Open Access
Add to BibSonomyAbstract
Although a breakthrough in the fabrication of green laser diodes has occurred, the high costs associated with the difficulty of manufacture still present a great obstacle for its practical application. Another approach for producing a green laser, by combining a laser device and a nonlinear crystal, entails the fabrication of complex structures and exhibits unstable performance due to interface contact defects, thus limiting its application. In this work, we report the fabrication by domain engineering of high quality periodically poled LiNbO3, co-doped with Nd3+ and Mg2+, which combines a laser medium and a high efficiency second harmonic conversion crystal into a single system that is designed to overcome the above problems. An 80mW self-frequency doubling green laser was constructed for the first time from a periodically poled Nd:Mg:LiNbO3 crystal of 16 mm in length. This crystal can be used for developing compact, stable, highly efficient mini-solid-state-lasers, which promise to have many applications in portable laser-based spectroscopy, photo-communications, terahertz wave generation, and laser displays.
© 2015 Optical Society of America
1. Introduction
Since the first report of the successful operation of a laser in 1960 by Maiman [1], lasers have played an important role in our daily lives. Compared with other light sources, lasers have an extremely high degree of monochromaticity, coherence, directionality and brightness. Lasers are present everywhere around us in applications such as printers, DVDs, medical instruments, communications, radar, etc [2–4 ]. At the same time, the demand for new types of lasers seems never ending. The various demands include the need for enhancement of the operating mode, power, wavelength, efficiency, size, etc. Solid state lasers have attracted a great deal of attention due to their high efficiency, high power, small size and low cost. Solid state lasers cover wavelengths ranging from the deep-ultraviolet to the terahertz wave range. Among all these, visible lasers, especially in the red, green, and blue (RGB) are of great importance in our daily life and for military uses.
Developing an RGB laser with high efficiency and high power output is a current area of research interest. The laser diode (LD) was thought to be the most effective way to generate visible laser light. Nowadays, watt-level red and blue laser light can be generated from AsGaIn and GaN semiconductors, respectively [3]. This kind of light is so important that the blue LED and LD won the 2014 Nobel Prize for Physics. However, a high power green laser is still hard to produce using semiconductors. It is very difficult to find appropriate semiconductor materials for high power green light emission. Although a watt-level green LD was reported in 2013 [5], the device still exhibits many problems such as poor beam quality, large astigmatism, high mode stability, high heat load, low efficiency, unstable power output and wavelength, and most importantly, very high cost. There is a practical way to obtain green laser light by means of second harmonic generation from a diode pumped solid-state-laser (DPSSL). Nd3+ doped materials like Nd:YVO4 or Nd:Y3Al5O12 (Nd:YAG) can emit light at 1064 nm. Using nonlinear optical methods, the 1064 nm light can be converted to 532 nm, which is in the middle range of green light. The optical-to-optical and electrical-to-optical efficiencies of this kind of green laser reach 60% and 20%, respectively [6,7 ]. Almost all commercially available green lasers are frequency-doubling Nd:YVO4 or Nd:YAG lasers using a nonlinear crystal, such as LiBO3 or KTiOPO4. These lasers exhibit high power output and high efficiency at an acceptable cost. As a result, it is commercially widely used as green laser source. However, this type of laser typically has a large volume, large weight, and involves too many interfaces. For some portable laser devices, like laser projectors, laser rangefinders and laser radar, where the apparatus must be portable, this kind of laser is still too complicated and unstable to be useful. It is therefore necessary to develop a green lasers having a more compact volume. Self-frequency-doubling (SFD) represents an effective solution. The idea behind SFD is to dope rare earth ions (usually Nd3+ and Yb3+) into a suitable nonlinear optical crystals. By combining the laser host and the nonlinear optical process into a single crystal, SFD makes the laser more stable and more compact. The most successful SFD crystals are Nd:GdCa4O(BO3)3 (Nd:GdCOB) [8,9 ] and Yb:YAl3(BO3)4 (Yb:YAB) [10]. These two crystals can produce watt-level SFD green laser light with an optical-to-optical efficiency of about 15%. Nevertheless due to a small nonlinear coefficient, the length of the Nd:GdCOB and Yb:YAB crystals must be more than 10mm. SFD occurs only when the light passes along a specific direction, which poses difficulties in machining. Thus, it is necessary to search for another more suitable SFD crystal.
LiNbO3 (LN) is the first crystal that was observed to generate SFD laser radiation. T.Y. Fan achieved the first SFD laser with Nd:Mg:LiNbO3, which was heated to a phase-matching temperature of 152 °C [11]. A pulsed Nd:Mg:LiNbO3, self-frequency-doubling laser pumped by a Xe lamp was demonstrated at room temperature with SHG green output of 400 pJ/shot [12]. R.N. Li reported a 598 nm dye laser-pumped CW Nd:Mg:LiNbO3 self-frequency-doubling laser that operates at room temperature [13]. In all the above research the self-frequency-doubling laser can be obtained only at a special cut angle, called the phase-matching angle, and does not utilize the large nonlinear optical coefficient necessary to obtain high power second harmonic light. The main challenge in realizing SFD with Nd:Mg:LiNbO3 is to grow crystals of high quality and large size. To use the largest nonlinear optical coefficient, a periodically poled domain structure is usually fabricated for LN series crystals for satisfying quasi-phase-matching (QPM). X.Y. Chen [14] modeled self-frequency-conversion (SFC) lasers in a rare-earth doped super-lattice crystal (periodically poled domain structure), and predicted that high efficiency conversion lasers can be generated from a SFC laser emitting over the entire visible range. Periodically poled or aperiodically poled Nd:Mg:LiNbO3 was reported to generate visible light in the blue [14–19 ], green [15, 17, 20 ], yellow [21] and red [16] ranges. However, all the previously reported domain structures were fabricated using the off-center Czochralski method. It is very difficult to precisely control the periodically poled lithium niobate (PPLN) crystal period with this method, thus leading to a very low conversion efficiency. All the lasers generated from SFD in QPM were at output power levels of several microwatts or less. An effective way to improve the efficiency is to prepare the domain structure using an electric field-induced domain inversion technique. The only example of fabricating Nd:PPLN by electric-field-induced domain inversion was demonstrated by L. Barraco [22]. They reported optical self-oscillation in a 29.45μm period Nd:PPLN. A 1.55μm laser was obtained with 24% conversion efficiency. The period of Nd:PPLN in their experiment was several tens of micrometers, which is much larger than that used to generate visible light. The Nd:PPLN period used for the generation of visible light is normally several micrometers, and is much more difficult to fabricate. The main challenge in fabricating short period domains in Nd:Mg:LiNbO3 is the difficulty of the crystal growth technique. As a result, large size, high quality Nd:Mg:LiNbO3 samples have not yet been reported. Another problem is that domains often merge together during the poling process because of the small distance between the electrodes. Therefore, it is very difficult to obtain periodically poled Nd:Mg:LiNbO3 using the conventional domain inversion method. No research on periodically poled Nd:Mg:LiNbO3 has been able to achieve visible laser light via self-frequency-doubling or self-frequency-summing.
In this letter, we report that a large size Nd3+ and Mg2+ co-doped LiNbO3 crystal of 3 inches in diameter was successfully grown. Watt-level infrared lasers of different wavelength were for the first time constructed using a diode-end-pumped Nd:Mg:LiNbO3 crystal, demonstrating that this crystal is of high quality with excellent lasing properties. Using a speically designed electric field-induced domain inversion method, the domain merging problem was overcome, and uniform periodically poled Nd:Mg:LiNbO3 (PPNdMgLN) with a period of 7.41μm was produced. An 80 mW green laser emitting at 542 nm was obtained by SFD in a 16 mm long PPNdMgLN crystal. This kind of self-doubling Nd:Mg:PPLN laser will have a great many applications in fields such as laser-based sensors, communications, military electronics, and laser displays.
2. Experimental section
A high quality single crystal of Nd:Mg:LiNbO3 with doping proportions of 0.5 at.% Nd3+ and 5 at.% Mg2+ was grown using the Czochralski method. The as-grown crystal was annealed at 1100 °C in an air atmosphere for 10 hours to release the stress formed in the crystal during growth, and then a voltage of 50~80 V × cm−1 was applied and maintained for 2 h. When the crystal was cooled down to 800 °C, the voltage was removed and the crystal was slowly cooled down to room temperature. The single domain poled crystal was sliced into several samples in order to perform the corresponding measurements. The as grown crystal and wafers are shown in Fig. 1 .
A cuboid with dimensions of 4 mm × 5 mm × 6 mm (a × b × c) was cut from the poled crystal. All six surfaces were optically polished. Measurements of the polarized fluorescence spectrum and fluorescence decay spectrum were carried out using a high-sensitivity fluorescence spectrometer (TRIAX550).
The experimental laser arrangement is shown in Fig. 1. The pump source employed in the experiment was an 810 nm CW fiber-coupled module with a core diameter of 400 μm and a numerical aperture of 0.22. The pump light was focused onto the crystal using two identical plano-convex lenses with a focal length of 10 mm. Both sides of the lenses had an anti-reflection coating at 810 nm. The input mirror M1 was a double-sided mirror that had an anti-reflection coating at 810 nm and a high-reflection coating at 1050-1100 nm. Two different double-sided plane mirriors that had respective transmission at 1050-1100 nm of 5.8% or 10% were used as the output coupler M2. The sample was an a-cut or c-cut 6-mm-long Nd:Mg:LiNbO3 crystal without any coating on either side. Its side surfaces were immersed into water at a temperature of 20 °C to remove the heat generated. The average output power was monitored by a laser power meter. The spectrum was measured using a micro-spectrum meter.
A 1 mm thick Nd:Mg:LiNbO3 wafer was cut from the crystal. Pure Al was evaporated on both surfaces to serve as electrodes. The electrode on the + Z face was patterned using conventional photolithography. The samples were immersed in silicon oil at 200 °C, and the domains were inverted by applying an electric field with voltage amplification.
The self-frequency-doubling arrangement is shown in Fig. 1. The pump source employed in the experiment was an 810 nm CW fiber-coupled module with a core diameter of 200 μm and a numerical aperture of 0.22. The pump light was focused onto the crystal using two identical plano-convex lenses with a focal length of 10 mm. Both sides of the lenses had an anti-reflection coating at 810 nm. The input mirror M1 was a double-sided mirror that had an anti-reflection coating at 810 nm and high-reflection coating at 1050-1100 nm and 525-550 nm. The output coupler M2 was a plano-concave mirror with a radius of curvature of 100 mm. The concave surface of M2 had a high-reflection coating at 1050-1100 nm and an anti-reflection coating at 525-550 nm. The sample was a 16 mm-long PPNdMgLN crystal with both sides anti-reflection coated at 1050-1100 nm and 525-550 nm. The crystal was wrapped with indium foil and mounted in a copper holder.
3. Results and discussions
A high-quality Nd:Mg:LiNbO3 single crystal 3 inches in diameter with doping proportions of 0.5 at.% Nd3 + and 5 at.% Mg2 + was grown by the Czochralski method, as shown in Fig. 2 .
In order to find a suitable pump source, the room temperature polarized absorption spectrum of the crystal was measured, as shown in Fig. 3 . The absorption spectrum does not exhibit any noticeable polarization characteristics, but σ-polarized light has a higher absorption than π-polarized light. The crystal has three main absorption regions at around 600 nm, 750 nm and 810 nm. The 810 nm region is the most important, because it corresponds to the emission band of the AlGaAs laser diode, a device which is generally employed as a pump source. The absorption centers for σ-polarization and π-polarization are 809 nm and 813 nm, respectively. Both polarizations have a full-width at half-maximum (FWHM) of more than 20 nm. Because the emission wavelength of the laser diode drifts with temperature, the wide FWHM enables the crystal to have broad temperature adaptability.
Fig. 3 (a)(b) σ-polarized absorption spectrum and (c)(d) π-polarized absorption spectrum of Nd:Mg:LiNbO3.
Figure 4 shows the polarized fluorescence spectrum of Nd:Mg:LiNbO3 under excitation by 810nm light. The fluorescence spectrum consists of three main peaks, at 920 nm, 1080 nm and 1400 nm. The strongest peak is in the 1080 nm region, which corresponds to the 4F3/2→4I11/2 transition. To further clarify the spectrum structure, a Lorentz multipeak decomposition procedure was applied to the peaks, the results of which are shown in Fig. 4. The π-polarization fluorescence component is much stronger than the σ-polarization component, similar to the case of Nd:Mg:LiTaO3 [23,24 ]. As can be seen from Fig. 4, the σ-polarization component emission contains 5 separate peaks, while the π-polarization component emission contains only 3 peaks, which are listed in Table 1 . In the σ-polarization component, the 1078.21 nm and 1092.08 nm peaks are the strongest and sharpest, while in the π-polarization component, the 1083.64 nm peak is much stronger than the others. The polarized fluorescence spectrum indicates that different wavelengths of laser radiation can occur when the crystal is pumped along different polarization.
Fig. 4 (a)(b) σ-polarized fluorescence spectrum and (c)(d) π-polarized fluorescence spectrum of Nd:Mg:LiNbO3.
Table 1. The results of Lorentz multi-peak decomposition of emission spectrum
The fluorescence decay curve of the 4F3/2 level is illustrated in Fig. 5 . The measured lifetime in Nd:Mg:LiNbO3 is 107.64 μs, which is similar to previously reported results [25]. This value is almost the same as that of Nd:YVO4, and indicates that Nd:Mg:LiNbO3 is suitable for laser generation.
A fundamental laser performance experiment was designed to determine if this crystal is suitable for SFD. The crystal was cut into a cuboid with dimensions of 3 × 3 × 6 mm3 to be used as the laser host medium. A plano-planar cavity was used. The output power of the c-cut Nd:Mg:LiNbO3 crystal as a function of absorbed pump power, and the output laser spectrum are shown in Fig. 6(a) and Fig. 6(b), respectively. Two different mirrors with respective transmissions of T = 5.8% and T = 10% were chosen for an output coupler M2 in this experiment. In the case where T = 5.8%, the maximum output CW power was 1.97 W with an incident absorbed pump power of 10.16 W. The pump threshold value is 2.56 W, corresponding to a slope efficiency of 26.8% and an optical-to-optical efficiency of 19.4%. In the case where T = 10%, the maximum obtained CW output power was 1.38 W with an incident absorbed pump power of 10.16 W. The pump threshold value is 3.32 W, corresponding to a slope efficiency of 20.2% and an optical-to-optical efficiency of 13.6%. Two sharp peaks at 1093.6 nm and 1077.5 nm, were obtained using either output mirror.
Fig. 6 (a) output power versus absorption pump power of c-cut Nd:Mg:LiNbO3; (b) typical laser spectrum with c-cut Nd:Mg:LiNbO3; (c) output power versus absorption pump power of a-cut Nd:Mg:LiNbO3; (d) typical laser spectrum with a-cut Nd:Mg:LiNbO3.
The output power of a-cut Nd:Mg:LiNbO3 as a function of the absorbed pump power, and the output laser spectrum are shown in Fig. 6(c) and Fig. 6(d), respectively. The same two output mirrors used with the c-cut crystals were chosen. In the case where T = 5.8%, the maximum output CW power was 1.24 W at 1084.7 nm, with an incident absorbed pump power of 7.17 W. The pump threshold value was 0.99 W, corresponding to a slope efficiency of 20.1% and an optical-to-optical efficiency of 17.3%. In the case where T = 10%, the maximum output CW power was 1.01 W at 1084.7nm, with an incident absorbed pump power of 7.17 W. The pump threshold value was 1.91 W, corresponding to a slope efficiency of 19.2% and an optical-to-optical efficiency of 14.1%.
The CW laser performance is summarized in Table 2 . For both the a-cut and c-cut crystals, the highest power is generated from the mirror at M2 with a transmission value of 5.8%. Because the crystals are uncoated, there is a large scattering loss in the cavity. So, the output coupling mirror with lower transmission can generate higher output power. A 1.97 watt dual-wavelength laser was obtained from the c-cut sample. This kind of laser has many applications, such as in communications systems [26], homeland security, and terahertz (THz) wave spectroscopy [18, 27, 28 ]. The a-cut crystal has a lower threshold, and the pump direction is also the same as the potential self-frequency-conversion direction, which means that the Nd:Mg:LiNbO3 crystal is a promising candidate for the use of PPNdMgLN to realize self-frequency-conversion.
Table 2. CW Laser Performance
An attempt was made to fabricate periodically poled Nd:Mg:LiNbO3 using the conventional electric field poling method. A schematic of the equipment for applying multiple pulses is shown in Fig. 7 . Voltage pulses were applied to the crystal. The coercive field of Nd:Mg:LiNbO3 is 2.2 KV/mm, a little greater than in MgO doped LiNbO3, which has a value of 1.8 KV/mm [29,30 ]. The structure obtained exhibited a necklace-like pattern with domain-inverted dots or merged islands, rather than a domain-inverted grating structure, as shown in Fig. 8(a) . A large leakage current was observed during the poling process. It is reported in [31] that when domain inversion was complete, the resistivity of the inverted region decreased. The large leakage current is caused by the low resistivity in the domain-inverted dot regions [31].
Fig. 8 Microphotographs ( + Z face, after etching) of the structures formed with (a): only positive voltage; (b) with periodically changing voltage.
Another main reason for leakage in this experiment is the presence of crystal defects.
The radius of Nd3+ is much larger than the radius of Li+, Nb5+ or Mg2+. It is difficult to invert such a large ion in this lattice. So sites without Nd3+ were inverted after the first few pulses, and this restricted uniform poling of the whole crystal. An effictive way to obtain uniform PPLN is to apply a negative voltage before the poling pulse occurs [32]. On one hand, the negative voltage activates the non-inverted region. On the other hand, the negative electric field also stabilizes the poled region. So a variable duty cycle voltage was applied the crystal. The positive and negative voltages are both 2.2 KV/mm, and the pulse width of the positive and negtive voltage is 500 μs and 300 μs, respectively. The sample used to periodically poling is 32mm in length and 10mm in width. Uniformly periodically poled domain structure was produced at the whole sample, as can be seen in Fig. 8(b).
The reflective indices of Nd:Mg:LiNbO3 at different light wavelengths was measured by the minimum-deviation method, as shown in Fig. 9 . The measurement system is HR SpectroMaster UV-VIS-IR(Trioptics, Germany) with a high accuracy of 1*10−5.
The fitting curve for no is:
The fitting curve for ne is:The refractive indices of extraordinary light at 1084 nm and 542 nm are 2.1458 and 2.2187, respectively. The data fit well to previously reported results [12, 33 ]. For example, the refractive indices of extraordinary light in Ref. 12 at 1084 nm and 542 nm are 2.1404 and 2.2187, respectively, while the refractive indices of extraordinary light in Ref. 33 at 1084 nm and 542 nm are 2.1517 and 2.2243, respectively. According to the Sellmeier equation for Nd:Mg:LiNbO3, a period of 7.41 μm is required to accomplish frequency doubling of 1084 nm laser light. Using an 810 nm laser diode as the pump source, a SFD laser was obtained from a 16 mm long PPNdMgLN crystal. Figure 10(a) shows the overall input-output laser curves. Figure 10(b) is a plot of a typical spectrum of the SHG green laser. A single peak at 542 nm is obtained, indicating that frequency doubling of only the 1084 nm line is generated in the crystal. This result coincides well with the fundamental laser performance in Fig. 4(c). The system has a threshold at 1.5 W from the 810 nm pump laser. The maximum green laser output power is 80 mW under 12.4 W of pump power. This is, to the best of our knowledge, the best performance of a SFD green laser generated from QPM materials. The output power in this experiment is greater by at least one order of magnitude than the values that were previously reported [14, 15, 21 ]. The power of this laser has attained a level where practical applications are feasible in laser-based sensors and in photo-communications. Although the obtained laser output in this work is still lower than other methods, we believe high power laser can be obtained by increasing the Nd3+ doping concentration in crystal, improving preparation quality of the PPLN, selecting better pump light with higher beam quality, and further optimization of the laser device. Work devoted to improving the quality of PPNdMgLN, and the performance the laser devices constructed with it is continuing.
Fig. 10 (a) Output power versus pump power of PPNdMgLN crystal; (b) typical laser spectrum of SFD green laser.
4, Summary
A 3 inch Nd:Mg:LiNbO3 crystal was grown by the Czochralski method. The crystal exhibited multi-emission peaks near 1080nm. The high quality of this crystal produces a high continuous wave output power of 1.97 W at both 1093.6 nm and 1077.5 nm with a c-cut crystal, and a continuous wave output power of 1.24 W at 1084.7 nm with an a-cut crystal. PPNdMgLN with a 7.41 μm poling period was fabricated using a specially designed electric-field poling method. A uniform domain structure was obtained using variable duty cycle voltage pulses. An 80 mW green laser emission at 542 nm was generated from a 16 mm length of PPNdMgLN crystal. This is the highest power generated in SFD from QPM materials. This work demonstrates that SFD generated in QPM materials is an attractive candidate for applications like laser displays, visible light communications, etc.
Acknowledgments
This work was supported by the National High Technology Research and Development Program of China (863 Program, Grant No. 2013AA030501). The authors also appreciate the kind assistance from Prof. Jin Yu and Dr. Zeqiang Mo for helping installing the SFD in the laboratory. The authors would like to thank Prof. Robert I. Boughton from Bowling Green State University for his professional suggestion.
References and links
1. T. H. Maiman, “Stimulated optical radiation in ruby,” Nature 187(4736), 493–494 (1960). [CrossRef]
2. J. W. Li, N. Xu, J. Li, Z. X. Zhang, and H. J. Yao, “Application of Laser Radar in Atmosphere Environment Monitoring,” Adv. Mater. Res. 760, 88–91 (2013).
3. T. Matsuoka, “InGaAlN and II-VI Systems for Blue-Green Light-Emitting Devices,” Adv. Mater. 8(6), 469–479 (1996). [CrossRef]
4. R. Xia, C. Cheung, A. Ruseckas, D. Amarasinghe, I. D. Samuel, and D. D. Bradley, “Wavelength conversion from silica to polymer optical fibre communication wavelengths via ultrafast optical gain switching in a distributed feedback polymer laser,” Adv. Mater. 19(22), 4054–4057 (2007). [CrossRef]
5. S. Nagahama, S. Masui, T. Miyoshi, D. Kasahara, E. Okahisa, T. Hirao, and T. Yanamoto, “Watt-class AlInGaN blue and green laser diodes,” in The 2nd Laser Display Conference. 2013, The Japan Society of Applied Physics: Yokohama, Japan.
6. M. Zhou, B. Yan, G. Bao, Y. Zhang, C. Gawith, D. Wang, Y. Qi, and Y. Bi, “52% optical-to-optical conversion efficiency in a compact 1.5 W 532 nm second harmonic generation laser with intracavity periodically-poled MgO:LiNbO3,” Laser Phys. 20(7), 1568–1571 (2010). [CrossRef]
7. C. Jung, B. A. Yu, I. S. Kim, Y. L. Lee, N. E. Yu, and D. K. Ko, “A linearly-polarized Nd:YVO4/KTP microchip green laser,” Opt. Express 17(22), 19611–19616 (2009). [CrossRef] [PubMed]
8. J. Wang, H. Zhang, Z. Wang, H. Yu, N. Zong, C. Ma, Z. Xu, and M. Jiang, “Watt-level self-frequency-doubling Nd:GdCOB lasers,” Opt. Express 18(11), 11058–11062 (2010). [CrossRef] [PubMed]
9. H. Yu, N. Zong, Z. Pan, H. Zhang, J. Wang, Z. Wang, and Z. Xu, “Efficient high-power self-frequency-doubling Nd:GdCOB laser at 545 and 530 nm,” Opt. Lett. 36(19), 3852–3854 (2011). [CrossRef] [PubMed]
10. P. Dekker, J. M. Dawes, J. A. Piper, Y. Liu, and J. Wang, “1.1 W CW self-frequency-doubled diode-pumped Yb: YAl3(BO3)4 laser,” Opt. Commun. 195(5-6), 431–436 (2001). [CrossRef]
11. T. Y. Fan, A. Cordova-Plaza, M. J. F. Digonnet, R. L. Byer, and H. J. Shaw, “Nd:MgO:LiNbO3 spectroscopy and laser devices,” J. Opt. Soc. Am. B 3(1), 140–149 (1986). [CrossRef]
12. M. Gong, G. Xu, K. Han, and G. Zhai, “Nd:Mg:LiNbO3 self-frequency-doubled laser pumped by a flashlamp at room temperature,” Electron. Lett. 26(25), 2062–2063 (1990). [CrossRef]
13. R. Li, C. Xie, J. Wang, X. Liang, K. Peng, and G. Xu, “CW Nd: MgO: LiNbO3 self-frequency-doubling laser at room temperature,” IEEE J. Quantum Electron. 29(9), 2419–2420 (1993). [CrossRef]
14. X. Chen, Z. Luo, and Y. Huang, “Modeling of self-frequency-conversion lasers in rare-earth doped optical superlattice crystal,” J. Opt. Soc. Am. B 19(5), 1067–1077 (2002). [CrossRef]
15. J. Capmany, V. Bermudez, D. Callejo, J. Garcia Sole, and E. Diéguez, “Continuous wave simultaneous multi-self-frequency conversion in Nd3+-doped aperiodically poled bulk lithium niobate,” Appl. Phys. Lett. 76, 1225–1227 (2000).
16. J. Capmany, D. Callejo, V. Bermudez, E. Dieguez, D. Artigas, and L. Torner, “Continuous-wave self-pumped optical parametric oscillator based on Yb3+-doped bulk periodically poled LiNbO3:MgO,” Appl. Phys. Lett. 79(3), 293–295 (2001). [CrossRef]
17. G. D. Laptev, A. A. Novikov, and V. V. Firsov, “Q-switched quasi-phase-matched self-frequency doubling and summing in a laser based on a periodically poled Nd:Mg:LiNbO3 crystal,” Quantum Electron. 34(3), 233–235 (2004). [CrossRef]
18. D. Jaque, J. Sanz Garcia, and J. García Solé, “Continuous-wave laser oscillation at 929 nm from a Nd3+-doped LiNbO3: ZnO nonlinear laser crystal: A powerful tool for blue laser light generation,” Appl. Phys. Lett. 85(1), 19–21 (2004). [CrossRef]
19. G. D. Laptev, A. A. Novikov, V. V. Firsov, in: K. L. Schepler, D. D. Lowenthal, and J. W. Pierce, “Quasi-phase-matched self-frequency summing in a periodically poled Nd:Mg:LiNbO3,” High-Power Lasers and Applications. International Society for Optics and Photonics, 2003: 42–49.
20. H. Maestre, A. J. Torregrosa, C. R. Fernández-Pousa, M. L. Rico, and J. Capmany, “Dual-wavelength green laser with a 4.5 THz frequency difference based on self-frequency- doubling in Nd3+ -doped aperiodically poled lithium niobate,” Opt. Lett. 33(9), 1008–1010 (2008). [CrossRef] [PubMed]
21. M. L. Rico, E. Moya, J. L. Valdés, J. Martinez-Pastor, V. Bermúdez, and J. Capmany, “Continuous-Wave Yellow Laser Based on Nd-Doped Periodically Poled Lithium Niobate,” IEEE J. Sel. Top. Quantum Electron. 13(3), 750–755 (2007). [CrossRef]
22. L. Barraco, A. Grisard, E. Lallier, P. Bourdon, and J. P. Pocholle, “Self-optical parametric oscillation in periodically poled neodymium-doped lithium niobate,” Opt. Lett. 27(17), 1540–1542 (2002). [CrossRef] [PubMed]
23. D. Sun, S. Liu, D. Wang, Y. Sang, X. Kang, H. Liu, Y. Bi, B. Yan, J. He, and J. Wang, “High output power of differently cut Nd: MgO: LiTaO3 CW lasers,” Laser Phys. 23(4), 045802 (2013). [CrossRef]
24. P. Zhang, J. Gong, Y. Hang, H. Li, R. Zhang, and J. Xu, “A Nd: MgO: LiTaO3 crystal as a gain medium for 1083 nm lasers,” Laser Phys. 23(4), 045803 (2013). [CrossRef]
25. R. Burlot, R. Moncorgé, H. Manaa, G. Boulon, Y. Guyot, J. Garcia Solé, and D. Cochet-Muchy, “Spectroscopic investigation of Nd3+ ion in LiNbO3, MgO: LiNbO3 and LiTaO3 single crystals relevant for laser applications,” Opt. Mater. 6(4), 313–330 (1996). [CrossRef]
26. Y. Li, A. J. Vieira, S. M. Goldwasser, and P. R. Herczfeld, “Rapidly tunable millimeter-wave optical transmitter for lidar-radar,” IEEE T. Microw. Theory 49(10), 2048–2054 (2001). [CrossRef]
27. R. Miles, P. Harrison, and D. Lippens, “Terahertz Sources and Systems,” (Springer Science & Business Media, 2001)
28. K. Sakai and J. T. K. Kikō, Terahertz Optoelectronics (Springer, 2005).
29. A. Kuroda, S. Kurimura, and Y. Uesu, “Domain inversion in ferroelectric MgO: LiNbO3 by applying electric fields,” Appl. Phys. Lett. 69(11), 1565–1567 (1996). [CrossRef]
30. H. Ishizuki, T. Taira, S. Kurimura, J. H. Ro, and M. Cha, “Periodic poling in 3-mm-thick MgO: LiNbO3 crystals,” Jpn. J. Appl. Phys. 42(2), L108–L110 (2003). [CrossRef]
31. N. Horikawa, T. Tsubouchi, M. Fujimura, and T. Suhara, “Formation of domain-inverted grating in MgO: LiNbO3 by voltage application with insulation layer cladding,” Jpn. J. Appl. Phys. 46(8A), 5178–5180 (2007). [CrossRef]
32. K. Mizuuchi, A. Morikawa, T. Sugita, and K. Yamamoto, “Polarisation-switching-induced resistance change in ferroelectric Mg-doped LiNbO3 single crystal,” Electron. Lett. 40(13), 819–890 (2004). [CrossRef]
33. S. Ishibashi, H. Itoh, T. Kaino, I. Yokohama, and K. Kubodera, “New cavity configurations of Nd:MgO:LiNbO3 self-frequency-doubled Lasers,” Opt. Commun. 125(1-3), 177–185 (1996). [CrossRef]
