• Home
  • Sitemap
  • Contact us
Article View

Research Paper

Applied Science and Convergence Technology 2022; 31(6): 141-144

Published online November 30, 2022


Copyright © The Korean Vacuum Society.

Structural and Optical Properties of InN Nanowires Formed on Si(111)

Jaehyeok Shina , Siyun Noha , Jinseong Leea , Jaewon Ohb , Mee-Yi Ryub , and Jin Soo Kima , *

aDepartment of Electronic and Information Materials Engineering, Division of Advanced Materials Engineering, and Research Center of Advanced Materials Development, Jeonbuk National University, Jeonju 54896, Republic of Korea
bDepartment of Physics, Kangwon National University, Chuncheon 24341, Republic of Korea

Correspondence to:kjinsoo@jbnu.ac.kr

Received: August 4, 2022; Revised: October 20, 2022; Accepted: October 22, 2022

We report the structural and optical properties of InN nanowires (NWs) formed on p-type Si(111) substrates using a plasma-assisted molecular-beam epitaxy. The InN NWs were formed on a Si(111) substrate using a new growth method, the indium (In) pre-deposition (InPD) method, in which In droplets are initially formed by supplying only In flux to the substrate, where they work as nucleation sites for the formation of subsequent InN NWs. Field-emission scanning-electron microscopy images show that the InN NWs have symmetric shapes along the vertical direction. In addition, most of the InN NWs were unidirectionally grown in the direction perpendicular to the substrate. Strong and narrow peaks corresponding to InN(0002) can be clearly observed in the double crystal X-ray diffraction rocking curves of the NW samples. In the transmission electron microscopy images of the InN NWs, stacking faults, typically observed in Si-based III-nitride semiconductors, are rarely observed. A strong free-exciton peak was observed from the InN NWs at the wavelength of 1297 nm with a narrow linewidth at room temperature. Structural and optical characterizations of the NW samples indicate that highly crystalline InN NWs were formed on Si(111) using the new InPD growth method.

Keywords: Plasma-assisted molecular-beam epitaxy, InN nanowires, Si, In pre-deposition method, High crystallinity

Recently, III-nitride semiconductors are being actively studied for optoelectronic applications including solar cells, photodetectors, lightemitting diodes, and optical sensors operating in the wavelength window from ultraviolet to near infra-red (NIR). The wide operation wavelength is typically obtained by manipulating the composition of group III elements in (Ga,In,Al)N [1]. Among the III-nitride materials, InN has the highest electron mobility and saturation velocity due to its smallest effective electron mass. Because the InN epitaxial layer shows a band gap from 0.6 to 1.1 eV [24], it is suitable for optoelectronic devices operating in the NIR wavelength window, especially those operating at 1.3 and 1.55 µm [5,6]. According to previous reports, InN is typically formed on Si or sapphire [7,8]. However, the formation of highly crystalline InN layers is very difficult due to significant differences in material properties, including lattice parameters and coefficient of thermal expansion between InN and substrate. In addition, dissociation of In atoms may be significantly higher than those of other group-III atoms, Ga and Al, during epitaxial growth. More seriously, there is significant In segregation even at low growth temperature [79]. These intrinsic growth-limiting factors make it is difficult to obtain highly crystalline InN films on Si or sapphire. To overcome this intrinsic problem of the epitaxial layer, InN nanowires (NWs) are attracting much attention because their small footprints can relax the strain originating from lattice mismatch through the NW sidewall [9]. Also, for nanophotonic and nanoelectronic device applications, onedimensional structures such as NWs and nanorods are promising due to their strong anisotropic optical absorption, directionality of light emission, and confined charge-carrier transport [10,11]. For the formation of InN NWs on a Si substrate, there has been wide adoption of the vapor-liquid-solid method, in which metal catalysts such as Au and Ni are used as nucleation sites for NWs. However, the metal catalysts show high probability of remaining inside or on the top surfaces of the InN NWs, resulting in the introduction of deep trap levels for carriers and consequent degradation of optical and electrical signals [12,13]. To overcome the problems caused by the use of metal catalysts, the Volmer-Weber mode has been used to promote self-catalyst and catalyst-free formation of InN NWs on Si. However, the formation of highly crystalline InN NWs with high structural symmetry remains difficult. Moreover, the growth direction of the NWs is random with respect to the substrate, causing difficulty in the device fabrication process [14]. According to previous reports, Si-based InN NWs had asymmetric, conical or tapered structures [15]. For example, Stoica et al. reported that InN NWs formed on Si were tapered and nonuniform along the vertical direction [16]. The asymmetric shapes of the NWs caused significant Fermi-level fluctuations along the NW axial direction, resulting in degradation of one-dimensional properties and of overall device performance [1720]. However, the previously reported InN NWs still had asymmetric and non-uniform shapes with random growth direction [21,22]. Considering these results, it is essential to obtain a comprehensive understanding of the overall growth mechanism so that highly crystalline InN NWs having symmetric and uniform shapes can be formed.

In this paper, we report a newly proposed growth method, the In pre-deposition (InPD) method, to improve the structural properties, including the shape uniformity, of InN NWs. In droplets were initially formed using the InPD method on a Si(111) substrate and act as nucleation centers for subsequent growth of InN NWs. Morphology of InN NWs was manipulated by controlling growth temperature and V/III flux ratio, which is defined as the ratio of the relative amount of nitrogen (N) flux to In flux. Field-emission scanning electron microscopy (FE-SEM), double-crystal X-ray diffraction (DCXRD), and aberration-corrected (Cs) transmission electron microscopy (TEM) measurements were used to analyze the structural properties of the InN NWs. Photoluminescence (PL) spectroscopy was used to analyze the optical properties of the InN NWs.

A schematic diagram of the growth of InN NWs using InPD method is shown in Fig. 1. InN NWs were formed on a p-type Si(111) substrate using a plasma-assisted molecular-beam epitaxy system. In the first growth process of InN NWs, because there is a naturally formed native oxide layer on the surface of the Si(111) substrate, annealing was performed at 900 °C for 1 h inside a chamber to remove it. Because the bonding energy between In and N atoms is stronger than that between In and Si atoms, a nitridation process, in which N-plasma flux was provided only to the Si(111) substrate at a growth temperature of 800 °C, was performed [17,23]. After the nitridation process, SiNx was formed on the surface of Si. Then, for the formation of In droplets, which act as nucleation sites for subsequent growth of InN NWs, only In flux was provided for 6 s to the SiNx/Si(111) surface, without supply of N-plasma flux. As a last step, InN NWs were grown by concurrently providing In and N-plasma fluxes. For the growth of highly crystalline InN NWs with high uniformity and symmetry, the growth temperature and V/III ratio were controlled. The InN NWs were grown at growth temperatures of 450 (NW1), 500 (NW2), 550 (NW3), and 600 °C (NW4) at a fixed V/III ratio of 180 for 4 h. The InN NWs were formed at V/III ratios of 180 (NW2), 145 (NW5), and 111 (NW6) at a fixed growth temperature of 500 °C for 4 h. The V/III ratio was controlled by varying the N-plasma flux at a fixed In flux of 5.7 × 10−8 torr. For clarity, details of the growth conditions of InN NWs are given in Table I. The structural properties of the InN NWs were investigated using FE-SEM (SU-70, Hitachi, installed in the Future Energy Convergence Core Center at Jeonbuk National University), DCXRD (Max-2500, Rigaku), and Cs-TEM (JEM-ARM200F, Jeol). For the PL measurements, a diode-pumped solid-state laser with a wavelength of 1064 nm was used as an excitation source.

Table 1 . Growth conditions and average lengths for InN NWs..

SampleGrowth temperature (°C)V/III ratio (Arb. Units)Length (nm)

Figure 1. Schematic illustration for the formation of InN NWs using InPD method.

Figures 2(a)-(d) shows plan-view (top) and cross-sectional (bottom) FE-SEM images of the NW1, NW2, NW3, and NW4 samples, respectively. In the plan-view FE-SEM images, the surfaces of the InN NWs show hexagonal shapes, indicating that InN NWs have wurtzite (WZ) crystal structures. The average lengths of the NW1, NW2, NW3, and NW4 samples were found to be 308, 407, 529, and 565 nm, respectively. Average length of the InN NWs increases with increasing growth temperature, largely due to the increase in the effective diffusion length of In adatoms. This result can be explained by the following equation [24].

Figure 2. Plan-view and cross-sectional FE-SEM images for the InN NWs; (a) NW1, (b) NW2, (c) NW3, and (d) NW4. (e) Schematic illustration for the growth behavior of In adatoms.

Diffusion length of In adatoms= 1RexpQkT

where R is growth rate, Q is activation energy, k is Boltzmann constant, and T is growth temperature. In this equation, the diffusion length of the In adatoms increases as the growth temperature increases. As a result, the probability for the In atoms to reach the top surface of the InN NWs increases, and, consequently the NW length increases. There are two major growth components of In adatoms that significantly contribute to the growth of InN NWs. These are that In adatoms directly reach the top surface of the NWs and also reach the substrate. Most of the In atoms that impinged on the top surface of the InN NWs are directly nucleated with N atoms, crystallizing the additional InN. On the other hand, the In atoms that arrived at the surface of the Si substrate can move to the top of the NWs through the sidewalls [25]. At the relatively higher growth temperature of 600 °C, because more In adatoms reached the substrate and are able to move to the top region of the NWs due to their longer migration length, their crystallization probability with N adatoms at the bottom region of the NWs may be low [20]. Consequently, InN NWs with slightly tapered shapes were grown, as shown in Fig. 2(d). The growth mechanism of the InN NWs with tapered shapes is schematically illustrated on the left-hand side of the Fig. 2(e). When the growth temperature decreased from 600 to 500°C, the shape uniformity along the vertical direction of the InN NWs improved because the crystallization probability of In adatoms with N atoms increased at the bottom NW region, as schematically illustrated on the right-hand side of Fig. 2(e). The InN NWs of the NW2 sample grown at 500 °C have more uniform shape than those of the NW3 and NW4 samples. However, when the growth temperature decreased below 500 °C, the diffusion length of In adatoms became shorter, and the length of InN NWs decreased, as shown in Fig. 2(a). In addition, for all the NW samples, most of the InN NWs were unidirectionally grown in a direction perpendicular to the substrate, quite different from previous reports [15,16]. The unidirectional growth of InN NWs will surely lead to ease of device fabrication and consequently improve device performance.

The V/III ratio was reduced from 180 to 111 at a fixed growth temperature of 500 °C, to evaluate its effect on the shape and uniformity of InN NWs. Figures 3(a)-(c) shows plan-view (top) and cross-sectional (bottom) FE-SEM images of the NW2, NW5, and NW6 samples, respectively. The average lengths of the NW2, NW5, and NW6 samples were 407, 466, and 490 nm, respectively. As the V/III ratio decreased, the average length of the InN NWs increased and the degree of change to tapered shape intensified because the N-blocking effect on the migration of In adatoms from the substrate to the top surface of NWs through their sidewall decreased [1].

Figure 3. Plan-view and cross-sectional FE-SEM images for the InN NWs; (a) NW2, (b) NW5, and (c) NW6.

Figure 4(a) shows DCXRD patterns of InN NW samples according to growth temperature. DCXRD peaks corresponding to the (0002) crystal plane of InN were observed at 31.30, 31.26, 31.23, and 31.22 ° for the NW1, NW2, NW3, and NW4 samples, respectively. These results are consistent with InN of WZ crystal structure. The peak position corresponding to the (0002) crystal plane of InN increased slightly with decreasing average length of NWs. This peak shift was due to the decrease in tensile strain resulting from the difference in material properties between InN and Si(111) [26]. The full width at half maximum (FWHM) values of the peaks for the (0002) crystal plane of InN were measured as 0.18, 0.16, 0.17, and 0.19 ° for the NW1, NW2, NW3, and NW4 samples, respectively. All FWHM values of InN NWs were significantly narrower than those (~ 0.46 °) in previously reported results [27,28]. This result indicates that InN NWs formed on Si(111) have high crystallinity. Figure 4(b) shows DCXRD patterns of the InN NW samples grown at three different V/III ratios. DCXRD peaks corresponding to the (0002) crystal plane of InN were observed at 31.26, 31.24, and 31.24 ° for the NW2, NW5, and NW6 samples, respectively. With decreasing V/III ratio, the peak position decreased slightly; this is also attributed to the increase in the average length of the NWs. FWHM values of the peaks of the (0002) crystal plane of InN were measured as 0.16, 0.19, and 0.18 ° for the NW2, NW5, and NW6 samples, respectively. In the DCXRD patterns of the NW samples, the NW2 sample had the narrowest FWHM value, indicating that the NW2 sample had the best crystal quality.

Figure 4. DCXRD rocking curves of the InN NW samples according to the (a) growth temperature (b) V/III ratio.

A Cs-TEM image of a single InN NW of the NW2 sample is shown in Fig. 5. High-resolution TEM (HRTEM) images and selective-area electron diffraction (SAED) patterns of the InN NW, measured from three different positions, are shown on the right-hand side of the TEM image. Typically, Si-based III-V compound-semiconductor NWs have significant quantities of stacking faults [14,29]. However, stacking faults were rarely observed in the HRTEM images of the InN NWs in this work. From this investigation, we can conclude that InN NWs with high crystallinity were grown on Si(111) substrates via the InPD method. SAED patterns obtained for the three different regions show the growth directions of [0002] and [01-10] originated from InN NW with WZ crystal structure [30]. This result is consistent with the DCXRD rocking curves shown in Fig. 4.

Figure 5. Cs-TEM image (left) and HRTEM images measured at three different positions along the growth direction of the InN NW, where the inset is the corresponding SAED patterns.

Figure 6 shows the PL spectra of the NW2 sample measured at 10 K and room temperature (RT). The excitation power was set to 3 mW/cm2. A free-exciton (FX) peak was found at the wavelength of 1288 nm with a linewidth of 21 nm at 10 K. Observation of FX peaks from Si-based III-nitride NWs at RT is quite difficult because there are typically many stacking faults and defects inside NWs [14,31]. In other words, a considerable number of carriers generated by light absorption can become trapped in the stacking faults inside NWs. As a result, radiative recombination becomes insufficient and non-radiative recombination increases [32,33]. However, in this paper, a strong FX peak was found at the wavelength of 1,297 nm with a narrow linewidth of 25 nm at RT. The PL peak of the NW2 sample at RT was slightly red-shifted from that at 10 K, because atomic vibration due to heat increased with increasing temperature [34]. The valley observed around the wavelength of 1,330 nm is associated with water absorption [35]. Compared with the previously reported emission wavelength of InN (~ 1.63 µm), the emission wavelength of the InN NWs was significantly blue-shifted. This result can be explained as a combined effect of the quantum confinement effect of the InN NWs and the Moss- Burstein effect [36,37].

Figure 6. PL spectra of the InN NWs measured at 10 K and RT.

In conclusion,highly crystalline InN NWs with symmetric shapes were successfully formed on a Si(111) substrate by adapting the InPD method and varying the growth parameters, growth temperature, and V/III ratio. In FE-SEM images of the InN NWs, symmetric shapes along the vertical growth direction were observed. Strong and narrow DCXRD rocking curves corresponding to InN(0002) were observed. In the HRTEM images, stacking faults and defects were rarely observed. The SAED patterns of the InN NWs show the growth directions of [0002] and [01-10]. A strong FX peak was observed from InN NWs with narrow FWHM values. These results indicate that highly crystalline InN NWs having WZ crystal structure formed on Si(111).

  1. S. Noh, S. Han, I. Choi, J. S. Kim, and M.-Y. Ryu, J. Korean. Phys. Soc. 77, 247 (2020).
  2. T. Inushima, V. V. Mamutin, V. A. Vekshin, S. V. Ivanov, T. Sakon, M. Motokawa, and S. Ohoya, J. Cryst. Growth 227-228, 481 (2001).
  3. Y. Nanishi, Y. Saito, and T. Yamaguchi, Jpn. J. Appl. Phys. 42, 2549 (2003).
  4. M. Higashiwaki and T. Matsui, J. Cryst. Growth 269, 162 (2004).
  5. D. Khandal and S. Jain, Int. J. Inf. Comput. Technol. 4, 1687 (2014).
  6. R. Karthika and S. Balakrishnan, SSRG Int. J. Electron. Commun. Eng. 2, 32 (2015).
  7. T. Inushima, M. Higashiwaki, and T. Matsui, Phys. Rev. B 68, 235204 (2003).
  8. J. Grandal, M. A. Sánchez-Garcia, and E. Calleja, Appl. Phys. Lett. 91, 021902 (2007).
  9. S. Weiszer, A. Zeidler, M. D. L. Mata, and M. Stutzmann, J. Cryst. Growth 510, 56 (2019).
  10. A. T. M. Golam Sarwar, S. D. Carnevale, F. Yang, T. F. Kent, J. J. Jamison, D. W. McComb, and R. C. Myers, Small 11, 5402 (2015).
    Pubmed CrossRef
  11. K. Peng and M. B. Johnston, Appl. Phys. Rev. 8, 041314 (2021).
  12. D. W. Park, S. G. Jeon, C.-R. Lee, S. J. Lee, J. Y. Song, J. O. Kim, S. K. Noh, J.-Y. Leem, and J. S. Kim, Sci. Rep. 5, 16652 (2015).
    Pubmed KoreaMed CrossRef
  13. Y. Wang, V. Schmidt, S. Senz, and U. Gösele, Nat. Nanotechnol. 1, 186 (2006).
    Pubmed CrossRef
  14. S. Eftychis, et al, J. Cryst. Growth 442, 8 (2016).
  15. C.-H. Shen, H.-Y. Chen, H.-W. Lin, and S. Gwo, Appl. Phys. Lett. 88, 253104 (2006).
  16. T. Stoica, R. J. Meijers, R. Calarco, T. Richter, E. Sutter, and H. Lüth, Nano Lett. 6, 1541 (2006).
    Pubmed CrossRef
  17. F. Gao, L. Wen, Z. Xu, J. Han, Y. Yu, S. Zhang, and G. Li, Small 13, 1603775 (2017).
    Pubmed CrossRef
  18. E. Calleja, J. Grandal, and M. A. Sánchez-Garcia, Appl. Phys. Lett. 90, 262110 (2007).
  19. D. Segev and C. G. Van de Walle, Surf. Sci. 601, L15 (2007).
  20. T. Stoica, R. Meijers, R. Calarco, T. Richter, and H. Lüth, J. Cryst. Growth 290, 241 (2006).
  21. J. Segura-Ruiz, N. Garro, A. Cantarero, C. Denker, J. Malindretos, and A. Rizzi, Phys. Rev. B 79, 115305 (2009).
  22. P.-C. Wei, L.-C. Chen, and K.-H. Chen, J. Appl. Phys. 116, 124301 (2014).
  23. S. Feng, J. Tan, B. Li, H. Song, Z. Wu, and X. Chen, J. Alloys Compd. 621, 232 (2015).
  24. X. W. Zhou, D. A. Murdick, B. Gillespie, and H. N. G. Wadley, Phys. Rev. B 73, 045337 (2006).
  25. R. K. Debnath, R. Meijers, T. Richter, T. Stoica, R. Calarco, and H. Lüth, Appl. Phys. Lett. 90, 123117 (2007).
  26. V. M. Kaganer, B. Jenichen, M. Ramsteiner, U. Jahn, C. Hauswald, F. Grosse, S. Fernández-Garrido, and O. Brandt, J. Phys. D: Appl. Phys. 48, 385105 (2015).
  27. M. Zeghouane, et al, CrystEngComm 21, 2702 (2019).
  28. V. Dumitru, S. Costea, M. Brezeanu, G. E. Stan, C. Besleaga, A. C. Galca, G. Ionescu, and O. Ionescu, Sensors 13, 16940 (2013).
    KoreaMed CrossRef
  29. R. S. Chen, H. Y. Tsai, C. H. Chan, Y. S. Huang, Y. T. Chen, K. H. Chen, and L. C. Chen, J. Electron. Mater. 44, 177 (2015).
  30. H. J. Joyce, J. Wong-Leung, Q. Gao, H. Hoe Tan, and C. Jagadish, Nano Lett. 10, 908 (2010).
    Pubmed CrossRef
  31. M. Musolino, A. Tahraoui, S. Fernández-Garrido, O. Brandt, A. Trampert, L. Geelhaar, and H. Riechert, Nanotechnology 26, 085605 (2015).
    Pubmed CrossRef
  32. T. Auzelle, B. Haas, M. D. Hertog, J.-L. Rouvière, B. Daudin, and B. Gayral, Appl. Phys. Lett. 107, 051904 (2015).
  33. Y.-H. Ra, R. Navamathavan, J.-H. Park, and C.-R. Lee, Nano Lett. 13, 3506 (2013).
    Pubmed CrossRef
  34. Y.-F. Wu, J. C. Lee, T.-E. Nee, and J.-C. Wang, J. Lumin. 131, 1267 (2011).
  35. E. Thimsen, B. Sadtler, and M. Y. Berezin, Nanophotonics 6, 1043 (2017).
  36. W. C. Ke, C. P. Fu, C. Y. Chen, L. Lee, C. S. Ku, W. C. Chou, W.-H. Chang, M. C. Lee, and W. K. Chen, Appl. Phys. Lett. 88, 191913 (2006).
  37. K. K. Madapu and S. Dhara, CrystEngComm 18, 3114 (2016).

Share this article on :

Stats or metrics