Open Access
Add to BibSonomyAbstract
Multiple light spots can be generated by modulating the spatial phase distribution of laser beam with a spatial light modulator (SLM). In this paper, we demonstrate the fabrication of three-dimensional 1 × 4 splitter waveguides inside a glass by focusing multiple light spots of femtosecond (fs) laser pulses, which can be controlled by switching spatial phase distributions on an SLM. In the conventional fs laser writing technique, a highly precise positioning of a substrate is essential for fabricating a branched waveguide in a splitter. Using the technique proposed in this paper, a continuously branched waveguide can be produced easily by translating a glass substrate only one time; therefore this technique can eliminate the need for a high precision in positioning of a substrate and save a fabrication time.
©2010 Optical Society of America
1. Introduction
The Y-junction coupler in optical waveguides is one of the essential components in optical networks and small photonic devices [1,2]. For example, it is used as a splitter in passive optical networks that distributes optical signals to subscribers [3], and it is also used for splitting and coupling optical signals in a Mach-Zehnder interferometer [4,5]. A number of techniques have been developed for fabricating waveguides, for example, ion-exchange [6], lithography [7], and laser direct writing [8,9]. Among these techniques, femtosecond laser direct writing is a promising technique for fabricating three-dimensional (3D) optical waveguides inside transparent materials, and the fabrications of Y-splitter have been demonstrated with the technique [10–13]. However, the problem in the fabrication of 3-D Y splitter with a conventional fs-laser direct writing is that it requires very high accuracy in the 3D positioning of the substrate. This problem can be solved by parallel writing of multiple waveguides with a spatial light modulator (SLM). With an SLM, multiple beam spots can be created in 3 D by modulating the spatial phase distribution of a fs laser beam [14]. Recently, fabrications of a directional coupler and bent waveguides in a glass with an SLM have been reported [15–18]. However, these structures are restricted to two dimensions, and 3 D couplers had not yet been fabricated with an SLM. In this study, we succeeded in fabricating three dimensional 1 × 4 splitters inside silica glass with a fs laser and SLM. The notable points of the fabrication technique demonstrated in this study are (i) various kinds of 1 × N splitters can be fabricated and (ii) a waveguide structure is continuous in the branching region.
2. Method
The principle of parallel laser writing of multiple waveguides with an SLM is shown in Fig. 1 . When a fs laser beam is reflected on an SLM, the spatial phase distribution of the laser beam is modulated. The spatial phase modulation given by an SLM is called a computer generated hologram (CGH), which is calculated by a computer. The laser beam reflected on the SLM is focused inside a glass substrate with an objective lens. Multiple focused beam spots can be created after focusing, depending on the spatial phase modulation. At each beam focus, the glass is photoexcited through nonlinear light absorption, which leads to a local refractive index modulation. Multiple refractive index lines can be written by translating the glass substrate with respect to the beam foci. By selecting the optimum laser writing conditions, it is possible to achieve refractive index change at the photoexcited region inside the glass, and the multiple refractive index lines work as optical waveguides [9].
A variety of waveguide structures such as bent or branched waveguides can be written by switching from one CGH to another during the substrate is being translated because the number and positions of focused beam spots can be controlled by selecting a CGH. The fabrication of a Y-splitter [or branched waveguide; shown in Fig. 2(a) ] requires that one waveguide structure gradually be divided into two during the translating a substrate. The waveguide division can be realized by gradually separating two focused beam spots, as shown in Fig. 2(b). However, the interference between two adjacent beam spots makes it difficult to write a branched structure. For example, the simulated light intensity distribution of two adjacent beam spots along the beam propagation direction is shown in Fig. 2(c). It shows clearly that the interference could distract the light intensity distribution that is necessary for writing waveguides. To solve this problem, we wrote branched waveguides with multiple beam spots separated in the translating direction. This method is represented by Fig. 2(d). A branched structure can be written by changing the transversal separation between two beam spots. By this method, a waveguide structure can be created without faults, because the separation between beam spots is large enough for the beams not to interfere with each other.
Fig. 2 (a) Structure of a Y splitter. (b) Separation of two adjacent beam spots. (c) Simulated light intensity distribution along the beam propagation direction in the case of (b). (d) Variation of two spatially separated beam spots in writing a splitter in this study.
3. Experimental setup
The experimental setup for writing waveguides is shown in Fig. 3 . Amplified Ti: Al2O3 fs laser pulses (Mira-Legend, Coherent Inc.) with a pulse duration of about 120 fs (FWHM), central wavelength of 800 nm and repetition rate of 1 kHz were used for writing waveguides. The laser pulses were reflected on a liquid crystal on silicon (LCOS-SLM, X10468-02, Hamamatsu Photonics K.K.), which had 800 × 600 square pixels with a pixel pitch of 20 μm and that can provide a reflected beam with an independent phase change up to 2π at each pixel [19]. The laser pulses after reflecting on an SLM passed through a telescope, which consisted of two concave lenses (magnification of M = 0.3), and were focused inside a silica glass substrate with an objective lens (f = 10 mm, NA = 0.45; LU-Plan, Nikon). The glass substrate was placed on a 3D translation stage. The spatially phase modulated beam was focused at multiple positions inside the glass. Multiple waveguide structures were written by translating the silica glass with the translation stage. Writing of the waveguides was observed with a CCD camera and back illumination of the glass with a white LED lamp. The pulse energy of the laser beam was controlled with a neutral density (ND) filter, and the exposure time was controlled with a mechanical shutter. The translation stage, LCOS-SLM and mechanical shutter were controlled by a personal computer.
Fig. 3 Experimental setup for laser writing of optical waveguides with an SLM. M1, M2: dielectric mirrors; L1: a lens of f = 500 mm; L2: a lens of f = 150 mm; DM: a dichroic mirror which refracts light of 750-850 nm; OL: an objective lens; MP: a metal plate.
CGHs were calculated by the optimal rotation angle (ORA) method [20]. The amplitude of the electromagnetic field at the desired position (X,Y,Z) was calculated by
where (xm, yn) is the position on an SLM, F(xm, yn) and ΔΦ(xm, yn) are the amplitude and phase of the laser electromagnetic field at (xm, yn), respectively, λ is the wavelength of the laser beam, and f is the focal length of a lens [21]. In the ORA method, {ΔΦ(xm, yn)} are altered to increase A(X,Y,Z) at the desired positions, and the alteration of {ΔΦ(xm, yn)} is iterated until A(X,Y,Z) at all the desired positions become uniform within an accuracy of less than 0.5%. The detail of the algorithm has been described in ref 20. A CGH with 512 × 512 pixels, which was calculated by the ORA method, was displayed in the central region of the active area of the LCOS-SLM. It took about 20 seconds to obtain a CGH that produced four focused points with a commercial personal computer.It is important to note the damage threshold of an SLM, because it could limit the number of waveguides written simultaneously. We have checked that the LCOS-SLM worked correctly during the exposure of 100 fs laser pulses of above 1 mJ at 1 kHz for 10 hours. Considering that the active area of the SLM was 2 cm2, the damage threshold of the LCOS-SLM should be larger than 5 × 109 Wcm−2. In waveguides writing in the present study, the pulse energy per one beam spot was up to 2.0 μJ and four beam spots were generated. Therefore, the resulting fluence on the SLM was only 4 × 107 Wcm−2, which is smaller than 1/125 of the damage threshold. This means that the damage threshold of the SLM is not a critical factor limiting the number of waveguides.
4. Results and discussion
A layout of a 3D 1 × 4 splitter is illustrated in Figs. 4(a) and 4(b). The single waveguide in the region ‘AB’ is branched into four waveguides at point ‘B’, and the four waveguides gradually separate in the region ‘BC’. In the region ‘CD’, the four waveguides are straight and mutually parallel. In the writing experiment, the translation direction of the glass was –y, i.e., the waveguides were written from ‘A’ to ‘D’. Figure 4(c) shows the distribution of the focused beam spots inside a silica glass (refractive index~1.46) for writing the waveguide (P1-P4). Because each focused spot is spatially separated from the other spots, no interference occurs between them. The positions of the focused spots by the n-th CGH are P1(n): [-(n/N)(Lx/2), ΔY, d + (n/N)(Lz/2)], P2(n): [(n/N)(Lx/2), 2ΔY, d + (n/N)(Lz/2)], P3(n): [-(n/N)(Lx/2), 3ΔY, d-(n/N)(Lz/2)], and P4(n): [(n/N)(Lx/2), 4ΔY, d-(n/N)(Lz/2)], where d is the focal depth from the glass surface without a CGH, ΔY is the distance along the y axis between the adjacent focused spots, n is an integer from 0 to N, and N + 1 is the number of CGHs prepared for writing the splitter. In this experiment, we chose N = 256, Lx = 83.3 μm, Lz = 81 μm, d = 360 μm and ΔY = 10.4 μm. For example, a CGH of n = 256 and simulated light intensity distribution were shown in Figs. 4(d) and 4(e), respectively. The simulated light distribution shows clearly that beam spots ‘1’, ‘2’ and ‘3′, ‘4’ are focused at different depths.
Fig. 4 (a) Birds-eye view of 3D 1 × 4 splitter fabricated in this study. (b) Top and side views of the splitter. (c) Distribution of the focused beam spots for writing a splitter. (d) Example of a CGH used in writing a splitter, and (e) the simulated light intensity distributions.
Figure 5 shows optical microscope images of a 1 × 4 splitter in which LAB = 5 mm, LBC = 10 mm and LCD = 5 mm. The fabrication of the splitter can be seen in the media [Media 1]. The structures in the single straight region (a), branching region (b), separating region (c), and four-straight region (d) are shown. Cross sections of the waveguides are also shown in Figs. 5(e)-5(h). In the fabrication of this splitter, the pulse energy at one waveguide was about 1.6 μJ and the translation velocity of the glass was 40 μm/s. No discontinuous structure was observed in all the regions, although CGHs were switched during the writing of the waveguides [17]. In the branching region [Fig. 5(b)], the waveguide was branched gradually in the x-direction, which reflects that two adjacent beam spots did not interfere with each other. However, the waveguides are not completely split in the z-direction at the same region [Fig. 5(f)], because the intensity distribution of the fs laser pulse spreads in the z-direction due to the low NA of the focusing lens ( = 0.45). In the separating and four-straight regions [Figs. 5(c) and (d)], the waveguides look defocused, because the z-positions of the four waveguides are changed from the original ones in these regions. In the output of the waveguide [Fig. 5(h)], there are four spatially separated waveguide structures. The transversal and vertical distances between the central positions of the waveguide structures are 86 μm and 91 μm, respectively. These values are slightly longer than those of the designed waveguides, Lx = 83.3 μm and Lz = 81 μm. The difference in the transversal distance could be due to the fact that the distance between two lenses (L1 and L2) in the telescope was not perfectly aligned, because the magnification of the fs laser beam depends on the distance. On the other hand, the difference in the vertical distance was probably due to the refractive index of the glass sample, because the focal depth depends on the refractive index. These differences are not so critical, because they can be corrected based on an experimental result.
Fig. 5 Optical microscope image of a 1 × 4 splitter. (a) Single-straight region, (b) branching region, (c) separating region, and (d) four-straight region. (e)-(h) Cross section images focused at each region. The fabrication of this splitter can be seen in Media 1.
The cross sections of the waveguides were elongated in the propagation direction of the excitation laser beam [Figs. 5(e)-5(h)]. This shape is the result of the depth of focus with loosely focusing and the self-focusing. Because the elongated cross section of the waveguide could cause high fiber coupling loss, it is desirable to make the cross section circular. Other researchers have reported that a waveguide with a circular cross-section can be fabricated with a high repetition rate fs laser (>200 kHz) [22,23]. If we use a high repetition rate fs laser, the cross sections will be circular and the fiber coupling loss will be improved. The laser writing with a high repetition rate fs laser is under investigation.
The splitting of a laser beam at 635 nm was observed by coupling the laser beam into the input of the fabricated 1 × 4 splitter with an optical fiber (the core diameter was 12 μm). The near-field intensity distribution at the exit of the splitter is shown in Fig. 6(a) . The guided laser beam was spatially separated into four by the splitter [Fig. 6(b)], and the splitting ratio was 28:23:25:24. Although the splitting ratio is not equal, it can be improved by optimizing the structure of the splitter and the processing parameters, such as writing speed, pulse energy and repetition rate of fs laser pulses. The transversal separation of the guided beams was about 86 μm, which is almost the same as the transversal distance between the waveguide structures. On the other hand, the vertical separation was about 100 μm, which is longer than the vertical distance between the waveguides [91 μm shown in Fig. 5(h)]. This should be attributed to the elongated structure in the z-direction and the bent structure in the splitter [2]. Further study is required to reduce the shift of the beam exit positions by optimizing the structure of the splitter.
Fig. 6 (a) Intensity distribution of the laser beam guided through the fabricated 1 × 4 splitter. (b) Three dimensional representation of the intensity distribution in (a).
Although the losses of the splitter were not evaluated, we compared the guided light intensity through a splitter and that through a single waveguide. The averaged light intensity through the 1 × 4 splitter was about 21% of that through a single waveguide. Therefore, about 16% of the light through a single waveguide was lost due to light splitting. This loss is sufficiently small for application.
Other researchers also have demonstrated the simultaneous writing of multiple waveguides with an SLM [15–18]. For example, Pospiech et al. fabricated a Y coupler inside a glass [16,18], and Mauclair et al. fabricated various photonic devices inside a glass by the similar method [17]. One difference between our works and theirs is that more than two waveguides can be written three-dimensionally and simultaneously by our method, while only two waveguides by their method. It is because our calculation method for obtaining a CGH is based on searching a spatial phase distribution which generates light spots at the desired positions, on the other hand, their methods are simply based on a phase grating and Fresnel lens. Another difference is that the structure of the waveguide was continuous in the branching region of a 1 × 4 splitter in our work, while the waveguides was disconnected in the branching region in their work [18]. The fabrication of a continuously branched structure is very important, because continuous line structure is essential for other applications such as fabrication of 3D micro-channels inside transparent materials [24]. While their method has a critical limitation by which continuous branched lines cannot be written, our method has a much broader range of applications, i.e. various kinds of 1xN splitter can be fabricated [Figs. 7(a) -7(c)] and it is applicable for making 3D branched micro-channels.
Fig. 7 Other examples of splitters written by the same method. (a) 1x3 splitter, (b) 1x4 splitter with different shape of output, and (c) 1x6 splitter.
5. Summary and conclusion
We demonstrated the fabrication of 1 × 4 splitter waveguides by creating four beam spots of fs laser pulses inside a glass substrate with an SLM. With this method, continuously branched waveguides can be written easily by translating a substrate only one time, therefore, we do not have to focus on precise position control of the translation stage, which was essential in the conventional fabrication method of splitters. Although the structures of a splitter must be optimized to get an equal splitting ratio and to increase the quality of the intensity distribution of the guided beam, this method is a major step forward in developing a more efficient method to produce more complicated optical devices such as 1 × N splitters and directional couplers, as well as three-dimensionally branched microchannels inside various transparent materials.
Acknowledgements
The authors would like to thank Dr. T. Hara, Mr. H. Ito, Mr. N. Matsumoto, Mr. N. Fukuchi and Mr. T. Inoue from HAMAMATSU Co. and Prof. Y. Hayasaki from Utsunomiya university for giving us good suggestion and helpful discussion. This research was carried out in the High Efficiency Processing Technology for Three-Dimensional Optical Devices Project supported by the New Energy and Industrial Technology Development Organization (NEDO).
References and links
1. D. B. Keck, A. J. Morrow, D. A. Nolan, and D. A. Thompson, “Passive Components in the Subscriber Loop,” J. Lightwave Technol. 7(11), 1623–1633 (1989). [CrossRef]
2. M. H. Hu, J. Z. Huang, R. Scarmozzino, M. Levy, and R. M. Osgood Jr., “A low-loss and compact waveguide Y-branch using refractive-index tapering,” IEEE Photon. Technol. Lett. 9(2), 203–205 (1997). [CrossRef]
3. H. Takahashi, Y. Ohmori, and M. Kawachi, “Design and fabrication of silica-based integrated-optic 1X128 power splitter,” Electron. Lett. 27(23), 2131–2133 (1991). [CrossRef]
4. N. Takato, K. Jinguji, M. Yasu, H. Toba, and M. Kawachi, “Silica-based single-mode wave-guides on silicon and their application to guide-wave optical interferometers,” J. Lightwave Technol. 6(6), 1003–1010 (1988). [CrossRef]
5. B. J. Luff, R. D. Harris, J. S. Wilkinson, R. Wilson, and D. J. Schiffrin, “Integrated-optical directional coupler biosensor,” Opt. Lett. 21(8), 618–620 (1996). [CrossRef] [PubMed]
6. T. G. Giallorenzi, E. J. West, R. Kirk, R. Ginther, and R. A. Andrews, “Optical waveguides formed by thermal migration of ions in glass,” Appl. Opt. 12(6), 1240–1245 (1973). [CrossRef] [PubMed]
7. S. M. Garner, V. Sang-Shin Lee, A. Chuyanov, A. Chen, W. H. Yacoubian, Steier, L. R. Dalton, S. S Lee, V Chuyanov, A Chen, A Yacoubian, W. H. Steier, and L. R Dalton, “Three-Dimensional Integrated Optics Using Polymers,” IEEE J. Quantum Electron. 35(8), 1146–1155 (1999). [CrossRef]
8. M. Svalgaard, C. V. Poulsen, A. Bjarklev, and O. Poulsen, “Direct UV writing of buried singlemode channel wave-guides in Ge-doped silica films,” Electron. Lett. 30(17), 1401–1403 (1994). [CrossRef]
9. K. M. Davis, K. Miura, N. Sugimoto, and K. Hirao, “Writing waveguides in glass with a femtosecond laser,” Opt. Lett. 21(21), 1729–1731 (1996). [CrossRef] [PubMed]
10. K. Minoshima, A. M. Kowalevicz, I. Hartl, E. P. Ippen, and J. G. Fujimoto, “Photonic device fabrication in glass by use of nonlinear materials processing with a femtosecond laser oscillator,” Opt. Lett. 26(19), 1516–1518 (2001). [CrossRef]
11. M. Ams, G. D. Marshall, D. Spence, J. M. Dawes, J. A. Piper, and M. J. Withford, “Direct writing of guided wave devices inside bulk glass using femtosecond lasers,” AOS News 19, 8–11 (2005).
12. S. Nolte, M. Will, J. Burghoff, and A. Tuennermann, ““Femtosecond waveguide writing: a new avenue to three-dimensional integrated optics,” Appl. Phys,” Adv. Mater. 77, 109–111 (2003).
13. W. Yang, C. Corbari, P. G. Kazansky, K. Sakaguchi, and I. C. Carvalho, “Low loss photonic components in high index bismuth borate glass by femtosecond laser direct writing,” Opt. Express 16(20), 16215–16226 (2008). [CrossRef] [PubMed]
14. Y. Hayasaki, T. Sugimoto, A. Takita, and N. Nishida, “Variable holographic femtosecond laser processing by use of a spatial light modulator,” Appl. Phys. Lett. 87(3), 031101 (2005). [CrossRef]
15. C. Mauclair, G. Cheng, N. Huot, E. Audouard, A. Rosenfeld, I. V. Hertel, and R. Stoian, “Dynamic ultrafast laser spatial tailoring for parallel micromachining of photonic devices in transparent materials,” Opt. Express 17(5), 3531–3542 (2009). [CrossRef] [PubMed]
16. M. Pospiech, M. Emons, A. Steinmann, G. Palmer, R. Osellame, N. Bellini, G. Cerullo, and U. Morgner, “Double waveguide couplers produced by simultaneous femtosecond writing,” Opt. Express 17(5), 3555–3563 (2009). [CrossRef] [PubMed]
17. M. Sakakura, T. Sawano, Y. Shimotsuma, K. Miura, and K. Hirao, “Parallel drawing of multiple bent optical waveguides by using a spatial light modulator,” Jpn. J. Appl. Phys. 48(12), 126507 (2009). [CrossRef]
18. M. Pospiech, M. Emons, B. Väckenstedt, G. Palmer, and U. Morgner, “Single-sweep laser writing of 3D-waveguide devices,” Opt. Express 18(7), 6994–7001 (2010). [CrossRef] [PubMed]
19. T. Inoue, H. Tanaka, N. Fukuchi, M. Takumi, N. Matsumoto, T. Hara, N. Yoshida, Y. Igasaki, and Y. Kobayashi, “LCOS spatial light modulator controlled by 12-bit signals for optical phase-only modulation,” Proc. SPIE 6487, Y11 (2007).
20. J. Bengtsson, “Kinoform design with an optimal-rotation-angle method,” Appl. Opt. 33(29), 6879–6884 (1994). [CrossRef] [PubMed]
21. K. Iizuka, Engineering optics (Springer-Verlag, Berlin; Tokyo, 1985)
22. C. B. Schaffer, A. Brodeur, J. F. García, and E. Mazur, “Micromachining bulk glass by use of femtosecond laser pulses with nanojoule energy,” Opt. Lett. 26(2), 93–95 (2001). [CrossRef]
23. S. M. Eaton, H. Zhang, P. R. Herman, F. Yoshino, L. Shah, J. Bovatsek, and A. Y. Arai, “Heat accumulation effects in femtosecond laser-written waveguides with variable repetition rate,” Opt. Express 13(12), 4708–4716 (2005). [CrossRef] [PubMed]
24. Y. Kondo, J. Qiu, T. Mitsuyu, K. Hirao, and T. Yoko, “Three-dimensional microdrilling of glass by multiphoton process and chemical etching,” Jpn. J. Appl. Phys. 38(Part 2, No. 10A), L1146–L1148 (1999). [CrossRef]
