Open Access
Add to BibSonomy
Abstract
The traditional polishing method inevitably results in subsurface cracks in the fused silica, which seriously degrades their ultraviolet laser damage resistance. CO2 laser irradiation can melt these cracks and improve their laser induced damage threshold (LIDT). Photoluminescence spectrum and SEM-FIB were employed to investigate the changes in the material microstructure at the crack location with CO2 laser melting. The density of the oxygen-deficient centers of type II (ODC II) defects decreases, while the density of the non-bridging oxygen hole center (NBOHC) defects increases after high-temperature melting. The reason for this change is related to the dihydroxylation reaction and the participation of environmental oxygen in the defect type conversion. The reduction of ODC II defects is most likely the reason for the improvement of LIDT.
© 2024 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Fused silica components are important optical components for ultraviolet lasers. Traditional mechanical processing methods inevitably cause subsurface cracks in fused silica. Cracks can cause a reduction in laser damage threshold in multiple ways. Early studies believed that cracks cause field enhancement [1,2], contain absorptive contaminant [3,4] and weaken mechanical strength [5]. Recently, research supposed that the sub-bandgap absorption induced by point defects at crack region should be responsible for laser damage [6,7]. These point defects provide additional initial excited state electrons, making the material more susceptible to laser energy absorption through sub-bandgap ionization, thereby causing damage. There are mainly two ways to eliminate point defect in cracks. One is to directly remove the point defect layer at the crack by acid etching. This method is simple and efficient, but it is insufficient for deeper sub-surface cracks. Another way is to change the material structure by high temperature, resulting in the annihilation of point defects and macroscopically manifesting as crack healing, such as laser annealing. Silica exhibits strong absorption light in the mid-to-far infrared long-wave band, producing significant thermal effects. Therefore, the 10.6 µm wavelength CO2 laser is suitable as a light source for melting, such as laser polishing [8–10,13] and laser damage mitigation process [11–15]. Studies have demonstrated that CO2 lasers can effectively repair cracks to significantly enhance its damage threshold [16,17]. However, there are relatively few studies on the evolution of point defects at crack locations and their effects on laser resistance [18]. The objective of this work is to better understanding of the defect annihilation mechanism after CO2 laser melting. We identified the types and distribution characteristics of point defects in cracks through photoluminescence (PL) spectra, discovered the changes in the distribution of point defect types before and after laser melting, and their underlying physical mechanisms.
2. Experimental method
We use artificial scratches to simulate sub-surface crack defects that occur in actual machining. The material of the samples is JGS-1 fused silica. All samples are mechanically ground and polished to a surface roughness of less than 1 nm, and then etched with 5% HF acid for 3 µm to eliminate the influence of contaminants in the surface redeposition layer on the experimental results. After completing the sample polishing and cleaning, a nano-indentation instrument with a conical indenter is used to prepare scratches on the sample surface. Two different loads of 0.75 N and 1.5 N were used during the scratch preparation process to obtain sub-surface cracks with different geometric shapes. The specific scratch preparation parameters are shown in Table 1. CO2 laser was used to melt the artificial scratches. Temperature monitoring during the melting process was achieved through a high-sensitivity infrared camera. To ensure sufficient fusion of the scratches, the center temperature of the laser spot was maintained at 2000°C. The CO2 laser parameters and scanning parameters in the experiment are shown in Fig. 1 and Table 2. Optical microscopy, SEM-FIB and fluorescence excitation spectroscopy other technical means are used to characterize the structural changes of the scratches before and after CO2 laser melting. The excitation laser wavelength used in the fluorescence excitation spectrum test is 266 nm. The spectrum analysis uses a grating with a groove density of 150 l/mm, and the center wavelength is set at 500 nm. To ensure enough fluorescence intensity is collected, the exposure time for each spectrum capture is set at 5 s. It should be noted that all fluorescence spectral curves in this experiment are the raw results after subtracting background noise and have not undergone any smoothing treatment.
Fig. 1. Intensity distribution of CO2 laser on the sample surface. (a) 2D intensity distribution and (b) intensity distribution along the X and Y axes, respectively.
Table 1. Parameters for scratch preparation
Table 2. Main parameters of CO2 laser used for melting the scratch
3. Experimental results
3.1 Healing effect of CO2 laser melting on artificial scratch
The Fig. 2 shows a comparison between the sub-surface morphology of artificially made scratches and actual machining defects. It can be observed that when the load is 0.75 N, the scratched surface appears as a well-defined groove with a smooth inner wall (as shown in Fig. 2(a)). However, upon observing the SEM images of the scratch cross-section, it was found that there existed an arc-shaped crack of depth around 10 µm (as indicated by the red dashed line in Fig. 2(b)). This is consistent with the larger sub-surface crack depths observed in actual polished components, as shown in Fig. 2(c). When the load was increased to 1.5 N, the extent of cracking inside the scratch increased, with the maximum crack depth reaching around 40 µm (as shown in Fig. 2(e)). At this load, the internal cracks extended to the surface, leading to fracture and fragmentation of the scratched surface along the sliding direction (as shown in Fig. 2(d)). Therefore, the depths and morphologies of scratches made using 0.75 N and 1.5 N loads match well with the sub-surface defects generated after traditional polishing and grinding, respectively.
Fig. 2. Comparison of artificial scratches and actual processed sub-surface defects in morphology. (a) Frontal microscopic image of a 0.75 N scratch, (b) crack distribution inside the 0.75 N scratch, (c) cross-sectional morphology of the polished sub-surface defect of fused silica, (d) Frontal microscopic image of a 1.5 N scratch, (e) crack distribution inside the 1.5 N scratch, (f) cross-sectional morphology of the ground sub-surface defect of fused silica.
After characterizing the depth and morphology of the artificial scratch cracks, we conducted melting process on the scratches using the CO2 laser parameters in Table 2. To verify the melting effect, we observed the healing of surface and internal cracks before and after melting using SEM-FIB. Figure 3 shows the melting results of a 1.5 N scratch after CO2 laser irradiation. Figure 3(a) is a schematic diagram of the CO2 laser scanning path and the relative position of the scratch. The CO2 laser scans through the middle of each scratch. The dotted line in Fig. 3(b) is the boundary line of the CO2 laser scanning area. It can be clearly seen that the two scratches gradually merge from left to right under the melting effect of the CO2 laser. Specifically, as shown in Fig. 3(c), there are a large number of fracture features on the area that has not been melted by the CO2 laser. As we transition to the melted area, these fracture features gradually disappear, and the scratch surface becomes smoother. In boundary of melting area, most of the cracks can be fused, and a few unfused cracks are occasionally observed, as shown in Fig. 3(d). In the fully melted area, no crack residue was observed inside the original scratch position, and all cracks were fused. The scratch surface is similar to the surrounding material, as shown in Fig. 3(e). Therefore, it can be concluded that the CO2 laser melting at a temperature of 2000°C can fuse the 1.5 N load scratch. As the subsurface cracks of the 0.75 N load scratch are shallower, it can be reasonably inferred that the 0.75 N scratch can also be completely fused under the same CO2 laser parameters.
Fig. 3. Microscopic images of the internal structure of scratches at different locations: (a) schematic diagram of CO2 laser scanning path for scratch, (b) surface morphology of the melted scratch after laser irradiation, (c) position without CO2 laser melting, (d) boundary position between CO2 laser melted and non-melted regions, and (e) position with CO2 laser melting.
3.2 Point defect distribution at crack sites
During the CO2 laser melting process, in addition to the visible crack healing caused by material melting and flowing, the changes in its micro-structure are also one of our points of concern. Numerous studies have shown that various types of point defects are generated at crack defect locations due to the fracture of the silicon-oxygen network, which causes defect-states to appear in the forbidden band of silica, leading to ionization transitions of ground-state electrons under low-flux laser irradiation and ultimately resulting in material damage. By analyzing the scratch PL spectra before and after CO2 laser irradiation, the distribution and density changes of point defect types can be identified. Figure 4 shows the PL spectra result. The PL spectrum of scratches under 266 nm laser excitation is a dual-band structure. The first band is located near 430 nm, and its intensity may be related to the density of known oxygen deficiency centers (ODC II, 2.7 eV ∼ 2.8 eV [19,20]) point defects; the second band is located near 650 nm, and its intensity is related to the density of non-bridging oxygen hole centers (NBOHC, 1.85 eV ∼ 1.95 eV [21]) point defects. The band intensity at 430 nm is greater than that at 650 nm, indicating that the main contribution to fluorescence comes from ODC II defects. Difference from Wong et al [22], there are also a large number of ODC II defects at the fracture region. There are numerous cracks and broken materials on the scratched surface and subsurface. Point defects are enriched at the broken structures, emitting characteristic fluorescence under excitation.
As the scratched material was melted and cracks were healed by CO2 laser high-temperature heating, its PL spectrum also underwent prominent changes (Fig. 5): the intensity of the 430 nm-band decreased, corresponding to the decrease in the density of ODC II point defects; while the intensity of the 650 nm-band increased, corresponding to the increase in the density of NBOHC point defects. It can be seen that CO2 laser has a significant effect on eliminating ODC II type defects, but the side effect is an increase in the number of NBOHC defects.
Fig. 5. Emission spectra comparison of scratched area ((a): 1.5 N, (b): 0.75 N) and unscratched area (c) before and after CO2 laser melting.
In order to analyze the effect of CO2 laser on scratch defects more accurately, we collected fluorescence spectra of scratches at different positions after CO2 laser melting, and extracted the fluorescence intensities at 460 ± 1 nm and 652 ± 1 nm. Assuming that the intensities at these wavelengths are proportional to the density of ODC II and NBOHC point defects, respectively, we can indirectly obtain the distribution of ODC II and NBOHC point defects at different positions. The results are shown in Fig. 6, where the horizontal axis represents the different positions of the scratch and the vertical axis represents the fluorescence intensity at the 460 nm or 652 nm position. The gray area in the Fig. represents the CO2 laser melting area, with a width of 4.5 mm (the size of the CO2 laser spot is about 4 mm), located near the center of the scratch. The total length of the scratch is 10 mm, and spectra sampling tests were performed at intervals of 0.5 mm.
Fig. 6. Fluorescence intensity at 460 nm and 652 nm wavelengths at different positions of scratches after CO2 laser melting.
The results show that the degree of change in point defect density is closely related to the fusion effect of cracks. The temperature at the center position of the CO2 laser spot can reach 2000°C, which is higher than the melting point of the material (about 1750°C), resulting in a flow of material and a significant reduction in ODC II point defect density and increase in NBOHC point defect density. The temperature at the edge of the CO2 laser spot is below 2000°C, and its value may be near the softening point (about 1400°C). The material flow is insufficient, and some cracks are not fused (as shown in Fig. 3(c)), so the change in point defect density is not as significant as in the central area. When moving further away from the CO2 laser spot, the material temperature drops to several hundred degrees Celsius, which is insufficient to cause material flow, and there is no sign of crack fusion (as shown in Fig. 3(d)). However, the low-temperature annealing effects may still have an impact on the distribution of point defect density in this area.
Table 3 provides a quantitative analysis of the effect of CO2 laser melting on the type and density distribution of point defects. For ODC II point defects, the PL intensity increases with increasing indentation load: 204 ± 76 with no scratch, 420 ± 99 with 0.75 N scratch, and 1633 ± 285 with 1.5 N scratch. However, after CO2 laser melting, it decreased to a range of 200∼240 for all cases. Similarly, for NBOHC point defects, the PL density also increases with increasing indentation load: 105 ± 25 with no scratch, 115 ± 34 with 0.75 N scratch, and 236 ± 69 with 1.5 N scratch. However, after CO2 laser melting, it increased instead, with the final value being greater for the 1.5 N scratch than for the 0.75 N scratch.
Table 3. Comparison of fluorescence intensities of point defects before and after CO2 laser melting for scratches with different loads
4. Discussion
In the fluorescence spectrum measurement, a change was observed in which the density of ODC II point defects at the scratch location decreased and the density of NBOHC point defects increased after CO2 laser melting. We suppose that there may be three physical processes that have led to this phenomenon. The first physical process names the transformation of point defect types, which can be described as follows: CO2 laser heats the material to near the melting point, and under high temperature, active oxygen atoms in the environment enter the silicon-oxygen network and bond with ODC II point defects ($= Si:$). If both two shell electrons of active oxygen atoms bond with silicon atoms in ODC II, a normal $= Si = $ structure is formed, the density of ODC II-point defects decreases, and the silicon-oxygen network is repaired. In the meanwhile, if only one shell electron of active oxygen atoms bonds with silicon atoms in ODC II, and the other electron remains unbound, a ${\equiv} Si - O \cdot $ structure is formed, named NBOHC point defect. This process can be explained by Eq. (1). At this condition, the density of ODC II point defects decreases, and the density of NBOHC point defects increases.
The core assumption of this is that O2 in the ambient atmosphere participates in the point defect type conversion reaction. To verify it, we placed the sample in a low vacuum chamber (<10 Pa) to eliminate the influence of ambient O2, and repeated the CO2 laser melting experiment. The result is shown in Fig. 7. It can be found that in vacuum environment, there was little increase in fluorescence spectrum at the 652 nm position in the area melted by CO2 laser. This means that the process of ODC II converting to NBOHC is greatly inhibited due to a lack of environmental oxygen, and the content of NBOHC no longer increases significantly. In addition, it can be observed that the number of ODC defects increases significantly after high-temperature melting under vacuum conditions. ODC II is a typical oxygen-deficiency-related defect. The negative pressure environment is more likely to cause the escape of oxygen atoms, resulting in an increase in the number of ODC II.
Fig. 7. Comparison of PL spectra of non-scratch areas in vacuum environments before and after CO2 laser melting
Another physical process is the generation of point defects due to the breaking of silicon-oxygen bonds in the network. At high temperatures, the vibrations of the silicon-oxygen network atoms intensify significantly, causing some silicon-oxygen bonds to break, directly forming NBOHC and E’ center (${\equiv} Si \cdot $). defects pairs. The highest temperature generated by CO2 laser is around 2000°C, and NBOHC is hard to further convert to ODC II. This mechanism results in an increase in NBOHC density while ODC II density remains mostly unchanged or slightly increased. This process can be described by Eq. (2):
The third physical process is named dihydroxylation. There is a certain concentration of hydroxyl groups in the fused silica material, which will undergo dihydroxylation reaction at high temperatures as shown in Eq. (3). One product of the reaction is NBOHC. With the reduction of hydroxyl groups after laser melting, the NBOHC point defects will increase.
This hypothesis was validated through Fourier infrared spectroscopy. Figure 8 shows the comparison of transmission spectra curves before and after CO2 laser melting. The thickness of the test sample is 1 mm. The main absorption band of the hydroxyl group in the fused silica is 3673 cm-1. After CO2 laser melting, with the reduction of hydroxyl group content due to dihydroxylation, the transmittance at 3673 cm-1 increases from 75.77% to 83.52% in the same region.
Finally, we will discuss the relationship between laser damage and point defects. Laser photons transfer energy to valence electrons of the material atoms, causing the electrons to transition to higher energy levels. When the excited electrons return to the ground state, the energy relaxation can be realized through radiative and non-radiative path. The fluorescence lifetime ${\tau _{PL}}$ is determined by the radiative transition rate ${k_{PL}}$ and the non-radiative transition rate ${k_{NR}}$, as described by Eq. (4).
Defect-induced laser damage is essentially a point explosion behavior that occurs when defects absorb laser energy and reach a certain temperature. Therefore, during the laser pulse irradiation process, the laser energy absorbed by defects must mainly depend on non-radiative transitions to cause a rapid temperature rise. This process is similar to the case of continuous excitation of electrons in metal. At this condition, $k_{NR}\gg k_{PL}$, $\tau_{PL}\approx1/k_{NR}$, and ${\tau _{PL}}$ should be very short and the intensity should be very weak. Typically, the fluorescence lifetime of point defects such as NBOHC in fused silica materials is 10 μs~20 μs [23, 24], with a low non-radiative transition rate. On the other hand, ODC II has a short life component in its fluorescence lifetime, with a value of about 4 ns [25], and a high non-radiative transition rate, which contributes to lattice heating during nanosecond laser pulse action [26]. Therefore, ODC II is the key type of point defect leading to laser damage.
5. Conclusion
In summary, this paper investigates the influence of CO2 laser melting on point defects located in sub-surface crack of fused silica. Fluorescence characterization of artificial scratch cracks and in-situ laser damage testing experiments were conducted. The results are as follows: Choosing appropriate CO2 laser parameters can result in good healing of sub-surface cracks up to a maximum depth of 40 µm without material ablation. During the high-temperature melting process of cracks, O2 in the environment participates in the point defect type conversion, resulting in a decrease in ODC II density and an increase in NBOHC density. In addition, the dehydroxylation reaction at high temperatures also promotes the increase in NBOHC density.
Funding
Outstanding Youth Talents Project of China (2017-JCJQ-ZQ-024).
Acknowledgments
The work of this article was completed with the support of the funding of Outstanding Youth Talents Project of China (2017-JCJQ-ZQ-024).
Disclosures
The authors declare 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. F. Y. Génin, A. Salleo, and T. V. Pistor, “Role of light intensification by cracks in optical breakdown on sub-surfaces,” J. Opt. Soc. Am. A 18(10), 2607–2616 (2001). [CrossRef]
2. N. Bloembergen, “Role of cracks, pores and absorbing inclusions on laser induced damage threshold at surfaces of transparent dielectrics,” Appl. Opt. 12(4), 661–664 (1973). [CrossRef]
3. F. Bonneau, P. Combis, J. L. Rullier, A. During, J. Y. Natoli, M. J. Pellin, M. R. Savina, E. Cottancin, and M. Pellarin, “Observation by photothermal microscopy of increased silica absorption in laser damage induced by gold nanoparticles,” Appl. Phys. Lett. 83(19), 3855–3857 (2003). [CrossRef]
4. J. Y. Natoli, L. Gallais, B. Bertussi, D. During, and P. Combis, “Localized pulsed laser interaction with submicronic gold particles embedded in silica: a method for investigating laser damage initiation,” Opt. Express 11(7), 824 (2003). [CrossRef]
5. S. Wang, J. Wang, X. Lei, Z. Liu, and Q. Xu, “Simulation of the nanosecond-pulse laser damage of KDP surface by the smoothed particle hydrodynamics method,” Opt. Lett. 44(21), 5338 (2019). [CrossRef]
6. P. E. Miller, J. D. Bude, T. I. Suratwala, N. Shen, and T. A. Laurence, “Fracture-induced sub-bandgap absorption as a precursor to optical damage on fused silica surface,” Opt. Lett. 35(16), 2702 (2010). [CrossRef]
7. D. Yang, J. Cheng, L. Zhao, M. Chen, H. Liu, J. Wang, C. Han, Z. Liu, S. Wang, F. Geng, Y. Sun, and Q. Xu, “Evolution of the point defects involved under the action of mechanical forces in mechanically machined fused silica surfaces,” Opt. Express 31(5), 7684–7706 (2023). [CrossRef]
8. S. Heidrich, A. Richmann, P. Schmitz, E. Willenborg, K. Wissenbach, P. Loosen, and R. Poprawe, “Optics manufacturing by laser radiation,” Optics & Lasers in Engineering 59, 34–40 (2014). [CrossRef]
9. W. Christian, U. Emrah, S. Andreas, et al., “Glass processing with pulsed CO2 laser radiation,” Appl. Opt. 56(4), 777–783 (2017). [CrossRef]
10. S. Heidrich, E. Willenborg, C. Weingarten, et al., “Laser polishing and laser form correction of fused silica optics,” Materialwissenschaft Und Werkstofftechnik 46(7), 668–674 (2015). [CrossRef]
11. C. Tan, L. Zhao, M. Chen, et al., “Experimental and theoretical investigation of localized CO2 laser interaction with fused silica during the process of surface damage mitigation,” Results Phys. 16, 102936 (2020). [CrossRef]
12. M. D. Feit and A. M. Rubenchik, “Mechanisms of CO2 laser mitigation of laser damage growth in fused silica,” Laser-Induced Damage in Optical Matls. Proc. of SPIE4932 (2002).
13. M. Matthews J, T. Steven, Nan Shen, et al., “Micro-Shaping, Polishing, and Damage Repair of Fused Silica Surfaces Using Focused Infrared Laser Beams,” Adv. Eng. Mater. 17(3), 247–252 (2015). [CrossRef]
14. L. Gallais, P. Cormont, and L. Rullier J, “Investigation of stress induced by CO2 laser processing of fused silica optics for laser damage growth mitigation,” Opt. Express 17(26), 23488–23501 (2009). [CrossRef]
15. Z. Fang, Y. Zhao, W. Sun, et al., “CO2 laser mitigation of the ultraviolet laser damage site on a fused silica surface,” Opt. Eng. 53(8), 084107 (2014). [CrossRef]
16. P. Cormont, A. Bourgeade, S. Cavaro, et al., “Relevance of carbon dioxide laser to remove scratches on large fused silica polished optics,” Adv. Eng. Mater. 17(3), 253–259 (2015). [CrossRef]
17. J. Bude, P. Miller, S. Baxamusa, et al., “High fluence laser damage precursors and their mitigation in fused silica,” Opt. Express 22(5), 5839–5851 (2014). [CrossRef]
18. R. N. Raman, M. J. Matthews, J. J. Adams, et al., “Monitoring annealing via CO2 laser heating of defect populations on fused silica surfaces using photoluminescence microscopy,” Opt. Express 18(14), 15207 (2010). [CrossRef]
19. S. O. Kucheyev and S. G. Demos, “Optical defects produced in fused silica during laser-induced breakdown,” Appl. Phys. Lett. 82(19), 3230–3232 (2003). [CrossRef]
20. Y. Sakurai and K. Nagasawa, “Green photoluminescence band in γ-irradiated oxygen-surplus silica glass,” J. Appl. Phys. 86(3), 1377–1381 (1999). [CrossRef]
21. L. Skuja, K. Tanimura, and N. Itoh, “Correlation between the radiation-induced intrinsic 4.8 eV optical absorption and 1.9 eV photoluminescence bands in glassy SiO2,” J. Appl. Phys. 80(6), 3518–3525 (1996). [CrossRef]
22. J. Wong, J. L. Ferriera, E. F. Lindsey, D. L. Haupt, I. D. Hutcheon, and J. H. Kinney, “Morphology and microstructure in fused silica induced by high fluence ultraviolet 3ω (355 nm) laser pulses,” J. Non-Cryst. Solids 352(3), 255–272 (2006). [CrossRef]
23. G. Pacchioni, L. Skuja, and D. L. Griscom, “Defects in SiO2 and related dielectrics: science and technology,” J. Non-Cryst. Solids 179(51), 5 (1994).
24. L. Skuja, T. Suzuki, and K. Tanimura, “Site-selective laser-spectroscopy studies of the intrinsic 1.9-eV luminescence center in glassy SiO2,” Phys. Rev. B 52(21), 15208–15216 (1995). [CrossRef]
25. L. N. Skuja, N. Streletsky A, and B. Pakovich A, “A new intrinsic defect in amorphous SiO2: Twofold coordinated silicon,” Solid State Commun. 50(12), 1069–1072 (1984). [CrossRef]
26. M. R. Beversluis, A. Bouhelier, and L. Novotny, “Continuum generation from single gold nanostructures through near-field mediated intraband transitions,” Phys. Rev. B 68(11), 115433 (2003). [CrossRef]
