Scholarly article on topic 'Evaluation of antidiabetic, antihyperlipidemic and antioxidant effects of Boehmeria nivea (L.) Gaudich., Urticaceae, root extract in streptozotocin-induced diabetic rats'

Evaluation of antidiabetic, antihyperlipidemic and antioxidant effects of Boehmeria nivea (L.) Gaudich., Urticaceae, root extract in streptozotocin-induced diabetic rats Academic research paper on "Materials engineering"

0
0
Share paper
Academic journal
Revista Brasileira de Farmacognosia
OECD Field of science
Keywords
{""}

Academic research paper on topic "Evaluation of antidiabetic, antihyperlipidemic and antioxidant effects of Boehmeria nivea (L.) Gaudich., Urticaceae, root extract in streptozotocin-induced diabetic rats"

Materials Expre

Copyright © 2015 by American Scientific Publishers All rights reserved.

Printed in the United States of America

2158-5849/2015/5/129/008 doi:10.1166/mex.2015.1220 www.aspbs.com/mex

Control of AlN single crystal nucleation: An insight into the crystal growth habit in the initial stages of the physical vapor transport method

Lei Jin1, Hua-Yu Zhang1, Jie-Cai Han2, Chao-Liang Zhao2, Tai Yao3'*, and Bo Song3,*

1 Shenzhen Graduate School, Harbin Institute of Technology, Shenzhen 518055, China

2 Centre for Composite Materials, Harbin Institute of Technology, Harbin 150080, China

3Academy of Fundamental and Interdisciplinary Sciences, Harbin Institute of Technology, Harbin 150080, China

ABSTRACT

The nucleation of AlN single crystals was investigated by heterogeneous nucleation theory, and by physical vapor transport experiments on tungsten (W) and tantalum (Ta) substrates in the temperature range of 21502250 °C. The theoretical analysis showed that AlN nucleation readily occurred on the substrate surface where m the heterogeneous nucleation energy of AlN crystal was low. The morphologies, crystal structure and chemical ^ compositions of AlN grains were characterized using laser scanning confocal microscopy, field emission scanning electron microscopy, energy-dispersive X-ray spectroscopy and X-ray diffraction. This revealed that the nucleation density on the W substrate was higher than that on the Ta substrate, and the grains were distributed in rows along concavities on both scrubbed W and Ta substrate surfaces. It was found that at the initial stages of crystal growth, the number of nuclei and the nucleation site can be controlled by varying the substrate type and morphology which will further influence the crystal growth from nucleation point to the final crystal.

Keywords: AlN, Nucleation, Substrate, Initial Stages.

1. INTRODUCTION

The III-V wide band-gap semiconductor AlN, which has a band-gap of 6.2 eV, displays outstanding properties, such as high breakdown field, high thermal conductivity, high saturation velocity, and high thermal stability. Its excellent physical properties make AlN a promising candidate for the fabrication of surface acoustic wave devices,® UV-light-emitting devices (UV-LED)® and power semiconductor devices.® In particular, a good match in lattice constants and thermal expansion coefficients with GaN make it an excellent candidate for the fabrication of GaN-based electric devices.1(4'5

At present, one of the most promising method of producing large size AlN bulk single crystals is physical

* Authors to whom correspondence should be addressed. Emails: yaotai@hit.edu.cn, songbo@hit.edu.cn

vapor transport (PVT) method, and by far, three strategies for AlN bulk single crystal growth via the PVT method have been extensively investigated, that is: (1) Homo-epitaxial seeding on AlN substrate; (2) hetero-epitaxial seeding on silicon carbide (SiC) substrates; and (3) spontaneous nucleation. It is well accepted that the homo-epitaxial route was considered as the pathway to obtain the samples with the high crystalline quality. However, owing to the lack of large size of AlN seeds, large size of AlN single crystal has been yet approached by this method.® In contrast, although there are already some successful demonstrations in hetero-epitaxial seeding on SiC substrates/7-9' the C, Si impurities originated from the decomposition of SiC or the high defects density inherited from the SiC substrates restrict their practical applications.®10-12 To solve this problem, the researchers turn their attention to the spontaneous nucleation process.

Recently, Hartmann et al.(13) reported that AlN single crystals without stress-induced defect with a size exceeding 10 x 10 x 12 mm3 was obtained by the spontaneous nucleation, revealing the great potential of this route. Nevertheless, the spontaneous nucleation process normally are associated with certain disadvantages and suffer from excess nucleation sites which will introduce redundant crystal boundaries and defects that impede the growth of AlN crystals from nucleation point to final crystals.(14-16) Moreover, the surrounded parasitic polycrystalline rim will result in the formation of stress-induced defects in AlN single crystal. Thus, to obtain high-quality AlN single crystals is highly desirable and is of special interest owing to its significant applications, but it is still a challenge for researchers. Exploring the initial growth habit of AlN spontaneous nucleation will help us further understand its growth habit and benefit to obtain high-quality single crystals. So far, the efforts towards improving AlN crystalline quality have been mainly based on trial and error to adjust the experimental parameters.(17-22) In this work, we performed detailed investigations of the initial stages of AlN spontaneous nucleation via PVT method. The theoretical analysis focused on the role of surface potential energy (SPE) and heterogeneous nucleation energy (HNE) to give 0 a relationship between substrate species and morphology

Power Crucible

Source

this, we hope to reconcile the conflicting objectives of simultaneously achieving high-quality AlN single crystal and controlling excess nucleation sites in the AlN spontaneous nucleation process.

2. EXPERIMENTAL DETAILS

2.1. Preparation

The AlN crystal nucleation process was analyzed using both theoretical and experimental methods. The growth model was simply established based on nucleation theory. Experimentally, AlN crystals were grown by the PVT method at 2150-2250 °C on W and Ta substrates. Before being heated to the desired temperature of 2250 °C, the furnace was evacuated to ~3.0 x 10-4 Pa to remove the residual gas. In the typical sublimation configuration, as shown in Figure 1(a), AlN polycrystalline powder (Alfa Aesar, 99.999%) decomposed in the high-temperature region (2250 °C) in 6.0 x 104 Pa N2 (99.999%), and then the decomposition products were transported to the low-temperature region (~2200 °C) where they adsorbed on the substrate surface to form AlN crystals. The estimated temperature gradient between the source and crucible lid was 3-7 ° C/mm with a distance of 8 mm between the source and crucible lid. At the end of crystal growth, the furnace chamber was filled with 105 Pa N2 (99.999%) to avoid secondary nucleation.

(c) 1 2 3

Fig. 1. (a) Typical PVT sublimation configuration. (b) The enlarged substrate morphology as shown in (a) with assumed substrate surface. (c) Schematic diagram of atom movement from site 1 to site 3 on the assumed surface. (d) SPE corresponding to the different sites shown in (c) in which gT is the intrinsic SPE barrier (site 1), gv is the energy barrier in the vicinity of the concavity, gS is the potential energy for site 2, and gL is the potential energy barrier for site 3. (e) The nucleation volume on concave, smooth and convex surfaces denoted by A, B' and C, respectively. 0 is the equilibrium contact angle.

2.2. Characterization

Three-dimensional (3D) morphology details of the as-grown products were characterized on a laser scanning confocal microscope (LSM, OLS 3100/SZX16). Further morphology was examined by a field emission scanning electron microscope (FE-SEM, Hitachi S4700) and semiquantitative composition analysis was conducted using an energy-dispersive X-ray spectrometer (EDX) attached to the FE-SEM with an Oxford energy-dispersive X-ray analysis system. The crystal structure of the samples was characterized using an X-ray diffractometer (Rigaku D/max 2500 pc) with mono-chromatized CuKa (A = 1.5406 Â) as the incident radiation and Raman spectra were measured at 300 K on a Lab RAM HR 800 spectrometer (Jobin Yvon) using a 532-nm excitation source. The grain density was given by number of grains per unit area.

3. RESULTS AND DISCUSSION

3.1. Growth Model

Theoretical analysis of the SPE and HNE was carried out to investigate the effects of substrate surface morphology on the nucleation process. To analyze the basic AlN nucleation mechanism, a typical substrate surface made up of an array of straight steps separated by concavities was assumed, as shown in Figure 1(b). On this surface,

a typical motion process of an atom from site 1 to 3 is shown in Figure 1(c).

First, as an intrinsic feature of substrate materials, the SPE distribution tendency corresponding to various sites has pictured in Figure 1(d) which effects on the atom movement process and can be tuned via morphology modifications, as proposed by Gilmer.(23) In Figure 1(c), the atom located at site 1 is constrained by the periodic SPE (gT). When it shifts towards site 2, it must first overcome gT and then get across the energy barrier gu in the vicinity of the concavity before entering site 2. At site 2, the SPE value (gS) is less than the SPE value at site 1.(23) Then, the atom continues moving from site 2 to 3, and the energy barrier for migration out of the concavity (site 2) is gu + gS. Before reaching site 3, it must enter into the concavity close to site 3 (via a similar movement process from site 1 to 2), and then overcome the maximum SPE difference (gL) prior to approaching site 3. Thus, the stability of atoms on different sites can be ranked as: site 3 (gL) > site 2 (gS + gu) > site 1 (gT). Notably, here, SPE distribution tendency is utilized to analyze the atoms stability on different sites as we only care about the influence tendency of substrate morphology.

Second, the nucleation needs to overcome the potential barrier of the HNE, which is determined by the type of substrate material, and the classical theory of crystal

Г .J^B

Fig. 2. (a) Optical image of the substrate with (1) zone 1 without scrubbing, (2) zone 2 scrubbed with #320 corundum abrasive paper, (3) zone 3 scrubbed with #80 corundum abrasive paper. (b) SEM image of zone 1. (c) SEM image of zone 2. (d) SEM image of zone 3.

nucleation has been proposed by Turnbull.(24'25) Moreover, the HNE can be changed by modification of the substrate morphology, as shown in Figure 2. It is well known that the wetting angle 0 of AlN is constant on a particular substrate material/24 thus, substrate morphology is the key factor affecting nucleation. Presumed the shape of the crystal nucleus is round, the nucleation volume will be smaller in a concavity (A) than on a smooth surface (B'), conversely, the HNE is larger on a convexity (C) than on a smooth surface as shown in Figure 1(e). Therefore, it is much easier for AlN to nucleate on concavities than convexities where the HNE is lower.

The aforementioned theoretical analysis demonstrates that both SPE and HNE, the intrinsic properties determined by substrate materials, have a significant influence on the initial nucleation process. For a given substrate material, it is speculated that SPE and HNE could be tuned by surface modification which could control the number of nuclei and the nucleation site.

3.2. Nucleation Habits on W and Ta Substrate

To verify these theoretical predictions, the W substrate was first selected to examine the effect of both SPE and HNE on the number of nuclei and nucleation site. To probe the effect of various morphologies, the substrate surface was scrubbed with commercial abrasive papers to divide it into three regions:

(i) zone 1 without scrubbing,

(ii) zone 2 scrubbed with #320 corundum abrasive paper, and

(iii) zone 3 scrubbed with #80 corundum abrasive paper (shown in Fig. 2(a)).

Figures 2(b)-(d) shows the SEM images corresponding to the three zones as shown in Figure 2(a). Straight steps separated by concavities were easily created through a simple scrubbing process to explore the influence of morphology on SPE and HNE.

The structure of the as-grown products grown on the W substrate has been determined by X-ray diffraction (XRD). As shown in Figure 3(a), all of the diffraction peaks in the XRD pattern can be identified as wurtzite AlN with cell constants of a = 0.3111 nm and c = 0.4978 nm (JCPDS No. 76-0565), and no impurity phases were detected in the XRD pattern. The Raman peaks of as prepared AlN crystal on W substrate are presented in Figure 3(b). Raman measurement was performed at room temperature. The peaks located at 247.1, 656.3, 668.5, and 889.1 cm-1 can be assigned to E2 (Low), E2 (high), Ex (TO) and A (LO), respectively, which provide further evidence that AlN grains was prepared.(26)

Figures 4(a)-(c) shows 3D LSM images of products as grown for 30 min on W substrates to show the information of grain-distribution. In Figure 4(a), the 3D image show that grains on smooth substrate distribute randomly. And, in Figures 4(b) and (c), the grains on scrubbed substrate

50 60 29 (deg.)

Raman shift(cm-1)

Fig. 3. (a) XRD pattern of the as-synthesized products and (b) Raman spectrum of the as-synthesized products.

distribute in rows. Red dotted lines through the peaks of the grains are provided to guide the eye which agrees well with the direction of concavity shown in zones 2 and 3 (see Figs. 3(c) and (d)). Figures 4(d)-(f) shows typical low-magnification SEM images of the as grown AlN grains corresponding to Figures 4(a)-(c). The crystal nuclei can be clearly seen on the substrates, which generate numerous pyramid-shaped or hillock grains that completely covered the substrate surface. The nucleation densities calculated on zone 1 (Fig. 4(d)), zone 2 (Fig. 4(e)) and zone 3 (Fig. 4(f)) are 1.5 x 106 cm-2, 0.8 x 106 cm-2, and 1.0 x 106 cm-2, respectively. From zone 1 to zone 3, the increased convexity of the scrubbed surface make the nucleation densities decreased in zone 2 and 3 than zone 1 which provides convincing evidence that variation of the morphology has an effect on the nucleation density. To reveal the influence of SPE and HNE on the crystal growth habit more clearly, the grains with similar morphology features on the scrubbed substrate are marked by circles with the same color in Figures 4(e) and (f). Similar to Figures 4(b) and (c), the circles with the same color distributed in rows in Figures 4(e) and (f). AlN nucle-ation occurred in rows along the concavity, which is in good agreement with previous results illustrated that the nucleation distribution also could be effect by morphology

Fig. 4. 3-D LSM images of AlN grains grown for 30 min on the W substrate. (a) Without scrubbing, (b) scrubbed with #320 corundum abrasive paper, and (c) scrubbed with #80 corundum abrasive paper. And, SEM images of AlN crystal grown for 30 min on W: (d) without scrubbing, (e) scrubbed with #320 corundum abrasive paper, and (f) scrubbed with #80 corundum abrasive paper. Red dotted lines through the peaks of the grains are provided to guide the eye in (b) and (c); grains with similar morphology features are marked by circles with the same color in (e) and (f).

modification. But, there are still many grains which may impede further expansion of AlN crystal.

Further verification of the theoretical analysis was carried out on another case of the Ta substrate. The EDX spectrum (Fig. 5(a)) reveals that composition of the grains is N and Al with a ratio ~1:1. Figures 5(b)-(d) shows SEM images of AlN grains grown for 30 min on the Ta substrate. Compared with the W substrate (Fig. 4(d)), the nucleation density on the Ta substrate (Fig. 5(b)) is lower, and it is speculated that the HNE of AlN on Ta substrate is thus higher than that on W substrate. In fact, AlN grains were found on smooth Ta substrate as shown in the inset in Figure 5(b), the formation of AlN grains were

contribution to unavoidable defect on substrate surface where AlN nucleation easily occur. On scrubbed surfaces, irregularly shaped grains arranged in rows were observed in Figures 5(c) and (d). It was found that the as-grown AlN grains were uniformly distributed without agglomeration in the concavities on the scrubbed Ta substrate. Here, as shown in Figures 5(c) and (d), the AlN grains grown on the Ta substrate were smaller than those grown on the W substrate, indicating that the SPE barrier of the Ta substrate is smaller than that of the W substrate. Note that the nucleation density greatly increased from the smooth to the scrubbed surface, which is different to the W substrate. The increase in the number of nuclei is considered

Fig. 6. Images of AlN crystal grown for 90 min and 60 h: (a) on W substrate scrubbed with #320 corundum abrasive paper for 90 min; (b) on Ta substrate scrubbed with #320 corundum abrasive paper for 90 min; (c) on W substrate scrubbed with #320 corundum abrasive paper for 60 h; (d) on Ta substrate with #320 corundum abrasive paper for 60 h. A typical grain with typical hexagonal (0001) facets, quadrilateral (10-10) and (10-1n) facets is enlarged as shown in the inset of Figure 6(b).

to be induced by smaller nucleation volume as a result of the concavity. This provides further evidence that the HNE could be changed by modifying the morphology features.

Based on our theoretical analysis and experiments on W and Ta substrates, it was found that:

(1) nucleation density on the W substrate was higher than that on the Ta substrate;

(2) both SPE and HNE, which can be altered by surface modifications, have a significant influence on the nucle-ation density and site of AlN crystal nuclei.

3.3. Preparation of AlN Boules

Figure 6 shows the image of AlN crystal grains grown for 90 min and 60 h on W and Ta substrate scrubbed with #320 corundum abrasive paper. Grains grown on scrubbed W substrate as shown in Figure 6(a) exhibited much grain boundary as the large grain density. In Figure 6(b), zonar structures with typical hexagonal (0001) facets, and quadrilateral (10-10) and (10-1n) facets were observed on Ta. Notably, comparing Figure 6(b) with Figure 5(d), the grain density decreased with increasing growth duration from 30 min to 90 min. It can be explained by crystalline lattice structure of AlN.(27'28) Moreover, Bickermann et al.(29) have reported that the ratio of growth rates is estimated to be roughly 1:3:10 for growth on prismatic, pyramidal, and basal plane facets. In this case, the lower grain density may decrease the grain boundary greatly and will benefit the further expansion of AlN crystal for a longer growth duration. The evaluation process from nucleation point to the final crystal has been reported.113'16) Comparing Figures 6(c) and (d), the enlargement of AlN grains on scrubbed Ta substrate is faster than that on scrubbed W substrate, as nucleation density on scrubbed Ta substrate is much less than that on scrubbed W substrate. The largest AlN single crystal obtained on scrubbed Ta substrate reached 7 mm in diameter, which is two times larger than that on scrubbed W substrate (~ 3 mm). From the top view of AlN boules, no typical grains structure was obtained on W substrate which is similar to previous report.(15) Whereas, the grains on Ta with typical hexagonal (0001) facets and quadrilateral (10-10) are obtained which are the same as AlN grains in Figure 6(b). The polycrystalline AlN boule on Ta substrate is much different from previous report(20) which is attributed to the substrate modification.

4. CONCLUSION

In this work, the initial stages of AlN nucleation and early growth processes were studied theoretically and experimentally. Thermodynamic and kinetic theory analysis showed that the SPE and HNE of the substrate have a significant effect on the number of nuclei and nucleation site. Experimental results provide convincing evidence to confirm the theoretical analysis results. Nucleation occurred

more easily on the W than the Ta substrate. The grains distributed in rows along concavities on both scrubbed W and Ta surfaces. One can see clearly that the number of nuclei and site of nucleation have great influence the further crystal growth from nucleation point to the final crystal which can be controlled by tailoring the substrate materials and substrate morphology via the PVT route.

Acknowledgments: This work was supported financially by the National Natural Science Foundation of China (Grant Nos. 50902037, 51172055, 51372056), Fundamental Research Funds for the Central University (Grant Nos. HIT.BRETIII.201220, HIT.NSRIF.2012045, HIT.ICRST.2010008), the Foundation of National Key Laboratory of Science and Technology on Advanced Composite in Special Environment in HIT, International Science and Technology Cooperation Program of China (2012DFR50020) and the Program for New Century Excellent Talents in University (NCET-13-0174).

References and Notes

1. C. J. Zhou, Y. Yang, Y. Shu, H. L. Cai, T. L. Ren, M. Chan, J. Zhou, H. Jin, S. R. Dong, and C. Y. Yang; Visible-light photoresponse of AlN-based film bulk acoustic wave resonator; Appl. Phys. Lett. 102, 191914 (2013).

2. T. Kinoshita, T. Obata, T. Nagashima, H. Yanagi, B. Moody, S. Mita, S. Inoue, Y. Kumagai, A. Koukitu, and Z. Sitar; Performance and reliability of deep-ultraviolet light-emitting diodes fabricated on AlN substrates prepared by hydride vapor phase epitaxy; Appl. Phys. Express 6, 092103 (2013).

3. M. H. Wong, Y. Pei, J. S. Speck, and U. K. Mishra; High power N-face GaN high electron mobility transistors grown by molecular beam epitaxy with optimization of AlN nucleation; Appl. Phys. Lett. 94, 182103 (2009).

4. Y. Taniyasu, M. Kasu, and T. Makimoto; An aluminium nitride light-emitting diode with a wavelength of 210 nanometres; Nature 441, 325 (2006).

5. J. R. Grandusky, S. R. Gibb, M. C. Mendrick, and L. J. Schowalter; Properties of Mid-ultraviolet light emitting diodes fabricated from pseudomorphic layers on bulk aluminum nitride substrates; Appl. Phys. Express 3, 072103 (2010).

6. Z. G. Herro, D. Zhuang, R. Schlesser, and Z. Sitar; Growth of AlN single crystalline boules; J. Cryst. Growth 312, 2519 (2010).

7. R. R. Sumathi; Bulk AlN single crystal growth on foreign substrate and preparation of free-standing native seeds; Cryst. Eng. Comm. 15, 2232 (2013).

8. S. B. Zuo, X. L. Chen, L. B. Jiang, H. Q. Bao, J. Wang, L. W. Guo, and W. J. Wang; Crystal growth of AlN: Effect of SiC substrate; Materials Science in Semiconductor Processing 15, 401 (2012).

9. R. Dalmau, R. Schlesser, and Z. Sitar; Polarity and morphology in seeded growth of bulk AlN on SiC; Phys. Stat. Sol. C 2, 2036 (2005).

10. P. Lu, J. H. Edgar, C. Cao, K. Hohn, R. Dalmau, R. Schlesser, and Z. Sitar; Seeded growth of AlN on SiC substrates and defect characterization; J. Cryst. Growth 310, 2464 (2008).

11. M. Bickermann, O. Filip, B. M. Epelbaum, P. Heimann, M. Feneberg, B. Neuschl, K. Thonke, E. Wedler, and A. Winnacker; Growth of AlN bulk crystals on SiC seeds: Chemical analysis and crystal properties; J. Cryst. Growth 339, 13 (2012).

12. C. Hartmann, M. Albrecht, J. Wollweber, J. Schuppang, U. Juda, C. Guguschev, S. Golka, A. Dittmar, and R. Fornari; BulkAlN growth under the influence of residual oxygen; J. Cryst. Growth 344, 19 (2012).

13. C. Hartmann, J. Wollweber, A. Dittmar, K. Irmscher, A. Kwasniewski, F. Langhans, T. Neugut, and M. Bickermann; Preparation of bulk AlN seeds by spontaneous nucleation of freestanding crystals; Jpn. J. Appl. Phys. 52, 08JA06 (2013).

14. Z. Gu, L. Du, J. H. Edgar, N. Nepal, J. Y. Lin, H. X. Jiang, and R. Witt; Sublimation growth of aluminum nitride crystals; J. Cryst. Growth 297, 105 (2006).

15. B. M. Epelbaum, M. Bickermann, and A. Winnacker; Approaches to seeded PVT growth of AIN crystals; J. Cryst. Growth 275, e479 (2005).

16. V. Noveski, R. Schlesser, B. Raghothamachar, M. Dudley, S. Mahajan, S. Beaudoin, and Z. Sitar; Seeded growth of bulk AlN crystals and grain evolution in polycrystalline AlN boules; J. Cryst. Growth 279, 13 (2005).

17. W. J. Wang, S. B. Zuo, H. Q. Bao, J. Wang, L. B. Jiang, and X. L. Chen; Effect of the seed crystallographic orientation on AlN bulk crystal growth by PVT method; Cryst. Res. Technol. 46, 455 (2011).

18. Y. X. Li and D. W. Brenner; Influence of trace precursors on mass transport and growth rate during sublimation deposition of AlN crystal; J. Appl. Phys. 100, 084901 (2006).

19. V. Noveski, R. Schlesser, S. Mahajanb, S. Beaudoinc, and Z. Sitar; Mass transfer in AlN crystal growth at high temperatures; J. Cryst. Growth 264, 369 (2004).

20. E. N. Mokhov, O. V. Avdeev, I. S. Barash, T. Yu. Chemekova, A. D. Roenkova, A. S. Segal, A. A. Wolfson, Y. N. Makarova, M. G. Ramm, and H. Helava; Sublimation growth of AlN bulk crystals in Ta crucibles; J. Cryst. Growth 281, 93 (2005).

21. B. M. Epelbaum, C. Seitz, A. Magerl, M. Bickermann, and A. Winnacker; Natural growth habit of bulk AlN crystals; J. Cryst. Growth 265, 577 (2004).

22. R. Schlesser, R. Dalmau, D. Zhuang, R. Collazo, and Z. Sitar; Crucible materials for growth of aluminum nitride crystals; J. Cryst. Growth 281, 75 (2005).

23. C. Roland and G. H. Gilmer; Kinetics of nucleation-dominated step flow; Phys. Rev. B 54, 2931 (1996).

24. D. Turnbull; Formation of crystal nuclei in liquid metals; J. Appl. Phys. 21, 1022 (1950).

25. D. Turnbull; Kinetics of solidification of supercooled liquid mercury droplets; J. Chem. Phys. 20, 411 (1952).

26. V. Y. Davydov, Y. E. Kitaev, I. N. Goncharuk, A. N. Smirnov, J. Graul, O. Semchinova, D. Uffmann, M. B. Smirnov, A. P. Mirgorodsky, and R. A. Evarestov; Phonon dispersion and Raman scattering in hexagonal GaN and AlN; Phys. Rev. B 58, 12899 (1998).

27. X. H. Xu, H. S. Wu, C. J. Zhang, and Z. H. Jin; Morphological properties of AlN piezoelectric thin films deposited by DC reactive magnetron sputtering; Thin Solid Films 388, 62 (2001).

28. X. P. Kuang, H. Y. Zhang, G. G. Wang, L. Cui, C. Zhu, L. Jin, R. Sun, and J. C. Han; Effect of deposition temperature on the microstructure and surface morphology of c-axis oriented AlN films deposited on sapphire substrate by RF reactive magnetron sputtering; Superlattices and Microstructures 52, 931 (2012).

29. M. Bickermann, B. M. Epelbaum, O. Filip, B. Tautz, P. Heimann, and A. Winnacker; Faceting in AlN bulk crystal growth and its impact on optical properties of the crystals; Phys. Status Solidi C 9, 449 (2012).

Received: 18 September 2014. Revised/Accepted: 17 December 2014.