Open Access
Add to BibSonomy
Abstract
In this work, the performance of Ca5(BO3)3F (CBF) single crystals was investigated for the third harmonic generation at 355 nm. A high energy conversion efficiency of 16.9% at 355 nm was reached using a two-conversion-stage setup. First, using a high peak power, passively Q-switched Nd3+:YAG/Cr4+:YAG microlaser based gain aperture in micro-MOPA, the second harmonic at 532 nm was achieved with lithium triborate (LBO) crystal, reaching 1.35 MW peak power. On a second step, laser pulses at 355 nm were generated using a 5 mm-long CBF crystal growth by TSSG method with energy, pulse duration and peak power of 479 µJ, 568 ps and 0.843 MW, respectively. These results are currently the highest reported for CBF material.
© 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Nowadays, high-power UV lasers are useful in a wide range of applications. Knowing a lot of materials (metals, ceramics or organic polymers) exhibit high absorption in the UV range, these devices can be used in many domains including micromachining, spectroscopy or photolithography of semiconductors [1,2]. With the recent improvements in microlaser amplification we can obtain compact, high peak power and good beam quality near-IR systems that are useful for high power frequency conversion [3]. Therefore, an attractive way to produce UV-range wavelength is to develop solid-state lasers that uses frequency conversion through suitable nonlinear optical (NLO) crystal. To generate the third harmonic from a 1 µm Nd-based microchip laser, three nonlinear crystals are commercially proposed: LBO (LiB3O5) crystal, BIBO (BiB3O6) crystal and BBO (BaB2O4) crystal [4–6]. It has been demonstrated around 20% conversion efficiency for BBO and LBO crystals [7] and up to 39% conversion from IR to UV for BIBO crystal [8]. However, these crystals present some drawbacks such as hygroscopy in the case of LBO and BBO which can altered the surface and damaged it under long-term UV operation [9], or photo-induced damage in the case of BIBO crystal [10]. All of these effects can limit the lifetime of crystals and their application in compact and high-power UV sources. In this paper, we investigate a non-hygroscopic Ca5(BO3)3F (CBF) single crystal [11] as a new promising material for UV light generation. CBF is a biaxial nonlinear crystal that crystallizes in the space group Cm (Z = 2) [12]. Figure 1 shows the crystal sample used, which was grown by top seeded solution growth method (TSSG) using a 20 wt% LiF flux [13]. The nonlinear properties of CBF crystal was investigated in a previous work [14]. In the present work, the third harmonic generation (THG) of a 1064 nm Nd-based microlaser was realized using a 5 mm-long CBF crystal oriented along XY axis and the efficiency was compared with the THG obtained with a LBO crystal in the same experimental conditions. Table 1 compares some nonlinear crystals used for third harmonic generation at 355 nm with CBF crystal. Even if BBO and BIBO crystal exhibit high effective nonlinear coefficient deff, their angular acceptance is very narrow and their walk-off is quite high, which can limit their application in commercial devices. Moreover, it can be seen that both LBO and CBF crystal have similar nonlinear coefficient. As a consequence, similar UV conversion performances can be expected for both crystals.
Fig. 1. Ca(BO3)3F crystal growth by TSSG method using 20 wt% LiF flux. The white dots inside the crystal are LiF inclusions which appear during the growth. The circle shows the area where the highest conversion efficiency was obtained.
Table 1. Commercial nonlinear crystals parameters compared to CBF parameters for type-II THG at 355 nm.
2. Experimental setup for third harmonic generation
A passively Q-switched Nd3+:YAG/Cr4+:YAG-based gain aperture of micro-MOPA (Master Oscillator Power Amplifier) was build using 4 mm-thick Nd3+:YAG crystal 1.1at-% doped. The saturable absorber (SA) was a 3 mm-thick Cr4+:YAG crystal with initial transmission of T0 = 40%. An output coupler (OC) mirror with 50% reflectivity was used. The total cavity length was 12 mm. The cavity was pumped by 808 nm laser diode (600 µm fiber core diameter) with 120 W peak power in quasi-CW mode at 100 Hz repetition rate and 215 µs pump pulse duration. The optical amplifier was composed of a 4 mm Nd3+:YAG crystal pumped by the same type laser diode as the oscillator. The principle of this system is to improve the output energy and the beam quality by amplifying only the single transverse electro-magnetic mode TEM00 of the beam and not the higher order ones as shown in Fig. 2 [15]. The M2 factor decreases from 2.5 before amplification to about 1.1 after the gain aperture.
After double-pass amplification through the crystal, the beam was extracted by a combination of quarter waveplate and polarizing beam splitter (PBS). This laser produced pulses over 4 mJ with 1.1 ns pulse duration. The fundamental IR beam is then focused with a 175 mm plano-convex lens inside an LBO crystal to generate the second harmonic (SHG) of the laser at 532 nm. The SHG is performed using critical phase matching (θ = 90° and Ф = 11.6° cutting angles) for the LBO crystal which is placed on a XYZ-3D rotational stage. Both residual IR light and green light pass through the CBF crystal to generate the third harmonic at 355 nm. The UV-converter crystal is also set on a XYZ-3D rotational stage to tune the phase matching angles. As shown in Fig. 3, the UV beam was finally separated from the IR and green beams using three dichroic mirrors with 99% reflection (HR) coating for 355 nm and high transmission (HT) for 1064 nm and 532 nm. Indeed, it was noticed that despite high transmission for green light, some residual light from the SHG was reflected on the dichroic mirrors and could affect the measurement.
3. Experimental results and discussion
First, the best conditions for THG were determined: the CBF crystal aperture was scanned to find the area where the highest UV output is obtained for a fixed IR and SHG input energy. Then, the SHG energy conversion efficiency (η2ω = E2ω / Eω, with E2ω the energy of the SHG wave and Eω the energy of the fundamental IR wave) was gradually changed by modifying the phase matching angles of the LBO crystal set on the 3D rotational stage to find the maximum UV output energy. This maximum THG conversion efficiency (η3ω = E3ω / Eω, with E3ω the energy at 355 nm) was reached for 37% SHG energy efficiency and then decrease with SHG efficiency as it can be seen in Fig. 4. Whereas the THG process required one IR photon for one green photon, the results show that the maximum UV output is reached for around 37% SHG conversion efficiency, which correspond to around 4 IR residual photons for only one green. This difference could be explained by the beam quality degradation of the residual IR beam after the SHG process under high intensity whereas the quality of the green beam generated is closed to the quality of the input IR beam []. In that case, the interaction between the residual IR beam and the nonlinear crystal will be weaker. As a consequence, more IR photons will be required for the THG process. All the next measurements were conducted using 37% conversion efficiency. This efficiency was kept constant by tuning the phase matching angles of the LBO crystal for each measured point while increasing the IR input energy.
The maximum THG energy efficiency for CBF crystal reached 16.9% after taking to account loss from reflections on the crystal surfaces, with an output energy of 479 µJ and 568 ps pulse duration. These results were compared with the ones obtained using LBO crystal cut for THG with AR coating for 1064 nm, 532 nm and 355 nm. The LBO maximum output reached 428 µJ of UV energy with 1.0 ns pulse duration as presented in Fig. 5(a). The difference between LBO and CBF pulse duration could be explained by saturation effects occurring in LBO crystal in these experimental conditions. Indeed, the THG efficiency for LBO crystal reached 20.2% for around 1 mJ of input energy and then decrease and stabilize around 13% at the same conditions as CBF material in Fig. 5(b). Also, the crystal length of the CBF was only 5 mm whereas the LBO crystal was twice longer. In that case, higher conversion could be expected using a longer CBF crystal with better optical quality.
Fig. 5. Experimental data on THG for uncoated CBF crystal and comparison with LBO crystal (anti-reflection coating for 1064 nm, 532 nm and 355 nm) in terms of output UV energy (a) and of conversion efficiency (b). The Rayleigh length was 403 mm and the fundamental beam size at the waist position was 2w0 = 0.75 mm.
Moreover, the same THG measurement was conducted on LBO and CBF crystals but with wider focusing conditions (the 175 mm lens was replaced by longer focal distances lenses). The results in Fig. 6(a) show that for the CBF crystal, the tighter the focusing of the input beam is, the higher the UV conversion efficiency is. In that case, high intensity is suitable for obtaining high UV conversion with CBF crystal. Concerning the LBO crystal, when the input beam size increases (i.e. the input intensity decrease), the decreasing effect on conversion efficiency disappear and the saturation does not occur anymore for the larger beam size as show in Fig. 6(b). The results show that the saturation effect in LBO crystal is an intensity-related effect. This effect is probably due to nonlinear effects such as multi-photons absorption occurring at high intensities inside LBO crystal. Because of the large bandgap of LBO (7.7 eV) any two-photon absorption process at 355 nm (energy of two-photon process at 355 nm is 6.98 eV) is impossible. However, some three photons absorption is still possible and have already been observed in LBO [16]. These results show that CBF crystal exhibits good performances for high power THG compared to commonly used LBO crystal.
Fig. 6. Experimental data on THG for different focusing conditions in CBF (a) and LBO (b) crystals. The Rayleigh lengths was 403 mm, 861 mm and 1325 mm for the fundamental beam sizes at the waist position of 2w0 = 0.75 mm, 1.08 mm and 1.34 mm, respectively.
However, after THG measurement, the CBF crystal exhibited very slight yellow color as compared to samples which were not used for high power measurement. The transmission spectrum of this crystal was recorded using a Cary 6000i spectrometer in Fig. 7. First a wide absorption band can be noticed around 450 nm in Fig. 7 inset. This band probably comes from a colored center created under high intensity illumination for long time, but the exact mechanism of this phenomenon is still under investigation. The CBF crystal was put in an oven at 300°C during 2 hours in order to try to recover the initial state of the crystal. As it can be seen in Fig. 6, after this thermal treatment the absorption band disappears. The transmission of CBF crystal increased by 1.72% after the heat treatment. That highlight the fact that the colored center which appear inside the crystal after long time irradiation could limit the performances of the crystal. Also, two narrow absorption bands can be observed around 235 nm and 288 nm, respectively. These two bands have already been observed [17] in borates-based crystals and in particular in YAl3(BO3)4 (YAB) crystal [18] and in KABO crystal [19]. These bands probably correspond to the charge transfer transitions between O2- and Fe3+ impurities coming from some pollution of the crucible during the crystal growth. However, all of these absorption bands are not in the wavelength ranges of interest (1064 nm, 532 nm and 355 nm). Therefore, the UV conversion performances of CBF crystal should not be seriously affected by these absorptions as seen in Table 2. These results observe for the first time the formation of some color centers under high intensity laser irradiation. Some deeper study has to be done to fully understand the mechanism of the formation of these centers and this also highlight some improvements to be done on the crystal quality of the CBF in addition of the reduction of the LiF inclusions in order to increase the life time of the crystal.
Fig. 7. Transmission spectrum of CBF crystal after thermal treatment at 300°C for 2 hours. The cutting wavelength in UV region was measured at 200 nm. The inset shows the transmission spectrum in the visible region before and after the thermal treatment. Before heating the CBF crystal, a large absorption band centered around 450 nm can be noticed due to defects induced by laser tests.
Table 2. Transmission of CBF crystal for THG of Nd3+:YAG laser at 1064 nm.
4. Conclusion
In this work THG of a Nd3+:YAG-based microchip laser was performed using type-II CBF crystal for frequency-tripling of 1064 nm fundamental light. Using an amplifier system, a high UV light generation efficiency was achieved for CBF single crystal with 16.9% THG energy efficiency, 479 µJ output energy and 0.843 MW peak power. Also, a yellow color center formation was noticed for the first time inside the CBF crystal after being used under high intensity UV conversion tests. The exact mechanism is still under investigation, but it shows some new requirement in increasing CBF crystal quality. Indeed, better performances can be expected from CBF crystal owing to its actual poorer optical quality compared to LBO regarding the comparative results obtained under high intensity pumping. The CBF crystal is shown as a promising candidate for high energy and high repetition rate UV microlaser. Future improvement in CBF crystal quality aims at reaching several mJ-level UV light generation at 355 nm.
Funding
Japan Science and Technology Agency (JPMJMI17A1); Institute for Molecular Science.
Acknowledgments
The authors acknowledge supports from Dr. Kawasaki and Dr. Lim from Institute for Molecular Science (IMS).
Disclosures
The authors declare no conflicts of interest.
References
1. S. Stefan, V. Frank, and B. Igor, “High-power excimer lasers for 157-nm lithography,” Proc. SPIE 5040, 1344–1351 (2003). [CrossRef]
2. E. A. Waddell, L. E. Locascio, and G. W. Kramer, “UV Laser Micromachining of Polymers for Microfluidic Applications,” JALA 7(1), 78–82 (2002). [CrossRef]
3. R. Bhandari and T. Taira, “Palm-top size megawatt peak power ultraviolet microlaser,” Opt. Eng. 52(7), 076102 (2013). [CrossRef]
4. V. P. Gapontsev, V. A. Tyrtyshnyy, O. I. Vershinin, B. L. Davydov, and D. A. Oulianov, “Third harmonic frequency generation by type-I critically phase matched LiB3O5 by means of optically active quartz crystal,” Opt. Express 21(3), 3715–3720 (2013). [CrossRef]
5. M. Ghotbi, Z. Sun, A. Majchrowski, E. Michalski, I. V. Kityk, and M. Ebrahim-Zadeh, “Efficient third harmonic generation of microjoule picosecond pulses at 355 nm in BiB3O6,” Appl. Phys. Lett. 89(17), 173124 (2006). [CrossRef]
6. R. Wu, “High-efficiency and compact blue source: intracavity frequency tripling by using LBO and BBO without the influence of birefringence,” Appl. Opt. 32(6), 971–975 (1993). [CrossRef]
7. A. Borsutzky, R. Brünger, C. Huang, and R. Wallenstein, “Harmonic and Sum-Frequency Generation of Pulsed Laser Radiation in BBO, LBO, and KD*P,” Appl. Phys. B 52(1), 55–62 (1991). [CrossRef]
8. M. Peltz, J. Bartschke, A. Borsutzky, R. Wallenstein, S. Vernay, T. Salva, and D. Rytz, “Harmonic generation in bismuth triborate (BiB3O6),” Appl. Phys. B 81(4), 487–495 (2005). [CrossRef]
9. H. Hong, Q. Liu, L. Huang, and M. Gong, “Improvement and formation of UV-induced damage on LBO crystal surface during long-term high-power third-harmonic generation,” Opt. Express 21(6), 7285–7293 (2013). [CrossRef]
10. J. H. Jang, I. H. Yoon, and C. S. Yoon, “Cause and repair of optical damage in nonlinear optical crystal of BiB3O6,” Opt. Mater. 31(6), 781–783 (2009). [CrossRef]
11. K. Xu, P. Loiseau, G. Aka, and J. Lejay, “A new promising nonlinear optical crystal for ultraviolet light generation: Ca5(BO3)3F,” Cryst. Growth Des. 9(5), 2235–2239 (2009). [CrossRef]
12. L. Shirong, H. Qinzhen, Z. Yifan, J. Aidong, and C. Chuangtian, “Structure of Calcium Fluoroborate, Ca5(BO3)3F,” Acta Crystallogr., Sect. C: Cryst. Struct. Commun. 45(12), 1861–1863 (1989). [CrossRef]
13. L. Deyra, S. Ilas, X. Délen, P. Loiseau, F. Balembois, G. Aka, F. Salin, and P. Georges, “Third harmonic generation at 343 nm in nonlinear Ca5(BO3)3F (CBF) crystals,” Opt. Mater. Express 3(11), 1798–1802 (2013). [CrossRef]
14. K. Xu, P. Loiseau, G. Aka, R. Maillard, A. Maillard, and T. Taira, “Nonlinear optical properties of Ca5(BO3)3F crystal,” Opt. Express 16(22), 17735–17744 (2008). [CrossRef]
15. V. Yahia and T. Taira, “High brightness energetic pulses delivered by compact microchip-MOPA system,” Opt. Express 26(7), 8609–8618 (2018). [CrossRef]
16. C. Mühlig and S. Bublitz, “Absolute absorption measurements in nonlinear optical crystals,” Laser Congress 2019 (ASSL, LAC, LS&C), OSA Technical Digest (Optical Society of America, 2019), paper JTu3A.24.
17. J. Ren, “Elaboration et caractérisation de monocristaux de borate pour la conversion de fréquences laser dans le domaine UV”, PhD, UPMC (2014).
18. S. Ilas, P. Loiseau, G. Aka, and T. Taira, “240 kW peak power at 266 nm in nonlinear YAl3(BO3)4 single crystal,” Opt. Express 22(24), 30325–32 (2014). [CrossRef]
19. L. Liu, C. Liu, X. Wang, Z. G. Hu, R. K. Li, and C. T. Chen, “Impact of Fe3+ on UV absorption of K2Al2B2O7 crystal,” Solid State Sci. 11(4), 841–844 (2009). [CrossRef]
