, were found in the region of ∼310 nm from the interface because the dislocation lines cannot finish inside bulk material unless they combine with dislocations of opposite sign, in other words, forming a loop.30 The high density of stacking faults (SF) introduced either by a translation of the crystal lattice along close-packing directions in the (0001) plane or by condensation of vacancies or interstitials to form a dislocation loop accompanied with partial dislocations (PD) are produced to mediate the misfit relief.31 The stacking faults delimited by partial dislocations were observed in the ZnO epi-layer, as shown in Figure 6. In wurtzite structure, three different types of basal stacking faults (BSFs) exist, two intrinsic ones named I1 and I2 and one extrinsic named E.32 34 The displacement vectors R associated with I1, I2, and E types of BSFs are Æ2203æ/6, Æ1100æ/3, and Æ0001æ/2, respectively. A stacking fault of displacement vector R is out of contrast as the diffraction vector g satisfies g 3 R = 2πn, where n is an integer. Therefore, all three types of BSFs are visible in images with g = (1011) and out of contrast as g = (0002). In the case of g = (1010), only intrinsic stacking faults of types I1 and I2 are in contrast. Figure 7a c shows the images taken near the surface of the 700 nm thick ZnO layer under different diffraction vectors. Lateral lines of average length approximate 10 nm were observed in Figure 7b for g = (1010) and become out of contrast in Figure 7a for g = (0002), agreeing with the contrast variation for BSFs. The lines are also clearly visible in Figure 7c for g = (1011), revealing the intrinsic nature of these BSFs. However, we cannot further determine which type of intrinsic fault, I1 or I2, it is because these BSFs belonging to other diffraction vectors such 2849
dx.doi.org/10.1021/cg1016774 |Cryst. Growth Des. 2011, 11, 2846–2851
Crystal Growth & Design
’ CONCLUSION In this work, we demonstrated the growth of high-quality crack-free ZnO epitaxial films by PLD on Si(111) substrates with a nanometer thick GGO buffer layer. The in-plane epitaxial relationship between the wurtzite ZnO, cubic GGO, and cubic Si follows {1010}ZnO {422}Gd2O3 {422}Si, indicating alignment of 2850
)
as (1121) cannot be reached due to the angular limitation of the rotational stage. To examine the spatial distribution of the BSFs, another image was taken near the ZnO/GGO interface with g = (1010) and is shown in Figure 7d, having much fewer BSFs. The densities of the intrinsic type BSFs are approximately 2.5 105 and 1.2 105 cm 1 near the surface and interface, respectively, as derived from the images. The BSF density near the surface is 2 times larger than that near the interface. This result contrasts with the previous reports of epitaxial GaN films grown on sapphire or silicon substrates, in which high SF densities were found at and near the nucleation or buffer layer.35 37 This suggests that the crack formation may be quenched by the production of the interstitial dislocation loop with the intrinsic BSF near the surface, resulting in the strain relaxation in the inplane direction. Temperature-dependent PL spectra were employed from 13 (LT) to 300 K (RT) to inspect the optical transition mechanism and performance of the ZnO epi film on Si using GGO buffer layer. The PL spectra of the ZnO epi film measured at LT, shown in Figure 8a, is dominated by a strong near-band-edge (NBE) emission peak at 3.358 eV with a narrow line width of 10.2 meV. Furthermore, the deep-level emission (DLE) near 2.2 eV, generally recognized as the defect emission, is negligible in the PL spectrum. Both low DLE and the narrow and intense NBE emission are signatures of good optical performance. The strongest
peaks at 3.358 eV and the shoulder on the high-energy side, 3.375 eV, are assigned to the neutral donor bound exciton (D0XA) and the free A-exciton (FXA), respectively, as shown in Figure 8(b). By fitting the temperature-dependent intensity variation of the FXA emission by using the Arrhenius expression, we obtained A-exciton binding energy of 57.67 meV, in good agreement with the 60 meV for bulk ZnO crystal.38 Compared with the position of the bound exciton peak D0XA from a high quality single-crystalline ZnO wafer, 3.364 eV,39 the observed ∼6 meV red shift is attributed to the tensile strain, consistent with the XRD and Raman results. The other major spectral features include two relatively strong emission lines in the NBE region and two sets of equally distanced peaks (up to five) in the PL spectra.40 The equally distanced PL features are attributed to the longitudinal optical (LO) phonon replicas resulting from FXA and D0XA, respectively. Due to the increase in the thermal occupation of the free exciton states at the cost of the bound exciton states, the intensity of the two series of peaks shows an opposite dependence on temperature. The intensities of the FXA phonon replicas increase with temperature whereas the intensities of the D0X replicas decrease with temperature. The D0XA nLO (FXA nLO) replicas for n = 1, 2, 3, and 4 were observed at 3.288 (3.303), 3.216 (3.232), 3.144 (3.161), and 3.073 (3.089) eV; the periodic energy separation yields the LO-phonon energy ∼71 meV, in good agreement with the value reported on bulk ZnO.39 The emission of the Y line located at 3.328 eV in LT spectra, which persists up to RT, is ascribed to an exciton bound to structural defects with the corresponding localized energy of 47 meV. The behavior of the Y line with temperature variation is different from that of the two-electron satellite (TES) emission line because the intensity of the TES line will vanish when lattice temperature exceeds ∼60 K, observed in our ZnO/Y2O3/Si(111) sample or refs 41 and 42. Dean et al. proposed a model for the origin of the Y line to be due to a localized transition at extended defects such as dislocation loops.43 This result demonstrates that dislocation loops play an important role in crackingfree ZnO epi film on Si(111) using a GGO epitaxial buffer layer, which is consistent with the result from TEM. In comparison with other ZnO layers grown on Si, our ZnO epitaxial films buffered with GGO exhibit very good quality in both structure and optical performance as evidenced by XRD and PL results. Lately, besides GGO, other high-quality rare earth oxides with bixbyite structure, such as Y2O3 (lattice mismatch with ZnO = 15.5%),20 Lu2O3 (13.3%),22 and Sc2O3 (7.3%),23 have also been adopted as the buffer layer for ZnO epitaxial growth on Si. Excellent structural, electrical, and optical properties of so obtained ZnO films were demonstrated. A 100 nm thick Sc2O3, which has the smallest lattice mismatch with ZnO among the four, further reduces structural defects and yields the ZnO films (1 μm thick) with the smallest XRD line width (tilt/twist = 0.09/0.11) and TD density (in the order of 108 cm 2). All the works demonstrate that these nanometer thick high-k oxides with bixbyite structure facilitate the overgrowth of high-quality ZnO films.
)
Figure 8. Temperature-dependent PL spectra of the ZnO film on GGO/Si(111). (a) The spectrum measured at 13 K is dominated by the narrow NBE emission with negligible defect related green emission near 2.2 eV. (b) Temperature-dependent PL spectra plotted in semilogarithm scale. The spectra are vertically offset for clarity. The solid and dashed arrows indicate the emission lines involving phonon replicas of the bound excitons and free excitons, respectively.
ARTICLE
dx.doi.org/10.1021/cg1016774 |Cryst. Growth Des. 2011, 11, 2846–2851
Crystal Growth & Design the ZnO lattice with the O sublattice in GGO. The ZnO/GGO interfacial structure can be nicely described by domain matching epitaxy with seven(six) planes of ZnO matching with six(five) planes of GGO across the interface along the ZnO Æ1120æ directions. XRD and TEM results reveal that the major defect structures in the ZnO films are edge TDs and intrinsic BSFs. The development of the interstitial dislocation loops coupled with intrinsic BSF in the region near the surface may be responsible for the strain relaxation in the in-plane direction and subsequently the quench of crack formation. The PL results show the excellent optical properties of the obtained ZnO layers. These results provide valuable information for the integration of ZnObased multifunctional devices on Si substrates.
’ AUTHOR INFORMATION Corresponding Author
*C.-H. Hsu: e-mail
[email protected]; tel 886-3-578-0281, ext 7118; fax 886-3-578-3813. W. F. Hsieh: e-mail wfhsieh@mail. nctu.edu.tw; tel 886-3-571-2121, ext 56316; fax 886-3-571-6631.
’ ACKNOWLEDGMENT The National Science Council (NSC) of Taiwan supported this work under Contracts NSC-99-2112-M-006-017-MY3 and NSC-97-2112-M-213-003-MY3. ’ REFERENCES (1) Tsukazaki, A.; Ohtomo, A.; Onuma, T.; Ohtani, M.; Makino, T.; Sumiya, M.; Ohtani, K.; Chichibu, S. F.; Fuke, S.; Segawa, Y.; Ohno, H.; Koinuma, H.; Kawasaki, M. Nat. Mater. 2005, 4, 42. (2) Park, J. Y. P.; Je, J. H.; Kim, S. S. J. Cryst. Growth 2010, 312, 3588. (3) Kim, J.-H.; Cho, D.-H.; Lee, W.; Moon, B.-M.; Bahng, W.; Kim, S.-C.; Kim, N.-K.; Koo, S.-M. J. Alloys Compd. 2010, 489, 179. (4) Du, X.; Mei, Z.; Liu, Z.; Guo, Y.; Zhang, T.; Hou, Y.; Zhang, Z.; Xue, Q.; Kuznetsov, A. Yu. Adv. Mater. 2009, 21, 4625. (5) Vigue, F.; Vennegues, P.; Deparis, C.; Vezian, S.; La€ugt, M.; Faurie, J.-P. J. Appl. Phys. 2001, 90, 5115. (6) Yang, S.; Lin, B. H.; Liu, W.-R.; Lin, J.-H.; Chang, C.-S.; Hsu, C.-H.; Hsieh, W. F. Cryst. Growth Des. 2009, 9, 5184. (7) Fong, D. D.; Eastman, J. A.; Kim, S. K.; Fister, T. T.; Highland, M. J.; Baldo, P. M.; Fuoss, P. H. Appl. Phys. Lett. 2010, 97, No. 191904. (8) Gorla, C. R.; Emanetoglu, N. W.; Liang, S.; Mayo, W. E.; Lu, Y. J. Appl. Phys. 1999, 85, 2595. (9) Munuera, C.; Zu~niga-Perez, J.; Rommeluere, J. F.; Sallet, V.; Triboulet, R.; Soria, F.; Mu~noz-Sanjose, V.; Ocal, C. J. Cryst. Growth 2004, 264, 70. (10) Takahashi, N.; Kaiya, K.; Nakamura, T.; Momose, Y.; Yamamoto, H. Jpn. J. Appl. Phys. Part 2 1999, 38, L454. (11) Huang, T.-H.; Chang, L.; Chou, M. M. C.; Jahn, U. J. Electrochem. Soc. 2010, 158, H38. (12) Lee, S.; Kim, D. Y. Mater. Sci. Eng. B 2007, 137, 80. (13) Pecz, B.; El-Shaer, A.; Bakin, A.; Mofor, A.-C.; Waag, A.; Stoemenos, J. J. Appl. Phys. 2006, 100, No. 103506. (14) Zhou, H. L.; Shao, P. G.; Chua, S. J.; van Kan, J. A.; Bettiol, A. A.; Osipowicz, T.; Ooi, K. F.; Goh, G. K. L.; Watt, F. Cryst. Growth Des. 2008, 8, 4445. (15) Shi, L.; Xu, Y.; Hark, S.; Liu, Y.; Wang, S.; Peng, L.-m.; Wong, K.; Li, Q. Nano Lett. 2007, 7, 3559. (16) Ding, Y.; Wang, Z. L. Micron 2009, 40, 335. (17) Shi, L.; Bao, K.; Cao, J.; Qian, Y. CrystEngComm 2009, 11, 2009. (18) Lin, C. C.; Chen, S. Y.; Cheng, S. Y.; Lee, H. Y. Appl. Phys. Lett. 2004, 84, 5040.
ARTICLE
(19) Yoo, Y. Z.; Sekiguchi, T.; Chikyow, T.; Kawasaki, M.; Onuma, T.; Chichibu, S. F.; Song, J. H.; Koinuma, H. Appl. Phys. Lett. 2004, 84, 502. (20) Liu, W.-R.; Li, Y. H.; Hsieh, W. F.; Hsu, C.-H.; Lee, W. C.; Lee, Y. J.; Hong, M.; Kwo, J. Cryst. Growth Des. 2009, 9, 239. (21) Liu, W.-R.; Li, Y. H.; Hsieh, W. F.; Hsu, C.-H.; Lee, W. C.; Hong, M.; Kwo, J. J. Phys. D: Appl. Phys. 2008, 41, No. 065105. (22) Guo, W.; Allenic, A.; Chen, Y. B.; Pan, X. Q.; Tian, W.; Adamo, C.; Schlom, D. G. Appl. Phys. Lett. 2008, 92, No. 072101. (23) Guo, W.; Katz, M. B.; Nelson, C. T.; Heeg, T.; Schlom, D. G.; Liu, B.; Che, Y.; Pan, X. Q. Appl. Phys. Lett. 2009, 94, No. 122107. (24) Kwo, J.; Hong, M.; Kortan, A. R.; Queeney, K. T.; Chabal, Y. J.; Mannaerts, J. P.; Boone, T.; Krajewski, J. J.; Sergent, A. M.; Rosamilia, J. M. Appl. Phys. Lett. 2000, 77, 130. (25) Lin, T. D.; Hang, M. C.; Hsu, C.-H.; Kwo, J.; Hong, M. J. Cryst. Growth 2007, 301, 386. (26) Narayan, J.; Larson, B. C. J. Appl. Phys. 2003, 93, 278. (27) Ding, Y.; Wang, Z. L.; Sun, T.; Qiu, J. Appl. Phys. Lett. 2007, 90, No. 153510. (28) Fultz, B.; Howe, J. Transmission Electron Microscopy and Diffractometry of Materials; Springer: Berlin, 2000. (29) Liu, W.-R.; Hsieh, W. F.; Hsu, C.-H.; Liang, Keng S.; Chien, F. S.-S. J. Appl. Crystallogr. 2007, 40, 924. (30) Liliental-Weber, Z.; Chems, D. J. Appl. Phys. 2001, 89, 7833. (31) Guo, W.; Allenic, A.; Chen, Y. B.; Pan, X. Q.; Che, Y.; Hu, Z. D.; Liu, B. Appl. Phys. Lett. 2007, 90, No. 242108. (32) Stampfl, C.; Van de Walle, C. G. Phys. Rev. B 1998, 57, R15052. (33) Vennegues, P.; Chauveau, J. M.; Korytov, M.; Deparis, C.; Zuniga-Perez, J.; Morhain, C. J. Appl. Phys. 2008, 103, No. 083525. (34) Gerthsen, D.; Litvinov, D.; Gruber, T.; Kirchner, C.; Waag, A. Appl. Phys. Lett. 2002, 81, 3972. (35) Komninou, Ph.; Kioseoglou, J.; Dimitrakopulos, G. P.; Kehagias, Th.; Karakostas, Th. Phys. Status Solidi A 2005, 202, 2888. (36) Pankove, J. I.; Moustakas, T. D. In Semiconductors and Semimetals; Willardson, R. K., Weber, E. R., Eds.; Academic Press: New York, 1998; Vols. 50 57. (37) Morkoc-, H. In Nitride Semiconductors and Devices; Springer Series in Materials Science, Vol. 32; Springer: Berlin, 1999. (38) Jiang, D. S.; Jung, ::H.; Ploog, K. J. Appl. Phys. 1998, 64, 1371. € ur, U.; Dogan, S.; Gu, X.; Morkoc-, H.; Nemeth, (39) Teke, A.; Ozg€ B.; Nause, J.; Everitt, H. O. Phys. Rev. B 2004, 70, No. 195207. (40) Wang, L.; Giles, N. C. J. Appl. Phys. 2003, 94, 973. (41) Liu, K. W.; Sakurai, M.; Aono, M. J. Appl. Phys. 2010, 108, No. 043516. (42) Ohno, R Y.; Koizumi, H.; Taishi, T.; Yonenaga, I.; Fujii, K.; Goto, H.; Yao, T. J. Appl. Phys. 2008, 104, 073515. (43) Dean, P. J. Phys. Status Solidi A 1984, 81, 625.
2851
dx.doi.org/10.1021/cg1016774 |Cryst. Growth Des. 2011, 11, 2846–2851
