Open Access
Add to BibSonomy
Abstract
Periodically poled second-order nonlinear materials with submicrometer periods are important for the development of quasi-phase matched backward-wave nonlinear optical processes. Interactions involving counter-propagating waves exhibit many unique properties and enable devices such as backward second harmonic generators, mirrorless optical parametric oscillators, and narrow-band quantum entangled photon sources. Fabrication of dense ferroelectric domain gratings in lithium niobate remains challenging, however, due to lateral domain spreading and merging. Here, we report submicrometer periodic poling of ion-sliced x-cut magnesium oxide doped lithium niobate thin films. Electric-field poling is performed using multiple bipolar preconditioning pulses that improve the poling yield and domain uniformity. The internal field is found to decrease with each preconditioning poling cycle. The poled domains are characterized by piezoresponse force microscopy. A fundamental period of 747 nm is achieved.
© 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Periodically poled ferroelectric materials enable quasi-phase matching (QPM) of nonlinear optical processes such as second harmonic generation (SHG). The period of the inverted domains determines the wavelength of the generated light and is typically several to tens of micrometers in the conventional co-propagation configuration, where all waves (pump, signal, and idler) travel in the same direction [1]. If the poling period is small enough, however, QPM interactions involving counter-propagating waves are possible. These backward QPM (B-QPM) processes display interesting phenomenon which are important for many applications including wavelength conversion, all-optical signal processing, and quantum optics. For instance, B-QPM processes inherently have narrow-band outputs and low sensitivity to temperature compared to the typical co-propagating configuration [2]. B-QPM processes also give rise to a built-in distributed feedback that enables optical parametric oscillators (OPOs) without the need for an external cavity [3], and is predicted to induce optical bistability, limiting, and self-pulsing in backward second harmonic generators (B-SHG) [4–6]. Other B-SHG configurations, utilizing two counter-propagating pumps, can have higher conversion efficiencies than co-propagating schemes [7], and have been studied for efficient all-optical switching [8,9], and soliton generation [10]. B-QPM interactions have also been proposed for pulse shaping [11], and slow light generation [12]. For quantum optic applications, spontaneous parametric down conversion (SPDC) with counter-propagating signal and idler photons has been investigated for single photon sources [13,14], and recently used to demonstrate the generation of photon pairs [15] and quantum entanglement [16]. In addition to nonlinear QPM devices, submicrometer poling enables electrically tunable Bragg reflectors [17], and photonic band gaps [18].
Fabrication of nanoscale domains, suitable for counter-propagating interactions, is challenging. Poling typically spans several centimeters in length and requires a uniform domain size with 50% duty cycle and sufficient depth to maximize conversion efficiency. On the nanoscale, crystalline defects, poling electrode imperfections, and fringing electric-fields all lead to domain spreading and non-uniformities. Significant progress has been made in poling KTiOPO4 (KTP) and its isomorphs, enabling B-SHG [19,20], and the first mirrorless OPO [21], with periods currently reaching as small as 500 nm [22]. The ability to pole KTP is aided in part by its chiral crystal structure that promotes anisotropic domain growth [23]. In bulk lithium niobate (LN), a number of poling techniques have been reported including the electric field method with various waveforms such as backswitching [24], surface “over-poling” [25], and multiple unipolar pulses [26], as well as direct-write methods using electron-beams [27], ion-beams [28], and scanning probe microscopy [29]. B-SHG has been achieved in bulk LN [30–33], but was limited to 16th-order QPM with 3.3 µm periods. Recently, 5th-order QPM counter-propagating SPDC was reported in Ti in-diffused LN waveguides with 1.7 µm poling periods [15]. As the conversion efficiency scales as m-2, where m is the QPM order, there is a desire to continue reducing the poling period.
Over the last decade, the thin film LN-on-insulator (LNOI) platform has become popular for photonic integrated circuits. The single-crystal films, formed by ion-slicing, have submicrometer thicknesses and provide high optical confinement which enhances the nonlinear conversion efficiency and reduces the domain aspect ratio compared to bulk LN. Isolated domains with 50 nm diameter and small area poled gratings with 300 nm period have been created in z-cut LNOI using scanning-probe microscopy [34]. Poled gratings with 2.4 µm period, suitable for co-propagating QPM, have been reported in x-cut LNOI using electric field poling [35]. Recently, we demonstrated electric field poling of x-cut magnesium oxide doped LN (MgO:LN) thin films with 7.5 µm period using a SiO2 insulation layer under the electrodes to reduce leakage current [36]. Here, we utilize a bipolar multi-pulse poling waveform to reduce the period by one order of magnitude [37].
2. Methods
The fabricated LNOI samples, shown in Fig. 1, consists of a 700 nm thick (t1) x-cut congruent 5 mol% MgO:LN thin film bonded to a LN substrate with 1 µm of SiO2 (t2). Poling electrodes are formed from 100 nm thick chromium on a 100 nm thick SiO2 insulation layer by electron beam lithography and plasma etching. An array of different electrode designs is fabricated with periods (Λ) equal to 1500 nm, 750 nm, and 375 nm and gap widths (G) equal to 15 µm, 10 µm, and 6 µm. All of the electrodes are 80 µm in length (L) with 40% duty cycle and have rounded tips to promote nucleate on at the apex. The gap widths G are chosen to minimize the domain length while also being able to accommodate strip-loaded or rib waveguides without inducing significant optical loss.
Fig. 1. Schematic of poling electrodes on 700 nm thick x-cut MgO:LN thin film. A 100 nm thick SiO2 insulation layer is located under the Cr electrodes.
Poling is performed at room temperature with the sample covered with a few drops of silicone oil to prevent dielectric breakdown of air. Electrical contact probes are used to apply the voltage pulses to the electrodes. The capacitive charging current of the circuit is measured without contacting the sample and is subtracted out to isolate the poling current. Details of the poling circuit and device fabrication are reported elsewhere [36]. After poling, the domains are imaged by piezoresponse force microscopy (PFM) using platinum coated silicon probes on the + x surface.
3. Results
We first attempted poling of devices with a nominal period of 750 nm by applying a single pulse, as used in our prior work with micrometer-scale periods [36]. It was found that in some regions along the length of the electrode, the domains fully merged together, while in other areas no poling occurred. Increasing the voltage or pulse duration could improve the yield, but the degree of domain merging worsened. The best results were achieved by applying the waveform shown in Fig. 2(a) to an electrode with a gap width G of 15 µm. The voltage ramps up to 510 V in 1 ms, then is held for 1 ms, and then ramped down in 5 ms. A 25-point moving average filter has been applied to the current in Fig. 2(a) to reduce the noise. The resulting domains imaged by PFM are shown in Fig. 2(b) in which the dark regions have been inverted. Our testing found the different gap widths have little to no impact on the domain quality after scaling the voltage. The poling voltage agrees with the semi-empirical Janovec-Kay-Dunn (JKD) scaling law (coercive field Ec $\propto$ G-2/3 [38,39]).
Fig. 2. Poling waveform and piezoresponse force microscope image of inverted domains on + x surface formed by a single pulse. The electrode has a 750 nm period Λ and 15 µm gap width G. Cr electrodes are visible at the top and bottom. The dark vertical bands are the inverted domains.
To overcome the tradeoff between yield and domain merging, we adopted a multi-pulse preconditioning technique which has been shown to improve the nucleation uniformity and reproducibility in KNbO3 [40]. In this method, the sample is poled forward (+V) and back (-V) repeatedly prior to a single forward poling pulse. For the 10 µm gap G devices, we use a series of 10 symmetrical ±560 V preconditioning cycles, followed by a poling pulse with voltage VP as shown in Fig. 3. The ramp up, ramp down, and holding times are all 1 ms.
Fig. 3. Measured poling voltage and current showing the first and last preconditioning cycles, followed by a final poling pulse with voltage VP. The electrode has a 750 nm period Λ and 10 µm gap width G.
The domains were imaging by PFM after the forward poling segment of cycle 1 and verified that the voltage is large enough to cause the domains to fully merge together. PFM imaging also verified that the poling is fully erased after all 10 preconditioning cycles. Therefore, the final single pulse is responsible for forming the domain grating. The voltage at which the peak current occurs decreases with each preconditioning cycle, indicating a reduced Ec that allows for a relatively low voltage final pulse. To optimize the voltage of the final pulse, a series of devices with 1500 nm nominal period Λ were poled at different voltages VP. Figure 4 shows the poling current, transferred charge (Q), and poled area (A) steadily decreases with voltage. The charges Q are found by integrating the current from 63 to 66 ms and are in relatively good agreement with the values predicted by Q = 2PsA considering the level of noise (spontaneous polarization Ps = 80 µC/cm2 [41]). For instance, the device in Fig. 4(e)-(f) with a measured Q of 45 pC has an average duty cycle of 64% and therefore the predicted Q is 57 pC (A = 80 µm length L × 64% duty cycle × 700 nm film thickness).
Fig. 4. Poling waveforms and PFM images of domains on + x surface formed by 10 ± 560 V preconditioning cycles followed by a final pulse with voltages VP. Only the final pulse is shown in the waveforms. Electrodes have a period Λ of 1500 nm and gap width G of 10 µm. (a)-(b) VP = 325 V, (c)-(d) VP = 280 V, (e)-(f) VP = 235 V, (g)-(h) VP = 190 V.
The optimized waveform with VP equal to 235 V was applied to a 750 nm period Λ electrode. The measured voltage and current waveforms are shown in Fig. 3 and (a) PFM image of the domains are shown in Fig. 5(a)-(b). The electrode tips are visible at the top and bottom of the PFM image, but are not fully resolved by the probe tip. A scanning electron microscopy (SEM) image of the electrode tips is shown in Fig. 5(c). The domains have an average duty cycle of 56% and standard deviation of 14%. The waveform was also applied to devices with a 375 nm period Λ, but the domains were largely merged together.
Fig. 5. (a) PFM image of inverted domains on + x surface across full electrode length L with 750 nm nominal period Λ. (b) Zoomed-in view of domains. Electrode tips at top and bottom of image are not fully resolved by AFM probe. (c) SEM image of electrode tip.
The domain gratings produced by a single pulse, Fig. 2, and by preconditioning pulses, Fig. 5(a), are characterized using the Fourier transform method [42]. In the ideal case, the spatially varying nonlinear coefficient, d(z), is a square wave with 50% duty cycle and amplitude of ±1 whose Fourier series is given by
Fig. 6. Fourier spectrums computed from PFM data slice across entire 80 µm long electrode of a device poled with a single pulse and multiple preconditioning pulses. Inset: Zoomed-in view of fundamental component corresponding to a poling period of 747 nm.
Polarization hysteresis curves shown in Fig. 7(a) help reveal the effects of the preconditioning cycles. The curves result from integrating the poling current (I) over each switching cycle and dividing by the domain-inverted area (A) i.e. P = A-1 ∫ I dt where A = 80 µm length L × 700 nm LN film thickness t1. The electric fields are approximated by dividing the applied voltage by the 10 µm gap width G. The coercive field in the forward (EF) and reverse (ER) poling directions are the magnitude of the positive and negative horizontal-intercepts, respectively. Each curve in Fig. 7(a) is an average of 26 devices from the same LNOI sample. The first cycle shows the largest EF which is decreased significantly on the second cycle. The coercive field in congruent LN is determined by the pinning of domain walls from defects [43], and therefore the reduced EF implies less pinning and smoother domain movement. The reduced EF is important for achieving submicrometer poling periods.
Fig. 7. (a) Ferroelectric polarization hysteresis resulting from integration of poling currents during each preconditioning voltage cycle. (b) Coercive field in the forward (EF) and reverse (ER) poling directions and the internal field, equal to (EF - ER)/2. The data is an average of 26 devices and the error bars show the standard deviation. The hysteresis curves in (a) show that the charge (Q = P/A) in the forward and backward switching directions are nearly the same and remain constant over the 10 cycles.
Figure 7(a) also shows that the entire loop shifts to the left along the horizontal-axis with each additional cycle. The offset of the hysteresis loop from the origin is characterized by the internal field (Eint), equal to (EF - ER)/2, and is plotted in Fig. 7(b) along with the coercive fields from each cycle. The plot shows EF decreases by 28% while ER increases only 11%, producing an overall decrease in Eint. When the first preconditioning cycle, with anomalous behavior, is excluded, EF still changes nearly twice as much as ER. The hysteresis curves also show asymmetry between the poling directions. In the forward direction, the polarization increases sharply after exceeding a threshold field, while in the reverse direction, the polarization decreases more gradually beginning near zero electric field.
4. Discussion
The internal field is known to be related to point defects and its magnitude varies with stoichiometry, doping, UV illumination and temperature [44]. The internal field is typically in the range of 2-3 V/µm in bulk congruent LN [45], and about 0.5 V/µm in bulk 5 mol% MgO:LN [46]. In our devices, the internal field is 15.0 ± 1.5 V/µm in cycle 1 and reaches 7.3 ± 0.5 V/µm at cycle 10, however, Eint also depends on the time between forward and reverse poling [41]. LN thin films have been reported to exhibit larger internal fields than bulk [47].
In congruent LN, there is good agreement that the Li deficiency causes intrinsic defects involving a niobium antisite $({Nb_{Li}^{4 + }} )$ compensated by four lithium vacancies $({V_{Li}^ - } )$, however, their particular configuration is still debatable [48]. The defect configuration in 5 mol% MgO:LN, as used in this work, is not as well studied, but Mg ions are known to replace the Nb antisites and the most probable defect complexes are $({4Mg_{Li}^{2 + } + 4{V_{Li}}} )$ [49]. Kim et al. [50], proposed that point defect complexes in intrinsic LN have electrical dipoles which prefer to align with the spontaneous polarization (Ps). After forward poling, the direction of Ps is inverted and the defect dipoles begin to realign with Ps into a high energy frustrated state which stabilizes the domains. Reverse poling before the dipoles have finished realigning will result in a lower coercive field (ER) and a change in internal field. A similar dipole defect model, referred to as the volume effect, describes de-aging in polycrystalline ferroelectric (FE) ceramics [51]. In de-aging, bipolar cycling leads to a dynamic equilibrium between defect dipoles that are aligned parallel and anti-parallel to Ps. Hirohashi et al. used the concept of reaching dynamic equilibrium to explain a very similar decrease in internal field observed during bipolar cycling of KNbO3 [40]. Hirohashi et al. also reported similar trends of coercive field in a number of other z-cut materials including 1 mol% MgO-doped and non-doped stoichiometric LiTaO3, 1 mol% MgO-doped stoichiometric LN, and KTP [52].
Moreover, while point defects certainly have a role, they do not fully explain the observations in Fig. 7. Surfaces, interfaces, and screening charges are also important. Shur et al. [53], explains that the internal field is a result of incomplete screening of the self-generated depolarization field which arises from the discontinuity of Ps at crystal surfaces. An intrinsic non-ferroelectric surface layer below electrodes prevents complete compensation of the depolarization field by fast redistribution of charges on electrodes. The residual depolarization field is instead compensated by slower bulk mechanisms including the reorientation of dipole defects, the injection of charge carriers through the surface layer, and movement of charge carriers made possible by bulk conductivity.
Interestingly, ferromagnetic (FM) materials experience a training effect with repeated switching and show remarkably similar trends to those observed in Fig. 7, where the exchange bias is analogous to the internal field in ferroelectrics [54]. An exchange bias arises at interfaces of FM and anti-FM layers, which pin magnetic dipoles. The training effect is characterized by a decreasing exchange bias with cycle number, asymmetric hysteresis loops, a larger decrease of the forward coercive field than the reverse, and an anomalous first cycle. The similarities between FM materials and the FE thin films used in this work suggest that our observed trends stem from more general phenomena and also reinforce that surfaces and interfaces are important. Continuing to understand the connections between FE and FM materials may help to provide a better understanding of the mechanism behind the reduced internal field and improved poling quality achieved with preconditioning pulses.
5. Conclusion
In summary, we have periodically poled MgO:LN thin films and achieved a fundamental period of 747 nm by utilizing multiple bipolar preconditioning poling cycles. The preconditioning pulses were found to reduce the forward coercive field which produced less domain wall pinning and resulted in improved poling yield and domain uniformity, compared to a single pulse. The preconditioning cycling also reduced the internal field and followed a trend very similar to the training effect in ferromagnetic materials. Future work that focuses on a further understanding of the defect configurations and screening mechanisms may help to optimize the preconditioning pulse waveform and allow for even smaller poling periods. Quasi-phase matched backward-wave nonlinear interactions are envisioned in photonic integrated circuits with applications in wavelength conversion, all-optical signal processing, and quantum optics.
Funding
National Science Foundation (1809894).
Disclosures
The authors declare no conflicts of interest.
References
1. M. M. Fejer, “Nonlinear frequency conversion in periodically-poled ferroelectric waveguides,” in Guided Wave Nonlinear Optics, D. B. Ostrowsky and R. Reinisch, eds. (Springer Netherlands, 1992), pp. 133–145. [CrossRef]
2. V. Pasiskevicius, G. Strömqvist, F. Laurell, and C. Canalias, “Quasi-phase matched nonlinear media: Progress towards nonlinear optical engineering,” Opt. Mater. 34(3), 513–523 (2012). [CrossRef]
3. S. E. Harris, “Proposed backward wave oscillation in the infrared,” Appl. Phys. Lett. 9(3), 114–116 (1966). [CrossRef]
4. M. Matsumoto and K. Tanaka, “Quasi-phase-matched second-harmonic generation by backward propagating interaction,” IEEE J. Quantum Electron. 31(4), 700–705 (1995). [CrossRef]
5. G. D’Alessandro, P. S. J. Russell, and A. A. Wheeler, “Nonlinear dynamics of a backward quasi-phase-matched second-harmonic generator,” Phys. Rev. A 55(4), 3211–3218 (1997). [CrossRef]
6. M. Lauritano, A. Parini, G. Bellanca, S. Trillo, M. Conforti, A. Locatelli, and C. D. Angelis, “Bistability, limiting, and self-pulsing in backward second-harmonic generation: a time-domain approach,” J. Opt. A: Pure Appl. Opt. 8(7), S494–S501 (2006). [CrossRef]
7. Y. J. Ding and J. B. Khurgin, “Second-harmonic generation based on quasi-phase matching: a novel configuration,” Opt. Lett. 21(18), 1445–1447 (1996). [CrossRef]
8. G. D. Landry and T. A. Maldonado, “Switching and second harmonic generation using counterpropagating quasi-phase-matching in a mirrorless configuration,” J. Lightwave Technol. 17(2), 316–327 (1999). [CrossRef]
9. G. D. Landry and T. A. Maldonado, “Efficient nonlinear phase shifts due to cascaded second-order processes in a counterpropagating quasi-phase-matched configuration,” Opt. Lett. 22(18), 1400–1402 (1997). [CrossRef]
10. K. Y. Kolossovski, A. V. Buryak, and R. A. Sammut, “Quadratic solitary waves in a counterpropagating quasi-phase-matched configuration,” Opt. Lett. 24(12), 835–837 (1999). [CrossRef]
11. M. Conforti, C. de Angelis, U. K. Sapaev, and G. Assanto, “Pulse shaping via backward second harmonic generation,” Opt. Express 16(3), 2115–2121 (2008). [CrossRef]
12. J. B. Khurgin, “Slowing and stopping photons using backward frequency conversion in quasi-phase-matched waveguides,” Phys. Rev. A 72(2), 023810 (2005). [CrossRef]
13. A. Christ, A. Eckstein, P. J. Mosley, and C. Silberhorn, “Pure single photon generation by type-I PDC with backward-wave amplification,” Opt. Express 17(5), 3441–3446 (2009). [CrossRef]
14. C.-S. Chuu and S. E. Harris, “Ultrabright backward-wave biphoton source,” Phys. Rev. A 83(6), 061803 (2011). [CrossRef]
15. K.-H. Luo, V. Ansari, M. Massaro, M. Santandrea, C. Eigner, R. Ricken, H. Herrmann, and C. Silberhorn, “Counter-propagating photon pair generation in a nonlinear waveguide,” Opt. Express 28(3), 3215–3225 (2020). [CrossRef]
16. Y.-C. Liu, D.-J. Guo, R. Yang, C.-W. Sun, J.-C. Duan, Y.-X. Gong, and S.-N. Zhu, “Narrow-band photonic quantum entanglement with counterpropagating domain engineering,” arXiv:1905.13395 (2019).
17. C. Canalias, V. Pasiskevicius, R. Clemens, and F. Laurell, “Submicron periodically poled flux-grown KTiOPO4,” Appl. Phys. Lett. 82(24), 4233–4235 (2003). [CrossRef]
18. Z. Zhou, J. Shi, and X. Chen, “Electrically induced and tunable photonic band gap in submicron periodically poled lithium niobate,” Appl. Phys. B 96(4), 787–791 (2009). [CrossRef]
19. X. Mu, I. B. Zotova, Y. J. Ding, and W. P. Risk, “Backward second-harmonic generation in submicron-period ion-exchanged KTiOPO4 waveguide,” Opt. Commun. 181(1-3), 153–159 (2000). [CrossRef]
20. C. Canalias, V. Pasiskevicius, M. Fokine, and F. Laurell, “Backward quasi-phase-matched second-harmonic generation in submicrometer periodically poled flux-grown KTiOPO4,” Appl. Phys. Lett. 86(18), 181105 (2005). [CrossRef]
21. C. Canalias and V. Pasiskevicius, “Mirrorless optical parametric oscillator,” Nat. Photonics 1(8), 459–462 (2007). [CrossRef]
22. R. S. Coetzee, A. Zukauskas, C. Canalias, and V. Pasiskevicius, “Low-threshold, mid-infrared backward-wave parametric oscillator with periodically poled Rb:KTP,” APL Photonics 3(7), 071302 (2018). [CrossRef]
23. A. Zukauskas, G. Strömqvist, V. Pasiskevicius, F. Laurell, M. Fokine, and C. Canalias, “Fabrication of submicrometer quasi-phase-matched devices in KTP and RKTP [Invited],” Opt. Mater. Express 1(7), 1319–1325 (2011). [CrossRef]
24. R. G. Batchko, V. Y. Shur, M. M. Fejer, and R. L. Byer, “Backswitch poling in lithium niobate for high-fidelity domain patterning and efficient blue light generation,” Appl. Phys. Lett. 75(12), 1673–1675 (1999). [CrossRef]
25. A. C. Busacca, C. L. Sones, V. Apostolopoulos, R. W. Eason, and S. Mailis, “Surface domain engineering in congruent lithium niobate single crystals: A route to submicron periodic poling,” Appl. Phys. Lett. 81(26), 4946–4948 (2002). [CrossRef]
26. K. Mizuuchi, A. Morikawa, T. Sugita, and K. Yamamoto, “Efficient second-harmonic generation of 340-nm light in a 1.4-µm periodically poled bulk MgO:LiNbO3,” Jpn. J. Appl. Phys. 42(Part 2, No. 2A), L90–L91 (2003). [CrossRef]
27. E. O. Vlasov, D. S. Chezganov, L. V. Gimadeeva, E. A. Pashnina, E. D. Greshnyakov, M. A. Chuvakova, and V. Y. Shur, “E-beam domain patterning in thin plates of MgO-doped LiNbO3,” Ferroelectrics 542(1), 85–92 (2019). [CrossRef]
28. D. S. Chezganov, V. Y. Shur, E. O. Vlasov, L. V. Gimadeeva, D. O. Alikin, A. R. Akhmatkhanov, M. A. Chuvakova, and V. Y. Mikhailovskii, “Influence of the artificial surface dielectric layer on domain patterning by ion beam in MgO-doped lithium niobate single crystals,” Appl. Phys. Lett. 110(8), 082903 (2017). [CrossRef]
29. G. Rosenman, P. Urenski, A. Agronin, Y. Rosenwaks, and M. Molotskii, “Submicron ferroelectric domain structures tailored by high-voltage scanning probe microscopy,” Appl. Phys. Lett. 82(1), 103–105 (2003). [CrossRef]
30. X. Gu, R. Y. Korotkov, Y. J. Ding, J. U. Kang, and J. B. Khurgin, “Backward second-harmonic generation in periodically poled lithium niobate,” J. Opt. Soc. Am. B 15(5), 1561–1566 (1998). [CrossRef]
31. J. U. Kang, Y. J. Ding, W. K. Burns, and J. S. Melinger, “Backward second-harmonic generation in periodically poled bulk LiNbO3,” Opt. Lett. 22(12), 862–864 (1997). [CrossRef]
32. S. Stivala, A. C. Busacca, L. Curcio, R. L. Oliveri, S. Riva-Sanseverino, and G. Assanto, “Continuous-wave backward frequency doubling in periodically poled lithium niobate,” Appl. Phys. Lett. 96(11), 111110 (2010). [CrossRef]
33. A. C. Busacca, S. Stivala, L. Curcio, A. Tomasino, and G. Assanto, “Backward frequency doubling of near infrared picosecond pulses,” Opt. Express 22(7), 7544–7549 (2014). [CrossRef]
34. T. Volk, R. Gainutdinov, and H. Zhang, “Domain patterning in ion-sliced LiNbO3 films by atomic force microscopy,” Crystals 7(5), 137 (2017). [CrossRef]
35. L. Chang, Y. Li, N. Volet, L. Wang, J. Peters, and J. E. Bowers, “Thin film wavelength converters for photonic integrated circuits,” Optica 3(5), 531–535 (2016). [CrossRef]
36. J. T. Nagy and R. M. Reano, “Reducing leakage current during periodic poling of ion-sliced x-cut MgO doped lithium niobate thin films,” Opt. Mater. Express 9(7), 3146–3155 (2019). [CrossRef]
37. J. T. Nagy and R. M. Reano, “Fabricating periodically poled lithium niobate thin films with sub-micrometer fundamental period,” in Frontiers in Optics + Laser Science APS/DLS (Optical Society of America, 2019), p. JTu3A.10. [CrossRef]
38. H. F. Kay and J. W. Dunn, “Thickness dependence of the nucleation field of triglycine sulphate,” Philos. Mag. 7(84), 2027–2034 (1962). [CrossRef]
39. V. Janovec, “On the theory of the coercive field of single-domain crystals of BaTiO3,” Czech. J. Phys. 8(1), 3–15 (1958). [CrossRef]
40. J. Hirohashi, K. Yamada, H. Kamio, and S. Shichijyo, “Embryonic nucleation method for fabrication of uniform periodically poled structures in potassium niobate for wavelength conversion devices,” Jpn. J. Appl. Phys. 43(2), 559–566 (2004). [CrossRef]
41. M.-L. Hu, L.-J. Hu, and J.-Y. Chang, “Polarization switching of pure and MgO-doped lithium niobate crystals,” Jpn. J. Appl. Phys. 42(Part 1, No. 12), 7414–7417 (2003). [CrossRef]
42. X. J. lv, L. N. Zhao, J. Lu, G. Zhao, H. Liu, Y. Q. Qin, and S. N. Zhu, “Poling quality evaluation of optical superlattice using 2D Fourier transform method,” Opt. Express 17(20), 18241–18249 (2009). [CrossRef]
43. D. Lee, V. Gopalan, and S. R. Phillpot, “Depinning of the ferroelectric domain wall in congruent LiNbO3,” Appl. Phys. Lett. 109(8), 082905 (2016). [CrossRef]
44. H. Steigerwald, “Influence of UV light and heat on the ferroelectric properties of lithium niobate crystals,” Ph.D. Dissertation, University of Bonn (2011).
45. M. de Angelis, S. De Nicola, A. Finizio, G. Pierattini, P. Ferraro, S. Grilli, and M. Paturzo, “Evaluation of the internal field in lithium niobate ferroelectric domains by an interferometric method,” Appl. Phys. Lett. 85(14), 2785–2787 (2004). [CrossRef]
46. J. H. Yao, Y. H. Chen, B. X. Yan, H. L. Deng, Y. F. Kong, S. L. Chen, J. J. Xu, and G. Y. Zhang, “Characteristics of domain inversion in magnesium-oxide-doped lithium niobate,” Phys. B 352(1-4), 294–298 (2004). [CrossRef]
47. 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]
48. Y. Kong, F. Bo, W. Wang, D. Zheng, H. Liu, G. Zhang, R. Rupp, and J. Xu, “Recent progress in lithium niobate: Optical damage, defect simulation, and on-chip devices,” Adv. Mater. 32(3), 1806452 (2020). [CrossRef]
49. A. V. Yatsenko, S. V. Yevdokimov, D. Y. Sugak, and I. M. Solskii, “Investigation of the defect complexes in highly Mg-doped LiNbO3 crystals by 93Nb NMR method,” Funct. Mater. 21(1), 31–35 (2014). [CrossRef]
50. S. Kim, V. Gopalan, K. Kitamura, and Y. Furukawa, “Domain reversal and nonstoichiometry in lithium tantalate,” J. Appl. Phys. 90(6), 2949–2963 (2001). [CrossRef]
51. Y. A. Genenko, J. Glaum, M. J. Hoffmann, and K. Albe, “Mechanisms of aging and fatigue in ferroelectrics,” Mater. Sci. Eng., B 192, 52–82 (2015). [CrossRef]
52. J. Hirohashi, “Characterization of domain switching and optical damage properties in ferroelectrics,” Ph.D. Dissertation, Royal Institute of Technology (2006).
53. V. Y. Shur, A. R. Akhmatkhanov, I. S. Baturin, M. S. Nebogatikov, and M. A. Dolbilov, “Complex study of bulk screening processes in single crystals of lithium niobate and lithium tantalate family,” Phys. Solid State 52(10), 2147–2153 (2010). [CrossRef]
54. A. Hochstrat, C. Binek, and W. Kleemann, “Training of the exchange-bias effect in NiO-Fe heterostructures,” Phys. Rev. B 66(9), 092409 (2002). [CrossRef]
