Open Access
Add to BibSonomy
Abstract
We propose a kind of wavelength division multiplexer (WDM) based on asymmetric directional couplers (ADCs). Wavelength signals are multiplexed into a bus waveguide by converting to different modes using the cascaded ADCs. We design the WDMs with conventional and tapered ADCs and analyze the performance in terms of insertion loss, crosstalk, and fabrication tolerance. Experimental results indicate that the WDM with tapered ADCs has both lower insertion loss (less than 1dB) and crosstalk (less than -23dB), and at the same time it has better fabrication tolerance than conventional ADCs. The ADC-based WDMs, which demonstrate good performance of multiplexing wavelengths, own the advantages of simple structure, high flexibility, and scalability. Since different wavelengths perform mode conversion on the same plane, the proposed WDM can be conveniently integrated on-chip with active and passive devices. Our approach combines both wavelength and mode dimensions, which is different from previous schemes, so it is expected to provide an alternative to developing a promising WDM in integrated silicon-based circuits.
© 2018 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
The technology of wavelength division multiplexing plays a crucial role in increasing the communication capacity of optical interconnects. For the system of multi-channel multiplexed optical interconnect, the on-chip integrated multiplexer is one of the key components. Over the last decade, various kinds of wavelength division multiplexers (WDMs) have been reported, such as arrayed waveguide grating (AWG) [1–5], etched diffraction grating (EDG) [6,7], micro-ring resonator (MRR) [8,9], Mach-Zehnder interferometer (MZI) [10–12], multi-mode interference (MMI) [13–18], coupled vertical grating [19,20] and innovative WDM using inverse design algorithm [21].
AWG and EDG are conventional planar waveguide multiplexers that allow multiple channels to be placed in parallel; therefore, they are widely applied in current wavelength division multiplexing systems. However, these two types of WDMs cannot perform their merits when multiplexing few wavelength channels, owing to their complicated design, large size, and inconvenient integration with other devices. Moreover, it is difficult for them to achieve extremely low insertion loss. MRR is a common component for implementing passive and active devices [22–24], due to its simple structure and flexible scalability. Cascaded MRRs can be designed as a (de)multiplexer for multiple wavelength channels, but in consideration of fabrication tolerance, there is still upside potential in obtaining uniform channel spacing. MZI-based WDM excels in spectral flatness compared with the above multiplexers, but it has a large footprint. Meanwhile, we need more efforts to decrease crosstalk [11] and insertion loss [12]. The MMI, which makes use of the self-imaging principle, has become one of the most common ways of achieving wavelength division multiplexing due to easy production and a large fabrication tolerance. In addition to silicon nanowire waveguide [13–15], MMI with slot waveguide structure has also been proposed [16–18]. Compared with silicon nanowire waveguide, the unique structure of slot waveguide offers some benefits, such as a small beat length and a strong restriction within the slot area which gives rise to low loss. However, for MMI-based WDM, the large device size and insertion loss are still the issues to be solved. WDMs with coupled vertical gratings have good performance in terms of device size and insertion loss, but it would be better if further work were made to reduce the channel crosstalk. The WDM designed with an inverse design algorithm has ultra-small dimension with an area of only 2.8µm×2.8µm [21], but its insertion loss is not ideal. In short, although all of the aforementioned WDMs manifest their advantages, they also have some drawbacks, for instance, large footprints, high insertion loss, or large crosstalk.
In recent years, asymmetric directional couplers (ADCs) have attracted much attention because of their simple structure, scalability, and flexibility [25–27]. In addition, tapered ADCs manifest better fabrication tolerance than conventional ADCs [28,29]. Therefore, this paper demonstrates two types of design schemes which are conventional ADCs and tapered ADCs to gain the four-channel silicon-based wavelength division multiplexer. Here, the proposed WDMs employ the principle of directional coupling to achieve the ability of multiplexing different wavelength signals. The four wavelength signals are converted into different modes (TE0, TE1, TE2, TE3) and multiplexed into a bus waveguide by cascaded ADCs. Since the multiplexed wavelength signals are transformed to different modes that are orthogonal in the bus waveguide by ADCs, the wavelength division multiplexer designed by exploiting the mode conversion method not only retains the advantages of simplicity and miniaturization of the ADC, but also has small crosstalk between different wavelength channels. We compare these two types of WDMs by the means of 3D finite-difference time domain (FDTD) simulations and experimental verification. The experimental results show that the WDM made of tapered ADCs has lower insertion loss, lower crosstalk and larger manufacturing tolerance in comparison with the WDM composed of conventional ADCs. We believe that it is feasible to offer an alternative method by exploiting our scheme to achieve the technology of wavelength division multiplexing in integrated silicon-based circuits.
2. Principle and device design
The WDM consisting of three conventional ADCs is illustrated in Fig. 1
Fig. 1 (a) Schematic structure of the proposed WDM based on conventional ADCs. (b) Top view of the wavelength λ2 channel.
In our proposed schemes, the height of the silicon waveguide is taken as 220 nm, and the width of the input single-mode waveguide is set as 400 nm. The gap between the two waveguides of each ADC is fixed as 200 nm in view of the practical fabrication tolerance. In order to obtain high coupling efficiency as much as possible, in particular, we design and optimize an S-bend structure at the input port of each ADC. In the optimizing scheme, as shown in Fig. 2(a)
Fig. 2 (a) Schematic structure of S-bend waveguide. (b) The transmission of S-bend waveguide as a function of the length Lx.
The width W1, W2, and W3 of each ADC should be accurately determined to obtain high coupling efficiency based on the phase matching condition. For example, when the input wavelength signal is λ2, it is converted into TE1 mode in the bus waveguide. Therefore, we change the waveguide width W1, while calculating the effective refractive index of the fundamental mode TE0 in the input waveguide and that of TE1 mode in the bus waveguide, which are presented in Fig. 3
Fig. 3 The effective refractive indices of TE0 mode and TE1 mode vary with the waveguide width at a wavelength of 1549.2 nm.
Fig. 4 (a) The coupling efficiency varies with the width W1 of the bus waveguide. (b) The coupling efficiency varies with the coupling length L1 under the optimal designed parameters.
When designing the wavelength channels λ3 and λ4, we also scan the refractive index of different waveguide widths, as shown in Fig. 5
Fig. 5 (a) The effective refractive indices of TE0 mode and TE2 mode vary with the waveguide width at a wavelength of 1548.4 nm. (b) The effective refractive indices of TE0 mode and TE3 mode vary with the waveguide width at a wavelength of 1547.6 nm.
Table 1. Optimal Parameters of WDM Based on Conventional ADCs
The WDM using tapered ADCs is presented in Fig. 6
Fig. 6 (a) Schematic structure of the proposed WDM based on tapered ADCs. (b) Top view of the wavelength λ2 channel.
Fig. 7 The effective refractive indices of TE0 mode and TE1 mode vary with the waveguide width at a wavelength of 1549.2 nm.
We also show that the effective refractive index between TE0 mode and TE2 mode at the wavelength of 1548.4nm, and that of TE0 mode and TE3 mode vary with the waveguide width of 1547.6nm in Fig. 8
Fig. 8 (a) The effective refractive indices of TE0 mode and TE2 mode vary with the waveguide width at a wavelength of 1548.4 nm. (b) The effective refractive indices of TE0 mode and TE3 mode vary with the waveguide width at a wavelength of 1547.6 nm.
Table 2. Optimal Parameters of WDM Based on Tapered ADCs
3. Simulation results and comparison
The simulation results of mode conversion by ADCs are exhibited in Fig. 9
Fig. 9 Field evolution from fundamental mode TE0 to high-order modes TE1, TE2, and TE3 at the wavelengths of (a) λ2, (b) λ3, and (c) λ4 by the conventional ADC1, ADC2, and ADC3, respectively. Those mode conversions by the tapered ADCs at the wavelengths of (d) λ2, (e) λ3, and (f) λ4。
Fig. 10 The transmission of the four wavelengths multiplexed in the WDMs. (a) conventional ADCs and (b) tapered ADCs.
We next analyze the fabrication tolerance of conventional ADCs and tapered ADCs, because their sensitivity to the actual manufacturing error may lead to the deterioration of device performance. The simulation includes input waveguide width, coupling length, waveguide spacing and etch depth between waveguides, whose changes with respect to the optimal parameters are expressed as ΔW, ΔL, ΔH and Δgap, respectively. Figure 11
Fig. 11 The insertion loss of each ADC varies with (a) ∆W, (b) ∆L, (c) ∆H and (d) ∆gap, including the WDMs based on conventional ADCs and tapered ADCs.
From Fig. 11(a) and Fig. 11(b), we can see that for width and length deviation of the input waveguide, the tapered ADC-based WDM has better fabrication tolerance than the WDM using conventional ADCs. In particular, the improvement in fabrication tolerance is more distinct for the WDM with tapered ADCs when multiplexing wavelengths couple to lower order modes, such as the TE1 mode. Numerical calculation shows that the insertion loss does not exceed 3 dB for the WDM with tapered ADCs in the case of width deviation of |∆W|=0~10 nm, while the WDM with conventional ADCs even exceeds 6 dB at the same degree of width deviation. As for the tolerance caused by the deviation of the coupling length of ADCs, the WDM composed of tapered ADCs can generally improve about 0.5 dB of insertion loss under the length deviation of 5 µm, relative to the WDM based on conventional ADCs.
As can be seen from Fig. 11(c), when the change in the etching depth of the gap between the two waveguides is less than 10 nm, the insertion loss of each wavelength channel increases slowly and does not exceed 1 dB. However, when ΔH is larger than 10 nm, the insertion loss of each wavelength channel sharply increases. The simulation results manifest that the etching speed should be tested before the inductively coupled plasma (ICP) etching process to ensure that the etching depth between the two waveguides is close to 220nm as far as possible. From Fig. 11(d), we can find that for the conventional ADCs, the insertion loss of each wavelength channel is increased by about 0.2 dB when the spacing between the two waveguides is increased or decreased by 10 nm. In contrast, for the tapered ADCs, the insertion loss of each channel is increased by about 0.1 dB only when the spacing between the two waveguides is increased by 10 nm, while remaining almost unchanged under other conditions. Therefore, from the above comparison, we find that the ADC-based WDM can effectively improve the fabrication tolerance by taking advantage of the tapering structures in the bus waveguide.
4. Device fabrication and experimental results
We fabricate these two types of WDMs on the same silicon-on-insulator (SOI) wafer with a silicon layer of 220 nm and an under-cladding silicon dioxide of 2 µm to compare their performance of multiplexing wavelengths. The whole fabrication process mainly includes E-beam lithography (EBL), inductively coupled plasma (ICP) etching, and deposition of 1 µm thick SiO2 upper-cladding layer via a plasma-enhanced chemical vapor deposition (PECVD) technique. Additionally, both the coupling gratings at the input and output ports and the referenced straight waveguide have also been made on the same SOI for convenient measurement and assessment of the fabricated devices. The scanning electron micrographs (SEMs) of devices are shown in Fig. 12
Fig. 12 SEM top view of fabricated whole device. (a) The test devices include three parts: the multiplexer, the demultiplexer, and focusing grating couplers (FGCs). The local enlargement of conventional ADCs for multiplexing the wavelengths (b) λ2, (c) λ3, (d) λ4. The local enlargement of tapered ADCs for multiplexing the wavelengths (e) λ2, (f) λ3, (g) λ4. (h) Overall image of the FGC for vertical coupling between the fiber and the input waveguide. (i) An enlarged view of the manufactured grating.
As exhibited in Fig. 13
, we illustrate the schematic diagram of experimental setup which is used to measure the fabricated WDMs. When respectively injecting different wavelengths (λ1, λ2, λ3, λ4) into the four input ports by a tunable laser and a polarization controller and receiving the wavelength signals in the output ports through a power meter and optical spectrum analyzer (OSA), we can get the spectral response of the WDMs as represented in Fig. 14Fig.14 When injecting λ1, λ2, λ3, λ4 single wavelength light sources into the four channels respectively, the measured spectral response of four wavelength channels of WDMs based on (a) the conventional ADCs and (b) the tapered ADCs.
We compare the insertion loss and crosstalk for each wavelength channel of the two types of wavelength multiplexers, as shown in Table 3
Table 3. Insertion Loss and Crosstalk of the WDMs
5. Conclusion
In this paper, we investigate two kinds of WDMs based on conventional ADCs and tapered ADCs, respectively. From the results of simulation and experiment, the WDMs in both schemes exhibit excellent performance in terms of insertion loss, channel crosstalk, fabrication tolerance, device size, design flexibility, and scalability. Based on experimental results, we find that, although the WDM with conventional ADCs has relatively smaller size, the tapered ADC-based WDM has lower insertion loss (less than 1dB), smaller crosstalk (less than -23dB) and higher fabrication tolerance. In our schemes, the different wavelength signals are converted into different modes, which can reduce the crosstalk between different wavelength channels, since the different modes are orthogonal to each other in the bus waveguide. Moreover, it is easy for the design scheme to implement a scalable WDM to multiplex more wavelengths in the scheme, as long as a wider bus waveguide is designed to support more high-order modes. Additionally, since the input waveguide and the bus waveguide are coupled on the same plane, the proposed WDMs can be conveniently integrated on-chip with other active and passive devices. Our design combines both wavelength and mode dimensions which is different from previous schemes, thus this scheme may be used in high-capacity silicon-based communications systems in the future.
Funding
National Natural Science Foundation of China (NSFC) (Grant No. 61435004).
References
1. L. Chen, C. R. Doerr, L. Buhl, Y. Baeyens, and R. A. Aroca, “Monolithically integrated 40-wavelength demultiplexer and photodetector array on silicon,” IEEE Photon. Technol. Lett. 23(13), 869–871 (2011). [CrossRef]
2. K. Okamoto and K. Ishida, “Fabrication of silicon reflection-type arrayed-waveguide gratings with distributed Bragg reflectors,” Opt. Lett. 38(18), 3530–3533 (2013). [CrossRef] [PubMed]
3. J. Wang, Z. Sheng, L. Li, A. Pang, A. Wu, W. Li, X. Wang, S. Zou, M. Qi, and F. Gan, “Low-loss and low-crosstalk 8 × 8 silicon nanowire AWG routers fabricated with CMOS technology,” Opt. Express 22(8), 9395–9403 (2014). [CrossRef] [PubMed]
4. S. Chen, X. Fu, J. Wang, Y. Shi, S. He, and D. Dai, “Compact dense wavelength-division (de)multiplexer utilizing a bidirectional arrayed-waveguide grating integrated with a Mach–Zehnder interferometer,” J. Lightwave Technol. 33(11), 2279–2285 (2015). [CrossRef]
5. D. Dai, J. Wang, S. Chen, S. Wang, and S. He, “Monolithically integrated 64-channel silicon hybrid demultiplexer enabling simultaneous wavelength and mode-division-multiplexing,” Laser Photon. Rev. 9(3), 339–344 (2015). [CrossRef]
6. J. Brouckaert, W. Bogaerts, P. Dumon, D. Van Thourhout, and R. Baets, “Planar concave grating demultiplexer fabricated on a nanophotonic silicon-on-insulator platform,” J. Lightwave Technol. 25(5), 1269–1275 (2007). [CrossRef]
7. B. B. C. Kyotoku, L. Chen, and M. Lipson, "Broad band 1 nm channel spacing silicon-on-insulator wavelength division multiplexer," in Proc. IEEE Conf. Laser Electro-Optics (2009), paper JWA41. [CrossRef]
8. P. Chen, S. Chen, X. Guan, Y. Shi, and D. Dai, “High-order microring resonators with bent couplers for a box-like filter response,” Opt. Lett. 39(21), 6304–6307 (2014). [CrossRef] [PubMed]
9. P. De Heyn, J. De Coster, P. Verheyen, G. Lepage, M. Pantouvaki, P. Absil, W. Bogaerts, J. Van Campenhout, and D. Van Thourhout, “Fabrication-tolerant four-channel wavelength-division-multiplexing filter based on collectively tuned Si microrings,” J. Lightwave Technol. 31(16), 2785–2792 (2013). [CrossRef]
10. F. Horst, W. M. Green, S. Assefa, S. M. Shank, Y. A. Vlasov, and B. J. Offrein, “Cascaded Mach-Zehnder wavelength filters in silicon photonics for low loss and flat pass-band WDM (de-)multiplexing,” Opt. Express 21(10), 11652–11658 (2013). [CrossRef] [PubMed]
11. S.-H. Jeong, S. Tanaka, T. Akiyama, S. Sekiguchi, Y. Tanaka, and K. Morito, “Flat-topped and low loss silicon-nanowire-type optical MUX/DeMUX employing multi-stage microring resonator assisted delayed Mach-Zehnder interferometers,” Opt. Express 20(23), 26000–26011 (2012). [CrossRef] [PubMed]
12. L. Chang, Y. Gong, L. Liu, Z. Li, and Y. Yu, “Low-loss broadband silicon-on-insulator demultiplexers in the O-Band,” IEEE Photon. Technol. Lett. 29(15), 1237–1240 (2017). [CrossRef]
13. D. Chack, V. Kumar, S. K. Raghuwanshi, and D. P. Singh, “Design and analysis of O–S–C triple band wavelength division demultiplexer using cascaded MMI couplers,” Opt. Commun. 382, 324–331 (2017). [CrossRef]
14. C. Yao, H.-G. Bach, R. Zhang, G. Zhou, J. H. Choi, C. Jiang, and R. Kunkel, “An ultracompact multimode interference wavelength splitter employing asymmetrical multi-section structures,” Opt. Express 20(16), 18248–18253 (2012). [CrossRef] [PubMed]
15. S. Zhang, W. Ji, R. Yin, X. Li, Z. Gong, and L. Lv, “Full bandwidth wavelength division multiplexer/ demultiplexer based on MMI,” IEEE Photon. Technol. Lett. 30(1), 107–110 (2018). [CrossRef]
16. J. Xiao, X. Liu, and X. Sun, “Design of an ultracompact MMI wavelength demultiplexer in slot waveguide structures,” Opt. Express 15(13), 8300–8308 (2007). [CrossRef] [PubMed]
17. B. B. Ben Zaken, T. Zanzury, and D. Malka, “An 8-channel wavelength MMI demultiplexer in slot waveguide structures,” Materials (Basel) 9(11), 881 (2016). [CrossRef] [PubMed]
18. D. Malka, Z. Zalevsky, and Y. Sintov, “Design of a 1×4 silicon wavelength demultiplexer based on multimode interference in a slot waveguide structures,” J. Opt. Technol. 17(12), 1–4 (2014).
19. D. T. H. Tan, K. Ikeda, S. Zamek, A. Mizrahi, M. P. Nezhad, A. V. Krishnamoorthy, K. Raj, J. E. Cunningham, X. Zheng, I. Shubin, Y. Luo, and Y. Fainman, “Wide bandwidth, low loss 1 by 4 wavelength division multiplexer on silicon for optical interconnects,” Opt. Express 19(3), 2401–2409 (2011). [CrossRef] [PubMed]
20. D. T. H. Tan, A. Grieco, and Y. Fainman, “Towards 100 channel dense wavelength division multiplexing with 100GHz spacing on silicon,” Opt. Express 22(9), 10408–10415 (2014). [CrossRef] [PubMed]
21. A. Y. Piggott, J. Lu, K. G. Lagoudakis, J. Petykiewicz, T. M. Babinec, and J. Vučković, "Inverse design and demonstration of a compact and broadband on-chip wavelength demultiplexer,” Nat. Photonics 9(6), 374–377 (2015). [CrossRef]
22. Y. Hu, X. Xiao, H. Xu, X. Li, K. Xiong, Z. Li, T. Chu, Y. Yu, and J. Yu, “High-speed silicon modulator based on cascaded microring resonators,” Opt. Express 20(14), 15079–15085 (2012). [CrossRef] [PubMed]
23. Y. Liu, Y. Li, M. Li, and J. J. He, “High-sensitivity and wide-range optical sensor based on three cascaded ring resonators,” Opt. Express 25(2), 972–978 (2017). [CrossRef] [PubMed]
24. L.-W. Luo, N. Ophir, C. P. Chen, L. H. Gabrielli, C. B. Poitras, K. Bergmen, and M. Lipson, “WDM-compatible mode-division multiplexing on a silicon chip,” Nat. Commun. 5(1), 3069 (2014). [CrossRef] [PubMed]
25. J. Wang, S. He, and D. Dai, “On-chip silicon 8-channel hybrid (de)multiplexer enabling simultaneous mode- and polarization-division-multiplexing,” Laser Photon. Rev. 8(2), L18–L22 (2014). [CrossRef]
26. Y.-D. Yang, Y. Li, Y.-Z. Huang, and A. W. Poon, “Silicon nitride three-mode division multiplexing and wavelength-division multiplexing using asymmetrical directional couplers and microring resonators,” Opt. Express 22(18), 22172–22183 (2014). [CrossRef] [PubMed]
27. T. Mulugeta and M. Rasras, “Silicon hybrid (de)multiplexer enabling simultaneous mode and wavelength-division multiplexing,” Opt. Express 23(2), 943–949 (2015). [CrossRef] [PubMed]
28. Y. Ding, L. Liu, C. Peucheret, and H. Ou, “Fabrication tolerant polarization splitter and rotator based on a tapered directional coupler,” Opt. Express 20(18), 20021–20027 (2012). [CrossRef] [PubMed]
29. Y. Luo, Y. Yu, M. Ye, C. Sun, and X. Zhang, “Integrated dual-mode 3 dB power coupler based on tapered directional coupler,” Sci. Rep. 6(1), 23516 (2016). [CrossRef] [PubMed]
30. K. Okamoto, Fundamentals of Optical Waveguides (Academic Press, 2006), Chap.4.
31. 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]
