Influence of ZnO seed layer precursor molar ratio on the density of interface defects in low temperature aqueous chemically synthesized ZnO nanorods/GaN light-emitting diodes , Hatim Alnoor , Galia Pozina, Volodymyr Khranovskyy, Xianjie Liu, Donata Iandolo, Magnus Willander, and Omer Nur
Citation: J. Appl. Phys. 119, 165702 (2016); doi: 10.1063/1.4947593 View online: http://dx.doi.org/10.1063/1.4947593 View Table of Contents: http://aip.scitation.org/toc/jap/119/16 Published by the American Institute of Physics
JOURNAL OF APPLIED PHYSICS 119, 165702 (2016)
Influence of ZnO seed layer precursor molar ratio on the density of interface defects in low temperature aqueous chemically synthesized ZnO nanorods/GaN light-emitting diodes Hatim Alnoor,1,a) Galia Pozina,2 Volodymyr Khranovskyy,2 Xianjie Liu,2 Donata Iandolo,1 Magnus Willander,1 and Omer Nur1 1
Department of Science and Technology (ITN), Link€ oping University, SE-601 74 Norrk€ oping, Sweden Department of Physics, Chemistry and Biology (IFM), Link€ oping University, SE-583 81 Link€ oping, Sweden
2
(Received 25 November 2015; accepted 14 April 2016; published online 27 April 2016) Low temperature aqueous chemical synthesis (LT-ACS) of zinc oxide (ZnO) nanorods (NRs) has been attracting considerable research interest due to its great potential in the development of light-emitting diodes (LEDs). The influence of the molar ratio of the zinc acetate (ZnAc): KOH as a ZnO seed layer precursor on the density of interface defects and hence the presence of non-radiative recombination centers in LT-ACS of ZnO NRs/GaN LEDs has been systematically investigated. The material quality of the as-prepared seed layer as quantitatively deduced by the X-ray photoelectron spectroscopy is found to be influenced by the molar ratio. It is revealed by spatially resolved cathodoluminescence that the seed layer molar ratio plays a significant role in the formation and the density of defects at the n-ZnO NRs/p-GaN heterostructure interface. Consequently, LED devices processed using ZnO NRs synthesized with molar ratio of 1:5 M exhibit stronger yellow emission (575 nm) compared to those based on 1:1 and 1:3 M ratios as measured by the electroluminescence. Furthermore, seed layer molar ratio shows a quantitative dependence of the non-radiative defect densities as deduced from light-output current characteristics analysis. These results have implications on the development of high-efficiency ZnO-based LEDs and may also be helpful in understanding the effects of the ZnO seed layer on defect-related non-radiative recombination. Published by AIP Publishing. [http://dx.doi.org/10.1063/1.4947593]
I. INTRODUCTION
One-dimensional (1D) zinc oxide (ZnO) nanostructures are promising material for many optoelectronic devices due to its excellent optical and electrical properties.1–3 Among the 1D ZnO nanostructures, the low temperature aqueous chemically synthesized (LT-ACS) ZnO nanorods (NRs) and nanowires (NWs) have recently attracted considerable research interest, due to their potential for the development of intrinsic white light-emitting diodes (LEDs).1,4,5 However, due to the difficulty of producing stable and reproducible p-type ZnO, an alternative approach is utilized to synthesize n-type ZnO NRs on p-type substrate such as GaN to realize ZnO-based heterojunction LEDs.6–10 However, LEDs based on such configuration usually exhibit low electroluminescence (EL) efficiency, resulting from a high density of defects and different energy barriers for electrons and holes at the heterojunction interfaces.10 It is generally known that the interface defect states can act as non-radiative centers that will drastically degrade the performance of ZnO-based heterojunction LEDs.11,12 Recently, much efforts have been made to reduce the energy barriers by introducing doping elements to overcome the potential barriers between ZnO and GaN in order to improve ZnO-based heterojunction LEDs EL efficiency.8,13–15 In addition, it has been reported that introducing an intermediate
a)
Author to whom correspondence should be addressed. Electronic mail:
[email protected]
0021-8979/2016/119(16)/165702/7/$30.00
layer between ZnO and GaN can reduce the potential barriers and hence improve the EL efficiency of the ZnO-based heterojunction.16–19 A typical LT-ACS consists of two main steps: the p-GaN substrates are first coated with a ZnO seed layer and then NRs are subsequently grown from a zinc ion containing solution. The initial presence of the ZnO seed layer is required in order to induce the synthesis nucleation site and improve the vertical alignment of the NRs.20,21 Recently, we have performed a systematic study to optimize the emission intensity by choosing the best synthesis nutrients stirring process.22 Previously, it has been reported that the type of the seed layer significantly affects the density of interface defects, and hence different color emissions were observed.23–25 However, until now, manipulating the ZnO seed layer synthesis, which can be a simple and an effective approach to tune the defects (and hence also non-radiative defects) density at the n-ZnO NRs/p-GaN interfaces, has not been investigated. Particularly, to the best of our knowledge, no studies have been performed to investigate the effect of the density of nonradiative defects at the heterojunction interface on the ZnObased heterojunction LEDs EL efficiency. Thus, in this work, we demonstrate the possibility to tune the density of the defects at the n-ZnO NRs/p-GaN interface utilizing different ZnO seed layer precursor molar ratios. We have demonstrated an improvement in the yellow EL intensity at the ratios of 1:5 M. Different analytical techniques for structural, optical, and electrical characterization were used. The observed results have a potential for further
119, 165702-1
Published by AIP Publishing.
165702-2
Alnoor et al.
development of high-efficiency ZnO-based LEDs and are also helpful in understanding the effect of the ZnO seed layer on the defect-related non-radiative recombination. II. EXPERIMENTAL
ZnO NRs based heterojunction LEDs were synthesized via the low temperature aqueous chemical synthesis on commercially available p-GaN substrates (which were magnesium-doped GaN (0001)-oriented layer grown on sapphire). ZnO seed layers were prepared by mixing zinc (II) acetate dihydrate (Zn (Ac)) and KOH in methanol as described in Ref. 21. In our experiments, conditions (mainly the precursor molar ratio of ZnAc/KOH) were selected in order to prepare ZnO seed layers with 1:1, 1:3, and 1:5 M molar ratios. The resulting three ZnO seed layers were spin coated on the p-GaN substrates, and the NRs were subsequently grown from a synthesis solution of 0.05 M zinc nitrate hexahydrate and 0.075 M HMTA heated to 80 C for 6 h. Details on the synthesis of the ZnO NRs are described in Ref. 22. For the fabrication of the LED, the as-synthesized n-ZnO NRs/p-GaN heterostructures were spun-coated with an insulating layer of Shipley 1805 photo-resist to isolate electrical contact on the ZnO NRs from the p-GaN substrates. Then, reactive ion etching (RIE) was used to remove the photo-resist from the ZnO NRs surface and expose the NRs tip. Finally, Ni/Au (15/ 35 nm) were thermally evaporated onto the p-GaN substrates and Ag (40 nm) on the exposed ZnO NRs tip to serve as p-type and n-type contact electrode, respectively. The p-type and n-type contacts were evaporated in circular area of diameter 2 mm. In the following, the as-fabricated heterojunction LEDs are labeled as device 1, 3, and 5 for the ZnO NRs synthesized with molar ratios of 1:1, 1:3, and 1:5 M, respectively. The morphology and the surface roughness of the samples were characterized using field-emission scanning electron microscopy (FE-SEM) Gemini LEO 1550 and Atomic force microscopy (AFM, Veeco-3100), respectively. The crystal structure and the properties of the as-prepared ZnO
J. Appl. Phys. 119, 165702 (2016)
seed layers with different molar ratios were characterized using X-ray powder diffraction (XRD) using a Philips PW1729 diffractometer equipped with Cu-Ka radiation ˚ ) and X-ray photoelectron spectroscopy (XPS) (k ¼ 1.5418 A measurements recorded by Scienta ESCA-200 spectrometer using monochromatic Al Ka x-ray source (1486.6 eV). The defect concentration at the n-ZnO NRs/p-GaN interface was investigated by spatially resolved cathodoluminescence (CL) using Gatan Mono CL4 system combined with Gemini LEO 1550 FE-SEM. The current-voltage (I-V) characteristics were measured using a semiconductor parameter analyzer (Keithley 2400-SCS). The EL measurements were performed using Keithley 2400 source to provide a fixed voltage and the emission spectra were collected using a SR-303i-B detection system. The macro-photoluminescence (M-PL) spectra were acquired using a 266 nm continuous wave laser as the excitation source. All the measurements were carried out at room temperature (RT). III. RESULTS AND DISCSSION
Figure 1 shows the top-view FE-SEM images of the as-prepared ZnO seed layers with different molar ratios and the corresponding ZnO NRs images grown on top of them. The as-prepared seed layers are polycrystalline and composed of ZnO nanoparticles (NPs) with a spherical like shape, as shown in Figs. 1(a) and 1(c). The average diameter of ZnO NPs has been estimated from the top-view FE-SEM images, and it was found to be decreased from 50 to 30 nm as the molar ratios increased from 1:1 to 1:5 M. Moreover, the as-prepared seed layers have the same thickness ˚ ), as deduced from cross-sectional view FE-SEM (55.0 A images as presented in Table I. As can be seen from Figs. 1(d) and 1(f), there is slight difference in the density of the as-synthesized ZnO NRs. This can be attributed to the dominating nucleation mechanism during the synthesis process of NRs. The seed layer with a relatively small grain boundary and low surface roughness favors for grain boundary
FIG. 1. Top-view FE-SEM images of the as-prepared ZnO seed layers with a molar ratio of (a) 1:1 M, (b) 1:3 M, and (c) 1:5 M, respectively. (d), (e), and (f) are the corresponding ZnO NRs SEM images grown on top of the seed shown in (a), (b), and (c), respectively.
165702-3
Alnoor et al.
J. Appl. Phys. 119, 165702 (2016)
TABLE I. Some structural properties of the as-prepared ZnO seed layer. Molar Seed layer average Seed layer Seed layer average ratio (M) diameter (nm) thickness (nm) RMS roughness (nm) OII/OI 1:1 1:3 1:5
50 40 30
55.5 55.0 56.0
1.85 6 0.02 1.84 6 0.11 2.65 6 0.17
0.56 0.64 0.75
nucleation, while large grain boundary and high surface roughness favors free surface nucleation.26 The seed layers’ root-mean square (RMS) roughness was investigated using AFM measurements for a scanned area of 2 2 lm as shown in Fig. 2. As can be seen from Fig. 2 and Table I, the asprepared seed layer with molar ratio of 1:5 M displays the highest surface roughness compared to the 1:1 and 1:3 M, respectively. In addition, the prepared seeds with molar ratio of 1:1 M and 1:3 M seem to have the same nanoparticle size distribution, whereas smaller nanoparticles can be seen for 1:5 M. Moreover, as it is possible to appreciate from Fig. 2(c), particles’ aggregates were present on the surface of the seed prepared with 1:5 M. This might be the reason to observe relatively larger measured roughness. As confirmed by X-ray powder diffraction measurement (not shown here), the as-synthesized ZnO NRs were hexagonal wurtzite structure (JCPDS Card No. 36–1451) and have a preferred growth orientation along the c-axis. To investigate the influence of the molar ratios on the material quality of the as-prepared ZnO seed layers, the X-ray photoelectron spectroscopy (XPS) measurements were performed. Figure 3(a) shows the O 1s core level XPS spectra of the as-prepared ZnO seed layers. As shown in Fig. 3(a), the O 1s peak for all samples exhibits an asymmetric profile, which can be decomposed to three Gaussian peaks. The main peak at low binding energy centered at 529.95 eV is related to the Zn-O (OI) bond within the ZnO crystal lattice.27–29 The peak at medium binding energy centered at 531.33 eV is related to the oxygen-deficient regions (OII) within the ZnO crystal lattice.27 Finally, the peak at higher binding energy centered at 532.37 eV is related to the absorbed oxygen on the ZnO surface (OIII), e.g., H2O and O2.27–29 The relative intensity ratios of the O/Zn calculated using integrated XPS peak areas and the element sensitivity of the O and Zn are summarized in Table I for all samples. It is found that the
relative intensity ratio of the OII/OI increases as the molar ratio increased from 1:1 M to 1:5 M, suggesting the presence of different oxygen-point defects sites. The concentration of the oxygen point defects is relatively higher for seed layer prepared with molar ratio of 1:5 M. The optical band gap of the as-synthesized ZnO NRs with different molar ratios was obtained using the following equation:8 1
ðahtÞ ¼ Bðht Eg Þ2 ;
(1)
where a is the optical absorption coefficient, ht is the photon energy, B is a constant coefficient, and Eg is the optical band gap energy. The optical band gap values were estimated by the Tauc plot as shown in Fig. 3(b) and were found to be 3.10, 3.15, and 3.20 eV for the NRs synthesized with molar ratios of 1:1, 1:3, and 1:5 M, respectively. Furthermore, the influence of the ZnO seed layer molar ratios on the density of defects of the as-synthesized n-ZnO NRs/p-GaN heterostructure interfaces was investigated using spatially resolved CL spectra. Figure 4 shows a number of specific CL spectra acquired in cross-sectional view. The electron beam at the acceleration voltage of 5 kV was focused to a spot with the diameter of 20 nm. The inset in Fig. 4(a) shows a typical cross-sectional SEM image of the n-ZnO NRs/p-GaN heterostructure with the indications of the points, where the CL spectra were taken, i.e., at the p-GaN substrate (black circle), at the n-ZnO NRs/p-GaN interface (red circle) and at the bottom of the ZnO NRs (blue circle). Interestingly, our results demonstrate that the seed layer molar ratios have significant effect on the density of defects at the n-ZnO NRs/p-GaN heterostructure interface. As can be observed, the as-synthesized n-ZnO NRs/p-GaN heterostructure obtained for the seed layer produced with a molar ratio of 1:3 M has a high defect emission (and hence high defect concentration) at the interface (red curve). This result indicated that the seed molar ratio has significant effect on the formation of the density on the interface defects. To evaluate the influence of the ZnO seed layer molar ratios on carrier transport and the quality of the interface at the heterojunction, the as-synthesized n-ZnO NRs/p-GaN heterostructures were used to fabricate LED devices. The
FIG. 2. AFM of the as-prepared ZnO seed layers with molar ratio of (a) 1:1 M, (b) 1:3 M, and (c) 1:5 M. Scanned area: 2 2 lm.
165702-4
Alnoor et al.
J. Appl. Phys. 119, 165702 (2016)
FIG. 3. (a) O 1s XPS spectra of the asprepared ZnO seed layers with different molar ratios indicated for each spectrum and (b) plot of ðahtÞ2 versus ht of the as-synthesized ZnO NRs.
FIG. 4. CL spectra of n-ZnO NRs/p-GaN heterostructures measured in cross-sectional view. The inset shows a typical cross-sectional SEM image of the n-ZnO NRs/p-GaN heterostructure with the indication of the point where the CL spectra were taken.
schematic structure of the device is illustrated in the inset of Fig. 5. The resulting heterojunction devices exhibited diodelike (I-V) behavior with a low turn-on voltage of 1.2, 1.6, and 2 V for the devices 5, 3, and 1, respectively, as shown in Fig. 5. As can be seen, device 5 is displaying the best rectification properties. The difference in the turn-on voltage and the injection current among the three devices is likely attributed to the difference in the defect concentration at the junction interface,9 as suggested from the results of Figs. 4(a) and 4(c). Under reverse bias, the results indicate the presence of leakage breakdown. This may be due to the defectmediated tunneling caused by the high defect concentration or trap centers in the n-ZnO NRs/p-GaN interface.12,30,31 Furthermore, to investigate the influence of the ZnO seed layer molar ratios on the EL emission, the EL spectra from all the three LED devices were monitored at different forward voltages/injection-currents as shown in Fig. 6. The insets in Figs. 6(a) and 6(c) show the corresponding light emission images at 24 V. As can be seen, the EL spectra of all the three devices exhibit a distinct yellow emission peak
FIG. 5. I-V curves of all the three devices. The inset shows the schematic structure of the device.
165702-5
Alnoor et al.
J. Appl. Phys. 119, 165702 (2016)
FIG. 6. EL spectra as function of the forward biased voltage of device (a) 1, (b) 3, and (c) 5, respectively. The insets in (a), (b), and (c) show the corresponding light emission images at 24 V. (d) The integrated EL intensity of three devices as a function of the forward biased voltage.
centered at 575 nm which can be attributed to the intrinsic deep level defect VO and Oi in ZnO8,11,13,15,18 or be explained by the interface defects related emission.8,20 However, besides yellow band, a weak blue emission centered at 420 nm was observed for device 1 (with a 1:1 M seed layer) as shown in Fig. 6(a). This observation can be ascribed to the transitions from the conduction band or shallow donors (in ZnO) to the Mg acceptor levels in the p-GaN.9 As the bias voltage is increased, the EL intensity of the dominant emission peak for all the three LEDs increases
rapidly without significant shift in the peak position as shown in Fig. 6(d). The improvement in the EL intensity with increasing the biased voltage is probably attributed to the reduction in the band bending of the n-ZnO NRs and the p-GaN.11,13 As a result, the kinetic energy of the electrons and holes is increased and they have much higher probability to go across the interface barrier and recombine on the opposite side of the junction.11,13 For comparison, the EL spectra at 24 V from all the three devices are shown in Fig. 7(a). Interestingly, a significant improvement in the yellow EL intensity is observed
FIG. 7. (a) EL spectra of all the three devices under forward bias voltage of 24 V. (b) The PL spectra of the p-GaN substrate and the three ZnO NRs/pGaN heterostructures. For clarity, all the PL spectra are normalized to the DLE and shifted in vertical direction.
165702-6
Alnoor et al.
from device 5 compared with devices 1 and 3. This occurs probably due to both the low density of the interface defects, which is favorable for increasing carrier injection through the heterojunction,32 and the low density of non-radiative recombination centers in the space-charge region. All the devices exhibited good stability, with light emission still observed after 3 months of storage in air. To explain the origin of the EL emission of the heterojunction devices, the PL spectra of the p-GaN substrate and n-ZnO NRs/p-GaN heterostructure were obtained and the corresponding data are shown in Fig. 7(b). The PL spectra of all the three n-ZnO NRs/p-GaN heterostructures exhibited weak UV emission peak centered at 381 nm, and a dominant broad yellow emission peak centered at 585 nm. The 381 nm UV emission is due to the near-band-edge (NBE) of ZnO, and the yellow emission is most likely attributed to the intrinsic deep level defect VO, Oi in ZnO.8,20,23 The PL spectra of the p-GaN substrate exhibited blue emission at 400 nm, which is ascribed to the transitions from the conduction band or shallow donors to the Mg acceptor levels.8,9,22 In view of the obtained PL results, we can attribute the blue emission at 420 nm in EL to the Mg acceptor levels in p-GaN or the interface defects related emission and the yellow emission at 575 nm to the intrinsic deep level defect VO and Oi in ZnO NRs. However, the peak position of the yellow EL emission is red shift by 10 nm compared with the yellow PL emission. This is possibly due to the fact that the EL and PL have different mechanisms. The recombination occurs at the interface between ZnO NRs and p-GaN in the EL measurements, while in the PL measurements, it takes place within the optical penetration depth from the top of the nanorods surface.23,32 To clarify the mechanism of the improved EL intensity of the yellow emission from the n-ZnO NRs/p-GaN heterostructure based LEDs by varying the ZnO seed layer molar ratios, the integrated intensity of the EL peak as a function of the injection current in a log-log scale is plotted in Fig. 8 for all the three LEDs. The results can be fitted with the law L ¼ cIm, where m accounts for the influence of non-radiative defects in the characteristics of light emission.32 A linear increase in L with I, i.e., m 1 can be expected when
J. Appl. Phys. 119, 165702 (2016)
radiative recombination dominates, while non-radiative recombination becomes dominant, when L shows a superlinear dependence on I, i.e., when m > 1.33 Our result shows a superlinear dependence with m ¼ 2.3, 4.0, and 1.9 for the devices 1, 3, and 5, respectively. This implies a role of the defect-related non-radiative recombination centers for all the three LEDs, which is in good agreement with the CL spectra in Fig. 4 and the PL spectra in Fig. 7(b). However, the value of m determined from the fitting curve for the device 5 (with 1:5 M) is the lowest (m ¼ 1.9). This reduced value indicates a low presence of introduced non-radiative recombination centers in the space-charge region. Since the low density of the interface defects is favorable for improving carrier injection through the heterojunction, the device 5 shows an improved EL performance compared to both devices 1 and 3. These results suggest that the molar ratios of the ZnO seed layer have significant effect on the density of interface defects and, in particular, playing a role of non-radiative recombination centers. IV. CONCLUSION
The influence of the molar ratio of zinc acetate: KOH used as precursors to prepare the ZnO seed layer on the density of interface defects and hence the presence of nonradiative recombination centers in low temperature aqueous chemically synthesized n-ZnO NRs/p-GaN LEDs has been investigated. The results show that, indeed, the molar ratio of the ZnAc: KOH in seed layer preparation has a significant effect on the density of interface defects and hence on the current-voltage (I-V) curves of the LED devices, while no effect of varying the molar ratio on the morphology and crystal structure has been observed. Electroluminescence (EL) measurements show that the LED devices based on ZnO NRs synthesized with a molar ratio of 1:5 M exhibit a stronger yellow emission centered at 575 nm than the LEDs fabricated using ZnO NRs based on 1:1 and 1:3 M, which is attributed to the dominant role of defect-related non-radiative recombination. Furthermore, the seed layer molar ratio shows a quantitative dependence of the non-radiative defect densities as deduced from light-output current (L-I) characteristics analysis. The findings in the study are of a potential for the development of high-efficiency ZnO-based LEDs and may also be helpful in understanding the effect of the ZnO seed layers on the formation of defect-related non-radiative recombinations. ACKNOWLEDGMENTS
Partial financial support from the Swedish Government Strategic Research Area in Materials Science on Functional Materials at Link€oping University (Faculty Grant No. SFOMat-LiU # 2009-00971) is highly appreciated. 1
FIG. 8. Integrated EL intensities of all the three devices as a function of the forward injections current. The solid lines represent the fitting results based on the power law L ¼ cIm.
M. Willander, O. Nur, Q. X. Zhao, L. L. Yang, M. Lorenz, B. Q. Cao, J. Z u~ niga Perez, C. Czekalla, G. Zimmermann, M. Grundmann, A. Bakin, A. Behrends, M. Al-Suleiman, A. El-Shaer, A. Che Mofor, B. Postels, A. Waag, N. Boukos, A. Travlos, H. S. Kwack, J. Guinard, and D. Le Si Dang, Nanotechnology 20, 332001 (2009). 2 A. B. Djurisic´, A. M. C. Ng, and X. Y. Chen, Prog. Quantum Electron. 34, 191 (2010).
165702-7
Alnoor et al.
€ ur, Ya. I. Alivov, C. Liu, A. Teke, M. A. Reshchikov, S. Do U. Ozg€ gan, V. Avrutin, S.-J. Cho, and H. Morkoc¸, J. Appl. Phys. 98, 041301 (2005). 4 A. Zainelabdin, S. Zaman, G. Amin, O. Nur, and M. Willander, Nanoscale Res. Lett. 5, 1442 (2010). 5 A. Baltakesmez, S. Tekmen, and S. T€ uzemen, J. Appl. Phys. 110, 054502 (2011). 6 W. I. Park and G. C. Yi, Adv. Mater. 16, 87 (2004). 7 O. Lupan, T. Pauporte, B. Viana, I. M. Tiginyanu, V. V. Ursaki, and R. Corte`s, ACS Appl. Mater. Interfaces 2, 2083 (2010). 8 A. Echresh, C. O. Chey, M. Z. Shoushtari, O. Nur, and M. Willander, J. Appl. Phys. 116, 193104 (2014). 9 X. Chen, A. M. C. Ng, F. Fang, Y. H. Ng, A. B. Djurisic´, H. L. Tam, K. W. Cheah, S. Gwo, W. K. Chan, P. W. K. Fong, H. F. Lui, and C. Surya, J. Appl. Phys. 110, 094513 (2011). 10 Z. Shi, Y. T. Zhang, X. J. Cui, B. Wu, S. W. Zhuang, F. Yang, X. T. Yang, B. L. Zhang, and G. T. Zhang, Appl. Phys. Lett. 104, 131109 (2014). 11 S. Xu, C. Xu, Y. Liu, Y. F. Hu, R. Yang, Q. Yang, J. H. Ryou, H. J. Kim, Z. Lochner, S. Choi, R. Dupuis, and Z. L. Wang, Adv. Mater. 22, 4749 (2010). 12 H. Y. Lee, C. T. Lee, and J. T. Yan, Appl. Phys. Lett. 97, 111111 (2010). 13 G. C. Park, S. M. Hwang, S. M. Lee, J. H. Choi, K. M. Song, H. Y. Kim, H.-S. Kim, S.-J. Eum, S.-B. Jung, J. H. Lim, and J. Joo, Sci. Rep. 5, 10410 (2015). 14 O. Lupan, T. Pauporte, L. Chow, G. Chai, B. Viana, V. V. Ursaki, E. Monaico, and I. M. Tiginyanu, Appl. Surf. Sci. 259, 399 (2012). 15 O. Lupan, T. Pauporte, B. Viana, V. V. Ursaki, I. M. Tiginyanu, V. Sontea, and L. Chow, J. Nanoelectron. Optoelectron. 7, 712 (2012). 16 J.-J. Don, H.-Y. Hao, J. Xing, Z.-J. Fan, and Z.-L. Zhang, Nanoscale Res. Lett. 9, 630 (2014). 17 Z. Guo, H. Zhang, D. Zhao, Y. Liu, B. Yao, B. Li, Z. Zhang, and D. Shen, Appl. Phys. Lett. 97, 173508 (2010). 18 H. H. Huang, G. J. Fang, Y. Li, S. Z. Li, X. M. Mo, H. Long, H. N. Wang, D. L. Carroll, and X. Z. Zhao, Appl. Phys. Lett. 100, 233502 (2012). 3€
J. Appl. Phys. 119, 165702 (2016) 19
S. G. Zhang, X. W. Zhang, Z. G. Yin, J. X. Wang, J. J. Dong, Z. G. Wang, S. Qu, B. Cui, A. M. Wowchak, A. M. Dabiran, and P. P. Chow, J. Appl. Phys. 109, 093708 (2011). 20 L. E. Greene, M. Law, D. H. Tan, M. Montano, J. Goldberger, G. Somorjai, and P. Yang, Nano Lett. 5, 1231 (2005). 21 A. Zainelabdin, S. Zaman, G. Amin, O. Nur, and M. Willander, Cryst. Growth Des. 10, 3250 (2010). 22 H. Alnoor, C. O. Chey, G. Pozina, X. Liu, V. Khranovskyy, M. Willander, and O. Nur, AIP Adv. 5, 087180 (2015). 23 A. M. C. Ng, X. Y. Chen, F. Fang, Y. F. Hsu, A. B. Djurisic´, C. C. Ling, H. L. Tam, K. W. Cheah, P. W. K. Fong, H. F. Lui, C. Surya, and W. K. Chan, Appl. Phys. B 100, 851 (2010). 24 A. M. C. Ng, Y. Y. Xi, Y. F. Hsu, A. B. Djurisic´, W. K. Chan, S. Gwo, H. L. Tam, K. W. Cheah, P. W. K. Fong, H. F. Lui, and C. Surya, Nanotechnology 20, 445201 (2009). 25 X. Y. Chen, A. M. C. Ng, F. Fang, A. B. Djurisic´, W. K. Chan, H. L. Tam, K. W. Cheah, P. W. K. Fong, H. F. Lui, and C. Surya, J. Electrochem. Soc. 157(3), H308 (2010). 26 S.-W. Chen and J.-M. Wu, Acta Mater. 59, 841 (2011). 27 J. D. Ye, S. L. Gu, F. Qin, S. M. Zhu, S. M. Liu, X. Zhou, W. Liu, L. Q. Hu, R. Zhang, Y. Shi, and Y. D. Zheng, Appl. Phys. A 81, 809 (2005). 28 A. B. Djurisic´, Y. H. Leung, K. H. Tam, Y. F. Hsu, L. Ding, W. K. Ge, Y. C. Zhong, K. S. Wong, W. K. Chan, H. L. Tam, K. W. Cheah, W. M. Kwok, and D. L. Phillips, Nanotechnology 18, 095702 (2007). 29 E. De la Rosa, S. Sep ulveda-Guzman, B. Reeja-Jayan, A. Torres, P. Salas, N. Elizondo, and M. Jose Yacaman, J. Phys. Chem. C 111, 8489 (2007). 30 S. Y. Liu, T. Chen, Y. L. Jiang, G. P. Ru, and X. P. Qu, J. Appl. Phys. 105, 114504 (2009). 31 B. You, X. W. Zhang, S. G. Zhang, H. R. Tan, J. Ying, Z. G. Yin, Q. S. Zhu, and P. K. Chu, J. Appl. Phys. 107, 083701 (2010). 32 J. J. Dong, X. W. Zhang, Z. G. Yin, J. X. Wang, S. G. Zhang, F. T. Si, H. L. Gao, and X. Liu, Appl. Phys. Lett. 100, 171109 (2012). 33 S. G. Zhang, X. W. Zhang, J. X. Wang, J. B. You, Z. G. Yin, J. J. Dong, B. Cui, A. M. Wowchak, A. M. Dabiran, and P. P. Chow, Phys. Status Solidi RRL 5, 74 (2011).
