Open Access
Add to BibSonomyAbstract
Novel suspended SiO2 ridge optical waveguides on silicon are fabricated and characterized. The present suspended SiO2 ridge optical waveguide has a SiO2 ridge core surrounded by air. The propagation loss and the bend loss measured are about 0.385dB/cm and 0.037dB/90° respectively for the fabricated 1μm-wide waveguides with a bending radius of 100μm when operating at the wavelength of 1550 nm. With the present suspended SiO2 optical waveguides, a small racetrack resonator with a radius of 100μm is also demonstrated and the measured Q-factor is about 3160.
©2012 Optical Society of America
1. Introduction
The demand of photonic integrated circuits (PICs) keeps increasing for optical communication, optical interconnects as well as optical sensing. In order to satisfy the demands, various material systems and optical waveguide types have been developed, like LiNbO3 [1], silica [2–5], silicon-on-insulator [6–8], III-V semiconductor [9, 10], and polymers [11].
Among them, silicon-on-insulator (SOI) provides a good platform to have ultra-dense PICs by utilizing SOI nanowires which have a submicron cross section and ultra-high index-contrast [6, 12]. Thus, silicon photonics has been developed rapidly in the past years for optical interconnects [13], as well as optical sensing [14, 15]. However, because silicon is not transparent in the wavelength range of less than 1.1μm [4], SOI nanowires do not work in the visible range, which is very important for some applications like optical.
In contrast, silica is one of the most attractive materials for passive PICs because of its mature fabrication process, as well as low propagation loss. One of the most popular silica optical waveguide is the SiO2-on-Si buried rectangular waveguides which has a low index-contrast (∆~0.75%) and a Ge-doped core as large as size 6 × 6μm2 so that it has high coupling efficiency with a single mode fiber [3]. However, for a conventional SiO2-on-Si buried rectangular waveguides, the bending radius is as large as some millimeters even centimeters [5], due to its low index-contrast (∆). This is not good for the future dense PICs. Besides, the buried SiO2-on-Si waveguide is also not a good option for optical sensing because the cladding prevents the medium to contact the evanescent field.
A deeply-etched SiO2 optical waveguide is a potential choice for solving these issues because it has an air cladding so that it enables sharp bending as well as optical sensing with relatively high-sensitivity [4, 5, 16]. However, it requires very deep etching (~16μm or more), which makes it not easy to achieve a low-loss light waveguiding because the deeply-etched sidewall is relatively rough.
In the present paper, we propose a novel suspended SiO2 optical waveguide on silicon substrate so that it could work in a broad wavelength band ranging from the visible light to the infrared light. Suspended optical waveguides have been demonstrated before with the materials of silicon [17], III-V semiconductor [10, 18], etc. Ultra-high Q suspended micro-disks on silica has been also demonstrated [19]. Kei Watanabe, et al realizes ultralow power consumption silica-based PLC-VOA/switches using suspended narrow ridge structure [20]. Our proposed suspended SiO2 optical waveguide has a small SiO2 ridge region surrounded by air, which introduces a relatively high index-contrast ∆ and consequently enables the bending radius as small as 100μm (or smaller). The surrounding air region also makes it very attractive for optical sensing. Furthermore, for the present suspended SiO2 optical waveguides, only a layer of SiO2 thin film whose thickness is around 1μm is needed to be formed on a silicon substrate. This makes the fabrication very simple and cheap potentially because a simple and short-time thermal oxidation process is enough for forming the SiO2 thin film and a shallowly etching is needed only. In contrast, for the conventional SiO2-on-Si buried rectangular waveguide and the structure in [20], which has ~40μm-thick SiO2 films (including an under-cladding, a core region as well as an upper-cladding), one needs some expensive technologies. For example, in order to form the thick SiO2 film, one usually needs the plasma-enhanced chemical vapor deposition (PECVD) or flame hydrogen deposition (FHD) technologies, and Ge-doping is required for the core layer with a higher index than the cladding layer. Furthermore, the deep etching technology is also needed. In this paper, we have designed and fabricated the proposed suspended SiO2 optical waveguides as well as race-track resonators as an example.
2. Structure and fabrication of the suspended SiO2 optical waveguide
Figure 1(a) shows the cross section of the proposed suspended SiO2 ridge optical waveguide, which has an air-cladded SiO2 ridge core on a silicon substrate. The air under-cladding is formed by removing the silicon beneath partially by using a second ICP dry etching process with the gases SF6, O2, CHF3 after the windows in the slab layer are open, as shown in Fig. 1(b). The slab layer is a very important part not only for forming a ridge waveguide but also supporting the suspended waveguides. It can be seen that silicon substrate is not removed in the region far away from the ridge so that the slab could be supported by the Si substrate. Therefore, the slab thickness should be thick enough to have good mechanical strength. In our design, we choose the slab thickness h to be around 300nm. The height hr of the air under-cladding is determined by the time of dry etching process. From Fig. 1(b), it can be seen that there are a row of windows at each side of the ridge. The distance dwin between the window edge and the ridge edge should large enough so that the light propagation along the ridge waveguide does not be influenced by the windows. According to the mode profile shown in Fig. 1(c), which gives the calculated TE-polarization mode profiles for a 1μm-wide straight waveguide, one sees that the distance dwin should be larger than e.g., 4μm. In our case, we choose dwin = 7μm as an example (see Fig. 1(b)). Here the waveguide parameters are chosen as follows: the SiO2 refractive index nSiO2 = 1.444, the ridge width wco = 1μm, the ridge height hr = 660nm, the slab height h = 310nm. Figure 1(d) shows the pure bending loss and the transition loss as the radius varies. One can see that the bending loss and transition loss is very small when the radius is 100μm.
Fig. 1 (a) The cross section of the proposed suspended SiO2 optical waveguide; (b) the SEM top-view picture; (c) the TE mode profile for a straight waveguide; (d) the calculated bending loss and transition loss. The width wco = 1μm, the height hr = 660nm, the slab height h = 310nm.
Figure 2 shows the fabrication process for the present SOW. First, a thin SiO2 film is formed on a <100> silicon substrate. It is well known that there are many ways to form a SiO2 film, e.g., using the plasma-enhanced chemical vapor deposition (PECVD) technology [2–4, 21], the flame hydrolysis deposition (FHD) technology [22], as well as the thermal oxidation [3]. The former two are very popular for forming very thick SiO2 film, however, which usually is porous (not dense). Consequently the mechanical strength is not strong enough for our present suspended waveguide. In contrast, the thermal oxidation method is pretty good and low-cost choice to form dense SiO2 thin film with very good mechanical robustness. The drawback is that the SiO2 thickness formed by the dry thermal oxidation is usually no more than 1~2μm and the rate is very slow. It takes 8h to grow about 400nm at normal pressure and 1050°C with dry thermal oxidation. For the present case, the SiO2 film is as thin as only hundreds of nanometers. Therefore, we choose wet thermal oxidation process to form SiO2 thin film on a silicon substrate. In our fabrication, a ~970nm-thick SiO2 film is formed through two-hour wet thermal oxidation (i.e., the O2 heated in 98°C hot water) at normal pressure and the temperature is around 1100°C. Then electron-beam negative electron-resist (ma-N 2405) is spin-coated on the ~970nm-thick SiO2 thin film. The thickness of the photo-resist film is about 470nm at the spinning speed of 3000rpm (See Fig. 2(a)). The sample with the electron-beam negative electron-resist is then baked (90 second @ 90°C) to evaporate the solvent. Then the pattern is generated by using the Raith150-two ultra-high resolution E-beam writer. After the development (See Fig. 2(b)), a dry-etching with the STS ICP (Inductive Coupled Plasma) etching system is proceeded to etch the SiO2 thin film (See Fig. 2(c)). Fluorine-based gas such as CF4 and CHF3 is used for this dry etching process. Then a SiO2 rib optical waveguide is formed on silicon substrate. In order to remove Si beneath the SiO2 optical waveguide, we open some rectangular windows about 5 × 7μm2 at both sides of the SiO2 ridge (the distance is about 7μm) with a second lithography (Fig. 2(d)) and dry-etching process. Si beneath the SiO2 optical waveguide is then removed by using ICP dry etching with the gases such as SF6, O2 and CHF3 through the open windows, as shown Fig. 2(e). In order to remove the residual Si, we soaked the sample in the TMAH (with a concentration of 10 wt%) at the temperature of 85°C for about 10min. Finally, we soak the Si wafer in H2SO4 and H2O2 mixed solution to remove the residual photoresist and fluorocarbon formed during the etching process.
Figure 3(a)-(d) show the SEM (scanning electron microscope) pictures for the fabricated suspended waveguides and structures. Figure 3(a) shows the coupling region consisting of two paralleled suspend optical waveguides for a racetrack resonator. Figure 3(b)-(c) shows the sidewall and the cross section of the fabricated suspended optical waveguide, respectively. From Fig. 3(b), it can be seen that the waveguide has quite smooth sidewalls, which is beneficial to obtain a low propagation loss. From cross section of the fabricated suspended optical waveguide shown in Fig. 3(c), it can be seen clearly that the silicon substrate beneath the waveguide ridge is removed. The height of the air region under the silica ridge is about 20μm, which could be varied by controlling the time of ICP etching with SF6. It can also be seen that there is significant buckling in the SiO2 after it is released from the from the silicon, which is due to the high temperature during oxidation around 1100°C [23], as shown in Fig. 3(c). The estimated stress is about 1.2 × 109 dyn/cm2 according to the result given in [23]. This caused a large coupling loss partially due to the numerical aperture mismatch when light is coupled between a fiber and the waveguides, which will be seen from the measured loss given below.
Fig. 3 SEM pictures of the structures. (a) the top view for the coupling region of a racetrack resonator. The waveguide width w = 1μm, and the gap width wgap = 1μm; (b) a straight waveguide; (c) the cross section of a fabricated waveguide; (d) the enlarged view for the cross section.
3. Characterization of the fabricated suspended SiO2 optical waveguides and devices
A bent waveguide is one of the basic elements for various PICs. It is desirable to achieve a small bending radius with low bend loss especially for dense PICs. For the present suspended SiO2 optical waveguide, small bending radius is expected because of its relatively high index-contrast. In order to measure the propagation loss as well as the bending loss of the suspended SiO2 optical waveguides, we have designed a series of spiral structures with different lengths. All the bends in the spiral structure has the same radius of 100μm, as shown by the inset in Fig. 4 . A tunable laser is used for the measurement and a lens fiber is used to couple the light into the suspended optical waveguide. Since the measured total loss Ltot is not sensitive to the wavelength in the C-band, we only show the measured total losses Ltot at 1550nm for these spirals in Fig. 4. The total loss for a spiral waveguide is given by Ltot = Ls l + N LB + 2Lc, where Ls is the propagation loss per unit length, l is the total length of the spiral, N is the 90°-bend number, LB is the bend loss, and Lc is the coupling loss per facet between a fiber and the suspended waveguide. In order to extract the loss values, we assume that the bending section has the same propagation loss LS over a unit length (dB/cm) as the straight waveguide while the bending section has an addition pure bending loss (i.e., LB). Such an assumption is reasonable for the present case because the bending radius is very large so that the mode field profile in the bending section is very similar with that in the straight section. With such an assumption, we extracted the value for the losses LS, LB, and Lc by fitting the measured data shown in Fig. 4 with the least square method. Finally we obtain Ls = 0.38dB/cm, LB = 0.037dB/90° and Lc = 17.6dB/facet. The large coupling loss comes from the following sources. The first source is from the mode mismatch loss between the lens fiber and the suspended waveguide. For the present case, the lens fiber and the suspended waveguide have diameters of about 3~4μm and 1μm, respectively. Consequently the estimated mode mismatching loss is about 10dB. The second source for the coupling loss is the loss due to the facet reflection, which is about 0.236dB. The third source is the waveguide bending at the facet due to the thermal strain, as shown in Fig. 3(b).
Fig. 4 The transmission for the suspended waveguides with different lengths. Inset is the micrographic photograph of a spiraled waveguide for measuring propagation loss and bend loss.
Regarding that a ring resonator is an essential versatile block for various functionalities like optical modulators [7], optical logic and switch [8, 9], optical sensors [14, 15], wavelength filters [24], power splitters [25], lasers [26], and so on, we design and fabricate a race-track resonator with our proposed suspended SiO2 optical waveguide, as shown in the inset of Fig. 5(b) . The length for the coupling region is chosen as Ldc = 200μm, the gap width is wgap = 1μm, and the bending radius R = 100μm. Figure 5 shows the measured spectral responses at the through port of the fabricated racetrack resonators. The insert shows the SEM picture for the fabricated racetrack resonator. There are many small square windows locating at both sides of the waveguide (inside as well as outside of the racetrack) for partially removing the silicon substrate beneath. From this figure, it can be seen that the FSR is about 1.5nm, which is very close to the theoretical prediction. The extinction ratio is about 20dB, and the 3dB bandwidth is about 0.48nm, which corresponds to a loaded Q factor of about 3200. This relatively low loaded Q-value is mainly due to the large coupling ratio between the resonator and the access waveguides. It is possible to enhance the Q-factor by improving the design and fabrication of the coupling region.
Fig. 5 The measured spectral response at the through port of the fabricated racetrack resonator. Inset is the SEM picture of the fabricated race-track resonator.
4. Conclusion
In this paper, we have designed and fabricated novel suspended optical ridge waveguides on silica. The measured propagation loss and bending loss are about Ls = 0.38dB/cm, and LB = 0.037dB/90°, respectively. A racetrack resonator has been also fabricated by using the proposed novel suspended optical waveguide. It has shown that the fabricated racetrack ring resonance has shown that the present suspended waveguide can be used to realize compact PICs in comparison with those traditional buried silica waveguides and the performance can be improved with the optimized fabrication process and waveguide parameters.
Acknowledgments
This project was partially supported by the National Nature Science Foundation of China (No. 61077040), a 863 project (Ministry of Science and Technology of China, No. 2011AA010301), Zhejiang provincial grant (Z201121938, No. 2011 C11024) of China, and also supported by the Fundamental Research Funds for the Central Universities.
References and links
1. A. Majkic, M. Koechlin, G. Poberaj, and P. Günter, “Optical microring resonators in fluorineimplanted lithium niobate,” Opt. Express 16(12), 8769–8779 (2008). [CrossRef] [PubMed]
2. T. Miya, “Silica-based planar lightwave circuits: passive and thermally active devices,” IEEE J. Sel. Top. Quantum Electron. 6(1), 38–45 (2000). [CrossRef]
3. M. R. Poulsen, P. I. Borel, J. Fage-Pedersen, J. Hubner, M. Kristensen, J. H. Povlsen, K. Rottwitt, M. Svalgaard, and W. Svendsen, “Advances in silica-based integrated optics,” Opt. Eng. 42(10), 2821–2834 (2003). [CrossRef]
4. Z. Sheng, B. Yang, L. Yang, J. Hu, D. Dai, and S. He, “Experimental demonstration of deeply-etched SiO2 ridge optical waveguides and devices,” IEEE J. Quantum Electron. 46(1), 28–34 (2010). [CrossRef]
5. D. Dai and Y. Shi, “Deeply etched SiO2 ridge waveguide for sharp bends,” J. Lightwave Technol. 24(12), 5019–5024 (2006). [CrossRef]
6. W. Bogaerts, D. Taillaert, B. Luyssaert, P. Dumon, J. Van Campenhout, P. Bienstman, D. Van Thourhout, R. Baets, V. Wiaux, and S. Beckx, “Basic structures for photonic integrated circuits in Silicon-on-insulator,” Opt. Express 12(8), 1583–1591 (2004). [CrossRef] [PubMed]
7. Q. Xu, B. Schmidt, S. Pradhan, and M. Lipson, “Micrometre-scale silicon electro-optic modulator,” Nature 435(7040), 325–327 (2005). [CrossRef] [PubMed]
8. Q. Xu and M. Lipson, “All-optical logic based on silicon micro-ring resonators,” Opt. Express 15(3), 924–929 (2007). [CrossRef] [PubMed]
9. T. Ibrahim, W. Cao, Y. Kim, J. Li, J. Goldhar, P.-T. Ho, and C. Lee, “All-optical switching in a laterally coupled microring resonator by carrier injection,” IEEE Photon. Technol. Lett. 15(1), 36–38 (2003). [CrossRef]
10. T. H. Stievater, W. S. Rabinovich, D. Park, J. B. Khurgin, S. Kanakaraju, and C. J. K. Richardson, “Low-loss suspended quantum well waveguides,” Opt. Express 16(4), 2621–2627 (2008). [CrossRef] [PubMed]
11. B. Yang, L. Yang, R. Hu, Z. Sheng, D. Dai, Q. Liu, and S. He, “Fabrication and characterization of small optical ridge waveguides based on SU-8 polymer,” J. Lightwave Technol. 27(18), 4091–4096 (2009). [CrossRef]
12. T. Tsuchizawa, K. Yamada, H. Fukuda, T. Watanabe, J. Ichi Takahashi, M. Takahashi, T. Shoji, E. Tamechika, S. Itabashi, and H. Morita, “Microphotonics devices based on silicon microfabrication technology,” IEEE J. Sel. Top. Quantum Electron. 11(1), 232–240 (2005).
13. N. Sherwood-Droz, H. Wang, L. Chen, B. G. Lee, A. Biberman, K. Bergman, and M. Lipson, “Optical 4x4 hitless silicon router for optical networks-on-chip (NoC),” Opt. Express 16(20), 15915–15922 (2008). [CrossRef] [PubMed]
14. D. X. Xu, A. Densmore, A. Delâge, P. Waldron, R. McKinnon, S. Janz, J. Lapointe, G. Lopinski, T. Mischki, E. Post, P. Cheben, and J. H. Schmid, “Folded cavity SOI microring sensors for highsensitivity and real time measurement of biomolecular binding,” Opt. Express 16(19), 15137–15148 (2008). [CrossRef] [PubMed]
15. D.-X. Xu, M. Vachon, A. Densmore, R. Ma, A. Delâge, S. Janz, J. Lapointe, Y. Li, G. Lopinski, D. Zhang, Q. Y. Liu, P. Cheben, and J. H. Schmid, “Label-free biosensor array based on silicon-on-insulator ring resonators addressed using a WDM approach,” Opt. Lett. 35(16), 2771–2773 (2010). [CrossRef] [PubMed]
16. M. Popovic, K. Wada, S. Akiyama, H. A. Haus, and J. Michel, “Air trenches for sharp silica waveguide bends,” J. Lightwave Technol. 20(9), 1762–1772 (2002). [CrossRef]
17. L. Martinez and M. Lipson, “High confinement suspended micro-ring resonators in silicon-on-insulator,” Opt. Express 14(13), 6259–6263 (2006). [CrossRef] [PubMed]
18. M. Garrigues, J. Danglot, J.-L. Leclercq, and O. Parillaud, “Tunable high-finesse InP/air MOEMs filter,” IEEE Photon. Technol. Lett. 17(7), 1471–1473 (2005). [CrossRef]
19. D. K. Armani, T. J. Kippenberg, S. M. Spillane, and K. J. Vahala, “Ultra-high-Q toroid microcavity on a chip,” Nature 421(6926), 925–928 (2003). [CrossRef] [PubMed]
20. K. Watanabe, Y. Hashizume, Y. Nasu, M. Kohtoku, M. Itoh, and Y. Inoue, “Ultralow power consumption silica-based PLC-VOA/switches,” J. Lightwave Technol. 26(14), 2235–2244 (2008). [CrossRef]
21. L. Martinu and D. Poitras, “Plasma deposition of optical films and coatings: A review,” J. Vac. Sci. Technol. 18(6), 2619–2645 (2000). [CrossRef]
22. A. Kilian, J. Kirchof, B. Kuhlow, G. Przyrembel, and W. Wischmann, “Birefringence free planar optical waveguide made by flame hydrolysis deposition (FHD) through tailoring of the overcladding,” J. Lightwave Technol. 18(2), 193–198 (2000). [CrossRef]
23. E. Kobeda and E. A. Irene, “Intrinsic SiO2 film stress measurements on thermally oxidized Si,” J. Vac. Sci. Technol. B 5(1), 15–19 (1987). [CrossRef]
24. S. Chu, B. Little, W. Pan, T. Kaneko, S. Sato, and Y. Kokubun, “An eight-channel add-drop filter using vertically coupled microring resonators over a cross grid,” IEEE Photon. Technol. Lett. 11(6), 691–693 (1999). [CrossRef]
25. D. Dai and S. He, “Proposal of a coupled-microring-based wavelength-selective 1×N power splitter,” IEEE Photonic. Tech. L. 21(21), 1630–1632 (2009). [CrossRef]
26. H. Rong, A. Liu, R. Jones, O. Cohen, D. Hak, R. Nicolaescu, A. Fang, and M. Paniccia, “An all-silicon Raman laser,” Nature 433(7023), 292–294 (2005). [CrossRef] [PubMed]
