Open Access
Add to BibSonomy
Abstract
Chirped and apodized grating couplers were designed, fabricated and characterized in lithium niobate thin film for fiber-to-chip coupling. The maximum coupling efficiency of -1.8 dB and -6.9 dB for TE mode at a wavelength of 1550 nm was simulated and measured, respectively. The discrepancies were mainly attributed to the different fabrication errors of local periods and groove widths, which hampered the mode matching condition heavily. In addition, when a metal bottom reflector was added, the simulated and measured coupling efficiency were improved to -0.8 dB and -5.5 dB, respectively.
© 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Thin film materials with high refractive index contrast are attractive and useful for high-density photonic integrated circuits. As one of them, lithium niobate thin film (lithium niobate on insulator, LNOI) is warmly used in nonlinear optics and integrated optics devices, due for its extraordinary nonlinear and excellent electro-optical properties [1–5]. However, Unlike thick films and block materials, coupling between a LNOI waveguide and a single mode fiber is especially difficult, since the guided mode is strongly confined in the waveguide core, and the mismatch between the waveguide mode and the fiber mode is obvious.
End-face coupling approaches and surface coupling approaches have been adopted in fiber-to-chip (LN chip) coupling for years. The end-face coupling approaches, such as inversed tapers [6], tapered optical fibers [7–10], and lensed optical fibers [11,12], are good at realization high coupling efficiency (CE) and broad bandwidth with low polarization sensitivity, although some essential post-fabrication treatment is in need, such as edge polishing, and high-resolution alignment. Conversely, surface coupling approaches, for example, the grating coupler can perform well in flexibility, alignment tolerance, and wafer-scale testing, they needn’t edge polishing and can be placed anywhere on a chip. By now, the main problem to grating coupler, especially grating coupler in LNOI, are the high polarization sensitivity, narrow bandwidth, and particularly the low CE.
A few researches on grating couplers in LNOI have been reported over recent years, but the measured CEs were not so high [13–21]. By now, the maximum CE reported was about -3.1 dB at wavelength 1660 nm by a combination process of electron beam lithography (EBL) and optimized lift-off process, followed by reactive ion beam etching (RIE) [20]. Except for fabrication errors, the backward reflection, downward radiation, and mode-mismatch should be responsible for the limited CE for grating couplers in LNOI platform, and some effective approaches adopted in other materials or platforms could be borrowed, such as a deep-etched groove in front of the gratings, a slanted angle of fiber, L-shaped grating structures, dual-layer structures, and Bragg or metal bottom mirrors [22–30], since they all effective and feasible approaches to reduce the backward reflection and downward radiation. The CE of a uniform grating coupler can reach to the maximum of 80% in theory [31], the lost 20% power is attributed to the mode-mismatch between fibers and waveguides. By reshaping the upward diffraction field distribution, chirped or apodized grating couplers can solve the mode-mismatch problem and minimize the back-reflection obviously [32].
In this work, grating couplers on LNOI with chirped and apodized configurations are designed, fabricated and characterized for effective fiber-to-chip coupling. Two dimensional finite-different times domain method (2D-FDTD) was applied in simulation, and focused ion beam (FIB) was used in fabrication to propose a validation of the design. By chirping the period and apodizing the filling factor (defined as the ratio of the groove width over the period, FF), the output decaying power is shaped like the Gussian beam and surely has a good mode-matching with the fiber mode. The measured CE of this grating configuration is -6.9 dB, while the simulated CE is -1.8 dB. In addition, when a metal bottom reflector is added between the oxide layer and lithium niobate substrate during the fabrication of LNOI platform, the measured CE is improved to -5.5 dB, while the simulated CE is -0.8 dB. The discrepancies are attributed to the imperfect fabrication of gratings, especially to the imperfect and nonuniform fabrication errors of the first few periods and filling factors, which has a considerable impact on coupling efficiency and should be improved in future fabrications.
2. Design of the coupler
The coupling problem was considered between a waveguide on LNOI platform and a single mode fiber with a core diameter of 9 μm and a cladding diameter of 125 μm. The thickness of lithium niobate thin film (H) and silicon dioxide layer (D) was set to be 494 nm and 1.96 μm to keep with the experiment, respectively. 2D-FDTD and perfect layer matching (PML) bonding condition were used in the simulation and optimization. The fundamental TE mode was injected in the waveguide, and monitors were placed in the fiber, as well as in the oxide layer and the forward and backward directions, to monitor the transmissions of this mode. The normalized transmission upward to the fundamental TE mode of fiber is defined as the CE. Fiber angle was tilted at 8° to reduce the backward reflection.
The initial simulations were performed on a uniform grating configuration and a filling factor of 0.4 was chosen to keep the power diffracted slowly by each grooves. The optimized CE for fundamental TE mode at wavelength 1550 nm was 45% (-3.5 dB), when the period was 940 nm and etch depth was 320 nm. Moreover, the related backward reflection, the upward transmission, and the downward diffraction were approximately 0.0026%, 64%, and 33%, respectively. The above result was then used as an initial setting for the following simulations. Considering the diffracted field as a superposition of fields scatted by each grooves, the constructive interference could be achieved by adjusting the parameters, such as periods, etch depth, and filling factor. Although the etch depth could be tuned, considering the complexity of the fabrication process, only the period and filling factor were tuned and hundreds of simulations were performed on the basic of the initial result. Because power was diffracted rapidly by the first few grating grooves, only parts of periods and FFs were chosen to be tuned, and so a chirped and apodized grating coupler was formed. Schematic diagram of the uniform and proposed grating coupler are shown in Fig. 1(a) and (b).
Fig. 1. Schematic diagram of the uniform (a) chirped and apodized grating coupler (b), and E field distribution profile in uniform (c) chirped and apodized grating coupler (d).
As shown in Fig. 1(b), the periods (Λi) and filling factors (Fi) in chirped part were tuned. The purpose is to make a slowly increasing diffraction, in order to shape the first half of Gaussian field distribution, while the radiated field distribution in the uniform grating was an exponential one along the propagation direction. Λi and Fi were varied and satisfied the following linear formulas among the optimization, which were designed to modulate the radiated field distribution:
Fig. 2. Simulated upward transmission (a) and coupling efficiency (b) for different grating configurations. The initial period and filling factor are 940 nm and 0.4, respectively. The blue lines denote the values of uniform grating configuration; the black lines denote the values of chirped and apodized grating configuration; the red lines denote the values of chirped, apodized, and metal-reflector-deposited grating configuration.
3. Experiment results
The gratings were firstly fabricated on a Z-cut LNOI chips purchased from the research center of Nanoln with H = 494 nm and D = 1.96 μm. FIB etching system was used to fabricate the lithium niobate waveguide and grating structure simultaneously. During the etching, the acceleration voltage, the extraction voltage, the emission current, and the suppress voltage were set to be 30 kV, 6 kV, 2 μA, and 9 V, respectively. The strip waveguides were etched by the mid current of 730 pA to save fabrication time while the grating grooves were etched by the fine current of 240 pA to ensure a good fabrication quality. The FIB microscope image of one grating fabricated is shown in Fig. 3. From Fig. 3(a),a strip waveguide and a grating structure etched on bottom of the waveguide can be seen, and another grating coupler was etched on the other end of the 12-μm-wide waveguide. The measured width of the waveguide was about 12 μm, and the measured area of grating zone was about $16 \times 13$ μm2. Figure 3(b) gives the enlarged version of grating shown in Fig. 3(a), the grating grooves (from top to bottom) were highlighted by some white solid lines. It could be seen that, for the first eight grating grooves, the width increased strip by strip, and so this part was described as the chirped part; the rest was consistent with each other, and so this part was described as the uniform part.
The CEs were determined by a transmission measurement system, more details had been demonstrated in [17]. The measurements were performed by two single mode fibers which tilted symmetrically up the two gratings at an angle of 8°, and used as the input fiber and output fiber to import TE mode power in and export power out the 12-μm-wide waveguide, respectively. In addition to this, the 12-μm-wide waveguide can be tapered to a thinner waveguide (1-3 μm in width) in the future fabrication to achieve the transmission of fundamental TE mode. Simulated result showed that the length of this taper should longer than 300 μm to insure a low transmission loss (less than 0.05 dB/taper). To further determine the CE and the propagation loss of the waveguides (12-μm-wide), transmissions of a 400-μm-long and an 800-μm-long waveguides in same grating configuration were fabricated and measured, respectively. There were no obvious difference between the two measured values, and so the propagation loss of the waveguide was neglected in the following calculation. Since the collected transmitted power T was obtained after twice coupling by gratings, the coupling efficiency η = √T.
The coupling efficiency with respect to wavelength is shown in Fig. 4. As shown in Fig. 4, for couplers on LNOI without a bottom reflector (blue line and black line), since the maximum transmission T measured by the power meter were about -15.6 dB (uniform) and - 13.7 dB (chirped and apodized) with an etch depth of 320 nm, the maximum CE for one coupler were -7.8 dB (uniform) and -6.9 dB (chirped and apodized), respectively. There is about 1 dB difference in CE between the two grating configurations, and it could be attributed to the chirped and apodized structure which might improve the mode-matching condition. Compared the results with the simulated ones (Fig. 2), it could be seen that the experiment results were in similar shapes with simulated ones, although about 4.3 dB (uniform) and 5 dB (chirped and apodized) value-discrepancies existed, respectively. Since the etch depth was fixed and controlled more precisely, these discrepancies could be mainly attributed to other fabrication errors, such as the grating periods and groove widths. It should also be noted that the discrepancy in chirped and apodized grating coupler is larger than that in the uniform one, perhaps the fabrication errors of the first few grating structures should be responsible for this, since it played an important role in the diffraction field distribution. As shown in Fig. 3(b), the widths of the first two grating grooves were extended larger than the optimal values (∼16 nm/period), and from the third to the eighth, the groove widths were about 10-33 nm less than the optimal values. The nonmonotonic changes in groove width surely destroy the designed mode-matching condition, and hampers the realization of the effective CE introduced by the optimized design, introducing a possible coupling loss in the measured CE (The additional coupling loss is about 1.2 dB in simulation). The measured 3-dB-bandwidth for the chirped and apodized grating was 90 nm.
Fig. 4. Measured coupling efficiencies for different grating periods. The blue line denotes the CE of a uniform coupler with a period of 940 nm fabricated on a pure chip; the magenta, black, and dark yellow lines denote the CEs of couplers with initial periods of 930 nm, 940 nm, and 950 nm fabricated on the above pure LNOI platform, respectively; the red line denotes the CEs of coupler with initial period of 940 nm fabricated on another LNOI chip which deposited with a metal bottom reflector.
To verify the effect of metal bottom reflector in coupling, a metal bottom layer (10 nm Cr/100 nm Au/30 nm Cr) was deposited by electron beam evaporation during the chip fabrication process, and a novel chip of LNOI with metal bottom layer was formed and used to etch chirped and apodized gratings. The coupling performance of this grating coupler was also characterized. The measured CE and 3-dB bandwidth was -5.5 dB (28.6%) and 82 nm, respectively. The CE was improved by 1.4 dB when the metal bottom reflector was added. It should be noted that this CE was achieved on another LNOI chip which had some differences in H and D. Dimensions for grating couplers mentioned above were listed in Table 1. (H = 482 nm and D = 2.23 μm).
Table 1. Dimensions of grating couplers fabricated.
The measured CEs for different periods are also shown in Fig. 4. The CEs were measured and calculated from the grating couplers with a period of 930 nm, 940 nm, and 950 nm, respectively. Result showed that 10-nm deviation of initial period from the optimal value of 940 nm had introduced less than 0.05 dB coupling loss, indicating an acceptable period fabrication tolerance of the proposed coupler.
In addition, a comparison between grating couplers ever reported in LNOI and the coupler in this work is listed in Table 2. For grating couplers in LNOI, the maximum simulated CE, the maximum experimental CE, and the broadest 3-dB-bandwidth for TE mode ever achieved were -1.5 dB [21], -3.1 dB [20], and 102 nm [18], respectively. In this work, they were about -1.8 dB, -6.9 dB, and 90 nm for the coupler in a pure LNOI, and they were -0.8 dB, -5.5 dB, and 82 nm for the coupler in LNOI with a metal bottom reflector. It could not be denied that chirped and apodized grating coupler is a good solution to the fiber-to-chip coupling, although the experiment results were not so good enough.
Table 2. Comparison of grating couplers ever reported in LNOI and in this work.
4. Conclusion
In this work, chirped and apodized grating couplers on LNOI platform were studied. The simulated and measured CE of this grating configuration was -1.8 dB and -6.9 dB, respectively. Although there exist discrepancies between the above values, this approach is considered a commendable solution to the coupling problem between single mode fibers and LNOI devices, if the fabrication precision can be controlled more accurately.
Funding
National Natural Science Foundation of China (61575111).
Disclosures
The authors declare no conflicts of interest.
References
1. C. Wang, M. Zhang, X. Chen, M. Bertrand, A. Shams-Ansari, S. Chandrasekhar, P. Winzer, and M. Lončar, “Integrated lithium niobate electro-optic modulators operating at CMOS-compatible voltages,” Nature 562(7725), 101–104 (2018). [CrossRef]
2. A. Boes, B. Corcoran, L. Chang, J. Bowers, and A. Mitchell, “Status and Potential of Lithium Niobate on Insulator (LNOI) for Photonic Integrated Circuits,” Laser Photonics Rev. 12(4), 1700256 (2018). [CrossRef]
3. M. He, M. Xu, Y. Ren, J. Jian, Z. Ruan, Y. Xu, S. Gao, S. Sun, X. Wen, L. Zhou, L. Liu, C. Guo, H. Chen, S. Yu, L. Liu, and X. Cai, “High-performance hybrid silicon and lithium niobate Mach–Zehnder modulators for 100 Gbit/s and beyond,” Nat. Photonics 13(5), 359–364 (2019). [CrossRef]
4. C. Wang, M. Zhang, M. Yu, R. Zhu, H. Hu, and M. Loncar, “Monolithic lithium niobate photonic circuits for Kerr frequency comb generation and modulation,” Nat. Commun. 10(1), 978 (2019). [CrossRef]
5. M. Zhang, B. Buscaino, C. Wang, A. Shams-Ansari, C. Reimer, R. Zhu, J. M. Kahn, and M. Lončar, “Broadband electro-optic frequency comb generation in a lithium niobate microring resonator,” Nature 568(7752), 373–377 (2019). [CrossRef]
6. L. He, M. Zhang, A. S. Ansari, R. Zhu, C. Wang, and M. Loncar, “Low-loss fiber-to-chip interface for lithium niobate photonic integrated circuits,” Opt. Lett. 44(9), 2314–2317 (2019). [CrossRef]
7. S. Diziain, R. Geiss, M. Steinert, C. Schmidt, W. K. Chang, S. Fasold, D. Füßel, Y. H. Chen, and T. Pertsch, “Self-suspended micro-resonators patterned in Z-cut lithium niobate membranes,” Opt. Mater. Express 5(9), 2081–2089 (2015). [CrossRef]
8. J. Wang, F. Bo, S. Wan, W. Li, F. Gao, J. Li, G. Zhang, and J. Xu, “High-Q lithium niobate microdisk resonators on a chip for efficient electro-optic modulation,” Opt. Express 23(18), 23072–23078 (2015). [CrossRef]
9. C. Wang, M. J. Burek, Z. Lin, H. A. Atikian, V. Venkataraman, I. C. Huang, P. Stark, and M. Lončar, “Integrated high quality factor lithium niobate microdisk resonators,” Opt. Express 22(25), 30924–30933 (2014). [CrossRef]
10. J. Lin, Y. Xu, Z. Fang, M. Wang, J. Song, N. Wang, L. Qiao, W. Fang, and Y. Cheng, “Fabrication of high-Q lithium niobate micro resonators using femtosecond laser micromachining,” Sci. Rep. 5(1), 8072 (2015). [CrossRef]
11. I. Krasnokutska, J. J. Tambasco, X. J. Li, and A. Peruzzo, “Ultra-low loss photonic circuits in lithium niobate on Insulator,” Opt. Express 26(2), 897 (2018). [CrossRef]
12. C. Wang, X. Xiong, N. Andrade, V. Venkataraman, X. Ren, G. Guo, and M. Lončar, “Second harmonic generation in nano-structured thin-film lithium niobate waveguides,” Opt. Express 25(6), 6963–6973 (2017). [CrossRef]
13. M. Mahmoud, S. Ghosh, and G. Piazza, “Lithium Niobate on Insulator (LNOI) Grating Couplers,” in Proceedings of IEEE Conference on Lasers and Electro-Optics (IEEE, 2015), pp. 1–2.
14. Z. Chen, R. Peng, Y. Wang, H. Zhu, and H. Hu, “Grating coupler on lithium niobate thin film waveguide with a metal bottom reflector,” Opt. Mater. Express 7(11), 4010–4017 (2017). [CrossRef]
15. M. S. Nisar, X. Zhao, A. Pan, S. Yuan, and J. Xia, “Grating Coupler for an On-Chip Lithium Niobate Ridge Waveguide,” IEEE Photonics J. 9(1), 1–8 (2017). [CrossRef]
16. M. A. Baghban, J. Schollhammer, C. E. Herranz, K. B. Gylfason, and K. Gallo, “Bragg gratings in thin-film LiNbO3 waveguides,” Opt. Express 25(26), 32323–32332 (2017). [CrossRef]
17. Z. Chen, Y. Wang, H. Zhang, and H. Hui, “Silicon grating coupler on lithium niobate thin film waveguide,” Opt. Mater. Express 8(5), 1253–1258 (2018). [CrossRef]
18. J. Jian, P. Xu, H. Chen, M. He, Z. Wu, L. Zhou, L. Liu, C. Yang, and S. Yu, “High-efficiency hybrid amorphous silicon grating couplers for sub-micron-sized lithium niobate waveguides,” Opt. Express 26(23), 29651–29658 (2018). [CrossRef]
19. L. Cai and G. Piazza, “Low-loss chirped grating for vertical light coupling in lithium niobate on insulator,” J. Opt. 21(6), 065801 (2019). [CrossRef]
20. I. Krasnokutska, R. J. Chapman, J.-L. J. Tambasco, and A. Peruzzo, “High coupling efficiency grating couplers on lithium niobate on insulator,” Opt. Express 27(13), 17681–17685 (2019). [CrossRef]
21. X. Ma, C. Zhuang, R. Zeng, J. J. Coleman, and W. Zhou, “Polarization-independent one-dimensional grating coupler design on hybrid silicon/LNOI platform,” Opt. Express 28(11), 17123 (2020). [CrossRef]
22. F. V. Laere, G. Roelkens, M. Ayre, J. Schrauwen, D. Taillaert, D. V. Thourhout, T. F. Krauss, and R. Baets, “Compact and Highly Efficient Grating Couplers Between Optical Fiber and Nanophotonic Waveguides,” J. Lightwave Technol. 25(1), 151–156 (2007). [CrossRef]
23. G. Roelkens, D. Van Thourhout, and R. Baets, “High Efficiency grating coupler between silicon-on-insulator waveguides and perfectly vertical optical fibers,” Opt. Lett. 32(11), 1495–1497 (2007). [CrossRef]
24. Y. Ding, C. Peucheret, H. Ou, and K. Y. Vind, “Fully etched apodized grating coupler on the SOI platform with -0.58 dB coupling efficiency,” Opt. Lett. 39(18), 5348–5350 (2014). [CrossRef]
25. D. Benedikovic, C. Alonso-Ramos, D. Pérez-Galacho, S. Guerber, V. Vakarin, G. Marcaud, X. L. Roux, E. Cassan, D. Marris-Morini, P. Cheben, F. Boeuf, C. Baudot, and L. Vivien, “L-shaped fiber-chip grating couplers with high directionality and low reflectivity fabricated with deep-UV lithography,” Opt. Lett. 42(17), 3439–3442 (2017). [CrossRef]
26. D. Benedikovic, C. Alonso-Ramos, S. Guerber, X. L. Roux, P. Cheben, C. Dupré, B. Szelag, D. Fowler, É. Cassan, D. Marris-Morini, C. Baudot, F. Boeuf, and L. Vivien, “Sub-decibel silicon grating couplers based on L-shaped waveguides and engineered subwavelength metamaterials,” Opt. Express 27(18), 26239–26250 (2019). [CrossRef]
27. Mark T. Wade, F. Pavanello, R. Kumar, Cale M. Gentry, A. Atabaki, R. Ram, V. Stojanović, and Miloš A. Popović, “75% efficient wide bandwidth grating couplers in a 45 nm microelectronics CMOS process,” in Proceedings of IEEE Conference on Optical Interconnects (IEEE, 2015), pp. TuB4.
28. J. Notaros, F. Pavanello, Mark T. Wade, Cale M. Gentry, A. Atabaki, L. Alloatti, Rajeev J. Ram, and Miloš A. Popović, “Ultra-Efficient CMOS Fiber-to-Chip Grating Couplers,” in Proceedings of IEEE Conference on Optical Fiber Communication (OFC) (IEEE, 2016), pp. M2I.5.
29. C. Baudot, D. Dutartre, A. Souhaite, et al., “Low cost 300 mm double-SOI substrate for low insertion loss 1D & 2D grating couplers,” in Proceedings of IEEE Conference on 11th International on Group IV Photonics (GFP) (IEEE, 2014), pp. 137–138.
30. A. Mekis, S. Gloeckner, G. Masini, A. Narasimha, T. Pinguet, S. Sahni, and P. D. Dobbelaere, “A grating coupler-enabled CMOS photonics platform,” IEEE J. Sel. Top. Quantum Electron. 17(3), 597–608 (2011). [CrossRef]
31. D. Taillaert, P. Bienstman, and R. Baets, “Compact efficient broadband grating coupler for silicon-on-insulator waveguides,” Opt. Lett. 29(23), 2749–2751 (2004). [CrossRef]
32. A. Bozzola, L. Carroll, D. Gerace, I. Cristiani, and L. C. Andrean, “Optimising apodized grating couplers in a pure SOI platform to -0.5 dB coupling efficiency,” Opt. Express 23(12), 16289–16304 (2015). [CrossRef]
