Open Access
Add to BibSonomyAbstract
We propose and experimentally demonstrate an ultracompact transverse magnetic (TM) mode pass filter based on a rectangularly-shaped one-dimensional (1-D) photonic crystal silicon waveguide with an extremely high polarization extinction ratio (PER) of >30 dB and a low insertion loss (IL) of ~1 dB. The device structure of the TM mode-pass filter is numerically simulated using a three-dimensional (3-D) finite difference time domain (FDTD) method. The proposed device supports its fundamental TM mode only, whereas the transverse electric (TE) mode is reflected by the 1-D photonic crystals (PhCs). The measured PER of the fabricated TM mode-pass filter is ~34 dB, and the IL is about 1 dB. The entire device length is about 4 μm. Our simulation results predict that the device bandwidth of 30 dB PER is about 200 nm.
© 2016 Optical Society of America
1. Introduction
Silicon photonics based on the silicon-on-insulator (SOI) platform have been recognized as a key technology to realize photonic integrated circuits (PICs). Applications include optical nonlinear optics [1], interconnects [2], and optical computing [3]. However, nano-silicon waveguides with a high refractive index contrast of the core section with respect to their clad counterpart, and conventional rectangularly-shaped cross sectional geometry causes large polarization mode dispersions (PMDs) due to structural birefringence. Many types of passive polarization controlling devices based on nano-silicon waveguides, such as polarization beam splitters (PBSs) [4,5], polarization rotators (PRs) [6,7], and polarization mode-pass filters (PMPFs) [8–11], have been shown to be indispensable in order to overcome PMD issues. Planar-type PMPFs with high polarization isolating properties are key polarization control devices for PIC applications.
Various planar-type PMPFs have been proposed and demonstrated with the goal of achieving linearly polarized light with a high polarization extinction ratio (PER) and low insertion loss (IL). A transverse electric (TE) mode-pass filter based on a shallow etched silicon waveguide has been reported with a hybrid plasmonic waveguide [8]. Their calculated PER is ~20 dB when the device length is 8 μm. Another type of TE-pass polarizer based on a hybrid plasmonic Bragg grating only 5 μm long has been proposed in [9]. Its simulated PER value is > 17 dB over a wavelength range of 160 nm (from 1,480 to 1,640 nm). Recently, the subwavelength structure effect has been reported to demonstrate PMPF devices of high PER and low IL in a compact size [10,11]. A TE mode-pass filter based on a subwavelength grating has been proposed and demonstrated experimentally in [10]. The measured PER value is ~30 dB and the average IL is 0.4 dB. However, the device length is 60 μm. A transverse magnetic (TM) mode-pass filter based on a silicon subwavelength grating waveguide has been demonstrated for a measured PER value of ~27 dB with a device length of about 9 μm [11].
In this study we propose and experimentally demonstrate a simple and compact TM mode-pass filter based on a one-dimensional (1-D) photonic-crystal silicon waveguide (PhCW). The device provides an extremely high PER of ~34 dB in a relatively compact device length of ~4 μm which corresponds to a higher PER in a smaller size than those of the previous works. The TM mode-pass filter was numerically simulated using a three-dimensional (3-D) finite difference time domain (FDTD) method. The device can be fabricated in a single etch step after patterning lithography without a complicated metal coating or additional processes.
2. Device design and numerical simulations
The proposed TM mode-pass filter is based on a 1-D PhCW with subwavelength structures. The device is realized on an SOI wafer that has a 250 nm-thick silicon layer on a 3 μm-thick buried oxide (BOX) layer. Silicon dioxide is used for the upper cladding. 1-D photonic crystals (PhCs) of rectangular shape were placed at the center of the waveguide. Figure 1(a) shows a schematic configuration of the proposed TM mode-pass filter based on a 1-D PhCs. Figure 1(b) shows a scanning electron microscope (SEM) image of the 1-D PhC section.
Fig. 1 (a) Schematic diagram of the proposed TM mode-pass filter based on a 1-D PhCW. (b) SEM image of the fabricated device.
In this study, 1-D PhCWs with subwavelength structures were used to support a Bloch wave for the TM mode. In order to realize a TM mode-pass filter, the silicon waveguide with subwavelength structure should support the Bloch wave for the TM polarization mode only by choosing the structural parameters such that the TM polarized light passes through the 1-D PhCs region. However, the 1-D PhCs act as a Bragg reflector for TE polarization mode. Thus, most of the TE polarized light is reflected, and the proposed device satisfies the Bragg grating equation as follows [12]:
where nb = nwave(d/Λ) + nPhC(1-d/Λ) is the Bloch mode effective index, and nwave and nPhC are the effective indices in the waveguide and in the PhC region, respectively. Λ represents the lattice constant of the 1-D PhCs, and d is the length of the PhC hole as shown in Fig. 1(b). We used a finite difference eigenmode (FDE) solver (Lumerical’s mode solutionsTM) to calculate nwave and nPhC. Device parameters of Wwave = 450 nm, WPhC = 200 nm, d = 210 nm, and Λ = 420 nm were used for the proposed device. For these parameters, the calculated values of nTMwave and nTMPhC are 1.940 and 1.587, respectively, which satisfy the Bragg grating condition of Eq. (1). Figure 2 shows the calculated band diagrams of the PhC waveguide of the given parameters for the TM and TE modes. The band diagrams were calculated using the dispersion relation described in [13]. The band diagram indicates that the wavelength region around 1,550 nm is located in a wide bandgap for the TE polarization mode. However, a propagation mode supported by the Bloch mode can be available in a wide wavelength range near 1,550 nm for the TM polarization mode.Fig. 2 Calculated band diagram of the TM mode-pass filter shown in Fig. 1. Λ and c are the lattice constant of the 1-D PhCs and the speed of the light in a vacuum, respectively.
The characteristics of the proposed TM mode-pass filter were evaluated using 3-D FDTD (Lumerical’s FDTD solutionsTM) simulations. From the simulation results, the optimum device parameters are determined. The mesh sizes used at the 1-D PhCs region are Δx = 10 nm, Δy = 25 nm, and Δz = 10 nm. A perfectly matched layer was applied to eliminate reflection from the simulation boundaries. The operating wavelength is 1,550 nm. To determine the optimum condition, we changed parameters of the 1-D PhCs to other values. The width of the PhCs, WPhC, was varied from 50 to 375 nm with a 25 nm interval. The fill factor of the PhCs, FF ( = d/Λ), was varied from 0.2 to 0.8 nm with a 0.5 interval. The lattice constant of the PhCs Λ and the number of PhCs N were fixed at 420 nm and 10, respectively. PER and IL of the TM mode-pass filter are defined as
andwhere TTM and TTE are transmittances of the TM and TE modes at the end of the PhC waveguide.Figure 3 shows 2-D contour plots of the calculated PER and IL values as functions of WPhC and FF. We note that the calculated PER values are higher than 30 dB when WPhC ranges from 175 to 350 nm and FF ranges from 0.45 to 0.65. Within these ranges, the calculated IL values are also < 2 dB. Thus, the proposed devices with suitable values of WPhC and FF can provide a high PER despite minor fabrication errors encountered during device processing. In various simulation conditions, the calculated PER values are greater than 20 dB. For WPhC = 200 nm, FF = 0.5 and N = 10 (with a corresponding length of ~4 μm), the calculated PER and IL are 36 dB and 0.8 dB, respectively, at a wavelength of 1,550 nm. Thus, this condition was taken to represent our optimized device parameters.
Fig. 3 Calculated 2-D contour plots of the TM mode-pass filter with a 1-D PhCW of Λ = 420 nm and N = 10 (at λ = 1,550 nm) for PER and IL as functions of WPhC and FF.
Figure 4 shows simulated results of light propagation along the 1-D PhC waveguide with optimized parameters for the TE and TM polarized light inputs, respectively. The TM polarized light passes through the 1-D PhC region with help of Bloch wave surpport, while the TE polarized light is reflected.
Fig. 4 Simulated light propagation of TE and TM polarization mode inputs in the TM mode-pass filter made from a 1-D PhCW for WPhC = 200 nm, Λ = 420 nm, FF = 0.5, and N = 10 (at λ = 1,550 nm).
3. Device fabrication and characterization results
The simulated TM mode-pass filters were fabricated on an SOI wafer with a 250 nm top Si layer and a 3 μm BOX layer. Electron beam lithography was used to pattern the device structure. Inductively-coupled plasma-reactive ion etching (ICP-RIE) was used to form the waveguide devices. In order to have input/output coupling of the light beams of both polarization modes into/from the 1-D PhCW, two different grating couplers, each designed for the TM and TE modes, were formed at the ends of the waveguides. The grating periods used for the TE and TM mode couplings were 580 and 870 nm, respectively. FF and the etching depth were 0.5 and 70 nm, respectively, and were kept the same for both types of gratings. Bridged directional coupler-type polarization beam splitters (DC-type PBSs) [5] were formed on both sides of the 1-D PhCW to combine and split to the TE and TM mode lights. The chip was finally covered by an 800 nm-thick SiO2 layer as shown in Fig. 1(a). The coupled TM mode light from the TM mode grating coupler is transferred to the core waveguide after propagation through a coupling length at the PBS. Then, the light passes through the 1-D PhC region, and exits through the output TM-mode grating coupler. On the other hand, the TE mode light coupled into the 1-D PhCW from the input TE mode grating coupler after passing through the DC-type PBS is reflected mostly by the 1-D PhCs.
To measure the performance of the fabricated devices experimentally, a tunable laser (Agilent 8168F) was used as an optical source. A polarization controller (PC) was used to adjust the polarization of the input source light. The output light was measured with a power meter (EXPO FPM-300), whose minimum detection power is about −60 dBm. Various 1-D PhCWs of the PhC size WPhC (ranging from 100 to 300 nm) with a fixed FF = 0.5 and N = 10 were fabricated. Other devices with various FF values ranging from 0.2 to 0.8 were also fabricated. In this case, WPhC and N were fixed at 200 nm and 10, respectively. The reference device without the 1-D PhC structure on the same waveguide chip was used to calculate the relative transmittance of the 1-D PhCWs. The transmittance for both polarization modes is expressed as
where Pref,TE/TM and Pfil,TE/TM are measured output powers of the reference device and the TM-mode-pass filter for input lights of both polarization modes, respectively. Measured transmittances of the fabricated devices for the TE and TM mode inputs at a 1,550 nm wavelength are plotted as functions of WPhC and FF in Fig. 5. Filled and open symbols indicate the measured and simulated results, respectively. The measured results show that variation of WPhC and FF affects the transmittance of the TE polarization mode more significantly than that of the TM polarization mode, which is in good agreement with the simulation results. We note that the measured PER values are higher than 30 dB when the WPhC values are greater than 150 nm. Some excess losses are also observed, which are mainly attributed to the variation of the fiber chip coupling loss and the loss in the 1-D PhC section due to fabrication error. For the device with optimized parameters, PER and IL were measured to be about 34 dB and 1 dB, respectively. The measured transmittances of the fabricated devices are shown in Fig. 6(a) for input lights of both polarization modes as functions of 1-D PhC number N when WPhC and FF are fixed at 200 nm and 0.5, respectively. The simulation results indicate that the excess loss of the device for the TM mode is about 1 dB, and does not vary significantly as N increases. The maximum PER value is ~40 dB with N = 20 (the corresponding length of the PhC section is ~8 μm). The device transmittance for the TE polarized light input saturates as N increases. The excess loss of the device also increases as N varies. This means that the PER is saturated or decreases as N increases further. Thus, we can choose the proper value of N in order to achieve a high PER and proper IL. Finally, we measured the transmission spectrum of the fabricated device using a broadband light source (EDFA ASE) and an optical spectrum analyzer (Anritsu MS9710B). Measured results are shown in Fig. 6(b). The solid lines and symbols correspond to the 1-D PhC device and to the reference waveguide without any 1-D PhCs, respectively. Both the measured and simulated PER values are about >30 dB over the entire measurement wavelength range. The simulated 30 dB bandwidth is about 200 nm (1,480 nm to 1,680 nm).Fig. 5 Simulated and measured transmittances for TE and TM mode inputs as functions of (a) WPhC and (b) of FF when N = 10 (at λ = 1,550 nm).
Fig. 6 (a) Simulated and measured transmittances of the TM mode-pass filter with optimized parameters for input lights of TE and TM modes as functions of 1-D PhC number N. (b) Measured transmittances spectra of fabricated reference and TM mode-pass filter devices.
4. Conclusion
We propose and demonstrate experimentally an ultracompact TM mode-pass filter based on a 1-D PhCW with subwavelength structures on an SOI wafer covered by an SiO2 upper clad. The proposed device can be fabricated in a single lithography and etching process, and can deliver high device performance with an extremely high PER value and a low IL value. When N, the number of PhCs, equals 10, the measured PER and IL values of the device are about 34 dB and 1 dB at a 1,550 nm wavelength. The overall device length is only about 4 μm. Although the IL increases slightly, higher PER values can be achieved when N is increased further. This suggests that another option to achieve an integrated polarization filter with an extremely high PER (> 40 dB), a relatively low IL (~2 dB), and a wide bandwidth is possible.
Funding
This work was supported in part by the Future Semiconductor Device Technology Development Program (project number 10044735) funded by the Ministry of Trade, Industry & Energy (MOTIE) of the Korean government and the Korea Semiconductor Research Consortium (KSRC), and in part by the Basic Science Research Programs through the National Research Foundation of Korea (NRF), funded by the Ministry of Science, ICT & Future Planning of the Korean government (no. 2013R1A1A2012409).
References and links
1. X. Liu, R. M. Osgood, Y. A. Vlasov, and W. M. J. Green, “Mid-infrared optical parametric amplifier using silicon nanophotonic waveguides,” Nat. Photonics 4(8), 557–560 (2010). [CrossRef]
2. G. T. Reed, G. Mashanovich, F. Y. Gardes, and D. J. Thomson, “Silicon optical modulators,” Nat. Photonics 4(8), 518–526 (2010). [CrossRef]
3. L. Liu, R. Kumar, K. Huybrechts, T. Spuesens, G. Roelkens, E. J. Geluk, T Vries, P Regreny, D. V. Thourhout, R Baets, and G Morthier, “An ultra-small, low-power, all-optical flip-flop memory on a silicon chip,” Nat. Photonics 4, 182–187 (2010).
4. H. Fukuda, K. Yamada, T. Tsuchizawa, T. Watanabe, H. Shinojima, and S. Itabashi, “Ultrasmall polarization splitter based on silicon wire waveguides,” Opt. Express 14(25), 12401–12408 (2006). [CrossRef] [PubMed]
5. D. W. Kim, M. H. Lee, Y. Kim, and K. H. Kim, “Planar-type polarization beam splitter based on a bridged silicon waveguide coupler,” Opt. Express 23(2), 998–1004 (2015). [CrossRef] [PubMed]
6. J. N. Caspers, M. Z. Alam, and M. Mojahedi, “Compact hybrid plasmonic polarization rotator,” Opt. Lett. 37, 4615–4617 (2012). [CrossRef] [PubMed]
7. M. Yin, Q. Deng, Y. Li, X. Wang, and H. Li, “Ultrashort and low-loss polarization rotators utilizing hybrid plasmonic-dielectric couplers,” IEEE Photonics Technol. Lett. 27(3), 229–232 (2015). [CrossRef]
8. M. Z. Alam, J. S. Aitchison, and M. Mojahedi, “Compact and silicon-on-insulator-compatible hybrid plasmonic TE-pass polarizer,” Opt. Lett. 37(1), 55–57 (2012). [CrossRef] [PubMed]
9. J. Zhang, E. Cassan, and X. Zhang, “Wideband and Compact TE-Pass/TM-Stop Polarizer Based on a Hybrid Plasmonic Bragg Grating for Silicon Photonics,” J. Lightwave Technol. 32(7), 1383–1386 (2014). [CrossRef]
10. Y. Xiong, D.-X. Xu, J. H. Schmid, P. Cheben, and W. N. Ye, “High extinction ratio and broadband silicon TE-pass polarizer using subwavelength grating index engineering,” IEEE Photonics J. 7(5), 7802107 (2015). [CrossRef]
11. X. Guan, P. Chen, S. Chen, P. Xu, Y. Shi, and D. Dai, “Low-loss ultracompact transverse-magnetic-pass polarizer with a silicon subwavelength grating waveguide,” Opt. Lett. 39(15), 4514–4517 (2014). [CrossRef] [PubMed]
12. R. Halir, P. J. Bock, P. Cheben, A. Ortega-Moñux, C. Alonso-Ramos, J. H. Schmid, J. Lapointe, D.-X. Xu, J. G. Wangüemert-Pérez, Í. Molina-Fernández, and S. Janz, “Waveguide sub-wavelength structures: a review of principles and applications,” Laser Photonics Rev. 9(1), 25–49 (2014). [CrossRef]
13. B. E. A. Saleh and M. C. Teich, Fundamentals of Photonics, 2nd edition, (John Wiley and Sons, Inc., 2007) Ch. 7.
