Open Access
Add to BibSonomyAbstract
The electric-optical property of the proton exchanged phase modulator in an x-cut single-crystal lithium niobate thin film was studied. Proton exchanged waveguides generally suffered from a deteriorated electric-optical coefficient. By introducing a shallow proton exchange layer (thickness = 0.165 μm), most energy of the optical mode was allowed to guide in the untouched single-crystal lithium niobate film, making contribution to the effective electric-optical coefficient as high as 29.5 pm/V, which was very close to that of the bulk lithium niobate (r33 = 31 pm/V). A 12 V voltage applied to the electrodes located on the two sides of the waveguide induced a 0.097 nm shift of the Fabry-Perot resonant peak. Considering the wavelength difference of the neighboring resonant peaks (0.228 nm) and the length of the electrodes (2.3 mm), the voltage-length product was as low as 6.5 V·cm, indicating the efficient electric-optical modulation.
© 2016 Optical Society of America
1. Introduction
Lithium niobate (LN) is a remarkable optical material due to its broad transparent window in visible and infrared range, significant nonlinear effect and acousto-optical properties [1]. It has extensive applications in integrated optics including nonlinear wavelength converters, modulators, and amplifiers [2–4]. Electric-optical (E-O) modulators originate from the optical phase variation by the change of the refractive index in the applied electric field in E-O materials. Due to the high E-O coefficient (r33 = 31 pm/V) in LN, high efficient E-O modulators in such material is very promising and always an interesting topic in optical interconnect technology [5–8]. In recent years, the emergence of the single-crystal LN thin film on insulators (LNOI) have attracted much interests because the high optical confinement in the thin film structure leads to strong guiding of light, enabling high performance and small footprint photonic devices, such as submicron size waveguides for second harmonic generation, Mach-Zehnder (MZ) modulators, and microresonators [9–15]. E-O modulation is usually based on channel waveguides which carry micron sized optical mode to facilitate the efficient interaction between the light and the E-O material. The methods of fabricating waveguides in LNOI include ion milling, deposition of high index strips, and proton exchange (PE) [16–18]. PE is a very simple, mature fabrication method and compatible with the LN optical waveguide industry. Furthermore, PE waveguides in LNOI can have a very low propagation loss [18]. However, for nonlinear photonic devices, PE waveguides generally suffer from dramatically reduced E-O and nonlinear coefficients in the previous studies on bulk LN [19–23]. Therefore, it is necessary to study and develop a method to get a prominent E-O effect in PE waveguides in LNOI.
In this paper, we fabricated a phase modulator in an x-cut LNOI to measure the E-O effect in the PE waveguide. The effective E-O coefficient was as high as 29.5 pm/V, which was very close to r33 of the bulk material (~31 pm/V). A short PE time made most of the guiding light (70%) confined in the pure LN region thus enabling using the untouched E-O effect. The measured VπL was 6.5 V·cm.
2. Simulation and calculation
The schematic image of the PE LNOI phase modulator is shown in Fig. 1(a). LNOI consisted of an ion-sliced LN film, which had a thickness T, bonded on a SiO2 layer on a LN substrate.
Fig. 1 (a) Configuration of the PE phase modulator in LNOI. (b) Optical mode profile with W = 3.2 μm, D = 0.165 μm and T = 0.52 μm for the quasi-TE guided mode at 1.55 μm. (c) Electrostatic field (Ez) after 1 V voltage was applied to the electrodes. The gap (G) between the electrodes was 10 μm.
The PE region was assumed to have a rectangular shape and step-like index distribution because the lateral diffusion could be ignored when the PE region width (W) was much larger than the exchange depth (D) [24]. The ordinary and extraordinary refractive index changes were −0.05 and 0.08, respectively [18]. The gap between the two electrodes was G. Figures 1(b) and 1(c) show the optical mode profile (at a wavelength of 1.55 μm) and the electrostatic field (Ez) in the waveguide. The parameters related to the structure were set as T = 0.52 μm, D = 0.165 μm, W = 3.2 μm and G = 10 μm, which were consistent with the following experiment fabrication. The optical mode size was 2.54 μm (horizontal 1/e intensity of mode distribution) × 0.42 μm (vertical 1/e intensity of mode distribution) = 1.1 μm2. According to the simulation, about 70% of the light energy was in the pure LN layer region, and 20% was in the PE region. The electrostatic field had high intensity in the LN film and gradually decreased downwards. Since the optical mode was mostly confined in the thin LN layer, it could have a larger overlap between the optical field and electrostatic field than the conventional Ti diffused LN waveguide whose mode profile had several microns vertical size [25].
The single-mode (SM) condition of the waveguide was simulated. The cut-off dimension of PE region for the quasi-TE10 mode (two electric field intensity peaks in the lateral direction and one electric field intensity peak in the vertical direction) was calculated and shown in Fig. 2. Any dimensions beneath the curves fulfilled the SM condition in the horizontal direction. In the vertical direction, when T > 0.59 μm, the quasi-TE01 mode (one electric field intensity peak in the lateral direction and two electric field intensity peaks in the vertical direction) existed.
Fig. 2 Cut-off dimensions of the PE region for quasi-TE10 mode. The dimensions below the curves corresponded to SM conditions in the horizontal direction.
Low VπL led to low drive power or small length. The small length enabled high bandwidth and low loss [26]. VπL was calculated as follows [27]:
Where λ was the wavelength and n was the refractive index of LN. The overlap factor, Γ, was defined as follows [27]:Where V was the applied voltage, and Eele and Eopt were the electrostatic field and optical field, respectively. The integral should be done over the LN film, excluding the PE region, because PE without anneal had nearly vanished E-O coefficient [21]. The dependence of VπL of the phase modulator with the configuration in Fig. 1(a) on D and W in different thickness LN films are simulated and shown in Fig. 3. Smaller G led to lower VπL but would result in higher loss of the waveguide due to the absorption of the metal electrodes. In the simulations, G was selected to be a value at which the absorption loss was 1 dB/cm. In [18], the relations between the mode size and parameters W, D and T were simulated. The mode size had minimal values when W or D varied and the mode size increased with T. VπL had a complex relationship with W, D and T. First, as W decreased from a large value, the mode size reduced and G had a chance to decrease. The decreased G led to a smaller VπL. As W decreased further, the mode size reached a minimal value and then increased [18], resulting in a larger G and thus larger VπL. Moreover, the light energy in the PE region would increase with W [18], that would reduce Γ and make VπL large. Second, large D resulted in high confinement of light and so the mode size decreased with the increasing D. G and VπL would then decrease. However, large D made more energy of light guide in the PE region which had low E-O coefficient, resulting in low Γ and high VπL. In addition, the propagation loss of the PE channel waveguide increased with D [18]. So D should be limited to some small values in the fabrications. Finally, large T would make more light energy guide in the untouched LN film, enabling significant E-O effect. However, as shown in [18], T should be selected below 0.72 μm to avoid optical leakage. In addition, when T was larger than 0.59 μm, the quasi-TE01 mode with two electric field intensity peaks in the vertical direction would exist. The upper intensity peak mainly located in the PE region with low E-O coefficient and so Γ would much smaller than that of the fundamental mode. As a result, VπL would be high.Fig. 3 Dependence of VπL of the phase modulator on W in (a) 0.4 μm, (b) 0.5 μm, (c) 0.6 μm and (d) 0.7 μm thick LN film. Difference color lines corresponded to different D.
3. Experiment
In experiment, a 0.52 μm thick LNOI sample was selected to avoid the leakage loss and the appearance of quasi-TE01 mode. The sample was prepared by crystal ion slicing and wafer bonding technologies [28]. D and W were set to be 0.165 μm and 2.5 μm according to the simulations. G was set to be 10 μm. The fabrication process was as follows: A 150 nm thick chromium (Cr) film was deposited onto the LNOI surface by an electron-beam evaporator. Then the 2.5 μm wide open channels along the y-axis of the LN film were defined by photolithography. After wet etching the exposed Cr, W expanded to about 3.2 μm due to the lateral erosion of Cr. After removing the residual photoresist, the sample was immersed in the benzoic acid for 3 minutes at 200°C. This PE process led to D = 0.165 μm. Then, 10 μm wide open channels were aligned with the above channel waveguides and patterned by a second photolithography process, followed by wet etching the exposed Cr. Finally, the two end-faces of the waveguide were polished to facilitate the end-face coupling. The fabrication process is shown in Fig. 4(a). Figure 4(b) shows the scanning electron microscope (SEM) of the cross-section of the phase modulator. The cross-section was obtained by focused ion beam (FIB) etching and the etching depth was about 1.1 μm. So there was a boundary about 1.1 μm under the surface. Transmission-Mode (light was injected from the back of the sample and then collected by the objective) of an optical microscope was used to show the structure of the phase modulator in Fig. 4(c). A dark strip (PE waveguide) could be seen between the two electrodes. The whole length of the waveguide and the electrodes were 2.3 mm.
Fig. 4 (a) Fabrication procedure of the PE phase modulator in LNOI. (b) SEM image of the cross-section (etched by FIB) of the phase modulator. (c) Optical microscope image (captured in the Transmission-mode) of the top view of the phase modulator.
The waveguide with the two polished end-faces formed a Fabry-Perot (FP) cavity. The applied electric field led to the variation of the refractive index, and so the optical phase through the waveguide changed. This would cause the shift of the FP resonant wavelength. Therefore, the E-O effect could be measured by detecting the movement of the FP resonant peak [29]. A linear polarized light at 1.55 μm was emitted from a tunable semiconductor laser and transmitted through a polarization maintaining fiber. Then the light was rotated to be TE polarized by a rotator and coupled into the waveguide through the lensed tip at the end of the fiber. A 40 × objective was used to collected the light from the waveguide and focused it to a germanium photodiode. The transmissions of the phase modulator before (black curve) and after (red curve) applying a 12 V voltage were shown in Fig. 5. The wavelength difference of neighboring resonances (Δλ) was 0.228 nm and the applied voltage induced a 0.097 nm shift of the resonant wavelength (Δλ'). So the Vπ of the phase modulator, that was the voltage needed to induce an optical phase change of π after transmitting through the waveguide, was 12 × Δλ/Δλ' = 26.1 V and thus VπL was 6.5 V·cm. The wavelength shifts by other voltages (including inversion and different values) were also measured and their VπL had similar values. The propagation loss of the waveguide, including the electrode absorption, was evaluated to be 3.5 dB/cm, which was larger than the previous reported waveguide without electrodes [18]. The thinner LN film we used (0.52 μm) made more light energy confined in the PE region, probably leading to a higher loss. The absorption loss caused by the electrodes was less than 1 dB/cm in the simulation. However, as shown in Fig. 4(c), due to technological constraints, the electrodes was not perfectly aligned with the waveguide (the waveguide was closer to the right electrode), resulting in an additional absorption loss. The effective E-O coefficient was evaluated by Eq. (1) by replacing r33 with reff and assuming the whole waveguide area (LN film and PE region) had uniform E-O coefficient reff. The obtained value was reff = 29.5 pm/V, which was very close to that of the bulk material, indicating the preservation of the excellent E-O property in LNOI after the ion-slicing and PE process. In some previous works, different levels of recovery of E-O coefficient after anneal were reported in PE of bulk LN. For example, in [19], there was no degradation of the E-O coefficient in the anneal proton exchanged (APE) sample, while, in [20], a restoration of up to 75% of the E-O coefficient of the bulk material was obtained. The method of short time PE in LNOI was simpler than the APE process. It allowed most light to guide in the untouched LN film to avoid the influence of the PE region. If we assumed that the E-O coefficients were 0 and r33 in the PE region and LN film, respectively, the calculated VπL by Eq. (1) was 8.4 pm/V, which was larger than the measurement (6.5 pm/V). The discrepancy could be attributed to the following aspects: 1. The E-O coefficient probably did not decreased to zero in the PE region. 2. D was probably smaller than 0.165 μm due to the fabrication errors like the PE temperature and the deviation of the diffusion coefficient we used in calculation from [24]. 3. The lattice strain had influence on the E-O coefficient [30]. There might be some strain in the LN film due to the stress at the interface between LN film and SiO2, arising from the different thermal expansion coefficients of the two materials. The strain might change the E-O coefficient. 4. The numerical errors in evaluating the overlap factor Γ would also make some contribution.
Fig. 5 (a) Measured transmissions of the phase modulator before (black) and after (red) applying 12 V voltage. A 0.097 nm shift of the resonant wavelength (Δλ') occurred. The wavelength difference of neighboring resonances (Δλ) was 0.228 nm.
4. Conclusions
In conclusions, the E-O effect was studied in the PE LNOI phase modulator which consisted of a PE waveguide and two electrodes on its two sides. Short time (3 minutes) PE was performed to fabricate waveguide with a very shallow PE layer. This process made most energy of the optical mode transmit in the untouched LN film, avoiding the influence of the PE region with high loss and low E-O coefficient. The waveguide with two polished end-faces formed a FP cavity. Through detecting the shift of the FP resonant peak by the applied voltage, VπL was measured to be 6.5 V·cm, and the effective E-O coefficient was evaluated to be 29.5 pm/V. If such design was applied to the MZ modulators with push-pull type electrodes, VπL would be as low as 3.25 V·cm, which was much smaller than the conventional Ti diffused LN MZ modulator [31]. Also, the high E-O coefficient indicated the good quality of the ion sliced single-crystal LN film. The low VπL phase modulator had potential to realizing high sensitivity sensors, and the efficient tuning of the optical phase can be used to enable phase matching between different waveguide modes to realize high conversion efficiency and small footprint wavelength convertors.
Acknowledgments
This work is supported by the National Natural Science Foundation of China (Grants No. 11275116, No. 61575111 and No. 11475105). We would like to thank Dr. G. Lee and Dr. R. Yin for their helpful discussions.
References and links
1. L. Arizmendi, “Photonic applications of lithium niobate crystals,” Phys. Status Solidi, A Appl. Res. 201(2), 253–283 (2004). [CrossRef]
2. S. Diziain, R. Geiss, M. Zilk, F. Schrempel, E.-B. Kley, A. Tünnermann, and T. Pertsch, “Second harmonic generation in free-standing lithium niobate photonic crystal L3 cavity,” Appl. Phys. Lett. 103(5), 051117 (2013). [CrossRef]
3. X. Wang and C. K. Madsen, “Design of a hybrid As2S3-Ti:LiNbO3 optical waveguide for phase-matched difference frequency generation at mid-infrared,” Opt. Express 22(22), 27183–27192 (2014). [CrossRef] [PubMed]
4. H.-C. Huang, J. I. Dadap, G. Malladi, I. Kymissis, H. Bakhru, and R. M. Osgood Jr., “Helium-ion-induced radiation damage in LiNbO3 thin-film electro-optic modulators,” Opt. Express 22(16), 19653–19661 (2014). [CrossRef] [PubMed]
5. P. Rabiei and W. H. Steier, “Lithium niobate ridge waveguides and modulators fabricated using smart guide,” Appl. Phys. Lett. 86(16), 161115 (2005). [CrossRef]
6. H. Lu, B. Sadani, N. Courjal, G. Ulliac, N. Smith, V. Stenger, M. Collet, F. I. Baida, and M.-P. Bernal, “Enhanced electro-optical lithium niobate photonic crystal wire waveguide on a smart-cut thin film,” Opt. Express 20(3), 2974–2981 (2012). [CrossRef] [PubMed]
7. L. Chen, M. G. Wood, and R. M. Reano, “12.5 pm/V hybrid silicon and lithium niobate optical microring resonator with integrated electrodes,” Opt. Express 21(22), 27003–27010 (2013). [CrossRef] [PubMed]
8. H. Jin, F. M. Liu, P. Xu, J. L. Xia, M. L. Zhong, Y. Yuan, J. W. Zhou, Y. X. Gong, W. Wang, and S. N. Zhu, “On-Chip Generation and Manipulation of Entangled Photons Based on Reconfigurable Lithium-Niobate Waveguide Circuits,” Phys. Rev. Lett. 113(10), 103601 (2014). [CrossRef] [PubMed]
9. R. V. Gainutdinov, T. R. Volk, and H. H. Zhang, “Domain formation and polarization reversal under atomic force microscopy-tip voltages in ion-sliced LiNbO3 films on SiO2/LiNbO3 substrates,” Appl. Phys. Lett. 107(16), 162903 (2015). [CrossRef]
10. R. Geiss, S. Saravi, A. Sergeyev, S. Diziain, F. Setzpfandt, F. Schrempel, R. Grange, E. B. Kley, A. Tünnermann, and T. Pertsch, “Fabrication of nanoscale lithium niobate waveguides for second-harmonic generation,” Opt. Lett. 40(12), 2715–2718 (2015). [CrossRef] [PubMed]
11. P. Rabiei, J. Ma, S. Khan, J. Chiles, and S. Fathpour, “Heterogeneous lithium niobate photonics on silicon substrates,” Opt. Express 21(21), 25573–25581 (2013). [CrossRef] [PubMed]
12. A. Rao, A. Patil, J. Chiles, M. Malinowski, S. Novak, K. Richardson, P. Rabiei, and S. Fathpour, “Heterogeneous microring and Mach-Zehnder modulators based on lithium niobate and chalcogenide glasses on silicon,” Opt. Express 23(17), 22746–22752 (2015). [CrossRef] [PubMed]
13. A. Guarino, G. Poberaj, D. Rezzonico, R. Degl’Innocenti, and P. Günter, “Electro-optically tunable microring resonators in lithium niobate,” Nat. Photonics 1(7), 407–410 (2007). [CrossRef]
14. 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] [PubMed]
15. 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 microresonators using femtosecond laser micromachining,” Sci. Rep. 5, 8072 (2015). [CrossRef] [PubMed]
16. H. Hu, R. Ricken, and W. Sohler, “Lithium niobate photonic wires,” Opt. Express 17(26), 24261–24268 (2009). [CrossRef] [PubMed]
17. S. Li, L. Cai, Y. Wang, Y. Jiang, and H. Hu, “Waveguides consisting of single-crystal lithium niobate thin film and oxidized titanium stripe,” Opt. Express 23(19), 24212–24219 (2015). [CrossRef] [PubMed]
18. L. Cai, R. Kong, Y. Wang, and H. Hu, “Channel waveguides and y-junctions in x-cut single-crystal lithium niobate thin film,” Opt. Express 23(22), 29211–29221 (2015). [CrossRef] [PubMed]
19. P. G. Suchoski, T. K. Findakly, and F. J. Leonberger, “Stable low-loss proton-exchanged LiNbO3 waveguide devices with no electro-optic degradation,” Opt. Lett. 13(11), 1050–1052 (1988). [CrossRef] [PubMed]
20. I. Savatinova, S. Tonchev, R. Todorov, M. N. Armenise, V. M. N. Passaro, and C. C. Ziling, “Electro-Optic Effect in Proton Exchanged LiNbO3 and LiTaO3 Waveguides,” J. Lightwave Technol. 14(3), 403–409 (1996). [CrossRef]
21. A. Méndez, G. De la Paliza, A. Garcia-Cabanes, and J. M. Cabrera, “Comparison of the electro-optic coefficient r33 in well-defined phases of proton exchanged LiNbO3 waveguides,” Appl. Phys. B 73(5-6), 485–488 (2001). [CrossRef]
22. M. L. Bortz, L. A. Eyres, and M. M. Fejer, “Depth profiling of the d33 nonlinear coefficient in annealed proton exchanged LiNbO3 waveguides,” Appl. Phys. Lett. 62(17), 2012–2014 (1993). [CrossRef]
23. K. R. Parameswaran, R. K. Route, J. R. Kurz, R. V. Roussev, M. M. Fejer, and M. Fujimura, “Highly efficient second-harmonic generation in buried waveguides formed by annealed and reverse proton exchange in periodically poled lithium niobate,” Opt. Lett. 27(3), 179–181 (2002). [CrossRef] [PubMed]
24. J. M. M. M. de Almeida, “Design methodology of annealed H+ waveguides in ferroelectric LiNbO3,” Opt. Eng. 46(6), 064601 (2007). [CrossRef]
25. E. Strake, G. P. Bava, and I. Montrosset, “Guided modes of Ti: LiNbO3 channel waveguides: A novel quasi-analytical technique in comparison with the scalar finite-element method,” J. Lightwave Technol. 6(6), 1126–1135 (1988). [CrossRef]
26. D. Petousi, L. Zimmermann, A. Gajda, M. Kroh, K. Voigt, G. Winzer, B. Tillack, and K. Petermann, “Analysis of optical and electrical tradeoffs of traveling-wave depletion-type Si Mach-Zehnder modulators for high-speed operation,” IEEE J. Sel. Top. Quantum Electron. 21(4), 3400108 (2015). [CrossRef]
27. M. García-Granda, H. Hu, J. R. García, and W. Sohler, “Design and fabrication of navel ridge guide modulators in lithium niobate,” J. Lightwave Technol. 27(24), 5690–5697 (2009).
28. G. Poberaj, H. Hu, W. Sohler, and P. Günter, “Lithium niobate on insulator (LNOI) for micro-photonic devices,” Laser Photonics Rev. 6(4), 488–503 (2012). [CrossRef]
29. S. Rao, “Hydrogenated amorphous silicon phase-change device based on a p-i-p waveguiding configuration,” Opt. Laser Technol. 53, 17–21 (2013). [CrossRef]
30. M. Minakata, K. Kumagai, and S. Kawakami, “Lattice constant changes and electro-optic effects in proton-exchanged LiNbO3 optical waveguides,” Appl. Phys. Lett. 49(16), 992–994 (1986). [CrossRef]
31. J. Kondo, A. Kondo, K. Aoki, S. Takatsuji, O. Mitomi, M. Imaeda, Y. Kozuka, and M. Minakata, “High-speed and low-driving-voltage X-cut LiNbO3 optical modulator with two step backside slot,” Electron. Lett. 38(10), 472–473 (2002). [CrossRef]
