Open Access
Add to BibSonomyAbstract
Highly integrated optical components are strongly demanded because they enable wavelength-division multiplexing optical communication systems to achieve smaller footprints, lower power consumption, and enhanced reliability. Variable optical attenuator (VOA) arrays are often used with optical switches in cascaded form for reconfigurable optical add-drop multiplexer systems. Although VOAs and optical switches based on polymer waveguide technology are commercially available, it is still not viable to integrate these two array devices on a single chip because of significant interchannel crosstalk. In this work, we resolved the issue of crosstalk and integrated the arrays of optical switch and VOA on a single chip by incorporating a self-assembled scattering monolayer (SASM). The SASM was effective for scattering the planar guided mode; consequently, the crosstalk into an adjacent channel was significantly reduced, to less than −35 dB.
© 2014 Optical Society of America
1. Introduction
Integrated optic devices have been deployed for various applications including optical communications, optical sensors, and optical signal processing. To increase the transmission capacity in wavelength-division multiplexing (WDM) optical communications, optical signal transmitters with highly integrated optical components have become indispensable [1–3]. Integrated optic devices have provided the core technology for optical sensors used in label-free biomolecule detection, electrical current monitoring at power plants, and resonator micro-optic gyros [4–6].
As the integration density of optical components increases, the control of stray light radiated out of the waveguide becomes important for preventing interchannel crosstalk. In waveguide devices, any light that leaks onto the cladding region is difficult to remove because it is confined in the thick planar cladding layer. To reduce the crosstalk signal, a vertically integrated structure in silicon-on-insulator waveguide devices and a self-assembled microsphere in polymeric waveguide devices have been investigated [7, 8].
Thermo-optically controlled polymer waveguide devices have shown significant advantages over silica devices owing to their large thermo-optic (TO) effect and strong thermal isolation [9]. Although polymer devices cannot yield nanoscale waveguide dimensions as InP and silicon waveguides, they still have the advantage of good mode matching with the optical fiber and smaller Fresnel reflection due to the refractive index similarity. Hence, variable optical attenuators (VOAs) and optical switches made of polymer waveguides have been widely deployed in WDM systems [10]. Moreover, the flexibility of polymer waveguides has been exploited for demonstrating unique devices such as strain-imposed tunable lasers and optical waveguide touch panels [11, 12]. Recently, a hybrid waveguide device with SiN core and polymer cladding was also reported to exhibit more efficient TO device [13].
Among various optical devices, VOAs and optical switches are essential components, especially for the reconfigurable optical add-drop multiplexer (ROADM), in which the two components are used in cascaded array form [14]. Hence, it is natural to try to integrate these two array devices into a single chip [15]. However, significant crosstalk was observed in the integrated version of the device because the light radiated from the switch array was crossing over into the adjacent VOA. When the two devices were prepared separately and packaged individually, the radiated light could be filtered out by the optical fibers connecting the two components, but it remained when the two devices were integrated on a single chip.
For the ROADM system to have strong competitiveness in the commercial market, an integrated version of switch-VOA device is eagerly anticipated. Hence, in this work, we demonstrate a switch-VOA integrated chip with significantly reduced crosstalk by incorporating a self-assembled scattering monolayer (SASM). The crosstalk into the adjacent channel is reduced to less than −35 dB, which is acceptable for commercial ROADM applications.
2. Operating principle
Fluorinated polymer materials were developed for low-loss optical waveguide devices useful for optical communication systems operating at IR wavelengths. Due to the reduction of the vibration overtone absorption and the increase in the material’s homogeneity, the polymer material achieved an absorption loss of less than 0.1 dB/cm at 1550 nm. Optical switches and variable optical attenuators are currently the most popular polymeric devices used in optical communication systems. The polymer waveguide optical switch is operating based on adiabatic mode coupling in an asymmetric Y-branch structure, and the switch provides digital-like operation, so precise control of the refractive index is not required [16]. The VOA used in this work operates by radiation mode coupling due to the refractive index perturbation introduced by the TO effect [17, 18]. The refractive index of polymer is reducing when the temperature is increased, which is mainly occurred by the volume expansion of the polymer material. The TO coefficient of polymer material is on the order of 1~3 x 10−4 /°C [19]. Because of the superior TO effect and excellent thermal isolation of polymers, polymeric optical waveguide devices exhibit a very low operating electrical power, which is important for producing highly integrated optical devices.
A cascaded array of polymeric switch and VOA devices could be realized as shown in Fig. 1(a). During ROADM operation, the original signal light should be blocked to pass the new added signal. At the 2 × 1 channel selection switch, the blocked light radiates out of the channel waveguide and spreads into the cladding layer resulting in a crosstalk. Moreover, in the VOA device, when the signal is attenuated, additional radiation is produced to increase the crosstalk. After the light radiates out of the channel waveguide, it still remains inside the planar waveguide polymer layers. Then it reaches the output fiber of adjacent channels, causing crosstalk. To prevent propagation of the radiated light, we introduce scattering particles in the middle of the device, as shown in Fig. 1(b), which are very effective for scattering and diffracting the radiated light propagating through the planar waveguide and removing the source of crosstalk. The crosstalk in a ROADM system increases as the signal is transmitted over many steps of add-drop multiplexing, so each ROADM module should provide crosstalk of no more than −35 dB for practical usage.
Fig. 1 Schematics of cascaded array, integrated optic device consisting of polymer waveguide switch and VOA: (a) reference device and (b) device incorporating SASM for crosstalk reduction.
To introduce the scattering particles into the device, we developed a self-assembly technique applicable to polymeric optical waveguide devices. The efficiency of the SASM for scattering of the planar guided mode was simulated using a finite-difference time-domain (FDTD) method (Omnisimm, Photon Design Co.). The planar waveguide consists of a core layer of a 32 μm thick polymer, an upper cladding of air, and a substrate of Si layer. In the simulation, the input light was launched at the center of the waveguide through a single-mode waveguide; then it was coupled into a multimode planar waveguide consisting of three layers: air, polymer, and a Si substrate. Because the polymer–Si interface produces frustrated total internal reflection, the power of the guided light decreased as it propagates. Hence, even for the reference structure with no SASM, the total guided power was decreasing along the propagation distance, as shown in Fig. 2(a). The SASM was located at 4.5 μm above the center line as appeared in the fabricated device. Initial diameter of particle was 2.0 μm, and it determines the period of scattering element in the simulation. The final particle size was controllable by plasma etching during fabrication, and it was varied from 1.0 to 2.0 μm in the simulation. As shown in Fig. 2(a), because of the enhanced scattering by the SASM, the total guided power decreased more rapidly than that in the reference device. By comparing the power remaining inside the multimode waveguide to that in the reference, the additional loss was calculated, as shown in Fig. 2(b). The scattering efficiency depends on the microsphere size, reaching a maximum of 18 dB/cm for a particle size of 1.6 μm. The refractive index of PS particle was 1.59 in the simulation, and the absorption coefficient of the particle was not included. However, the PS particle has a considerable absorption for the 1550 nm wavelength [20], so that the crosstalk signal will be further reduced by the absorption of PS particle.
Fig. 2 FDTD simulation results for SASM efficiency: (a) total power remaining inside the planar waveguide as a function of propagation distance for various scattering microsphere sizes and (b) additional scattering loss introduced by the SASM, which reaches a maximum of 18 dB/cm for 1.6-μm microspheres.
3. Fabrication of switch-VOA array device with SASM
To fabricate the device, we selected two low-loss fluorinated polymers (ZPU series) with refractive indices of 1.430 and 1.440 for the cladding and core layers, respectively. The fabrication procedure is outlined schematically in Fig. 3. The lower cladding layer was formed on a Si wafer by spin coating and UV curing of the cladding polymer to have a thickness of 18 μm. Then, a photoresist pattern was formed to cover a certain area where the waveguide core would be placed. To form the SASM, 10 wt% solution of suspended polystyrene microsphere (latex microsphere suspension, Thermal Scientific Co.) was spin coated over the ZPU polymer after an oxygen plasma surface treatment. By controlling the solution thickness and spin coating speed, a monolayer of microspheres could be obtained [21, 22]. However, aggregation of the microparticles often produced multiple stacks. Hence, to obtain a monolayer of self-assembled particles, it was necessary to rinse the coated surface by immersing the sample in a surfactant solution (2 wt% tween20 mixed in DI water) for 3 min. Oxygen plasma etching was used to adjust the microparticle size in order to yield the highest scattering efficiency as shown in the FDTD simulation. The original size of 2.0 μm was reduced to 1.6 μm after 3 min of plasma etching. Then, the photoresist was lifted off along with the particles remaining over it. A scanning electron microscopy (SEM) image of the etched SASM is shown in Fig. 4(a).
Fig. 4 (a) SEM image of the polystyrene microsphere SASM covered on the lower cladding layer and plasma etched to control the size of particle, and (b) microphotograph of fabricated waveguide structure illustrating the difference between the SASM-covered area and the clean surface.
Over the patterned scattering microspheres, another layer of lower cladding with a thickness of 4.8 μm was added. Then, a waveguide core pattern with a groove thickness of 2.8 μm was formed by oxygen plasma etching through a photoresist mask. The core polymer was coated to produce a film thickness of 4.2 μm; then it was etched 2.6 μm in oxygen plasma to obtain a final core thickness of 4.4 μm. The upper cladding layer was formed of the same polymer as the lower cladding with a thickness of 9.0 μm. The heating electrode was fabricated by thermal evaporation of Cr-Au with a thickness of 10-100 nm. The widths of the heating electrodes were 5 and 10 μm, and the resultant resistance was 2370 and 330 Ω for the switch and VOA, respectively. A microphotograph of the fabricated waveguide structure is shown in Fig. 4(b) to illustrate the difference between the SASM-covered area and the clean surface.
The uniformity of the SASM coating was excellent, and a densely packed SASM was produced over a 4-in. silicon substrate, as observed in Fig. 5(a). To compare the performance of the SASM and reference devices, the two devices were fabricated on a single substrate adjacent to each other, as shown in Fig. 5(b). The SASM device shows a hazy surface due to scattering. After the sample was polished at 8°, each of the devices was packaged by connecting V-groove fiber arrays. Among the many devices in the array, to facilitate the packaging, only three devices in the center area were wire bonded. The length of final device was 2.3 cm.
Fig. 5 (a) Photograph of switch-VOA chips on a wafer, exhibiting excellent uniformity of the SASM covering a 4-in. area, and (b) comparison of reference and SASM-covered devices fabricated adjacent to each other on a single wafer; SASM device exhibits a hazy surface.
Before the pigtail, we observed the output mode profile of the devices. Because the switch has no control signal, when the light source was connected to one of the two input ports of the 2 × 1 switch, half of the input light was radiated and coupled to the planar modes. Hence, at the output end of the device, a significant amount of planar guided light was observed, as shown in Fig. 6(a). In contrast, in the SASM device, the planar guided light was significantly reduced, as shown in Fig. 6(b), due to the scattering and an additional absorption of PS particle.
Fig. 6 Output mode images from (a) reference device and (b) SASM device; planar guided mode does not appear for the SASM device because it is suppressed.
4. Characteristics of the SASM incorporated switch-VOA device
The performance of SASM incorporated switch-VOA device was first observed and compared to that of the reference device. With a single 1550-nm distributed feedback laser (DFB) laser diode connected to one of the two input ports of the 2 × 1 switch, the output signal after the VOA was measured. Figure 7(a) and 7(b) show the transmission power as a function of the applied heating power on the switch and the VOA, respectively. In the switch, depending on the active electrode, one could select or block the input signal, and the transmission power increased or decreased, respectively. In the VOA, the output power was attenuated by more than 30 dB for the applied heating power of 40 mW. Regardless of the SASM, the two devices exhibited almost the same characteristics, which indicates the SASM does not affect the device characteristics.
Fig. 7 Comparison of basic characteristics of reference device and switch-VOA device with SASM: transmission power measured as function of applied heating power on (a) switch and (b) VOA.
To observe the reduction in crosstalk, we measured the crosstalk of the reference and SASM devices for three cases. First, with the switch set to pass the input light launched at the center channel, as shown in Fig. 8(a), the crosstalk into the upper and lower adjacent channels was measured during VOA heating operation. In the reference device, the initial crosstalk was −42 dB and −54 dB for the channels, and it increased by about 10 dB by the VOA operation to cause additional crosstalk. However, in the SASM device, the crosstalk was well below the measurement limit of −70 dB, and it barely appears in Fig. 8(a). In the second measurement, shown in Fig. 8(b), the crosstalk was measured during switch heating to block the input light (the VOA was off). Similar to the case during VOA operation, the SASM device exhibited a much lower crosstalk level, which increased slightly but remained less than −50 dB until the heating power of 100 mW. In the third case, the effect of adjacent VOA heating operation was observed with the switch set to block the signal, because we were concerned that the radiated light might couple into the adjacent channel waveguide through the index-perturbed region of the VOA. However, as shown in Fig. 8(c), the crosstalk did not increase even during adjacent VOA operation, which means that the index perturbation of the VOA does not contributes to the coupling between radiated and guided lights.
Fig. 8 Crosstalk measured from reference and SASM devices for three cases: (a) light was launched to the center channel, and crosstalk occurred in the upper and lower adjacent channels during VOA operation; crosstalk in the SASM device was lower than the detection limit, (b) with the switch heated to block the signal, stronger crosstalk was observed in the reference device, whereas the SASM device still had significantly lower crosstalk, and (c) while the switch blocked the input signal, the VOA of the adjacent channel was operated to confirm whether the radiated light was coupled into the VOA.
To simulate crosstalk in the WDM communication, two DFB light sources with wavelengths of 1550.6 and 1552.5 nm were connected to each channel, as shown in Fig. 9(a). In this setup, the λ2 signal would pass into the output of the upper channel along with the λ1 signal as the crosstalk. A comparison of the output spectra measured from the reference and SASM devices, as shown in Fig. 9(b) and 9(c), respectively, reveals that the crosstalk into the upper channel was reduced by more than 20 dB in the SASM device. The crosstalk toward the lower channel was also measured, using the configuration shown in Fig. 9(d). In this case, the crosstalk was more significant than that for the upper channel, as shown in Fig. 9(e). The reason is that when the signal blocking occurred at the lower branch of the 2 × 1 switch, the light radiates more strongly toward the bottom. The crosstalk reduction in the SASM device in this case was 16 dB, and the resultant crosstalk was −36 dB, as shown in Fig. 9(f); this is acceptable for practical ROADM applications.
Fig. 9 Crosstalk measured using two lasers with different wavelengths: input launched and the crosstalk toward upper channel to be monitored as (a), and the output spectra for the reference (b) and the SASM device (c); input launched and the crosstalk toward lower channel to be monitored as (d), and the output spectra for the reference (e) and the SASM device (f). The crosstalk was improved by 16 - 20 dB in the SASM device, and the resultant crosstalk was less than −36.5 dB.
5. Conclusion
To resolve the problem of crosstalk in switch-VOA integrated array devices based on polymer waveguide technology, we employed a scattering monolayer formed by self-assembled monolayer technique. The crosstalk was caused mainly by the light radiation at the 2 × 1 channel selection switch, which propagated toward adjacent channels through the planar waveguide structure. By using the self-assembly and lift-off technique, densely packed microspheres were formed on a certain area of the device to scatter the propagating radiated light. Simple spin coating of the particle dispersed solution enables the formation of an SASM, which covers the 4-in. substrate with excellent uniformity. With no changes in the basic characteristics, the SASM exhibited significant reduction of the adjacent channel crosstalk by as much as 16 – 20 dB; the resultant crosstalk was −36 to −50 dB, which would be acceptable for practical ROADM applications. Because we found the mechanism of crosstalk increase, further reduction would be possible by increasing the scattering efficiency of the SASM and modifying the propagation direction of radiated light.
Acknowledgments
This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MEST) (2012-001697), and the ATC project of ChemOptics funded by the Ministry of Knowledge Economy, Korea.
References and links
1. F. A. Kish, D. Welch, R. Nagarajan, J. L. Pleumeekers, V. Lal, M. Ziari, A. Nilsson, M. Kato, S. Murthy, P. Evans, S. W. Corzine, M. Mitchell, P. Samra, M. Missey, S. DeMars, R. P. Schneider, M. S. Reffle, T. Butrie, J. T. Rahn, M. Van Leeuwen, J. W. Stewart, D. J. Lambert, R. C. Muthiah, H. Tsai, J. S. Bostak, A. Dentai, K. Wu, H. Sun, D. J. Pavinski, J. Zhang, J. Tang, J. McNicol, M. Kuntz, V. Dominic, B. D. Taylor, R. A. Salvatore, M. Fisher, A. Spannagel, E. Strzelecka, P. Studenkov, M. Raburn, W. Williams, D. Christini, K. K. Thomson, S. S. Agashe, R. Malendevich, G. Goldfarb, S. Melle, C. Joyner, M. Kaufman, and S. G. Grubb, “Current status of large-scale InP photonic integrated circuits,” IEEE J. Sel. Top. Quantum Electron. 17(6), 1470–1489 (2011). [CrossRef]
2. L. Schares, J. A. Kash, F. E. Doany, C. L. Schow, C. Schuster, D. M. Kuchta, P. K. Pepeljugoski, J. M. Trewhella, C. W. Baks, R. A. John, L. Shan, Y. H. Kwark, R. A. Budd, P. Chiniwalla, F. R. Libsch, J. Rosner, C. K. Tsang, C. S. Patel, J. D. Schaub, R. Dangel, F. Horst, B. J. Offrein, D. Kucharski, D. Guckenberger, S. Hegde, H. Nyikal, C.-K. Lin, A. Tandon, G. R. Trott, M. Nystrom, D. P. Bour, M. R. T. Tan, and D. W. Dolfi, “Terabus: Terabit/second-class card-level optical interconnect technologies,” IEEE J. Sel. Top. Quantum Electron. 12(5), 1032–1044 (2006). [CrossRef]
3. 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]
4. X. Fan, I. M. White, S. I. Shopova, H. Zhu, J. D. Suter, and Y. Sun, “Sensitive optical biosensors for unlabeled targets: A review,” Anal. Chim. Acta 620(1-2), 8–26 (2008). [CrossRef] [PubMed]
5. M.-C. Oh, W.-S. Chu, K.-J. Kim, and J.-W. Kim, “Polymer waveguide integrated-optic current transducers,” Opt. Express 19(10), 9392–9400 (2011). [CrossRef] [PubMed]
6. H. Mao, H. Ma, and Z. Jin, “Polarization maintaining silica waveguide resonator optic gyro using double phase modulation technique,” Opt. Express 19(5), 4632–4643 (2011). [CrossRef] [PubMed]
7. A. M. Jones, C. T. DeRose, A. L. Lentine, D. C. Trotter, A. L. Starbuck, and R. A. Norwood, “Ultra-low crosstalk, CMOS compatible waveguide crossings for densely integrated photonic interconnection networks,” Opt. Express 21(10), 12002–12013 (2013). [CrossRef] [PubMed]
8. J.-W. Kim, N.-S. Son, J.-H. Jang, K.-J. Kim, and M.-C. Oh, “Ultra-low inter-channel crosstalk in array waveguide device incorporating self-assembled microsphere diffraction layer,” Opt. Express 19(21), 20904–20909 (2011). [CrossRef] [PubMed]
9. Y.-O. Noh, H.-J. Lee, J. J. Ju, M. S. Kim, S. H. Oh, and M.-C. Oh, “Continuously tunable compact lasers based on thermo-optic polymer waveguides with Bragg gratings,” Opt. Express 16(22), 18194–18201 (2008). [CrossRef] [PubMed]
10. M.-C. Oh, K.-J. Kim, W.-S. Chu, J.-W. Kim, J.-K. Seo, Y.-O. Noh, and H.-J. Lee, “Integrated photonic devices incorporating low-loss fluorinated polymer materials,” Polymers 3(4), 975–997 (2011). [CrossRef]
11. K.-J. Kim, J.-W. Kim, M.-C. Oh, Y.-O. Noh, and H.-J. Lee, “Flexible polymer waveguide tunable lasers,” Opt. Express 18(8), 8392–8399 (2010). [CrossRef] [PubMed]
12. B.-J. Cheon, J.-W. Kim, and M.-C. Oh, “Plastic optical touch panels for large-scale flexible display,” Opt. Express 21(4), 4734–4739 (2013). [CrossRef] [PubMed]
13. Z. Zhang, D. Liu, D. Felipe, A. Liu, N. Keil, and N. Grote, “Polymer embedded silicon nitride thermally tunable Bragg grating filters,” Appl. Phys. Lett. 102(18), 181105 (2013). [CrossRef]
14. S. Tibuleac and M. Filer, “Transmission impairments in DWDM networks with reconfigurable optical add-drop multiplexers,” J. Lightwave Technol. 28(4), 557–598 (2010). [CrossRef]
15. J.-U. Shin, Y.-T. Han, S.-P. Han, S.-H. Park, Y.-S. Baek, Y.-O. Noh, and K.-H. Park, “Reconfigurable optical add-drop multiplexer using a polymer integrated photonic lightwave circuit,” ETRI J. 31(6), 770–777 (2009). [CrossRef]
16. Y.-O. Noh, H.-J. Lee, Y.-H. Won, and M.-C. Oh, “Polymer waveguide thermo-optic switches with −70 dB optical crosstalk,” Opt. Commun. 258(1), 18–22 (2006). [CrossRef]
17. Y.-O. Noh, C.-H. Lee, J.-M. Kim, W.-Y. Hwang, M.-C. Oh, Y.-H. Won, H.-J. Lee, and S.-G. Han, “Polymer waveguide variable optical attenuator and its reliability,” Opt. Commun. 242(4-6), 533–540 (2004). [CrossRef]
18. Y.-T. Han, J.-U. Shin, S.-H. Park, S.-P. Han, C.-H. Lee, Y.-O. Noh, H.-J. Lee, and Y. Baek, “10 channel polymer variable optical attenuator array for power monitoring and equalization in integrated PLC ROADM module,” in Proceedings of IEEE Conference on Photonics in Switching, 2008, Sapporo, (2008). [CrossRef]
19. Z. Zhang, P. Zhao, P. Lin, and F. Sun, “Thermo-optic coefficients of polymers for optical waveguide applications,” Polymers 47(14), 4893–4896 (2006). [CrossRef]
20. X. Ma, J. Q. Lu, R. S. Brock, K. M. Jacobs, P. Yang, and X. H. Hu, “Determination of complex refractive index of polystyrene microspheres from 370 to 1610 nm,” Phys. Med. Biol. 48(24), 4165–4172 (2003). [CrossRef] [PubMed]
21. C. L. Cheung, R. J. Nikolic, C. E. Reinhardt, and T. F. Wang, “Fabrication of nanopillars by nanosphere lithography,” Nanotechnology 17(5), 1339–1343 (2006). [CrossRef]
22. Y. Koide, K. Fujisawa, and M. Nakane, “Preparation of non-contact ordered array of polystyrene colloidal particles by using a metallic thin film of fused hemispheres,” Colloids Surf. A Physicochem. Eng. Asp. 330(2-3), 108–111 (2008). [CrossRef]
