Experimental performance of DWDM quadruple Vernier racetrack resonators

We demonstrate that one can meet numerous commercial requirements for filters used in dense wavelength-division multiplexing applications using quadruple Vernier racetrack resonators in the silicon-oninsulator platform. Experimental performance shows a ripple of 0.2 dB, an interstitial peak suppression of 39.7 dB, an adjacent channel isolation of 37.2 dB, an express channel isolation of 10.2 dB, and a free spectral range of 37.52 nm. © 2013 Optical Society of America OCIS codes: (230.7408) Wavelength filtering devices; (230.5750) Resonators. References and links 1. O. Schwelb, “The nature of spurious mode suppression in extended FSR microring multiplexers,” Opt. Commun. 271, 424–429 (2007). 2. O. Schwelb and I. Frigyes, “Vernier operation of series-coupled optical microring resonator filters,” Micro. Optical Tech. Lett. 39, 257–261 (2003). 3. Y. Goebuchi, T. Kato, and Y. Kokubun, “Optimum arrangement of high-order series-coupled microring resonator for crosstalk reduction,” Jpn. J. Appl. Phys. 45, 5769–5774 (2006). 4. D. Zhang, Y. Huang, X. Ren, X. Duan, B. Shen, Q. Wang, X. Zhang, and S. Cai, “Add-drop filters based on asymmetric high-order microring resonators,” Proc. SPIE 8555, 85550U-1–85550U-7 (2012). 5. S. Dey and S. Mandal, “Modeling and analysis of quadruple optical ring resonator performance as optical filter using Vernier principle,” Opt. Commun. 285, 439–446 (2012). 6. C. Chaichuay, P. P. Yupapin, and P. Saeung, “The serially coupled multiple ring resonator filters and Vernier effect,” Opt. Appl. 39, 175 (2009). 7. H. Yan, X. Feng, D. Zhang, and Y. Huang, “Integrated optical add-drop multiplexer based on a compact parentsub microring-resonator structure,” Opt. Commun. 289, 53–59 (2013). 8. V. M. N. Passaro, B. Troia, and F. De Leonardis, “A generalized approach for design of photonic gas sensors based on Vernier-effect in mid-IR,” Sensor. Actuat. B-Chem. 168, 402–420 (2012). 9. G. Ren, T. Cao, and S. Chen, “Design and analysis of a cascaded microring resonator-based thermo-optical tunable filter with ultralarge free spectrum range and low power consumption,” Opt. Eng. 50, 0746010746016 (2011). 10. F. Boffi, L. Bolla, P. Galli, S. Ghidini, and L. Socci, “Method and device for tunable optical filtering using Vernier effect,” U.S. Patent EP2181348 B1 (2012). 11. E. J. Klein, “Densely integrated microring-resonator based components for fiber-to-the-home applications,” Ph.D. dissertation, University of Twente (2007). 12. S. Suzuki, K. Oda, and Y. Hibino, “Integrated-optic double-ring resonators with a wide free spectral range of 100 GHz,” J. Lightwave Technol. 13, 1766–1771 (1995). 13. S.-J. Choi, Z. Peng, Q. Yang, S. J. Choi, and P. D. Dapkus, “Tunable narrow linewidth all-buried heterostructure ring resonator filters using Vernier effects,” IEEE Photon. Technol. Lett. 17, 106–108 (2005). 14. S. T. Chu, B. E. Little, V. Van, J. V. Hryniewicz, P. P. Absil, F. G. Johnson, D. Gill, O. King, F. Seiferth, M. Trakalo, and J. Shanton, “Compact full C-band tunable filters for 50 GHz channel spacing based on high order micro-ring resonators,” in Optical Fiber Communication Conference (Optical Society of America, 2004). #183816 $15.00 USD Received 28 Jan 2013; revised 25 Mar 2013; accepted 27 Mar 2013; published 4 Apr 2013 (C) 2013 OSA 8 April 2013 | Vol. 21, No. 7 | DOI:10.1364/OE.21.009103 | OPTICS EXPRESS 9103 15. T. Chu, N. Fujioka, S. Nakamura, M. Tokushima, and M. Ishizaka, “Compact, low power consumption wavelength tunable laser with silicon photonic-wire waveguide micro-ring resonators,” in 35th European Conference On Optical Communication (ECOC), 1–2 (2009). 16. P. Prabhathan, Z. Jing, V. M. Murukeshan, Z. Huijuan, and C. Shiyi, “Discrete and fine wavelength tunable thermo-optic WSS for low power consumption C+L band tunability,” IEEE Photon. Technol. Lett. 24, 152–154 (2012). 17. W. Fegadolli, G. Vargas, X. Wang, F. Valini, L. Barea, J. Oliveira, N. Frateschi, A. Scherer, V. Almeida, and R. Panepucci, “Reconfigurable silicon thermo-optical ring resonator switch based on Vernier effect control,” Opt. Express 20, 14722–14733 (2012). 18. R. Boeck, N. A. F. Jaeger, N. Rouger, and L. Chrostowski, “Series-coupled silicon racetrack resonators and the vernier effect: theory and measurement,” Opt. Express 18, 25151–25157 (2010). 19. R. Boeck, J. Flueckiger, H. Yun, L. Chrostowski, and N. A. F. Jaeger, “High performance Vernier racetrack resonators,” Opt. Lett. 37, 5199–5201 (2012). 20. B. Timotijevic, G. Mashanovich, A. Michaeli, O. Cohen, V. M. N. Passaro, J. Crnjanski, and G. T. Reed, “Tailoring the spectral response of add/drop single and multiple resonators in silicon-on-insulator,” Chinese Opt. Lett. 7, 291–295 (2009). 21. M. Mancinelli, R. Guider, P. Bettotti, M. Masi, M. R. Vanacharla, J. Fedeli, D. V. Thourhout, and L. Pavesi, “Optical characterization of silicon-on-insulator-based single and coupled racetrack resonators,” J. Nanophotonics 5, 051705 (2011). 22. Y. Yanagase, S. Suzuki, Y. Kokubun, and S. T. Chu, “Box-like filter response and expansion of FSR by a vertically triple coupled microring resonator filter,” J. Lightwave Technol. 20, 1525–1529 (2002). 23. Y. Goebuchi, T. Kato, and Y. Kokubun, “Expansion of tuning range of wavelength selective switch using Vernier effect of series coupled microring resonator,” in The 18th Annual Meeting of the IEEE Lasers and Electro-Optics Society, 2005. LEOS 2005, 734–735 (2005). 24. K. Oda, N. Takato, and H. Toba, “A wide-FSR waveguide double-ring resonator for optical FDM transmission systems,” J. Lightwave Technol. 9, 728–736 (1991). 25. L. Jin, M. Li, and J.-J. He, “Highly-sensitive silicon-on-insulator sensor based on two cascaded micro-ring resonators with vernier effect,” Opt. Commun. 284, 156–159 (2011). 26. J. Hu and D. Dai, “Cascaded-ring optical sensor with enhanced sensitivity by using suspended Si-nanowires,” IEEE Photon. Technol. Lett. 23, 842–844 (2011). 27. T. Claes, W. Bogaerts, and P. Bienstman, “Experimental characterization of a silicon photonic biosensor consisting of two cascaded ring resonators based on the Vernier-effect and introduction of a curve fitting method for an improved detection limit,” Opt. Express, 18, 22747–22761 (2010). 28. X. Jiang, “Silicon nanowire waveguide sensor based on two cascaded ring resonators,” in Asia Communications and Photonics Conference, OSA Technical Digest (online) (Optical Society of America, 2012), paper AS4E.3. 29. “Single channel DWDM (100 GHz),” Alliance Fiber Optic Products, Inc. 30. “Optical add/drop multiplexers 100 GHz OADM (1x2),” Photonics-USA. 31. “Fiber optic dwdm single add/drop device,” AOXC Technologies. 32. R. Ding, T. Baehr-Jones, T. Pinguet, J. Li, N. C. Harris, M. Streshinsky, L. He, A. Novack, E.-J. Lim, T.-Y. Liow, H.-G. Teo, G.-Q. Lo, and M. Hochberg, “A silicon platform for high-speed photonics systems,” in Optical Fiber Communication Conference, OSA Technical Digest (Optical Society of America, 2012), paper OM2E.6. 33. D. Minoli, Telecommunications Technology Handbook (Artech House, 2003). 34. “DWDM and CWDM three port device optical parameter definition and test requirements,” Alliance Fiber Optic Products, Inc. 35. R. S. Romaniuk, “Optical fiber transmission with wavelength multiplexing: faster or denser?,” 5484, 19–28 (2004). 36. R.J. Bojko, J. Li, L. He, T. Baehr-Jones, M. Hochberg, and Y. Aida, J. Vac., “Electron beam lithography writing strategies for low loss, high confinement silicon optical waveguides,” Sci. Technol. B 29, 06F309 (2011). 37. “High isolation OADM (100 GHz),” Alliance Fiber Optic Products, Inc. 38. C. Manolatou, M. A. Popovic, P. T. Rakich, T. Barwicz, H. A. Haus, and E. P. Ippen, “Spectral anomalies due to coupling-induced frequency shifts in dielectric coupled-resonator filters,” in Optical Fiber Communication Conference, 2004, paper TuD5. 39. P. Prabhathan, V. M. Murukeshan, and J. Zhang, “Optimal detuning combinations in a series coupled silicon micro ring resonator thermo optic-wavelength selective switch,” Opt. Eng. 51, 044604 (2012). 40. S. J. Mason, “Feedback theory-further properties of signal flow graphs,” Proc. IRE 44, 920–926 (1956).


Introduction
The Vernier effect has been used extensively in the research community to expand the free spectral range (FSR) in ring resonators, and has been demonstrated both theoretically [1][2][3][4][5][6][7][8][9][10] and experimentally [11][12][13][14][15][16][17][18][19][20][21][22][23][24][25][26][27][28].Ideally, one would like a box-like drop port spectral response, which can be achieved using series-coupled ring resonators [22].Increasing the number of seriescoupled rings, allows one to increase the clear window (channel bandwidth) and thus the data rate that the filter can handle.However, to date, no one has experimentally demonstrated the Vernier effect in more than three series-coupled rings and no one has experimentally shown whether series-coupled Vernier ring resonators can be used to meet typical commercial filter requirements.Timotijevic et al. [20] demonstrated double Vernier silicon-on-insulator (SOI) series-coupled ring resonators which had an FSR larger than the span of the C-band but minimal interstitial peak suppression (IPS).Fegadolli et al. [17] fabricated thermally tunable double Vernier SOI series-coupled ring resonators exhibiting the Vernier effect in the through port and drop port.Mancinelli et al. [21] used double Vernier SOI series-coupled resonators that had an extended FSR of 20 nm.Prabhathan et al. [16] fabricated double SOI cascaded ring resonators exhibiting the Vernier effect at the drop port for use as a thermally tunable wavelength selective switch.The authors' device had an FSR of approximately 50 nm and a drop port out-of-band extinction greater than 15 dB [16].However, ring resonators in cascaded configurations only exhibit the Vernier effect in the drop port response and not the through port response [11].Yanagase et al. [22] have shown triple-ring resonators exhibiting the Vernier effect, where the coupling was done vertically and the material used for the waveguides was Ta 2 O 5 -SiO 2 .However, their devices show minimal IPS and their extended FSR was less than the span of the C-band.Also, they do not show the through port responses.Previously published research, both experimental and theoretical, show the benefit (FSR expansion) of using the Vernier effect within series-coupled ring resonators, as compared to the case where each resonator is identical, but minimal research has been done regarding meeting commercial requirements as regards to the IPS (see [3,4,10,19]).Here, we experimentally demonstrate that it is, in fact, possible to meet many commercial requirements when using quadruple SOI series-coupled racetrack resonators exhibiting the Vernier effect including the necessary IPS.The commercial requirements that are met are the drop port ripple (R depth ), the drop port adjacent channel isolation (A i ), the drop port non-adjacent channel isolation (nA i ), and the express channel isolation (EC i ).It should be noted that the EC i has the same definition as the through port channel extinction ratio and pass channel residual at express port.The IPS is similar to that of the nA i , however, since our device involves the Vernier effect there are multiple peaks between two dominant resonant peaks and, thus, it is important that the IPS meets the commercial specification for the nA i .The target specifications come from commercial dense wavelength-division multiplexing (DWDM) data sheets [29][30][31].

Theory
A schematic of our quadruple series-coupled racetrack resonators exhibiting the Vernier effect is shown in Fig. 1 which is similar to that found in [2][3][4] (same arrangement of the resonators but different resonator lengths, field transmission and coupling factors, and propagation loss).Here, L a , L b , L c , and L d are the total lengths of racetrack resonators a, b, c, and d, respectively.L y is the length of the straight coupling regions, r is the radius of the racetrack resonators, and L is the length of the straight sections (other than those in the coupling regions) for racetrack resonators c and d. α is the total field loss coefficient for the racetracks resonators.κ 1 , κ 2 , κ 3 , κ 4 , and κ 5 are the symmetric (real) point field coupling factors.t 1 , t 2 , t 3 , t 4 , and t 5 are the The waveguides are strip waveguides with a top silicon dioxide cladding having widths and heights of 502 nm and 220 nm, respectively.The propagation loss for each ring was assumed to be 2.4 dB/cm [32].The modeling and analysis of the quadruple Vernier racetrack resonator was done using a mixture of numeric and analytic methods.Specifically, the effective index of the strip waveguides and the field coupling and transmission factors were determined using MODE Solutions by Lumerical Solutions, Inc., and everything else was done analytically.The gap distances are g 1 = 125 nm, g 2 = 350 nm, g 3 = 410 nm, g 4 = 350 nm, and g 5 =125 nm, respectively.The value of their field coupling and field transmission factors were determined by applying a third order polynomial curve-fit to the wavelength dependent even and odd effective indices.At The derivation of the drop port and through port transfer functions can be found in the Appendix.The spectral characteristics are defined based on typical commercial definitions that describe the performance of DWDM filters.We choose a 100 GHz (≈ 0.8 nm) channel spacing and a 6 GHz (≈ 0.048 nm) clear window centered at a desired wavelength (our choice of clear window assumes there will be no laser wavelength drift).The FSR should be greater than or equal to 35.89 nm (the span of the C-band, 1528.77nm to 1563.86 nm [33], plus one adjacent channel).The definitions for A i [34,35], nA i [34,35], R depth [33,34], and EC i [34] can be found in the citations given.The spectral characteristics are determined within the clear windows of the desired channel and the 44 clear windows to the left and right of the desired channel.The spectra shown in Fig. 2(a) and 2(b) meet the typical commercial values for the target specifications as shown in Table 1.It should be noted that there is some variance in target values depending on the DWDM vendor.For example, [29] specifies an EC i value of 12 dB whereas [30] specifies a value of 10 dB.

Fabrication results and discussion
The device was fabricated at the University of Washington using electron beam lithography, as described in [36].Figures 3(a) and 3(b) show the experimental through port and drop port responses of one of our quadruple Vernier racetrack resonators.Figure 3(a) clearly shows significant IPS.The maximum through port insertion loss (IL thru ) [34] and the drop port insertion loss (IL drop ) [33][34][35] have not been included since we were unable to measure them accurately.The spectral characteristics meet numerous commercial requirements as shown in Table 2.The EC i , R depth , and IPS are within 0.9 dB of the theoretical results.The experimental A i is 37.2 dB whereas the theoretical A i is 52.7 dB, which is likely due to the experimental filter line shape being asymmetric and to increased field coupling factors due to the bend regions of the couplers.To be able to simultaneously drop and add signals using just one instance of the device shown in Fig. 1, the target values shown in Table 2 would be needed except that EC i would need to be greater than or equal to 25 dB [37].Here, we have defined the IPS as the contrast between the minimum transmission within the clear window of the desired channel and the maximum transmission within the clear windows of all non-adjacent channels.However, it should be noted that in [19], the IPS was defined as the difference (in dB) between the maximum value of a major peak and the maximum value of the largest interstitial peak.Based on this definition, our measured IPS would be 37.1 dB.The much larger notches within the pass band of the through port as compared to the theoretical results are possibly due to fabrication variations and coupling-induced frequency shifts (CIFS) [38,39], which can be corrected by thermally tuning each racetrack resonator [39].However, the notches are not located within any of the adjacent or non-adjacent channels as shown in Figs.3(c) and 3(d).The passband of the through port to the left of the major peak shows that there are actually 4 notches (two small notches and two large notches) as shown in Fig. 3(c).However, our theoretical results showed that there are only two small notches as shown in Fig. 2(a).The likely reasons for this difference between the theory and experimental results are fabrication variations, in which the effective indices of the resonators are not all exactly the same, and CIFS.For example, if the effective index of racetrack resonator a is decreased by 0.003 (as shown in Fig. 3(e)) each of the notches separates into 2 notches (one small and one large), where the larger notch is located to the right of the smaller notch (as shown in Fig. 3(f)), which is in agreement with the experimental results.In addition to the device presented here, we fabricated 48 other devices in which the gap distances were varied.The device presented here showed the best performance.However, future designs based on this device can be made to be thermally tunable to compensate for fabrication variations and effects such as CIFS.[29,30], 30 [31] nA i , IPS (dB) 39.7 ≥ 35 [31], 40 [29,30] R depth (dB) 0.2 ≤ 0.5 [30] EC i (dB) 10.2 ≥ 10 [30], 12 [29]

Summary
In summary, we have experimentally shown that it is possible to meet numerous commercial requirements for dense wavelength-division multiplexing filters using quadruple series-coupled racetrack resonators exhibiting the Vernier effect.We have demonstrated a Vernier filter having a free spectral range greater than the span of the C-band (37.52 nm), a ripple of 0.2 dB, an adjacent channel isolation of 37.2 dB, an interstitial peak suppression of 39.7 dB, and an express channel isolation of 10.2 dB.

Appendices
Appendix: Quadruple series-coupled racetrack resonator transfer functions Here, we will derive the drop port and through port transfer functions of quadruple seriescoupled racetrack resonators using Mason's rule [40].In [5], Dey et al. derived the drop port transfer function using Mason's rule [40] but did not derive nor present the through port transfer function.For completeness, here we have re-derived the drop port transfer function as well as derived the through port transfer function.Since in our configuration, there are four racetrack resonators coupled in series, there are 33 loop gains.There are 10 loop gains of the 10 possible combinations of 1 non-touching loop, (2) The gain and co-factor of the first forward path that is used to determine the drop port transfer function is, The determinant for the entire system is given Thus, the transfer function for the drop port is given by [5], The gains of the second to sixth forward path that are used to determine the through port transfer function are, and the corresponding co-factors are, (47) Thus, the transfer function for the through port is given by,

#Fig. 3 .
Fig. 3. (a) measured through port and drop port spectral response.(b) zoom-in of the measured major resonance.(c) zoom-in of the measured through port passband to the left of the major peak.(d) zoom-in of the measured through port passband to the right of the major peak.(e) zoom-in of the theoretical notch splitting when the effective index of racetrack resonator a decreases and (f) zoom-out of Fig. 3(e) (showing the increase in notch depth as the effective index of racetrack resonator a decreases).
) where X a,b,c,d = exp(−α a,b,c,d L a,b,c,d − jβ a,b,c,d L a,b,c,d ), where the field loss coefficients and propagations constants for the racetrack resonators are represented by α a,b,c,d , and β a,b,c,d , respectively.κ 1 , κ 2 , κ 3 , κ 4 , and κ 5 are the symmetric (real) point field coupling factors.

Table 1 .
Theoretical and Target Specifications