Open Access
Add to BibSonomyAbstract
A reconfigurable two-mode mux/demux device in planar waveguide was proposed. The simulated mux/demux extinction ratio was ≥ 35 dB with ≤ 0.4 dB excess loss. The device was realized in polymer materials using the thermo-optic effect. It was characterized via a tunable laser source at 1550 nm. Its mux/demux performance in both routes was demonstrated and compared with the theoretical prediction. The device is easy to implement and has applications in future multimode optical communication systems to further extend transmission capacity.
© 2014 Optical Society of America
1. Introduction
Given the rapid expansion and continuous growth of bandwidth usage, dense wavelength division multiplexing (DWDM) technology with single-mode fiber may not be able to cope with future demand. Space division multiplexing that involves the use of multicore fibers is one of the techniques used to extend transmission capacity [1, 2]. In contrast, mode division multiplexing has recently attracted attention as an alternative technique for expanding transmission capacity [3–18]. This technique uses modes in multimode fiber or waveguide to carry sets of DWDM and modulated signals. The transmission capacity expanded through this technique hence exceeds that for a single-mode waveguide.
Mode mux/demux (MMD) is the key device for handling mode multiplexing and demultiplexing processes. Different types of MMD devices are available. In the optical fiber platform, couplers based on long period grating are often employed as MMD devices [3–5]. A well-designed fiber core structure can also be used to perform multiplexing and demultiplexing [6,7]. In the bulk-optic platform, modes are multiplexed and demultiplexed by lenses with phase masks or filters [8–10], but devices that use the aforementioned platforms are often relatively large. Such devices are also not rugged and incompatible with optical integrated circuits. A planar lightwave circuit (PLC) is the third alternative platform to realize compact MMD devices that can be easily integrated into other PLC devices to achieve high packing device density. PLC-based MMD devices that use a T-shape coupler [11], a multimode interferometer [12] and asymmetric coupler [13], an asymmetric Y-junction [14], a tapered coupler [15], and a hollow tapered waveguide [16] have been proposed recently.
However, most of the reported MMDs are passive in nature. System designers can thus only achieve flexibility if the MMD channels are dynamically reconfigurable. In this study, we propose a reconfigurable two-mode mux/demux (RMMD) device in polymer materials that offer numerous advantages, including structural flexibility, easy processing, and mass fabrication capability. The two routes carried by the fundamental mode and the first-order mode of the device can be swapped through the thermo-optic effect of the polymer. We believe that this study is the first work to report an RMMD device in planar waveguide applied in MMD. The properties of the device are simulated using commercially available a 3D Beam Propagation Method (BPM) simulation tool, and its reconfiguration performance is demonstrated experimentally.
2. Operation principle
Figure 1 shows the schematic diagram of the RMMD, which consists of an unbalanced Mach-Zehnder interferometer (MZI) cascaded with a two-mode interferometer (TMI). One of the ports (Port A) of the MZI has a two-mode region. The MZI has a path difference of ∆L between its unbalanced arms, and the TMI has an interactive length Z over its two-mode region where a heater is placed on top for switching. Both two-mode regions of the RMMD are designed to support only the first two propagating modes, whereas all other parts of the waveguide (i.e., the single-mode region) are designed to support only the fundamental mode.
The overall transfer function [17, 18] of the RMMD offers for the fundamental and first-order modes launching without significant mode coupling loss as:
where AP1 is the amplitude of e-field at port 1 (P1), AP2 is the amplitude of e-field at port 2 (P2), APA is the amplitude of e-field at port A, ∆β is the difference in propagation constants between the two propagation modes in the two-mode interactive region, and β is the propagation constant of a single-mode waveguide at the arms of the MZI. “±” signs in (1) representing the phase difference of two fields entering MZI. “+” sign is used for fundamental mode launching as it gives two in-phase fields. In contrast, first order mode launching gives two out-of-phase fields and “-” sign is usedAssuming APA is equal to 1, the output powers at P1 and P2 can be derived from Eq. (1) as follows:
where PP1, fun is the output power at P1 when launching the fundamental mode, PP2, fun is the output power at P2 when launching the fundamental mode, PP1, fir is the output power at P1 when launching the first order mode, PP2, fir is the output power at P2 when launching the first order mode, and θ is β∆L/2.The output powers at P1 and P2 are shown to be dependent on the phase differences contributed by MZI and TMI in Eqs. (2) and (3). For the appropriate path difference in MZI and the interactive length of TMI, i.e. ∆βZ/2 becomes π/4 and θ becomes π/4, light launched at Port A in the fundamental mode is then retrieved at P1 and light launched at Port A in the first order mode is retrieved at P2. The demultiplexing function is realized as the superposition of the e-field of two different modes. Given that the light path is reciprocal, the multiplexing of the fundamental and first-order modes is achieved when the light is launched simultaneously at ports 1 and 2.
If ∆βZ/2 is changed to 3π/4, the launching light in the fundamental mode will then be retrieved at P2, whereas the launching light in the first order mode will be retrieved at P1. This condition can be achieved by heating up the interactive region of the TMI base on the thermo-optic effect. The mux/demux functions are still valid because the superposition and the reciprocal properties of the two modes still hold. The RMMD performance will be further discussed in sections 3 and 4.
3. Design and simulation
Figure 2 shows the cross-sectional view of a two-mode region and a single-mode region, which employ an embedded waveguide structure supported by a silicon substrate. The core material is Dow Chemical benzocyclobutene (BCB), which has a refractive index of 1.537 at 1.55 μm. The cladding material is ZP51 (ChemOptics Inc., Daejeon, Republic of Korea), which has a refractive index of 1.512 at 1.55 μm. Considering the material index, the dimensions of the two-mode region (W1 × T) and the single-mode region (W2 × T) are set at 6 μm × 2.5 μm and 3 μm × 2.5 μm, respectively. They were designed to ensure that only the first order and the fundamental modes in the two-mode waveguide regions, and only a single mode in the other region of the device will exist within the C + L band. The half-branching angle of all the Y-junctions was designed as 1°. With this half-branching angle, the simulated losses for the fundamental and first-order modes around the transition region of the Y-junction were less than 0.07 dB. Such losses should have an insignificant effect on the RMMD performance. The interactive length Z of the TMI is set at 3516 µm, and the path difference ∆L of the MZI is set at 7.2 µm for the best operation in transverse electric (TE) mode at 1.55 μm. To demonstrate the function [17, 18] of the RMMD, therefore, TE mode operated at 1.55μm is considered first in the simulation.
Figure 3 shows the mux function of the device, which is simulated using the 3D full vector BPM of RSoft Photonic CAD suite, before (Figs. 3(a)–3(b)) and after (Figs. 3(c)–3(d)) reconfiguration through the heater on the TMI. Based on given thermo-optics coefficient of the cladding material ZP51, we assumed its refractive index change per °C be −8x10−5. The heating effect in BeamProp, then, was simulated by reducing the cladding refractive index of TMI by 0.00008 for every °C increment.
Fig. 3 Simulation of the basic mux function of the RMMD at 1.55 μm (TE mode) when light is launched at (a) P1 and (b) P2 when the heater is turned off, or at (c) P1 and (d) P2 when the heater is turned on.
As shown in Figs. 3(a) and 3(b), a fundamental mode or a first-order mode output obtained at Port A depends on whether the light is launched at P1 or P2, respectively. This condition demonstrates the basic mux function of the RMMD. Figures 3(a) and 3(c) shows that the mode at Port A changes from the fundamental mode to the first-order mode for the same input mode launched at P1 when the heater on the TMI is turned on. This finding demonstrates the reconfiguration function of the RMMD. A similar situation can be observed when a signal is launched at P2, as shown in Figs. 3(b) and 3(d).
Figure 4 demonstrates the superposition properties of the fundamental and first order modes in the RMMD when the heater is either turned ON or OFF. When two independent signals at a given power ratio are launched simultaneously at P1 and P2, a fundamental mode and a first-order mode output with the same power ratio can be obtained at Port A.
Fig. 4 Simulation of the mux functions by simultaneously launching power into P1 and P2 at an arbitrary power ratio.
Figure 5 shows the demux function of the device before (Figs. 5(a)–5(b)) and after (Figs. 5(c)–5(d)) the heater is turned ON. The demux function of the device is achieved simply by launching the light at Port A instead of at P1 or P2. When a fundamental mode (Fig. 5(a)) or a first-order mode (Fig. 5(b)) is launched at Port A, a single-mode output is obtained at P1 or P2, respectively. For the demux case, the mode output is switched from P1 to P2 (or vice versa) for the same fundamental mode (or first-order mode) launched at Port A when the heater is turned on (Figs. 5(a), 5(c) and Figs. 5(b), 5(d)).
Fig. 5 Simulation of the basic demux function of the RMMD at 1.55 μm (TE mode) when light is launched at port A in the (a) fundamental mode and (b) first-order mode, when the heater is turned off, and in the (c) fundamental mode and (d) first-order mode, when the heater is turned on.
As shown in Fig. 6, the superposition also applies when two modes are launched at Port A simultaneously, fully demonstrating the demux function of the RMMD. The minimum extinction ratio and device excess loss for all the above cases are 35 dB and 0.4 dB, respectively. Yet, they are relatively thermal sensitive and dropped to around 18-20dB when the reconfiguration temperature had ± 2 °C deviation.
Fig. 6 Simulation of the demux functions by simultaneously exciting the fundamental and the first order modes at an arbitrary power ratio.
For RMMD operated in transverse magnetic (TM) mode at 1.55 μm, however, the simulated demultiplexing and multiplexing extinction ratio dropped by 15-20 dB. This shows a relatively high polarization dependence of our design for RMMD.
4. Fabrication and experimental results
The proposed device was fabricated in the following procedure (Fig. 7). A layer of 4-μm thick ZP51 and 2.5-μm thick BCB core were spin-coated and cured on top of the silicon one by one. The RMMD pattern was transferred to the BCB layer using photolithography followed by reactive ion etching techniques to remove the unwanted BCB materials. Finally, another layer of 8-μm thick Z51 was spin-coated on the BCB layer to form the upper cladding. The RMMD fabrication was performed after placing a micro-heater over the two-mode region of the TMI. The device was then characterized using an end-fired setup with a TE linear polarized laser source. The output mode patterns of the device were captured by an infrared camera. The total insertion loss of the device was 10 dB, which included 7 dB (~3.5 dB per facet) of coupling loss at both ends of the device. It was deduced from the measured output power when fundamental mode was launched at Port 1, Port 2 or Port A. The device’s excess loss was approximately 3 dB. This figure is higher than the simulated value because the material loss (~1.5 dB/cm) was not included in our simulation. The excess loss without considering the material loss was hence close to our simulated result.
Fig. 7 Overview of the fabrication process that consists of (a) the formation of BCB and ZP51 as cores and lower cladding layers by spin coating, (b) deposition of a Cr layer by RF sputtering, (c) patterning of photo resist by photolithography, (d) patterning of a Cr layer by wet etching, (e) patterning of BCB by reaction ion etching, (f) removal of Cr and photo resistance by wet etching, (g) formation of ZP51 as an upper cladding layer by spin coating, and (h) placing a heater on top of the waveguide.
Figures 8(a) and 8(b) show the output mode field patterns corresponding to the launching and operation conditions described in Figs. 3(a) and 3(b), respectively. A fundamental mode pattern and a first-mode output pattern could be observed at Port A when the light source was launched at P1 and P2, respectively. This result confirms the mux function of the RMMD. However, the extinction ratio could not be obtained experimentally because of the unavailability of a tool that can separately measure the fundamental mode power and the first-mode power directly from Port A. Connecting Port A to a mode sorter is a possible mechanism to separate the fundamental and the first order modes for individual mode power measurement. Comparing Figs. 8(a) and 8(c), the mode at Port A changed from the fundamental mode to the first-order mode (while keeping the same input mode launched at P1) when the heater on the TMI was turned on. This result experimentally demonstrates the mode switching function of the RMMD. A similar situation was observed when a signal was launched at P2 (Figs. 8(b) and 8(d)).
Fig. 8 Captured output mode profile when light is launched at (a) P1 and (b) P2 when the heater is turned off, and at (c) P1 and (d) P2 when the heater is turned ON in the mux process; when light is launched at (e) Port A with the fundamental mode, and at (f) Port A with the first-order mode when the heater is turned OFF; and when light is launched at (g) Port A with the fundamental mode, and at (h) Port A with the first-order mode when the heater is turned ON in the demux process.
We need to excite a pure fundamental mode and a pure first-order mode at Port A to experimentally demonstrate the demux function. In this work, we attempted to excite a fundamental mode at Port A by launching the light exactly at the center of the input waveguide. Figure 8(e) shows that a light spot was observed at P1, whereas light was not seen at P2. The extinction ratio measured by optical spectrum analyzer was 25 dB. To excite a first-order mode at Port A, the light source was launched at an offset from the waveguide center in order. It is because Port A was designed to support only the fundamental mode (symmetric mode) and the first order mode (asymmetric mode), if the light source is launched at an offset from the waveguide center, asymmetric mode, i.e. the first-order mode, should be primarily excited at Port A. However, a small amount of fundamental mode power was still unavoidably being excited in practice. As shown in Fig. 8(f), strong and weak light spots are observed at P2 and P1, respectively. The measured extinction ratio is approximately 10 dB in our case. If we can excite a purer first-order mode, we expect that the discrepancy between the theoretical prediction and the measured extinction ratio should be closer.
Comparing Figs. 8(e) and 8(g), the mode output at P1 and P2 was interchanged (while keeping the same launching condition at Port A) when the heater was turned ON. This result experimentally demonstrates the mode switching function of the RMMD. A similar phenomenon (Figs. 8(b) and 8(d)) was observed when a first-order mode signal was launched at Port A and the heater was turned ON. Given that the mode launched at Port A is not a pure first-order mode, the relative spot intensity between P1 and P2 remains the same. This indicates that the RMMD follows the superposition property predicted in Section 3.
Figure 9 shows the experimental de-multiplexing extinction ratio over the wavelength range of 1.546 μm to 1.554 μm. As expected from our design, the RMMD operated best at 1.55 μm with the measured extinction ratio of 25 dB. The acceptable extinction ratio of over 20 dB can also be obtained within ± 2.5 nm of the best operating wavelength. The wavelength dependence of the RMMD was observed because RMMD is operated by phase that is sensitive to the operating wavelength. A drop of ~1.25 dB/nm was observed when the operating wavelength shifted away from 1.55 μm. Minimizing the wavelength dependence of the RMMD is an interesting topic in future studies.
Fig. 9 Measured de-multiplexing extinction ratio varies with the wavelength when the fundamental mode was excited at Port A in TE polarization.
5. Conclusion
We proposed an RMMD for use in MMD. The mux/demux performance of the device was demonstrated through simulation and experiment. The simulated mux/demux extinction ratio was ≥ 35 dB with a ≤ 0.4 dB excess loss. The device was fabricated using polymer materials and characterized via a tunable laser source at 1550 nm. The experimental performance of the device closely agreed with the theoretical prediction. The mux/demux and reconfiguration function of the device could be clearly observed in the experiment. The measured demux extinction ratios for the fundamental mode in TE before and after reconfiguration were 25 dB and 24 dB, respectively. However, the measured demux extinction ratio for the first-order mode was significantly lower than the expected value because of the lack of a suitable signal source to generate a pure first order mode for testing. This device can be applied in mode-multiplexing systems to enhance the transmission capacity of future optical communication systems.
Acknowledgments
This work was supported by SRG NO. 7004052 of the City University of Hong Kong.
References and links
1. J. Sakaguchi, B. J. Puttnam, W. Klaus, Y. Awaji, N. Wada, A. Kanno, T. Kawanishi, K. Imamura, H. Inaba, K. Mukasa, R. Sugizaki, T. Kobayashi, and M. Watanabe, “305 Tb/s Space Division Multiplexed Transmission Using Homogeneous 19-Core Fiber,” J. Lightwave Technol. 31(4), 554–562 (2013). [CrossRef]
2. J. Sakaguchi, Y. Awaji, N. Wada, A. Kanno, T. Kawanishi, T. Hayashi, T. Taru, T. Kobayashi, and M. Watanabe, “Space DivisionMultiplexed Transmission of 109-Tb/s Data Signals Using Homogeneous Seven-Core Fiber,” J. Lightwave Technol. 30(4), 658–665 (2012). [CrossRef]
3. N. Hanzawa, K. Saitoh, T. Sakamoto, T. Matsui, S. Tomita, and M. Koshiba, “Demonstration of mode-division multiplexing transmission over 10 km two-mode fiber with mode coupler,” in OSA/OFC/NFOEC 2011 (Optical Society of America 2011), paper OWA4.
4. A. Li, A. Al Amin, X. Chen, and W. Shieh, “Transmission of 107-Gb/s mode and polarization multiplexed CO-OFDM signal over a two-mode fiber,” Opt. Express 19(9), 8808–8814 (2011). [CrossRef] [PubMed]
5. A. Al Amin, A. Li, S. Chen, X. Chen, G. Gao, and W. Shieh, “Dual-LP11 mode 4×4 MIMO-OFDM transmission over a two-mode fiber,” Opt. Express 19(17), 16672–16679 (2011). [CrossRef] [PubMed]
6. F. Saitoh, K. Saitoh, and M. Koshiba, “A design method of a fiber-based mode multi/demultiplexer for mode-division multiplexing,” Opt. Express 18(5), 4709–4716 (2010). [CrossRef] [PubMed]
7. J. Xu, C. Peucheret, J. K. Lyngsø, and L. Leick, “Two-mode multiplexing at 2 × 10.7 Gbps over a 7-cell hollow-core photonic bandgap fiber,” Opt. Express 20(11), 12449–12456 (2012). [CrossRef] [PubMed]
8. R. Ryf, S. Randel, A. H. Gnauck, C. Bolle, A. Sierra, S. Mumtaz, M. Esmaeelpour, E. C. Burrows, R. J. Essiambre, P. J. Winzer, D. W. Peckham, A. H. McCurdy, and R. Lingle, “Mode-Division Multiplexing Over 96 km of Few-Mode Fiber Using Coherent 6x6 MIMO Processing,” J. Lightwave Technol. 30(4), 521–531 (2012).
9. M. Salsi, C. Koebele, D. Sperti, P. Tran, H. Mardoyan, P. Brindel, S. Bigo, A. Boutin, F. Verluise, P. Sillard, M. Bigot-Astruc, L. Provost, and G. Charlet, “Mode-Division Multiplexing of 2 100 Gb/s Channels Using an LCOS-Based Spatial Modulator,” J. Lightwave Technol. 30(4), 618–623 (2012). [CrossRef]
10. X. Chen, A. Li, J. Ye, A. Al Amin, and W. Shieh, “Demonstration of Few-Mode Compatible Optical Add/Drop Multiplexer for Mode-Division Multiplexed Superchannel,” J. Lightwave Technol. 30(4), 641–647 (2012).
11. H. Kubota, H. Takara, and T. Morioka, “T-shaped mode coupler for two-mode mode division multiplexing,” IEICE Electron. Express 8(22), 1927–1932 (2011). [CrossRef]
12. T. Uematsu, Y. Ishizaka, Y. Kawaguchi, K. Saitoh, and M. Koshiba, “Design of a Compact Two-Mode Multi/Demultiplexer Consisting of Multimode Interference Waveguides and a Wavelength-Insensitive Phase Shifter for Mode-Division Multiplexing Transmission,” J. Lightwave Technol. 30(15), 2421–2426 (2012). [CrossRef]
13. S. Bagheri and W. M. J. Green, “Silicon-on-Insulator Mode-Selective Add-Drop Unit for On-Chip Mode-Division Multiplexing,” in Proceedings of IEEE 6th International Conference on Group IV Photonics (Institute of Electrical and Electronics Engineers, 2009), pp. 166–168. [CrossRef]
14. J. B. Driscoll, R. R. Grote, B. Souhan, J. I. Dadap, M. Lu, and R. M. Osgood Jr., “Asymmetric Y junctions in silicon waveguides for on-chip mode-division multiplexing,” Opt. Lett. 38(11), 1854–1856 (2013). [CrossRef] [PubMed]
15. Y. Ding, J. Xu, F. Da Ros, B. Huang, H. Ou, and C. Peucheret, “On-chip two-mode division multiplexing using tapered directional coupler-based mode multiplexer and demultiplexer,” Opt. Express 21(8), 10376–10382 (2013). [CrossRef] [PubMed]
16. H. Dalir, Y. Yokota, and F. Koyama, “Spatial mode multiplexer/demultiplexer based on tapered hollow waveguide,” IEICE Electron. Express 8(9), 684–688 (2011). [CrossRef]
17. R. M. de Ridder and C. G. H. Roeloffzen, Wavelength Filters for Fibre Optics, (Springer, 2006), Chap. 9.
18. J. D. Love and N. Riesen, “Single-, Few-, and Multimode Y-Junctions,” J. Lightwave Technol. 30(3), 304–309 (2012). [CrossRef]
