Open Access
Add to BibSonomy
Abstract
Large GaSe crystals were grown and various antireflection microstructures (ARMs) were fabricated on their cleaved surfaces using optimized femtosecond laser ablation, which provided the antireflection effect in a wide wavelength range of 4-16 µm. The influence of ARMs created on the GaSe surface on the change of the laser-induced damage threshold (LIDT) of the crystal at a wavelength of 5 μm was evaluated. The 5-µm Fe:ZnMgSe laser with the pulse duration of 135 ns was used for the LIDT test in conditions close to single pulse exposure. The measured values of LIDT of 56 ± 6 MW/cm2 and 51 ± 9 MW/cm2 for two GaSe substrates, respectively, were comparable with the known data of single pulse LIDT of GaSe. The average LIDT intensities of 54 ± 6 MW/cm2 and 52 ± 7 MW/cm2 for the ARMs at two GaSe plates, respectively, were close to LIDT intensities for the corresponding GaSe substrates. The ARMs with lower structural quality had lower LIDT (50-52 MW/cm2) in comparison with the high-quality ARMs (58-60 MW/cm2). High LIDT for high-quality ARMs can be caused by increased selenium content in the ARMs. In any case, all the tested ARMs on the GaSe plates with different surface quality are workable for development of widely tunable mid-infrared nonlinear optical converters.
© 2024 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Currently, one of the most promising nonlinear crystals designed for widely tunable conversion of laser radiation into the mid (from 2 to 15 µm) and far (> 15 µm) infrared spectral ranges is gallium selenide (GaSe) [1]. The structure of this crystal has a space group 62 m, it is transparent in a wide spectral range (0.62-20 µm). GaSe crystals have a high nonlinear susceptibility (54 pm/V at 10.6 µm), a wide range of phase matching and high LIDT (> 30 MW/cm2 at 10.6 µm, 125 ns) [1]. The refractive index no is about 2.9, and the Fresnel losses from a single surface in the mid-IR are about 23%. Therefore, the task of reducing the reflection losses for GaSe and improving the efficiency of nonlinear optical frequency converters is very relevant. A standard method of adjusting reflectance and transmittance is to apply a thin-film coating on the optical element surfaces [2]. There are some attempts dedicated to the development of antireflection coatings for GaSe [3]. But since GaSe is characterized by perfect cleavage and high plasticity, poor surface adhesion, this method is not widely applied for this crystal. An alternative solution is to create special antireflection microstructure (ARM) on the crystal surface [4]. Many optical crystals with ARMs, including ZnSe and Cr:ZnSe, demonstrate LIDT corresponding to the level of a well-polished surface, which in some cases is many times higher than the value achievable with thin-film antireflection coatings [5–7]. Recently, effect of ARM on the optical properties of GaSe was studied in dependence on different laser ablation modes allowing to increase the crystal transmittance in the range from 5 to 16 µm [8,9]. ARM operational wavelength range is determined by surface morphology of microstructures [10] and can be efficiently tuned by variation of structure period. At the same time, it is important to ascertain that the use of ARM does not reduce the LIDT values, especially when using GaSe in widely tunable high-power laser systems.
In present work, large GaSe crystals with different surface quality were grown and various ARMs were fabricated on their surfaces using femtosecond laser ablation, which provided the antireflection effect in a wide wavelength range of 4-16 µm. The influence of ARMs created on the GaSe surface on the change of the laser-induced damage threshold (LIDT) of the crystal was evaluated at a wavelength of 5 μm which is a possible pump wavelength for nonlinear down-conversion in GaSe.
2. Preparation of test samples
Large GaSe crystals were grown by the modernized Bridgman-Stockbarger method with a controlled heat exchanger under conditions of low temperature gradients [9]. GaSe boules up to 70 mm in length were obtained, plane-parallel plates 10 × 10 × 1.0 mm3 (ARM-GaSe plate 1) and 5 × 5 × 2.0 mm3 (ARM-GaSe plate 2) in size were chipped perpendicular to z axis of the boule and used for further ARM production.
A femtosecond Yb:KGW laser Pharos-PH1-SP was used to create an ARM on the surface of a GaSe crystal by femtosecond laser ablation. This technique is based on local material removal due to the high energy density of a femtosecond pulse transferred to the super-heated electron plasma induced in the GaSe volume and then to the atomic lattice of the material [9]. For the production of the ARM, the 1026-nm fundamental laser radiation or the 513 nm second harmonic with pulse duration of 200 fs and a repetition rate of 200 kHz was used. The average optical power of the fundamental radiation was up to 1000 mW and the second harmonic– up to 500 mW. Three-axis Aerotech ANT-90 nanopositioners were used for positioning points on the surface of the GaSe crystal when ARMs manufacturing. A 100×NA=0.5 lens (Mitutoyo Corporation, Japan) was used to focus the laser beam on the sample surface and exceed the ablation threshold of the material.
ARMs were created by creating holes of a given shape and depth at a certain step on the crystal surface using femtosecond laser ablation. The typical depth of the holes was 1-2 microns, whereas the period and type of packaging were determined by the required wavelength range in which it was necessary to increase the transmission. The technology of creating microstructures on the surface was described earlier [5,7,9].
Figure 1 shows images of the samples. The images were obtained with an MBS10 microscope. ARM-GaSe plate 1 had four variants of structures: ARM11-ARM14. The size of each ARM was about 2 × 2 mm2. For ARM-GaSe plate 2, structures ARM21-ARM24 with the size about 0.8 × 0.8 mm2 were investigated.
Fig. 1. An image of GaSe plates with different variants of ARMs:(a) ARM-GaSe plate 1, circles show the areas of transmission measurements; (b) ARM-GaSe plate 2.
Because of the layering of GaSe due to its perfect cleavage along (001) plane, the surface of the plates, obtained by chipping and used as optical elements, may often be non-ideal, with layer boundaries on it [9]. Generally, this doesn’t affect significantly the conversion efficiency and such samples are successfully used in laser systems. However, surface quality is critical for creating microstructures. In order to evaluate the effectiveness of ARM antireflection effect on different working samples, plates with different surface quality as well as different laser ablation modes were used. Thus, in Fig. 1 one can see inhomogeneities associated with imperfections (layer boundaries, bending deformation due to the plasticity of a thin plate with a large aperture) on the surfaces.
Antireflection microstructures were obtained with the parameters specified in Table 1. Some samples were obtained without a 1.5× magnification optics which is inserted just before objective pupil. In normal case this demagnifier reduced the beam spot to provide optimal illumination of 2.8 mm objective aperture. Without it the objective aperture was overfilled and flatter intensity distribution on ablation site was achieved at cost of maximum intensity. For these experiments, average power of the second harmonic was ramped up to 500 mW to compensate an intensity decrease (marked as “without telescope” in Table 1).
Table 1. Laser ablation parameters (* without telescope).
Morphology of the sample surface was studied using a Pioneer scanning electron microscope (SEM) manufactured by Raith. Figure 2 demonstrates SEM images of the ARM11-ARM14 microstructures in the areas marked with the circles in Fig. 1. The configuration of ARM was described in [8]. As one sees the microstructure cavities are deeper for ARM11-ARM13 fabricated by the 513-nm radiation than for ARM14 fabricated by the 1026-nm radiation. An increase in the wavelength of radiation while maintaining its power density should lead to deepening the microstructure cavities [11]. However, the fact is that the beam diameters in focus at 1026 nm is 2 times larger than at 513 nm due to diffraction, and so the power density is lower at 1026 nm. Moreover, the laser power had to be significantly reduced down to 80 mW at 1026 nm (see Table 1). This was done so that the walls between the microstructure cavities would not break through with such a small period of microstructure (2 µm) at larger beam diameter, which gave shallow microstructure cavities at 1026 nm. The surface chemical composition of the ARM-GaSe plates was determined using the Hitachi SU 8020 SEM [9]: the Ga/Se ratio for the GaSe substrates was 1.08, but for the ARMs it was 0.83 and 0.94 in the microstructure cavities and in the points around the cavities, respectively, i.e. increased selenium content was recorded in ARMs.
Figure 3 shows mid-IR transmission spectra of the GaSe samples with/without ARMs measured using a Bruker Vertex 70 FTIR spectrometer combined with a microscope Hyperion 2000, with beam diameter of about 100 µm and spectral resolution of 4 cm−1. It should be emphasized that firstly various modes were tested for microstructure fabrication on ARM-GaSe plate 1, which led to a very different surface structure and transmission as well. The comparative evaluation of properties at different sites was interesting. It can be seen from Fig. 3 that the shortest wavelength of the antireflection effect is increased from 4 µm for ARMs with high structural quality (deep ARMs without debris and crushed inter-cavities walls) up to 5 µm for ARM12 and ARM13 having lower structural quality and also for ARM14 having the highest period of the microstructure (Fig. 2). ARM-GaSe plate 2 was obtained after with optimized regimes that provided high quality and transmission of ARM21-ARM24. In any case, the antireflection effect is at wavelengths from 4-5 µm up to 16 µm. For wavelengths shorter than 4 µm, low transmission of the ARMs does not allow these to be used for nonlinear conversion and should lead to a LIDT decrease. Therefore, it is important to test LIDT at 5 µm being a possible pump wavelength for nonlinear down-conversion in GaSe.
Fig. 3. Transmission spectra in the mid-IR obtained for (a) ARM-GaSe plate 1 and (b) ARM-GaSe plate 2 in comparison with GaSe without ARMs.
3. Laser system characteristics and focusing for LIDT
Typically, solid-state lasers efficiently generate radiation with a wavelength of less than 5 µm, for example, Fe:ZnSe (4.4-4.7 µm) and Dy:PbGa2S4 (4.3-4.7 µm) lasers [12]. However, the present experiment requires a laser source with a wavelength of at least 5 µm, because otherwise lightwaves diffract on the ARMs and thus the transmittance of zero-order is reduced. Recently, such a source was obtained using an Fe-doped ZnMgSe laser crystal [13]. This 5-µm Fe:ZnMgSe laser optically pumped by a 2.94-µm Er:YAG laser was used for the LIDT experiments. The laser operated at a repetition rate of 1 Hz. The laser oscillation wavelength depended on temperature. By adjusting the temperature using a nitrogen cryostat, we have fixed the laser oscillation wavelength to λ0 = 4.98 µm with a spectral width of Δλ = 0.08 µm. Figure 4 shows (a) spectrum, (b) oscillogram, and (c) beam profiles at the laser output and (inset) in a focus of a CaF2 lens with a focal length of 20 mm focusing the output radiation onto the sample under investigation. The laser spectrum was measured using a monochromator (Oriel 77250) together with a liquid nitrogen cooled mercury-cadmium-telluride (MCT) detector (Judson-Teledyne J15). The laser pulse duration (FWHM) measured by Vigo PVI-6 photodetector + Tektronix DPO 4042 oscilloscope was as high as 135 ns. The maximum pulse energy (at the exposure site) was about 1.2 mJ. The type of installation for LIDT measurement is shown in Fig. 5. During the LIDT measurement, the 5 µm radiation was attenuated by calibrated filters.
Fig. 4. (a) Spectrum, (b) oscillogram, and (c) beam profile at the Fe:ZnMgSe laser output and at the focus of a 20-mm focal length lens (in the inset).
A small part of the ∼5 µm radiation was reflected by a CaF2 plate onto the energy probe and therefore energy of each shot was measured. This ensured overcoming the problem of instability of the laser pulse energy from pulse to pulse. The output beam profile measured by a beam profiling camera (SpiriconPyrocam III) was close to Gaussian on both axes. The radiation was focused by the CaF2 lens onto the sample under investigation. The focused beam waist radius of rx = 70 µm (beam quality factor Mx2 = 4.3) and ry = 86 µm (beam quality factor My2 = 6.4) at a 1/e2 level was measured using an alternative knife-edge technique [14] because the camera pixel (80 µm) was close to the focused beam size (see the inset in Fig. 4(c)). The tested plate was moved by a three-coordinated positioner. LIDT was registered mostly by plasma generation that appeared on the camera.
4. Experimental results
Statistical processing of the experimental results was carried by the R-on-1 LIDT test [15] because the testing area was too small for a detailed analysis. The measured values are presented in Fig. 6 for the ARM-GaSe plate 1, Fig. 7 for the ARM-GaSe plate 2, and Table 2 for both the ARM-GaSe plates.
Fig. 6. The measured results of LIDT for the ARM-GaSe plate 1 (directly measured points and average lines) and photos of the damaged samples.
Fig. 7. The measured results of LIDT for the ARM-GaSe plate 2 (directly measured points and average lines).
Table 2. Measured results: the average LIDT intensity ILIDT; the number of damaged sites N; the Student’s coefficient K at the confidence probability of 0.95; the mean confidence interval ΔILIDT.
The LIDT intensity at ith site of the ARM-GaSe plate was calculated as Ii = 2Ei/(τp π rx ry), where Ei is the measured laser pulse energy of LIDT at ith site of the ARM-GaSe plate, τp is the laser pulse duration (FWHM), rx,y is the laser beam radius (HW1/e2 M), coefficient 2 is due to the Gaussian beam profile. The laser pulse energy was measured by a high-sensitive energy probe (Coherent J25 MB-LE) in each shot to minimize a measurement error. The τp and rx,y values were measured previously (see Section 2.2). The average LIDT intensity ILIDT was calculated as
where N is the number of damaged sites on the ARM-GaSe plate. The mean confidence interval was calculated as5. Discussion
The known data of mid-IR LIDT of GaSe for a long 125-ns pulse duration with a 20-Hz repetition rate at a wavelength of 10.6 µm is as high as 33 MW/cm2 [16]. For shorter pulse duration of 30 ns, the known data of single pulse LIDT of GaSe at a wavelength of 9.55 µm is as high as 121 MW/cm2 [17]. For high repetition rates of 10 and 20 kHz, the GaSe LIDT data are as high as 95 MW/cm2 (2.1 µm, 15 ns) and 50 MW/cm2 (2.1 µm, 23 ns) respectively [18]. The decrease in LIDT with increasing the repetition rate can be explained by the cumulative effect of damage.
Our measured value for the GaSe ARM’s substrate 1 is 56 ± 6 MW/cm2 (Fig. 6). It is in agreement with the known value at 9.55 µm, 30 ns [17] if we took into account the theoretical proportionality to the square root of the pulse duration [19], i.e. $121 \cdot \sqrt {30/135} \approx 57$. MW/cm2. But our value is 1.7 times higher than the earlier result (33 MW/cm2) at 10.6 µm for almost the same pulse duration [16]. This discrepancy can be explained by the fact that the earlier result was obtained from the cumulative effect of damage under exposure to 1000 pulses per site, whereas our LIDT was tested under conditions close to single pulse exposure.
For the GaSe ARM’s substrate 2 we had LIDT of 51 ± 9 MW/cm2. It was slightly lower than for the GaSe ARM’s substrate 1 and also had higher variation of values around the average value (Fig. 7), and so we had to measure higher number of points. This is due to lower quality of the GaSe substrate 2 in comparison with the GaSe substrate 1.
Measurement of LIDT for ARMs of the ARM-GaSe plate 1 gave slightly different values from ARM11 to ARM14. For ARM11 and ARM14 we had higher LIDT of 60 ± 11 and 58 ± 10 MW/cm2, respectively, but for ARM12 and ARM13 we had lower LIDT of 52 ± 6 and 50 ± 9 MW/cm2, respectively. It can be explained by lower quality of ARM12 and ARM13 that is demonstrated in Fig. 2. Average LIDT for all four ARMs (ARM11-ARM14) of 54 ± 6 MW/cm2 is close to the value for the GaSe ARM’s substrate 1 (56 ± 6 MW/cm2).
The same measurement for ARMs of the ARM-GaSe plate 2 showed a very close average value of LIDT of 50-53 MW/cm2 both for all four ARMs (ARM21-ARM24) and for the substrate 2, but with higher variation of values around the average value as for the GaSe substrate 2. This correlates with high quality of all fabricated ARMs on the ARM-GaSe plate 2.
Our observations suggest that slightly increased (within the mean confidence interval) average LIDT for the high-quality ARMs in comparison with average LIDT for the corresponding GaSe substrate might be due to the increased selenium content in the ARMs. This hypothesis will be addressed in our future research including SEM imaging for the regions with laser induced damage.
6. Conclusion
Large GaSe crystals with different surface quality were grown and various ARMs were fabricated on their surfaces using femtosecond laser ablation, which provided the antireflection effect in a wide wavelength range of 4-16 µm. The influence of ARMs created on the GaSe surface on the change of the laser-induced damage threshold (LIDT) of the crystal was evaluated at a wavelength of 5 μm. The Fe:ZnMgSe laser operating at 5 µm with the pulse duration of 135 ns was used for the LIDT test. The measured values of LIDT of 56 ± 6 MW/cm2 and 51 ± 9 MW/cm2 for the GaSe substrate 1 and 2, respectively, obtained in conditions close to single pulse exposure are higher than the known value (33 MW/cm2) obtained at a cumulative effect of 1000 pulses [16]. However, our values are comparable with the known data of single pulse LIDT [17]. The average LIDT intensities of 54 ± 6 MW/cm2 and 52 ± 7 MW/cm2 for ARMs at the ARM-GaSe plate 1 and 2, respectively, were close to the LIDT intensities for the GaSe substrates. The ARMs with lower quality had lower LIDT (50-52 MW/cm2) in comparison with the high-quality ARMs (58-60 MW/cm2). High LIDT for high-quality ARMs can be caused by the increased selenium content in the ARMs. In any case, all the tested ARMs on the GaSe plates with different surface quality are applicable for development of mid-infrared nonlinear optical converters.
Funding
Ministry of Science and Higher Education of the Russian Federation (122041400031-2); Russian Science Foundation (20-72-10027-P).
Acknowledgments
The authors thank Michal Jelínek (Czech Technical University in Prague) for technical support and valuable discussions and A.A. Shklyaev (“VTAN” (ATRC) of the NSU Physics Department) for SEM images. P.K., L.I., A.Y., A.G, S.S., A.G., A.B., A.A. and M.T. conducted ARM fabrication, transmission and LIDT investigations supported by the Russian Science Foundation (20-72-10027-P). S.L. conducted crystal growth supported by the Ministry of Science and Higher Education of the Russian Federation (state assignment of IGM SB RAS № 122041400031-2).
Disclosures
The authors declare that they have no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. D. N. Nikogosyan, Nonlinear optical crystals: a complete survey (Springer Science + Business Media, 2005).
2. H. K. Raut, V. A. Ganesh, A. S. Nair, et al., “Anti-reflective coatings: A critical, in-depth review,” Energy Environ. Sci. 4(10), 3779–3804 (2011). [CrossRef]
3. R. A. Redkin, D. A. Kobtsev, S. A. Bereznaya, et al., “GaSe crystals with antireflection coatings for terahertz generation,” Mater. Res. Express 6(12), 126201 (2019). [CrossRef]
4. D. S. Hobbs and B. D. MacLeod, “SPIE Proceedings,” Proc. SPIE5786, 349–364 (2005). [CrossRef]
5. M. K. Tarabrin, A. A. Bushunov, V. A. Lazarev, et al., “Fabrication of anti-reflection microstructures on znse single crystal by using femtosecond laser pulses,” in OSA Technical Digest, Frontiers in Optics 2017, Washington, USA, 18-21 September 2017.
6. D. Hobbs, B. MacLeod, E. Sabatino, et al., “Laser testing of anti-reflection micro-structures fabricated in ZnSe and chromium-ion doped ZnSe laser gain media,” Opt. Mater. Express 7(9), 3377–3388 (2017). [CrossRef]
7. A. A. Bushunov, M. K. Tarabrin, V. A. Lazarev, et al., “Fabrication of anti-reflective microstructures on chalcogenide crystals by femtosecond laser ablation,” Opt. Mater. Express 9(4), 1689–1697 (2019). [CrossRef]
8. A. A. Bushunov, A. A. Teslenko, M. K. Tarabrin, et al., “Fabrication of antireflection microstructures on the surface of GaSe crystal by single-pulse femtosecond laser ablation,” Opt. Lett. 45(21), 5994–5997 (2020). [CrossRef]
9. A. P. Yelisseyev, L. I. Isaenko, S. I. Lobanov, et al., “Effect of antireflection microstructures on the optical properties of GaSe,” Opt. Mater. Express 12(4), 1593–1608 (2022). [CrossRef]
10. A. A. Bushunov, M. K. Tarabrin, and V. A. Lazarev, “Review of surface modification technologies for mid-infrared antireflection microstructures fabrication,” Laser Photonics Rev. 15(5), 2000202 (2021). [CrossRef]
11. B. H. Christensen and P. Balling, “Modeling ultrashort-pulse laser ablation of dielectric materials,” Phys. Rev. B 79(15), 155424 (2009). [CrossRef]
12. M. E. Doroshenko and V. V. Osiko, “Compact solid-state mid-IR lasers. Devoted to the 100th anniversary of academician A M Prokhorov,” Laser Phys. 26(8), 084009 (2016). [CrossRef]
13. M. E. Doroshenko, V. V. Osiko, H. Jelínková, et al., “Spectroscopic and laser properties of bulk iron doped zinc magnesium selenide Fe:ZnMgSe generating at 45 – 51 µm,” Opt. Express 24(17), 19824–19834 (2016). [CrossRef]
14. M. A. C. de Araujo, R. Silva, E. de Lima, et al., “Measurement of Gaussian laser beam radius using the knife-edge technique: improvement on data analysis,” Appl. Opt. 48(2), 393–396 (2009). [CrossRef]
15. Lidaris. Laser-Induced Damage Testing (LIDT) Services. https://lidaris.com/tests/r-on-1-lidt-test/, 2023 (accessed 2 February 2023).
16. G. B. Abdullaev, K. R. Allakhverdiev, M. E. Karasev, et al., “Efficient generation of the second harmonic of CO2 laser radiation in a GaSe crystal,” Sov. J. Quantum Electron. 19(4), 494–498 (1989). [CrossRef]
17. Y. M. Andreev, V. V. Badikov, V. G. Voevodin, et al., “Radiation resistance of nonlinear crystals at a wavelength of 9.55 μm,” Quantum Electron. 31(12), 1075–1078 (2001). [CrossRef]
18. Ch. Zhu, V. Dyomin, N. Yudin, et al., “Laser-induced damage threshold of nonlinear GaSe and GaSe:In crystals upon exposure to pulsed radiation at a wavelength of 2.1 μm,” Appl. Sci. 11(3), 1208 (2021). [CrossRef]
19. M. H. Niemz, “Threshold dependence of laser-induced optical breakdown on pulse duration,” Appl. Phys. Lett. 66(10), 1181–1183 (1995). [CrossRef]
