We investigated ultraviolet (UV) laser-induced degradation of nonlinear optical crystal CsLiB6O10 (CLBO) and β-BaB2O4 (BBO) using a high-repetition-rate pulsed laser. In this research, we found that the degradation of CLBO was caused by a UV-induced refractive index change at a peak power density higher than a few tens of MW/cm2. On the other hand, BBO exhibited lower UV transmittance at a higher intensity of 10 MW/cm2 and the UV absorption increased with time due to the formation of an absorption center. Moreover, we confirmed that the degradation resistance of CLBO with fewer light scattering defects was improved.
© 2014 Optical Society of America
There is an increasing demand for high-power deep ultraviolet (UV) sources in industrial fields such as high-resolution inspection and advanced material processing [1,2]. CsLiB6O10 (CLBO) and β-BaB2O4 (BBO) are known nonlinear optical crystals suitable for generating high-power deep UV output with wavelengths below 300 nm [3–7]. Laser-induced damage is a critical issue associated with UV power scaling. The single-shot bulk damage threshold has been reported to be 23.1–23.6 GW/cm2 at 266 nm in CLBO and 25 GW/cm2 at 355 nm in BBO [8,9]. It is important to research UV output degradation in nonlinear optical crystals occurring at a lower peak power density than the bulk laser-induced damage threshold because the degradation seriously limits power scaling and the lifetime of UV lasers. However, only a few studies have investigated the degradation of crystals [10–12]. In addition, there have been no reports on the degradation in CLBO and BBO using the same experimental conditions. The research into degradation in nonlinear optical crystals is useful for gaining a better understanding of the behavior and the mechanism of laser-induced damage, which are valuable in the design and application of UV laser systems. Deki et al. in  conclude that a UV-induced refractive index change near the focal point in CLBO results in the 266 nm output degradation. We found that the UV-induced degradation in CLBO and BBO occurred after long-term operation under conditions usual in practical applications. Therefore, instead of UV light generation in each crystal, the UV beam was highly focused in the samples to evaluate the degradation behavior over a short time. We have recently investigated UV-induced degradation of CLBO using high-repetition-rate pulsed lasers with a wavelength of 266 nm . The degradation behavior of CLBO is similar to the light-induced scattering of LiNbO3 (LN). Moreover, nonlinear-induced absorption (NLA) also limits UV power scaling because NLA can cause self-heating, which results in phase-mismatching. In this experiment, by using the same experimental conditions, we compared for the first time UV-induced degradation and the NLA of CLBO and BBO. Because we have recently grown high-quality CLBO crystals with fewer light scattering defects, we also investigated the degradation resistance of the crystals.
2. Experimental methods
A Nd:YVO4 laser (Iridex Corporation; TEM00 mode; wavelength: 1064 nm; pulse width: 10 ns; pulse repetition frequency: 30 kHz) was used as the fundamental light source. The fourth harmonic beam (wavelength: 266 nm; pulse width: 8.2 ns) was generated in nonlinear crystals as the input source for the degradation measurement. The shape of the 266 nm beam was adjusted to be circular by a pair of cylindrical lenses as shown in Fig. 1.Samples of commercial CLBO and BBO with dimensions of 5 × 5 × 10 mm3 were cut along the phase-matching direction for type-I second harmonic generation of 532 nm laser light (CLBO: (θ, φ) = (61.9°, 45°), BBO: (θ, φ) = (64.8°, 90°)). We evaluated the BBO crystal grown using the flux method. Water in a CLBO crystal significantly reduces its degradation resistance . Therefore, before the experiment, water impurities inside the CLBO sample were eliminated though dehydration treatment. From the experimental results of the focal position dependence of the degradation resistance , CLBO exhibited the lowest resistance at the center of the sample. Therefore, the beam with extraordinary polarization was focused at the center of the sample with a focusing lens of f = 70 mm to induce degradation. The degradation measurements with extraordinary ray for both CLBO and BBO are valuable for practical frequency conversions. The beam waist (1/e2 intensity diameter) was measured to be about 2ω = 27.5 ± 0.24 μm in air using the scanning knife-edge method. The average input power was 10–100 mW. Therefore, the peak power density at the focal point was calculated to be 14–137 MW/cm2 with an effective radius for a Gaussian beam. The beam patterns through the sample were observed on fluorescent paper A and B in the far-field, which were about 60 and 10 cm from the beam focus, respectively.
First, we investigated the initial transmittance T0 at 266 nm for each sample. The initial transmittance is defined as an average transmittance within a few seconds after the UV illumination began. Figure 1(a) shows the experimental setup for the UV transmittance measurements. The power transmitted through the sample was measured by the power meter. A small part of the beam was separated from the main beam by a beam splitter to monitor the beam transmitted through the sample. The beam patterns were observed on fluorescent paper A. Each measurement was performed after changing the irradiation point on the sample. We measured the UV peak power density dependence for 266 nm transmittance for CLBO and BBO to investigate the NLA with a high pulse repetition frequency. For a beam with a Gaussian profile (TEM00 mode), the nonlinear transmission can be described by the following formula :Eq. (1), we estimated the NLA coefficient of each crystal.
Next, we investigated the UV-induced degradation. Figure 1(b) shows a schematic of the experimental setup for this accelerated testing of the degradation. In the case of LN, the light-induced scattering such as interference patterns and speckle patterns is observed using an experimental setup similar to that shown in Fig. 1(b). To determine not only the change in transmitted power but also the change in refractive index, we set the aperture at 70 mm from the focal point. The aperture size was about 3 mm (1/e2 intensity diameter). The distorted beam patterns were observed on fluorescent paper A and B. The power transmitted through the aperture was measured to determine the degradation that occurred near the focal point in the sample. The 266 nm transmittance T(t) was defined as , where T0 was the initial transmittance of the sample (see Fig. 2), P(t) was the power transmitted through the aperture, and P(0) was the initial power transmitted through the aperture. The resistance to degradation was measured for each experimental condition after changing the irradiation point. We measured the change in transmitted power through the aperture and the temperature dependence of the degradation resistance for both CLBO and BBO. We also investigated the UV peak power density dependence of the degradation resistance for CLBO to determine the degradation threshold.
3. Results and discussion
3.1 Transmittance at 266 nm of CLBO and BBO
Figure 2 shows the initial transmittance of CLBO and BBO samples at 30 and 150 °C as a function of the UV peak power density at the focal point. We corrected for the effect of Fresnel reflection loss on both surfaces. CLBO exhibited high transmittance (> 97%) at 30 and 150 °C regardless of the UV peak power density. No pattern distortion occurred in this experiment. This means that CLBO does not induce a large optical loss in this experimental power density range. On the other hand, the initial transmittance of BBO decreases with increasing UV intensity. As shown in Fig. 2(b), the far-field pattern through BBO was distorted to a greater extent as the transmittance decreased. From Fig. 2 and Eq. (1), we estimated the NLA coefficient. The values of the NLA coefficient for the CLBO and BBO samples were about 0.5 and 15.0 cm/GW, respectively, regardless of the sample temperature. The NLA coefficient for BBO is the same as that given in  (wavelength: 262 nm; pulse width: 44 ns; pulse repetition frequency: 10 kHz) at 151 °C. We also confirmed that the bulk laser-induced damage threshold of the BBO sample was about 120 MW/cm2. We consider the bulk damage threshold in this experiment to be lower than the value of 270–300 MW/cm2 reported in  (wavelength: 266 nm; pulse width: 8 ns; pulse repetition frequency: 10 Hz) because NLA increases as pulse repetition frequency increases . There is a possibility that the BBO in  exhibited a higher bulk damage threshold because a flux-free crystal was used.
3.2 UV-induced degradation of CLBO
Figure 3 shows typical results obtained for a CLBO crystal at 30 °C. The peak power density was 55 MW/cm2. The pattern distortion occurred after 0.5 minutes of illumination, and the size of the scattered pattern increased until reaching a steady state after 5 minutes as shown in Fig. 3(c). As a result, the power transmitted through the aperture gradually decreased. The UV absorption did not increase in this experiment because the power transmitted without an aperture did not change. We also confirmed that the distorted beam pattern recovered after the experiment . This indicates that the degradation is not caused by permanent damage inside the crystal and on the surfaces. As the beam patterns on fluorescent paper A in Figs. 3(a)–3(c) show, the distortion produced an interference pattern and developed from the center of the beam. On the other hand, the beam pattern on fluorescent paper B in Fig. 3(c) shows that a speckled pattern is observed at the periphery of the beam. The intensity of the speckled pattern increases continuously until reaching a steady state as shown in Fig. 3(c) while forming the interference pattern as Fig. 3 (b). The maximum scattering angle of the speckled pattern is larger than that of the interference pattern. Similar far-field diffraction patterns resulted from the light-induced scattering of LN [16,17]. Moreover, we have recently found that, similar to LN, the UV-induced degradation resistance of CLBO was improved by increasing the sample temperature . In the case of LN, the light-induced refractive index change can be explained by the band transport model . The charge separation in the crystal occurs through the diffusion and retrapping of light-induced carriers, resulting in a space charge field. Therefore, the refractive index is changed via the electro-optic effect. Negative light-induced lensing gradually occurs due to the light-induced refractive index change and results in self-defocusing. Although the electro-optic coefficient of CLBO has not been reported and the physical mechanism of this degradation is currently not well understood, the results in this experiment show that the degradation of CLBO is caused by a UV-induced refractive index change.
3.3 Improvement of light scattering and UV-induced degradation of newly developed CLBO
CLBO has already been put to practical use in several UV laser sources developed for semiconductor inspection. However, light scattering is observed in commercial CLBO. A green laser (wavelength: 532 nm; average power: 30 mW) was passed through a CLBO sample along the a-axis, and the resulting light scattering, as viewed in the c-axis, was photographed. The shape of the 532 nm light was adjusted to form the elliptic beam by a cylindrical lens. The beam diameter was measured to be about 2 mm × 84 μm in air using the scanning knife-edge method. The polarization is vertical to the scattering plane. Figure 4 shows the results for various CLBO samples. CLBO has different type of light scattering as seen in Fig. 4(a). One is a bright spot categorized as Mie scattering and/or geometric index scattering and the other is an optical path categorized as Rayleigh scattering. Typical commercial CLBO contains such optical paths as shown in Fig. 4(b). This means that the point defects are uniformly distributed inside the crystal. In spite of such defects, CLBO is useful for use in commercial laser systems with present power levels. We have successfully grown high quality CLBO crystals with fewer light scattering defects. In spite of using a 30 mW bright green laser, the light scattering of the crystal was invisible to the naked eye as is apparent from Fig. 4(c).
Next, we compared the UV-induced degradation resistance of conventional CLBO (Sample A) and the newly developed CLBO with fewer light scattering defects (Sample B). The cutting and polishing processes of CLBO were conducted by Kogakugiken under identical conditions. In this experiment, the lifetime (see Fig. 3) is defined as the time taken for the transmission to drop to 90% of the initial transmission through the aperture. Figure 5 shows the lifetime dependency on the UV peak power density for the conventional and newly developed CLBO samples at 150 °C. The lifetime decreases with increasing UV peak power density. Moreover, the newly developed CLBO with fewer light scattering defects exhibits a longer lifetime than conventional CLBO for a power density range of 69–137 MW/cm2. Using this result, we estimated the degradation threshold (for a lifetime of 10,000 hours). The thresholds of the conventional and newly developed CLBO were about 20–30 and 35–45 MW/cm2, respectively. This shows that the reduction in light scattering defects improves the degradation resistance. Although the lifetime of the newly developed CLBO decreased with decreasing the sample temperature, the crystal showed longer lifetime at 30 °C than the conventional CLBO. So, we consider that the light scattering defects distributed inside the crystal may have a relationship with the formation and/or trapping of photo-induced carriers.
3.4 UV-induced degradation of BBO
Figure 6 shows typical results for the degradation measurements in BBO crystal at 30, 70, 100, and 150 °C. The peak power density was 55 MW/cm2. In the case of 30 °C, although pattern distortion was not confirmed as shown in Fig. 6(b), the transmitted power monotonically decreased after the UV illumination began. UV absorption increased in this experiment because the power transmitted without an aperture changed. After that, pattern distortion occurred as shown in Fig. 6(c), and the transmitted power through the aperture decreased gradually. When the input UV power decreased after the UV absorption increased, as shown in Fig. 6(b), the transmittance did not recover. Therefore, the degradation of BBO is caused by the formation of an absorption center. The beam pattern is distorted because a thermally induced refractive index change may be induced by the formation of an absorption center. We also confirmed that the resistance to degradation increased with increasing sample temperature. The dependence of the absorption of BBO on the sample temperature is similar to that of KTiOPO4 (KTP) [19,20]. In the case of KTP, because the recombination of the unstable laser-induced defect pairs is accelerated at a higher temperature region, the formation of color centers (absorption centers) can be effectively suppressed by increasing the temperature. Therefore, some recombination effects of laser-induced defects in BBO at high temperature may increase the degradation resistance. As shown in Figs. 2 and 6, the initial transmittance of BBO did not change regardless of the sample temperature, although the degradation resistance improved as sample temperature increased. This means that the formation of an absorption center does not directly affect the initial transmittance loss. In general, the Na component of the flux is incorporated as an impurity in commercial BBO crystals and results in a decrease in optical properties. In this experiment, the concentration of Na incorporated into the sample was 72.5 wt. ppm. Therefore, we consider that the degradation of BBO may be suppressed by reducing such flux elements.
4. Discussions for practical applications
According to Stamm et al. , for high-repetition-rate pulsed lasers with a wavelength of 266 nm, the second harmonic generation conversion efficiency from 532 nm to 266 nm of BBO is gradually saturated with increasing input power. As shown in Fig. 2, the NLA of BBO increases as UV intensity increases. Therefore, in the case of high-power UV generation in BBO, self-heating at the focal point may be induced by the NLA, resulting in phase-mismatching. From the present experimental results, BBO is expected to be a high UV conversion efficiency at a power density lower than about 10 MW/cm2. On the other hand, at 150 °C, CLBO shows superior performance in terms of long-term operation at a power density lower than about 50 MW/cm2 because the degradation occurs at high-power levels as shown in Fig. 5. If the degradation is suppressed at high-power levels, CLBO is expected to generate high-power UV light and to operate long term at a UV peak power density of over 100 MW/cm2. In general, the laser-induced degradation such as refractive index change in LN and color centers in KTP can be effectively suppressed by increasing the temperature. From  and Fig. 6, both CLBO and BBO should be used at a high temperature to improve the degradation resistance, as is the case with LN and KTP. Although we investigated the preliminary lifetime test of CLBO at temperatures over 150 °C, the lifetime monotonically decreased in the higher temperature region. It means that some thermal-induced defects and/or carriers may have harmful effects on the degradation at temperatures over 150 °C. Therefore, we empirically determine that the best temperature for CLBO is 150 °C. The mechanism of this degradation acceleration of CLBO at a higher temperature region is currently unclear.
The UV conversion efficiency of BBO grown using a flux-less solution is higher than that using a flux solution . We now consider impurity reduction in BBO crystals essential for generating high-power UV light capable of long-term operation. Therefore, it is important that BBO crystals are grown from melt without any flux elements in the future .
For the first time we investigated UV-induced degradation of CLBO and BBO at 266 nm using the same setup. The degradation of CLBO was similar to the light-induced scattering in LN and was caused by a UV-induced refractive index change. We also found that CLBO with fewer light scattering defects had higher degradation resistance. BBO exhibited decreasing transmittance with increasing UV intensity in the range of 10–120 MW/cm2; the degradation was caused by the formation of an absorption center, and the degradation resistance was improved with increasing sample temperature.
References and links
1. S. Imai, K. Matsuki, N. Kikuiri, K. Takayama, O. Iwase, Y. Urata, T. Shinozaki, Y. Wada, and S. Wada, “Highly reliable 198-nm light source for semiconductor inspection based on dual fiber lasers,” Proc. SPIE 7580, 75800H (2010). [CrossRef]
2. A. Tamhankar and R. Patel, “Optimization of UV laser scribing process for light emitting diode sapphire wafers,” J. Laser Appl. 23(3), 032001 (2011). [CrossRef]
3. Y. Mori, I. Kuroda, S. Nakajima, T. Sasaki, and S. Nakai, “New nonlinear optical crystal: Cesium lithium borate,” Appl. Phys. Lett. 67(13), 1818–1820 (1995). [CrossRef]
4. T. Sasaki, Y. Mori, and M. Yoshimura, “Progress in the growth of a CsLiB6O10 crystal and its application to ultraviolet light generation,” Opt. Mater. 23(1-2), 343–351 (2003). [CrossRef]
5. J. Sakuma, K. Deki, A. Finch, Y. Ohsako, and T. Yokota, “All-solid-state, high-power, deep-UV laser system based on cascaded sum-frequency mixing in CsLiB6O10 crystals,” Appl. Opt. 39(30), 5505–5511 (2000). [CrossRef] [PubMed]
6. C. Chen, “A new-type ultraviolet SHG crystal-β-BaB2O4,” Sci. Sin. Ser. B 28, 235–243 (1985).
7. H. Masuda, N. Umezu, K. Kimura, and S. Kubota, “High-repetition-rate, 192–197 nm pulse generation in β-BaB2O4 by intracavity sum-frequency-mixing of a Ti:Sapphire laser with a frequency-quadrupled Nd:YAG laser,” OSA Trends in Optics and Photonics Advanced Solid State Lasers 26, 63–69 (1997).
8. M. Nishioka, A. Kanoh, M. Yoshimura, Y. Mori, and T. Sasaki, “Growth of CsLiB6O10 crystals with high laser-damage tolerance,” J. Cryst. Growth 279(1-2), 76–81 (2005). [CrossRef]
9. H. Nakatani, W. R. Bosenberg, L. K. Cheng, and C. L. Tang, “Laser-induced damage in beta-barium metaborate,” Appl. Phys. Lett. 53(26), 2587–2589 (1988). [CrossRef]
10. K. Deki, Y. Kagebayashi, N. Kitatochi, M. Horiguchi, Y. K. Yap, Y. Mori, T. Sasaki, and K. Yoshida, “Fourth harmonic generation of Nd:YAG Laser with CLBO crystal: The properties of the long term operations,” IEICE Tech. Rep. LQE97–74, 41–46 (1997) [in Japanese].
11. K. Takachiho, M. Yoshimura, Y. Fukushima, Y. Takahashi, M. Imade, T. Sasaki, and Y. Mori, “Al doping of CsLiB6O10 for high resistance to ultraviolet-induced degradation,” Appl. Phys. Express 6(2), 022701 (2013). [CrossRef]
12. K. Kondo, M. Oka, H. Wada, T. Fukui, N. Umezu, K. Tatsuki, and S. Kubota, “Demonstration of long-term reliability of a 266-nm, continuous-wave, frequency-quadrupled solid-state laser using β-BaB2O4.,” Opt. Lett. 23(3), 195–197 (1998). [CrossRef] [PubMed]
13. K. Takachiho, M. Yoshimura, K. Masuda, Y. Takahashi, M. Imade, T. Sasaki, and Y. Mori, “Impurity and defect control of nonlinear optical crystal CsLiB6O10 for improving ultraviolet laser-induced damage tolerance,” The Review of Laser Engineering 41, 830–834 (2013).
14. M. Takahashi, A. Osada, A. Dergachev, P. F. Moulton, M. C. Raduban, T. Shimizu, and N. Sarukura, “Effects of pulse rate and temperature on nonlinear absorption of pulsed 262-nm laser light in β-BaB2O4,” Jpn. J. Appl. Phys. 49(8), 080211 (2010). [CrossRef]
17. G. Nava, P. Minzioni, W. Yan, J. Parravicini, D. Grando, E. Musso, I. Cristiani, N. Argiolas, M. Bazzan, M. V. Ciampolillo, A. Zaltron, C. Sada, and V. Degiorgio, “Zirconium-doped lithium niobate: photorefractive and electro-optical properties as a function of dopant concentration,” Opt. Mater. Express 1(2), 270–277 (2011). [CrossRef]
18. N. V. Kukhtarev, V. B. Markov, S. G. Odulov, M. S. Soskin, and V. L. Vinetskii, “Holographic storage in electro-optic crystals. I. Steady state,” Ferroelectrics 22(1), 949–960 (1978). [CrossRef]
19. V. Mürk, V. Denks, A. Dudelzak, P. P. Proulx, and V. Vassiltsenko, “Gray tracks in KTiOPO4: Mechanism of creation and bleaching,” Nucl. Instrum. Methods Phys. Res. B 141(1-4), 472–476 (1998). [CrossRef]
20. S. Wang, V. Pasiskevicius, and F. Laurell, “Dynamics of green light-induced infrared absorption in KTiOPO4 and periodically poled KTiOPO4,” J. Appl. Phys. 96(4), 2023–2028 (2004). [CrossRef]
21. U. Stamm, W. Zschocke, T. Schröder, N. Deutsch, and D. Basting, “High efficiency UV-conversion of a 1 kHz diode-pumped Nd:YAG laser system,” OSA Trends in Optics and Photonics Advanced Solid State Lasers 10, 7–9 (1997).
22. R. Bhandari, T. Taira, A. Miyamoto, Y. Furukawa, and T. Tago, “> 3 MW peak power at 266 nm using Nd:YAG/Cr4+:YAG microchip laser and fluxless-BBO,” Opt. Mater. Express 2(7), 907–913 (2012). [CrossRef]
23. K. Itoh, F. Marumo, and Y. Kuwano, “β-barium borate single crystal grown by a direct Czochralski method,” J. Cryst. Growth 106(4), 728–731 (1990). [CrossRef]