Open Access
Add to BibSonomy
Abstract
Multilayer dielectric gratings (MLDGs) have been widely used for pulse compression in chirped pulse amplification technology, and encounter amplified nanosecond (ns), picosecond, or femtosecond laser pulse irradiation. Damage behavior in the ns regime is statistically significant; however, only the 1-on-1 test method was employed in previous studies to identify the damage precursors. Here, we adopted a raster scan procedure with mass test samplings to comprehensively evaluate the damage characteristics of MLDGs. The damage experiment was conducted at 1064 nm with a pulse width of 8 ns. The laser-induced damage thresholds (LIDTs) for the MLDGs were shown to be approximately 30% lower than those of multilayer dielectric films (MLDFs). The normalized electric field intensity |E |2 (EFI) enhancement caused by the surface-relief grating structure and incomplete grating cleaning contributed to this LIDT reduction. Three discrete damage-initiation morphologies near the LIDT were found: nodular ejection, nano absorbing defect damage, and plasma scalding. In addition to the nodular defect damage that usually occurs in the fundamental frequency high reflectors, the strong absorption of nano defects and the poor interfacial quality make the interface nano absorbing defects of the MLDG also easily triggered. The interface differences between the MLDG and MLDF should be related to multiple annealing processes during MLDG fabrication. The plasma scalding behaves as a color change and is only involved at the surface of the grating pillar. The slight dependence of damage morphology on the EFI peak was first observed.
© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Chirped pulse amplification (CPA) technology has allowed the generation of petawatt peak power pulses and focal intensities on the order of 1021 W/cm−2 [1–5]. While the CPA technique is a highly effective method for eliminating short-pulse intensity-dependent damage to the amplifier components, it does not eliminate the final optic damage. The laser damage resistance of the final pulse compression grating limits the final energy output of an ultra-high-intensity laser system [6,7]. Multilayer dielectric gratings (MLDGs), which exhibit a high laser-induced damage threshold (LIDT) and high diffraction efficiency (DE), are popular solutions for CPA-based laser systems [8,9].
In CPA, the low-energy ultra-short pulse generated from the oscillator is stretched to a long pulse of nanoseconds (ns) or several hundred picoseconds (ps), and then the stretched long pulse is amplified and recompressed as an ultrashort pulse of a femtosecond (fs) or several ps duration [10,11]. Therefore, the MLDGs in the final compressor encounter amplified ns, ps, or fs laser pulse irradiation. Many studies have been carried out on fs and ps short-pulse laser damage of MLDGs. For ultrashort pulses, it is evident that the damage is deterministic, and the initial damage morphology is localized to the back edges of the grating pillar [12], in correspondence with the electric field intensity (EFI) maximum. For fixed incident angles and materials, the measured LIDT was directly related to the EFI maximum in the material [13] and was reported to agree well with the value simulated in multiphoton ionization mode. For sub-ps laser pulses, a raster scan procedure was introduced to qualify the damage density of MLDG, and sparse precursors with lower damage resistances were found, which were not detected by the 1-on-1 or R-on-1 test methods [14]. However, few studies have focused on the damage characteristics of MLDGs in the ns regime. Only the 1-on-1 procedure was employed to evaluate the ns LIDTs of MLDGs in previous studies. Rough studies on the whole grating pillar damage and the groove bottom damage were observed under ns laser irradiation. [15,16]. However, unlike the deterministic damage induced by ultra-short laser pulses, damage in the nanosecond regime has obvious statistical behavior that is related to the initiation of precursor defects [18]. Therefore, the 1-on-1 test method with a small number of samplings was unable to evaluate the LIDT as well as the critical defects induced by the ns laser pulse effectively.
In this study, raster scan [17,18] with mass samplings was first adopted in ns laser irradiation to identify low-density defects and comprehensively evaluate the laser damage resistance of the MLDGs. The detailed results obtained in this study suggest that the LIDTs of MLDGs are approximately 30% lower than those of MLDFs. Three types of damage characteristics were observed near the LIDT. They were nodular ejection, nano absorbing defect damage, and plasma scalding. Nodular ejection damage in MLDG is similar to that observed in the fundamental frequency high-reflection film damage [19,20]. Besides, the strong absorption of nano defects located at the poor-quality interface also easily induced damage. Plasma scalding was found to affect only the surface of the grating pillars. In addition, we also found that the damage process in the ns regime depends slightly on the normalized EFI ${|E |^2}$ distribution.
2. Experiment
2.1 Sample preparation
The MLDGs were designed to have a groove density of 1740 l/mm, for which the Littrow angle was 67°. MLDFs for manufacturing the MLDGs were deposited on 50 mm ${\times} $ 50 mm ${\times} $ 1.5 mm fused silica substrates using E-beam evaporation. The main stack was an HLL design, as shown in Fig. 1(a), where H and L refer to HfO2 and SiO2, respectively. An approximately 540 nm thick SiO2 layer was used to fabricate the surface-relief grating structure. The spectral characteristics of the MLDF at an incidence angle of 67° with TE polarization are shown in Fig. 2(a). The photoresist material was spun onto the MLDFs and a photoresist grating structure was manufactured using a holographic method on the mirror, and the structures of the photoresist gratings were transferred to the SiO2 top layer of the MLDF by reactive ion-beam etching.
Fig. 2. (a) Spectral characteristic of the MLDF at an incidence angle of 67°. (b) -1st order DE of MLDG.
A schematic diagram of the MLDG is shown in Fig. 1(b), where W1, W2, W, D, T, f, and ${\mathrm{\theta }_\textrm{g}}$ represent the upper width of the trapezoidal grating, bottom width of the trapezoidal grating, pillar thickness at half-height, groove depth, period, duty cycle, and base angle of the grating pillar, respectively. The measured duty cycle f, defined as $\textrm{W}/\textrm{T}$, is 0.38 with a base angle of 87°. The -1st order DE of the well-cleaned samples is greater than 97% at an incident angle of 67° for TE-polarized laser light at a wavelength of 1064 nm, as shown in Fig. 2(b).
2.2 Experimental setup and method
A schematic of the laser damage test setup is shown in Fig. 3. Laser damage experiments were performed using an 8 ns pulse from a 1064 nm Nd: YAG laser at an incidence angle of 67° with TE polarization, which generated a near-Gaussian spatial profile. The measured effective area of the focused beam spot is 0.61 mm2 at normal incidence. A raster scan with 825 test samplings in an area of 1 cm2 per fluence was performed. The damage is monitored by online plasma flash detection and a charge-coupled device (CCD). LIDT was defined as the maximum fluence at which no damage appeared in the test area and was given in the beam normal in this study.
The detailed features of the damage morphologies were characterized using scanning electron microscopy (SEM, Zeiss Auriga) combined with a focused ion beam (FIB). An atomic force microscopy (AFM, Dimension Icon) was used to determine the grating groove and structural parameters. X-ray photoelectron spectroscopy (XPS, Thermo Scientific K-Alpha) was performed on different samples to detect residual elements. Observation and analysis of the available cross-sectional morphology of MLDF and MLDG were performed using transmission electron microscopy (TEM, Thermo Scientific Talos F200X).
3. Results
3.1 LIDT results
Considering the sample differences, multiple samples prepared under the same process were carried out for experiments, and the LIDT results are shown in Table 1. The LIDTs for the MLDFs are 30.0 J/cm2 or 35.0 J/cm2 while the LIDTs for the MLDGs are 20.5 J/cm2 or 23.1 J/cm2. The LIDTs of the MLDGs were shown to be approximately 30% lower than those of the MLDFs. This result is different from the previous 1-on-1 test result, which stated that the LIDTs for the MLDGs were reduced by up to approximately 70% compared with those of MLDFs [14,21].
Table 1. LIDT results for MLDFs and MLDGs
3.2 Typical damage morphologies for MLDGs
The pristine morphologies of the MLDG without laser irradiation are shown in Figs. 4(a) and (b). The original surfaces showed a clear boundary between the pillars and grooves. The statistical results after laser irradiation suggest that there are three typical damage morphologies under near the LIDT laser irradiation, which can be classified as nodular ejection, nano absorbing defect damage, and plasma scalding.
Fig. 4. (a) Top-view image of the initial morphology for MLDG. (d) Cross-sectional view of the initial morphology for MLDG.
Figure 5(a) shows the typical morphology of the first damage observed under ns laser irradiation. The features observed are similar to the damage of HfO2/SiO2 high-reflection mirrors and behave as nodular ejection pits. The diameter of these pits is of the order of micrometers. The cross-sectional view of the damage in Fig. 5(a) is shown in Fig. 5(b). It reveals that the nodular seed originated from the substrate. In addition, the damage pit is surrounded by collateral damage, such as cracks or spalling of the first layers.
Fig. 5. (a) Typical damage morphology of nodular ejection. (d) Cross-sectional view of nodular ejection (where F denotes the fluence and Fth is the threshold fluence).
The second damage morphology was induced by nano absorbing defects, as shown in Fig. 6(a). It is a shallow pit, similar to a swelling area, containing a venting hole in the center. Figure 6(b) shows a cross-sectional view of Fig. 6(a), which shows that the damage originates from the first SiO2/HfO2 interface. This implies that the defects at the interface absorb laser energy, resulting in a temperature rise and then vaporization, which breaks through the bondage of the surrounding film, and finally forms the venting hole and the surrounding film bulge. Higher-resolution interface details characterized by TEM are introduced in the following discussion.
Fig. 6. (a) Typical damage morphology of nano absorbing defect damage. (b) Sectional view of nano absorbing defect damage (where F denotes the fluence and Fth is the threshold fluence).
Figure 7(a) shows a top-view SEM image of the third damage site, called plasma scalding, which can be clearly observed because of the color change. Similar morphologies have also been reported in a variety of single layer or multilayer systems [22,23]. For the characteristics of plasma scalding in MLDG, three positions outward from the center of the damage area are chosen for local magnification observation. Figures 7(b)–(d) depict local magnified views of the positions marked by rectangles in Fig. 7(a). Figure 7(b) shows the morphology of the area closest to the damage center, where many tiny molten holes and pits appear at the surface of the grating pillar. As we move further away from the damage center, the molten characteristics on the top of the grating pillar are weakened, as shown in Fig. 7(c). Nano-scale sheds on the grating pillar surface become a typical damage characteristic at the edge of the damage area, as shown in Fig. 7(d).
Fig. 7. (a) Top-view SEM image of plasma scalding of MLDG. (b)–(d) Local magnified views of the positions marked by rectangles in Fig. 7 (a) (where F denotes the fluence and Fth is the threshold fluence).
4. Discussion
4.1 EFI enhancement and surface contamination
The experimental measurements show that the LIDTs of MLDGs are approximately 30% lower than those of MLDFs. The presence of the surface-relief grating structure changes the original EFI distribution of the MLDF significantly, which is closely related to the laser damage resistance of the MLDG [13]. The theoretically calculated normalized EFI ${|E |^2}$ distributions of MLDG and MLDF are shown in Figs. 8(a) and (b). The maximum peak of the EFI for MLDF was within the top SiO2 layer, and the corresponding value was 0.59, as shown in Fig. 8(b). However, the strongest EFI for MLDG exists in the grating pillar of the backlight side, and the maximum ${|\textrm{E} |^2}$ is 2.95, as shown in Fig. 8(a). Therefore, the significant EFI enhancement induced by the surface-relief grating structure should be one of the main factors causing the reduction in LIDT.
Fig. 8. (a) The theoretically calculated EFI distribution within MLDG. (b) The theoretically calculated EFI distribution within MLDF.
Residues and debris on the grating’s surface are left by the grating fabrication process, which can also dramatically reduce its laser damage resistance [24,25]. Therefore, a final cleaning process that removes a broad spectrum of contaminant materials is essential [26,27]. In this study, XPS was used to analyze the surface composition of MLDGs. The element name determined according to the binding energy is shown in Fig. 9(a), and the corresponding atomic percentage of each element is shown in Fig. 9(b). The graph indicates that surface contamination, such as F and Ta, remained on the surface after the cleaning process, which can increase energy absorption during laser irradiation, leading to further reduction of LIDT.
4.2 Laser damage induced by the defects and EFI distribution
For the first type of nodular ejection damage, visible nodular defects have been widely studied in multilayer dielectric films [28], which are formed by the seeds from particles during film deposition, contaminations on the substrate, and so on. The presence of a nodular defect enhances EFI. This enhanced EFI creates localized thermally induced stress and causes nodular ejection.
For the second type of nano absorbing defect damage, this type of damage morphology rarely appears in the MLDFs, but it is abundant in the MLDGs. However, it is not possible to characterize the defect morphology at the interface owing to the absence of visible defects. To further investigate these absorbing defects, we performed cross-sectional TEM analysis, as shown in Fig. 10. Compared with the interface of the MLDFs in Fig. 10(c), the differences between the SiO2/HfO2 and HfO2/SiO2 interfaces for MLDGs were observed, as shown in Fig. 10(a). Local magnified views of Figs. 10(a) and (c) are shown in Figs. 10(b) and (d), respectively. The SiO2/HfO2 interface of the MLDGs in Fig. 10(b) shows distinctly poorer quality than that of MLDF, as presented in Fig. 10(d). This interface difference should play an important role in the second type of damage. Although the EFI of the SiO2/HfO2 interface is low, it is still not equal to zero, as shown in Fig. 10 (e). The strong absorption of nano defects make the deposited laser energy sufficient. In addition, poor interfacial quality can weaken the interfacial strength. Both cause the interface nano absorbing defects of the MLDG being triggered easily. The interface of the MLDG and MLDF differences should be related to the multiple-annealing process during MLDG fabrication.
Fig. 10. (a) and (b) Poor interface quality of MLDG. (c) and (d) Interface of MLDF corresponding to the positions shown in Figs. 10 (a) and (b). (e) The calculated EFI distribution of MLDG along the yellow line in Fig. 8 (a).
The formation of the third damage site is called plasma scalding, which is related to plasma formation. Figures 11(a)–(c) show the grating pillars of the light-facing surface, and Figs. 11(d)–(g) correspond to the backlight surface. Each corresponds to the three positions shown in Fig. 7(a). All images were taken at an inclination of 52°. The results show that only the surface of the grating pillar is molten, while no damage is observed on the sidewall of the grating pillar on either the light-facing surface or the backlight surface. The surface of the MLDG appears to have “burnt” [22]. They are initiated as defects, contaminations, or air breakdown during a laser shot and are caused by an increase in temperature on the surface during plasma formation.
Fig. 11. (a)–(c) SEM images for the grating pillar of the light-facing surface. (d)–(g) SEM images for the grating pillar of the backlight surface.
The damage of MLDG irradiated by a fs pulse laser is mostly affected by EFI enhancement, and damage usually occurs on the sidewall of the grating pillars, which faces away from the incident light [8,21,22]. We found a similar phenomenon under the irradiation of a ns pulse laser. Figure 12(a) shows the cross-sectional SEM image of the laser irradiation area, and Fig. 12(b) shows a cross-sectional SEM image of the undamaged area. The difference between the two images is mainly in the grating pillar on the right and the top surface of the grating pillar, as marked by the red rectangle. A slight depression was observed corresponding to the position where the EFI enhancement was at its maximum, as shown in Fig. 12(a). Besides, the thermal melting process at the surface marked by the red rectangle shows different characteristics from ps and fs pulses laser irradiation.
5. Conclusions
A raster scan with mass samplings can identify low-density defects and comprehensively evaluate the ns laser damage characteristics of MLDG. The measured LIDT results suggest that the LIDT of MLDF decreased by approximately 30% after being fabricated into MLDG, where modulation of the EFI distribution by the surface-relief grating structure and the residual contamination on the MLDG surface are inevitably responsible for this phenomenon. Nodular ejection, nano absorbing defect damage and plasma scalding were found near the LIDT. The strong absorption of nano defects and the poor interfacial quality make the interface nano absorbing defects of the MLDG easily triggered. The interface of the MLDG and MLDF differences should be related to multiple annealing processes during MLDG fabrication. The plasma scalding involved only the top of the grating pillar for the MLDG. In addition, a slight dependence of damage morphology on the EFI peak was first observed. In this work, the damage characteristics and initiation of MLDG irradiated by a nanosecond laser have been studied thoroughly, which provides a new reference for the design optimization and guidance of the preparation process. This improved understanding, in turn, can be used to help design and fabricate materials with higher LIDTs.
Funding
National Key Research and Development Program of China (2018YFE0115900); National Natural Science Foundation of China (11704285, 11774319, 11874369, U1831211); Strategic Priority Research Program of the Chinese Academy of Sciences (XDB1603); CAS Special Research Assistant Project.
Acknowledgment
We thank Lifeng Li of Tsinghua University for providing the theoretical EFI data presented in Figs. 8(a), 8(b), and Fig. 10(e).
Disclosures
The authors declare that there are no conflicts of interest related to this study.
Data availability
The 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. Strickland and G. Mourou, “Compression of amplified chirped optical pulses,” Opt. Commun. 56(3), 219–221 (1985). [CrossRef]
2. D. A. Alessi, C. W. Carr, R. P. Hackel, R. A. Negres, K. Stanion, J. E. Fair, D. A. Cross, J. Nissen, R. Luthi, G. Guss, J. A. Britten, W. H. Gourdin, and C. Haefner, “Picosecond laser damage performance assessment of multilayer dielectric gratings in vacuum,” Opt. Express 23(12), 15532–15544 (2015). [CrossRef]
3. L. Gallais, M. Rumpel, M. Moeller, T. Dietrich, T. Graf, and M. A. Ahmed, “Investigation of laser damage of grating waveguide structures submitted to sub-picosecond pulses,” Appl. Phys. B 126(4), 69 (2020). [CrossRef]
4. Z. X. Zhang, F. X. Wu, J. B. Hu, X. J. Yang, J. Y. Gui, P. H. Ji, X. Y. Liu, C. Wang, Y. Q. Liu, and X. M. Lu, “The 1 PW/0.1 Hz laser beamline in SULF facility,” High Power Laser Sci. Eng. 8, e4 (2020). [CrossRef]
5. M. Galletti, P. Oliveira, M. Galimberti, M. Ahmad, and C. Hernandez-Gomez, “Ultra-broadband all-OPCPA petawatt facility fully based on LBO,” High Power Laser Sci. Eng. 8, e31 (2020). [CrossRef]
6. N. Bonod and J. Néauport, “Optical performance and laser induced damage threshold improvement of diffraction gratings used as compressors in ultra high intensity lasers,” Opt. Commun. 260(2), 649–655 (2006). [CrossRef]
7. T. J. Kessler, J. Bunkenburg, H. Huang, A. Kozlov, and D. D. Meyerhofer, “Demonstration of coherent addition of multiple gratings for high-energy chirped-pulse-amplified lasers,” Opt. Lett. 29(6), 635–637 (2004). [CrossRef]
8. J. L. Zhang, S. K. Shi, H. F. Jiao, X. C. Ji, Z. S. Wang, and X. B. Cheng, “Ultra-broadband reflector using double-layer subwavelength gratings,” Photonics Res. 8(3), 426 (2020). [CrossRef]
9. C. H. Zhou, “Chirped pulse amplification: review and prospective from diffractive optics,” Chin. Opt. Lett. 18(11), 110502 (2020). [CrossRef]
10. P. Poole, S. Trendafilov, G. Shvets, D. Smith, and E. Chowdhury, “Femtosecond laser damage threshold of pulse compression gratings for petawatt scale laser systems,” Opt. Express 21(22), 26341–26351 (2013). [CrossRef]
11. Y. C. Shao, Y. A. Zhao, H. Ma, C. Li, D. W. Li, C. Wang, Y. X. Leng, and J. D. Shao, “Influences of nanosecond pulse pre-irradiation on femtosecond laser damage resistance of gold pulse compression grating,” Opt. Commun. 461, 125258 (2020). [CrossRef]
12. L. Y. Xie, J. L. Zhang, Z. Y. Zhang, B. Ma, and X. B. Cheng, “Rectangular multilayer dielectric gratings with broadband high diffraction efficiency and enhanced laser damage resistance,” Opt. Express 29(2), 2669–2678 (2021). [CrossRef]
13. J. Neauport, E. Lavastre, G. Razé, G. Dupuy, and F. Desserouer, “Effect of electric field on laser induced damage threshold of multilayer dielectric gratings,” Opt. Express 15(19), 12508–12522 (2007). [CrossRef]
14. Z. L. Li and Z. L. Xia, “The damage mechanism and process of metal multi-layer dielectric gratings induced by ps-pulsed laser,” Appl. Surf. Sci. 494, 977–982 (2019). [CrossRef]
15. F. Y. Kong, Y. X. Jin, D. W. Li, W. X. Chen, M. P. Zhu, T. Wang, C. Y. Li, H. B. He, G. Xu, and J. D. Shao, “Effect of pulse duration on laser induced damage threshold of multilayer dielectric gratings,” Proc. SPIE 8530, 85300L (2012). [CrossRef]
16. K. P. Wu, Z. L. Xia, F. Y. Kong, and Y. X. Jin, “Nanosecond laser induced damage mechanism of multilayer dielectric gratings,” J. Optoelectron Adv. Mater. 20, 258–263 (2018).
17. M. Sozet, J. Néauport, E. Lavastre, N. Roquin, L. Gallais, and L. Lamaignère, “Laser damage density measurement of optical components in the sub-picosecond regime,” Opt. Lett. 40(9), 2091–2094 (2015). [CrossRef]
18. M. R. Borden, J. A. Folta, C. J. Stolz, J. R. Taylor, J. E. Wolfe, A. J. Griffin, and M. D. Thomas, “Improved method for laser damage testing coated optics,” Proc. SPIE 5991, 59912A (2005). [CrossRef]
19. X. F. Liu, D. W. Li, Y. A. Zhao, X. Li, and J. D. Shao, “Characteristics of nodular defect in HfO2/SiO2 multilayer optical coatings,” Appl. Surf. Sci. 256(12), 3783–3788 (2010). [CrossRef]
20. C. J. Stolz, J. E. Wolfe, J. J. Adams, M. G. Menor, N. E. Teslich, P. B. Mirkarimi, J. A. Folta, R. Soufli, C. S. Menoni, and D. Patel, “High laser-resistant multilayer mirrors by nodular defect planarization [Invited],” Appl. Opt. 53(4), A291–296 (2014). [CrossRef]
21. W. J. Kong, Z. C. Shen, J. Shen, J. D. Shao, and Z. X. Fan, “Investigation of laser-induced damage on multi-layer dielectric gratings,” Chin. Phys. Lett. 22(7), 1757–1760 (2005). [CrossRef]
22. F. Genin and C. Stolz, “Morphologies of laser-induced damage in hafnia-silica multilayer mirror and polarizer coatings,” Proc. SPIE 2870, 439–448 (1996). [CrossRef]
23. J. R. Schmidt, M. J. Runkel, K. E. Martin, and C. J. Stolz, “Scattering-induced downstream beam modulation by plasma scalded mirrors,” Proc. SPIE 6720, 67201H (2007). [CrossRef]
24. B. Ashe, K. L. Marshall, D. Mastrosimone, and C. McAtee, “Minimizing contamination to multilayer dielectric diffraction gratings within a large vacuum system,” Proc. SPIE 7069, 706902 (2008). [CrossRef]
25. I. Jovanovic, C. G. Brown, B. C. Stuart, W. A. Molander, N. D. Nielsen, B. Wattellier, J. A. Britten, D. M. Pennington, and C. P. J. Barty, “Precision damage tests of multilayer dielectric gratings for high-energy petawatt lasers,” Proc. SPIE 5647, 34–42 (2005). [CrossRef]
26. H. P. Howard, A. F. Aiello, J. G. Dressler, N. R. Edwards, T. J. Kessler, A. A. Kozlov, I. Manwaring, K. L. Marshall, J. B. Oliver, and S. Papernov, “Improving the performance of high-laser-damage-threshold, multilayer dielectric pulse-compression gratings through low-temperature chemical cleaning,” Appl. Opt. 52(8), 1682–1692 (2013). [CrossRef]
27. B. Ashe, K. L. Marshall, C. Giacofei, A. L. Rigatti, T. J. Kessler, A. W. Schmid, J. B. Oliver, J. Keck, and A. Kozlov, “Evaluation of cleaning methods for multilayer diffraction gratings,” Proc. SPIE 6403, 64030O (2006). [CrossRef]
28. X. F. Liu, D. W. Li, Y. A. Zhao, and X. Li, “Further investigation of the characteristics of nodular defects,” Appl. Opt. 49(10), 1774–1779 (2010). [CrossRef]
