Room temperature 90° phase-matching in zirconium and magnesium co-doped lithium niobate crystals

Laser has been widely used in many aspects, by now it is difficult to get each frequency that we want, and frequency conversion is an effective way to obtain different frequency laser through a nonlinear optical crystal. MgO-doped LiNbO3 (Mg:LN) crystal has usually been used for second harmonic generation (SHG) through temperature-matching configuration with a stove, till now a room temperature 90° phase-matching is still lacking. Here we find that the SHG of Nd:YAG laser is achieved at 26.1 °C while the optical damage resistance is higher than 6.5 MW/cm2 in the ZrO2 and MgO co-doped LiNbO3 (Zr,Mg:LN) crystal. Moreover, the monotonic decrease of phase-matching temperature is firstly found with the increase of doping concentration. These unusual properties may be attributed to the formation of MgLi+\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{\bf{Mg}}}_{{\bf{Li}}}^{{\boldsymbol{+}}}$$\end{document}  + ZrNb−\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{\bf{Zr}}}_{{\bf{Nb}}}^{{\boldsymbol{-}}}$$\end{document} defect pairs. Our work suggests that Zr,Mg:LN crystal may be an attractive candidate for nonlinear optical applications.

1 School of Physics, Nankai University, Tianjin, 300071, China. 2 MOE Key Laboratory of Weak-Light Nonlinear Photonics and TEDA Institute of Applied Physics, Nankai University, Tianjin, 300457, China. 3 R&D Center, Taishan Sports Industry Group, Leling, 253600, China. 4 Collaborative Innovation Center of Chemical Science and Engineering (Tianjin), Tianjin, 300072, China. 5  close to one [24][25][26] . Up to now, investigations on Zr:LN crystals mainly focus on optical waveguide, defect structure and co-doping with photorefractive impurities [27][28][29] , but their refractive indices and nonlinear optical properties (e.g., phase-matching temperature) are rarely reported. In fact, no sign of phase matching in Zr:LN crystals was found when they were heated from room temperature to above 200 °C in our pre-experiments, which implies that the phase-matching temperature of Zr:LN crystals may be lower than room temperature. If that is true, we may obtain room temperature 90° phase-matching by doubly doping with ZrO 2 and MgO in LiNbO 3 . And co-doping with two optical damage resistant ions is conducive to finely tune the optical properties of LiNbO 3 19,30 .
In this paper, we grew a series of ZrO 2 and MgO co-doped LiNbO 3 (Zr,Mg:LN) crystals with various ZrO 2 dopants, and the doping concentration of MgO was chosen as 5.0 mol.% because Mg:LN has high phase-matching temperature in this doping level. Our experimental results demonstrate that the phase-matching temperature of Zr,Mg:LN monotonically decreases with increased ZrO 2 concentration. And the efficient 90° phase-matching is achieved in Zr,Mg:LN crystal at room temperature without a stove, meanwhile it also has a very high optical damage resistance.

Results
Temperature tuned 90° phase-matching. Temperature tuned 90° phase-matching was achieved by using a Q-switched Nd:YAG laser with a wavelength of 1064 nm. The dependence of the phase-matching temperature T PM on the ZrO 2 concentration in the melt for Zr,Mg:LN crystals is depicted in Fig. 1(a). From this figure, we can see that the T PM decreases as the ZrO 2 concentration increases, and a straight line can be fitted well to the experimental data. For comparison, the T PM versus the doping concentration of MgO for Mg:LN crystals is presented in Fig. 1(b), and the data are referenced from the previous literature 19 . The relationship between the T PM and the impurity concentration is similar to a parabola going downwards, and the maximum temperature stays within the concentration range of 4~6 mol.%. Generally, the T PM versus the doping concentration exposes a more or less expressed threshold behavior corresponding to the sharp change of optical properties, and this similar behavior can be found in other mono or dual doped LN crystals, such as Zn:LN 21 , Sc:LN 22 and Zn,In:LN 30 . In contrast, Zr,Mg:LN crystals exhibit a significant monotonic, and a simple linear extrapolation from existing data holds over a wider concentration range.
Please note that the phase-matching temperature in Zr 1.7 ,Mg 5.0 :LN crystal is 26.1 °C, which is close to room temperature (25 °C). Figure 2 clearly shows its normalized temperature-tuning curve for doubling 1064 nm using 90° phase-matching. The dots are the measured second harmonic output power, and the solid curve is a fit to the x sinc( ) function, which almost perfectly overlaps the experimental data. The full width at half maximum (FWHM) of the temperature-tuning curve is 1.2 °C. Moreover, the conversion efficiency is plotted as a function of the incident fundamental power density in Fig. 3. An average second harmonic power of 91.5 mW is obtained with a conversion efficiency of 28.6% at the peak-power density of 50 MW/cm 2 , and maintaining this conversion efficiency for two hours, there is no significant degradation. Overall, we should point out emphatically that if considering the temperature increase (about 2~3 °C) in the continuous harmonic output 31,32 , Zr 1.7 ,Mg 5.0 :LN crystal is particularly well-suited for practical application of laser frequency doubling at room temperature.
Optical damage resistance. In order to measure the optical damage resistance, the distortion of the transmitted light beam through the wafer was observed with a 532 nm laser. Figure 4 shows the transmitted laser beam spots after 5 min of irradiation. As the concentration of ZrO 2 increases from 0.5 to 1.7 mol.%, none of Zr,Mg:LN crystals appears noticeable beam smeared, even under the highest focused intensity of 6.5 × 10 6 W/cm 2 in our laboratory, and the optical damage resistance is the same magnitude as that of Zr 2.0 :LN. However, Mg 5.0 :LN crystal can only withstand a maximum intensity of 4.1 × 10 5 W/cm 2 under the same conditions. The above results indicate that the optical damage resistance of these Zr,Mg:LN crystals is improved by at least an order of magnitude than that of Mg 5.0 :LN. To quantitatively characterize the optical damage resistance, the light-induced changes of the refractive index Δn of these crystals were measured by two-beam holography 33 . Two e-polarized coherent beams at 532 nm were intersected in the wafers with equal intensity of 400 mW/cm 2 . The change of refractive index Δn was calculated by the equation 34 Here, max η is the maximum diffraction efficiency; λ is the wavelength, 532 nm; d is the crystal thickness, 3.0 mm; and θ cry is the intersection half-angle of the two coherent beams outside the crystal, 2 cry θ = 30°. The photoconductivity ph σ was also estimated through the relationship, , where ε 0 is the vacuum dielectric constant, ε = 28 is the relative dielectric constant of the crystal 35 , and the erasure time constant τ e is defined as the time when the diffraction efficiency decays to 1/e of its initial value.
The change of refractive index Δn and the photoconductivity ph σ versus the ZrO 2 concentration for all samples are presented in Fig. 5. We can see that the change of refractive index reduces rapidly with doping 0.5 mol.% ZrO 2 into Mg 5.0 :LN, then changes slightly as the ZrO 2 concentration increases. Moreover, the Δn of Zr,Mg:LN is considerably less than that of Mg 5.0 :LN and even lower than that of Zr 2.0 :LN. In addition, the σ ph of Zr,Mg:LN is larger than that of Mg 5.0 :LN but close to that of Zr 2.0 :LN. It is well known that the increase of the photoconductivity is primarily responsible for the increase of the optical damage resistance 36 . Therefore, the results demonstrate again that adding some ZrO 2 into Mg 5.0 :LN can further enhance the optical damage resistance, which is consistent with the results of the transmitted light beam distortion.
Infrared absorption spectra. The infrared absorption spectra, referring mainly to OH − absorption spectra, sensitively reflect the change of defect structure in LiNbO 3 , which have become an important tool in studying the properties of dopant-related defects. Figure 6 shows the OH − absorption spectra of CLN and Zr,Mg:LN crystals. As no obvious OH − band shift is observed, a three-peak model 37 is employed by Lorentz fitting, and the results are listed in Table 1. For comparison, that of Mg 5.0 :LN and Zr 2.0 :LN crystals are also listed. It can be seen from the      Our results on SHG of Nd:YAG lasers have shown that the additional ZrO 2 doping can significantly influence the phase-matching temperature of Mg 5.0 :LN. In particular, the monotonic decrease of phase-matching temperature on ZrO 2 concentration is found for the first time, which is different from the previous reports on doped LiNbO 3 . As we know, so-called optical damage resistant impurities in visible region such as Mg, Zn, In and Hf can enhance the UV photorefractive effect 41,42 , but Zr exhibits excellent optical damage resistance in both visible and UV region. It is thought that the enhanced UV photorefractive effect has direct relationship with doped ions occupying Nb sites 43 , which means that Zr 4+ ions in Nb sites can greatly alter defect structures and properties of LiNbO 3 . In Zr,Mg:LN crystals, Mg Li + + − Zr Nb neutral pairs may play an important role in this monotonic decrease relationship. However, further investigation is greatly needed to clarify the micro-mechanism.

Conclusion
We grew a series of LiNbO 3 co-doping with 5.0 mol.% MgO and various ZrO 2 concentrations. The experimental results indicate that the phase-matching temperature of Zr,Mg:LN decreases with increased ZrO 2 concentration for the first time. And 90° phase-matching of 1064 nm radiation is achieved at room temperature in Zr 1.7 ,Mg 5.0 :LN crystal, while it holds a high resistance of optical damage at 532 nm, and does not suffer any dark trace damage when exposed to high power laser irradiation for two hours, which will be greatly valuable for engineering applications in compact and efficient high-power green lasers. These excellent properties of Zr,Mg:LN may be attributed to the formation of Mg Li + + Zr Nb − neutral defect pairs. Spectra characterization. The infrared absorption spectra and UV-Visible absorption spectra of 1.0 mm thick y-plates were measured at room temperature with a Magna-560 Fourier transform infrared spectrophotometer and a U-4100 spectrophotometer, respectively. The resolution value of this infrared spectrometer was 4.0 cm −1 , and the step-length of the UV-Vis. spectrometer was 1.0 nm.

Second harmonic generation.
The experiments for second harmonic generation were performed with a Q-switched Nd:YAG laser 1064 nm at a 1 Hz repetition rate, 8 ns pulse width. The laser facula diameter was 5.0 mm, and the maximum average output energy was 320 mJ. The fundamental light was directed to the crystal in a geometry with the c-axis of the crystal perpendicular to the polarization direction of the light, so-called 90° phase-matching. The second harmonic energy was detected with a band-pass filter and a pulse laser energy meter. The bulk crystal was mounted in an oven thermally controlled to a temperature better than ±0.2 °C. The temperature was measured with a Pt-100 thermistor placed in direct contact with the crystal, and a slow heating rate was used about 0.5 °C/min to minimize temperature gradients with the sample. In addition, the distance between the entrance and the exit windows of the crystal surfaces was at least 5.0 cm to minimize the possible temperature gradients along the crystal y-axis. The formula for the conversion efficiency was η = E 2 /E 1 , where E 1 was the energy of the fundamental wave and E 2 was that of the second harmonic wave.
Data availability. The datasets generated during the current study are available from the corresponding author on reasonable request.