Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group

Formation of c-axis oriented ZnO optical waveguides by radio-frequency magnetron sputtering

Open Access Open Access

Abstract

ZnO/Mg0.16Zn0.84O (ZnO/MgZnO) films are fabricated on x-cut and z-cut LiNbO3 (LN) substrates by radio-frequency magnetron sputtering. High transparencies are confirmed by a spectrophotometer. X-ray diffraction (XRD) spectra show that all the films are c-axis oriented. The waveguiding properties, as well as the refractive indices and thickness of the films are demonstrated and determined by prism coupling. Both transverse electric (TE) and transverse magnetic (TM) modes are measured at λ=0.633 µm and 1.539 µm, respectively. The waveguide loss is measured at λ=0.633 µm with a fiber probe technique. The experimental results show that high optical quality ZnO films can be obtained with MgZnO buffer layers.

©2005 Optical Society of America

1. Introduction

As a wide bandgap wurtzite semiconductor, zinc oxide (ZnO) is an attractive material for potential applications in electro-optic and acousto-optic devices. It has a wide direct band gap (~3.37 eV) and a large exciton binding energy (~60 meV), which makes it one of the promising candidates for the development of ultra-violet (UV) optoelectronics [1]. Several methods have been used for the fabrication of ZnO films, such as radio-frequency magnetron sputtering, pulsed laser deposition, metalorganic chemical vapor deposition and laser molecular beam epitaxy. The corresponding structures and optical characteristics of the films have been intensively investigated in recent years [24]. Recently, ZnO-based waveguide structures on different substrates have been demonstrated, and suggest possible applications of ZnO films in optoelectronics and integrated optics [58]. In addition to a wide bandgap, ZnO film also exhibits a large piezoelectricity and nonlinearity as a consequence of the noncentrosymmetry of wurtzite. In most of the reported experiments, the refractive indices of ZnO films, important optical parameters, were demonstrated to be close to and slightly lower than those of bulk ZnO. Generally, lower-index substrates (e.g., glass and sapphire) or buffer layers grown on higher-index substrates (e.g., silicon dioxide grown on silicon by thermal oxidation) are prevalently used as the substrates of ZnO-based waveguide structures. However, the quality of such ZnO film is limited because of the inherent in-plane lattice mismatch between ZnO film and the substrate, although in most cases ZnO film exhibits a tendency to grow along its c-axis, irrespective of surface crystallinity. The ternary alloy MgxZn1-xO is of great interest due to its application in wide bandgap engineering and quantum well heterostructure. It is believed that it may eventually compete with the group III nitrides material. Having a hexagonal structure, the lattice constant of MgxZn1-xO (x<0.33) film is only slight different from that of ZnO (only 1%), usually with a small expansion in a-axis and compression in c-axis comparing to those of ZnO, which changes of lattice constants are attributed to the incorporated Mg substituting Zn site in original ZnO lattice structure [9]. The refractive indices of MgxZn1-xO were demonstrated to be lower than those of ZnO [10]. Since MgxZn1-xO exhibits a more preferable lattice-matched to ZnO, a guiding ZnO film with better optical properties can be expected if grown on MgxZn1-xO substrate. In this work, the low-index Mg0.16Zn0.84O (MgZnO) film is firstly grown on LiNbO3 substrate to provide an optical confinement for ZnO waveguide. Thereafter the ZnO guiding film is deposited on MgZnO buffer layer. The optical and waveguiding properties are measured and discussed.

2. Experiments

ZnO/Mg0.16Zn0.84O (ZnO/MgZnO) films are grown on x-cut and z-cut LiNbO3 (LN) substrates by radio-frequency magnetron sputtering, respectively. The ZnO target (99.99% purity, 80 mm diameter) is fixed on a magnetron-effect cathode, and the substrate holder is situated in front of the target at 50 mm distance. All the depositions are performed in argon-oxygen (1:1) mixture ambient under a pressure of 1.0 Pa. The temperature in chamber is at about 350 K, the radio-frequency (RF) power is at 200 W. Before depositing ZnO film, a MgZnO alloyed layer is sputtered on LN substrate and subsequently annealed to obtain highly textured film, which acts as a basis for the following deposition. Later, the same process is carried out for the preparation of ZnO film. The annealing treatment is performed at 400°C for 60 minutes in oxygen ambient. The sputtering times are 60 minutes for the deposition of MgZnO film, and 120 minutes for that of ZnO, respectively. The film structures are studied by XRD patterns. Prism coupling technique is employed to characterize the waveguide performance of all the films and to determine the waveguide parameters (such as refractive indices and thickness). In the following discussion the ZnO/MgZnO films deposited on x-cut and z-cut LN substrates are assigned as films S1 and S2, respectively.

3. Results and discussion

Figure 1 shows the wavelength dependence of optical transmittance spectra of both ZnO and MgZnO films deposited on z-cut LN substrates. Each film shows an average transmittance of about 90% in wavelength region 430–2000 nm for ZnO film and 360–2000 nm for MgZnO film, respectively. The oscillations in both spectra attest a good quality of the deposited films [11], which is also consistent with the result in Ref. [2]. The absorption edges are at about 389nm and 346nm for ZnO and MgZnO films deposited, respectively According to relation Ea=1.24/λ, the corresponding band energy gaps are calculated to be 3.18eV and 3.58eV, respectively. So a blue shift of 0.4eV is obtained.

 figure: Fig. 1.

Fig. 1. Optical transmittance diagram of ZnO film (solid line) and Mg0.16Zn0.84O film (dotted line) grown on z-cut LiNbO3 substrates by radio-frequency magnetron sputtering under a pressure of 1.0 Pa at 200 W RF-power at 350 K.

Download Full Size | PDF

 figure: Fig. 2.

Fig. 2. X-ray diffraction patterns of ZnO/MgZnO films deposited on (a) x-cut LN and (b) z-cut LN. The inlet figures exhibit the x-ray diffraction patterns of MgZnO films.

Download Full Size | PDF

Figure 2 shows the XRD θ-2θ scans of ZnO/MgZnO films, where the inlet figures show the x-ray diffraction patterns of annealed MgZnO films deposited on (a) x-cut and (b) z-cut LN substrates, respectively. For the inlet figure in Fig. 2(a), the MgZnO (002) peak is obtained at 34.14°, which coexists with a LN (110) peak at 34.78°. In the inlet figure in Fig. 2(b), a strong MgZnO (002) peak is exhibited at 34.42° as well as a weak peak from LN (006) at 38.92°. The obvious strong peak intensity of MgZnO (002) in the inlet Fig. 2(b) shows a better crystal orientation of MgZnO grown on z-cut LN than that grown on x-cut LN. In the two figures, it can be found that both films exhibit strong ZnO (002) diffractive peaks, which indicates the completely c-axis oriented growth. The ZnO (002) diffractive peaks for annealed films S1 and S2 are at 34.06° and 34.34°, and the corresponding full width at half maximum (FWHM) of both peaks are 0.68° and 0.33°, respectively. The small value of FWHM indicates a better crystalline orientation of S2 compared with that of S1. The results are in accordance with the experimental results from Yamamoto et al [12]. The measured values of the lattice constant c are 5.2602 Å and 5.2068 Å for S1 and S2, respectively, which are slightly larger than that of the bulk material. The c-axis lattice constants of annealed films are contracted in comparison with those of as-grown films. In fact, in spite of the highly c-axis orientation, the deposited films are demonstrated to be polycrystalline with in-plane grains. The grain size can be calculated via Scherrer’s formula [13],

d=0.9λBcosθB

where, λ, θB and B are the x-ray wavelength (1.5405 Å), Bragg diffraction angle and line width at half-maximum, respectively. For the presented films the typical grain size is calculated to be 26 nm in annealed film S2.

The prism coupling measurement (with Metricon Model 2010 Prism Coupler) is carried out to confirm the formation of waveguide and characterize the optical waveguiding performance. According to waveguide theory, the confined light in the waveguide can travel only at certain angles, called ‘mode angles’ (θm, m=0, 1, 2….), where m is the mode order. In any particular waveguide structure, the number of possible modes will depend on the thickness, the refractive index of the waveguide and the wavelength of employed light. For a given employed wavelength and polarization, when two or more modes are excited by prism, the refractive indices and thickness of a waveguide can be determined from the TE or TM mode eigenvalue equations (2) and (3), which are given as follows [14]:

kzd=mπ+tan1[nm2n02n12nm2]12+tan1[nm2n22n12nm2]12
kzd=mπ+tan1[n12n02][nm2n02n12nm2]12+tan1[n12n22][nm2n22n12nm2]12

where kz is given by the following equation:

kz=k(n12nm2)12

k is the wave vector; d the thickness of guiding region; m the mode order; n0, n1 and n2 are the refractive indices of air(surface), waveguide and substrate; respectively. In the two equations above, nm is the mode effective index, which depends on the mode angle and can be determined by using the following equation [15]:

nm=n1sinθm=npsinθm

where θm is the ‘mode angle’ at which the m-mode is excited in guiding film, ′θm the incident angle at the interface between the film and the prism and np is the refractive index of the rutile prism. From Eqs. (2), (4) and (5), the ordinary refractive index (no) and thickness of a film can be solved when two or more TE modes are excited. Similarly, the extraordinary refractive index (ne) and thickness can be determined from two or more excited TM modes using Eqs. (3), (4) and (5). The accuracy of determination of refractive indices and thickness of a film is within 0.01% and 0.5% by using prism coupling method, respectively.

By using laser beams at 0.633 µm and 1.539 µm, both TE and TM guiding modes are measured. The corresponding refractive indices (no and ne) and thicknesses of both guiding films are deduced according to the former analysis. In general, in the measured TE or TM modes patterns, the guiding modes can be distinguished from the so-called ‘substrate mode’ due to their higher mode effective refractive indices and relatively sharp peaks. When the guiding modes are excited by laser beam at 0.633 µm, six TE or TM modes are detected in both guiding films. Since waveguide devices working at 1.54 µm are most preferred in optical telecommunication, we investigate the guiding properties of both annealed films S1 and S2. Figure 3 shows the TE and TM modes measured at wavelength 1.539 µm.

Figure 3(a) and 3(c) show the measured modes plots with TE polarization in both films, and Fig. 3(b) and 3(d) in TM case. As we can see from Fig.e 3, only the first three modes are demonstrated to be the guiding modes because of the higher mode effective refractive indices than those of MgZnO buffers. It is also noted that the propagation light with TM polarization has a better optical confinement than that with TE polarization. By analyzing experimental results, it can be concluded that MgZnO buffer layers play an important role in characterizing guiding region in ZnO/MgZnO films. The fine ZnO films can only be obtained with the MgZnO buffer layers that have proper concentration of Mg in the structure. Based on the measurements of excited TM and TE modes, the extraordinary refractive index (ne) and ordinary index (no) in waveguide are obtained for both films. The experimental result shows that the difference in refractive index between ZnO and MgZnO films is at about 3%. The values of refractive indices and thicknesses of films S1 and S2 measured at 1.539 µm are given in Table 1.

 figure: Fig. 3.

Fig. 3. Mode plots vs. effective refractive indices at λ=1.539µm for annealed films S1 with (a) TE polarization, (b)TM polarization; and for annealed film S2 with (c) TE polarization, (d) TM polarization, respectively.

Download Full Size | PDF

Tables Icon

Table 1. Experimental values for samples S1 and S2, all the data are measured at 1.539µm.

As can be seen from Table 1, both ordinary and extraordinary refractive indices show the similar tendency to the annealing treatment, i.e., both refractive indices decrease after annealing. In the previous work reported, two factors, namely packing density and the lattice constant (c-axis constant) are responsible for high temperature-induced changes of refractive indices (Ref. [8]). The possible explanation is that the packing density increases as a result of the further particle conglomeration during annealing treatment at proper temperature and the increased density results in the enhancement of refractive indices [16]. For the case of our as-grown film, the refractive indices are increased because the c-axis lattice constant is elongated in comparison with the bulk value(c=5.2040 Å), then it decreases due to the contraction of c-axis lattice constant in the process of annealing treatment [17]. From Table 1 it can be found that the effect of annealing on refractive indices is dominated by lattice contraction. Another fact from Table 1 is that the crystalline birefringence is enhanced after annealing treatment. The larger birefringence implies the better crystalline nature, which means that film S2 will show an improved nonlinearity in contrast to film S1.

Propagation loss is an important parameter in characterizing the quality of waveguide structure. The fiber scanning technique is used to measure the loss of waveguides at λ=0.633 µm. For as-deposited films, no distinct propagation line can be observed from the top surface of both films after TE and TM modes are excited. This phenomenon is considered as the result of strong scattering and absorption loss, which limits the propagation of light in waveguide. After annealing at 400°C for 60 minutes, the loss measurements are carried out for both films by moving the fiber detector along the propagation line and recording the intensity of light. For the excited TM0 in film S1, the loss is measured to be 7.21dB/cm. A more uniform propagation line is observed in film S2 when TM0 mode is excited, the loss is measured to be 3.64 dB/cm. The lower loss of S2 means the relative better confinement of light in film S2 than that in S1. The loss measurement results are shown in Fig. 4(a) for annealed film S1 and (b) for annealed film S2, respectively. The high waveguide loss may be due to the still poor film quality and the non-optimized film growth condition, e.g., growth temperature, air pressure, thickness and annealing condition etc. The further research will be carried on to improve it.

 figure: Fig. 4.

Fig. 4. Loss measurements of annealed films (a) S1 and (b) S2 at λ=0.633µm with the excited TM0 modes.

Download Full Size | PDF

4. Summary

In summary, highly transparent and c-axis orientated ZnO films on MgZnO buffer layers have been fabricated by radio-frequency magnetron sputtering on x-cut and z-cut LN substrates, which provides an alternative method for fabricating high quality optical waveguide ZnO films. The properties of the films are characterized by XRD θ-2θ scans, spectrophotometer and prism coupling technique. The results of mode plots and loss measurement show that high optical quality ZnO/MgZnO films deposited on z-cut LN substrates are better than that on x-cut LN substrate. Both films S1 and S2 can act as waveguide structures and the waveguiding performances are demonstrated significantly to depend upon post-deposited annealing treatment.

Acknowledgments

This work was supported by the National Natural Science Foundation of China under Grant No.10375037.

References and links

1. Y. F. Chen, D. M. Bagnall, H. Ko. K. Park, K. Hiraga, Z. Zhu, and T. Yao, “Plasma assited molecular beam epitaxy of ZnO on c-plane sapphire: Growth and characterization,” J. Appl. Phys. 84, 3912–3918 (1998). [CrossRef]  

2. R. Ondo-Ndong, F. Pascal-Delannoy, A. Boyer, A. Giani, and A. Foucaran, “Structural properties of zinc oxide thin films prepared by r. f. magnetron sputtering,” Mat. Sci. Eng. B 97, 68–73 (2003). [CrossRef]  

3. K. Matsubara, P. Fons, A. Yamada, M. Watanabe, and S. Niki, “Epitaxial growth of ZnO thin films on LiNbO3 substrates,” Thin Solid Films 347, 238–240 (1998). [CrossRef]  

4. H. Kato, K. Miyamoto, M. Sano, and T. Yao, “Polarity control of Zno sapphire by the MgO buffer layer thickness,” Appl. Phys. Lett. 84, 4562–4564 (2004). [CrossRef]  

5. W. S. Hu, Z. G. Liu, X. L. Guo, C. Lin, S. N. Zhu, and D. Feng, “Preparation of c-axis oriented ZnO optical waveguiding films on fused silica by pulsed laser reactive ablation,” Mater. Lett. , 25, 5–8 (1995). [CrossRef]  

6. N. Mails, J. P. Reithmaier, A. Forchel, M. Kohls, L. Spanhel, and G. Müller, “Er doped nanocrystalline ZnO planar waveguide structures for 1.55µm amplifier applications,” Appl. Phys. Lett. 75, 2005–2007 (1999). [CrossRef]  

7. S. F. Yu, C. Yuen, S. P. Lau, Y. G. Wang, H. W. Lee, and B. K. Tay, “Ultraviolet amplified spontaneous emission from zinc oxide ridge waveguides,” Appl. Phys. Lett. 83, 4288–4290 (2003). [CrossRef]  

8. N. Mehan, V. Gupta, K. Sreenivas, and A. Mansingh, “Effect of annealing on refractive indices of radio-frequency magnetron sputtered waveguiding zinc oxide films on glass,” J. Appl. Phys. 96, 3134–3139 (2004). [CrossRef]  

9. R. D. Shannon, “Revised effective ionic radii and systematic studies of interatomic distances in halides and chalcogenides,” Acta Crystallogr. A 32, 751–767 (1976). [CrossRef]  

10. C. W. Teng, J. F. Muth, Ü. Özgür, M. J. Bergmann, H. O. Everitt, A. K. Sharma, C. Jin, and J. Narayan, “Refractive indices and absorption coefficients of MgxZn1-xO alloys,” Appl. Phys. Lett. 76, 979–981 (1999). [CrossRef]  

11. F. Reizman, “Optical Thickness Measurement of Thin Transparent Films on Silicon,” J. Appl. Phys. 36, 3804–3807 (1965). [CrossRef]  

12. H. Yamamoto, N. Saiga, and K. Nishimori, “ZnO thin films deposited on various LiNbO3 substrates by RF-sputtering,” Appl. Surf. Sci. 169170, 517–520 (2001). [CrossRef]  

13. L.V. Azaroff, Elements of X-ray Crystallography (McGraw-Hill, New York, 1968)

14. P. D. Townsend, P. J. Chandler, and L. Zhang, Optical Effects of Ion Implantation (Cambridge University Press, Cambridge, UK, 1994). [CrossRef]  

15. A. Ghatak and K. Thyagarajan, Integrated Optics in Optical Electronics (Cambridge University Press, New Delhi, 1994).

16. V. Gupta and A. Mansingh, “Influence of postdeposition annealing on the structural and optical properties of sputtered zinc oxide film,” J. Appl. Phys. 80, 1063–1073 (1996). [CrossRef]  

17. A. Dietrich, K. Schmalzbauer, and H. Hoffman, “The effect of annealing on the optical properties of indium tin oxide films,” Thin Solid Films 122, 19–29 (1984). [CrossRef]  

Cited By

Optica participates in Crossref's Cited-By Linking service. Citing articles from Optica Publishing Group journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (4)

Fig. 1.
Fig. 1. Optical transmittance diagram of ZnO film (solid line) and Mg0.16Zn0.84O film (dotted line) grown on z-cut LiNbO3 substrates by radio-frequency magnetron sputtering under a pressure of 1.0 Pa at 200 W RF-power at 350 K.
Fig. 2.
Fig. 2. X-ray diffraction patterns of ZnO/MgZnO films deposited on (a) x-cut LN and (b) z-cut LN. The inlet figures exhibit the x-ray diffraction patterns of MgZnO films.
Fig. 3.
Fig. 3. Mode plots vs. effective refractive indices at λ=1.539µm for annealed films S1 with (a) TE polarization, (b)TM polarization; and for annealed film S2 with (c) TE polarization, (d) TM polarization, respectively.
Fig. 4.
Fig. 4. Loss measurements of annealed films (a) S1 and (b) S2 at λ=0.633µm with the excited TM0 modes.

Tables (1)

Tables Icon

Table 1. Experimental values for samples S1 and S2, all the data are measured at 1.539µm.

Equations (5)

Equations on this page are rendered with MathJax. Learn more.

d = 0.9 λ B cos θ B
k z d = m π + tan 1 [ n m 2 n 0 2 n 1 2 n m 2 ] 1 2 + tan 1 [ n m 2 n 2 2 n 1 2 n m 2 ] 1 2
k z d = m π + tan 1 [ n 1 2 n 0 2 ] [ n m 2 n 0 2 n 1 2 n m 2 ] 1 2 + tan 1 [ n 1 2 n 2 2 ] [ n m 2 n 2 2 n 1 2 n m 2 ] 1 2
k z = k ( n 1 2 n m 2 ) 1 2
n m = n 1 sin θ m = n p sin θ m
Select as filters


Select Topics Cancel
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.