We report on NIR efficient end-coupling in single-mode silicon on insulator waveguides. Efficient coupling has been achieved using Polymer-Tipped Optical Fibers (PTOF) of adaptable radius of curvature (ROC). When compared with commercial micro lenses, systematic studies as a function of PTOF ROC, lead for subwavelength PTOF to a coupling factor enhancement as high as 2.5. This experimental behavior is clearly corroborated by radial FDTD simulations and an absolute coupling efficiency of about 50% is also estimated.
© 2009 Optical Society of America
Highly integrated optical waveguide devices are expected to play an increasing role in telecommunication systems, therefore, how to couple light into diffraction-limited waveguide becomes an important issue. Among the years, a large number of methods of light coupling into optical waveguide have been investigated, these include the grating coupler [1–3], polarization insensitive coupling technique [4–5] and inverted lateral taper [6–7] and dual-grating assisted directional coupler . Among these methods, the end coupling with single-mode fiber is a usual method to inject light into integrated optical waveguides. However, it is not really suitable for sub-micrometer-thick waveguides like SOI technology based waveguides.
Recently Polymer-Tipped Optical Fibers (PTOF) based on free-radical photo polymerization on the top end of optical fibers was introduced in our group by Bachelot et al . In the case of a semiconductor NIR edge emitting laser diode, a coupling efficiency, as high as 70%, was obtained . In this paper, we investigate the end coupling in SOI integrated device using Polymer-Tipped Optical Fibers (PTOF). We show that in the case of SOI waveguides, PTOF can lead to very efficient coupling, 2.5 times higher than with conventional microlensed fibre.
2. Devices description
The developed photopolymerization technique is a very attractive technology allowing for tight focusing of the light with a very small working distance. One of the main advantages is that the radius of curvature (ROC) of the polymer tip apex can be varied from about 200 nm to several microns by simply changing the exposure power. For a given waveguide structure, this parameter, which is directly connected to both the working distance and the spot diameter, can be adjusted to obtain the maximum coupling efficiency, i.e the best matching between the focus spot and the waveguide mode(s). In this paper, we focus on four types of PTOF, characterized by four different radius of curvature at the tip apex (ROC = 0.58 μm, 1.05μm, 1.55μm and flat). These PTOF have been realized on standard SMF28 fiber manufactured by Corning, using a formulation similar to the one described in the reference , an exposure time of few seconds and an optical power ranging from 80 nW to 500 nW. Two SMF28-PTOF examples are shown in Fig. 1. More details on the PTOF fabrication can be found in .
The coupling between these four PTOF and two types of single mode SOI strip buried waveguide are studied in the following parts. The first single mode SOI waveguide is a straight silicon core having a rectangular cross-section of 0.25*0.3 μm , completely embedded in a silica layer of 2 μm. The second sub-micron waveguide has the same design but a linear taper (2*0.3 μm2 ) was added at the entrance in order to increase the coupling efficiency with commercial microlensed fibers. Structures were fabricated at the LETI/DIHS & DOPT-LPM (France).
3. Experimental setup and results
The experimental setup sketched in Fig. 2 is based on the end-coupling method. Laser beam at a wavelength of 1.55μm from a non polarized ASE light source is injected into the single-mode fiber. The PTOF position is accurately adjusted in front of the SOI waveguide with an x-y-z piezoelectric stage. The output light is collected by AR treated microlensed fiber of 2.5μm spot size (1/e2) provided by OZ-optics.
With this setup, the maximum coupling efficiency and the optimal working distance (WD) are obtained by scanning successive (X, Z) transverse plans along the Y direction. In Fig. 3, examples of detected intensity distributions are compared in the case of a commercial micro lens and a polymer tip with an ROC equals to 0.58μm.
Notably, the detected intensity exhibits a much sharper maximum in the case of the polymer probe which can be attributed to a high field confinement near the tip apex. From these intensity maps, the WD distance can be retrieved and an enhancement factor can be estimated. The table 1 summarizes the experimental results obtained on the untapered waveguide for the four PTOF.
The enhancement factor IPTOF /IML is here defined as the ratio between the maximum detected output power IPTOF obtained with the PTOF and the maximum detected power with the commercial microlens IML. Highest enhancement factors are obtained for the smallest tip apexs. This is not really surprising as tightly focused spots are required to maximize the overlap between the incident field and the strongly confined guided mode. Depending on the light confinement of the considered structure, other ROC could however be selected in order to maximize the light coupling. In order to gain a better insight on these experimental results, and demonstrate the versatility of the PTOF, the optical intensity distribution is studied using numerical simulation tools in the following part.
4. Simulation results and discussion
In this part, the Finite Difference Time Domain (FDTD) method  is used to simulate both the focusing properties of the single polymer microtip and the coupling efficiency with SOI waveguide.
The light propagation inside the micro-tip fibers is mapped for different ROC, in order to obtain important parameters such as working distance and spot diameter and to explain the results obtained on the untapered waveguide. Owing to the azimuthal symmetry of the PTOFs, a radial FDTD (2.5D) in full vectorial mode was used (Fullwave commercial software from Rsoft). The wavelength λ is fixed to 1.55μm with a grid step size of 0.07μm for all the simulations. The refractive index of the polymer is chosen equals to 1.52. From careful SEM images analysis of PTOF, the polymer shape is found to be well described by a monomial function of the form:
where L is the PTOF’s length, w 0 is the width of the PTOF’s base and α is a fitted parameter which describes the general shape of the taper which is closely related to the final ROC. The Figs. 4(a)–4(d) shows the spatial intensity distribution for the four PTOFs, when the fundamental fiber mode is excited. From these results, the focus spot diameter and a working distance can be estimated as shown in Figs. 4(e)–4(f). As it could be expected, the working distance and the spot diameter increase with the radius of curvature (from 0.58μm to flat). For smaller radius of curvature, a higher confinement near the apex is observed as the evanescent field contribution becomes important. It is clear that the maximum coupling efficiency with a waveguide is not necessarily obtained when the waveguide’s front end is placed onto the focused spot, as it may not correspond to the best matching with the guided optical modes . However in the case of the submicron SOI waveguides under study, the optical mode is highly confined and the coupling is indeed expected near the focus spot as observed in the experimental part, where the maximum enhancement factor is obtained when the waveguide is brought in the near-field of the sharpest polymer taper.
Simultaneously, to illustrate the influence of the PTOF geometry on the coupling efficiency, the PTOF-waveguide system was simulated. The coupling efficiency was numerically measured for different ROC. The Fig. 5 shows the light distribution in the whole system in the case of a sharp PTOF.
Note that in the experiment the coupling efficiency is lower because the waveguide doesn’t have a radial symmetry so that the modes overlaps are weaker. Table 2 summarizes simulated results for various radius of curvature compared to the value obtained for a Gaussian injection profile of 2.5 μm of diameter (1/e2).
As for the experiment the coupling enhancement factor strongly increases for the two sharpest PTOF to reach a maximum value of 3. These results together with the exponential increase of the injected intensity for the sharpest tip (cf. Fig. 3(d)) confirm the evanescent nature of the coupling. Although experimental and theoretical factors are in a rather good agreement, they cannot be straightforwardly compared. While giving the general trend, the radial FDTD simulation cannot account for the injection in a channel waveguide due the mismatch between the axial symmetric PTOF modes and the asymmetric SOI waveguide mode. Indeed compared to the untapered waveguide experiments, the guided mode profile has here a wider extension since there is no air superstrate. This could explain the larger enhancement factor obtained in the simulation. On other consequence is that an optimum coupling is less critical to achieve and can actually be already obtained for wider PTOF (ROC=1.55μm) and at a distance where the evanescent field has a smaller contribution.
While a qualitative interpretation of the experimental results can be done using the radial simulation, it is clear that only a full vectorial 3D-calculation would quantitatively reproduce the experiment, as the considered waveguide is rectangular. For the time being such a simulation was not possible in the laboratory. However, to estimate the amount of light coupled to the waveguide, the “microlensed fiber” experiment was simulated using 3D FDTD by considering a Gaussian input field having a beam waist of 2.5μm. A coupling efficiency of about 20% was obtained (± 1% for the two orthogonal polarization states ). Considering the maximum experimental enhancement factor, this leads to an efficiency of about 50% in the case of the 1.05 μm ROC PTOF . This value is comparable with the most efficient integrated coupling techniques, such as the grating coupler technique [1–3].
In conclusion, we have proposed and demonstrated that a PTOF can efficiently couple light between a submicron size SOI-waveguide and a single-mode fiber. Compared to a commercial microlense, for a SOI waveguide with a taper (2×0.3μm2) and a ROC (1.55μm), the coupling efficiency is enhanced by 1.3. In the case of a SOI waveguide without taper (0.2×0.3μm2), the coupling efficiency using sub-wavelength PTOF exceeds by 2.5 times the coupling efficiency obtained with a commercial lensed fiber. Experimental results are in agreement with numerical simulations confirming the evanescent nature of the coupling in the case of these untappered waveguides. PTOF appear as excellent candidates for coupling light in sub-micrometer SOI waveguide using the end-fire technique.
SOI samples have been fabricated at the CEA/LETI under the coordination of J. M. Fedeli. This work was partially supported by the “Région Champagne-Ardenne” and the European Social Found.
References and links
1. R. Orobtchouk, A. Layadi, H. Gualous, D. Pascal, A. Koster, and S. Laval, “High-Efficiency Light Coupling in a Submicrometric Silicon-on-Insulator Waveguide,” Appl. Opt. 39, 5773–5777 (2000), http://www.opticsinfobase.org/abstr act.cfm?URI=ao-39-31-5773. [CrossRef]
2. D. Taillaert, W. Bogaerts, and R. Baets, “Efficient coupling between submicron SOI- waveguides and single-mode fibers,” IEEE/LEOS Benelux Chapter, Enscheda (2003).
3. L. Vivien, D. Pascal, S. Lardenois, D. Marris-Morini, E. Cassan, F. Grillot, S. Laval, J. Fedeli, and L. El Melhaoui, “Light Injection in SOI Microwaveguides Using High-Efficiency Grating Couplers,” J. Lightwave Technol. 24, 3810–3815(2006), http://www.opticsinfobase.org/JLT/abstract.cfm7URWLT-24-10-3810. [CrossRef]
4. L. Vivien, S. Laval, E. Cassan, X. L. Roux, and D. Pascal, “2-D Taper for Low-Loss Coupling Between Polarization-Insensitive Microwaveguides and Single-Mode Optical Fibers,” J. Lightwave Technol. 21, 2429–(2003), http://www.opticsinfobase.org/JLT/abstract.cfm7URWLT-21-10-2429. [CrossRef]
5. J. V. Galan, P. Sanchis, G. Sánchez, and J. Marti, “Polarization insensitive low-loss coupling technique between SOI waveguides and high mode field diameter single-mode fibers,” Opt. Express 15, 7058–7065 (2007), http://www.opticsinfobase.org/oe/abstract.cfm?URI=oe-15-l 1-7058. [CrossRef] [PubMed]
6. Almeida, R. R. Panepucci, and M. Lipson, “Nanotaper for compact mode conversion,” Opt. Lett. 28,1302–1304 (2003), http://www.opticsinfobase.org/abstract.cfm?URI=OL-28-15-1302. [CrossRef] [PubMed]
7. T. Wahlbrinka, W. S. Tsaic, M. Waldowb, M. Forstb, J. Boltena, T. Mollenhauera, and H. Kurza, “Fabrication of high efficiency SOI taper structures,” Mic. Eng. in press.
8. G. Masanovic, G. Reed, W. Headley, B. Timotijevic, V. Passaro, R. Atta, G. Ensell, and A. Evans, “A high efficiency input/output coupler for small silicon photonic devices,” Opt. Express 13, 7374–7379 (2005), http://www.opticsinfobase.org/oe/abstr act.cfm?URI=oe-13-19-7374. [CrossRef] [PubMed]
9. R. Bachelot, C. Ecoffet, P. Deloeil, P. Royer, and D. J. Lougnot. ”Integration of Micrometer-Sized Polymer Elements at the End of Optical Fibers by Free-Radical Photopolymerization,” Appl. Opt. 40, 5860–5871 (2001). [CrossRef]
10. R. Bachelot, A. Fares, R. Fikri, D. Barchiesi, G. Lerondel, and P. Royer,“ Coupling semiconductor lasers into single-mode optical fibers by use of tips grown by photo polymerization,” Opt. Lett. 29, 1971–1973 (2004), http://www.opticsinfobase.org/ol/abstract.cfm?URI=ol-29-17-1971. [CrossRef] [PubMed]
11. A Taflove and S. C. Hagness, “The Finite-Difference Time-Domain Method in Computational Electrodynamics,” 2nd edition, (Boston Artech House2000).
12. M. Skorobogatiy, S. Jacobs, S. Johnson, and Y. Fink, “Geometric variations in high index-contrast waveguides, coupled mode theory in curvilinear coordinates,” Opt. Express 10,1227–1243 (2002), http://www.opticsinfobase.org/oe/abstr act.cfm?URI=oe-10-21-1227. [PubMed]
13. The difference observed between the two polarizations is small, mainly du to the fact that the waveguide’s cross-section is almost square (250nm×300nm) and also because the influence of the air/ sample surface is weak since the waveguide is notably buried.